Skip to main content

Full text of "Mediæval romance in England: a study of the sources and analogues of the noncyclic metrical romances"

See other formats 


UNIVERSITY 
OF  FLORIDA 
LIBRARIES 


Digitized  by  the  Internet  Archive 

in  2011  with  funding  from 

LYRASIS  Members  and  Sloan  Foundation 


http://www.archive.org/details/medivalromanceinOOIoom 


MEDIAEVAL    ROMANCE    IN 
ENGLAND 


MEDIAEVAL    ROMANCE   IN 

ENGLAND 


A  STUDY  OF  THE  SOURCES 
AND  ANALOGUES  OF  THE  NON-CYCLIC 
METRICAL  ROMANCES 


By 

LAURA  A.  HIBBARD 
(MRS.  LAURA  HIBBARD  LOOMIS) 

Formerly  Professor  of  English  Literature, 
Wellesley  College 


New  Edition  with  Supplementary 

Bibliographical  Index 

(1926-1959) 


Burt  Franklin  Bibliographical  and  Reference  Series  #  XVII 


BURT  FRANKLIN 

New  York  25,  N.  Y. 

1963 


The  first   edition   of  this  work   was  published   by   Oxford   University 
Press  (N.  Y.)   1924  and  copyrighted  by  Wellesley  College.  All  rights 

reserved. 


Published  by 

BURT  FRANKLIN 

514  West  113th  Street 
New  York  25,  N.  Y. 


"**<** 


Printed  in  the  United  States  of  America  by 

NOBLE  OFFSET  PRINTERS,  INC. 


TO 
THE  MEMORY  OF 

GERTRUDE  SCHOEPPERLE  LOOMIS 

"Truth  lies  hid  in  the  trappings  of  a  tale" 


PREFACE 

At  the  end  of  the  twelfth  century  there  were,  according  to 
Jean  Bodel,  but  the  three  "  matieres,"  of  France,  of  Britain, 
and  of  "  Rome  la  grant,"  of  which  men  wished  to  hear.  This 
observation  was  as  remarkably  wrong  as  are  most  of  the  gener- 
alizations offered  by  critics  concerning  the  literary  tastes  of 
their  own  time.  For  court  circles,  for  the  literary  elect,  the  fa- 
mous cycles  of  Carolingian,  of  Arthurian,  and  of  pseudo-classical 
tales,  were  certainly  the  fashion,  but  so,  likewise,  and  for  a  much 
longer  period,  if  we  may  judge  from  the  number  of  surviving 
manuscripts  and  allusions  to  stories  of  wholly  different  prove- 
nance, were  what  may  be  called  the  non-cyclic  romances.  These 
were  the  romans  d'aventure:  the  local  legends,  the  traditional 
tales,  which  poets  could  transform  into  romantic  guise.  Men 
who  used  materials  such  as  these  ranged  free  in  them  as  life  and 
language  and  taste  itself;  they  drew  at  will  on  the  great  store- 
house of  folk  lore,  of  local  legend,  of  religious  and  patriotic 
tradition,  —  to  say  nothing  of  individual  invention.  They  imi- 
tated the  style,  epic  or  romantic,  of  the  traditional  cycles,  and 
they  developed  to  the  full  the  formulas  of  speech  and  theme,  of 
incident  and  character,  in  short,  the  stock  materials  of  mediaeval 
fiction. 

Investigation  in  this  great  body  of  miscellaneous  romance  is 
no  new  thing.  Single  legends  and  special  themes  have  long  been 
studied  with  scrupulous  care.  In  the  case  of  certain  legends 
such  as  those  of  Crescentia  {Florence  of  Rome),  or  of  Con- 
stance, the  number  of  versions  discovered  has  been  so  large  that 
scholars  have  sometimes  referred  to  the  group  itself  as  a  saga 
or  a  cycle.  But  versions  of  these  stories  are  always  essentially 
the  same ;  they  have  none  of  that  continuity  of  character  and  of 
incident  which  is  found  in  a  series  of  romances  connected  with 
a  favorite  hero  like  Gawain  or  with  a  court  like  Charlemagne's. 
They  are  not  cyclic  in  any  true  sense  of  the  word,  but  they  do 
afford,  in  the  very  multiplicity  of  their  texts,  irrefutable  proof 

Hi 


iv  PREFACE 

of  their  mediaeval  popularity.  Of  group  study  of  these  non- 
cyclic  romances,  there  has  been,  however,  comparatively  little. 
Langlois  recognized  that  their  value  for  the  social  historian  was 
superior  to  that  of  the  cyclic  romances  and  he,  accordingly, 
based  his  study  of  La  Societe  Frangaise  au  XIII6  Steele  (3rd  ed., 
191 1 )  on  ten  rowans  d'aventure,  on  Joufroi,  Guillaume  de  Dole, 
L'Escoufle,  on  the  Chdtelain  de  Couci,  and  others.  In  her  study 
of  Le  Roman  Idyllique  dans  le  Moyen  Age  (1913),  Madame 
Lot-Borodine  made  use  of  Floire  et  Blanche flor  and  Guillaume 
de  Palerme  for  illustration  of  a  literary  type,  of  a  certain  style 
and  spirit,  that  can  hardly  be  paralleled  among  the  more  stand- 
ardized cycles  of  romantic  literature.  Deutschbein  in  his 
Sagengeschichte  Englands  (1906)  emphasized  the  historical  as- 
pects of  legends  that  culminated  in  such  romances  as  Horn, 
Havelok,  and  Guy  of  Warwick. 

The  amount  of  intensive  work  that  has  been  done  on  the  in- 
dividual non-cyclic  romances  and  the  lack  of  any  comprehensive 
effort  to  summarize  the  results  have  led  almost  equally  to  the 
present  undertaking.  For  the  purpose  of  selection  the  Middle 
English  romances  of  this  type  offered  a  convenient  and  somewhat 
neglected  group  with  which  to  begin.  Though  a  few  of  them 
are  relatively  unimportant,  others  are  versions  of  some  of  the 
most  famous  European  stories  of  the  Middle  Ages,  and  for  the 
sake  of  convenient  reference  it  has  seemed  well  to  include  the 
less  with  the  more  important.  Each  romance  is  treated  indi- 
vidually in  a  section,  one  part  of  which  deals  with  Versions,  and 
one  with  the  Origins  of  the  tale.  Under  Versions  I  have  at- 
tempted to  give  something  of  a  life  history  of  each  legend  by 
listing  all  of  the  literary  versions  which  were  composed  before 
1500,  and  by  indicating  so  far  as  possible  their  relationship.  In 
the  section  on  Origins  I  have  recorded  the  opinions  of  scholars 
on  the  historical  and  legendary  elements  that  gave  rise  to  the 
story  and  have  tried  to  set  forth  the  most  important  motifs 
which  characterize  the  different  versions.  The  Index  of  Matters 
and  Literature  at  the  close  of  the  book  is  designed  to  co-ordinate 
this  material  and  to  reveal  for  these  romances,  as  the  similar  in- 
dex in  Volume  V  of  Professor  Child's  monumental  edition  of 
The  English  and  Scottish  Ballads  revealed  for  the  ballads,  the 
recurrent   themes,   the   stock   situations,   characters,    incidents, 


PREFACE  v 

properties,  the  dominating  conceptions,  which  mark  the  favorite 
patterns  of  mediaeval  story-tellers  as  diverse  in  purpose  and 
ability  as  the  authors  of  these  romances. 

The  plan  and  scope  of  this  book  obviously  differentiate  it 
from  earlier  studies  on  Middle  English  romance.  Dr.  Billings's 
Guide  to  the  Middle  English  Metrical  Romances,  1901,  1905, 
the  plan  of  which  was  in  part  suggested  by  Korting's  Grundriss 
der  Geschichte  der  engl.  Literatur  (3rd  ed.,  1899),  took  up  only 
seven  of  the  romances  treated  here.  Professor  Wells's  invalu- 
able Manual  of  the  Writings  in  Middle  English  (1916)  treated 
all  but  two.  The  fact,  however,  that  he  was  surveying  the  whole 
of  Middle  English  literature  naturally  limited  his  discussion  of 
each  romance  to  the  briefest  possible  statement.  Especially  is 
this  true  of  his  reference  to  its  sources  and  history,  the  two  as- 
pects which  are  chiefly  developed  here.  Between  his  book  and 
this  there  is  a  certain  necessary  duplication  of  information 
concerning  the  manuscripts,  dates,  and  dialects,  of  the  Middle 
English  texts,  but  in  each  case  the  work  is  of  wholly  independent 
character.  In  the  present  instance  all  this  material  was  pre- 
sented as  a  doctoral  dissertation  to  the  University  of  Chicago 
in  19 1 6,  shortly  before  the  publication  of  the  Manual.  Since 
then  the  text  has  been  amplified,  as  has  the  whole  bibliography, 
by  the  inclusion  of  references  extending  through  January,  1923. 

A  word  must  be  said  concerning  a  departure  here  in  biblio- 
graphical method,  the  wisdom  of  which  remains  to  be  proved. 
In  general  the  bibliographies  appended  to  each  romance  have 
been  confined  to  studies  published  since  1900.  The  amazing 
volume  and  progress  of  investigations  in  this  field  is  thus  made 
plain,  and  the  overweighting  of  these  lists  with  much  that  is 
now  discredited,  is  avoided.  On  the  other  hand,  it  is  hoped  that 
no  earlier  study  of  enduring  worth  has  been  neglected,  for  to 
such  works  full  reference  is  made  in  the  course  of  the  appro- 
priate discussion  or  in  supplementary  bibliographical  lists.  In 
these  discussions  a  citation  with  the  author's  name  and  page 
reference  only,  indicates  a  work  published  since  1900  for  which 
complete  information  is  given  in  the  final,  alphabetically  ar- 
ranged bibliography  at  the  end  of  each  section.  Works  of 
general  reference,  cited  by  the  author's  name  and  a  brief  title, 
are  fully  described  in  the  Table  of  Abbreviations. 


Vi  PREFACE 

"  The  fruit  of  every  tale  is  for  to  seye  "  one's  thanks.  My 
own  go  most  warmly  to  the  librarians  of  the  British  Museum, 
of  Harvard  University,  of  Wellesley,  and  of  the  University  of 
Chicago,  who  courteously  facilitated  my  work;  to  Professor 
Manly  and  to  Professor  Nitze  of  the  University  of  Chicago  who 
gave  valuable  counsel  on  my  dissertation;  to  my  colleagues, 
Professor  Martha  Hale  Shackford  and  Miss  Anne  K.  Tuell; 
and  to  my  friend  and  comrade,  Gertrude  Schoepperle  Loomis, 
who  followed  this  study  through  all  but  its  latest  stages.  To 
the  memory  of  her  who  was  as  distinguished  an  Arthurian 
scholar  as  she  was  keen  a  lover  of  all  romance,  the  book  is  now 
dedicated. 

In  conclusion  I  wish  to  express  my  grateful  appreciation  for 
the  honor  of  having  this  volume  included  in  the  Wellesley  College 
Semi-Centennial  Series. 

Laura  A.  Hibbard 


CONTENTS 

PAGE 

I.  Romances  of  Trial  and  Faith 

Isumbras 3 

Florence  of  Rome 12 

Emare 23 

Erie  of  Tolous 35 

King  of  Tars 45 

Gowther  (Robert  the  Devil) 49 

Robert  of  Cisyle 58 

Amis  and  Amiloun 65 

Amadas      73 

Cleges 79 

II.  Romances  of  Legendary  English  Heroes 

King  Horn 83 

Horn  Childe 87 

Havelok  the  Dane 103 

Beves  of  Hampton 115 

Guy  of  Warwick 127 

Reinbrun 140 

Athelston 143 

Richard  Coeur  de  Lion 147 

Gamelyn 156 

III.  Romances  of  Love  and  Adventure 

Apollonius  of  Tyre 164 

Seven  Sages  of  Rome 174 

Floris  and  Blauncheflur 184 

Orfeo      195 

Partonope  of  Blois 200 

William  of  Palerne 214 

Ipomedon      224 

Generides 231 

Chevalere  Assigne  (Knight  of  the  Swan) 239 

Knight  of  Courtesy  (Chatelain  de  Couci) 253 

Squyr  of  Lowe  Degre 263 

vii 


viii  CONTENTS 

PAGE 

Octavian 267 

Eglamour 274 

Torrent  of  Portyngale 279 

Triamour 283 

Roswall  and  Lillian 290 

Lay  le  Freine 294 

Degare 301 

Degrevant 307 

Eger,  Grime,  and  Graysteele 312 

Table  of  Abbreviations  and  References 321 

Index  of  Matters  and  Literature 327 

I  Author's  note  to  new  edition  343 

II  Reviews  of  Medieval  Romance  in  England  345 

III  General  Bibliographies  346 

IV  Index  of  Non-Cyclic  Romances  and  Bibliographical 

references  (1926-1959)  347 


MEDIAEVAL  ROMANCE 
IN  ENGLAND 

I.    ROMANCES   OF   TRIAL   AND    FAITH 

SIR   ISUMBRAS 

Versions.  The  mediaeval  mind  was  not  often  free  from  some 
preoccupation  with  spiritual  destiny.  The  problem  of  salvation 
was  immediate  and  terrible,  and  the  nature  of  extremest  sin 
and  extremest  virtue  was  frequently  a  subject  for  frightened 
speculation.  Out  of  such  meditations,  which  had  in  them 
much  that  was  akin  to  Eastern  habits  of  thought,  grew  the 
legend  of  the  Man  Tried  by  Fate.  Its  hero  was  Job-like; 
he  was  suddenly  bereft  of  home  and  wealth  and  family,  but 
he  lived  with  uncomplaining  patience  until  he  was  at  last  re- 
possessed of  all  that  he  had  lost.  The  theme  made  an  obvious 
appeal  to  the  pious  and  to  the  fatalistic,  and  its  development 
almost  inevitably  suggested  romantic  possibilities.  Gerould 
(pp.  354-72)  has  listed  eight  mediaeval  literary  versions  belong- 
ing to  western  Europe,  and  of  these  all  but  two,  the  various 
texts  of  the  famous  legend  of  Saint  Eustache,  and  the  exemplum 
in  the  Gesta  Romanorum  (EETSES,  xxxiii,  87),  are  highly 
developed  romances.  They  include  the  long  twelfth-century 
French  romance  Guillaume  d'Angleterre  (ed.  Foerster,  Roman 
Bibliothek,  xx,  Halle,  191 1) ;  a  Middle  High  German  romance, 
Wilhelm  von  Wenden  (Toischer,  Prague,  1876),  by  Ulrich 
von  Eschenbach,  who  wrote  between  1287  and  1297,  a  thir- 
teenth-century Swabian  romance,  Die  Gute  Frau  (Sommer,  Zts. 
f.  deut.  Alterthum,  11,  385  ff.) ;  the  early  fourteenth-century 
Spanish  romance,  El  Cavalier 0  Cijar  (Michelant,  Tubingen, 
1872),  the  Middle  English  romance  of  Sir  Isumbras,  and  a  Ger- 
man poem  of  the  fifteenth  century,  Der  Graf  von  Savoien 
(Eschenburg,  Denkmaler,  1799,  p.  347).    To  the  same  group 

3 


4  MEDIEVAL   ROMANCE    IN   ENGLAND 

belong  the  folk  versions,  the  Danish  ballad  of  Sakarias 
(Grundtvig,  n,  605),  the  Breton  ballads  of  the  King  of  Romani 
(Luzel,  Chants  populaires,  1868,  1,  179  ff.),  and  the  less  truly- 
popular  German  folk-song  about  St.  Hubert  (Simrock,  Deut. 
Sagen,  p.  46). 

In  comparison  with  many  of  these  versions  Isumbras  occupies 
a  modest  place.  It  contains  804  lines,  written  in  the  twelve-line 
tail-rime  stanza  and  in  well-worn  minstrel  phrases.  Schleich 
(p.  97)  believed  it  must  have  originated  in  the  northeastern 
Midland  district,  and  as  it  is  mentioned  in  the  Cursor  Mundi 
it  must  have  attained  some  popularity  before  1320.  It  survives 
in  one  late  fourteenth-century  manuscript,  that  of  Caius  Col- 
lege, in  five  manuscripts  of  the  fifteenth  and  one  of  the  six- 
teenth century,  as  well  as  in  William  Copland's  undated  edition 
and  two  fragmentary  prints. 

Though  the  condensation  of  the  story  emphasizes  some  "  pre- 
posterous "  elements  in  the  original  ecclesiastical  romance,  and 
though  the  humble  minstrel  author  of  Isumbras  makes  no  pre- 
tenses at  originality  of  diction,  the  poem  is  not  without  redeem- 
ing touches.  There  is  occasionally  a  blunt  realism  about  it. 
Thus  we  are  told,  after  the  hero's  troubles  begin,  of  the  dreary 
"  hirdemen  "  who  accost  Isumbras  with  the  news  that  there  is 
not  left  "  a  stotte  unto  youre  plowghe " ;  we  hear  that  the 
wretched  family  of  the  once  proud  nobleman  suffer  in  the 
wood  where  they  see  nothing  that  "  come  of  corn  "  but  only, 
on  "  the  holtes  hare,"  "  the  floures  of  the  thorns."  In  many  ver- 
sions of  the  original  legend,  the  hero  performs  manual  labor 
during  the  period  of  his  humiliation,  but  it  is  only  the  North 
Midland  poet  of  Isumbras  who  makes  his  hero  labor  in  a 
quarry  as  a  "  smethyman  "  carrying  out  "  irynstone,"  and  blow- 
ing the  bellows  at  a  yeoman's  hire  until  at  last  he  himself 
becomes  a  "  smethyman "  able  to  forge  his  own  "  grymly 
growndyn  gare  "  and  a  complete  suit  of  armor.  With  typical 
English  delight  in  physical  prowess  the  poet  describes  how 
Isumbras  can  put  the  stone  and  how,  when  some  jesters  at  court, 
thinking  him  only  a  poor  palmer,  mount  him  on  "  ane  crokede 
stede  "  and  send  him  to  a  tournament,  he  does  so  mightily  that 
the  queen  laughs  out,  "  My  palmere  es  styffe  enoghe,  —  he  is 
worthi  to  fede!  " 


SIR    ISUMBRAS  5 

Origin.  It  is  with  the  earliest  western  literary  version,  i.e. 
the  legend  of  St.  Eustache,  that  the  study  of  origins  must  begin. 
Placidus,  according  to  the  story,  is  a  Roman  general  who  in 
the  time  of  Trajan  is  converted  by  the  sight  of  a  crucifix  which 
appears  between  the  horns  of  a  stag  which  he  is  hunting.  He 
is  told  by  the  stag  that  he  must  be  tried  by  sufferings.  Mis- 
fortunes immediately  overtake  him;  he  loses  his  servants  and 
cattle  through  pestilence,  and  his  wife  and  children  through 
robbery,  sailors  taking  his  wife  and  wild  animals  his  children. 
For  fifteen  years  he  works  as  a  laborer  and  then  again  becomes 
a  commander  and  is  served  by  his  own  two  sons  who  are  ig- 
norant of  their  parentage.  Their  mother,  happening  to  lodge 
the  two  boys,  overhears  their  talk ;  she  recognizes  them  and  the 
three  return  to  Rome  where  they  are  reunited  to  Placidus,  or 
Eustathius,  according  to  the  name  he  receives  at  baptism.  Later, 
under  Hadrian,  the  entire  family  suffers  martyrdom. 

In  this  story  Delehaye  (p.  182)  distinguished  three  sections, 
the  miraculous  conversion,  the  adventures  of  the  separated 
family,  and  the  martyrdom.  Though  Eustache  has  been  con- 
sidered a  Roman  saint,  there  seems  no  historical  basis  even  for 
the  concluding  "  passion,"  and  Delehaye  (p.  209)  was  probably 
right  in  calling  the  legend  simply  a  literary  composition  in 
hagiographical  form.  W.  Meyer  (1915)  contended  that  the 
oldest  version  of  this  legend  is  the  Latin  text  printed  by  him 
from  manuscripts  dating  from  the  ninth  century.  He  believed 
(1916,  p.  766)  that  the  original  Latin  version  was  composed  in 
the  fifth  or  sixth  century,  was  translated  after  700  a.d.  into 
Greek  and  so  passed  into  the  East  where  the  story,  bereft  of  the 
conversion  and  martyrdom  episodes,  entered  upon  an  indepen- 
dent life.  This  view  was  in  flat  contradiction  to  the  one  held 
by  Gerould,  Monteverdi,  and  Bousset,  who  thought  that  the 
original  Eastern  story  of  the  Man  Tried  by  Fate  was  translated 
into  Greek,  was  known  to  John  of  Damascus  in  the  eighth  cen- 
tury as  the  legend  of  St.  Eustache,  and  was  translated  from 
Greek  into  Latin.  Gerould  (pp.  344-53)  listed  some  sixteen 
eastern  versions  of  the  story  of  the  Man  Tried  by  Fate.  These 
were  in  Sanskrit,  Singhalese,  Arabic,  Persian,  Hebrew,  Turkish, 
and  Armenian,  and  to  them  others  have  been  added  (Ludtke, 
1917;  Meyer,  1917,  p.  81).     Certain  correspondences  between 


6  MEDIAEVAL   ROMANCE    IN   ENGLAND 

some  of  the  latest  forms  of  these  tales  and  the  Greek  Placidas 
furnished  Hilka  and  Meyer  (191 7,  pp.  92-93)  with  an  argument 
for  their  view  of  the  Latin-Greek-Oriental  origin  of  the  story, 
but  it  does  not  appear  that  they  considered  the  possible  retro- 
active influence  of  the  saint  legend  itself,  which  Gerould  (p.  385) 
suggested  in  connection  with  an  Armenian  tale  narrated  in  the 
nineteenth  century.  Nor  did  they  consider  the  story  which  may 
well  have  been  the  kernel  of  the  saint  legend,  the  tale  of  the 
future  Buddha  who  appeared  in  the  form  of  a  golden  stag  to  a 
king  over-fond  of  hunting  (Garbe,  pp.  538-46). x  The  pre- 
Christian  story  offers  the  earliest  instance  known  of  this  type  of 
Miraculous  Conversion,  and  predicates  an  Eastern  origin  for 
part  of  the  Eustache  legend.  Bousset's  recent  studies  of  the 
second  part  of  the  legend,  the  "  Wiedererkennungs  Marchen," 
strengthen  previous  arguments  for  its  Oriental  origin. 

Whatever  its  earlier  history,  by  the  tenth  century  the  legend 
of  St.  Eustache  was  well  known  in  western  Europe.  The  Latin 
versions  were  translated  into  most  of  the  vernacular  languages.2 
In  England  the  history  of  the  legend  may  be  traced  from  ^Elfric's 
Anglo-Saxon  version,  written  about  996,  to  the  play  of  Placidus 
acted  at  Braintree,  Essex,  in  1534  (Chambers,  Med.  Stage,  11, 

1  Cf.  Jataka  xn  of  the  collection  in  Pali  of  the  stories  of  the  former  exist- 
ences of  Buddha  (trans.  Cambridge,  1895-1907).  Three  scenes  from  this 
conversion  story  were  represented  on  the  great  stiipa  of  Bharhut  in  India  as 
early  as  200  B.C.  The  citation  of  this  story  does  away  with  the  difficulty  noted 
by  Gerould  (p.  386)  in  explaining  the  stag  incident  in  the  Eustache  legend.  In 
later  times  it  was  connected  with  SS.  Hubert,  Julian,  Felix  of  Valois  and  others 
(Garbe,  pp.  538-46).  Cf.  Pschmadt,  Die  Sage  von  der  verfolgten  Hind,  Greifs- 
wald,  ion;  Ogle,  "The  Stag  Messenger,"  Amer.  Jour.  Phil,  xxxvn,  411  ff. 
(1916).  Delehaye  (pp.  193-210)  rejected  the  idea  of  any  direct  connection 
of  the  western  legend  with  the  eastern  tales,  although  acknowledging  their 
astonishing  similarity.  But  his  explanation  of  them  as  unrelated  folk-tales  is 
not  convincing. 

2  Cf.  Gerould,  p.  354;  Meyer,  pp.  227  ff.;  Delehaye,  1919,  p.  176.  Mon- 
teverdi, Studi  Med.  1910,  pp.  392  ff.,  gave  the  fullest  account  of  the  Greek  and 
Latin  texts;  see  also  p.  418  for  notes  on  the  relics  of  St.  Eustache,  and  App. 
Ill  for  some  account  of  the  legend  in  art.  For  the  versions  in  French,  Italian, 
Spanish,  English,  German,  Irish,  see  Monteverdi,  n,  27-105  (Bergamo). 
Fisher,  1917,  p.  2,  recorded  eleven  versions  in  French  verse  (i3th-i5th  cen- 
turies) and  four  in  French  prose.  In  Rom.  xxxvi,  Meyer  printed  the  text 
of  a  short  thirteenth -century  French  monorimed  poem  (Egerton  1066)  which 
shows  the  influence  of  the  chansons  de  geste.  The  French  poem  of  1572 
verses  in  quatrains  printed  by  Ott  (Bibl.  Nat.  Paris,  MS.  i374)>  is  more  closely 
related  to  the  Latin  versions. 


SIR    ISUMBRAS  7 

342),  or  to  Henry  Chettle's  in  1599  (Gerould,  Saints  Legends, 
Index).  With  such  widespread  diffusion  both  in  England  and 
on  the  Continent  as  the  extant  texts  reveal,  it  is  not  surprising 
that  the  legend  became  the  source  of  many  romances,  or  at 
least  a  dominating  influence  upon  them.  In  the  case  of  the 
Spanish  romance,  El  Cavallero  Cifar,  the  dependence  of  the  ro- 
mance on  the  legend  is  shown  not  only  by  the  order  and  nature 
of  identical  incidents,  but  by  direct  reference  to  the  legend  itself 
(Wagner,  pp.  13-29).  The  romances,  like  the  Eastern  tales,  in 
general  parallel  only  the  second  part  of  the  Eustache  legend, 
the  story  of  the  separation  and  reunion  of  the  family. 

In  Isumbras  the  hero  is  pictured  as  "  kynge  of  curtasye,"  rich 
and  generous,  but  over-proud.  To  him  divine  warning  comes 
through  a  bird,  as  it  does  also  to  the  hero  in  the  Gesta  version 
and  in  the  Breton  ballad.  The  fact  that  these  three  versions 
are  alike  in  having  the  Bird  Warning,  and  also  agree  in  many 
subsequent  details,  suggests  some  possible  relationship  between 
them  (Gerould,  p.  416).  The  Choice  of  Woe,  of  poverty  and 
woe  in  youth  or  in  age  in  Isumbras,  eternal  sorrow  or  ten  years 
of  misery  in  Der  Graf  von  Savoien,  woe  on  earth  or  in  the  here- 
after in  the  Danish  ballad,  Sakarias,  is  found  only  in  the  later 
versions  of  the  Eustache  legend,  such  as  that  given  by  Vincent 
of  Beauvais,  Speculum  Historiale,  x,  cap.  58-61,  or  in  the  Le- 
genda  Aurea.  Since  this  motif  appears  in  at  least  seven  inde- 
pendent tales  in  which  the  choice  is  presented  to  a  woman  by 
the  Virgin,  Gerould  (pp.  355,  384,  424-36)  believed  that  it  was 
an  originally  separate  story  which  was  later  absorbed  by  the 
St.  Eustache  legend. 

The  most  characteristic  incidents  linking  the  legend  and  the 
romances  are  those  which  tell  of  the  separation  of  the  family. 
In  Isumbras  the  knight  carries  one  son  across  the  water  but 
when  he  returns  a  lion  carries  the  first  child  away,  a  leopard 
the  second,  and  ultimately  a  unicorn  takes  the  third.  In  the 
Breton  ballad  also  three  children  are  mentioned;  in  other  ver- 
sions there  are  but  two.  In  Guillaume,  Wilhelm,  and  Gute  Frau 
the  children  (twins)  are  born  after  their  parents'  departure 
from  home,  and  this  feature  connects  the  motif  of  the  Loss  of 
the  Children  with  such  romances  as  Oct  avian,  Valentine  and 
Orson,  La  belle  Helene  de  Constantinople,  Beves  of  Hampton, 


8  MEDIEVAL   ROMANCE    IN   ENGLAND 

Ogier,  Eglamour,  and  Torrent  of  Portyngale.  In  these  tales  a 
young  wife,  after  she  has  been  driven  from  her  home,  gives  birth 
to  two  children  who  are  stolen  from  her  by  wild  animals.  The 
vogue  of  this  wildly  improbable  episode  in  all  these  European 
variants  contrasts  oddly  with  the  more  sober  Oriental  versions 
in  which  the  children  are  usually  separated  from  their  parents 
by  robbers,  by  shipwreck,  or  by  the  father's  act  of  penance  in 
deliberately  giving  them  away.  In  every  case  the  loss  of  the 
children  seems  an  integral  and  natural  part  of  the  story  of  the 
Man  (or  Woman)  Tried  by  Fate.3  The  Loss  of  the  Wife  takes 
place  in  Isumbras,  Cifar,  and  Wilhelm  after  the  loss  of  the  chil- 
dren; in  the  Eustache  legend,  the  Gesta,  Sakarias,  and  the 
Breton  ballad  before  the  children  are  stolen.  The  Sale  of  the 
Wife  (Verkauf  der  Frau)  to  which  the  unfortunate  husband  is 
forced  by  violence  in  Isumbras,  Guillaume,  and  Graf  is,  perhaps, 
as  Jordan  (p.  365)  surmised,  a  concession  to  European  taste, 
for  in  certain  Oriental  tales  and  in  even  Die  Gute  Frau,  the  wife 
herself  advises  this  extreme  measure.  Likewise  the  preservation 
of  the  heroine's  chastity  when  she  is  in  the  hands  of  her  captors 
is  on  the  whole  a  distinctive  Occidental  trait  (Gerould,  p.  372). 
But  it  is  not  possible,  as  Monteverdi  (Studi  Med.,  in,  227  ff.) 
pointed  out,  to  accept  Jordan's  belief  that  this  motif  (Scheinehe) 
can  determine  the  relationship  of  the  versions.4 

The  European  stories  are  distinguished  by  a  characteristic 
subsidiary  theme  which  in  Guillaume,  Wilhelm,  Die  Gute  Frau, 
Isumbras,  and  Graf,  was  grafted  on  the  legend  of  the  Man  Tried 
by  Fate.  Eustache  has  no  trace  of  it.  In  several  stories,  Cama- 
ralzaman  in  the  Arabian  Nights    (Chauvin,  Bibliographie  des 

3  In  general,  stories  of  the  Woman  Tried  by  Fate  belong  to  the  type  best 
known  as  the  Calumniated  Wife.  Gerould  (p.  441)  noted  the  danger  of  de- 
riving these  diverse  stories  from  Eustache  with  which  they  often  have,  in 
common,  only  the  incident  of  the  Loss  of  the  Children.  Helene  alone  seems 
clearly  to  have  borrowed  from  the  legend.  In  the  others  the  use  of  the  in- 
cident is  due  to  the  influence  of  the  romances  themselves.    See  Octavian  here. 

4  Jordan  argued  from  the  fact  that  in  one  group  of  versions  the  Pretended 
Marriage  is  contracted  by  the  wife,  and  in  Cifar,  in  the  Arab  tale  of  the 
"king  who  lost  all"  (Chauvin,  Bibl.  des  Ouvrages  Arabes,  vi,  p.  164),  and 
in  Beves,  by  the  husband,  who  marries  the  ruler  of  a  country  but  who,  in  the 
hope  of  recovering  his  wife,  puts  off  the  consummation  of  his  marriage  with 
the  princess  of  the  land.  Since  these  three  versions  have  hardly  anything 
more  than  this  one  trait  in  common,  no  argument  can  be  based  upon  them. 
The  suggested  derivation  of  the  Civile  episode  in  Beves  from  the  lost  source 
of  Cifar  is  altogether  fictitious. 


SIR    ISUMBRAS  g 

Ouvrages  Arabes,  iv,  204-12,  1900),  and  other  Eastern  variants, 
in  the  old  French  romance,  UEscoufle,  c.  1204  (ed.  Meyer,  SATF, 
1894),  in  the  various  versions  of  Pierre  de  Provence  et  la  belle 
Maguelonne  (ed.  A.  Biedermann,  Paris,  Halle,  191 5),  one  im- 
portant theme  is  that  of  a  Lost  Treasure.     In  Isumbras  it  is 
the  gold  given  to  the  hero  for  his  wife  by  the  heathens  who 
had  carried  her   off.     Isumbras  wraps   it   in  his  "  mantill  of 
skarlet  rede,"  and  this  an  eagle  carries  away.     In  the  stories 
just  listed,  though  the  circumstances  of  the  hero  and  heroine 
are  differently  motivated  from  those  in  the  Man  Tried  by  Fate, 
they  are  alike  in  telling  of  the  separation  of  the  lovers,  and 
of  their  loss  through  a  bird's  theft  of  some  object  of  value. 
Gerould  (p.  412)  believed  that  the  Lost  Treasure  story  must 
have  come  independently  from  Asia  to  Europe,  and  that  at 
least  as  early  as  the  twelfth  century,  it  had  been  incorporated 
in  the  Eustache  story  as  represented  by  Guillaume  d'Engleterre. 
The  adventures  of  the  separated  family  are  told  with  con- 
siderable variation.     In  Isumbras,  after  serving  as  a  smith  for 
seven  years  the  hero  fights  victoriously  in  his  self-made  arms 
against  the  Saracens.     After  many  wanderings  he  comes  as  a 
palmer  to  the  land  where  his  wife  reigns  as  queen.     Unrecog- 
nized, he  serves  humbly  in  her  court  until  he  is  identified  by 
the  gold  wrapped  in  his  red  mantle.    This  Lost  Treasure,  in  a 
fashion  peculiar  to  such  stories,  he  had  recovered  from  a  bird's 
nest.    He  is  crowned  king  and  later,  in  a  great  battle  against 
the  heathen,  is  aided  by  his  sons  who  ride  on  animals  like  those 
which  had  stolen  them.    The  youths  announce  their  parentage 
and  the  long  severed  family  is  at  last  reunited.    In  Isumbras, 
Guillaume,    Wilhelm,    Gute    Frau,    Graf,    and    Sakarias,    the 
heroine,  when  she  is  separated  from  her  husband,  suffers  no 
hardship  but  through  a  Marriage  in  Name  or  by  popular  elec- 
tion speedily  rises  to  great  place.    In  this  detail  the  romances 
differ    notably    from    the   St.    Eustache   legend    in    which    the 
heroine's  humble  life  of  self-support  suggests  the  influence  of 
the  legendary  story  of  Helen,  the  mother  of  Constantine.5    The 
scene   in   Isumbras   in   which   the  husband,   unrecognized,   re- 

5  See  Emare  here.  In  the  Gesta  tale  and  the  Breton  ballad  the  wife  serves 
as  a  beggar  and  a  servant.  It  is  a  realistic  detail  in  Isumbras  that  she  is 
pictured  as  going  into  battle  clad  "  in  armour,  als  scho  were  a  knyghte  "  (1. 
746). 


10  MEDIEVAL   ROMANCE    IN    ENGLAND 

ceives  the  bounty  of  his  wife  may  be  paralleled  by  similar  ones 
in  Beves  of  Hampton,  Guy  of  Warwick  and  Die  Gute  Frau. 
The  Treasure  serves  as  a  Recognition  Token  in  Isumbras,  and 
various  other  tokens  appear  in  the  different  versions;  in  Guil- 
laume,  for  instance,  a  horn  and  ring  are  mentioned.  In  the 
Eustache  legend  the  recognition  is  accomplished  through  the 
wife's  overhearing  her  two  sons  tell  how  they  were  stolen  by 
animals,  and  then  through  her  going  to  Placidus,  their  com- 
manding general,  to  ask  for  help. 

Despite  all  the  romantic  and  fabulous  elements  in  Isumbras, 
its  structural  likeness  to  the  legend  of  St.  Eustache  remains 
apparent.  In  spirit  too  it  keeps  the  insistent  note  of  piety,  of 
joy  "  Goddes  werkkes  for  to  wyrke." 


BIBLIOGRAPHY 

Texts:  (i)  C,  Caius  College,  Cam.  175  (A,  ix)  summarized  by  Ellis, 
Specimens,  pp.  479-91  (1848);  MS.  described,  J.  Zupitza,  Eng.  Stud. 
xrv,  321;  (2)  T,  Thornton  MS.  Lincoln  Cathedral,  ed.  Halliwell, 
Thornton  Romances,  pp.  88-120,  Camden  Soc,  1844;  see  also  p. 
268  for  selection  from  C;  reed.  F.  S.  Ellis,  Kelmscott  Press,  1897;  (3) 
L,  Cott.  Cal.  A  11,  Brit.  Mus.,  cf.  Ward,  Cat.  of  Romances,  1,  180,  760; 
(4)  A,  Ashmole  61,  Oxf.  cf.  Halliwell,  p.  268;  (5)  E,  Advocates  Lib.  19, 
3,  1,  Edin.,  cf.  Halliwell,  p.  267;  (6)  N,  Royal  Library,  Naples,  xm,  B. 
29,  (dated  1457),  ed.  Kolbing,  Eng.  Stud,  in,  200;  (7)  G,  Gray's  Inn, 
Lond.  Early  Editions:  (8)  c,  Copland,  undated,  repr.  Utterson,  1,  77 
ff. ;  (9)  d,  Douce  fragment  78,  Oxf.;  (10)  D,  Douce  261;  (11)  M, 
Malone  941,  Bodleian,  Oxf.,  a  few  lines;  cf.  Percy  Folio,  ed.  Hales  and 
Furnivall,  1,  532;  Zupitza,  Archiv.  88,  72.  A  critical  edition  was  pub- 
lished by  Zupitza  and  G.  Schleich  in  Palaestra  xv,  128  pp.,  Berlin,  1901, 
Stammbaum,  p.  87.  Rev.  Liter aturbl.  xxm,  p.  18  (1903). 

Studies:  Cf.  Wells,  Manual,  p.  781. 
Bousset,  W.    "  Die  Geschichte  eines  Wiedererkennungsmarchens."    Nach- 

richten  von  der  Konig.      Gesellschaft zu  Gottingen,  Philol.  hist. 

Kl.  pp.  460-551  (1916):  pp.  703-45  (1917). 
Delehaye,  H.     "La  Legende  de  Saint  Eustache."    Bull,  de  la  classe  des 

lettres,  Acad.  roy.  de  Belgique,  iv,  175-210  (1919).    Bibliog.  p.  176. 
Esposito,  M.     "La  vie  de  S.  Eustache"  (Anglo-Norman,  Dublin  MS.). 

Florence,  192 1. 
Fisher,  J.  R.     "  La  Vie  de  St.  Eustache  par  Pierre  de  Beauvais."    Rom. 

Rev.  viii,  1-67  (191 7). 
Garbe,  R.    "  Contributions  of  Buddhism  to  Christianity "   (St.  Eusta- 


SIR    ISUMBRAS 


II 


chius,  pp.  538-50).     The  Monist,  xxi,  Chicago,  191 1.    Trans,  from 

Deutsche  Rundschau,  1910-11. 
Gerould,  G.  H.     "  The  Eustache  Legend,  Forerunners,  Congeners,  and 

Derivatives."  .  .  .  PMLA.  xix,  335-448  (1904). 
Hilka,  A.  u.  W.  Meyer.     "  Ueber  die  neu-aramaische  Placidas-Wander- 

geschichte."     Nachrichten  .  .  .  zu  Gottingen,  pp.  80-95    (1917). 
Jordan,  L.     "  Die   Eustachiuslegende,   Christians  Wilhelmsleben,    Boeve 

de   Hanstone   u.   ihre    orientalischen   Verwandten."     Archiv,    cxxi, 

341-67   (1908). 
Liidtke,    W.    "  Neue    Texte    zur    Geschichte    eines    Wiedererkennungs- 

marchens  u.  zum  Text  der  Placidas-Legende."    Nachrichten  .  .  .  zu 

Gottingen,  pp.   746-760   (191 7). 
Meyer,  P.    "Fragment  d'une  Vie  de  S.  Eustache."     Rom.  xxxvi,  12- 

28  (1907). 
Meyer,   W.    "  Der   Rythmus   liber   den   h.    Placidas-Eustasius."     Nach- 
richten .  .  .  zu  Gottingen,  pp.  226-87  (I0I5);   I0I6,  pp.  745—799. 

See  Hilka. 
Monteverdi,  A.    /  testi  delta  Leggenda  di  S.  Eustachio,  Bergamo,  1909- 

10   (2  vols.);   Studi  Medievali,  in,   169-226,  392-498. 
Ogden,  P.     A  Comparative  Study  of  the  Poem  Guillaume  d'Angleterre. 

Diss.  Baltimore,  1900. 
Ott,  A.  C.     "  Das  altfrz.  Eustachiusleben."    Rom.  Forsch.,  xxxn,  481- 

607    (1912). 
Schleich.     See  Texts. 
Speyer,  G.  S.     "  Buddhistische  elementen  in  eenige  episoden  uit  de  Leg- 

enden   van   St.    Hubertus   en   St.    Eustachius."      Theologisch    Tijd- 

schrift,  xl,  427-53.    Rev.  Delehaye,  Le  Museon,  N.S.  xm,  91-100, 

Lou  vain,    191 2. 
Wagner,  C.  P.     "  The  Sources  of  El  Cavallero  Cifar."     Revue  Hispa- 

nique,  x,  5-104  (1903). 


FLORENCE   OF  ROME 

Versions.  Among  the  mediaeval  stories  of  innocent  women 
who  suffer  a  succession  of  trials  and  misfortunes,  the  Crescentia 
story,  to  which  the  various  versions  of  Florence  of  Rome  be- 
long, is  made  distinctive  by  two  traits:  (i)  in  the  absence  of 
the  heroine's  husband  it  is  always  his  own  brother  who  first 
approaches  her  with  offers  of  love  (Le  conte  de  la  femnie  chaste 
convoitee  par  son  beau-frere) ;  and  (2)  it  is  the  lady's  fame 
as  a  healer  which  ultimately  brings  together  those  who  have 
wronged  her  and  who  then  confess  their  crimes  against  her 
(Wallenskold,  2,  p.  105).  The  story  enjoyed  a  popularity  which 
makes  difficult  the  classification  or  even  the  enumeration  of  the 
various  versions.  Of  these  more  than  one  hundred,  dating  from 
the  twelfth  to  the  nineteenth  century,  have  been  listed  by  various 
scholars  (Stefanovic,  p.  467).  Karl  (p.  164),  indeed,  referred  to 
two  hundred  and  sixty  versions  but  it  is  probable  that  such  a 
list  would  involve  a  large  amount  of  repetition.  Here  only  the 
most  important  groups  of  texts  can  be  noted. 

The  type  name  for  the  story  comes  from  what  is  commonly 
believed  to  be  the  oldest  European  text,  the  story  of  Crescentia 
in  the  Old  High  German  Kaiser chronik,  v.  11,352  ff.  (ed.  E. 
Schroedef,  1892).  This  was  written  about  11 50,  and  other  ver- 
sions in  verse  and  prose  of  the  same  text  appeared  in  the  thir- 
teenth and  fourteenth  centuries.  The  episode  was  incorporated 
in  practically  the  same  form  as  that  in  the  Kaiserchronik  in  the 
Sachsische  Weltchronik,  1237-1251  (ed.  L.  Weiland,  Mon.  Germ. 
Hist.,  1877,  P-  w),  and  in  later  times  had  a  long  history  in 
Volksbuch  form.  In  Wallenskold's  opinion  (1,  p.  60)  the  Cres- 
centia story  itself  was  but  a  variant  of  an  anonymous  Miracle  de 
Vierge  which  other  scholars  have  been  inclined  to  think  repre- 
sented a  second  and  separate  group  of  stories. 

The  earliest  text  of  the  Miracle  is  a  twelfth-century  prose  ac- 
count (ed.  Wallenskold,  1,  p.  116  ff.),  but  in  all  some  thirty-eight 
texts  are  known.    These  include  not  only  versions  of  the  Miracle 

12 


FLORENCE   OF    ROME  I3 

proper,  but  also  the  abbreviated  account  of  it  given  in  Les  Vies 
des  Peres  (Le  Grand  d'Aussy,  Fabliaux,  1829,  v,  125)  and  in  the 
legend  of  St.  Guglielma  (Stefanovic,  pp.  552-54).  The  anony- 
mous Miracle  itself,  according  to  Wallenskold's  enumeration 
(1,  pp.  32-36),  exists  in  four  forms,  in  Latin,  in  French,  in 
Dutch,  and  in  Icelandic.  To  these  Hilka  (p.  136)  added  three 
other  Latin  texts  of  the  fourteenth  and  the  fifteenth  centuries. 
The  Miracle  was  introduced  into  the  Speculum  Historiale,  1244- 
54,  Bk.  VII,  cap.  90-92,  of  Vincent  of  Beauvais,  and  so  into  the 
many  later  translations  in  French,  Dutch,  and  Italian  of  Vin- 
cent's work.  It  was  also  used  as  an  exemplum  by  Etienne  de 
Bourbon  (1261)  in  his  Liber  de  Septem  Donis ;  by  Humbert  de 
Romans  (d.  1277)  in  his  Liber  de  abundantia  exemplorum ;  by 
the  author  of  the  Alphabetum  Narrationum,  &  book  subsequently 
translated  into  Spanish  and  English ;  by  Jean  de  Garlande  in  his 
Stella  Maris  (1288);  by  Johannes  Junior  in  the  Scala  Celi  in 
the  early  part  of  the  fourteenth  century;  by  Johannes  Herold 
in  his  Promptuarium  de  Miraculis,  1435-40;  and  by  the  Italian, 
Gabriel  Bareleta  (before  1480),  in  his  Sermones.  As  an  exem- 
plum the  story  was  usually  given  under  the  heading  Castitas.  In 
French  verse  the  Miracle  was  retold  before  1222  by  Gautier  de 
Coincy  (Meon,  Nouv.  Recueil,  1823,  11,  1-128)  in  the  famous 
Miracles  de  Notre  Dame  of  which  so  many  manuscripts  survive 
(Wallenskold,  1,  p.  37;  2,  pp.  118-123).  A  fourteenth-century 
Spanish  prose  translation  of  Gautier's  version  of  Ulmperatrice 
de  Rome  is  known  (Mussafia,  Sitz.  d.  kais.  Akad.  Ph.  hist.  CI. 
Vienna,  liii,  508  ff.).  In  the  fourteenth  century  the  Miracle  was 
dramatized  in  France  (Miracles,  1879,  SATF.,  iv,  234),  and  again 
in  the  fifteenth  century  turned  into  French  prose  by  Jean  Mielot. 
In  Germany  the  story  appeared  in  Der  Seelen  Trost  (Pfeiffer, 
Die  deut.  Mundarten,  1856,  11,  7-9),  a  religious  book  of  exempla 
which  was  translated  into  Dutch,  Swedish,  and  Danish;  in  a 
fifteenth-century  poem  of  Hans  Rosenbliit  of  Nuremberg;  in  a 
"  comedie,"  Die  Unschuldig  Keyserin  von  Rom  of  Hans  Sachs, 
1 551;  and  in  a  poem  of  the  meistersinger  of  the  sixteenth  cen- 
tury, Albrecht  Baumholtz  (Wallenskold,  1,  pp.  57-59;  App.  M). 
In  Italy  the  content  of  the  Latin  miracle  was  well  known  in 
popular  and  dramatic  form  in  the  sixteenth  and  seventeenth 
centuries,  but  the  earliest  extant  version  of  the  story  is  the  four- 


I4  MEDIEVAL   ROMANCE    IN    ENGLAND 

teenth-century  Istoria  di  Santa  Guglielma,  which  differs  consid- 
erably from  the  older  Miracle.  In  the  fifteenth  century  the 
dramatization  of  this  Istoria  by  Antonia  Pulci  gave  a  new  im- 
petus to  this  particular  form  of  the  story  (Wallenskold,  i,  pp. 
46-54;  2,  pp.  120-21). 

The  third  group  of  stories  includes  the  various  redactions  to 
which  the  heroine  gives  her  name,  Florence  of  Rome.  Of  a 
French  chanson  d'aventure  of  6410  lines  (ed.  Wallenskold,  Paris, 
1909),  composed  in  the  first  quarter1  of  the  thirteenth  century, 
there  are  extant  two  thirteenth-century  manuscripts  (PM)  and 
a  late  thirteenth-century  fragment  (L;  cf.  Ward,  Cat.  of  Ro- 
mances, 1,  711).  Besides  this  there  are  four  versions  of  approxi- 
mately the  same  type:  an  early  fourteenth-century  Dit  de  Flor- 
ence (D)  in  quatrains  (ed.  Jubinal,  Nouv.  Recueil,  1839)  5  a 
long  fifteenth-century  version  of  4562  alexandrine  verses  (Q,  ed. 
Wallenskold,  Florence,  1909,  pp.  130  ff.) ;  the  fifteenth-century 
Middle  English  poem  (R) ;  and  a  prose  version  in  Spanish  (S) 
(ed.  Amador  de  los  Rios,  Historia  critica,  1864,  v,  391-468)  of 
the  late  fourteenth  or  early  fifteenth  century.  The  relations  of 
these  versions  have  been  studied  with  varying  results  by  Wenzel 
(p.  62),  Knobbe,  and  Wallenskold.  The  last  (Flor.,  p.  14;  26) 
agreed  with  Knobbe  in  finding  that  the  texts  LMPS  (D?)  form 
one  group  which  may  be  distinguished  from  R  and  L,  the  two 
texts  which  apparently  represent  independent  derivatives  of  the 
lost  original. 

The  single  Middle  English  version,  Le  Bone  Florence,  which 
is  written  in  the  North-Midland  dialect  (Knobbe,  p.  49),  is 
what  Ritson  terms  "  an  excellent  old  romance."  It  contains  2187 
lines  in  the  twelve-line,  tail-rime  stanza;  and  a  comparison  with 
the  extant  French  texts,  three  of  which  extend  to  over  four 
thousand  lines,  shows  a  resolute  shortening  of  the  original.  To 
this  the  English  version  often  refers  as  the  "  boke  "  (11.  84,  491, 

1  Wallenskold,  2,  pp.  99  ff.,  noted  that  the  earliest  French  text  of  Florence 
contains  a  reference  to  Guillaume  de  Dole  or  the  Roman  de  la  Rose,  c.  1200, 
and  that  the  Roman  de  la  Violette,  c.  1225-30,  refers  to  Florence.  These  dates 
serve  to  place  the  chanson  d'aventure  between  1200  and  1230.  The  earliest 
reference  to  the  story  of  Florence  is  found  in  La  Naissance  de  la  Chevalier 
au  Cygne  (ed.  Todd,  PMLA.  1889,  1.  3098),  which  is  ascribed  to  the  end  of 
the  twelfth  or  the  beginning  of  the  thirteenth  century.  Wallenskold,  p.  104, 
thought  the  reference  was  not  to  an  extant  text  of  Florence,  but  to  the  lost 
original  of  the  extant  versions. 


FLORENCE    OF    ROME  15 

869),  the  "  romance  "  (11.  643,  1164,  1539),  etc.  The  author  even 
suggests  that  he  had  consulted  more  than  one  version  of  the  story 
(1.  84),  and  in  1.  2173  states  that  the  story  was  written  down  by 
"  Pope  Symonde  "  in  the  "  cronykyls  of  Rome,"  a  pseudonym, 
perhaps,  for  that  mysterious  "  boke  of  Rome  "  of  which  some 
scholars  are  now  inclined  to  think  the  Gesta  Romanorum  is  in 
some  sort  an  imitation  (Rickert,  Emare,EETSES  99,  p.  48,  n. 
2).  The  Middle  English  poet  seems  to  have  taken  something 
from  Benoit's  Roman  de  Troie  (Knobbe,  p.  7),  but  the  large 
number  of  fantastically  spelled  classical  names  which  he  intro- 
duces, he  may  well  have  derived  from  his  French  original.  A 
reference  in  1.  1888  to  Sir  Lucius  Ibarnyus  as  the  founder  of 
"  Beverfayre,"  in  the  French  versions  Beau-Repaire,  —  a  con- 
vent said  to  have  been  founded  by  Julius  Caesar,  —  has  been 
taken  as  an  indication  that  the  author  had  in  mind  the  Lucius 
Iberius  of  the  alliterative  Morte  Arthur  e,  1.  86  (Knobbe,  p.  9). 
More  convincing  is  the  evidence  that  he  was  familiar  with  an- 
other Middle  English  poem,  the  King  of  Tars,  which,  like  Flor- 
ence, sets  forth  the  grief  of  a  Christian  princess  who  mourns 
when  a  heathen  suitor  attacks  her  father's  land  that  so  many 
men  should  die  for  her  sake  (Siefkin,  p.  43).  It  is  possible, 
however,  that  this  passage  in  the  King  of  Tars  or  its  original 
"  geste,"  was  itself  influenced  by  the  French  original  of  Florence. 
The  Middle  English  redactor  of  Florence  was  of  a  strongly 
religious  cast  of  mind  and  he  tells  his  story  not  for  the  sake  of 
diversion,  but  for  the  picture  it  gives  of  Christian  fortitude. 
The  chastity  of  his  heroine,  for  instance,  is  not  saved  by  a  magic 
brooch  as  in  the  French  versions,2  but  simply  by  the  heroine's 
prayer  to  the  Virgin,  who  makes  the  persecutor  forget  his  pas- 
sion. Similarly,  prophetic  dreams,  weird  portents,  fantastic  epi- 
sodes, such  as  the  successive  attacks  of  wild  animals  on  the 
wicked  brother-in-law,  Miles,  which  are  found  in  the  eminently 
pious  but  far  more  romantic  French  texts,  are  omitted.3    Even 

2  Wallenskold,  2,  p.  37,  n.  1,  referred  to  other  instances  of  the  jewel  (Aye 
d'Avignon,  p.  62,  Charles  le  Chauve,  1.  10620)  or  some  other  object  (girdle, 
Boeve  de  Haumtone ;  herb,  Orson  de  Beauvais)  which  is  similarly  protective 
of  chastity. 

3  Wallenskold,  2,  ch.  m,  Caractere  de  la  Chanson,  found  that  the  romance 
falls  into  two  parts:  (1)  that  which  tells  of  the  attack  on  Rome  led  by 
Garcy,  the  old  king  of  the  Greeks  and  the  brutal  suitor  of  Florence,  of  its 
failure,  and  of  her  marriage  to  Esmere  who,  with  his  wicked  older  brother 


1 6  MEDIEVAL   ROMANCE    IN   ENGLAND 

though  the  English  version  keeps  the  essential  miraculous  ele- 
ments of  the  older  story,  such  as  the  successive  escapes  of  the 
heroine  from  her  persecutors  on  sea  or  on  land,  and  the  mar- 
velous ringing  of  the  bells  of  the  convent  when  the  holy  heroine 
draws  near,  a  trait,  as  Wallenskold  pointed  out,4  peculiarly  dear 
to  the  hagiographical  literature  of  the  Middle  Ages,  still  in  gen- 
eral it  is  clear  that  the  author  was  trying  to  make  his  narrative 
as  sober  and  unromantically  moral  as  possible.  On  the  whole 
his  restraint  achieves  a  more  readable  result  than  the  too  long- 
winded  chanson  d'aventure. 

Three  groups  of  versions  of  the  Crescentia  saga  remain  to  be 
noted,  two  of  them  European,  and  one,  Oriental.  In  the  early 
fourteenth  century,  if  not  before,  a  condensed  version  of  the 
story,  derived  from  the  same  source  as  Florence  of  Rome,  was 
incorporated  into  the  Gesta  Romanorum  and  from  that  time  it 
appeared  in  the  Continental,  the  Anglo-Latin,  and  Late  Middle 
English  redactions.  About  142 1  it  was  turned  into  English  rime- 
royal  verse  by  Hoccleve  (ed.  Furnivall,  1892,  1,  pp.  140  ff.).  In 
the  fifteenth  century  a  Bavarian  schoolmaster,  Johannes  Birck, 
introduced  the  story,  again  told  at  some  length,  into  his  chronicle 
of  the  Abbey  of  Kempten.  He  attributed  the  founding  of  the 
Abbey  to  Hildegard,  reputed  the  second  or  third  wife  of  Charle- 
magne, and  made  her  the  heroine  of  a  story  frequently  resem- 
bling Crescentia's.5  Twenty-eight  later  texts,  chiefly  of  the  six- 
teenth or  seventeenth  century,  make  the  same  identification,  and 
to  this  group,  therefore,  the  name  of  the  Hildegard  saga  is  given. 
Its  chronological  relations  are  a  matter  of  grave  dispute.    One 


Miles,  had  come  to  her  aid;  (2)  that  which  relates  her  adventures  after  she 
had  been  separated  from  her  husband  by  Miles.  In  attempting  to  make  the 
story  a  chanson  de  geste,  the  French  author  has  described  great  battles  and 
terrific  single  combats;  given  much  detail  in  regard  to  houses  and  armor; 
boldly  contrasted  his  good  and  evil  characters;  and  infused  the  whole  with 
fervent  piety  in  the  manner  of  the  literary  type  he  imitates. 

4  Wallenskold,  2,  p.  39,  n.  2.  Cf.  also  Child,  Ballads,  Index,  Bells;  J. 
Tatlock,  "Bells  rung  without  Hands,"  MLN.  xxix,  98  (1914),  and  P.  Barry, 
MLN.  xxx,  28  (1915).  Church  bells  are  first  mentioned  by  Gregory  of  Tours. 
The  first  reference  to  their  miraculous  ringing  comes  in  the  Vita  Bonifatii. 
By  the  middle  of  the  tenth  century  the  literary  tradition  seems  to  have  been 
established,  and  in  the  eleventh  century  the  motif  passed  from  religious 
legend  into  the  chansons  de  geste  (Barry). 

5  Cf.  G.  Paris,  Hist.  Poetique  de  Charlemagne,  2d.  ed.,  1905,  p.  395;  Birck's 
Chronik,  ed.  K.  Reiser,  1899,  1,  442. 


FLORENCE   OF    ROME  17 

point  of  view  is  represented  by  Wallenskold  (2,  p.  127;  3,  p.  72) 
who  held  that  Birck's  version  was  simply  an  unhistoric  adapta- 
tion of  the  Miracle  de  la  Vierge.  This  view  was  opposed  by 
Stefanovic,  who  argued  (pp.  500-11)  that  the  later  versions  of 
Hildegard's  story  differ  too  greatly  from  Birck's  work  not  to 
have  been  derived  from  an  independent  and  presumably  much 
earlier  tradition. 

The  Oriental  versions  of  the  story  fall  into  three  groups 
(Wallenskold,  1,  p.  9;  2,  pp.  109-22).  The  first  and  earliest  of 
those  now  extant  is  found  in  a  fourteenth-century  Persian  collec- 
tion of  tales  called  the  Touti-Nameh  (German  trans,  by  R. 
Schmidt,  Stuttgart,  1899)  which  in  part  at  least  goes  back  to  a 
very  ancient  Sanskrit  original  now  imperfectly  represented  by  the 
Soukasaptati  (or  the  Sixty-Six  Tales  of  a  Parrot).  A  fifteenth- 
century  Turkish  version  of  the  tale  of  the  chaste  Merhuma  in 
the  Touti-Nameh  is  extant  (G.  Rosen,  1858,  1,  89-108).  In  the 
famous  Arabian  collection  of  the  Thousand  and  One  Nights, 
there  are  three  versions  of  the  story  of  the  Chaste  Wife,  and  in 
the  Thousand  and  One  Days  (1 710-12),  of  which  the  earliest 
known  manuscript  is  a  Turkish  redaction  written  in  1450,  occurs 
the  tale  of  the  chaste  Repsima. 

Origin.  In  this  highly  romantic  and  supernatural  fiction  con- 
cerning a  saintly  heroine,  there  is  little  that  seems  of  local  or 
racial  character.  In  general  when  attempting  to  discover  the 
origin  of  the  story,  one  must  depend  primarily  on  a  considera- 
tion of  the  nine  principal  incidents  in  the  different  versions: 
( 1 )  the  wooing  of  the  heroine  by  her  brother-in-law ; 6  ( 2 )  the 
accusation  of  adultery  brought  against  her  by  him;7   (3)  her 

6  In  the  Florence  of  Rome  versions  two  young  nobles,  dispossessed  by 
their  stepfather,  come  as  soldiers  of  fortune  to  aid  the  Emperor  of  Rome 
against  the  attack  of  his  daughter's  barbarous  suitor.  The  brothers  become 
rival  suitors  for  the  daughter  and  ultimately  the  younger  brother  Esmere 
wins  the  lady  and  thereby  becomes  Emperor  of  Rome.  As  Florence  will  be 
his  wife  in  name  only  until  he  has  destroyed  her  enemy,  Esmere  goes  at 
once  in  pursuit  of  the  routed  suitor,  leaving  his  wife  in  his  brother's  care. 
The  wicked  Miles  practises  various  stratagems  to  deceive  her,  once  even 
attempting  to  pass  off  a  mutilated  dead  body  as  Esmere's.  For  this  deed 
she  has  Miles  shut  up  in  a  tower  from  which,  in  her  joyous  anticipation  of 
Esmere's  return,  she  later  releases  him,  thereby  giving  him  the  opportunity 
to  bring  her  new  suffering. 

7  In  other  versions  than  Florence  the   husband   commonly   believes  this 


1 8  MEDIEVAL   ROMANCE    IN   ENGLAND 

condemnation  to  death  or  exile;  (4)  her  flight;  (5)  her  refuge 
in  a  household  where  she  is  again  wooed  by  a  rejected  suitor 
who  in  revenge  murders  the  child  of  her  protectors  and  accuses 
her  of  the  crime;  8  (6)  her  second  flight;  (7)  her  adventure  with 
a  debtor  whom  she  frees  from  debt  but  who  sells  her  to  a  ship's 
captain;  (8)  her  escape  from  the  captain  through  a  storm  that 
wrecks  the  boat  or  causes  the  captain  to  put  her  ashore;  (9)  her 
life  as  a  holy  woman  whose  fame  as  a  healer  brings  to  her  ulti- 
mately all  her  stricken  persecutors ;  her  restoration  when  by 
their  confession  her  innocence  is  finally  established. 

There  are  four  theories  in  regard  to  the  original  narrative 
embodying  all  or  part  of  these  episodes.9  Grundtvig  (Danmarks 
gatnle  folkeviser,  1853,  1,  195;  in,  782;  iv,  730)  thought  that  it 
was  of  Germanic  origin  and  that  the  Oriental  versions  were  but 
importations  from  the  west.  His  hypothesis  rested  on  an  inade- 
quate classification  of  the  extant  versions,  and  a  consequent 
confusion  of  our  story  with  the  type  represented  by  the  Danish 
ballad  of  Ravengaard  og  M entering  (No.  13),10  which  is  cousin 
to  the  English  ballad  of  Sir  Aldingar.  The  first  part  of  his 
conclusion  may,  on  this  ground,  be  disregarded.  In  1865  Mus- 
safia,11  convinced  despite  the  lateness  of  the  extant  Oriental 
texts,  of  their  actual  priority,  set  forth  the  theory  that  the  story 
was  of  eastern  origin,  and  that  it  was  introduced  into  the  west 
in  the  abbreviated  form  now  found  in  the  Kaiserchronik  and  in 
the  Miracle  group.  In  this  form  the  seventh  and  eighth  episodes 
are  omitted.  Later,  he  thought,  there  was  a  second  importation 
of  the  story  from  the  East  from  which  the  complete  versions 
were  derived.  This  theory  was  in  part  discredited  by  Wallens- 
kold,  although  in  general  he  believed  in  the  Oriental  origin  of 


unsupported  accusation,  and  incidents  3  and  4  follow  directly.  In  Florence 
the  falsity  of  the  accusation  is  known  almost  at  once,  but  Miles  gets  the  lady 
into  his  power  by  pretending  that  he  has  been  sent  as  an  escort  to  bring  her 
to  Esmere.  After  new  attempts  to  force  his  love  upon  her  he  abandons  her 
in  a  forest. 

8  See  here  Emare,  n.  5. 

9  The  history  of  the  discussion  of  the  origins  of  Florence  is  summarized 
to  1909  by  Wallenskold,  Florence,  pp.  106  ff. 

10  Cf.  Child,  Ballads,  u,  34  ff.,  and  Erie  of  Tolous,  here. 

11  "Uber  eine  italienische  metrische  Darstellung  der  Crescentia  Sage." 
Sitzungsberichte  der  phil.  hist.  Classe  der  Kais.  Akad.  der  Wiss.,  Vienna,  li, 
pp.  589  ff- 


FLORENCE   OF    ROME  19 

the  story.  He  argued  that  since  the  European  versions  have  an 
incident,  the  imprisonment  of  the  brother-in-law  in  a  tower, 
which  is  lacking  in  all  the  Eastern  tales,  it  is  improbable  that 
in  two  successive  western  adaptations  of  the  Eastern  tale,  the 
same  invention  should  have  been  made.  This  episode  and  the 
fact  that  the  husband  of  the  heroine  is  invariably  in  the  western 
versions  a  person  of  exalted  rank,  an  emperor  or  king,  indicate 
that  they  had  a  common  source  which  Wallenskold  believed  was 
an  Oriental  tale  introduced  into  Europe  about  the  end  of  the 
eleventh  century.  In  his  opinion  this  tale  was  represented  by 
the  longer  western  versions  and  the  shorter  forms  were  simpli- 
fications of  it  that  were  due  to  oral  tradition. 

Wallenskbld's  theory  was  stoutly  opposed  by  Stefanovic  who 
returned  to  the  idea  of  a  Germanic  origin,  though  on  different 
grounds  from  those  proposed  by  Grundtvig.  He  asserted,  very 
much  as  Bedier  did  in  connection  with  the  fabliaux,  the  lateness 
of  the  extant  Oriental  texts,  and  the  difficulty  of  finding  actual 
evidence  of  their  transmission  to  Europe.  To  him  the  fact  that 
both  the  longer  European  versions  and  the  Oriental  texts  con- 
tained the  episodes  (7  and  8)  of  the  Man  Freed  from  the  Gal- 
lows 12  and  the  Ship's  Captain,  suggested  not  a  common  Oriental 
source,  such  as  Wallenskold  conjectured,  but  rather  the  proba- 
bility that  these  episodes  were  added  to  a  European  original 
of  the  type  represented  by  the  Kaiserchronik,  and  that  it  was 
this  expanded  version  which  passed  to  the  East.  This  theory 
adheres  at  any  rate  to  the  chronology  of  the  extant  texts  and 
offers  a  fairer  interpretation  of  the  relation  of  the  Crescentia- 
Miracle  versions  than  that  suggested  by  Wallenskold  (Florence, 
123-24).  The  latter  argued  that  the  Crescentia  story,  in  which 
St.  Peter  rescues  the  heroine  after  she  has  been  thrown  into  the 
Tiber,  was  in  fact  simply  a  variant  of  the  true  Miracle  type  in 
which  it  was  the  Virgin  herself  who  saved  the  heroine  and  en- 
dowed her  with  healing  power,  or  gave  her  a  magic  healing  herb. 
Yet  stories  of  such  miracles  were  current  long  before  the  Mary- 

12  Cf.  Kohler,  Kleinere  Schriften,  11,  284-86  (1900),  who  cited  numerous 
proverbs  showing  how  widespread  was  the  popular  belief  that  anyone  who 
freed  a  criminal  justly  condemned  to  the  gallows,  thereby  made  an  enemy. 
Stefanovic,  p.  490,  urged  that  the  episode  of  the  Man  Freed  came  from  this 
belief.  In  this  episode  and  that  of  the  Ship's  Captain,  he  admitted  the  possi- 
bility in  the  Florence  story  of  Oriental  influence,  but  thought  that  influence 
improbable. 


20  MEDIAEVAL   ROMANCE    IN    ENGLAND 

cult  of  the  twelfth  century  drew  them  round  her  name.  Stef- 
anovic (p.  572)  referred  to  legends  told  by  Gregory  of  Tours 
(chs.  lxix,  cxxx)  in  which  the  heroine,  who  is  accused  of  adul- 
tery, is  thrown  into  the  river  and  is  saved  by  St.  Genesius  or  by 
Christ.  It  is  also  important  to  notice  in  regard  to  the  chronology 
of  the  Crescentia  and  the  Kaiser chronik  stories,  that  the  Tower 
episode,  already  noted  as  a  distinctive  feature  of  the  Western 
versions  and  appearing,  of  course,  in  the  Kaiser  chronik,  could 
not  have  been  an  original  element  in  the  original  Miracle  version 
since  it  is  lacking  in  several  derivative  texts.  This  fact  also 
favors  the  priority  of  the  Crescentia  tale.  Stefanovic  (p.  541) 
found  no  evidence  of  any  connection  between  the  nameless  Tower 
in  that  story  and  the  actual  tower  of  the  Roman  Crescentius 
(c.  985).  The  many  details  in  the  versions  of  Florence  which 
describe  the  building  and  somewhat  fantastic  appearance  of  the 
Tower,  do  not  conceal  that  it  was  originally  conceived,  as  in  the 
Kaiser  chronik,  simply  as  a  prison. 

In  further  refutation  of  the  Oriental  hypothesis,  Stefanovic 
(P-  535)  pointed  to  some  minor  traits  differentiating  the  two 
types.  The  punishment  to  which  the  heroine  is  subjected  differs 
somewhat  in  each  group:  it  is  attempted  assassination,  drown- 
ing, burning,  in  the  European  tales ;  stoning,  burning  or  hanging 
in  the  Eastern.  So  also  is  the  manner  of  her  delivery  different. 
In  the  oldest  European  texts  she  is  saved  by  supernatural  inter- 
vention ;  in  the  Eastern  tales  she  escapes  by  natural  means.  The 
torture  which  the  heroine's  first  cruel  lover  inflicts  upon  her 
when  he  hangs  her  up  by  the  hair,  is  an  incident  found  only  in 
the  western  versions  and  is  suggestive  of  the  fate  of  the  holy 
Juliana.  These  characteristic  differences,  like  the  details  which 
are  unique  in  the  western  versions,  prove  little  in  themselves, 
but  when  parallels  can  be  drawn  between  them  and  the  themes 
and  incidents  in  European  story  which  precede  by  two  centuries 
or  more  the  possibility  of  Oriental  influence  —  granting  that  this 
did  not  become  effective  in  fiction  until  the  end  of  the  eleventh 
century  —  the  possibility  of  European  origin  becomes  more 
convincing.  In  Stefanovic's  opinion  the  terrible  severity  of  the 
old  Germanic  laws  for  the  punishment  of  adultery  brought 
into  existence  numerous  folk-tales,  which  were  made  all  the 
more  dramatic  by  the  innocence  of  the  accused.     From  these 


FLORENCE   OF   ROME  21 

developed  such  tales  as  those  related  in  the  stories  of  Hildegard 
or  Crescentia,  or  suggested  in  the  Old  English  poem,  the  Wife's 
Complaint.13 

The  last  theory  in  regard  to  the  development  of  the  Florence 
legend  is  that  of  Karl.  He  noted  that  three  of  the  five  groups 
into  which  the  European  versions  of  the  story  may  be  divided, 
the  Gesta,  the  long  French  poem  Florence,  and  the  Miracle, 
make  mention  of  Hungary.  In  the  first  the  heroine  is  a  princess 
of  Hungary;  in  the  second  her  husband  is  a  prince  of  that  land, 
though  by  his  marriage  he  becomes  Emperor  of  Rome;  and  in 
the  third  he  is  a  king  of  Hungary.  These  local  references  Karl 
explained  as  being  due  to  the  influence  of  the  story  of  St.  Eliza- 
beth (d.  1 231)  of  Hungary  whose  actual  experiences  were  easily 
adapted  to  those  in  the  type  story  of  the  Innocent  Persecuted 
Woman.  After  the  death  of  her  husband  on  his  way  to  a  Cru- 
sade, Elizabeth  endured  calumny  and  persecution;  she  was 
exiled  from  her  home,  and  for  some  years  devoted  herself  to  the 
care  of  the  sick.  After  a  life  of  signal  piety,  she  was  canonized 
in  1238.  Before  1236,  however,  the  influence  of  her  story  is 
perceptible  in  the  Miracle  of  Gautier  de  Coincy.  In  large  out- 
line at  least  St.  Elizabeth's  life  accords  with  the  story  told  of 
Florence's  persecutions,  her  saintliness,  and  healing  powers. 
Whether  the  romantic  story  was  influenced,  as  Karl  (p.  176) 
also  urged,  by  the  legend  of  Aimeri,  another  saint  (canonized 
in  1083)  of  the  royal  house  of  Arpad,  in  whose  name  and  piety 
Karl  would  recognize  the  prototype  of  Florence's  husband  Es- 
mere,  is  more  problematical. 

BIBLIOGRAPHY 

Texts:  (i)  Cam.  Univ.  Lib.  Ff.  11,  38  (Bishop  More's  Coll.,  No.  690), 
ed.  Ritson,  1802,  111,  1-92;  W.  Vietor,  Marburg,  1893. 

Studies:  Cf.  Edwardes,  Summary,  p.  252;  Gautier,  Bibliographie,  p. 
103;  Wallenskold  (see  below),  pp.  81-95;  Stefanovic,  pp.  552-56;  Wells, 
Manual,  p.  782. 

Hilka,  A.    "   Zum  Crescentiastoff,"  Archiv.,  cxxxm,  135-41   (1915). 
Karl,  L.     "  Florence  de  Rome  et  la  Vie  de  deux  saints  de  Hongrie." 
Revue  des  Langues  Romanes,  lii,  163-80  (1909). 


13  Stefanovic,  Anglia,  xxxn,  398-433    (1909).    See  here,  Emare,  note  2. 


22  MEDIAEVAL   ROMANCE    IN   ENGLAND 

Knobbe,  A.    Ueber  die  mittelengl.  Dichtung  "  Le  Bone  Florence."   Diss. 

59  pp.     Marburg,  1899. 
Siefken,  O.     Das  gediildige  Weib,  pp.  34-47. 
Stefanovic,   S.     "  Die   Crescentia-Florence   Sage,   eine   kritische   Studie 

ueber  ihren  Ursprung  u.  ihre  Entwicklung."     Rom.  Forsch.  xxix, 

461-556   (1911). 
Teubert,    S.     Crescentia-Studien.     Diss.  Halle,    1906. 
Wallenskold,  A.     (1)  Le  Conte  de  la  Femme  Chaste  convoitee  par  son 

beau-frere.    Etude  de  litterature  Comparee.     Acta  Societatis  Scien- 

tiarinm  Fennicae,  xxxiv,  1-172.     Helsingfors,  1907. 

(2)  "Florence  de  Rome,  Chanson  d'Aventure."    SATF.    (Introd., 
pp.   1-130,  Text).  Paris,  1909. 

(3)  "L'Origine    et    Evolution   du    Conte    de   la    Femme    Chaste." 
Neuphilol.     Mitteilungen,  pp.   67-78.    Helsingfors,    191 2. 

Wenzel,  R.    Die  Fassungen  der  Sage   von  Florence   de  Rome  u.  ihr 
gegenseitiges  Verhaltnis.    Diss.  Marburg,  1890. 


EMARE 

Versions.  The  romance  of  Emare  is  one  of  the  many  branches 
of  that  widespread  "  Constance  Saga  "  of  which  twenty-three 
literary  and  more  than  forty  popular  versions  have  been  listed.1 
In  these  tales  an  innocent  maiden  flees  from  or  is  banished  by 
an  unnatural  father;  she  reaches  a  foreign  land  and  is  there 
married  to  a  prince.  Accused  in  her  husband's  absence  of  bear- 
ing monstrous  offspring,  she  is  banished,  usually  through  the 
machinations  of  her  wicked  mother-in-law.  Letters  are  forged 
and  the  young  wife,  instead  of  being  kindly  treated,  as  her 
husband  has  commanded,  is  exposed  in  a  forest  or  set  adrift  in 
a  rudderless  boat.  In  the  one  case  the  Outcast  Wife  and  her 
two  sons  are  saved  by  a  hermit;  in  the  other,  the  mother  and 
her  one  son  drift  across  the  sea  and  when  they  arrive  at  last  in 
Rome,  find  refuge  in  the  house  of  a  noble  senator  or  merchant 
(Suchier,  Beautnanoir,  i,  pp.  xxiv-lv).  In  all  versions  the  hero- 
ine is  ultimately  reunited  with  her  husband  and  in  some  cases 
with  her  father  also. 

To  England  belong  three  notable  versions  of  this  story.  The 
earliest  text 2  is  that  found  in  the  Vitae  Duorum  Off  arum  (Cot- 
ton Nero  D  I,  Brit.  Mus.).  This  Latin  chronicle  emanated  from 
St.  Albans  and  may  be  ascribed  to  the  abbacy,  if  not  to  the 
actual  authorship  of  the  learned,  pious,  and  somewhat  credulous 
John  de  Cella,  1 195-12 14  (Rickert,  2,  30-39).    The  Vita  Offae 

1  See  note  8. 

2  Occasional  attempts  have  been  made  to  relate  the  Old  English  poem, 
The  Wife's  Complaint  (ed.  W.  Sedgefield,  Anglo-Saxon  Verse  Book,  Man- 
chester, 1922,  p.  35)  to  the  Offa  Saga  (Wulker,  Grund.  d.  ags.  Liter aturgesch., 
pp.  224-27;  E.  Rickert,  "The  OE.  Offa  Saga,"  Mod.  Ph.  11,  29  ff.)  The  poem 
alludes  to  the  exile  of  a  wife,  to  her  sorrow  for  her  husband,  to  the  treacherous 
kinsmen  who  have  separated  them,  to  her  own  life  in  what  seems  to  be  a 
forest  cave.  Wm.  Lawrence,  "  The  Banished  Wife's  Lament,"  Mod.  Ph.  v,  401 
(1908),  thought  it  "equally  impossible  to  prove  that  the  Lament  is  or  is  not 
based  on  the  Offa-saga."  Stefanovic,  Anglia,  xxxii,  431,  argued  that  the 
Complaint  was  connected  not  with  the  Constance,  but  with  the  original  of 
the  Crescentia  or  Hildegard  legend. 

23 


24  MEDIEVAL   ROMANCE    IN   ENGLAND 

Prinii3  (Originals  and  Analogues,  Chaucer  Society,  Lond.,  1872) 
relates  that  the  king  of  York  loves  his  own  daughter,  that  her 
executioners  abandon  her  in  the  forest,  that  she  is  found  by 
Off  a  and  that  she  becomes  his  wife.  After  the  birth  of  her 
children  and  the  lapse  of  many  years,  Offa  goes  to  aid  the  vassal 
king  of  Northumbria  against  the  Scots.  For  Offa's  message  of 
victory  his  son-in-law  substitutes  a  letter  commanding  that  the 
queen  and  her  children  should  be  left  to  die  in  the  woods.  The 
hands  and  feet  of  the  children  are  cut  off  but  are  later  miracu- 
lously restored  by  a  pious  hermit.  The  mother  and  children 
stay  with  him  until  they  are  found  by  the  despairing  Offa.  The 
hermit  suggests  that  in  gratitude  Offa  should  erect  an  abbey  by 
the  hermitage,  but  this  is  not  accomplished  until  the  time  of  one 
of  Offa's  descendants,  who  at  last  begins  the  building  of  St. 
Albans. 

A  century  or  so  after  this  inspired  account,  a  second  version 
of  the  story  was  introduced  by  Nicholas  Trivet,  a  learned 
Dominican  friar,  into  his  Chronique  Anglo-N  ormande,  1334- 
1347  {Originals  and  Analogues,  pp.  3  ff.).  From  this  prose  text, 
in  which  the  gentle  heroine  is  called  Constance,  comes  the 
generic  name  of  the  legend.  Trivet's  account  is  important  in 
itself  and  as  the  source  of  both  Gower's  story  in  the  Conjessio 
Amantis,  Liber  II,  587  ff.  (ed.  G.  Macaulay,  Oxford,  1901),  and 
of  Chaucer's  Man  of  Law's  Tale,4  a  beautiful  version  in  which 
all  the  art  of  the  poet  is  lavished  on  the  tender  pathos,  the 
devoutness,  and  spiritual  fortitude  of  the  heroine's  character. 
In  Trivet's  account  and  its  derivatives,  the  Incestuous  Father 
episode  is  omitted.  Constance,  the  daughter  of  Emperor  Tiberius 
of  Rome,  leaves  her  maiden  home  in  order  to  wed  and  convert 
a  heathen  Sultan.  At  the  wedding  feast  there  is  a  massacre  of 
the  Christian  guests,  a  crime  instigated  by  the  first  of  the  two 
incredibly  similar  and  wicked  mothers-in-law  in  the  story.  The 
young  bride  is  set  adrift  on  the  sea.    She  reaches  Northumbria 

3  The  first  Offa,  it  is  thought,  reigned  in  Schleswig  in  the  fourth  century. 
Allusions  to  him  are  found  in  Widsith,  v.  335~45>  and  in  Beowulf,  v.  1931-62. 
See  F.  Klaeber,  Beowulf,  N.  Y.,  1922,  pp.  187-91;  E.  Rickert,  2,  pp.  53  ff.; 
below,  note  13. 

4  Cf.  E.  P.  Hammond,  Chaucer,  A  Bibliographical  Manual,  N.  Y.,  1908, 
p.  282;  Wells,  Manual,  p.  877;  Liicke,  "  Das  Leben  der  Constanze  bei  Trivet, 
Gower  u.  Chaucer,"  Anglia,  xiv,  77  ff. 


EMARE 


25 


and  after  enduring  various  misadventures,5  she  is  at  last  happily 
married  to  Alle,  the  King.  At  this  point  the  type  story  begins. 
When  Constance's  child  is  born,  her  mother-in-law  prepares  a 
letter  for  Constance's  husband  in  which  the  girl  is  accused  of 
being  a  witch  and  of  bearing  evil  offspring,  and  another  letter, 
purporting  to  come  from  him,  in  which  the  exposure  of  the 
queen  and  her  child  is  commanded.  They  are  set  adrift  and 
come  at  last  to  Rome.  There  the  long  separated  family  is  at 
last  reunited.  Trivet's  story  is  somewhat  dull  but  edifying,  and 
is  of  course  written  in  conventional  chronicle  fashion  with  many 
pseudo-historic  details.  Both  in  material  and  style  the  text 
might  easily  be  turned  into  an  exemplum  on  the  virtue  of  resig- 
nation to  the  will  of  God,  or,  with  the  addition  of  a  martyrdom 
episode,  into  a  saint  legend. 

What  it  became  in  romance  is  shown  in  the  third  version  of 
the  story  that  arose  in  England,  namely,  the  romance  of  Emare, 
a  poem  of  1033  lines.  This  was  written  in  the  last  half  of  the 
fourteenth  century,  or  possibly  as  late  as  1400,  since  it  is  notably 
lacking  in  archaic  forms  (Rickert,  1,  p.  xxviii).  Its  popular 
style,  its  twelve-line,  tail-rime  metre,  its  familiar  allusions  to 
minstrels  (1.  13  ff.),  its  North-East  Midland  dialect,  suggest 
that  its  author  was  a  minstrel  belonging  to  the  Mid- Yorkshire 
district  in  which  Trivet  localized  his  version  (Rickert,  1,  pp. 
xviii,  xxviii).  The  single  extant  manuscript  of  the  poem  was 
written  between  1446  and  1460  (ibid.,  p.  x).  The  story  curiously 
combines  the  motifs  of  the  two  earlier  versions.  It  begins  with 
the  episode  of  the  Incestuous  Father.6     He  presently  orders 

5  Notable  among  these  is  the  episode  of  the  Cruel  Lover  who,  when  his 
love  is  rejected  by  the  heroine,  kills  the  wife  of  her  kind  protector,  puts  the 
bloody  knife  beside  the  maiden,  and  accuses  her  of  the  crime.  This  episode 
is  practically  identical  with  one  in  Florence  of  Rome.  In  the  various  ver- 
sions of  this  romance,  it  is  the  child  of  the  heroine's  protector  who  is  killed. 
The  episode  is  found  in  the  mid-twelfth  century  Kaiserchronik  and  must  have 
been  borrowed  from  some  subsequent  version  of  the  story  by  Trivet.  The 
accusation  of  an  innocent  person  by  the  true  assassin  is  a  motif  frequently 
found  in  popular  tales  (Wallenskold,  La  Femme  Chaste,  p.  10,  n.  2 ;  see 
Florence  of  Rome  bibliography). 

6  The  Incestuous  Father  appears  in  many  versions  of  the  Constance  story. 
In  Manekbie  the  king  has  promised  to  wed  no  one  save  a  woman  like  his  dead 
wife;  he  is  reluctant  when  his  nobles  wish  to  make  him  marry  his  daughter. 
In  the  Catalan  tale,  Historia  del  rey  de  Hungrie,  the  father  loves  the  daughter 
because  of  the  beauty  of  her  white  hands,  and  for  this  reason  she  cuts  them 
off    (Suchier,  Beaumanoir,  1,  p.  xlii).    For  folk  tales  embodying  the  theme 


26  MEDIAEVAL   ROMANCE    IN   ENGLAND 

that  his  recalcitrant  daughter  should  be  set  adrift  on  the  sea. 
Arrived  in  "  Galys,"  she  is  kindly  received  by  the  royal  steward, 
Sir  Kadore,  and  is  promptly  wedded  by  the  king.  The  episode 
of  the  cruel  Mother-in-law  and  the  two  Forged  Letters  is  the 
same  as  in  Trivet,  and  similar  also  is  the  account  of  the  heroine's 
second  Exposure  on  the  Sea,  her  arrival  in  Rome,  and  reunion 
with  her  husband.  Much  in  this  version  is  made  of  Emare's 
beauty,  which  equalled  that  of  her  dead  mother.  The  pseudo- 
historic  details  and  the  accusation  that  the  heroine  has  mur- 
dered the  child  or  wife  of  a  protector  are  omitted,  and  the  super- 
natural element  is  reduced  to  the  two  voyages  in  which  the 
heroine  is  marvellously  preserved.  There  are  no  references  in 
Emare  to  any  mutilation  of  the  heroine,  or  to  the  heavenly 
vengeance  on  her  murderous  lover,  such  as  occur  in  so  many 
versions  of  Constance.  To  a  large  extent  the  story  has  been 
rationalized  and  its  earlier  barbarity  softened.7 

The  varied  development  and  popularity  of  the  Constance  leg- 
end is  best  shown  by  its  history  on  the  Continent.  Suchier 
(Beaumanoir,  i,  xxv  ff.)  analyzed  some  seventeen  literary  ver- 
sions, and  these  have  been  added  to  and  differently  classified  by 
later  scholars;  chronologically,  by  Gough  (2,  pp.  2  ff.)  and  by 
Daumling  (pp.  17  ff.) ;  geographically,  and  by  types,  by  Dr. 
Rickert  (1,  pp.  xxxiii  ff.).8     On  this  basis  it  appears  that  five 


of  the  Father  Who  Wishes  to  Marry  his  Daughter,  see  C.  M.  Cox,  Cinderella 
(Catskin).  In  Sophie  Jewett's  Folk  Ballads  of  Southern  Europe,  N.  Y., 
1913,  p.  23,  is  translated  the  Roumanian  ballad  of  the  Sun  and  Moon,  in 
which  the  Sun  wishes  to  marry  his  sister,  "  Helen  of  the  long  gold  hair." 
The  Incest  motif  was  familiar  enough  in  mediaeval  story.  Cf.  Apollonius, 
the  legend  of  Arthur,  La  Fille  du  roi  de  Hongrie  (Petit  de  Julleville,  Les 
Mysteres,  u,  300-03),  etc.  Rickert,  2,  p.  357,  suggested  that  Charlemagne's 
inordinate  pride  in  refusing  suitors  for  his  daughters  may  have  caused  legends 
of  this  type  to  be  associated  with  him. 

7  Off  a  I  tells  of  the  barbarous  mutilation  of  the  children  and  their  mother; 
Trivet  of  how  Alle  hacked  his  mother  to  pieces  in  vengeance  for  her  crime 
against  his  wife.  In  this  last  chronicle,  the  heroine,  when  she  is  assailed  by 
a  wicked  suitor,  pushes  him  overboard  with  her  own  hands.  Chaucer's 
version,  like  Emare,  minimizes  these  elements  of  horror. 

8  As  full  bibliographical  details  have  been  given  by  the  scholars  men- 
tioned above,  they  are  here  omitted  except  in  cases  of  editions  later  than 
1908.  For  folk  versions  see  Suchier,  Beaumanoir,  1,  lviii-lxxii,  forty-two 
tales;  Rom.  xxx,  xxxix;  Klapper,  Mitteilungen,  for  two  Latin  exempla,  one 
from  a  fifteenth-century  manuscript  in  Breslau  and  one  from  the  Scala  Celi, 
dr.  1300;  Popovic  for  a  Slavic  marchen  which  he  believed  was  derived  from 
Beaumanoir's  Manekine.  See  also  Hudepohl,  mentioned  here  under  Amadas, 
note  2. 


EMARE 


27 


versions  were  written  in  France  between  the  thirteenth  and 
fifteenth  centuries:  La  Manekine,  c.  1270,  a  French  metrical 
romance  by  Philippe  de  Remi,  Sire  de  Beaumanoir,  which  was 
dramatized  in  the  Parisian  Miracles  de  Nostre  Dame  (No.  29, 
SATF,  1880) ;  a  dit  (13 13-16)  by  Jehan  Maillart  (Mart)  called 
La  Contesse  d'Anjou  (ed.  B.  Schumacher  &  E.  Zubke,  Roma- 
nisches  Mus.  I,  Greifswald,  1920) ;  a  romance,  La  belle  Uelhne 
de  Constantinople?  presumably  of  the  late  thirteenth  century, 
though  no  known  manuscript  antedates  the  fifteenth;  an  in- 
edited  fifteenth-century  Latin  play,  Columpnarium,  which  makes 
unique  combination  of  the  tale  of  the  Persecuted  Wife  with  that 
of  Orestes  and  Clytemnestra  (Rickert,  1,  p.  xlix) ;  the  story,  Be 
Alixandre,  Roy  de  Hongrie,  qui  voulut  espouser  sa  fille,  which 
is  found  in  a  fifteenth-century  manuscript  (Langlois,  Nouvelles 
frang.,  Paris,  1908,  pp.  61  ff.).  To  Germany  belong  three  ver- 
sions dating  from  the  thirteenth  to  the  fifteenth  century:  a 
German  metrical  romance,  Mai  und  Beaflor  (c.  1260) ;  the  versi- 
fied tale  by  Jansen  Eninkel  in  his  Weltbuch  (c.  12 77-1300), 
Die  Konigstochter  von  Reussen,  which  is  also  known  in  a  prose 
version  (Daumling,  p.  18) ;  and  another  German  metrical  ro- 
mance, Die  Konigstochter  von  Frankreich  by  Hans  von  Biihel 
(c.  1401).  Three  versions  belong  to  Spain:  a  late  fourteenth- 
century  Catalan  tale,  Historia  del  Rey  de  Hungria ;  a  tale  in  the 
prose  chronicle,  Le  Victorial,  by  Guitierre  Diez  de  Games  (cir. 

9  New  edition  announced  by  E.  Rickert,  Emare,  p.  xxxiii.  For  a  detailed 
analysis  of  the  long  fifteenth-century  poetic  version  in  monorimed  laisses  see 
Albert  Leon,  Une  Pastorale  Basque,  Etude  historique  et  critique,  Paris,  1909, 
pp.  118-77;  for  an  account  of  the  fifteenth-century  prose  romance  see  pp. 
179-94.  Cf.  H.  Bussmann,  Grammatische  Studien  uber  den  Roman  de  la 
belle  Helaine,  Diss.  Greifswald,  1907 ;  W.  Soderhjelm,  "  St.  Martin  et  le  roman 
de  la  belle  Helene,"  Memoires  de  la  Soc.  neophilologique  a  Helsingfors,  1893; 
R.  Ruths,  Die  franz.  Fassungen  des  Roman  de  la  belle  Helene,  Diss.,  Greifs- 
wald, 1897.  For  reproductions  of  the  illuminations  of  the  prose  romance  by 
Jean  Wauquelin  see  J.  van  den  Gheyn,  L'Histoire  de  Relay ne,  Brussels,  1913. 
Suchier,  Beaumanoir,  1,  xxvii-xxxii,  dated  the  original  romance  in  the  thirteenth 
century;  he  noted  the  doubtful  attribution  of  the  romance  to  Alexander  de 
Bernay;  the  evident  familiarity  of  the  author  with  Tours  and  certain  localities 
in  Flanders,  and  with  such  legends  as  those  of  St.  Alexis,  of  St.  Eustache,  and 
of  la  reine  Sibille.  In  this  version  H61ene  bears  two  sons.  They  are  stolen 
from  her  by  animals,  are  saved  by  a  hermit,  and  are  named  by  him  Lyon 
and  Bras,  —  the  last,  because  the  boy  carries  always  with  him  his  mother's 
severed  arm,  and  the  first,  because  the  lad  had  been  stolen  by  a  lion.  When 
they  are  baptized  they  receive  the  names  Martin  and  Brice.  Martin  subse- 
quently becomes  the  famous  saint  of  Tours. 


28  MEDIEVAL   ROMANCE    IN    ENGLAND 

1400),  and  another  fifteenth-century  Catalan  version,  La  lstoria 
de  la  Filla  de  VEmperador  Contasti  (ed.  Suchier,  Rom.  xxx, 
519).  To  Italy  belong  seven  versions  dating  from  the  four- 
teenth to  the  seventeenth  century:  the  Ystoria  Regis  Francho- 
rum  et  Filie  in  qua  Adulterium  Comitere  Voluit  (Rickert,  Emare, 
p.  xxxiv),  a  Latin  prose  tale  written  in  1370  (ed.  Suchier,  Rom. 
xxxix,  61)  ;  an  Italian  novella,  Dionigia,  in  II  Pecorone,  x,  1,  by 
Fiorentino  (1378);  the  late  fourteenth-century  Novella  della 
Figlia  del  Re  di  Dacia ;  an  Italian  romance  in  ottava  rima,  His- 
toria  de  la  Regina  Oliva  (cir.  1400),  on  which  was  based  an  early 
Italian  drama,  La  Rappresentazione  di  Santa  Uliva  (Gough, 
Constance,  p.  5) ;  a  prose  tale  in  the  Miraculi  de  la  Gloriosa 
Verzene  Maria,  cap.  xi,  printed  in  Venice,  1475,  and  later  drama- 
tized (Daumling,  p.  20) ;  a  Latin  prose  version  in  chronicle 
style,  De  Origine  inter  Gallos  et  Britannos  Belli  Historia, 
written  before  1457  by  Bartolomeo  Fazio;  and  a  novella,  La 
Penta  Manomozza,  in  Basile's  Pentamerone  (n.  22),  written 
before  1637. 

These  versions,  as  Dr.  Rickert  (1,  p.  xxxiv)  noted,  show 
clearly  the  progress  of  the  legend.  "  Spreading  from  England, 
by  the  end  of  the  thirteenth  century  it  had  passed  through  France 
and  Germany,  during  the  fourteenth  century  it  reached  Italy 
and  Spain,  died  out  in  Spain  in  the  fifteenth,  but  continued  in 
Italy  until  the  seventeenth ;  in  the  fourteenth  also  it  was  revived 
in  England  in  English,  but  is  not  known  to  have  persisted  long 
after  1400." 

Origin.  The  direct  source  of  Emare  must  have  been  a  French 
lay,  which  may  or  may  not  have  been  of  Breton  origin,  but  was 
very  probably  called,  as  our  poem  asserts,  LEgaree  (1.  1032). 
In  Emare  are  preserved  many  common  words  of  French  origin, 
besides  numerous  proper  names.  Among  these  the  two  names  of 
the  heroine,  Emare,  which  comes  either  from  esmarie  (afflicted, 
troubled)  or  from  esmaree  (in  the  sense  of  one  of  rare  worth), 
and  Egare  from  esgaree  (outcast)  10  indicate  a  French  source. 

10  For  the  derivations,  see  Rickert,  1,  p.  xxix.  Dr.  Rickert,  p.  xxxi,  sug- 
gested that  the  marvelous  robe  embroidered  with  the  stories  of  lovers,  of 
Amadas  and  Ydoyne,  of  Trystram  and  Isowde,  of  Florys  and  Blawncheflour, 
which  in  Emare  is  said  to  have  been  given  to  Emare's  father  by  a  king  of 
Sicily,  may  have  been  inspired  in  the  original  French  version  of  the  poem 
by  the  wonder  felt  over  the  pannis  sericis  which  were  presented  in  iiqi  to 
Richard  Coeur  de  Lion  by  Tancred  of  Sicily. 


EMARE  29 

This,  it  is  believed,  must  have  antedated  the  late  thirteenth- 
century  French  and  German  versions,  but  was  not  earlier  than 
1200. 

In  its  extant  form  Emare  shows  the  influence  of  motifs  popular 
in  romantic  story  from  the  middle  of  the  twelfth  century.  By 
the  end  of  the  century,  in  the  Vita  Offce  Primi,  the  central  theme 
of  Emare,  the  Innocent  Persecuted  Wife,  had  been  combined 
with  that  of  the  Incestuous  Father.  This  last  was  known  not 
only  through  such  a  romance  as  Apollonius  of  Tyre,  but  also,  in 
all  probability,  through  folk-tales  of  the  type  represented  by 
Cats  kin  (Peau  d'dne,  or  Allerleirauh) .  To  this  combination  of 
the  Persecuted  Wife  and  the  Cat  skin  type  of  story,  Suchier  (1, 
p.  lxxix)  ascribed  the  origin  of  the  Constance  legend.  The  em- 
phasis in  Emare  on  the  unearthly  beauty  of  the  heroine,  her 
strange  reluctance  to  explain  herself  even  to  her  rescuers,  —  a 
trait  true  of  practically  all  the  versions,  —  the  accusation  brought 
by  her  mother-in-law  that  the  young  queen  has  given  birth  to 
monstrous  offspring,  indicate  that  the  type  story  of  Constance 
had  also  to  some  extent  become  confused  with  that  of  the  Swan- 
Maiden  legend.11 

The  earliest  known  instance  of  the  device  of  the  Forged  Letters 
in  the  Constance  legend,  is  in  the  Vita  Offce  I  where,  following 
the  king's  absence  from  home,  it  was  a  natural  enough  invention 
and  probably  owed  nothing  to  the  Uriasbrief  and  the  Substituted 
Letters  which  appear  in  Li  Dis  de  VEmpereur  Constant  and 
Beves  of  Hampton.  The  details  of  Constance's  life  in  Rome, 
where  she  is  said  to  support  herself  by  beautiful  needlework, 
seem  reminiscent  of  the  legend  of  Helena,  the  mother  of  Con- 
stantine.  By  the  end  of  the  eighth  century,  genuine  romantic 
coloring  had  been  given  to  this  legend  (Rickert,  1,  p.  xxxix). 
It  told  of  the  laborious,  humble  life  of  Helena  and  her  child,  of 
the  winning  of  the  father's  attention  by  the  boy's  grace  and 
charm,  of  the  revelation  of  the  lad's  identity,  and  of  the  re- 
union of  the  humbly  situated  mother  with  the  royal  father. 
The  influence  of  this  legend  is  especially  perceptible  in  the  va- 
rious versions  of  the  Constance  saga  (with  the  exception  of  the 
Vita  Off 03  I)  through  the  use  of  the  recurrent  names,  Helena, 
Constantine  or  Constans. 

11  See  below,  note  14.    Also  here,  Chevalere  Assigne. 


30  MEDIEVAL   ROMANCE    IN    ENGLAND 

The  attempt  to  find  out  the  original  nature  of  the  Constance 
legend  must  be  based  chiefly  on  the  considerations  which  follow. 
The  more  important  early  versions  localize  the  story  in  England, 
Scotland,  or  Wales,  and  commonly  in  the  ancient  kingdoms  of 
Northumbria  and  Mercia.  The  hero  of  Trivet's  version,  which 
professes  to  be  based  on  ancient  Anglo-Saxon  chronicles  and 
actually  does  contain  a  number  of  typically  Old  English  names, 
is  Alle  (Aella),  the  first  king  of  Deira,  559-88  (Gough,  2,  pp.  21, 
51).  This  ascription  of  the  legend  to  an  historic  personage  is  of 
interest,  but  the  historical  facts  which  might  explain  it  are  ex- 
ceedingly meagre.12  In  the  Vita  Offce  I,  the  story  is  associated 
with  an  Offa  who  is  said  to  be  king  of  the  West  Angles.  The 
two  Offas  of  whom  legend  and  history  know  are  the  fourth- 
century  Offa,  king  of  the  continental  Angles,  and  the  historic 
Offa  of  Mercia  (757-94).  With  each  one  was  associated  a  curi- 
ous marriage  legend  of  a  strange  woman  who  came  to  the  king 
after  her  exposure  on  the  sea.  But  the  Valkyrie-like  Thryth, 
the  woman  who  is  mentioned  in  Beowulf  (1.  1950)  as  having 
come  "  ofer  fealone  flod  "  to  Offa  of  the  Angles,  and  the  ener- 
getic Cynethryth,13  the  actual  wife  of  Offa  of  Mercia,  and  the 
violent  legendary  Drida  who,  according  to  the  Vita  Offce  II, 
suffered  exposure  on  the  sea  before  she  became  his  wife,  were 
women  whose  traditional  characters  do  not  in  the  least  suggest 
that  of  the  patient  and  holy  Constance. 

Yet  a  comparison  of  these  facts  and  legends  suggests  certain 
possibilities  of  important  connections.  It  is  evident  that  by  the 
twelfth  century  the  stories  of  the  two  Offas  had  been  confused ; 
the  marriage  legend  of  the  first  Offa  had  been  more  or  less  com- 
pletely transferred  to  Offa  II,  and  the  latter's  character  and  rule 
had  in  turn  been  confused  with  that  of  his  supposed  ancestor. 

12  In  his  attempt  to  prove  that  there  was  an  Alle-Eadwine  saga,  Gough, 
Constance  (pp.  37  ft".),  had  to  rely  chiefly  on  the  Vita  Offa?  I.  The  available 
facts  seem  to  be  simply  that  ^lla  warred  with  the  Scots,  that  his  famous 
son  Eadwine  was  born  during  these  wars,  and  that  the  latter  endured  a  long 
exile.     This  did  not  begin,  however,  until  after  Ella's  death. 

13  Gough's  study  of  this  personage  brought  together  much  interesting 
historical  material.  Dr.  Rickert,  2,  pp.  327-47,  argued  that  the  lady's  actual 
history  was  confused  with  the  legend  of  Thryth,  with  that  of  Offa's  infamous 
daughter,  Eadburga,  of  whom  Asser  tells,  and  with  the  crimes  of  Queen 
Quoenthryth,  the  slayer  of  St.  Kenelm  (819).  Cf.  R.  W.  Chambers,  Six 
thirteenth-century  drawings  illustrating  the  story  of  Offa  and  Thryth  (Drida), 
Lond.,  1912. 


EMARE  3! 

A  Carolingian  legend  may  also  have  been  introduced  into  this 
complex  tradition.  Offa  II  was  the  actual  contemporary  of 
Charlemagne  and  there  was  at  least  raised  between  them  the 
question  of  the  marriage  of  Charlemagne's  daughter  Bertha  to 
Offa's  son  (Rickert,  2,  p.  348).  This  may  account,  in  the  in- 
extricable legendary  confusion  of  the  two  Offas  and  their  wives, 
for  the  possible  association  with  the  wife  of  Offa  I  of  the  famous 
Carolingian  legend  concerning  Berte  aus  grans  pies,  the  mother 
of  Charlemagne.  She  was  falsely  accused  and  was  condemned 
to  death  in  a  forest.  Her  executioners  were  to  bring  back  as 
evidence  of  her  death  her  heart  or  tongue.14  They,  however, 
had  pity  upon  her  and  she  was  at  last,  after  many  trials,  reunited 
with  her  husband.  Though  this  story  of  an  Innocent  Persecuted 
Wife  may  have  been  partly  instrumental  in  suggesting  a  similar 
tale  for  the  Offa  tradition,  and  though  the  influence  of  the  Berte 
legend,  especially  in  those  elements  suggestive  of  its  Swan- 
Maiden  origin,  is  evident  in  various  versions  of  the  Constance 
legend,  it  is  improbable  that  the  latter  was  in  any  ultimate  sense 
derived  from  the  other.  Their  essential  features  are  too  different. 
The  fact  that  the  Constance  story  in  its  earliest  version,  the 
Vita  Offce  I,  in  Trivet's  version  and  its  derivatives,  and  to  a  less 
extent  in  Helene  de  Constantinople,  in  Beaumanoir's  Manekine, 
in  von  Blind's  poem,  Die  Konigstochter,  is  so  definitely  localized 
in  England,  emphasizes  the  possibility  that  there  was  an  ancient 
local  tale  of  this  general  type.  The  extant  Anglo-Saxon  poem, 
the  Wife's  Complaint,15  is  too  fragmentary,  too  blindly  allusive 
to  the  facts  which  would  explain  the  wife's  presence  in  the 
forest  and  her  grief  for  her  husband,  to  be  mentioned  as  more 
than  an  interesting  possibility.  At  best  it  could  serve  only  to 
suggest  Anglo-Saxon  prototypes  for  the  local  and  romantic  as- 
pects of  the  Constance  story,  but  in  no  way  could  it  account  for 
the  religious  element  which  in  the  character  of  the  heroine  and 

14  The  thirteenth-century  romance,  Berte  aus  grans  pies  (ed.  Scheler,  1874) 
by  Adenes  le  Roi,  was  confessedly  based  on  much  older  versions.  Cf.  Gaston 
Paris,  Hist.  Poet,  de  Charlemagne,  1905,  pp.  166,  526;  Reinhold,  "  Berte  aus 
grans  pies  dans  la  litteratures  germaniques  et  romanes  et  Berte  dans  la 
mythologie,"  Memoires  de  VAcademie  des  Sciences  de  Cracovie,  3e  Ser.,  1910, 
pp.  1-194;  also  Zts.  f.  rom.  PhU.  XXXV.  The  story  of  the  Hired  Murderers 
and  the  Evidence  of  Death  which  they  fabricate,  is  current  in  many  forms. 
See  Schoepperle,  Tristan,  1,  208;  Cox,  Cinderella,  p.  475. 

15  See  above,  note  2. 


32  MEDIEVAL   ROMANCE    IN    ENGLAND 

the  incidents  of  her  life  is  so  strongly  marked  a  feature  of  her 
story. 

This  religious  bias  is  usually  given  to  the  legend  by  the  epi- 
sodes in  which  the  heroine  suffers  mutilation  and  miraculous 
restoration.  Such  episodes  belong  with  a  large  group  of  folk- 
tales known  as  La  Fille  sans  mains.16  This  motif  appears  but 
confusedly  in  the  Vita  Offce  I,  where  it  is  said  that  only  the 
hands  and  feet  of  the  heroine's  children  are  cut  off  and  not  her 
own.  But  the  miraculous  restoration  of  the  children's  limbs  and 
the  heroine's  association  with  a  holy  man  and  a  holy  place,  are 
strongly  emphasized.  Likewise  in  the  other  versions  in  which 
the  Severed  Hand  motif  is  missing,  in  Mai  und  Beaflor,  in  La 
Contesse  d'Anjou,  in  Trivet's  chronicle  and  its  derivatives,  in 
Pecorone,  in  von  Blind's  romance,  in  Emare,  in  Fazio's  novelle, 
its  absence  may  be  explained  by  the  relationship  of  these  ver- 
sions to  each  other,  or  by  surviving  traits  in  them  which  sug- 
gest that  something  has  been  lost.  Daumling  (p.  21)  noted  that 
in  twenty-three  European  versions  of  Constance,  the  motif  ap- 
pears ten  times;  in  these  instances  the  heroine  cuts  off  her 
handfs]  in  order  to  avoid  a  criminal  marriage.17  He  believed 
that  Beaumanoir,  who  was  the  first  to  make  plain  use  of  the 
theme,  probably  borrowed  it  from  some  version  of  the  story 
told  in  the  Legenda  Aurea  (No.  lxxxiii)  concerning  Pope  Leo, 
or  possibly  from  such  a  tale  as  that  of  Jacques  de  Vitry  in  his 
Exempla  (ed.  Crane,  1890,  pp.  22,  158).  In  the  first  there  is 
literal  fulfillment  of  the  command,  "  Si  manus  tua  scandalizat 
te,  abscinde  earn,"  for  the  Pope  cuts  off  the  hand  that  has 
aroused  passion  through  a  woman's  kiss ;  and  in  the  second  tale, 
a  beautiful  nun  whose  eyes  have  excited  the  desire  of  a  prince, 
plucks  them  out  and  casts  them  before  him.  In  La  Manekine 
the  legalistic  Beaumanoir  does  not  state  that  the  girl  sacrificed 
her  hands  because  their  beauty  has  aroused  her  father's  love, 

16  For  the  modern  folk  versions  see  Suchier,  i,  lviii  iL  and  his  articles 
in  Romania;  J.  Bolte  und  G.  Polivka,  Anmerkungen  zu  den  Kinder-  und 
Hausmarchen  der  Bruder  Grimm,  Leipzig,  1913,  I,  297;  for  bibliography  of 
texts  and  studies  to  191 2,  Daumling,  pp.  14-17.  For  a  Philippine  version  of 
Constance,  see  Amer.  Jour.  Folk-Lore,  xxix,  p.  222  (1916). 

17  In  La  Manekine  and  the  Lion  de  Bourges  the  severed  hand  of  the 
heroine  is  swallowed  by  a  fish  and  miraculously  preserved,  a  detail  which 
seems  to  have  been  borrowed  from  the  Polykrates  legend  (Daumling,  p.  28). 


EMARE  33 

but  because  she  knows  he  cannot  take  for  queen  a  woman  "  Qui 
n'ait  tous  ses  membres  "  (1.  798).  In  effect,  however,  the  pietistic 
intention  is  the  same,  for  the  final  Divine  restoration  of  the 
severed  hand  is,  in  the  exempla  and  the  romance,  a  reward  for 
chastity  and  religious  zeal.  The  miraculous  element,  if  we  are 
to  judge  from  the  romance  versions  of  the  Constance  legend  and 
the  many  folk-tales  of  La  Fille  sans  Mains  and  the  representa- 
tions of  the  story  in  art,18  made  the  greatest  appeal  to  the  mind 
of  the  Middle  Ages  and  may  thus  be  held  to  account  for  the 
popularity  of  the  legend  and  the  saintly  character  of  its  heroine. 


BIBLIOGRAPHY 

Texts:  (i)  Cotton  Caligula  A,  II,  Brit.  Mus.,  ed.  Ritson,  11,  204-47 
(1802);  A.  B.  Gough,  Old  and  Middle  Eng.  Texts,  n,  Lond.,  N.  Y., 
Heidelberg,  1901;  E.  Rickert,  EETSES,  xcix,  Lond.,  1908  (rev.  Eng. 
Stud.,  xl,  413). 

Studies  :  Cf .  Wells,  Manual,  p.  783 ;  see  note  8. 

Daumling,  H.  Studie  uber  den  Typus  des  Madchens  ohne  Hande  inner- 
halb  des  Konstanzezyklus.  Diss.  Miinchen,  191 2. 

Gough,  A.  B.  (1)  On  the  Middle  English  Romance  of  Emare,  A  Study 
of  the  Metrical  and  Grammatical  Aspects  of  the  Texts.  Diss.  1, 
Kiel,  1900.  (2)  The  Constance  Saga,  A  Study  of  the  Literary  Ver- 
sions of  the  Saga  and  of  Its  Occurrence  in  English  Historical  Tra- 
ditions.   Diss.  11,  Berlin,  1902;  Palaestra,  xxiii. 

Holthausen,  F.     "Zu  Emare,  v.  49  ff.,"  Anglia  Beiblatt,  1902,  xiii,  46. 

Huet,  G.  "  Les  Sources  de  La  Manekine  de  Philippe  de  Beaumanoir." 
Rom.,  xlv,  94-99  ( 1 918-19). 

Klapper,  J.  "  Das  Marchen  von  dem  Madchen  ohne  Hande  als  Predigt- 
exempel."  Mitteilungen  der  Schlessischen  Gesell.  f.  Volkskunde, 
Heft,  xix,  29-75,  Breslau,  1908;  "  Sagen  u.  Marchen  des  Mittel- 
alters."    Heft  xx,  1-29. 

Popovic,  P.  "  Die  Manekine  in  der  Sudslav  Lit.,"  Zts.  f.  rom.  Phil, 
xxxii,  312-22   (1908). 

Rickert,  E.  (1)  Emare,  see  Texts.  (2)  "The  Old  English  Offa  Saga." 
Mod.  Ph.,  11,  29-76;  321-76  (1904-05). 

18  Suchier,  1,  p.  xv,  referred  not  only  to  the  illuminations  in  the  single 
extant  manuscript  of  Beaumanoir's  poetic  works,  but  also  (p.  liii)  to  the 
ivory  carvings  at  St.  Germain-des-Pres  which  represent  scenes  from  the 
Fille  sans  Main  story  and  from  Florence-Crescentia,  for  illustration  of  scenes 
of  women's  goodness  in  contrast  to  those  of  women's  wickedness  as  represented 
in  the  stories  of  Aristotle  ridden  by  a  woman  and  of  Virgil  in  his  basket. 


34  MEDIEVAL   ROMANCE   IN   ENGLAND 

Stefanovic',  S.  "Das  angelsachsische  Gedicht,  Die  Klage  der  Frau" 
Anglia,  xxxn,  399"433  (i9°9)- 

Suchier,  H.  (1)  Les  CEuvres  Poetiques  de  Philippe  de  Remi,  Sire  de 
Beaumanoir.  SATF.  Paris,  1884-85.  (2)  "La  Fille  sans  Mains." 
Rom.,  xxx,  519-38  (1901);  xxxix,  61-76  (1910). 


THE    ERLE    OF    TOLOUS 

Versions.  The  story  of  the  Erie  of  Tolous  turns  on  the 
theme  of  the  Innocent  Persecuted  Wife.  Left  by  her  husband 
in  the  charge  of  two  false  knights,  the  Empress  of  Almayne  is 
traitorously  wooed  by  them.  When  she  rejects  their  advances, 
they  in  vengeance  introduce  a  youth  into  her  room,  kill  him  in 
the  presence  of  various  nobles,  and  accuse  her  of  infidelity  to 
her  lord.  She  is  condemned  to  death,  her  husband  concurring  in 
the  judgment,  but  is  saved  at  the  last  moment  by  a  champion 
who  kills  in  judicial  combat  one  of  her  accusers  and  forces  the 
other  to  confess.  The  Emperor  richly  rewards  the  rescuer  of 
his  wife  and  only  then  discovers  him  to  be  his  own  former 
enemy,  the  chivalrous  Earl  of  Toulouse. 

In  his  study  of  the  versions  of  this  most  characteristically 
mediaeval  tale,  Ludtke  distinguished  four  main  groups  or  types.1 
To  the  oldest,  the  Catalan,  belong  three  Catalan  chronicle  ver- 
sions, the  first  written  at  the  end  of  the  thirteenth  century  by 
Bernat  Desclot  (Buchon,  Paris,  1840),  the  second  at  the  end  of 
the  fifteenth  century  by  Pere  Miguel  Carbonell  (Barcelona, 
1547),  the  third  in  the  sixteenth  century  by  Pedro  Anton  Beuter 
(Valencia,  1556).  With  these  may  be  grouped  a  fifteenth-century 
Spanish  romance,  El  Conde  de  Barcelona  (Duran,  Romancer 0 
General,  11,  210),  and  two  French  chronicles  written  in  Provence 
in  the  seventeenth  century,  the  first  by  Cesar  de  Nostredame 
(Lyon,  1 6 14),  and  the  second,  La  Royalle  Couronne  des  Roys 
d' Aries,  1641  (Ludtke,  pp.  78  ff.). 

The  representative  of  the  second  group  is  our  Middle  English 
Lay,  a  poem  of  1224  verses  in  twelve-line  stanzas.    Of  this  there 

1  To  Ludtke's  list  of  versions  several  others  must  be  added.  Paris,  p.  8, 
n.  3,  cited  the  fifteenth-century  Catalan  romance,  Curial  y  Guelfa  (ed.  A. 
Rubio  y  Lluch,  1901),  as  one  of  the  Catalan  group  of  the  Erie  of  Tolous 
stories.  Thomas  added  the  story  of  Gaufier  de  las  Tors,  and  Stefanovic, 
that  of  Philopertus  (see  below,  n.  6).  Bolte  (pp.  viii-lx)  added  a  number 
of  versions  of  Bandello's  story  dating  from  the  sixteenth  to  the  eighteenth 
century. 

35 


36  MEDIEVAL   ROMANCE   IN   ENGLAND 

are  four  texts.  The  oldest  and  best  is  the  early  fifteenth-century 
manuscript  (A)  now  in  the  University  Library,  Cambridge ;  the 
next  oldest  (C)  is  in  Lincoln  Cathedral;  and  the  two  Ashmolean 
manuscripts  (BD)  are  both  of  the  sixteenth  century.  These 
four  texts  seem  to  be  independent  derivatives  (AB  and  CD)  of 
two  versions  (xy)  which  had  a  common  source.  In  Ludtke's 
opinion  (p.  41)  the  original  version  was  composed  about  1400 
in  the  North-East  Midland  district,  or,  according  to  Sarrazin 
(p.  136),  who  did  not  credit  Ludtke's  contention  that  the  poem 
showed  signs  of  Chaucerian  influence,  about  a  half  century 
earlier.  The  "  Lay  "  purports  to  be  derived  from  a  "  romance," 
"a  Lay  of  Bretayne,"  2  a  "  geste  —  cronyclyd  in  Rome."  Its 
source  was  probably  a  French  poem  written  in  the  last  part  of 
the  twelfth,  or  the  early  thirteenth  century.  In  the  Erie  of 
Tolous,  the  heroine's  name,  Dame  Beulybon,  is  a  translation  of 
the  phrase  dame  belle  et  bonne  from  this  lost  original  (Lot,  p. 

152). 
Of  about  the  same  period  as  the  English  version  but  very 

different  in  characterization  and  detail  and  in  its  introduction 
of  scenes  of  divine  intervention  in  which  the  Virgin,  Gabriel, 
and  Michael  appear,  is  the  representative  of  another  group,  the 
Miracle  de  la  Marquise  de  la  Gaudine.  This  is  a  typical  miracle 
play,  a  characteristic  instance  of  the  transformation  for  religious 
purposes  of  romantic  themes.  It  is  found  in  the  famous  Parisian 
repertoire  of  the  Miracles  de  Nostre  Dame  in  the  Cange  manu- 
script (Paris  and  Robert,  SATF.  1877,  IJ>  121). 

The  fourth  group  of  versions  seems  to  have  been  derived, 
though  indirectly,  from  the  same  source  as  the  English  poem. 
It  includes  a  Danish  poem  of  the  fifteenth  century,  Den  Kydske 
Dronning  by  Jeppe  Jensen  (Brandt,  Romantisk  Digtning,  1870, 
11,  89  ff.) ;  a  Latin  prose  narrative,  Philopertus  et  Eugenia,  1470 
(ed.  Schuddekopf,  1891,  Zts.  /.  vergl.  littgesch.  iv,  342) ;  the 
sixteenth-century  French  prose  romance  UHistoire  de  Palanus, 
Comte  de  Lyon  (ed.  A.  de  Trebasse,  1833),  which  was  composed 
before  1539;  a  German  Volksbuch,  Eine  schone  .  .  .  History 
vom  edlen  .  .  .  Ritter  Galmien,  c.  1539,  which  is  now  ascribed 

2  Paris,  p.  7,  n.  2,  discounted  this  reference  which  was  probably  taken 
over  from  the  lost  French  original.  Liidtke,  p.  89,  believing  in  the  historic 
fidelity  of  the  Breton  lays,  was  inclined  to  credit  it. 


THE    ERLE   OF   TOLOUS  37 

to  Georg  Wickram  (Bolte,  p.  v)  and  had  long  popularity  (ibid., 
pp.  xvi-xxvi) ;  and  a  tale  of  Bandello's,  Amove  di  Don  Giovanni 
di  Mendozza  e  delta  Duchessa  di  Savoia,  printed  in  1554.  This 
last  was  translated  in  Painter's  Palace  of  Pleasure,  1,  no.  45 
(1566)  and  also  into  Spanish  and  Dutch  (Bolte,  p.  viii).  On 
Wickram's  version  Hans  Sachs  founded  his  play,  Der  Ritter 
Galmi  mit  der  Hertzogin  auss  Britanien,  1552  (Keller,  viii,  261). 

Origin.  There  are  many  versions  of  the  Innocent  Persecuted 
Wife  story  in  which  an  accusation  similar  to  that  brought  against 
the  Empress  of  Almayne  forces  upon  the  heroine  either  sub- 
mission to  an  ordeal  or  endurance  of  many  hard  adventures.3 
Among  such  stories  the  Erie  of  Tolous  has  a  special  place  be- 
cause it  contains  the  distinctive  episode,  the  Judicial  Combat, 
in  which  Gaston  Paris  recognized  the  true  kernel  of  the  story. 

The  earliest  instance  in  western  literature  of  a  story  concern- 
ing a  falsely  accused  queen  and  her  champion  is  the  legend  of 
Gundeberg,  wife  of  the  Lombard  king  Arioald  (cir.  630),  whose 
story  was  recorded  by  Fredegarius,  Paul  the  Deacon,  and 
Aimoinus.4  The  source  was  probably  a  lost  Frankish  or  Lom- 
bard poem.  In  this  tale  the  champion  is  called  Pitto  or  Carellus 
and  the  significance  of  the  diminutive  is  made  clear  in  the  legend 
attached  to  the  name  of  Gunhild,  daughter  of  Canute,  and  wife 
of  him  who  became  the  Emperor  Henry  III  (1036).  In  this 
legend,  according  to  William  of  Malmesbury  (De  Gestis,  11,  c. 
12),  the  champion  of  the  falsely  accused  queen  is  a  mere  boy 
whom  she  has  brought  with  her  from  her  English  home.  In 
the  late  English  ballad  Sir  Aldingar  (Child,  no.  59),  the  champion 
has  become  a  mysterious  unknown  little  creature  of  super- 
natural character.    In  several  of  the  allied  Scandinavian  ballads 

3  Cf.  the  story  of  Richarda,  887,  told  by  Regino  of  Prum,  etc.;  of  St. 
Cundegund,  1024;  of  Emma,  wife  of  Canute  (cf.  Athelstan  here) ;  of  Gunhild, 
told  by  William  of  Malmesbury  and  others;  of  Olif,  sister  of  Charlemagne 
in  the  Karlamagnus  Saga  and  of  Pepin  in  the  Spanish  prose  romance  Oliva, 
1498  (Child,  vol.  11,  37-39).  In  Tristan  (Beroul,  3032  ff.)  to  the  interest  of 
the  Ordeal  is  added  that  of  the  Ambiguous  Oath.  Cf.  Schoepperle,  Tristan,  1, 
223-26;  n,  446-55,  Unlawful  Love  in  0.  F.  Literature.  Child  (n,  43,  n.) 
cited  also  Le  Lai  du  Corn,  1.  325,  in  which  Arthur's  queen  is  willing  to  submit 
to  the  ordeal  by  fire. 

4  For  bibliographical  details,  cf.  Potthast,  Bibliotheca  historica  medit  cevi, 
2d  ed.,  1896.  Rajna,  Le  Origini  dell  'Epopea  francese,  p.  191,  and  Paris,  p. 
27,  both  accepted  the  authenticity  of  the  Gundeberg  story. 


38  MEDIEVAL   ROMANCE    IN   ENGLAND 

the  champion  is  of  diminutive  size,  but  not  supernatural  (Child, 
vol.  II,  34).  Though  the  Erie  of  Tolous  lacks  this  slightly 
ludicrous  aspect  of  the  David-Goliath-like  combat,5  it  is  never- 
theless of  the  same  familiar  story  type  which  had  from  the 
seventh  century  down  thus  associated  together  the  false  accu- 
sation of  a  wife's  adultery,  a  terrorizing  accuser,  and  a  combat 
won  by  an  unexpected  champion.  Within  this  type  of  narrative 
minor  variations  are  of  small  consequence :  whether  the  accusers 
are  one,6  as  in  all  the  stories  just  noted  and  in  the  Miracle  de 
la  Marquise,  or  two,  as  in  the  versions  belonging  to  the  first, 
second,  and  third  groups;  or  what  is  the  size  of  the  champion; 
or  whether  he  does  or  does  not  conceal  his  identity.  Liidtke, 
however,  on  the  basis  of  these  differences,  would  not  grant  that 
there  is  more  than  a  general  resemblance  between  the  Gunhild- 
Gundeberg  legend  and  that  which  he  conceived  to  be  the  primi- 
tive version  of  the  Erie  of  Tolous  as  it  developed  in  legendary 
form  from  certain  historical  personages  and  events. 

The  heroine  of  the  story,  an  Empress  in  the  Catalan  and 
English  groups,  was  identified  by  Liidtke  (pp.  98  ff.,  209  ff.) 
with  the  Empress  Judith,  daughter  of  Wolf  I,  Count  of  Ba- 
varia, and  the  second  wife  of  Louis  the  Pious  (778-840).  She 
was  a  brilliant,  beautiful,  and  masterful  woman,  whose  exertions 
to  secure  a  kingdom  for  her  son  (Charles  the  Bald)  led  to  strange 
chances  and  changes  of  fortune.  Twice  at  least  Judith  was 
exiled  from  the  imperial  court,  charged  with  illicit  relations  with 
Bernard,  Count  of  Barcelona,  son  of  that  famous  William  of 
Toulouse  who  is  known  in  romance  as  William  of  Orange  and 
in  religious  legends  as  St.  Guillaume  de  Gellone.7     She  main- 

5  The  champion  story  was  not  infrequently  associated  with  men  of  nota- 
ble prowess  but  of  ordinary  shape  and  size.  Cf.  Bernard  of  Septimania, 
Gaufier  de  Lastours,  Ramon  Berenger  III,  the  hero  of  Carbonell's  account, 
etc. 

6  Cf.  the  story  of  Philopertus,  who  champions  the  Duchess  Eugenia  of 
Burgundy  against  her  accuser,  Medardus,  a  Latin  tale  which  is  found  in  a 
German  student  notebook  of  the  fifteenth  century.  Ward,  Cat.  of  Romances, 
it  7X3>  wrongly  ascribed  this  to  the  Chaste  Duchess  type  of  story.  Stefanovic, 
Rom.  Forsch.  xxxix,  464,  pointed  out  that  the  combat  scene  makes  it  one 
of  the  Erie  of  Tolous  group. 

7  Cf.  Paris,  p.  15;  Bedier,  Les  Legendes  Epiques,  2d  ed.  (1914),  ch.  iv, 
St.  Guillaume  de  Gellone;  ch.  v,  Guillaume,  Comte  de  Toulouse;  J.  Calmette, 
La  Famille  de  St.  Guilhem,  Annales  du  Midi,  xviii,  145-65  (1906).  Calmette 
showed  that  Bernard  was  the  youngest  son  of  Guillaume  by  his  second  wife, 


THE    ERLE   OF   TOLOUS  39 

tained  her  innocence  before  an  Assembly  of  the  States  at  Aix 
in  831;  and  Bernard  himself,  though  exonerated  by  her  oath, 
later  challenged  any  one  to  maintain  in  battle  the  accusation 
against  him.  When  no  one  accepted  his  offer,  he  withdrew  to 
Barcelona.  The  identity  of  this  Bernard  with  the  Bernard, 
Count  of  Toulouse,  of  the  English  poem  and  of  its  Old  French 
prototype  is  not  to  be  doubted. 

It  is  evident  that  the  historical  situation  was  dramatic  enough 
to  have  appealed  to  the  popular  imagination  and  that  it  was 
peculiarly  capable  of  romantic  transformations.  History  tells  of 
two  political  enemies  of  the  Empress,  Hugo,  Count  of  Tours, 
and  Matfrid,  Count  of  Orleans,  partisans  of  her  stepson  Lothair ; 
the  legend  describes  two  accusers  fired  by  guilty  love  and  fear. 
History  records  Bernard's  militant  offer;  legend  tells  of  bloody 
accomplishment.  In  neither  case  is  the  change  other  than  what 
might  well  be  expected  in  a  romance-loving  age,  and  especially 
in  a  story  as  obviously  made  up  of  romantic  accretions  as  the 
Erie  of  Tolous.  Notable  among  these  is  the  change  from  the 
unsupported  accusation,  which  it  seems  probable  belonged  to 
the  original  story,  to  the  rather  elaborate  conspiracy  in  which 
the  evidence  against  the  heroine  is  fabricated.  This  "  stratageme 
a  la  fois  infame  et  naif  "  by  which  a  pretended  lover  is  dis- 
covered in  the  bed  of  the  Chaste  Wife  is  widely  recurrent.8    In- 


Witburge;  that  Bernard  himself  was  married  in  824  to  a  lady  named  Dhuoda 
and  was  condemned  and  executed  in  844.  Lot,  in  his  valuable  review  of 
Calmette's  book,  De  Bernardo,  found  ch.  vn,  De  Bernardo  in  jabulis,  most 
open  to  criticism.  Calmette  believed  that  the  legend  of  the  Empress's  ex- 
oneration was  derived  from  two  independent  sources,  one  in  Catalan,  and 
one  from  the  South  of  France.  It  is  a  matter  of  unnoted  interest  that  the 
first  wife  of  St.  William  was  named  Cunegonde  (Calmette,  Annates,  xn,  147). 
If,  as  Child  (11,  38)  suspected,  the  exoneration  story  was  told  of  Gunhild, 
1036,  daughter  of  Canute,  because  after  her  marriage  she  was  called  Cunigund 
and  so  was  confused  with  St.  Cunigund  (1002-24),  it  may  be  that  the  same 
name,  borne  by  the  first  wife  of  Bernard's  father,  brought  the  story  into 
association  with  the  great  Frankish  family. 

8  Paris,  p.  12,  n.  1,  thought  this  stratagem  drawn  from  the  Chansons  de 
Geste.  It  occurs  in  the  French,  Italian,  and  English  forms  of  Octavian;  in 
the  various  versions  of  La  Reine  Sibille  (Kohler,  Kleinere  Shriften,  11,  276). 
See  THamour  here.  Together  with  the  stratagem  motif  may  also  be  noted 
the  Prophetic  Dream  of  the  Emperor.  He  dreams  his  wife  is  attacked  by 
wild  beasts  and  so  hastens  home,  only  to  be  met  there  by  the  false  accusation 
of  her  infidelity.  Cf.  Mentz,  Die  Traume,  p.  53;  Beneze,  Das  Traummotiv, 
P.  30. 


4o  MEDIEVAL   ROMANCE    IN   ENGLAND 

deed  it  is  one  of  the  regular  devices  of  that  favorite  villain  of 
mediaeval  romance,  the  False  Seneschal.  The  luckless  dupe  in 
the  different  versions  of  the  Erie  of  Tolous  story  is  variously 
described  as  a  dwarf  {Miracle),  a  scullion  (Volksbuch),  a  servi- 
tor (Danish),  and  in  Bandello's  tale  and  the  Middle  English 
poem  as  a  young  gentleman. 

A  second  important  addition  to  the  original  story,  according 
to  Gaston  Paris  (p.  n),  was  the  introduction  of  the  love  element 
between  the  heroine  and  her  champion.  Liidtke  thought  this  a 
part  of  the  primitive  story  because  there  was  some  suggestion 
for  it  in  the  historic  tradition.  But  the  idea  of  an  amorous 
relation  was  conveyed,  it  must  be  remembered,  only  in  an  accu- 
sation that  was  in  the  eyes  of  mediaeval  law  adequately  dis- 
proved.9 Moreover,  if  the  romantic  element  were  thus  early 
made  an  essential  part  of  the  story,  how  account  for  its  insig- 
nificance in  the  French  prose  romance  of  Palanus  and  the 
German  story  of  Galmi?  It  is  not  strange  that  the  love  theme 
should  not  be  found  in  the  pietistic  Miracle  but  its  absence  from 
the  important  version  discovered  since  Liidtke  and  Paris  wrote, 
goes  far  to  confirm  the  latter's  opinion.  This  version  explains 
that  the  right  to  bear  the  royal  fleur-de-lys  was  granted  to  the 
Provencal,  Goufier  de  Lastours,  because,  under  the  same  circum- 
stances as  those  described  in  the  Erie  of  Tolous,  he  had  saved 
the  life  of  the  Queen  of  France  (Thomas,  p.  59).  This  legend 
is  now  found  only  in  an  eighteenth-century  manuscript  copy  of 
a  lost  text  dating  from  the  end  of  the  fifteenth  or  early  sixteenth 
century,  but  there  is  no  reason  to  doubt  that  the  association  of  it 
with  Goufier  is  of  much  older  antiquity.  Goufier  himself  was  a 
warrior  of  the  First  Crusade ;  he  was  made  famous  by  his  con- 
temporary, Bechada,  and  was  the  hero  of  numerous  stories, 
among  them  a  Knight  of  the  Lion  episode.10  In  the  story  of 
his  championship  of  the  Queen,  there  is  but  one  accuser,  and 
his  service,  performed  after  hearing  the  Queen's  confession,  is 

9  Calmette  (p.  106)  agreed  with  Liidtke  that  the  love  element  was  an 
original  part  of  the  story.  Paris  (pp.  17,  29)  and  Lot  (p.  151)  pointed  out 
that  the  love  motif  is  absent  from  the  oldest,  the  Catalan  versions  of  the 
story,  and  that  historical  references  to  Bernard's  supposed  passion  for  the 
Empress  came  only  from  his  enemies. 

10  Cf.  Paris,  Rom.,  xxn,  358,  n.  1 ;  Thomas,  Rom.,  xxxrv,  56;  F.  Blondeux, 
"La  Legende  du  Chevalier  du  Lyon,  1,  Les  Debuts  de  la  Legende,"  Revue  de 
Belgique,  2d  Ser.,  xxxvm,  xxxix.    See  Guy  of  Warwick  here. 


THE    ERLE   OF   TOLOUS  41 

purely  chivalric  and  not  performed  for  love's  sake.  M.  Thomas 
has  raised  pertinent  questions  as  to  when  and  where  this  legend 
was  thus  ascribed  to  Goufier,  and  has  suggested  that  it  may 
have  been  derived  from  a  Provencal  version  of  the  Erie  of 
Tolous  which  preceded  the  French  original  of  the  English  poem. 
Its  simplicity  at  any  rate  points  to  a  fairly  early  date. 

We  may  also  note  indications  in  other  versions  that  the  rela- 
tion between  the  heroine  and  her  champion  was  originally 
Platonic  and  not  romantic.  In  the  Miracle  the  knight  Anthenor 
performs  his  service  simply  out  of  gratitude  to  the  Marquise ; 1X 
in  the  version  by  Desclot,  the  Count  of  Barcelona  has  never 
seen  the  Empress  but  is  aroused  by  a  minstrel's  story  of  her  sad 
fate;  in  no  version  does  the  lady  grant  a  greater  favor  than  a 
ring  or  a  kiss.  In  short,  the  story  does  seem  to  represent 
"  l'incarnation  du  plus  noble  ideal  chevaleresque,"  of  generosity, 
justice,  and  feudal  loyalty  (Paris,  p.  9).  Even  in  the  Erie  of 
Tolous,  the  most  romantic  of  all  the  versions,  this  disinterested 
service  of  chivalry  dominates  the  situation.  The  Earl,  who  has 
fallen  in  love  with  the  Empress  from  the  account  of  her  given 
by  a  captive,  Sir  Tralabas,  risks  his  life  in  the  enemy's  country 
for  the  sake  of  one  glimpse  of  the  lady,  and  is  then  content  to 
dream  of  her  from  afar.  When  danger  threatens  her,  unlike 
more  passionate  lovers  for  whom  it  is  a  principle  of  courtly  love 
to  make  no  question  of  right  or  wrong  in  regard  to  the  Beloved, 
he  pauses  to  assure  himself  of  her  innocence  before  attempting 
her  defense.12  Having  saved  her,  he  withdraws,  his  identity 
still  concealed ;  13  and  it  is  evident  that  in  this  version  the  con- 
venient death  of  the  Emperor  and  the  marriage  of  his  wife  to 
the  Earl  are  but  happy  after-thoughts.    The  poet  brings  them 

11  By  kissing  Anthenor  in  the  king's  presence,  the  queen  allows  the  latter 
to  suspect  she  is  Anthenor's  love. 

112  In  monk's  disguise  he  hears  her  confession  of  innocence.  Paris,  p.  17, 
n.  2,  referred  to  Baudouin  de  Sebourc;  to  Comte  Claros  (Ludtke,  p.  86), 
and  La  tradition  d'Eginhard  et  d'Emma,  MLN.  vn,  450  ff.  (1892).  Cf.  with 
the  devout  and  idealistic  use  of  the  motif  of  the  Lover's  Confession  in  the 
Erie  of  Tolous,  its  scornful  and  satiric  use  in  the  fabliau,  L'Homme  qui  fist 
sa  femme  confesse  (see  Bedier,  Les  Fabliaux,  pp.  253,  409),  or  in  such  ballads 
as  Queen  Eleanor's  Confession.  For  a  comparative  study  of  the  last  two 
types  see  Hart,  "  The  Fabliau  and  Popular  Literature,"  PMLA.  xxin,  330  ff. 

13  The  concealment  of  the  champion's  identity  is  necessitated  in  the 
Erie  of  Tolous  by  the  mortal  enmity  between  himself  and  the  Emperor. 
Paris,  p.  9,  believed  that  this  was  an  original  element  in  the  story. 


42  MEDIAEVAL   ROMANCE    IN    ENGLAND 

in  almost  incidentally;  his  real  concern  and  achievement  is  to 
make  lifelike  and  touching  the  stock  character  of  the  Chaste 
Wife,  which  in  other  stories,  like  those  of  Constance  or  Florence 
of  Rome,  inclines  to  saintliness  rather  than  credibility  (Siefken, 
p.  66).  Memorable  in  truth  in  the  Erie  of  Tolous  is  this  un- 
named Empress  of  Almayne  for  the  vigor  of  her  scorn  against 
her  false  guardians  and  the  treacherous  Trylabas,  and  for  the 
grave  and  beautiful  dignity  with  which  she  requites  the  reckless 
gallantry  of  the  Earl's  attempt  to  see  her.  In  no  version,  per- 
haps, is  she  a  finer  or  purer  or  more  vitalized  character  than  in 
this,  but  none  the  less  her  nobility  is  not  conceived  here  more 
than  elsewhere  for  romantic  purposes. 

The  allusion  in  the  Erie  of  Tolous  to  an  original  Breton  lay 
as  its  source  must,  obviously,  be  taken  as  a  conventional  refer- 
ence, for  there  is  nothing  in  the  poem  characteristically  Celtic. 
The  widespread  diffusion  of  stories  of  the  same  type  led  Child 
(vol.  ii,  34)  to  believe  that  to  seek  a  single  source  for  them  all, 
too  much  emphasizes  "  the  poverty  of  human  invention."  But 
there  is,  it  must  be  admitted,  a  certain  persistent  resemblance 
in  the  stories  associated  not  only  with  Gundeberg  and  her  de- 
fender Pitto  and  with  the  Empress  Judith  and  Bernard,  but 
with  other  ladies  such  as  the  wife  of  Sancho  the  Great,  King  of 
Navarre  (1001),14  with  Gunhild  (1036),  daughter  of  Canute, 
and  in  later  times  with  Marie  de  Brabant,  second  wife  of 
Philippe  III  of  France  (1276).15  Fact  and  fiction  are  always 
strangely  mingled  in  mediaeval  tradition,  and  in  some  cases  the 
one  influences  the  other.  In  Paris's  opinion  (p.  28),  the  authen- 
ticated seventh-century  history  of  Gundeberg,  itself  perhaps  in- 
fluenced by  some  earlier  imaginary  poem  of  a  queen  delivered 
by  a  champion  in  a  judicial  combat,  passed  into  France  as  a 
poem  which  was  in  its  turn  to  influence  the  authentic  story  of 
Judith  and  Bernard  when  their  fortunes  delivered  them  into  the 
mouths  of  story-tellers.  In  England  there  is  hardly  any  doubt 
that  the  earlier  history  of  Gundeberg  was  substantially  the  base 
of  the  story  told  of  Gunhild  and  her  dwarf  defender,  Mimecan 
{Lives  of  Edward  the  Confessor,  ed.  Luard,  11.   506-31).     In 

14  Mila  y  Fontanals,  Poesia  heroico  —  popular  castellana,  1874,  p:  200. 
Paris,  p.  22,  thought  it  probable  that  this  is  the  oldest  extant  trace  of  the 
legend. 

15  See  Paris,  p.  31. 


THE    ERLE   OF   TOLOUS  43 

Spain  the  story  of  Sancho's  wife  keeps  clearly  the  outlines  of 
the  romanticized  Judith  narrative.  Here  the  champion  has  to 
fight  against  two  enemies ;  here  a  monk  and  a  confession,  though 
differently  introduced,  play,  as  in  the  romantic  version,  an  im- 
portant part  (Paris,  p.  22,  n.  i).  In  England  again,  the  Erie 
of  Tolons,  which  alone  of  all  the  extant  versions  of  the  story 
keeps  the  name  Bernard,  shows  the  survival  of  this  bit  of  fact 
in  all  the  changes  which  fiction  imposed  on  the  actual  story  of 
Judith  and  Bernard. 

The  original  form  of  the  story  of  Judith  was  supposed  by 
Liidtke  to  have  been  a  ninth-century  Latin  account,  written, 
perhaps,  by  a  partisan  of  Bernard's.  Paris  (p.  19)  agreed  with 
this  conjecture  and  suggested  that  the  name  Palanus  in  the 
French  prose  version  may  well  go  back  to  some  form  of  this 
Latin  story  in  which  the  hero  may  have  been  styled  "  Comes 
quidam  palatinus."  Lot  (p.  153)  accepted  the  explanation,  think- 
ing that  the  word  palatinus  may  have  been  introduced  as  a  gloss 
into  the  text.  He  also  reemphasized  the  necessary  anonymity  of 
the  characters  in  any  contemporary  story  in  which  Bernard  of 
Septimania  played  his  part. 


BIBLIOGRAPHY 

Texts:  (i)  A,  Cambridge  •  Univ.  Lib.  Ff.  11,  38,  ed.  Ritson,  hi,  93- 
144;  from  1.  895  to  end,  by  O.  F.  Emerson,  Middle  English  Reader, 
N.  Y.,  1908,  1915;  (2)  B,  Oxford,  Ashmole  45;  (3)  D,  Ashmole  61; 
(4)  C,  Lincoln  Cathedral,  Thornton  MS.  A.  5.  A  critical  edition  of  all 
the  texts  was  made  by  G.  Liidtke,  Berlin,  1881,  Sammlung  englischer 
Dankmaeler  in  kritischer  Ausgaben,  vol.  in;  rev.  Sarrazin,  Eng.  Stud., 
vii,  136-40;  Anglia  V,  Anz.,  pp.  4-6  (1882).  Trans.  E.  Rickert,  Ro- 
mances of  Love,  pp.  80  ff. 

Studies:  Cf.  Korting,  Grundriss,  §125;  Wells,  Manual,  pp.  137,  784. 

Bolte,  J.  "  Georg  Wickrams  Werke,  Galmy."  Bibliothek  des  litteraris- 
chen  Vereins  in  Stuttgart,  vol.  222.    Tubingen,  1901. 

Calmette,  J.  "  De  Bernardo  Sancti  Guillelmi  filio  "  (c.  844).  117  pp. 
Toulouse,  1902.  Rev.  by  F.  Lot,  Le  Moyen  Age,  2e  Ser.,  vm,  148- 
54  (1904).     Cf.  Calmette,  Annates  du  Midi,  1906,  pp.  145-65. 

Child,  F.  J.     Ballads,  n,  33-44  (1886). 

Lot,  F.     See  Calmette. 

Liidtke,  G.     See  Texts. 


44 


MEDIAEVAL   ROMANCE    IN    ENGLAND 


Paris,  S.     "Le  Roman  du  Conte  de  Toulouse."    Annates  du  Midi,  xn, 

1-3 1    (1900). 
Sarrazin,  G.     See  Texts,  Ludtke. 
Siefken,  O.     Das  geduldige  Weib,  pp.  66-68. 
Thomas,  A.    "  Le  Roman  de  Goufier  de  Lastours."    Rom.  xxxrv,  55- 

65   (1905). 


KING    OF    TARS 

Versions.  The  earliest  known  version  of  the  story  embodied 
in  the  King  of  Tars  is  found  in  the  Reimchronik  (Scriptores 
Rerum  Austriacarum,  in,  c.  192-93),  which  was  written  before 
1290  by  Ottokar  von  Horneck  (Krause,  p.  28).  Like  the  English 
poem  this  tells  of  the  love  of  a  heathen  king  for  a  Christian 
princess,  of  their  marriage,  of  the  birth  of  a  strange  offspring, 
of  its  transformation  at  baptism,  and  of  the  consequent  conver- 
sion of  the  father.  In  this  version,  however,  the  princess  is  given 
to  the  Tartar  king  in  the  hope  that  she  may  convert  him;  the 
child  which  she  bears  is  beautiful  on  one  side,  rough  and  hairy 
on  the  other;  because  of  it  she  is  accused  of  adultery  and  sen- 
tenced to  death;  she  then  demands  that  the  child  be  baptized, 
and  her  husband,  overcome  by  its  transformation,  himself  im- 
mediately receives  baptism  together  with  twelve  of  his  knights. 

In  contrast  to  this  first  version,  which  is  told  in  255  lines, 
the  Middle  English  version  in  1228  lines,  is  greatly  amplified. 
It  has  some  notable  changes  of  detail,  and  shows  a  strong  ten- 
dency to  turn  the  story  into  a  pietistic  romance.  That  the  story 
enjoyed  a  certain  real  popularity  in  English  is  suggested  by  the 
three  fourteenth-century  manuscripts  in  which  it  is  found.  Of 
these  the  oldest  is  the  Auchinleck  manuscript  (1330-40),  which 
is  probably  not  much  later  than  the  original  version.  The  Ver- 
non manuscript  (1370-80)  was  derived  from  the  same  source  as 
the  Auchinleck  (A)  and  is  closely  related  to  Additional  manu- 
script 22283  of  the  British  Museum  (Ward,  Catalogue,  1,  763). 
All  three  texts  were  written  in  the  twelve-line,  tail-rime  stanza 
with  the  rime  scheme  aabaabccbddb,  and  with  considerable  use 
of  alliteration  (Krause,  p.  10).  The  diction  is  full  of  the  con- 
ventional formulas  of  expression:  the  poem  begins  and  ends 
with  prayers  and  contains  an  unusual  number  of  purely  religious 
phrases,  such  as  "  For  Marie,  3at  swete  <5ing,"  "  Jesus  3at  dv3ed 
on  tre},"  "  Of  Jesu  Crist  in  trinite,"  "  bi  Jesu  ful  of  mi3t,"  etc. 

45 


46  MEDIAEVAL   ROMANCE    IN    ENGLAND 

Although  its  substance  is  ecclesiastical,  a  minstrel  origin  for 
the  poem  is  suggested  by  its  style,  by  the  appreciative  praise 
of  gift-giving  to  minstrels  (1.  556),  and  by  the  frequent  allusions 
to  the  teller  of  the  tale  or  to  those  hearing  it.  The  original  poem 
was  probably  composed  in  the  Midland  dialect,  perhaps  in  that 
North-Eastern  district  from  which  came  Amis  and  Amiloun.  The 
A  and  V  manuscripts  show  some  mixture  of  forms,  northern 
ones  predominating  in  A  and  southern  ones  in  V  (Krause,  p.  19). 

Origin.  The  immediate  source  of  the  King  of  Tars  seems  to 
be  a  story  told  under  the  year  1299  in  the  Flores  Historiarum. 
This  chronicle  was  long  wrongly  ascribed  to  Matthew  of  West- 
minster, but  is  in  reality  the  work  of  Matthew  Paris,  whose 
Chronicle  from  the  Creation  to  1265  was  included  in  the  Flores  1 
and  extended  to  1303  by  some  monk  of  Westminster  Abbey. 
In  this  account  we  are  told  of  Paganus,  brother  of  the  king  of 
the  Tartars,  and  his  love  for  a  Christian  princess  of  Armenia, 
of  the  refusal  of  Paganus  to  become  Christian  at  the  insistence 
of  the  maiden's  father,  of  the  threatened  war,  and  of  the  maiden's 
sacrifice  of  herself,  "  salute  gentis  suae,  velut  Hester  altera,"  in 
order  to  prevent  such  woe.  With  the  exception  of  this  last  de- 
tail the  story  is  identical  with  that  in  the  Reimchronik  and  un- 
doubtedly represents  the  same  Eastern  tale  brought  home  to 
Germany  and  to  England  by  returning  travelers  or  Crusaders. 
It  is  even  possible  that  the  Templar  whose  stories  of  the  East 
are  recorded,  also  under  the  year  1299,  in  the  Annates  Anglioe  et 
Scotice,2  may  have  been  the  means  of  transmission  so  far  as 
England  was  concerned. 

The  story  is  briefly  alluded  to  in  the  Chronica  (12  59-1306)  of 
William  Rishanger  and  is  retold  in  terms  practically  identical 
with  those  in  the  Flores  in  the  Historia  Anglicana  (Rolls  Series, 
pp.  77  and  113)  of  Thomas  of  Walsingham  who  died  about  1422. 
As  Krause  has  pointed  out,  the  priority  of  the  King  of  Tars  to 
this  version  and  the  omission  of  the  suggestive  passage  about 
the  voluntary  sacrifice  of  the  princess,  preclude  consideration 
of  this  text  as  a  source  of  the  romance. 

1  Cf.  Krause,  p.  28.  For  bibliographical  discussion  of  the  Flores  and.  its 
true  authorship  see  Gross,  Sources  of  Eng.  Hist.,  §  1774. 

2  Ed.  by  H.  T.  Riley,  W.  Rishanger,  Chronica  (Rolls  Ser.  1865,  p.  400). 
Krause  did  not  refer  to  this  possibility  nor  to  the  version  of  the  story  in 
Rishanger's  chronicle,  p.  189. 


KING    OF    TARS  47 

From  the  twenty-two  lines  of  the  Latin  text  of  the  Flores 
the  Middle  English  poet  has  developed  his  story  in  characteristic 
fashion.  The  beauty  of  the  heroine  is  stressed  and  the  heathen 
Sultan  is  said  to  fall  in  love  with  her  simply  from  hearsay,  a 
situation  which  recalls  that  in  Trivet's  version  of  the  Constance 
legend 3  and  the  still  more  romantic  ardor  of  the  plaintive 
troubadour,  Jaufre  Rudel.4  The  Sultan  is  at  once  the  haughty 
suitor  who  offers  the  choice  between  himself  and  death  to  the 
heroine  and  the  typical  Saracen  of  mediaeval  fiction.5  He  flies 
into  ungovernable  rages,  smashes  tables,  and,  when  his  prayers 
remain  unanswered,  beats  the  images  of  his  gods  "  till  he  gan 
to  swete."  Romantic  scenes  are  elaborated  with  the  usual  de- 
scriptive detail :  there  is  a  great  tournament  and  a  wedding  feast 
to  which  the  bride  comes  clad  "  in  riche  palle "  ;  there  are 
two  battle  scenes,  that  between  the  King  of  Tars  and  the  Sultan 
before  the  marriage,  and  the  great  conflict  which  later  the  con- 
verted Sultan  wages  against  his  own  recalcitrant  subjects.  Like 
many  another  heroine  of  romance  the  princess  dreams  (1.  418  ff.) 
that  hounds  attack  her,  but  it  is  a  characteristic  touch  in  this 
particular  story  that  makes  her  fancy  that  one  of  them  turns 
into  a  white-clad  knight  and  addresses  her  in  terms  of  pious 
consolation.  The  change  is  supposed  to  be  prophetic  of  the 
moment  when  her  heathen  husband,  having  received  baptism, 
is  transformed  from  black  to  white. 

The  naive  piety  of  the  tale  is  perhaps  its  most  striking  feature. 
Indeed,  piety  seems  to  have  been  the  author's  chief  concern, 
for  he  scatters  religious  allusions  broadcast  through  the  poem, 
emphasizes  the  heroine's  saintly  resignation  and  fortitude,  con- 
trasts the  saving  power  of  the  Christian  Triune  God  with  the 
false  helpless  gods  of  the  Saracens,  and  sets  forth  the  articles 
of  Christian  faith  in  what  is  practically  a  sermon  preached  by 
the  princess  to  her  penitent  husband.  The  misconceptions  of 
Mohammedanism,  characteristic  of  the  period,  and  the  fanatic 
zeal  of  the  romance  bear  witness  to  its  connection  with  the 
Crusading  era.     To  its  primary  religious  impulse  may  be  as- 

3  Cf.  Emare. 

4  Cf.  O.  H.  Moore,  "  Rudel  and  the  Lady  of  Dreams,"  PMLA.  xxix,  527 

ff.  (1914). 

5  Cf.  the  Saracen  characters  in  Richard  Coeur  de  Lion,  Beves  of  Hamp- 
ton, etc. 


48  MEDIEVAL   ROMANCE    IN   ENGLAND 

cribed  some  of  those  changes  which  differentiate  the  romance 
from  the  legend.  In  all  the  other  versions,  for  instance,  the 
disfigured  child6  is  supposed  by  the  father  to  be  the  result  of 
the  mother's  sin.  In  the  English  story,  in  which  the  Sultan 
delays  his  marriage  with  the  princess  until  she  has  at  least  out- 
wardly accepted  his  faith,  the  shapeless  and  inert  offspring  which 
she  bears  is  thought  by  each  parent  to  be  the  result  of  the  other's 
lack  of  faith. 

BIBLIOGRAPHY 

Texts:  (i)  Auchinleck  MS.  f.  7,  described  by  Kolbing,  Eng.  Stud. 
vii,  178;  (2)  Vernon  MS,  Bodleian,  3938,  f.  304,  ed.  Ritson,  11,  156- 
203  (1802).  Both  MSS.  were  printed  by  Krause,  Eng.  Stud,  xi,  33,  ff. 
(1887).  (3)  Additional  MS.  22283,  f-  126,  Brit.  Mus.  described  by 
Ward,  Catalogue  of  Romances,  1,  767  ff.  Cf.  Brown,  Register,  11,  No. 
745- 

Studies:  Cf.  Wells,  p.  122;  782. 
Krause.     See  Texts. 

Schoneld,  H.    Eng.  Lit.  .  .  .  to  Chaucer,  p.  312. 
Ward.     See  Texts. 


6  The  episode  of  the  Misbegotten  Child  is  paralleled  in  Theseus  of  Co- 
logne, a  chanson  de  geste  of  which  the  first  recension  must  have  been  com- 
posed early  in  the  fifteenth  century  (Ward,  Catalogue  of  Romances,  1,  771). 
Theseus,  the  son  born  to  King  Floridas  of  Cologne  and  his  wife,  a  princess 
of  France,  is  ugly  and  deformed.  A  rejected  suitor  of  the  queen  convinces 
Floridas  that  the  child  is  the  son  of  the  queen  and  the  royal  dwarf.  A 
miracle  ultimately  restores  the  child's  beauty  and  he  is  recognized  as  the 
true  royal  heir. 


GOWTHER    (ROBERT   THE    DEVIL) 

Versions.  Despite  the  changes  in  the  hero's  name,  in  his 
ultimate  fate,  and  in  the  localization  of  his  story,  the  Middle 
English  romance  of  Sir  Gowther  is  easily  recognized  as  a  ver- 
sion of  the  famous  legend  of  Robert  the  Devil,  a  man  so  pos- 
sessed of  evil  that  he  commits  every  crime  ere  repentance  comes 
to  him.  His  story  is  known,  according  to  Breul's  list  (pp.  198- 
207),  in  one  hundred  and  six  texts,  though  many  of  these  are  but 
modern  reprints  of  older  versions.  Of  these  fifty-three  belong  to 
France,  eleven  of  them  antedating  the  sixteenth  century,  sixteen 
to  Spain,  —  the  earliest  a  sixteenth-century  text,  —  three 
eighteenth-century  texts  to  Portugal,  eleven  to  England,  none 
of  which  are  earlier  than  the  fifteenth  century,  five  to  the 
Netherlands,  thirteen  to  Germany,  these  chiefly  in  the  form  of 
Volksbucher,  and  five  are  related  French  and  English  legends. 
The  modern  popularity  of  this  typically  mediaeval  story  is  one 
of  its  most  amazing  features,  and  is  best  explained  by  the  uni- 
versal love  for  melodramatic  story  that  combines  excitement  with 
unforgettable  "  doctrine." 

The  original  version  of  the  long  roman  d'aventures,  Robert  le 
Diable  (Loseth,  Paris,  1902),  is  ascribed  by  its  editor  on  lin- 
guistic grounds  to  the  late  twelfth  century.  In  this  text  there 
are  references  to  an  earlier  written  source  which,  indeed,  one 
would  presuppose  from  the  style  of  the  narrative  itself.  The  ro- 
mance, containing  over  five  thousand  lines  of  verse,  is  extant  in 
two  manuscripts,  one  of  the  thirteenth  and  the  other  of  the 
fourteenth  century.  In  this  version  the  hero's  name  is  Robert, 
and  his  parents  are  the  Duke  and  Duchess  of  Normandy.  The 
story  is  localized  in  Normandy,  and  Normandy  is  the  heritage 
which  Robert  rejects  at  the  end  in  order  to  continue  in  Italy 
his  pious  penances  and  solitude.  After  his  death  he  is  buried 
in  the  cathedral  of  St.  John  Lateran  at  Rome,  but  his  bones  are 
later  stolen  by  a  rich  man  from  Puy  in  Velay  and  placed  in  the 

49 


5o  MEDIEVAL   ROMANCE    IN   ENGLAND 

great  abbey  of  St.  Robert  founded  in  his  honor.  In  another 
version,  the  brief  Latin  prose  exemplum  (pr.  Breul,  p.  208) 
which  the  Dominican  monk,  Etienne  de  Bourbon,  recorded  was 
told  him  "  a  duobus  fratribus,  a  fratre  qui  hoc  se  legisse  as- 
serebat,"  the  legend  is  baldly  treated.  The  whole  story  of 
Robert's  iniquitous  early  life,  a  matter  of  some  five  hundred 
lines  in  the  romance,  is  given  in  eighteen;  and  in  general  no 
names  of  persons  or  places  appear  except  the  name  Robert  and 
the  mention  of  his  penitent  journey  to  Rome.  Etienne's  collec- 
tion of  exempla  seems  to  have  been  made  about  1261  (Loseth, 
pp.  xvii,  xlvii). 

From  the  same  original  as  the  romance  and  the  exemplum 
came  two  prose  versions,  one  in  the  Croniques  de  Normandie, 
Rouen,  1487  (Breul,  p.  57),  which  was  written  late  in  the  thir- 
teenth century  and  several  times  reprinted,  and  the  other  in  a 
fifteenth-century  German  prose  redaction  (Borinski,  Gerrnania, 
1892,  xxxvii,  44).  As  Loseth  (p.  xviii)  pointed  out,  all  these 
texts  are  characterized  by  the  pietistic  ending  which  tells  of 
Robert's  death  as  a  holy  hermit.  In  the  romantic  versions  the 
hero  regularly  marries  a  princess  and  dies  as  the  ruler  of  his 
own  land.  The  earliest  of  these  romantic  texts  is  the  Dit  de 
Robert  le  Diable  (Breul,  Tobler-Abhandlungen,  1895),  a  four- 
teenth-century poem  of  over  two  hundred  four-line  stanzas. 
From  this  were  derived  the  Miracle  de  Robert  le  Dyable  (Paris, 
Miracles  de  Nostre  Dame,  1881,  vi)  and  an  inedited  Vie  de 
Robert  written  in  French  verse  by  Jacques  de  la  Hague  early 
in  the  sixteenth  century.  In  France  the  prose  romance  printed 
at  Lyons,  1496,  and  Paris,  1497,  became  the  most  famous  ver- 
sion of  the  story.  It  was  also  widely  known  outside  of  France 
and  served  as  the  basis  for  the  popular  Spanish,  Portuguese, 
Dutch,  and  German  versions.  In  the  sixteenth  century  the  story 
was  again  versified  in  French,  in  the  eighteenth  it  was  revived, 
and  in  the  nineteenth  century  the  legend,  in  one  form  or  another, 
served  for  a  French  pantomime,  a  ballet,  a  mystfre,  a  ballad, 
and  a  grand  opera  (Breul,  pp.  50-67). 

In  England  the  story  seems  to  have  made  its  way  slowly. 
Its  influence  is  apparent  in  the  fourteenth-century  poem,  Roberd 
of  Cesile,  and  in  the  fifteenth-century  poem,  Sir  Gowther,  but 
not  until  the  sixteenth  century  did  the  Robert  story  itself  be- 


GOWTHER    (ROBERT   THE   DEVIL)  51 

come  well  known.  It  was  then  translated  from  the  French,  and 
published  in  two  editions  by  Wynkyn  de  Worde  (Esdaile,  Eng. 
Tales,  p.  120).  This  English  version  was  turned  into  a  metrical 
romance  (Hazlitt,  Remains,  1864,  1,  217)  by  some  anonymous 
writer,  and  into  a  long  dull  prose  romance  by  Thomas  Lodge, 
1591  (Breul,  p.  98). 

The  Middle  English  romance  of  Sir  Gowther  is  found  in  two 
manuscripts  of  the  latter  half  of  the  fifteenth  century,  both 
written  in  the  twelve-line,  tail-rime  stanza  form.  To  Breul  (p. 
5)  the  text  of  B  (Royal  17),  which  is  less  alliterative  than  that 
of  A  (Advocates  Libr.)  and  somewhat  more  learned  in  the  con- 
ventions of  romance  parlance,  showed  signs  of  revision  for  a 
more  cultured  audience.  Breul  did  not  think  that  the  original 
poem,  probably  composed  in  the  North-East  Midland  district, 
was  written  much  before  the  beginning  of  the  fifteenth  century. 
No  reference  to  Sir  Gowther  in  contemporary  literature  has  been 
noted;  and  since  the  two  extant  manuscripts  belong  to  prac- 
tically the  same  district,  the  story  itself  perhaps  enjoyed  no 
widespread  popularity.  To  its  connection  with  the  Robert 
legend,  Sir  Gowther  makes  no  allusion.  It  purports  rather  to 
come  from  "  a  lai  of  Breyten  "  for  which,  the  poet  remarks,  he 
had  long  to  seek.  On  his  own  account,  apparently,  he  introduces 
the  information  that  Gowther,  while  in  fool's  guise,  was  called 
Hob,  a  popular  name  which  Breul  (note  to  1.  371)  thought  might 
be  connected  with  that  in  a  popular  song,  "  Nou  kyng  Hobbe 
in  5e  mures  3onge5."  The  Middle  English  poet  also  states  that 
after  Gowther's  death  and  burial  at  an  abbey,  where  he  was 
"  a  varre  corsent  parfytt  "  and  where  his  shrine  became  a  place 
of  healing  miracles,  he  was  called  Seynt  Gotlake.  This  shows 
an  evident  confusion  of  the  hero's  name  with  that  of  St.  Guthlac, 
founder  of  Croyland  Abbey. 

Origin.  Until  after  the  publication  of  Breul's  conclusive 
study  of  the  marchen  elements  in  the  Robert  legend,  the  story 
was  supposed  to  have  originated  in  Normandy,  the  scene  of  its 
action,  and  the  hero  was  identified  with  various  early  dukes. 
One  by  one  these  conjectural  identifications  have  been  given  up, 
for,  as  Breul  (pp.  107-17)  pointed  out,  the  facts  concerning  the 
historical  characters  do  not  agree  with  those  set  forth  in  the 


52  MEDIEVAL   ROMANCE    IN   ENGLAND 

legend.  Since  Breul's  work  on  the  subject,  one  further  attempt 
has  been  made  to  identify  Robert  with  a  historical  personage.1 
It  is  not  impossible  that  the  name  and  Norman  origin  of  Robert 
Guiscard  (1015-1089),  the  remembrance  of  his  savage  raids  in 
Apulia,  and  of  his  warfare  on  the  invading  Turks,  may  have  in- 
fluenced the  literary  forms  of  the  Robert  legend.  But  it  is  dif- 
ficult to  believe  that  the  historical  character  actually  gave  rise 
to  the  legend ;  for  in  Italy,  where  in  that  case  the  legend  must 
have  originated,  there  is  no  trace  of  it,  and  the  actual  details 
of  Guiscard's  life  and  death  cannot  be  identified  with  those  of 
Robert. 

The  legend  begins  with  an  account  of  the  marriage  of  Robert's 
parents,  of  their  long  childlessness,  of  the  mother's  appeal  for  a 
child  whether  from  God  or  the  Devil,  and  her  promise  to  give 
the  child  to  the  devil  if  it  should  be  born  through  his  aid.  The 
child  thus  born  comes  into  the  world  already  possessed  by  evil. 
He  has  extraordinary  strength  and  precocity,  and  from  the  first 
gives  evidence  in  his  violence  and  wickedness  of  his  diabolic 
origin.  After  falling  to  the  utmost  depths  of  human  depravity, 
he  is  roused  to  a  sense  of  the  horror  he  inspires,  forces  his 
mother  to  tell  the  story  of  his  birth,  and  then  begins  a  long  and 
arduous  penance. 

The  Wish-Child  or  Wonder-Child  folk-tales 2  seem  to  have 
influenced  the  beginning  of  the  legend.  They  tell  of  a  mar- 
vellous child  born  from  the  union  of  a  mortal  woman  with  an 
Otherworld  being,3  —  a  theme  of  immemorial  fairy  lore,  frankly 

1  Borinski,  Germania,  1892,  xxxvn,  60;  Zts.  fur  Volkerpsychologie,  1889, 
xix,  77;  Loseth,  p.  xxx. 

2  "  Grindkopf,"  Breul,  p.  115;  "  Teigneux,"  Loseth,  p.  xxx;  cf.  Crane, 
p.  53,  for  other  names  applied  to  the  type. 

3  For  Celtic  instances  of  the  Semi-Supernatural  Son,  see  T.  P.  Cross, 
"  The  Celtic  Origin  of  Lay  of  Yonec,"  Univ.  of  North  Carolina,  Stud,  in 
Phil.,  1913,  pp.  54  ff. ;  also  Revue  Celtique,  xxxi  (1910).  Ogle,  "Some 
Theories  of  Irish  Influence  and  the  Lay  of  Yonec,"  Rom.  Rev.,  x,  123-48 
(1919),  attacked  the  theory  of  Celtic  influence  and  cited  a  large  number  of 
stories  from  classical  mythology,  from  Oriental  and  Christian  sources  in  which 
this  theme  appears.  Of  special  interest  for  illustration  of  the  mediaeval  trans- 
formation of  a  god-like  father  into  a  devil  are  the  accounts  of  the  birth  of 
Merlin.  In  Layamon's  Brut  the  supernatural  father  appears  as  a  knight  in 
golden  armour;  in  the  prose  Merlin  as  a  devil  who  is  sent  to  ruin  an  inno- 
cent maiden.  Cf .  Toldo,  "  Leggenda  dell'  amore  che  trasforma,"  Zts.  f.  rom. 
Phil.,  1903,  pp.  279  ff.;  "Yonec,"  Rom.  Forsch.,  xvi,  609-29  (1903-4). 


GOWTHER    (ROBERT   THE    DEVIL)  53 

pagan  and  unmoral,  —  or  of  a  child  born  through  the  instru- 
mentality of  a  supernatural  being  to  whom  he  is  thereby  prom- 
ised. After  being  delivered  into  the  keeping  of  this  creature, 
a  demon,  wild  man,  or  sorcerer,  the  child  touches  some  forbidden 
object,  and  his  hair  turns  golden  (Panzer,  pp.  251-54).  The 
motif  of  the  "  Child  Vowed  to  the  Devil  "  appeared  in  ecclesiasti- 
cal guise  in  the  thirteenth  century,  but  in  Paul  Meyer's  4  opinion 
these  versions,  in  which  the  outwitting  of  the  Devil  is  accom- 
plished by  the  innocent  youth  himself  or  through  the  intercession 
of  the  Virgin,  have  nothing  to  do  with  the  Robert  legend.  In 
the  latter  it  is  a  question  not  of  physical  but  of  spiritual  cap- 
tivity, and  the  crux  of  the  tale  is  the  hero's  spiritual  redemp- 
tion. The  earliest  text  showing  an  adaptation  along  the  line 
which  the  Robert  legend  was  to  take,  of  the  initial  motif  of  the 
"  Child  Vowed  to  the  Devil  "  is  the  lmram  Hui  Corra  (ed.  by 
Stokes,  Revue  Celt,  xiv,  22  ff.),  an  Irish  tale  preserved  in  the 
fifteenth-century  Book  of  Fermoy,  but  possibly  as  old  as  the 
eleventh  century.  In  this  a  couple  rashly  promise  their  offspring 
to  the  Devil,  and  the  three  sons  who  are  born  begin,  after  the  dis- 
covery of  their  demoniac  origin,  a  wild  outlaw  life.  The  three 
are  subsequently  converted,  endure  arduous  penances,  and 
eventually  start  off  on  those  saintly  wanderings,  the  account  of 
which  is  the  true  purpose  of  the  story  (Crane). 

The  romance  of  Sir  Gowther  follows  closely  the  Robert  legend 
in  the  account  of  the  childless  parents,  the  prayer  of  the 
Duchess  of  Austria,  etc.  Here  only  is  found  the  scene  in  which 
the  Devil,  taking  the  form  of  the  lady's  husband,  as  did  Uther 
in  winning  the  mother  of  Arthur,  woos  the  Duchess  in  the 
orchard.  On  leaving  he  reveals  himself  and  prophesies  the 
demoniac  nature  of  his  son.  These  departures  have  been  traced 
by  some  scholars  to  the  influence  on  Sir  Gowther  of  the  Breton 
lays,  from  one  of  which  it  claims  descent  (Ravenel).  Similarly 
in  Tydorel  (ed.  Paris,  Rom.  vm,  66  ff.),  the  queen  is  wooed  in 
her  garden  by  a  splendid  Otherworld  stranger  who  at  parting 
foretells  the  birth  of  his  son.     Despite  its  thirteenth-century 

4  Meyer's  list  of  eight  versions  superseded  that  of  Breul.  The  folk- tales 
of  the  Wish-Child,  Robert,  and  the  legends  of  the  Child  Vowed  to  the  Devil, 
have  the  same  point  of  departure  in  the  vow  of  the  mother  to  give  her  child 
to  the  Devil.  An  extraordinary  modern  instance  of  this  theme  is  given  by 
Jane  Addams,  "The  Devil  Baby  at  Hull  House,"  Atlantic  Monthly,  1916. 


54  MEDIEVAL   ROMANCE    IN    ENGLAND 

polish,  Tydorel  represents  the  Wonder  Child  story  in  its  dis- 
tinctly pagan  form.  No  touch  of  moral  obloquy  is  laid  upon  the 
mother  for  her  liaison  with  the  Otherworld  lover;  no  evil  effect 
is  traced  in  the  nature  of  the  child.  If,  however,  this  tale  came 
under  the  grey  influence  of  ecclesiastical  thought,  the  splendid 
lover  might  readily  be  transformed,  as  he  is  in  Gowther,  into 
the  Devil  of  Christian  theology,  the  godlike  child  into  a  monster 
of  iniquity,  and  his  joy  and  pride  in  his  high  lineage  into 
loathing. 

But  this  view,  that  the  pagan  Celtic  story  was  thus  Christian- 
ized, has  been  sharply  attacked  by  Ogle  on  the  ground  that  the 
orchard  scene  in  both  Tydorel  and  Gowther  should  be  derived 
from  the  Apocryphal  legend  of  Anna,  wife  of  Joachim  and 
mother  of  the  Virgin  Mary.  This  legend  was  current  in  western 
Europe  from  the  fifth  century,  and  made  widely  known  the  type 
story  in  which  the  grief  of  a  childless  pair  leads  the  wife  to 
desperate  prayers  and  to  the  coming  to  her  in  an  orchard 
(pomerium)  of  an  Angelic  Visitor  who  promises  her  a  child  of 
grace.  Though  it  is  difficult  to  see  how  the  pious  author  of 
Gowther  could  ever  have  borrowed  this  story  directly,  as  Ogle 
(p.  43)  was  inclined  to  think  he  did,  since  such  borrowing  would 
involve  equating  the  Angel  with  the  Devil  and  the  Blessed  Virgin 
with  the  devil-born  boy,  it  is  nevertheless  certain  that  this  scene 
in  the  two  romances  is  more  closely  paralleled  by  that  in  the 
Anna  legend  than  in  any  other  yet  cited.  From  this  point  of 
view  Tydorel  must  be  regarded  simply  as  a  wholly  secularized 
fairy  tale5  and  Gowther  as  an  off-shoot  from  the  secular  type. 

The  central  portion  of  the  Robert-Gowther  story  tells  of  the 
hero's  adventures  after  his  conversion.  In  this  section  Breul 
recognized  as  the  most  fundamental  part  of  the  story  the  folk- 
tale of  the  Male  Cinderella.  Living  for  penance  as  the  humblest 
menial  or  fool  at  the  court  of  the  Emperor  (of  Rome,  Robert; 
of  Almayn,  Gowther),  the  hero  is  nevertheless  able  by  his  master- 
ful deeds  in  three  great  battles  to  save  the  realm  from  invading 
Saracens  or  from  the  angry,  disappointed  wooer  of  the  Em- 

5  For  an  instance  of  the  complete  secularization  of  early  Apocryphal 
legends  concerning  the  Virgin  Mary  see  C.  B.  Lewis,  PMLA.  xxxvn,  141-81 
(1922).  He  argued  that  from  these  legends  came  the  passionate  and  wholly 
secular  Old  French  lyrics  known  as  the  Weaving  Songs  (Chansons  de  Toile) 
and  the  Fountain  Songs. 


GOWTHER    (ROBERT   THE    DEVIL)  55 

peror's  daughter.  In  answer  to  his  prayer  for  divine  assistance 
he  is  provided  for  each  battle  with  horses  and  armor  of  dif- 
ferent color.  Thus  equipped,  he  rides  forth,  unrecognized,  and 
saves  the  day.  The  dumb  Princess,  whose  power  of  speech  is 
miraculously  restored,  subsequently  proves  the  identity  of  the 
fool  with  the  hero  and,  in  Gowther,  his  absolution  having  been 
pronounced,  marries  him.  The  episode  of  the  Three  Days' 
Battle  indicates  the  parent  group  of  folk-tales  for  this  part  of 
the  story  (Weston,  pp.  2 1  ff.) :  a  Wish-Child,  marked  by  his 
golden  hair,  or  some  similar  attribute,  takes  service  as  a  menial 
in  a  king's  palace.  He  is  loved  by  the  Princess,  who  alone  sees 
his  golden  hair,  and  ultimately  wins  her  hand,  in  a  contest  of 
her  suitors,  or  by  such  service  against  foreign  enemies  that  the 
Princess  is  given  in  reward,  or  by  rescuing  her  from  a  fairy 
castle  or  a  dragon's  den.  Magical,  strangely  colored  horses  are 
found  for  him  or  he  obtains  them  from  his  supernatural  parent 
or  captor,  or  from  Grateful  Animals.  In  the  romance  versions 
of  the  Three  Days'  Battle 6  the  contest  is  always  a  tournament 
or  battle,  and  the  three  disguises  are  provided  by  the  hero  him- 
self, or,  as  in  the  Robert -Gowther  legend,  by  an  Angel.  From 
fiction  this  picturesque  disguise  seems  to  have  passed  into  real 
life,  for  in  141 7-18  Malory's  patron,  Richard  Beauchamp,  Earl 
of  Warwick,  during  his  governorship  of  Calais,  under  such  names 
as  the  Chevalier  Vert,  for  three  days  challenged  French  knights 
to  a  tourney.7 

To  adapt  this  folk-tale  to  ecclesiastical  purposes,  it  was  only 
necessary  to  identify  the  menial  service  of  the  disguised  and 
heroic  youth  with  the  ardent  penance  of  a  contrite  sinner.  The 
special  nature  of  the  punishment  which  forced  a  man  to  give  up 
speech  and  apparent  understanding  and  to  live  with  animals,8 
in  short,  to  degrade  the  human  to  the  animal  state,  had  its  in- 

6  See  Ipomedon. 

7  Cf.  Maynadier,  Arthur  of  the  English  Poets,  1907,  p.  223,  for  comment 
on  this  episode  and  its  romance  parallels. 

8  Eating  with  dogs  is  the  penance  of  a  faithless  wife  in  the  Forty  Viziers, 
Behrnauer,  Die  Vierzig  Veziere,  p.  325  (tale  of  the  39th  Vizier)  ;  E.  Gibb,  Forty 
Vezirs,  p.  331  (tale  of  34th  Vizier).  For  other  references  establishing  the 
oriental  origin  of  the  motif,  and  indicating  its  wide  distribution,  see  Kit- 
tredge,  Arthur  and  Gorlagon,  pp.  251-53,  notes.  The  penances  of  St.  Alexius 
and  of  Valentine  in  Valentine  u.  Namelos  (W.  Seelmann,  1884)  were  of  almost 
equal  severity. 


56  MEDI/EVAL    ROMANCE    IN    ENGLAND 

ception  in  Oriental  ideas.  From  Buddhistic  practices,  from  the 
legend  of  the  Prodigal  Son  and  his  swine,  from  the  legend  of 
Nebuchadnezzar  9  eating  grass,  the  theme  passed  from  the  East 
to  the  West.  Here  one  of  its  earliest  appearances  in  vernacular 
form  must  have  been  the  lost  Carolingian  poem  on  the  proud 
king,  Guisbert,  a  version  of  which  is  preserved  in  the  Reali  de 
Francia,  v.  (Rotn.,  n,  355).  Breul  (p.  130)  noted  that  the  beast 
penance  appeared  in  Robert  of  Sicily,  in  the  legends  of  St.  Al- 
bano  and  of  St.  Giovanni  Boccardoro  (ed.  d'Ancona,  Bologna, 
1865;  G.  Paris,  Revue  Celtique,  in,  54),  and  in  the  Dit  des  III 
Chanoines  (ed.  Jubinal,  Nouv.  Recueil,  1,  266).  Appropriately, 
therefore,  it  found  its  way  into  the  legend  which  was  to  be  the 
preeminent  expression  of  monkish  meditation  over  the  problem 
of  extremest  sin  and  the  possibility  of  atonement. 


BIBLIOGRAPHY 

Texts:  Mss.  (i)  Advocates  Library,  Edin,  19,  3,  1;  (2)  Royal,  17, 
B,  in,  ed.  Utterson,  1,  157  ff.  (1817).  Both  MSS.  ed.  by  K.  Breul, 
Oppeln,  1886,  1895.    Rev.  Eng.  Stud.,  xn,  78-83  (1889);  Rom.  xv,  160. 

Studies:    Cf.  Breul,  pp.  198-207;  Wells,  Manual,  p.  784. 
Crane,  R.  S.     "An  Irish  Analogue  of  the  Legend  of  Robert  the  Devil." 

Rom.  Rev.  v,  55-67    (1914). 
Deister,  B.     Sprachliche   Untersuchung  des  abenteuerromans  Robert  le 

Diable.     Diss.  Gottingen,  1918. 
Kippenderg,  A.     "  Die  Sage  von  Robert  dem  Teufel  in  Deutschland  und 

ihre  Stellung  gegeniiber  der  Faustsage."     Stud,  zur  vergl.  Lit.  iv, 

308-33  (1904). 

Loseth,  E.  Robert  le  Diable,  Roman  d'Aventure.  SATF.  Paris,  1902. 
(Introduction,  i-xlviii;  Text,  1-198). 

Meyer,  P.  "  L'Enfant  Voue  au  Diable."  Romania,  xxxiii,  162-78 
(1904). 

Ogle,  M.  "  The  Orchard  Scene  in  Tydorel  and  Sir  Gowther."  Rom.  Rev. 
xin,  37-43  (1922). 

Panzer,  F.  Hilde-Gudrun:  eine  sagen-und  liter argeschichtliche  Unter- 
suchung. Halle,  1 901.   (Variants  of  the  Goldenermarchen,  pp.  251  ff.) 

9  For  fourteenth-century  English  versions  of  this  story,  see  Gower,  Con- 
fessio  Amantis  (ed.  G.  Macaulay,  Lond.,  1900),  bk.  1;  Chaucer,  Monk's 
Tale,  1.  153;  the  alliterative  poem,  Purity  or  Clannesse  (ed.  R.  Menner, 
New  Haven,  191 7),  1.  1676.  As  early  as  1100  the  Nebuchadnezzar  story  was 
carved  in  the  cloister  of  the  abbey  church  at  Moissac  (A.  Tilley,  Mediaeval 
France,  1922,  pp.  336,  393). 


GOWTHER    (ROBERT   THE    DEVIL)  57 

Ravenel,  F.  L.  "  Tydorel  and  Sir  Gowther,"  PMLA.  xx,  152-77  (1905). 

Tardel,  H.  (1)  Die  Sage  von  Robert  dem  Teufel  in  neueren  deutschen 
Dichtungen  u.  in  Meyerbeer s  Oper.  Berlin,  1900.  Rev.  Stud.  z. 
vergleich.  Lit.  11,  503  (1902).  (2)  "  Neuere  Bearbeitungen  der 
Sage,"  Stud,  zur  vergl.  Lit.  iv,  334-45  (1904). 

Thorns,  W.  The  Lyfe  of  Robert  the  Deuyll,  A  Romance  from  the  edi- 
tion by  Wynkyn  de  Worde,  reprinted,  Edin.,  1904;  Lond.,  1907. 

Weston,  J.     The  Three  Days'  Tournament,  Lond.,  1902. 


ROBERT    OF    CISYLE 

Versions.  The  Middle  English  version  of  Robert  of  Cisyle 
may  be  considered  either  a  romance  or  an  ecclesiastical  legend. 
In  the  many  forms  1  of  this  story  current  in  western  Europe 
the  hero  had  various  names  and  titles  but  these  are  negligible 
for  purposes  of  classification.  In  one  type  of  story,  most  notably 
represented  by  the  tale  of  the  Emperor  Jovianus  in  the  Gesta 
Romanorum  (c.  1300),  the  humiliation  of  the  proud  man  who 
thinks  his  power  supreme  is  accomplished  by  means  of  a  bath. 
While  on  a  hunting  expedition  he  becomes  over-heated  and 
plunges  into  a  stream.  Deprived  of  his  clothes  and  insignia  and 
impersonated  in  court  by  the  one  who  has  taken  them,  the 
ruler  is  unrecognized  and  suffers  in  utter  destitution  until  he  has 
learned  humility  and  wisdom.  The  story  appears  in  the  various 
texts  of  the  Latin  Gesta  (ed.  Oesterley,  1872,  p.  360)  and  in 
various  vernacular  translations.2  Closely  related  to  it,  though 
often  with  many  changes  of  detail,  are  two  Icelandic  versions, 
one  of  the  fourteenth  and  one  of  the  seventeenth  century,  and 
a  number  of  sixteenth-century  variants,  among  them  a  French 
moralite  printed  at  Lyons  in  1584,  a  song  and  comedy  by  Hans 
Sachs  (1549-56),  an  anonymous  Dutch  poem  probably  of  the 
fourteenth  century,  a  Hungarian  poem  by  Stephen  Poli  (1583), 
a  German  version  printed  by  Valentine  Schumann  (1559)  of 
Leipzig,  and  various  popular  versions  in  Italian  and  Bohemian. 
In  a  sixteenth-century  Spanish  play  by  Rodrigo  de  Herrera  the 
theme  of  the  Proud  King  Humiliated  is  interestingly  related  to 
the  Emperor  Frederick  II  and  the  story  is  localized  in  Sicily 
(Varnhagen,  1,  p.  38). 

1  Varnhagen's  Stammtajel  of  European  variants,  (2,  p.  161)  included 
forty-five  texts.  This  was  somewhat  amplified  by  Kummel.  Both  discussed 
composite  forms  of  the  two  main  types  of  the  story  but  only  with  reference 
to  late  modern  forms. 

2  Cf.  Dick,  Erlanger  Beit.  z.  engl.  Phil.  1890;  Herbert,  Cat.  of  Romances, 
in,  202  ff.,  214;  Graesse,  Gesta,  1905,  n.  262.  In  the  Anglo-Latin  Gesta  the 
tale  is  numbered  29  and  is  sometimes  called  Ponnius  in  Civitate. 

58 


ROBERT   OF    CISYLE  59 

As  the  type  name  "  Konig  im  Bade  "  is  derived  from  an  early- 
scene  in  the  first  version  of  the  story,  so  is  the  "  Magnificat " 
title  for  an  older  version,  though  this,  in  some  of  its  later  vari- 
ants, does  not  always  exclude  the  bath  scene.  In  this  version 
the  king  hears  at  church  the  solemn  verse,  "  Deposuit  potentes 
de  sede  et  exaltavit  humiles  "  (Luke,  1,  52),  and  proudly  denies 
its  application  to  himself.  He  falls  asleep,  his  place  is  taken  by 
an  Angel,  clad  in  his  garments,  and  the  king,  waking,  rushes 
forth  to  find  himself  mistaken  for  a  beggar  and  a  madman.  As 
in  the  first  group  of  tales,  after  his  bitter  lesson  has  been  learned, 
he  is  restored  to  his  former  state.  The  earliest  version  in  this 
group  is  a  Middle  High  German  poem  3  ascribed  to  Der  Strieker 
(1240),  which  may  have  been  derived  in  part  from  that  same 
German  prose  chronicle,  now  lost,  in  which  Herrand  von 
Wildonie  said  he  found  the  source  for  his  story  Der  Nackte 
Kaiser  (Corneus),  written  in  the  last  half  of  the  thirteenth 
century.  Early  in  the  fourteenth  century  Jean  de  Conde  com- 
posed a  poem  on  a  proud  king  of  "  Sezile,"  Li  dis  dou  Magnificat 
(Scheler,  Brussels,  1866,  11,  355  ff.).  Jean's  source,  it  is  now 
supposed,  was  the  same  "  boke "  as  that  which  inspired  the 
Middle  English  version.4  In  1335  Don  Juan  Manuel  inserted  a 
Spanish  prose  version  in  El  Conde  Lucanor  (ed.  Knust-Hirsch- 
feld,  1900,  Ex.  ll.).    About  1374  Giovanni  Sercambi  of  Lucca5 

3  Ed.  by  H.  von  der  Hagen,  Gesammtabendteuer,  1850,  Nr.  71.  For 
further  bibliographical  details  about  Der  Strieker  and  Herrand,  see  Edwardes, 
Summary  of  Literatures,  Index.  Cf.  Varnhagen,  1,  p.  53.  In  Herrand's  version 
the  king  is  forced  to  become  a  kitchen  servant  and  to  receive  the  abuse  of 
his  fellows. 

4  Varnhagen,  2,  p.  38,  expressed  disbelief  in  his  earlier  idea  that  Jean's 
poem  was  the  source  of  the  ME.  version.  The  basis  for  this  change  was  a 
consideration  of  the  fact  that  two  ancient  traits  appear  in  the  latter  that 
are  not  found  in  Jean's  poem,  namely,  that  the  king  has  the  Deposuit  trans- 
lated to  him,  and  that  the  king  is  mistreated  by  his  own  palace  porter.  On 
the  other  hand  there  seem  to  be  even  more  important  points  of  cor- 
respondence. The  question  needs  further  investigation,  especially  in  con- 
nection with  the  more  exact  determination  of  the  relationship  between  the 
German  and  French  versions  of  the  late  thirteenth  or  the  fourteenth-century 
and  the  Middle  English  romance. 

5  The  hero  of  Sercambi's  story  is  Ambrotto,  King  of  Navarra.  Not 
only  the  name  but  also  the  type  of  the  story,  which  belongs  to  the  Magnificat 
group  yet  introduces  the  bath  episode,  makes  impossible  any  relation  between 
it  and  the  ME.  poem,  interesting  as  it  would  be,  in  view  of  Young's 
suggestion,  Kittredge  Anniversary  Papers,  p.  405,  (ioi4)>  concerning 
Chaucer's  possible  indebtedness  to  Sercambi,  to  find  other  evidence  of 
knowledge  of  the  novelle  in  England. 


60  MEDIEVAL   ROMANCE    IN    ENGLAND 

included  the  story  in  his  novelle  and  at  least  twice  in  the  fif- 
teenth century  it  was  again  retold  by  Italian  writers,  one,  an 
anonymous  dramatist  (d'Ancona,  Sacre  Rappresentazione,  m, 
175)  and  the  other,  Antoninus,  Archbishop  of  Florence  (d.  1459), 
who  wrote  in  Latin  prose  (Varnhagen,  1,  pp.  45-59).  A  German 
play  on  the  subject  was  written  by  John  Romoldt  in  1563,  and 
a  Danish  play  by  Rudolph  Schmidt,  Den  Forvandlede  Konge, 
was  acted  in  1876  in  Copenhagen. 

In  England  the  legend  seems  to  have  had  special  popularity. 
Three  of  the  eight  manuscripts  which  contain  the  Middle 
English  poem  are  of  the  last  half  of  the  fourteenth  century,  i.e., 
Vernon  manuscript,  Additional  manuscript  22283,  and  Trinity 
College  manuscript;  the  others  are  all  of  the  fifteenth  century. 
It  is  probable  that  the  original  poem  was  not  composed  much 
before  1370,  the  date  of  the  Vernon  manuscript.  According  to 
Horstmann,  the  Vernon  and  Trinity  College  manuscripts  most 
nearly  represent  that  original;  in  Nuck's  opinion  the  Vernon 
manuscript  should  be  in  a  class  by  itself,  but  he  was  not  aware 
of  Additional  Manuscript  22283,  which  is  closely  related  to  the 
Vernon  manuscript,  though,  as  Ward  (Catalogue  1,  763)  pointed 
out,  not  a  copy  of  it.  Nuck  believed  the  original  poem  was 
written  by  an  unknown  poet  in  the  southern  part  of  the  East 
Midland  district  but  beyond  noting  a  few  stereotyped  expres- 
sions, he  made  no  effort  to  characterize  the  author's  style.  Few 
Middle  English  poets,  however,  tell  their  stories  with  more  fresh- 
ness and  even  poignancy  of  phrasing.  The  author  was  no  min- 
strel with  a  repertoire  of  stock  phrases  and  themes,  but  a  poet 
in  whom  the  best  of  monastic  influences  is  discernible.  Tender, 
devout,  wistfully  credulous  about  that  blessed  time  when  an 
Angel  ruled  upon  earth,  he  tells  the  story  with  moving  sweetness 
and  unusual  dramatic  power.  Though  he  uses  throughout  the 
short  riming  couplet,  he  falls  occasionally  into  the  use  of  refrain 
lines,  as  in  the  Angel's  question,  "  Where  is  now  <5i  dignite?  ",  or 
the  stricken  king's  piteous  prayer,  "  Lord,  on  di  fol  Jou  have 
pite,"  with  almost  liturgical  dignity  and  obvious  stanzaic  effect. 
He  stresses  churchly  seasons  and  Scriptural  legends.  It  is  on 
St.  John's  Night,  Midsummer  Eve,  that  the  Angel  comes;  on 
Holy  Thursday  the  Angel's  splendid  gifts  are  given  in  Rome. 
From  some  version  of  the  Book  of  Judith  the  poet  paraphrases 


ROBERT   OF    CISYLE  6l 

the  story  of  "  Sire  Olyferne  " ;  again  he  tells  of  that  "Nabgodon- 
osare  "  on  whose  shame  Roberd  meditates,  and  carefully  quotes 
and  translates  the  Latin  text  of  the  Deposuit. 

The  most  distinctive  influence  on  the  Middle  English  poem, 
however,  is  that  of  the  famous  conversion  legend  of  Robert  the 
Devil.  In  no  other  version  of  the  King  Deposed  story  does  the 
hero  bear  the  name  of  Robert  and  in  no  other  are  the  humilia- 
tions suffered  by  the  king  so  reminiscent  of  the  penance  of  the 
converted  "  Devil."  In  both  legends  the  hero  is  treated  as  a 
fool,  in  Roberd  he  wears  a  hateful  garment  "  With  foxes  tayles 
wyuen  aboute  ",  he  endures  buffets  and  jeers  from  those  who 
should  most  do  him  honor,  and  he  eats,  like  a  beast  himself,  with 
hounds.  On  Holy  Thursday  he  is  made  to  enter  Rome  where 
his  true  spiritual  penance  is  to  begin.  It  is  probable  that  the 
assimilation  of  the  two  legends  had  taken  place  in  the  thirteenth 
century,  but  in  no  version  is  there  a  deeper  perception  of  the 
tragic  irony  of  the  situation  than  in  this  Middle  English  Roberd 
of  Cisyle ;  nowhere  is  the  language  phrasing  it  more  adequate. 
One  does  not  lightly  forget  the  brief  stern  speech  of  the  Angel 
to  the  raving  king  nor  the  plaintiveness  of  the  poor  fool's  prayer. 

The  legend  was  known  in  England  not  only  through  these 
various  texts  of  the  poetic  narrative  but  also  through  versions  of 
the  Gesta  Romanorum  and  through  dramatic  representations. 
In  145 2-1453  a  Ludus  de  Robert  de  Cesill  was  acted  at  Lincoln 
and  in  the  time  of  Henry  VII  another  play  on  the  same  subject 
was  given  at  Chester.  Since  no  texts  of  these  plays  have  been 
preserved,  their  existence  is  known  only  from  contemporary 
allusions  and  town  records.0  In  the  nineteenth  century  another 
revival  of  interest  resulted  from  Leigh  Hunt's  A  Jar  of  Honey 
from  Mount  Hybla  (1848),  and  from  the  Sicilian's  Tale  in 
Longfellow's  Tales  of  a  Wayside  Inn  (1863),  a  version  destined 
to  give  new  life  in  America  to  the  old  story. 

Origin.  The  ultimate  origin  of  the  legend  of  Roberd  of  Sicily 
is,  according  to  Varnhagen  (1,  pp.  1-33),  to  be  found  in  ancient 
Hindoo  beliefs  in  metempsychosis.  He  and  Kohler  have  cited  a 
number  of  Indian  tales  which  turn  on  the  idea  of  the  trans- 

6  Cf.  E.  K.  Chambers,  The  Mediaeval  Stage,  Oxford,  1903,  n,  151,  356, 
378. 


62  MEDIAEVAL   ROMANCE    IN   ENGLAND 

ference  of  a  man's  soul  from  his  own  body  to  that  of  another 
man  or  to  that  of  some  bird  or  beast.  In  one  version  the  hero 
is  a  weary  old  king  longing  for  the  renewal  of  youth.  On  a 
hunting  expedition  a  wily  magician  persuades  him  to  enter  the 
body  of  a  dead  youth  and  immediately  enters,  himself,  into  the 
body  of  the  king  and  assumes  his  royal  power.  But  this  tale, 
despite  the  royal  rank  of  its  hero  and  the  hunting  scene,  is  much 
less  close  to  the  legend  of  the  King  Deposed  than  are  the 
various  Jewish  accounts  of  Solomon.  The  Babylonian  Talmud, 
for  instance,  relates  that  Ashmodei,  king  of  demons,  gained 
Solomon's  ring 7  and  place,  and  that  Solomon,  after  various 
sufferings,  regained  his  throne.8  Levi,  in  an  article  on  "  L'Orgueil 
de  Solomon  "  {Revue  des  Etudes  Juives,  xvii,  1888,  pp.  62-63), 
quoted  an  even  older  story  from  the  Talmud  of  Jerusalem  in 
which  it  is  set  forth  that  an  angel,  by  God's  command,  took  the 
form  of  Solomon,  and  that  the  king  was  then  driven  forth  to 
wander  from  place  to  place,  mocked  always  by  those  to  whom 
he  cried,  "  I,  the  Preacher,  was  king  over  Jerusalem  "  (Ecclesi- 
astes  1,  12).  The  type  traits  which  characterize  this  story,  the 
divine  purpose  of  humiliating  mortal  pride,  the  angelic  usurper, 
the  mockery  of  the  king,  are  those  which  define  the  European 
legend  of  the  King  Deposed.  Levi's  belief  that  the  latter  offers 
a  striking  example  of  direct  mediaeval  adaptation  through  the 
medium  of  converted  Jews  or  of  clerks  interested  in  Rabbinical 
lore,  seems  on  the  whole  more  probable  than  Varnhagen's  belief 
(2,  p.  47)  that  some  form  of  the  Solomon  story  passed  by 
way  of  Byzantium  into  South  Slavic  lands  and  was  transmitted 
through  translations  from  Bulgarian  or  Servian  texts  to  Europe. 

7  Cf.  Kohler's  long  note  on  the  story  of  the  lost  ring  recovered  from 
the  stomach  of  a  fish. 

8  For  summaries  of  the  Solomon  legend  see  The  Jewish  Encyclopcedia,  XI, 
443  (1905) ;  R.  Farber,  Konig  Solomon  in  der  Tradition,  Vienna,  1902. 
Kohler,  p.  208  ff.  gave  a  useful  resume  of  the  Solomon  story  from  both 
the  Jewish  and  the  Mohammedan  sources.  He  believed  the  story  of 
Schehabeddin  {Thousand  and  One  Nights,  Forty  Viziers)  was  closely  related 
to  other  Oriental  sources.  A  sultan  who  is  to  be  punished  for  his  sin  of 
doubt,  is  told  to  immerse  his  head  in  water  and  he  will  see  marvels.  He 
does  so  and  immediately  passes  into  a  kind  of  trance  in  which  he  thinks  him- 
self bereft  of  his  kingdom,  and  finds  himself  a  stranger  in  a  strange  land 
where  he  lives  many  years.  On  recovering  consciousness  he  finds  the  entire 
experience  has  occupied  but  a  moment.  This  is  a  curiously  rationalized 
version. 


ROBERT   OF    CISYLE  63 

In  their  various  stages  of  composition  the  Talmudic  legends  of 
Solomon  were  probably  influenced  by  the  ancient  Biblical  ex- 
ample of  the  humiliation  of  King  Nebuchadnezzar.  In  the 
European  development  of  the  King  Deposed  legend  the  essential 
kinship  of  the  two  stories  was  often  recognized.  In  an  obscure 
Viennese  version,9  an  anonymous  poem  of  the  thirteenth  or 
fourteenth  century  derived  from  the  poem  ascribed  to  Der 
Strieker,  the  king's  name  is  actually  given  as  Nabochodnoser, 
although  in  general  the  namelessness  of  the  king  is,  as  Varnhagen 
(2,  p.  32)  pointed  out,  a  characteristic  trait  of  the  western  ver- 
sions. In  the  Middle  English  texts  the  interpolated  summary, 
already  noted,  of  the  Nebuchadnezzar  story  is  distinctive.  As 
an  English  version  of  that  story,  it  is  of  special  interest  for 
comparison  with  other  almost  contemporary  versions  by  such 
poets  as  Gower  and  Chaucer  10  and  the  author  of  Purity.  Earlier 
than  this  the  Nebuchadnezzar  theme  had  certainly  affected  such 
legends  as  that  of  King  Guisbert X1  of  France  who,  having  boasted 
that  he  did  not  fear  God,  thereupon  became  a  leper  and  lived 
like  a  wild  beast  in  the  forest  until  he  repented  of  his  pride. 
It  had  also  influenced  the  legend  of  Robert  the  Devil.  From  the 
beast  penance  in  this  seems  to  have  come  the  specific  sug- 
gestion for  the  details  in  the  French  and  English  versions  of 
Roberd  of  Cisyle. 

BIBLIOGRAPHY 

Texts:  (i)  Caius  Coll.  Cbg.  174,  edit.  Horstmann,  Archiv  lxii,  421; 
(2)  Univ.  Lib.  Cbg.  Ff.  II,  38,  pr.  Halliwell,  Nugce  Poeticce,  Lond.  1844; 
Horstmann,  op.  cit.  p.  426;  Hazlitt,  Remains  of  Eng.  Pop.  Poetry,  1,  264, 
ff.  (1864);  (3)  Univ.  Lib.  Cbg.  ji,  iv,  9,  Horstmann,  Archiv,  lxii, 
417;  (4)  Trinity  Coll.  Oxf.  57,  variants  given  by  Horstmann  in  his 
edition;  for  date  see  Anglia  1,  287;  (5)  Vernon  MS.,  Bodl.  3938,  Oxf., 
edited  by  Horstmann,  Sammlung  ae.  Legenden,  1878,  p.  209  ff. ;  selec- 
tions pr.  Cook,  Reader,  pp.  168-73;  (6)  Harl.  1701,  Br.  Mus.,  pr.  in  part 
Ellis,  Specimens,  III,  143  ff.;  Halliwell-Ellis,  p.  474  ff.;  variants  pr. 
Horstmann,  Sammlung  ae.  Leg.   209-19;    (7)   Harl.   525,  pr.   Utterson, 

9  Cf.  J.  M.  Schottky,  Jahrbiicher  der  Literatur,  Wien,  1819,  v,  31.  The 
poem  has  240  verses. 

10  See  here,  Gowther,  n.  9.  Nuck,  note  to  lines  181-84,  suggested  that 
Chaucer  knew  the  ME.  Roberd. 

11  Cf.  Gaston  Paris,  Romania,  n,  355,  in  a  review  of  Rajna's  Reali  di 
Francia.    Varnhagen,  2,  p.  57,  did  not  discuss  the  legend  adequately. 


64  MEDIEVAL   ROMANCE    IN    ENGLAND 

1839;  Extracts  in  Warton,  Hist.  Eng.  Poetry,  1,  183  ff.  (1840);  described 
Ward,  Catalogue,  1,  765;  (8)  Additional  MS.  22283,  Br.  Mus.,  desc.  Ward, 
ibid.  A  critical  edition  of  the  poem  was  printed  by  R.  Nuck,  Berlin, 
1887.  Trans.  F.  J.  Darton,  A  Wonder  Book  of  Romance,  N.  Y.  1907. 
Cf.  Brown,  Register,  11,  no.  1711. 

Studies:    Cf.  Wells,  Manual,  pp.  162-163. 
Herbert,  J.  Catalogue  of  Romances,  III,  202,  214,  etc. 
Jacob,  G.     Xoros  Kardash,  pp.  104-08.     Berlin,  1906. 
Kohler,  R.  "  Der  nackte  Konig,"  Kleinere  Schriften,  11,  207-13   (1900). 
Kiimmell,  K.  Drei  italienischen  Pro  sale  genden.  Halle,  1906.     See  Konig 

im  Bade. 
Nuck,  R.  See  above. 
Varnhagen,  H.  (1).  Ein  indisches  Mdrchen  auf  seiner  Wander ung  durch 

die  asiatischen  u.  europaischen  Litteraturen.  123  pp.  Berlin,  1882; 

rev.  Eng.  Stud,  vi,  259-60. 

(2).     Longfellow's   Tales  of  a  Wayside  Inn  und  ihre  Quellen, 

Berlin,  1884.     See  Robert  of  Sicily,  pp.  16-80.    Reviews,  Eng.  Stud. 

viii,  324-27;  Anglia  vn,  Anz.  143-46. 


AMIS    AND    AMILOUN 

Versions.  The  numerous  versions  of  the  famous  story  of 
Amis  and  Amiloun  fall  into  two  groups  so  nearly  contemporary 
in  their  earliest  forms  that  it  is  difficult  to  determine  their 
relative  priority.  The  most  ancient  text  of  all  is  the  Epistola 
written  in  1090-1100  by  Raoul  le  Tourtier  (Radulfus  Tortarius), 
a  monk  of  Fleury-sur-loire,  who  celebrated  in  Latin  distichs 
various  famous  friends,  among  them  Ami  and  Amile  (Bedier, 
p.  175,  n.  2).  Raoul's  account  is  little  more  than  a  learned 
resume  of  a  story  which  he  says  was  even  then  widely  known. 

His  account  belongs  to  the  group  of  romantic  versions  x  repre- 
sented first  by  three  manuscripts,  the  earliest  about  1200,  of 
an  Anglo-Norman  version,  Amis  e  Amilun  (ed.  E.  Kolbing,2 
Amis,  pp.  lxxiii,  11 1-87),  written  in  short  riming  couplets; 
second  by  a  French  chanson  de  geste,  extant  in  a  single  thirteenth- 
century  Paris  manuscript  (ed.  C.  Hofmann,  Erlangen,  1882),  in 
which  the  heroes  are  attached  to  the  Charlemagne  cycle;  third 
by  an  inedited  and  much  longer  fifteenth-century  manuscript; 
fourth  by  a  fourteenth-century  drama,  Un  Miracle  de  Nostre 
Dame  d'Amis  et  Amile,  (ed.  G.  Paris  &  U.  Robert,  187 1,  iv)  ; 
and  fifth  by  the  Middle  English  romance.  This  last,  which  is 
closely  related  to  the  Anglo-Norman  form,  seems  to  have  origin- 
ated in  the  latter  half  of  the  thirteenth  century.  It  was  written 
down  in  209  twelve-line,  tail-rime  stanzas  in  the  dialect  of  the 
Northeast-Midland  district.  Four  manuscripts  attest  its  popu- 
larity.    Of  these  the  best  and  earliest  is  the  Auchinleck  text 

1  The  most  complete  bibliography  for  the  French  versions  is  to  be  found 
in  L.  Gautier,  Les  Epopees  Francaises,  1897,  vol.  v,  52-55.  Cf.  Hist.  Lift. 
xxn,  288;  P.  Schwieger,  Die  Sage  von  Amis  u.  Amiles,  Berlin,  1885;  C. 
Hofmann,  Amis  et  Amiles  u.  Jourdains  de  Blaivies,  Erlangen,  1882,  pp.  iv-vi, 
listed  twenty-three  versions. 

2  Besides  the  1884  edition  of  the  Middle  English,  Old  French,  Norse,  and 
Latin  texts,  Kolbing's  most  important  publications  on  the  Amis  story  were 
in  Paul  u.  Braune,  Beitr.  w,  271-314  (1877) ;  in  Engl.  Stud.  11,  205-310 
(1878-9).  In  this  he  argued  against  Ten  Brink's  theory  that  the  English 
poem  was  derived  from  the  chanson  de  geste. 

65 


66  MEDIEVAL   ROMANCE    IN   ENGLAND 

(1330-40),  which  was  independently  derived  from  the  common 
source  of  the  other  three  manuscripts  (Kolbing,  Amis,  pp.  ix- 
xiv).  Of  these  Egerton  2862  is  ascribed  to  the  fourteenth  cen- 
tury, Douce  326  to  the  fifteenth,  and  Harley  2386  to  the 
sixteenth. 

The  distinction  of  the  legendary  as  opposed  to  the  romantic 
versions  of  the  story,  is  that  the  heroes,  under  the  names  Amicus 
and  Amelius,  are  honored  as  martyrs.  The  longer  version  of 
this  group  includes  a  twelfth-century  Latin  prose  legend,  the 
Vita  Sanctorum  Amid  et  Amelii  (Kolbing,  Amis,  pp.  xcvii  ff.), 
and  its  old  French  prose  translation,  Li  Amitiez  de  Ami  et  Amile 
(ed.  L.  Moland,  C.  d'  Hericault,  Nouvelles  frangoises  en  prose 
du  /je  Siecle,  pp.  35-82,  1856) ;  a  mediocre  Latin  versifying  of 
the  prose  life  (Faral),  of  which  a  portion  was  edited  with  the  Vita 
by  Kolbing  (p.  cxi),  and  a  fourteenth-century  version  in  Welsh 
(ed.  Gaidoz,  Rev.  Celt.,  iv,  203-44,  1879).  A  shortened  form  of 
the  Latin  legend,  De  duobus  pueris,  Amico  et  Amelio,  used  by 
Vincent  of  Beauvais  in  the  Speculum  Hist  oriole,  lib.  xxiii,  162-66, 
169,  served  as  the  basis  for  later  versions.  Among  these  may  be 
mentioned  the  thirteenth-century  Dutch  version  of  the  Speculum 
by  Jacob  van  Maerlant  and  the  Norse  Amicus  Saga  (ed.  Kolbing, 
Ger mania,  1874,  xix,  184  ff.),  the  material  of  which  was  trans- 
mitted via  England  and  translated  in  Iceland  in  the  thirteenth 
century  (Leach,  Angevin  Britain,  p.  263).  The  story  appears  in 
the  series  of  religious  dialogues  known  as  Der  Seelen  Trost 
(Wackernagel-Stadler,  pp.  181-87),  written  in  German  prose  at 
the  end  of  the  fourteenth  or  beginning  of  the  fifteenth  century, 
a  book  of  which  there  were  soon  versions  in  High  German,  Dutch, 
Swedish,  and  Danish ;  in  a  German  poem  of  about  the  same  date 
by  Andreas  Kurzman,  a  monk  of  Neuberg;  and  in  an  inedited 
prose  version  in  the  Swabian  dialect  (Kolbing,  Amis,  p.  cxx). 

Curious  versions  of  the  story  of  Amis  et  Amile,  combined  with 
other  material,  are  found  in  the  short  Old  French  poem,  Le  Bit 
des  Trois  Pommes  (ed.  Ulrich),  in  which  the  three  apples  serve 
as  a  test  for  determining  a  true  friend,  and  in  the  UYstoire  des 
Sept  Sages  (ed.  G.  Paris,  SATF.  1876,  xxxm,  167-92),  in  the 
story  of  Alexander,  Prince  of  Egypt,  and  his  friend,  Prince  Louis 
of  France.  It  was  somewhat  closely  imitated  by  the  German 
poet,  Konrad  von  Wiirzburg,  in  the  thirteenth  century  in  a  long 


AMIS   AND   AMILOUN  67 

poem,  Englehart  und  Engletrue  3  (ed.  Haupt  and  Gereke,  Halle, 
1912),  which  makes  an  unexplained  change  of  names  and  tales, 
and  refers  (v.  211,  6492)  to  an  unidentified  Latin  source.  The 
thirteenth-century  chanson  de  geste,  Jourdains  de  Blaivies  (ed. 
Hofmann,  Erlangen,  1882),  links  itself  loosely  to  the  Amis  story, 
as  Jourdains  is  said  to  be  the  grandchild  of  Amis.  An  important 
free  adaptation  of  the  Amis  story  appeared  in  the  French  prose 
romance  of  Olivier  de  Castile  et  Artus  d'Algarbe,  printed  at 
Geneva  in  1482  and  1492.  In  this  the  Two  Friends,  Olivier  and 
Arthur,  are  stepbrothers,  and  Olivier  is  forced  to  flee  because  of 
the  incestuous  love  of  his  stepmother.  After  various  adventures 
he  weds  a  princess  and  has  by  her  two  children.  He  is  im- 
prisoned, and  his  stepbrother  becomes  aware  of  his  danger 
through  the  change  of  color  of  a  magic  phial.  Arthur  goes  in 
quest  of  his  friend,  is  mistaken  for  him  by  Olivier's  wife, 
achieves  Olivier's  release,  and  himself  falls  gravely  ill.  Through 
a  dream  Olivier  learns  that  the  blood  of  his  children  will  restore 
his  friend.  He  slays  them,  but  they  are  miraculously  restored  to 
life.  English  translations  of  this  text  were  printed  in  15 18  and 
1695  (Esdaile,  Eng.  Tales,  p.  104),  a  frequently  reprinted  Spanish 
translation  in  1499  (repr.  A.  M.  Huntington,  N.  Y.,  1902),  a  Ger- 
man translation  from  the  French  by  Ziely4  in  1521,  and  a 
popular  Italian  version  in  1552.  Gerould  (The  Grateful  Dead, 
1908,  pp.  92-94)  discussed  the  relation  of  the  Spanish  version  to 
Lope  de  Vega's  Don  Juan  de  Castro,  1623,  and  to  the  play  El 
Mejor  Amego  el  Muerto,  c.  1627,  attributed  in  part  to  Calderon 
and  based  on  Don  Juan.  The  Spanish  Romance  de  la  linda 
Melisenda  (Belissant)  in  which  Amiles  bears  the  name  Airuelos, 
has  also  been  recognized  as  a  version  of  the  Amis  legend  (Hof- 
mann, Amis,  p.  v,  n.). 

Origin.  The  earliest  extant  reference  to  the  legend  is  con- 
tained in  the  poem  by  Raoul  de  Tortaire  to  which  reference  has 
already  been  made.    Raoul  not  only  asserted  that  the  legend  was 

3  For  studies  on  the  work  of  Konrad  von  Wiirzburg,  see  G.  Janson, 
Studien  iiber  die  legendendichtungen  Konrads,  Marburg,  1902 ;  Schroeder, 
Studien  zu  Konrad,  Gottingen,  1906;  H.  Landan,  Die  Chronologie  der 
Werke  des  Konrads,  Gottingen,  1906. 

4  Ziely's  work  was  translated  into  English  by  Leighton  and  Barrett,  The 
History  of  Oliver  and  Arthur,  1903. 


68  MEDIEVAL   ROMANCE    IN   ENGLAND 

widely  known,  but  he  identified  the  heroes  with  local  personages 
in  Auvergne  and  Gascony.  In  Bedier's  opinion  (n,  179)  and 
that  of  Huet  (p.  163)  both  Raoul's  poem  and  the  Latin  Vita 
Sanctorum  Amici  et  Amelii  were  derived  from  a  French  chanson 
de  geste  of  a  character  more  archaic  than  the  version  now  known. 
This  literary  version,  the  work  probably  of  a  French  jongleur, 
originated,  it  would  seem,  in  the  little  Lombard  city  of  Mortara. 
Here  traditionally  was  the  scene  of  the  great  battle  between 
Charlemagne  and  the  Lombard  king  Didier,  in  which  Ami  and 
Amile  were  killed.  Here  Charlemagne  gave  them  noble  burial 
and  here  the  miracle  of  the  tombs  took  place  when  the  sar- 
cophagus of  Amile  mysteriously  rejoined  that  of  Ami  from  which 
it  had  been  separated.5  This  was  a  marvel  of  course  even  to 
Raoul.  For  centuries  at  Mortara  the  tombs  of  the  two  heroes, 
canonized  as  warrior  saints,  were  shown  in  the  church  of  St. 
Albin's  to  the  pious  pilgrims  who  were  travelling  the  Via 
Francesca  on  their  way  to  Rome.  The  churchly  legend  which 
developed  about  the  two  friends  emphasized,  according  to 
Bedier's  theory,  the  pietistic  theme,  "  Omnem  filium  quern  Deus 
recepit,  corripit,  fiagellat  et  castigat,"  especially  in  regard  to  the 
humiliation  of  Amile.  But  the  jongleur  who  took  up  this  story 
and  who  was  evidently  familiar  not  only  with  places  along  the 
pilgrim  route  but  with  France  and  its  epic  traditions,0  secular- 
ized it  straightway.  In  his  hands  in  the  last  years  of  the 
eleventh  century  the  legend  became  essentially  feudal  and  mili- 
tant and  its  fame  was  spread  by  jongleurs  and  pilgrims  alike. 

But  if  the  mediaeval  secularization  of  the  legend  can  thus  be 
accounted  for,  there  yet  remains  a  nucleus  of  very  different 
origin.     In  the  romance  the  two   friends 7  are  said  to  be  so 

5  The  tombs  at  Mortara  are  thus  referred  to  not  only  in  the  Vita  but 
also  by  Godfrey  of  Viterbo  (died  1190).  Cf.  Bedier,  pp.  173-74.  The  fullest 
account  is  in  the  Chevaler*"  Ogier  de  Danemarche  (ed.  Barrois,  1842,  w. 
5847  ff.).  In  this,  Amis  and  Amile,  riding  unarmed  on  their  way  to  join 
Charlemagne's  army  after  a  pilgrimage  to  Rome,  are  brutally  killed  by  Ogier. 
Bedier  (p.  186  ff.)  showed  the  probable  influence  of  the  eleventh  century 
Vita  Hadriani  on  the  authors  of  the  Vita  Amici,  on  the  original  Ogier,  and  on 
the  Italian  chronicles  which  preserve  the  Carolingian  legends. 

6  Cf.  Korner's  study  of  place  names  in  Amis.  The  actual  Carolingian 
references  in  the  Vita  are  supplemented  in  the  chanson  de  geste  by  such 
indirect  allusions  as  that  to  the  traitor  Harde,  who  is  represented  as  one  of 
Ganelon's  kin. 

7  R.    Mertz,    Die    deut.   Bruchstiicke   von    Athis   u.    Prophilias    in   ihrem 


AMIS   AND    AMILOUN  69 

remarkably  alike  that  the  wife  of  one  of  them  mistakes  the 
other  for  her  husband.  They  are  depicted  as  men  of  equally 
noble  rank,  named  somewhat  similarly  and  perhaps  allegorically, 
and  are  said  to  have  been  born  on  the  same  day,  to  have  been 
brought  up  in  the  closest  intimacy.  All  this  seems  referable  to 
the  type  of  story  known  as  the  Two  Friends  (Bolte  and  Polivka, 
Hausniarchen,  1,  528)  in  which  two  brothers  who  exactly  re- 
semble each  other  experience  various  vicissitudes.  One  falls 
victim  to  enchantment  and  the  other,  going  in  quest  of  him,  is 
mistaken  by  his  brother's  wife  for  her  husband.  His  use  of  the 
Sword  of  Chastity  is  in  this  instance  paralleled  by  the  action  of 
Ami  who,  likewise  acting  as  the  husband  of  his  friend's  wife, 
places  at  night  his  unsheathed  sword  between  himself  and  the 
lady.8  The  combination  of  the  "  Sosie  "  motif  with  that  of  the 
Sword  of  Chastity  in  both  the  folk-tale  and  the  romance,  makes 


Verhaltnis  zum  aljtrz.  Roman,  Leipzig,  1904,  p.  1,  thus  classified  the  two 
principal  groups  of  mediaeval  Friendship  Romances:  (1)  Athis  and  Prophilias, 
Titus  and  Gisippus  (Boccaccio,  Decameron,  10,  8),  Alexander  (Alcander) 
and  Septimus;  (2)  Amis  and  Amile,  Engelhart  and  Dietrich,  Alexander  and 
Louis  {Sept  Sages),  Olivier  and  Artus.  For  Athis  see  A.  Hilka's  edition, 
Dresden,  1912-16;  L.  Stael  von  Holstein,  Le  Roman  d' Athis,  Etude  litt.  sur 
ses  deux  versions,  Upsala,  1909;  Liese,  Vergleich  der  Erzahlung  x,  8.  bet 
Boccaccio  mit  Athis,  Progr.  d.  stadt.  Realschule  zu  Gorlitz,  1901 ;  Hist.  litt. 
xv,  179-93;  W.  Grimm,  Kleinere  Schriften,  ni,  212-366  (Berlin,  1883)  ;  Ward, 
Catalogue  of  Romances,  1,  173.  On  p.  929  ff.  Ward  pointed  out,  as  others  had 
done  before  him,  the  source  story  of  Athis  in  the  Disciplina  Clericalis  (ed.  A. 
Hilka,  1912;  cf.  Ward,  Catalogue,  11,  247).  From  this  it  was  copied  into 
various  exempla  collections  such  as  the  Gesta  Romanorum,  the  Alphabet  of 
Tales,  etc.  For  a  long  bibliography  for  these  and  other  analogues  to  Boc- 
caccio's story  see  A.  C.  Lee,  The  Decameron,  Its  Sources  and  Analogues,  pp. 
330-43  (Lond.  1909)  ;  F.  L.  Jones,  Boccaccio  and  his  Imitators,  p.  39 
(Chicago,  1910) ;  G.  Groeber,  Ueber  die  Quellen  von  Boccaccio's  Dekameron, 
pp.  85-87  (Strassburg,  1913). 

8  B.  Heller,  "  L'Epee  Symbole  et  Gardienne  de  Chastete,"  Rom.  xxxvi, 
36-49  (1908)  ;  also  K.  Campbell,  Seven  Sages,  p.  cxii,  Heller  (p.  37)  be- 
lieved that  it  was  by  way  of  the  Vita  Amid  as  adapted  into  the  Roman  des  Sept 
Sages,  that  the  sword  motif  passed  into  the  literature  and  popular  traditions  of 
Europe.  Cf.  the  Sigurd-Brynhild  story,  the  Edda,  the  Volsung  Saga,  Tristan 
et  Iseult,  Bovon  de  Haumtone,  Le  Roman  de  la  Poire  (ed.  Stehlich,  1881) 
etc.,  and  the  religio- romantic  legend,  Le  Prevot  d'Aquilee  (ed.  Kohler,  Kleinere 
Schriften,  I).  For  romance  references  see  G.  Schoepperle,  Tristan,  Index, 
Separating  Sword.  Ogle,  Rom.  Rev.  x,  131,  n.  6  (1919),  noted  that  the  situa- 
tion in  the  Mabinogi  of  Pwyll  was  similar  to  that  in  Amis  and  "  was  no  doubt 
derived  from  the  same  source."  Pwyll  appears  in  the  guise  of  Arawn  but  lives 
in  perfect  chastity  with  the  wife  of  his  friend.  In  this  version  there  is  no 
reference  to  the  sword. 


70  MEDIAEVAL   ROMANCE    IN   ENGLAND 

certain  some  relationship  between  them.  Since  the  resemblance 
motif  in  the  folk-tale  is  clearly  and  regularly  explained  by  the 
fraternal  relationship,  the  folk-tale  version  is  probably  the  earlier 
form  (Huet,  p.  168). 

Even  more  important,  since  it  concerns  the  major  portion  of 
the  story,  is  the  evidence  of  the  popular  origin  of  the  central 
theme  of  the  story.  It  concerns  the  testing  of  the  loyalty  of  two 
men  who  are  called  upon  to  make  supreme  sacrifices  for  each 
other,  and  corresponds  closely  with  the  well-known  marchen 
known  as  Faithful  John  or  the  Faithful  Companion  (Bolte  and 
Polivka,  Hausmdrchen,  i,  42).  In  this  a  servant  secures  a  bride 
for  his  royal  master,  saves  him  from  various  dangers  though  at 
the  sacrifice  of  being  himself  turned  to  stone,  and  recovers  life 
through  being  bathed  with  the  blood  of  his  master's  child.  The 
folk-tales  do  not  explain  the  doom  visited  upon  the  faithful 
servant;  but  Potter  (p.  484),  on  the  evidence  of  a  song  from 
southern  Siberia,  believed  it  probably  due  to  the  folk  idea  of  the 
danger  in  breaking  a  taboo  of  silence.  The  servant  learns  in  a 
vision  or  through  the  speech  of  birds  or  animals  of  the  perils 
awaiting  his  master  and  suffers  petrifaction  when  he  betrays  this 
supernatural  knowledge.  In  the  literary  versions  of  Amis  in 
which  essentially  this  same  situation  is  reproduced,  leprosy9 
takes  the  place  of  petrifaction,  and  the  variety,  as  well  as  the 
unsatisfactory  nature  of  the  reasons  given  for  this  affliction, 
seem  to  indicate  that  it  was  an  old  traditional  element  which  was 
not  readily  understood  by  the  western  romancers  and  so  was 
interpreted  in  various  ways.  In  any  case,  as  Huet  (p.  180) 
pointed  out,  the  great  antiquity  and  the  consistency  of  the 
Eastern  versions  in  regard  to  the  petrifaction  motif,  make  it 
impossible  to  believe  that  in  this  episode  the  East  borrowed  from 
the  West.  Moreover  since  the  antiquity  of  the  Amis  story  makes 
evident  the  circulation  in  France  of  the  folk-tale  at  a  time  long 

9  In  Raoul's  poem  the  affliction  is  accidental;  in  the  Vita  Amici  it  is  a 
test  of  piety;  in  the  chanson  de  geste  and  the  Miracle  it  is  because  Amis, 
wedding  the  king's  daughter  under  a  false  name,  commits  the  crime  of  bigamy ; 
in  the  English  version  it  is  because  he  enters  falsely  into  a  judicial  combat. 
Cf.  Bedier,  p.  180.  The  leprosy  motif,  as  it  appears  in  Hartmann  von  Aue's 
Der  arme  Heinrich  and  analogous  tales,  is  fully  studied  by  Wackernagel  and 
Stadler,  pp.  217-23.  For  Hartmann  see  F.  Piquet,  Etude  sur  Hartmann  von. 
Aue,  Paris,  1898 ;  E.  Gierach,  Der  arme  Heinrich,  Uberlieferung  u.  Herstellung, 
Heidelberg,  19 13. 


AMIS   AND   AMILOUN  7I 

antedating  Hartmann  von  Aue's  Der  arme  Heinrich  (c.  1200), 
the  earliest  known  German  analogue  to  the  Amis  story,  Vore- 
tzsche's  idea  (p.  246)  of  the  folk-tale  as  originally  German  can 
hardly  be  accepted. 

In  the  chivalric  transformation  of  the  Amis  legend  in  France, 
numerous  romantic  characters  and  incidents  were  woven  into  the 
secularized  legend.  In  the  romance  a  Jealous  Seneschal  betrays 
the  love  of  Amis  and  the  king's  daughter.  The  lady  herself  is 
of  the  type  known  as  the  Wooing  Princess  or  the  Forth-Putting 
Lady.  A  Judicial  Combat  takes  the  place  of  the  dangers  which 
threatened  the  Faithful  Servitor,  and  the  lover  of  the  princess 
gallantly  though  falsely  undertakes  to  prove  her  innocent  of  a 
liaison  with  himself.  A  prophetic  dream,  in  which  Amiloun  sees 
his  friend  beset  by  wild  beasts,  warns  him  of  the  latter's  trouble. 
A  voice  from  Heaven  tells  him  of  the  danger  of  taking  his 
friend's  place  in  the  combat.  Golden  cups  which  the  two  friends 
exchange  at  their  first  parting  serve  later  as  Recognition  Tokens 
when  Amiloun,  hideously  disguised  by  leprosy  and  poverty, 
comes  to  his  friend's  castle.  The  characters,  such  as  Libias,  the 
Cruel  Wife  of  Amiloun,10  are  developed  with  power  and  almost 
sinister  realism,  and  mediaeval  customs  and  conditions  are  pic- 
tured in  colorful  ways.  The  most  graciously  romantic  quality  in 
the  story  comes  from  its  emphasis  on  friendship  as  an  ideal 
human  relationship,  and  it  is  for  this  quality  that  the  tale  is 
remembered  longest  as  a  "  changoun  d'amur,  de  leaute,  et  de 
grant  dougur." 

BIBLIOGRAPHY 

Texts:  (i)  A,  Auchinleck  MS.,  Advocates  Libr.  Edin.,  ed.  Weber,  11, 
360-473  (1810);  E.  Kolbing,  Altengl.  Bibl,  Heilbronn,  1884  (a  critical 
edition  of  this  and  other  MSS.) ;  cf.  Eng.  Stud,  rx,  175,  456,  477; 
xiii,  134;  (2)  S,  Duke  of  Sutherland's  MS.,  desc.  Eng.  Stud,  vn,  191; 
now  Egerton  2862,  desc.  Brit.  Mus.  Cat.  of  Add.  MSS.,  p.  239  (1905- 
10);  (3)  D,  Douce  326;  (4)  H,  Harley  2186,  desc.  Ward,  Catalogue, 
I,  677.  Trans.  E.  Rickert,  Romances  of  Friendship;  F.  J.  Darton, 
A  Wonder  Book  of  Old  Romance,  N.  Y.  1907. 

Studies  and  Analogues:  Cf.  n.  1;  Edwardes,  Summary,  pp.  74,  176, 
224;  Wells,  Manual,  p.  787.    Trans,  from  the  French:  Amis  and  Amile, 

10  Cf.  Comfort,  "  Character  Types  in  the  Old  French  Chansons  de  Geste," 
PMLA.  xxi,  p.  380. 


72  MEDIEVAL   ROMANCE    IN    ENGLAND 

adapted  from  the  chanson  de  geste  and  retold  in  modern  French  by 
J.  G.  Frazer,  Lond.  1903;  in  modern  English  by  William  Morris, 
Portland,  Maine,  1909,  reprinted  from  ed.  of  1894;  in  modern  German 
verse,  by  H.  Grein,  Kiel,  1902;  W.  Pater,  The  Renaissance,  Lond.  1898, 
1902,  gives  a  partial  translation  of  the  thirteenth-rentury  prose  version 
in  the  Bibliotheque  Elzevirienne ;  trans,  from  the  text  of  Moland  by 
E.  Mason,  Aucassin,  Everyman  Series,  1910,  191 2,  pp.  173,  ff. 
Ayres,  H.   M.   "  The  Faerie  Queen  and  Amis  and  Amiloun."     MLN. 

xxiii,  177-80  (1908). 
Bedier,  J.  Les  Legendes  Epiques,  u,  170-86.  Paris,  1908. 
Brechtefeldt,  W.  Der  Bau  des  Nomens,  u.  Verbums  in  den  Chanson  de 

Geste  "  Amis  et  Amile  "  und  "  Jourdains  de  Blaivies,"  Ein  Beitrag 

zur  altfrz.     Dialectkunde.     Diss.   175  pp.  Kiel,  1904. 
Faral,  E.  "  Le  MS.  Latin  3718  de  la  Bibliotheque  Nat.  Paris."  Rom. 

xvli,  244-46  (1920). 
Haupt,  M.  Engelhard  und  Engeltrut.  Leipzig,  1844,  1900;  Altdeut.  Text 

Bibl.  xvii.  Halle.  191 2. 
Huet,  G.    "  Amis  et  Amile,  Les  Origines  de  la  Legende."  Le  Moyen  Age, 

xxx,  162-84  (1919). 
Kolbing,  E.  See  Texts. 
Korner,  K.  "  Ueber  Die  Ortsangaben  in  Amis  et  Amiles"  Zts.  frz.  Spr. 

u.  Lit.  xxxm,   195-205    (1908). 
Potter,  M.  A.  "Ami  et  Amile."  PMLA.  xxiii,  471-85  (1908). 
Ulrich,  J.  "Le  Dit  des  Trois  Pommes."  Rom.  Forsch.  xix,  622-32  (1906). 
Voretzsch,  Einfuhrung,  p.  244-47,  I9I3>  2d  ed. 
Wackernagel,  W.  u.  Stadler.  Der  arme  Heinrich  u.  zwei  jungere  Prosale- 

genden  verwandten  Inhaltes,  S.  Silvester,  u.  Amicus  u.  Amelius  aus 

Der  Seelen  Trost  {Amicus,  pp.  181-87;  226-32)  Basle,  191 1. 


AMADAS 

Versions.  There  are  few  Middle  English  romances  for  which 
an  immediate  French  source  is  neither  known  nor  very  positively 
conjectured,  but  of  these  Amadas  is  one.  The  hero  bears  the 
name  of  the  lover  in  the  romance  of  Amadas  et  Idoine,1  but 
his  story  is  of  an  altogether  different  type.  It  is  found  in  a 
fifteenth-century  manuscript  in  the  Advocates  Library,  Edin- 
burgh, and  in  the  Ireland  manuscript  (Hale,  Lancashire)  in  which 
various  local  records  were  commenced  in  1413  by  William  Ire- 
land. The  two  texts  are  written  in  the  twelve-line,  tail-rime 
stanza  in  the  dialect  of  the  North-Western  part  of  England 
(Stephens,  p.  5),  possibly  in  that  of  Lancashire  (Robinson,  pp.  x, 
xlii).  They  seem  to  have  been  independently  derived  from  a 
common  source. 

The  story  of  Amadas  is  told  briefly  and  vigorously  in  less 
than  a  thousand  lines  and  concerns  chiefly  the  service  which 
Amadas  renders  to  a  dead  man  and  the  Grateful  Dead's  subse- 
quent actions.  Didactic  as  it  is  in  theme,  the  story  is  enlivened 
by  touches  of  blunt  and  almost  humorous  realism.  The  poor 
wife,  who  has  sat  for  sixteen  weeks  beside  her  husband's  corpse, 
says  frankly  that  he  "  wroght  more  lyk  a  fole  "  in  giving  his  goods 
to  all  who  asked.  The  merchant  who  will  not  let  the  body  be 
buried  until  his  debt  is  paid,  like  any  canny  Scot,  wishes  a  sorry 
grace  to  all  such  "  wastars,"  and  grieves  for  the  thirty  pounds  of 
which  he  will  not  see  a  penny  more.  The  ruined  Sir  Amadas, 
who  gives  his  last  cent  to  secure  burial  for  the  dead  man,  sits  at 
last  alone  in  the  forest,  lamenting  his  own  old  follies  with  re- 
freshing honesty.  Pure  romance,  however,  finds  expression  in 
such  scenes  as  that  in  the  woodland  chapel  with  its  bier  and 
burning  candles  where  Amadas  first  finds  the  dead  man  and  his 
grieving  lady,  in  the  appearance  of  the  ghost  as  a  White  Knight 

1  Cf.  G.  Paris,  Furnivall  Miscellany,  Lond.  1901.  Though  no  Middle 
English  version  exists,  the  number  of  allusions  to  this  romance  shows  how 
widely  it  was  known  in  England.  The  French  text  of  Amadas  et  Idoine 
(ed.  C.  Hippeau.  Paris,  1863)  is  summarized  in  the  Hist.  Litt.  xxn,  758  and 
is  the  subject  of  an  unpublished  Harvard  dissertation  by  J.  Reinhard. 

73 


74  MEDIAEVAL   ROMANCE    IN   ENGLAND 

who  comes  to  the  rescue  of  the  hero,  and  in  the  latter's  discovery 
of  a  shore  where  from  a  great  wreck  the  waves  have  washed  a 
chest  of  gold,  dead  knights  in  armor,  rich  robes,  and  live  horses, 
brown,  white,  and  grey.  Amadas,  when  he  comes  to  a  king's 
castle,  is  taken  by  his  white  hand  and  given  royal  welcome.  In 
true  romantic  fashion  he  wins  in  the  course  of  time  lands,  wealth, 
and  a  princess  for  his  wife. 

In  effect,  it  must  be  admitted,  the  story  is  less  a  romance 
than  a  moral  tale.  Its  moral  purport,  that  kindness  even  to  the 
dead  does  not  go  unrewarded,  is  reenforced  by  a  simple  piety  of 
spirit,  sure  that  "  Goddes  help  his  ay  nere."  Pious  ejaculations 
and  allusions,  especially  to  the  Rood,  are  many.  The  great 
scene  of  the  romance,  in  which  the  White  Knight,  the  ghost  of 
the  buried  debtor,  comes  to  demand  the  promised  half  of 
Amadas's  new  possessions,  is  strongly  suggestive  of  the  Abraham 
and  Isaac  story.  The  White  Knight  will  not  take  the  "  londes 
wyde,"  the  towered  castles,  the  wood  and  waters,  the  wild  deer, 
the  forests,  jewels,  silver,  or  red  gold,  but  only  half  of  that  which 
is  dearest,  the  wife  and  child  of  Amadas.  The  lady's  gentle 
resignation  as  she  lies  meekly  down,  covering  her  eyes  with  her 
handkerchief  and  taking  her  babe  into  her  arms,  Amadas's  grief- 
stricken  obedience  as  he  raises  his  sword  to  smite  them  both,  are 
surely  reminiscent  of  the  sacrifice  of  Isaac  in  mediaeval  drama- 
tizations. The  test  accomplished,  the  White  Knight,  after  a 
moral  peroration,  "  glod  away  as  dew  in  son." 

The  oldest  extant  version  of  the  theme  of  the  Grateful  Dead 
is  Cicero's  tale  of  Simonides  (De  Divinatione,  i,  c.  27),  who 
finds  a  corpse  on  the  seashore,  buries  it,  and  is  later  saved  from 
drowning  by  the  dead  man's  ghost.  Gerould  (p.  26),  who  has 
made  the  most  exhaustive  study  of  the  theme,  suggested  that 
this  was  an  independent  anecdote  which  passed  down  to  the 
Middle  Ages  through  Valerius  Maximus  (Facta  et  Dicta,  1,  7), 
Robert  Holkot  (Super  Libros  Sapientie,  Lect.  103),  and  Chaucer 
(Nun's  Priest's  Tale).  Another  early  version  appeared  in  the 
Apocryphal  Book  of  Tobit,  c.  76.  In  this  Tobit  buries  a  dead 
man  at  night,  is  imprisoned,  and  sends  his  son  Tobias  to  seek 
help;  in  disguise  the  angel  Raphael  aids  the  boy  in  gaining  a 
bride  and  delivering  her  from  the  demon  by  which  she  is 
possessed;  and  finally  the  angel  and  Tobias  return  and  release 


AMADAS  75 

Tobit.  The  simple  treatment  of  the  theme  was  here  greatly 
complicated  by  the  addition  of  other  motifs,  bat  as  "  perhaps 
the  best  loved  story  of  the  Apocrypha,"  it  had  undoubtedly  great 
influence. 

Most  nearly  relate-  to  Sir  Amadas  is  the  group  of  six  stories 
represented  by  the  foui  teenth-centtiry  German  poem  Ritiertriuwe 
(von  der  Hagen,  Gesamt  i    '850,  1,  105  ff.)  and  by  the 

Old  French  romances,  the  thirteenth-century  Richars  li  Biaus 
(W.  Foerster,  Vienna,  1874)  and  the  curious  fourteenth-century 
Lion  de  Bourgcs,2  which  was  translated  into  German  prose3  in 
the  fifteenth  century  and  printed  in  1514  (Hippe,  p.  154).  A 
literary  source  in  French  must,  be  postulated  for  these  two  ro- 
mances and  presumably  alst  the  story  of  Duke  Pippin  of 
Lorraine,  told  in  the  Old  Swedish  Legendariurn,  1256-70 
(Stephens,  p.  73),  for  the  fourteenth-century  Novella  di  Messer 
Dianese  (d'Ancona,  Rom.  in,  191),  and  for  the  English  Sir 
Amadas.  In  all  six  stories  a  knight  starts  for  a  tourney;  in  all 
but  the  Old  Swedish  he  is  a  spendthrift  who  pays  for  the  burial 
of  a  dead  man  at  the  cost  of  practically  all  he  possesses;  in 
Richars,  Lion,  and  Amadas,  he  is  later  provided  for  by  the  ghost 
in  the  form  of  a  White  Knight;  and  in  all  but  Richars  he 
promises  to  share  his  winnings.  In  Richars,  in  the  Old  Swedish 
Legendariurn,  and  in  Amadas,  the  ghost  demands  half  the  lady, 
and  in  Lion  and  in  Dianese  either  the  lady  or  the  property.  As 
Gerould  (p.  33)  pointed  out,  the  distinctive  trait  of  this  group  of 
versions  is  the  combination  of  the  Grateful  Dead  theme  with 
that  of  the  Spendthrift  Knight. 

The  Grateful  Dead  theme  was  used  somewhat  incidentally  in 
the  long  thirteenth-century  romance  of  Walewein  (Jonckbloet, 

2  See  H.  Wilhelmi,  Studien  uber  die  Chanson  de  Lion  de  Bourges, 
Marburg,  1894;  B.  Scholvien,  Weitere  Studien  z.  Chanson  de  Lion  de  Bourges 
Teil  1,  Diss.  Greifswald,  1905;  R.  Krickmeyer,  Weitere  Studien  z.  Chanson 
de  Lion  de  B.,  Teil  1,  Greifswald,  1905;  E.  Hudepohl,  Weitere  Studien  z.  Chan- 
son de  Lion  de  B.;  analyse  des  schlussteiles,  text  der  Joieuse-Tristouce-episode 
(sage  vom  Madchen  mit  der  abgehauenen  hand),  Diss.  Greifswald,  1906;  E. 
Stein,  Sprache  u.  heimat  der  jiingeren  fassung  der  Chans,  de  Lion  de  B.  (Hs.B), 
Diss.  Greifswald,  1908;  H.  Zeddies,  Weitere  Studien  z.  Chans,  de  Lion  de 
B.  Diss.  Greifswald,  1907;  W.  Zorn,  Sprache  u.  heimat  des  Lion  de  B.  Diss. 
Greifswald,  1907. 

3  Cf.  I.  Beth,  "  Federzeichnungen  der  Herpin-Handschrift  in  der  K. 
Bibliothek  zu  Berlin."  Jahrb.  d.  Kon.  preusz.  Kunstsamml.  xxix,  264-75 
(Berlin  1908) ;  E.  Muller,  Uberlieferung  des  Herpin  von  Burges,  Halle,  1905. 


76  MEDIEVAL   ROMANCE    IN    ENGLAND 

1846;  cf.  Ker,  Folk-Lore,  v,  121-27)  in  which  the  hero  mortally 
wounds  a  Red  Knight  who  then  asks  and  receives  Christian 
burial.  By  the  aid  of  a  thankful  beast  no  less  than  by  the  help 
of  the  Red  Knight's  ghost,  Walewein  is  enabled  to  accomplish 
the  various  impossible  tasks  assigned  to  him.  In  the  fifteenth 
century  the  theme  was  combined  with  that  of  the  Two  Friends, 
although  even  in  Sir  Amadas  itself  the  romantic  eagerness  of 
Amadas  to  welcome  the  White  Knight,  to  share  with  him  his  all, 
and  his  willingness  to  give  even  unto  the  life  of  his  child,  suggest 
the  influence  of  Amis  and  Amiloun.  In  this  connection  Gerould 
(p.  92  ff.)  referred  to  the  fifteenth-century  prose  romance,  Olivier 
de  Castile  et  Artus  d'Algarbe,  and  the  derivatives  which  are  here 
discussed  under  Amis  and  Amiloun.  In  sixteenth-century 
English  the  Grateful  Dead  motif  again  appeared  in  that  extraor- 
dinary composite  of  folk-lore,  farce,  and  romance,  Peele's  Old 
Wives'  Tale,  pr.  1595  (Gayley's  Representative  Eng.  Comedies, 
1903).  There  is  also  a  confused  but  recognizable  use  of  the 
theme  in  Massinger's  Fatal  Dowry,  1632,  and  in  Nicholas  Rowe's 
The  Fair  Penitent,  1720. 

The  folk-lore  versions  of  the  Grateful  Dead  number  more 
than  ninety  and  are  of  almost  universal  distribution.  Of  special 
interest  to  English  readers  are  the  numerous  versions  of  Jack  the 
Giant  Killer  which  show  absorption  of  the  theme  (Gerould, 
pp.  24,  70  ff.) 

Origin.  It  is  probable  that  the  origin  of  the  Grateful  Dead 
story  is  to  be  sought  in  ancient  beliefs  about  the  sacred  duty  of 
burial,  and  in  those  customs,  as  old  as  the  ancient  Egyptian  law 
of  which  Herodotus  tells,  which  permitted  a  son  to  pledge  for 
debt  even  his  father's  body,  and  as  modern  as  that  still  practised 
in  Scotland  in  the  sixteenth  century,  by  which  a  corpse  might  be 
left  unburied  until  the  relatives  had  paid  the  dead  man's  debts 
and  the  expenses  of  burial  in  the  churchyard  (Gerould,  p.  162 
ff.).  These  widespread  practices  and  beliefs  are  sufficiently 
similar  to  allow  the  supposition  that  the  basic  tale  was  produced 
sporadically  here  and  there,  but  the  long  integrity  of  the  theme 
"  in  lands  whose  inhabitants  are  connected  by  blood  or  social 
intercourse,"  may  indeed  entitle  it  to  consideration  as  "  an  organ- 
ism with  a  life  history  of  its  own."    Against  those  who  believe 


AMADAS  77 

in  a  Germanic  or  at  least  a  European  origin  Gerould  (p.  167) 
argued,  on  the  evidence  of  the  distinctly  Semitic  origin  and  color- 
ing of  Tobit,  and  of  modern  Asiatic  folk-tales  coming  from  a 
district  practically  unvisited  by  Europeans,  that  it  was  originally 
an  Oriental  folk-tale. 

In  tracing  the  development  of  the  simple  theme  of  the  duty 
of  burial  and  the  gratitude  of  the  ghost,  Gerould  noted  its 
gradual  combination  with  others.  The  reward  obtained  by  the 
hero  is  a  wife;  sometimes  she  is  possessed  by  a  demon;  some- 
times she  is  herself  a  Poison  Maiden  4  from  whose  fatal  embrace 
the  ghost  saves  his  friend ;  sometimes  she  is  a  Woman  Ransomed 
from  slavery  by  the  hero  to  whom  he  is  united  only  after  the 
ghost  saves  him  from  dire  peril  of  enemies  (Gerould,  ch.  iv,  v). 
In  other  cases  the  ghost  obtains  for  the  hero  the  Water  of  Life 
or  achieves  for  him  some  other  equally  Impossible  Task.5  In  these 
more  elaborated  forms  of  the  reward-story  appears  the  motif  of 
the  Bargain  Contract  which  calls  for  a  literal  division  of  all  the 
hero's  gains.  In  the  Poison  Maiden  group  of  stories  the  sharing 
of  the  bride  comes  from  the  desire  either  to  purify  her  or  to 
save  the  hero  from  her  embrace.  Gerould  (p.  75)  disagreed  with 
Hippe's  belief  (p.  181)  that  this  motivation  was  more  primitive 
than  that  in  which  the  ghost's  demand  is  a  test  of  the  hero's 
willingness  to  fulfill  an  obligation.6 

The  combination  of  the  Grateful  Dead  theme  with  that  of  the 
Spendthrift  Knight 7  must  have  been  effected,  according  to  the 
literary  variants,  by  the  middle  of  the  thirteenth  century;  that 
with  the  much  older  motif  of  the  Woman  Ransomed,  before  the 
fourteenth  century.8 

4  The  term  "  Poison-Maiden  "  comes  specifically  from  the  story  in  the 
pseudo-Aristotelian  Secretum  Secretorum  of  the  maiden  reared  on  venom  and 
sent  as  a  Greek  gift  to  Alexander,  who  was  saved  from  her  by  Aristotle.  Cf. 
W.  Hertz,  Die  Sage  vom  Giftmadchen,  1893. 

5  The  Impossible  Tasks  are  achieved  not  only  by  the  aid  of  the  ghost  but 
often  by  helpful  animals  similarly  inspired  by  gratitude.  Cf.  Gerould,  p.  159; 
A.  Wunsch.  Die  Sagen  vom  Lebensbaum  u.  Lebenswasser,  1904;  E.  W.  Hopkins, 
"  The  Fountain  of  Youth,"  Jour.  American  Oriental  Soc.  xxvi.  Gerould 
(p.  119)   noted  the  need  of  a  new  study  of  the  Water-of-Life  theme. 

0  Gerould  (p.  158)  thought  that  stories  of  the  type,  "  The  Child  Vowed 
to  the  Devil,"  may  have  suggested  the  idea  of  including  the  child  in  the 
sacrifice   to  the  ghost.     See   Gowther  here   and  Meyer,  Rom.  xxxni,   163. 

7  See  Cleges. 

8  Gerould,  p.  82,  171.     The  combination  appears  in  the  Scala  Celi  com- 


78  MEDIAEVAL   ROMANCE   IN   ENGLAND 


BIBLIOGRAPHY 

Texts:  (i)  Ireland  Ms.  ed.  Robson,  Camden  Soc.  1842;  (2)  MS. 
Jac.  v,  7,  27,  Advocates  Library,  Edin.,  ed.  Weber,  1810,  ill,  243-75.  The 
two  MSS.  were  printed  together  by  G.  Stephens,  Ghost  Thanks,  or  the 
Grateful  Dead,  Cheapinghaven,  i860.  Trans.  E.  Rickert,  Romances  of 
Friendship,  pp.  40-67;  J.  Weston,  Chief  Middle  Eng.  Poets,  p.  216  ff. 

Studies:  Cf.  Wells,  Manual,  p.  787. 
Benary,  W.  "  Hervis  von  Metz  u.  die  sage  von  dankbaren  Toten."  Zts.  f. 

rom.  Phil,  xxxvn,  57-92;    xxxvm,  229   (1914). 
Cock,  A.  De  Sage  van  den  te  gast  denooden  doode.  46  pp.  Ghent,  1909. 
Dutz,  H.  "  Der  Dank  des  Todten  in  der  engl.  Literatur."  Jahresber.  d. 

Staats-Oberrealschule.  Troppau,  1894. 
Gerould,  G.  H.  The  Grateful  Dead,  The  History  of  a  Folk  Story.    Folk 

Lore  Soc.  Lond.  1907.    Bibliography,  pp.  7-25. 
Hippe,  M.  "  Unterschungen  zu  d.  mittelengl.    Romanze  von  Sir  Amadas." 

Archiv.  lxxxi,  141-183  (1888).    Rev.  Rom.    xviii,  197. 
Holthausen,  F.  "  Sir  Amadas  und  Peele's  Old  Wives  Tale."  Archiv.  cxvii, 

177  (1907). 
Kohler,   R.   "  Die  dankbaren   Toten   u.   der  gute   Gerhard."     Kleiner e 

Schriften,  1,  5-29.     Weimar,  1898,  1900. 
Stephens,  G.  See  Texts. 


posed  by  Johannes  Junior  (Gobius).  In  this  Saint  Nicholas  plays  the  part  of 
the  ghost.  To  rebuild  the  church  of  the  saint  a  merchant  impoverishes  him- 
self. He  wins  and  loses  a  wife,  the  daughter  of  a  sultan.  He  regains  her 
through  the  help  of  the  saint. 


CLEGES 

Versions.  Unique  in  Middle  English  for  its  combination  of 
humor,  piety,  and  romance,  is  the  pleasant  little  poem  of  Sir 
Cleges,  which  is  preserved  in  two  fifteenth-century  manuscripts. 
Common  rimes  show  the  derivation  of  these  texts  from  an  earlier 
English  original,  but  in  detail  they  differ  widely,  as  if  written 
down  from  memory  or  oral  recitation  (McKnight,  p.  lxxiv). 
The  familiar  verse  tag,  "  so  seyeth  9e  boke  "  (1.  248),  may  or 
may  not  refer  to  this  lost  version,  but  it  is  at  least  possible  from 
a  comparison  of  the  two  extant  texts  to  conclude  that  it  was 
composed  in  the  North  Midland  district  about  the  beginning  of 
the  fifteenth  century  (Treichel,  p.  371).  The  verse  is  the  twelve- 
line  stanza,  riming  usually  aabccbddbeeb. 

Origin.  Like  so  many  others,  Cleges  is  clearly  a  minstrel 
tale.  Naively  it  praises  gift-giving  to  minstrels  and  Christmas 
feasts ;  naively  it  enjoys  the  punishment  of  the  surly  porter,  the 
traditional  enemy  of  minstrels ;  and  a  minstrel  "  geste  "  is  made 
partly  responsible  for  the  impoverished  hero's  restoration  to 
fortune.  In  verse,  as  in  spirit,  one  hears  the  voice  of  some 
shrewd,  singing  wayfarer.  With  catholic  taste,  he  combines  in 
somewhat  motley  sort  themes  proper  to  entirely  different  literary 
types.  Influenced  probably  by  Sir  Atnadas,  he  begins  with  the 
motive  of  the  "  Spendthrift  Knight,"  l  and  touches  it  with 
homely  tenderness  in  the  picture  of  Cleges,  grieving  over  his 
ruined  state,  being  comforted  by  his  gentle  wife  and  little 
children.  As  in  a  conte  devot,  he  tells  of  Cleges's  prayer  for 
help  and  the  answering  miracle  of  the  cherry  tree2  hung  with 

1  McKnight,  pp.  Ixiii  ff.,  noted  as  other  examples  of  the  Spendthrift  Knight 
motif  Chestre's  Launfal  (ed.  Eng.  Stud,  xvm)  and  the  later  Sir  Lambewell 
(Percy  Folio  MS.  n),  the  fifteenth-century  Knyght  and  his  Wyfe  (Hazlitt, 
Remains,  n),  and  a  True  Tale  of  Robin  Hood  (Child,  no.  154).  See  here 
under  Amadas. 

2  The  miracle  of  getting  fruit  out  of  season  appears  in  saint  legends,  in 
stories  of  Impossible  Tasks,  and  of  magical  accomplishment.  Cf.  McKnight, 
p.  lxv;  Chambers,  Medieval  Stage,  1,  252-53;  Tatlock,  Chaucer  Soc.  2nd 
Ser.  51,  p.  55,  note  3. 

79 


80  MEDIAEVAL   ROMANCE    IN    ENGLAND 

fruit  on  Christmas  day.  This  is  at  once  the  knight's  reward  for 
his  previous  generosity  and  the  means  of  his  rehabilation.  The 
minstrel  draws  on  romance  for  the  dim  Arthurian  setting  of  the 
court  of  King  Uther,  whose  famous  gallantries  to  Igerne  are 
faintly  suggested  in  the  king's  gift  of  Cleges's  cherries  to  a 
"  bryght  lady  "  of  Cornwall.  Even  the  name  Cleges,  which 
appears  in  the  stories  of  Chretien  and  Malory,  belongs  to  ro- 
mantic tradition.  But  as  the  poet  goes  on  to  tell  of  the  hero's 
encounter  with  the  greedy  porter,  the  usher,  and  the  steward, 
who  each  demand  a  third  of  his  reward  before  admitting  him  to 
the  king's  presence,  and  on  whose  account  he  asks  for  a  reward 
of  twelve  blows,  romance  gives  way  to  a  widely  known  folk- 
tale. As  jest,  biographical  anecdote,  fabliau,  or  moralized 
exemplum,  the  story  of  the  "  Blows  Shared  "  or  "  Greed  Re- 
quited "  appears  in  the  English  Gesta  Romanorum 3  and  in 
popular  French,  German,  Spanish,  Latin,  Greek,  Swedish,  Italian, 
Turkish,  and  Arabic  versions.  It  is  one  of  those  universal  tales 
common  to  all  races  and  all  times  to  which  no  date  or  home 
can  be  assigned.  So  far  as  is  now  known,  Sir  Cleges  is  simply 
one  of  many  independent  versions  (McKnight,  p.  lxxii  ff.). 


BIBLIOGRAPHY 

Texts:  (i)  Jac  V,  7,  27,  Advocates  Library,  Edin.,  ed.  Weber,  1,  331 
(1810);  Ashmole  61,  Bodleian,  Oxf.  (576  lines),  ed.  G.  H.  McKnight, 
Middle  English  Humorous  Tales  in  Verse,  pp.  39-59,  Boston,  1913;  The 
two  MSS.  were  ed.  by  A.  Treichel,  Eng.  Stud,  xxn,  374  ff.  Trans,  by 
J.  Weston,  Libeaus  Desconus  and  Sir  Cleges,  Lond.  1902. 

Studies:    McKnight,  Introd.  lxi-lxxv,  and  Bibliography,  p.  89  ff. 

3  For  the  "  Blows  Shared  "  story  in  England  see  the  English  Gesta  (EETSES. 
xxxin,  413,  no.  xc) ;  the  Latin  story  of  Invidia  of  John  Bromyard,  Summa 
Praedicantium,  f.  cxin,  b,  and  the  Pleasant  Conceites  of  Old  Hobson,  the 
Merry  Londoner  (Hazlitt,  Shakespeare's  Jest  Books,  p.  40,  no.  24).  Cf. 
McKnight,  pp.  lxix-lxxi.  As  the  latter  points  out,  the  story  is  akin  to  the 
tale  of  the  Three  Wishes,  (Bedier,  Les  Fabliaux,  1895,  p.  220),  to  the  Dit  du 
Buffet  (Montaiglon  et  Raynaud,  Fabliaux,  1872-90,  Notes)  and  to  the  one 
called  "  Lucky  they  are  not  peaches "  (W.  Clouston,  Popular  Tales,  n, 
467  ff.). 


ROMANCES  OF  LEGENDARY  ENGLISH  HEROES 


KING   HORN 

Versions.  Horn  is  one  of  the  best  stories  that  came  out  of 
Anglo-Norman  England.  The  manner  of  its  telling  varies  in 
different  versions,  but  its  narrative  vigor  does  not  greatly  change. 
"  Blithely  "  enough  it  sets  forth  the  story  of  the  boy  Horn  and 
the  loss  of  his  kingdom,  of  his  bringing  up  at  the  court  of  a 
foreign  king,  of  his  winning  the  love  of  a  foreign  princess,  of  a 
false  accusation  made  against  Horn  and  his  second  exile,  of  his 
adventures  at  the  court  of  another  king,  of  his  two  rescues  of  his 
sweetheart  from  unwelcome  suitors,  and  of  his  final  recovery  of 
his  heritage.  Outworn  as  most  of  these  situations  were  destined 
to  become  in  the  development  of  mediaeval  fiction,  in  the  Horn 
legend  they  are  freshly  told  and  altogether  diverting. 

The  earliest  manuscripts  of  the  story  in  either  French  or 
Middle  English  belong  to  the  thirteenth  century.  The  Anglo- 
Norman  version1  consists  of  5250  alexandrines  rimed  in  "ti- 
rades." Its  original  has  in  general  been  ascribed  to  the  twelfth 
century,  though  some  scholars,  on  account  of  the  elaborated 
picture  of  court  life  presented  in  the  poem,  have  placed  it  in  the 
first  half  of  the  thirteenth  century  (Hartenstein,  pp.  19-20). 
The  increasing  realization  of  the  influence  upon  it  of  the  Anglo- 
Norman  Tristan 2  lends  likelihood  to  this  later  date.     In  the 

1  This  is  extant  in  three  long  manuscripts  and  in  the  Cambridge  fragments 
printed  in  192 1  by  Braunholtz.  The  poem  was  first  edited  in  1845  by  Michel 
under  the  title  Horn  et  Rimenhild  (RH).  It  was  re-edited  in  1883  by 
Brede  and  Stengel,  Ausgaben  u.  Abhandlungen,  vni,  and  a  new  critical 
edition  is  under  consideration  (P.  Studer,  Study  of  Anglo-Norman,  Oxford, 
1920,  p.  28). 

2  Thomas's  Tristan  was  written  between  1155  and  11 70  (Bedier,  Tristan, 
H,  p.  55) ;  or  possibly  as  late  as  1189  (Loomis,  MLR.  xw,  39;  xvn,  26).  It 
is  to  be  regretted  that  no  complete  historical  study  of  the  names  in  RH 
has  yet  been  made,  for  behind  them  may  lurk  a  good  deal  of  historical  evi- 
dence, to  say  nothing  of  connections  with  the  chansons  de  geste.  For  a 
list  of  the  proper  names  in  RH,  see  Mettlich,  Bemerkungen  zu  d.  agn.  Lied 
vom  wachern  Ritter  Horn,  Munster,  1896;  rev.  Eng.  Stud,  xvi,  306.  Heuser, 
pp.  1 1 2-1 5,  emphasized  the  vital  importance  of  the  name-evidence  in  the 
French  version  for  any  source  study  of  the  legend. 

83 


84  MEDIAEVAL   ROMANCE   IN   ENGLAND 

extant  version  the  author  alludes  several  times  to  a  written 
source,  his  "  escrit  "  (1.  192),  his  "parchment  "  (11.  2993,  3981), 
and  refers  to  himself  as  "  Mester  Thomas  "  and  to  his  "  fiz 
Gilimot  /Ki  la  rime  apre  mei  bien  controuerat  "  (1.  5240).  Who 
this  Thomas  was  has  not  been  determined.  Hartenstein  (pp.  23- 
25)  believed  that  he  was  a  native  of  southern  England,  who 
had  possibly  been  in  Brittany;  that  his  native  speech  was 
Anglo-Norman,  but  that  he  was  not  altogether  ignorant  of 
English  since  he  introduced  such  words  as  horn  (1.  4206)  and 
God  wite  (1.  4013)  into  his  text;  that  he  was  a  poet  familiar 
not  only  with  the  Horn  story  but  with  legends  about  Horn's 
father;  finally,  that  he  was  old  enough  when  he  finished  Horn 
to  have  a  son  capable  of  continuing  his  work.  Hartenstein's 
idea  (p.  34)  that  this  Thomas  was  a  traveling  minstrel  depended 
chiefly  on  the  many  references  to  minstrels  and  their  accom- 
plishments which  come  into  the  story.  But  its  length  and 
elaboration  argue  that  its  author  was  of  more  leisurely  habit  of 
life  than  most  minstrels.  He  may  well  have  been  a  courtly 
clerk  like  his  countryman  and  possible  contemporary,  Hue  de 
Roteland,  the  author  of  Ipomedon. 

The  French  Horn,  as  Soderhjelm  observed  (Rom.  1886,  xv, 
579  ff.),  has  in  general  the  spirit  of  a  chanson  de  geste  and  the 
romantic  character  of  a  story  of  the  Table  Round.  Epic  ten- 
dencies are  shown  in  lusty  fights  between  Christians  and  pagans, 
in  the  glorification  of  prowess,  the  enthusiastic  descriptions  of 
battle  after  battle  in  which  Christian  heroes  are  individually 
portrayed,  and  in  the  practical  absence  of  all  supernatural  ele- 
ment save  the  prophetic  "  epic  dream."  Horn  himself  is  con- 
ceived primarily  as  a  fighter,  and  the  romance  has  as  a  whole 
an  air  more  militant  than  romantic.  Yet  it  was  definitely 
influenced  by  romances  of  the  "  courtois  "  type ;  its  manners  and 
customs  are  altogether  feudal  and  chivalric;  it  lingers  over  all 
the  scenes  of  courtly  life,  over  the  elaborate  feasts,  the  tourna- 
ments, the  games  and  sports  of  the  young  nobles,  over  the 
description  of  the  rooms  of  the  Irish  princess  Lemburc,  over 
details  of  costume  and  feeling  (Stimming,  Eng.  Stud.  1,  357-60). 
The  courtly  line,  "  De  la  belte  de  Horn  tute  la  chambre  re- 
splent  "  (1.  1053),  which  recalls  a  similar  gracious  bit  of  fantasy 
in  Aucassin,  proved  too  charming  for  even  the  abrupt  Middle 


KING   HORN  85 

English  redactor  to  resist  (Hall,  p.  117).  The  influence  of 
Tristan  seems  most  palpable  in  the  scenes  in  which  the  disguised 
young  Horn  sings  before  Lemburc  or  Rimenhild  (Schofield,  p. 
60).  The  women  characters  of  the  French  Horn  are  wholly 
without  the  emotional  subtlety  of  those  in  Tristan,  but  it  is  sug- 
gestive of  the  sophisticated  character  of  the  romance  that  Rimel 
has  her  confidante  Herselot,  and  is  attended  by  a  retinue  of 
lovely  maidens. 

The  Middle  English  version  of  King  Horn  (KH),  like  that 
in  French,  has  been  preserved  in  three  manuscripts.  The  oldest 
and  probably  the  nearest  to  the  original  text  is  the  Cambridge 
manuscript  (C),  written  about  1260  by  an  Anglo-Norman  scribe 
(Hall,  p.  x).  Wissmann  (1,  p.  15)  believed  that  the  three  texts 
were  independent  variants  derived  from  oral  transmission,  but 
Zupitza  (Anz.  /.  deut.  Alt.  ix,  181-92)  and  Hall  (p.  xiv) 
emphasized  the  textual  relationship  of  Laud  and  Harleian  2253, 
that  famous  anthology  of  lyrics,  satires,  fables,  and  saint  legends 
which  may  have  emanated  from  the  priory  of  Leominster  (Hall, 
p.  viii).  McKnight  (p.  xxviii)  thought  the  original  Middle 
English  version  of  Horn  must  have  been  composed  in  the  last 
half  of  the  thirteenth  century,  probably  in  the  Middle  South; 
in  Essex,  according  to  Wissman  (Untersuch.  p.  33) ;  in  North- 
West  Surrey,  according  to  Hall  (p.  xliv).  The  poem  is  appar- 
ently transitional  in  form,  for  it  is  rimed  but  is  not  in  the  "  beat 
verse  "  of  romance  rhythm.  Some  have  seen  in  it  "  the  coming 
to  light  again  of  the  primitive  Teutonic  measure  song  verse  "  of 
the  four-stress  "  Otfrid  "  type  (Luick,  Paul's  Grundriss,  1893,  11, 
994,  1004;  1007;  Wissman,  Untersuch,  p.  56);  others,  like 
Schipper  (Grund.  d.  Eng.  Metrik,  Bonn,  1885),  recognized  in 
it  the  natural  development  of  the  Old  English  alliterative  verse 
under  the  influence  of  French  prosody  with  its  insistence  on 
syllabic  regularity  and  rime.  McKnight  (p.  xxiii)  felt  that  in 
the  verse  of  Horn  alliteration  has  become  an  unessential  element, 
and  agreed  with  Schipper  (pp.  71-72)  that  in  about  1300  verses 
the  prevailing  form,  a  variety  of  Siever's  A  type  in  Anglo-Saxon 
verse,  has  three  accents  and  feminine  rime.  Schipper's  scansion 
of  the  poem  was  also  accepted  by  Hall.  West  (ch.  iv)  argued, 
however,  that  Horn  shows  a  two-stress  movement  in  free  rhythm 
and  that  in  its  internal  structure  the  Horn  couplet  is  only  a 


86  MEDLEVAL   ROMANCE    IN   ENGLAND 

natural  Middle  English  development  of  the  Anglo-Saxon  four- 
stress  long  line  when  rimed  in  equalized  short  lines  (pp.  35,  50, 
88). 

In  comparison  with  the  French  version,  King  Horn  (KH) 
seems  as  abrupt  as  it  is  virile  and  primitive.  Its  poet  cares 
nothing  for  the  knowing  courtliness  of  Thomas.  Though  he  has 
the  same  story  to  tell,  he  hurries  through  it  as  to  the  swift 
twanging  of  his  own  harp.  His  Horn  is  "  blithe  "  to  be  alive, 
blunt  to  the  point  of  rudeness,  brutal  and  gay,  impatient  of 
everything  save  fighting,  a  wholly  unsentimental  lover  who  is 
more  interested  in  rescuing  his  sweetheart  than  in  remaining 
with  her.  The  bareness  of  scene,  the  simplicity  of  motive,  and 
lively  vigor  of  action,  give  the  English  Horn  a  popular, 
ballad-like  quality. 

Because  of  such  archaic  effects  as  these,  many  German  and 
English  scholars  have  thought  the  English  poem  older  than  the 
Anglo-Norman  version.3  Such  a  conclusion  not  only  contradicts 
such  evidence  as  the  manuscripts  give,  but  ignores  the  French 
influence  in  the  metre,  vocabulary,  and  spirit  of  the  poem.4 
As  Schofield  (p.  3)  pointed  out,  romance  scholars  are  unanimous 
in  declaring  that  the  extant  French  version  is  earlier  than  the 
English.  The  essential  identity  of  the  two  stories  has  always 
been  recognized,  but  in  all  the  study  lavished  on  their  individual 
traits,  aside  from  the  matter  of  terminology,  not  more  than  eight 
passages  have  been  noted  which  are  peculiar  to  the  English  Horn 
(McKnight,  p.  x).  Of  these  not  one  changes  the  story  or  is 
more  than  a  natural  enough  variation  for  an  author  as  different  in 
poetic  character  and  purpose  as  the  English  minstrel  was  from 
Mestre  Thomas.    In  their  specialized  study  of  the  names  in  the 

3  Child,  Ballads,  1,  192;  Billings,  Guide,  p.  5;  McKnight,  p.  xii;  Hall, 
p.  liv;  Wissman,  Untersuch.,  p.  113;  Stimming,  Eng.  Stud.  1,  352.  Harten- 
stein,  pp.  19-20,  78,  109,  recognized  the  priority  of  the  French  version. 

4  For  the  romance  words  in  the  vocabulary  see  Hall,  pp.  xxix-xxxi;  for 
metre  West,  passim.  Hartenstein,  p.  116,  recorded  the  ninety-five  French 
rimes  of  KH,  but  thought  the  number  too  small  to  be  significant.  McKnight, 
p.  x,  who  assumed  that  RH  borrowed  from  KH,  listed  a  number  of  close 
verbal  parallels  between  the  two  poems.  Though  the  unromantic  and 
uncourtly  style  of  KH  gives  the  effect  of  much  greater  primitiveness  than 
does  the  style  of  the  French  poem,  it  must  be  noted  that  KH  depends 
absolutely  for  content  on  romantic  themes  which  were  net,  as  far  as  we  know, 
familiarized  in  England  until  after  the  Conquest.  Cf.  Schofield,  p.  52; 
McKnight,  Horn,  p.  xx. 


KING   HORN  37 

English  poem,  both  Morsbach  (p.  297)  and  Schofield  (pp.  51-53) 
became  convinced  that  it  had  come  directly  from  some  earlier 
French  version.  Schofield,  however,  continued  to  believe  that 
behind  this  inevitable  "  lost  "  French  version,  there  must  have 
been  a  still  earlier  one  in  English.5  Though  Heuser  did  not 
credit  the  name  studies  of  Morsbach  and  Schofield,  he  admitted 
(p.  129)  the  possibility  that  an  early  Bretonized  lay  existed,  and 
that  an  English  redaction  was  made  in  the  first  half  of  the 
eleventh  century. 

In  addition  to  the  French  and  Middle  English  versions  already 
noted,  the  story  of  Horn  is  also  preserved  in  the  poem  known 
as  Horn  Child  (HC),  in  certain  Scotch  ballads  of  Horn,  and  in 
the  prose  romance,  Ponthus  et  la  belle  Sidoyne,  written  between 
1372  and  1390  in  honor  of  the  famous  Tour  Landri  family  of 
Anjou  (G.  Paris,  Rom.  xxvi,  468-70).  The  earliest  manuscript 
of  Ponthus  is  in  the  magnificent  volume,  Royal  15,  E,  vi,  of  the 
British  Museum,  which  was  presented  to  Margaret  of  Anjou  in 
1445  (Ward,  Catalogue,  1,  130).  Not  all  the  French  manuscripts 
of  this  romance  have  been  noted,  but  the  favor  in  which  it  was 
held  is  suggested  by  the  seven  manuscripts  listed  by  Hartenstein 
(p.  144)  and  by  the  seven  French  editions  appearing  between 
1478  and  1548  (Brunet,  Manuel  du  Libraire,  Paris,  1863).  In 
England  the  story  was  translated  about  the  middle  of  the 
fifteenth  century  in  a  version  of  which  two  copies  survive  in 
manuscripts.  The  same  version  is  also  found  in  Wynkyn  de 
Worde's  edition  of  15 11  and  in  another  edition  of  1548  (Esdaile, 
List  of  Eng.  Tales  before  1740,  p.  113).  About  1465  Ponthus 
was  translated  from  French  into  German  by  Eleanor,  daughter 
of  James  I  of  Scotland  and  wife  of  Archduke  Sigismund  of 
Austria.  Eleanor's  translation  was  frequently  copied  and  printed 
in  various  revisions,  among  them  the  sixteenth-century  Buch  der 

5  The  basis  for  this  belief  is  not  clear.  In  passing,  Schofield,  p.  36, 
referred  to  the  English  names  in  the  story  and  implied,  pp.  29,  51,  not  only 
that  the  recurrent  pun  on  the  hero's  name  in  RH  and  KH  was  in  an  original 
English  version  of  the  story  but  that  this  version  preceded  the  Conquest. 
The  evidence  for  this  was  an  allusion  in  the  French  Waldej  to  English  versions 
of  Tristan,  Waldej,  and  Aalof,  a  hero  who  is  mentioned  in  RH  as  the  father 
of  Horn.  The  unsatisfactory  nature  of  this  reference  is  discussed  below 
(Horn  Child,  note  6),  but  at  best  it  offers  an  argument  from  analogy  only. 
Even  the  proved  existence  of  an  Old  English  Aalof  saga  would  not  establish 
the  existence  also  of  a  Horn  saga. 


88  MEDIEVAL   ROMANCE    IN    ENGLAND 

Liebe.  Through  the  sixteenth  and  seventeenth  centuries  the 
story  in  romance  or  chap-book  form  continued  its  popularity 
(Hartenstein,  pp.  146-48).  From  Germany  it  passed  into  Ice- 
land, and  there  in  the  sixteenth  century  an  Icelandic  Ponthus 
was  begun  (Mather,  p.  xli,  ff.). 

The  relation  of  Ponthus  to  the  Anglo-Norman  Horn  was  first 
noted  by  Weber  {Met.  Romances,  18 10,  in,  361).  Mather,  (pp. 
vi-xvii)  showed  how  carefully  the  French  author  used  every 
essential  element  of  the  earlier  plot  of  Horn  but  how  he  changed 
the  localization  of  the  story,  substituting  Spain  for  Suddene, 
England  for  Ireland,  and  intensified  by  the  introduction  of  many 
French  scenes  and  historical  characters  the  interest  for  a  French 
audience.  His  special  purpose  in  retelling  it,  apart  from  glorify- 
ing the  Tour  Landri  family,  was  to  make  it  a  book  of  intro- 
duction in  courtesy ;  and  for  this  purpose  he  dwelt  on  the  beauty, 
accomplishments,  and  virtues  of  his  hero  Ponthus,  who  is  repre- 
sentative of  the  later,  more  complicated  ideal  of  knighthood. 
Mather  (p.  xlvii)  noted  a  certain  sweetness  and  gaiety  of  spirit 
which  save  Ponthus,  despite  his  virtues,  from  being  "  a  Grandison 
out  of  due  time."  In  the  French  author's  style  there  was  no  dis- 
tinction, but  in  the  English  version  Mather  felt  some  improve- 
ment. Though  he  would  not  compare  it  with  Malory's  subtly 
beautiful  cadences,  he  found  in  it  the  brisk  and  fluent  virtue  of 
unaffected  talk. 

The  latest  versions  of  the  Horn  story  are  found  in  nine  or 
ten  Scotch  ballads  of  Hind  Horn?  and  in  one  version  known  as 
Colhorn  (MacSweeney).  In  these  texts  the  old  story  is  reduced 
to  a  matter  of  seventy  lines  or  less.  The  lovers  exchange  gifts ; 
Horn  receives  a  magic  ring;  when  it  warns  him  of  his  love's 
danger  he  returns,  hears  from  a  beggar  of  her  imminent  mar- 
riage, enters  the  castle,  and  begs  a  drink  in  Horn's  name.  The 
bride  recognizes  the  ring  he  drops  in  the  cup,  and  joyously  offers 
to  go  beg  with  him  "  frae  town  to  town."  Traits  such  as  the  dis- 
coloration of   the   ring,   the   beggar   disguise,   the   ironic   final 

6  Child,  Ballads,  no.  17;  Hartenstein,  p.  88.  Nelles  believed  the  likeness 
between  Horn  and  the  debased  ballad  of  the  Ritchie  Boy  (Child,  Vol.  iv, 
400)  is  more  a  matter  of  stock  stanzas  than  of  actual  relationship.  Mac- 
Sweeney,  (p.  210)  spoke  of  the  influence  of  Hind  Horn  in  Young  Beichan, 
Lady  Diamond,  Robin  Hood  rescuing  three  Squires. 


KING   HORN  89 

couplet  (A  24),  seem  to  link  the  ballads  with  Horn  Child  rather 
than  with  King  Horn.  Hartenstein  (p.  124)  believed  the  ballads 
originated  in  a  North-English  folk-saga  from  which  all  extant 
versions  were  derived,  and  Nelles,  who  has  made  the  closest 
study  of  the  ballads,  accepted  the  theory  of  a  common  source  for 
them  and  Horn  Child. 

Origin.  Whether  the  original  Horn  story  came  from  the 
Anglo-Saxons,  the  Danes,  the  Norse  Vikings,  or  the  Anglo- 
Normans,  is  still  a  matter  of  dispute.  Germanic  elements  in  the 
story  were  noted  as  early  as  181 1  by  Grimm  (Mus.  f.  altd. 
kunst  u.  Litt.  Berlin,  11,  303).  Stimming  (Eng.  Stud.  1,  355) 
likewise  thought  that  the  names  were  Germanic,  and  McKnight 
made  the  same  claim  for  such  dominant  motifs  as  the  exile-and- 
return  of  the  hero  and  the  separation  and  reunion  of  faithful 
lovers.  In  191 1  Grass  (p.  38)  too  urged  that  Horn  as  a  bride- 
winning  story  was  closely  related  to  Kudrun  and  other  Germanic 
tales.  But  Wissman's  observations  (Anglia  iv,  398)  that  there 
is  nothing  in  Horn  which  can  be  called  exclusively  Germanic 
appear  still  to  hold  good.  In  general  the  topography  of  the 
poem  has  come  to  be  the  decisive  test  of  critical  opinion.  Those 
who  identify  Suddene,  Horn's  own  home,  as  Surrey  (Michel, 
RH,  p.  454)  or  Sussex  (Ward,  Catalogue,  I,  450)  or  as  South 
Devon  (Su5defne,  Heuser,  p.  119)  or  any  other  part  of  the 
southern  coast  of  England,  believe  in  the  English  origin  of  the 
story  and  relate  its  events  to  the  Anglo-Danish  period.  Harten- 
stein (p.  29)  and  the  latest  editors  of  the  Middle  English  poem, 
Hall  (p.  Iv)  and  McKnight  (p.  xvii),  expressed  themselves  in 
favor  of  this  English  derivation. 

The  belief  in  the  Danish  origin  of  the  legend  rests  on  the 
identification  of  Suddene  as  Suddene,  the  land  of  the  South- 
Danes,  who  are  mentioned  in  Beowulf.  Ten  Brink  (Hist.  Eng. 
Lit.  1,  150)  was  convinced  that  "  the  North  Sea,  its  neighboring 
waters,  and  their  shores,  were  the  scene  of  the  action."  Horn 
himself  was  tentatively  identified  by  Suchier  (Gesch.  d.  frz.  Lit., 
p.  in)  with  the  Danish  Horm  who  went  in  851  to  Ireland  where 
he  was  hospitably  received  and  fought  victoriously  against  the 
Norwegian  Vikings.  Deutschbein  (pp.  15-26)  made  the  same 
suggestion  and  went  on  to  trace  the  gradual  Christianization  of 


90  MEDIEVAL   ROMANCE   IN   ENGLAND 

the  narrative  and  its  absorption  of  the  names  and  characters  of 
later  Norse-Irish  history.  For  instance,  the  "  Arald,  crown  prince 
of  the  foreigners  of  Erinn,"  who  was  killed  at  the  great  battle  of 
Glenmama  in  iooo,  was  equated  by  Deutschbein  with  Arild,  son 
of  King  Thurston  of  Ireland,  in  the  romance,  and  in  a  later 
article  Deutschbein  (3,  p.  55)  discovered  Horn's  true  historic 
prototype  to  have  been  the  Viking,  Eyvindr  Urarhorn.  But  the 
evidence  concerning  Horm  or  Eyvindr  or  the  Norse  Orns  men- 
tioned by  Schofield  (p.  29)  is  slight  at  best,  and  the  assumption 
that  the  story  of  Horn's  life  at  the  Irish  court  is  the  oldest  part 
of  the  legend,  is  seriously  open  to  question.  The  "  duplication 
of  climax,"  once  regarded  as  evidence  that  independent  tradi- 
tions about  Horn  had  been  linked  together,  was  shown  by 
McKnight  (p.  224)  to  be  a  familiar  enough  feature  in  mediaeval 
fiction.  Since  in  all  versions  the  most  vital  part  of  the  story  is 
Horn's  relation  to  Rimenhild  and  his  rescues  of  her,  it  seems 
difficult  to  believe  that  the  whole  story  of  their  early  love  was 
invented  as  mere  duplication  of  the  Irish  episode.  Deutsch- 
bein's  belief  (p.  4)  in  the  priority  of  the  Irish  adventures  was 
due  to  what  seems  a  mistaken  emphasis  on  the  detail  that  in 
Ireland  Horn  had  a  chance  to  kill  some  of  the  enemies  who  had 
slain  his  father.7  But  this  is  only  a  palpable  device  for  linking 
unrelated  episodes  together ;  the  real  vengeance  of  Horn  belongs 
to  the  later  part  of  the  story  in  which  he  returns  to  his  kingdom 
and  regains  it  after  a  general  destruction  of  his  foes.  If  the 
secondary  character  of  the  Irish  adventures  be  admitted,  then 

7  Ward,  Catalogue,  1,  448,  also  believed  that  the  Irish  episode  was  the 
oldest  part  of  the  story.  But  the  reason  he  gave,  i.e.,  that  Horn  regains  his 
love  and  his  heritage  through  the  help  of  Irish  knights,  is  as  open  to  question 
as  Deutschbein 's.  In  romance  heroes  acquire  troops  with  incredible  ease; 
troops  are  mere  accessories.  In  the  traditional  situation  the  lover  who  returns 
to  rescue  his  sweetheart  from  an  enemy's  power  is  usually  alone,  just  as  in 
the  ballad  versions  Horn  is  alone.  In  any  primitive  version  this  was  probably 
the  case.  To  the  present  writer  it  seems  as  futile  to  argue  the  priority  of  the 
Irish  Rimel-Lemburc  story  to  the  rest  of  RH  as  it  would  be  to  urge  that 
in  the  Tristan  legend  the  episode  of  the  second  Isolt  was  the  earliest.  The 
practical  identity  in  KH  of  the  names  Rymenhild  and  Reynild  shows  that 
even  the  English  poet  was  familiar  with  this  device  of  writers  using  in  whole  or 
in  part  the  theme  of  the  Man  with  Two  Wives.  (Cf.  Matzke,  Mod.  Phil. 
1907-8,  v,  and  here,  Lay  le  Freine,  note  6.)  On  different  grounds,  Grass, 
p.  50,  also  expressed  disbelief  in  the  "  historical  "  origin  of  Horn,  and  in  the 
priority  of  the  Reynild  story. 


KING   HORN  9I 

the  theory  that  they  represent  the  historic  kernel  of  the  romance 
is  hardly  tenable. 

In  1903  Schofield  set  forth  an  ingenious  argument  to  prove 
that  the  original  Horn  story  was  an  old  Norse  tale  of  the 
tenth  century  orally  transmitted  to  some  Old  English  poet,  and 
that  his  Old  English  version  was  translated  by  an  Anglo-Norman 
into  the  simple  version  which  was  the  source  of  the  extant 
Middle  English  Horn.  Schofield  identified  Suddene  as  the  Isle 
of  Man,  from  which  a  strong  north-west  wind  might,  as  the  story 
tells,  drive  Horn's  boat  in  twenty-four  hours  to  England,  Westir 
as  Ireland,  and  Westernesse  (Western  Ness)  as  the  Wirral  penin- 
sula. Topographically  Schofield 's  theory  is  in  many  ways  reason- 
able, but  unfortunately  it  rests,  as  did  the  identification  of 
Suddene  as  Surrey,  on  the  single  recorded  instance  in  the  manu- 
script of  Gaimar's  Estorie  des  Engles  in  which  the  one  word  is 
used  for  the  other.  Schofield  (p.  7,  12)  pointed  out  that  in  form 
the  old  name  of  Surrey,  Sudreie,  would  be  identical  with  the 
name  given  by  the  Norse  Vikings  to  the  Isle  of  Man,  Su  rey, 
as  one  of  the  Su  reyjar,  South  Isles,  and  that  it  would  not  be 
impossible  for  French  writers  to  turn  Surrey  into  Sudene  even 
as  they  turned  Orkneye  into  Orceine.  He  argued  further  (p.  39) 
that  the  events  and  places  in  KH  were  of  a  sort  familiar  to 
the  Norsemen  of  the  ninth  and  tenth  centuries,  and  that  the 
Horn  story  itself  was  of  the  same  type  as  that  told  of  the 
Viking,  Gunnlaug  Serpent-Tongue  (983-1009),  whose  actual 
history  received  poetic  embellishment  and  was  orally  transmitted 
for  several  generations.  The  realistic  impression  given  by  the 
English  Horn,  Schofield  (pp.  43-45)  believed,  was  due  to  its 
reminiscences  of  actual  Norwegian  depredations  of  the  tenth 
century,  introduced  by  the  Old  English  poet  to  whom  the  story 
of  a  Viking's  adventures  had  come. 

The  Namenuntersuchung  set  forth  by  Heuser  in  1908  did 
more  to  advance  the  true  understanding  of  the  Horn  legend  than 
earlier  studies  in  so  far  as  he  emphasized  the  importance  of  the 
earliest  extant  text,  the  French  Horn  et  Rimenhild,  which  other 
scholars,  obsessed  by  the  idea  of  the  primitiveness  of  King  Horn, 
have  neglected.  Heuser  insisted  that  all  the  supposed  historical 
elements  in  the  different  versions  are  variable,  whereas  the  love 
story  of  Horn  and  Rimenhild  remains  constant,  recognizable  in 


92  MEDIEVAL   ROMANCE    IN    ENGLAND 

whatsoever  form,  and  persistently  dear  to  people  living  in 
England.8  The  roots  of  the  story,  he  believed,  must  have  been 
English  and  have  had  their  growth  in  Cornwall,  where  the 
romantic  adventures  of  Hereward,  which  Deutschbein  (p.  55), 
Wissman  (Untersuch.  p.  no),  and  others,  have  recognized  as  a 
variant  of  the  Horn  story,  are  clearly  localized.  Of  all  explana- 
tions of  Horn's  name,  that  of  Heuser's,  which  connected  it  by 
way  of  the  Celtic  corn  (Latin  cornu)  with  Cornwall,  is  perhaps, 
the  most  convincing.  Deutschbein's  assertion  (2,  18)  of  the  lack 
of  any  literary  connections  between  the  Anglo-Saxons  and  the 
people  of  Cornwall  was  disproved  by  the  evidence  concerning  the 
early  history  of  Cornwall  in  J.  Loth's,  Contributions  a  VEtude 
des  Romans  de  la  Table  Ronde  (Paris,  1912,  p.  65  ff.).  The 
Ur-Horn  may  indeed  have  been  influenced  by  tales  of  Scandi- 
navian-Irish wars,  but  it  seems  probable  that  it  was  in  the  form 
practically  of  a  lay  and  that  its  nucleus  was  romantic  rather 
than  historical. 

The  significant  motifs  and  details  in  King  Horn  are  for  the 
most  part  easily  defined.  The  Exposure  of  the  boy  hero  in  a 
rudderless  boat  recalls  that  of  the  boy  Sceaf,  of  Arthur's  ex- 
posure of  the  children  born  on  May  day,  and  that  of  many 
another  unfortunate  in  history  and  romance  (Hall,  pp.  102-03). 
The  lad's  arrival  at  the  court  of  the  Foreign  King  may  be 
matched  with  that  of  Beves  (Boje,  Beves,  p.  79  ff.).  The  ac- 
count of  the  instructions  given  for  his  Education,  the  description 
of  his  Dubbing,  the  gift  to  him  of  the  wonder-working  Ring,  are 
most  fully  paralleled  by  Hall  (pp.  108-09;  124-26;  129-30)  and 
by  Wissman  (Anglia  iv,  352-57).  Horn's  rescue  of  Rimenhild 
from  the  unwelcome  wooing  of  King  Modi  and  again  from  his 
own  false  friend,  Fikenild,  are  incidents  appearing  in  a  great  num- 
ber of  romantic  tales  which  have  been  studied  by  Child  (Ballads, 
1,  194)  and  by  Splettstosser  (Der  heimkehrende  Gatte,  Berlin, 

8  Nelles,  like  Heuser,  was  content  to  recognize  the  primarily  romantic 
character  of  Horn.  Many  other  writers,  even  those  most  insistent  on  the 
ancient  historical  origin  of  the  poem,  have  admitted  the  impress  of  the  age 
of  chivalry   on  King  Horn. 

Of  importance  for  the  further  study  of  names  in  Horn,  Havelok,  Beves, 
the  so-called  "  Viking  sagas,"  is  E.  Bjorkman's  Nordische  Personennamen 
in  England  in  alt.  u.  friihmittel.  Zeit,  Halle,  iqio.  Interesting,  too,  is  such  a 
study  as  H.  O.  Wyld's,  "  Old  Scandinavian  Personal  Names  in  England,"  MLR., 
v,  280-96  (iqio),  though  this  is  limited  to  Lancashire  names. 


KING   HORN  93 

1899).  In  these  the  regularly  recurrent  elements  are  the  long 
absence  of  the  lover,  his  sudden  return  and  appearance  under 
disguise  at  the  feast  celebrating  the  nuptials  of  his  love,  and 
his  revelation  of  himself  to  the  bride  by  means  of  a  ring  dropped 
into  a  wine-cup.  The  Disguise  as  a  minstrel  or  a  beggar  is  a 
device  which  may  have  borrowed  no  more  from  fiction  than 
from  life  itself.9  Of  special  interest  is  the  similarity  already 
noted  between  the  Horn  story  and  the  Gesta  Herewardi  (ed. 
T.  H.  Hardy,  Gaimar's  Estoire,  Rolls  Ser.  1888,  1,  349-53),  which 
tells  of  the  disguised  Hereward's  rescue  of  a  Cornish  princess  at 
her  marriage  feast.  Layamon's  Brut  (ed.  Madden,  11.  30,728- 
828)  tells  of  Brian,  disguised  as  a  pilgrim,  who  sought  his  sister 
at  the  court  of  Edwine  (Hartenstein,  pp.  137-38).  Episodes  of 
this  sort  seem  to  have  been  especially  popular  during  the  Cru- 
sades, and  in  this,  as  well  as  in  the  typical  animosity  to  "  hecfene 
houndes,"  the  Horn  legend  seems  to  show  the  influence  of 
crusading  times.  Like  other  heroes  in  such  "  cheerful  edifying 
romances  "  as  Beves  of  Hampton  and  Guy  of  Warwick,  Horn 
during  his  long  absence  remains  loyal  to  Rimenhild,  and  the 
second  lady,  who  is  either  offered  to  him  (KH)  or  is  won  by 
his  own  charm  (RH),  is  not  treated  as  a  psychological  problem 
but  simply  as  a  test  of  the  hero's  faith  (Schoepperle,  Tristan, 
1,  166-73). 

The  most  notable  folk-lore  elements  in  the  legend  are  the 
riddles  or  parables  which  appear  in  both  the  French  and  the 
Middle  English  versions.  In  King  Horn  the  disguised  hero  tells 
Rimenhild  he  has  come  back  to  see  if  his  net  is  still  as  he  left 
it,  his  net  meaning  Rimenhild  herself.  In  the  French  version 
Horn  meets  his  rival  Modin  and,  riddling  so  that  the  latter 
thinks  him  a  fool,  tells  of  a  net  for  which  he  has  now  returned. 
Later  to  Rimel  he  tells  the  parable  of  the  Hawk  which  he  hopes 
to  find  as  good  as  he  left  it.    Child  (Ballads,  1,  191)  noted  the 

9  No  one  seems  to  have  noted  in  this  connection  the  story  quoted  by 
Dugdale,  Monasticon  Anglicanum,  Lond.,  1830,  vi,  501,  from  the  Register  of 
Lacock  Abbey,  concerning  William  Talbot.  In  11 86  William  assumed  the 
guise  of  a  pilgrim  and  sought  for  Ela,  heiress  of  the  Earl  of  Salisbury,  who 
had  been  taken  to  Normandy  and  was  there  strictly  guarded.  Having  found 
the  right  castle,  William,  who  was  skilled  in  songs,  put  on  a  minstrel's  guise 
and  so  won  his  way  within.  He  succeeded  in  escaping  with  the  lady  and  in 
bringing  her  back  to  England,  where  she  was  given  in  marriage  to 
William  Longespee. 


94  MEDIAEVAL   ROMANCE    IN    ENGLAND 

similarity  of  the  French  parable  of  the  net  to  a  story  told  in  the 
Gesta  Romanorum  in  which,  after  exchanging  vows  of  fidelity 
with  an  emperor's  daughter,  a  soldier  goes  to  the  holy  land  for 
seven  years,  returns,  meets  a  king  who  is  a  rival  suitor,  and  tells 
the  same  net  story.  The  Hawk  parable  is  found  in  the  romance 
of  Jehan  et  Blonde  (ed.  Suchier  n,  1.  2821),  written  between 
1270  and  1280  by  Philippe  de  Remi,  Sire  de  Beaumanoir.  In 
the  fifteenth  century  a  revised  version  was  known  as  Romant  de 
Jehan  de  Paris  (Hartenstein,  pp.  138  ff.).  Except  for  these 
tales  and  for  variants  of  Apollonius,  riddles  are  of  comparatively 
rare  occurrence  in  romance. 

The  century  and  more  of  criticism  that  has  been  devoted  to 
King  Horn  has  established  few  facts,  though  it  has  brought  forth 
stimulating  discussion  of  the  verse  form  and  topography  of  the 
poem  and  the  relationship  of  the  versions.  At  present  the  ten- 
dency is  to  believe  that  the  Horn  legend  was  possibly  influenced 
in  the  course  of  its  development  by  Anglo-Danish  affairs  but 
that  its  origin  was  in  romantic  rather  than  historical  tradition, 
and  that  the  early-seeming  Middle  English  version  is  actually 
later  than  the  elaborated  Anglo-Norman  romance  by  Mestre 
Thomas.  Though  in  this  last  version  lies  probably  the  most 
fruitful  field  for  future  research,  in  King  Horn  itself  lives  one  of 
the  earliest  and  most  virile  of  Middle  English  romances. 


BIBLIOGRAPHY 

Texts:  (i)  H,  Harley  2253  (1546  lines),  ed.  Ritson,  1802,  11,  91-155; 
(2)  O,  Laud  Misc.  108,  Bodleian,  desc.  Archiv.  xlix,  395-414;  ed. 
Horstmann,  Herrig's  Archiv.  l,  39-58  (1872) ;  (3)  Univ.  Libr.  Cbg.,  Gg. 
rv,  27,  2  (1530  lines),  ed.  Michel,  Horn  et  Rimenhild,  Paris,  1845; 
J.  R.  Lumby,  EETS.  xiv,  1866;  reed.  G.  McKnight,  1901;  R.  Morris, 
Specimens,  1887,  1,  237-86;  Maetzner,  Alteng.  Sprachproben,  1867,  1. 
209  ff.  A  critical  edition  was  published  by  T.  Wissman,  Quellen  u.  Forsch. 
xlv  (1881).  The  three  MSS.  were  printed  by  McKnight,  op.  cit.  and  by 
Jos.  Hall,  Oxford,  1901.  See  Hall,  p.  xv,  for  reviews  of  Wissman;  Wells, 
p.  762,  for  reviews  of  Hall.  An  edition  by  L.  Morsbach  of  Gottingen 
was  announced  in  1902.  Trans.  L.  A.  Hibbard,  Three  Middle  Eng. 
Romances,  191 1;  J.  Weston,  Chief.  Middle  Eng.  Poets,  p.  93  ff. ;  H.  Lin- 
demann.  King  Horn,  eine  mittelengl.  Romane  aus  dem  13  Jhdt.  ins. 
deutsche  iibertragen,  Koln.    1904. 


KING   HORN  95 

Studies:  Billings,  Guide,  pp.  23-24;  Hartenstein  (see  below),  pp. 
3-14;  Wells,  Manual,  p.  762;  West  (see  below)  for  verse  studies, 
pp.  xi-xiv. 

Azzalino,  W.  Die  Wortstellung  im  King  Horn.  Diss.  194  pp.  Halle,  191 5. 

Braunholtz,  E.  G.  "  Cambridge  Fragments  of  the  Anglo-Norman  Roman 
de  Horn,"  MLR.,  xvi,  23-33  (l921)- 

Breier,  W.  "  Zur  Lokalisierung  des  King  Horn,"  Eng.  Stud,  xui,  307- 
09  (1910). 

Dahms,  0.  Der  Formenbau  des  Nomens  u.  Verbums  in  dent  anglo- 
normanischen  Gedichte  "  Das  Lied  von  wackern  ritter  Horn."  Diss. 
87  pp.  Kiel,  1906. 

Deutschbein,  M.  (1)  Studien  zur  Sagengeschichte  Englands  (Horn, 
Havelok,  Tristan,  Boeve,  Guy  of  Warwick).  264  pp.  Cothen,  1906. 
Rev.  Liter ar.  Zentralblatt,  1906,  sp.  1276;  Deut.  Liter aturzeit.  1906, 
sp.  1578;  Liter aturblatt  j.  ger.  u.  rom.  Phil.  1907,  sp.  280;  MLR., 
1907,  p.  176;  Anglia  Bl.  xvm,  1-13;  (2)  "  Beitrage  z.  Horn  und 
Havelocsage."  Anglia  Bl.  xx,  16-24;  (3)  "  Zur  historischen  Horn- 
sage  (=Horn  B),"  ibid.  pp.  55-59  (1910). 

Grass,  P.  Horn  u.  Hilde  in  ihrer  Stellung  z.  germ.  Sagengeschichte.  Diss. 
51  pp.  Minister,  Borna-Leipzig,  191 1. 

Hall,  J.  See  Texts. 

Hartenstein,  0.  Studien  zur  Horn  Sage,  ein  Beitrag  z.  Litteraturgeschichte 
des  Mittelalters.     (Kieler  Stud.  z.  engl.  Phil,  iv.)  Heidelberg,  1902. 

Heuser,  W.  "  Horn  u.  Rigmel  (Rimenhild) ;  eine  Namenuntersuchung," 
Anglia  xxxi,  105-31  (1908). 

Luick,  K.  "  Berichtigung  zu  King  Horn,"  Anglia  Beiblatt  xm,  332-33 
(1902). 

MacSweeney,  J.  J.  "Hind  Horn,"  MLR.  xrv,  210-11  (1919). 

Mather,  F.  J.  "  King  Ponthus  and  the  Fair  Sidone,"  PMLA.  xn,  1-150 
(1897).    Rev.  G.  Paris,  Rom.  xxvi,  468-70. 

McKnight,  G.  H.  "  Germanic  Elements  in  the  Story  of  Horn,"  PMLA. 
xv,  221-32  (1900).    See  also  Texts. 

Morsbach,  L.  Die  angebliche  Originalitat  des  fruhmitteleng.  King  Horn," 
Beitr.  z.  rom.  u.  engl.  Phil.,  Festgabe  f.  Foerster,  p.  297  ff.  Halle, 
1902. 

Nelles,  W.  C.  "The  Ballad  of  Hind  Horn,"  Jour,  of  American  Folk- 
Lore,  xxn,  42-63  (1909). 

Northup,  C.  S.  "  King  Horn,  Recant  Texts  and  Studies  "  (with  a  com- 
parative table  of  line  numbers  for  nine  editions),  Jour.  Ger.  Phil. 
tv,  529-41   (1902). 

Schofield,  W.  H.  "The  Story  of  King  Horn  and  Rimenhild/'  PMLA. 
xvm,  1-84  (1903). 

Tamson,  G.  "  A  Passage  in  the  Middle  Eng.  King  Horn,"  v.  701-04, 
Anglia  xix,  460  (1896-97). 

Vising,  J.  Studier  i  den  franska  romanen  om  Horn.  I.  Goteborg,  1903; 
11  (1904),  Prolegomena  zu  einer  edition  des  roman  von  Horn  (1905). 


96  MEDIAEVAL   ROMANCE    IN   ENGLAND 

West,  H.  S.  The  Versification  of  King  Horn.    Diss.  92  pp.  Baltimore, 

1907. 
Wissman,  T.  (1)  Untersuchungen  zur  ME.  Sprach-  u.  Litt.  Geschichte, 

Strassburg,  1876;  Quellen  u.  Forsch.  xvi.     (Cf.  Wells,  p.  762,  for 

reviews.)   (2)  "  Studien  zu  Horn,"  Anglia  iv,  342-400  (1881).     (3) 

See  also  Wissman's  editions  of  Horn,  Texts. 
Wiist,  P.  Die  deut.  Prosaromane  von  Pontus  v.  Sidonia.  Diss.  Marburg, 

1903. 


HORN    CHILDE 

Versions.  Chaucer's  reference  in  Sir  Thopas  to  Hornchilde 
has  given  it  a  bad  eminence,  justified,  so  most  critics  have  felt, 
by  the  "  degenerate  minstrelsy  "  of  the  poem  itself.1  It  con- 
sists of  1 136  lines  in  the  tail-rime  stanza  (aab  aab  ccb  ddb),  and 
the  single  extant  text  is  in  the  Auchinleck  Manuscript  (1330- 
40).  The  poem  therefore  must  have  been  composed  before  1340, 
and  its  borrowings  from  Sir  Tristrem  (ed.  Kolbing,  p.  lxiv),  to 
which  it  alludes  by  name  (1.  311),  show  that  it  was  written  after 
1290  (Hartenstein,  p.  78).  The  author,  who  refers  to  his 
"  boke  "  (1.  277)  and  who  makes  Horn  study  both  "  harpe  and 
romance,"  was  familiar  with  the  stock  phrasings  of  Middle  Eng- 
lish diction.  Caro  (pp.  347-50)  noted  a  large  number  of  these 
and  indicated  also  the  extensive  use  in  the  poem  of  assonance 
and  alliteration.  At  times  the  poet  rather  helplessly  repeats  his 
own  expressions,  and  there  is  little  variety  or  vigor  in  his  style. 
In  general  critics  assign  his  home  to  the  Northern  district,  but 
Caro  (p.  342),  who  alone  has  made  any  serious  study  of  lin- 
guistic evidence,  assigned  it  to  the  southern  part  of  North 
England,  near  the  East  Midland  boundaries   (Hartenstein,  pp. 

78-79). 

As  early  as  Bishop  Percy's  time  Horn  Childe  (HC)  was  recog- 
nized as  "  an  altered  and  somewhat  modernized  version  of  King 
Horn,"  or,  more  accurately,  of  the  Horn  legend  (Hartenstein, 
p.  77).  Its  story  of  Horn's  adventures  is  essentially  the  same  as 
that  in  the  earlier  versions,  though  it  is  varied  by  some  individual 
touches.  Like  so  many  of  the  lesser  poets  of  his  time,  the 
author  has  an  unhappy  genius  for  the  concrete  and  trivial  at  the 
expense  of  the  imaginative  effect.  Horn,  he  tells  us,  discovers 
Rimnhild,  whom  her  father  has  cruelly  beaten,  "  liggeand  on  hir 
bede,  Mou5e  and  nose  al  forbled;"  Horn  himself  on  another 
occasion  stays  at  home  for  a  blood-letting.     The  Irish   king 

1  Cf.  Hartenstein,  p.  85;   Schofield,  p.  66;   Wissman,   Untersuch.  pp.  94- 
100;  Ward,  Catalogue  1,  458-60. 

97 


98  MEDIEVAL   ROMANCE    IN   ENGLAND 

Finlac,  wounded  so  that  the  blood  "  ran  ouer  his  eise  "  orders 
that  his  daughter  "  schuld  a  plaster  ta."  The  poet  goes  to  excess 
in  localizing  the  scenes  of  his  story;  the  Danes  invade  North- 
umberland at  "  Clifland  bi  Tiseside,"  fight  with  Horn's  father  at 
"  Alertonmore,"  and  leave  their  bones  where  men  yet  see  them 
lie  "  bi  Seyn  Sibiles  kirke."  HaQeolf  hunts  on  "  Blakeowemore  " 
heath,  holds  feasts  at  Pikering  and  at  York,  fights  and  dies  at 
"  Stainesmore."  Horn  as  a  child  is  carried  far  south  into  Eng- 
land, and  later  as  an  exile,  goes  to  Snowdon  in  Wales.2  The 
same  excess  is  shown  in  the  poet's  interest  in  unimportant  char- 
acters such  as  Horn's  young  comrades,  whose  individual  fortunes 
are  unnecessarily  narrated.  Yet  with  all  this  detail  the  poem  is 
not  prolix,  and  makes  a  real  attempt  to  combine  local  color  and 
patriotic  feeling  with  true  romantic  spirit.  One  scene  at  least 
is  pleasantly  told:  Rimnild  in  her  bower  breaks  a  pomegran- 
ate and  offers  Horn  true  knightly  gifts,  a  goshawk,  greyhounds, 
a  black  steed,  an  ivory  horn  with  silk  and  golden  baldric,  and 
Weland's  sword,  "  Bitterfer."  At  parting  she  gives  Horn  a  ring 
which  will  show  him  by  its  change  of  color  how  she  keeps  faith. 
He  in  return  bids  her  watch  a  well,  overgrown  with  ivy,  which 
will,  as  long  as  it  is  shadowless,  be  symbolic  of  his  truth.3  In 
the  typical  vein  of  popular  romance  too  are  the  beggar  who 
brings  Horn  word  of  his  lady's  danger,  and  the  surly  porter 
whose  bones  Horn  so  cheerfully  breaks.  In  somewhat  courtly 
fashion  the  poet  amplifies  the  king's  speech  in  regard  to  the 
fealty  of  Horn's  comrades  or  the  account  of  Horn's  joust  in  the 
wood  or  of  the  tournament  in  Wales. 

No  other  version  of  Horn  Childe  is  known,  though  the  Scotch 
ballads,  already  discussed  in  connection  with  King  Horn,  were 
certainly  in  some  way  related  to  the  poem. 

Origin.  The  first  two  hundred  and  fifty  lines  of  Horn  Childe, 
according  to  Schofield  (p.  75),  embody  genuine  historical  tradi- 
tions not  connected  with  the  Horn  legend  until  the  composition 
of  this  particular  poem.     Its  remaining  portion  has  a  large 

2  For  proposed  identifications  of  these  places  see  Deutschbein,  p.  90; 
Hartenstein,  pp.  82-3;  D.  Haigh,  Anglo-Saxon  Sagas,  Lond.  1891,  pp. 
62-70. 

3  For  similar  tests  of  faith  or  of  chastity  see  Percy  Folio  MS.  ed.  Furn- 
ivall,  1868,  11,  300-04;  Child,  Ballads,  v.  Index. 


HORN    CHILDE  99 

number  of  important  traits  which  connect  it  with  the  Anglo- 
Norman  version  of  Horn.  Many  names  are  the  same  for  both ; 
alike  too  are  the  scenes  in  which  Rimnild  gives  her  gifts  and 
that  in  which  Horn,  disguised  as  a  beggar,  meets  his  rival,  King 
Modiun  and  "  riddles  "  him  the  parable  of  the  net.  Whatever 
the  minor  changes  in  Horn  Childe,  these  parables  with  Horn  et 
Rimenhild  make  indisputable  its  author's  knowledge  of  the 
French  text.4  With  King  Horn,  on  the  other  hand,  there  are  few 
resemblances  in  diction  and  none  in  plot  detail  not  also  to 
be  found  in  the  French  text.  The  features  listed  by  McKnight 
as  peculiar  to  HC,5  may,  with  two  exceptions,  be  regarded  as 
due  to  omission.  In  an  abridgment  such  as  this,  therefore,  they 
prove  nothing  in  regard  to  source  material. 

The  two  matters  in  which  Horn  Childe  most  differs  from  the 
French  version  are  its  transformation  of  the  scene  of  the  story 
and  its  historical  introduction.  This  deals  first  with  a  victory  of 
King  Hacleolf  over  Danish  raiders,  and  next  with  his  defeat  and 
death  at  the  hands  of  King  Malkan  and  Irish  foes.  The  poet 
was  evidently  familiar  with  the  territory  the  traditions  of  which 
he  wished  to  use.  Schofield  (p.  68)  noted  that  most  of  the 
places  mentioned  in  HC  belong  to  the  North  Riding  of  York- 
shire, and  identified  Horn's  father,  King  Hacfeolf,  with  Earl 
Eadulf  (c.  966)  of  Northumberland;  Malkan  with  King  Malcolm 
I  or  II  of  Scotland;  Thorbrand,  who  usurped  the  land  after 
HaQeolf  's  death,  with  the  Thorbrand,  who,  in  a  pseudo-historical 
tract  ascribed  to  Simeon  of  Durham  (Ed.  T.  Arnold,  Rolls  Ser. 
1882, 1,  215),  was  killed  by  Aldred,  son  of  Uchtred,  the  valorous 
successor  of  Eadulf.  Details  from  different  texts  of  wholly  dif- 
ferent date  and  origin,  have  to  be  fitted  together  to  make  these 
identifications  possible.  Schofield  (p.  72)  thought  that  the  poet 
derived  his  details  not  entirely  from  oral  traditions  but  from 
Anglo-Saxon  poems  of  the  same  type  as  the  Battle  of  Maldon 
(991,  Anglo-Saxon  Chronicle).  Just  what  use  a  fourteenth- 
century  poet  could  have  made  of  these  songs  is  not  clear. 
Deutschbein  accepted  Schofield's  placing  of  the  story  but  pointed 

4  Cf.   Hartenstein,   p.    118;    McKnight,   p.   xv,   for   variations   of    critical 
opinion. 

5  McKnight,  Horn,  p.  xiv,  mistakenly  listed  as  peculiar  to  HC  the  scene 
of  Rimmeld's  gift-giving  which  is  directly  based  on  that  in  RH. 


100  MEDIEVAL   ROMANCE    IN   ENGLAND 

out  that  the  tract  ascribed  to  Simeon,  which  tells  at  length  of  the 
raid  of  Malcolm  II  and  the  siege  of  Durham,  describes  an  Eng- 
lish victory  under  Uchtred,  whereas  Horn  Childe  tells  of  the 
English  defeat  in  which  Hacteolf  was  killed.  Deutschbein  (p.  91) 
found  in  Fordun's  late  fourteenth-century  Chronica  Gentis 
Scotorum  (ed.  Skene,  Historians  of  Scotland,  Edin.  1871-72,  1, 
182)  a  more  satisfactory  reference  to  a  defeat  inflicted  on  the 
English  by  Malcolm  II  in  the  neighborhood  of  Stanmore.  As 
for  the  account  in  Horn  Childe  in  which  Hacfeolf  is  first  repre- 
sented as  conquering  the  Danish  invaders,  an  episode  which 
Schofield  (p.  68)  had  somewhat  tentatively  connected  with  the 
thirteenth-century  saga  of  Ola]  Tryggvason  (trans.  Sephton, 
Lond.  1895,  cn-  64) ,  Deutschbein  (p.  93)  found  its  historical 
antecedent  in  the  account  of  a  victory  achieved  by  Athelred  in 
the  neighborhood  of  Alverton  (=AUerton)  and  celebrated  by  a 
feast  in  York.  Although  Langtoft,  who  recorded  this  episode 
{Chronicle  ed.  T.  Wright,  Rolls  Ser.  1866-68,  1,  310),  himself 
ascribed  it  to  the  reign  of  Athelred  I  (866-71),  Deutschbein 
believed  that  it  should  be  dated  in  the  reign  of  Athelred  II  (978— 
1016).  It  is  no  wonder  that  by  this  method  "  all  difficulties 
disappear." 

The  modicum  of  fact  so  far  established  is  simply  that  the 
introductory  portion  of  Horn  Childe  does  not  refer  to  a  special 
locality  and  does  contain  apparent  reminiscences  of  Danish  and 
Irish  raids.  But  it  seems  altogether  improbable  that  the  author 
drew  his  material  from  oral  tradition,  from  Anglo-Saxon  songs, 
or  from  any  of  the  texts  yet  cited.  He  was,  as  the  rest  of  the 
poem  shows,  of  wholly  unoriginal  mind  and  art,  and  it  is  safe 
to  say  that  this  puzzling  introduction  had  some  single  French 
source  as  easy   to   abridge   as   was   the   French  Horn   itself.6 

A  final  word  may  be  said  about  the  evidence  that  there  was 

6  Hartenstein  (p.  121)  believed  that  there  was  a  northern  Volkssage  of 
the  Horn  story,  that  the  author  of  HC  used  it  and  also,  perhaps,  the  lost 
"  Chanson  Aalof."  Schofield,  pp.  74-75,  noted  the  lack  of  any  evidence  before 
HC  which  would  connect  the  Hafeolf  story  in  HC  with  the  Horn  legend. 
The  Aalof  story,  as  Thomas  tells  it,  is  as  courtly  and  sophisticated 
as  HR  itself.  This  the  HC  poet  might  have  abridged  as  he  did  HR, 
with  change  of  the  original  names  and  places.  He  might  also  in  a  twelfth-  or 
thirteenth -century  French  chronicle  have  found  some  account  of  Ha3eolf. 
But  in  any  case  the  immediate  source  used  by  this  poet  writing  about  1300 
cannot  have  been  of  very  "  primitive  "  character. 


HORN    CHILDE  IOi 

once  a  "  saga  "  of  some  sort  about  Horn's  father.  The  oldest 
extant  text  of  the  Horn  legend,  Horn  et  Ritnenhild,  refers  ex- 
tensively to  Aaluf  (Aalof).  He  was  born  in  a  forest,  the  son  of 
an  unknown  father  and  of  Goldeburc,  daughter  of  Baderolf, 
emperor  of  Germany.  He  was  brought  as  a  foundling  to  King 
Silaf  (Silaus),  happily  reared,  and  became  distinguished  for 
his  feats  of  prowess.  He  was  slandered  by  the  traitor  Deneray, 
but  nevertheless  received  for  his  wife  the  Princess  Samburc,  and 
later  became  king.  During  his  ten  years'  rule  he  became  the 
father  of  Horn,  and  the  "  cumpagnun  "  and  sworn  friend  of  King 
Gudereche  who  was  later  to  welcome  Horn  and  tell  him  some- 
thing of  his  father's  history.  After  a  heroic  fight  with  invaders 
Aalof  was  killed  (Hartenstein,  p.  60). 

It  has  always  been  a  question  whether  the  story  of  Aalof  was 
of  independent  origin  and  attached  to  the  Horn  legend  by  Mestre 
Thomas,  or  whether  he  simply  elaborated  hints  found  in  his 
source  story.  Schofield  (p.  56-58)  favored  the  first  theory, 
since  the  names  in  the  Aalof  story  are  predominantly  Germanic 
and  not  of  Norse  origin.  He  thought  them  as  foreign  to  the 
primitive  story  as  would  have  been  the  adventures  of  Hadermod, 
Horn's  son,  which  Thomas  said  his  own  son  Gilimot  would 
record.  Thomas's  reference  (1.  192)  to  his  escrit  would  seem  to 
imply  an  earlier  written  account  of  Aalof,  for  which  Schofield 
found  evidence  in  a  passage  of  the  French  Waldef.  The  argu- 
ment seems  hardly  to  take  into  account  the  fact  that  the  Aalof 
story  in  Horn  Ckilde,  so  far  as  Hartenstein  (p.  60)  and  others 
have  been  able  to  reconstruct  it,  is  not  merely  a  "  suitable  "  but 
an  almost  exact  counterpart  of  the  Horn  story,  and  that  the 
device  of  duplication,  not  merely  of  climax  but  of  theme,  which 
so  often  appears  in  mediaeval  fiction,  could  fully  account  for  it. 
By  the  time  when  Mestre  Thomas  wrote,  it  was  an  established 
custom  to  write  romances  dealing  with  fathers,  sons,  and  even 
grandsons.7    Thomas  himself  intended  such  a  triology  in  writing 

7  Schofield  (p.  58)  referred  to  the  stories  of  Galahad,  son  of  Lancelot,  of 
Lohengrin,  son  of  Parzival.  It  may  be  noted  that  nearly  all  of  the  thirteenth- 
century  romances  which  introduce  the  enfance  theme  develop  the  history  of 
the  hero's  parents  almost  as  fully  as  that  of  the  hero's  son.  This  is  true 
not  only  of  the  prose  Tristan  and  of  the  prose  Lancelot,  but  also  of  the 
French  and  Latin  romances  about  the  enfance  of  Gawain.  Cf.  Meyer,  "  En- 
fances  Gauvain,"  Rom.  xxxix,  1-32  (1910) ;  J.  D.  Bruce,  "  De  Ortu  Wal- 
uuani,"  Hesperia,  1013. 


102  MEDIEVAL   ROMANCE    IN   ENGLAND 

of  Aalof,  Horn,  and  Hadermod.  Thomas  may  have  had  his 
earlier  written  account  of  Aalof,  but  the  reference  in  Waldef8 
is  doubtful  confirmation,  since  its  only  trustworthy  statement  is 
that  the  current  French  romances  of  Tristan,  Waldef,  and  Aalof, 
were  "  molt  amees."  In  the  absence  of  any  other  known  French 
version  of  the  Aalof  story,  the  Waldef  passage  may  as  well  refer 
to  Thomas's  own  account  as  to  anything  else.  So  voluminous  a 
romancer  was  Thomas  that  mere  references  to  Horn's  father  in 
any  antecedent  text  might  well  have  been  elaborated  by  him 
into  the  story  which  he  tells.  He  was  amply  familiar  with  con- 
temporary romance;9  he  had  in  the  Horn  story  itself  a  good 
model  to  imitate,  and  his  Aalof  story  must  not,  therefore,  despite 
its  Germanic  names,  be  taken  as  serious  witness  either  to 
antiquity  or  to  reality. 


BIBLIOGRAPHY 

Texts:  (i)  Auchinleck  MS.  f.  317-22,  ed.  Ritson  1810,  111,  282-320; 
Michel,  Horn  et  Rimenhild,  1845,  pp.  341-89;  J.  Caro,  Eng.  Stud,  xn, 
323-66  (1888-89);  J.  Hall,  King  Horn,  Oxford,  1901,  pp.  179-92. 

Studies.     Cf.   Billings,  Guide,  pp.   11-12;   Hartenstein   (see  below); 

Wells,  Manual,  p.  763. 
For  references  to  Deutschbein,  Hartenstein,  McKnight,  Nelles,  Scho- 
field,  see  Horn  bibliography. 

8  Cf.  Hartenstein,  p.  no;  Schofield,  p.  50.  For  the  fifteenth-century 
Latin  translation  by  John  Bremis,  the  monk  of  Thetford,  see  Historia  Regis 
Waldef,  ed.  R.  Imelmann,  Bonner  Stud.  z.  Eng.  Phil,  rv,  1912;  "Vom  ro- 
mantischen  u.  geschichtlichen  Waldef,"  Eng.  Stud,  liii,  362-39   (1919). 

9  Heuser  (p.  113,  n.)  referred  to  the  likeness  of  two  of  the  women's 
names  to  those  in  the  Havelok  legend.  Deutschbein  (pp.  254-62)  discussed 
the  relations  of  "Sage  u.  Lit.  Deutschlands  u.  Englands  im  n,  12,  u.  13 
Jahrhundert."  See  also  Langlois,  Table  des  Noms  Propres  dans  les  chansons 
de  Geste. 


HAVELOK   THE    DANE 

Versions.  The  legend  of  Havelok  tells  of  a  prince  become 
pauper.  When  his  father  is  killed  and  his  kingdom  lost,  the  boy 
Havelok  is  taken  to  England  where  he  becomes  the  sturdy 
scullion  of  an  English  lord  and  is  married  perforce  to  an  English 
princess.  Her  wicked  guardian,  who  is  Havelok's  lord,  hopes 
thus  to  fulfill  his  oath  to  wed  her  to  the  strongest  man  about  and 
also,  by  this  seeming  mesalliance,  to  degrade  the  maiden.  After 
many  adventures  Havelok  regains  his  own  heritage  and  his  wife's 
and  dies  as  king  of  England  and  of  Denmark. 

Of  this  legend  there  are  four  principal  versions  (Sisam,  p. 
xi-xx).  The  earliest,  an  episode  introduced  in  Gaimar's  Estorie 
des  Engles,  11.  41-818  (ed.  Hardy  and  Martin,  Rolls  Ser.  1888, 
Part  1,  1-34),  was  written  between  1145  and  1151  and  is  pre- 
served in  four  manuscripts  (Heyman,  p.  139;  Gross,  Gaimar ; 
die  komposition  seiner  Reimchronik,  Erlangen,  1902);  the 
second,  a  French  metrical  romance  of  eleven  hundred  lines,  the 
Lai  d'Havelok  (ed.  Hardy  and  Martin,  op.  cit.  1,  290  ff.)  was 
composed  late  in  the  last  half  of  the  twelfth  century  (Bell,  p. 
23) ;  the  third  is  the  Middle  English  romance  known  as  Havelok 
the  Dane ;  and  the  fourth  is  a  summary  of  eighty-two  lines  (pr. 
Sisam,  p.  xvii)  interpolated  before  1400  in  the  Lambeth  copy  of 
the  translation  of  the  chronicle  of  Pierre  de  Langtoft.  This 
translation  was  made  by  Robert  Manning  of  Brunne  who  him- 
self interpolated  a  passage  which  shows  that  in  1338,  when  he 
finished  his  work,  he  was  familiar  with  a  Middle  English  version 
of  Havelok  (Sisam,  p.  xvi). 

Short  minor  versions  of  the  story  are  found  in  some  of  the 
later  chronicles:  in  the  Petit  Brut  d'Angleterre,  written,  accord- 
ing to  the  prologue,  in  13 10  by  Raouf  de  Boun(e)  of  Lincoln- 
shire; in  the  anonymous  Brut  in  fourteenth-century  prose  which 
was  later  translated  into  English  and  printed  in  part  by  Caxton 
in  1480 ;  in  the  Scala  Chronica  x  written  in  French  prose  about 

1  Selections  from  the  two  Bruts  are  given  by  Skeat,  1902.  For  the  Scala 
Chronica,  see  Heyman,  p.  116. 

103 


I04  MEDIEVAL   ROMANCE    IN    ENGLAND 

1335  by  Sir  Thomas  Grey  of  Northumberland;  in  the  Eulogium 
Historiarum  (Rolls  Ser.  1858-63)  of  the  late  fourteenth  century; 
and  in  the  Chronicon  (Rolls  Ser.  1889-95)  written  by  Henry 
Knighton  in  the  same  century  (Hey man,  pp.  109-22).  These 
are  unimportant  in  determining  the  relationship  of  the  earlier 
versions  but  suggest  some  interesting  connections  with  other 
legends.  Heyman  (p.  112)  noted  that  in  Raouf's  Brut,  Guy  of 
Warwick  is  connected  with  Havelok's  son,  and  that  Knighton 
tells  Guy's  story  immediately  after  Havelok's.  Both  Raouf's 
Brut  and  the  anonymous  Brut  in  French  prose  (cf.  Brie,  p. 
362)  give  Havelok's  father  the  name  "  Birkebeyn  "  in  agreement 
with  the  Middle  English  romance.  The  nickname  "  Bircke- 
beinar  "  (Birch-legged  fellows),  which  first  appeared  in  England 
in  Roger  of  Hoveden's  Chronica,  was  Norse  in  origin  and  was 
originally  given  in  Norway  to  a  group  of  outlaws  who  succeeded 
in  1 1 84  in  making  Sverre  Sigurdson  their  king.  In  Roger's 
chronicle,  which  was  written  after  1192,  this  king  was  referred  to 
as  Swerus  Birkebein.2 

The  relationship  of  the  extant  French  versions  was  first  seri- 
ously studied  in  1880  by  Kupferschmidt  {Roman.  Stud,  iv,  411- 
30).  He  believed  that  a  lost  French  version  in  verse  was  the 
common  source  of  Gaimar  and  the  French  Lai.  Putnam  in 
1900  accepted  these  results  with  one  exception.  He  urged  that 
the  Lambeth  Interpolation  was  not  derived  from  Gaimar,  as 
Kupferschmidt  thought,  but  was  an  independent  summary  of 
the  lost  version.  In  the  main  these  views  prevailed  until  Dr. 
Fahnestock's  reconsideration  of  the  whole  problem  in  191 5.  She 
derived  the  Lai  from  Gaimar,  and  her  conclusion  was  confirmed 
by  Bell's  independent  studies  in  1923.  After  a  comprehensive 
sketch  of  the  history  of  the  question,  Dr.  Fahnestock  enumerated 
in  parallel  columns  the  lines  in  Gaimar's  account  and  in  the  Lai 
that  are  in  close  resemblance,  and  found  (p.  105)  that  one 
hundred  and  seventy  lines  are  identical.  The  fact  that  in  several 
instances  these  lines  occur  in  large  groups,  suggests  that  passages 
in  the  later  version  were  taken  bodily  from  Gaimar's  text.    To 

2  Heyman,  p.  87.  Were  Raouf  de  Bourfs  text  of  more  reliable  character 
(according  to  Heyman,  p.  104,  it  is  found  in  only  one  seventeenth-century 
manuscript,  Harley  902),  it  would  establish  the  fact  that  the  English  Havelok 
was  in  any  case  composed  before  13 10,  for  Raouf  took  this  name  Birkenbayne 
from  the  English  romance. 


HAVELOK    THE    DANE  IOj 

the  romancer's  conscious  imitation  besides  of  the  Lais  of  Marie 
de  France,  Dr.  Fahnestock  (pp.  115-36)  traced  such  evidences 
of  characteristic  technique  as  the  prologue  and  epilogue  in  the 
Lai,  its  references  to  Arthur  and  to  the  Bretons,  who  are  said 
to  have  made  a  lay  about  Havelok,  and  its  introduction  of 
scenes  and  characters  which  belong  to  a  courtly  environment. 
Notable  among  these  personages  is  Grim,  Havelok's  friend  and 
rescuer,  who  is  represented  as  a  lordly  baron.  In  few  instances 
can  the  process  by  which  an  estoire  developed  into  romance  be 
better  studied  than  in  these  two  texts.  As  far  as  the  Lai  is 
concerned,  no  Ur-Havelok  earlier  than  Gaimar's  version  need  be 
supposed. 

The  Middle  English  romance  of  Havelok  is  preserved  in  four 
brief  fragments  of  a  late  and  corrupt  text  discovered  by  Skeat 
in  191 1  {MLR.  vi)  and  in  Laud  Misc.  108,  the  early  fourteenth- 
century  manuscript  which  also  contains  King  Horn.  In  regard 
to  the  date  of  Havelok,  Hales  (Athenaeum,  Feb.  1889)  argued 
that  the  reference  (1.  139)  to  "  Rokesburw  "  as  the  northern 
frontier  proved  the  poem  after  1296,  the  year  in  which,  as  he 
supposed,  Roxburgh  first  became  a  border  fortress.  He  believed 
also  that  the  reference  (1.  11 78)  to  a  Parliament  at  Lincoln  must 
be  after  130 1,  the  year  in  which  the  first  Parliament  was  held 
there.  But  it  is  now  known  that  Roxburgh  was  in  the  hands  of 
the  English  as  early  as  11 74  (Deutschbein,  p.  159),  and  a 
Parliament  was  held  at  Lincoln  in  1226  (Van  der  Gaaf,  p.  319). 
Skeat  (1902,  notes,  11.  679,  819)  thought  the  poem  was  written 
before  1303,  asserting  that  he  caught  echoes  in  it  from  Robert 
Manning's  Handlyng  Synne.  The  two  parallel  passages  cited  by 
him  seem,  however,  somewhat  meagre  evidence.  Both  Sisam 
(p.  xxiv)  and  Schmidt  (pp.  89-97)  quoted  with  approval  Skeat's 
argument  that  the  extensive  use  of  final  e  in  Havelok,  contrasted 
with  the  more  limited  use  in  Handlyng  Synne,  proves  the  earlier 
date  of  the  romance.  But  it  must  be  remembered  that  the  earli- 
est manuscript  of  the  Handlyng  Synne  is  later  than  Laud  108, 
and  that,  as  far  as  we  know,  Robert's  work  may  have  been  only 
begun,  not  necessarily  finished,  in  1303.  The  dialect  of  the 
romance  is  that  of  Lincolnshire  where  the  story  itself  is  localized 
(Schmidt,  p.  80  ff.).  On  the  evidence  of  rime  and  verb  inflection 
Hupe  (Anglia,  xm,  193)   argued  for  Norfolk  as  the  home  of 


106  MEDIEVAL   ROMANCE   IN   ENGLAND 

the  poet.  The  3001  lines  of  Havelok  are  for  the  most  part  rimed 
in  short  couplets  in  clear  imitation  of  French  models,  but  occa- 
sional passages  occur,  such  as  11.  87-105,  which  have  one  rime 
only  (Sisam,  p.  xxvii-xxxix). 

Havelok  begins  with  an  appeal  for  a  "  cuppe  of  god  ale,"  and 
the  many  personal  comments  of  the  author,  his  quaint  proverbs, 
his  hearty  curses,  his  homely  figures  of  speech,  the  easy  casual 
swing  of  his  verse,  suggest  that  he  was  a  minstrel  poet  of  the 
same  sort  as  the  one  who  probably  once  packed  into  his  saddle- 
bags the  small  compact  manuscript  which  still  preserves  the 
poem.  Havelok  was  certainly  meant  for  minstrel  recitation,  not 
for  reading,  and  for  an  altogether  popular  audience.  The  author 
knows  and  likes  plain  names  such  as  Griffin  Galle,  William 
Wendut,  and  Huwe  Rauen,  and  enjoys  scenes  in  kitchen  or 
cottage  or  tavern.  He  describes  with  gusto  the  hard  ways  of  a 
fisherman's  life,  and  of  a  Lincoln  laborer  such  as  Havelok  be- 
came. In  such  passages  the  poet  amplified  his  French  source; 
two  detailed  descriptions,  for  instance,  one  of  forty-five  lines  and 
one  of  two  hundred,  correspond  to  passages  in  the  French  version 
of  thirteen  and  fifty-six  lines  (Creek,  p.  205).  The  English 
Havelok,  who  is  utterly  unlike  the  conventional  heroes  of  ro- 
mance except  in  his  prowess  and  the  physical  marks  of  his  royal 
origin,  is  a  hard-working,  simple-minded  person  who  is  frankly 
horrified  at  marriage  with  a  wife  whom  he  is  too  poor  to  support. 
Barefoot  he  goes  to  Lincoln,  clad  in  the  rough  coat  which  Grim 
the  fisherman  has  fashioned  from  a  sail.  Rarely  is  romance  so 
frankly  realistic  as  is  Havelok,  so  familiar  in  its  portrayal  of 
humble  life  and  of  actual  environment.  Because  these  traits  are 
distinctive  and  consistent,  Creek  rightly  refused  to  accept  the 
old  belief  that  they  were  all  due  to  popular  tradition.  He 
insisted  that  one  feels  in  Havelok  an  author  who  had  a  definite 
attitude  toward  his  story,  who  deliberately  changed  what  he 
wanted  to  change  in  his  French  source,  and  who  is  recognizable 
as  a  "  clear-eyed,  kindly  practical  man,  interested  in  common 
people,  thoroughly  patriotic,  and  very  religious  in  character,  if 
not  in  profession." 

The  relationship  to  its  source  of  the  English  version  of  the 
Havelok  story  is  still  a  matter  of  controversy  of  which  Dr. 
Fahnestock  (pp.  27-32)  has  given  the  fullest  report.     Skeat's 


HAVELOK   THE    DANE  I0? 

elaborate  theory  (1902,  p.  xlviii)  of  the  derivation  from  a  lost 
English  original  of  the  English  Havelok  involved  four  hypo- 
thetical stages  of  development  through  which  the  extant  text 
passed.  Skeat  took  heed  of  the  textual  influence  of  Norman 
scribes  but  not  of  any  possible  contacts  between  the  English  and 
the  French  versions.  Heyman  (p.  147)  vigorously  denied  that 
the  English  poem,  in  which  none  of  the  names  are  French  and 
in  which  passages  occur  each  of  which  "  is  found  only  in  one 
of  the  three  other  versions  respectively,"  could  be  derived  from 
anything  save  independent  English  tradition.  He  neglected  the 
possibility  that  changes  of  name  may  be  the  result  of  deliberate 
intention  or  even  of  mere  translation,  and  his  argument  from 
parallel  passages  is  unconvincing  until  the  relationship  of  the 
versions  has  been  finally  established.  Because  of  the  known  der- 
ivation of  most  Middle  English  romances  from  French  originals, 
and  of  the  indisputable  existence  of  the  French  versions  of  the 
Havelok  story,  Creek  (p.  196)  and  others  have  admitted  the 
probability  that  the  source  of  Havelok  was  French  (Fahnestock, 
p.  31).  Of  special  interest  was  Creek's  own  study  (pp.  197-202) 
showing  the  remarkable  uniformity  of  geographical  references 
throughout  all  the  versions,  and  more  important  still,  the  evi- 
dence that  the  English  version  is  "  honeycombed  with  inconsis- 
tencies and  difficulties  which  point  directly  to  the  French  version 
for  explanation." 

A  certain  resemblance  between  Havelok  and  the  Historia 
Meriadoci,  a  Latin  romance  of  the  thirteenth  century,  has  given 
rise  to  the  conjecture  that  the  latter  in  its  original  form  was  a 
Welsh  version  of  the  Havelok  story  (Deutschbein,  p.  134;  Bruce, 
Historia  Meriadoci,  Hesperia,  19 13,  p.  xxx).  The  resemblance 
is  limited  to  the  first  episodes  of  the  two  stories,  in  which  royal 
children  are  deprived  of  their  heritage  by  a  wicked  regent,  and 
saved  from  the  death  to  which  he  condemns  them  by  compas- 
sionate executioners.3 

Origin.  It  is  evident  that  by  the  middle  of  the  twelfth  cen- 
tury, when  Gaimar  wrote  his  Estorie,  the  Havelok  story  was 
fully  formulated.     That  it  was  localized  in  Lincolnshire  may 

3  A.  Olrik,  Heroic  Legends  of  Denmark,  tr.  L.  Hollander,  N.  Y.  iqiq, 
p.  310,  noted  that  the  Hrolfssaga  is  connected  but  slightly  with  the  Havelok 
legend  but  closely  with  Meriadoc. 


108  MEDIEVAL   ROMANCE    IN    ENGLAND 

have  been  one  reason  for  Gaimar's  telling  it,  as  he  was  writing 
under  the  patronage  of  Constance  FitzGilbert,  a  Lincolnshire 
lady  (Sisam,  p.  xxii).  Before  1338,  according  to  Robert  Man- 
ning, himself  of  Lincolnshire,  men  indicated  in  Lincoln  as  places 
of  interest  the  stone  that  Havelok  was  supposed  to  have  thrown 
at  some  enemies,  and  the  chapel  where  he  was  married ;  Grimsby 
too  was  held  to  be  the  town  founded  by  Grim  the  fisherman, 
(Chronicle,  ed.  Hearne,  Oxf.  1725,  1,  25).  The  seal  of  the 
town,  which  is  at  least  as  old  as  the  time  of  Edward  I,  portrays 
the  figures  of  "  Grym,"  "  Habloc,"  and  "  Goldeburgh  "  in  com- 
memoration of  the  English  legend.  Between  the  middle  of  the 
twelfth  and  the  middle  of  the  fourteenth  century,  the  story  was 
then  vitally  alive.    What  was  its  origin? 

The  historical  conditions  suggested  by  the  various  versions  of 
the  story  are  not  enlightening.  The  account  of  Athelwold's  reign 
in  the  English  Havelok  is  simply  idealized  description  but  cer- 
tain other  reminiscences  in  the  legend  strongly  suggest  the 
Anglo-Scandinavian  period.4  Since  in  all  versions  the  central 
idea  of  the  story  is  the  elevation  of  a  Danish  prince  to  the  king- 
ship of  Denmark  and  of  all  or  part  of  England,  the  origin  of 
the  legend  as  a  whole  must  be  subsequent  to  the  time  when  a 
Danish  sovereign  actually  ruled  England.  Heyman  (p.  88,  91) 
noted  that  the  first  Dane  to  hold  this  office  was  Sven  Tveskaeg 
(1013)  and  that  Canute  was  the  first  Danish  king  to  be 
crowned  in  London  as,  in  the  English  romance,  Havelok  is  said 

4  For  brief  but  interesting  comment  on  Horn,  Beves,  and  Havelok,  as 
Viking  Sagas  see  H.  Leach,  Angevin  Britain  and  Scandinavia,  Index;  for 
Havelok  in  particular  see  Bjorkman,  Bugge,  and  Whistler.  Miss  Ashdown, 
p.  113  ff.,  believed  that  the  episode  in  the  Lai  in  which  Havelok  fights  to 
decide  the  fate  of  his  kingdom  with  the  usurper  Odulf,  formed  part  of  the 
original  legend  of  Anlaf-Havelok,  and  that  the  duel  of  Canute  and  Edmund 
Ironside  to  decide  the  fate  of  England,  an  episode  recorded  by  Henry  of 
Huntingdon  and  others,  was  perhaps  inspired  by  the  tradition  associated  with 
Anlaf.  But  as  Bell,  p.  27,  pointed  out,  the  one  clear  thing  about  the  battle 
recorded  by  Gaimar  and  the  author  of  the  Lai  is  that  it  occurred  in  Denmark, 
not  in  England,  as  it  would  if  it  had  been  part  of  the  local-historical  tradition 
connected  with  Anlaf.  This  episode,  moreover,  is  far  from  being  a  regular 
part  of  the  Havelok  tradition.  If  it  is  difficult  to  accept  the  legend  of  the 
Single  Combat  even  in  Guy  of  Warwick,  in  which  it  is  a  very  important 
part  of  the  story,  as  symbolic  of  Brunanburh  and  its  subsequent  traditions, 
it  is  still  more  difficult  to  believe  in  the  Havelok-Odulf  episode  as  in  any 
way  related  to  the  Anlaf  who  was  defeated  at  Brunanburh.  See  Guy  of 
Warwick,  note  11. 


HAVELOK   THE    DANE  109 

to  be.  The  large  number  of  Norse  words  in  the  English 
Havelok  "  argues  that  it  was  composed  in  some  stronghold  of 
Scandinavian  influence  such  as  Lincolnshire."  But  this,  as  has 
been  said,  is  rather  a  suggestion  of  the  distinctive  character  of 
the  English  version,  than  an  argument  that  this  version  pre- 
serves, more  independently  than  the  French  versions,  the  tradi- 
tions of  that  place. 

The  quest  for  historical  "  kernels  "  in  a  story  so  apparently 
realistic  has  gone  on  apace  with  the  name  of  Havelok  as  the 
crux  of  many  discussions.  In  the  French  versions  and  the 
Lambeth  Interpolation,  Havelok  bears  the  nickname  Cuaran 
(Cuheran,  Coraunt),  which  the  Lai  (11.  258-60)  says  was  what 
"  li  Breton  "  called  a  cook.  Originally  this  meant,  in  Irish, 
brogue  or  sandal.  Because  of  the  use  of  this  nickname  and 
because  the  French  Avelok  is  the  same  as  the  Anglo-Saxon  Anlaf, 
the  Scandinavian  Olaf  (Irish,  Amhlaibh;  Welsh,  Abloyc),  Storm 
(Eng.  Stud,  in,  533)  identified  the  Havelok  of  romance  as  the 
Viking,  Olaf  Sictricson,  who  was  also  known  as  Anlaf  Cuaran. 
Chroniclers  and  scholars  alike  have  confused  this  man  with  his 
more  famous  cousin,  Anlaf  Guthfrithson,  who  was  defeated  by 
the  English  at  the  battle  of  Brunanburh  in  937  (Beaven,  p.  6). 
Both  Anlafs  experienced  vicissitudes  of  fortune,  gaining  and 
losing  royal  power  in  Northumbria,  and  their  efforts  to  regain 
their  lands  parallel  to  some  extent  those  of  Havelok.  But  neither 
man  ever  had  any  connection  with  Denmark  and  the  resemblance 
between  them  and  the  hero  of  the  romance  is  chiefly  a  matter 
of  name  (Hey wood,  p.  71,  80  ff. ;  Beaven,  p.  6,  n.  22).  However 
tempting  it  is  to  find  the  historic  prototype  of  Havelok  in  Anlaf 
Cuaran,  it  must  be  remembered  that  the  earlier  the  texts  of  the 
legend,  the  more  romantic  they  are.  It  is  in  the  later  texts  that 
the  tendency  grows  to  emphasize  historical  aspects  and  con- 
nections. In  the  chronicles  of  Rauf  de  Boun  and  of  Knighton, 
Havelok's  story  is  told  as  affording  evidence  of  Canute's  claim 
to  the  English  throne  by  virtue  of  Havelok's  earlier  rule  (Ash- 
down,  p.  115,  n.  5).  The  unreliability  and  confusion  worse  con- 
founded of  these  "  historical  "  accounts  can  be  shown  even  in 
Gaimar's  Estorie ;  for  in  this,  written  little  more  than  one 
hundred  years  after  the  close  of  the  Anglo-Danish  period  in 
England,  the  Havelok  story  is  placed  in  what  would  correspond 


HO  MEDIEVAL   ROMANCE    IN    ENGLAND 

to  the  years  495-556  in  the  Anglo-Saxon  Chronicle,  and  Havelok 
is  endowed  with  a  father  Gunter  who  was  conquered  by  King 
Arthur.  In  Gunter  still  later  chroniclers,  such  as  Pierre  de 
Langtoft,  recognized  the  Guthrum  who  was  King  Alfred's  op- 
ponent and  the  later  king  of  East  Anglia  (Heyman,  p.  82). 

Another  attempt  to  find  historical  antecedents  for  the  Havelok 
legend  was  made  by  Deutschbein.  He  agreed  with  Heyman  that 
the  connection  between  Anlaf  and  Havelok  was  merely  nominal, 
but  in  the  heroine  Goldeboru  (Argentine  in  the  French  versions) 
he  recognized  (pp.  103-17)  Alfwyn,  daughter  of  Athelflaed,  the 
famous  Lady  of  Mercia.  After  her  mother's  death  in  918  Alfwyn 
was  dispossessed  of  her  inheritance  by  her  uncle,  Edward  the 
Elder.  In  Caradoc's  History  of  Wales,  1697,  is  recorded  the 
tradition  that  Edward's  wrath  had  been  incurred  by  Alfwyn 's 
secret  marriage  to  Reynald,  king  of  the  Danes.  Although  it  is 
true  that  the  Anglo-Saxon  Chronicle  and  other  authoritative 
texts  refer  to  the  disinheritance  of  Alfwyn  and  to  certain 
activities  of  this  Reynald  (Reginwald),  there  is  unhappily  no 
other  evidence  than  Caradoc's  for  the  romantic  story  of  Reynald 
as  the  husband  of  Alfwyn  and  of  Edward's  anger.  Rather  are 
we  told  in  the  Chronicle  that  Reginwald  concluded  in  924  a  firm 
alliance  with  Edward  whom  he  chose,  according  to  Florence  of 
Worcester,  for  his  father  and  lord.  Deutschbein  (p.  109)  ad- 
mitted that  even  on  Caradoc's  evidence  Edward  was  represented 
as  opposing  Alfwyn's  marriage  whereas  in  Havelok  the  heroine's 
uncle  wickedly  forces  her  into  what  seems  a  wretched  union. 
If  Edward  were  historically  the  friend  of  Reginwald  and  the 
opposer  of  Alfwyn's  marriage,  it  is  difficult  to  recognize  in  him 
the  Villain-Uncle  of  Havelok's  wife,  the  man  on  whom  the  hero 
took  such  hearty  vengeance.  In  the  historical  accounts  cited  by 
Deutschbein  to  justify  his  contention  that  the  legend  of  Havelok 
grew  out  of  the  history  and  deeds  of  Reginwald,  the  uncle  of 
Anlaf  Cuaran,  there  is  little  agreement  in  the  names  of  historical 
characters  5  with  those  in  the  legend,  and  no  agreement,  save  in 

5  Deutschbein  did  not  discuss  the  Raegnald  Guthfrithson  who  was  received 
at  baptism  in  943  by  King  Edmund  and  who  was  for  a  time  the  successful 
rival  of  Anlaf  Cuaran.  (Cf.  Beaven,  pp.  7-9,  for  an  account  of  this  person- 
age who  would  seem  to  have  almost  equal  claims  with  those  of  the  earlier 
Reginwald  for  consideration.)  In  order  to  strengthen  his  Reginwald  theory 
Deutschbein  noted  that  various  chroniclers  report  the  name  of  Reginwald's 


HAVELOK   THE    DANE  m 

fragmentary  details,  in  the  sequence  of  events.  The  fundamental 
assumption  that  a  legend  of  the  famous  Anlaf  might  usurp  the 
adventures  of  his  uncle,  is,  of  course,  entirely  possible;  but  as 
Reginwald's  supposed  connection  with  Alfwyn  was  Deutschbein's 
principal  reason  for  asserting  that  the  life  of  Reginwald  more 
nearly  paralleled  that  of  Havelok  than  did  Anlaf's  own,  this 
explanation  of  the  "  historical  "  elements  in  the  Havelok  legend 
must  stand  or  fall  with  the  Alfwyn  story. 

In  its  general  outline  the  exile-and-return  type  of  the  Havelok 
tale  proves  nothing  save  the  popularity  of  the  formula.6  Have- 
lok, like  many  another  dispossessed  young  prince,  lives  humbly  in 
the  service  of  a  foreign  king.  Panzer,  in  his  study  of  the 
"  Goldener-marchen "  listed  (Hilde-Gudrutn,  p.  252-54)  many 
bright-haired  kitchen  boys  who,  like  Havelok,  serve  their  lords. 
A  more  literary  parallel  may  be  found  in  the  second  part  of  the 
French  epic  Aliscans,  of  the  kindly  giant  Rainouart  who  serves 
in  the  royal  kitchen,  outdoes  his  fellows  in  feats  of  strength  and 
in  eating,  and  finally  marries  a  royal  princess  (Heyman,  p.  97). 
Bugge  (pp.  272-95)  set  forth  the  notable  likeness  between  the 
story  of  Havelok  and  the  saga  of  Olaf  Tryggvason,  who  was 
spirited  on  board  ship  by  his  mother,  was  captured  by  pirates, 
as  Havelok  was  in  the  French  Lai  and  the  Interpolation,  who 
served  as  a  slave  even  as  Havelok  served  as  a  scullion,  and 
who  was  distinguished  by  a  strange  light  which  spread  over  him, 
even  as  Havelok  was  by  the  light  that  came  from  his  mouth 
when  he  slept.    Zenker  (p.  97)  somewhat  elaborately  compared 


father  as  Guthred  or  Guthrum,  a  king  of  East  Anglia.  This  Guthred  Deutsch- 
bein  identified  as  the  Gunter  of  the  romance  and  cited  in  connection  with 
him  a  story  told  by  Simeon  of  Durham  to  the  effect  that  Guthred  (son  now 
of  Hardacnut)  had  been  sold  as  a  slave  to  a  widow  and  had  been  rescued  at 
the  instigation  of  St.  Cuthbert.  This  story  Deutschbein  thought  (p.  118) 
might  have  suggested  the  scullion  boy  episode  in  Havelok.  It  would  seem 
then  that  Havelok  owed  his  name  to  Anlaf  Cuaran,  the  account  of  the  loss 
and  recovery  of  his  kingdom  and  his  marriage  to  a  disinherited  woman  to 
Reginwald,  and  his  youthful  experiences  as  a  scullion  to  Guthred.  If  Havelok 
were  the  result  of  this  strange  compilation,  then  truth  is  even  stranger  than 
the  fiction  created. 

6  Cf .  Nutt,  "  The  Arian  Expulsion-and-Return  Formula  in  the  Folk-and- 
Hero  Tales  of  the  Celts,"  Folk-Lore  Record,  Lond.  1881;  A.  Olrik,  op.  cit. 
p.  309,  for  Danish  parallels;  Deutschbein,  pp.  120-31,  cited  examples  from 
the  legends  of  Olaf  Tyggvason,  Tristan,  of  Aurelius  Ambrosius,  brother  of 
Uther-Pendragon,  of  Waldef,  of  Beves. 


II2  MEDIEVAL   ROMANCE    IN    ENGLAND 

this  early  portion  of  Havelok  with  Livy's  similar  tale  (Lib.  I, 
cap.  xxxix)  of  Servius  Tullius  (Cf.  Heyman,  p.  99).  Despite 
his  noble  birth,  Servius,  like  Havelok,  grew  up  as  a  servant; 
was  recognized  as  royal  by  a  flame  playing  about  his  head ;  won 
fame  as  a  warrior  and  ultimately  married  a  king's  daughter. 
Deutschbein  (p.  168)  pointed  out  the  inconsistency  between 
Zenker's  acknowledgment  that  Servius's  flame  simply  corre- 
sponds to  the  gold  hair  of  the  boy  in  the  folk-story,  and  his 
insistence  that  Havelok's  flame  must  be  a  literary  borrowing. 
In  the  French  versions  of  the  story,  Havelok  is  distinguished 
not  only  by  the  flame  but  by  his  ability  to  blow  a  great  horn 
and  thus  prove  himself  the  true  heir.  In  the  English  romance  he 
bears  on  his  shoulder  a  bright  "  kingmark  "  which  more  than 
satisfied  Grim  of  his  origin.  In  Richars  It  Biaus  (1.  670),  as 
Heyman  noted  (p.  101),  the  hero's  lineage  is  recognized  from 
the  light  in  his  face  and  the  shining  crosses  on  his  shoulder.  In 
the  related  poem,  the  Lion  of  Bourges,7  the  heir  is  known  both 
by  a  cross  and  by  his  power  to  blow  the  horn,  but  it  is  im- 
probable that  this  late  romance  is,  even  in  these  details,  an  off- 
shoot of  the  Havelok  story.  Fiction  has  always  identified  true 
heirs  by  these  and  similar  devices.  In  French  and  English 
versions  of  Havelok  alike  the  hero's  lineage  is  revealed  to  his 
wife  through  a  dream  or  vision.  In  the  French  versions  the 
dream  of  animals  attacking  the  hero  is  a  portent  of  dangers  to 
come  familiar  enough  in  romance,  but  the  later  account  of  the 
homage  paid  by  animals  and  of  the  trees  that  bowed  to  the  hero, 
is  a  bit  of  elaboration  not  commonly  met.8  In  the  English 
romance  Havelok's  dream  of  himself,  with  his  arms  stretched 
out  to  embrace  all  England,  is  curiously  matched  by  the  dream 
of  William's  mother  in  both  the  French  and  the  English  versions 
of  William  of  Palerne  (Heyman,  p.  107). 

A  certain  similarity  between  the  legend  of  Hamlet  and  that 
of  Havelok  has  been  used  by  Zenker  to  support  his  theory  that 
the  two  stories  are  fundamentally  related.  They  both  concern 
Danish  princes  who  marry  in  England  and  return  and  regain 

7  For  Richars  and  Lion  see  Amadas,  n.  2. 

8  Cf.  W.  Baake,  Die  Verwendung  des  Traummotivs  in  der  engl.  Dichtung 
bis  auf  Chaucer,  Diss.  Halle,  1906,  ch.  in;  W.  Henzen,  Ueber  die  Traume  in  der 
altnord.     Sagalitteratur .     Leipzig,    1890. 


HAVELOK   THE    DANE  U3 

their  heritage.  In  particular  the  legends  introduce  the  same 
stratagem  to  deceive  an  enemy.  In  both  cases  dead  men  are  set 
up  on  stakes  in  order  to  suggest  the  arrival  of  new  forces. 
Hey  man  (p.  96  ff.)  noted  how  the  same  motif  appears  in  Ogier 
le  DanoiSj  in  which  the  hero  made  dummies  of  wood  and  horse- 
hair and  placed  them  on  the  walls  of  Castlefort  (Gautier, 
Epopees,  in,  240  ff.),  in  the  Provengal  Philomena,  and  in  other 
widely  distributed  literary  and  popular  tales. 


BIBLIOGRAPHY 

Texts:  (i)  Laud  Misc.  108,  ed.  Madden,  Roxburghe  Club,  Lond. 
1828;  Skeat,  EETSES.  rv,  1868,  Oxford,  1902  (rev.  Anglia  BL.  xiv,  10); 
Skeat,  revised  by  K.  Sisam,  Oxford,  1915  (rev.  MLN.  xxxi,  252); 
Holthausen,  Old  and  Middle  Eng.  Texts,  Lond.  1901  (rev.  JGP.  ill, 
510);  Heidelberg,  2nd  ed.  1910;  (2)  Cbg.  Univ.  Lib.  4409  (19),  three 
fragments,  ed.  Skeat,  MLR.  vi,  455-7  (191 1),  pr.  Sisam,  p.  103.  Trans. 
E.  Hickey,  Lond.  1902;  F.  J.  Darton,  Wonder  Book  of  Romance,  N.  Y. 
1907;  L.  A.  Hibbard,  Three  ME.  Romances,  Lond.  191 1;  J.  Weston, 
Chief  ME  Poets,  Boston,  19 14. 

Textual  and  Linguistic  Studies: 

Bradley,  E.  "On  Havelok,"  1.  2333,  Trans.  Phil.  Soc.  1903-04. 
Browne,  W.  H.  "  Havelok's  Lament,"  v.  570-04,  MLN.  xxi,  23  (1906). 
Foerster,  W.  "  Zu  Havelok,"  v.  2461,  Archiv,  107,  107   (1901)., 
Grattan,  J.  "Minor  Notes  on  Havelok,"  MLR.  iv,  91  (1908). 
Holthausen,  F.  "  Zum  Havelok,"  v.  321,  504,  Eng.  Stud,  xxx,  343  (1902)  ; 

Archiv,  CI,  100;  Anglia  BL  XI,  306,  359;  xn,  146;   "Emendations 

to  the  text  of  Havelok,"  Furnivall  Miscellany,  Oxford,  1901,  p.  176. 
Horn,  W.  "  On  Havelok,"  v.  247,  Anglia  xxix,  132. 
Koeppel,  E.  "  Havelok-randglossen,"  Anglia  BL  xxm,  294   (1912). 
Littlehale,  H.  "  On  Havelok,"  v.  2495,  Trans.  Phil.  Soc.  1902-03,  p.  161. 
Morsbach,  L.  "  Bemerkungen  zum  Havelok,"  Eng.  Stud,  xxix,  368-74 

(1901). 
Napier,  A.  "Havelok  Notes,"  MLR.  XI,  74  (1916). 
Schmidt,  F.  Zur  H eimatbestimmnng  des  Havelok.  Diss.  98  pp.  Gottingen, 

1900. 
Wolff,  A.  Zur  Syntax  des  Verbums  im  alteng.  Lay  of  Havelok.  Diss.  69 

pp.  Leipzig,  1909. 

Studies  and  Analogues:  Cf.  Billings,  Guide,  pp.  23-4;  Edwardes, 
Summary,  pp.  72-74;  Fahnestock  (see  below),  pp.  31-32;  Heyman  (see 
below),  pp.  i-ix;  Sisam,  (see  texts),  pp.  v-viii;  Wells,  Manual,  pp. 
763-64. 


II4  MEDIAEVAL   ROMANCE   IN   ENGLAND 

Ashdown,  M.  "The  Single  Combat  in  Certain  Cycles  of  English  and 
Scandinavian  Tradition  and  Romance,"  MLR.  xvn,  113-30  (1922). 

Beaven,  M.  "  King  Edward  I  and  the  Danes  of  York,"  Eng.  Hist.  Rev. 
xxxiii,  1-9  (ig1^). 

Bell,  A.  "The  Single  Combat  in  the  Lai  d'Havelok,"  MLR.  xvm,  22-29 

(1923)- 
Bjorkman,   E.   "  Nordiska  Vikingasagor  i  England,"  Nordisk   Tidskrift, 

1906,  pp.  440-50. 
Brie,  F.  "  Zum  Fortleben  der  Havelok-sage,"  Eng.  Stud,  xxxv,  359-71 

(1905). 
Bugge,  A.  "  Havelok  and  Olaf  Tryggvason,"  Saga  Book  of  the  Viking 

Club,  vi,  257-95  (1910). 
Creek,  H.   C.   "  The  Author  of   Havelok,"  Eng.  Stud,  xlviii,   193-212 

(1915). 
Deutschbein,  M.  See  Horn  bibliography. 

Fahnestock,  E.  A  Study  of  the  Sources  and  Composition  of  the  Old 

French  Lai  d'Havelok.  Diss.  138  pp.  Bryn  Mawr,  191 5. 
Gaaf,  W.  van  der.  "  Parliaments  held  at  Lincoln,"  Eng.  Stud,  xxxn, 

319-20  (1903). 
Heyman,  H.  Studies  on  the  Havelok-Tale.  Diss.  153  pp.  Upsala,  1903. 
Putnam,  E.  K.  (1)  "  The  Lambeth  Version  of  Havelok,"  PMLA.  xv,  1-16 

(1900).      (2)    "The   Scala   Chronica  Version,"   Trans.  Amer.  Phil. 

Assoc.  1903,  p.  xci. 
Sisam,  K.  See  Texts. 
Skeat,  W.  See  Texts. 
Whistler,  C.  "  Saga  of  Havelok  the  Dane."  Saga  Book  of  the  Viking  Club, 

in,  395-412    (1902). 
Zenker,  R.  Boeve-Amlethus,  Die  Havelok  Saga,  pp.  91-111.    See  Beves 

bibliography. 


BEVES    OF   HAMPTON 

Versions.  The  hero  who  bears  the  name  of  Beves  of  Hamp- 
ton (Boeve  de  Hamptone,  Hanstone)  might  well  be  described 
as  an  international  character.  The  wide  wandering  of  his  story 
was  like  his  own  fabled  adventuring  from  Hampton  to  Damascus. 
Versions  in  English,  Welsh,  Irish,  French,  Dutch,  Scandinavian, 
Italian,  attest  the  popularity  of  him  who  became  even  in  Russia 
the  most  acclimated  hero  of  the  chivalric  epic  (Wesselofsky ;  cf. 
Rom.  xvm,  313).  The  story  of  the  loss  and  recovery  of  his 
inheritance,  his  fights  with  Saracens  and  dragons,  his  marriage 
with  a  converted  princess,  his  gaining  of  innumerable  possessions, 
is  distinctive  chiefly  for  its  amazing  absorption  of  familiar  motifs 
and  for  its  blending  of  elements  drawn  from  romance,  fairy  tale, 
saint  legend,  and  heroic  epic.  Few  stories  better  illustrate  the 
catholicity  of  mediaeval  taste ;  and  in  this,  perhaps,  lay  the  secret 
of  an  influence  which  may  be  traced,  not  only  through  the  wealth 
of  manuscript  material  but  through  many  literary  allusions  to 
the  poem  and  through  the  representation  of  its  incidents  in  dif- 
ferent artistic  forms.1 

The  length,  the  number,  and  the  variety  of  the  vernacular 
versions  of  Beves  make  the  problem  of  their  classification  ex- 
tremely difficult.  Since  the  publication  in  1899  of  Stimming's 
edition  of  the  Anglo-Norman  version  of  Beves,  the  story  has 
been  the  subject  of  many  elaborate  investigations,  but  for  the 
purpose  of  enumeration  it  is  convenient  to  disregard  the  maze 
of  controversy  and  to  note  as  the  three  principal  versions  the 
Anglo-French  (AF),  the  Continental  French  (CF),  and  the 
Italian  (Matzke,  Mod.  Phil,  x,  20). 

1  Scenes  from  Beves  appear  in  the  Smithfield  Decretals  and  in  the  Tay- 
mouth  Horae  (See  here  Guy,  n.  10).  Notes  and  Queries,  8th  ser.  xi  (1897) 
referred  to  the  hangings  of  Juliana  de  Leybourne,  1362,  which  were  worked 
with  the  legend.  W.  G.  Thompson,  Tapestry  Weaving,  p.  26,  mentioned  two 
pieces  of  arras  of  Beves  of  the  time  of  Henry  V.  The  Bull,  de  la  Soc.  des 
Antiquaires  de  France,  1909,  p.  237,  shows  a  small  stone  mould  (c.  1359)  of 
the  Musee  de  Cluny  on  which  Beves  and  two  lions  appear.  An  inscription 
refers  to  "  Bueve." 

115 


n6  MEDIAEVAL   ROMANCE    IN    ENGLAND 

The  first  group,  as  Stimming  made  clear,  has  four  branches,  a 
thirteenth-century  Anglo-Norman  poem  extant  in  two  long  sup- 
plementary fragments  (ed.  Stimming,  1899),  a  fourteenth-century 
prose  version  in  Norse  (ed.  Cederschiold,  1884),  another  of  the 
thirteenth-century  in  Welsh  (R.  Williams,  1892),  and  one  in 
Middle  English  verse  (ed.  Kolbing).  The  last,  in  the  Auchinleck 
manuscript,  has  the  first  474  verses  in  a  six-line  stanza  and  the 
remaining  4146  lines  in  short  couplets.  The  popularity  of  the 
version,  belonging  originally,  it  would  seem,  to  the  south  of 
England  (Kolbing,  xin  ff.),  is  attested  by  the  six  existing  texts 
and  by  the  six  which  Kolbing  assumed  as  antecedent  in  order 
to  explain  the  extant  readings.  These  six  manuscripts  fall  into 
two  classes  (A  and  SN;  Mo-ME-C),  in  which  the  earliest,  the 
(A) Auchinleck  manuscript,  is  less  near  to  the  lost  thirteenth- 
century  Middle  English  original  than  is  the  fifteenth-century 
(M)  Manchester  manuscript,  or  even  Pynson's  old  print.  This 
original,  from  which  the  later  manuscripts  take  over  numerous 
references  to  a  French  original,  was,  in  Stimming's  opinion, 
derived  from  a  lost  Anglo-Norman  version  (x),  the  source  also, 
through  various  lost  intermediaries,  of  the  extant  Anglo-Norman 
and  Welsh  texts,  and  of  the  Norse  account.  The  Middle  English 
poet  seems  to  have  shortened  his  original  at  will,  to  have  elab- 
orated certain  episodes,  and  to  have  made  three  important  addi- 
tions :  (1)  the  account  of  Beves's  first  battle  fought  on  Christmas 
day  for  the  honor  of  God ;  (2)  his  great  fight  with  the  dragon  of 
Cologne,  an  episode  which  suggests  to  the  poet  comparison  of 
his  hero  with  Lancelot,  Wade,  and  Guy  of  Warwick;  and  (3) 
the  heroic  defense  made  by  Beves  and  his  sons  against  the 
London  citizens  when  they  are  roused  against  him  by  the  accusa- 
tion that  Beves  has  killed  the  king's  son,  a  scene  graphic  enough 
to  suggest  some  contemporary  riot.  Despite  its  prolixity  and 
its  constant  borrowings  from  the  commonplaces  of  Middle 
English  romantic  diction,  which  Schmirgel  pointed  out  in 
Kolbing's  edition,  (pp.  xlv-lxvi),  the  poem  has  a  certain  vigor  of 
its  own.  Its  popularity  with  a  mediaeval  audience  is  not  to  be 
wondered  at,  nor  is  it  strange  that  the  traditional  delight  in  this 
hero  persisted  even  in  the  Elizabethan  period.2    An  instance  of 

2  Beves  and  Guy  had  an  almost  equal  popularity,  and  the  heroes  were 
often  mentioned  together.     See  Crane,  Bibliog.  of  Guy. 


BEVES   OF   HAMPTON  n7 

the  foreign  interest  in  the  Middle  English  Beves  is  a  fifteenth- 
century  Irish  translation  (ed.  Robinson). 

The  Anglo-Norman  (AF)  text  is  generally  thought  to  represent 
an  independent  version  of  the  same  story  as  that  told  by  the 
continental  French  texts.  Of  these,  nine  manuscripts  in  verse 
and  two  in  prose  are  now  known.  They  fall  apparently  into 
three  groups.  The  first  is  represented  by  the  thirteenth-century 
Paris  manuscript  (P1)  published  by  Stimming  in  191 1.  This 
version,  Behrens  (p.  77)  believed,  originated  between  1230  and 
1250,  on  the  southern  borders  of  Picardy.  The  second  version, 
represented  by  an  inedited  and  incomplete  manuscript  in  Rome 
(R),  another  (W)  of  the  fifteenth  century  in  Vienna,  and  by 
another  thirteenth-century  Paris  manuscript  (ed.  Stimming, 
1913),  was  thought  by  Oeckel  (p.  78)  and  Meiners  (p.  239)  to 
have  been  by  the  scribe,  Pierot  du  Ries.  The  possibility  that 
Pierot  might  have  been  the  author  was  dismissed  by  Stimming 
(2,  p.  4,  200),  This  version  tends  constantly  to  amplify  the 
original  by  new  episodes  and  so  much  delights  in  ecclesiastical 
detail  that  its  author  was  presumably  of  the  clergy.  "  Lokal 
patriotismus,"  however,  gives  now  and  then  a  secular  touch  to 
his  story.  The  third  group  comprises  the  Beves  texts  of  the 
thirteenth  and  the  fourteenth  centuries  found  in  manuscripts  at 
Carpentras  (C),  Turin  (T),  and  Venice  (V);  and  finally  a 
fragment  now  at  Modena  (Wolff  and  Paetz).  Of  these  conti- 
nental texts  Boje  (pp.  136-37)  believed  the  oldest  and  truest 
form  to  be  represented  by  the  Rome  and  Paris  manuscripts  of 
the  second  group,  and  the  original  text  to  be  the  work  of  one 
man  only.  As  a  whole  this  continental  French  version  is  some- 
what longer  than  AF,  and,  unlike  it,  places  the  hero's  home  on 
Gallic  soil  and  names  his  stepfather  Doon  de  Mayence.  In  the 
AF  version  Doon  is  Emperor  of  Almayn,  and  Beves's  home  is  at 
Littlehampton  (Hampton-sur-Mer,  v.  2811),  not  more  than  two 
and  one  half  miles  from  Arundel,  the  city  named,  according  to 
the  English  romance,  in  honor  of  the  race  won  by  Arundel, 
Beves's  famous  horse.  Finally  it  may  be  noted  that  the  two 
fifteenth-century  French  prose  versions  of  Beves  and  the  five 
known  sixteenth-century  editions  belong  to  the  same  redaction 
as  the  manuscripts  P,  R,  W  (Boje,  p.  13). 

The  Italian  version  is  preserved  in  at  least  six  texts,  of  which 


n8  MEDIEVAL   ROMANCE    IN   ENGLAND 

the  earliest  is  the  fragmentary  thirteenth-century  Venetian 
manuscript  (ed.  Reinhold).  The  only  complete  form  is  the 
Buova  d'Antona  in  the  Reali  di  Francia,  a  late  fifteenth-century 
composite  which  draws  on  the  French  as  well  as  the  Italian 
versions.  The  Italian  version  is  shorter  than  the  French;  it 
differs  in  names  and  in  sequence  of  events ;  and  is,  in  the  opinion 
of  Rajna  (Ricerche  intorno  at  Reali  di  Francia,  pp.  135-40, 
Milan,  1872),  of  Jordan,  and  Matzke  (3,  p.  32),  the  prior  form 
"  independently  transmitted  from  the  original  version  of  which 
the  common  source  of  AF  and  CF  is  another  offspring." 

Of  the  later  popular  versions  of  Beves,  the  first  Dutch  edition, 
printed  at  Antwerp,  1504,  was  derived  from  the  CF  version;  and 
the  sixteenth-century  Russian  and  Jewish  folk-books  were  from 
the  Italian  (Wesselofsky,  Rom.  xvm,  302-14,  1889).  In  1881 
the  Italian  was  translated  into  Roumanian  (Groeber,  1901,  11,  3, 
386).  The  fullest  account  of  these  and  all  the  other  versions 
is  given  by  Boje  (pp.  1-13). 

The  influence  of  Beves  has  been  traced  in  the  Middle  High 
German  poem,  Graf  Rudolph  (cf.  Bethmann,  Palcestra,  xxx; 
Deutschbein,  p.  191),  but  the  similar  scenes  are  of  the  fairly 
conventional  type  concerning  a  Christian  hero  and  a  heathen 
princess.  The  Provencal  poem,  Daurel  et  Beton  (ed.  P.  Meyer, 
Paris,  1880),  is  in  part  clearly  a  sequel  to  Beves  (Jordan,  1, 
102).  Brockstedt's  account  (pp.  96-103)  of  this  relationship  is 
more  convincing  than  his  idea  that  the  Siegfriedlied  and  the 
Nibelungenlied  are  variations  of  the  Anglo-Norman  Beves.  The 
forest  death  of  Beves's  father,  Beves's  fight  with  the  dragon  of 
Cologne,  and  the  bridal  of  Josian  with  Earl  Miles,  are  in  truth 
analogous  to  scenes  in  the  German  poems,  but  the  inference 
made  from  the  resemblance  is  over-large.  Boje  (p.  137)  believed 
that  the  influence  of  the  French  forms  of  Beves  was  to  be  clearly 
traced  in  certain  incidents  in  five  poems ;  in  Florent  et  Octavian 
{Hist.  Litt.,  xxvi,  316),  in  P arise  et  Vienne  {Rom.  Forsh.,  xv, 
1904),  in  Ciperis  {Hist.  Litt.,  xxvi,  31),  in  Valentin  u. 
Namelos  (ed.  Seelmann,  1884,  p.  68)  and,  most  interesting  of  all, 
in  Aucassin  (ed.  Suchier),  in  the  episode  in  which  the  heroine, 
disguised  as  a  maiden  minstrel,  goes  in  search  of  her  lost  lover. 
On  the  whole,  however,  the  influence  of  Beves  is  best  attested  by 
the  long  line  of  its  own  self-perpetuating  versions. 


BEVES   OF   HAMPTON  n9 

Origin.  Beves  of  Hampton  is  a  typical  roman  d'aventure 
which  moves  within  a  certain  "  Ideenkreis  "  of  a  well-defined 
character.  In  his  comparison  of  it  with  one  hundred  and  eighty- 
seven  Old  French  romances  Boje  distinguished  the  following 
characteristic  details  and  incidents :  the  forest  hunt,  p.  62  ;  the 
murder  of  Beves's  father,  the  marriage  of  his  mother  with  her 
husband's  murderer,  the  stepfather's  hostility  to  Beves,  pp.  62- 
64;  the  disguise  of  Beves,  coloring  his  face,  p.  67,  etc.,  to  save  his 
life;  the  exhibition  of  his  blood-stained  clothes  as  a  proof  of 
death,  p.  66;  the  rude  porter,  p.  71 ;  the  feast  broken  up  by  a 
tumult,  p.  66 ;  the  selling  of  the  boy  and  his  stay  at  the  court  of 
a  foreign  king,  the  love  for  him  of  the  Saracen  princess,  the 
defeat  through  Beves  of  her  cruel  suitor,  the  false  accusation 
brought  against  the  lovers,  the  letter  of  death  carried  by  Beves  to 
a  heathen  king,  pp.  74-80 ;  the  overthrow  of  the  idols  by  Beves, 
p.  82  ;  his  imprisonment  in  Damascus,  his  escape  and  the  vain 
pursuit,  pp.  91-100 ;  the  beating  of  the  idols  by  the  heathen  king, 
p.  100;  Josian's  forced  marriage  and  the  magic  protection  of 
her  virginity,  p.  106;  Beves's  disguise  as  a  palmer  and  his  horse's 
recognition  of  his  master,  pp.  108-09 )  the  drugging  of  Josian's 
guard,  p.  112;  the  elopement  of  the  lovers,  pp.  109-12;  the 
grotesque  giant  Escopart  and  his  comic  baptism,  pp.  1 13-14; 
Josian's  second  forced  marriage,  the  killing  of  her  husband,  and 
Beves's  rescue  of  Josian  from  the  stake,  pp.  11 5-1 7;  Beves's 
homecoming,  the  rage  of  the  usurper  who  throws  a  knife  at  the 
messenger,  p.  90;  the  overthrow  of  the  usurper  by  Beves  in 
battle  or  by  a  judicial  combat,  pp.  82-88 ;  the  great  race  won  by 
Beves's  horse,  p.  118;  the  horse  theft  attempted  by  the  king's 
son,  p.  131 ;  the  killing  of  the  king's  son,  pp.  120-23 ;  the  second 
exile  of  Beves,  the  forest  birth  of  Josian's  twin  sons,  the  separa- 
tion of  the  family,  pp.  123-24;  Beves's  nominal  marriage  with 
another  lady,  Josian's  disguise  as  a  minstrel,  her  search  for  her 
lost  love,  the  recognition  and  reunion  of  husband  and  wife,  pp. 
128-31 ;  the  old  age  of  Beves,  the  angelic  warning  and  his  death, 
pp.  132-33- 

As  no  text  of  Beves  antedates  the  thirteenth  century,  as  lin- 
guistic studies,  no  less  than  a  literary  study  of  motifs  such  as 
Boje's,  suggest  nothing  antecedent  to  1200,  it  is  probable  that 
the  original  poem  was  not  composed  before  that  date.     But 


120  MEDIAEVAL   ROMANCE    IN    ENGLAND 

numerous  attempts  have  been  made  to  find  in  the  extant  versions 
the  signs  of  much  more  ancient  origin.  Suchier's  belief  (i,  p. 
cxcv)  based  on  the  evidence  of  such  names  as  Ivor,  Bradmund, 
Rudefoun,  etc.,  that  the  poem  was  basically  a  Viking  saga,  may  be 
offset  by  reference  to  Langlois's  Tables  des  noms  propres  dans  les 
chansons  de  geste,  Paris,  1904,  from  which  it  appears  that  these 
names  appear  in  Old  French  poems  for  which  no  Viking  origin 
can  possibly  be  alleged.  Deutschbein  (p.  198)  sought  to  connect 
the  story  with  certain  historical  German  antecedents  and  sug- 
gested identification  of  Doon,  represented  in  Beves  as  the  Em- 
peror of  Almayne  who  murders  Beves's  father  in  the  forest  in 
order  to  marry  his  mother,  with  Otto  (Odon)  the  Great  (929-947) 
who  exiled  his  step-son,  Duke  Ernst  of  Swabia,  or  with  the 
father  of  Ernst  II  of  Swabia  who  was  killed  on  a  hunt  and  whose 
son  revolted  against  his  step-father,  the  Emperor  Conrad  II. 
Boje  (pp.  62  ff.),  however,  proved  the  essentially  literary  char- 
acter of  this  introductory  part  of  the  romance. 

The  question  of  origin  has  been  constantly  associated  with 
the  localization  of  the  story.  The  apparently  ample  evidence  of 
English  place-names,3  which  led  Stimming  (pp.  183-85)  to  be- 
lieve the  poem  of  Anglo-Norman  origin,  has  been  brought  into 
dispute  by  the  contention  that  the  Italian  version,  in  which  the 
English  are  supplanted  by  Continental  names,  is  representative 
of  the  oldest  and  most  authoritative  version.  Rajna  in  1872 
was  one  of  the  first  to  point  out  in  his  studies  on  the  Reali  di 
Francia  that  Hamtone  or  Hanstone  might  better  be  identified 
with  Hunstein  or  Hammerstein  on  the  Rhine  than  with  South- 
ampton, and  others  have  stressed  the  importance  of  the  clearly 
non-English  elements  in  the  romance.  Nevertheless,  Matzke, 
who  did  most  to  establish  the  independent  value  of  the  Italian 
version,  thought  {Mod.  Phil.,  x,  54)  the  question  of  insular  or 
continental  origin  still  an  open  one.  Less  cautious  scholars, 
by  considering  limited  portions  of  the  story  in  the  AF  or  CF 
group,  which  they  take  to  represent  the  original  nucleus  of  the 
story,  have  arrived  at  interestingly  varied  opinions.  Settegast 
(pp.  282,  383)  derived  the  history  of  Beves's  first  exile  from  an 
Armenian  tale  in  which  a  king  was  killed  on  a  hunting  expedi- 

3  Cf.  J.  Westphal,  Englische  Ortsnamen  im  Altfranzosischen.    Diss.  Strass- 
burg,  1 89 1. 


BEVES   OF    HAMPTON  12 1 

tion,  the  throne  was  seized  by  an  usurper  and  a  young  prince, 
the  true  heir,  escaped  in  disguise  as  a  shepherd  boy.  By  the 
most  dubious  sort  of  etymology  (p.  354)  the  names  in  this  tale 
were  made  in  some  instances  to  coincide  with  those  in  Beves, 
and  so  made  to  argue  an  eastern  origin  for  the  romance. 
Deutschbein  (p.  182),  emphasizing  different  elements  in  this 
same  part  of  Beves,  the  ill  treatment  of  the  boy  by  his  relatives, 
the  feast  which  he  breaks  up  by  shaming  his  enemy,  was  re- 
minded of  Karl  Mainet  and  of  an  episode  in  Jourdain  de 
Blaivies.  The  account  of  Beves's  relations  with  his  royal  step- 
father still  further  suggested  (p.  198)  the  twelfth-century 
German  poem,  Herzog  Ernst,  (ed.  Bartsch,  1869),  which  relates 
the  adventures  of  Ernst  of  Swabia,  traditionally  the  rebellious 
stepson  of  Otto  the  Great.  In  Graf  Rudolph,  c.  11 70  (Palaestra 
xxx)  the  eastern  adventures  of  the  hero,  his  escape  from  prison, 
his  rescue  of  his  beloved  from  a  forced  marriage,  parallel  to 
some  degree  similar  incidents  in  Beves.  These  stories  of 
Mainet  and  Ernst  and  Rudolph,  which  were  known  in  their 
earliest  versions  in  the  district  between  Flanders  and  Picardy, 
were  supposed  by  Deutschbein  (p.  204)  to  have  been  carried  to 
England  by  Flemish  colonists  who  settled  in  Pembrokeshire  in 
the  neighborhood  of  Haverford  (Aberford,  in  AN.  Beves). 
There  the  stories  were  localized,  and  to  some  extent,  perhaps, 
influenced  by  tales  of  the  Horn  type.  The  commonplace  like- 
ness between  Beves  and  Horn  in  the  hero's  expulsion  from  home, 
his  adventures  at  the  foreign  court,  his  banishment,  his  rescue 
of  his  betrothed,  led  Hoyt,  on  wholly  insufficient  grounds,  to 
conclude  that  the  home  of  the  two  stories  must  have  been  in 
England  and  that  Beves  was  "  but  a  romantically  developed 
form  of  the  Horn  Saga." 

The  historical  kernel  for  the  story  of  Beves's  second  exile  is 
to  be  found,  according  to  Jordan  (Archiv,  cxin,  98),  in  the  story 
recorded  under  the  year  870  by  Regino  of  Priim  (Mon.  Germ. 
I)  of  Carolus,  the  Frankish  prince.  In  this  anecdote  a  courtier, 
whose  horse  has  been  stolen  in  jest  by  the  prince,  unluckily 
wounds  the  royal  youth  and  has  to  flee  for  his  life.  Deutschbein 
(p.  209)  accepted  Jordan's  view  and  noted  that  Priim  was  not 
far  from  the  district  from  which  he  fancied  some  episodes  in 
the  first  part  of  Beves  to  have  been  originally  drawn.    The  theft 


122  MEDIAEVAL   ROMANCE    IN   ENGLAND 

of  a  famous  horse  as  an  episode  in  itself  was,  as  Boje  (p.  131) 
indicated,  a  popular  incident. 

Legendary  sources  for  Beves  have  been  found  far  and  near. 
Zenker  (p.  44)  maintained  that  Beves  and  the  Hamlet  (Am- 
lethus)  legend  told  by  Saxo  Grammaticus  were  versions  of  the 
same  story  (p.  32),  and  that  the  common  source  probably 
originated  in  England.  In  the  two  stories  the  hero  becomes  the 
stepson  of  his  father's  murderer,  vows  vengeance,  has  a  violent 
altercation  with  his  mother,  is  sent  (but  for  different  causes)  to 
a  foreign  court  bearing  a  letter  of  death  (Uriasbrief),  escapes, 
and  finally  returns  to  accomplish  his  revenge  on  the  step-father, 
the  usurper  of  his  heritage.  Zenker  believed  that  of  these  inci- 
dents the  most  distinctive  was  the  use  of  the  Uriasbrief,  and 
paralleled  it  (p.  45)  with  numerous  oriental  tales,  with  the  Greek 
Bellerophron  story  (pp.  283,  313),  and  the  French  Bit  de 
VEmpereur  Constant  (Rom.,  vi,  162  ff.).  But  later  students 
have  shown  that  in  most  of  these  instances,  with  the  exception 
of  the  Greek  story,  the  letter,  so  rewritten  as  to  command  great 
rewards  for  the  bearer,  opened  to  him  a  new  career  of  successful 
adventure.  Such  is  the  tale  twice  found  in  the  Amlethus  legend, 
but  in  Beves  the  original  letter  was  delivered  by  the  hero,  and 
almost  caused  his  death.  This  simpler  use  of  the  motif  seems  to 
be  derived  either  from  the  ancient  Biblical  story  (2  Sam.  xi, 
15)  of  David  and  Uriah  or  from  "  a  folk-lore  tale  current  in  the 
East  and  introduced  into  Beves  in  the  time  of  the  Crusades." 

A  second  important  argument  of  Zenker's  that  Amlethus  is 
the  source  of  Beves,  rested  on  the  supposedly  similar  incidents 
of  the  double  marriage  of  the  two  heroes.  In  Beves  the  hero, 
separated  from  his  wife  and  children,  comes  to  a  city  (AF, 
Aumberforce,  CF,  Civile)  ;  its  ruler,  one  of  many  "  Forth- 
Putting  "  ladies,  offers  herself  to  him,  having  been  attracted  by 
his  military  prowess;  he  enters  reluctantly  into  a  pretended 
marriage  with  her  (AF  version)  ;  and  his  true  wife  appears  in 
time  to  prevent  its  consummation.  In  Amlethus  the  hero  enters 
willingly  into  the  second  marriage,  and  the  interest  of  the  epi- 
sode lies  entirely  in  the  Valkyrie-like  character  of  the  lady  who, 
because  of  her  vow  of  chastity,  has  long  caused  the  death  of 
all  her  suitors.  The  essential  unlikeness  of  the  episodes  makes 
it  improbable  that  one  was  derived  from  the  other.    To  Jordan 


BEVES   OF    HAMPTON  I23 

(2)  the  distinctive  element  in  Beves  was  the  hero's  separation 
from  his  family  —  the  separation  and  reunion  motif  that  domi- 
nates such  stories  as  Guillaume  d'Angleterre,  Sir  Isumbras,  Die 
Gute  Frau,  etc.,  narratives  which  are  always  in  this  episode  in 
some  way  related  to  the  Eustachius  legend. 

Although  Zenker  believed  that  the  larger  portion  of  Beves 
was  to  be  derived  from  the  northern  Hamlet  legend  or  its 
variants  in  the  stories  of  Havelok,  Urol]  Hraka,  or  the  Icelandic 
iAnlodi,  which  he  thought  basically  related,  he  accounted  for 
many  of  its  eastern  elements  by  traces  which  he  detected  in  the 
Hamlet  legend  itself  of  the  ancient  Persian  Chosro  story  found 
in  the  "King's  Book"  of  the  poet  Firdausi  (cir.  ton).  This 
Chosro  account  in  turn  seems  to  show  a  fusion  of  the  Brutus  and 
Bellerophron  legends.  Beves's  childhood  resembles  that  of 
Chosro;  for  each  has  a  faithful  protector  in  the  person  of  his 
father's  friend,  each  acquires  a  wonderful  horse  whose  recogni- 
tion of  his  master  is  sometimes  of  vital  consequence,  each  hero 
marries  a  king's  daughter. 

The  Eastern  names,  the  localization  of  so  many  incidents  in 
eastern  places,  the  perceptible  flavor  of  the  Crusading  spirit  in 
Beves,  have  led  to  other  attempts  to  identify  special  incidents. 
Beves's  imprisonment  in  Damascus  was  traced  by  Settegast  (pp. 
282,  338)  to  the  similar  experience  of  Bischen  as  recorded  in 
Firdausi 's  book,  and  more  significantly  by  Brockstedt  (p.  35)  to 
the  French  Floovent.  In  the  Italian  Bovo  (Jordan,  1,  p.  17)  the 
princess  Malgaria  loves  and  protects  the  imprisoned  Beves;  in 
Floovent  the  princess  Maugalie,  similarly  tender-hearted,  aids 
the  hero  to  escape.  The  possible  influence  on  the  Ur-Bueve  of 
Floovent  or  other  stories  of  this  exceedingly  popular  type  must 
be  admitted.  Warren's  study  (PMLA.  xxix,  340-59)  of  the 
Enamoured  Moslem  Princess,  showed  that  the  type  story  greatly 
antedated  the  Crusading  era,  as  he  traced  its  earliest  western 
form  to  the  sixth  Controversia  of  Seneca,  the  Rhetorician,  and 
the  earliest  Crusade  version  to  the  account  of  Bohemond  in  the 
Historia  Ecclesiastica,  c.  1135,  of  Orderic  Vitalis.  Brockstedt's 
argument,  however,  that  the  Italian  version  of  Beves,  because  it 
borrowed  the  episode  from  Floovent,  is  a  late  form,  was  disputed 
by  Matzke  (Mod.  Phil.,  x,  25)  who  urged  that  the  role  of  Mal- 
garia must  have  belonged  to  the  French  source  of  the  Italian 


I24  MEDIEVAL   ROMANCE    IN   ENGLAND 

poem,  since  she  is  to  be  recognized  as  the  necessary  second  hero- 
ine of  the  story  (which  he  believed  the  fundamental  one  in 
Beves),  the  so-called  Legend  of  the  Man  with  Two  Wives  ("  Lay 
of  Eliduc,"  Mod.  Phil,  v,  211-39).  In  this  type  a  youth  exiled 
from  his  own  home  wins  through  his  valor  the  love  of  a  princess. 
He  is  slandered  and  is  again  forced  to  go  into  exile.  In  another 
court  he  wins  the  love  of  another  lady  but  remains  loyal  to  the 
first.  He  returns  in  time  to  rescue  her  from  an  unwelcome  mar- 
riage, or  she  appears  in  time  to  prevent  his  marriage  to  the 
second  lady.  To  Matzke  (3,  p.  41  ff.)  the  starting  point  of  the 
legend  is  simply  the  doubling  of  the  exile-and-return  formula, 
and  the  consequent  doubling  of  the  love  adventure  of  the  hero. 
The  doubled  form  appeared  in  such  tales  as  Horn,  Me  et  Galeron, 
and,  with  certain  variations  in  Tristan,  Eliduc,  Lai  del  Fraisne, 
and  its  derivative,  Roman  de  Galeran.  A  comparison  of  the 
different  versions  of  Beves  seems  to  show  that  its  original  form 
was  structurally  of  the  same  type  as  these. 

Some  of  the  earliest  processes  of  accretion  in  Beves  are  set 
forth  in  Matzke's  study  of  the  St.  George  legend.  In  its  ancient 
Eastern  forms  this  legend  had  known  only  the  monster-killing 
and  martyrdom  episodes,  but  in  the  course  of  its  development  in 
the  west  it  absorbed  the  Beves  story  and  became  a  typical 
roman  d'aventure,  as  it  appears,  for  instance,  in  Richard 
Johnson's  Seven  Champions  of  Christendom,  London,  1592.  In 
Beves,  on  the  other  hand,  the  influence  of  the  saint  legend  is 
especially  obvious  in  the  scene  in  which  Beves  overthrows  the 
heathen  idol,  in  the  account  of  his  sufferings  in  the  prison  of 
Damascus,  and  his  fight  with  the  dragon  of  Cologne. 

In  regard  to  the  authorship  of  Beves,  the  most  important  sug- 
gestion of  recent  years  was  that  made  by  Boje.  He  urged  that 
the  original  French  version  was  the  work  of  a  single  author  suffi- 
ciently acquainted  with  contemporary  romance  to  borrow  from 
it  freely.  His  belief  that  Beves  was  not  a  racial  saga,  that  it 
was  not  of  German,  Anglo-Saxon,  Celtic,  or  Viking  origin,  nor  a 
gradual  combination  of  elements  drawn  from  Persian-Armenian, 
nor  Graeco-Roman  story,  but  a  literary  romance,  the  work  of 
one  man,  is  in  line  with  the  whole  tendency  of  modern 
criticism.4 

4  Cf.  Bedier,  Les  Legendes  J&piques,  Paris,  1908-13;  L.  Foulet,  Roman  de 
Renard,  Paris,  1914;  F.  Lot,  Etude  sur  le  Lancelot  en  prose,  Paris,  1918. 


BEVES   OF    HAMPTON 


BIBLIOGRAPHY 


125 


Texts,  Middle  English:  (i),  A,  Auchinleck  MS.,  ed.  Turnbull, 
Maitland  Club,  Edin.,  1838,  rev.  Eng.  Stud.  11,  317;  E.  Kolbing,  EETSES. 
xlvi,  xlviii,  lxv,  1885-86,  1894,  rev.  Anglia,  xi,  325;  Eng.  Stud,  xix, 
261;  Rom.  xxm,  486;  (2)  C,  Caius  Coll.  Cbg.  175,  desc.  Eng.  Stud. 
xiv,  321;  (3)  S,  Egerton  2862,  desc.  Brit.  Mus.  Catalogue  of  Add.  MSS, 
1905-10,  p.  238,  formerly  the  MS.  of  the  Duke  of  Sutherland,  desc. 
Eng.  Stud.,  vii,  191  ff. ;  (4)  N,  Royal  Library,  Naples,  MS.  xm,  B,  29; 
(5)  C,  Cbg.  Univ.  Libr.  MS.  Ff.  11,  38;  (6)  M,  Chetham  Library, 
Manchester,  MS.  8009,  ed.  Kolbing,  op.  cit.;  cf.  Eng.  Stud,  vn,  198. 
Early  printed  editions:  L,  "Douce  fragments,"  no.  19,  Bodleian;  0, 
undated  edition  by  Pynson,  Bodleian.  Editions  from  1 689-1 711  listed  by 
Esdaile,  English  Tales,  pp.  163-64.  Trans.  L.  Hibbard,  Three  Middle 
Eng.  Romances. 

French:  Stimming,  A.  (1)  "Der  Anglo-Normannische  Boeve  de 
Haumton,"  Bibliotheca  Normannica,  vn,  Halle,  1899;  (2)  "  Der  fest- 
landische  Bueve  de  Hantone,"  Fassung  1,  Gesellschaft  f.  rom.  Lit.  xxv 
(1911);  (3)  Fassung  11,  ibid,  xxx  (1912);  xli  (1918);  Fassung  hi, 
ibid,  xlii  (1920). 

Irish:  Robinson,  F.  N.  "The  Irish  Lives  of  Guy  of  Warwick  and 
Bevis  of  Hampton."  Zts.  f.  celt.  Phil,  vi,  180-320  (1907).  Text  and  trans. 
See  also,  Eng.  Stud,  xxiv,  463. 

Italian  :  Reinhold,  J.  "  Die  f ranko-italienische  Version  des  Bovo 
d'Antone."  Zts.  /.  rom.  Phil,  xxxv,  555-607;  683-714;  xxxvi,  1-32 
(1912). 

Studies:    Billings,  Guide,  pp.  40-1;  Boje  (see  below)  for  MSS.,  pp. 
1-13;   Studies,  pp.  43-49;  Wells,  Manual,  pp.   765-66. 
Behrens,  L.  Ort  u.  Zeit  der  Entstehung  der  Fassung  I  des  festlandischen 

Beuve  de  Hantone.  Diss.   135  pp.  Gottingen,   1913. 
Bodtker,  A.     "  Ivens  Saga  u.  Bevis  Saga  in  Cod.  Holm.  Chart.  46,"  PB. 

Beitrdge  xxxi,   261-71    (1906). 
Boje,  C.  "  Ueber  den  altfrz.  roman  v.  Bueve  de  Hamtone."    Beihefte  z. 

Zts.  f.  rom.  Phil,  xix,  145  pp.  Halle,  1909.     Rev.  Rom.  xlii,  314; 

Zts.  /.  frz.  Spr.  u.  Lit.  xxxv,  49. 
Brockstedt,    1.    Floovent  Studien.    Kiel,    1907.    2.    Von  mittelhochdeut. 

Volksepen  franzosischen  Ursprungs.  Kiel,  191 2.  Beves,  pp.  60-159. 

Rev.  Archiv.  cxxi,  170-72. 
Deutschbein,    M.   Studien   z.   Sagengeschichte   Englands.     Die   Wiking- 

ersagen:    Horn,    Havelok,    Tristan,    Boeve,   pp.    181-215,    Guy-  of 

Warwick.     Cothen,  1906. 
Favaron,  G.  U element 0  italiano  nel  period  popolare  toscano  del  epopea 

romanzesca;  Saggio  sul  Buovo  d'Antona.  61  pp.  Bologna  (1900). 


I26  MEDIEVAL   ROMANCE    IN   ENGLAND 

Gerould,  G.  See  Isnmbras  here. 

Groeber,  G.  Grundriss,  II,  386   (1901). 

Hibbard,  L.  A.  "  Beves  of  Hampton  and  the  Nibelungenlied,"  MLN. 
xxvi,  159-60  (1911).  "  Jaques  de  Vitry  and  Boeve  de  Haumtone," 
MLN.  xxxiv,  408-11  (19 19). 

Hoyt,  P.  C.  "The  Home  of  the  Beves  Saga,"  PMLA.  xvn,  237-46 
(1902). 

Jordan,  L.  1.  "  Ueber  Boeve  de  Hanstone,"  Beihejte  z.  Zts.  j.  rom.  Phil. 
xiv.  197  pp.  Halle,  1908.  Rev.  Archiv.  cxxn,  412,  Zts.  j.  jrz.  Spr. 
u.  Lit.  xxxiv,  25.  2.  "  Die  Eustachiuslegende,  Christians  Wilhelm- 
sleben,  Boeve  de  Hanstone  u.  ihre  orientalischen  Verwandten," 
Archiv.  cxxi,  340-62   (1908). 

Kuhl,  H.  Das  gegenseitige  Verh'dltnis  der  Handschrijten  der  Fassung  II 
des  jestlandischen  Bueve  de  E  ant  one.  Diss.  63  pp.  Gottingen,  191 5. 

Matzke,  J.  E.  1.  "  Contributions  to  the  Legend  of  St.  George,"  PMLA. 
xvii,  464-535  xviii,  99-171  (1902-03).  2.  "The  Legend  of  St. 
George;  Its  Development  into  a  Roman  d'A venture,"  PMLA.  xix, 
449-78  (1904).  3.  "The  Oldest  Form  of  the  Beves  Legend."  Mod. 
Phil,  x,  19-54   (1912-13). 

Meiners,  J.  E.  Die  Handschrijten  P  (RW),  Fassung  II  d.  jestlandischen 
Bueve  de  Hantone.  Diss.  268  pp.  Gottingen,  1914. 

Oeckel,  F.  Ort.  u.  Zeit.  d.  Entstehung  der  Fassung  II  d.  jestlandischen 
Boeve  v.  Hantone.  Diss.  88  pp.  Gottingen,  191 1. 

Paetz,  H.  "  Ueber  das  gegenseitige  Verhaltnis  d.  venetianischen,  d.  franko- 
italienischen  u.  d.  franzosischen  gereimten  Fassungen  d.  Bueve  de 
Hantone."  Beihejte  z.  Zts.  f.  rom.  Phil.  L.  133  pp.  Halle,  1913. 

Reinhold,  See  Texts,  Italian. 

Robinson,  See  Texts,  Irish. 

Sander,  G.  Die  Fassung  T  des  jestlandischen  Fassung  d.  Bueve  de  Han- 
tone. Diss.  Gottingen,  1913. 

Settegast,  F.  Quellenstudien  z.  gallo-rom.  Epik.  Leipzig,  1904.  Ch. 
xvi,  338-69,  Beves,  Generides. 

Schiiltsmeier,  F.  Die  Sprache  d.  Handschrijt  C  d.  jestldnd.  Bueve  de 
Hantone.  Diss.  200  pp.  Gottingen,  1913. 

Stimming,  See  Texts,  French. 

Wolf,  S.  Das  gegenzeitige  Verh'dltnis  d.  gereimten  Fassungen  d.  jest- 
land.    Bueve  de  Hantone.    Diss.  Gottingen,  191 2. 

Zenker,  R.  "  Boeve-Amlethus,  Das  altfrz.  Epos  Boeve  de  Hantone  u.  der 
Ursprung  der  Hamletsage."  Liter arhist.  Forschungen,  xxxn,  480 
pp.  Berlin,  1905.    Rev.  Archiv.,  cxvin,  226;  Eng.  Stud.,  xxxvi,  284. 


GUY  OF   WARWICK 

Versions.  Though  Chaucer  in  the  fourteenth  century  was 
already  jesting  at  Guy  of  Warwick  as  a  "  romance  of  pris,"  it  is 
ironically  true  that  in  sheer  popular  favor  the  story  for  long 
years  outlasted  anything  of  his.  Frequently  re-issued  by  the 
early  printers,1  treated  as  serious  history  by  the  Elizabethan 
chroniclers  known  to  Skelton,  Udall,  Puttenham,  Drayton, 
Shakespeare,  dramatized  by  Day  and  Dekker,  1620,  turned  into 
stall  ballads  (1592)  and  again  into  heroic  poems  such  as 
Samuel  Rowlands's  Famous  History  of  Guy  Earle  of  Warwicke, 
1608,  revived  by  eighteenth-century  antiquarians,  and  happily 
read  as  pure  fairy  tale  by  nineteenth-century  children,  the  ro- 
mance had  a  history  of  which  greater  works  might  well  be  proud. 
Crane's  study  of  the  vogue  in  England  of  this  "  popular  classic  " 
gave  a  valuable  picture  not  only  of  the  extraordinary  persistence 
of  a  mediaeval  story  in  modern  times,  of  a  popularity  that 
stretched  from  the  thirteenth  to  the  nineteenth  century,  but  of 
the  successive  transformations  by  which  the  story  kept  pace 
with  the  changing  tastes  of  different  audiences.  Though  it 
passed  from  the  hands  of  one  rather  dull  redactor  to  another,  its 
diversity  of  episode  continued  to  attract  the  adventurous,  the 
romantic,  the  pious  alike.  Moreover,  until  the  end  of  the 
Renaissance,  the  story  made  a  strong  patriotic  appeal  to  all 
Englishmen,  for  Guy,  by  his  fight  with  Colbrond,  was  thought 
to  have  saved  his  country  from  the  Danes.  By  the  middle  of 
the  thirteenth  century  he  was  one  of  the  most  accredited  heroes 
of  English  legend.    By  14 10,  according  to  Dugdale  (Baronage,  1, 

1  Cf.  Crane,  p.  128  ff.:  Pynson's  edition,  printed  shortly  before  1500,  was 
"  in  all  essentials  identical  with  a  version  in  short  couplets  composed  as  early 
as  the  fourteenth  century";  Wynkyn  de  Worde,  c.  1500;  Copland,  1562-69; 
Caswood.  "After  1575  the  metrical  versions  ceased  to  be  reprinted"  (Crane, 
p.  141).  Reeves,  MLN.  xi,  col.  405,  disposed  of  Morley's  unsupported  as- 
sertion {Early  Prose  Romances,  1889,  p.  27)  that  the  earliest  edition  of  Guy 
in  English  prose  was  printed  by  Copland.  Brown  (p.  15)  showed  that  the 
source  of  Morley's  chapbook  version  was  Rowlands's  Famous  History.  See 
also  Crane,  p.  130,  and  Esdaile,  Eng.  Tales  printed  147 5-1642. 

127 


128  MEDIEVAL   ROMANCE    IN   ENGLAND 

243),  his  fame  had  spread  even  to  Jerusalem,  for  there  a  reported 
descendant  of  his  was  received  with  great  honor. 

The  mediaeval  versions  of  Guy  begin  with  those  in  French  verse 
of  which  thirteen  manuscripts  are  now  known.2  Winneberger's 
study  of  seven  of  these  manuscripts  led  him  to  believe  that  they 
fall  into  two  groups;  and  Weyrauch  (p.  81),  on  comparing  them 
with  the  Middle  English  manuscripts  of  Guy,  found  that  the 
latter  were  in  general  translated  from  the  second  group  of  French 
texts.3  Since  of  the  five  Middle  English  manuscripts,  the 
earliest  is  the  Auchinleck  manuscript  (1330-40),  the  first  Middle 
English  version  was  probably  composed  about  1300.  The  first 
English  references  to  Guy  are  to  be  found  in  Robert  of  Brunne's 
translation  of  Langtoft's  Chronicle,  c.  1338,  and  in  the  Speculum 
Vitce,  c.  1350  (Percy  Folio  MS.  11,  510,  512).  Of  the  English 
versions  of  the  romance  one,  in  short  riming  couplets,  concluded 
somewhat  abruptly  in  the  Auchinleck  manuscript  with  Guy's 
killing  of  the  dragon  that  came  "  out  of  Irlond  "  and  his  pres- 
entation of  its  head  to  Athelstan  at  Warwick.  This  couplet 
version  (A)  served  as  the  basis  for  the  fourteenth-century 
copies  made  in  the  Auchinleck  manuscript  (f.  107-46),  in  the 
fragmentary  Sloane  manuscript,  and  in  the  fifteenth-century 
manuscript  of  Caius  College,  Cbg.  107,  which  included  Guy's 
later  history.  According  to  Brandl,  Grundriss,  §  37,  it  was, 
perhaps,  composed  in  southern  Warwickshire. 

Another  version  (a),  of  a  slightly  more  northern  origin,4 
fashioned  in  twelve-line  tail-rime  stanzas,  is  found  only  in  the 
Auchinleck  manuscript  (ff.  146-67).  Beginning  practically  at 
the  point  where  the  story  in  A  had  stopped,  except  for  a  brief 

2  Seven  OF.  MSS.  were  listed  by  Tanner,  pp.  49-51,  of  which  one  (Bibl. 
Nat.  Paris,  MS.  fr.  FR.  1476)  was  a  fifteenth-century  prose  translation. 
Twelve  OF.  MSS.  were  given  by  Winneberger,  p.  2,  and  Weyrauch,  p.  67. 
These  lists  were  corrected  by  Herbert,  Rom.  35,  68,  who  added  an  account  of 
the  former  Edwardes  MS.,  now  additional  MS.  38662  of  the  British  Museum, 
(Acher).  Jenkins,  Mod.  Phil,  vii,  described  a  fragmentary  MS.  found  in  the 
Library  of  York  Minster.  Among  the  OF.  prose  versions  of  Guy  should  be 
listed  Royal  MS.  15  E  VI  of  the  British  Museum. 

3  This  includes  O  (MS.  Bennet  50.  6,  CCCCbg.  late  thirteenth  century) ; 
R  (Regent  MS.  8  F  IX,  Br.  Mus.,  early  fourteenth  century) ;  0  and  f  (Rawlin- 
son  D,  913,  Bodl.  Oxf.,  early  fourteenth  century.) 

4  Wilda,  Ueber  die  oertliche  Verbreitung  der  12-zeiligen  Schweifreimstrophe 
in  Eng.  Breslau,  1888,  pp.  46-55,  thought  this  version  arose  in  a  region  border- 
ing upon  Essex. 


GUY   OF   WARWICK  1 29 

recapitulation  of  Guy's  early  life,  this  version  told  of  the 
hero's  marriage  to  his  lord's  daughter  Felice,  the  proud- 
spirited lady  who  had  insisted  that  he  win  fame  enough  to  com- 
pensate for  his  lower  birth,  of  his  pious  resolve  to  abandon  his 
bride  for  whose  sake  he  had  so  long  forgotten  the  service  of 
God,  and  of  his  subsequent  adventures  until  his  death's  day  as 
penitent  warrior  and  hermit.  After  this  came  the  story  of  Guy's 
son  Reinbrun  (ff.  167-75),  a  short  romance  in  itself  of  127 
twelve-line  tail-rime  stanzas.  In  addition  to  these  two  versions 
Zupitza  in  1873  distinguished  two  others  in  couplet  form,  a  third 
represented  by  the  fragmentary  fourteenth-century  text,  now 
Add.  MS.  14408,  and  a  fourth  found  in  the  fifteenth-century 
manuscript,  Ff.  2,  38,  Cbg.  Univ.  Library.  Weyrauch  in  his 
study  of  the  relationship  of  all  the  versions,  decided  that  the 
second  or  a  version  was  the  result  of  an  independent  translation 
of  the  French  original.  His  conjecture  that  it  was  composed  by 
the  scribe  who  copied  A,  has  been  disputed  by  Moller  (p.  5)  on 
the  basis  of  differences  in  style,  phraseology,  rime,  and  dialect. 
In  Moller's  opinion  (p.  105),  A,  a,  and  ao,  as  he  designated 
Reinbrun,  were  by  different  authors.  He  also  pointed  out  (pp. 
47-81)  that  though  the  correspondence  between  a,  the  strophic 
part  of  Guy,  and  the  Middle  English  version  of  Amis  and 
Amiloun  was  much  greater  than  Kolbing  had  supposed,  it 
pointed  not  to  identity  of  authorship,  but  to  the  conscious  use  of 
Guy  by  the  author  of  Amis. 

The  number  of  extant  French  and  Middle  English  manuscripts 
but  approximately  suggests  those  once  known.  Guy  is  frequently 
listed  in  such  mediaeval  catalogues  5  as  have  come  down  to  us, 
two  texts,  for  instance,  Guy,  Earl  of  Warwick,  and  Guy  and 
Colbronde,  appearing  in  the  little  library  of  English  books  of 
John  Paston  (Crane,  p.  126).  The  popularity  of  the  story  was 
far  from  insular.  In  the  fifteenth  century  it  was  translated  into 
Irish  from  a  Middle  English  version  (Robinson).  It  strongly 
influenced  the  famous  Catalonian  tale  of   Tirant  lo  Blanch? 

5  Savage,  Old  Eng.  Libraries,  pp.  229-31 ;  cf.  also  PFMS.  11,  510,  for 
reference  to  texts  at  Bruges  and  Brussels. 

6  Reproduced  in  facsimile  by  the  Hispanic  Soc,  N.  Y.,  1904.  See  Swan, 
Gesta  Romanorum,  11.  527,  and  Spence,  Legends  and  Romances  of  Spain, 
p.  193  ff-5  J.  A.  Vaeth,  Tirant  lo  Blanch,  Columbia  Univ.  Diss.  N.  Y.,  1918, 
p.  72,  97-111;  A.  Thomas,  Spanish  and  Portuguese  Romances  of  Chivalry, 
Cambridge.  1920,  pp.  32-40. 


I3o  MEDIEVAL   ROMANCE    IN   ENGLAND 

Valencia,  1490,  which  in  turn  was  translated  into  Castilian, 
Italian,  and  French.  Though  the  hermit  warrior  whom  the 
young  Tirant  of  Brittany  met  near  Windsor  was  called  William, 
Earl  of  Warwick,  his  old  adventures  were  those  of  Guy.  Like 
Guy  he  had  gone  to  the  holy  land,  had  returned  to  live  as  a 
hermit  near  his  own  castle,  had  been  forced  to  fight  mightily 
against  an  invader  of  his  country,  etc.  The  eastern  adventures 
of  the  young  Tirant  suggest  those  of  the  youthful  Guy. 

Not  only  in  romances  but  in  exempla  and  in  chronicles  are 
versions  of  Guy  to  be  traced.  In  the  Latin  and  vernacular  texts 
of  the  Gesta  Romanorum  appeared  a  curious  synopsis  of  the 
romance  which  abbreviated  into  a  page  or  so  the  thousands  of 
lines  devoted  to  Guy's  early  history.  It  related  in  detail, 
however,  the  story  of  his  meeting  with  his  dispossessed  friend, 
Tirri,  of  the  latter's  dream  of  treasure,  of  the  soul-animal  which 
ran  from  Tirri 's  mouth  into  a  knoll  where  the  friends  were  to 
discover  weapons  for  Guy,  of  the  treachery  by  which  Guy  was 
thrown  into  the  sea,  and  of  his  victory  over  Tirri 's  enemy.  As 
Tanner  observed  (pp.  41-42),  this  amazing  tale  with  its  absurd 
moral  was  certainly  not,  as  Warton  (Hist.  Eng.  Poetry,  1,  286) 
first  supposed  it  to  be,  an  early  outline  of  the  champion's  his- 
tory, but  a  condensation  based  on  the  known  romance  itself. 
In  a  number  of  manuscripts  of  the  Gesta  the  hero's  name  was 
lost  and  he  was  called  Josias  or  Rosias.  This  version  was  popular 
enough  to  have  a  life  independent  even  of  the  Gesta  Romanorum. 
It  appears  as  a  separate  narrative  in  the  fifteenth-century  Ger- 
man prose  narrative  published  by  Mau,  and  served  as  the  basis 
for  the  mystere  of  Jean  Louvet  which  was  written  in  Paris  in 
1537  (Hibbard,  p.  183). 

The  English  chronicles  make  frequent  allusion  to  Guy  and 
especially  to  his  fight  with  the  Danish  giant  Colbrond,  but  since 
no  reference  of  this  sort  antedates  the  fourteenth  century,  it  may 
be  safely  inferred  that  the  chronicles  borrowed  from  the  ro- 
mances. The  earliest  of  these  pseudo-historical  accounts  seems 
to  have  been  that  written  by  Gerard  of  Cornwall  in  his  Historia 
Regum  Westsaxonum.  Of  this  the  eleventh  chapter,  the  His- 
toria Guidonis  Warwick,  is  still  preserved.7    Nothing  is  known 

7  In  a  manuscript  of  Higden's  Poly  chronic  on,  now  MS.  147  of  Magdalen 
College,  Oxf.,  and  in  Cott.  Vesp.  D  IX  f.  41  (10)  in  the  British  Museum. 


GUY   OF   WARWICK  13 1 

of  the  writer  save  that  he  is  referred  to  in  the  Liber  de  Hyda,  a 
late  fourteenth-century  compilation,  and  in  Thomas  Rudborne's 
Historia  Wintoniensis,  c.  1454.  These  references  are  Kingsford's 
only  reasons  for  suggesting  in  the  Dictionary  of  National  Biog- 
raphy that  Gerard  lived  about  1350.  Tanner  believed  (p.  33) 
that  Gerard  should  be  identified  with  that  shadowy  Walter  of 
Exeter  of  whom  Bale,  in  the  Index  Brittannicce  Scriptorunt, 
recorded  that  he  (Walter)  had  written  in  1301  "  apud  S.  Caradoc 
in  Cornubia  "  a  Vitam  Guidonis  Comitis  de  Warwyck  at  the 
instance  of  Baldwin,  "  civis  Excestriensis  urbis."  More  acutely 
Tanner  (p.  29)  noted  that  Gerard's  text  was  simply  a  Latin 
translation  of  the  French  romance.  What  relation,  if  any,  this 
Latin  prose  version  had  to  the  "  canticum  Colbrondi  "  which  the 
minstrel  Herebertus  8  sang  in  1338  in  the  hall  of  St.  Swithin's,  is 
of  course  impossible  to  say.  After  Gerard  the  story  of  Guy's 
fight  was  told  by  Knighton  (Chronicon,  before  1366),  by  Rud- 
borne,  1454,  by  Hardyng  (Chronicle,  before  1465),  and  by  John 
Rous.9  Even  in  the  sixteenth  century,  as  Crane  noted  (p.  134), 
it  was  seriously  recorded  by  such  chroniclers  as  Fabyan,  Grafton, 
Holinshed,  and  Stow.  It  is  of  interest  to  find  that  between  1442 
and  1468  Gerard's  text  was  translated  at  the  request  of  Margaret, 
Countess  of  Shrewsbury,  who  claimed  descent  from  the  old  Eng- 
lish hero.  The  poet  in  this  case  was  John  Lydgate,  who,  despite 
a  few  pseudo-Chaucerian  flourishes,  managed  to  retell  the  story 
in  eight-line  stanzas  with  equal  fidelity  and  dullness.  The  poem 
survives  in  six  manuscripts  (Robinson).  A  little  Geste  of  Guy 
and  Colbrond  is  in  the  Percy  Folio  MS.  (11,  527-49). 

Illustrations  from  the  romance  of  Guy  of  Warwick  have  not 
infrequently  appeared  in  mediaeval  art.  A  number  of  scenes 
were  introduced  in  the  Taymouth  Horae,  c.  1330,  and  others  in 
the  Smithfield  Decretals  (Br.  Mus.,  MS.  10  E  IV).10    Dugdale 

8  Hazlitt-Warton,  Hist.  Eng.  Poet.  1871,  n,  97,  in,  168;  cf.  E.  K.  Cham- 
bers, Medieval  Stage,  1,  56.  Kolbing  (Germania,  Neue  Reihe,  xxn,  193) 
thought  the  poem  Guy  and  Colbrande  (PFMS.  11,  527)  was  derived  from  an 
older  version  closely  related  to  that  used  by  the  poet  of  the  A  version. 

9  For  Rous  see  Diet.  Nat.  Biog.  Two  versions  are  extant  of  his  Roll  of 
the  Earls  of  Warwick.  The  Rous  Rol,  1477-85,  was  published  privately  (Rows 
Roll)   in  1845,  and  by  Wm.  Courthope,   1859. 

10  These  references  are  due  to  the  kindness  of  R.  S.  Loomis,  whose  studies, 
especially  in  connection  with  the  Tristan  legend,  are  doing  much  to  illumine 
the  relationship  of  mediaeval  narrative  and  art.    See  below,  notes  14  and  17. 


I32  MEDIEVAL   ROMANCE    IN    ENGLAND 

(Antiquities  of  Warwickshire,  i,  237)  recorded  that  a  suit  of 
arras  containing  the  story  of  Guy  and  hanging  in  Warwick 
Castle,  was  given  by  a  special  grant  to  Thomas  Holland,  Earl  of 
Kent  (Pat.  21,  Ric.  2).  In  Rowlands's  poem  (Ch.  ix),  reference 
was  made  to  the  arras  at  Warwick  which  depicts  the  fight  of 
Guy  with  the  dragon  (Brown,  p.  19).  Warton  (Hist.  Eng. 
Poetry,  1824,  1,  93)  recorded  that  within  his  memory  a  rude 
painting  of  the  fight  of  Guy  and  Colbrand  was  to  be  seen  on  the 
walls  of  the  north  transept  of  Winchester  Cathedral. 

Origin.  In  the  traditionally  famous  fight  of  Guy  with  Col- 
brond,  Tanner  (p.  21)  and  others  sought  to  find  the  historical 
nucleus  of  the  romance.  Anglo-Saxon  chronicle  and  song  record 
the  great  victory  of  Brunanburh,  937,  won  by  Athelstan  over  the 
Viking  invader,  Anlaf,  and  since  the  romance  uses  the  names  of 
both  Athelstan  and  Anlaf,  the  temptation  to  identify  the  historic 
and  the  fictitious  event  has  been  strong.  But  the  romance  local- 
izes the  scene  near  Winchester,  which  was  certainly  not  the  site  of 
Brunanburh,  and  makes  the  fate  of  England  dependent  on  the 
issue  of  a  single  combat.11  Deutschbein  (p.  221)  furthermore 
pointed  out  that  the  duel  in  fact  and  fiction  was  wholly  unfamiliar 
to  the  Anglo-Saxons  and  that  its  presence  in  the  romance  precludes 

11  G.  Neilson,  Trial  by  Combat,  Lond.  1890,  pp.  23-30,  and  Deutschbein, 
p.  223,  show  the  flimsiness  of  the  evidence  purporting  to  prove  that  the 
judicial  single  combat  was  known  in  England  before  the  Norman  Conquest. 
If  the  duel  itself,  if  the  names  of  the  adversaries  and  the  place  of  the 
Guy-Colbrand  fight  cannot  be  associated  with  Brunanburh,  it  is  impossible  to 
accept  the  legendary  account  as  even  symbolic  of  the  historic  battle.  For 
romance  texts  describing  the  Island  Combat  see  G.  Schoepperle,  Tristan,  II, 
339-67;  for  the  Single  Combat  see  O.  Leibecke,  Der  verabredete  Zweikampf 
in  der  altjrz.  Lit.,  Gottingen,  1905 ;  M.  Pfeffer,  "  Die  formalitaten  des 
gottesgerichtlichen  Zweikampfs,"  Zts.  f.  rom.  Phil,  ix  (1885),  1-75;  G. 
Fundenburg,  Feudal  France  in  the  French  Epic,  Princeton,  1918,  pp.  92-99; 
Ashdown,  p.  128.  Miss  Ashdown,  p.  126,  n.  2,  disputed  Deutschbein's  con- 
tention that  Guy  was  after  the  Conquest;  she  argued  that  the  Single  Combat 
was  widely  known  in  Scandinavia  before  the  Norman  Conquest  and  that  the 
Colbrand  episode,  in  which  Colbrand  represents  the  Danish  nation,  may  as 
well  have  been  inspired  by  Scandinavian  as  by  Anglo-French  tradition.  But 
the  fact  remains  that  the  Anglo-Normans  were  the  first  shapers  of  the  Guy 
legend  and  that  they  gave  it  the  form  of  pure  romance.  The  single  combat 
in  Middle  English  romance  appears  in  Guy  of  Warwick,  in  Beves  of  Ham- 
toun,  in  Tristrem,  in  Torrent  of  Portyngale,  in  Partonope  of  Blois,  in  King 
Alisaunder,  in  Duke  Rowlande  and  Sir  Otuel,  and  traces  of  it  are  to  be  found 
in  King  Horn  (Cf.  ed.  by  Hall,  p.  143)  and  in  the  English  Troy  Book 
(EETS.  cxxi,  1.  8477). 


GUY   OF    WARWICK  133 

the  possibility  that  the  episode  was  drawn  from  the  tenth-century 
tradition.  Conscious  archaism  and  patriotic  purpose  in  Guy, 
as  in  the  romance  of  At  heist  on  itself,  sufficiently  explain  the  use 
of  the  historic  names,  and  the  popularity  of  the  fateful  Single 
Combat  in  mediaeval  story  accounts  for  its  association  with  Guy. 
Ward's  attempts  (Cat.  of  Romances,  1,  475)  to  identify  Rohaud 
of  Warwick,  the  father  of  Felice,  as  that  actual  Thurkill  of 
Warwick  (Domesday  Book,  ff.  240-41)  who  had  a  granddaughter 
Felice,  and  Siward  of  Ardern  as  that  Siward  of  Wallingford  who 
in  the  romance  was  Rohaud's  steward,  explained  nothing  in 
regard  to  the  motivation  of  the  romance  but  did  possibly  indicate 
a  certain  amount  of  antiquarian  information  on  the  part  of  the 
French  poet.  The  romance  is  evidently  intended  to  exalt  the 
lords  of  Wallingford  and  Warwick;  and  records  of  the  two 
founders  of  those  families,  of  Thurkill  of  Warwick  and  of  Wigod 
of  Wallingford,  himself  the  Cupbearer  of  Edward  the  Confessor, 
and  the  most  probable  prototype  of  Guy  in  the  romance,  must 
have  been  many  in  both  family  and  monastery,  in  the  monas- 
tery of  Abingdon,  for  instance,  to  which  Thurkill  and  his  son, 
Robert  d'Oily,  had  been  benefactors,  and  in  Oseney  Abbey, 
where  the  Abbot  Wigod  (1138-69)  was  possibly  a  descendant  of 
the  earlier  Wigod  (Freeman,  Conquest,  iv,  App.  C;  Deutschbein, 
p.  218).  It  is  precisely  in  tracing  the  shrewdness  with  which  such 
fabricators  as  the  author  of  Gut  de  Warwick  adapted  historic 
names 12  to  present  purposes,  that  some  of  the  most  interesting 
problems  in  connection  with  mediaeval  romance  have  yet  to  be 
worked  out.  In  general,  however,  the  author  of  Gut  patched  to- 
gether his  borrowings  from  purely  romantic  and  pietistic 
material. 

No  source  study  of  this  long-winded  romance  has  yet  been 
made,  but  it  is  possible  to  define  certain  familiar  motifs  and 
details  and  to  illustrate  the  author's  method  of  work.  The  initial 
description  of  Felice,  taught  the  Seven  Arts  by  wise  men  of 
Toulouse,  and  of  Guy,  with  whom  thirty  maidens  fell  simul- 
taneously in  love,  is  an  obvious  courtly  elaboration.  Felice  is  a 
typical  Proud  Princess  who  rejects  a  more  lowly  born  suitor, 
and  Guy,  swooning  with  love  sickness,  an  altogether  French  and 

12  Brown,  p.  22,  noted  a  possible  Irish  original   (Collbran)   for  the  name 
Colbrand  and  found  in  it  confirmation  of  the  Brunanburh  theory. 


I34  MEDIAEVAL   ROMANCE    IN   ENGLAND 

not  a  mediaeval  English  lover.  After  he  and  his  young  friends 
are  knighted,  in  a  scene  which  Hall  {Horn,  note  to  1.  499)  thought 
was  directly  imitated  from  that  in  Horn,  Guy  in  his  quest  for 
fame  goes  to  Rouen  and  wins  there  in  tourney  a  priceless  falcon, 
a  steed,  and  greyhounds,  to  say  nothing  of  the  Princess  Blaunch- 
flour,  whom  he  rejects.  Her  name  here  is  used  incidentally,  as 
later  on  is  that  of  Amis,  a  gallant  young  lord  who  aids  Guy  in 
rescuing  friends  from  prison,  but  the  use  of  the  names  suggests 
memories  of  the  two  famous  romances,  Floris  and  Amis..  What- 
ever the  relationship  between  the  Middle  English  versions  of 
Guy  and  Amis,  the  fundamental  influence  of  the  older  legend  is 
evident  besides  in  the  way  in  which  the  Friendship  motif  is 
embodied  in  Guy.  Though  less  moving  and  more  militant  than 
Amis,  Guy  is  as  much  a  romance  of  friendship  as  of  love,  for  it 
tells  at  length  of  the  sworn  fellowship  of  Guy  and  Tirri,  a  devo- 
tion which  began  with  Guy's  rescue  of  Oisel,  Tirri's  lady,  from 
outlaws,  and  went  on  through  other  rescues  of  Tirri  from  prison, 
and  of  Oisel  from  a  forced  marriage  to  the  treacherous  Duke 
Otun  of  Pavia.  In  the  Gesta  Romanorum  and  versions  related 
to  it,  the  story  of  Guy  and  Tirri  was  of  primary  importance. 

The  adventures  of  Guy  on  the  Continent  and  in  the  East  be- 
long to  a  far-ranging,  geographical  type  of  story  of  which  Beves 
of  Hampton  is  a  notable  example.  Most  realistic,  perhaps,  are 
those  episodes  which  tell  of  Guy's  capture  of  the  German 
Emperor,  of  the  fight  with  his  host,  Earl  Florentin,  and  of  his 
meeting  with  Tirri  near  the  city  of  Spires. 

The  first  adventure  takes  place  in  Argonne.  Duke  Segwyn  is 
unjustly  attacked  by  the  German  Emperor,  and  Guy  goes  to  his 
assistance.  After  numerous  battles  the  Duke  learns  from  a  spy 
that  the  Emperor  plans  a  hunt,  and  Guy  vows  to  bring  him  to 
dine  with  Segwyn.  The  next  day,  at  the  moment  when  the 
Emperor  sees  himself  surrounded  with  armed  men,  Guy  ap- 
proaches, olive  twig  in  hand.  The  Emperor  somewhat  sur- 
prisingly accepts  the  invitation  to  enter  the  city  and  there  he  is 
royally  served  by  his  formidable  vassal.  The  next  day  the  Duke, 
clad  only  in  his  shirt,  and  with  a  rope  around  his  neck,  pleads 
before  the  Emperor  for  peace.  He  is  forgiven  and  is  presently 
wedded  to  the  Emperor's  daughter.  Throughout  all  the  narra- 
tive of  Guy,  German  scenes  and  characters  are  stressed,  but  the 


GUY   OF    WARWICK  13  5 

penance  of  the  Duke  recalls  a  French  incident.  His  appearance 
and  his  plea  suggest  comparison  with  the  historic  story  of  the 
burghers  of  Calais  before  Edward  III  in  1347,  although  earlier 
instances  of  similar  scenes  can  be  cited.13 

The  second  adventure  is  likewise  sufficiently  graphic  to  sug- 
gest a  definite  though  not  yet  identified  source.  Guy,  who  has 
pursued  a  boar  into  Brittany,  kills  the  beast  and  slays  at  the 
same  time  the  arrogant  son  of  Earl  Florentin  who  has  been  sent 
to  bring  the  hunter  to  court.  Guy  thereafter  accepts  the  hos- 
pitality of  Florentin  and  is  in  his  hall  when  the  dead  youth  is 
brought  in.14  There  follows  a  dramatic  scene.  Florentin  learns 
the  identity  of  the  slayer  and  hurls  an  andiron  at  Guy's  head. 
Guy  escapes  with  difficulty  from  the  castle  and  in  a  subsequent 
fight  with  his  pursuers  generously  remounts  the  old  earl  for  the 
sake  of  the  dinner  that  has  been  given  him.  There  are  many 
parallels  to  the  fury  of  Florentin  in  his  attack  on  a  guest  pro- 
tected by  the  laws  of  hospitality  or  of  truce,15  but  no  parallel  to 
the  episode  as  a  whole  has  yet  been  noted.  Nearest  to  it  is  the 
episode  in  Floovent,  11.  1046-1194,  in  which  Richier,  who  has 
unwittingly  slain  the  son  of  his  host,  Emelons,  proves  his  inno- 
cence by  fighting  and  receives  the  pardon  of  Emelons. 

It  is  futile  to  attribute  anything  in  Guy,  save  the  mere  art  of 
compilation,  to  its  author.     His  method  of  borrowing  may  be 

13  G.  Neilson,  "  Submission  of  the  Lord  of  the  Isles  to  James  I:  Its  Feudal 
Symbolism,"  Scottish  Antiquary,  xv,  113,  noted  the  surrender  of  Milan  in 
1162  as  told  by  a  Flemish  chronicler,  Rerum  Germ.  Belgicum,  vi,  183  (1607): 
the  rabble  appeared  in  their  shirts  and  with  ropes  around  their  necks.  In 
romance  literature  the  chivalric  penance  appears  in  the  Voeux  du  Paoun, 
1300-05,  in  the  Scotch  translation,  the  Buik  of  Alexander,  in  Richard  Coeur 
de  Lion,  etc. 

14  This  adventure  is  one  of  those  illustrated  in  the  Decretals,  fol.  16,  17. 
The  OF.  text  may  be  found  in  MS.  Regent  F  ix,  fol.  133. 

15  Boje,  Beuve  de  Hantone,  p.  90,  cited  thirty-one  instances.  Especially 
important  are  those  in  Ogier,  p.  174  (ed.  Barrois,  1874)  and  in  Octavian. 
Further  correspondences  between  Beves  and  Guy  as  noted  by  Boje  are  as 
follows:  (here  §  numbers  refer  to  Boje,  page  references  to  Guy) ;  §  2,  Guy's 
coloring  of  his  face,  p.  328;  §  3,  Herhaud's  disguise  as  a  palmer,  p.  415;  Guy's 
stay  at  the  court  of  a  foreign  king,  p.  167  ff.,  a  long  story  which  includes  the 
episode  in  which  Guy  is  sent  on  a  Message  of  Death,  p.  217  (Boje,  Uriasbrief, 
79) ;  §  7,  the  carbuncle  gleaming  pn  a  helmet,  p.  590,  or  on  the  roof,  p.  657; 
§  12,  the  knife-throwing;  §  13,  the  dreadful  prison,  p.  334;  §  15,  the  vain  pur- 
suit, p.  313;  §  16,  the  idols  beaten,  p.  212;  §  25,  the  freeing  of  the  beloved,  p. 
343;  §  27,  the  killing  of  the  king's  son  at  a  game  of  chess,  p.  427;  §  31,  the 
old  age  and  death  of  the  hero,  p.  612. 


136  MEDIAEVAL   ROMANCE    IN    ENGLAND 

illustrated  by  his  account  of  the  meeting  of  Guy,  the  penitent, 
and  his  unfortunate  friend  Tirri,  the  episode  which  became  so 
well-known  through  the  version  already  referred  to  in  the  Gesta. 
Of  the  antiquity  of  this  tale  there  is  no  question,  for  it  is  the 
story  of  King  Guntram's  Dream  told  in  the  eighth  century  by 
Paul  the  Deacon  in  his  History  of  the  Langobards.  Paul's  story 
remained  essentially  unchanged  in  all  the  mediaeval  chronicles 
recording  it  and  slipped  from  one  of  them  into  the  French  Gut.16 
The  romance  version  simply  altered  the  names  of  the  personages 
and  omitted  the  description  of  the  watching  soldier  who  draws 
his  sword  to  make  a  bridge  across  the  stream  for  the  soul  of  the 
sleeping  Guntram.  Likewise  the  Treasure  incident  in  the  ro- 
mance is  an  unmitigated  bit  of  borrowing. 

The  whole  paraphernalia  of  romance  appears  in  Guy.  For 
supernatural  elements  the  author  uses  chiefly  dreams,  dragons, 
and  giants.  In  dreams  angels  give  convenient  information  or 
wild  animals,  attacking  the  hero,  give  intimation  of  his  danger. 
Gigantic  champions  appear,  such  as  the  Saracen  Amoraunt  or  the 
African  Colbrond.  In  the  tale  of  Guy's  fight  with  Amoraunt,  a 
story  still  popular  in  the  seventeenth  century  (Percy  Folio  MS. 
ii,  136-43),  the  hero's  magnanimity  is  made  apparent  by  the  per- 
mission he  grants  Amoraunt  to  drink,  a  permission  later  refused 
to  Guy.  A  certain  almost  humorous  exaggeration  occurs  in  the 
account  of  Colbrond's  coming  to  combat  "  so  michel  5at  non  hors 
mist  him  bere,"  and  bringing  more  than  two  hundred  weapons. 
Amazing  arms  are  possessed  by  Guy  and  his  adversaries,  Alexan- 
der's helmet,  King  Clarel's  hauberk,  the  swords  of  Hector  and 
Hercules.  The  hero  fights  later  with  a  mighty  dragon  come 
"  out  of  Irlonde."  Like  Ywain,  Guy  saves  a  lion  from  a  dragon 
and  henceforth,  until  it  is  treacherously  slain,  it  serves  him  with 
doglike  devotion.17 

The  Eastern  elements  in  the  story  are  commonplace.  Guy 
goes  to  the  relief  of  Constantinople  when  it  is  besieged  by  a 
cruel  Sultan;  inevitably  Guy  defeats  the  heathen  hordes;  the 

16  Hibbard,  Romanic  Review,  1913,  iv,  189. 

17  The  lion  episodes  are  illustrated  in  the  Taymouth  Horae,  fol.  12-14, 
and  in  the  Smithfield  Decretals,  fol.  80-85.  For  studies  on  Grateful  Animals 
see  O.  M.  Johnston,  "  Ywain,  the  Lion  and  the  Serpent,"  Zts.  f.  frz.  Spr.  u. 
Lit.  xxxi,  157-66  (1907);  A.  C.  L.  Brown,  "Knight  of  the  Lion,"  PMLA. 
xx,  702;  G.  Baist,  "  Der  dankbare  Lowe,"  Rom.  Forsch  xxix,  317-19  (I911)- 


GUY   OF    WARWICK  I37 

Sultan  rails  upon  his  gods  and  breaks  his  idols ;  Guy  is  sent  on 
a  message  of  death  to  the  Sultan  but  beheads  him  as  he  sits  in 
his  splendid  pavilion.  Guy's  last  eastern  fight  is  at  Alexandria 
where  he  serves  as  champion  of  King  Triamour  whose  son  has 
killed  the  son  of  another  Sultan  at  a  game  of  chess.  In  all  this 
the  setting  and  the  abuse  of  the  Saracens  are  characteristic  of 
the  Crusading  spirit  in  romance,  but  the  episodes  have  nothing 
of  Eastern  character.  The  offer  to  Guy  of  the  hand  of  the 
Christian  Emperor's  daughter,  the  jealousy  of  the  false  steward, 
Morgadour,  his  accusation  to  the  Emperor  that  Guy  is  dishonor- 
ing the  Princess,  make  a  somewhat  absurd  combination  of  con- 
tradictory motifs,  some  of  which  seem  borrowed  from  Horn  and 
Beves. 

The  latter  part  of  the  story  changes  wholly  from  the  romantic 
tone  to  the  pietistic.  As  one  of  the  most  famous  knights  of  the 
world  Guy  weds  Felice,  only  to  leave  her  during  the  wedding 
festivities  from  a  sudden  overwhelming  sense  of  contrition  for 
the  worldliness  of  his  life.  In  this  part,  as  in  the  account  of 
Guy's  return  as  a  poor  and  unknown  pilgrim  to  beg  at  his  own 
castle  gates,  the  influence  of  the  St.  Alexis  legend  on  the  probably 
monastic  author  has  been  recognized.18  Whatever  may  have 
been  the  relationship  of  the  old  French  texts  of  legend  and  ro- 
mance, it  is  possible  in  the  Middle  English  version  to  feel  some 
direct  connection.  For  instance,  in  the  Laud  MS.  of  St.  Alexis 
(EETS.  lxix)  the  departing  young  husband  says: 

"  And  half  5e  godenesse  <5at  I  do 
Graunte  <5ee  god  almi3th."    (Alexis,  1.  233) 

In  Guy  he  uses  almost  the  same  words : 

"  And  of  all  the  goodnesse  that  I  doo  shall 
I  graunte  the  euere  haluendell."     (Guy,  Auch.  MS.  7430) 

It  is,  perhaps,  merely  a  humorous  chance  that  brings  into  the 
subsequent  story  of  St.  Alexis  an  account  of  Jonah  and  into  that 
of  Guy  the  meeting  with  Earl  Jonas. 

18  Groeber,  Grundriss,  n,  776.  For  the  Alexius  legend  see  Wells,  Manual, 
p.  809.  The  Middle  Eng.  texts  were  listed  by  C.  Brown,  Register,  vol.  n, 
Index. 


I38  MEDIEVAL   ROMANCE   IN   ENGLAND 

The  story  of  the  ultimate  return  of  Guy  to  England,  of  the 
king's  entreaty  that  the  poor  pilgrim  undertake  the  fight  with 
Colbrand,  of  the  old  man's  heroic  victory  and  swift  withdrawal 
to  a  hermitage,  was  referred  by  Deutschbein  (p.  225)  to  the  epic 
type  preserved  in  the  Montage  Guillaume.19  In  this  the  old 
warrior  William  hears  in  his  hermitage  of  King  Louis's  desperate 
need  of  a  champion  against  the  heathen  leader  whose  army  is 
besieging  Paris.  William  overthrows  the  enemy  and  returns 
unrecognized  to  his  hermitage,  though  the  king  learns  subse- 
quently the  name  of  his  deliverer.  The  same  type  of  story 
appears  in  the  account  of  the  monk  Ogier's  last  fight  against  the 
heathen  besieging  Meaux.20  In  actual  life  many  a  war-worn  old 
hero  must  have  similarly  renounced  the  world,  but  it  is  probable 
that  the  tales  of  such  triumphant  exploits  were  chiefly  the 
product  of  monastic  imagination  at  work  on  epic  legend.  The 
resulting  type  of  story  seems  to  have  been  clearly  defined  by  the 
middle  of  the  twelfth  century  and  to  have  been  amply  familiar 
to  the  original  French  author  of  Guy  of  Warwick.21 

BIBLIOGRAPHY 

Texts:  (i)  A,  a,  Auchinleck  MS.  ed.  Turnbull,  Abbotsford  Club, 
1840;  Zupitza,  EETSES.  42,  49,  59;  (2)  C,  Caius  College,  Cbg.  107,  ed. 
Zupitza,  ibid.;  (3)  S,  Sloane  MS.  1044,  ed.  Zupitza,  Sitz.  d.  Wiener, 
Akad.  d.  Wiss.  Ph.  Kl.  74,  624-29  (216  lines);  (4)  P,  Add.  MS.  14408, 
ed.  Phillips,  Middle  Hill,  1838;  reprinted  by  Turnbull,  1840;  (5)  C, 
MS.  Ff.  2,  ^S,  Cbg.  Univ.  Libr.  ed.  Zupitza,  EETSES.  25,  26  (11,  976 
lines).  Early  printed  editions:  W,  Douce  fragments,  no.  20,  Bodleian; 
d,  Copland's  edition.  Cf.  Zupitza,  Sitzungsberichte,  p.  641,  632-39; 
Schleich,  Palaestra,  139. 


19  For  bibliography  to  1908  see  Bedier,  Les  Legendes  fipiques,  1,  90.  The 
whole  volume  is  devoted  to  an  epochal  study  of  Le  Cycle  de  Guillaume 
d' Orange. 

20  See  Bedier,  "  Ogier  le  Danois  et  l'Abbaye  de  St.  Faron  de  Meaux," 
Ltgendes,  u,  305-310.  Of  special  interest  is  the  evidence  of  the  familiarity 
of  Alexander  Neckham,  the  Englishman  who  taught  in  Paris  in  1180-86, 
with  this  Ogier  legend  (cf.  De  naturis  rerum,  ch.  CLVII,  Rolls  Series,  1863). 
See  T.  Walker,  Die  altfrzt  Dichtungen  vom  Helden  int  Kloster,  1910. 

21  All  the  Middle  Eng.  texts  of  Guy  refer  to  Tirri's  coming  to  England 
after  Guy's  death  and  to  his  taking  Guy's  body  back  to  Lorraine  where  he 
built  a  splendid  abbey  in  which  monks  were  forever  to  sing  for  Guy's  soul. 
See  B6dier's  remarks  on  the  tombs  of  Amis  and  Amiloun,  Legendes,  n,  170  ff., 
and  of  Ogier  and  his  friend  Benoit,  n,  294  ff. 


GUY   OF   WARWICK 


139 


Studies   and    Analogues:     Billings,    Guide,    25-32;   Wells,  Manual 

15-19,  764. 

Acher,  J.  "  Acquisition  des  MSS.  de  Mr.  Dunn  par  le  musee  Brit- 
tanique,  Gui  de  Warwick,  Add.  MS.  38662,"  Revue  des  langues 
romanes,  lvi,  5i3~i4  (i9i3)- 

Ashdown,  M.  "  Single  Combat  in  English  and  Scandinavian  Romance." 
MLR.  xvii,  113-130  (1922). 

Brown,  A.  C.  L.  "  The  Source  of  a  Guy  of  Warwick  Chap-Book,"  Jour. 
Ger.  Phil,  ill,  14-23  (1900). 

Chap-Books  and  Broadsides,  Harvard  Univ.  Bibliographical  Contributions, 
No.  56,  p.  29,  No.  484-89  (1905). 

Crane,  R.  "  Vogue  of  Guy  of  Warwick  from  the  Close  of  the  Middle 
Ages  to  the  Romantic  Revival,"  PMLA.  xxx,  125-94  (191 5). 

Deutschbein,  M.  Sagengeschichte,  pp.  214-34  (1906). 

Herbert,  J.  A.  "  An  Early  Manuscript  of  Guy  of  Warwick,"  Romania, 
xxxv,  68-81  (1906). 

Hibbard,  L.  A.  "  Guy  of  Warwick  and  the  Second  Mystere  of  Jean 
Louvet,"  Mod.  Phil,  xiii,  181-87  (191 5). 

Kolbing,  E.  "  Amis  and  Amiloun  und  Guy  of  Warwick,"  Eng.  Stud,  ix, 
477-78  (1886). 

Jenkins,  T.  A.  "  A  New  Fragment  of  the  Old  French  Gui  de  Warwick," 
Mod.  Phil,  vii,  593-96  (1910). 

Liebermann,  F.  "  Guy  of  Warwicks  Einfluss,"  Herrigs  Archiv.  107,  107. 

Mau,  P.  Gydo  und  Thyrus,  ein  deut.  Ausldufer  d.  altfr. -mitt  elengl, 
Freundschaftsromans  Guy  v.  Warwick.  Diss.   (69  pp.)  Jena,  1909. 

Moller,  W.  Uritersuchungen  uber  Dialekt  u.  Stil  des  mitteleng.  Guy  of 
Warwick  in  der  Fassung  der  Auchinleck-Handschrift  u.  uber  das 
Verhaltnis  des  strophischen  Teiles  des  Guy  zu  der  mitteleng.  Ro- 
manize Amis  und  Amiloun.  Diss.  Konigsberg,  191 7  (107  pp.).  For 
Reinbrun,  see  pp.  36-47. 

Penn,  H.  C.  "  On  the  Dialect  of  the  Auchinleck  and  the  Caius  MSS.  of 
Guy  of  Warwick,"  PMLA.  xx,  p.  xxviii  (1905). 

Reeves,  W.  "  The  So-Called  Prose  Version  of  Guy  of  Warwick,"  MLN. 
xi,  202-04  (1896). 

Robinson,  F.  N.  "  On  Two  Manuscripts  of  Lydgate's  Guy  of  Warwick," 
Harvard  Studies,  v,  177-220  (1897).  "Irish  Lives  of  Guy  of  War- 
wick," Zeit.  f.  celt.  Phil.  vi.  Halle,  1907. 

Tanner,  A.  Die  Sage  v.  Guy  von  Warwick.  Diss.  68  pp.  Heidelberg, 
1877. 

Weyrauch,  M.  Die  mitteleng.  Fassungen  der  Sage  v.  Guy  of  Warwick 
u.  ihre  altfr.  Vorlage.  Diss.  Breslau,  1901.  Rev.  Eng.  Stud,  xxxii, 
405   (1903).  ## 

Winneberger,  O.  Uber  d.  handschriftenverhaltnis  d.  altfr.  Guy  de  War- 
wick.    Frankf.  a.M.  1889. 
Zupitza,  J.  "  Zur  Literaturgesch.  des  Guy  of  Warwick."     Wien,   1873 
(Sitzungsber,  d.  Kais.  Akad.  d.  Wiss.  Phil.  Hist.  Kl.  lxxiv,  623-45). 


REINBRUN 

Versions.  The  romance  of  Reinbrun,  Gij  sone  of  Warwicke, 
is  found  in  Middle  English  only  in  the  Auchinleck  manuscript 
(1330-1340).  It  contains  1524  lines  in  twelve-line  tail-rime  stan- 
zas and  was,  perhaps,  written  in  South  Warwickshire  (Brandl, 
Grundriss,  §  37).  Like  Guy  of  Warwick  it  is  a  redaction  of  an 
older  inedited  version  in  Old  French.  Moller  (pp.  36-45)  urged 
that  its  author  should  not  be  identified  with  the  redactor  of  the 
a  version  of  Guy. 

After  a  pious  plea  for  blessing  on  those  who  read,  and  a  rapid 
summary  of  the  events  leading  to  Reinbrun's  birth,  the  story 
begins  with  a  colorful  account  of  the  "  richesse  "  brought  to  Lon- 
don by  the  merchants  to  whom  Athelston  gives  permission  to  fare 
through  his  land.  Arrived  at  Wallingford,  they  send  a  Spanish 
mule  to  Heraud  who  was  left,  after  Guy's  death,  in  charge  of  the 
town  and  of  Guy's  seven  year  old  son.  After  they  have  spoken 
"  stille  "  with  the  porter,  they  steal  the  child  and  set  sail  for 
Russia.  But  a  great  storm  drives  them  to  Africa  where  Reinbrun 
is  given  to  King  Argus's  daughter,  a  maid  who  knows  much  of 
"  menstralcie  "  and  "  of  romance  reding." 

In  England,  meanwhile,  Heraud,  grieving  desperately  for  the 
lost  child,  goes  to  the  king's  parliament  and  advises  Athelston 
concerning  the  defence  of  the  realm  against  a  threatened  attack 
of  the  Danes.  Heraud  is  accused  by  envious  lords  of  having  sold 
Reinbrun  for  his  weight  in  gold  and  of  being  a  traitor  to  his  lord. 
Edgar,  Heraud's  steward,  challenges  for  his  master's  sake  the 
Duke  of  Cornwall  and  Heraud  himself  starts  forth  to  seek  Rein- 
brun. Arrived  in  Africa,  he  is  so  long  imprisoned  by  the  Emir 
Parsan  that  his  hair  grows  down  to  his  girdle,  "  Grisliche  he 
was  of  siste."  When  the  Emir  at  last  learns  that  his  prisoner 
once  served  Guy  of  Warwick,  he  arrays  Heraud  richly  and  sends 
him  forth  to  fight  the  forces  of  King  Argus.  In  a  great  battle  he 
almost  kills  Argus  but  the  latter  is  saved  by  the  youth  Reinbrun. 
In  the  midst  of  a  great  fight  Heraud  learns  the  name  of  the  boy 

140 


REINBRUN  I4I 

and  recognition  follows.    They  return  to  the  Emir,  Argus  is  slain, 
and  Reinbrun  and  Heraud  start  for  England. 

The  next  adventure  tells  of  the  coming  of  these  two  knights 
to  a  castle  owned,  as  the  porter  tells  them,  by  a  most  sorrowful 
lady.  Her  lord,  Amis  of  Mountayne,  once  Guy's  young  friend, 
has  disappeared,  captured  presumably  by  an  old  enemy  of  Guy's. 
Reinbrun  goes  to  seek  Amis,  swims  a  river  "  sterne  and  grim,'*' 
sees  a  castle  with  crystal  walls,  rafters  of  cypress,  jasper  posts, 
a  great  carbuncle  shining  from  the  front,  and  with  a  tree  full  of 
singing  birds  standing  near  the  gate.  Reinbrun  learns  from 
Amis  whom  he  finds  alone  in  the  hall  that  this  is  the  home  of  a 
fairy  knight,  Gayer,  and  that  in  this  house  one  may  stay  without 
ever  growing  old.  Reinbrun  catches  up  a  magic  sword  and  the 
room  shines  with  light.  Pursued  by  Gayer,  Reinbrun  fights  with 
him  as  "  fresch  ase  grehonde  to  hare,"  until  Gayer  begs  in  Guy's 
name  for  mercy.  Reinbrun  restores  Amis  to  his  lady,  and  again 
the  youth  and  Heraud  start  on  their  way. 

In  Burgundy  the  country  is  devastated  and  the  Earl,  shut  up 
in  one  castle,  is  aided  by  a  gallant  youth  of  twenty.  None  may 
pass  him  without  fighting.  In  a  great  combat  between  Reinbrun 
and  the  youth  it  is  discovered  that  the  latter  is  Heraud's  son, 
Haslak.  Taunted  with  his  father's  long  absence,  the  boy  had 
dubbed  himself  knight  and  gone  forth  to  seek  Heraud.  After 
a  joyful  recognition,  the  three  men  return  to  the  Earl,  defeat  his 
foes  and  finally  set  sail  for  England. 

Origin.  The  wholly  unoriginal  quality  of  Reinbrun  is  evi- 
dent from  this  summary.  The  account  of  the  merchants  who 
steal  the  child  and  the  tale  of  Heraud's  long  search  for  him,  are 
in  palpable  imitation  of  the  Tristan  legend.  The  selling  of  the 
boy  to  the  Saracen  king  and  the  princess's  care  for  him  recall 
the  story  of  Beves.  The  fights  of  Reinbrun  first  with  his  foster 
father  and  later  with  Heraud's  son,  are  somewhat  unhappily 
duplicated  variants  of  the  Father  and  Son  Combat.  Reinbrun's 
adventure  in  releasing  Amis  from  the  fairy  castle  is  interesting 
chiefly  because  of  its  clear  indications  of  the  fairy  landscape 
(the  dark  river,  the  impossible  richness  of  the  castle,  the  tree 
of  singing  birds)  and  for  the  English  poet's  mild  attempts  at 
rationalization.     In  the  Old  French  text  of  Gut  in  the  College 


1 42  MEDIEVAL   ROMANCE   IN   ENGLAND 

of  Arms,  Amis  tells  his  friend  that  without  a  kiss  from  Reinbrun, 
Amis  will  be  turned  into  a  serpent.  This  suggests,  though  in 
oddly  inverted  form,  Le  Bel  Inconnu  and  the  variants  sum- 
marized by  Schofield  (Libeaus  Desconus,  Harv.  Stud,  iv,  1896) 
which  tell  of  the  serpent  lady  who  can  only  be  disenchanted  by 
a  kiss.  But  until  the  Old  French  texts  have  been  more  ex- 
tensively printed  it  is  impossible  to  say  how  far  the  process 
of  omission  and  of  rationalization  has  gone  in  the  Middle  English 
Reinbrun, 

BIBLIOGRAPHY 

For  texts  and  studies  see  Guy  of  Warwick. 


ATHELSTON 

Versions.  A  single  manuscript  written  in  the  North  Midland 
dialect  in  the  last  half  of  the  fourteenth  century,  contains  the 
only  known  version  of  Atheist  on  (Zupitza,  p.  xiv).  The  initial 
prayer  for  grace,  the  familiar  appeal,  "  lystnes,  lordyngs,"  the 
rime  phrase,  "  I  wol  Sow  tel,"  the  use  of  the  popular  twelve-line, 
tail-rime  stanza  with  occasional  lapses  into  stanzas  of  eight,  six 
or  even  four  lines,  and  the  use  likewise  of  a  great  number  of 
well-worn  alliterative  phrases,  are  signs  of  professional  min- 
strelsy.1 The  author  refers  to  his  source  as  "  in  book  iwreten  " 
or  "  in  romaunce  as  we  rede,"  and  also  as  a  tale  "  men  me  told." 
This  discrepancy  may  well  indicate,  despite  the  conventionality 
of  the  phrases,  the  use  of  both  oral  and  written  sources. 

The  plot  of  Atheist  on  is  strikingly  simple  and  unified.  A 
jealous  courtier  makes  accusation  against  King  Athelston's 
relatives.  The  king  orders  their  execution  but  is  forced  by 
Alryke,  Archbishop  of  Canterbury,  to  allow  his  sister,  his  brother- 
in-law,  and  their  children,  the  ordeal  by  fire.  This  they  pass 
successfully;  the  traitor's  wickedness  is  revealed  and  he  suffers 
the  doom  he  had  prepared  for  others.  In  this  direct  and  graphic 
story  there  are  no  foreign  elements.  The  names  are  typically  old 
English  for  Athelston  and  his  sworn  brothers,  Alryke,  Wymound, 
and  Egelan(d),  and  the  poem  abounds  in  local  color.  No  ro- 
mance refers  more  familiarly  to  places  of  interest  in  London,  to 
Westminster,  to  Charing  Cross,  to  Fleet  Street.  The  detailed 
mention  of  towns  through  which  royal  messengers  take  their  way 
from  London  to  Canterbury  or  Dover,  shows  that  the  author 
must  have  had  more  than  a  hearsay  knowledge  of  the  famous 

1  Zupitza 's  invaluable  list  of  minstrel  commonplaces,  in  conjunction  with 
those  given  by  Schmirgel,  Beves  of  Hampton  (EETS.),  by  Kolbing,  Arthour 
and  Merlin,  Sir  Tristrem,  by  Fuhrmann,  Die  allit.  Sprachformeln  in  Morris' 
E.  E.  Allit.  Poems  and  Sir  Gawayne  and  the  Grene  Knight,  Kiel,  1896,  Hall's 
Horn,  Schleich's  Ywain  and  Gawain,  offer  much  material  for  that  possible 
classification  through  diction  of  the  different  schools  of  romance  makers  in 
England  which  was  suggested  by  Dr.  Rickert,  Emare,  note  to  1.  9. 

143 


I44  MEDIEVAL   ROMANCE    IN    ENGLAND 

pilgrim  route.  In  style  also,  in  its  omission  of  all  the  elements 
of  chivalry  and  romance,  in  its  rude  vigor,  its  occasional  bru- 
tality, its  liking  for  scenes  of  tumult  and  rapid  action,  in  its 
simple  motives  and  naive  credulity,  the  tale  has  distinctive 
qualities  of  English  popular  fiction  which  are  found  in  such 
Middle  English  romances  as  Havelok,  Richard  Coeur  de  Lion, 
and  Gamelyn.  An  occasional  line  like  that  which  tells  how  the 
queen  in  a  moment  of  danger  casts  off  her  "  gerlondes  of  cheryes," 
has  the  apt  brevity  of  a  ballad  phrase.  Ballad-like  also  are  the 
wild  horse-killing  rides,  the  hasting  messengers,  and  the  effective 
suspense  of  the  three  ordeals  by  fire. 

Origin.  In  name  at  least  the  Athelston  of  the  romance  has 
been  identified  with  the  famous  yEthelstan,  king  of  England 
from  925  to  939,2  conqueror  at  Brunanburh  (937),  and  the 
storied  king  for  whom  Guy  of  Warwick  fought  with  the  Danish 
giant  Colbrand.  He  was  the  hero,  according  to  William  of 
Malmesbury,  of  many  old  songs  and  legends,  and  among  these 
Zupitza  (p.  xiv)  noted  two  which  were  concerned  with  con- 
spiracies against  the  king.  In  these  tales,  however,  there  is  no 
single  point  of  agreement  with  Athelston  except  in  the  common 
likeness  of  all  stories  of  kings,  jealous  courtiers,  and  false  ac- 
cusations. As  a  matter  of  fact  Athelston  in  the  romance,  though 
he  bears  so  famous  a  name,  is  not  the  real  hero  of  the  story; 
that  place  is  taken  by  Archbishop  Alryke,  whose  fearless  defiance 
of  the  king  was  possibly  modeled  on  the  great  traditional  ex- 
ample of  Thomas  Becket  in  his  quarrel  with  Henry  II.  In  any 
case,  as  Gerould  has  pointed  out,  Athelston  has  a  strong  ecclesi- 
astical bias;  it  exalts  the  clergy  above  royalty  itself,  and  its 
climactic  scene  is  in  the  nature  of  religious,  rather  than  romantic, 
marvel. 

These  facts  give  the  clue  to  the  origin  of  the  romance  in  the 
famous  Winchester  legend  of  Queen  Emma  and  the  Ploughshares 
(Hibbard,  p.  227).  The  tale  was  first  told  (c.  1200)  by  Richard 
of  Devizes,  a  monk  of  Winchester,  and  was  copied  from  him  by 
various  Benedictine  chroniclers  such  as  Ranulf  Higden  of  Chester 
(1327)   and  Richard  of  Cirencester,  a  monk   of  Westminster 

2  See   Beaven,   "  The    Regnal   Dates   of   Alfred,   Edward   the    Elder,   and 
Athelstan,"  Eng.  Hist.  Review,  191 7,  xxxii,  Si 7-31. 


ATHELSTON  145 

(1355-1400).  In  this  legend  Emma,  mother  of  Edward  the  Con- 
fessor, was  accused  of  various  crimes  and  conspiracies  by 
Robert  of  Jumieges,  Edward's  special  favorite.  She  exculpated 
herself  by  the  ordeal  of  the  fiery  ploughshares  and  together  with 
the  good  Bishop  Alwyn  of  Winchester,  who  had  been  accused 
with  her,  received  the  king's  penitent  submission. 

In  structure  Atheist  on  is  practically  identical  with  this  story. 
In  details  the  romance  was  altered  for  the  deliberate  purpose  of 
disguise.  The  part  of  the  wicked  churchman,  Robert  of 
Jumieges,  was  given  to  the  layman,  Wymound;  the  part  of  the 
gullible  King  Edward  to  Athelston,  who  was  made  not  only  gul- 
lible but  brutal ;  Queen  Emma's  part  was  divided  between 
Athelston 's  wife  and  his  sister,  and  the  "whole  legend  was  shifted 
from  Winchester  to  Westminster.  In  this  transformation  of  the 
story  it  is  possible  to  detect  certain  details  which  could  only 
have  come  from  Westminster  monks  and,  in  all  probability,  from 
Richard  of  Cirencester  himself.  Of  such  sort  is  the  unusual 
name  Alryke  which,  in  an  obscure  Westminster  chronicle  used 
by  Richard,  is  the  name  of  the  monk,  elected  in  1050  by  his 
fellows  to  be  ruler  over  them,  who  was  displaced  by  Robert  of 
Jumieges.  With  true  ironic  fitness  it  is  in  the  romance  Alryke, 
as  Archbishop  of  Canterbury,  who  brings  to  nought  the  plans  of 
Wymound,  the  fictitious  counterpart  of  Robert. 

This  clever  manipulation  of  the  original  story  shows  clearly 
enough  that  Athelston  was  influenced  by  the  learning  and  the 
taste  of  some  Westminster  monk,  presumably  Richard.  But  the 
characteristic  style  of  the  romance,  the  introduction  of  such 
popular  incidents  as  the  swearing  of  "  brotherhood  " 3  by  the 
four  "  messengeres,"  Athelstan,  Wymound,  Egelan,  and  Alryke, 
and  the  rude  violence  of  the  king  to  his  wife,  must  be  referred  to 

3  For  general  discussions  of  blood  brotherhood  and  sworn  brotherhood  see 
Gerould,  p.  194;  Peebles,  "Blood  Brotherhood,"  pgr.  PMLA.  1913;  J.  Flach, 
Le  Compagnonnage  dans  les  Chansons  de  Geste,  Etudes  Romanes',  Paris, 
1891 ;  Stowell,  "  Personal  Relationships  in  Mediaeval  France,"  PMLA.  xxvin, 
388-416  (1913)  ;  G.  F.  Fundenburg,  Feudal  France  in  the  French  Epic, 
Princeton,  1918,  p.  67,  79-84.  Flach's  article,  which  also  appeared  in  his 
Origines  de  Vancienne  France  (Paris,  1893,  II,  427-90),  set  forth  Le  Comitat 
germain,  then  the  primitive  Scandinavian  rite  of  blood  brotherhood,  11,  439, 
then  Compagnonnage  sous  les  rois  francs,  and  in  the  last  two  chapters,  ex- 
amples of  poetic  friendships  such  as  those  of  Roland  and  Oliver  or  Amis  and 
Amile. 


I46  MEDLEVAL   ROMANCE   IN   ENGLAND 

minstrel  authorship.  Westminster,4  like  other  great  monasteries 
of  the  day,  kept  minstrels  in  pay,  and  at  Westminster,  we  must 
believe,  lived  the  unknown  minstrel  who  produced  the  present 
version  of  Atheist  on.  At  Winchester  itself,  as  late  as  1338,  the 
original  Emma  story,  the  "  gestum  Emmae  reginae  a  iudico  ignis 
liberatae,"  was  sung  by  the  minstrel  Herbert  in  the  hall  of  the 
priory  of  St.  Swithin's.5 


BIBLIOGRAPHY 

Texts:  MS.  175  Caius  College,  Cambridge,  ed.  Hartshorne,  Ancient 
Metrical  Tales,  Lond.  1829;  Th.  Wright,  Reliquiae  Antiquae,  Lond.  1895, 
11,  85  ff.;  J.  Zupitza,  Eng.  Stud,  xni,  331-414  (1889);  Lord  Francis  Har- 
vey, Corolla  S.  Eadmundi,  Lond.  1907,  p.  525  ff.  Trans.  Rickert,  Ro- 
mances of  Friendship,  p.   67  ff. 

Studies:  Cf.  Billings,  Guide,  pp.  32-36;  Wells,  Manual,  pp.  23-25. 
Gerould,  G.  H.  "  Historical  Reminiscences  in  the  ME.  Athelston,"  Eng. 

Stud,  xxxvi,  193-208  (1906). 
Hibbard,  L.   A.   "  Athelston,  A   Westminster  Legend,"  PMLA.   xxxvi, 

223-244  (1921). 
Zupitza,  J.  "  Die  Romanze  von  Athelston,"  Eng.  Stud,  xiv,  321-344 

(1890). 


4  Chambers,  Mediaeval  Stage,  I,  56:  L.  Pound,  "  The  Eng.  Ballads  and  the 
Church"  PMLA.  xxxv,  182   (1920). 

6  Warton,  Hist,  of  Engl.  Poetry,  1840,  p.  81 ;  see  also  Chambers,  op.  cit. 


RICHARD   COEUR   DE   LION 

Versions.  Three  specific  allusions  to  a  French  source  and  at 
least  ten  references  to  an  original  "  geste  "  or  "  boke,"  indicate 
that  the  Middle  English  romance  of  Richard  was  preceded  by 
an  Anglo-Norman  version.  The  allusion  to  a  French  source  is 
confirmed  by  the  preservation  of  a  number  of  phrases,  "  Seyn- 
yours,  tuez,"  and  even  of  whole  verses,  such  as  the  Angel's  cry 
to  the  Christians,  "  Suse,  Seynours,  has  tost  armes  "  (1.  3012), 
and  by  the  large  number  of  French  rimes  (Paris,  Rom.  xxvi, 
361,  n.).  That  this  original  poem  was  written  in  England  is 
evident  from  the  blatantly  partisan  nature  of  Richard  with  its 
praise  of  the  English  and  its  derision  of  the  French  as  braggarts 
and  cowards  (1.  3849-65).  As  a  date  for  it,  Paris  (p.  361)  sug- 
gested "about  1230";  Loomis  (JEGP.  xv,  456-8)  about  1250. 
Before  the  end  of  the  century  it  was  translated  into  English,  and 
it  was  known  presumably  in  this  form  to  the  author  of  the 
chronicle  called  Robert  of  Gloucester's  (B runner,  p.  73).  The 
evidence  of  distinctively  Kentish  forms  suggests  that  the  author 
was  a  man  of  Kent  (Loomis,  ibid,  p.  463).  Kolbing,  in  his 
edition  of  Arthour  and  Merlin  (Leipzig,  1890,  p.  lxxiii  ff.),  set 
forth  his  belief  that  this  Kentish  translator  was  identical  with 
the  author  of  the  two  Kentish  poems,  Arthour  and  Merlin  and 
King  Alisaunder,  other  biographical  romances  of  strangely-born 
heroes.1  In  these  texts  a  number  of  passages  are  parallel  to 
those  in  Richard.  Especially  notable  is  a  lyric  passage  on  the 
"  merye-tyme  of  May,"  on  bird-song  and  ladies'  bowers,  strewn 
with  "  red  roses  and  lylye  floures  "  (1.  3759-72),  which  is  closely 
related  to  similar  effusions  in  the  Kentish  poems.  But  it  must 
be  noted  that  the  Kentish  author  was  not  the  only  one  to  use 
such  passages,  and  that  in  the  muscular  and  occasionally  rather 
brutalized  narrative  of  Richard,  the  passage  stands  out  with  an 

1  Kolbing's  argument  from  the  similarity  of  theme,  so  far  as  this  depended 
on  the  supernatural  birth  of  the  hero,  is  discredited  if  Paris's  theory  be 
accepted  that  the  birth  story  in  Richard  is  a  late  accretion. 

147 


I48  MEDIEVAL   ROMANCE   IN   ENGLAND 

effect  of  startling  difference.  It  was  more  probably  due  to  a 
redactor  familiar  with  the  Kentish  romances  than  to  the  man 
who  translated  the  Anglo-Norman  original  of  Richard.  With 
the  exception  of  the  first  twenty-four  lines,  this  first  English 
version  is  represented  by  the  fragments,  696  lines,  in  the  (L) 
Auchinleck  manuscript  ( 1330-42 ).2 

In  addition  to  this  manuscript  there  are  six  others  which  pre- 
serve the  story  of  Richard  in  Middle  English.  All  these,  except 
(E)  Egerton  2862,  a  late  fourteenth-century  manuscript,  are  of 
the  fifteenth  century.  In  1509  and  1528  the  poem  was  printed 
by  Wynkyn  de  Worde,  and  again  in  1568-9  by  Thomas  Purfoote 
(Arber,  Stationers'  Registers,  Lond.  1875,  1,  179).  According  to 
Brunner,  the  latest  editor  of  the  romance,  these  texts  present 
two  versions,  a  represented  by  manuscripts  CB-W,  and  b,  by 
L-ED-AH,  which  differ  greatly  in  content.  The  nearest  ap- 
proach to  a  complete  version  is  Wynkyn  de  Worde's  edition  of 
1509.  After  a  brief  prologue  the  Auchinleck  manuscript  passes 
directly  to  a  resume  of  the  events  leading  up  to  the  Third  Cru- 
sade. The  manuscripts  of  the  a  type  amplify  the  story  by  be- 
ginning with  an  account  of  the  fairy  lady,  Cassodorien,  who 
reaches  England  in  a  marvelous  ship  with  ropes  of  silk,  samite 
sails,  and  ivory  mast ;  she  becomes  Richard's  mother,  and  stays 
until  she  is  forced  to  behold  the  Sacrament ;  she  then  flies  away. 
This  version  tells  also  of  the  Salisbury  tournament  where  Richard 
receives  the  stout  blows  of  two  Lincolnshire  knights,  Thomas  de 
Moulton  and  Fulke  Doilly,  and  welcomes  them  as  his  comrades 
in  arms,  of  Richard's  pilgrimage  to  the  Holy  Land,  of  his  im- 
prisonment by  the  King  of  Almayne,  and  of  Richard's  revenge. 
These  accretions  amount  to  1233  lines.  They  seem  due  chiefly 
to  the  redactor's  liking  for  oral  and  written  tradition  of  a  fabu- 
lous sort,  and  to  an  especial  desire  to  glorify  the  two  South 
Lincolnshire  knights,  whose  names  appear  in  records  from  1190 
to  1240.     In  what  has  been  accepted  as  the  translation  of  the 

2  Brunner  (p.  17  ff.)  did  not  accept  this  estimate  of  the  Auchinleck  MS. 
a9  indicative  of  the  original  character  of  the  romance.  Although  in  general 
he  agreed  with  Paris  that  it  was  approximately  faithful  to  historical  facts, 
he  accepted  the  wildly  unhistorical  account  of  Richard's  birth  as  part  of  the 
original  tale,  mainly,  it  would  seem,  because  MSS.  of  the  b  version  include 
lines  35-1268.  See  J  EG?,  xv,  458-62,  for  further  refutation  of  this  point  of 
view. 


RICHARD    COEUR   DE  LION  i49 

original  Anglo-Norman  version  these  names  do  not  occur,  nor 
does  history  know  of  them  as  famous  crusaders.  Their  glory 
in  Richard  must  have  been  inspired  by  a  minstrel,  who  was 
enjoying,  as  Ward  {Catalogue,  i,  946)  suggested,  the  patronage 
of  their  descendants.3  Loomis  (JEGP.  xv,  p.  465)  noted  that  the 
two  families  were  actually  connected  through  the  marriage  of 
Lambert  de  Multon  (d.  1247)  with  the  widow  of  Geffrey  de 
Oilli.  In  addition  to  these  earlier  episodes,  those  enumerated 
below  in  the  discussion  of  the  fabulous  elements  of  the  romance, 
are  likewise  to  be  ascribed  to  the  same  poet,  who  is  known  as 
the  "  interpolator  "  or  the  "  Lincolnshire  minstrel."  This  seems 
a  somewhat  unfortunate  nomenclature  since,  whatever  his  inter- 
ests, his  dialect  was,  in  Brunner's  opinion  (p.  48),  of  the  south- 
east, and  not,  as  Lincolnshire  would  suggest,  of  the  northeast. 
The  verse  form  of  the  two  versions,  a  and  b,  is  the  short  riming 
couplet.  At  the  beginning  of  the  Auchinleck  copy  there  are  two 
twelve-line,  tail-rime  stanzas. 

Origin.  Between  the  death  of  Richard  I  in  1199  and  the 
earliest  English  romance  accounts  of  his  life,  many  chronicles 
of  the  Third  Crusade  were  written  by  men  of  English,  French, 
and  Eastern  origin.  In  these  histories,  to  greater  or  less  degree 
according  to  the  national  bias  of  the  writers,  was  preserved  the 
record  of  the  wildly  reckless  deeds,  the  boisterous  courage,  the 
brutality  and  fantastic  chivalry,  which  made  Richard  of  England 
one  of  the  most  picturesque  of  royal  Crusaders.  Fact  in  his 
case  was  often  stranger  than  fiction,  as  subsequent  story-tellers 
were  quick  to  perceive.  In  consequence  their  tales  have  a  cer- 
tain historical  value,  sometimes  as  the  record  of  an  attitude  of 
mind  and  the  growth  of  a  personal  myth,  and  again  as  distorted 
but  recognizable  images  of  actual  fact.  The  shorter,  more  sober 
b  version  of  the  Middle  English  romance  of  Richard,  like  its 
various  antecedents,  undoubtedly  omits  much,  is  inexact  in 
chronological  detail,  and  somewhat  subject  to  patriotic  exaggera- 
tion concerning  its  hero,  and  to  depreciation  of  his  rivals  and 
enemies,  but  on  the  whole  the  narrative  is  fairly  authentic.  It 
tells  with  measurable  fidelity  of  the  preparations  for  the  Cru- 

3  For   the   complicated  history    of   the   Multone   family   see   N.    Neilson, 
A  Terrier  of  Fleet,  Lond.  1920,  p.  lxxxii  ff. 


l5o  MEDIEVAL   ROMANCE   IN   ENGLAND 

sades  in  Europe,  of  Richard's  stay  in  Sicily,  of  the  conquests  of 
Cyprus,  of  the  taking  of  Acre  (1191),  the  massacres  of  Saracen 
prisoners,  the  march  to  Jaffa,  the  attempted  fortification  of 
Ascalon,  the  relief  of  Jaffa,  largely  through  the  personal  prowess 
of  Richard,  and  his  summons  home  when  news  comes  of  the 
treacherous  activities  of  his  brother  John.  The  question, 
however,  as  to  how  much  of  this  record  is  due  to  the  extant 
chronicles  of  the  thirteenth  century,  such  as  Ambroise's  L'Estoire 
de  la  guerre  sainte,  11 90-2,  or  the  Itinerarium  Peregrinorum  et 
Gesta  Regis  Ricardi,  or  Richard  of  Devize's  De  Rebus  Ricardi 
I,  1189-92,  or  the  Imagines  Historiarum,  1 188-1202,  of  Ralph  of 
Diceto,  how  much  to  lost  chronicles,  and  how  much  to  memory 
of  the  actual  events,  has  long  been  a  matter  of  dispute.  Jentsch's 
attempt  to  find  in  the  Itinerarium  the  principal  source  of  the 
romance  was  disposed  of  by  Gaston  Paris  {Rom.  xxvi,  369-85), 
who  showed  that  the  alleged  correspondences  between  the  two 
are  necessarily  incidental  to  the  treatment  of  the  same  historic 
events  and  personages,  and  that  the  differences  in  the  order  of 
events,  the  respective  omissions  and  elaborations,  are  inexplicable 
if  the  Itinerarium  be  considered  the  source  of  Richard.  In 
Paris's  opinion  the  Anglo-Norman  poet  made  up  his  story  from 
oral  tradition  and  was  in  general  independent  of  written  docu- 
ments. He  may,  however,  have  made  use  of  one  text  treating  of 
the  siege  of  Acre  before  the  arrival  of  Richard  and  the  French 
king.  In  Brunner's  study  (pp.  51-70)  of  the  romance  a  long  list 
was  given  of  the  details  which  in  his  opinion  were  to  be  classified 
as  historic,  pseudo-historic,  or  purely  fabulous.  Chiefly  with 
regard  to  the  latter,  his  work  on  the  sources  supplemented  that 
of  Gaston  Paris,  who  disregarded  these  fictitious  details  on  the 
ground  that  they  did  not  appear  in  the  original  Anglo-Norman 
version  with  which  he  was  primarily  concerned. 

It  is  evident  that  in  the  thirteenth  century  the  story  of 
Richard,  stimulated  by  patriotic  enthusiasm  for  the  memory  of 
a  king  whom  popular  imagination  was  making  into  a  kind  of 
victorious  Roland,  was  constantly  increasing  its  content  and 
popularity.  Though  Richard's  own  mother,  to  whom  reference  is 
made  in  the  b  version  of  the  romance  (1.  2041),  was  a  sufficiently 
famous  subject  for  song,  the  legend  of  his  demon  birth  was  in 
circulation  at  an  early  date.    Richard's  contemporary,  Giraldus 


RICHARD    COEUR   DE   LION  i5l 

Cambrensis  (De  principis  instructione,  in,  cap.  27,  ed.  Warner, 
1 89 1,  p.  300),  quoted  the  King's  story  of  the  demon  lady  who 
was  one  of  his  Angevin  ancestors  (Paris,  Rom.  xxvi,  357). 
She,  like  his  mother  in  the  a  version,  may  be  related  to  all  the 
Fairy  Mistresses  of  romance  but  is  more  particularly  like  the 
Lamia  type  of  lady  best  known,  perhaps,  in  the  Lusignan  leg- 
end of  Melusine  (Edwardes  Summary,  p.  284)  as  set  forth 
about  1387  by  Jean  d'Arras.  Legends  of  Richard's  prowess 
rapidly  became  wide-spread  and  were  represented  in  art  as  well 
as  story.  Thus  his  duel  with  Saladin,  which  is  recounted  in 
early  fourteenth-century  chronicles  such  as  Peter  de  Langtoft's 
(Rolls  Ser.  11,  102)  and  Walter  de  Hemingburgh's  (ed.  H. 
Hamilton,  Lond.  1848,  p.  183),  was  sufficiently  well-known  in 
1250  for  Henry  III  to  order  the  story  painted  on  the  walls  of 
Clarendon  palace.  It  also  appeared  in  the  Chertsey  Tiles,  in 
the  Louterell  Psalter  (c.  1340),  and  in  various  other  manuscripts 
(Loomis,  pp.  513-19).  The  famous  story  of  the  lion  fight  from 
which  Richard  drew  his  nickname,  was  represented  in  the  Chert- 
sey Tiles,  in  the  late  thirteenth-century  Peterborough  Psalter, 
and  on  a  boss  in  the  cloister  of  Norwich  Cathedral  (ibid.,  pp. 
519-22).  The  date  at  which  this  incident  became  a  part  of  the 
tale  of  Richard's  captivity  is  undetermined.  Historically  the 
captivity  episode  belongs  to  his  return  from  the  Crusade  in 
1 1 92;  in  the  romance  it  is  an  incident  of  his  return  from  a 
pilgrimage  made  supposedly  before  the  Crusade.  He  travels  as 
a  palmer,  is  imprisoned  by  King  Modard  of  Almayne,  and  is 
exposed  to  the  attack  of  a  lion  sent  into  his  cell.  Richard  wraps 
around  his  arm  the  forty  silk  handkerchiefs  brought  him  by  the 
king's  daughter.  When  the  lion  attacks  him,  he  tears  out  its 
heart,  stalks  into  the  king's  hall,  and  salts  and  eats  the  heart 
before  the  astounded  court.  The  feat  suggests  comparison  with 
traditional  exploits  of  Samson  and  David,  but  in  such  pictures, 
and  carvings  as  are  known,  the  scene  represents  "  Richard,  that 
robbed  the  lion  of  his  heart  "  in  fairly  recognizable  fashion. 

The  elaborated  a  version  shows  how  strong  a  magnet  the  story 
of  Richard  was,  not  only  for  floating  scraps  of  tradition  about 
the  king,  but  also  for  anecdotes  and  motifs  which  had  originally 
no  connection  with  him.  The  barbaric  account  of  Richard's 
cannibalism  when  he  dines  on  Saracen  heads  "  all  hot  "  (1.  3427) 


1 52  MEDIAEVAL   ROMANCE    IN    ENGLAND 

and  forces  imprisoned  foes  to  eat  of  Saracen  flesh,  recalls  the 
grim  account  given  by  Ademar  de  Chabannes  (in,  55,  ed.  Cha- 
vanon,  p.  178)  of  an  eleventh-century  Norman  who  in  Spain 
forced  imprisoned  Moors  to  eat  the  flesh  of  countrymen  (Paris, 
Rom.  xxvi,  359,  n.).  In  another  passage  in  Richard  (1.  3089) 
the  substitution  of  a  "  Sarezyn3onge  and  flat  "  for  pork,  suggests 
comparison  with  the  Chanson  d'Antioche  (ed.  G.  Paris,  1848, 
11,  3).  Similarly  antedating  Richard's  own  time  is  the  old 
anecdote  of  the  bee-hives  (1.  2906)  thrown  into  a  besieged  city, 
and  the  account  (1.  2150)  of  the  steward  whose  nose  was  cut  off 
by  the  Emperor  of  Cyprus.  Paris  (p.  389)  has  likened  this  un- 
fortunate to  Estatin  Vesnase  of  the  first  Crusade.  On  the  other 
hand,  later  than  Richard's  time  was  the  Castel  Pilgrim  of  the  ro- 
mance, for  the  Castle  was  not  made  or  named  until  12 18;  and 
later  too  was  the  concept  of  St.  George  as  patron  saint  of  the 
English.  The  saint  was  not,  officially,  thus  recognized  until 
after  1222  (Matzke,  Beves,  PMLA.  xvm,  155). 

The  process  of  accretion  is,  however,  most  clearly  indicated  in 
the  use  of  various  romance  motifs.  The  boat  which  brings  the 
fairy  lady  to  England  is  matched  in  its  strange  splendor  by 
those  in  Guigemar  (Warnke,  Lais,  p.  lxxix),  Partenopeus,  etc., 
where  its  purpose  is  likewise  to  bring  lovers  together.  In  his 
Salisbury  tournament  Richard  wears  successive  disguises  of 
black,  red,  and  white,  as  do  most  of  the  heroes  of  the  Three 
Days'  Tournament  theme.  When  he  goes  on  his  first  journey  to 
the  East,  he  wears  the  usual  pilgrim  disguise.  English  ballad 
stories  of  Kings  Incognito  parallel,  if  they  do  not  anticipate,  the 
rude  speech  between  Richard  and  the  minstrel  who  comes  begging 
to  his  fire.  Betrayed  by  the  minstrel  and  imprisoned  by  King 
Modard  of  Almayne,  Richard  manages  in  an  Exchange  of  Buf- 
fets 4  to  kill  the  King's  son,  —  a  deed  which  commonly  intro- 
duces in  romantic  narrative  the  hero's  flight  or  exile, — 
and  to  engage  in  a  love  affair  with  Modard 's  daughter, 
a  typical  "  wooing  princess." 5     In   the   account   of   Richard's 

4  For  the  ancient  and  widespread  game  of  pluck  buffet,  see  G.  L.  Kit- 
tredge,  Sir  Gawain  and  the  Green  Knight,  Cambridge,  1916,  p.  21  ff.  no, 
123,  221.    See  also  Child,  Ballads,  III,  55,  77. 

5  McKnight,  Horn,  n.  to  1.  264,  thought  that  the  scene  in  which  the 
Princess  bids  the  jailer  bring  her  his  captive,  disguised  as  a  squire,  is  possibly 
borrowed  from  Horn. 


RICHARD    COEUR    DE   LION  153 

revenge  journey  into  "  Almayn,"  of  the  edict  of  Modard 
who  forbids  the  inhabitants  to  sell  Richard  any  food,  and 
of  Richard's  circumvention  of  the  order,  the  Middle  English 
poet  has  apparently  made  a  somewhat  stupid  adaptation 
of  a  motif  used  in  Aimeri  de  Narbonne,  and  repeated  in  chron- 
icles which  variously  ascribe  the  adventure  to  Robert  the 
Magnificent  (1034),  to  the  Norwegian  kings,  Harold  (1034)  and 
Sigurd  (mi),  to  the  Swabian  founder  of  Donanwerth  Abbey 
(1027),  and  to  the  Emperor  Frederic  II,  (Paris,  Rom.  ix,  515). 
In  each  case  the  edict  of  a  rival  king  or  noble  is  evaded  by  the 
unparalleled  wealth  and  recklessness  of  the  newcomers.  No 
price  for  food  is  too  great  for  them;  if  fuel  fails,  they  burn 
houses,  etc.  Again  the  scene  in  which  the  infuriated  Emperor  of 
Cyprus  hurls  his  knife  at  Richard's  messenger  (1.  2120)  is 
paralleled  in  Beves  of  Hampton,  in  Generides,  and  many  other 
romances.  Richard's  overthrow  of  a  marble  image  (1.  6265)  in  a 
Saracenic  city  recalls  Beves's  overthrow  of  the  idol  in  Damascus, 
a  deed  which  harks  back  to  the  legend  of  St.  George.  The 
defeated  defenders  of  Castel  Orgylous  appear,  as  do  so  many 
other  unfortunates  enacting  in  fact  or  legend  the  so-called 
"  Chivalric    Penance," 6    "  barffoot,    ungyrt,    withouten    hood " 

(L4181). 

Throughout  the  romance  the  supernatural  element  is  but 
rarely  of  distinctive  sort.  When  the  English  hero  departs  finally 
from  Almayn,  he  is  given  two  Magic  Rings  for  protection  from 
fire  and  water,  though  this  bit  of  popular  detail  has  no  further 
significance  in  the  narrative.  Prophetic  dreams,  angelic  warn- 
ings or  commands,  save  Richard  from  danger  or  send  him  into 
battle  at  the  propitious  moment.  In  his  direst  need  St.  George 
comes  to  his  aid.  Of  more  unique  character  is  the  description  of 
the  Demon  Horse  presented  to  Richard.  The  presentation  itself 
seems  a  reminiscence  of  an  incident  in  the  Estoire  de  la  Guerre 
Sainte  which  tells  of  a  chivalrous  Moslem  who  gave  a  horse  to 
his  heroic  enemy  (Paris,  Jour,  des  Savants,  p.  489,  n.  3).  As  for 
the  demon  colt  itself,  demon  steeds  are  familiar  enough  in  folk- 
lore, and  even  in  Arthurian  romance  Perceval  almost  comes  to 
his  doom  through  a  demon  horse,  but  in  no  other  case  is  the 
creature   treated   with   such   graphic   and   comic   detail   as   in 

6  See  here,  Guy  of  Warwick,  n.  13. 


I54  MEDIEVAL   ROMANCE    IN   ENGLAND 

Richard.  The  colt's  ears  are  stuffed  with  wax,  a  forty  foot  tree 
is  trussed  across  its  head,  and  it  is  a  sorry  demon  indeed  that 
Richard  rides  to  a  duel  with  its  master  (1.  5535  ff.).  According 
to  the  evidence  of  the  Chertsey  Tiles,  these  fantastic  details 
must  have  appeared  in  the  duel  story  about  1275  (Loomis,  p. 
512). 

Richard  is  one  of  the  most  militant  of  the  Middle  English 
romances.  Throughout  its  7212  lines  it  is  so  filled  with  the 
pitiless  and  fanatic  spirit  of  the  Crusades  that  its  pages  teem 
with  abuse  of  the  Saracen  "  dogs,"  caricatures  of  their  leaders, 
and  absurd  laudations  of  the  Christians.  It  is  so  militant  that 
it  seems  almost  untouched  by  courtly  or  chivalric  influence.  The 
one  passage  in  praise  of  May,  to  which  reference  has  already 
been  made,  and  a  single  brief  love  affair  of  Richard's,  are  the 
only  romantic  elements  in  the  story.  Few  heroes,  even  in 
Carolingian  epic,  are  represented  in  more  violent  mood  than  the 
Lion-Heart.  He  smashes  tables  in  his  rage;  he  cuts  through 
iron  chains  before  the  gates  of  beleaguered  cities;  he  is  the 
very  "  scourge  of  God  "  upon  his  enemies ;  he  abuses  his  allies 
at  the  first  sign  of  cowardice  or  indecision.  Of  all  this,  and  of 
the  business  of  mediaeval  siege  warfare,  of  fights  on  land  or 
sea,  of  the  great  catapults  with  which  Richard  won  his  victories, 
of  poisoned  wells,  and  trenches  choked  with  dead,  the  romance 
tells  with  frankest  relish.  That  the  author's  taste  was  shared  by 
others,  the  number  of  extant  manuscripts  would  seem  to  show, 
but  there  is  hardly  enough  evidence  as  yet  to  indicate  the 
existence  of  a  local  cult  of  Richard  in  the  South  Lincolnshire 
district.7 

BIBLIOGRAPHY 

Texts:  (i)  L,  Auchinleck,  MS.  pr.  Laing,  Owain  Miles  and  Other 
Fragments,  Edin.  1837;  Kolbing,  Eng.  Stud,  vm,  115-9;  (2)  E,  formerly 
Duke  of  Sutherland's,  now  Egerton  2862,  desc.  Brit.  Mus.  Catalogue  of 
Add.  MSS.,  1906-10,  p.  238;  (3)  C,  Caius  College  175/96,  desc. 
M.  R.  James,  Cat.  of  Library  of  Gonville  and  Caius  Coll.  Cbg.,  1907; 
largely  printed  by  Weber,  1803,  11,  148-278  (11 136  lines),  and  used  as 
a  basis  for  the  critical  edition  published  by  Karl  Brunner  (see  below) ; 

7  For  further  study  of  Richard's  place  in  fiction,  see  Needler's  disserta- 
tion, Brunner,  pp.  73-75,  and  E.  A.  Baker's  Guide  to  Historical  Fiction, 
Lond.  1914,  p.  18. 


RICHARD    COEUR   DE  LION  I5S 

(4)  H,  Harley  4690,  fragment;  (5)  A,  College  of  Arms,  HDN.  58  desc. 
W.  H.  Black,  Cat.  of  Arundel  MSS.,  Lond.  1829,  pp.  104-110;  (6)  B, 
Add.  MS.  31042,  desc.  Heritage,  Sege  of  Melayne,  EETS.  xxxv,  p. 
viii;  (7)  D,  Douce  MS.  228,  Bodleian,  fragment.  The  British  Museum 
MSS.  are  described  by  Ward,  Catalogue,  1,  944-50.  Trans,  in  part, 
J.  Weston,  Chief  ME.  Poets. 

Studies:    Wells,  Manual,  p.  786. 

Brunner,  Karl.     Der  mittelengl.    Versroman  uber  Richard  Lowenherz, 

Kritische  Ausgabe  nach  alien  Handschriften,  mit  Einleitung,  Anmer- 

kungen  u.  deutscher  Ubersetzung.     604  pp.  Wiener  Beitrdge  z.  eng. 

Phil,  xli,  Wien  u.  Leipzig,  1913.     Rev.  Eng.  Stud,  xlix,  126-42; 

N.  Y.  Nation,  July  30,  1914;  JEGP.  xv,  455-66. 
Jentzsch,  F.  "  Die  me.    Romanze  Richard  Coeur  de  Lion  u.  ihre  Quellen," 

Eng.  Stud,  xv,  161-247  (1891). 
Lodeman,  F.  E.  "  Le  Pas  Saladin,"  MLN.  xn,  21-34,  Introduction:  84- 

96,  Text:    209-229;  273-81  (1897). 
Loomis,  R.  S.  "  Richard  Coeur  de  Lion  and  the  Pas  Saladin  in  Medieval 

Art,"  PMLA.  xxx,  509-28  (1915).     Illus. 
Needier,  G.  H.  Richard  Coeur  de  Lion  in  Literature.     Diss.  Leipzig, 

1890. 
Paris,  G.  "  Sur  un  Episode  d'  Aimeri  de  Narbonne,"  Rom.  ix,  515-46 

(1880).    Cf.  Richard,  version  a,  1475  ff. 

"  La  Legende  de  Saladin,"  Journal  des  Savants,  1893,   (mai-aout). 

"  Le   Roman    de   Richard    Coeur    de    Lion,"   Rom.   xxvi,   353-93 

(i897). 
Thomas,  A.  "  La  Legende  de  Saladin  in  Poitou,"  Jours,  des  Savants,  pp. 
467-71  (1908). 


GAMELYN 

"  Gone,  the  merry  morris  din ; 
Gone  the  song  of  Gamelyn." 

—  Keats. 

Versions.  The  tale  of  Gamelyn,  the  dispossessed  youth  who 
sturdily  regains  his  heritage,  is  found  in  sixteen  Chaucerian 
manuscripts.  All  apparently  are  derived  from  the  same  original 
and  none  are  of  earlier  date  than  the  fifteenth  century  (Ham- 
mond, Chaucer  Manual,  p.  425).  In  the  Canterbury  Tales  the 
story  is  commonly  inserted  after  the  fragmentary  Cook's  Tale, 
and  critical  opinion  has  varied  as  to  whether  it  was  intended 
for  the  Cook,  or,  as  seems  more  probable,  for  the  Yeoman.  But 
even  if  Chaucer's  interest  in  the  story  may  be  assumed  from  its 
presence  among  the  manuscripts  of  his  own  works,  it  is  certain 
that  the  poem  had  received  no  revision  at  his  hands.  Since 
Tyrwhitt's  Introductory  Discourse  to  his  edition  of  the  Canter- 
bury Tales  (1775-8),  Gamelyn  has  been  rejected  from  the 
Chaucerian  Canon.  Skeat  {Chaucer,  III,  400)  noted  that  its 
dialect  was  more  northern  than  that  used  by  Chaucer  and  that 
words  of  French  origin  were  in  smaller,  and  those  of  Scandi- 
navian origin  in  larger  number,  than  in  Chaucer's  usage.  The 
name  of  Gamelyn,1  apparently  derived  from  the  Scandinavian 
Gammel-ing  (son  of  the  old  man),  is  found  in  records  in  England 
as  early  as  the  tenth  or  eleventh  century  (Bjorkman,  Nordische 

1  Prideaux,  Notes  and  Queries,  Ser.  7,  11,  423,  suggested  that  the  name 
comes  from  that  of  Candelou,  a  family  related  to  that  of  Fulke  Fitzwarin, 
the  famous  outlaw  in  whom  Prideaux  recognized  the  original  of  Robin 
Hood.  His  reason  was  that  Leland's  spelling  for  Candelou  was  Gandeline 
and  that  this  spelling  is  found  also  in  Robin  and  Gandeleyn,  a  ballad  which 
Prideaux  thought  represented  figuratively  the  struggle  between  Fulke  and 
an  early  rival.  In  the  ballad  the  name  of  the  adversary  of  Robin  and 
Gandelyn  was  Wrennok  and  this  name  apparently  was  also  that  of  the  son 
of  Fulke's  temporary  successor.  But  since  the  ambushed  and  easily  mur- 
dered Robin  of  the  ballad  was  certainly  not  the  Robin  Hood  of  tradition, 
neither  was  he  representative  of  Fulke,  who  had  a  long  and  prosperous 
career  after  his  early  troubles,  nor  was  Gandelyn  representative  of  a  friendly 
Candelou. 

156 


GAMELYN  I57 

Personnenamen,  p.  45).  In  temper,  nc  less  than  in  language, 
Gamelyn  suggests  that  its  author  was  of  a  neighborhood  or  of  a 
nature  in  which  the  old  native  strain  had  been  little  changed  by 
French  influence. 

The  extant  texts  of  Gamelyn  must  have  been  derived  from  a 
fourteenth-century  version.  The  language  is  certainly  not  more 
archaic  than  that  of  Havelok,  the  romance  with  which  Gamelyn 
has  been  most  frequently  compared.  Skeat  {ibid)  argued  that 
the  latter  should  be  dated  after  1340  because  it  borrows  (lines 
277,  764)  from  a  poem  on  the  Evil  Times  of  Edward  II  (Percy 
Soc.  28,  1 )  the  commonplace  line,  "  By  Seint  Jame  in  Galys  that 
many  a  man  hath  sought."  This  poem,  found  in  the  Auchinlek 
manuscript  (1330-40),  is  generally  ascribed  to  the  year  1320, 
but  its  priority  to  the  lost  original  version  of  Gamelyn  has  so  far 
only  been  asserted,  not  proved.  Such  historical  filiations  as  the 
narrative  of  Gamelyn  suggests,  point,  as  will  presently  be  shown, 
to  the  early  years  of  the  fourteenth  century ;  but  realistic  as  the 
poem  seems,  its  descriptive  details  offer  no  historical  data. 
Lindner  (Eng.  Stud.  11,  321)  tried  to  find  in  the  description  of 
the  manor  house,  about  the  possession  of  which  so  much  of  the 
story  centers,  features  suited  to  a  thirteenth-century  house;  but 
as  a  matter  of  fact  neither  here,  nor  in  the  passage  describing 
the  moot-hall  where  Gamelyn's  trial  was  held,  is  the  evidence 
sufficiently  detailed  to  do  more  than  add  to  the  general  vividness 
of  the  setting.  So  local  is  this  in  character,  with  its  references 
to  village  wrestling  green,  to  the  woods  and  fen  beyond  the 
manor,  to  the  villagers  "  ryding  jolily "  in  "  cartes  and  in 
waynes,"  so  devoid  of  the  French  romantic  touch,  that  it  is 
difficult  to  believe  that  it  emanates  from  any  French  original. 
Nevertheless  Skeat  (ibid.)  on  the  basis  of  the  French  names  of 
Gamelyn's  father,  Sir  Johan  of  Boundes,  and  of  his  brother 
Ote(s),  was  inclined  to  believe  in  an  Anglo-French  original. 

Gamelyn  is  roughly  divided  into  six  parts  by  such  minstrel 
admonishments  to  his  hearers  as,  "  litheth  and  lesteneth,"  and 
occasionally,  "  holdeth  your  tonge."  The  story  is  told  in  rimed 
couplets.  The  single  lines  have  an  irregular  number  of  syllables 
as  in  the  old  alliterative  verse,  but  alliteration  itself  is  found 
chiefly  in  the  numerous  stock  phrases.     In  the  first  half-line 


IS8  MEDIEVAL   ROMANCE    IN    ENGLAND 

there  are  commonly  four  stresses,  and  in  the  second,  three.  The 
couplets,  therefore,  sometimes  approximate  the  ballad  stanza, 
and  touches  of  ballad  style  are  not  infrequent.  We  find  the 
verve  and  simplicity  of  phrasing,  the  use  throughout  more  than 
half  the  poem  of  the  dialogue  form,  the  frequency  of  minstrel 
tags,  the  repetition  of  easy  rimes,  as  Lindner  (ibid.,  p.  101)  has 
shown  in  his  rime  index,  and  well-worn  adages  of  popular  wisdom, 
such  as  "  after  bale  cometh  bote."  In  style,  spirit,  setting, 
Gamelyn  is  as  surely  English  as  the  Geste  of  Robin  Hood. 

Although  the  early  printers  were  so  aware  of  the  popular 
appeal  of  the  Geste  that  they  frequently  reprinted  it,  they  seem 
altogether  to  have  neglected  Gamelyn,  even  in  the  black  letter 
editions  of  Chaucer,  and  it  must  have  been  from  one  of  the 
later  manuscript  forms  that  Thomas  Lodge  refashioned  it  into 
the  first  part  of  his  prose  romance,  Rosalynde  or  Euphues 
Golden  Legacie,  1590.  True  Elizabethan  that  he  was,  Lodge, 
under  the  influence  of  Euphuistic  style  and  of  Italian  pastoral 
convention,  refurbished  and  refined  the  old  story  out  of  its  un- 
couth roughness.  Stout  Gamelyn  became  the  graceful  Rosader ; 
his  brothers  John  and  Ote,  the  fanciful  Saladyne  and  Fernan- 
dyne ;  their  English  manor  a  house  in  Bordeaux.  Fighting  gave 
way  to  a  romance  which  was  altogether  lacking  in  the  blunt 
old  story,  and  Rosalynde,  beloved  of  Rosader,  and  her  friend 
Alinda  began  those  dainty  wanderings  in  the  forest  of  Arden 
which  drew  at  last  Shakespeare  himself  to  sing  of  them.  In 
As  You  Like  It  (1599-1600)  the  pretty  idealized  love  story 
introduced  by  Lodge  reached  its  ultimate  fulfillment  in  the 
beauty  and  mirth  of  Rosalind  and  Orlando. 

Origin.  Though  traditional  character  types  and  incidents 
play  so  large  a  part  in  Gamelyn,  there  are  many  elements  in  it 
which  savor  chiefly  of  fact.  The  tale  begins,  to  be  sure,  with  a 
true  folk-tale  situation,  for  Gamelyn  is  that  youngest  son  who 
is  perennially  subjected  to  the  malice  of  his  older  brothers.  But 
Gamelyn's  story  is  developed  with  the  literalness  of  life.  His 
oldest  brother  sordidly  cheats  him  out  of  a  comfortable  country 
property,  and  the  youth's  complaints  about  his  unsown  lands, 
his  oaks  cut  down,  his  broken  parks,  his  decayed  houses,  and  his 
slain  deer,  remind  one  of  the  accounts  in  the  Patent  Rolls  of 


GAMELYN  159 

Edward  I,  II,  and  III,  which  record  similar  abuses  wrought 
against  property  in  private  feud  or  by  careless  keepers.  There 
is  nothing  in  the  poem,  as  there  is  in  Havelok  about  kingdoms 
or  dispossessed  royal  heirs,  nothing  of  battles  and  knights.  The 
parallels  between  the  two  stories,  the  facts  that  the  two  heroes 
are  wrestlers,  that  both  use  clubs  instead  of  weapons  in  their 
rough  and  tumble  fights,  that  the  two  poems  share  certain 
colloquial  expressions,  such  as  the  quaint  "  so  brouke  I  myn  ye  " 
(1.  334),  rise  from  a  common  source  in  popular  taste  and  not 
from  literary  influence.  Havelok  draws  clearly  on  romance 
tradition,  Gamelyn  on  everyday  realities.  Gamelyn  broods  over 
his  wrongs,  pulls  his  young  beard,  flies  into  an  ungovernable 
rage,  chases  his  brother  round  his  own  manor  yard,  and  up  the 
stairs  into  a  loft  where  hastily  "  he  schette  the  dore  fast."  The 
scene  is  differentiated  by  its  homely  realism,  as  is  the  whole  dis- 
inheritance story  in  Gamelyn,  from  the  conventional  use  of  the 
same  situations  in  romance  and  folk-tale.  In  the  records  of  the 
thirteenth  and  fourteenth  centuries  the  cheating  of  minors  is  a 
constantly  recurrent  offense ;  and  though  no  exact  historic  paral- 
lel to  Gamelyn  has  been  found,  there  is  absolutely  nothing  in 
this  part  of  the  story  for  which  contemporary  circumstance 
could  not  have  furnished  ample  suggestion. 

Gamelyn's  exploit  in  overcoming  a  famous  wrestler  is  a  popular 
detail  which  may  very  possibly  go  back  to  some  local,  well-known 
tale.  The  Geste  of  Robin  Hood  (Fit  11,  st.  135-42)  tells  of  a 
wrestling  match  between  all  the  best  yeomen  of  the  West  Coun- 
tree  (West  Riding  of  Yorkshire),  of  a  stranger  who  wins  the 
match,  of  his  danger  from  angry  competitors,  and  of  his  rescue 
by  a  noble  knight  for  love  of  Robin  Hood.  Within  twenty- 
eight  lines  in  the  Geste  there  is  an  amount  of  detail  concerning 
the  prizes  and  the  wrestling  much  greater  than  in  the  hundred 
lines  devoted  to  the  incident  in  Gamelyn.  The  lines  in  the  Geste 
sound  like  the  summary  of  an  older,  longer  ballad  on  the  same 
subject.2  Though  the  Geste  is  too  different  from  Gamelyn  to  be 
identified  with  it,  the  resemblance  in  this  incident  points  to  the 
common  use  of  an  older  traditional  tale. 

Other  traditional  elements  may  be  easily  recognized.  On 
Gamelyn's  return  from  the  match  he  is  refused,  by  his  brother's 

2  W.  H.  Clawson,  The  Geste  of  Robin  Hood,  Toronto,  1909,  pp.  47-8. 


160  MEDIEVAL   ROMANCE    IN    ENGLAND 

orders,  admittance  to  his  home.  Like  Beves  and  many  another 
hero,  he  pitches  the  Surly  Porter  into  a  well.3  Sir  John  pretends 
that  he  has  sworn  to  have  Gamelyn  bound  hand  and  foot,  and  in 
the  moment  of  their  apparent  reconciliation  asks  Gamelyn  to 
permit  the  act  so  that  he  (John)  may  not  be  foresworn.  Thus 
bound,  Gamelyn,  like  a  humble  Samson,  endures  the  insults  of 
his  brother's  friends  as  they  feast  about  him.  Later,  when, 
though  released  by  the  faithful  old  servant,  Adam  (the) 
Spenser,  he  still  stands,  feigning  captivity,  the  irony  of  the  situa- 
tion suggests  the  many  popular  tales  in  which  a  man,  supposedly 
poor  and  friendless,  is  taunted  by  those  from  whom  presently, 
to  their  dire  discomfiture,  he  is  to  regain  his  inheritance.  The 
incident  has  always  had  the  same  appeal  of  grim  humor  whether 
in  the  tale  of  royal  Odysseus,  unrecognized  among  the  evil 
suitors  in  his  own  Ithacan  hall,  or  in  the  legend  of  the  Heir  of 
Linne  (Fit  n,  84-129),  who  recovers  his  lands  from  the  greedy 
Abbot  of  St.  Mary's.4  The  sarcastic  mirth  with  which  the  author 
describes  Gamelyn  sprinkling  holy  water  "  with  an  oaken  spire  " 
on  the  unlucky  heads  of  his  foes,  Leach  (p.  657)  noted  as  one 
of  the  qualities  suggestive  of  the  kinship  in  temper  of  Gamelyn 
and  Scandinavian  story. 

Gamelyn's  fight  with  the  Sheriff,  his  flight  to  the  wild  wood, 
his  fellowship  there  with  amiable  and  sportive  outlaws  whose 
master  he  presently  becomes,  his  hostility  to  the  rapacious 
clergy,  his  feud  with  the  unjust  Sheriff,  his  loyalty  to  the  king, 
illustrate  incidents  and  feelings  which  seem  closely  akin  to  those 
in  the  Geste  of  Robin  Hood  or  to  the  earlier  ballads  which  were 
its  source.  The  well-known  allusion  in  Piers  Plowman  (B,  v, 
401)  proves  that  these  were  current  before  1377,  the  date  of 
the  B  text,  but  the  question  of  their  relationship  to  the  lost 
fourteenth  century  original  of  Gamelyn  has  so  far  remained  an 
open  one.  In  his  edition  of  the  tale  Skeat  (p.  xi)  suggested 
that  the  namelessness  of  the  outlaw  chieftain,  whose  place 
Gamelyn  took,  implied  that  the  romance  represented  an  earlier 
stage  than  the  ballads  with  their  hero  of  familiar  name.  This 
was  to  argue  that  every  outlaw  figure,  however  shadowy,  must 
be  identified  with  Robin  Hood  alone  and  to  ignore  the  many 

3  Child,  Ballads,  III,  95,  n. 

4  Cf.  Clawson,  p.  45. 


GAMELYN  jfa 

sources  in  contemporary  fact  and  fiction  for  such  a  figure.5 
Furthermore  the  outlaw  incident  in  Gamelyn  is  frankly  inci- 
dental to  the  disinheritance  theme;  it  is  a  distinct  popular  em- 
bellishment which  in  style,  incident,  and  theme  reflects  the  type 
of  narrative  established  in  the  Robin  Hood  ballads.  Since  it  is 
impossible  to  believe  that  Gamelyn  ever  gave  rise  to  the  ballads, 
it  only  remains  possible  to  find  in  Gamelyn  the  definite  influence 
of  the  ballads,  and  to  recognize  their  priority.  Later  on,  when 
the  Robin  Hood  ballads  were  yet  more  widely  known,  popular 
fancy  seems  to  have  striven  to  glorify  the  outlaw  character  of 
Gamelyn,  and  to  have  associated  his  name  with  that  of  Robin 
Hood  somewhat  to  the  disparagement  of  the  latter.  In  the 
ballad  of  Robin  and  Gandelyn  (Child,  No.  115)  the  outlaw 
leader  who  is  murdered  by  Wrennok  and  avenged  by  Gandelyn, 
is  certainly  not  the  traditional  Robin  Hood,  although  the  bor- 
rowing of  the  name  is  a  significantly  popular  touch.  In  the  much 
later  ballad,  Robin  Hood  Newly  Revived  (Child,  No.  128),  the 
ballad  itself  is  built,  as  Child  remarked,  "  on  the  ruins  of  the 
fine  old  tale  of  Gamelyn"  In  this  Gandelyn  outdoes  Robin 
Hood,  is  recognized  by  him  as  "  his  sister's  son  "  and  is  adopted 
into  the  outlaw  troup  under  the  name  of  Will  Scadlock  or 
Scarlet.6 

The  concluding  scenes  in  Gamelyn  have  a  distinctive  realism 
which  merits  attention.    When  Gamelyn  has  fled  to  the  green- 

5  A  parallel,  for  instance,  to  the  fact  that  Gamelyn  after  living  as  an 
outlaw  leader,  made  his  peace  with  the  king,  and  was  later  appointed  by 
him  Chief  Justice  of  the  Forest,  lies  in  the  well-known  history  of  Adam 
Gurdon  (d.  1306)  who  was  "  disinherited "  by  Henry  III  and  became  a 
famous  outlaw,  but  who,  on  being  reconciled  to  Edward  I,  was  appointed 
by  him  for  many  years  Justice  of  the  Forest  on  circuits  in  Berkshire  and 
Wiltshire.  (Cf.  E.  Foss,  The  Judges  of  England,  1870;  Diet.  Nat.  Biog.) 
The  episode  of  Adam's  famous  single-handed  fight  with  Prince  Edward 
(1266),  which  was  known  to  nearly  all  the  chroniclers  of  the  reign  of 
Edward  I,  is  of  special  interest  for  comparison  with  the  story  of  Robin 
Hood's  buffeting  with   "  our  comely   king  Edward." 

6  The  name  of  a  light-fingered  Walter  Scarlet,  committed  for  burglary 
in  the  house  of  a  chaplain,  is  found  in  a  commission  for  clearing  the  jail 
at  Norwich,  Nov.  8,  1286,  and  in  one  for  Estedam,  Jan.  28,  1287,  Patent 
Rolls,  Edward  I,  Vol.  II.  A  Will  Skarlet  de  Shategrave  appeared  as  the 
accuser  in  a  trial  of  1289  in  Norfolk.  See  T.  F.  Tout  and  H.  Johnstone, 
State  Trials  of  the  Reign  of  Edward  I,  1 288-1 293,  Camden  Soc,  1906,  p. 
120.  This  adds  two  more  to  the  rather  profitless  endeavors  to  equate  his- 
torical names  with  those  found  in  the  Robin  Hood  ballads. 


t.62  MEDIAEVAL   ROMANCE   IN   ENGLAND 

wood,  his  brother  John  proclaims  him  "  wolf's  head."  At  the  next 
"  schire  "  Gamelyn  goes  boldly  to  the  moot-hall,  is  arrested  and 
delivered  over  to  the  keeping  of  his  good  brother,  Sir  Ote.  From 
him  he  returns  to  the  greenwood,  and  at  the  time  set  for  his 
trial  Ote  is  bound  in  his  place.  For  this  trial  Sir  John  packs 
the  jury:  "He  was  faste  aboute  — For  to  hyre  the  quest  to 
hangen  his  brother"  (1.  785).  At  the  trial  itself  there  is  a 
wild  scene.  Gamelyn  fiercely  rebukes  the  Justice,  saying,  "  Thou 
hast  yeven  domes  that  ben  yvel  dight,"  and  thereupon  throws 
him  over  the  bar  and  breaks  his  arm.  Then  seating  himself  in 
the  Justice's  place,  he  makes  of  his  own  men  a  jury  and  dooms 
the  false  Justice  to  be  hanged,  together  with  his  brother,  the 
sheriff,  and  the  twelve  false  "  sisours."  Allowing  for  popular 
extravagance  and  for  the  irresistible  temptation  from  the  story- 
teller's point  of  view  to  make  the  accused  become  the  accuser, 
and  the  victim  the  judge,  we  may  recognize  in  all  this  something 
which  again  seems  reminiscent  of  fact  rather  than  fancy.  The 
account  of  the  false  sheriff,  the  bribed  jury,  the  trial  and  punish- 
ment which  judge  and  jury  are  made  to  suffer,  recalls  the  no- 
torious scandals  of  1289  when  the  charges  of  bribery  and  cor- 
ruption brought  against  certain  chief  justices  and  sheriffs  led  to 
a  series  of  trials  lasting  over  three  years  and  to  the  infliction 
on  the  guilty  of  heavy  fines  and  penalties.7  Many  of  the  con- 
temporary chronicles  refer  to  the  affair,  and  in  Norfolk,  es- 
pecially, whence  so  many  complaints  came,  the  memory  of  the 
whole  sorry  business  must  have  been  long  lived.  Legal  injus- 
tice, of  course,  was  a  sufficiently  familiar  commonplace  of 
mediaeval  life,  but  the  rarity  with  which  popular  writers  treated 
it,  gives  the  more  importance  to  its  bold  presentment  in  Gamelyn, 
and  the  more  reason  to  find  for  this  episode  some  possible  sub- 
stratum of  fact.  So  far  as  story  goes,  as  Leach  (p.  354)  sug- 
gested, the  relish  for  court  proceedings  in  Gamelyn  is  best 
paralleled  by  the  interest  in  the  doings  of  the  Thing  shown  by 
the  authors  of  Njals  Saga  and  Grettis  Saga. 

7  Cf.  Tout  and  Johnstone,  passim.  Various  cases  are  given,  indicating 
unfair  trials,  removal  and  substitution  of  jurors,  intimidation,  cheating  of 
minors,  etc. 


GAMELYN  163 

BIBLIOGRAPHY 

Texts:  (i)  Br.  Mus.  Harley  7334,  pr.  Wright,  Canterbury  Tales, 
1847;  by  Bell  from  Wright,  1854;  Chaucer  Society,  1885.  For  descrip- 
tion, see  Hammond,  Chaucer  Manual,  pp.  177-8.  Skeat,  Oxford 
Chaucer,  ill,  403,  considered  this  much  the  best  copy  of  Gamelyn  and 
made  it  the  basis  of  his  editions,  Oxford,  1884,  1893;  Oxford  Chaucer 
(Text)  iv,  645  ff.,  1894;  (notes)  v,  497.  (2)  Harley,  1758;  (3)  Royal 
18  C  ii;  (4)  Sloane,  1685;  (5)  Sloane,  1686;  (6)  Lansdowne  851;  (7) 
Corpus  Christi,  Oxford,  196;  (8)  Petworth,  owned  by  Lord  Leconfield. 
2-7  are  pr.  by  the  Chaucer  Society,  Appendix  to  Fragment  A  of  the 
Six  Text  Edition;  Appendices  to  separate  issues  of  these  six  manuscripts. 
(9)  Royal  17,  D  xv;  (10)  Egerton  2726  contains  "a  Jacobean  copy 
of  Gamelyn";  (11)  Barlow  20,  Bodleran;  (12)  Rawlinson  149;  (13) 
Laud  600;  (14)  Christ  Church  152,  Oxf.;  (15)  Trinity  49,  Oxf.;  (16) 
Ashmole  45,  Oxf.  For  details  concerning  these  manuscripts,  see  Ham- 
mond, Chaucer  Manual,  p.  425  and  pp.  173-192.  The  first  printed  edi- 
tion of  Gamelyn  was  that  by  Urry,  Chaucer,  1721.  Trans.  Rickert, 
Romances  of  Friendship,  pp.  85  ff. 

Studies:   For  bibliography  see  Hammond,  Manual,  pp.  425-26;  Wells, 
Manual,  p.  766;  25-7. 
Leach,  H.  S.    Angevin  Britain  and  Scandinavia,  Cambridge,   1920,  pp. 

351-355. 
Lindner,  F.     "The  Tale  of  Gamelyn,"  Eng.  Stud,  u,  94-114,  321-343 
(1878-79). 


ROMANCES  OF  LOVE  AND  ADVENTURE 


APOLLONIUS 

Versions.  The  Apollonius  story  is  a  famous  instance  of 
a  tale  that  for  more  than  a  thousand  years  has  kept  an  almost 
undiminished  integrity  and  vigor.  A  Florentine  manuscript  of 
the  tenth  century  (Laurentius  lxvi)  represents  a  stage  undoubt- 
edly late  in  the  development  of  the  story  itself,  but  it  is  with 
this,  the  oldest  extant  text,  that  the  history  of  the  European 
versions  begins.  Over  one  hundred  manuscripts  of  this  Latin 
prose  version,  the  Historia  Apollonii  regis  Tyri  (ed.  A.  Riese, 
Leipzig,  1893),  are  known,  but  many  are  yet  inedited  and  their 
classification  is  still  a  matter  of  some  dispute.  They  appear  to 
fall  into  three  groups:  the  redaction  RA,  which  includes  the 
Laurentian  manuscript  and  a  fourteenth-century  one  in  Paris; 
the  redaction  RB,  represented  by  manuscripts  of  the  tenth, 
eleventh,  and  fourteenth  centuries  in  Leyden,  Oxford,  and  Paris ; 
and  a  redaction  made  of  mixed  texts  from  which  the  mediaeval 
versions  are  chiefly  derived.1 

Of  mediaeval  Latin  versions  in  addition  to  the  Historia  the 
principal  ones  are:  (1)  the  Gesta  Apollonii  (ed.  E.  Diimmler, 
Mon.  Germ.  Hist.  1888,  11,  483-506),  a  fragmentary  poem  of 
792  verses  in  leonine  hexameter,  contained  in  an  eleventh- 
century  manuscript ;  ( 2 )  the  abbreviated  version  in  tercets  in- 
troduced about  1 1 86  by  Godfrey  of  Viterbo  into  his  chronicle 
history  of  the  world,  the  Pantheon  (Singer,  2,  pp.  150-177)  ;  and 
(3)  the  prose  narrative  in  that  famous  collection  of  stories,  the 
Gesta  Rotnanorunt,  compiled  early  in  the  fourteenth  century 
(Klebs,  p.  334  ff.).  Although  the  Apollonius  (ed.  Singer,  2,  pp. 
68-105;  Smyth,  pp.  93-112)  is  found  in  only  one  of  the  late 
fourteenth-century  manuscripts  (Colmar  10)  of  the  Gesta,  its 
appearance  there  and  in  the  widely  distributed  Latin  editions  of 

1  Cf.  Klebs,  pp.  18-47.  Schreiber  noted  Klebs'  disproof  of  Riese's  asser- 
tion of  the  superiority  of  the  RA  redaction.  It  is  in  fact  no  more  "  original " 
than  is  the  second  group  of  versions  of  the  Historia.  Both  RA  and  RB  appear 
to  be  independent  versions  of  a  lost  original.  The  Viennese  manuscript  edited 
by  Schreiber  belongs  to  the  "  mixed  "  texts. 

164 


APOLLONIUS  ^5 

the  next  century,  shows  that  it  must  have  been  known  better  and 
at  an  earlier  date  than  would  appear  from  the  extant  manuscript 
evidence  (Klebs,  p.  353).  It  was  according  to  Singer  (2,  pp. 
68-149),  the  Gesta  version  of  Apollonius  that  served  as  the 
basis  for  the  Dutch  translation  of  the  Gesta  that  was  printed 
in  148 1,  1483,  and  1484,  for  the  oldest  French  version  of  the 
Gesta,  the  free  fifteenth-century  French  translation  of  the  Gesta, 
Le  Violier  des  Histoires  Romaines  Moraliseez,  and  for  the  Eng* 
lish  version  by  Lawrence  Twine,  The  Patterne  of  Painejull 
Adventures,  London,  1576,  1595,  1607.  Free  treatments  of  the 
Gesta  version  of  Apollonius  appear  also  in  the  Danish  Volks- 
buch,  Copenhagen,  1627,  in  a  similar  Swedish  version  in  1747, 
in  an  Hungarian  version  of  1591,  and  in  Polish  and  Russian 
texts  of  the  seventeenth  and  eighteenth  centuries  (Murko, 
Archiv  f.  slavische  Phil.  1892,  xiv,  405  ff. ;  Klebs,  pp.  362-83). 
No  less  than  the  numerous  Latin  texts  of  the  Historia  do  the 
vernacular  versions  of  Apollonius  attest  its  popularity  through 
the  Middle  Ages.  In  England  its  history  began  at  an  excep- 
tionally early  date.  From  some  Latin  manuscript  of  Group  in, 
in  either  the  tenth  or  the  eleventh  century,  was  made  an  Old 
English  prose  translation,  now  extant  in  two  long  fragments.2 
In  the  Middle  English  period  a  version  was  fashioned  from  the 
RB  redaction  (Klebs,  p.  460).  Of  this  only  a  fragment  (ed. 
Smyth,  pp.  49-55)  of  some  two  hundred  and  fifty  lines  in  iambic 
tetrameter  verse  is  left.  It  is  written  on  two  vellum  leaves  once 
used  as  the  cover  of  a  book.  The  language  of  this  text  seems  to 
antedate  that  of  Gower  and  to  be  of  the  Southwest  Midland 
district  (Smyth,  p.  49).  The  concluding  lines  state  that  the 
poem  was  "  translatyd  Almost  at  Engelondes  ende,"  by  one  who 
was  vicar  at  "  Wymborne  mynstre  "  (Dorsetshire).  In  style  the 
poem  resembles  the  usual  rimed  chronicle.  Gower  produced  in 
1390-1393  a  long  and  fairly  well  told  version  of  Apollonius  in 
his  Confessio  Amantis,  Bk.  vin,  271  ff.  (ed.  Macaulay,  Oxford, 

2  Corpus  Christi  College,  Cambridge,  MS.  201,  ed.  J.  Zupitza,  Archiv, 
xcvii,  17-35  (1896);  cf.  Anglia  1,  463;  also  "  Welcher  Text  liegt  d.  altengl. 
Bearbeitung  der  Erzahlung  v.  Apollonius  z\i  Grunde,"  Rom.  Forsch.  m, 
269-79;  R.  Markisch,  "Die  ae.  Bearbeitung  der  Erzahlung  v.  Apollonius," 
Palaestra,  vi  (Berlin,  1899).  The  Old  English  version  of  Apollonius  has 
special  interest  as  presaging,  before  the  Norman  Conquest,  the  introduction 
into  England  of  the  spirit  of  romance.  See  Peters,  p.  51,  for  monastic  cata- 
logues of  the  eighth  or  ninth  century  referring  to  copies  of  Apollonius. 


1 66  MEDIEVAL   ROMANCE    IN    ENGLAND 

1901).  This  was  professedly  based  an  Godfrey's  Pantheon,  but 
it  is  evident  from  the  number  of  details  omitted  in  the  Pantheon 
but  found  in  the  Historia,  that  Gower  was  drawing  on  the 
earlier  Latin  narrative  (Macaulay,  p.  537).  It  was  Gower's  ver- 
sion, presumably,  which  brought  forth  Chaucer's  much  debated 
fling  at  "  swich  cursed  stories."  3  A  fifteenth-century  English 
verse  translation  of  Apollonius,  surviving  in  MS.  Douce  216  in 
a  fragment  of  one  hundred  and  forty  lines,  was  noted  by  Warton- 
Hazlitt  (Hist.  Eng.  Poetry,  1871,  11,  303)  and  by  Klebs  (p.  472). 
A  prose  romance,  Kynge  Appolyn  of  Thyre,  translated  from  the 
French  by  Robert  Copland  and  printed  by  Wynkyn  de  Worde 
in  1 5 10,  Twine's  prose  novel,  already  mentioned,  the  partly 
Shakespearean  play,  Pericles,  Prince  of  Tyre,  1609,  in  which  the 
change  of  the  hero's  name  from  Apollonius  to  Pericles  was  a 
reminiscence  of  the  great  Athenian  statesman  and  of  Pyrocles, 
one  of  the  heroes  in  Sidney's  Arcadia  (Lee,  p.  10,),  the  novel, 
Pericles,  by  George  Wilkins,  1608,  which  was  based  in  part, 
according  to  Baker  (PMLA.  1908,  xxm,  103),  on  Wilkins 's  own 
dramatic  version  of  Twine's  book,  and  finally  George  Lillo's 
play,  Marina,  1738,  make  up  the  list  of  the  eight  or  nine  ver- 
sions of  the  story  in  England  (Smyth,  p.  48).  The  most  im- 
portant single  influence  on  these  sixteenth  century  versions  was 
Gower's  poem  which,  having  been  printed  by  Caxton  in  1483  and 
again  reprinted  in  1532  and  1554,  was  the  most  readily  accessible 
version. 

The  extant  Continental  texts  of  Apollonius  represent  probably 
but  a  small  part  of  those  once  in  existence.  Numerous  allusions 
in  Old  French  and  Provengal  poetry  of  the  late  twelfth  and  the 
early  thirteenth  centuries  show  knowledge  of  the  story.4     An 

3  Cf.  E.  P.  Hammond,  Chaucer  Manual,  pp.  278-79;  Wells,  Manual, 
p.   699. 

*  Cf.  Marden,  p.  xxv-vi;  Lewis,  pp.  147-50;  Singer,  2,  passim.  One  of 
the  earliest  Old  French  references  to  Apollonius  is  in  Philomena  which  refers 
to  the  delights  of  Apoloines  and  of  Tristanz.  Cf.  Faral,  Rom.  xliii, 
443-45.  The  former  romance  may  actually  have  influenced  the  Tristan. 
Like  Apollonius,  Tristan  after  a  great  storm  and  a  shipwreck  comes  to 
land,  is  received  by  a  kindly  king,  plays  before  the  court,  and  later  in  the 
Irish  court  is  admired  by  a  princess  and  becomes  her  instructor.  In  the 
Apollonius  the  wicked  Dionysias,  jealous  of  the  loveliness  of  the  young 
Tarsia,  who  had  been  left  by  her  father,  Apollonius,  in  the  care  of  Dionysias, 
orders  her  slave,  Theopilus,  to  kill  Tarsia,  but  the  maiden  is  carried  off  by 
pirates  and  the  slave  reports  her  death.     In   Tristan  Iseult  orders  two  serfs 


APOLLONIUS  ^7 

allusion  in  the  Poeme  Moral  (ed.  W.  Cloetta,  1887),  cir.  1200, 
to  "  les  vers  d'Apoloine  u  d'Aien  d'Avinion,"  has  been  happily 
confirmed  by  the  discovery  of  a  fragment  telling  of  the  riddle- 
solving  of  Apollonius  in  an  Old  French  metrical  version  of  the 
tale  (Schulze,  p.  226).  A  French  prose  translation  (ed.  Lewis, 
Rom.  Forsch.  xxxiv)  of  the  Latin  romance  exists  in  four  manu- 
scripts of  the  fourteenth  and  fifteenth  centuries  and  in  three  free 
versions,  of  which  one,  in  a  Brussels  manuscript,  is  unconvinc- 
ingly  ascribed  by  Lewis  (p.  222)  to  Adam  de  la  Hale  {Rom. 
xliii,  444).  Neither  the  versions,  however,  nor  the  references, 
fully  indicate  the  influence  of  the  romance  in  French  literature. 
In  the  early  thirteenth-century  chanson  de  geste,  Jourdains  de 
Blaivies,  many  adventures  of  the  hero,  his  shipwreck,  his  arrival 
in  a  foreign  land,  his  test  of  skill  with  the  king,  his  marriage 
with  the  princess,  their  separation  and  reunion,  and  the  rescue 
of  their  daughter  from  a  fate  similar  to  Tarsia's,  appear  to  be 
borrowed  from  Apollonius  (Smyth,  p.  80  ff.).  A  belated  French 
version  purporting  to  come  from  "  une  histoire  tiree  du  Grec," 
was  inserted  by  Belleforest  in  his  Histoires  Tragiques,  1582 
(Klebs,  p.  421). 

In  Germany  the  story  was  known  as  early  as  the  twelfth 
century,  for  the  poet  Lamprecht  alluded  to  it  in  his  Alexander- 
lied,  v.  1009  (ed.  Kinzel,  1884),  and  the  author  of  the  Middle 
High  German  poem,  Orendel  (cf.  Edwardes,  Summary,  p.  321), 
combined  episodes  drawn  from  our  story  with  those  of  a  ficti- 
tious Crusading  king,  Orendel  of  Treves.  The  earliest  complete 
and  independent  version  (ed.  Singer,  1906)  was  that  of  a 
Viennese  physician,  Heinrich  von  Neustadt  (1300),  who  stated 
at  the  end  of  some  twenty  thousand  lines  of  verse  that  he  was 
the  first  to  translate  the  story  from  Latin  into  German.  The 
value  of  this  poem  has  been  variously  estimated ;  Klebs  pre- 
ferred the  older  Latin  version,  but  Bockhoff  (pp.  1-25)  found 
Heinrich's  an  important  mediaeval  embellishment  of  the  ancient 
story.  The  adventures  of  Apollonius  with  the  gigantic  peoples, 
Gog  and  Magog,  with  centaurs  and  sirens,  with  which  Heinrich 


(in  Eilhart's  version,  two  poor  knights)  to  take  Brangwein  into  the  forest 
and  bring  back  evidence  of  her  death.  Cf.  Emare  here,  note  14.  Teubert, 
Crescentia  Studien  (see  Florence  of  Rome  bibliography),  p.  7,  asserted  that 
the  Crescentia  legend  showed  the  influence  of  Apollonius  in  the  account  of 
the  shipwreck  of  the  heroine  and  of  her  meeting  with  a  fisherman. 


1 68  MEDIEVAL   ROMANCE    IN    ENGLAND 

filled  in  the  fourteen  years  of  Apollonius's  wanderings  after  he 
had  left  his  infant  daughter,  Singer  (3,  pp.  29-70)  believed 
were  drawn  from  a  lost  Byzantine  romance.  As  there  is  no  ex- 
ternal evidence,  however,  to  support  this  conjecture,  Pettingill's 
citation  (3)  of  at  least  one  instance  in  which  the  poet  proved 
himself  possessed  of  a  measure  of  originality  has  special  interest. 
In  the  episode  of  the  combat  between  a  man  and  a  woman  which 
takes  place  before  the  hero  and  his  knights,  Heinrich  evidently 
drew  for  the  rules  of  the  whole  strange  procedure  on  contempo- 
rary law  and  the  possible  reminiscence  of  an  incident  recorded 
in  the  Chronik  de  Berne  (1288).  But  despite  the  elaboration 
of  Heinrich's  story,  it  was  not  until  the  fifteenth  century  that 
Apollonius  was  really  popularized  in  Germany.  This  happened 
through  the  increasing  number  of  texts  of  the  Gesta,  through  the 
very  popular  version  (Augsburg,  147 1)  by  Heinrich  Steinhowel, 
who  asserted  that  he  followed  "  Doctor  Gotfrids  von  Vitterben  " 
but  more  probably  used  both  the  Gesta  and  the  Historia,  and 
through  two  fifteenth-century  German  prose  translations  of  the 
Historia  (Smyth,  pp.  25-30). 

Among  other  early  vernacular  versions  of  Apollonius  the  fol- 
lowing are  the  most  important:  an  old  Danish  ballad  (Grundtvig, 
Danmarksgamle  Folkeviser,  11,  466;  Klebs,  p.  379)  which  con- 
cerns itself  with  the  shipwreck  of  the  hero  and  has  certain  affilia- 
tions with  the  Old  French  poem,  Jourdain  de  Blaivies,  and  the 
German  Orendel  (cf.  Meyer,  Zts.  /.  deut.  Altertum,  xxxvn, 
325-56) ;  a  Spanish  poem  of  the  early  thirteenth  century,  the 
Libro  de  Apolonio  (ed.  Marden,  191 7),  based  on  Latin  prose  ver- 
sions and  significant  chiefly  for  its  peculiar  metrical  form,  "  de 
nueva  maestria,"  and  for  its  moralizing  tendencies ;  references  in 
Alfonso  the  Wise's  Grande  e  General  Estoria,  showing  that  it 
was  intended  to  introduce  the  story  there  (Marden,  p.  xxxiii) ; 
a  Portuguese  version  made  in  the  fourteenth  century  by  Roberto 
Paym  of  Gower's  Confessio  Amantis  and  later  turned  into 
Spanish  prose  (ed.  Birch-Hirschfeld,  Leipzig,  1909) ;  a  Spanish 
prose  version  in  the  Patrahuela,  1576,  of  Juan  de  Timoneda 
(Klebs,  pp.  384-411) ;  three  fourteenth-century  Italian  prose 
translations  based  on  the  Historia  (Cf.  L.  del  Prete,  Apollonio 
in  vol  gar  e  italiano,  Lucca,  1861 ;  Salvioni,  Versione  Tosco- 
Veneziana,  Turin,  1889),  one  of  which  served  as  the  source  of  a 


APOLLONIUS  it  69 

Greek  metrical  version  of  the  fifteenth  century ;  and  a  fourteenth 
century  Italian  poem  attributed  to  Antonio  Pucci  (Klebs,  pp. 
422-450).  This  poem  was  often  reprinted  down  to  the  eighteenth 
century.  In  modern  Greek  there  are  two  sixteenth-century  ver- 
sions of  the  Apollonius  story  (W.  Wagner,  Mediaeval  Greek 
Texts,  Lond.  1871,  1)  and  a  modern  folk-tale  (Klebs,  pp.  451-58). 

Origin.  The  earliest  reference  to  the  story  of  Apollonius 
is  found  in  sacred  lyrics  of  Venantius,  Bishop  of  Poitiers,  566— 
568,  who  sadly  compared  his  wanderings  in  Gaul  to  those  of 
Apollonius.  A  grammatical  index,  Tract  at  de  dubiis  Nominibus, 
of  somewhat  later  date,  makes  passing  mention  of  the  "  balneum 
in  Apollonio  "  (Smyth,  pp.  21-23).  It  would  seem  then  that 
a  Latin  version  of  the  story  must  have  been  known  in  western 
Europe  as  early  as  the  end  of  the  fifth  or  the  beginning  of  the 
sixth  century.  This  was  slightly  Christianized,  though  Klebs 
(p.  189  ff.),  in  his  very  comprehensive  study  of  the  romance, 
showed  that  some  features,  such  as  the  episode  of  the  fisher- 
man's sharing  his  garment  with  Apollonius,  supposedly  inspired 
by  the  legend  of  St.  Martin  (Riese,  p.  xviii),  or  of  Apollonius's 
vow  not  to  cut  his  beard  or  hair,  a  vow  considered  by  some  to 
hint  of  Judaic  custom,  do  not  prove  any  pietistic  influence.5 
Klebs  (p.  189  ff.)  found  the  only  evidence  of  Christian  influence 
in  the  oldest  texts  in  the  recurrent  use  of  "  Deus,"  and  in  the 
description  of  "  quendam  angelico  habitu,"  who  warns  Apollonius 
to  return  to  Ephesus.  The  many  allusions  to  pagan  gods,  to 
Neptune,  Lucina,  Apollo,  Diana,  in  whose  temple  the  wife  of 
Apollonius  is  supposed  to  serve  as  a  priestess,  the  descriptions 

5  Rohde,  p.  447,  conjectured  that  the  whole  of  the  Antiochus  episode 
was  introduced  by  a  Christian  scribe  who  thus  motived  Apollonius's  long 
absence  from  home  through  the  hero's  fear  of  the  incestuous  king  whose 
criminal  secret  he  had  discovered.  In  the  Greek  romance  of  Antheia  and 
Habrocomes,  which  closely  parallels  a  large  part  of  Apollonius,  the  hero's 
absence  from  home  is  due  to  the  command  of  an  oracle.  But  when  so  many 
pagan  features  were  left  unchanged  in  Apollonius  it  would  seem  incredible 
that  this  long  episode  should  have  been  introduced  simply  to  do  away  with 
the  pagan  oracle.  Burger,  p.  17,  n.  22,  suggested  that  the  wicked  Antiochus 
and  his  daughter  from  the  first  belonged  to  the  romance  as  figures  of 
dramatic  contrast  to  the  virtuous  Apollonius  and  Tarsia.  On  the  Incest 
theme  itself  see  Baum,  "  The  Legend  of  Judas  Iscariot,"  PMLA.  xxxi,  593 
ff.  (1916) ;  O.  Rank,  Das  Inzest-Motij  in  Dichtung  u.  Sage,  Leipzig,  1912; 
See  Emare  here,  note  6. 


l7o  MEDIAEVAL   ROMANCE    IN    ENGLAND 

of  pagan  feasts  and  rites  of  burial,  the  atmosphere  and  localiza- 
tion of  the  story  in  Antioch,  Tyre,  Tarsus,  Cyrene,  Ephesus, 
Mitylene,  unquestionably  point  to  pagan  antiquity  for  the  origin 
of  the  story.  To  Rohde  (p.  441),  as  to  the  majority  of  scholars, 
the  civilization  suggested  by  the  romance,  the  love  of  arts  and 
letters,  the  actual  resemblance  of  the  tale  to  Greek  sophist  ro- 
mances, indicate  a  pagan  Greek  original  of  the  second  or  third 
century.  Greek  tales  of  that  period 6  drew  upon  much  the 
same  "  universal  apparatus  of  romance,  pirates,  sea  storms,  ap- 
parent death,  the  separation  and  reunion  of  lovers  "  (Smyth, 
p.  11).  Klebs  (p.  191),  however,  attacked  this  theory  of  Greek 
origin  on  the  ground  that  the  references  to  Roman  coins,  the 
nature  of  the  inscriptions  which  in  the  story  were  put  upon 
monuments  raised  to  Apollonius  and  his  daughter,  Tarsia,  the 
incidental  allusions  to  specifically  Roman  customs,  the  literary 
style  of  the  narrative,  with  its  rhetorical  plays  on  words,  its 
Virgilian  descriptions  of  the  storm,  of  the  awakening  love  of 
the  Princess  for  the  storm-tossed  hero,  like  Dido's  for  ^neas, 
were  possible  only  for  a  pagan  Latin  author  of  the  first  part  of 
the  third  century.  This  author,  in  the  opinion  of  Klebs  (p.  299), 
was  aware  of  the  literary  effectiveness  of  the  Incest  theme  used 
by  Seneca  and  by  Ovid,  and  of  the  farcical  relief  afforded  by  the 
character  of  the  bawd  in  Plautine  comedy.  From  such  sources 
then  he  drew  ideas  for  the  episode  of  the  Incestuous  Father 
which  begins  the  story,  and  for  the  scene,  thrice  occurring  in 
the  plays  of  Plautus,  in  which  an  innocent  maiden  preserves  her 
purity  in  a  brothel.  From  this  original  Klebs  (pp.  215  ff.)  be- 
lieved there  were  made  various  intermediate  versions  in  which 
the  story  gradually  acquired  that  popular  tone  of  a  Volksbuch 
evident  in  the  earliest  extant  text. 

The  conclusion  that  there  was  a  Latin  version  of  Apollonius 
in  the  third  century  is  now  generally  accepted.  But  the  dis- 
covery in  1890-19007  of  new  fragments  of  the  Greek  romance 

6  Cf.  the  Babylonian  Histories  of  the  Syrian,  Iamlichus,  166-180  a.d.; 
the  Ethiopian  tales  of  Heliodorus,  250-300;  the  Lencippe  and  Clitophon  of 
Achilles  Tatius  (after  the  fourth  century).  Convenient  references  are  in 
S.  Wolff's  Greek  Romances  in  Elizabethan  Fiction,  N.  Y.  1912.  Cf.  Rohde, 
p.  441 ;  Smyth,  p.  12. 

7  Cf.  Williamowitz's  review  of  Fayum  Towns  and  Their  Papyri  by 
Grenfell,  Hunt,  and  Hogarth,  London,  1912,  in  Gottinger  Gelehrte  Anzeigen, 
1,  30  (1901);  also  Burger.  For  the  text  of  Chaereas  see  Fayum  Towns,  pp. 
76-80. 


APOLLONIUS  I7I 

of  Chaereas  and  Callirrhoe  which  seem  to  prove  that  its  com- 
position must  have  been  before  150  a.d.,  has  again  raised  the 
question  of  an  even  earlier  text  for  Apollonius.  Since  these  frag- 
ments antedate  any  known  to  Rohde,  there  is  a  heightened  pos- 
sibility of  a  Greek  original  for  Apollonius.  In  this,  in  Burger's 
opinion  (p.  26),  the  familiar  marchen  motifs  and  the  literary 
style  were  sufficiently  similar  to  the  Greek  tale  which  was  the 
source  of  Apuleius's  version  of  the  Amor  and  Psyche 8  story 
in  the  Golden  Ass,  to  justify  considering  them  as  a  special,  and 
probably  contemporary,  group  of  the  second  century.  From 
this  Greek  Apollonius  Burger  thought  Zenophon  of  Ephesus  in 
the  third  century  drew  the  episodes  in  his  romance,  Antheia 
and  Habrocomes,  which  have,  on  any  other  supposition,  a  wholly 
unexplained  resemblance  to  those  in  Apollonius.2 

Whatever  the  "  original  "  Apollonius  may  have  been,  it  is  gen- 
erally conceded  that  the  riddles  which  appear  in  the  mediaeval 
versions  did  not  belong  to  it.  In  the  RA  redaction  there  are 
ten  of  these  riddles,  in  the  RB  group  seven,  in  the  mixed  texts 
a  number  which  varies  upwards  from  three  (Klebs,  p.  178). 
The  riddles  were  derived  from  the  one  hundred  Enigmata  of 
Symposius  and  were  probably  composed  in  the  fourth  or  fifth 
century.  In  the  story  the  riddles  are  proposed  to  her  father 
by  the  unrecognized  Tarsia  who  thus  endeavors  to  drive  away 
his  melancholy.10 

Strange,  shapeless,  improbable,  as  in  its  entirety  is  "  the 
mouldy  tale,"  as  Jonson  called  it,  of  Apollonius,  there  is  need 
in  any  final  estimate  to  pay  tribute  to  the  venerableness  of  its 
history  and  to  the  enduring  appeal  made  by  what  might  well  be 
called  the  first  of  our  western  romans  d'aventure.     The  words 

8  For  the  folk-tale  elements  in  Amor  and  Psyche  see  Partonope  here, 
note  12.  Burger  noted  that  in  Apollonius  the  Wicked  Father,  the  Dangerous 
Riddle,  the  Fate  of  the  Suitors,  the  Jealous  Guardians  of  the  heroine,  the 
plan  for  her  murder,  all  occur  in  well-known  marchen.  See  also  Rohde, 
pp.  448  ff. 

9  Cf.  Rohde,  pp.  409,  ff.  440;  Smyth,  p.  11,  for  analysis  of  the  story. 
The  likenesses  of  the  two  romances  forced  even  Klebs,  pp.  298,  306,  to 
admit  the  possibility  of  a  Greek  source.  Singer's  attempt  (3,  191 1)  to 
prove  the  existence  of  a  Byzantine  original  from  supposed  traces  in  the 
Apollonius  of   Heinrich   von   Neustadt  was  not  convincing. 

10  See  Smyth,  Pericles,  p.  88  ff.  on  the  relation  of  this  peculiar  style  of 
dialogue  to  that  employed  in  the  Solomon-Markolf  type  of  story. 


1 72  MEDIEVAL   ROMANCE    IN    ENGLAND 

that  Shakespeare  or  another  put  on  the  lips  of  "  ancient  Gower  " 
still  apply  to  it: 

"  It  hath  been  sung  at  festivals, 
On  ember-eves,  and  holy-ales; 
And  lords  and  ladies  in  their  lives 
Have  read  it  for  restoratives. 
The  purchase  is  to  make  men  glorious; 
Et  bonum  quo  antiquius,  eo  melius." 

—  Pericles. 

BIBLIOGRAPHY 

Texts:  (i)  Fragments  of  two  leaves  printed  by  J.  0.  Halliwell- 
Philipps,  A  New  Boke  about  Shakespeare  and  Stratford-upon-Avon, 
1850;  printed  also  by  A.  H.  Smyth  (see  below).  For  the  Old  English 
version  see  note  2.     (2)  Gower's  Apollonius,  see  Macaulay  below. 

Studies  and  Analogues:    Cf.  Betz,  La  Litt.  Comparee,  Index;  Ed- 
wardes,  Summary,  Index;  Wells,  Manual,  p.   784. 
Beck,  J.  W.     Quaeritur  an  recensio  Christiana  Historiae  Apollonii  Regis 

Tyri  in  Gallia  orta  esse  possit.     Album  —  in  honorem  H.  von  Her- 

werden.     Rheims,   1902. 
Bockhoff,  A.  und  S.  Singer  (see  below).     "  Heinrichs  v.  Neustadt  Apol- 
lonius u.  seine  Quellen,  Ein  Beitrag  z.  mittelhochdeut.  u.  byzantin- 

ischen  Literaturgesch."  Sprache  u.  Dichtung,  vi,  191 1.  Rev.  Zts.  f. 

rom.  Phil,  xxxvm,  381  ff. ;  Literaturbl.  xxxiv,  363. 
Burger,  K.     Studien  zur  Geschichte  des  griechischen  Romans,  Teil  n, 

Die     litteraturgeschichtliche     Stellung  .  .  .  d.     historia     Apollonii. 

Blankenburg,   1903. 
Herbert,   J.     Catalogue   of   Romances    (The    Gesta   Romanorum),   vol. 

in,  pp.   183-282.     Lond.,   1910. 
Klebs,    E.     Die    Erzahlung    v.    Apollonius,    Eine    geschichtliche    Unter- 

suchung    uber   ihre    lateinische    Urform    u.    ihre  .s  pater  en   Bearbei- 

tungen.     Berlin,   1899.     Rev.   Anglia  Bbl.   x,   233. 
Lee,  Sidney.    Shakespeare's  Pericles.    (Facsimile,  Introd.,  Notes)  Oxford, 

1905. 
Lewis,  C.  B.     "Die  altfrz.     Prosaversionen  des  Appollonius  —  Romans 

nach  alien  bekannten  Handschriften."  Rom.  Forsch.  xxxiv,  1-277 

(1913).     Rev.  Rom.  xliu,  443. 
Macaulay,  G.     Works  of  John  Gower   (Confessio  Amantis)   11,  393  ff. 

Oxford,   1901. 
Marden,    C.     Libro    de   Apolonio,   An   Old   Spanish    Poem.     Baltimore, 

Paris,    1917-1922. 
Patterne   of   Paineful  Adventures  by  Lawrence   Twine,    1576,   repr.   J, 

Collier,  1843,  repr.  New  Rochelle,  N.  Y.,  1903. 


APOLLONIUS  173 

Peters,  R.  Die  Geschichte  des  Konigs  Apollonius,  Der  Lieblingsroman 
des  Mittelalters.  (Introduction  and  German  translation.)  Leipzig, 
1904. 

Pettengill,  R.  W.  (1)  The  Apollonius  von  Tyrland  of  Heinrich  v. 
Neustadt,  A  Study  of  Its  Sources.     Unpublished  Harvard  diss.  19 10. 

(2)  "Zu  den  Ratseln  im  Apollonius  des   Heinrich  v.   Neustadt," 
JEGP.  xii,  248-51  (1913)- 

(3)  "The  Source  of  an  Episode  in  Heinrich's  v.  Neustadt's  Apol- 
lonius," JEGP.  xiii,  45-50  (1914)- 

Rohde,  E.    Der  griechische  Roman  u.  seine  Vorldufer.  3rd  ed.  Leipzig, 

1914. 
Ropohl,    F.    Das   Verhaltnis   d.   Assonanzteiles   z.   Reimteile   im   altfrz. 

Apolloniusroman  (Jourdains  de  Blaivies).     Diss.  95  pp.  Kiel,  1909. 
Schreiber,    E.    W.    Zum    Texte    d.    Historia    Apollonii    (Summary    and 

extracts  from  a  twelfth  century  Viennese  MS.)   Korneuburg,   1909. 
Schulze,  A.     "  Ein  Bruchstiick  des  altfrz.   Apolloniusromanes."  Zts.  f. 

rom.  Phil,  xxxiii,   226-29   (1909). 
Singer,    S.      (1)    Heinrichs    von    Neustadt    Apollonius    von    Tyrland. 

Deutsche   Texte   des  Mittelalters,  vn.     Berlin,    1906. 

(2)  Apollonius,    Untersuchungen   uber   das   Fortleben   des   antiken 
Romans  in  spat  em  Zeiten.    Halle,  1895. 

(3)  See  Bockhoff  above. 

Smyth,  A.  H.  Shakespeare's  Pericles  and  Apollonius  of  Tyre,  A  Study 
in  Comparative  Literature.    Philadelphia,   1898. 


THE    SEVEN    SAGES    OF    ROME 

Versions.  The  great  mediaeval  vogue  of  the  various  ver- 
sions of  the  Seven  Sages,  which  in  actuality  is  a  collection  of 
tales  rather  than  a  romance,  once  rivalled  that  of  Apollonius. 
Literally  hundreds  of  texts  are  known,  representing  in  fact  or 
theory  almost  every  age  and  race.  Bewildering  as  even  Gaston 
Paris  admitted  this  chaos  of  versions  to  be,  it  is  nevertheless 
possible  to  note  certain  clear  lines  of  demarcation.  The  primary 
division  is  between  the  versions  of  the  East  and  the  West.  In 
the  Eastern  group  eight  versions  are  found  of  the  Book  of 
Sindibad.  In  general  they  tell  of  the  philosopher,  Sindibad,  to 
whom  the  education  of  a  young  prince  is  entrusted.  Death 
nearly  befalls  this  young  man  when  his  royal  stepmother  falsely 
accuses  him  of  attempting  to  seduce  her,  but  his  life  is  saved 
by  seven  sages  who  for  seven  days  rival  with  tales  of  false  women 
those  concerning  false  counsellors  which  the  Queen  tells  her  hus- 
band. On  the  eighth  day  the  Prince  speaks  in  his  own  defense 
and  the  Queen  is  exiled  or  killed.  The  eight  versions  are  as 
follows:1  Sindban  (English  translation,  Folk-Lore,  vin,  99  ff.), 
a  tenth-century  Syriac  text ;  the  Greek  Syntipas  (ed.  Eberhard, 
Fabulae,  1872),  possibly  of  the  last  half  of  the  eleventh  century; 
a  Spanish  text,  Libro  de  los  Engannos  (ed.  A.  Bonilla  y  San 
Martin,  Barcelona,  1914),  written  in  1253  and  seemingly  de- 
rived from  the  same  lost  Arabic  text  which  was  the  source  of 
Sindban;  a  Hebrew  version,  Mischle  Sindbad;2  a  Persian  prose 

1  Cf.  Chauvin,  vm,  1-2 1 ;  Campbell,  pp.  xii-xrv,  for  full  bibliographical  de- 
tails; also  the  standard  study  of  Comparetti,  Ricerche  intorno  al  libro  di 
Sindibad,  Milan,  1870;  Eng.  trans.  Folk-Lore  Soc.  ix,  Lond.  1882;  Ward, 
Catalogue  of  Romances,  u,  192   ff.;  Hilka,  1,  pp.  vin,  xxi. 

2  Ed.  with  German  translation  by  P.  Cassel,  Berlin,  1888.  Questions  of 
date  and  origin  are  much  disputed  in  regard  to  this  version.  Hilka  (1,  p.  x) 
agreed  with  those  who  believed  it  to  be  of  Arabian  origin,  and  put  its 
composition  in  the  first  half  of  the  thirteenth  century.  Faral,  Rom.  xlii, 
147,  asked  whether  Hilka's  15th  century  Latin  version  was  not  a  late 
translation  from  the  Hebrew  by  an  author  familiar  with  and  influenced 
by  a  western  form  of  the  story.     Campbell,   p.  xvi,  noted  as  characteristic 

174 


THE    SEVEN    SAGES    OF   ROME  175 

text  of  As-Samarquandi  of  the  late  twelfth  century;  another  in 
the  Touti-Nameh  (Eighth  Night)  of  NachsebT  (d.  1329),  and, 
most  important  of  the  Persian  texts,  the  poem,  Sindibad-nameh, 
1375  (Campbell,  pp.  xii-xiv).  A  late  Arabic  version  known  as 
the  Seven  Vezirs  is  extant  in  three  versions  of  the  Arabian 
Nights  Tales.  Of  the  eight  texts  the  first  four  are  the  oldest 
and  most  authentic,  and  of  these  the  Mischle  Sindbad,  from  the 
point  of  view  of  European  story,  is  the  most  interesting,  for  it 
has  been  held  to  be  the  most  probable  intermediary  between 
the  Eastern  and  the  Western  versions.  In  Hilka's  opinion  (1, 
p.  x)  the  early  fifteenth-century  Latin  manuscript,  discovered 
by  him  in  191 1,  was  derived  from  the  Hebrew  text  and  most 
truly  represents  the  intermediate  stage. 

In  the  Western  versions  of  the  Seven  Sages  there  is  no  men- 
tion of  Sindibad,  and  the  Sages  have  individual  names,  a  trait 
which  in  the  East  appears  only  in  the  Hebrew  version.  The 
Sages  tell  only  one  story  each  instead  of  two  or  more,  as  in  the 
Eastern  texts ;  the  instruction  of  the  Prince  is  entrusted  to  them, 
not  to  the  one  philosopher,  Sindibad ;  and  only  four  of  the  origi- 
nal tales  (Canis,  Aper,  Senescalus,  Avis)  reappear  in  the  fifteen 
tales  that  are  normally  told  in  the  Western  redactions.3  In 
Europe  there  are  at  least  forty  different  versions,  preserved 
in  upwards  of  two  hundred  manuscripts  and  nearly  two  hun- 
dred editions  (Campbell,  p.  xvii).  In  1866  began  the  first 
serious  attempts  at  the  classification  of  this  material,  a  task 
more  nearly  achieved  in  regard  to  certain  versions  by  Gaston 
Paris  in  his  edition  of  Deux  Redactions  du  Roman  des  Sept 
Sages,  SATF.,  1876,  than  by  any  one  else.  The  most  complete 
summaries  of  the  literature  relating  to  the  subject  were  given 
by  Chauvin  in  1904  and  by  Campbell  in  his  1907  edition  of  the 
Middle  English  versions. 


features  of  the  Hebrew  version:  the  individual  names  of  the  sages;  the 
rivalry  of  the  sages  in  attempting  to  procure  the  task  of  instructing  the 
prince;  the  defense  of  the  prince  by  the  sages,  not  by  the  king's  counsellors; 
peculiarities   in    the   stories    of   Aper   and   Avis. 

3  The  traits  differentiating  the  versions  may  be  roughly  summarized  as 
follows:  number  and  order  of  stories;  the  subject-matter  of  the  stories;  the 
localization  of  the  framework  tale;  the  names  of  the  personages,  of  the  king, 
queen,  prince,  sages.  These  details  differ  widely.  For  comparative  tables 
of  the  tales  in  the  Eastern  versions  see  Bedier,  Les  Fabliaux,  p.  136;  for 
those  in  the  Western  versions,  Campbell,  p.  xxxv;  Smith,  p.  5. 


I76  MEDIEVAL  ROMANCE   IN  ENGLAND 

The  first  European  version  of  the  Seven  Sages  is  found  in  the 
Dolopathosy  a  Latin  prose  work  now  extant  in  six  manuscripts. 
It  was  written  about  1190  by  Joannes  de  Alta  Silva  and  trans- 
lated not  long  afterwards  by  the  French  poet,  Herbert.4  In  the 
Dolopathos,  as  in  the  Eastern  versions,  the  prince  has  only  one 
instructor,  here  the  poet,  Virgil,  but  of  the  nine  tales  told  only 
one,  Cants,  is  found  also  in  the  Book  of  Sindibad.  Four  of  the 
stories  {Cants ,  Gaza,  Puteus,  Inclusa)  in  the  Dolopathos  agree 
with  those  in  the  later  western  redactions  of  the  Seven  Sages, 
and  it  seems  probable  that  the  Latin  text  was  simply  an  inde- 
pendent derivative  of  a  version  which  must  have  been  current 
before  John  wrote  (Campbell,  p.  xx).  The  second  group  of  ver- 
sions is  represented  by  an  Old  French  poem  (K),  Les  Sept  Sages 
de  Rome  (ed.  Keller,  Tubingen,  1836),  which  Paris5  believed 
was  composed  about  1155,  though  the  one  extant  manuscript 
is  of  the  thirteenth  century.  From  the  same  source  as  K  came 
a  third  version  in  a  single  French  prose  text,  the  "  version 
derimee  "  (ed.  G.  Paris,  Deux  Redactions,  pp.  1-55),  and  also 
possibly  the  large  fourth  group  (A*)  of  versions  of  the  type  rep- 
resented by  the  fifteenth  century  Italian  prose  texts  of  the  Libro 
dei  Sette  Savj  de  Roma  (ed.  d'Ancona,  Pisa,  1864).  To  this 
fourth  group  also  belong  two  other  early  Italian  versions,  an 
Old  French  prose  version  still  preserved  in  a  large  number  of 
manuscripts  (cf.  Plomp,  p.  18  ff.),  two  early  Swedish  versions 
(ed.  Klemming,  Stockholm,  1887),  a  Welsh  version,6  and  one  in 
Dutch  verse  (Campbell,  pp.  xxxii-iv).  The  earliest  manuscripts 
of  the  Swedish,  Welsh,  and  Dutch  versions  do  not  antedate  the 
fourteenth  century,  nor  is  the  Auchinleck  manuscript,  the  earliest 
of  the  nine  texts  containing  the  story  in  Middle  English,  of 
earlier  date. 

From  his  work  on  these  texts,  Campbell,  however,  concluded 

4  The  Dolopathos  was  edited  by  H.  Oesterley,  Strassburg,  1873  (rev. 
Gaston  Paris,  Rom.  n,  481-503)  and  by  Hilka,  1913.  For  Herbert's  version 
see  Li  Romans  de  Dolopathos,  ed.  Montaiglon,  Paris,  1856.  For  bibliography 
see  Chauvin,  vm,  pp.  30-31 ;   Campbell,  pp.  xvin-xxi. 

6  Litt.  jrc.  an  Moyen  Age,  2e.  ed.  p.  247. 

6  This  was  a  prose  redaction  written  by  the  Welsh  priest,  Llewelyn. 
He  borrowed  almost  verbatim  the  scene  in  which  the  young  Queen  visits 
an  old  witch  to  ask  about  the  King's  son,  from  a  passage  in  Kulhwch  and 
Olwen  in  the  Mabinogion.  The  Welsh  version  appears  in  five  manuscripts, 
the  earliest  dating  from  the  fourteenth  century.  Cf.  Campbell,  p.  xxxm; 
Loth,  p.  3Si. 


THE    SEVEN    SAGES    OF   ROME  jyy 

that  there  was  originally  a  thirteenth-century  English  manu- 
script (V)   from  which  eight  of  the  extant  texts  (C,  R,  A,  E, 

B,  F,  D,  Ar),  were  derived,  though  of  these  only  two,  C  and  R, 
were  copies  of  the  same  manuscript  (Ibid.,  p.  xl).  The  Aslone 
manuscript  (As),  was,  he  thought  (p.  xli),  derived  from  an  Old 
French  manuscript  of  the  A*  group.  All  the  Middle  English 
versions  are  in  the  octosyllabic  couplet,  and  it  seems  probable 
that  their  parent  version  was  written  in  Kent;  the  three  manu- 
scripts, A,  Ar,  E,  are  most  clearly  Kentish  in  dialect;  B,  F,  are 
southern,  perhaps  Kentish ;  D,  is  of  the  Southeast  Midland ;  CR 
is  northern,  and  As,  Scottish  (Campbell,  pp.  xxxvi-xl).  Nothing 
is  known  of  the  authors  of  these  versions,  and  there  seems  little 
to  support  Kolbing's  conjecture  (Art hour  and  Merlin,  1890,  p. 
civ)  that  A  was  written  by  the  author  of  the  Kentish  versions 
of  Art  hour  and  Merlin,  Alisaunder,  and  Richard  Coeur  de  Lion 
(Campbell,  p.  lviii).  A  chronological  classification  of  the  manu- 
scripts shows  that  A,  R,  D,  were  of  the  fourteenth  century, 

C,  Ar,  F,  E,  of  the  fifteenth,  and  R  and  As  of  the  sixteenth. 

In  addition  to  the  four  versions  already  noted,  Gaston  Paris 
classified  the  remaining  Latin,  French,  and  Italian  versions  of 
the  story  in  five  groups.7  Of  these  the  large  group  represented 
by  the  Historia  Sept  em  Sapientum  (H)  is  of  most  interest  in 
its  relations  to  English  literature.  The  oldest  of  the  twenty- 
eight  Latin  manuscripts  of  the  Historia  dates  from  1342  8  and 
the  Latin  text,  printed  first  at  Cologne  in  1475,  was  frequently 
reprinted  and  almost  endlessly  translated    (Buchner,  p.   301). 

7  The  five  groups  are:  (1)  the  fourteenth-century  abridgment  of  a 
lost  Latin  text  made  by  the  Dominican,  Joannes  Junior,  in  the  Scala  Cell; 
(2)  a  Latin  redaction  known  as  the  Historia  Septem  Sapientum,  the  largest 
of  all  the  groups;  (3)  the  Versio  Italica;  (4)  the  French  redactions  of  the 
type  of  the  first  Leroux  de  Lincy  text,  Roman  des  Sept  Sages,  Paris,  1838; 
(5)  the  Male  Marrastre  (Paris,  Deux  Redactions,  p.  xxv).  Paris  considered 
the  Dolopathos  apart  from  the  eight  types  represented  by  the  other  European 
versions  discussed  by  him. 

8  In  Oesterley's  edition  of  the  Gesta  Romanorum  at  least  thirty-five  manu- 
scripts of  the  Gesta  are  described  in  which  all  or  a  part  of  the  Seven  Sages 
appears  in  its  Latin  form  (Campbell,  p.  xxiv).  It  should  be  noted  that 
the  earliest  known  text  of  the  two  famous  collections  of  stories  is  the 
Innsbruck  codex  of  1342.  Cf.  G.  Buchner,  Die  Historia  septem  sapientum 
nach  der  Innsbrucker  Handscrift,  Erlangen,  1889;  Herbert,  Catalogue  of 
Romances,  1910,  111,  184  ff.  The  literary  history  of  the  two  collections  is 
remarkably  parallel.  In  the  Middle  English  versions  of  each  collection  appear 
two  of  the  same  stories,  Virgilius  and  Canis. 


I78  MEDIEVAL   ROMANCE    IN    ENGLAND 

Schmitz  enumerated  twelve  manuscripts,  Fischer  seventy-two 
editions  of  the  German  prose  translation,  and  Campbell  (p.  xxv) 
noted  the  metrical  German  version  of  Hans  von  Buhel  (1412), 
a  tragedy  by  Sebastian  Wildt  (1560)  and  others.  The  Historia 
was  translated  into  French  prose  and  at  least  eight  French  edi- 
tions dating  from  1492  (ed.  Paris,  Deux  Redactions,  pp.  55- 
205)  survive.  A  Spanish  version  (Burgos,  1530)  ran  through 
six  editions,  and  a  Dutch  translation  through  fifteen  or  more  edi- 
tions after  1479.  Two  Swedish  versions  were  made  in  the  fif- 
teenth century,  a  Danish  version  in  the  seventeenth  century,  an 
Icelandic  version  (Campbell,  p.  xxvi),  and  numerous  translations 
into  Slavic  languages.  Of  these  last  the  oldest  extant  text  is 
a  Bohemian  manuscript  of  the  fourteenth  century.  From  a 
Polish  translation,  which  itself  passed  through  eight  editions, 
came  a  Russian  version  now  represented  by  some  forty  manu- 
scripts of  the  seventeenth  and  eighteenth  centuries  (Murko,  Sitz. 
Ph.-hist.  KL.  Abhandl.  x,  Vienna,  1890,  pp.  12-92).  An 
Armenian  version  translated  from  one  of  the  printed  texts  of 
the  Historia  was  made  in  1614  (Macler-Chauvin,  pp.  xiv-xx). 
The  Continental  popularity  of  the  Historia  in  the  fifteenth  and 
sixteenth  centuries  especially  is  reflected  by  numerous  contempo- 
rary editions  in  England.  From  the  Latin  Historia  was  made 
the  translation  printed  by  Wynkyn  de  Worde,  an  edition  which 
served  as  the  source  for  many  of  the  later  English  versions  in 
verse  and  prose. 

Origin.  Whatever  the  sources  of  the  widely  varied  single 
tales  in  the  different  versions  of  the  Seven  Sages,  the  frame- 
work 9  itself  is  Eastern  in  origin.  Not  only  is  the  oldest  Oriental 
text,  the  Syriac  Sindban,  two  centuries  older  than  the  earliest 
European  version,  but  even  in  the  western  versions  of  the  story, 
the  motivating  impulses  remain  inherently  Eastern.  These  bal- 
anced tales  of  designing  women  and  evil  counsellors  are  typical 
of  Oriental  cynicism,  and  typical,  too,  of  the  East  are  the  Brah- 

9  For  other  European  instances  of  stories  in  a  frame-work  see  Ham- 
mond, Chaucer  Manual,  1908,  p.  150;  Wells,  Manual,  p.  185;  Schofield,  Eng. 
Lit. —  to  Chaucer,  p.  337;  Voretzsch,  Einfiihrung,  p.  416.  The  frame-work 
device  is  found  in  the  Marques  de  Rome  (ed.  J.  Alton,  Tubingen,  1889),  one 
of  the  five  continuations  of  the  Roman  des  Sept  Sages  (Paris,  La  Litt.  frc., 
1882,  p.  109;  Tatlock,  MLN.  xxxvi,  96.  n.  5). 


THE    SEVEN    SAGES   OF   ROME  179 

minical  glorification  of  wise  teaching,  and  the  despotism  which 
makes  the  issue  of  life  or  death  dependent  on  mere  story-telling. 
But  where  or  when  the  original  Eastern  version  was  composed 
we  can  only  conjecture.  Some  scholars  have  found  an  ultimate 
basis  for  it  in  the  Indian  story  of  Kunala  and  Acoka,  and  con- 
jectured that  in  India  a  parent  text  was  fashioned  as  early  as 
the  fifth  century  b.c.  (Campbell,  p.  xi).  But  the  available  evi- 
dence shows  us  no  more  than  that  the  extant  texts  of  the  Sages 
were  descendants  in  some  sort  from  a  lost  Arabic  version  written 
by  one  Musa  (Moses)  in  the  eighth  century. 

There  are  various  theories  concerning  the  transmission  of  the 
story  to  Europe.  Paris 10  thought  that  after  it  had  passed 
through  various  forms,  Persian,  Syriac,  Arabic,  Greek,  it  finally 
received  in  the  Byzantine  kingdom  a  new  form,  and  that  this 
Greek  version,  after  it  had  passed  from  the  Eastern  Empire  into 
Italy,  became  the  source  of  the  Western  versions.  In  the  opin- 
ion of  other  scholars,  it  was  not  the  Greek  but  the  Hebrew  ver- 
sion, or,  as  stated  above,  a  Latin  translation  of  this  text,  which 
formed  the  bridge  between  the  Eastern  and  the  Western  ver- 
sions. In  Campbell's  opinion  (p.  xvi)  the  changes  in  the  latter 
were  too  great  to  be  accounted  for  if  the  author  of  the  original 
Western  text  had  been  working  from  any  actual  text.  Rather, 
he  thought  some  returning  Crusader  heard  the  story  in  the  East 
and  told  of  it  on  his  return.  Campbell  granted  (p.  xvii)  the 
"  slight  probability  "  that  it  was  some  form  of  the  Hebrew  ver- 
sion which  was  thus  transmitted.  However  that  may  be,  it  is 
probable  that  the  legend  had  reached  western  Europe  by  the 
middle  of  the  twelfth  century,  and  possibly  somewhat  earlier  if 
Paris's  date  for  the  original  version  of  the  Old  French  metrical 
romance  be  accepted.  In  this,  as  also  in  the  majority  of  the 
derivative  versions,  the  setting  of  the  tale  is  laid  in  Rome,  and 
names  for  the  principal  characters  are,  like  that  of  Diocletian 
the  Emperor,  drawn  from  Roman  history  and  legend  (Campbell, 
p.  xxii).  In  the  later  texts  names  from  Romance  literature  be- 
gin to  intrude  themselves.  In  the  Middle  English  versions,  for 
instance,  the  name  Florentine  for  the  Prince  is  probably  derived 
from  the  romance  of  Octavian  (Campbell,  p.  150).  Of  far 
greater  importance,  of  course,  is  the  substitution,  in  all  but  four 

10  La  Lilt.  frc.  au  Moyen  Age,  p.  82. 


180  MEDIEVAL   ROMANCE   IN   ENGLAND 

instances,  of  new  tales  for  the  fifteen  Oriental  stories  in  the 
Book  of  Sindibad. 

The  sources  and  analogues  of  these  popular  tales  show  upon 
investigation  little  connection  with  mediaeval  romance.  They 
belong  primarily  to  the  fabliau  or  exemplum  type  and  are  often 
anecdotal  and  bourgeois  in  character.  Only  in  three  or  four 
instances  in  any  of  the  versions  have  romance  analogues  been 
pointed  out,  and  these  in  general  show  the  retroactive  influence 
of  romance  on  the  tales.  The  story  (Canis)  of  the  faithful  dog 
which  saves  his  master's  child  somewhat  remotely  parallels  an 
episode  in  the  Latin  romance  of  Arthur  and  Gorlagon.11  The 
story  of  Virgilius  in  the  Middle  English  versions  describes  brazen 
men,  one  with  a  bow  and  arrow,  one  with  a  magic  mirror,  which 
are  said  to  have  been  set  up  by  Virgil  for  the  protection  of 
Rome.  Human  automata,  as  Hertel  and  Bruce  12  have  shown, 
are  familiar  features  in  mediaeval  romance,  but  it  is  probable, 
as  Bruce  (p.  521)  observed,  that  the  conception  of  automata 
had  become  a  commonplace  in  the  romances  before  it  influenced 
the  Virgil  legend.  In  the  story  of  the  Children  turned  into 
Swans  (Cygni),  which  first  appeared  in  the  Dolopathos,13  we 
have  the  earliest  version  of  what  became  the  introduction  to 
the  famous  Knight  of  the  Swan  legend.  In  the  story  of  Alex- 
ander and  his  friend,  Prince  Louis  of  France  (Paris,  Deux  Re- 
dactions, pp.  167-92),  there  is  direct  borrowing  from  the  equally 
well-known  legend  of  Amis  et  Aniile}* 

A  final  word  may  be  said  concerning  Vidua,  the  twelfth  story 
in  the  Seven  Sages.   Despite  its  gruesome  horror  Vidua  has  been 

11  Campbell,  p.  lxxix.  In  the  Latin  romance  a  tame  werwolf  attacks 
the  king's  steward,  who  is  the  lover  of  the  Queen.  The  beast's  attack  is 
motived  by  loyalty  to  his  master  just  as  in  the  more  simple  tale  the  Faithful 
Animal  rescues  his  master's  child.  Cf.  Kittredge,  "Arthur  and  Gorlagon," 
Harvard  Studies,  vm,  ch.  vni,  "Defense  of  the  Child";  Chauvin,  vni,  no.  31. 

12  Cf.  A.  Hertel,  Verzauberte  Ortlichkeiten  u.  Gegenstande  in  d.  altfrz, 
erz'dhlenden  Dichtung,  Diss.  Gottingen,  1908;  Bruce,  Mod.  Phil,  x,  511  ff.; 
Comparetti,  Virgilio  nel  medio  evo,  1896;  G.  Leland,  Unpublished  Legends 
of  Virgil,  N.  Y.,  1900;  Campbell,  p.  xcv;  and  for  further  notes  on  the 
general  subject  of  automata  see  Floris  here,  note  20.  Bruce  (p.  521)  found 
the  earliest  instance  in  which  the  invention  of  automata  is  ascribed  to  Virgil 
in  the  Image  du  Monde,  which  was  not  composed  until  1245.  Comparetti 
thought  the  composition  of  the  Virgilian  automata  was  Oriental  in   origin. 

13  See  Knight  of  the  Swan  here,  note  3. 

14  See  Amis  and  Amiloun  here.     Cf.  Chauvin,  vrn,  no.  235. 


THE    SEVEN   SAGES   OF   ROME  181 

called  "  perhaps  the  most  popular  of  all  stories."  15  It  tells  of 
a  woman  whose  grief  for  her  husband  is  so  assuaged  at  his  very- 
grave  that  she  will  presently  allow  his  body  to  be  subjected  to 
indignity  for  the  sake  of  a  new  lover.  The  tale  goes  by  the 
type  name  of  the  Widow  or  Matron  of  Ephesus,  the  heroine 
of  the  classical  versions.  Since  1879,  much  controversial  argu- 
ment has  been  expended  to  prove  some  connection  between  this 
brief,  cynical  tale  and  the  long,  highly  romantic  and  chivalric 
romance,  Iwain,  of  Chretien  de  Troyes.  Without  necessarily 
accepting  all  of  Brown's  contention  16  that  the  latter  is  a  Celtic 
Fairy  Mistress  tale,  one  may  recognize  from  his  discussion  of 
the  essential  incongruity  between  Vidua  and  Iwain  that  the  two 
stories  are  of  altogether  different  provenance. 


BIBLIOGRAPHY 

Texts:  (i)  A,  Auchinleck  MS.  (2646  lines),  ed.  Weber,  111,  1-153; 
cf.  Kolbing,  Eng.  Stud,  vi,  443  ff. ;  extracts,  Ellis,  Specimens,  m,  1  ff.; 
(2)  R,  Rawlinson  Poet.  175  (new  number  14667),  Bodleian;  cf.  PMLA. 
xiv,  459;  (3)  D,  Camb.  Univ.  Libr.  Dd.  I,  17,  ed.  Wright,  Percy  Soc. 
xvi  (1845);  cf.  Eng.  Stud,  vi,  448;  (4)  C.  Cotton  Galba  E.  ix  (4328 
lines),  ed.  Weber,  1810,  in,  1  ff. ;  K.  Campbell,  Seven  Sages  of  Rome, 
pp.  1-145,  Boston,  1907;  rev.  MLN  xxiv,  153;  (5)  F,  Camb.  Univ.  Libr. 
MS.  Ff.  11,  38  (2555  lines),  extracts  pr.  Halliwell,  Camden  Soc.  xxx, 
p.  xliii;  Wright,  Percy  Soc.  1845,  xvi,  p.  lxx  ff.;  Petras,  Uber  die 
mittelengl.  Fassungen  d.  Sage  von  den  Sieben  Weisen  Meistern,  p.  60 
ff.  Breslau,  1885;  (6)  Ar,  Arundel  140,  Brit.  Mus.;  E,  Egerton  1995, 
Brit.  Mus.  (complete,  3588  lines);  (8)  B,  Balliol  Coll.  Oxf.  354, 
(complete,  3708  lines);  (9)  As,  Asloan,  Malahide  Castle,  Ireland, 
desc.  Varnhagen,  Eng.  Stud,  xxv,  321  ff.  The  later  English  versions 
include:  (1)  a  prose  translation  printed  by  Wynkyn  de  Worde,  1505- 
1515,  repr.  Villon  Soc,  Lond.,  1885;  (2)  a  lost  edition  by  Wm. 
Copland,  1548-61;  (3)  a  metrical  version  by  John  Rolland  of  Dalkeith, 
composed  about  1560,  reprinted  seven  times  before  1631;  (4)  a  lost 
dramatic  version  (cf.  Henslowe's  Diary,  March,  1599-1600);  and 
twenty-six  prose  editions  from  1653  to  the  early  part  of  the  nine- 
teenth century  (Campbell,  pp.  lx-lxv). 

16  Cf.  Joseph  Jacobs,  Caxton's  ZEsop,  1,  13;  Campbell,  p.  ci;  E.  Grisebach, 
Die  Wanderung  der  Novelle  von  der  treulosen  Wittwe  durch  die  Weltlitteratur, 
2  ed.  Berlin,  1889;  Chauvin,  vm,  No.  254. 

16  "  Iwain,  A  Study  in  the  Origins  of  Arthurian  Romance,"  Harvard 
Studies,  vm,  Boston,  1903,  ch.  1. 


!S2  MEDIEVAL   ROMANCE    IN   ENGLAND 

Studies  and  Analogues  :   Cf .  Chauvin  (below) ;  Edwardes,  Summary, 
pp.  no,  191,  427;  Wells,  Manual,  p.  792. 
Bertoni,   G.     "  Un   Manuscrit  du  Roman  des  Sept  Sages  en  prose." 

Zts.  f.  rom.  Phil,  xxxi,  713-15  (1907). 

(2)  "  Sulla  lingua  del  Roman  des  Sept  Sages  in  versi  (ediz.  Keller, 

1836)."    Studj  Romanzi  vi  (1910). 
Bonilla  y  San  Martin,  A.    Libro  le  los  enganos  &  los  asayamientos  de  las 

mugeres.    Barcelona,  1904. 
Botermans,  A.  J.    Die  Hystorie  van  die  seven  wijse  mannen  van  Romen, 

(ed.  1479).    Haarlem,  1898.    Rev.  Rom.  xxvm,  448. 
Buchner,    G.     "  Beitrage    z.    Geschichte    der   Sieben    Weisen   Meister." 

Archiv,  cxm,  297-301   (1904). 
Campbell,  K.     (1)  See  above,  Texts  (4).    Unless  it  is  otherwise  stated, 

references  are  to  the  Introduction  in  this  edition. 

(2)  "A  Study  of  the  Seven  Sages  with  Special  Reference  to  the 
Middle  English  Versions."  PMLA.  xiv,  1-107  (1899);  Rom. 
xxviii,  166;  Anglia  Bbl.  x.  38. 

(3)  "  The  Source  of  the  Story  of  Sapientes."  MLN.  xxm,  202. 
Chauvin,    V.      Bibliographie    des    Ouvrages    Arabes,   vol.    vm,    1-2 13. 

Leipzig,  Liege,  1904. 
Evans,    J.    G.    "  Report    on    Manuscripts    in    the    Welsh    Language." 

Historical  MSS.  Commission,  pp.  3-4,  33-34,  101   (1902). 
Fischer,   H.     Beitrage   z.   Litter atur   der   Sieben    Weisen   Meister.    Die 

hands chrijtliche  uberliejerung  der  Historia  septem  sapientum.  Diss. 

Greifswald,   1902.  Rev.  Archiv  cxm,  29. 
Hilka,  A.  H.     (1)   "  Historia  septem  sapientum.  1,  Eine  bisher  unbe- 

kannte    lateinische    Ubersetzung    einer    orientalischen    Fassung    der 

Sieben  weisen  Meister    (Mischle   Sendabar)."   Sammlung   Mittella- 

teinischer  Texte,  iv.  Heidelberg,  191 2.  Rev.  Rom.  xlii,  147. 

(2)  11.  Johannis  de  Alta  Silva,  Dolopathos,  nach  den  festland. 
Handschriften.  Sammlung,  v  (191 3). 

(3)  Die  Wanderung  der  Erzdhlung  von  der  Inclusa  aus  dem  Volks- 
buch  der  Sieben  Weisen  Meister.  Breslau,  191 7. 

Loth,  J.     "La  Version  galloise  des  Sept  Sages  de  Rome  et  le  Mabinogi 

de  Kulhwch  et  Olwen."    Revue  Celt,  xxiii,  349-52  (1902). 
Mackinnon,  D.     Catalogue  of  Gaelic  Manuscripts  in  Scotland,  p.   152. 

Edin.,  1912. 
Macler,  F.     (1)    Contes  Syriaques:  Histoire  de  Sindban.  Paris,   1903. 

(2)  La  Version  Armenienne  de  V Histoire  des  Sept  Sages  de  Rome. 

Mise  en  franchise.  Introduction  par  V.  Chauvin.   Paris,   1919. 
Massey,  Isabella.     Text-u.  Quellenstudien  zu  dem  anonymen  mitteldeut. 

Gedicht  von  den  Sieben  Weisen  Meistern.  Diss.  Marburg,  1913. 
Mikolajczak.     De  Septem  Sapientum  fabulis  quaestiones  selectae.  Diss. 

Breslau,  1902. 
Napier,  A.  S.     "  A  Hitherto  Unnoticed  Middle  Eng.  Manuscript  of  the 

Seven  Sages  (Rawl.  Poet.   175)."  PMLA.  xrv,  459-64   (1899). 
Paris,  Gaston.    Deux  Redactions  des  Sept  Sages  de  Rome.  SATF.  Paris, 

1876. 


THE    SEVEN    SAGES    OF   ROME  1 83 

Plomp.  H.  P.     De  middelnederlandsche  B  ewer  king  van  het  gedicht  van 

den  vii  Vroeden  van  binnen  Rome.  Utrecht,  1899.  Rev.  Rom.  xxviii, 

449. 
Schmitz,  J.     Die  dltesten  Fassungen  des  deut.  Romans  von  den  Sieben 

Weisen  Meistem.  Diss.  Greifswald,  1904. 
Smith,  H.  A.     "  A  Verse  Version  of  the  Sept  Sages  de  Rome."    Rom. 

Rev.  in,  1-67   (1912). 
Turkish  version,   English  translation.   The  History   of   the  Seven  Wise 

Masters  of  Rome.  Kasan,  1900. 
Ulrich,  J.     Proben  der  lateinischen  novellistik  des  Mittelalters  ausge- 

wahlt  u.  mit  anmerkungen  versehen.  Leipzig,   1906. 
Voretzsch,   C.    Einfuhrung  in  das  Studium  d.  altfrz.  Lit.    (1913),  pp. 

414-17;  p.  457. 


FLORIS   AND    BLAUNCHEFLUR 

Versions.  The  gracious  tale  of  Floris  and  Blauncheflur 
was  once  widely  known  and  loved  in  mediaeval  Europe.1  The 
oldest  version  is  a  French  poem  composed  between  1160-70, 
now  found  in  three  Paris  manuscripts  (A,  B,  C).  Of  these  the 
oldest  (A,  Bibl.  Nat.  375),  containing  3342  verses,  was  written 
at  the  end  of  the  thirteenth  century.  Du  Meril  (Floire  et 
Blancheflor,  Paris,  1856)  called  this  the  "  version  aristocratique  " 
(A),  and  a  second  version  (B),  also  composed  in  the  late  twelfth 
century,  but  now  found  only  in  one  manuscript  of  the  first  half 
of  the  fourteenth,  the  "  version  populaire  "  (ed.  Du  Meril,  p. 
125  ff.).  Fundamentally  the  story  of  the  two  versions  is  the 
same,  a  tale  of  two  children  who  become  passionately  attached 
to  each  other,  of  their  separation  lest  the  royal  lad  wed  the  girl 
of  inferior  station,  of  his  search  for  her  and  their  ultimate  re- 
union in  a  far-off  land.  But  the  two  versions  differ  in  style  and 
individual  incidents.2  The  first  is  an  idyllic  romance  with  the 
emphasis  on  sentiment  and  aesthetic  detail ;  the  second,  full  of 
action  and  stirring  incident,  is  somewhat  obviously  modelled  on 
the  contemporary  chansons  de  geste? 

The  nature  of  the  origin  of  these  versions  and  of  their  rela- 

1  Reinhold,  p.  9,  listed  sixteen  allusions  to  the  story  by  Provencal  poets 
but  there  is  no  proof  that  there  was  ever  a  Provencal  version  of  the  story. 
The  best  comparative  studies  of  the  various  versions  of  Floris  are:  H. 
Herzog,  Die  beiden  Sagenkreise  von  Flore  u.  Blanscheflur,  Germania,  1884; 
H.  Sundmacher,  Die  altfrz.  u.  mhd.  Bearbeitung  der  Sage,  Gottingen,  1872 ; 
and  those  by  Reinhold  and  Ernst. 

2  The  episode  of  Floire's  attempted  suicide  in  the  lion  pit  is  usually 
cited  as  the  distinctive  episode  of  the  extant  first  version.  Distinctive  in  the 
second  are  the  false  accusation  brought  against  Blanchefleur  that  she  had 
attempted  to  poison  the  king;  the  judicial  combat  which  Floire  fought  for 
her  as  her  champion,  an  incident  due  perhaps  to  the  influence  of  stories  of 
the  type  of  the  Erie  of  Tolous;  and  the  warfare  in  which  Floire  becomes  in- 
volved in  the  East.    Cf.  Reinhold,  ch.  ni. 

3  Cf.  Gaston  Paris,  Rom.  xxvm,  443.  It  is  of  interest,  in  connection 
with  the  Carolingian  cycle,  to  note  that  Floire  and  his  love  are  said  to  be 
the  parents  of  Berte,  the  mother  of  Charlemagne. 

184 


FLORIS   AND    BLAUNCHEFLUR  185 

tion  to  each  other,  has  long  been  a  matter  of  heated  controversy.4 
Were  they  independent  derivatives  of  a  common  source?  Was 
the  second  version  derived  from  some  more  primitive  version  of 
the  first  or  from  the  extant  version  in  combination  with  another 
group?  Was  the  second  version,  as  Du  Meril  thought,  a  re- 
working and  transformation  of  the  first  in  order  to  adapt  it 
to  an  audience  of  less  cultivated  and  courtly  taste?  In  1906 
Reinhold  sharply  attacked  this  traditional  classification  of  the 
French  versions  and  maintained  (p.  82,  ff.)  that  the  popular 
version  was  not  less  chivalric  than  the  first,  and  that  variations 
in  it  were  simply  because  of  a  redactor's  imperfect  memory  of 
the  first  version.5  From  a  prayer  for  "  bons  ostax  "  in  the  B 
version,  he  thought  (p.  115)  that  this  one  may  have  been  espe- 
cially destined  for  the  entertainment  of  pilgrims  to  Compos- 
tella.  In  both  versions  Blancheflor's  mother  is  captured  by 
Saracens  from  pilgrims  similarly  bound,  and  it  may  have  been 
the  potential  interest  of  this  episode  which  inspired  some 
enterprising  jongleur  to  recount  the  story.  Interesting  as  is  this 
hypothesis  in  connection  with  Bedier's  theory  for  the  develop- 
ment of  the  chansons  de  geste,  the  second  French  version  is  in 
itself  generally  felt  to  be  of  inferior  literary  quality. 

The  antiquity  of  the  French  versions  is  indicated  not  only 
by  internal  evidence  but  by  the  fact  that  the  story  had  passed 
before  the  end  of  the  twelfth  century  from  France  into  Germany. 
About  1 1 70  was  composed  a  Low  Rhenish  poem  (ed.  Steinmeyer, 
Zts.  f.  deut.  Alterturn,  1877,  xxi,  307  ff.),  of  which  only  368 
verses  remain.  It  was  based  undoubtedly  on  French  sources. 
Early  in  the  thirteenth  century  the  German  poet,  Konrad  Fleck, 
freely  paraphrased  the  tale  in  8006  lines  of  Middle  High  Ger- 
man verse  (ed.  W.  Golther,  Kiirschners  Deut.  National-Lit.  iv, 
Abt.  3,  Berlin,  1889)  in  a  version  known  through  two  fifteenth- 
century  manuscripts,  through  a  fifteenth-century  prose  version 
(ed.  Herzog,  Gerniania,  xxix,  218-26),  and  through  two  thir- 
teenth-century fragments  discovered,  one  in  1898  (Lambel, 
Festschrift  z.  vm  allg.  Neuphil.-tage  in  Wien,  pp.  37-58)  and 
another  by  Zwierzina  in  1904   (cf.  Rischen,  1913).     From  the 

4  Cf.  Reinhold,  p.  82  ff.,  for  a  history  of  the  question. 

6  In  the  opinion  of  Gaston  Paris,  op.  cit.  p.  444,  Version  n  was  a  hybrid 
made  from  memory  by  a  poet  familiar  with  Version  1  and  with  the  source 
of  the  Spanish-Italian  group.    Cf.  Herzog,  op.  cit.  pp.  4-1 1. 


1 86  MEDLEVAL   ROMANCE   IN    ENGLAND 

same  source  that  Fleck  used,  another  poet,  Diederic  von 
Assenede,  about  1250  composed  a  simpler  version  of  3983  verses 
in  Middle  Low  German  (ed.  Moltzer,  Groningen,  1879;  Leen- 
dertz,  1912).  A  Low  German  poem,  Flos  unde  Blankeflos  extant 
in  five  manuscripts  of  the  fourteenth  and  sixteenth  centuries,  was 
thought  by  its  most  recent  editor,  Decker  (1913),  to  represent  a 
unique  and  primitive  form  of  the  Ur-floire.  It  refers  explicitly 
to  a  "  fransoische  bokelin  "  as  its  source.  On  linguistic  grounds 
Decker  (p.  131)  dated  the  poem  in  the  last  decade  of  the  thir- 
teenth century.  Ernst  (pp.  50-53)  contended  that  a  portion 
of  this  version  was  translated  from  a  Ripuarian  poem,  a  frag- 
ment of  184  lines  (ed.  Schafstadt,  1906),  with  which  it  has  some 
verbal  similarities.  Finally,  among  the  German  versions  may 
be  noted  two  groups  of  Volksbucher,  one  an  adaptation  of 
Fleck's  poem,  and  one  from  Boccaccio's  Italian  rendering  of 
the  story  of  Floris.  With  the  exception  of  this  last  group,  all 
these  German  texts  presuppose  a  French  source  antedating  the 
extant  A  version.  But  it  is  evident  from  the  problems  brought 
up  by  new  texts  and  new  studies  that  the  last  word  has  not 
been  said  on  their  history  or  on  their  special  relationships  with 
the  French  and  English  versions. 

The  story  of  Floris  and  Blauncheflur  appeared  in  England 
about  the  middle  of  the  thirteenth  century,  in  the  last  half  of 
which  were  written  the  oldest  extant  manuscripts,  Cott.  Vitell. 
and  C  (McKnight,  p.  xli).  These  texts  are  unfortunately  very 
fragmentary,  but  their  gaps  may  be  supplemented  by  two  other 
manuscripts  (A  and  T),  which  are  respectively  of  the  fourteenth 
and  the  fifteenth  century.  After  collating  these  texts  for  his 
edition  of  the  poem  (1885)  Hausknecht  concluded  that  C  rep- 
resents one  group,  and  the  other  three  manuscripts  a  group  X 
in  which  A  and  Cott.  Vitell.  form  a  sub-group.  The  manuscript 
C  agrees  more  frequently  with  the  French  source  than  do  the 
others.  In  Hausknecht 's  text  the  poem  contains  1296  verses 
written  in  short  couplets  and  in  the  dialect  of  the  East  Midland 
district  (McKnight,  p.  xli). 

The  omission  in  the  English  romance  of  any  of  the  traits 
peculiar  to  the  second  French  version  indicates  that  its  source 
is  to  be  sought  in  the  oldest  French  form.  Of  the  three  texts 
(A,  B,  C)  belonging  to  that  version,  the  English  poem  shows 


FLORIS   AND    BLAUNCHEFLUR  jgy 

the  closest  agreement  with  B.  But  as  Hausknecht  (pp.  137- 
140  ff.)  pointed  out,  its  agreement  in  three  passages  not  con- 
tained in  either  manuscript  A  or  B,  but  found  in  the  German 
texts  of  Fleck  and  Diederic,  suggests  that  it  was  derived  from 
their  common  French  source.6 

Whatever  text  was  its  direct  antecedent,  the  English  version, 
which  closely  follows  the  order  of  events  and  in  some  places 
the  very  phraseology  of  the  extant  French  texts,  is  nevertheless 
a  free  and  characteristic  rendering.  McKnight  (p.  xxxix)  7  gave 
an  interesting  list  of  the  passages  omitted,  modified,  or  con- 
densed by  a  poet  who  had  in  view  a  practical-minded  audience 
and  who  evidently  felt  that  for  Englishmen  the  tender  senti- 
mentality of  the  French  text,  its  ornate  descriptions  of  gems 
and  precious  stuffs,  of  the  wonderful  cup,  to  which  the  French 
poet  devoted  sixty-seven  lines  and  the  Englishman  but  seventeen, 
or  of  the  marvellously  lovely  garden  of  the  Emir,  would  be 
somewhat  over  elaborate.  But  even  so  he  keeps  much  of  the 
charming  idyllic  quality  of  the  original.  The  atmosphere  of 
love  and  youth  and  innocence  is,  if  anything,  enhanced  by  the 
English  poet's  simplicity  of  style.  His  Floris  is  as  boyishly 
inexpressive  in  his  grief  at  the  (false)  news  of  Blauncheflur's 
death  as  is  Chaucer's  young  Dreamer ;  8  his  little  Blauncheflur 
sums  up  all  her  resolution  not  to  wed  the  Emir  by  vowing,  in 
familiar  English  phrase,  not  "  to  change  old  love  for  new." 
Throughout  the  poem,  in  fact,  the  speech  of  the  characters  is 
singularly  fresh  and  natural,  and  the  descriptive  passages,  though 
so  much  condensed,  keep  enough  gaily  colored  detail  to  make 
them  vivid.  One  of  the  earliest  of  the  extant  Middle  English 
romances,  Floris  and  Blauncheflur,  is  also  one  of  the  best.  The 
continued  popularity  of  the  legend  in  England  is  suggested  by 
the  allusion  to  it  in  the  fourteenth-century  romance,  Emare.  In 
this  a  beautiful  robe  is  described  which  has  been  embroidered 
with  the  stories  of  famous  lovers,  Amadas  and  Idoine,  Tristan 

6  Cf.  Hausknecht,  p.  134  ff.;  Reinhold,  p.  32;  Ernst,  p.  62  ff.,  for  vary- 
ing views  concerning  the  source   of  the  English  poem. 

7  McKnight  followed  Hausknecht's  list,  p.  143  ff.;  Reinhold,  p.  33> 
criticized  severely  the  lack  (to  him)  of  imaginative  and  poetic  feeling  in 
the   English  version. 

8  Cf.  v.  239  (Trentham  MS.)  with  the  Book  of  the  Duchesse,  v.  1309. 
In  Floris  himself  no  one  can  suspect  that  "  artless  artfulness "  which  Kit- 
tredge   (Chaucer's  Poetry,  1915,  p.  53-54),  attributed  to  Chaucer's  Dreamer. 


1 88  MEDLEVAL   ROMANCE    IN   ENGLAND 

and  Isolt,  and,  as  no  less  famous,  with  that  of  Floris  and 
Blauncheflur. 

The  remaining  groups  of  Continental  versions  of  the  story 
belong  to  Italy,  to  the  Scandinavian  countries,  and  to  Spain. 
In  Italy  it  developed  in  one  version  known  as  the  Cant  are  di 
Fiorio  e  Biancifiore  (Crescini,  Bologna,  1899;  Crocioni,  1903) 
written  in  ottava  rima  by  a  popular  poet,  probably  early  in  the 
fourteenth  century.  (Crescini,  11,  25-6.)  From  a  version  de- 
rived from  the  same  source  as  the  Cant  are  Boccaccio  fashioned 
between  1338-41  the  Filocolo?  his  first  prose  romance.  It  is 
generally  supposed  that  the  lost  common  source  was  a  Franco- 
Italian  poem  (Crescini,  1,  163;  11,  10,  239),  probably  nearer  in 
some  instances  to  the  original  French  version,  and  certainly 
more  complete  than  the  abridged  Cantare  version.10 

With  the  Italian  versions  belongs  the  Spanish  prose  romance, 
Flores  y  Blancaflor,  printed  at  Alcala  in  15 12,  and  translated 
into  French  in  1554  (Ernst,  p.  4).  The  Spanish  text  was  orig- 
inally of  much  greater  antiquity,  though  the  name  Flores,  pre- 
serving the  s  of  the  French  nominative,  indicates,  as  Gaston 
Paris  pointed  out  (Rom.  xxvm,  446),  a  date  subsequent  to  that 
of  the  French  versions.  Paris's  summary  n  of  the  traits  which 
distinguish  the  Spanish  romance,  the  Cantare,  and  the  Filocolo, 
from  either  of  the  two  French  versions,  argued  that  a  lost  French 
text,  as  distinctive  in  character  as  the  two  that  are  extant,  was 

9  Ed.  by  E.  De  Ferri,  Turin,  1921-22.  Cf.  Hausknecht,  p.  37;  H.  Cum- 
mings,  Indebtedness  of  Chaucer's  Works  to  the  Italian  Works  0}  Boccaccio, 
Univ.  of  Cincinnati  Studies,  x,  1916,  ch.  1. 

10  Cf.  Paris,  Rom.  xxvm,  439.  A  fuller  but  slightly  different  version 
found  in  the  fourteenth-century  Greek  romance,  Flores  and  Platziaflore 
(ed.  W.  Wagner,  Mediaeval  Greek  Texts,  Lond.  1870).  Herzog  believed  the 
first  French  version  and  the  Italian- Spanish  groups  to  be  variants  of  a  more 
ancient  Greek  romance  which  had,  before  the  fourteenth  century,  experienced 
two  redactions.  Crescini,  11,  4,  noted  that  the  name  Blancheflour,  common 
to  the  two  twelfth -century  French  versions,  practically  makes  this  hypothesis 
impossible. 

11  Rom.  xxvm,  441.  In  the  Spanish-Italian  group  Blancheflur's  parents 
are  of  Roman  nationality;  her  mother  (in  the  French  versions  an  unnamed 
French  lady)  is  called  Topazia;  widowed  and  captured  at  the  same  time, 
she  later  dies  in  giving  birth  to  Blancherlur;  the  maiden  gives  Floire  a  ring 
which  will  indicate  to  him  when  she  is  in  danger;  the  master,  not  the  father, 
of  Floire  discovers  the  boy's  love  for  the  maiden;  she  is  held  captive  in 
Babylon  in  Egypt  (Cf.  p.  444,  n.  1) ;  she  has  for  servant  a  girl  named  Gloris 
(in  the  French  versions  her  friend  and  equal  is  called  Claris) ;  she  and 
Floire  at  the  end  of  the  story  receive  the  imperial  power  of  Rome. 


FLORIS   AND    BLAUNCHEFLUR  189 

once  known;   but  again  the  complex  question  of  their  inter- 
relation is  open  to  doubt. 

From  Scandinavian  countries  come  the  following  versions: 
two  Norse  sagas,  one  written  before  13 19  and  still  preserved  in 
complete  form  (ed.  Kobling,  Halle,  1896),  and  another  of  which 
only  a  fragment  remains  (ed.  G.  Storm,  Copenhagen,  1874) ;  a 
Swedish  poem  (ed.  E.  Olson,  Lund,  192 1)  which  was  translated 
about  13 1 1  from  a  lost  Icelandic  original  and  later  translated 
into  Danish  (ed.  C.  Brandt,  Romantisk  Digtning,  Copenhagen, 
1869-77).  All  of  these  versions  were  derived  from  a  French 
source,  but  their  variety  and  distribution  indicate  the  popularity 
of  the  story  in  the  North. 

Origin.  Before  1897  the  story  of  Floire  et  Blanche flor  was 
variously  supposed  to  have  come  from  Provence,  from  Ger- 
manic myth,  from  Spain,  from  a  lost  romance  of  Byzantine 
origin,  and  from  Persia.12  In  1897-99  Ten  Brink  and  Huet 13 
independently  advanced  the  belief  that  it  was  originally  an 
Arabic  tale.  They  derived  the  Tower  of  Maidens,  in  which 
Floire  discovered  his  lost  sweetheart,  from  an  eastern  harem, 
and  found  a  striking  parallel  in  at  least  four  Arabic  tales  for 
the  episode  in  which  Floire,  getting  himself  concealed  in  a  basket 
of  flowers,  gains  admittance  to  the  Tower.  The  incident  is  prac- 
tically the  same  in  the  Arabic  tales,  though  certain  details  vary 
widely.  The  girl  is  a  harem  slave,  the  youth  a  merchant  or 
money  changer;  in  all  the  stories  but  one  he  enters  the  harem 
disguised  as  a  woman  or  as  the  Khalif  himself.  There  are  be- 
sides such  correspondences  as  these:  in  one  story  the  ruse  epi- 
sode is  preceded  by  an  account  of  the  lovers'  childhood  and  of 
their  youthful  passion  for  each  other ;  and  in  two  stories,  as  also 
in  Floire,  the  hero,  when  he  arrives  within  the  harem,  goes  or  is 
taken  to  the  wrong  door.  In  the  oldest  French  version  of  Floire 
and  in  the  derivative  versions,  the  tower  with  its  many  rooms, 
its  armed  guardians,  its  lovely  inmates,  their  services  to  the 
Emir,  their  equality  with  each  other,  and  the  geographical  allu- 
sions (in  the  French  versions  to  Babylon  of  Asia),  seem  to  point 

12  The  most  recent  summary  of  these  opinions  is  given  by  Johnston, 
2,   pp.   126-29. 

13  See  respectively,  Geschiedenes  der  Nederlandsche  letter  kunde,  p.  115 
ff.,  Amsterdam,  1897,  and  Huet,  Rom.  xxvm,  349-59   (1899). 


1 9o  MEDIEVAL   ROMANCE    IN   ENGLAND 

to  the  customs,  the  locality,  and  the  story  motifs  of  the  Arabic 
East.  Especially  notable  from  this  point  of  view  is  the  fact  that 
Floire,  in  starting  on  his  long  quest  to  regain  Blancheflor,  as- 
sumes a  merchant  disguise,  a  kind  rare  indeed  in  the  chivalric 
romances  of  the  west,  but  used  innumerable  times  in  eastern 
story  (Huet,  Rom.  xxvin,  355;  xxxvi,  95). 

The  Arabic  hypothesis  for  the  origin  of  Floire  is  engaging 
but  not  wholly  satisfactory.  In  1906  Reinhold  pointed  out  that 
no  single  Arabic  tale  is  of  the  same  structural  type  as  Floire, 
although  single  incidents  may  be  paralleled  here  and  there;  he 
urged  strongly  the  improbability  that  the  tales,  especially  those 
of  the  Arabian  Nights  collection,  were  known  in  twelfth-century 
France ;  14  and  finally  he  asserted  that  the  idea  of  the  Tower  of 
Maidens  as  a  supposedly  Eastern  harem  is  as  anomalous  as  that 
of  the  monagamous  but  murderous  Emir  who  in  the  romance  is 
said  to  kill  his  wife  each  year  before  taking  a  new  bride.  Though 
the  modification  of  a  harem  into  a  Maidens'  Tower  would  not 
seem  a  very  startling  modification  for  a  western  poet  to  make, 
Reinhold  preferred  to  think  the  idea  came  from  the  Book  of 
Esther  in  which  the  House  of  Maidens  is  clearly  distinguished 
from  that  of  the  king's  concubines.  He  found  still  other  par- 
allels between  the  Book  of  Esther  and  the  romance  in  the  ac- 
counts of  the  feasts  of  Ahasuerus  and  the  Emir,  which  were 
broken  up,  in  one  case  by  Vashti's  disobedience,  in  the  other  by 
the  discovery  of  Blancheflor's  lover  in  the  Tower,  and  in  the 
description  of  the  assembling  of  a  court  of  judges.  Though 
Reinhold's  destructive  criticism  has  much  to  commend  it,  there 
is  nothing  in  this  suggested  derivation  of  Floire  from  Esther 
that  brings  conviction.  Neither  the  characters  nor  the  incidents 
of  the  two  stories  are  in  any  genuine  sense  alike. 

To  explain  the  actual  story  of  the  lovers,  Reinhold  turned  to 
that  of  Cupid  and  Psyche  as  told  by  Apuleius  in  the  Golden 
Ass.15  In  this  as  in  Floire  the  love  story  includes  the  mesalli- 
ance motif,  the  separation  of  the  lovers  through  parental  influ- 
ence, the  quest  of  one  lover  for  the  other,  the  entrance  of  the 

14  Huet,  Rom.  xxxv,  97,  assembled  the  evidence  showing  that  the  col- 
lection of  the  Thousand  and  One  Nights  was  certainly  anterior  to  the 
fifteenth  century.  The  composition  of  the  oldest  portion  has  been  assigned 
to   the   tenth    century. 

15  See  Partonope  of  Blois,  note  13. 


FLORIS   AND    BLAUNCHEFLUR 


191 


seeker  into  a  dangerous  place  (Psyche  into  the  house  of  the  irate 
goddess  Venus,  her  greatest  foe,  Floris  into  the  Emir's  Tower), 
and  the  winning  of  that  foe's  forgiveness  through  the  lover's 
great  devotion.  But  the  comparison  offers  at  best  a  far-fetched 
likeness  which  ignores  the  fundamental  marchen  elements  in 
Psyche's  story.  Her  marriage  with  the  god,  her  breaking  of  the 
prohibition  which  he  laid  upon  her,  her  painful  quest  and 
achievement  of  impossible  tasks,  have  no  parallel  in  the  romantic 
and  thoroughly  mediaeval  adventures  of  the  opulent  young 
Floris. 

Far  nearer  to  Floris  in  chivalric  and  aesthetic  interest  and  in 
actual  detail  is  Aucassin  and  Nicolette,16  the  unique  and  lovely 
chantejable  of  thirteenth-century  France.  Each  romance  tells 
of  the  passionate  devotion  to  each  other  of  a  high-born  lad  and 
a  girl  in  lowly  case,  of  the  cruel  lordly  father  who  separates 
them,  and  of  the  youthful  lover's  romantic  quest.  These  corre- 
spondences convinced  Brunner  (Ueber  Aucassin  and  Nicolette, 
Halle,  1880)  that  Floire  was  the  source  of  Aucassin,  and  made 
Johnston  (1911,  2,  p.  129)  suspect  the  opposite.  He  noted  that 
in  several  instances  in  the  initial  pilgrimage  scene,  in  the  attack 
of  the  Saracens,  the  account  of  the  birth  of  the  two  children 
and  the  linking  of  their  destinies,  like  those  of  Amis  and  Amile, 
by  their  simultaneous  birth  and  by  their  similar  names,17  the 
author  seems  to  have  elaborated  mere  suggestions  for  these 
things  in  Aucassin.  In  the  main,  however,  scholars  believe  in 
a  common  source  for  the  two  poems.18  It  seems  probable  that 
this  source  was  influenced  by  some  twelfth  century  version  of 
Ovid's  tale  of  Pyramus  and  Thisbe.     In  one  extant  text 19  the 

16  Cf.  the  seventh  edition  by  H.  Suchier,  Paderborn,  1909,  p.  vii,  for 
bibliography  to  1909;  also  ninth  edition,  1921.  In  regard  to  the  date  of 
Aucassin  Madame  Lot-Borodine  pointed  out  (p.  86)  that  as  Nicolette's  dis- 
guise as  a  jongleur  is  borrowed  from  that  of  Josian  in  Beuve  de  Hanstone, 
Aucassin  must  be  of  later  date  than  Beuve.  As  regards  its  source,  Bullock, 
in  an  able  review  (MLN.  xxxvi,  497)  of  the  1921  edition  of  Aucassin, 
showed  that  "  the  whole  question  of  Arabic  influence  on  the  story  is  still 
open." 

17  Cf.  A.  Krappe,  "The  Legend  of  Amicus  and  Amelius,"  MLR.  xvm, 
152  (1923)  ;  J.  Harris,  Cult  of  the  Heavenly  Twins,  Cambridge,  1906,  p.  58, 
on  the  use  of  similar  names,  especially  for  twin  children. 

18  Cf.  Lot-Borodine,  p.  79;  Johnston    (2),  p.  129. 

19  Pyrame  et  Thisbe,  poeme  du  XIIe  siecle,  ed.  C.  de  Boer,  Paris,  1921. 
Cf.  E.  Faral,  Rom.  xli,  32-57  (1912),  and  Sources  Latines  des  Contes  et 
Romans  Courtois  du  Moyen  Age,  Paris,   1913,  pp.   5-63.     Faral   (Sources,  p. 


I92  MEDIEVAL  ROMANCE    IN   ENGLAND 

author  clearly  develops  the  idyllic  note,  making  much  of  the 
childhood  devotion  of  the  pair,  and  so  turns  the  classic  tale  into 
a  little  tragic  idyll,  the  prototype  of  the  happy  idylls  in  Floire 
and  in  Aucassin  (Lot-Borodine,  pp.  93-98). 

Notable  in  the  Floire  legend  are  three  descriptive  passages. 
One  concerns  the  beautiful  tomb  shown  to  the  hero  as  the  grave 
of  his  sweetheart.  In  the  French  A  text  (Du  Meril,  p.  24)  this 
is  described  as  shining  like  the  sun,  so  rich  is  it  in  precious 
stones.  On  it  are  carved  the  figures  of  two  beautiful  children 
holding  flowers  in  their  hands  and  moving  when  the  wind  blows 
so  that  they  embrace  each  other  and  speak  "  par  nigremance."  20 
This  passage  Reinhold  derived  from  the  account  in  the  ancient 
romance  of  Apollonius  of  the  pretended  tomb  of  Tarsia  which 
was  shown  to  her  grieving  father.  It  has  also  been  compared 
with  the  tomb  of  Didon  and  Camille  in  the  Eneas.21    Another 


32)  thought  that  Floire  was  influenced  by  Pyramus  in  the  episode  of  the 
hero's  attempted  suicide  and  in  his  apostrophe  to  his  "  grafe  "  (v.  785-98) . 
For  later  vernacular  texts  of  Pyramus1,  see  Faral,  pp.  35-36. 

It  is  perhaps  of  some  interest  to  note  that  the  Pyramus  story  was  retold 
in  English  not  only  by  Chaucer  and  Gower  but  also  by  John  Metham,  a 
self-styled  scholar  of  Cambridge.  He  was  writing  about  1448-49  under  the 
patronage  of  Sir  Miles  and  Lady  Stapleton.  His  romance,  Amoryus  and 
Cleopes,  in  seven-line  stanzas,  is  recognizable  as  a  version  of  the  Pyramus 
story,  though  in  it  the  tragic  is  changed  to  a  happy  ending,  for  a  hermit 
miraculously  restores  the  two  dead  lovers.  The  tale  is  absurdly  amplified 
by  the  introduction  of  much  pseudo-learning.  The  author  was  interested  in 
"  palmestrye,"  physiognomy,  fortunate  days,  astrology,  etc.  Only  one  manu- 
script of  this  romance  is  known.  It  is  summarized  by  Furnivall,  Political, 
Religious,  and  Love  Poems,  EETS.  xv,  301-08  (1866).  For  other  versions 
see  Fliigel,  "  Pyramys  and  Tysbe,"  Anglia  xn,  13-20,  631    (1889). 

20  Cf.  J.  D.  Bruce',  "  Human  Automata  in  Classical  Tradition  and  Medi- 
aeval Romance,"  Mod.  Phil,  x,  511-26  (1913).  Bruce  did  not  discuss  these 
figures  on  the  tomb  but  referred  (p.  518)  to  a  golden  image  of  a  harper 
which  appears  to  Floire  through  the  entertaining  magic  of  an  enchanter, 
"  Et  harpe  le  lai  d'Orphey."  Other  automata  noted  by  Bruce  are  in  the 
Pelerinage  de  Charlemagne,  the  Eneas  (an  archer  on  the  tomb  of  Camilla, 
v.  7691),  Tristan.  Alexandre,  Huon  de  Bordeaux,  Diu  Crone,  Cleomades,  and 
various  versions  of  the  Grail  romances.  Many  of  these  automata  resemble 
those  ascribed  to  the  magic  art  of  Virgil.  Cf.  Comparetti,  Vergilio  nel  medio 
evo,  2nd  ed.  Florence,  1896;  Faral,  Sources  latines,  pp.  328-35,  les  auto- 
mates; arbres;  oiseaux;  personnages  humains.  This  section  of  Faral's  book 
is  part  of  an  interesting  study  on  "  Le  Merveilleux  et  ses  Sources  dans  les 
Descriptions  Romans  Frc.  du  XIIe  Siecle,"  pp.  307-78.  On  this  general 
subject  see  also  Easter,  A  Study  of  the  Magic  Elements  in  the  Romans 
d'Aventure,  Diss.,  Baltimore,  1906.     Cf.  £even  Sages,  note   12. 

21  Cf.  A.  Dressier,  Der  Einfiuss  des  altfrz.    Eneas-romanes  auf  die  altfrz. 


FLORIS   AND    BLAUNCHEFLlTR  193 

passage  (Du  Meril,  p.  19),  describing  a  cup  on  which  was 
wrought  the  judgment  of  Paris  and  the  beauty  of  Helen  of 
Troy,  shows  that  interest  in  classical  story  characteristic  of  the 
oldest  French  version  but  almost  lost,  unfortunately,  from  the 
English  texts.  The  third  passage  describes  the  garden  about  the 
Tower  of  Maidens  (Du  Meril,  p.  65  ff.).  The  garden  has 
streams  of  water  from  Paradise,  gravel  formed  of  jewels,  a  Tree 
of  Love,  a  Well  of  Chastity.  Spring  there  is  eternal.  Johnston 
suggested  the  possible  influence  on  this,  as  on  other  descriptions 
of  twelfth  century  French  romance,  of  conventionalized  descrip- 
tions of  the  Celtic  Otherworld.  But  again  such  indications  as 
we  have  point  to  the  East  (Patch,  PMLA.  191 8,  xxxiii,  623- 
4).  The  marvels  of  the  East  were  becoming  known  through  the 
Alexander  romances  and  through  Crusaders'  and  travelers'  tales 
of  the  sort  that  ultimately  gave  rise  to  such  accounts  as  those 
of  Friar  Oderic  or  Mandeville  concerning  the  unearthly  mag- 
nificence of  Prester  John,  of  the  Great  Khan,  and  the  Old  Man 
of  the  Mountain  (Rickert,  p.  xlii).  Like  the  Emir,  they  have 
marvellous  palaces  and  gardens,  light-giving  jewels,  a  Well  of 
Youth,  and  companies  of  young  maidens  for  their  service. 
Readily  indeed  would  such  picturesque  details  blend  with  a 
story  to  which  the  earliest  European  redactors  obviously  wished 
to  give  the  charm  and  mystery  of  the  far-off  East,  however 
much  they  imbued  it  with  the  sentimentality  and  chivalry  of  the 
West. 

BIBLIOGRAPHY 

Texts:  (i)  Cott.,  Cotton  Vitellius  D  m,  ed.  J.  R.  Lumby,  EETS. 
xiv,  1866;  reed,  and  all  texts  printed,  G.  McKnight,  EETS.  1901;  (2) 
A,  Auchinleck  MS.,  ed.  C.  H.  Hartshorne,  Ancient  Metrical  Tales,  Lond., 
1829;  D.  Laing,  A  Penni  Worth  of  Witte,  Abbotsford  Club,  Edin.,  1857; 
(3)  C,  Cambridge  Univ.  MS.  Gg.  4,  27,  2;  (4)  T,  Trentham  MS. 
belonging  to  the  Duke  of  Sutherland,  Trentham  Hall,  Staffordshire,  now 
MS.  Egerton  2862,  desc.  Brit.  Mus.  Cat.  of  Add.  MSS.  1905-10,  pp. 
238-89.  A  critical  edition  of  the  MSS.  was  made  by  Hausknecht, 
Sammlung  engl.  Denkmaler,  Berlin,  1885.  Rev.  Eng.  Stud,  ix,  92. 
Trans.  E.  Rickert,  Romances  of  Love;  F.  G.  Darton,  A  Wonder  Book 
of  Romance. 


Lit.,  Leipzig,   1907,   pp.   135-39.     The   Middle   English   texts   preserve   but  a 
mere  mention  of  the  tomb. 


194 


MEDIAEVAL   ROMANCE    IN    ENGLAND 


Studies   and   Analogues:    Cf.    Edwardes,   Summary,   Index;    Wells, 
Manual,  p.   785. 
Basset,  R.     "  Les  Sources  Arabes  de  Floire  et  Blancheflor"  Revue  des 

Traditions  Populaires,  xxn,  24  ff.   (1907). 
Boekenoogen,  G.  J.     De  Historie  van  Floris  ende  Blancefluer  (naar  den 

Amsterdamschen    druk    van    Ot    Barentsz.    Smient    uit    het    1642 

uitgegeven).  Leiden,  1903. 
Bongini,  D.     Noterelle  critiche  sul  Filocolo  di  Boccaccio,  precedute  da 

una    introduzione    storico-bibliografica    sulla    leggenda    di    "  Florio 

e  Bianco fiore."  Aosta,   1907. 
Bonilla  y  San  Martin,  A.     La  Historia  de  los  dos  enamorados  Flores  y 

Blancaflor.  Madrid,   1916. 
Crescini,  V.    La  redazione  velletrana  del  cantare  de  "  Fiorio  e  Bianci- 

fiore,"  11.  Rome,   1905.     See  also  //  cantare  di  "  Fiorio  e  Bianci- 

fiore,"  Bologna,   1889-99.  Rev.  Rom.  xxvm,   1899,  pp.   439-47. 
Crocioni,   G.   "  II  Cantare   di  '  Fiorio   e  Biancofiore '  secondo  un   MS. 

velletrano."     Soc.   filol.   romana.   Rome,    1903. 
Decker,    O.    Flos    unde    Blankeflos.    Krit.    Ausg.d.    mittelniederdeut. 

Gedichtes.   166  pp.   Diss.  Rostock,   1913. 
Ernst,  L.     "  Floire  u.  Blantscheflur:   Studie  zur  vergleichenden  Litera- 

turwissenschaft."  Quellen  u.  Forschungen,  cxviii.  Strassburg,  191 2. 
Huet,  G.     "Encore  Floire  et  Blanche fleur "  Rom.  xxxv,  95-100  (1906). 

See  also  Rom.  xxviii,  349-59   (1899). 
Johnston,   O.    M.     (1).    "The    Description   of   the   Emir's    Orchard   in 

Floire  et  Blancheflor,"  Zts.  f.  rom.  Phil,  xxxii,  705-10  (1908). 

(2).  "The  Origin  of  the  Legend  of  Floire  and  Blanchefleur,"  Matzke 

Memorial  Vol.,  pp.  125-38,  Leland  Stanford  Univ.,  California,  191 1. 

(3).  "  Notes  on  Floire  et  Blancheflor,"  Flilgel  Memorial  Vol.,  pp. 

193-99,  Leland  Stanford  Univ.,  California,    1916. 
Leendertz,   P.     "Floris  ende  Blancefloer  van  Diederic  van  Assenede." 

Bibl.  van  middelnederlansche  Letterkunde,  Leiden,  191 2.  Rev.  Rom. 

xlii,   155. 
Lot-Borodine,  M.    Le  Roman  Idyllique  au  Moyen-Age,  Floire,  pp.  9-74. 

Paris,  1913. 
McKnight,  G.  H.     See  Texts.  Introd.  pp.  xxx-xliv. 
Reinhold,   J.     Floire   et   Blancheflor,  Etude   de  Litterature   Comparee. 

Diss.  178  pp.  Paris,  1906.  Rev.  Rom.  xxxvii,  310-12;  Litteraturbl. 

f.  germ.  u.  rom.  Phil,  xxix,  156-57;  Zts.  f.  Rom.  Phil,  xxx,  153. 

Cf.  Ernst  (above),  p.  9,  n.  3. 
Rischen,  C.  H.     Bruchstiicke  von  Konrad  Flecks  "  Floire  u.  Blansche- 

flur."    Germanische  Bibliothek,  111,  4.  Heidelberg,  191 3. 
Schafstadt,  H.     Die  Mulheimer  Bruchstiicke  von  Flors  u.  Blanzeflors. 

Progr.  des  Mulheimer  Gym.,  1906. 
Zwierzina,  K.     "  Frauenfelder  Bruchstiicke  von  Flecks  Floire,"  Zts.  /. 

deut.  Alter,  xlvii,  161-82   (1904). 


SIR    ORFEO 

Versions.  The  earliest  extant  version  in  Middle  English  of 
the  Lay  of  Sir  Orjeo  is  found  in  the  early  fourteenth-century 
Auchinleck  manuscript.  The  poem  contains  602  lines  in  short 
riming  couplets  and  is  to  be  ascribed  to  the  South-Midland  dis- 
trict (Zielke,  p.  55).  The  two  fifteenth-century  manuscripts, 
Harleian  3810  and  Ashmole  61,  seem  to  be  minstrel  variants  of 
a  second  version  (y)  derived  from  the  same  source  as  the 
Auchinleck  text  (Zielke,  p.  25).  The  original  poem  was  prob- 
ably composed  about  the  end  of  the  thirteenth  century. 

Coming  at  a  time  when  "  imitation  and  not  originality  was 
the  rule  in  English  writing,"  the  grace  and  beauty  of  Orjeo  are 
the  more  exceptional.  Brief  yet  vivid,  the  little  tale  is  inimitably 
fresh  in  style  and  content.  So  artless  it  seems,  that  a  ballad- 
like quality  has  been  claimed  for  it  {Cambridge  Hist.,  1,  328). 
Ballad-like  it  is  in  the  simplicity  of  its  theme,  —  a  king's  rescue 
of  his  queen  out  of  fairy  land,  in  the  bright  distinctness  of  its 
few  characters,  Orfeo  himself,  his  queen,  Heurodis,  a  fairy  king, 
a  porter  and  a  faithful  steward,  and  in  an  occasional  humor- 
ously laconic  phrase.  But  the  poem  is  not  without  indications 
of  conscious  artistry.  Such  descriptions  as  that  of  the  hundred- 
towered,  crystal-shining  castle  of  the  fairy  king  (v.  387  ff.),  or 
of  the  fairy  company  riding  on  snow-white  steeds  (v.  109  ff.), 
show  deliberate  pictorial  sense,  and  in  the  passage  (v.  245) 
which  contrasts  Orfeo's  life  in  his  royal  hall  with  his  misery  on 
the  desolate,  freezing  moor,  there  is  conscious  pathos.  If  Orjeo 
is  minstrel  verse,1  it  is  of  very  high  order  and  far  removed  from 
such  a  true  offspring  of  popular  verse  as  the  ballad  into  which 
it  was  ultimately  fashioned. 

1  Ker's  illustrative  reference  (Eng.  Lit.  Mediaeval,  p.  127)  to  the  lines 
in  Orfeo  on  the  wandering  minstrels  who  must  proffer  their  glee,  however 
inhospitable  their  reception,  is  far  from  proving  that  the  author  of  Orfeo 
was  a  minstrel.  Necessarily  the  poem  says  much  of  minstrelsy,  but  so,  also, 
and  in  a  very  different  tone,  does  Sir  Cleges,  a  typical  minstrel  tale.  The 
quality  and  effect  of  Orfeo  is  far  less  popular  than   Cleges. 

195 


I96  MEDIEVAL  ROMANCE   IN   ENGLAND 

The  ballad  of  King  Orfeo  (Child,  Ballads,  No.  19)  was  not 
written  down  until  late  in  the  nineteenth  century,  in  Unst,  Shet- 
land; but  in  its  choral  and  dramatic  form  it  is,  as  Gummere 
(p.  224)  pointed  out,  of  ancient  structural  type,  its  story  evi- 
dently an  oral,  traditional  version  of  the  lay.  The  ballad  con- 
tains seventeen  two-line  stanzas  with  an  unintelligible  refrain 
that  may  originally  have  been  composed  of  Danish  words.  Of 
these  stanzas  two,  twice  repeated,  tell  of  Orfeo's  playing: 

"  And  first  he  played  da  notes  o  noy, 
An  dan  he  played  da  notes  o  joy. 

An  dan  he  played  da  god  gabber  reel, 
Dat  meicht  ha  made  a  sick  hert  hale." 

In  other  words  the  ballad  is  almost  exclusively  interested  in 
Orfeo,  the  music-maker,  and  presents  the  episode  in  which  his 
skill  recovers  for  him  Lady  Isabel,  in  the  most  abbreviated 
narrative  form. 

Origin.  A  few  specifically  English  touches  such  as  that 
which  turns  Thrace  into  Winchester,  "  a  cite  of  noble  defens," 
or  that  which  makes  Orfeo  summon  his  people  to  a  "  parlement  " 
to  appoint  a  new  king  when  he  shall  be  dead,  are  but  slight 
modifications  of  a  story  recognizably  classical  in  origin.  Through 
Ovid  (Metamorphoses,  x),  Virgil  (Georgics,  iv,  454  ff.),  and 
Boethius  (Philosophiae  Consolationis,  in,  metre  xn),  the  fa- 
vorite classic  authors  of  the  Middle  Ages,  the  tragic  Greek 
legend  of  Orpheus  and  Eurydice  was  widely  known.  In  Eng- 
land 2  as  early  as  the  ninth  century  King  Alfred  had  translated 
it  from  Boethius,  and  in  France  we  have  it  alluded  to  in  the 
tenth  century  by  the  monk,  Froudmont  of  Tegernsee  (Zielke, 
p.  130).  By  the  twelfth  century  it  had  been  turned  from  Latin 
into  French  verse  and  there  are  extant  two  Old  French  frag- 
ments of  the  classical  story  (Kittredge,  p.  182,  p.  1).  But  obvi- 
ously no  version  that  kept  to  the  classical  form  could  adequately 

2  For  the  history  of  the  Orpheus  legend  in  English  literature  see  Wirl's 
dissertation.  He  discussed  the  Alfredian  and  Chaucerian  versions  of  Boethius, 
the  Lay  of  Sir  Orfeo,  the  Orpheus  fable  of  Robert  Henryson  (Scottish  Text 
Soc,  vol.  Lvm,  1908),  which  was  entirely  uninfluenced  by  the  Lay,  and 
various  allusions  to  the  legend  by  writers  of  the  fifteenth,  sixteenth,  and 
seventeenth  centuries. 


SIR   ORFEO  I97 

account  for  the  Middle  English  version  nor  for  the  Breton  lay 
from  which  it  claims  descent.3 

The  English  poem  shows  that  the  Greek  legend  has  been  trans- 
formed under  the  influence  of  a  different  racial  culture  and  be- 
lief. The  dim  Hades  of  Greek  myth  has  become  a  glowing  en- 
chanting Otherworld ; 4  the  sad  Greek  gods  of  the  dead  have 
turned  to  beautiful,  passionate,  and  mysterious  fairy  beings. 
Folk  superstition  has  intruded  itself  in  the  scene  in  which  the 
Otherworld  king  gains  power  over  Heurodis,  the  mediaeval 
Eurydice,  not  because  she  dies,  but  because  she  falls  asleep  under 
a  "  fairy  tree."  5  In  his  study  of  Orfeo  Kittredge  indicated  not 
only  the  general  influence  of  what  seems  to  be  primarily  Celtic 
folk-lore,  but  the  specific  modification  besides  of  the  classic 
legend  under  the  probable  influence  of  one  of  the  most  famous 
tales  known  to  Irish  minstrels,  the  Wooing  of  Etain  (Tochmarc 
Etain),6  written  down  in  at  least  one  extant  manuscript  before 

3  Cf.  v.  1-24  and  v.  595.  The  opening  lines,  which  generalize  on  the 
usual  contents  of  Breton  lays,  are  practically  identical  with  those  which 
make  up  the  Prologue  of  the  Lay  le  Fresne.  Foulet,  p.  46  if.,  believed  they 
belonged  to  the  French  original  of  Sir  Orfeo  and  were  borrowed  from  Fresne. 
Miss  Guillaume,  p.  463,  noted  that  as  the  Prologue  occurs  only  in  the  two 
fifteenth-century  texts  of  Orfeo,  as  it  is  inferior  to  the  text  given  in  Fresne, 
and  as  the  text  of  Orfeo  shows  direct  borrowing  from  Fresne,  it  is  more 
reasonable  to  suppose  that  the  Prologue  itself  in  Orfeo  was  borrowed  from 
the  other  poem. 

4  The  pagan  Irish  Otherworld  was  a  fairy  realm  which  lay  beneath  or 
beyond  the  sea,  or  was  hidden  in  a  mound.  For  details  concerning  its 
pleasant  landscape  and  the  Perilous  Passage  which  commonly  led  to  it,  see 
A.  C.  L.  Brown,  "  Iwain,"  Harvard  Studies  in  Phil,  and  Lit.  vm  (1903);  L. 
Paton,  Studies  in  the  Fairy  Mythology  of  Arthurian  Romance,  p.  8s  ff., 
Boston,  1903;  T.  B.  Cross,  "The  Celtic  Origin  of  the  Lay  of  Yonec,"  Revue 
Celt,  xxxi,  461,  n.  3;  Hibbard,  "The  Sword  Bridge  of  Chretien  de  Troyes 
and  Its  Celtic  Original,"  Rom.  Rev.  rv,  178  ff.  (1913).  Even  H.  R.  Patch, 
who  discredited  in  his  study  of  "  Mediaeval  Descriptions  of  the  Otherworld," 
PMLA.  xxxni  (191 8)  the  idea  of  Celtic  influence  on  these  descriptions,  ad- 
mitted (p.  612)  that  the  idea  of  a  fairy  hill  was  peculiarly  Celtic,  and  noted 
that  in  the  Lay,  Orfeo  followed  the  fairy  throng  "  in  at  the  roche  "  as  the 
only  means  of  penetrating  the  fairy  hill.  There  is,  however,  a  genuine 
reminiscence  of  the  classic  legend  in  Orfeo  in  the  description  of  the  fairy- 
land as  a  place  of  the  dead  and  of  the  court  held  by  the  fairy  king. 

5  Kittredge,  p.  189,  thought  this  "  a  Celtic  survival,"  although  he  ad- 
mitted that  the  idea  of  danger  of  sleeping  under  special  trees,  because 
it  exposed  one  to  the  power  of  the  fairies,  was  not  an  exclusively  Celtic  idea. 
For  arguments  against  considering  this  "orchard  scene "  in  any  way  Celtic 
see  Ogle's  arguments  (Sir  Gowther  bibliography). 

6  See  Bibliography  of  Irish  Philology  and  Printed  Literature,  ed.  R.  I. 
Best,  Dublin,  1913;  Schoepperle,  Tristan,  u,  422,  n.  3. 


i98  MEDIEVAL   ROMANCE   IN   ENGLAND 

1106.  In  this,  as  in  Orfeo,  Etain,  the  happy  wife  of  Eochaid, 
high  king  of  Ireland,  is  stolen  away  by  Midir,  a  fairy  king,  to 
whom  in  a  former  life  she  has  been  wedded.7  Like  the  fairy 
king  in  Orfeo,  Midir  sings  to  her  of  his  marvellous  Otherworld 
realm;  like  Heurodis,  Etain,  though  guarded  by  her  mortal 
husband's  warriors,  is  spirited  away  through  the  air,  and  is  re- 
covered at  last  from  a  fairy  hill.  The  recovery  itself,  accom- 
plished in  the  Etain  story  by  a  siege  of  the  fairy  hill,  is  not 
paralleled  in  the  later  romance.  Orfeo  draws  for  this  part  of 
the  story,  it  would  seem,  on  an  earlier  episode  in  the  Irish  legend. 
At  his  first  coming  to  Eochaid's  court,  Midir,  disguised,  lures 
the  king  into  the  rash  promise  to  give  Midir  whatsoever  he 
should  desire  if  Midir  wins  a  game  of  chess.  Eochaid  admits 
the  sanctity  of  the  promise  even  when  Midir  asks  for  Etain. 
So  also  in  the  Lay,  Orfeo,  disguised  as  a  minstrel,  wins  the  rash 
promise  of  high  reward  from  the  fairy  king.  Like  the  mortal 
Eochaid,  the  fairy  king  keeps  his  promise  even  though  Orfeo 
promptly  demands  the  stolen  lady.  "  Stories  of  a  woman  thus 
won  and  lost  by  a  ruse  between  mortals  and  immortals  seem  to 
have  been  a  favorite  type  among  the  Celts "  (Schoepperle, 
Tristan,  11,  428).  The  same  theme  is  found  in  numerous  Celtic 
tales  of  which  the  most  famous,  perhaps,  are  the  Irish  Diarmid 
and  Grainne  story,  the  Welsh  tale  of  the  wedding  of  Pwyll  in 
the  Mabinogion,  and  the  French  and  Norse  versions  of  Tristan. 
In  Orfeo  the  abduction  episode  and  its  happy  sequel,  for  which 
there  is  no  parallel  in  the  classic  legend,  seem  to  represent  char- 
acteristic Celtic  adaptations.8 

The  evidence  of  Celtic  influence  makes  more  credible  the  ref- 
erence in  Orfeo  to  an  original  "  Breton  "  lay  as  its  source,  espe- 
cially as  this  is  supported  by  a  reference  in  the  Lai  de  VEspine 
(ed.  R.  Zenker,  Zts.  f.  rom  Phil,  xvn,  233)  to  a  musical  Lai 

7  There  is  no  suggestion  in  Orfeo  of  any  former  relationship  between 
Heurodis  and  the  fairy  king.  But  this  detail  from  the  Old  Irish  stories 
survives  in  some  versions  of  the  Rape  of  Guinevere  in  connection  with  the 
lover  who  appears  as  Meliagrance  in  Malory,  Morte  Darthur,  xix,  ch.  2. 
In  the  older  versions  of  this  episode  she  is  stolen  by  a  supernaturally  splen- 
did person  who  as  lover  or  husband  has  a  claim  to  her  prior  to  Arthur's. 
(Webster,  "Arthur  and  Charlemagne,"  Eng.  Stud.  1906,  xxxvi,  348  ff.; 
Schoepperle,  n,  528-31). 

8  Cf.  Schoepperle  Tristan,  11,  541-45  for  a  detailed  comparison  of  the 
abduction  episode  in  Orfeo  with  that  in  the  stories  of  Tristan  and  Guinevere. 


SIR   ORFEO  199 

d'Orphey  sung  by  an  Irish  harper  (Kittredge,  p.  201).  It  is  con- 
ceivable that  the  bilingual  Breton  minstrels  may  have  turned 
the  Orpheus  story  into  the  form  of  a  lay  which  the  Irish  min- 
strel learned  and  sang,  or  that  the  latter  himself,  knowing  the 
form  and  assured  popularity  of  the  "  Breton  lays,"  may  have 
worked  the  transformation  of  the  original  story.  In  Bretoniz- 
ing  the  original  legend  it  was  infused  with  Celtic  "  magic  "  and 
turned,  as  were  all  the  extant  lays,  into  swift-flowing  French 
couplets.  From  this  form  the  story  then  passed  into  Middle 
English.  It  is  probable  that  the  English  version  owes  to  its 
French  original  those  special  qualities  which  make  it  "  nearly 
perfect  as  an  English  representative  of  a  Breton  lay  —  its  brevity 
and  romantic  charm  "  (Ker,  English  Literature,  Mediaeval,  p. 
127). 

BIBLIOGRAPHY 

Texts:  (i)  Auchinleck  MS.  ed.  Laing,  Select  Remains  of  Ancient 
Popular  Poetry  of  Scotland,  Edin.  1822,  1884;  O.  Zielke,  Breslau,  1880 
(a  critical  edition) ;  M.  Shackford,  Legends  and  Satires,  p.  141  ff., 
Boston,  1913;  A.  Cook,  A  Literary  Middle  English  Reader,  Boston, 
1915;  (2)  Ashmole  61,  Bodleian,  ed.  Halliwell,  Illustrations  of  the 
Fairy  Mythology  of  A  Midsummer  Night's  Dream,  Lond.,  1895;  (3) 
Harleian  3910,  Brit.  Mus.,  Ritson,  n,  248  ff.,  1884.  Trans.  E.  E. 
Hunt,  Cambridge,  Mass.,  1909;  E.  Rickert,  Romances  of  Love,  p.  32 
ff. ;  Weston,  Chief  ME.  Poets,  p.  133. 

Studies  and  Analogues:  Korting,  Grundriss,  p.  160;  Wells,  Manual, 

p.  783. 

Foulet,  L.     "The  Prologue  of  Sir  Orfeo,"  MLN.  xxi,  46-50    (1906). 

Guillaume,  G.     "  The  Prologues  of  the  Lay  le  Freine  and  Sir  Orfeo," 

MLN.  xxxvi,  458-64  (1921). 
Gummere,  F.  B.     The  Popular  Ballad,  Boston,  1908. 
Ker,  W.     English  Literature,  Mediaeval,  p.  127-29,  N.  Y.,  191 2. 
Kittredge,  G.  L.     "Sir  Orfeo,"  Amer.  Jour.  Phil,  vn,  176-202   (1886). 
Marshall,  L.  E.     "Greek  Myths  in  Modern  English  Poetry,"  Studi  di 

filologia  moderne,  v,   203-32    (191 2). 
Schofield,  W.     English  Literature  —  to   Chaucer,  pp.    184-86.     N.   Y., 

1906. 
Wirl,  J.  "Orpheus  in  der  engl.  Literatur,"  Wiener  Beitrdge,  XL  (1913); 

rev.  Archiv,  cxxxn,  239. 


PARTONOPE   DE    BLOIS 

Versions.  The  story  of  Partenopeus  de  Blois  and  his  fairy 
love,  Melior,  has  been  called  one  of  the  most  beautiful  romances 
of  the  Middle  Ages.  The  Paris  Arsenal  manuscript  (A,  ed. 
Crapelet,  Paris,  1834),  containing  9744  verses,  is  the  oldest  ex- 
tant text.  It  begins  with  an  account  of  Partenopeus,  the  young 
nephew  of  King  Clovis  of  France,  and  ends,  after  a  great  combat 
between  the  hero  and  the  Sultan  of  Persia,  with  the  celebration 
of  a  triple  marriage,  of  Partenopeus  to  Melior,  of  the  young 
king  of  France  to  her  sister,  the  wise  Urake,  and  of  the  hero's 
faithful  friend,  Gaudin,  to  Persevis,  Urake's  maid  of  honor. 
Bodtker  (1,  p.  vi)  thought  that  the  vivid  style  and  picturesque 
descriptions  of  this  ending  make  it  one  of  the  most  striking 
passages  in  Old  French  literature,  but  he  held  that  it  was  added 
by  a  Picard  poet  to  the  original  version  of  the  poem. 

The  author  of  this  original  version  was  formerly  supposed 
to  be  Denis  Piramus,  the  undoubted  author  of  La  Vie  Seint 
Edmunt.  As  Ward  {Catalogue,  1,  700)  pointed  out,  a  misinter- 
preted reference  in  the  Vie  to  the  popularity  of  Partenopeus 
was  the  principal  basis  for- this  ascription,  and  Haxo  (p.  350  ff.) 
noted  that  a  comparison  of  the  language  of  the  two  poems  alone 
would  serve  to  show  that  they  could  not  have  been  written  by 
the  same  author.  Haxo  was  inclined  to  identify  Denis  with 
Magister  Dionisius,  a  monk  of  St.  Edmund's  Bury  from  11 73  to 
1200,  and  to  date  the  Vie  Seint  Edmunt  after  11 75,  or,  if  Marie's 
Lais,  to  which  it  also  refers,  were  not  written  until  as  late  as 
1 1 80,  after  1190.  Denis's  allusion  to  Partenopeus  is  the  earliest 
known  and  indicates  that  the  poem  must  be  dated  at  some  time 
before  his  own  work.1     Kawczynski   (1)  believed  the  romance 

1  Groeber,  Grundriss,  1902,  p.  589,  thought  that  Florimont,  a  romance 
written  about  1188  by  Aimon  de  Varennes,  shows  in  part  the  influence  in 
style  and  structure  of  Partenopeus.  Florimont,  similarly  written  in  praise 
of  a  lady,  brings  in  classical  names  and  allusions,  and  involves  the  hero  in 
a  love  affair  with  a  fee  who  gives  him  magic  gifts. 

200 


PARTONOPE   DE  BLOIS  201 

was  composed  in  1153  by  an  author  attached  to  the  Count  of 
Blois  (Rom.  xxxi,  475-6). 

Partenopeus  is  also  contained  in  six  other  French  manuscripts, 
the  relationships  of  which  have  been  studied  by  Pfeiffer.2  They 
constitute  the  B  redaction  which  does  not  diverge  notably  from 
that  of  the  A  text  until  v.  9163  of  the  latter.  This  divergent 
part  is  known  as  the  Continuation.  In  B  there  is  no  single  com- 
bat and  the  Sultan  departs  meditating  vengeance.  Partenopeus 
is  wedded  to  Melior  and  much  of  the  remaining  part  of  the 
story  is  devoted  to  telling  of  the  return  of  the  Sultan,  and  to 
recounting  the  adventures  of  Anselet,3  the  heathen  lad  whom 
Partenopeus  tricked  into  receiving  baptism,  and  then,  himself 
bent  on  suicide,  abandoned.  Of  the  six  manuscripts,  the  longest, 
G,  contains  11,848  lines  in  octosyllabic  couplets  and  768  lines 
in  Alexandrine  verse.  It  is  closely  related  to  P,  a  manuscript 
also  in  the  Bibliotheque  Nationale,  Paris,  but  the  two  texts  are 
independent  derivatives  of  a  common  source  (Pfeiffer,  p.  37). 
From  his  study  of  a  Tours  manuscript  in  which  the  story  is 
carried  further  than  in  any  other  French  text,  Sneyders  de  Vogel 
(p.  17  ff.)  attempted  to  prove  that  the  B  Continuation,  although 
preserved  only  in  manuscripts  later  than  A,  was  nevertheless 
the  work  of  the  original  poet.  His  argument  rested  chiefly  on 
analogies  in  style  and  in  point  of  view  between  the  first  part, 

2  Ueber  die  handschriften  des  altjrz.  romans  Partenopeus.  Diss.  Marburg, 
1884;  Ausgaben  u.  Abhandlungen  aus  d.  Gebiete  der  rom.  Phil,  xxv  (1885). 
Seven  MSS.  were  cited  by  Pfeiffer  and  by  Bodtker,  2,  p.  1.  In  connection 
with  Nr.  7516,  Nouv.  acq.,  Bibl.  Nat.,  Paris,  see  Rom.  ix,  509  and  xxxi, 
473  (1902).  This  MS.  was  listed  in  1407  among  the  books  of  Francesco 
Gonzaga  I  of  Mantua.  In  Pfeiffer's  opinion  the  oldest  redaction  of  Parti- 
nopeus  is  represented  by  A  (the  Arsenal  MS.),  the  younger  B  by  MSS.  BPG. 

3  Anselet  is  an  important  character  for  the  classification  of  the  different 
versions.  In  the  A  text  the  poet  promises  to  tell  more  of  Anselet's  adven- 
tures but  does  not  do  so  (English  version,  Bodtker,  v.  7069).  The  fact 
that  more  is  told  about  him  in  the  B  Continuation  was  one  of  Sneyder  de 
Vogel's  reasons  for  believing  in  its  authenticity.  He  suggested  that  the  epi- 
sode in  B  in  which  Anselet  kills  his  faithful  greyhound,  Noon,  may  have 
been  interpolated  by  some  one  familiar  with  the  Canis  story  in  Les  Sept 
Sages.  In  the  Norse  saga,  towards  the  end  of  the  story,  Anselet  is  identi- 
fied with  Gaudin  who  in  the  other  versions  is  represented  as  another  faithful 
friend  of  Partenopeus  (Eng.  version,  v.  9396  ff.).  Bodtker,  2,  p.  34,  expressed 
the  opinion  that  in  the  original  version  of  Partenopeus  Anselet  and  Gaudin 
were  two  distinct  personalities.  Their  early  history  and  relation  to  Parte- 
nopeus seem,  however,  amazingly  alike. 


202  MEDIEVAL   ROMANCE    IN   ENGLAND 

practically  identical  in  A  and  B,  and  the  Continuation  as  set 
forth  in  the  Tours  manuscript  and  the  still  longer  text  now 
preserved  only  in  a  Dutch  version.4 

The  first  effective  classification  of  the  Continental  versions  of 
Partenopeus  was  made  by  Kolbing  ("  Uber  die  verschiedenen 
Gestaltungen  d.  Partonopeus-sage,"  Bartschs  Germanist.  Studien, 
Suppl.  Gerntania,  n,  55  ff.).  There  seem  to  have  been  two  ver- 
sions of  the  story;  to  the  first  of  these  (Y)  belong  all  the  French 
manuscripts  of  the  A  and  B  group,  a  mid-thirteenth  century 
German  version,5  the  Dutch  version,  a  free  Italian  adaptation 6 
of  the  late  fourteenth  century,  and  the  longer  English  version.7 

This  English  version  (ed.  Bodtker)  is  contained  in  five  fif- 
teenth-century texts,  of  which  the  oldest  one,  now  in  University 
College,  Oxford,  was  written  about  the  middle  of  that  century. 
Wiilker's  suggestion  (Anglia  xn,  607  ff.)  that  Gower  knew  the 
English  version  of  Partonope,  was  disputed  on  chronological 
grounds  by  Kolbing  (Eng.  Stud,  xiv,  435  ff.).  In  the  absence 
of  any  other  evidence  than  that  of  the  manuscripts,  there  seems 
no  reason  to  suppose  the  English  version  antedated  the  fifteenth 
century.  From  his  study  of  the  manuscripts  of  Partenopeus 
printed  before  1888,  Weingartner  concluded  that  no  extant  Old 
French  manuscript  is  the  source  of  this  Middle  English  version. 
He  showed  by  detailed  comparison  with  Crapelet's  edition  many 
instances  of  practically  verbal  translation  from  the  French,  but 
noted  too  that  the  English  poet,  though  faithful  to  the  order 
and  content  of  the  French  story,  was  not  merely  a  translator. 
Weingartner  (p.  13)  and  Schofield  (p.  308)  both  comment  on 

4  The  extant  Dutch  fragments  (ed.  Bormans,  Brussels,  1871)  contain  about 
9000  lines.  Cf.  A.  van  Berkum,  De  middelnederlandsche  Bewerking  van  den 
Parthonopeus  Roman  en  hare  verhouding  tot  het  oudfransche  origineel.  Diss. 
Leyden,  1897. 

5  Written  by  Konrad  von  Wurzburg  from  a  German  version  made  for 
him  by  Heinrich  Marschant.  Kolbing  (Germ.  Stud.  11,  96)  thought  it  prob- 
able that  this  written  version  was  closely  related  to  the  extant  Dutch  ver- 
sion. He  did  not  agree  with  Bartsch's  assertion,  in  the  latter's  edition  of 
Konrad's  poem,  concerning  the  superiority  of  Konrad's  poem  to  its  French 
original.  Cf.  H.  van  Look,  Der  Partonopiar  Konrads  von  Wurzburg  u.  der 
Partonopeus  de  Blois.    Diss.  Strassburg,  1881. 

6  Cantare  de  lo  Bel  Gherardino,  ed.  F.  Zambrini,  Bologne,  1867.  The 
poem  has  been  attributed  to  Antonio  Pucci.     Cf.  Bodtker,  2,  pp.  2-4. 

7  Cf.  Kolbing,  "  Ueber  die  engl.  Versionen  der  Partonopeussage,"  pp. 
80-92,  Beitrage  z.  vergleich.  Geschichte  der  ram.  Poesie  u.  Prosa  des 
Mittelalters,  Breslau,  1876. 


PARTONOPE   DE   BLOIS  203 

the  poet's  occasional  willingness  to  omit  ornamental  descriptive 
detail,  and  they  signalize  the  passage  (v.  6168  ff.)  in  which  those 
who  care  for  hearing  about  the  minute  details  of  a  lady's  dress 
are  referred  to  the  French  original.  Such  adaptations  and 
abbreviations  in  the  English  version  were  made,  according  to 
Schofield,  to  meet  the  taste  of  one  who  was  "  neither  very  re- 
fined himself,  nor  wrote,  it  would  seem,  for  gentlefolk."  But  the 
very  length  and  nature  of  this  Middle  English  poem  in  which, 
as  Weingartner  (p.  45)  himself  observed,  the  author  shows  a 
special  predilection  for  reflective  and  allegorical  passages,  more 
or  less  invalidate  such  criticisms.  Like  Chaucer,  the  poet  re- 
gards his  "  olde  bokes  "  with  serious  deference ;  in  them  "  ys 
goode  doctrine"  (v.  34).  He  does  not  in  reality  condense  that 
"olde  booke,"  "In  ffrenshe  also,  and  fayre  endyted  "  (v.  501), 
which  was  his  source,  for  he  followed  it  in  such  leisurely, 
Lydgatian  fashion  that  his  own  text  runs  to  12,192  lines.  A 
careful  study  of  Partonope  would  show,  it  seems  to  the  writer, 
not  only  Chaucerian  influence  in  the  phraseology,  but  also  a 
real  appreciation  of  the  artistic  effectiveness  of  the  French  poem. 
The  poet  pauses  for  such  rhetorical  passages  as  the  anaphoric 
lines  in  which  Melior  laments  (v.  6046),  "  My  joye,  my  bolde- 
nes,  and  all  my  game " ;  he  lingers  over  the  description  (v. 
689  ff.)  of  a  moonlit  forest  night,  of  the  meadow  where  the  grass 
grew  stirrup-high,  and  later  of  that  "  delectabell  contre  "  ruled 
by  Melior.  Though  the  excess  of  detail  grows  tedious,  though 
there  is  little  vivacity  in  the  long-winded  conversations,  and  the 
adventures  are,  like  those  in  the  French  original,  too  intermin- 
ably drawn  out,  this  Middle  English  version  must  be  considered 
an  important  and  far  from  unworthy  rendering  of  this  particular 
story.  It  was  undertaken,  so  the  poet  tells  us  (v.  2335  ff.),  be- 
cause his  "  sovereyne  "  thought  the  story  too  little  known  and 
commanded  him  to  draw  it  from  French  into  English. 

The  second  and  shorter  version  of  Partonope  was  also,  in 
its  Middle  English  form,  derived  from  a  French  source  from 
which  Kolbing  (Beitrdge,  p.  90)  thought  it  preserved  the 
word  "  enchauntement,"  v.  95.  This  lost  French  version  was 
not  less  widely  dispersed,  according  to  the  same  scholar,  than 
was  the  longer  one,  as  is  shown  by  the  number  of  independent 
derivative  versions  to  which  it  gave  rise.     These  are  alike  in 


204  MEDIEVAL   ROMANCE    IN    ENGLAND 

that  the  opening  part  of  the  story  tells  of  the  father  of  Melior 
and  of  her  accession  to  the  throne.  The  scene  of  the  story  is 
Greece,  and  the  events  which  follow  Melior's  decision  to  take 
a  husband,  are  about  the  same  as  those  which  in  the  longer  ver- 
sion are  related  to  Partenopeus  by  Melior  after  her  magic  arts 
have  brought  him  to  her  side.  In  Middle  English  this  version 
is  preserved  in  a  fragment  of  308  lines  in  a  manuscript  (c.  1450) 
now  at  Vale  Royal  (Bodtker,  1,  p.  viii).  The  same  version  (Z) 
is  found  in  a  Danish  poem  (ed.  C.  Brandt,  Romantisk  Digtning, 
Copenhagen,  1870, 11,  33  ff.),  written  in  1484  and  twice  printed  in 
the  sixteenth  century,  and  in  the  two  redactions  of  an  Icelandic 
version  now  extant  in  two  fifteenth-century  manuscripts  and  in 
several  others  of  later  date.8  Bodtker  (2,  pp.  45-47)  believed 
the  Danish  and  the  Icelandic  forms  were  derived  from  a  lost 
Norwegian  version  of  the  thirteenth  century.  Like  the  short 
English  version,  this  Norwegian  Saga  was  in  all  probability 
derived  from  a  lost  Anglo-Norman  version.  In  general  these 
Scandinavian  texts  treat  the  brilliant  social  aspects  of  the  French 
romance  with  something  of  the  austerity  of  the  North.  They 
omit  the  love  complaints  and  the  more  elaborate  descriptive 
passages ;  they  change  innumerable  small  details,  —  for  instance, 
it  is  by  means  of  a  magic,  light-giving  jewel,  instead  of  by  a 
lantern,  that  the  hero  first  sees  his  love  (Bodtker,  2,  p.  26) ; 
they  diverge  more  widely  than  do  the  texts  of  the  Y  version 
from  the  original  tale.  In  Spain,  where  the  Z  version  became 
widely  known,  the  Spanish  redactor  introduced  into  his  story 
"  many  traits  that  did  honor  to  his  patriotism  and  Catholicism  " 
but  at  the  expense  of  fidelity  to  the  original  story.  Buchanan's 
list  of  eight  Spanish  editions  dating  from  15 14  to  1844  illus- 
trates the  popularity  of  the  romance.  It  is  probable  that  the 
original  Castilian  text  differed  very  little  from  the  extant  edi- 
tion printed  at  Toledo  in  1526  (Bodtker,  3,  p.  235). 

Origin.  The  romance  deals  chiefly  with  the  love  of  Parte- 
nopeus, the  young  nephew  of  King  Clovis  of  France,  for  Melior, 
a  queen  whose  fairy  nature  is  indicated  by  the  unparalleled 
richness  of  her  abode  and  by  certain  magic  arts  through  which 

8  Cf.    Kolbing",    Ueber   die   nordischen   Gestaltungen   de   Partonopeussage, 
Strassburg,  1873. 


PARTONOPE    DE    BLOIS  205 

she  can  make  herself,  her  servants,  and  even  Partenopeus,  in- 
visible.9 As  concerns  the  ultimate  origin  of  the  story,  the  most 
significant  episodes  are  the  following:  Partenopeus  is  brought 
to  a  mysterious  and  magnificent  abode  by  supernatural  means; 
is  served  by  invisible  but  assiduous  hands ;  becomes  the  lover 
of  Melior  on  condition  that  he  will  not  for  a  given  period  at- 
tempt to  see  her;  breaks  this  prohibition  at  the  instigation  of 
his  mother;  beholds  Melior's  beauty  in  the  momentary  light  of 
the  lantern  given  him  by  his  mother;  is  cast  forth  the  next 
day  in  shame  and  despair  from  Melior's  palace;  endures  great 
hardships;  and  ultimately  regains  his  lady's  favor.  All  this 
has  been  given  a  Christian  10  and  typically  mediaeval  character, 
even  to  the  introduction  of  a  gorgeous  tournament  as  the  final 
test  of  Partenopeus  for  Melior's  hand.  It  has,  nevertheless,  gen- 
erally been  recognized  as  a  mediaeval  transformation  of  the 
beautiful  legend  of  Cupid  and  Psyche,  first  found  in  Apuleius's 
Metamorphoses    (ed.   S.    Gaseler,   Loeb   Classical   Libr.,   Lond., 

9  The  rest  of  the  lady's  powers  are  merely  necromantic.  She  can  cause 
illusions  of  various  sorts,  make  a  room  seem  of  gigantic  size  or  the  sun 
seem  to  shine  at  midnight,  cause  the  apparition  of  great  tourneys,  of  wild 
beasts,  etc.  (cf.  Eng.  version,  v.  5946  ff.).  The  same  art  of  illusion  is  prac- 
tised by  the  fairy  lady  in  Le  Bel  Inconnu  (BI)  for  the  purpose  of  playing 
tricks  on  her  lover,  but  neither  in  her  case  nor  in  that  of  Melior,  does  the 
power  seem  to  be  regarded  as  more  than  a  special  accomplishment.  For 
such  arts  alone  Melior  could  no  more  be  considered  a  fee  than  could  the 
Orleans  clerk  in  Chaucer's  Franklin's  Tale  or  Colle  Tregetour  in  the  Rous  of 
Fame,  who  are  said  similarly  to  practice  the  art  of  illusion.  For  other  in- 
stances of  its  use,  see  Schofield,  PMLA.  xvi,  419  (1901)  ;  Tatlock,  "Astrology 
and  Magic  in  Chaucer's  Franklin's  Tale,"  Kittredge  Anntv.  Papers,  1913,  pp. 
341,  349.  Faral,  Sources  Latines,  p.  318,  pointed  out  that  Melior's  knowl- 
edge of  necromancy  might  be  compared  with  that  of  Medea  in  the  Roman 
de  Troie  (v.  12 16),  and  the  belief  that  it  was  based  on  her  mastery  of  the 
Seven  Arts  with  similar  ideas  expressed  in  Eneas  (v.  2199),  and  in  Troie 
(v.  1219). 

10  Melior,  like  Yonec  (Cross,  Revue  Celt,  xxxi,  p.  414),  is  at  great  pains 
on  the  occasion  of  her  first  meeting  with  her  lover  to  explain  her  own  com- 
plete orthodoxy.  But  the  mother  of  Partenopeus  is  certain  that  Melior  is 
a  devil  practising  enchantments  on  her  son.  She  explains  her  fears  to  the 
Bishop  of  Paris  who  thereupon  so  works  on  the  religious  fears  of  Parte- 
nopeus that  he  is  at  last  persuaded  to  take  his  mother's  magic  lantern  and 
try  to  behold  his  love  (Eng.  version,  5650-586,0.  In  Peter  von  Staufen- 
berg  (ed.  Schroeder,  Berlin,  1894),  a  Middle  High  German  romance  of  the 
fourteenth  century,  the  same  pietistic  treatment  of  a  similar  story  is  observ- 
able. On  the  advice  of  a  priest  Peter  comes  to  believe  that  the  fairy  wife 
who  has  so  richly  endowed  him  is  a  devil  of  hell.  (Cf.  Cross,  Mod.  Phil,  xn, 
592,  n.  2) 


2o6  MEDLEVAL  ROMANCE   IN   ENGLAND 

191 5,  p.  185  ff.).11  Divested  of  the  moral  and  allegorical  ele- 
ments added  by  the  African  Apuleius  or  by  his  unknown  Greek 
predecessors,  the  legend  itself  is  reducible  to  well-known  folk- 
tale types  12  which  are  combined  in  a  distinctive  fashion.  Parte- 
nopeus, as  is  shown  below,  might  be  explained  piecemeal  by 
reference  to  mediaeval  lays  and  romances,  but  in  the  Psyche 
legend  alone  is  there  an  indisputable  parallel  for  the  mysterious 
marriage,  for  the  nature  of  the  marriage  taboo,  for  the  night 
scene  in  which  it  is  broken,  and  the  anguished  separation  of  the 
lovers.  It  may  well  be  questioned  how  and  when  Apuleius's 
story  came  to  France,13  but  it  seems  impossible  to  doubt  that 

11  The  assertions  by  Groeber,  Voretzsch,  p.  384,  and  Schofield,  p.  307, 
that  this  is  not  the  case,  are  unconvincing.  The  differences  between  the 
classical  and  mediaeval  versions  of  the  story  may  be  fully  recognized  with- 
out invalidating  the  claim  that  the  episodes  of  Partenopeus,  as  listed  above, 
are  structurally  related  to  those  of  the  Psyche  legend  or  to  the  folk-tale 
from  which  it  came.  Pschmadt  in  his  study,  Die  Sage  von  der  verfolgten 
Hind,  iqii,  p.  97,  called  attention  to  the  unmistakable  similarity  between 
the  fairy  palace  as  described  in  the  Psyche  legend  and  in  Partenopeus.  B. 
Stumfall,  "  Das  Marchen  von  Amor  u.  Psyche  in  seinem  Fortleben," 
Munchener  Beitrage,  xxxix,  pp.  8-13,  Leipzig,  1907,  asserted  that  the  author 
of  Partenopeus  drew  on  a  folk-tale  of  the  Psyche  type  but  not  on  the  story 
of  Apuleius. 

12  One  of  the  fullest  discussions  of  these  folk-tale  types  is  to  be  found 
in  the  notes  by  A.  Gough  to  L.  Friedlander's  Roman  Life  and  Manners, 
Lond.,  1913,  vol.  rv,  99-123.  Roughly  speaking  these  tales  may  be  grouped 
under  such  captions  as  the  following:  The  Lady  and  the  Monster;  The 
Magic  House  and  the  Invisible  Servants;  the  False  Sisters;  The  Jealous 
Mother-in-law;  The  Broken  Prohibition;  Impossible  Tasks.  Gough  believed 
that  a  genuine  folk-tale  formed  the  basis  of  the  Apuleian  narrative  and  noted 
(p.  115)  that  the  group  of  Danish,  Norwegian,  and  Swedish  folk-tales  ana- 
lyzed by  him,  are  most  closely  akin  to  the  story  of  Apuleius.  They  all 
contain  the  important  feature  that  the  young  heroine  is  advised  to  look  at 
her  lover  by  a  light  at  night.  Cf.  Andrew  Lang's  introduction  to  the  re- 
print of  William  Aldington's  translation  of  Cupid  and  Psyche,  Lond.,  1897. 
The  theme  of  the  Secret  Lover  appears  in  the  folk-tales  discussed  by  Kohlef, 
Lais-Marie,  pp.  cvi-cxviii;  Schoepperle,  Tristan,  1,  150-51;  Cosquin,  Rom. 
x,  117-31   (1881). 

13  Kawczynski,  2,  p.  193  ff.,  urged  that  the  work  of  Apuleius  was  known 
to  the  author  of  Partenopeus.  He  also  believed  that  it  exerted  some  influ- 
ence on  Aucassin,  Floire  et  Blancheflor,  Le  Chevalier  au  Cygne,  Berte  aus 
grans  pies,  and  on  Huon  de  Bordeaux.  Huet  in  1909  strongly  questioned 
the  extent  and  even  the  fact  of  this  influence.  He  pointed  out  that  only 
two  MSS.  of  the  Metamorphoses  antedate  the  thirteenth  century,  and  that 
the  first  known  mention  of  it  is  that  by  Vincent  of  Beauvais.  He  believed 
that  the  Metamorphoses  was  practically  unknown  in  Europe  before  this 
period.  In  191 7  he  admitted  the  possibility  that  the  legend  might  have  been 
known    through    other   sources    than    the    Apuleian    narrative,    for    instance, 


PARTONOPE   DE  BLOIS  207 

it  was  known  in  some  form  to  the  author  of  Partenopeus.  His 
introductory  summary  of  the  Trojan  story  and  the  name  which 
he  gives  to  his  hero,  taken  from  that  of  one  of  the  Seven  Cham- 
pions against  Thebes,14  indicate  an  interest  in  classical  legend 
that  must  be  related  to  the  contemporary  popularity  of  romantic 
redactions  of  the  Matter  of  Antiquity.15  The  Roman  de  Troie, 
the  Roman  de  Thebes,  the  Roman  d'Eneas,  written  in  the  sixth 
or  seventh  decade  of  the  twelfth  century,  had  successfully  com- 
bined classic  legends  with  purely  romantic,  mediaevalized  stories 
of  love  and  courtship,  and  offered  not  only  the  incentive,  but 
the  model,  for  similar  attempts. 

The  striking  reversal  in  Partenopeus  of  the  specific  roles  of 
Psyche  and  the  God  of  Love  is  possibly  to  be  accounted  for 
by  reference  to  the  supposedly  Celtic  stories  which  exercised 
so  potent  an  influence  on  the  romance.  It  is  to  be  regretted 
that  no  detailed  study  of  Partenopeus  from  this  point  of  view 
has  yet  appeared,  though  the  essential  fact  has  been  recognized. 
Schofield  (p.  307)  remarked:  "  In  induction  and  other  features 
it  resembles  the  Breton  lays  of  Guingamor,  Guigemar,  and 
Lanval ;  in  development,  the  romances  of  Ivain  and  The  Fair 
Unknown."  The  resemblance  between  Partenopeus  and  the  lays 
is  especially  close  in  the  earlier  part  of  the  story.  Partenopeus, 
in  pursuit  of  a  fairy  boar  sent  by  Melior,  loses  himself  in  the 
forest  of  Ardennes,  until  at  last  he  finds  on  the  shore  a  mysteri- 
ous ship  which  carries  him  to  Melior's  magnificent  city.     The 


through  the  Mythologiarum  of  the  African  compiler,  Fulgentius,  who  drew 
his  abbreviated  story,  the  Fabula  deae  Psicae  et  Cupidnis,  from  Apuleius. 
Though  thus  admitting  the  possibility  of  Apuleian  influence,  Huet  still  in- 
clined to  the  belief  that  Partenopeus  represents  the  combination  of  a  Psyche 
folk-tale  with   the  history  of  a  fee. 

14  Kolbing  (Germ.  Stud,  u,  57)  suggested  that  the  hero's  name  comes 
from  the  city  of  Partenay,  the  lords  of  which  at  various  times  had  some 
connection  with  those  of  Lusignan  to  whom  was  attached  the  famous  legend 
of  their  fairy  progenitor,  Melusine,  the  serpent  woman.  But  the  Melusine 
legend  is  too  late,  it  would  seem,  to  have  influenced  Partenopeus.  Whatever 
was  the  origin  of  the  local  myth,  the  earliest  known  literary  treatment  of 
the  story  is  the  prose  romance  of  Melusine  by  Jean  d'Arras.  It  was  com- 
piled about  1387,  printed  at  Geneva,  1478,  "  englisht  "  about  1500  (EETSES. 
lxviii).  Cf.  J.  Kohler,  Der  Ursprung  der  Melusinensgage;  Eine  ethnologische 
Untersuchung,  Leipzig,  1895;  Baudot,  Les  Princesses  Yolande  et  Les  Dues  de 
Bar  de  la  famille  des  Valois,  I,  Melusine,  Paris,  1900. 

15  Cf.  Dressier,  pp.  130-35;  Otto  (op.  cit.  Ipomedon  Bibliog.),  p.  59,  for 
the  special  influence  of  the  romances  of  Eneas  and  of  Thebes  on  Partinopeus. 


2o8  MEDIEVAL   ROMANCE   IN   ENGLAND 

hero's  hunt  for  a  fairy  animal  as  the  preface  to  his  meeting 
with  the  fee  herself  is  found  in  Guigemar,  in  Guingamour,  in 
Graelent,  and  in  the  later  romance,  Generides.  The  incident  is 
combined  with  that  of  the  fairy  boat  in  Guigemar.16  In  Grae- 
lent, as  in  Partenopeus,  the  fee  makes  a  curious  pretense  of 
anger  and  of  helplessness  before  the  young  hero,  but,  having  at 
last  surrendered  herself,  she  confesses  that  it  was  by  her  wish 
and  means  that  he  has  been  brought  to  her.17.  She  admits  that 
she  had  long  since  heard  of  his  prowess  and  loved  him  even  when 
unseen,18  a  confession  which  closely  parallels  that  of  the  fee  in 
Marie's  Lay  of  Lanval,  of  the  fairy  dame  d'amour  in  Le  Bel 
Inconnu,  and  Libeaus  Desconus,  of  the  fairy  lover  in  Marie's 
Lay  of  Yonec,  and  of  the  fee  again  in  the  old  French  lay  of 
Mellon,  poems  for  which  a  Celtic  ancestry  has  been  claimed. 
In  all  these  instances  the  supernatural  lover  has  the  peculiar 
magnificence,  the  power,  the  lordly  generosity,19  which  are  char- 
acteristic attributes  of  Celtic  fairy  folk. 

16  Bodtker,  2,  pp.  7,  19,  noted  that  in  an  Icelandic  MS.  of  the  seventeenth 
century  the  boar  which  appears  in  the  French,  English,  and  Danish  versions, 
is  replaced  by  a  deer,  the  usual  Fairy  Messenger  of  fees  in  Old  French  ro- 
mances. See  Pschmadt,  note  11,  above;  Isumbras,  note  1;  Warnke-Kohler, 
Lais-Marie,  p.  lxxix.  Magic  Boats  are  familiar  properties  of  romance.  Cf. 
Brown,  Mod.  Phil,  xrv,  392,  n.  4   (1906). 

17  Schofield,  "  The  Lays  of  Graelent  and  Lanval,"  PMLA.  xv,  129  ff.  and 
Cross,  "  The  Celtic  Fee  in  Launfal,"  Kittredge  Anniv.  Papers,  1913,  p.  385, 
argued  that  the  powerful  Celtic  fee  who  wills  and  achieves  what  she  desires, 
is,  in  Graelent  especially,  confused  with  the  swan-maiden  type  of  fairy  who 
is  helpless  without  her  feather  garment.  In  Graelent,  the  hero  finds  the  fee 
bathing,  is  bitterly  reproached  when  he  takes  her  garments  and  wins  her 
only  by  force,  although  she  later  tells  him  she  had  foreordained  their  meet- 
ing. Cross,  "  Celtic  Elements  in  Lanval  and  Graelent,"  Mod.  Phil,  xn,  617 
(1914-15),  noted  that  the  swan-maiden  type,  "generally  regarded  as  dis- 
tinctively of  Germanic  tradition,  figured  in  Celtic  literature  before  the 
twelfth  century."  He  was,  therefore,  inclined  to  discredit  Schofield's  belief 
that  Graelent  had  been  influenced  by  some  form  of  the  Wayland  Smith 
story  (Cross,  p.  621). 

18  For  the  common  folk-lore  motif  of  Love  in  Absence  see  Ipomedon, 
n.  1;  Cross,  Mod.  Phil,  xn,  612,  n.  3.  He  compared  the  fees  in  Mane's  lays 
who  so  frankly  woo  their  mortal  lovers  with  the  Forth-Putting  Women  of 
Old  Irish  story. 

19  In  Partenopeus  the  hero,  during  his  stay  with  Melior,  is  provided  with 
all  possible  accessories  for  his  daily  hunt.  On  his  return  to  Blois  he  is  met 
by  twelve  sumpter  horses  laden  with  gold  sent  by  Melior.  With  this  great 
wealth  he  is  able  to  recruit  a  host  of  followers  and  to  become  the  foremost 
soldier  of  France  (Eng.  version  2508-3066).  In  the  Lays,  Launfal,  after  his 
meeting  with  the  fee,  is  provided  with  magnificent  clothes  and  returns  to  find 


PARTONOPE    DE   BLOIS  209 

Even  more  important  in  this  connection  is  the  command  which 
Melior  laid  upon  her  lover  not  to  attempt  to  see  her  for  a  given 
time.  On  the  one  hand  this  seems  merely  a  modification  of 
Cupid's  command  to  Psyche  and  may  hark  back  to  immemorial 
folk  superstitions  and  marriage  taboos.  The  idea  reappears 
in  one  form  or  another  in  numerous  European  folk-tales  and  in 
such  famous  legends  as  those  of  Lohengrin  and  Melusine,  but  for 
Partenopeus  the  closest  parallel  is  to  be  found  in  just  the  lays 
which  we  have  been  considering,  now  accepted,  at  least  by  the 
Celticists,  as  French  adaptations  of  a  Celtic  folk-tale  of  the 
Offended  Fee.  Regularly  in  this  tale  the  prohibition  is  laid 
upon  a  mortal  lover  by  a  regal  Fairy  Mistress  and  loss  of  her 
follows  when  he  breaks  her  command.20  Comparison  of  the 
fees  in  Lanval,  Desire,  Graelent,  and  Guingarnor,  with  Melior 
shows  that  she  is  essentially  of  their  sisterhood.  She  endows  her 
lover  bounteously ;  she  imposes  the  strange  command ;  she  is 
relentless  when  it  is  broken.  That  Melior  is  so  rationalized  as 
to  seem  only  an  independent  young  queen,  that  her  command 
is  no  more  than  her  own  whim,  and  not  a  law  of  her  being; 
that  she  herself  suffers  anguish  for  her  self-willed  separation 
from  her  lover,  are  inconsistencies,  to  be  sure,  but  they  may  be 
due  to  the  very  element  which  made  the  old  French  poets  so 
quick  to  seize  on  this  particular  type  of  Fairy  Mistress  story. 
To  minds  filled  with  the  precepts  of  courtly  love,  the  fee's  com- 
mand was  completely  in  accord  with  the  insistence  of  courtly 
love  doctrines  on  the  necessity  for  secrecy  in  love.  It  became  a 
test  of  love,  its  breaking  by  the  hero  a  failure  in  love  for  which 
the  direst  hardships,  love-sickness  running  even  into  madness, 
were  but  rightful  expiation  (Cf.  Cross.  Mod.  Phil,  xn,  641, 
1).  For  the  poet  of  Partenopeus,  himself  a  lover  who  paused, 
like  Renaud  in  Le  Bel  Inconnu,  for  numerous  long  digressions 
concerning  his  own  sorrowful  state  and  the  hauteur  of  his  lady,21 


his  men  have  been  finely  arrayed.  Graelent's  Fairy  Mistress  sends  him  the 
best  horse  in  the  world,  a  servant  who  provides  for  his  every  need.  Cf. 
Cross,  Mod.  Phil,  xn,  628  ff.,  The  Fairy  Gifts. 

20  On  the  gets  or  special  prohibition  laid  on  Old  Irish  heroes  see  Schoep- 
perle,  Tristan,  Index.  The  heroes  of  Lanval  and  Graelent  are  forbidden  to 
name  their  fairy  loves.  Some  form  of  taboo  is  almost  universally  character- 
istic of  stories  in  which  supernatural  beings  enter  into  relations  with  mortals. 

21  Cf.  G.  Paris,  Rom.  xv,  10;  Schofield,  Libeaus  Desconus,  p.  108.  It  is 
of  interest  to  note  that  the  English  poet  emphasizes  (v.  6759  ff.)   the  passage 


210  MEDIEVAL   ROMANCE    IN   ENGLAND 

this  courtly  phase  of  his  story  was  of  special  interest.  He  de- 
veloped it  in  accordance  with  the  typical  ideas  and  stylistic 
devices  known  to  him.  Thus  the  lovers  in  exchanging  their  first 
vows  so  expatiate  on  the  duties  and  obligations  of  lovers  that 
they  succeed  in  defining  the  cock  of  love  itself.  Thus  Melior 
exhorts  her  lover  to  achieve  great  deeds  for  love's  sake  and 
thousands  of  lines  are  subsequently  devoted  to  the  description 
of  the  wars  of  Partenopeus  against  the  enemies  of  France ; 22 
thus,  as  a  true  lover,  the  hero  implores  Melior's  mercy  because 
on  his  return  to  France,  he  had  under  the  influence  of  a  drug 
even  for  a  moment  "  falsely  forgot  "  her ;  thus  Melior  laments  the 
loss  of  all  her  good  when  Partenopeus  breaks  her  command,  and 
he  himself  in  wild  grief  over  his  exile  from  her,  starves  himself 
until  he  is  unrecognizable  and  goes  off  into  the  forest  to  die. 
Equally  typical  are  the  long  discussions  between  Melior  and  her 
sister  Urake  concerning  the  sin  and  suffering  and  even  the  death 
of  Partenopeus,  which  Urake  reports,  passages  obviously  de- 
signed for  their  emotional  effectiveness,  for  the  laying  of  love  in 
the  balance  against  pride  or  grief  or  anything  else  that  was  con- 
ceived as  an  enemy  to  love.  In  Partenopeus,  as  in  the  Lays, 
the  tragic  type  of  story  in  which  the  Offended  Fee  is  irreparably 
lost,  is  so  far  departed  from  that,  after  adequate  suffering,  the 
true  lover  regains  his  fairy  love  (cf.  Cross,  Mod  Phil,  xn,  641). 
This,  of  course,  was  in  accord  with  the  current  understanding  of 
folk-lore  and  romance  that  no  true  lover  should  go  forever  un- 
rewarded. It  was  also  in  harmony  with  the  Cupid  and  Psyche 
story. 

In  certain  structural  features  Partenopeus  seems  then  to  com- 
bine a  traditional  classical  legend  involving  a  marriage  taboo 
with  Celtic  narratives  of  similar  theme  as  they  had  been  modi- 
fied and  developed  in  the  so-called  Breton  lays.  But  in  style  and 
in  much  of  its  content,  Partenopeus  is  to  be  explained  by  refer- 
ence to  the  longer  contemporary  romances.  It  is  a  matter  for 
regret  that  the  relationship  between  this  romance  and  Chretien's 

in  his   original  lamenting   the   folly   of   these   "  olde    clerkes "   who   satirize 
women  and  their  love. 

22  Cf.  Brown,  "  Iwain,"  Harvard  Studies  vni,  p.  129.  One  of  Iwain's 
greatest  exploits  is  his  Single  Combat  with  Gawain.  Partenopeus  has  a 
tremendous  conflict  with  the  noble  heathen  king,  Sornegour,  a  fight  on 
which  the  fate  of  France  depends   (Eng.  version,  3225  ff.) 


PARTONOPE   DE   BLOIS  211 

I  wain  or  Renaud's  Le  Bel  Inconnu  or  Hue  de  Rotelande's  1  pome- 
don  has  not  been  more  exactly  indicated.  Between  Iwain  and  our 
romance  some  general  analogies  were  pointed  out  by  Voretzsch 
(pp.  384-86).  Laudine,  like  Melior  a  rationalized  fee,  simi- 
larly recognizes  the  claim  of  chivalric  honor  and  accedes  to  her 
lover's  wish  to  leave  her  and  return  to  court;  she  is  unforgiv- 
ing when  he  breaks  her  command  to  return  within  a  year,  as 
Melior  is  when  Partenopeus  breaks  her  command  to  refrain 
from  looking  upon  her  for  a  given  time.  Laudine  is  brought 
to  reconciliation,  after  Iwain  has  run  mad  in  the  woods,23  only 
through  the  long  efforts  of  her  friend  Lunet,  who,  like  Urake, 
teaches  the  hero  how  to  regain  his  lady's  favor.  In  Le  Bel  In- 
connu, besides  the  personal  digressions  already  noted  as  com- 
mon to  Renaud's  poem  and  to  Partenopeus,  the  two  romances 
have  in  part  the  same  fairy-like  setting,  the  He  d'Or  in  the  first, 
and  the  Chief  d'Oire  in  the  second.  The  fairy  ladies  make  simi- 
lar confessions  of  love  to  the  hero  and  pay  him  a  seductive  noc- 
turnal visit.  Each  lady  is  said  to  have  learned  in  youth  her 
magic  powers  and  to  be  possessed  of  an  abode  of  extraordinary 
richness.  In  each  romance  a  Faithful  Squire  figures  largely.24 
In  Ipomedon  the  heroine  is  a  capricious  young  duchess  whose 
pride  makes  her  as  difficult  to  win  as  the  Offended  Fee  of  Celtic 
or  other  lineage.  Like  Melior,  however,  she  too  falters  and  fails 
in  trying  to  pronounce  the  lover's  name.  The  scene  is  too  much 
alike  in  the  two  romances  not  to  suggest  specific  borrowing.25 
Very  similar  also  in  the  account  of  the  tournament 26  is  the  epi- 
sode in  which  the  hero  overthrows  his  greatest  opponent  and  is 

23  The  madness  of  the  hero  was  a  motif  which,  once  used  in  Iwain,  had 
before  it  "une  brillante  fortune"  (G.  Paris,  Furnivall  Misc.  1902,  p.  393). 
Cross,  Mod.  Phil,  xn,  p.  641,  n.  1. 

24  Cf.  Schofield,  Libeaus  Desconus,  p.  no,  on  the  Squire  Robert  in  Le 
Bel  Inconnu.  In  Partenopeus  there  are  two  figures  of  this  sort.  The  heathen 
boy  Fursin,  also  called  Gileamour,  allows  himself  to  be  christened  Anselet  (see 
above,  note  3)  out  of  his  devotion  to  Partenopeus  (Eng.  version,  v.  6877  ff.). 
Toward  the  end  of  the  story  the  elderly  knight,  Gaudin,  insists  on  taking 
service  with   Partenopeus  and   renders  him  great  service    (v.   9396). 

25  Cf.  Eng.  version,  v.  8817  ff.,  9063-65.  Kohler,  Kleinere  Schriften, 
in,  1  ff.,  "  Das  vom  Sterbenden  nicht  vollendete  Wort,"  discussed  the  epi- 
sode in  Orlando  Furioso.  Cf.  Ipomedon  here,  note  6;  Carter,  p.  254,  was 
inclined  to  believe  that  the  author  of  Partenopeus  borrowed  from  Ipomedon. 

26  Cf.  K.  G.  Webster,  "  The  Twelfth  Century  Tourney,"  Kittredge  An- 
niversary Papers,  pp.  227-342,  for  a  discussion  of  the  tourney  in  Partenopeus 
and  other  romantic  and  historical  sources. 


212  MEDIEVAL   ROMANCE    IN    ENGLAND 

ably  helped  by  a  friend.  In  style,  in  the  leisurely  elaboration 
with  which  the  picturesque  background  of  chivalric  life  is  de- 
scribed, in  spirit  and  tone  and  length,  these  last  two  romances 
especially  belong  to  the  same  sophisticated  and  delightful  genre. 


BIBLIOGRAPHY 

Texts:  (i)  Univ.  College,  Oxford,  C,  188  (7096  lines),  pr.  W.  E. 
Buckley,  Roxburghe  Club,  Lond.,  1862;  (2)  Rawlinson  Poet.  14, 
Bodleian,  partly  printed  by  Buckley;  (3)  Eng.  Poet.  C,  3  (158  lines), 
Bodleian,  pr.  by  Buckley;  (4)  Lord  Robartes'  MS.  now  belonging  to 
Viscount  Clifden,  pr.  Wulker,  Anglia  xn,  607-20  {cf.  Kolbing,  Eng. 
Stud,  xiv,  435-37  (1890));  (5)  Addit.  MS.  35,288,  Brit.  Mus.  (12,192 
lines) ;  (6)  Vale  Royal  MS.,  pr.  R.  C.  Nichols,  Roxburghe  Club,  1873. 
All  the  MSS.  are  printed  in  full  by  A.  T.  Bodtker,  Partonope  of  Blois, 
The  Middle  English  Versions,  EETSES.  cix.  191 2;  (7)  Addit.  MS.  4860 
(18th  cent.),  cf.  Ward,  Catalogue,  I,  707. 

Studies:  Wells,  Manual,  p.   785. 
Bodtker,  A.  T.     (1)    See  Texts. 

(2)  "  Partenopeus  de  Blois,  Etude  Comparative  des  versions  islan- 
daise  et  danoise,"  Videnskabs-Selskabets  Skrifter,  11,  Hist.-Filos. 
Kl.  No.  3.  Christiania,  1904.  Rev.  Rom.  xxxrv,  167;  Deut.  Lit. 
Zeitung,   1905,  col.   34~35- 

(3)  "Partenopeus  in  Catalonia  and  Spain,"  MLN.  xxi,  234-45 
(1906). 

Buchanan,  M.  A.    "  Partinuples  de  Bles  (A  Bibliography  of  the  Spanish 
Chapbook)"  MLN.  xxi,  3-8  (1906). 

Catalan  version,  1588,  reprinted.     Historia  de  Vesforcat  cavalier  Parti- 
nobles.    Barcelona,    191 2. 

Dressier,  A.     Der  Einfluss   des   altfrz.   Eneas-Romanes   auf   die  altfrz. 
Literatur.    Diss.    Borna-Leipzig,    1907. 

Haxo,  H.     "Denis  Piramus,"  Mod.  Phil,  xn,  345-66;  559-83  (1914-15). 

Huet,  G.       "  Le  Roman  d'Apulee:  Etait-il  connu  au  Moyen  Age?"  Le 
Moyen  Age,  xxn,  22-28  (1909);  xxix,  44-52   (1917). 

Kawczynski,  M.  (1)  Partenopeus  de  Blois.  162  pp.  Cracow,  1902. 
Cf.  Bull,  de  VAcademie  des  Sciences  de  Cracow,  1901.  A  German 
resume  of  the  Polish  scholar's  work  is  given  at  the  end  of  his 
book.  Literaturbl.  xxm,  28-33;  Rom.  xxxi,  475-76. 
(2)  "  1st  Apuleius  im  Mittelalter  bekannt  gewesen? "  Bausteine 
zur  rom.  Phil.,  Festgabe  f.  Mussafia,  pp.   193-210.  Halle,  1905. 

Miiller,    L.      Sprachliche    u.    text-kritische    JJntersuchungen    iiber    den 
altfrz.  Partenopeus  de  Blois.     Diss.  Gottingen,  1920. 

Schofield,  W.     English  Literature  to  Chaucer,  p.  307. 

Sneyders  de  Vogel,  K.     "  La  Suite  du  Parthenopeu  de  Blois  et  la  ver- 


PARTONOPE   DE   BLOIS  213 

sion   hollandaise,"   Revue   des   langues   romanes,   xlviii    (5e    Ser. 
vin),    5-29    (1905).     Rev.   Rom.    xxxv,    617;   Zts.   /.   rom.   Phil. 
xxx,  510  (1906). 
Weingartner,  F.    Die  mittelengl.  Fassungen  der  Partonopeussage  u.  ihr 
Verhdltnis  zum  altfrz.  Originate.  Diss.  Breslau,  1888. 


WILLIAM   OF   PALERNE 

Versions.  It  is  not  surprising  that  a  romance  of  the  type  of 
William  of  Palerne,  charming  as  it  is  in  places,  should  have 
achieved,  so  far  as  records  show,  but  moderate  recognition.1  It 
tells  too  strange  a  tale,  mingles  elements  too  diverse,  perhaps, 
even  for  mediaeval  taste.  A  prince  turned  into  a  werwolf,  a 
Roman  princess  in  love  with  an  unknown  foundling,  lovers  liv- 
ing in  the  skins  of  white  bears  and  picnicing  on  provisions 
brought  by  the  kindly  werwolf,  battles,  enchantments,  extrava- 
gant emotions,  —  out  of  such  extraordinary  things  as  these  is  the 
story  wrought. 

The  original  version,  Guillaume  de  Palerne  (ed.  H.  Michelant, 
SATF.,  Paris,  1876),  is  a  French  romance  of  9663  lines  written 
about  the  end  of  the  twelfth  century  2  "  after  the  manner  of  the 
older  romantic  school  of  11 50  to  1180."  Impregnated  with  the 
doctrines  of  Vamour  courtois,  it  constantly  analyzes  the  emo- 
tions and  emphasizes  the  agonies  of  love-sickness  and  the  joys 
of  lovers  in  one  another's  company.  In  style  it  is  somewhat 
"  precieuse,"  verbally  prolix,  full  of  formal  speeches,  of  inter- 
minable digressions,  and  marked  by  occasional  allegorical  ten- 
dencies, especially  in  the  consideration  of  love  (Lot-Borodine, 
p.  264).  Though  all  this  is  incongruous  when  combined  with 
the  rapid  action  and  fabulous  incidents  of  a  typical  roman 
d'aventure,  the  style,  no  less  than  the  content,  was  probably 

1  Three  copies  of  the  French  version  were  listed  in  the  inventories  of 
1467  and  1487  of  the  libraries  of  the  Dukes  of  Burgundy  (Skeat,  p.  xiv). 
See  below,  note  4. 

2  Suggested  dates  for  the  poem  are  as  follows:  11 78-1 200,  Skeat,  p.  xvi; 
about  1 20s,  Paris,  La  Litt.  jrc.  au  Moyen  Age,  3c  ed.  p.  276;  1212-25, 
Zingarelli.  The  only  copy  of  the  poem  is  found  in  the  same  thirteenth 
century  manuscript  which  contains  L'Escoufle.  Warren,  p.  97,  pointed  out 
that  Jean  Renart,  the  author  of  L'Escoufle,  Lai  de  VOmbre,  and  Guillaume 
de  Dole,  should  not  be  considered  the  author  of  Guillaume  de  Palerne,  which 
is  different  in  style  and  versification  from  the  others.  On  Jean  Renart,  see 
also  Bedier,  Lai  de  VOmbre,  SATF.  1913,  p.  x  ff.  For  the  influence  on 
Guillaume  of  Cliges  see  Lot-Borodine,  p.  247. 

214 


WILLIAM    OF   PALERNE  215 

designed  with  special  reference  to  the  taste  of  that  "  boine 
dame,"  the  Countess  Yolande,  for  whom,  at  the  end  of  his  tale, 
the  author  says  he  translated  it  "  de  latin  en  roumans."  The 
idea  of  a  Latin  source  may  be  discounted  for  the  romance  as 
a  whole,  but  the  deliberate  suggestion  of  a  classical  origin,  like 
the  allusions  to  the  power  and  wealth  of  the  Greek  empire,  came 
from  the  poet's  wish  to  please  this  special  patron.  She  was  the 
aunt  by  marriage  of  that  Baldwin  VI,  count  of  Flanders  and 
Hainault,  who  was  elected  in  1204  Emperor  of  Constantinople. 
From  her  brother,  Baldwin  V,  Count  of  Hainault,  she  once  re- 
ceived, we  are  told,  the  Latin  manuscript  of  a  life  of  Charlemagne 
which  she  had  translated  (Michelant,  p.  x-xi).  These  facts  go 
far  to  explain  the  courtly  elegance  and  the  liking  for  literary 
sophistication  which  are  palpably  reflected  in  Guillaume  de 
Palerne. 

In  contrast  with  the  French  romance,  the  English  version  has 
in  style  at  least  a  fresh  and  almost  homely  air,  though  likewise 
it  was  the  result  of  noble  patronage.  It  was  translated  from 
the  French  by  one  William  (v.  5521)  at  the  order  of  Sir  Hum- 
phrey de  Bohun,  the  nephew  of  Edward  II,  for  "  ese  of  Englysch 
men."  (v.  165).  Humphrey,  who  succeeded  to  the  earldom  in 
1335  and  died  in  1361,  was  in  France  in  1349  and  in  1359,  and 
might  on  either  occasion  have  brought  the  French  romance  home 
with  him  (Skeat,  p.  ix-xii).  The  translation  of  this  into  the 
English  long-line,  alliterative  verse  synchronized  with  the  re- 
vival of  English  alliterative  poetry  about  the  middle  of  the 
century  in  the  West  Midland  district.  Despite  the  confusion  of 
dialect  forms  in  William,  this  region  is  generally  accepted  as 
the  original  home  of  the  poem.3 

The  Middle  English  version  is  now  extant  in  only  one  mid- 
fourteenth-century  manuscript,  fragmentary  at  the  beginning 
and  in  one  or  two  other  places  but,  even  so,  containing  5540 
lines.  The  lines  5047-5317,  in  this  text,  correspond  with  those 
in  the  prose  fragment  printed  presumably  about  1520-35  by 
Wynkyn  de  Worde.  This  fragment  constitutes  the  second  known 
English  version  of  William.  In  Brie's  opinion  a  prose  version 
in  English  intervened  between  the  two  texts;  this  lost  version 

3  In  "The  'West  Midland'  of  the  Romances,"  Mod.  Phil,  xix,  1-16 
(1921),  J.  R.  Hulbert  disputed  the  traditional  assignment  of  the  alliterative 
romances  to  the  West  Midland  district.    Cf.  Menner,  PMLA.  xxxvn  (1922). 


2i6  MEDIEVAL   ROMANCE   IN   ENGLAND 

was  simply  a  prose  redaction  of  the  Middle  English  poem,  and 
not  a  translation  of  the  French  prose  version.4 

Impossible  as  it  is  not  to  recognize  the  essential  incongruity 
in  a  story  which  attempts  to  combine  a  courtly  love  intrigue 
with  a  typical  roman  d'aventure,  there  is  much  in  William  of 
peculiarly  mediaeval  charm  and  picturesqueness.  Dainty  pic- 
tures linger  in  one's  mind  of  young  William  hiding  in  the  wide- 
branched  apple  tree  to  see  his  love;  of  Melior  laughing  out  of 
her  white  bear-skin  disguise,  "Am  I  nou3t  a  bold  best?";  or 
of  the  two  lovers,  now  disguised  as  a  hart  and  hind,  slipping 
through  the  moonlight,  to  hide  themselves  on  a  ship  sailing 
from  Reggio,  or  of  their  joy  together  "  under  a  louely  lorel  tre  " 
when  they  are  safe  in  the  Queen's  garden  at  Palermo.  It  is  a 
matter  of  course  that  in  such  a  tale  there  should  be  much  of 
chivalry,  of  wars  and  heroic  combats,  of  magnificent  feasts,  of 
gift-giving  to  minstrels,  etc.,  but  the  familiar  setting  serves  only 
to  enhance  the  unusual  elements  in  the  story  itself,  and,  in  the 
Middle  English  version  especially,  its  actual  charm.  Of  its  kind 
nothing  is  better  in  Middle  English  romance  than  that  scene 
(vv.  16-70)  in  which  the  baby  William,  stolen  by  the  watchful 
werwolf  and  hidden  in  its  den,  is  tempted  into  the  open  by  the 
grasses  and  flowers  blowing  in  the  sunlight.  The  cowherd  and 
his  wife,  even  the  dog  which  discovers  the  boy,  and,  later  in  the 
story  (v.  2520),  the  colliers  who  talk  together  outside  the  quarry 
where  the  lovers  lie  hidden,  have  a  humorous  realism  rare  in 
chivalric  story,  —  perhaps  only  to  be  matched  in  the  incom- 
parable Aucassin.  In  other  ways,  too,  in  the  feeling  for  nature,5 
in  the  constant  dwelling  on  the  youthful  beauty,  of  the  young 
hero  no  less  than  of  the  heroine,  in  the  gay  romantic  adventur- 
ousness  of  spirit,  there  is  much  that  may  be  compared  in  the 
two  stories.    In  these  respects  the  simpler  naturalism  of  the  Eng- 

4  Brie  (p.  322)  remarked  that  thirty-five  English  prose  romances  in  addi- 
tion to  William  were  composed  between  the  years  iSSo-JSSo-  The  French 
prose  version  of  Guillaume  was  apparently  composed  in  the  fifteenth  century 
and  was  printed  at  Paris  by  Bonfons  (no  date),  at  Lyons,  1552,  at  Rouen, 
1620,  and  again  about  1634.  Skeat,  p.  xvii,  on  the  basis  of  an  acrostic 
signature,  ascribed  the  authorship  of  this  French  prose  version  to  Pierre 
Durand. 

5  For  studies  on  Nature  in  Middle  English  see  Weichardt,  Die  Entwicklung 
des  Naturgefuhls  in  der  me.  Dichtung  vor  Chaucer  (einschliesslich  des 
Gawain-Dichters) ,  Kiel,  1900. 


WILLIAM    OF   PALERNE  217 

lish  version  of  William  makes  it  gain  rather  than  lose  by  com- 
parison with  the  French  Guillaume. 

Origin.  The  fact  that  the  Middle  English  text  is  simply  a 
more  or  less  direct  translation  of  an  extant  French  original 
makes  the  latter  serve  as  a  point  of  departure  for  any  investi- 
gation of  the  origin  of  the  werwolf  legend.  However  much  dis- 
guised by  the  accretions  of  mediaeval  romantic  narrative,  by 
elaborate  setting,  by  combination  with  a  disproportionate  love 
story  with  which  it  can  originally  have  had  nothing  to  do,  the 
nucleus  of  Guillaume  is  that  story  which  centers  round  its  true 
hero,  Alphonse,  the  prince  of  Spain,  who  was  changed  in  his 
childhood,  through  the  magical  arts  of  a  witch-like  stepmother, 
into  a  werwolf.  His  adventures,  his  revenge,  his  ultimate  re- 
covery of  his  human  form,  constitute  the  material  for  an  inde- 
pendent story  of  a  type  paralleled  in  several  other  Old  French 
narratives.  Without  the  werwolf,  the  adventures  of  Guillaume 
and  the  Princess  Melior  would  fall  into  nothingness.  It  is  the 
werwolf  who  saves  the  baby  William  from  the  conspirators  who 
would  poison  him;  it  is  the  werwolf  who  guides  and  protects 
the  escaping  lovers;  it  is  he  who  brings  about  Guillaume's  res- 
toration to  his  heritage.  He  is,  in  short,  as  Michelant  remarked, 
the  deus  ex  mackina  of  the  love  story. 

The  belief  in  werwolves  6  is  of  too  great  antiquity,  and  of  too 
universal  distribution,  to  give  any  indication  of  its  origin.  But 
in  general  in  werwolf  stories  certain  more  or  less  primitive  fea- 
tures may  be  discerned  (Tibbals,  p.  361).  The  most  primitive 
type  is  that  in  which  the  man  is  a  true  loup-garou,  and  it  is  a 
necessity  of  his  nature  to  become,  at  recurrent  intervals,  a  wolf 
among  wolves,  a  "  constitutional  werwolf."  In  this  type,  best 
represented  by  the  Lai  de  Bisclavret  (ed.  Warnke,  1900,  p.  75 
ff.),  of  Marie  de  France,  the  transformation  is  effected  simply 
by  the  removal  of  the  man's  clothes,  the  sign  of  his  civilized 
nature.  A  second  type  is  represented  by  those  stories  in  which 
the  transformation  is  effected  by  the  putting  on  of  a  wolf  skin, 
as  in  the  story  of  Sigmund  and  Sinfiotli  in  the  Volsunga  Saga; 

6  On  werwolves  in  general  see  the  many  references  given  by  Kittredge, 
p.  169,  n.  1,  also  p.  173,  n.  3,  pp.  257  ff.;  Warnke-Kohler,  Lais  der  Marie, 
p.  xcix  (1900);  Smith,  p.  1  ff.;  E.  O'Donnell,  Werwolves,  Boston,  1912;  C. 
T.  Stewart,  The  Origin  of  the  Werwolf  Superstition,  Univ.  of  Missouri,  Social 
Science  Series  II,  1909. 


2i8  MEDLEVAL   ROMANCE   IN   ENGLAND 

and  a  third,  the  least  primitive,  in  which  it  is  caused  by  external 
magic,  and  is  entirely  involuntary.  To  this  type  the  story  of 
Alphonse  belongs. 

The  group  of  werwolf  stories  discussed  by  Kittredge,  the  Lai 
de  Bisclavret,  the  anonymous  Lai  de  Melion,  preserved  in  a 
thirteenth-century  manuscript,  the  Latin  romance  of  Arthur  and 
Gorlagon,  preserved  in  a  fourteenth-century  manuscript,  and 
apparently  derived  through  some  lost  Welsh  intermediary  from 
some  old  Irish  tale  of  the  type  represented  by  the  still  current 
Irish  tale  of  Morraha  or  the  Quest  for  the  Sword  of  Light 
(Larminie,  West  Irish  Folk  Tales,  1893,  p.  10  ff.),  embodies  a 
form  of  the  werwolf's  tale  undoubtedly  older  than  that  con- 
tained in  Guillaume  de  Palerne.1  In  these  stories  the  en- 
chantress is  a  faithless  wife  who  wishes  to  dispose  of  her  hus- 
band. She  wheedles  from  him  the  secret  of  that  which  will 
transform  him,  his  clothes  in  Bisclavret,  a  rod  made  from  his 
life-tree  in  Gorlagon,  in  Melion,  a  magical  ring,  "  a  congenital 
talisman  like  the  necklaces  in  the  Knight  of  the  Swan  "  (Kit- 
tredge, p.  171).  After  the  husband  has  been  transformed,  he 
takes  refuge  with  a  king  whom  he  serves  with  the  intelligence 
of  a  man  and  the  fidelity  of  a  beast.  In  the  Irish  tale  he  is  made 
the  special  guardian  of  the  King's  child,  and  preserves  it  from 
harm.  After  some  years  of  tame  domesticity  he  is  roused  to 
wolf-like  fury  by  the  appearance  of  his  wife  and  her  lover  and 
violently  attacks  them.  The  king  protects  him  from  those  who 
would  kill  him  as  a  wild  beast,  and  forces  the  lady  and  her 
paramour  to  confess.  Through  his  royal  friend  or  through  his 
wife,  the  werwolf  is  then  disenchanted.  In  Bisclavret  and 
Melion  he  is  taken  into  a  private  room  and  the  actual  trans- 
formation takes  place.  Instead  of  being  condemned  to  death, 
the  wife  is  forgiven  on  condition  that  she  keep  silence  (Irish 

7  The  story  of  the  knight  Biclarel,  found  in  the  early  fourteenth-century 
Roman  du  Renard  Contrefait,  practically  duplicates  that  of  Bisclavret 
(Warnke-Kohler,  Lais  der  Marie,  pp.  xcix-ciii).  Malory,  Morte  Darthur,  bk. 
XIX,  ch.  11,  referred  to  "  Sir  Marrok  the  good  knyghte  that  was  betrayed 
with  his  wyf,  for  she  made  hym  seuen  yere  a  werwolf."  Among  other 
parallels  Kittredge,  p.  254  ff.,  noted  the  Middle  Dutch  Walewein  in  which 
a  king's  son  after  a  scene  resembling  the  "  Potiphar's  wife  incident "  in 
the  Seven  Sages  is  changed  into  a  fox  by  his  stepmother;  also  the  Icelandic 
Alafiekkssaga  in  which  Ali  on  his  wedding  night  is  transformed  by  a  wizard 
into  a  wolf  doomed  to  ravage  the  lands  of  his  own  father  and  not  to  escape 
death  unless  some  one  asks  pardon  for  him  when  he  is  taken. 


WILLIAM    OF   PALERNE  21g 

tale),  or  she  is  divorced  and  made  to  suffer  various  humiliating 
penances. 

The  False  Wife  story  in  all  these  versions  is  certainly  prior 
to  that  of  the  Cruel  Stepmother  variant  in  Guillaume  de  Palerne, 
though  in  other  details  this  romance  is  closely  akin  to  the  original 
werwolf  story.8  The  likeness  becomes  clearly  recognizable  after 
v.  7205  in  Guillaume  (v.  4012  in  the  English  version).  The  wer- 
wolf's obeisance  to  the  King  of  Spain,  whom  William  has  cap- 
tured and  who  is  in  reality  the  father  of  Alphonse,  corresponds 
to  a  similar  act  in  the  other  stories.9  The  order  of  events  in 
the  romance  has  been  shifted  but  it  is  probable  that  the  wer- 
wolf's' initial  protection  of  the  baby  William  (vv.  1-120)  rose 
from  some  confused  reminiscence  of  the  independent  anecdote 
known  as  The  Defence  of  the  Child  which  appears  in  the  vari- 
ous versions  of  the  Seven  Sages  and  the  Gesta  Romanorum  and 
in  a  number  of  Oriental  forms.  In  the  Irish  tale  of  Morraha 
the  werwolf  saves  a  child  from  a  monster;  in  Gorlagon  he  dis- 
covers the  child  after  the  doubly  false  wife  and  mother  has 
reported  its  death  (Kittredge,  p.  234  ff.).  In  Guillaume  no 
explanation  is  given  of  how  the  werwolf  knew  conspirators  were 
plotting  against  the  royal  boy,  yet  obviously  such  knowledge 
could  have  come  only  if  the  werwolf  had  been  domesticated,  as 
in  the  werwolf  story  proper,  at  the  King's  court.  The  episode 
has  been  changed  from  the  middle  to  the  beginning  of  the  ro- 
mance in  order  to  provide  for  the  introduction  of  the  long  ac- 
count of  the  lovers'  adventures,  in  all  of  which  the  werwolf  plays 
the  part  of  a  Helpful  Animal.10 

8  This  conclusion  rests  on  Kittredge's  study,  though  he  himself  barely 
mentioned  (p.  184,  n.  2)  Guillaume  de  Palerne.  G.  Paris,  Litterature  frg. 
§  67,  thought  the  werwolf  part  of  the  romance  was  directly  derived  from 
the  Lais  of  Biscldvret  and  Melton.  Ahlstrom's  refutation  of  this  point 
(Studier,  p.  81)  was  quoted  with  approval  by  Kohler,  p.  xiv.  The  literary 
derivation  may  be  questioned  without  destroying  the  probability  of  a  common 
source. 

9  The  werwolf  is  hunted  by  a  king  who  is  his  father-in-law  (Melion, 
Irish  tale),  his  brother  (Gorlagon) .  Being  hard  pressed  the  wolf  seizes  the 
king's  stirrup  and  licks  his  foot  (Gorlagon).  In  Melion  the  werwolf  escapes 
from  the  hunt  led  by  his  father-in-law,  but  later,  when  Arthur  comes  to 
visit  this  Irish  king,  the  werwolf  falls  at  Arthur's  feet.  In  this  indoors 
submission  Melion  is  nearer  to   Guillaume. 

10  It  is  probable  that  the  swimming  feats  of  the  werwolf  in  Guillaume, 
swimming  across  the  straits  of  Messina  with  the  child  William  in  his  mouth, 


220  MEDIEVAL   ROMANCE   IN   ENGLAND 

In  Guillaume  de  Palerne  the  Enchantress  Stepmother  is  sum- 
moned to  Palermo  by  her  captive  husband;  she  is  violently 
attacked  by  the  wolf,  and  finally  made  to  confess.  Guillaume 
acts  as  the  werwolf's  protector  when  those  about  would  kill 
him;  in  fact  Guillaume's  role  corresponds  exactly  with  that  of 
the  king  in  the  werwolf  tale.  The  scenes  of  the  werwolf's 
attack  upon  the  woman  who  has  injured  him,  of  his  protection, 
of  his  recovery  of  his  human  form,11  are  practically  the  same 
as  those  of  the  Old  French  stories.  Finally  it  may  be  noted  that 
in  this  romance,  as  in  the  Irish  tale,  the  enchantress  is  forgiven 
and  escapes,  as  indeed  she  does  in  all  the  tales,  the  death  penalty 
for  her  crimes.  In  Melion  and  to  some  extent  in  the  Irish  tale 
Kittredge  (p.  176)  thought  she  preserved  the  character  of  a 
Celtic  fee.  In  Guillaume  this  Queen  Brande,  who  lives  in  a 
rich  city  by  the  sea,  who  is  "  sage  a  merveille  et  bien  letree  " 
and  skilled  in  sorceries  (v.  287),  has  herself  such  dignity  and 
sweetness  of  address  (v.  7469),  that  one  wonders  if  the  concept 
has  not  been  touched  by  some  faint  memory  of  the  Circe  myth. 

Although  only  the  more  obvious  parallels  of  incident  and  char- 
acter have  been  touched  on  here,  it  seems  sufficiently  clear  that 
the  author  of  Guillaume  de  Palerne  was  familiar  with  the  par- 
ticular story  of  which  Bisclavret  and  Melion  were  independent 
derivatives.  If,  as  Kittredge  conjectured  (p.  262),  a  Norman 
text  of  Melion  preceded  the  extant  Picard  text,  we  can  the  more 
readily  account  for  the  means  of  its  transmission  to  the  Normans 
of  Sicily.  Casual  as  are  many  of  the  geographical  references  in 
Guillaume,  the  author  must  have  had  a  certain  knowledge  of 
the  southern  land  in  which  he  localized  his  story,  —  or  how 
explain  his  references  to  Reggio,  to  Cefalu,  to  Santa  Maria  della 
Scala  (v.  4636),  near  Palermo,  to  Palermo  itself?  (Zingarelli, 
p.  257  ff.).    Whether  Yolande's  poet  at  sometime  lived  in  Sicily 


v.  11S,  or  leaping  from  the  ship  on  which  he  had  embarked  with  the  lovers 
as  stowaways,  to  swim  to  land  (French,  4604  ff.;  English,  vv.  2728  ff.)  are 
derived  from  those  in  other  werwolf  stories.  So  in  Melion  the  wolf  gets 
passage  as  a  stowaway  to  Ireland;  in  the  Irish  tale  the  wolf  swims  back 
to  his  own  country. 

11  In  Guillaume  the  wolf  is  disenchanted  by  means  of  a  ring,  itself  proof 
against  all  magic,  which  his  stepmother  ties  about  his  neck.  So  also  in 
Melion  the  ring  is  the  instrument  of  disenchantment.  In  all  these  stories 
the  werwolf  resumes  his  human  shape  in  closest  privacy. 


WILLIAM    OF   PALERNE  22I 

and  found  there  his  story,  or  heard  or  read  it  in  some  brief 
form  that  was  carried  north,  we  cannot  say  with  any  certainty.12 
In  combination  with  this  fundamental  werwolf  story  that  of 
the  lovers,  Melior  and  Guillaume,  is,  as  has  been  said,  both 
disproportionate  and  incongruous.  On  analysis  it  reduces  to 
little  more  than  a  patchwork  of  familiar  motifs,  an  account 
of  a  supposed  mesalliance  between  a  princess  and  her  page,  of 
their  flight,  their  disguises,  and  their  final  restoration  to  their 
families  and  rightful  dignities.  The  allusions  to  Guillaume's 
mother  as  a  princess  of  Greece  and  to  one  of  Melior's  suitors  as 
a  prince  of  that  land,  are  the  result  of  contemporary  interests 
on  the  part  of  the  author  and  do  not  imply  connections  with 
Greek  romance.  The  mesalliance  theme,  familiar  in  Greek 
story,  was  well-known  in  western  story  by  the  end  of  the  twelfth 
century.  Many  western  lovers,  especially  among  the  Celts,  had 
learned  in  story  at  least  the  joys  of  a  Forest  Life,13  though  none 
it  would  seem,  save  Guillaume  and  Melior,  went  so  far  in  their 
return  to  Nature,  as  to  don  the  actual  skins  of  animals.  Many 
a  lady  had  had  a  confidante  though  few  with  the  wit  and  tact  of 
Alexandrine,  Melior's'  lively  friend.  The  rescuing  of  such  be- 
leagured  ladies  as  Guillaume's  mother  and  sister,  who  were  be- 
sieged in  Palermo  by  the  latter's  too  ardent  suitor,  was  a  well- 
established  chivalric  incident.  The  reunion  and  recognition  of 
long-separated  families,  in  this  instance  of  Guillaume  with  his 
mother,  from  whom  he  had  been  stolen  in  infancy,  and  of 
Alphonse,  who  had  by  his  transformation  into  a  werwolf  been 
thus  separated  from  his  father,  is  used  in  a  way  that  illustrates 
not  only  the  mediaeval  fondness  for  this  simple  theme  but  the 
author's  special  liking  for  doubling  what  he  conceives  to  be 
effective  material.  To  this  liking  may  be  ascribed  his  doubling 
of  the  motif  of  animal  disguise,  for  the  lovers,  having  eloped  in 
the  guise  of  white  bears,  are  represented  as  presently  changing 
to  that  of  a  hart  and  hind.  A  quaint  but  singularly  inept  touch 
is  added  when  the  poet,  still  too  much  in  love  with  this  device 

12  In  his  short  essay,  "La  Sicile  dans  la  litterature  franchise,"  Rom.  v. 
108-13  (1876),  Gaston  Paris  barely  mentioned  Guillaume  de  Palerne.  He 
considered  it  a  Celtic  tale  localized  in  Sicily. 

13  Cf.  the  adventures  of  Aucassin  and  Nicolette,  of  Beves  of  Hampton 
and  Josian,  and  chiefly,  of  course  those  of  Tristan  and  Isolt.  On  this 
romance  and  the  many  Celtic  parallels  see  Schoepperle,  Tristan,  n,  391-400. 


222  MEDIEVAL   ROMANCE    IN    ENGLAND 

to  abandon  it,  makes  the  Queen  of  Palermo,  Guillaume's  un- 
recognized mother,  don  a  hind's  skin  before  she  attempts  to 
communicate  with  her  son.  Awkward  too  is  the  poet's  borrow- 
ing of  the  incident  of  the  Faithful  Horse,14  for  in  Guillaume, 
though  the  hero  is  stolen  when  a  baby,  he  is  nevertheless  recog- 
nized by  his  father's  horse,  Brunsaudebruel,  when  years  later 
he  returned  to  Palermo  (v.  5405).  In  his  use  of  Marvellous 
Dreams  the  poet  likewise  follows  patterns  long  current  in  saint 
legends  and  romance.15  The  dream  of  William's  mother  that 
her  right  arm  stretches  over  Rome  and  her  left  over  Spain,  a 
prophecy  fulfilled  when  her  son  becomes  Emperor  of  Rome  and 
her  daughter  Queen  of  Spain,  is  particularly  close  to  that  of 
the  hero  in  the  English  version  of  Havelok  who  dreams  of  one 
arm  reaching  over  Denmark  and  the  other  over  England.  In- 
cidentally it  may  be  suggested  that  the  author,  in  his  account 
of  Guillaume's  devotion  to  the  werwolf,  and  of  the  exquisite 
politeness  of  the  latter  at  the  court  of  Palermo,  may  have  been 
influenced  to  some  extent  by  Chretien's  I  wain,  the  story  of  the 
famous  Chevalier  au  Lyon. 

This  enumeration  of  literary  influences  on  the  love  affair  in 
Guillaume  and  the  remarking  of  various  flaws  in  the  story  as 
a  whole  should  not,  however,  destroy  a  final  sense  that  the  ro- 
mance originated  in  a  spirit  pleasantly  touched  by  the  gracious 
charm  of  unreality.  The  pure  idyllic  mood  is  expressed  in  the 
happy  certainty  of  Guillaume,  vowing,  as  he  and  his  love  betake 
themselves  to  the  forest: 

"  Bien  viverons  de  nos  amors, 
D'erbes,  de  fuelles  et  de  flors."    (v.  3033-34) 

14  French  version,  v.  5405;  English,  v.  3225  ff.  More  aptly  in  the  various 
versions  of  Beves  of  Hampton,  the  horse  Arundel  recognizes  his  master  after 
seven  years'  absence.  For  other  instances  see  F.  Settegast,  Quellenstudien  z. 
gallo-rom.  Epik.  Leip.  1904,  p.  343;  Boje  (see  Beves  bibliog.  here)  p.  108-9; 
Deutschbein,  Sagengeschichte,  p.  196;  L.  Jordan,  Die  Sage  von  den  vier 
Haimonskindern,  pp.  93,  140,  Anm.  2. 

15  Cf.  Gerould  Saints  Legends,  1916,  p.  37;  Heyman,  Studies  in  the 
Havelok  Legend,  p.  107. 


WILLIAM    OF   PALERNE 


BIBLIOGRAPHY 


223 


Texts:  (i)  King's  College,  13,  Cambridge,  ed.  F.  Madden,  Rox- 
burghe  Club,  1832;  ed.  Skeat,  EETSES.  1,  Lond.  1867;  cf.  Kaluza, 
Eng.  Stud,  rv,  280-7.  (2)  Prose  fragment  probably  printed  by  Wynkyn 
de  Worde,  pr.  F.  Brie,  Archiv,  cxvn,  318-25  (1907).  Trans,  by  F. 
G.  Darton,  A  Wonder  Book  of  Old  Romance,  N.  Y.,  1907. 

Studies:  Billings,  Guide,  pp.  45-46;  Edwardes,  Summary,  p.  109  for 
English  version,  p.  192  for  French;  Wells,  Manual,  p.   765. 
Brie,  F.     See  Texts. 
Buxton,   E.   M.    Stories  from  Old  French   Romance    (Trans,   of   Gull- 

laume),  N.  Y.,  1910. 
Delp,  W.  E.  Etude  sur  la  langue  de  "  Guillaume  de  Paleme."     Suivie 

d'un  glossaire.  103  pp.  Paris,  1907;  rev.  Rom.  xxxvi,  448-50. 
Kaluza,    M.     "  Das    mittelengl.    Gedicht    William    of    Palerne    u.    seine 

frz.  Quelle,"  Eng.  Stud,  iv,  199-287    (1881). 
Kittredge,  G.  L.     "  Arthur  and  Gorlagon,"  Harvard  Studies  in  Philology 

and  Literature,  viii,  149-273   (1903).     Cf.  Revue  Crit.  1905,  lix, 

5-6;  Moyen  Age,  1904,  xvn,  66. 
Lot-Borodine,  M.     Le  Roman  Idyllique  au  Moyen  Age  (Guillaume  de 

Palerne,  pp.   237-65).   Paris.   1913. 
Skeat,  W.     See  Texts. 
Smith,  K.     "  Historical  Study  of  the  Werwolf  in  Literature,"  PMLA. 

ix,   1-41    (1894). 
Tibbals,    K.      "  Elements    of   Magic   in   the   Romance   of   '  William    of 

Palerne,'"  Mod.  Phil.  1,  355-71   (1903). 
Warren,  F.     "  The  Works  of  Jean  Renart  and  their  Relation  to  Galeran 

de  Bretagne,"  Parts  I  and  II,  MLN.  xxm,  69-73;  97-100  (1908). 
Zingarelli,  N.     "  II  '  Guillaume   de   Palerne '  e  i  suoi  dati  di  luogo  e 

di  tempo,"  Miscellanea  di  Archeologia  dedicata  al  Prof.  A.  Salinas, 

pp.  256-72.    Palermo,  1906.  Rev.  Rom.  xxxvi,  151   (1907). 


IPOMEDON 

Versions.  The  Anglo-French  romance  of  Ipomidon  (ed. 
Kolbing  und  Koschwitz,  Breslau,  1889)  is  one  of  the  compara- 
tively few  of  which  the  author  is  known.  Hue  de  Rotelande 
(Flintshire)  alludes  to  himself  in  several  passages  of  the  poem, 
—  in  one  confessing  genially: 

"  Hue  dit  ke  il  ni  ment  de  ren 
Fors  aukune  feiz  neent  mut." 

He  states  that  his  home  was  at  Credehulle  (Credenhill),  about 
four  miles  from  Hereford,  and  refers  not  only  to  bits  of  local 
history,  but  to  celebrities  of  that  district,  to  Huge  de  Hungrie, 
probably  a  canon  of  Hereford,  and  to  that  famous  worthy, 
Walter  Map,  Archdeacon  of  Oxford  in  1197,  to  whom  Hue  gives 
the  palm  in  the  "  art  de  mentir"  (Ward,  Catalogue,  1,  734). 
At  the  end  of  the  Prothesilaus,  a  metrical  sequel  to  Ipornedon, 
Hue  refers  to  his  patron,  Gilbert  Fitz-Baderon,  fourth  Lord  of 
Monmouth,  "  dount  sis  chastels  est  mult  manauntz  e  de  latyn  e 
de  romaunz,"  from  whose  library  Hue  asserts  that  he  received 
the  Latin  source  of  Prothesilaus.  Gilbert's  death  in  1 190-91 
gives,  therefore,  a  terminus  ad  quern  for  the  composition  of 
these  romances,  and  if  Hue's  reference  in  Ipornedon  to  the  raid 
of  the  Welsh  king,  Rhees  ap  Gryffth,  into  the  English  counties, 
be  that  of  11 86,  as  Carter  (p.  237)  thought,  the  poem  is  prob- 
ably to  be  dated  shortly  after  that  year.  The  text  of  Hue's 
poem,  now  extant  in  three  manuscripts,  contains  over  ten  thou- 
sand lines  and  shows  that  the  author  was  a  graceful  writer,  well- 
learned  in  the  taste  of  a  chivalric  audience,  and  skilful  in  adapt- 
ing material  of  proved  popularity.  A  certain  gaiety  of  tone,  as 
in  his  amused  reference  to  Map,  is,  perhaps,  his  special  dis- 
tinction. 

The  three  Middle  English  versions  of  Ipomedon's  story  (ed. 
Kolbing,  1889)  are  based  on  Hue's  romance.  The  longest  and 
most  important  text  is  in  the  fifteenth-century  Chetham  manu- 
script (A).    The  author  of  this  version  must  have  had  before  him 

224 


IPOMEDON  225 

one  of  the  French  manuscripts,  so  exactly  does  he  at  times  trans- 
late and  even  borrow  the  very  words  of  his  original.  Yet  his 
Ipomadon,  written  in  751  twelve-line  tail-rime  stanzas,  is  not  a 
slavish  but  a  fairly  independent  piece  of  work.  It  gives  good 
illustration,  as  Kolbing  (pp.  lxv-clvii)  pointed  out,  of  the  English 
redactor's  methods  of  work,  of  his  free  transposition  of  phrases 
and  details,  though  he  adhered  so  closely  to  the  narrative  content, 
of  his  original  (Kolbing,  p.  xxxvi  ff.)  This  first  version  Kolbing 
(p.  clxxiii)  thought  originated  in  north  Lancashire  about  the 
middle  of  the  fourteenth  century.  The  second  version  (B), 
The  Lyfe  of  Ipomydon,  is  a  poem  of  2346  verses  in  short  riming 
couplets.  It  is  written  in  the  fifteenth-century  Harleian  manu- 
script by  the  same  scribe  who  wrote  the  first  part  of  the  stanzaic 
Morte  Arthur.  On  linguistic  grounds  it  seems  improbable  that 
the  two  poems  could  have  been  by  the  same  author  (Seyferth, 
p.  76).  This  rather  pithily  condensed  version  of  Ipomedon  was 
perhaps  made  from  memory  of  Hue's  long  and  complicated  ro- 
mance (Kolbing,  p.  lxv).  The  third  version  (C)  of  Ipomedon 
is  in  prose.  Although  the  manuscript  in  which  it  is  contained 
seems  to  be  older  than  that  of  A,  the  text  itself  is  too  brief  to 
be  the  source  of  that  version.  On  the  other  hand  it  cannot  be 
simply  a  prose  rendering  of  A,  for  it  has  numerous  parallels,  not 
included  in  A,  with  the  French  version  (Kolbing,  p.  xlvi). 

The  three  Middle  English  versions  do  not,  it  must  be  ad- 
mitted, preserve  very  much  of  the  special  excellencies  of  Hue's 
humorous  and  leisurely  romance.  They  do,  however,  testify  to 
the  long  continued  liking  for  the  story  in  which  he  had  woven 
together  with  satisfying  effectiveness  two  well-established 
themes,  that  of  the  Three  Days'  Tournament  and  the  Rescue  of 
a  Beleaguered  Lady. 

Origin.  Hue's  introductory  assertion  that  he  translated 
Ipomedon  out  of  Latin  into  "  romaunz  "  is  generally  regarded  as 
a  hoax ;  so  also  is  his  statement  at  the  end  of  the  romance  that 
his  hero's  further  adventures  may  be  found  in  the  story  of 
Thebes.  Though  Warren's  arguments  "  On  the  Latin  Sources 
of  Thebes  and  £neas"  (PMLA.  xvi,  375-87)  have  increased 
the  belief  in  possible  mediaeval  Latin  versions  of  these  Romances 
of  Antiquity,  Hue's  own  borrowings  from  the  twelfth-century 


226  MEDIAEVAL   ROMANCE   IN   ENGLAND 

versions  in  French  make  the  supposition  of  a  Latin  Ipomedon 
wholly  unnecessary.  Ward  (Catalogue  i,  732)  thought  it  im- 
probable that  he  knew  anything  of  Statius's  Thebaid.  As  a 
matter  of  fact  a  comparison  of  Hue's  Prologue  with  those  of 
the  Roman  de  Thebes  and  the  Roman  de  Troie  more  than  ex- 
plains the  source  of  his  inspiration  about  "  Latin  "  sources. 

The  first  part  of  Ipomedon  tells  of  the  love  which  overcomes 
the  hero  when,  like  a  second  Rudel,1  he  hears  of  the  beauty  of 
La  Fiere,  the  young  duchess  of  Calabria,  of  his  service  as  her 
squire  until  he  is  driven  away  by  her  capricious  scorn,  and  of 
his  return  when  a  tournament  is  proclaimed  for  the  winning  of 
her  hand.  In  this  Three  Days'  Tournament  Ipomedon  appears 
disguised  in  white,  red,  and  black  armor  and  has  horses  to  match. 
At  each  day's  end  he  disappears  from  the  field  and  returns  to 
her  uncle's  court.  Here  he  pretends  that  he  has  spent  the  day 
in  hunting  and  the  court  ladies  laugh  him  to  scorn  for  his  un- 
chivalric  tastes.  In  far  more  primitive  form  this  episode  is 
found  in  numerous  folk  tales.2  From  them  it  passed  into  such 
romances  as  Chretien's  Cliges  (1160),  into  Robert  le  Diable  and 
its  English  version,  Sir  Gowther,  into  the  Lanzelet  of  the  poet, 
Ulrich  von  Zatzikhoven,  into  the  French  prose  Lancelot,  Parte- 
nopeus,  Richard  Coeur  de  Lion,  and  Roswall  and  Lillian.  The 
special  similarity  between  this  episode  in  Ipomedon  and  one  in 
Lanzelet  suggests  that  Hue  knew  the  original  of  Lanzelet  which 
its  author  states  was  a  French  romance  taken  from  England 
in  1 1 94  by  Hugh  de  Morville,  one  of  the  hostages  for  Richard 
I.  But  it  is  also  clear  that  Hue  must  have  known  some  form 
of  the  folk-tale  version  in  which  the  episode  is  regularly  con- 
nected, as  it  is  in  Ipomedon  but  not  in  Lanzelet,  with  the  win- 
ning of  a  princess.3     In   this  respect,   likewise,   Ipomedon  is 

1  O.  H.  Moore,  "  Jauf  re  Rudel  and  the  Lady  of  Dreams."  PMLA.  xxix, 
527-8  (1914),  gave  a  number  of  interesting  references  to  literary  examples 
of  the  use  of  the  idea  of  falling  in  love  from  hearsay  or  from  a  dream. 
See  Partonope,  n.  18. 

2  Cf.  Gowther,  note  2.  Carter,  p.  239,  cited  a  number  of  collections 
in  which  these  tales  are  found.  See  also  for  the  popular  and  the  literary 
versions',  Weston,  Three  Days'  Tournament,  pp.  34-43;   Ward,  Catalogue,  1, 

734- 

3  Carter,  p.  248.  See  F.  Lot,  Etude  sur  le  Lancelot  en  prose,  Paris,  1918, 
p.  166,  p.  128,  for  brief  discussions  of  Lanzelet,  and  of  the  improbability  that 
Map  was  the  author  of  its  source. 


IPOMEDON  227 

nearer  the  original  form  of  the  story  than  is  Cliges.  Though 
the  latter  is  the  earliest  extant  romance  to  make  use  of  the 
theme,  it  seems  improbable  that  it  was  the  source  of  this  inci- 
dent in  Ipomedon.  Cliges  tells  of  a  Four  Days'  Tournament; 
the  episode  is  an  incidental  chivalric  adventure  on  the  part  of 
the  hero  and  has  no  romantic  significance. 

The  second  part  of  Ipomedon  turns  on  motifs  well  known  in 
Arthurian  romance.  On  hearing  that  La  Fiere  is  wooed  by  a 
barbarous  suitor,  Ipomedon  returns  to  her  uncle's  court.  Here 
dressed  and  shorn  like  a  fool,  he  extorts  from  King  Meleager 
the  promise  of  the  first  feat  of  arms  that  shall  offer.  He  is 
assigned,  accordingly,  to  the  service  of  the  Maiden  Messenger 
who  comes  with  her  dwarf  to  ask  Meleager's  aid  for  his  niece. 
Despite  her  scorn  and  bitter  tongue,  the  hero  follows  her  and 
performs  prodigies  of  valor.  With  varying  details  the  essential 
elements  of  this  episode  are  also  found  in  the  Old  French  poem 
of  Renaud  de  Beaujeu,  Le  Bel  Inconnu  or  Guinglain,  written 
about  1 1 90  or  a  little  earlier,  and  in  the  later  versions  of  the 
same  story,  the  Middle  High  German  Wigalois  by  Wirnt  von 
Gravenberg,  c.  12 10,  the  Middle  English  Libeaus  Desconus,  c. 
1350,  and  in  Carduino,  the  Italian  poem  ascribed  to  Antonio 
Pucci,  c.  1375.4  It  is  also  in  Malory's  tale  of  Sir  Gareth  (Beau- 
mains),  the  seventh  book  of  the  Morte  Arthur.  Ipomedon  and 
Le  Bel  Inconnu  are  approximately  of  the  same  date  and  are,  in 
action  and  even  occasionally  in  phraseology,  closely  related.  For 
these  reasons  Carter  (pp.  255-61)  believed  that  for  this  inci- 
dent Hue  had  before  him  either  Renaud 's  own  poem  or  its  French 
source. 

Other  episodes  in  Ipomedon  are  similarly  suggestive  of  other 
romances.  The  Combat  between  Relatives,  here  between  the 
hero  and  his  half-brother,  Campaneus,  a  climax  skilfully  pre- 
pared for  throughout  the  story,  is  simply  a  variant  of  the  famil- 
iar Father  and  Son  Combat  of  epic  and  romantic  literature.5 
Suggestive  of  the  hero  of  Tristan  is  Ipomedon's  skill  at  skinning 

4  For  these  versions  see  W.  Schofield's  "  Studies  in  Libeaus  Desconus," 
Studies  and  Notes  in  Philology  and  Literature,  iv,  Boston,  1895. 

5  M.  A.  Potter,  Sohrab  and  Rustum  (The  Epic  Theme  of  a  Combat  be- 
tween Father  and  Son.  A  Study  of  its  Genesis  and  Use).  London,  1902. 
Potter,  p.  207,  enumerated  twenty-seven  instances  from  folklore  and  ro- 
mance of  combats  between  brothers. 


228  MEDIEVAL   ROMANCE    IN   ENGLAND 

game,  his  disguise  as  a  fool,  and  the  manner  of  his  jesting, 
The  courtly  ideals,  the  emphasis  on  the  courtesy,  generosity, 
beauty,  and  courage  of  the  hero,  the  splendor  of  the  court  of 
Meleager,  the  scenes  of  hunting  and  tourney,  the  maiden-dwarf 
episode,  the  analyses  of  love-sickness,  in  many  ways  seem  rem- 
iniscent of  Arthurian  romance  as  developed  in  the  hands  of 
Chretien  and  his  contemporaries.  But  the  absolute  lack  of  ref- 
erence to  any  Arthurian  personage  and  the  lack  too  of  that 
"  paraphernalia  of  wonder,"  of  which  romances  like  the  Charette 
or  Iwain  are  full,  make  it  doubtful  whether  Hue  had  any  knowl- 
edge at  all  of  these  romances.  Kolbing  (Ipomadon,  1889,  p. 
xxviii  ff.)  thought  the  author  did  know  them  but  later  studies 
on  the  connection  of  Ipomedon  with  the  Romances  of  Antiquity 
have  called  this  view  into  question.  It  has  been  pointed  out 
that  at  least  twenty  names  in  Ipomedon  are  drawn  from  the 
Roman  de  Thebes,  that  every  important  character  in  Ipomedon, 
the  active  king  Meleager,  who  is  very  unlike  the  passive  Arthur 
of  Chretien's  stories,  the  great  hero,  Campaneus,  supposedly 
modelled  on  Gawain,  but  unlike  Gawain,  defeated  by  the  hero 
of  the  story,  the  seneschal  Caeminius  with  his  Kay-like  "  cus- 
tumers  de  mesdire,"  can  be  paralleled  in  Thebes  or  Eneas. 
Moreover  the  virtues  of  character  and  the  nature  of  the  love- 
code  in  Ipomedon  are  of  distinctly  less  chivalric  cast  than  in 
the  Arthurian  tales  (Gay,  p.  469  ff.).  So  far,  indeed,  is  Ipome- 
don from  being  a  conventional  lover  that  he  can  remark,  even 
when  the  lady  is  more  than  ready  to  be  his,  "  De  femme  aveir 
ne  dei  haster  "  (v.  6644),  and  leave  her  pining  for  yet  another 
year.  In  all  this  he  is  much  more  like  an  Anglo-Norman  Have- 
lok  than  he  is  like  one  of  Chretien's  superfine  heroes.  Ipome- 
don's  zest  for  adventure,  which  at  the  end  induces  him  to  ap- 
pear in  the  guise  of  his  lady's  foe  and  fight  almost  to  the  death 
with  her  champion,  is  far  more  convincing  than  is  his  zeal  as 
a  lover. 

With  Partenopeus  of  Blots,  Ipomedon  has  some  likenesses 
which  prove  that  the  author  of  one  romance  must  have  bor- 
rowed from  the  other,  though  the  order  of  the  two  poems  has 
not  yet  been  established.  In  each  story  there  is  a  proud  heroine 
who  by  her  own  capricious  pride  drives  away  her  lover.  She 
ultimately  confesses  her  love  to  a  confidante  but  in  trying  to 


IPOMEDON  229 

say  her  lover's  name  breaks  down  in  the  middle  of  it.6  Kolbing 
(p.  xxxi)  likewise  called  attention  to  the  further  parallel  be- 
tween the  two  stories  in  the  account  of  the  Three  Days'  Tourna- 
ment. La  Fiere,  like  Melior  in  Partenopeus,  is  urged  by  her 
lords  to  marry  and  the  tournament  is  called  for  the  purpose  of 
deciding  on  a  suitable  husband.  In  tone,  length,  and  elabora- 
tion, the  two  romances  have  much  in  common. 

Like  Partenopeus,  Guillaume  de  Palerne,  Floire  et  Blanche- 
flor,  lpomedon  was  formerly  placed  among  the  mediaeval  French 
romances  inspired  by  Byzantine  sources  (Paris,  La  Litt.  frg.  au 
Moyen  Age,  ch.  in).  The  story  is  now  recognized  as  a  clever 
"  manufactured  "  romance  which  admirably  illustrates  the  so- 
phistication of  literary  art  in  England  at  the  end  of  the  twelfth 
century. 

BIBLIOGRAPHY 

Texts:  (i)  A,  Chetham  Library,  8009,  Manchester,  desc.  Eng.  Stud. 
vii,  195;  (2)  B,  Harleian,  2252,  desc.  Ward,  Catalogue,  1,  755  ff.;  pr. 
Wynkyn  de  Worde,  1500;  ed.  Weber,  1810,  11,  281  ff.;  (3)  C,  MS.  25 
of  Marquis  of  Bath,  desc.  Kolbing,  Eng.  Stud,  x,  203.  The  MSS.  ABC 
were  edited  by  Kolbing,  Breslau,  1889.  Rev.  Eng.  Stud,  xiii,  482-93. 

Studies:   Cf.  Wells,  Manual,  p.  785-86. 

Carter,  C.  H.  "  lpomedon,  An  Illustration  of  Romance  Origins,"  Haver- 
ford  Essays,  pp.  239-70,  Haverford,  1909.  This  was  part  of  an 
unpublished  dissertation,  lpomedon:  A  Study  of  the  Poem,  Harvard, 
1904. 

Gay,  L.  M.  "  Hue  de  Roteland's  lpomedon  and  Chretien  de  Troyes," 
PMLA.  xxxii,  468-92   (191 7). 

Hahn,  W.  Der  Wortschatz  des  Dichters  Hue  de  Rotelande.  Diss.  Berlin, 
1910. 

Holthausen,  F.  "  Zu  mittelengl.  Romanzen,  Ipomadon,"  Anglia  xli, 
463-97    (1917). 

Kittredge,  G.  L.  "  Anmerkungen  zum  mittelengl.  Ipomadon,"  Eng. 
Stud,  xiv,  386-92    (1890). 

Kolbing,  E.     See  Texts. 

Koppel,  E.     "  Zur  Textkritik  des  Ipomadon"  Eng.  Stud,  xrv,  371-86. 

Otto,  S.    Der  Einfluss  des  "  Roman  de  Thebes  "  auf  die  altfrz.  Litera- 


6  Ipomadon,  v.  144°;  Eng.  version,  Partonope,  v.  5782.  Miss  Gay,  p. 
486,  n.  91,  noted  that  Hue  more  probably  borrowed  this  episode  (v.  1497) 
from  the  account  of  Lavinia  in  Eneas  (v.  8554)  than  from  ParUnopeus  (v. 
7247).     Cf.  Partonope  here,  n.  25. 


230 


MEDIAEVAL   ROMANCE    IN   ENGLAND 


ture.  Diss.  Gottingen,   1909.     For  the  influence  on  Ipomedon,  see 

p.  62  ff. 
Seyferth,  P.    Sprache  u.  Metrik  des  mittelengl.  strophisGhen  Gedichtes 

"  Le  Morte  Arthur  "  u.  sein  Verhdltnis  zu  "  The  Lyje  of  Ipomydon." 

Diss.  Berlin,  1895.  Berlinger  Beitrdge,  vin. 
Ward,  H.  L.     Catalogue  of  Romances,  1,  728-750  (1883). 
Weston,  J.     The  Three  Days'  Tournament,  Lond.,  1902. 
Willert,  H.    "  Ipomadon,  Strophe  37,"  Eng.  Stud,  xxxvm,  131-32  (1907). 


GENERIDES 

Versions.  In  the  Parlement  of  Thre  Ages  1  and  in  Gower's 
forty-third  Balade  {Works,  ed.  Macaulay,  Oxford,  1899,  1,  372) 
the  authors  list  with  other  great  lovers,  "  Genarid  and  Clarionas." 
Though  to  the  taste  of  a  later  time  the  two  wear  but  borrowed 
robes,  the  shreds  and  patches  of  more  authentic  personalities, 
the  literary  references  to  them  and  the  two  extant  fifteenth- 
century  texts  of  Generides  imply  a  fair  degree  of  popularity. 
That  this  lasted  into  the  next  century  is  shown  by  the  fact  that 
the  book  was  licensed  to  Thomas  Purfoote  in  1568-69,  and 
printed  in  an  edition  of  which  a  few  fragments  are  still  extant 
(Wright,  p.  vii). 

The  original  version  of  Generides,  whether  in  French  or 
Middle  English,  was  probably  a  fourteenth-century  compilation 
for  which  the  author  used  chiefly  the  French  texts  of  current 
romances.  In  the  A  version  of  Generides,  represented  by  the 
Helmingham  manuscript,  the  rimes,  the  vocabulary,  the  whole 
style  and  spirit  of  the  piece,  show  this  specifically  French  in- 
fluence. The  reference  in  the  Prologue  to  a  Latin  version  of 
Generides  made  by  a  monk  at  Hertford,  may  best  be  taken  as 
an  imitation  of  a  similar  assertion  in  the  French  Ipomedon,  a 
romance  with  which  Generides  has  several  connections.  The  A 
version  in  rimed  couplets  is  over  three  thousand  lines  longer 
than  the  B  version  (Trinity  College  MS.  Camb.)  in  rime  royal. 
Howe's  study  (pp.  46-97)  of  their  relationship  showed  a  num- 
ber of  small  points  peculiar  to  each  version,  a  good  deal  of 
variation  in  the  spelling  of  names,  and  the  absence  of  much  cor- 
respondence in  rime.     He  concluded  that  the  two  texts  repre- 

1  Gollancz  in  his  edition  of  the  poem  (Lond.,  1915)  dated  the  Parlement 
about  1350.  As  the  Middle  English  texts  of  Generides  bear  every  sign  of 
late  fourteenth  or  early  fifteenth  century  work,  so  that  Kolbing  even  sus- 
pected in  them  signs  of  Chaucerian  influence,  it  must  be  supposed  that  the 
allusion  in  the  Parlement  was  to  the  lost  French  version  of  Generides. 
Among  other  lovers  cited  in  the  Parlement  list  are  Amadase  and  Idoyne, 
Ipomadoun,  Eglamour,  Tristram,  and  Dame  Gaynore. 

231 


2 32  MEDIAEVAL   ROMANCE   IN   ENGLAND 

sented  entirely  independent  versions  of  the  original  story.  In 
general  he  thought  that  the  A  version  kept  closer  to  its  sources 
and  had  an  ampler,  less  flattened  style  than  the  B  text,  though 
in  this  last  he  found  signs  of  some  effort  towards  unity  and  pro- 
portion, and  a  sincere  if  uninspired  effort  to  follow  the  "  boke." 
Neither  version  really  escapes  mediocrity  and  it  is  clear  that 
the  original  author  was  more  than  willing  to  evade  any  respon- 
sibility for  either  novelty  or  invention. 

Origin.  The  romance  is  full  of  Eastern  names  and  locations, 
but  it  is  difficult  to  believe  that  these  have  any  more  significance 
than  the  names  drawn  from  Biblical 2  and  classical  sources. 
The  author  had  a  passion  for  both  personal  and  geographical 
names  but  it  does  not  appear  that  there  was  much  lore  behind 
his  usage.  As  Zirwer  (p.  18)  pointed  out,  the  poet  tells  of  an 
army's  sailing  from  Damascus  to  India.  The  confusion  and  un- 
reality of  the  Indian  and  Persian  names  in  Generides  seem  al- 
most enough  to  discredit  at  the  start  Settegast's  attempts  to  find 
the  origin  of  the  story  in  ancient  Eastern  legends. 

In  his  Quellenstudien  (pp.  245-47),  Settegast  argued  that  the 
initial  story  concerning  the  birth  of  Generides  should  be  derived 
through  various  unstated  intermediaries  from  the  sixth  ( ?)  cen- 
tury drama,  Sakuntala,  which  the  poet  Kalidasa  based  on  an 
opening  episode  of  the  Mahabharata,  the  national  epic  of  India. 
In  this  it  is  told  how  the  king,  Dushyanta,  when  hunting  through 
the  forest,  comes  upon  a  place  of  great  loveliness  where  he  finds 
and  wins  a  beautiful  maiden  who  was  living  under  the  care  of 
her  holy  hermit  father.  By  her  Dushyanta  becomes  the  father 
of  the  child  of  prophecy,  the  wondrous  Bharata.  The  meeting 
and  separation  of  the  lovers,  the  king's  forgetfulness,  the  woman's 
quest,  the  token  which  recalls  her  to  him,  are  to  some  extent 
parallel  incidents  in  the  two  stories,  though  there  is  nothing  in 
Generides  to  suggest  either  the  many  specifically  Indian  features 
of  its  supposed  original  or  its  characteristic  motivation.  Sette- 
gast's attempt  (pp.  258-66)  to  derive  such  names  as  Sereyne 
from  Selene,  or  Auferius  from  that  of  Avelius,  or  to  equate  the 

2  Among  the  Biblical  names  given  for  various  warring  kings  in  Gen- 
erides may  be  noted  Abell,  Balam,  David,  Galad  (Cf.  Bruce,  MLN.,  1918; 
Pauphilet,  Queste  del  Saint  Gral,  Paris,  1921,  p.  136),  Jonathas,  Ishmaell, 
Reuben,  Samson,  etc. 


GENERIDES  233 

name  of  Nathanel  with  that  of  Matali,  the  charioteer  of  the  god 
Indra,  weakens  rather  than  supports  a  parallel  that  is  interesting 
even  if  unconvincing.  In  his  opinion  the  further  stages  of  the 
story  were  as  follows. 

The  Indian  story  concerning  the  birth  of  a  marvellous  hero 
was  fused  in  Persia  with  the  stories  of  the  Persian  heroes 
Sijawusch  and  Bischen,  whose  fame  was  recorded  about  ion  by 
the  poet  Firdausi  in  the  Persian  Book  of  Kings,  the  Shahnama 
(Eng.  trans,  by  A.  G.  and  F.  Warner,  Lond.  1906,  vol.  II,  200). 
Like  Generides  Sijawusch  had  an  unfortunate  encounter  with  a 
stepmother  who  first  proffered  him  her  love  and  then  accused 
him  of  evil  intentions  against  her.3  He  went  to  the  court  of  a 
foreign  king,  fell  in  love  with  the  king's  daughter,  and  there  in- 
curred the  hostility  of  the  king  (Warner,  n,  200-32).  The  tragi- 
cal outcome  of  this  last  adventure,  in  which  Sijawusch  lost  his 
life  (Warner,  11,  296-323),  was  changed  to  a  happy  ending  in  the 
case  of  Bischen  who,  having  become  the  lover  of  a  princess  and 
having  been  betrayed  to  her  father,  was  rescued  from  his  prison 
pit  and  lived  to  marry  his  princess.  The  story  in  which  these 
episodes  and  characters  had  merged  drifted  then,  according  to 
Settegast  (pp.  236-41),  to  Syria,  where  it  absorbed  some  geo- 
graphical names,  and  thence  to  Constantinople,  where  it  took 
over  certain  elements  from  the  legendary  history  of  the  Emperor 
Zeno.  Between  475-77  the  Emperor  had  been  driven  from  his 
throne  by  a  conspiracy  of  his  wicked  stepmother  and  her  lover ; 
he  fled  and  in  company  with  his  faithful  wife,  waited  until  he 
could  gather  an  army  for  the  conquest  of  his  land.  Settegast 
(p.  238)  identified  Aufreus,  the  father  of  Generides,  with  Zeno; 
Sereyne,  the  mother  of  the  hero,  with  Zeno's  true  wife,  Ariadne ; 
and  Aufreus's  false  wife  with  Zeno's  wicked  stepmother.  In 
what  fashion  this  strange  composite  of  stories  thus  drawn  from 
Indian,  Persian,  and  Byzantine  tradition  came  to  western  Europe, 
Settegast  did  not  say,  but  in  accord  with  other  critics  he  agreed 
that  the  extant  English  versions  must  have  been  derived  from 
a  lost  French  original. 

As  a  matter  of  fact  it  needs  no  such  far-reaching  efforts  to 

3  Cf.  Warner,  op  cit.,  n,  212,  and  the  Seven  Sages  here.  The  motif  is 
commonly  known  as  that  of  Potiphar's'  Wife,  and  appears  in  such  medi- 
aeval romances  as  Graelent  and  Launfal. 


234  MEDIEVAL   ROMANCE    IN    ENGLAND 

account  for  most  of  the  characters  and  episodes  in  Generides. 
They  may  be  tracked,  as  Howe  has  shown,  in  familiar  paths  of 
mediaeval  romance. 

The  opening  episode  of  the  story  tells  of  the  magic  hart  which 
leads  King  Aufreus  apart  from  his  fellows  and  brings  him  to 
his  predestined  mistress.  There  is  much  here  to  suggest  direct 
borrowing  from  the  lay  of  Guigemar,  though  Pschmadt's  study 
(Die  Sage  von  der  verfolgten  Hind,  191 1)  and  Howe's  on  The 
Magic  Hunt  prove  how  commonly  this  motif  was  used  as  an  in- 
troductory episode  in  romance.  In  Generides  the  lady's  com- 
panion is  one  of  the  Seven  Sages,  and  it  is  he  who  prophesies 
concerning  the  child  she  will  bear.  In  Guigemar  the  lady  is 
married  to  a  cruel  old  Jaloux;  in  Generides  it  is  Auferius  who 
is  already  married  to  a  wife  both  false  and  cruel.  In  order  to 
explain  his  liaison  with  Sereyne,  it  is  not  necessary  to  turn  with 
Settegast  to  the  multiple  marriages  of  the  East,  nor,  with  the 
mediaeval  doctrines  of  courtly  love  in  mind,  to  explain  the  lady's 
claim  as  due  originally  to  the  peculiar  marriage  celebrated  be- 
tween Sakuntala  and  Dushyanta.  The  cheerfully  unmoral  little 
episode,  as  in  Guigemar,  is  "  moralized,"  if  at  all,  by  the  char- 
acter of  the  lovers.  The  decorated  room  and  bed,  the  magic 
pillow,  the  ivory  gates  of  the  beautiful  house,  in  the  A  version 
of  Generides,  suggest  deliberate  imitation  on  the  author's 
part  of  the  lay  (Howe,  pp.  310-55).  That  the  original  author 
of  Generides  was  at  this  point  also  influenced  by  the  Erec  of 
Chretien  de  Troyes  (ed.  Foerster,  1890,  v.  2490),  is  suggested 
by  the  early  morning  scene  between  the  lovers.  Like  Erec, 
Auferius  is  awakened  by  his  lady's  tears. 

In  the  matter  of  a  Recognition  Token  between  the  lovers  there 
is  in  Generides  a  curious  variation  from  Guigemar.  In  the  latter 
the  lady  plaited  a  fold  in  her  lover's  tunic  in  such  fashion  that 
only  she  could  undo  it.  In  Generides  the  Sage  explains  (B,  v. 
194)  that  the  lady's  tears  can  only  be  washed  out  by  the  lady 
herself.  The  variation  is  sufficient  to  suggest  that  at  this  point 
the  poet  departed  from  the  lay  in  order  to  follow  some  folk-tale 
of  the  type  now  represented  by  the  Black  Bull  of  Norroway.* 

4  The  Scotch  version  was  first  printed  by  Robert  Chambers  in  1826  in 
The  Popular  Rhymes  of  Scotland  and  has  attained  a  new  vogue  in  Andrew 
Lang's  Blue  Fairy  Book,  Lond.  1901.  Chambers  gave  a  Scotch  and  an  English 
version  called   respectively   The  Black  and  the   Red  Bull  of  Norroway,  the 


GENERIDES  235 

Magic  shirts  are  common  enough  in  folk-lore,5  but  in  these  two 
stories  alone,  it  would  seem,  does  the  same  detail  occur  in  the 
same  situation.  In  both  folk-tale  and  romance  the  heroine,  who 
has  been  long  separated  from  her  lover,  comes  at  last  to  the 
place  where  he  rules ;  she  alone  can  wash  the  magic  shirt  free 
from  its  stain  of  blood  or  tears,  and  by  this  means  is  recog- 
nized by  him.  In  any  case  the  peculiar  liking  which  the  author 
of  Generides  had  for  the  laundress  6  role  is  worth  noting.  Howe 
(p.  354)  observed  that  the  poet  makes  a  "  lavendere  "  serve  in 
three  important  incidents,  for  it  was  a  lavendere  who  received 
the  new-born  Generides  from  his  mother's  confidante;  it  was 
as  a  lavendere  that  his  mother  regained  her  lord ;  and  it  was 
another  lavendere  who  helped  Generides'  own  lady  to  escape 
from  a  threatening  suitor.  It  may  be  added  that  almost  the 
only  touch  of  realistic  humor  in  the  romance  comes  in  this  last 
scene,  for  the  honest  woman  who  is  trying  to  disguise  Clarionas 
as  a  true  lavendere  complains  outspokenly  of  the  unseemly 
whiteness  of  the  princess's  legs  (B,  v.  4403). 

It  is  characteristic  of  the  discursive  manner  of  the  Generides 
poet  that  so  large  a  portion  of  his  story  is  devoted  not  to  the 
hero  but  to  his  father.  There  are  in  fact  three  love  stories  in 
the  romance,  that  of  Auferius  and  Sereyne,  of  Generides  and 


latter  an  abbreviated  and  somewhat  rationalized  version  of  the  former.  The 
Scotch  version  tells  of  a  maid  carried  off  by  the  Black  Bull,  an  enchanted 
Prince  of  Norroway,  of  their  separation  because  she  breaks  his  command, 
of  her  seven  years  of  service  for  a  smith  from  whom  she  finally  obtains  the 
iron  shoes  which  enable  her  to  climb  the  glass  mountain  on  the  top  of 
which  is  the  Prince's  kingdom.  There  she  finds  a  witch  woman  and  her 
daughter  who  vainly  try  to  wash  the  Prince's  blood-stained  shirt,  since  he 
has  said  he  will  marry  the  one  who  can  cleanse  it.  The  maiden  does  it 
herself  and  is  at  last  restored  to  her  lover. 

A  Gaelic  version  of  this  story,  found  in  Campbell's'  Popular  Tales  of  the 
West-Highlands,  iv,  267  ff.,  was  told  about  181 2  by  a  serving  maid.  In  this 
version  the  hero  is  transformed  into  a  great  grey  dog.  The  dog-hero  of 
this  version  confirms  the  evidence  of  the  name  Norroway  in  the  Scotch  ver- 
sion which  led  Leyden,  in  his  edition  of  The  Complaint  of  Scotland  (EETSES. 
1872,  xvii,  63),  to  believe  that  The  Taill  of  the  Thre  Futted  Dog  of  Norro- 
way, to  which  the  Complaint  refers,  was  this  story.  As  the  "  ballads " 
listed  in  the  Complaint  were  still  current  in  1548,  the  folk-tale  can  be  traced 
back  to  within  a  century  of  the  earliest  text  of  Generides. 

5  Cf.  L.  A.  Hibbard,  "  Chaucer's  '  Shapen  was  my  Sherte,'  "  Philological 
Quarterly,  1922,  1,  222-25. 

6  Cf.  G.  Krapp,  "  Chaucer's  Lavendere,"  MLN.  1902,  xvn,  102-03,  for 
other  uses  of  the  word. 


236  MEDIAEVAL   ROMANCE   IN   ENGLAND 

Clarionas,  and  of  Generides'  young  friend,  Darell,  and  Lycidas, 
the  daughter  of  the  hero's  cruel  stepmother.  In  the  love  story 
of  the  hero,  one  recognizes  the  predominant  influence  of  the 
Tristan  legend.  When  Clarionas,  the  young  daughter  of  the 
Sultan  to  whose  court  Generides  comes,  falls  in  love  with  him, 
"  of  hir  cupp  she  offeryd  him  to  drynk  "  (B.  v.  693).  Whether 
this  be  a  faint  reminiscence  of  the  immortal  potion  or  not,  it  is 
clear  that  the  jealous  steward  who  hides  in  a  tree  (B,  v.  1360), 
and  who  persuades  the  king  to  spy  upon  the  lovers  from  a  win- 
dow, is  a  stupid  descendant  of  Tristan's  foe.  Later  on  in  the 
romance,  the  poet  reverts  to  the  same  legend  in  telling  how 
Generides  rescued  his  love  from  an  abductor  and  made  for  her 
a  lodge  in  the  wood.  In  the  B  version  he  builds  one  for  himself 
also  and  sleeps  in  it  with  his  drawn  sword  by  his  side  (Howe, 
p.  153).  Inept  as  the  imitation  is,  it  is  evidently  the  grotto 
scene  of  the  Tristan  that  the  poet  has  in  mind.  Like  Mark,  the 
angry  Sultan  comes  upon  his  daughter,  and  then,  moved  by  pity, 
exchanges  swords  with  Generides  instead  of  killing  him.  Howe 
(p.  153)  noted  that  the  inconsistency  with  which  the  incident  is 
treated  in  B,  is  proof  of  a  different  authorship  from  that  of  A, 
in  which  the  author  followed  his  original  with  at  least  a  fair 
amount  of  understanding.  But  it  is  to  be  noted  that  the  B  ver- 
sion makes  four  different  references  (B,  v.  4736,  etc.)  to  the  dog 
which  Generides,  like  Tristan,  gave  to  his  lady.  Thus  in  this 
respect  the  B  version  is  closer  to  the  old  legend  than  the  A  text 
which  makes  no  allusion  at  all  to  such  a  gift.  A  final,  almost 
ludicrous  bit  of  similarity  between  the  romances  comes  in  the 
scene  in  which  Clarionas,  like  a  second  Isolt,  hastens  to  her 
lover,  who  has  fallen  desperately  ill  upon  hearing  of  the  death 
of  his  parents. 

Aside  from  the  Tristan,  Generides  follows  many  other  familiar 
patterns  of  romance.  The  lovers  are  warned  in  dreams  of  com- 
ing trouble ;  Generides  is  imprisoned  and  kills  the  False  Steward 
who  had  accused  him  of  dishonoring  the  princess ;  he  is  released 
in  order  to  fight  against  Belen,  the  Haughty  Suitor  who  de- 
manded the  homage  of  the  Sultan  and  the  hand  of  his  daughter ; 
he  kills  this  King  of  Kings  in  single  combat ;  he  rescues  Clarionas 
from  the  emissary  sent  by  Belen's  son,  who  assumed  his  father's 
throne  and  also  his  claim  to  the  lady;  Generides  follows  when 


GENERIDES  237 

again  she  is  carried  off;  he  crosses  the  sea,  and  in  the  disguise 
of  a  leprous  beggar,  wins  access  to  her.  In  this  episode  there 
is  some  likeness  to  the  Horn  story  where  similarly  at  her  wed- 
ding feast  a  bride  recognizes  her  true  lover  by  his  ring.  The 
disguise  motif  is  used  three  times  in  Generides,  and  in  each  in- 
stance seems  only  a  variant  of  this  original  (Howe,  p.  445). 
The  elopement  of  the  lovers,  who  find  their  horses  at  a  garden 
gate,  is  slightly  suggestive  of  Eliduc.  The  pursuit  by  Sir  Yuell, 
the  king's  emissary,  and  his  treacherous  attempt  to  kill  Gen- 
erides with  a  secret  knife  in  his  hand,  is  a  feeble  variant  of  the 
familiar  Vain  Pursuit  of  which  Boje  (Beuve  de  Hatntone,  p. 
96)  gave  so  many  examples.  The  lovers  separate  in  order  to 
permit  Generides  to  go  to  his  father's  aid,  and  their  reunion  is 
rendered  difficult  by  the  machinations  of  the  Wicked  Step- 
mother. She  sends  word  to  each  of  them  that  the  other  is 
faithless,  and  it  calls  for  much  travelling  back  and  forth  on  the 
part  of  their  faithful  friends  to  bring  them  together  again. 

Both  hero  and  heroine  in  Generides  are  provided  with  effi- 
cient friends.  Mirabell,  as  a  confidante,  serves  her  lady  much 
more  arduously  than  Alisandrine  does  Melior  in  William  of 
Palerne,  though  she  has  little  of  the  gaiety  or  pertness  that  often 
characterizes  such  a  character  in  Old  French  romance.  Like  the 
hero  in  Ipomedon,  Generides  is  given  into  the  hands  of  a  gov- 
ernor who  accompanies  him  upon  his  travels.  It  is  Nathanell 
who  listens  to  the  love-sick  hero,  who  arranges  with  Mirabell 
to  have  him  meet  Clarionas  in  the  garden,  who  takes  to  her 
the  steeds  won  by  the  hero  in  battle,  who  waits  with  horses  for 
the  eloping  lovers,  and  is  ultimately  rewarded  with  the  hand  of 
Mirabell.7  In  the  course  of  the  story  other  friends  are  provided 
for  Generides;  they  divide  the  functions  properly  belonging  to 
Nathanell.  Darell,  for  instance,  rushes  from  India  to  Persia  to 
find  out  the  true  state  of  his  lord's  lady,  and  takes  the  place  of 
Generides  when  the  hero  himself  feels  impelled  to  go  to  Persia. 
This  easy-going  doubling  of  roles  and  the  absence  of  any  real 
attempt  at  characterization  leave  the  personages  of  the  romance 
simply  type  figures  moving  in  a  puppet  show. 

7  Cf.  K.  Young,  Origin  and  Development  of  Troilus,  1908,  pp.  46-47,  on 
the  Faithful  Squire  or  Friend  in  Eglamour,  Ipomedon,  Libeaus  Desconus, 
Amadas  et  Idoine,  Cleomades,  etc. 


238  MEDIEVAL  ROMANCE   IN   ENGLAND 


BIBLIOGRAPHY 

Texts:  (i)  Helmingham  MS.,  ed.  Furnivall,  Roxburghe  Club,  1866; 
(2)  Cambridge,  Trinity  College  o,  5,  2,  ed.  Wright,  EETS.  55,  70  (1873- 
78). 

Studies:    Cf.  Korting,  Grundriss,  §116;  Wells,  Manual,  p.  785. 
Holthausen,  F.    Beitrage  zur  Textkritik  der  ME.  Generydes-romanze  (ed. 

Wright).    Goteborg,  1898. 
Howe,  W.  D.    Sir  Generides ;  its  Origin,  History,  and  Literary  Relations. 

Diss.    Unpublished.    Harvard  University,  1899. 
Kolbing,   E.     "  Zur  Textkritik  der   strophischen   dichtung   Generydes/' 

Eng.  Stud,  xvii,  49-73  (1892). 
Settegast,  F.     Quellenstudien  zur  galloromanischen  Epik.    Leipzig,  1904. 
Zirwer,  O.     (1)     "  Zur  Textkritik  der  ME.  Generides-Romanzen,"  Eng. 

Stud,  xvii,  23-49  (1892). 

(2)     Untersuchungen    zu    den    beiden    ME.    Generides-Romanzen. 

Diss.  Breslau,  1889. 


CHEVALERE   ASSIGNE 

(Knight  of  the  Swan) 

Versions.  (I,  The  Swan-Children.)  It  is  one  of  the  un- 
fortunate chances  of  mediaeval  English  literature  that  of  the 
many  accounts  of  the  legend  of  the  Knight  of  the  Swan  once 
current  in  England,  only  two  English  versions  are  now  known  to 
exist,  one,  the  short  poem  of  three  hundred  and  seventy  lines 
called  the  Chevelere  Assigne,  and  the  other  the  prose  version 
printed  in  1570  by  Wynkyn  de  Worde  and  called  by  him  The 
History  of  the  noble  Helyas,  Knight  of  the  Swanne.  The  text 
was  a  free  translation  made  by  Robert  Copland  from  the  com- 
pilation known  as  La  genealogie  avecques  les  gestes  .  .  .  du  tres 
preux  .  .  .  prince  Godeffroy  de  Boutin:  et  des  ses  freres  .  .  . 
Baudouin  et  Eustace:  yssus  et  descendus  de  la  tres  noble  .  .  . 
lignee  du  .  .  .  vertueux  Chevalier  au  Cygne.  This  book  was  pre- 
pared by  Pierre  d'Esrey  or  Desrey  of  Troyes,  and  the  earliest  ex- 
tant edition  was  printed  at  Paris  in  1504.  From  the  date  1499  in 
the  preface  it  is  probable  there  was  an  earlier  edition  and  there 
were  certainly  many  subsequent  ones  (Jaffray,  p.  68).  Of  de 
Worde's  edition  and  of  the  reprint  by  William  Copland,  son  of  the 
translator,  Robert,  only  single  copies  are  now  known.1  Similarly  a 
single  fifteenth-century  manuscript  contains  the  English  poetic 
version.  This  was  composed  in  the  last  half  of  the  fourteenth 
century  and,  as  its  alliterative  verse  would  suggest,  was  perhaps 
written  in  the  north-west.  The  brief  extant  text,  however,  is 
not  of  the  north,  but  was  written  by  an  East-Midland  scribe 

1  Gibbs,  p.  x,  wrongly  thought  that  there  was  no  extant  copy  of  the 
edition  by  Wynkyn  de  Worde.  The  copy  belonging  to  Robert  Hoe  was 
reprinted  in  a  beautiful  facsimile  edition  by  the  Grolier  Club  in  iqoi. 

The  propriety  of  including  this  romance  among  the  non-cyclic  legends  is 
open  to  question.  It  has  been  included,  however,  because  the  original  story 
of  the  Swan-Children  had  an  origin  and  early  history  independent  of  the 
Swan-Knight  legend,  and  because  the  English  poetic  version  preserves  this 
non-cyclic  character. 

239 


240  MEDIEVAL   ROMANCE   IN   ENGLAND 

(Kruger,  i,  p.  180).  Gibbs  (p.  i),  who  edited  the  poem,  and 
Jaffray  (p.  53)  thought  it  merely  an  epitome  of  the  first  1083 
lines  of  the  French  poem  in  that  wonderfully  illuminated  manu- 
script of  the  British  Museum,  Royal  15  E  VI,  which  is  itself  a 
whole  collection  of  romances.  Kruger  (1,  p.  175)  somewhat 
more  carefully  derived  it  from  the  same  source  as  a  Latin  ver- 
sion now  in  the  Bodleian,  Rawlinson  Misc.  358,  in  which,  as  in 
the  English  poem,  the  hero's  name  appears,  not  as  Helyas,  but 
as  Enyas  (v.  270). 

The  Middle  English  poem  gives  a  fairly  complete  and  not 
unspirited  account  of  the  events  that  had  come  to  be  the  in- 
troductory portion  of  the  Knight  of  the  Swan  story.  It  is  the 
legend  of  the  children  turned  into  swans  and  of  the  return  of 
ail  but  one  of  them  to  human  form.  The  oldest  literary  version 
of  the  story  is  found  in  the  Dolopathos  (ed.  Hilka,  1913,  p. 
80  ff.),  which  was  written  about  1190  by  Johannes  de  Alta  Silva 
and  translated  into  French  verse  not  long  after  by  the  poet 
Herbert.2  John's  statements  that  his  tales  were  "  adhuc  scrip- 
toribus  intacta  vel  forsitan  incognita,"  that  they  were  "  non  ut 
visa  sed  ut  audita"  (Hilka,  p.  107),  and  the  unquestionably 
primitive,  not  to  say  barbaric,  character  of  this  particular  story, 
the  seventh  in  the  collection,  do  not  preclude  the  probability 
that  it  came  to  him,  not  as  a  folk,  but  as  a  jongleur's  tale  of 
which  there  must  have  been  more  than  one  version  in  order 
to  explain  the  differences  between  the  later  vernacular  versions 
(Huet,  p.  208).  These  were  classified  in  a  memorable  article 
by  Gaston  Paris  (1,  p.  315)  in  four  groups,  each  determined  by 
the  name  of  the  mother  of  the  swan-children.  In  the  first 
group,  represented  by  the  Dolopathos  story,  all  the  characters 
are  unnamed ;  the  mother  is  a  nympha  whom  a  young  lord  finds 
bathing  in  a  fountain,  takes  home,  and  marries.  She  is  clearly  a 
swan-maiden,  but  her  way  of  reading  the  stars,  and  the  astrolog- 
ical prophecy  which  she  makes  concerning  her  future  offspring, 
as  Huet  (p.  207)  pointed  out,  indicate  a  semi-learned  adapta- 
tion of  some  early  but  literary  version.  When  her  seven  chil- 
dren are  born,  the  wicked  mother-in-law  substitutes  dogs  for 
them  and  gives  the  children  to  a  serf  to  kill.     Believing  his 

2  See  the  Seven  Sages  here.  For  bibliography  of  the  versions  of  the 
Swan-Children  story  see  Chauvin,  Bibliographie,  vm,  206-08;  Bolte-Polivka, 
Anmerkungen-Hausmarchen,  1913,  1,  427-34. 


CHEVALERE   ASSIGNE  241 

wife  to  be  an  abhorrent  creature,  the  husband  has  her  half- 
buried  in  the  court-yard  and  left  for  seven  years  to  endure  the 
chance  mercies  of  passers-by.  The  children,  who  were  spared 
by  the  serf  and  later  rescued  by  an  old  man,  are  discovered  by 
a  servant  of  the  mother-in-law.  When  the  servant  takes  the 
golden  chains,  which  had  been  about  their  necks  from  the  time 
they  were  born,  the  six  boys  turn  into  swans.  Their  sister  leads 
them  to  their  father's  house  where,  as  a  little  beggar  herself, 
she  provides  for  them  and  for  their  tortured  mother.  Their 
father's  pity  is  aroused,  the  true  story  is  revealed,  and  the 
magic  chains  are  restored,  with  the  exception  of  one  that  had 
been  partly  used  in  the  making  of  a  beautiful  cup.  In  conse- 
quence one  child  was  left  in  his  swan  form.  The  author's  con- 
cluding comment,  "  hie  est  cignus  de  quo  fama  in  eternum  per- 
severat,  quod  catena  aurea  militem  in  navicula  trahat  arma- 
tum"  {Rom.  xix,  317),  shows  that  in  his  time  the  tale  of  the 
Swan-Children  was  already  linked  with  that  of  the  Swan- 
Knight.  Even  clearer  indication  of  this  fact  is  given  in  the 
poet  Herbert's  translation  of  the  Dolopathos  (Blondeaux, 
xxxviii,  169). 

The  second  group  in  Paris's  classification  receives  its  name 
from  Eloixe,  the  heroine  of  a  long  French  poem,  La  Naissance 
du  Chevalier  an  Cygne  (ed.  H.  A.  Todd,  PMLA.  iv,  1889),  ex- 
tant in  two  thirteenth-century  manuscripts.  In  this,  3500  lines 
are  devoted  to  a  version  of  the  Swan-Children  story  which  Paris 
(1,  p.  319)  thought  was  probably  derived  from  the  same  source 
as  the  short  Dolopathos  tale.  But  much  that  was  primitive  in 
the  earlier  version  is  softened  in  this:  the  mother's  character 
as  a  fee  is  less  clearly  recognizable,  and  her  barbarous  torture 
is  omitted,  as  she  dies  in  giving  birth  to  the  children.  The 
author  of  this  version  was  familiar  with  the  romances  of  chiv- 
alry; he  tells  of  jongleurs  singing  of  Oliver  and  Ogier  (v.  3226- 
8),  and  contrasts  the  "  fable  d'Artu  .  .  .  co  fu  faerie  "  (v.  3296) 
with  his  own  veracious  tale,  an  "  estoire  "  which  was  found  at  a 
church  founded  in  honor  of  "  Sainte  Marie  "  at  Nimeque  (v. 
3301).  These  references  indicate  that  the  poem  was  not  written 
before  the  end  of  the  twelfth  century  (Paris,  1,  p.  320). 

The  two  remaining  forms  of  the  Swan-Children  story  were 
independently  derived  from  the  source  of  the  Dolopathos  ver- 


242  MEDIAEVAL   ROMANCE    IN    ENGLAND 

sion.  The  third  version,  Isomberte,  though  it  undoubtedly  ex- 
isted at  one  time  in  the  form  of  a  French  chanson  de  geste,  is 
now  known  only  in  a  Spanish  prose  chronicle,  La  Gran  Con- 
quista  de  Ultranter,  ch.  xlvii-lxviii  (cf.  Rom.  xrx,  320;  xvn, 
522).  It  opens  with  a  situation  suggestive  of  the  initial  motif 
of  the  Mannekin  group  of  stories  in  which  a  royal  princess  flees 
from  home  to  escape  a  hated  marriage.3  Isomberte,  rescued  by 
Count  Eustache  from  a  wood  to  which  she  had  fled,  marries 
him  and  in  his  absence  bears  him  seven  children  at  once.  By 
forged  letters,  again  a  device  that  belongs  to  the  Mannekin 
group  of  stories,  the  wicked  mother-in-law  contrives  to  order 
Isomberte  and  her  children  put  to  death.  The  children's  life 
with  the  hermit,  their  metamorphosis  into  swans,  is  much  the 
same  as  in  the  earlier  versions,  but  it  is  one  of  the  most  dis- 
tinctive changes  in  this  and  the  following  version,  that  the  child 
who  escapes  transformation  is  not  the  girl,  but  a  boy  who  can 
later  serve  as  his  mother's  champion. 

The  fourth  version,  Beatrix*  is  found  in  several  manuscripts 
of  the  Chevalier  au  Cygne  (ed.  Hippeau,  Paris,  1874-77),  to 
which  story  it  has  been  completely  joined.  Beatrix  opens  with 
a  scene  based  upon  an  idea  suggested  in  Isomberte.  After  the 
heroine  of  that  version  had  borne  several  children  at  one  birth, 
she  was  in  consequence  accused  of  adultery  5  and  condemned 
to  death  unless  she  could  find  a  champion  to  prove  her  inno- 
cence. In  Beatrix  this  idea  appears  at  the  beginning  of  the 
story  in  the  conversation  between  Beatrix  and  her  husband, 
King  Oriant.  A  beggar  woman,  the  mother  of  twins,  is  taunted 
by  Beatrix  with  being  an  unfaithful  wife.  The  taunt  is  re- 
peated by  her  own  jealous  mother-in-law,  Matabrune,  when  in 
time  Beatrix  herself  gives  birth  to  six  sons  and  one  daughter. 
The  story  then  follows  the  same  course  as  Isomberte  but  with 

8  Paris,  Rom.  xix,  323,  n.  4,  thought  that  in  the  type  story  the  heroine 
escapes  from  the  wooing  of  her  own  father.     See  Emarc  here,  notes  6,  16. 

4  It  was  this  version  chiefly  which  found  representation  in  art.  Cf.  the 
account  of  the  fourteenth-century  ivory  casket  showing  thirty-six  scenes 
from  the  romance  which  is  given  by  Gibbs,  Chevelere  Assigne,  p.  vii  ff. 
Cf.  Jahrbuch  d.  Kunsthistorischen  Sammlungen  des  allerhochsten  Kaiser- 
hauses,  xx,  265;  xxxiv,  68-69;  Jahrbuch  der  K.  K.  Centralcommission  f. 
Denknwlpflege,  1912,  Beilage,  p.  118;  F.  Bond,  Misericords,  for  other  refer- 
ences. 

5  See  Lay  le  Freine  here. 


CHEVALERE   ASSIGNE  243 

many  additional  details.  Important  among  these  are  the  name 
Helias,  given  to  the  boy,  who  is  represented  as  a  typical  forest- 
reared  youth,  of  the  "  Great  Fool  "  type,  amazed  at  horses,  armor, 
etc.,6  the  long  account  of  the  flight  and  ultimate  punishment  of 
Matabrune,  and  the  angel's  command  which  starts  Helias  in  a 
boat  drawn  by  his  untransformed  brother-swan  off  on  new  ad- 
ventures. The  author  of  this  version  claims  at  least  to  be  the 
first  to  tell  this  story  as  the  true  account  of  the  origin  of  the 
Chevalier  au  Cygne  (v.  28-31).  He  gives  an  elaborate  account 
of  the  judicial  combat  in  which  Helias  in  his  mother's  defense 
overcomes  Mauquarre,  and  of  his  subsequent  fight  with  Mau- 
quarre's  brother.  In  all  this  the  original  fairy  tale  element  gives 
place  to  heroic  and  epic  elements. 

From  the  names  Oryens,  Bewtrys,  Matabryne,  as  well  as  from 
the  context  of  the  story,  it  is  evident  that  the  Middle  English 
poem  belongs  to  this  fourth  version.  Especially  to  the  taste  of 
the  English  poet  were  the  scenes  of  violence  and  brutality. 
With  gusto  the  old  Queen  accuses  Bewtrys  and  with  zeal  the 
servants  "slongen  here  deepe  —  in  a  dymme  prysoun  "  (v.  86). 
Young  Enyas,  coming  to  act  as  his  mother's  champion,  when 
after  many  years  she  is  at  last  brought  to  the  stake,  has  his 
hair  torn  out  by  the  furious  Matabryne  ere  he  can  assure  her: 
"  Thy  hedde  shalle  lye  on  thy  lappe  for  thy  false  turnes." 
To  the  young  hero,  Malkedras  vows  that  he  cares  not  the  value 
of  a  cherry  for  the  sign  of  the  cross.  The  poem  is  wholly  with- 
out courtliness  or  chivalry  and  so  popular  in  manner  that  a 
ballad-like  phrase  such  as,  "  she  nykked  him  with  nay,"  comes 
easily  into  the  text.  The  incident  itself  of  the  small  "  chyle  " 
fighting  with  the  burly  Malkedras  is  of  the  sort  that  made  popu- 
lar the  various  versions  of  the  ballad  of  Sir  Aldingar.  The 
pietistic  element  is  strongly  marked :  God,  who  saved  Susannah 
"  fro  sorwefulle  domus,"  saves  also  Queen  Bewtrys  from  death ; 
a  hermit  and  a  hind,  "  whylle  our  lorde  wolde,"  care  together 
for  the  infants  exposed  in  the  forest ;  an  angel  follows  the  boy 
hero;  bells  ring  without  hands  (v.  272)  during  his  fight  for  his 
mother,  and  an  adder  springs  from  the  cross  that  Malkedras 

6  See  Paris,  Rom.  xix,  322,  n.  2,  on  the  imitation  of  the  beginning  of  the 
Perceval  story.  For  a  detailed  treatment  of  the  different  versions  see  R. 
Griffith,  Sir  Perceval  of  Galles  (Ch.  I,  The  Hero's  Forest  Rearing),  Chicago, 
1912.     Cf.  Bruce,  Evolution  Arthur.  Romance,  pp.  306,  310,  339. 


244  MEDIEVAL   ROMANCE    IN   ENGLAND 

derides.  Roughly  but  effectively,  too,  the  English  poet  intro- 
duces the  note  of  humor,  especially  in  the  comments  of  the 
forest-reared  boy  who  asks  about  his  horse  that  "  etethe  on  yren  " 
and  of  the  shield  that  bears  heavily  down  on  his  own  young 
neck.  At  the  end  the  poet  gives  a  genuine  touch  of  pathos  to 
the  account  of  the  five  swans  that  again  become  boys  when 
their  chains  are  restored  to  them,  and  of  the  sorrow  of  the 
sixth  whose  chain  had  been  partly  destroyed.  "  He  bote  hym 
self  with  his  bylle,"  we  are  told,  "  his  feyre  federes  fomede  upon 
blode,"  because  "  for  losse  of  his  cheyne  "  he  must  always  re- 
main a  swan.  Nothing  in  this  version  anticipates  the  service  he 
later  renders  the  brother  who  is,  however,  even  here,  christened 
"  Chevelere  assygne." 

Versions.  (II,  The  Knight  of  the  Swan.)  The  story  of 
the  knight  who  came,  conducted  by  a  swan,  to  the  assistance 
of  a  lady  whom  he  freed  from  persecution  by  killing  her  enemy 
in  a  judicial  combat,  of  his  marriage  with  her  or  her  daughter, 
of  his  departure  when  asked  a  tabooed  question  concerning  his 
origin,  and  of  the  descendants  of  this  marriage,  was  known  in 
two  forms  in  early  French  and  German  literature.  In  the 
French  versions  the  originally  unrelated  tale  of  the  Swan-Chil- 
dren was  added  to  that  of  the  Swan-Guided  Knight.  The  Swan- 
Guide  was  identified  with  the  youth  in  the  fairy  tale  who  did 
not  regain  his  human  form.  The  brother  whom  he  served  was 
called  Helias,  and  it  was  Helias  who  rescued  the  widowed 
Duchess  of  Bouillon,  married  her  daughter,  Ida,  and  through  her 
became  the  ancestor  of  the  famous  Crusader,  Godfrey  of  Bouillon. 
The  story,  which  thus  accounted  for  the  origin  of  Godfrey,  be- 
came the  introductory  portion,  in  the  order  of  events,  of  the 
group  of  stories  about  the  Crusades  of  which  Godfrey  was  pre- 
eminently the  hero.  This  group  or  cycle,  as  it  ultimately  became, 
had  five  branches  represented  by  ( i )  La  Naissance  du  Chevalier 
au  Cygne f  (2)  Le  Chevalier  au  Cygne  et  les  Enfances  Godefroi, 
(3)  the  Chanson  d'Antioche  (Ed.  P.  Paris,  1848),  in  part  prob- 
ably the  oldest  portion  of  the  whole  cycle  as  it  seems  originally 
to  have  been  composed  by  Richard  le  Pelerin  between  1125- 
1138,  (4)  Les  Chetifs,  a  story  of  five  knights  of  the  first  Cru- 
sade, of  their  capture  by  the  pagan  king,  Corbaran  d'Oliferne, 
and  of  their  escape,  and   (5)   the  Chanson  de  Jerusalem  (ed. 


CHEVALERE   ASSIGNE  245 

Hippeau,   1868),  the  story  of  the  capture  of  the  city  by  the 
Crusaders  in  1099. 7 

There  are  extant  nine  manuscripts 8  of  the  metrical  version 
of  the  Chevalier  au  Cygne,  of  which  none  are  earlier  than  the 
thirteenth  century.  From  the  Paris  Arsenal  manuscript,  which 
bears  the  date  1268  and  includes  all  the  five  branches,  it  is  evi- 
dent that  by  that  time  all  these  divisions  of  the  work  were  com- 
plete. The  oldest  French  version  of  the  Swan-Knight  story 
itself  was  contained,  according  to  Gaston  Paris  (2,  pp.  407-08), 
in  the  Chanson  d'Antioche  (cf.  Blondeaux,  xxxviii,  165;  Jaffray, 
p.  6).  This  amounts  to  not  much  more  than  a  mere  account  of 
the  mysterious  coming  and  departure  of  the  Swan-Knight  and 
resembles  a  similarly  brief  version  introduced  into  the  Karla- 
magnus  Saga  concerning  a  knight  there  known  as  Gerard  Cygne. 
The  legendary  hero  is  claimed  in  the  Chanson  d'Antioche  as  an 
ancestor  of  Godfrey's  in  order  to  match  the  claim  put  forth  by 
his  military  rival,  the  Duke  of  Normandy,  that  as  the  latter  is 
a  descendant  of  Doon  de  Mayence  his  lineage  is  superior  to 
Godfrey's,  and  his  right  to  be  the  champion  of  the  Crusaders  is 
therefore  greater.  Neither  in  this  nor  in  two  earlier  allusions  to 
the  story  that  are  known,  is  there  any  specific  mention  of  the 
name  taboo  or  of  the  famous  traditional  combat  of  the  Swan- 
Knight.  But  the  allusions,  brief  as  they  are,  and  the  number  of 
extant  manuscripts,  indicate  that  by  the  end  of  the  twelfth 
century  his  story  was  popular  and  well-known  in  Lorraine, 
though  it  is  difficult  to  say  which  text  is  nearest  to  the  original 
version.  The  fact  that  the  Berne  manuscript,  which  was  consid- 
ered the  oldest  text  until  Smith  and  Miss  Einstein  brought  the 
matter  into  question,  refers  to  an  earlier  "  estoire  "  preserved  at 
Mayence,  has  raised  the  difficult  problem  concerning  the  locali- 
zation of  the  story  in  that  supposed  original.  The  Berne  Manu- 
script itself,  the  German  poem  of  Lohengrin,  and  the  Chronicle 
of  the  Abbey  of  Brogne    (cir.   12 10)    agree  with   these   Paris 

7  Cf.  Blondeaux,  Revue  de  Belg.  xxxviii,  160;  Gautier,  Bibliog.  des 
Chansons  de  Geste,  1897,  pp.  77-81. 

8  Miss  Einstein,  pp.  725-27,  gave  the  most  complete  list  but  she  omitted 
Add.  MS.  36615.  Cf.  Brit.  Mus.  Cat.  of  Add.  MSS.,  1905-10,  p.  157.  In 
her  opinion  the  Enfances  Godefroi  was  written  by  a  layman  between  1161 
and  1 187.  The  author  of  the  Chevalier  au  Cygne,  either  a  monk  or  a  clerk, 
she  thought  must  have  written  between   11 70  and   1182. 


246  MEDIEVAL   ROMANCE    IN   ENGLAND 

manuscripts  in  localizing  the  arrival  and  combat  of  the  Swan- 
Knight  in  Mayence;  other  manuscripts  put  these  scenes  at 
Nimeque  (Kriiger,  Rom.  xxm,  449).  B16te-(2,  p.  414)  believed 
that  the  use  of  the  name  Mayence  was  occasional  and  due  to 
the  exigencies  of  rime,  but  Paris  (2,  p.  405)  argued  strongly 
that  Mayence  was  in  the  original  French  version  and  that 
Nimeque  was  a  variation  carelessly  introduced  into  a  manuscript 
which  gave  rise  to  the  use  of  the  name  in  later  texts.  In 
his  opinion  the  Brogne  chronicle  followed  a  lost  French  manu- 
script of  more  ancient  date  than  any  that  is  now  extant. 

Leaving  the  French  versions  for  the  moment,  we  may  note 
that  in  the  German  versions  the  oldest  is  the  Lohengrin  story, 
written  about  1205,  in  Wolfram  von  Eschenbach's  Parzival  (ed. 
E.  Martin,  Halle,  1900-03).  The  name  of  the  Swan-Knight, 
Loherangrin,  i.e.,  Loheraingarin,  Garin  le  Loherin,  Garin  of  Lor- 
raine, is  suggestive  of  the  locality  from  which  came  the  French 
tale  that  was  presumably  Wolfram's  source.  By  making  Loheran- 
grin the  son  of  Parzival,  the  king  of  the  Grail  Castle  at  Mon- 
salvaesch  from  which  the  young  hero  himself  departs,  Wolfram 
loosely  linked  the  legend  of  the  Swan-Knight  with  that  of  the 
Grail.  The  poet  tells  of  the  writings  on  the  Grail  that  bid 
Loherangrin  go  forth  to  champion  the  young  Duchess  of  Brabant, 
of  the  coming  of  the  Swan-boat  for  him,  of  his  arrival  at  Ant- 
werp, of  his  great  combat  with  the  lady's  foe,  of  Loherangrin's 
marriage  to  the  lady,  and  of  his  departure  when  she  asks  him 
the  tabooed  question  concerning  his  name  and  origin.  At  a 
later  date  in  the  same  century  were  written  Der  Schwanritter 
(ed.  F.  Roth,  1861)  by  Konrad  von  Wiirzburg,  and  the  anony- 
mous poem,  Lohengrin  (ed.  H.  Riickert,  1858),  which  follows 
closely,  though  with  a  good  deal  of  incidental  elaboration,  the 
version  by  Wolfram.  In  these  poems  the  incidents  of  the  story 
take  place  at  Antwerp,  Nimeque,  Cologne  and  Mainz,  the  per- 
secuted lady  is  the  Duchess  of  Brabant  or  her  daughter,  and 
by  wedding  the  young  woman  the  hero  becomes  duke.  In  Kon- 
rad's  poem  the  two  sons  of  this  marriage  are  said  to  have  been 
the  founders  of  the  houses  of  Cleves  and  Gueldres.  This  is  the 
earliest  reference  to  a  tradition  which  grew  steadily  in  political 
importance  (Blote,  1;  Jaffray,  ch.  vm).  In  Brabant  itself,  from 
the  fourteenth  century,  an  independent  version  of  the  legend  was 


CHEVALERE   ASSIGNE  247 

current  in  which  the  Swan-Knight  was  identified  with  Brabon 
Silvius,  the  first  legendary  duke  of  Brabant.  In  this  version  the 
hero  is  made  the  contemporary  of  Julius  Caesar  and  Octavian, 
is  represented  as  delivering  Antwerp  from  the  power  of  a  giant 
who  took  the  tribute  of  a  cut  hand  from  travellers,  and  as  fol- 
lowing a  swan  which  led  him  to  Valenciennes  ("  dats  Swanen- 
dale  in  dietsche  "),  where  he  received  the  hospitality  of  the  two 
Swanes,  mother  and  daughter,  and  returned  bringing  the  mother's 
gift  to  Octavian  and  receiving  from  him  the  right  to  marry  the 
maiden  (Cornicke  van  Brabant,  summarized  by  Blote,  7,  pp.  24- 
28).  In  general  this  story  shows  the  influence  of  rationalization 
and  a  very  incomplete  mastery  of  traditional  material.  Its  fif- 
teenth-century author,  Hennen  von  Merchtenen,  refers  to  the 
"  Clarasien  van  Jacop  van  Merlant  "  as  his  source  for  the  ac- 
count of  Brabon  (Breboen)  from  which  Brabant  received  its 
name.  Blote  (7,  p.  30  ff.),  thought  this  source  was  actually 
written  between  1320  and  1330  by  some  unknown  author.  Many 
subsequent  versions  of  the  Brabon  legend  are  known. 

The  popularity  of  the  Chevalier  au  Cygne  in  its  non-political 
forms  continued  all  through  the  thirteenth  and  fourteenth  cen- 
turies. In  the  thirteenth-century  Karlamagnus  Saga,  to  which 
reference  has  already  been  made,  the  legend  is  connected  with 
the  Emperor,  and  its  hero,  Gerard  Swan,  is  represented  as  the 
friend  of  Roland  and  the  husband  of  Charles's  sister.  In  this 
version  nothing  is  said  of  the  tabooed  question  and  the  knight 
is  hailed  as  coming  from  God  (Blondeaux,  xxxviii,  231).  In  an 
Icelandic  saga  of  later  date  the  hero  is  called  Helis  and  is  said  to 
be  the  son  of  Julius  Caesar  (Jaffray,  p.  104).  In  his  Speculum 
Naturale  Vincent  of  Beauvais  briefly  summarized  the  Swan- 
Knight  story,  localizing  it  at  Cologne  and  telling  simply  of  the 
coming  and  departure  of  the  knight.  In  neither  of  these  texts 
is  any  reference  made  to  the  Swan-Children  nor  is  any  explana- 
tion offered  for  the  swan  and  the  knight  he  brings  (Blondeaux, 
ibid.,  p.  164).  In  the  fifteenth  century,  according  to  Krliger 
(2,  p.  424),  was  compiled  the  huge  enlargement  of  the  story 
found  in  the  Brussels  manuscript  published  by  Baron  de  Reiffen- 
berg  {Monuments  pour  servir  a  VHistoire  des  Provinces  de 
Namur,  de  Hainaut,  et  de  Luxembourg,  1846-50,  vols,  iv,  v, 
vi),  and  known  also  in  a  manuscript  of  1469  now  at  Lyons. 


248  MEDIEVAL   ROMANCE    IN   ENGLAND 

This  version  was  later  refashioned  as  a  prose  romance.  Among 
the  best  known  of  these  prose  redactions  was  that  of  Pierre 
Desrey  of  Troyes. 

Origin.  The  fairy-tale  theme  of  the  Swan-Children  seems 
to  be  of  very  ancient  date  and  of  unlimited  extent.  The  ancient 
Celtic  tale  of  the  Children  of  Lir,  now  known  only  in  eighteenth- 
century  texts,  attributes  the  transformation  of  the  children  to 
the  hostile  magic  of  a  stepmother  who  is  jealous  of  their 
father's  love  for  them.  The  subsequent  account  of  the  suffer- 
ings of  the  swans  and  the  religious  elements  that  are  introduced 
into  the  story,  differentiate  it  entirely  from  that  form  in  which 
the  theme  appears  in  the  Dolopathos  account.  Lot  (p.  63), 
however,  pointed  out  that  the  divine  origin  of  the  Children  of 
Lir  explains  in  a  sense  the  semi-supernatural  nature  of  the 
mother  in  the  Dolopathos  version,  and  that  the  gold  or  silver 
chain  by  which  the  transformation  of  each  swan  is  accomplished, 
far  from  having  been  suggested  by  the  episode  in  the  Swan- 
Knight  legend  in  which  the  swan  draws  the  swan-boat  by  means 
of  a  precious  chain,  was  an  ancient  and  original  element  in  the 
story  of  the  Swan-Children.  He  paralleled  it  by  the  Serglige 
Conculaind  in  which  two  goddesses  appear  to  the  hero  under  the 
guise  of  swans  linked  by  a  chain  of  gold.  To  this  might  be 
added  the  account  in  the  Tochmarc  Etain  in  which  a  god-like 
pair  take  flight  as  swans  similarly  linked  together.  Other  in- 
stances have  been  cited  by  Blote  (Zts.  filr.  deut.  Altertum, 
xxxviii,  272)  and  by  Poisson  (p.  186)  who  believed  the  Old 
Irish  legend  of  divine  beings  transformed  into  swans  was  intro- 
duced during  the  Carolingian  period  into  the  Rhine  monasteries 
by  Irish  monks.  To  one  of  these  monks,  among  whom  the 
knowledge  of  Greek  was  preserved  even  through  the  Dark  Ages, 
Poisson  (p.  195)  also  attributed  the  naming  of  the  Swan-Knight. 
The  name  Helias,  he  thought,  came  from  that  of  the  prophet 
Elias  whose  name  and  story  had  absorbed  some  suggestions  from 
the  myth  of  the  Sun  God  Helios.  Dechelette's  evidence  {Revue 
Archeologique,  Paris,  1909,  1,  305)  that  in  certain  primitive  rep- 
resentations the  solar  disc  was  associated  with  a  boat,  suppos- 
edly that  which  conveyed  the  Sun  across  the  ocean  paths  from 
night  to  day,  convinced  Poisson  (p.  193)  that  the  Swan-Knight 


CHEVALERE   ASSIGNE  249 

legend  itself  came  from  a  dim  surviving  memory  of  the  old 
pagan  myth,  and  that  it  was,  like  the  story  of  the  Swan-Children, 
Celtic  in  origin.9 

Another  explanation  for  the  origin  of  the  Swan-Knight  story 
is  that  offered  by  Blote  (Zts.  f.  rom.  Phil,  xxi,  179).  He  found 
it  in  the  history  of  Roger  of  Toeni,  also  called  Roger  the 
Spaniard,  whose  legendary  emblem  was  a  swan,  who  rescued  the 
Countess  of  Barcelona  from  a  Moorish  assault  (cir.  1018)  and 
who  married  her  daughter.  Roger's  granddaughter,  Godehild 
of  Toeni,  married  Baldwin,  the  brother  of  Godfrey  of  Bouillon, 
and  Roger's  story  and  emblem  thus  became  associated  with  the 
famous  Crusader.  To  Gaston  Paris,  on  the  other  hand,  who 
was  impressed  by  the  gaps  and  assumptions  in  Blote's  argu- 
ment and  the  lateness  of  the  evidence  for  it,  the  Swan-Knight 
story  was  an  ancient  totemic  legend  of  Lorraine,  arbitrarily 
attached  in  the  twelfth  century  to  the  house  of  Bouillon  and 
only  later  connected  with  the  Toeni  family  (Rom.  xxvi,  581). 
The  local  legend  would  naturally  be  drawn  into  the  cycle  of 
stories  growing  up  about  one  who  was  not  only  Duke  of  Lower 
Lorraine  but  also  one  of  the  most  famous  men  of  his  time.  A 
possible  confusion  of  the  word  Signe  (Crusader's  cross)  with 
Cygne,  has  also  been  suggested  in  explanation  of  the  association 
of  the  legend  with  Godfrey  (Edwardes,  Summary,  p.  179). 

The  early  history  and  associations  of  the  Swan-Knight  legend, 
quite  aside  from  the  extant  literary  versions,  are  full  of  interest. 
The  first  allusion  to  the  story  is  in  a  Latin  letter  written  about 
1 1 70  by  Gui  de  Bazoches  in  which  Baldwin,  Godfrey's  brother, 
is  spoken  of  as  "  nepos  militis  ejus,  Per  vada  cui  Rheni  dux 
fuit  albus  olor."  10  About  twenty  years  later  William,  Arch- 
bishop of  Tyre,  in  his  history  of  the  first  Crusade,  referred 
sceptically  to  the  legend  which  made  the  Swan-Knight  an  an- 
cestor of  Godfrey  of   Bouillon   as  a  well-known   fable.     The 

9  A  different  and  even  less  credible  folkloristic  theory  was  given  by 
Pestalozzi,  p.  150,  who  believed  in  the  Germanic  origin  of  the  tale.  To  him 
the  Swan-knight  was  originally  a  demon  or  elf  who  must  disappear  upon 
being  asked  his  name;  he  was  also  originally  identical  with  the  swan,  since 
this  was  the  form  he  assumed  in  attempting  to  enter  the  world  of  men. 
For  the  later  development  of  this  rationalized  fairy  tale  Pestalozzi  accepted 
Blote's  theory  of  its  association  with  Roger  of  Toeni. 

10  Cf.  Paris,  Rom.  xxx,  406;  Blondeaux,  Revue  de  Belg.  xxxvm,  163; 
Blote  5,  pp.  185-91 ;  Jaffray,  p.  4. 


2  5o  MEDIEVAL   ROMANCE    IN    ENGLAND 

brevity  of  the  allusion  to  the  story  in  the  Dolopathos  has  been 
interpreted  as  showing  that  the  learned  author  had  a  similar 
awareness  of  this  fact  and  so  wished  to  avoid  the  tale  (Huet,  p. 
309).  The  popularity  of  the  legend  increased  with  its  associa- 
tion with  the  house  of  Bouillon  and  ultimately  resulted,  it  would 
seem,  in  the  attachment  of  the  story  to  still  other  families.  In 
England  the  legend  appeared  in  connection  with  the  Norman 
Radulf  of  Toeni,  son  of  Roger  the  Spaniard,  and  founder  of  the 
house  of  Stafford,  to  whose  descent  from  the  Swan-Knight,  in  a 
passage  in  his  Lives  of  the  Abbots,  Matthew  Paris  (1250)  made 
somewhat  belated  allusion  (Blote,  4,  p.  342).  The  marriage 
before  1125  of  Matilde,  niece  of  Godfrey  of  Bouillon,  to  Stephen 
of  Blois,  may  well  have  stimulated  English  interest  in  the 
traditional  legends  of  her  family.  Extracts  copied  in  the 
thirteenth  century  in  the  Red  Book  of  the  Exchequer  from  a 
monastic  register  of  Feversham  Abbey,  founded  by  Matilda  and 
Stephen  in  1148,  show  that  the  monastery  possessed  a  Liber  de 
Cigno  which  was  possibly  a  royal  gift  (Liebermann,  p.  106). 
In  the  fourteenth  and  fifteenth  centuries  the  Tony  family,  the 
Beauchamps,  the  Bohuns,  and  the  Staffords,  used  a  swan  as  a 
heraldic  device,  and  in  some  instances  claimed  descent  from  the 
Swan-Knight.  This  was  notably  true  of  that  Edward,  Duke  of 
Buckingham  and  Earl  of  Stafford,  "  linially  dyscended  of  — 
Helyas,  the  Knight  of  the  Swanne,"  at  whose  instigation  Cop- 
land made  his  version  of  the  story  of  Plelyas  (Blote,  4,  p.  349). 
On  the  continent  similar  claims  were  made  from  time  to  time. 
These  have  been  investigated  with  interesting  results  by  Blote 
(3).  Through  the  marriage  in  11 79  of  Matilde  of  Boulogne  to 
Heinrich  IV  of  Brabant,  the  legend  came  to  be  associated  with 
the  rulers  of  Brabant.  In  the  twelfth  century  the  old  title  of 
Duke  of  Lower  Lorraine  was  discarded  for  that  of  Duke  of 
Brabant,  and  this,  it  will  be  remembered,  was  the  title  given  to 
the  Swan-Knight  after  his  marriage  with  the  Duchess  of  Bra- 
bant in  Wolfram's  story.  Blote  has  distinguished  three  periods 
in  the  development  of  the  story  in  Brabant,  and  to  the  second 
of  these  belongs  the  period  of  its  rationalization.  The  earliest 
specific  reference  to  the  Swan-Knight  as  the  ancestor  of  the 
Duke  of  Brabant  seems  to  be  that  of  the  Flemish  poet,  Jacob 
van  Maerlent,  in  the  Spiegel  Historial  1286-90  (Blote,  7,  p.  3). 


CHEVALERE  ASSIGNE  251 

A  similar  claim  made  by  the  house  of  Cleves,11  which  has  been 
held  to  antedate  even  that  of  the  house  of  Brabant,  has  been 
shown  by  Blote  (1)  to  be  equally  void  of  mythological  signifi- 
cance or  of  real  antiquity. 


BIBLIOGRAPHY 

Texts:  (i)  Cotton  Caligula  A  II,  ed.  Utterson,  Roxburghe  Club, 
Lond.,  1820;  ed.  H.  H.  Gibbs,  EETSES.  vr,  1868.  (2)  Helios,  Knyght 
of  the  Swanne,  pr.  Wynkyn  de  Worde,  151 2;  repr.  from  copy  in  the 
library  of  Robert  Hoe,  Grolier  Club,  N.  Y.,  1901. 

Studies:   Cf.  Billings,  Guide,  pp.  228-29;  Edwardes,  Summary,  Index; 
Wells,  Manual,  p.  777. 
Blondeux,  F.     "  La  Legende  du  Chevalier  au  Cygne,"  Revue  de  Bel- 

gique,  Brussels,  1903:    (1)   "  Les  Debuts  de  la  Legende,"  xxxviii, 

158-76;  (11)  "Les  Versions  de  la  Legende,"  pp.  231-42;  (111)  "  Les 

Destinees  de  la  Legende,"  xxxrx,  40-49,  371-80. 
Blote,  J.     (1)  "  Das  Aufkommen  des  clevischen  Schwanritters,"  Zts.  /. 

deut.  Alter,  xlii,  1-53  (1898).    Rev.  G.  Paris,  Rom.  xxvn,  334-5. 

(2)  "  Die  Sage  vom  Schwanritter  in  der  Brogner  Chronik  von  c. 
121 1,"  Zts.  f.  deut.  Alter,  xliv,  407-20  (1900). 

(3)  "Der  historische  Schwanritter,"  Zts.  f.  rom.  Phil,  xxi,  176-91; 
xxv,  1-44  (189  7-1 901). 

(4)  "  Der  Ursprung  der  Schwanrittertradition  in  englischen  Adels- 
familien,"  Erig.  Stud,  xxix,  337-68  (1901). 

(5)  "  Der  Schwanritterpassus  in  einem  Brief  des  Guido  von 
Bazoches,"  Zts.  f.  deut.  Alter,  xlvii,  185-91   (1903). 

(6)  "  Mainz  in  der  Sage  vom  Schwanritter,"  Zts.  f.  rom.  Phil. 
xxvn,  1-24  (1903). 

(7)  Das  Aufkommen  der  Sage  von  Brabon  Silvius,  dem  brabantischen 
Schwanritter.  127  pp.  Amsterdam,  1904.  Rev.  Litter aturbl. 
xxvii,  col.  1-3  (1906). 

Chauvin,  V.     Bibliographie  des  Ouvrages  Arabes.     Liege,  Leipzig,  1904. 

(Bibliography,  pp.   206-208). 
Einstein,  M.     Beitrage  z.  uberlieferung  des  Chevalier  au  Cygne  u.  der 

Enfances  Godefroi.    Diss  Berne,  1910.    Rom.  Forsch.  xxix,  721-63 

(1911). 
Hilka,  A.    Historia  septem  Sapientem,  II,  Johannis  de  Alta  Silva,  Dolo- 

pathos  sive  de  Rege  et  septem  sapientibus.     112  pp.     Heidelberg, 

1913.  [!,' 


11  Cf.  Jaffray,  Ch.  VIII,  The  Cleves  Legend,  especially  pp.  88-89,  for  an 
account  of  the  famous  Fete  du  Faisan,  1454,  of  the  entry  of  Adolph  of 
Cleve  as  Le  Chevalier  au  Cygne,  and  of  Adolph's  own  banquet  at  which 
there  was  a  representation  of  the  Swan-Knight,  his  boat,  and  swan. 


252  MEDIAEVAL   ROMANCE   IN   ENGLAND 

Huet,  G.    "  Sur  quelques  formes  de  la  legende  du  Chevalier  au  Cygne" 

Rom.  xxxiv,  206-14  (1905). 
Jaffray,  R.     The  Two  Knights  of  the  Swan,  Lohengrin  and  Helyas.    A 

Study  of  the  Legend  of  the  Swan  Knight.     121  pp.  N.  Y.,  Lond., 

1910. 
Kawczynski,  M.    "  Le  Chevalier  au  Cygne;  Huon  de  Bordeaux."    Bull,  de 

I  Acad,  de  Cracovie,  1902. 
Kleinschmidt,  W.    Das  Verhdltnis  der  Baudoin  de  Sebourc  zu  den  Cheva- 
lier au  Cygne.     Gottingen,  1908. 
Kriiger,  A.     (1)  "  Zur  mittelengl.  Romanze,  Cheulere  Assigne"  Archiv. 

lxxvii,  168-80  (1887).     (2)  "Les  Manuscripts  de  la  Chanson  du 

Chevalier  au  Cygne  et  de  Godefroi  de  Bouillon"  Rom.  xxvni,  421- 

26  (1899). 
Liebermann,  F.    "  Chevalier  au  Cygne  in  England,"  Archiv.  evil,  106-07 

(1901). 
Lot,  F.     "Le  Mythe  des  Enfants-Cygnes,"  Rom.  xxi,  62-67  (1892). 
Mazorriaga,   E.     La  Leyenda   del   Cavallero   del   Cisne.     Transcription 

anotada  del  codice   de  la  Biblioteca  Nacional.     Madrid,   1914.     1, 

Texte. 
Paris,  Gaston.     (1)  La  Naissance  du  Chevalier  au  Cygne  (ed.  Todd), 

Rom.  xix,  314-40  (1890). 

(2)  "  Mayence  et  Nimeque  dans  le  Chevalier  au  Cygne,"  Rom.  xxx, 

404-09  (1901). 
Pestalozzi,  R.     "  Geschichte  d.  deut.  Lohengrinsage."    Neue  Jahrbiicher 

f.  das  klass.  Alterthum,  Leipzig,  xxni  (1909),  147-58. 
Poisson,  G.     "  L'Origine  celtique  de  la  legende  de  Lohengrin,"  Revue 

Celt,  xxxrv,  182-202  (1913). 
Rank,  0.     "  Die  Lohengrinsage,  Ein  Beitrag  zu  ihrer  Motivgestaltung  u. 

Deutung."  Schriften  z.  angewandten  Seelenkunde,  xm.  Vienna,  191 2. 
Smith,  H.    "  Some  Remarks  on  a  Berne  MS.  of  the  Chanson  du  Cheva- 
lier au   Cygne   et   Godefroi  de   Bouillon,"  Rom.   xxxvni,    120-28 

(1909). 


KNIGHT   OF   COURTESY,   or  THE   CHATELAIN 

DE   COUCI 

Versions.  The  gruesome  final  incident  in  the  Knight  of 
Courtesy  and  stories  analogous  to  it,  gives  them  the  name  of 
the  Legend  of  the  Eaten  Heart.1  There  are  at  least  fourteen 
literary  versions  which  Matzke  (i,  p.  i)  divided  into  two 
groups.  In  the  first  the  husband  kills  his  wife's  lover  and  gives 
her  his  daintily  cooked  heart  to  eat.  She  thereupon  kills  her- 
self either  by  falling  from  a  high  place  or  by  self-imposed  star- 
vation. The  most  important  versions  in  this  group  are:  (i) 
the  two  Provencal  biographies,  one  much  longer  than  the  other, 
of  the  Provencal  troubadour,  Guillem  de  Cabestaing,2  accounts 
that  are  found  in  documents  of  the  thirteenth  and  fourteenth 
centuries;  (2)  Boccaccio's  story  in  the  Decanter one,  Day  iv, 
Tale  9,  of  Messer  Guiglielmo  Rossiglione  e  Messer  Guiglielmo 
Guardastagno,  which  the  author  asserts  he  took  from  a  Prov- 

1  See  G.  Cecioni,  "II  Cuore  mangiato,"  Rivista  Contemp.,  1,  Sept.  1888; 
H.  Patzig,  Zur  Geschichte  der  Herzmare  (progr.)  Berlin,  1891 ;  Ahlstrom, 
Studier  i  den  Fornfranska  Lais-Litteraturen,  pp.  127-29,  Upsala,  1892.  Matzke 
(1)  reduced  Patzig's  twenty-three  versions  to  fourteen.  Cf.  Gaston  Paris, 
Rom.  vin,  343-73  (1879);  Histoire  Litt.  de  la  France,  xxvm,  352-90  (1881); 
Child,  Ballads,  v,  33-35  (1894). 

2  Ed.  by  C.  Chabaneau,  Les  Biographies  des  Troubadours  en  langue 
Provencale,  p.  99  ff.,  Toulouse,  1885.  Cf.  E.  Beschnidt,  Die  Biographie  des 
Trobadors  Guillem  de  Capestaing,  Diss.,  Marburg,  1879.  Matzke,  1,  p.  4, 
strongly  questioned  Beschnidt's  conclusion  that  the  shorter  biography  was 
the  earlier;  to  Matzke  it  seemed  simply  an  abridgment  of  the  longer  text. 
Hauvette,  Rom.  xli,  187,  noted  another  derivative  of  the  Provencal  biog- 
raphy in  the  Comptes  Amoureux  de  Madame  Jeanne  Flore,  Lyons,  c.  1540. 
Another  variant  is  to  be  found  in  the  story  told  of  the  Spanish  Marquise  of 
Astorga  and  the  Countess  d'Aulnoys  (Memoires  de  la  Cour  d'Espagne,  1, 
203).  In  this  it  is  the  Marquise  who  in  a  jealous  rage  kills  her  husband's 
mistress,  serves  him  with  the  dead  woman's  heart,  and  shows  him  later  her 
severed  head.     Cf.  Rom.  vni,  362,  n.  4. 

253 


254  MEDIAEVAL   ROMANCE   IN   ENGLAND 

engal  source;3  (3)  Sercambi's  ninety-sixth  story,4  De  prava 
amicitia  vel  societate  (R.  Renier,  Novelle  inedite,  Turin,  1889, 
p.  338)  which  is  closely  related  to  Boccaccio's  tale  but  is  not 
identical  with  it;  and  (4)  an  Indian  folk-tale  concerning  the 
Punjab  hero,  the  Rajah  Rasalu  (Swynnerton).5  In  these  stories 
the  husband  kills  the  lover,  cuts  out  his  heart,  and  in  two  ver- 
sions, the  longer  biography  of  Guillem  and  the  Indian  tale,  shows 
to  the  lady  the  lover's  severed  head  or  his  mutilated  corpse  as 
proof  of  his  death.  Matzke  (1,  p.  8)  believed  all  these  versions 
had  a  common  source  but  that  the  western  forms  were  derived 
from  the  same  lost  Provencal  intermediary. 

Two  of  the  remaining  versions  in  this  group  are  the  oldest 
texts  of  the  story.  One  is  the  eight-line  Lay  of  Guirun  which 
Isolt  sings  in  the  Tristan6  (ed.  Bedier,  1,  295)  of  Thomas.  In 
another  late  twelfth-century  version,  the  lay  of  Ignaure  (Mon- 
merque  et  Michel,  Theatre  jrc.  au  moyen  age,  1842)  the  theme 
is  grotesquely  treated.  The  lay  tells  of  a  knight,  the  lover  of 
twelve  ladies,  who  is  killed  by  their  enraged  husbands  and  his 
heart  given  them  to  eat.  After  this  they  refuse  all  other  food. 
Coarse  derivatives  of  this  tale  are  the  story  of  Linaure,  a  Pro- 
vencal troubabour  (c.  1190)  whose  history  is  mentioned  by 
Arnaud  de  Mersan  (Raynouard,  Choix  de  Poesies,  Paris,  1816- 
21,  11,  308),  and  the  sixty-second  tale  in  the  Cento  Novelle 

3  Gaston  Paris,  Hist.  Litt.  xxviii,  378,  did  not  believe  that  Boccaccio 
followed  the  extant  Provencal  story  of  Guillem,  since  in  the  Italian  version 
the  hero  is  not  described  as  a  poet.  In  Hauvette's  opinion  the  differences 
in  name  and  detail  between  the  Italian  and  the  Provencal  text  were  best 
accounted  for  by  the  probability  that  Boccaccio  worked  from  his  memory 
of  an  abbreviated  version  of  Guillem's  life.  Cf.  Patzig,  p.  21.  For  a  com- 
pact interesting  list  of  versions  including  many  modern  analogues  see  A. 
C.  Lee,  The  Decameron,  Lond.,  1909,  pp.  143-52. 

4  This  did  not  appear  in  Matzke's  list  nor  in  any  of  the  usual  studies  of 
the  legend.  It  was  briefly  cited  by  Miss  E.  N.  Jones,  Boccaccio  and  His 
Imitators,  p.  23,  Chicago,  1910;  also  by  A.  C.  Lee,  op.  cit.,  p.  148.  (Tn  this 
version  by  Sercambi  the  names  are  unlike  those  given  in  the  biographies  or 
by  Boccaccio.  The  cruel  husband  is  called  Marsilio  of  Sivereto;  the  lady, 
Caterina  de'  Salimbeni  da  Siena;  the  lover,  Count  Guarnieri  di  Monte 
Scudaio.  In  this  the  unfaithful  wife  is  given  the  face  of  the  lover,  not  his 
heart,  and  stabs  herself  to  death. 

5  See  also  C.  Swynnerton,  Folk-Lore  Journal,  1,  Lond.  1883 ;  Clouston, 
Popular  Tales,  Eden,  1887,  n,  187.     Cf.  Hauvette,  p.  200,  for  other  variants. 

6  Cf.  Schofield,  PMLA.  xv,  122-25.  In  some  instances  the  severed  heart 
episode  was  associated  with  the  hero  Graelent. 


KNIGHT  OF   COURTESY  255 

Antiche.7  In  this  the  hero  is  a  mere  rustic,  and  the  amorous 
ladies  instead  of  dying  found  a  "  convent  "  where  the  most 
excessive  hospitality  is  practised. 

In  the  second  group  of  stories  the  hero,  dying  at  some  dis- 
tance from  the  lady,  commands  a  servant  to  take  her  his  dead 
heart.  But  the  husband  intercepts  the  servant,  takes  the  heart, 
and  has  it  served  to  his  wife  in  the  same  fashion  as  in  Version 
I.  The  earliest  extant  version  in  this  group  is  the  short  metri- 
cal tale,  Die  Herzmdhre  (F.  Roth,  Frankfurt  a.  M.  1846)  by 
Konrad  von  Wiirzburg  (d.  1287).  This  was  later  turned  into 
an  exemplum  in  the  sermon-book,  Sermones  Parati  de  tempore 
et  de  Sanctis  (No.  124  reprinted  by  Matzke,  2,  p.  18).  Konrad's 
source  according  to  Gaston  Paris  {Rom.  vm,  366)  was  also  the 
source  of  the  long  romance  of  the  late  thirteenth  or  early  four- 
teenth century  by  Jakemon  Maket.  This  romance,  the  well- 
known  Chatelain  de  Couci,8  was  the  most  elaborate  of  all  the 
literary  versions  of  the  story,  and  was  frequently  referred  to 
by  such  writers  as  Froissart,  the  Knight  of  La  Tour  Landri, 
and  Christine  de  Pisan.9  Its  influence  may  be  traced  in  the 
Dutch  adaptation,  Van  den  Borchgrave  van  Couchi  (De  Vries, 
Leiden,  1887),  a  poem  of  which  two  long  fragments  in  a  four- 
teenth-century manuscript  are  known. 

From  Maket's  romance  seems  to  have  been  derived  also  the 
short  Middle  English  version  which  contains  five  hundred  lines 
written  in  four-line  stanzas  riming  abab.  Brandl  (Paul's 
Grundriss  11,  697)  thought  its  dialect  that  of  the  South  Mid- 
land. It  was  composed  in  the  fifteenth  century  but  the  only 
extant  text  is  that  of  Copland's  edition.  Deferring  for  the 
moment  the  comparison  of  the  English  with  the  French  ro- 
mance, we  may  note  the  further  appearance  of  versions  related 

7  Ed.  J.  H.  Heitz,  Strassburg,  1908;  German  trans,  by  J.  Ulrich,  Leipzig, 
1905.     Cf.  Rom.  vm,  368;  xli,  193. 

8  Ed.  Crapelet,  1829;  summarized  by  Langlois,  pp.  186-221.  The  author's 
name  is  known  only  from  an  acrostic  signature.  Early  readings  gave  the 
name  Sakesep.  For  that  of  Maket,  see  Matzke,  2,  p.  12,  n.  28;  Langlois, 
p.  187. 

9  Cf.  Michel,  Les  Chansons  du  Chatelain  de  Couci,  Paris,  1830,  p.  xxxiii; 
cf.  Smythe,  p.  169.  The  continued  popularity  of  the  romance  is  shown  by 
the  manuscripts  listed  in  such  inventories  as  those  of  the  library  of  Charles 
V,  i373>  of  Marguerite  de  Male,  1405,  or  the  libraries  of  Brussels,  1487,  and 
of  Bruges,  1567.    Michel,  p.  xxvi. 


256  MEDIEVAL   ROMANCE    IN   ENGLAND 

to  Maket's  in  the  Chronique  de  France10  (to  1380),  printed  by 
Claude  Fauchet  in  1581,  and  again  in  the  Anecdotes  de  la  Cour 
de  Philippe- August e  (in,  262-320)  published  by  Mile,  de  Lussan 
in  1793.  This  typical  eighteenth-century  version  started  the 
story  upon  a  new  era  of  popularity.11  In  it  the  heroine  was 
given  the  name  of  Gabrielle  de  Vergi,  and  thus  was  completed 
the  confusion  which  had  already  existed  between  the  anonymous 
Lady  of  Fai'el  who  was  beloved  by  the  Chatelain  de  Couci,  and 
Gabrielle,  the  heroine  of  the  delightful  and  pathetic  thirteenth- 
century  romance,  La  Chatelaine  de  Vergi}2 

Of  the  remaining  versions  of  the  story,  the  most  important  is 
Boccaccio's  famous  story  in  the  Decamerone,  Day  iv,  Tale  1,  of 
Guiscardo  e  Ghismonda,  in  which  Tancred,  the  father  of  Ghis- 
monda,  takes  the  place  of  the  cruel  husband.  Through  the 
translation  of  this  story  in  Painter's  Palace  of  Pleasure,  1566 
(ed.  Jacobs,  Lond.,  1890,  1,  180),  or  Tuberville's  Tragical  Tales, 
1576,  or  a  chap-book  version,  it  became  the  foundation  of  many 
English  plays  13  and  poems.  Among  these  was  the  little  ballad 
of  Lady  Diamond  (Child,  No.  269,  vol.  v,  29)  in  which,  as  in 
the  majority  of  the  ballad  versions  in  Italian,  Scandinavian,  or 

10  Reprinted  by  Matzke,  2,  pp.  6-8,  from  Fauchet's  Recueil  de  VOrigine 
de  la  Langue  et  de  la  Poesie  frangoise.  Matzke  thought  this  text  represented 
a  version  older  than  the  romance,  and  closely  related  to  the  Provencal 
biography.  In  the  Chronique  the  hero,  Regnault  de  Coucy,  is  not  a  trouvere, 
but  a  warrior  who  willingly  joins  in  the  Crusade  of  Richard  the  Lion-Heart; 
he  receives  from  his  lady  "  ung  las  de  soye,"  instead  of  the  braid  of  hair 
which  in  the  Chatelain,  the  Lady  of  Fai'el  gave  to  her  lover;  he  dies  on  land, 
not  as  did  the  Chatelain  from  a  poisoned  arrow,  and  on  his  homeward 
journey.  These  and  other  details  make  it  hard  to  believe  that  the  Chatelain 
could  have  been  the  source  of  the  Chronique. 

11  Cf.  Michel,  p.  xxi;  Paris,  Rom.  vm,  371;  Lorenz,  pp.  119-38.  See 
also  Lorenz,  Die  altfrz.  Versnovelle  von  der  Kastellanin  von  Vergi  in  spatern 
Bearbeitungen,  Diss.   138  pp.  Halle,  1909. 

12  Ed.  Raynaud,  Paris,  1910;  1912  (rev.  Rom.  Rev.  vi,  112);  trans,  by 
Alice  Kemp-Welch,  Lond.  1903.  On  the  story  in  Art,  see  W.  Bombe,  La 
Chatelaine  de  Vergy  en  Italie,  Revue  des  Langues  Romans,  Lvn,  262^1 
(1914) ;  Gazette  des  Beaux-Arts,  Sept.  1911,  p.  231  ff;  see  also  E.  Bertaux, 
La  Femme  et  l'Art  du  Moyen-Age  frangais,  Revue  de  Paris,  Nov.  15,  1909, 
PP-  367-90.  In  this  romance  the  exquisitely  sensitive  heroine  dies  when  she 
thinks  that  her  lover  has  revealed,  despite  her  prohibition,  the  secret  of  her 
love.  The  hero  commits  suicide  when  he  finds  her  dead  body.  The  perfect 
constancy  of  the  two  pairs  of  lovers  in  the  Chatelain  de  Couci  and,  in  the 
Chatelaine  de  Vergi  associated  them  together  and  probably  facilitated  a  cer- 
tain confusion  about  them. 

13  Cf.  F.  Schelling,  Elizabethan  Drama,  N.  Y.,  1908,  Index. 


KNIGHT   OF   COURTESY  257 

German,  the  lover's  heart  is  sent  in  a  cup  of  gold  to  the  lady 
by  her  angry  father.  In  the  Swedish  song,  Hertig  Frojdenborg 
och  Fro  ken  Adelin,  an  eighteenth-century  broadside,  the  origi- 
nal trait  is  preserved,  and  the  lady  eats  the  heart  of  her  lover. 
The  popular  versions  seem  to  mingle  traits  that  belong  to 
Boccaccio's  two  stories  or  their  sources,  but  in  a  German  meis- 
terlied,  Vom  dem  Brernberger's  end  und  Tod  (von  der  Hagen, 
Minnesinger,  1838,  iv,  281),  of  the  sixteenth  century,  there  is 
a  curious  reversion  to  the  most  primitive  type  of  the  story. 
The  husband  kills  the  lover,  cuts  off  his  head,  and  gives  his 
heart  to  the  lady  to  eat  (Child,  p.  32).  The  lover  in  this  ver- 
sion is  said  to  be  the  minnesinger,  Reinmann  von  Brennenberg, 
who  lived  in  the  middle  of  the  thirteenth  century  and  who  loved 
a  duchess  of  Austria.  The  fact  that  the  hero  here  is  a  poet  is 
paralleled  in  Guirun,  in  the  Chdtelain  de  Couci,  in  Linaure  and 
Guillem  de  Cabestaing,  and  seems  to  represent  a  distinct  line 
of  tradition.  In  Matzke's  opinion  (1,  p.  8)  it  was  due  to  an 
innovation  made  by  the  author  of  the  longer  biography  of 
Guillem,  and  probably  did  not  appear  in  the  original  version. 

The  Knight  of  Courtesy  is  an  abbreviated,  indeed,  an  expur- 
gated, version  of  the  Chdtelain.  It  is  so  utterly  vague  as  to 
names  and  places  of  the  French  story,  that  it  seems  probable 
the  English  author  was  simply  writing  from  memory.  The 
worth  of  his  story  has  been  variously  estimated.  Gaston  Paris 
(Rom.  viii,  369)  called  it  "charmant";  Dr.  Rickert  (p.  li) 
considered  it  "  poor  in  everything  but  sentimentality  " ;  interest- 
ing only  as  "  representing  an  aspect  of  mediaeval  psychology, 
being  a  singular  combination  of  morbid  hyper-analysis  with 
sheer  brutality."  It  may  be  granted  that  the  English  poet  is 
poor  in  invention;  he  intrudes,  for  instance,  the  hackneyed  ac- 
count of  a  dragon  fight  among  the  adventures  befalling  the  hero 
after  his  separation  from  his  lady-love.  But  it  is  also  true  that 
he  avoids  the  prolixity  of  the  French  poet  without  losing  the 
essential  pathos  of  the  story.  The  eight  thousand  lines  of  the 
French  romance  present  a  brilliant  picture  of  the  pageantry  of 
life  in  the  late  thirteenth  century.  It  abounds  in  descriptions 
of  beautiful  caroles,  of  fetes  and  tourneys  and  heraldic  de- 
vices ; 14  it  is  a  characteristic  product  of  V amour  courtois  with 

14  Among  the  heroes  present  are  Simon  de  Montfort  and  the  Sire  de  Join- 
ville  and  Jehans  de  Niyelle  (Nesles).    Was  this  the  Jean  de  Nesles  for  whom 


258  MEDIAEVAL   ROMANCE    IN    ENGIAND 

the  usual  courtly  exaltation  of  lover  over  husband.  Much  in 
the  romance  is  made  of  the  way  in  which  the  Chatelain  blithely 
woos  the  wife  of  his  friend  and  neighbor,  of  the  rebuffs  and 
disappointments  he  must  endure  before  the  capricious  lady 
grants  him  her  favor,  of  the  songs  he  makes  for  her,15  of  the 
stratagems  and  disguises  which  are  necessary  for  the  lovers' 
meeting,  and  of  the  friends  to  whom  they  frequently  confide 
their  emotions.16 

All  the  picturesque  detail  and  all  the  elaborate  love-making 
are  omitted  by  the  English  poet.  His  lovers  love,  but  in  purity. 
After  their  one  mutual  confession,  they  meet  but  once  again 
when,  as  in  the  French  tale,  she  gives  him  locks  of  her  yellow 
hair  to  wear  henceforth  on  his  helmet.  In  the  French  version 
the  hero  comes  to  his  death  in  Palestine  as  a  Crusader  of 
Richard  the  First  of  England;  in  the  English  story  he  dies  at 
a  siege  of  Rhodes,  presumably  that  of   1443    (Rickert,  p.  l). 

Origin.  The  ultimate  origin  of  the  Legend  of  the  Eaten 
Heart  is  still  a  matter  of  debate.  Remotely  its  horror  may  be 
paralleled  in  the  story  of  Tantalus  serving  his  own  son  Pelops 
to  the  gods,  or  in  the  vengeance  of  Atreus  when  he  gave  his 
brother  Thyestes  the  flesh  of  the  latter 's  sons,  or  in  the  tale 
of  Procne  who  gave  her  faithless  husband  the  body  of  their 
child  Itys.  Likewise  in  Norse  legend  there  is  a  story  of  Gudrun, 
who,  in  vengeance  for  the  death  of  her  brothers,  gave  to  her 
lord  Atli  the  roasted  hearts  of  their  two  children  (Vigfusson 
and  Powell,  Corpus  Poet.  Boreale,  1883,  1,  51).  The  legend  in 
any  of  these  forms  is  as  ancient  as  it  is  terrible,  but  it  is  in- 
herently improbable  that  the  pre-Christian  tradition  had  any 


the  Perlesvaus  was  written?  The  last  historical  reference  to  Jean  among 
those  cited  by  Evans  (High  History  of  the  Holy  Grail,  Everyman  ecL,  p. 
xii)  is  for  1225.  Cf.  Nitze,  MLN.,  xiv,  498  (1899).  Prinet,  p.  170,  pointed 
out  that  the  description  of  the  arms  of  Jean  is  technically  at  fault. 

15  In  this  the  romance  follows  the  fashion  first  set  in  the  romance  of 
Guillaume  de  Dole  (Le  Roman  de  la  Rote),  SATF.,  1893,  1.  8-15,  —  at  least 
according  to  its  author's  own  assertion, 

16  Matzke,  1,  p.  15,  noted  the  interweaving  of  characters  and  themes  of 
typical  romances;  from  the  Tristan  legend,  the  husband  who  spies  on  the 
interviews  of  the  lovers;  the  wife's  devoted  friend  and  confidante;  the  ruses 
for  the  deceit  of  the  husband,  etc.  Gobert,  the  Chatelain's  faithful  squire, 
serves  the  lovers  by  pretending  to  be  the  husband's  spy. 


KNIGHT  OF   COURTESY  259 

direct  connection  with  the  mediaeval  versions.17  In  them  it  was 
drawn  into  the  usual  triangle  story  of  husband,  lover,  and  wife, 
and  found  credence  in  a  society  only  too  familiar  with  out- 
bursts of  barbarous  passion.  But  the  question  remains.  Did 
the  mediaeval  story  have  its  origin  in  fact  or  fiction,  and  if  the 
last,  did  it  come  from  the  East  or  the  West?  Ahlstrom  {op. 
cit.)  believed  that  the  tradition  showed  Germanic  elements 
but  thought  its  literary  form  derived  from  Guirun.  Gaston 
Paris,  chiefly  because  of  the  supposedly  "  Breton "  lay  of 
Guirun,  believed  in  a  Celtic  origin.  This  theory  he  aban- 
doned when  the  Indian  tale  of  Rasalu  came  to  light,  and  he 
voiced  his  agreement  {Rom.  xxi,  140)  with  Patzig  that  the 
story  was  originally  an  Eastern  tale  which  had  in  some  way 
come  to  be  localized  in  Provence.  The  name  Rasalu  may  have 
caused  the  story  to  be  localized  at  Castel-Rossello  in  the  duchy 
of  Roussilon  in  Provence.  Since  the  home  of  the  troubadour, 
Guillem  de  Cabestaing,  was  not  far  from  that  of  Raimon  de 
Castel-Rossello,  to  whom  he  is  known  to  have  addressed  some 
poems,  the  legendary  association  of  the  two  was  not  unnatural. 
If  the  part  of  Rasalu  were  ascribed  to  Raimon,  the  identifica- 
tion of  the  poet  with  the  lover  of  Rasalu's  wife  would  follow. 
But  all  this  is  conjecture  which  is  reducible  simply  to  the  fact 
that  some  form  of  the  name  Rossilon  must  have  stood  in  the 
source  of  the  Provencal  biographies  and  of  the  Indian  tale 
(Matzke,  2,  p.  8). 

The  most  serious  argument  against  the  belief  that  this  source 
was  Oriental  was  advanced  by  Hauvette  (pp.  199-205).  The 
Indian  tale  is  admittedly  dateless.  There  is  in  it  nothing  that 
is  specifically  Indian ;  indeed  it  belongs  to  that  northern  Punjab 
country  which  is,  on  account  of  its  Mohammedan  population, 
the  least  Indian  part  of  all  India.  In  the  extant  versions  the 
story  of  Rasalu  gives  but  comparatively  minor  place  to  the 
central  episode  of  the  Eaten  Heart;  its  originators  were  more 
interested  in  the  speaking  parrot  which  betrays  the  wife's  in- 
fidelity and  in  Rasalu's  long  preparations  for  revenge.18     It  is 

17  Hauvette,  p.  19S,  discussed  the  allegorical  idea  of  the  Eaten  Heart  in 
connection  with  its  use  by  such  poets  as  Sordello  or  Dante  {Vita  Nuova, 
hi).  In  these  and  other  references,  the  essential  thought  is  that  the  heart 
will  convey  to  another  its  own  passionate  virtue. 

18  In   an   Eastern    tale   of   the   type   known   as   the   Dog   and   the   Lady 


26o  MEDIAEVAL   ROMANCE    IN   ENGLAND 

characteristic,  Hauvette  thought,  that  in  this  single  version,  the 
wife's  suicide  is  prompted  by  her  despairing  anticipation  of  her 
husband's  cruelty;  in  the  western  versions  she  dies  so  that  she 
may  never  again  touch  food  after  tasting  that  which  was  noblest 
on  earth.  In  other  words,  the  Eaten  Heart  episode  is  not  vitally 
emphasized  in  the  Indian  tale  and  we  may  infer  that  it  was  not 
an  original,  but  a  borrowed,  element  in  the  story. 

This  borrowed  element,  in  Hauvette's  opinion,  was  a  version 
of  Boccaccio's  tale  brought  into  India  by  some  of  those  far- 
travelling  Italians  of  the  fourteenth  and  fifteenth  centuries 
whose  prolonged  residence  in  India  is  a  known  fact.  Matzke 
(i,  p.  6)  had  previously  dismissed  this  possibility  of  derivation 
on  the  ground  that  in  Boccaccio's  story  nothing  is  said  of  the 
lover's  severed  head,  and  yet  this  trait  appears  in  the  story  of 
Rasalu.  But  Matzke  himself  had  proved  that  the  detail  be- 
longed to  Boccaccio's  source  story,  and  this  might  have  been 
transferred  to  the  East  as  easily  as  the  Boccaccian  tale  itself. 

There  is  no  known  reason,  according  to  Langfors  (p.  353),  for 
the  original  association  of  the  legend  of  the  Eaten  Heart  with 
Guillem  de  Cabestaing,  a  man  of  whom  history  records  not 
much  more  than  the  fact  that  in  12 12  he  fought  at  the  battle 
of  Las  Nevas.  If  he  were  the  same  man  as  the  troubadour, 
Guillem,  to  whom  some  nine  love  lyrics  are  attributed,  he  was 
at  one  time  the  friend  of  Raimon  de  Roussilon  to  whom  two  of 
the  songs  are  dedicated  (Langfors,  p.  8).  But  no  text  before  the 
end  of  the  thirteenth  century  makes  allusion  to  any  tragic  event 
happening  to  Guillem  or  to  the  wife  of  Raimon,  a  lady  who  was 
a  widow  when  she  married  Raimon  and  who  lived,  apparently, 
to  take  still  a  third  husband.  The  transference  of  the  legend 
to  the  crusading  poet  known  as  the  Chatelain  de  Couci  was 
probably  due  to  the  invention  of  Jakemon  Maket.  By  choosing 
the  Chatelain  as  a  type  of  amorous  chevalier,  the  author  was 
enabled  to  introduce  into  his  story   the  actual   songs  of  the 


(Kittredge,  "  Gorlagon,"  Harvard  Studies,  vm,  247,  252,  n.  1),  the  faithless 
wife  is  betrayed  by  a  faithful  dog  and  is  made  to  suffer  a  variety  of  hor- 
rible punishments.  Only  in  a  western  version,  the  Latin  text  of  the  Gesta 
Romanorum  (Oesterley,  p.  355-6)  does  the  punishment  in  the  least  resemble 
that  of  the  Eaten  Heart  legend.  In  the  Gesta  the  lady  has  to  eat  from  her 
lover's  skull.  In  the  Latin  romance  of  Gorlagon  (Kittredge,  p.  245),  the 
lady  had  constantly  to  hold  before  her  at  table  the  bloody  head  of  her  lover. 


KNIGHT   OF    COURTESY  261 

Chevalier.19  He  localized  the  romance  in  Vermandois,  a  district 
of  which  he  himself,  to  judge  from  his  rimes,  his  knowledge  of 
the  country,  etc.,  was  a  native  (Langlois,  p.  187).  His  choice 
for  heroine  of  a  lady  of  Fai'el  (Fayet),  a  place  near  St.  Quentin, 
may  have  been  due  to  the  same  interest  in  locality.  In  Maket's 
day  Faiel  had  a  chateau  and  was  not  without  some  importance 
(Rom.  viii,  361).  But  whatever  the  realism  of  place  or  the 
accuracy  of  observation  of  thirteenth-century  manners  and  cus- 
toms or  the  sophistication  of  sentiment  in  the  Chdtelain  de 
Couciy  its  author  did  not  succeed  by  virtue  of  these  things  in 
impressing  his  story  on  the  imagination  of  men.  That  long- 
lasting  impression  came  from  the  terrible  and  tragic  theme  he 
had  borrowed  and  from  its  sad  celebration  of  the  loyalty  in  love 
"  de  quoi  on  doie  faire  conte." 

BIBLIOGRAPHY 

Texts:  (i)  Bodleian,  edition  printed  in  1568  by  William  Copland; 
ed.  Ritson,  1802,  in,  193-218;  Hazlitt,  Remains,  1866,  11,  64-87;  trans. 
Rickert,  Romances  of  Love,  1908,  p.  141  ff. 

Studies:    Cf.  Wells,  Manual,  p.  787. 
Hauvette,  H.    "  La  39c  Nouvelle  du  Decameron  et  la  legende  du  Coeur 

Mange,"  Rom.  xli,  184-205   (1912). 
Langfors,  A.    "  Le  Troubadour  Guilhem  de  Cabestanh,"  Annates  du  Midi, 

xxvi  (1914):  I,  Les  Chansons  attributes  a  Guilhem  de  Cabestanh, 

pp.  1-5 1 ;  189-99;  II,  Les  Quatre  Redactions  de  la  Biographie,  pp. 

199-225;   III,   Guilhem   de   Cabestanh  Personnage   Historique,   pp. 

349-356.     Rev.  Neuphil.  Mitteil.,  Helsingfors,   191 5,  p.  38. 
Langlois,    C.    V.     La   Societe   Frangaise   au   XHIe    Siecle    d'apres    dix 

romans  d'aventure,  2e  ed.  Paris,  1904;  La  Chdtelain  de  Coucy,  pp. 

186-221;  3e  ed.  1911. 


19  The  romance  refers  to  the  Chatelain's  name  as  Renault.  Gaston  Paris 
{Rom.  viii,  353)  accepted  this  as  the  name  of  the  Crusading  poet.  Fath, 
Die  Lieder  des  Castellans  von  Coucy,  Heidelberg,  1883,  showed  that  the 
name  of  the  historic  trouvere  was  Gui  de  Couci  who  died  and  was  buried  at 
sea  during  the  Fourth  Crusade.  Matzke,  2,  p.  14,  found  no  better  explana- 
tion for  the  use  of  the  name  Renault  than  that  it  was  "  probably  frequent 
in  the  well-known  Coucy  family."  Since  the  same  name  appeared  in  Fauchet's 
Chronique,  Matzke  pointed  out  that  Maket  could  not  have  been  the  first  to 
associate  a  Renault  de  Coucy  with  the  Eaten  Heart  tradition,  though  he  was 
probably  the  first  to  identify  Renault  with  the  poetic  Chatelain  de  Coucy. 
It  may  be  said  that  few  romances  are  more  in  need  of  historical  and  critical 
study  than  the  Chdtelain.    It  has  not  been  edited  since  1829. 


262  MEDIEVAL   ROMANCE    IN   ENGLAND 

Lorenz,  E.  Die  Kastellanin  von  Vergi  in  der  Literatur  Frankreichs, 
Italiens,  der  Niederlande,  Englands  u.  Deutschlands;  mit  einem 
Anhange,  Die  "  Kastellan  von  Couci "  sage  als  "  Gabrielle  de 
Vergi"  Legende,  pp.  117-38.    Halle,  1909. 

Matzke,  J.  (1)  "The  Legend  of  the  Eaten  Heart,"  MLN.  xxvi,  1-8 
(1911). 

(2)  "The  Roman  du  Chatelain  de  Couci  and  Fauchet's  Chronique," 
Studies  in  honor  of  A.  Marshall  Elliot,  1,  1-18.     Baltimore,  191 3. 

Prinet,  M.  "Les  Armoires  dans  le  doman  du  Chatelain  de  Coucy" 
Rom.  xlvi,  161-79  (1920). 

Siefken,  O.    Das  gediildige  Weib,  pp.  69-71. 

Smythe,  Barbara.  Trobador  Poets  (Guilhem  de  Cabestanh,  pp.  169- 
181).     N.  Y.,  Lond.,  1911. 

Swynnerton,  C.     Romantic  Tales  from  the  Panjab.    Westminster,  1903. 

Zanders,  J.  Die  altprovenzalische  Prosanovelle ;  eine  liter ar hist orische 
Kritik  der  Trobador-Biographien,  pp.  113-27.   Halle,  1913. 


THE   SQUYR   OF   LOWE   DEGRE 

Versions.  Late  but  not  least  in  fame  or  significance  among 
the  English  metrical  romances  is  the  Squyr  of  Lowe  Degre.  It 
was  printed  by  Wynkyn  de  Worde  about  1520,  by  William  Cop- 
land between  1555  and  1560,  was  licensed  for  printing  to  John 
Kynge  in  1560,  and  listed  in  1575  by  Robert  Laneham  with 
other  still  popular  romances.1  The  greater  Elizabethan  poets  not 
infrequently  referred  to  it,  though  often  in  somewhat  jocose 
vein  (Mead,  p.  xiii).  The  only  extant  texts,  however,  are  the 
short  fragments  of  de  Worde's  edition  (W),  the  Copland  edition 
(C),  1 132  lines  in  short  riming  couplets,  a  poem  which  closely 
follows  the  older  text,  and  a  short  version  presumably  derived 
from  oral  tradition,  in  Bishop  Percy's  Folio  Manuscript  (P). 
The  two  versions  (CP)  seem  to  be  independent  derivatives  of 
an  early  tale  (x)  which  had  probably  the  same  general  outlines 
as  the  version  P.  The  original  version  was  presumably  in  the 
form  of  a  short  tale  told  in  verse  which  gradually  developed 
through  romantic  accretions  into  the  form  of  C.  The  composi- 
tion of  this  original  poem  Mead  (pp.  lii-lxxvii)  ascribed  to  the 
fifteenth  century.  The  old  theory,  held  by  Tunk  even  as  late 
as  1900,  that  the  poem  belonged  to  the  fourteenth  century, 
rested  on  no  better  ground  than  the  supposition  that  Chaucer's 
amusing  list  of  garden  herbs  and  of  songless  birds  in  that  "  fair 
forest  "  through  which  Sir  Thopas  rode,  was  a  parody  on  the 
descriptive  catalogue  lists  of  trees  and  birds  in  the  Squyr  of 
Lowe  Degre.  Though  this  East-Midland  romance  is,  as  Mead's 
notes  show,  a  perfect  mosaic  of  the  romantic  conventions  which 
Chaucer  burlesqued  so  gaily  in  Sir  Thopas,  there  is  nothing 
which  proves  that  the  Squyr  was  antecedent  to  Chaucer's  poem. 
In  fact  the  linguistic  evidence  of  the  Squyr,  as  well  as  its  gen- 

1  Cf.  H.  S.  Murch,  Knight  of  the  Burning  Pestle,  Yale  Studies,  New 
York,  1908,  pp.  lxx  ff.  Laneham's  letter  contains  a  list  of  the  ballads  and 
story-books  owned  by  the  Coventry  mason,  Captain  Cox.  Among  his  ro- 
mances were  also  Bevis  of  Hampton,  Knight  of  Courtesy,  Eglamour,  Tria- 
mour,  Isumbras,  the  Seven  Wise  Masters. 

263 


264  MEDIEVAL   ROMANCE    IN   ENGLAND 

eral  imitative  quality  in  style,  point  to  a  fifteenth-century  date. 
There  are  at  least  twenty-six  words  which,  with  two  or  three 
exceptions,  do  not  appear  in  English  before  that  period,  and  the 
final  e  is  practically  silent. 

The  romance  tells  of  a  poor  squire  who  loves  the  daughter  of 
the  King  of  Hungary.  His  confession  of  love  is  betrayed  to  the 
King  by  an  eaves-dropping  steward;  the  squire  is  attacked  as 
he  comes  at  night  to  bid  farewell  to  his  lady;  but  instead  of 
being  killed,  as  she  thinks,  he  is  carried  to  the  King  and  is 
presently  sent  on  his  way  to  achieve  glory  in  far-off  lands  as 
the  Princess  has  previously  commanded  him.  Meanwhile  for 
seven  years 2  she  keeps  with  her  the  embalmed  body  of  the 
steward  whose  men  had  left  it,  so  mutilated  as  to  be  unrecog- 
nizable, at  her  door.  Just  as  she  is  about  to  become  an  an- 
choress, the  Squire  returns  and  their  marriage  is  celebrated. 
In  the  extant  texts,  at  least,  the  plot  elements  are  awkwardly 
managed.  The  mesalliance  theme  is  unconvincing,  for  the  King 
has  no  real  feeling  about  his  daughter's  marriage  to  the  Squire; 
the  Princess  makes  no  delay  about  loving  her  lowly  suitor,  but 
pauses  before  undoing  her  door  3  for  a  discourse  of  unconscion- 
able length  while  he  waits  in  peril  of  his  life;  her  father,  who 
is  supposed  to  be  kindly,  nevertheless  permits  her  to  suffer 
seven  years  of  needless  anguish.  But  it  is  not  by  the  structural 
flaws  in  the  poem  that  the  poet  should  be  judged  (Mead,  p. 
Ixxvii  ff.).  The  spirit  in  which  he  wrote  was  delicate  if  naive, 
and  his  art  vividly  and  charmingly  pictorial.  All  the  glamour 
of  the  mediaeval  pageant  of  life  is  there,  in  descriptions  rich 
in  details  concerning  lovely  fabrics  and  armour,  the  stately 
course  of  mediaeval  banquets,  the  sports  and  diversions  of  fif- 
teenth-century lords  and  ladies.  The  Princess's  anaphoric  fare- 
well to  the  glad  things  of  the  world   (vv.  940-55)  4  and  her 

2  On  seven  as  a  conventional  number,  see  Mead,  p.  48;  Child,  Ballads, 
Index. 

3  The  plea,  "  Undo  Your  Dore,"  seems  to  have  been  the  popular  title 
for  the  poem  before  this  was  displaced  by  the  present  one  (Mead,  p.  71,  n. 

534). 

4  This  has  the  tone  of  some  of  the  moral  lyrics  of  the  fifteenth  century. 
Cf.  Chambers  and  Sedgwick,  Early  Eng.  Lyrics.  The  entire  seriousness  of 
this  passage,  as  indeed  of  the  whole  romance  in  setting,  or  incident,  shows  how 
mistaken  was  Brandl's  idea  (Paul's  Grundriss  d.  germ.  Phil.  11,  1,  657)  that 
it  was  a  burlesque  on  romances  of  the  exile-and-return  type  (cf.  Mead,  p. 
lxxix) . 


THE    SQUYR   OF    LOWE   DEGRE  265 

father's  enumeration  of  the  delights  with  which  he  would  tempt 
her  from  her  sorrow  (vv.  738-852)  give  a  vision  of  what  was 
to  the  poet  a  kind  of  ideality.  Its  potent  appeal  for  a  later  age 
is  aptly  set  forth  in  the  appreciative  criticism  of  James  Russell 
Lowell  {Literary  Essays). 

Origin.  The  author's  familiarity  with  a  certain  number  of 
Middle  English  romances  is  indicated  by  his  references  to  such 
heroes  as  Libeaus,  Arthur,  Guy  of  Warwick  (vv.  73;  614),  by 
his  stereotyped  diction,  and,  as  already  suggested,  by  the  con- 
ventional nature  of  his  characters  and  the  incidents  in  which 
they  take  part.  In  other  words  it  seems  improbable  that  in  the 
Squyr  there  is  any  single  basic  fact  or  legend.  As  Mead  (pp. 
xxvii  ff.)  pointed  out,  the  supposedly  lowly  suitor  who  loves  a 
king's  daughter  appears  in  Apollonius,  in  William  of  Palerne,  in 
Horn,  in  Roswall  and  Lillian.  Closest  of  all,  perhaps,  among 
the  romances  is  the  opening  situation  in  Guy  of  Warwick  5  in 
which  the  heroine  likewise  imposes  a  quest  for  glory  on  her 
humble  suitor.  The  Treacherous  Steward  who  betrays  the  lovers 
is  a  purely  stock  character  (Mead,  p.  xxx),  even  as  absence  as 
a  Test  of  Fidelity  is  a  stock  situation. 

The  two  most  distinctive  motifs  in  the  Squyr  of  Lowe  Degre 
are  the  substitution  of  the  steward's  body  for  that  of  a  lover 
supposed  to  be  dead,  and  the  preservation  of  that  body  by  a 
heart-broken  lady.  The  nearest  parallel  to  the  first  is  the  epi- 
sode in  Florence  of  Rome  6  in  which  the  false  brother-in-law, 
Miles,  presents  to  Florence  the  mutilated  body  of  a  vassal  as 
that  of  her  husband.  He  does  it  in  hope  of  winning  her  him- 
self. In  the  Squyr  of  Lowe  Degre  the  lady's  father  permits  the 
deception  in  the  hope  of  breaking  her  attachment  for  the  Squire. 
The  narrative  setting  for  the  two  episodes  is  different,  and  this 
similar  detail  in  the  Squyr  may  well  be  an  independent  inven- 
tion. But  in  view  of  the  general  uninventiveness  of  the  author 
of  the  Squyr  it  seems  improbable. 

5  Mead,  p.  xxxvii-xliv,  believed  in  the  strong  probability  that  the  author 
of  the  Squyr  modelled  his  poem  on  that  of  Guy.  He  considered  it  proof 
that  the  author  of  the  Squyr  was  not  working  from  a  French  original  because 
the  Squyr  has  so  many  passages  similar  in  phraseology  to  Guy. 

6  See  Florence  of  Rome,  n.  6.  The  parallel  does  not  seem  to  have  been 
noted  before. 


266  MEDIEVAL   ROMANCE    IN   ENGLAND 

The  ghastly  notion  of  embalming  the  dead  and  keeping  the 
relic  of  love  or  hatred  in  intimate  fashion,  is  found,  as  Jefferson 
(p.  102)  has  pointed  out,  in  a  number  of  romantic  narratives. 
In  Boccaccio's  story  (Decameron,  iv,  5)  Isabella  keeps  the  head 
of  her  lover  in  a  pot  of  basil  until  she  herself  dies ;  in  the  fif^ 
teenth-century  romance  of  Eger  and  Grime,  Lady  Loospaine 
keeps  the  hand  of  her  enemy  Graysteel  in  a  coffer;  in  the 
Knight  of  Courtesy  a  dying  lover  orders  his  heart  to  be  carried 
to  his  lady.7  Other  parallels  between  this  story  and  that  of  the 
Squyr  are  to  be  found  in  the  earlier  scenes,  in  the  secret  love 
affair  of  the  lady  with  a  vassal,  in  their  stolen  interview  in  the 
garden,  their  betrayal  by  a  spy,  their  separation.  The  known 
ancestry  of  the  Knight  of  Courtesy  makes  it  probable  that  it  in- 
fluenced and  was  not  influenced  by  the  Squyr  of  Lowe  Degre,  the 
author  of  which  may  even  have  known  the  French  original  of 
the  former  poem.8  In  the  Chdtelain  de  Couci  the  combination 
of  decorative  and  pictorial  elements  with  frank  mediaeval  bru- 
tality corresponds  closely  with  the  descriptive  richness  and  mor- 
bidity of  the  Squyr. 

BIBLIOGRAPHY 

Texts:  (i)  W.  Wynkyn  de  Worde's  edition,  about  1520,  under  the 
title  "Undo  Youre  Dore  "  (180  lines  in  two  fragments),  Britwell  Court, 
Bucks.  Eng. ;  (2)  C.  Copland's  edition  printed  about  1555-60,  Brit. 
Mus.  (1132  lines),  repr.  by  Ritson,  m,  145-92,  1802;  by  W.  C.  Hazlitt, 
Early  Popular  Poetry,  11,  21-64  (1866);  (3)  P.  Percy  Folio  MS.  (170 
lines),  ed.  Hales  and  Furnivall,  in,  263-68.  The  texts  have  been  re- 
printed by  W.  E.  Mead,  The  Squyr  of  Lowe  Degre,  A  Middle  English 
Metrical  Romance,  Boston,  1904.    Trans.  E.  Rickert,  Romances  of  Love, 

P-  153  ff- 

Studies:    Cf.  Wells,  Manual,  p.  786. 
Jefferson,  B.  L.    "A  Note  on  the  Squyr  of  Lowe  Degre,"  MLN.  xxviii, 

102-103  (1913)- 
Mead,  W.  E.     See  Texts. 
Rickert,  E.    See  Texts. 
Tunk,  P.    Studien  zur  mittelengl.  Romanze,  the  Squyr  of  Lowe  Degre. 

Diss.  68  pp.     Breslau,  1900. 
Weyrauch,  M.     "  Zur  Komposition,  Entstehungszeit  u.  Beurteilung  der 

ME.  Squyr  of  Lowe  Degre,"  Eng.  Stud,  xxxi,  177-82   (1902). 


7  See  the  Knight  of  Courtesy  here,  note  18,  where  another  instance  of  an 
embalmed  head  is  noted  in  connection  with  Gorlagon. 

8  The  fact  that  the  Squire  is  told  to  go  to  Rhodes  implies  that  the  author 
was  here  following  the  Knight  of  Courtesy  and  not  its  French  original. 


OCTAVIAN 

Versions.  Octavian  begins  with  the  tale  of  a  Calumniated 
Wife  but  passes  swiftly  to  the  amorous  and  militant  adventures 
of  her  sons.  Through  the  many  texts  into  which  it  passed,  the 
romance  kept  always  something  of  the  hearty  humor  and  adven- 
turous vigor  of  the  chansons  de  geste..  Its  earliest  extant  ver- 
sion (A)  in  French,  seems  to  be  the  octosyllabic  poem  of  5371 
lines  copied  by  an  Anglo-Norman  scribe  in  an  early  fourteenth- 
century  manuscript  now  in  the  Bodleian  (Hatton  100,  ed.  K. 
Vollmoller,  Heilbronn,  1883).  In  Vollmoller's  opinion  (p.  iv)  its 
author  was  a  Picard  poet  familiar  with  Paris  and  its  environs, 
who  was  writing  between  1229-1244.1  Foerster  (Aiol,  Heil- 
bronn, 1876-82,  p.  xxvi)  noted  some  borrowings  from  the  two 
chansons  de  geste,  Aiol  and  Elie  de  Saint-Gille.  An  inedited 
second  French  version  (B)  of  Octavian,  known  as  Florent  et 
Octavian  de  Rome,  is  also  in  the  form  of  a  chanson  de  geste. 
Gautier  {Bibliographic,  p.  104)  believed  that  the  Anglo-Norman 
poem  (A)  is  simply  an  abridgment  of  the  18,576  lines  in  mono- 
rimed  laisses  in  Florent.  Florent  is  known  in  three  fifteenth- 
century  manuscripts  of  the  Bibliotheque  Nationale,  Paris,  and 
through  a  fifteenth-century  prose  version  which  passed  quickly 
into  print.2  A  French  Octavian  printed  in  1534  was  translated 
and  printed  in  German  in  1535  and  became  one  of  the  most 
popular  of  German  Volksbucher  (Weber,  1,  p.  lix).  The  story 
was  the  first  in  that  famous  collection  of  romances,  Das  Buch 
der  Liebe,  printed  at  Frankfurt  in  1587.  The  Volksbuch  was 
the  basis  of  a  play  by  Hans  Sachs,  1555,  a  poem  by  Sebastian 
Wilde,  1566,  and  of  popular  versions  in  Danish,  1597,  in  Dutch, 
1 62 1,  in  Icelandic,  1733.  Several  of  these  books  are  known  to 
have  existed  in  earlier  editions  than  those  now  extant.    Streve 

1  The  poet  seems  to  think  of  Jerusalem  as  being  in  the  possession  of 
the  Christians,  as  it  actually  was  during  these  years.  Paris  {Rom.  xi,  610) 
thought  the  reasons  given  by  the  editor  for  accepting  this  date  unconvincing. 

2  Cf.  Vollmoller,  p.  xvn;  Paris,  Rom.  xi,  611;  Gautier,  Bibliographie, 
p.  103. 

267 


268  MEDIAEVAL   ROMANCE    IN   ENGLAND 

(cf.  Stammtafel,  p.  50)  derived  them  all  ultimately  from  the 
French  A  version. 

In  a  third  version  similarly  derived  from  the  common  source 
of  the  A  and  B  versions,  Streve  placed  both  the  French  miracle 
play,  Le  Rot  Thierry  et  Osanne,  sa  femme  (ed.  Paris  et  Robert, 
SATF.,  Miracles,  v,  No.  xxxn),3  and  the  original  of  the  extant 
Italian  versions  of  the  story.  These  are  found  in  the  Reali  di 
Francia,  and  in  the  early  fourteenth-century  Italian  prose  ro- 
mance, Libro  di  Fioravante  (ed.  Rajna,  Collezione  di  Opere 
inedite,  1872),  which  Rajna  believed  was  the  source  of  the  first 
two  books  of  the  Reali  (Paris,  Rom.  111,  352).  In  chapters  17- 
60  of  the  Fioravante  are  set  forth  the  adventures  of  Fioravante ; 
in  61-77,  those  of  Drugiolina,  his  innocent  calumniated  wife, 
and  of  her  two  sons;  and  in  78-81,  the  Eastern  warfare  of 
Octavian,  one  of  those  sons.  The  Italian  stories  are  based  on 
French  originals,  and  the  Drugiolina  story  is  certainly  connected 
with  the  French  forms  of  Octavian. 

In  Middle  English  there  are  two  versions  of  the  Octavian 
story,  a  southern  one  now  contained  in  a  single  fifteenth-century 
manuscript  (A),  and  a  northern  one  found  in  two  manuscripts 
(C  L)  of  the  same  period.4  As  all  the  manuscripts  contain 
references  to  florins,  which  were  not  coined  in  England  until 
1343,5  Sarrazin  (pp.  xviii  and  xxxviii)  believed  that  the  two 
versions  were  composed  after  that  date,  but,  because  of  certain 
archaisms  in  the  text,  considerably  before  the  end  of  the  cen- 
tury. The  southern  version  of  1962  lines  is  written  in  six-line 
stanzas  riming  aaabab,  the  a  lines  in  iambic  tetrameter  and  the 
b  lines  in  iambic  dimeter,  the  form  subsequently  so  well  known 
in  Burns's  To  a  Mountain  Daisy.  The  northern  version  uses 
the  popular  twelve-line,  tail-rime  stanza,  and  frequent  allitera- 
tion. 

The  notable  differences  between  these  two  versions  in  dialect, 
verse  form,  and  to  a  minor  degree  in  context,  led  Sarrazin  (p. 
xxxix)  to  believe  that  they  were  entirely  independent  of  each 

3  Cf.  Petit  de  Julleville,  Les  Mysteres,  11,  306  ff.  In  the  South-English 
version,  the  supernatural  and  religious  element  appears  in  the  nightly  visions 
which  Florentyn  has  of  the  Virgin  who  bids  him  undertake  the  battle  with 
the  giant. 

4  Eule  (p.  16)   found  that  both  MSS.  were  from  the  same  source. 

5  Cf.  A.  Dodd,  History  of  Money,  1911,  p.  22.  Campbell,  Squyr  of  Lowe 
Degre,  1,  243,  note,  gives  the  date  1337. 


OCTAVIAN  269 

other,  but  in  his  later  study  Eule  (p.  16)  argued  that  both  were 
derived  from  the  same  Middle  English  source.  This  poem  Sar- 
razin  (pp.  xviii,  xxxviii)  and  Streve  thought  to  be  a  more  or  less 
close  translation  of  the  Anglo-Norman  Octavian.  Undoubtedly 
the  general  course  of  the  story  is  the  same,  but  Eule's  list  (pp.  8- 
1 1 )  of  traits  peculiar  to  the  English  versions  made  him  favor  the 
theory  of  a  lost  French  source.6  In  both  versions  the  characters 
and  tone  and  temper  of  the  story  are  Anglicized,  but  the  northern 
version  keeps  on  the  whole  closer  to  the  extant  French  poem  and 
in  Sarrazin's  opinion  is  fully  its  equal  in  quality.  He  was  inclined 
(p.  xliv)  in  fact  to  find  in  this  version  with  its  piety,  pathos, 
and  feeling,  one  of  the  best  pre-Chaucerian  poems,  and  to  attrib- 
ute it  to  the  clerical  author  who  wrote  Isumbras.  The  two 
poems  are  in  the  same  dialect,  are  found  in  the  same  Thornton 
manuscript,  have  a  certain  similarity  of  theme,  and  are  associ- 
ated in  the  condemnation  expressed  in  the  Speculum  Vitae. 
On  the  whole  this  ascription  seems  as  probable  as  Sarrazin's 
attribution  (p.  xxv-xxxi)  of  the  Southern  version  to  Thomas 
Chestre,  the  author  of  Sir  Launfal,  and  also,  according  to  the 
critic,  of  Libeaus  Desconus  (cf.  Billings,  Guide,  p.  142).  In 
Kaluza's  opinion  this  Southern  Octavian  was  Chestre's  best  and 
latest  work  (Eng.  Stud.,  xvm,  185-7). 

Origin.  After  an  account  of  the  long  childlessness  of  the 
rulers  of  Rome,  Octavian  tells  that  at  length  the  Empress  bears 
twins  and  is  accused  by  her  Cruel  Mother-in-law  of  adultery.7 
The  accusation,  the  stratagem  8  by  which  a  Pretended  Lover  is 
found  in  the  bed  of  the  Empress,  her  condemnation  and  expo- 
sure 9  with  her  children  in  a  forest,  follow  traditional  lines  of 

6  Sarrazin  (p.  xliv)  also  admitted  the  possibility.  He  could  not  other- 
wise explain  such  a  reference  as  that  in  the  northern  version  to  Borogh 
Larayne  (Bourg  la  Reine),  outside  of  Paris,  which  an  English  minstrel  was 
not  likely  to  know.  Eule  found  confirmation  of  his  theory  in  the  fact  that 
the  heroine's  confidant,  unnamed  in  the  French  versions,  was  called  in  the 
English  Olyue,  Olyuayne,  as  if  from  a  French  nominative  and  accusative 
case.  Streve  (p.  27),  noting  that  the  two  English  versions  alone  make  the 
Empress  daughter  of  King  Dagobert  of  France,  found  in  this  detail  an  evi- 
dence of  their  French  source. 

7  See  Lay  le  F reine  and  Knight  of  the  Swan,  note  5. 

8  See  Erie  of  Tolous,  note  8. 

9  See  Emare,  p.  23. 


2  7o  MEDIAEVAL   ROMANCE    IN    ENGLAND 

the  story  of  the  Innocent  Persecuted  Wife.  In  the  forest  she 
loses  her  children,  one  being  carried  away  by  an  ape,  the  other 
by  a  lion.  The  theft  of  the  children  by  robber  beasts,  an  epi- 
sode first  told  in  European  story  in  connection  with  the  Eustache 
legend,10  and  its  combination  with  the  story  of  the  Persecuted 
Wife  is  characteristic  of  Octavian.  The  remaining  part  of  the 
romance  tells,  with  details  varying  in  the  different  versions,  the 
adventures  of  the  mother  and  her  two  sons.  The  boys  grow 
up,  one  in  Paris,  and  one  in  the  East.  The  outstanding  features 
of  this  last  part  of  the  story  are  romantic  and  militant  with 
some  admixture  of  farce.  The  two  sets  of  stories  are  best 
considered  separately. 

In  the  story  of  Florent,  the  boy  is  rescued  by  a  knight  from 
the  ape  that  had  carried  him  off,  he  is  captured  by  robbers  and 
sold  by  them  to  Clement,  a  butcher  of  Paris.  The  boy's  noble 
lineage  is  revealed  by  his  impracticality ;  he  exchanges  good 
oxen  for  a  falcon  and  gives  an  absurd  price  for  a  young  colt. 
He  finally  proves  his  knightly  worth  by  the  prowess  with  which, 
despite  the  handicap  of  Clement's  wretched  armour,  he  over- 
throws a  great  giant,  the  champion  of  the  heathen  Sultan  who 
comes  to  besiege  Paris.  By  this  battle  Florent  wins  the  love 
of  the  Sultan's  daughter,  and,  presently,  with  her  own  conni- 
vance, he  carries  her  off  and  marries  her.11  Throughout  this 
part  of  the  story,  in  the  French  and  English  versions  at  least, 
comedy  is  introduced  by  way  of  the  good-hearted,  typical 
bourgeois  Clement,  who  lustily  beats  Florent  for  his  lordly  ex- 
travagance, swells  with  pride  over  his  victory,  and  in  canny 
peasant  fashion  (C  1285;  L  1070)  carries  off  the  mantles  of  the 
French  nobles  till  they  have  paid  for  the  feast.12  In  one  comic 
adventure  Clement  rides  off  with  the  wondrous  horned  horse 

10  See  Isumbras,  p.  7. 

11  Florent's  prowess  in  battle  and  his  love  affair  with  the  Sultan's  daughter 
were  possibly  drawn  from  the  French  chanson  de  geste,  Floovent.  Cf. 
Gautier,  Bibliographie,  p.  102 ;  Rajna,  Reali,  1,  77.  The  extant  MS.  of 
Floovent  is  of  the  fourteenth  century,  but  Darmsteter  (De  Floovante,  Paris, 
1877)  proved  that  the  original  version  was  of  the  twelfth  century.  The 
earliest  extant  foreign  redaction  is  a  thirteenth-century  Dutch  version.  Cf. 
Voretzsch,  Einfiihrung,  p.  207;  G.  Brockstedt,  Floovent  Studien,  Kiel,  1907, 
p.  6. 

12  This  scene  offers  a  comic  contrast  to  that  in  Auberi,  where  the  nobles 
not  only  pay  magnificently  for  their  expenses,  but  disdain  even  to  pick  up 
their  costly  mantles.    See  Richard  Coeur  de  Lion  here. 


OCTAVIAN  271 

of  the  Sultan  from  before  the  very  eyes  of  the  Saracen  (C  1440). 
The  contrast  between  knight  and  peasant  which  all  this  develops 
is  in  the  typical  manner  of  the  chansons  de  geste. 

Octavian,  the  son  who  is  carried  away  by  a  lion,  is  rescued 
by  his  own  mother.  The  lion,  which  had  been  carried  by  a 
griffin  to  an  island,  there  kills  the  griffin  and  henceforth  cares 
for  the  child  as  for  its  own  whelp.  On  a  pilgrimage  Octavian's 
mother,  hearing  from  sailors  of  the  child  in  the  lion's  den,  goes 
to  it,  finds  it  to  be  her  own  son,  and  takes  it  up.  The  lion  follows 
her,  to  the  consternation  of  the  sailors,  even  on  board  ship. 
From  this  time  on  he  serves  as  the  boy's  constant  companion 
until  years  later  he  is  killed,  fighting  in  battle  by  the  young 
knight's  side.  The  curious  story  of  the  Faithful  Lion  was  bor- 
rowed in  the  original  version  of  Octavian  either  from  one  of 
the  early  saint  legends,  or  from  some  Crusader's  story,  or,  as 
seems  most  probable,  from  contemporary  romances  in  which  the 
episode  of  a  knight  to  whom  a  faithful  lion  was  attached,  had 
achieved  increasing  popularity.13  It  is  the  most  characteristic 
feature  of  the  story  of  Octavian,  whose  subsequent  history  is, 
in  all  the  versions  except  the  French  chanson  de  geste,  briefly 
passed  over.  When  he  hears  of  the  great  defeat  in  which  the 
King  of  France,  his  visitor,  Octavian,  the  Emperor  of  Rome, 
and  hosts  of  Christians,  among  them,  the  young  hero  Florent, 
have  been  captured,  the  young  Octavian  leads  forth  the  armies 
of  the  King  of  Jerusalem,  his  own  kindly  protector.  His  vic- 
tory over  the  Saracens,  his  release  of  the  prisoners,  the  final 
defeat  of  the  Sultan,  the  reunion  of  the  long-separated  family 
of  the  Emperor  Octavian,  make  the  inevitable  sequel.  In  the 
French  forms  all  this  is  a  matter  of  some  five  thousand  lines; 
in  Middle  English  it  is  reduced  to  about  five  hundred  lines. 

The  comparative  insignificance  of  the  part  played  by  the  two 
Octavians,  father  and  son,  in  the  romance  of  Octavian,  is  the 
best  answer  to  any  attempt  to  find  there  the  romantic  trans- 
formation of  historical  characters  or  events.  Settegast's  argu- 
ment (pp.  52-57)  that  the  wars  of  Octavian  in  the  East,  as  set 

13  Cf.  Brown.  Twain,  p.  132  (Harvard  Studies,  vin),  who  referred  for 
the  saint  legend  to  Maury,  Croyances  et  Legendes  du  Moyen-Age,  Paris,  1897, 
p.  247.  See  also  the  Erie  of  Tolous,  note  10;  Guy  of  Warwick,  note  17.  In 
the  main  these  are  stories  of  Grateful  Beasts,  but  in  Octavian  the  lion's  de- 
votion is  not  thus  motived. 


272  MEDIEVAL   ROMANCE    IN    ENGLAND 

forth  in  the  Fioravante,  were  based  on  remembrances  of  the 
eastern  wars  of  the  Emperor  Julian,  rested  on  no  stronger  evi- 
dence than  that  history  and  romance  alike  describe  the  getting 
of  foreign  troops  and  the  holding  of  a  council  of  war,  and  on 
the  dubious  identification  of  Marzadonia  (Fioravante,  ch.  78) 
as  Marcianopolis,  which  was  besieged  by  the  Romans  in  376- 
77.  His  belief  (pp.  58-64)  that  the  chief  personages  of  the 
Octavian  story  are  to  be  identified  as  the  Emperor  Octavian 
Augustus  and  his  family  involved  such  peculiar  combinations  of 
historic  names  and  personalities  as  to  be  wholly  unconvincing. 
Octavian  is  made  up  of  pure  romance  themes,  the  very  use  of 
which  implies  a  long  previous  development  of  romantic  material. 
The  source  references  in  the  various  versions  of  Octavian 
offer  room  for  conjectures  rather  than  conclusions.  The  A  ver- 
sion in  Old  French  refers  to  "  merueilles  .  .  .  de  latin  en  ro- 
manz  traites "  concerning  the  reign  of  Dagobert,  and  to  an 
"  estoire  "  (1.  85)  concerning  his  contemporary,  Octavian,  the 
Emperor  of  Rome.  Similarly  the  South  English  version  twice 
(1.  935;  1.  1359)  refers  to  a  Latin  source,  "  as  seyd  the  Latyn," 
perhaps  by  way  of  reference  to  the  allusion  in  the  French  text, 
perhaps  because  of  the  convenience  of  "  Latyn  "as  a  rime  word. 
The  author  may,  however,  have  believed  sincerely  enough  that 
originally  the  story  was  told  in  Latin,  in  those  "  Bukes  of 
Rome  "  to  which  the  North-English  version  (1.  10)  makes  ref- 
erence. The  parallelism  between  the  episode  of  the  Loss  of  the 
Children  in  the  romance  and  in  the  Latin  legend  of  St.  Eustache 
may  have  lent  some  color  to  this  supposition. 


BIBLIOGRAPHY 

Texts:  (i)  A,  Southern  version  (1962  11.),  Brit.  Mus.,  Cott.  Cal.  A, 
11,  ed.  by  Weber,  in,  157-239  (1810);  cf.  Ward,  Cat.  of  Romances,  1, 
762;  (2)  C,  Cambridge  Univ.  Lib.  Ff.  11,  38  (1731  lines),  ed.  by  Halli- 
well,  The  Romance  of  the  Emperor  Octavian,  Lond.,  1848;  (3)  L, 
Thornton  MS.,  Lincoln  Cathedral,  A,  5  (1629  lines).  All  three  manu- 
scripts edited  by  G.  Sarrazin,  Altenglische  Bibliothek,  m,  Heilbronn, 
1885. 

Studies,  etc.  Cf.  Gautier,  Bibliog.  des  chansons  de  geste,  pp.  103-4; 
Wells,  Manual,  p.  782. 


OCTAVIAN 


2  73 


Eule,  R.     Untersuchungen  uber  die   nordengl.     Version  des  Octavian. 

Diss.  40  pp.    Halle,  1889. 
Sarrazen,  G.     (See  Texts)  Introd. 
Settegast,  F.    "  Floovent  u.  Julian  nebst  einem  anhang  liber  die  Oktavien- 

sage"  (pp.  52-64),  Beihefte  z.  Zts.  /.  rom.  Phil,  ix  (1906). 
Siefken,  0.    Das  gediildige  Weib,  pp.  48-61.   Dresden,  1904. 
Streve,  P.    Die  Octaviansage.    Halle,  1884. 


SIR   EGLAMOUR 

Versions.  The  Middle  English  Eglamour  makes  many  refer- 
ences to  an  original  book  or  written  source  and  in  two  passages 
(vv.  712,  859)  calls  this  the  "  Buke  of  Rome,"1  —  by  an  apoc- 
ryphal reference  probably  borrowed  along  with  much  else  from 
Octavian.  The  imitation  of  French  names,  such  as  the  hero's 
actual  or  assumed  name,  Eglamour  or  Auntour,  that  of  his  lord, 
Pryncesamour,  or  of  the  Emperor's  daughter,  Dyatour  (v.  771), 
is  too  naively  like  Chaucer's  humorous  Pleyndamour  (Sir 
Thopas,  2088)  to  make  necessary  the  assumption  of  a  French 
source.  In  short,  Eglamour,  compounded  of  incidents  familiar- 
ized in  Middle  English  romance  before  the  middle  of  the  four- 
teenth century,  is  best  considered  as  the  original  work  of  some 
unknown  Englishman.  His  stereoptyped  phrases,  his  frank 
commendation  of  gift-giving  to  minstrels  "  Dat  dey  myght  de 
better  bee,"  suggest  for  author  the  minstrel  rather  than  the 
monk  or  clerk  (Adam,  p.  xxvii).  A  certain  homely  flavor  is 
occasionally  given  to  his  style  by  his  liking  for  proverbial 
phrases  such  as,  "  De  man,  Dat  hewes  over-hey,  —  The  chyppes 
falles  in  his  eye,"  or  by  such  touches  of  rough  humour  as  come 
in  the  giant's  lament  for  the  great  bear  that  was  his  "  littill 
spotted  hogelyn." 

A  single  leaf  of  a  manuscript  (S)  written  late  in  the  fourteenth 
century  is  the  oldest  extant  text  of  Eglamour.  We  have  also 
three  fifteenth-century  manuscripts  (LCF),  five  sixteenth-cen- 
tury editions  (ebwad),  and  the  transcript  of  a  sixth  in  the  Percy 
Folio.  Of  all  these  the  Thornton  manuscript  (L)  has  the  best 
and  fullest  text.  It  contains  1335  lines  in  113  twelve-line 
stanzas  and  was  written  about  1440.  From  a  detailed  study  of 
this  text  Schleich  concluded  that  L  was  an  independent  version 
of  the  source  which  also  gave  rise  to  M,  the  conjectured  original 
of  all  the  other  texts.  The  source  of  the  two  primary  versions, 
M  and  L,  was,  it  is  believed,  a  poem  ascribed  by  Zielke  (p.  47) 

1  See  Emare,  ed.  Rickert,  p.  xlviii,  n.  2,  for  conjectures  about  the  "Buke 
of  Rome."     Cf.  Torrent  of  Partyngale  here,  note  4. 

374 


SIR  EGLAMOUR  275 

to  the  border  of  the  north-west  Midland  district,  and  written 
between  1380  and  1400.  The  date  might  be  more  exactly  de- 
termined if  it  could  be  decided  whether  the  author  in  lines 
1273-74,  "It  es  sothe  sayd  —  That  ofte  metis  men  at  unsett 
stevyn,"  was  quoting  Chaucer's  lines  (663,  667)  in  the  Knight's 
Tale,  "  But  sooth  is  seyd,  gon  sithen  many  yeres,  —  For  al-day 
meteth  men  at  unset  steuene."  But  it  is  hardly  more  safe  to 
assert  that  this  proves  Eglamour  to  have  been  written  after 
Chaucer's  poem  than  to  be  sure  with  Skeat  that  the  phrase  was 
used  independently  by  the  two  writers.2 

In  addition  to  the  narrative  versions  of  Eglamour  already 
noted,  the  history  of  the  romance  may  be  traced  in  various 
dramatic  forms.  The  play  of  Eglamour  and  Degrebelle  was 
given  at  St.  Albans  in  1444,  and  in  1580  Sidney's  satirical  formula 
for  a  popular  play  so  strikingly  parallels  the  plot  of  Eglamour 
that  it  is  not  improbable  that  he  was  familiar  with  some  drama- 
tization of  the  romance  (Baskervill,  pp.  467,  491).  The  play 
was  certainly  known  in  Germany  in  the  seventeenth  century 
though  generally  under  the  name  of  the  heroine,  Christabella. 
In  these  dramatic  versions  for  the  fairy  tale  motives  of  the  first 
half  of  the  romance  was  substituted  a  story  of  love  and  knightly 
adventure  (Baskervill,  p.  760). 

Origin.  The  first  half  of  the  romance  tells  of  the  three  great 
tasks  imposed  on  the  Love-Sick  Eglamour,  a  knight  "  of  lytill 
lande,"  before  he  can  win  Christabelle,  daughter  of  Pryncesamour, 
Earl  of  Artois.  The  hero  kills  a  giant  near  Artois,  goes  a  seven 
weeks'  journey  into  "  Sedoyne,"  kills  a  wild  bear  and  its  gigan- 
tic master,  Marasse,  goes  home  like  Bevis  with  his  victims' 
heads  on  his  spear ;  then  after  twelve  weeks'  stay,  his  "  bonys 
for  to  reste,"  he  slays  a  fiery  serpent  near  Rome.  Triumphal 
processions  are  made;  bells  are  rung  for  his  sake;  the  Princess 
Organata  of  Sidon  is  offered  to  him  and  is  refused;  he  receives 
from  her  a  safety-ensuring  ring ;  he  is  healed  of  a  deadly  wound 
by  the  Princess  Dyatoure  of  Rome.  As  the  tasks  and  these  asso- 
ciated details  are  the  veriest  commonplaces  of  folk-tales,  the 

2  Cf.  Haeckle,  Das  Sprichwort  bei  Chaucer,  Leipzig,  1890,  p.  22.  The 
New  Eng.  Diet.  (Steven)  has  a  number  of  instances  showing  the  ME.  use 
of  "  to  set  a  steven,"  or  "  at  unset  steven,"  but  there  is  no  other  example 
of  such  close  parallelism  as  that  between  Chaucer's  lines  and  Eglamour. 


276  MEDIEVAL   ROMANCE    IN   ENGLAND 

resemblance  between  this  part  of  Eglamour  and  the  ballad  of 
Sir  Cawline  (Child,  No.  61)  is  to  be  accounted  for  on  the  basis 
of  popular  common  sources  rather  than  by  any  consideration  of 
the  ballad  as  a  version  of  the  romance.3 

The  second  part  of  Eglamour  concerns  the  adventures  of 
Christabelle,  who  bears  a  child  in  Eglamour's  absence,  is  set 
adrift  on  the  sea  by  her  Cruel  Father,  and  is  rescued  only  after 
she  has  seen  a  griffin  carry  away  the  child  she  has  wrapped  in 
her  red  mantle.  Exhausted  and  speechless  she  comes  at  length 
to  her  uncle,  the  King  of  Egypt.  With  him  she  lives  for  fifteen 
years  until  she  is  won  in  a  tournament  and  is  married  to  her 
own  son,  Degrebelle,  who  has  been  nobly  reared  by  the  "  King 
of  Iraelle."  When  through  his  arms,  which  bear  the  picture  of 
a  griffin  carrying  a  child,  she  recognizes  her  son,  he  refuses  to 
give  her  up  except  to  a  man  who  can  overthrow  him ;  a  tourna- 
ment is  proclaimed,  and  thither  comes  Eglamour,  who  has  been 
fighting  for  fifteen  years  in  the  Holy  Land.  He  overthrows  his 
son  and  is  recognized  by  Christabelle  when  he  explains  why  his 
shield  bears  the  device  of  a  lady  and  child  in  a  boat.  The 
wedding  of  the  long-lost  lovers  is  celebrated,  and  young  Degre- 
belle receives  the  very  maiden  Organata  who  was  offered  to  his 
father  so  many  years  before.  Here,  as  always  in  mediaeval  ro- 
mance, ladies  are  immortally  young. 

The  "  patchwork  "  character  of  this  part  of  the  story  is  evi- 
dent (Gerould,  p.  440).  The  exposure  of  the  heroine  and  her 
child,  her  seven  days'  woe  on  the  sea,  her  kindly  reception  in  a 
foreign  land,  come  clearly  from  the  exposure  of  the  Calumni- 
ated Wife  in  the  Constance-Emare  type  of  story.  The  loss  of 
the  child4  wrapped  in  the  red  mantle  and  golden  girdle,  is  an 

3  Cawline  also  begins  with  the  episode  of  the  Love-Sick  Knight  visited 
by  his  lady  who  comes  at  her  father's  command.  She  reproaches  her  lover 
and  bids  him  rise  and  fight  with  the  Eldridge  knight.  The  description  of 
the  combat  in  Cawline  seems  to  be  connected  with  that  in  the  romance  of 
Eger;  it  is  followed  by  a  combat  with  a  giant  whom  Cawline  kills  with  his 
"  Eldridge  "  sword,  even  as  Eglamour  kills  his  giant  with  a  sword  brought 
from  "  the  Grekes  "  sea  and  given  him  by  Christabelle.  The  successive  fights 
of  a  hero  with  a  wild  beast  and  its  gigantic  master  are  found  in  Sir  Lionel 
(Child,  No.  18,  vol.  1,  209),  another  ballad  which  has  much  in  common  with 
Eglamour.  In  the  C  version  of  Lionel,  the  giant's  place  is  taken  by  a  "  wild 
woman  "  whose  lament  for  her  "  pretty  spotted  pig  "  may  be  compared  with 
that  of  the  giant  in  Eglamour  (1.  545). 

4  Gerould   (p.  441)    rightly  insisted  on  the  incidental  nature  of  the  epi- 


SIR   EGLAMOUR  277 

episode  which  represents  the  unification,  so  to  speak,  of  two 
episodes,  one,  the  Loss  of  the  Children,  a  regularly  recurrent 
incident  in  the  Eustache  legend  and  its  romantic  derivatives, 
and  the  other,  the  occasional  Treasure  theme  of  the  same  set 
of  stories.  But  one  need  go  no  further  back  than  the  Middle 
English  romance  of  Isumbras  to  find  in  close  sequence  the 
episodes  of  the  treasure  wrapped  in  a  red  mantle  carried  away 
by  an  eagle,  and  of  the  child  stolen  by  a  wild  animal.  The  in- 
cident also  occurs  in  Octavian,  and  from  this,  perhaps,  the  Egla- 
mour poet  derived  the  idea  of  making  the  robber  beast  a  griffin.5 
Like  Octavian,  Eglamour  similarly  combines  the  story  of  a  per- 
secuted lady  and  her  children  with  heroic  exploits  of  a  type  far 
more  romantic  than  the  Crusading  adventures  of  Isumbras  or 
his  prototype,  St.  Eustache.  In  Octavian  these  exploits  are  ac- 
complished by  the  sons  of  the  heroine;  in  Eglamour  by  her 
husband,  but  they  are  sufficiently  alike  to  suggest  direct  rela- 
tionship between  the  two  romances.  Between  Eglamour's  com- 
bat with  the  giant  Marasse,  for  instance,  and  that  of  Florent 
with  the  gigantic  champion  of  the  Sultan,  there  is  some  corre- 
spondence not  only  of  incident  but  even  of  phrase,  at  least  in 
the  versions  contained  in  the  Thornton  manuscript.6 

The  feebler  poet  of  Eglamour  seems  also  to  have  been  influ- 
enced by  the  Middle  English  romance  of  Sir  Degare  from  which 
he  drew  the  (Edipus-like  episode  of  the  marriage  of  a  mother 
and  son.  In  both  cases  the  discovery  of  the  relationship  pre- 
vents the  tragic  sequel  of  the  classic  story.  From  Degare,  too, 
it  seems  probable,  came  the  hint  for  the  Father  and  Son  combat 
which  in  Eglamour  follows  the  son's  proclamation  of  a  tourna- 
ment for  the  awarding  of  his  mother's  hand.7    Another  instance 


sode  of  the  Lost  Child  in  both  Eglamour  and  the  romance  of  Sir  Torrent. 
Neither  romance  has  any  direct  connection  with  the  Eustache  legend,  though 
both  have  often  been  said  to  be  derived  from  it.  Cf.  Adam,  Torrent,  p. 
xxiv;  Holland,  Chretien  de  Troyes,  Tubingen,  1854,  wno  held  that  Octavian, 
Eglamour,  and  Torrent,  were  all  derived  from  the  legend. 

5  Gerould,  ibid.,  noted  that  the  griffin  appears  only  in  Octavian,  in 
Uggieri  il  Danese  {Rom.  iv,  401-2),  which  has  borrowed  the  episode  from 
Fioravante,  in  the  Italian  version  of  Octavian,  and  in  Torrent. 

6  In  both  romances  the  Single  Combat  takes  place  before  the  walls  of 
a  city,  and  much  is  said  of  the  terror  of  the  townspeople  when  the  giant 
comes.  Cf .  Eglamour,  547,  "  the  giant  on  the  wallis  dange  " ;  Octavian,  v. 
740,  "  the  walles  doune  gan  he  dynge." 

7  Cf.  Murray,  Sohrab  and  Rustum,  p.  53,  for  Eglamour. 


278  MEDLEVAL   ROMANCE   IN   ENGLAND 

of  the  Eglamour  poet's  lack  of  inventive  faculty 8  is  his  repeti- 
tion of  the  scene  in  which  the  heroine  recognizes  her  husband 
as  she  has  recognized  her  son,  by  his  curious  symbolic  arms. 

BIBLIOGRAPHY 

Texts:  (i)  S,  Duke  of  Sutherland's  MS.  (lines  1-160)  now  Egerton 
2862,  cf.  Kolbing,  Eng.  Stud.,  vn,  193;  Schleich  (see  below)  p.  91; 
Brit.  Mus.  Cat.  of  Add.  MSS.  1910,  p.  239;  (2)  L,  Lincoln  Cathedral, 
A,  1,  17,  extracts  printed  by  Halliwell,  Thornton  Romances,  pp.  273-87; 
A.  S.  Cook,  Sir  Eglamour,  N.  Y.  191 1;  (3)  F,  Camb.  Univ.  Lib.  Ff.  11. 
38,  pr.  by  Halliwell,  ibid.,  pp.  121-76;  (4)  C,  Cott.  Cal.  A,  11.  (131 1 
lines);  cf.  Ward,  Cat.  of  Romances,  1,  766;  820;  (5)  P,  Percy  Folio,  ed. 
Hales  and  Furnivall,  11,  341-89.  Early  Editions:  (6)  e,  Advocates 
Library,  Edin.,  edition  printed  by  W.  Chepman  and  A.  Myllar,  Edin., 
1508;  repr.  Laing,  The  Knightly  Tale  of  Golagros  and  Gawane,  Edin., 
1827;  (7)  b,  fragments  of  a  book  printed  by  R.  Bankes,  Lond.  cir. 
1530,  repr.  Hall,  Archiv.  xcv,  308-11;  (8)  w,  Brit.  Mus.,  edition  pr. 
by  J.  Walley,  cir.  1540,  extracts  pr.  by  Laing  (see  under  e),  and  sum- 
marized by  Ellis,  Specimens,  pp.  527-38;  (9)  a,  Brit.  Mus.,  edition  by 
Wm.  Copland,  1548-61;  (10)  d,  Oxford,  Douce  261,  edition  pr.  in  1564; 
(11)  p,  Percy  Folio  MS.  ed.  Hales  and  Furnivall,  11,  338.  A  critical 
edition  of  the  poem  based  on  L  was  published  by  G.  Schleich,  in 
Palaestra,  Lin,  Berlin,  1906.  Rev.  Archiv.  cxvin,  441;  Eng.  Stud., 
xxxix,  433. 

Studies:   Wells,  Manual,  p.  781. 

Adam,  E.     (See  Torrent)  Introd.  pp.  xxiv-xxxii. 

Baskervill,  C.  R.  (1)  "  Some  Evidence  for  Early  Romantic  Plays  in 
England,"  Mod.  Phil,  xiv,  229  ff.  (1916).  (2)  "An  Elizabethan 
Eglamour  Play,"  ibid.,  p.  759  ff. 

Gerould,  G.  H.     "  The  Eustace  Legend,"  PMLA.  xix,  439-41   (1904). 

Schleich,  G.  See  Texts.  See  also  "  Ueber  die  Beziehungen  von  Egla- 
mour u.  Torrent."    Archiv.  XCII,  343-66   (1894). 

Siefken,  0.    Das  geduldige  Weib,  pp.  52-6  (1904)- 

Zielke,  A.     Untersuchungen  zu  Sir  Eglamour.    Diss.  60  pp.    Kiel,  1889. 


8  The  Eglamour  poet  repeats  somewhat  helplessly  such  hackneyed  phrases 
as  "white  as  foam,"  11.  25,  638;  "white  as  flour,"  11.  184,  920,  1210;  "white 
as  whalesbone,"  11.  680,  780,  1053;  "white  as  swan,"  1.  1284,  etc. 


TORRENT   OF   PORTYNGALE 

Versions.  A  single  fifteenth-century  manuscript  and  a  few 
fragments  of  a  sixteenth-century  edition  contain  the  only  known 
version  of  the  long-winded  romance,  Torrent  of  Portyngale.  It 
is  a  poem  of  2669  lines  written  in  the  tail-rime,  twelve-line 
stanza  form,  but  with  so  many  imperfections  of  rime,  metre, 
and  stanzaic  structure,  that  it  was  possible  for  Halliwell,  who 
first  edited  the  text,  so  to  misunderstand  its  nature,  as  to  print 
it  throughout  in  six-line  stanzas.  In  a  second  edition  of  the 
poem  Adam  restored  the  basic  stanza  structure  and  many  of 
the  original  rimes  which  the  scribe  who  wrote  the  Chetham  text 
had  often  misplaced.  The  scribe  was  either  superlatively  care- 
less and  more  to  be  cursed  than  Chaucer's  Adam;  or  else,  as 
Halliwell  conjectured  (Preface,  p.  v),  he  was  writing  the  poem 
down  from  oral  recitation.  It  has  innumerable  small  variations 
from  the  better  text  preserved  in  the  fragments  of  the  early 
edition,  but  unquestionably  scribe  and  printer  worked  from  the 
same  original  version  of  the  poem.  Adam  (p.  xvi)  ascribed  the 
original  to  the  eastern  border  of  the  East-Midland  district. 
The  text  contains  a  number  of  fairly  archaic  words,  but  the 
style  and  content  of  the  romance,  and  especially  its  connection 
with  Sir  Eglaniour,  make  it  probable  that  Torrent  was  a  rather 
late  fifteenth-century  composition.  As  it  is  so  full  of  references 
to  the  saints,  of  pious  invocations  and  prayers,  the  hero  never 
beginning  an  exploit  without  prefacing  it  with  a  proper  petition, 
as  it  makes  rather  frequent  mention  of  the  rites  of  the  church,1 
and  tells,  not  of  gift-giving  to  minstrels  but  of  thank  offerings 
to  churches,  Adam  (p.  xx)  thought  it  may  have  been  composed 
by  a  monk  or  some  other  pious  cleric. 

Origin.  The  presence  of  a  few  French  names  and  phrases 
in  Torrent  is  inadequate  evidence  that  there  was  ever  a  French 

1  Masses,  Confession,  Baptism,  etc.  Few  scenes  in  the  poem  are  more 
absurd  than  the  one  in  which  the  princess  grieves  for  her  lost  children  be- 
cause she  does  not  know  how  they  will  be  baptized  (1.  1892). 

279 


28o  MEDIEVAL   ROMANCE    IN   ENGLAND 

original,  especially  in  view  of  the  extraordinary  likeness  be- 
tween Torrent  and  Eglamour.  In  large  outline  the  plots  are 
identical ;  2  Torrent  loves  Desonelle,  Princess  of  Portugal,  and 
her  Cruel  Father  imposes  five  tasks,  the  killing  of  far-off  giants, 
upon  his  vassal  knight.  Incidentally,  the  hero  kills  dragons  and 
is  offered  the  hands  of  numerous  princesses.  In  his  absence 
Desonelle  is  delivered  of  twins,  is  exposed  on  the  sea,  loses  the 
children  to  robber  beasts,  and  is  ultimately,  after  the  boys  have 
grown  to  young  manhood,  reunited  to  them  and  to  Torrent. 
Despite  entire  change  of  names  and  setting,  and  the  intrusion 
of  many  superfluous  details,  such  as  the  description  in  every  in- 
stance of  the  giant's  castle  which  Torrent  wins  and  of  the  pris- 
oners whom  he  releases,  Torrent  is  so  close  to  Eglamour  in 
incident,  in  the  principal  characters,  even  in  phraseology,3  that 
it  must  be  either  from  the  same  source  as  Eglamour 4  or  simply 
an  amplified  redaction  of  the  romance  itself.  This  last  possi- 
bility was  accepted  by  Halliwell  {Thornton  Romances,  p.  xxii), 
disputed  by  Adam  (p.  xxvii)  in  favor  of  the  first  hypothesis, 
and  again  and,  it  would  seem,  authoritatively,  accepted  by 
Schleich.  In  the  matter  of  phraseology  Schleich  made  the  im- 
portant point  that  Torrent  is  closer  to  the  later  versions  of 
Eglamour  than  to  the  more  vigorous  version  preserved  in  the 
Thornton  manuscript,  the  oldest  of  the  complete  texts  of  that 
romance.  The  verbal  correspondences  are,  he  thought,  too  many 
to  justify  the  belief  that  the  Torrent  poet  was  working  merely 
from  memory  either  of  Eglamour  or  of  its  source. 

In  the  second  part  of  Torrent  there  is  greater  divergence  from 
Eglamour,  chiefly  because  the  Torrent  poet  apparently  wished 
to  omit  the  episode  of  the  marriage  of  the  mother  and  son.    In 

2  The  minor  differences  (changes  of  name,  setting,  order  of  incidents, 
nature  of  presents,  etc.)  which  Adam  (p.  xxviii)  noted,  are  insignificant  in 
comparison  with  the  fundamental  likeness  of  the  two  romances.  Certain 
additions  in  Torrent  can  be  fully  paralleled  elsewhere.  The  scene  in  which 
a  virgin  princess  leads  the  hero  safely  between  two  lions  (v.  286),  records 
a  bit  of  folk-lore  found  also  in  Beves,  2392.  The  consternation  caused  by 
Torrent's  appearance  with  these  lions  (v.  387)  is  like  that  in  Octavian  when 
the  mother  and  child  and  robber  lion  appear  together.  For  Torrent's  Island 
Combat  see  Guy  of  Warwick,  note  11.  For  the  supposed  connection 
of  Torrent  with  the  Eustache  legend  see  Eglamour,  note  4. 

3  See  Adam,  p.  xxxi;  Schleich,  p.  364. 

4  Like  Eglamour  and  Octavian,  Torrent  likewise  refers  to  its  source  as 
the  "  Buke  of  Rome." 


TORRENT   OF   PORTYNGALE  281 

Eglamour,  it  will  be  remembered,  this  leads  directly  to  the 
tournament  in  which  the  hero,  overthrowing  his  own  son,  re- 
gains his  lost  lady.  In  Torrent  one  battle  and  two  tournaments 
are  needed  to  accomplish  the  same  result.  In  each  of  them 
Torrent  tilts  with  his  own  son,  and  this  triple  use  of  the  modi- 
fied Father  and  Son  Combat  is  simply  a  final  instance  of  the 
constant  tendency  of  the  Torrent  poet  to  double  or  triple  what- 
ever he  conceived  to  be  a  good  point  in  his  original.  It  is 
surely  in  reminiscence  of  Eglamour  that  he  describes  the  sym- 
bolic devices  borne  by  Torrent  and  his  sons,  although  he  forgets 
to  make  these  arms  the  means,  as  they  are  in  Eglamour,  of  the 
heroine's  partial  recognition  of  her  relatives. 

The  Torrent  poet's  zest  for  trivial  and  unrelated  detail,  a 
characteristic  woefully  true  of  the  later  romancers,  is  probably 
the  explanation  for  the  surprising  number  of  proper  names  which 
come  into  his  story.  Giants,  swords,  forests,  as  well  as  cities 
and  kingdoms,  alike  have  their  names  in  greater  number  pro- 
portionally than  occur  elsewhere.  Most  interesting  among 
them  are  the  references  to  Saint  Anthony,  who  rescues  Torrent's 
child,  to  Saint  Nicholas  de  Barr  (Bari,  Italy),  to  whom  Torrent 
makes  offerings,  and  to  Weland,5  as  the  maker  of  Torrent's 
sword,  Adolake  (Hathloke). 


BIBLIOGRAPHY 

Texts:  (i)  C,  Chetham  Libr.  Manchester,  ed.  J.  O.  Halliwell, 
Lond.,  1842;  by  E.  Adam,  EETSES.  li,  1887;  cf.  E.  Kolbing,  Eng. 
Stud.,  vii,  195;  344;  ibid.,  xn,  432;  notes,  Zupitza,  ibid.,  xv,  1-12;  (2) 
Fragments  of  an  early  printed  edition,  Douce,  Bodl.,  repr.  Halliwell, 
Appendix. 


5  Cf.  Halliwell,  Torrent,  p.  vii,  and  Zupitza,  "  Ein  Zeugnis  f.  die  Wieland- 
sage,"  Zts.  f.  deut.  Alterthum,  xix,  129.  Other  references  to  Weland  in 
Middle  English  literature  are  found  in  Layamon's  Brat  (ed.  Madden,  1847, 
11.  463,  21129)  and  in  Horn  Childe.  See  for  full  study  of  the  Weland  legend 
in  mediaeval  and  modern  literature  P.  Maurus,  "  Die  Wielandsage  in  der 
Literatur,"  Munchener  Beitrdge  z.  rom.  u.  engl.  Philologie,  xxv,  1-224  (1902), 
and  Die  Wielandsage,  Weitere  neuzeitliche  Bearbeitungen,  Munich,  iqio;  also 
G.  B.  Depping  and  F.  Michel,'  Wayland  Smith,  A  Dissertation  on  a  Tradition 
of  the  Middle  Ages,  Lond.,  1847;  and  Schofield,  PMLA.  xv,  172  (1900). 


282  MEDIAEVAL   ROMANCE   IN   ENGLAND 

Studies:    Wells,  Manual,  p.  782;  Korting,  Grundriss,  §113. 
Adam,  E.     See  Texts. 
Gerould,  G.  H.     See  Isumbras  here. 
Holthausen,   F.    "  Zu  mittelengl.    Romanzen,    Torrent   of  Portyngale" 

Anglia  xlii,  429-49  (1918). 
Schleich,  G.     See  Eglamour  here. 

Spence,  L.     Dictionary  of  Mediaeval  Romance  and  Romance  Writers, 
(Full  outline  of  Torrent).    London,  1913. 


SIR   TRIAMOUR 

Versions.  The  first  part  of  Sir  Triamour  (to  1.  612)  is  of 
special  interest  as  it  preserves  the  only  Middle  English  version 
of  the  well-known  French  story  of  Sebilla,  the  persecuted  wife 
of  Charlemagne,  and  of  the  Dog  of  Montargis.  The  original 
version  of  this  tale  of  the  Chaste  Queen  and  the  False  Steward 
seems  to  have  been  one  of  the  best  productions  of  the  trouveres 
who  at  the  end  of  the  twelfth  century  were  writing  of  what  may 
have  been  ancient  epic  stories.  One  version  in  French  alexan- 
drines survives  in  fragments  of  two  manuscripts,  one  edited  by 
Scheler  (Bull,  de  VAcademie  royale  de  Belgique,  2nd  Ser.  1875), 
and  the  other,  which  was  not  discovered  until  191 5-19 17,  by 
Baker.  A  resume  "  a  cantoribus  gallicis  "  was  given  in  the 
early  thirteenth-century  chronicle  of  Alberic  de  Trois  Fontaines 
(quoted  by  Guessard,  p.  xii),  who  thought  it  a  beautiful  story 
which  could  move  those  who  heard  it  to  laughter  and  to  tears. 
Another  poem  on  the  same  subject,  but  so  greatly  inferior  that 
Paris  (p.  395)  called  it  "  d'une  secheresse  incroyable,  d'une 
grossierte  qui  indique  l'extreme  decadence  de  l'art,"  is  the 
Franco-Italian  poem  known  as  Macaire 1  extant  in  the  thirteenth 
century  Venetian  manuscript 2  which  has  from  time  to  time 
engaged  so  much  laborious  study.3  In  this  the  queen  bears  the 
name  Blanciflor.  A  version  in  French  prose  is  preserved  in 
MS.  3351  of  the  Arsenal  and  in  Spanish  prose  in  a  fourteenth 

1  For  further  references  in  regard  to  this  and  other  versions  see,  Paris, 
Charlemagne,  pp.  389-95,  and  Baker,  Rom.  xliv,  p.  7. 

2  Edited  by  Mussafia,  Altfrz.  Gedichte,  Vienna,  1864,  who  thought  (p.  4) 
that  the  author  of  Macaire  had  shortened  his  French  original  to  suit  himself; 
and  again  edited  by  Guessard,  Les  Anciens  Poetes  de  la  France,  1866,  who 
believed  (p.  xiii)  that  the  author  used  a  very  old  short  poem  to  which  the 
author  of  the  story  known  to  Alberic  must  have  added  extensively.  The 
Anglo-Norman  fragments  published  by  Baker  correspond  to  vv.  912-920  and 
1007-66  of  Macaire. 

3  Reinhold,  Litter aturblatt,  xxxni,  col.  150,  announced  a  critical  edition 
of  the  whole  MS.  Cf.  Guessard,  Macaire,  p.  ciii;  Reinhold,  Zts.  /.  rom. 
Phil,  xxxv,  xxxvi. 

283 


284  MEDIEVAL   ROMANCE    IN    ENGLAND 

century  manuscript  of  the  Escurial,  and  in  the  printed  editions 
of  1532  and  1551  (Gautier,  in,  686-87;  Kohler,  11,  273  ff.).  To 
the  same  prose  version  belongs  a  Dutch  Volksbuch  printed  at 
Antwerp  in  the  first  part  of  the  sixteenth  century.  To  these 
texts  Child  {Ballads,  11,  p.  40)  added  two  in  German:  a  metri- 
cal tale  of  uncertain  date,  Diu  Kunigun  von  Frankreich  und  der 
ungetriuwe  Marschalk  (von  der  Hagen,  Gesammtabenteuer ,  1, 
169)  found  in  many  manuscripts;  a  meisterlied,  Die  Kunigun 
von  Frankreich,  dy  der  marschalk  gegen  dem  Kunig  versagen 
wart,  which  was  printed  in  the  fifteenth  century  (Wolff's  Halle 
der  V biker,  \\,  255).  Later  Hans  Sachs  dramatized  the  story 
of  the  false  marshall  (Keller,  vm,  54).  In  the  nineteenth  cen- 
tury, it  may  be  noted,  appeared  various  plays  in  French,  Spanish, 
and  English  which  made  melodramatic  use  of  the  Dog  of  Mon- 
targis  story  (Guessard,  pp.  lxv,  lxxiv). 

In  fourteenth-century  England  the  story  was  certainly  known, 
as  an  entry  shows  in  the  catalogue  of  the  books  of  Peterborough 
(Guessard,  p.  lxxi).  "  Macharie  and  Queen  Sible  and  the  Dog 
of  Montargis  "  makes  the  seventy-eighth  tale  in  the  fifteenth - 
century  manuscript  of  the  Anglo-Latin  Gesta  Romanorum,  now 
known  as  Additional  MS.  9066  (Herbert,  Catalogue,  in,  259). 
From  some  such  abbreviated  version  as  this  it  is  probable  the 
English  poet  of  Sir  Triamour  got  the  idea  for  the  first  part  of 
his  story,  though  in  it  none  of  the  characters  bear  their  original 
names.  The  first  English  metrical  version  seems  to  have  been 
written  about  1400  in  a  twelve-line  tail-rime  stanza.  Three 
manuscripts  of  it  are  extant,  the  earliest  (C)  belonging  to  the 
early  part  of  the  fifteenth  century,  and  the  other  two  (RP)  to 
the  sixteenth  and  the  seventeenth  centuries.  Copland  in  the 
sixteenth  century  twice  printed  the  romance  (LB).  In  his 
study  of  the  relationship  of  these  texts,  Bauszus  decided  that  C, 
through  some  lost  intermediary  version  (y),  was  from  the  same 
source  (x)  as  that  which  gave  rise  to  the  source  (z)  of  LBP. 
Bauszus  (p.  50)  thought  the  dialect  of  the  poem  more  nearly 
that  of  the  North  Midland  than  other,  but  he  found  in  the  text 
both  northern  and  southern  forms. 

The  style  of  the  first  part  at  least  of  Triamour  is  neither  dull 
nor  prolix,  though  the  author  uses  the  undistinguished  diction 
of  Oct  avian  and  E glamour  and  many  other  equally  common- 


SIR   TRIAMOUR  285 

place  romances  (Bauszus,  p.  42).*  The  second  part  of  the  poem 
is  perhaps  best  described  in  the  words  of  G.  W.  Hales  as  a  "  fair 
specimen  of  the  old  romances  with  all  their  vices  and  virtues, 
prolixity,  improbabilities,  exaggeration,  with  their  wild  graces 
also,  their  chivalrousness,  their  pageantry."  It  is  chiefly  inter- 
esting for  its  reminiscent  use  of  motifs  familiar  in  much  older 
texts. 

Origin.  The  Sebilla  story  opens  with  an  incident  but  slightly 
indicated  in  Triamour.  The  queen  is  accused,  because  Charle- 
magne finds  a  dwarf  in  her  bed.5  In  order  to  incriminate  her 
the  dwarf  has  placed  himself  there  at  the  instigation  either  of 
his  own  revengeful  passion  which  the  queen  has  scornfully  re- 
jected, or  because  of  the  bribes  of  Macaire,  the  treacherous 
father  of  Ganelon,  whose  guilty  passion  has  been  similarly 
scorned.  This  scene  is  omitted  by  the  English  poet,  who  tells 
briefly  of  the  accusation  brought  by  the  false  steward  Marrok 
against  the  wife  of  King  Ardus  of  Aragon,  although  she  was 
"  true  as  the  turtle  on  tree."  The  king  has  confided  to  Marrok 
the  care  of  his  queen  and  kingdom  while  he  himself  goes  on  a 
pilgrimage  in  the  hope  that  God  will  hear  his  prayer  for  a 
child.  The  False  Seneschal,  who  has  wooed  the  lady  in  earnest, 
but,  finding  her  true,  pretends  to  have  but  tested  her  loyalty, 
accuses  her  on  her  husband's  return  of  having  sinned  with  a 
knight.  On  Marrok's  advice  Ardus  banishes  the  queen  with 
a  single  old  knight,  Sir  Roger,  for  escort.    The  older  versions 

4  Such  lines  as  those  in  Octavian,  v.  283, 

"  They  riden   forth  to   a  wylde  forest 
There  was  many  a  wylde  best," 
and  those  in  Triamour,  v.  1033, 

"  He  saw  many  a  wylde  beest 
Both  in  heth  and  in  wylde  forest," 
suggest  Chaucer's  derisive  lines  in  Sir  Thopas, 

"  He  pricketh  thurgh  a  fair  forest 
Ther-inne  is  many  a  wilde  best." 

5  The  episode  of  the  Pretended  Lover  is  found  also  in  the  various  ver- 
sions of  the  Erie  of  Tolous  (cf.  note  8),  in  those  of  Octavian,  and,  as 
Child  (n,  No.  59)  pointed  out  in  connection  with  Sir  Aldingar,  also  in 
Doon  Alemanz,  and  in  the  versions  of  the  Macaire  story.  In  the  Didriks 
Saga,  cc.  156-59,  the  queen  Sisibe  is  entrusted  to  two  nobles,  one  of  whom, 
Hartvin,  tries  to  win  her  favor  and  is  threatened,  as  Margaret  threatens 
Marrok,  with  the  gallows.  The  king  is  then  told  that  his  wife  has  had  a 
thrall  for  lover  (Child,  n,  41). 


286  MEDIEVAL   ROMANCE    IN   ENGLAND 

are  full  of  epic  traditions  concerning  the  constant  hostility  to 
Charlemagne's  royal  house  of  Ganelon's  treacherous  family. 
Much  is  made  therefore  of  the  part  his  relatives  play  in  bribing 
the  dwarf,  in  urging  the  instant  execution  of  the  queen,  and  of 
the  difficulties  which  Naime,  one  of  Charlemagne's  wisest  coun- 
sellors, has  in  restraining  them.  It  is  at  his  advice  that  Auberi 
de  Montdidier,  one  of  the  Emperor's  best  vassals,  is  sent  to 
take  the  queen  to  the  frontier.  The  sorrowful,  suffering  lady, 
who  is  about  to  give  birth  to  a  child,  and  this  young  knight 
are  attacked  in  the  forest  by  Macaire,  and  Auberi  is  killed,  but 
not  before  his  valiant  efforts  have  given  the  queen  a  chance  to 
escape  into  the  woods.  The  English  version 6  at  this  point 
seems  merely  to  condense  its  French  original  to  which  refer- 
ence is  made  in  1.  316,  —  "  as  it  is  in  the  romans  tolde." 

The  episode  that  follows  in  the  French  story  sets  forth  the 
story  of  an  animal  faithful  even  unto  death.  Auberi's  dog,  which 
has  vainly  attempted  to  aid  his  master,  does  his  best  for  the 
dead  body,  covers  it  with  dirt,  and  guards  the  grave  until,  sent 
back  by  hunger  to  the  royal  palace,  he  arouses  suspicion  by  a 
violent  attack  on  Macaire.  The  dog  leads  the  courtiers  to 
Auberi's  grave,  and  again  on  the  advice  of  Naime,  the  suspected 
Macaire  is  forced  to  fight  in  a  judicial  combat  with  the  dog. 
The  fight  ends  with  the  man's  defeat  and  confession.  The  Eng- 
lish poet,  by  stupidly  deferring  the  dog's  attack  on  Macaire  for 
seven  years,  by  making  it  fatal,  and  having  in  consequence  to 
omit  the  famous  combat  between  the  man  and  dog,  has  done 
but  scant  justice  to  his  original. 

The  earliest  literary  treatment  of  an  analogous  story  is 
Plutarch's  tale  of  the  dog  which  King  Pyrrhus  found  guarding 
the  dead  body  of  its  master.  He  took  it  away,  and  later  at  a 
review  of  the  King's  soldiers  the  dog  identified  and  attacked  the 
murderer  of  his  master.  In  the  Hexameron  of  St.  Ambrose,  writ- 
ten in  the  fourth  century,  there  is  a  very  similar  tale  localized  in 
Antioch.  It  was  this  version,  presumably,  which  Giraldus  Cam- 
brensis  knew  and  introduced  into  the  Itinerarium  Cambriae,  c. 
1 188  (Lond.,  1585,  1,  124).     Giraldus's  is  the  first  extant  text 

6  The  account  in  Triamour  of  the  king's  resolve  to  exile  the  queen,  and 
of  how  she  is  provided  with  a  horse  and  a  few  florins,  and  of  the  grief 
of  the  people,  is  obviously  close  to  that  in  the  ME.  Octavian  (Weber,  in, 
265  ff.). 


SIR   TRIAMOUR  287 

which  introduces  the  typically  mediaeval  idea  of  the  Judicial 
Combat,  but  it  is  believed  {Historical  Litt.,  xxvi,  373)  that  he 
did  not  know  the  French  story  of  Auberi's  dog.  This  was,  how- 
ever, evidently  known  to  Alberic  not  more  than  sixty  years  later. 
Alberic  states  that  the  reason  for  Charlemagne's  dismissal  of 
his  wife,  the  daughter  of  Dedier,  the  Lombard  king,  was  un- 
known, but  that  he  sent  her  off  in  company  with  Auberi,  that 
Auberi  was  killed  by  Macaire  and  avenged  by  his  faithful  dog 
in  a  judicial  combat.  The  dog  episode  appeared  not  only  in 
the  versions  of  the  Sebilla  story  already  noted  but  in  French  art 
at  least  as  early  as  the  fourteenth  century.7  The  title  "  dog  of 
Montargis  "  comes  from  the  painting  in  the  hall  of  the  chateau 
of  that  name  (Paris,  p.  392,  n.  2). 

In  the  Sebilla  story  the  fleeing  queen,  after  her  escape  from 
Macaire,  encounters  Varocher,  a  poor,  kindly  giant  of  a  fellow 
who  acts  henceforth  as  her  protector.  Her  child  is  born  at  an 
inn  in  Hungary.  Her  royal  father  presently  receives  her  and 
decides  on  a  war  with  Charlemagne  in  order  to  avenge  her  in- 
juries. The  subsequent  attack,  the  grotesque,  gallant  deeds 
of  Varocher,  and  the  reconciliation  of  the  king  and  his  wife, 
contain  those  comic  parts  of  the  story  to  which  Alberic  referred. 
Of  all  this  there  is  nothing  in  the  second  part  of  Triamour. 
The  queen  gives  birth  in  the  forest  to  Triamour ; 8  she  is  found 
and  taken  home  by  a  kindly  knight,  Sir  Bernard  Messengere. 
The  birth  in  the  wood  recalls  that  of  Tristan  and  of  Josian's 
children  in  Beves.    Triamour,  like  many  another  poor  and  un- 

7  Guessard,  Macaire,  p.  xxix,  enumerated  various  references  to  the  story 
in  French  literature.  In  the  fourteenth  century  Deduits  de  la  Chase  by  Gui 
de  la  Buigne,  there  is  a  reference  to  painted  scenes  of  the  story.  Guessard 
repudiated  the  theory  that  there  was  ever  at  Montargis  a  Celtic  dog  cult 
which  might  have  given  rise  to  a  story  thus  lauding  a  dog.  The  etymology 
of  the  name  was  thus  explained,  Mont;  Celtic  ar,  French  du,  Celtic  ki, 
French  chien.  Of  general  interest  is  Baugert's,  Die  Tiere  im  altfrz.  Epos, 
Marburg,  1885. 

8  Bauszus,  p.  32,  compared  with  this  name  various  similar  ones  in  ME. 
romance,  Triamour,  the  name  of  the  fee  in  Chestre's  Launfal,  Pryncesamour 
in  Eglamour,  Segramour  in  Emare,  Pleyndamour  in  the  lost  romance  men- 
tioned by  Chaucer  in  Sir  Thopas,  and  the  lady  in  Libeaus  Desconus  "  that 
highte  la  dame  d'amour."  In  the  first  part  of  Triamour  the  original  French 
names  were  changed.  Macaire  became  Marrok;  Auberi,  Roger;  and  Joseran, 
the  protector  who  cares  for  the  heroine  after  the  birth  of  her  child,  Bernard 
Messengere  (Mowswinge). 


288  MEDIEVAL   ROMANCE   IN   ENGLAND 

known  knight,  wins  a  princess  "  fresshe  and  amerous  "  at  his  first 
tournament  but,  being  wounded,  rides  away,  leaving  his  reward 
unclaimed  for  a  year  and  a  day.  A  mild  version  of  the  Father 
and  Son  Combat  appears  in  this  tournament,  for  the  young  hero 
strikes  down  his  own  father,  whom  he  knows  no  more  than 
Degare  knows  his  father  in  the  forest  combat.  Triamour,  with 
his  father's  aid,  attacks  and  kills  the  cowardly  son  of  the  Ger- 
man Emperor.  The  young  hero  kills  cowardly  foresters  out  of 
hand  and  also  a  great  hart  that  has  harmed  his  greyhounds. 
Later  Triamour  becomes  the  champion  of  his  unknown  father 
in  the  fight  to  be  waged  against  the  Emperor's  challenger,  Mar- 
radas.  In  the  description  of  this  battle  occurs  an  episode 
strongly  reminiscent  of  Florent's  fight  with  the  Saracen  giant 
(Octavian,  Weber,  in,  1095),  —  when  Triamour  kills  his  oppo- 
nent's horse  and  is  taunted  by  him  for  the  accident.  The 
mighty  champion  Burlond,  of  whom  we  are  told  that,  when  his 
legs  were  cut  off  at  the  knee,  he  then  "  on  his  stumpes  stood," 
suggests  the  famous  squire  in  Chevy  Chase  (st.  50) :  when  "  his 
leggis  were  smitten  off,  /  He  fought  upon  his  stumpes."  At  the 
end  of  the  romance,  when  he  marries  the  princess,  Triamour, 
like  Degare,  has  the  satisfaction  of  uniting  his  long  separated 
parents. 


BIBLIOGRAPHY 

Texts:  (i)  C,  Cbg.  Univ.  Libr.  Ff.  n,  38,  ed.  J.  0.  Halliwell,  Percy  Soc. 
xvi,  1846;  (2)  R.  Rawlinson,  a  fragment  of  75  verses;  (3)  Additional 
MS.  27879,  Br.  Mus.,  a  17th  c.  MS.  containing  10  articles  of  the  Percy 
Folio  MS.  ed.  Furnivall  and  Hales,  1868,  11,  78-135;  (4)  L,  Copland's 
edition,  1593,  reprinted  by  Utterson,  1816,  1;  (5)  B,  Copland's  edition, 
undated,  Bodl.  Abstract,  Ellis,  Spec.  491;  Ashton  p.  171  ff. 

Studies:  Cf.  Edwardes,  Summary,  p.  171;  404;  Gautier,  Bibliog. 
des  Chansons  de  Geste  (Macaire,  p.  143),  Paris,  1897;  Wells,  120,  782. 
Baker,  A.  T.    "  Fragments  de  la  Chanson  de  la  Reine  Sibile,"  Romania 

xliv,  1-13  (1915-1917). 
Bauszus,  H.    Die  mitteleng.  Romanze  Sir  Triamour.  Diss.  58  pp.  Konigs- 

berg,  1902.     (Critical  text,  11.  1-132.) 
Bonilla  y  San  Martin,  A.    Libros  de  Caballerias,  Madrid,  1907,  1,  503  ff. 
Gautier,  L.     Les  Epopees  Francoises,  Paris,   1880.     See  Macaire,  ill, 

684-719. 


SIR   TRIAMOUR  289 

Kohler,  R.  "  Zu  der  altspan.  Erzahlung  von  Karl  u.  Sibille,"  Kleinere 
Schriften,  Berlin,  1900,  11,  273-304,  repr.  from  Jahr.  f.  rom.  u.  eng. 
Lit.  xii,  286  ff.  (1871). 

Paris,  G.    Historie  Poetique  de  Charlemagne.    Paris,  1905. 

Siefkin.    Das  geduldige  Weib,  pp.  62-66. 

Spence,  L.  Dictionary  of  Romance,  pp.  358-362.  (Outline  of  Tria- 
mour.) 


ROSWALL   AND    LILLIAN 

Versions.  In  1804  Sir  Walter  Scott  (Works,  1868,  v,  407) 
remarked :  "  Within  the  memory  of  man  an  old  person  used  to 
perambulate  the  streets  of  Edinburgh  singing  in  a  monotonous 
cadence,  the  tale  of  Rosswal  and  Lilian."  It  is  probable  that 
this  tale  was  some  such  abbreviated  version  of  the  romance 
as  that  in  the  extant  stall  copies  and  prints  of  the  eighteenth 
century,  —  texts  which  contain  about  four  hundred  lines.  The 
earlier  and  longer  version  from  which  these  must  have  been  de- 
rived is  now  represented  by  the  edition  printed  at  Edinburgh 
in  1663  (A),  by  a  later  edition  of  1679,  by  that  printed  at  New- 
castle, and  by  various  later  reprints  such  as  David  Laing's  in 
1822.  The  first  text  (A),  written  in  the  dialect  of  southern 
Scotland,  contained  846  lines  in  short  riming  couplets  (Lengert, 
Eng.  Stud.,  xvii,  360).  The  rimes  and  the  style  of  this  version, 
its  many  allusions  to  earlier  heroes  and  heroines  of  romance, 
show  that  it  could  not  be  dated  before  the  fifteenth  century.1 
On  the  other  hand,  the  story  must  have  been  fairly  well  known 
early  in  the  sixteenth  century  and  possibly  before  that  date 
since  it  had  then  passed  into  ballad  form.  In  1580  it  was  en- 
tered as  "  The  Lord  of  Lome  and  the  False  Steward  "  in  the 
Stationers'  Registers  (Arber,  n,  379),  and  a  few  years  later 
was  referred  to  as  an  old  ballad  "  of  king  Harrie's'  day."  The 
oldest  extant  text  of  this  ballad  version  is  in  the  Percy  Folio 
(ed.  Hales  and  Furnivall,  1,  180-98).  It  contains  some  tradi- 
tional material  not  found  in  the  romance  version,  but  the  gen- 
eral similarity  of  the  two  texts  and  their  identical  use  of  the 

1  There  are  three  catalogue  lists:  verses  15-24  say  that  Roswall  sur- 
passes Ulisses,  Gandifer,  Achilles,  Troyalus,  Priamus,  Clariadus,  the  fair  Philmox, 
Florentine  of  Almanie,  Lancelot  du  Lake;  verses  343-4 S  that  Lillian 
was  fairer  than  the  lady  Pelicane,  than  Helen  or  the  true  Philippie^  or  the 
lady  Christian;  and  in  verses  391-401  the  hero  is  besought  to  take  the  name 
of  Hector  or  Oliver,  or  Sir  Porteusor,  of  the  worthy  Amedus,  or  the  noble 
Predicase,  Sir  Lion-dale,  Florent  of  Albanie,  or  Lancelot  du  Lake.  See 
Rickert,  Notes,  for  tentative  identifications. 

290 


ROSWALL   AND   LILLIAN  291 

name  Dissawar,2  which  in  each  version  the  hero  assumes,  per- 
suaded Child  {Ballads,  v,  47)  that  the  ballad  was,  at  least  in 
part,  derived  from  the  romance. 

Origin.  Despite  the  lateness  of  its  versions  and  its  conven- 
tionalized style,  Roswall  and  Lillian  is  far  closer  to  primitive 
folk-tales  than  many  a  romance  of  much  earlier  date.  Its  pri- 
mary theme  is  that  of  the  Male  Cinderella,  i.e.,  of  the  royal 
youth  who  is  forced  to  become  a  menial  servant;  but  this  is 
combined  with  various  distinctive,  popular  motifs.  To  begin 
with,  Roswall  is  exiled  from  his  home  because  he  imprudently 
releases  three  of  his  royal  father's  prisoners.  Despite  an  attempt 
to  rationalize  these  characters,  they  have  the  same  function  as 
the  supernatural  Helpful  Companions  who  appear  so  often  in 
folk-tales.  In  the  group  of  tales  analyzed  by  Lengert  (p.  347  ff.) 
and  by  Child  (v,  45-7)  for  the  sake  of  their  likeness  in  this  in- 
cident to  Roswall,  only  one  grateful  being  appears ;  he  is  a  wild 
man  (Bosnian),  a  peri  (Tartar),  an  iron  man  (Der  Eisenhans, 
Grimm,  Kinder -Mar  chen,  No.  136 ;  cf.  Bolte-Polivka,  Anmer- 
kungen,  in,  94-114),  a  robber  of  fabulous  strength  (Russian), 
an  invisible  knight  (Polish).  In  the  ballad  version  of  Roswall 
this  episode  of  the  Released  Prisoners  is  omitted  and  the  boy's 
absence  from  home  is  accounted  for  in  remarkably  non-popular 
fashion  by  saying  that  he  was  sent  from  Scotland  "  to  learne  the 
speeches  of  strange  londs." 

In  the  romance  Roswall  is  accompanied  into  exile  by  a  False 
Steward,  who  takes  advantage  of  the  boy  when  he  is  drinking 
from  a  brook,  threatens  him  with  death,  and  robs  him  of  his 
gold  and  letters.  The  romance  offers  no  explanation  for  this 
villainy,  but  it  can,  perhaps,  be  found  in  five  of  the  twenty 
analogues  cited  by  Lengert.  These  begin  with  a  charge  laid 
on  the  hero  not  to  travel  with  a  beardless  man  or  one  deformed, 
and  the  boy's  disobedience  brings  his  troubles  upon  him.  The 
forbidden  person  appears  and  gets  control  over  the  lad  in  pre- 
cisely the  same  fashion  as   that  used   by   Roswall 's   faithless 

2  The  name  has  not  been  satisfactorily  explained.  Does  it  mean  Unaware 
(cf.  Percy  Folio,  Disaware)  ?  Is  it  from  Dis-avoir,  i.e.,  without  possessions 
(Child,  Glossary),  from  disavow  (Rickert),  or  formed  by  analogy  with  such 
a  name  as  that  of  Libeaus  Desconus?  For  the  somewhat  similar  use  of  such 
a  cognomen  see  Degare  and  Emare. 


292  MEDIEVAL   ROMANCE    IN    ENGLAND 

servant.  In  other  words  it  seems  possible  that  the  episode  origi- 
nally developed  from  the  currency  of  certain  superstitions  con- 
cerning types  of  people  whom  it  was  considered  unlucky  to 
encounter.3  But  in  both  Roswall  and  the  Lord  of  Lome  any 
distinctive  attribute  in  the  villain's  appearance  has  been  lost 
and  the  mother's  warnings  to  her  son  are  couched  in  the  most 
general  terms  {Roswall,  11.  166-72). 

The  remaining  portion  of  Roswall  follows  in  general  the  for- 
mula of  stories  of  the  False  Princess,  a  type  well  represented 
by  Grimm's  Die  Gansemagd  (Kinder-Mat chen,  No.  89;  cf. 
Bolte-Polivka,  11,  273-85).  In  this  the  true  princess  is  forced 
to  become  a  goose-girl,  her  ugly  maid  marries  the  king,  and  only 
when  the  princess  is  overheard  telling  her  sad  story,  first  to 
her  horse's  head,  and  then  to  a  stove,  is  the  truth  discovered.4 
Arfert's  study  (op.  cit.)  of  the  Substituted  Bride  motif  in  folk- 
tale and  romance,  established  the  fact  of  its  wide  diffusion  and 
pointed  out  that  within  this  story-type  it  is  no  unusual  varia- 
tion to  have  a  royal  youth  in  the  part  of  the  princess.  In  the 
Goldenmarchen  studied  by  Panzer  (Hilde-Gudrun,  p.  251),  the 
gold  hair  of  the  boy,  accidentally  revealed,  brings  him  the 
notice,  then  the  love,  of  a  true  princess.  In  Roswall  the  hero, 
who  is  serving  humbly  at  court,  is  chosen  by  the  Princess  Lillian 
because  of  his  "  wonder  fair  bodie."  The  romance  omits  the 
traditional  detail  which  in  the  ballad  tells  how  the  hero  evaded 

3  The  influence  of  popular  superstitions  on  the  evolution  of  character 
types  is  a  subject  much  in  need  of  further  investigation.  Kohler  commented 
on  "  Der  jungling  u.  der  bartlose,"  Archiv  /.  litter  atugesch.  xn,  137;  Ger- 
mania,  xi,  398.  P.  Arfert,  Das  Motiv  von  der  untergeschobenen  Braut,  Diss., 
Rostock,  1878,  p.  32,  mentioned  tales  involving  warnings  against  beardless 
or  deformed  men.  In  many  folk-tales  red  is  the  villain's  color.  Jones,  Folk 
Tales  of  the  Magyars,  Folk  Lore  Soc.  1886,  p.  329,  quoted  the  Magyar 
jingle:  "A  red  dog;  a  red  nag;  a  red  man;  none  is  good."  Cf.  Argyllshire 
Hero  Tales,  Folk  Lore  Soc.  1889,  p.  475.  In  the  Three  Counsels  type  of 
story  studied  by  Greenlaw,  PMLA.  xxi,  589,  596,  the  prohibition  against 
trusting  to  or  travelling  with  a  red-bearded  man  is  not  infrequent.  Cf.  Baum, 
JEGP.  xxi,  520-29  (1922),  "  Judas's  Red  Hair." 

4  Child,  v,  48,  noted  as  genuine  traditional  material  this  confession  to  a 
horse's  head  or  to  some  other  inanimate  object.  Cf.  the  story  of  Midas's 
wife  telling  his  secret  to  the  reeds,  or  the  dwarf  in  Beroul's  Tristan  who  tells 
Mark's  secret  to  a  hawthorn.  Cf.  Schoepperle,  Tristan,  11,  269-70.  For  the 
theme  of  the  Substituted  Bride  see  P.  Arfert,  op.  cit.,  pp,  50-71 ;  Bolte- 
Polivka,  11,  284;  Schoepperle,  1,  206.  See  also  under  Emare,  note  14,  for 
references  to  the  Berte  legend  into  which  this  theme  enters  largely. 


ROSWELL   AND    LILLIAN  293 

his  oath  of  secrecy  to  the  False  Servant  by  bewailing  his  fate 
to  his  horse.  The  princess  overhears  him  and  ultimately  brings 
about  his  restoration  to  proper  place  and  fortune. 

In  the  romance  the  restoration  of*  Roswall  follows  a  tradi- 
tional pattern  but  of  more  elaborated  kind.  A  tournament  is 
proclaimed,  and  Lillian  begs  him  to  joust  for  his  lady;  Roswall 
pretends  that  he  would  rather  hunt  than  joust,  and  each  day 
rides  away.  Unquestionably  at  this  point  Roswall  shows  the 
influence  of  Ipomedon,  in  which  the  hero  similarly  chooses  to 
deceive  his  lady  and  to  win  incognito  the  Three  Days'  Tourna- 
ment. Roswall  is  provided  with  different  suits  of  armour,  white, 
red,  and  gold,  by  the  Grateful  Prisoners  whom  he  had  formerly 
released.  In  this  reference  to  these  helpful  beings  the  episode 
reverts  somewhat  from  its  romanticized  character  to  its  original 
folk-tale  type.  The  actual  phraseology  of  Roswall  is  full  of 
reminiscences  not  only  of  Ipomedon,  but,  as  Lengert's  notes 
show,  of  other  romances  such  as  Eger  and  Grime.  Definite 
ballad  imitation  is  suggested  by  such  lines  as : 

"  He  looked  east  and  looked  west, 
He  looked  over  the  bents  brown."    (11.  478-9) 


BIBLIOGRAPHY 

Texts:  (i)  A,  Black  Letter  (846  lines),  Edin.,  1663,  Advocates  Libr., 
ed.  0.  Lengert,  Eng.  Stud.,  xvi,  321-56  (1891);  (2)  M,  another  early 
print,  cir.  1679;  (3)  B,  an  undated  print,  Newcastle;  (4)  D,  an  Edin- 
burgh print,  cir.  1775,  Douce  Collection,  Bodleian;  summarized  by  Ellis, 
Specimens,  1848,  pp.  578-84;  (5)  C,  an  Edinburgh  stall  copy,  1785; 
(6)  L,  an  edition  based  on  A  and  other  prints,  by  David  Laing,  Early 
Metrical  Tales,  Edin.,  1826.  Trans.  E.  Rickert,  Romances  of  Love,  pp. 
116-37. 

Studies. 
Bolte,  J.  G.  Polivka,  Anmerkungen  zu  den  Kinder-u.    Hausmarchen  der 

Bruder  Grimm,  Leipzig,  191 3-1 8.     3  vols. 
Child,  F.     Ballads,  v,  43-48  (1898). 
Lengert,  0.    "  Die  schottische  romanze  Roswall  and  Lillian^  Eng.  Stud., 

xvii,  341-88  (1892). 
Rickert.    See  Texts. 


LAY  LE  FREINE 

Versions.  Le  Lai  del  Fraisne  (ed.  Warnke,  1901)  is  one  of 
the  most  engaging  stories  told  by  Marie  de  France.1  She  wrote 
it  presumably  about  1165  when  her  other  lais  were  composed, 
and  certainly  before  1190,  for  the  poem  contains  a  reference  to 
the  Archbishopric  of  Dol  which  was  suppressed  in  that  year 
(Rickert,  p.  180).  The  direct  statement  that  the  adventure 
took  place  "en  Bretaigne  "  might  be  held  to  localize  the  story 
and  to  indicate  that  by  those  who  made  and  named  the  lai  "  pur 
la  dame  "  (v.  536)  Marie  meant  the  Bretons,  were  these  ex- 
pressions of  a  less  conventional  character  than  they  have  been 
long  recognized  to  be  (Lot,  Rom.  xxiv,  1895,  527).  Her  poem 
contains  536  lines  in  octosyllabic  couplets.  In  the  thirteenth 
century  it  was  greatly  amplified  and  changed  in  the  Roman  de 
Galeran  de  Bretagne  (ed.  Boucherie,  Montpellier,  1888;  Lang- 
lois,  La  Soc.  frq.  au  XIIIe  Steele,  1904,  pp.  1-39),  and  in  the 
early  years  of  the  fourteenth  century  it  was  rather  closely 
translated  and  somewhat  condensed  in  a  Middle  English  version. 
This  poem  of  340  lines,  in  the  same  metre  as  the  original,  is  now 
preserved  in  the  early  fourteenth-century  Auchinleck  manu- 
script. It  is  as  charmingly  distinctive  in  style  as  is  Orjeo  and 
may,  indeed,  have  been  by  the  same  author.  The  two  poems 
have  the  same  freshness  of  touch  and  are  linked  to  each  other 
by  evident  verbal  borrowings  (Guillaume,  p.  463).  Though  the 
evidence  at  best  is  slight,  it  seems  probable  that  the  Middle 

1  Comparatively  little  is  known  of  Marie  de  France  beyond  the  facts 
which  she  herself  gives  in  her  various  works,  the  Lais  (c.  1160-70),  the 
Ysopet,  a  collection  of  over  one  hundred  fables  (c.  1170-80),  the  Espurga- 
toire  Seint  Patriz  (after  iiqo).  See  Miss  Rickert,  Lais  of  Marie,  pp.  137- 
64;  Warnke,  Lais.  Fox,  Eng.  Hist.  Review,  xxv,  303-06  (1910),  xxvi,  317, 
found  evidence  for  believing  that  Marie  was  in  her  later  years,  during  the 
reigns  of  Richard  I  and  John,  the  Abbess  of  Shaftesbury.  E.  Kinkier, 
Mark  de  France,  Sitzungsberichte,  Vienna  Acad.  Ph.  Hist.  Kl.  Band  188 
(1918),  attempted  to  identify  her  with  Marie  de  Champagne.  This  theory 
was  soundly  rejected  by  Bertoni,  Nuova  Antologia,  Sept.  1920,  pp.  18  fif. 
Cf.  Bruce,  Evolution  of  Arthurian  Romance,  Gottingen,  1923,  p.  56. 

294 


LAY  LE   FREINE  295 

English  poet,  a  devoted  reader  of  Marie's  lays,  which  he  hap- 
pily characterizes  in  a  little  Prologue  to  the  Lay  le  Freine,  first 
made  his  translation  of  Marie's  Fraisne  and  later,  in  even  more 
independent  mood  and  with  even  more  mature  grace,  fashioned 
the  Lay  of  Orfeo,  to  which  he  or  a  scribe  transferred  the  Pro- 
logue originally  written  for  the  Lay  le  Freine.2 

Marie's  Lai  offers  a  peculiarly  interesting  example  of  the 
transference  of  popular  themes  and  beliefs  to  the  setting  of 
twelfth-century  life  and  literature.  The  scene  of  the  story  shifts 
from  the  rich  home  of  the  parents  of  the  heroine  to  the  convent 
where  she  spends  her  girlhood  and  then  to  the  castle  where  she 
lives  as  the  lovely  and  respected  mistress  of  the  young  lord, 
Gurun.  Her  liaison  is  regarded  with  serene  unconcern  by  an 
author  accustomed  to  the  doctrines  of  courtly  love.  Marie  even 
pauses  for  a  bit  of  amused  jesting  over  the  young  lord's  gifts 
to  the  convent,  gifts  not  given  for  the  sake  of  his  soul's  good 
but  for  a  chance  to  see  the  maiden.  Deft  bits  of  characteriza- 
tion and  delightful  realistic  touches  distinguish  Marie's  version, 
but  beneath  this  artistry  of  expression  certain  primitive  themes 
may  be  recognized. 

The  opening  episode  of  the  story  depends  on  the  widespread 
superstition  that  no  virtuous  wife  could  give  birth  at  one  time 
to  more  than  one  child.  The  same  theme  appears  in  the  many 
versions  of  the  Octavian  story,  wherein  the  birth  of  twin  chil- 
dren provides  a  cruel  mother-in-law  with  excuse  for  charging 
her  son's  wife  with  adultery.  In  a  large  number  of  stories,  in- 
deed in  a  majority  of  those  cited  by  Kohler  (p.  lxxxvi  ff.),  not 
family  but  class  prejudice  expresses  itself.  A  noblewoman  taunts 
with  a  similar  accusation  a  poor  woman  who  is  the  mother  of 
twins.  Ultimately  the  proud  lady  herself  gives  birth  to  two  or 
more  children  and  to  save  her  own  repute  attempts  to  destroy 
all  but  one.  In  these  tales  the  poor  woman's  curse  is  fulfilled 
with  a  literalness  which  evinces  popular  satisfaction  in  that 
justice  of  fate  or  providence  which,  in  story  at  least,  so  commonly 
surpasses  that  of  men.  In  a  few  tales,  as  in  Marie's  Lai  and  in 
the  account  of  the  Countess  Margareta  of  Holland  (Eccard's 
Corpus  historicum  medii  aevi,  11,  955),  the  two  women  are  of 
equal  rank.    In  pseudo-historical  legends  into  which  the  theme 

2  See  Orfeo  here,  note  3. 


2 g6  MEDIEVAL   ROMANCE    IN    ENGLAND 

passed,  the  story  sometimes  serves  to  explain  a  family  name 
such  as  that  of  the  Guelphs  (Welpen).3  In  pure  fiction  the 
theme  was  to  greater  or  less  extent,  as  in  the  versions  of  the 
Swan-Children  story,  joined  to  that  of  the  Swan-Maiden  and 
the  motive  belief  that  a  woman  whose  children  are  abnormal 
in  form  or  number  must  necessarily  be  of  demoniac  or  fairy 
origin.4 

The  disposal  of  the  unwelcome  children  in  stories  of  this  gen- 
eral type  presents  all  possible  varieties  of  the  Exposure  motif. 
In  Marie's  Lai  we  are  told  that  the  proud,  humiliated  mother, 
having  given  birth  to  twin  daughters,  lets  her  maiden  carry  one 
away.  The  maiden  leaves  the  child  in  an  ash  tree  near  a  con- 
vent, and  after  it  is  found  the  child  is  henceforth  known  as  Le 
Fraisne.  Marie  does  not  explain  the  reason  for  giving  the 
name  La  Coldre  to  the  child  kept  by  the  mother,  but  later 
when  the  question  is  raised  by  Gurun's  vassals  concerning  his 
discarding  of  Freine  and  his  marriage  to  La  Coldre,  they  cleverly 
contrast  the  fruitless  ash  with  the  "  noiz  e  deduiz"  (1.  349)  of 
the  hazel.  It  is  possible,  as  Miss  Rickert  and  others  have  sug- 
gested, that  the  legend  originally  belonged  to  the  village  of  La 
Coudre,  which  was  not  far  from  Dol,  but  in  that  case  it  seems 
difficult  to  explain  the  secondary  place  of  La  Coldre  in  the 
story.  The  choice  of  this  particular  name  was  probably  due 
merely  to  Marie's  fertile  instinct  for  effective  contrast.5 

The  second  part  of  the  Lai  tells  of  the  heroine's  life  in  the 
convent,  of  her  elopement  with  the  nephew  of  the  Abbess,  of 
her  self-abnegation  when  her  lover  is  forced  to  discard  her  and 

3  Cf.  Kohler,  p.  Ixxxvii  ff.  Gibbs,  Chevelere  Assigne,  p.  xi,  quoted  the 
Guelph  story.  In  this  a  noblewoman  is  punished  for  her  pride  and  false 
accusation  of  a  poor  woman  by  becoming  herself  the  mother  of  twelve  chil- 
dren. She  sends  her  maid  to  drown  all  but  one  in  the  river.  The  father 
of  the  children,  meeting  the  woman,  inquires  what  she  carries  and  is  told 
that  she  carries  whelps.  Insisting  on  seeing  them,  he  forces  a  confession  from 
the  woman,  has  the  children  reared  in  seclusion,  and  ultimately  brought 
home.  On  account  of  this  episode  the  race  of  the  Guelphs  received  and  kept 
this  name. 

4  See  Emare  here,  note  11. 

5  It  is,  also,  as  Miss  Rickert  pointed  out,  Lais,  p.  179,  a  mark  of  popular 
origin  when  names  are  thus  derived  from  some  physical  peculiarity  or  from 
some  circumstance  connected  with  the  early  history  of  the  child.  Cf. 
Cinderella,  Snow-White,  Gold-Tree,  Tom  Thumb,  Little  One-Eye,  etc.  Cf. 
A.  Nutt,  " Eliduc  and  Snow-White,"  Folklore,  in,  26  (1892),  Matzke,  p.  230. 


LAY  LE    FREINE  297 

wed  another,  and  of  her  reunion  with  him  and  her  kindred  when 
she  finds  that  the  new  bride  is  her  own  sister.  Practically  the 
same  story  is  found  in  a  ballad  widely  known  in  at  least  eight 
versions  in  English,  also  in  German,  Dutch,  Danish,  and  Swedish 
texts  (Child,  11,  63-69).  The  ballad  accounts  for  the  maiden's 
separation  from  her  kindred  by  the  bald  statement  that  she  was 
stolen  away  as  a  child  by  pirates  or  by  the  future  lover.  Ulti- 
mately, though  in  some  versions  she  has  borne  him  seven  sons, 
he  decides  to  discard  her,  either  because  she  is  a  "  waif  woman  " 
or  because  he  can  get  great  wealth  with  a  new  bride.  Like  the 
little  abandoned  Fraisne,  with  whom  in  the  ash  tree  a  rich  robe 
and  ring  were  left,  in  several  of  the  ballad  versions  the  stolen 
maiden  has  with  her  certain  tokens  the  recognition  of  which, 
on  the  part  of  her  mother  or  sister,  brings  about  the  identifica- 
tion of  herself.  In  a  few  versions  her  sister  overhears  the  plain- 
tive lament  in  which  the  girl  names  her  parents,  and  in  others 
it  is  Fair  Annie's  resemblance  to  herself  that  first  arouses  the 
bride's  interest.  In  this  detail  may  be  preserved  a  trait  more 
primitive  than  that  found  in  Marie's  version.  Although  her 
Lai  antedates  by  four  centuries  the  earliest  known  ballad 
version  and  the  two  stories  are  obviously  the  same,  it  has 
generally  been  felt  that  the  ballad  is  not  a  derivative 
of  the  Lai,  but  that  the  two  "  have  a  common  source 
which  lies  further  back  and  too  far  for  us  to  find " 
(Child,  n,  67). 

The  idea  in  the  ballads  of  the  physical  resemblance  of  the 
two  maidens  appears  also  in  the  romance  of  Galeran  de 
Bretagne.  Galeran  is  so  moved  by  the  resemblance  of  Florie 
to  her  sister  Freine,  whom  he  has  loved  and  lost  in  youth,  that 
he  is  about  to  marry  Florie.  To  the  physical  likeness  of  the 
two  sisters  is  likewise  added  the  resemblance  in  name,  a  feature 
more  clearly  recognizable  here  than  in  Marie's  Lai  (Matzke, 
p.  226).  So  also  in  other  romances  the  maiden  offered  to  the 
hero  in  place  of  the  one  to  whom  he  has  been  married  or  be- 
trothed, frequently  bears  a  name  like  that  of  his  lost  love.  In 
the  English  Horn 6  the  maidens  are  Rimenhild  and  Reynild ; 
in  Marie's  Eliduc  they  are  Guildeluec  and  Guilliadun ;  in  Me  et 

6  See  Horn  here,  note   7,  and  Schofield,   PMLA.  xvm,  35;   Matzke,  pp. 
216-17,  225-26. 


2  98  MEDIEVAL   ROMANCE   IN   ENGLAND 

Galeron  7  by  Gautier  d 'Arras  they  are  Galeron  and  Ganor ;  in 
the  various  versions  of  Tristan  the  hero  is  drawn  to  Isolt  of 
Brittany  because  her  name  is  that  of  the  lovelier  Isolt  of  Ire- 
land.8 To  this  group  of  stories,  which  he  made  (pp.  227  ff.) 
also  to  include  parts  of  the  story  of  Guy  of  Warwick  and  of 
Beves  of  Hampton,  Matzke  gave  the  name  of  the  Legend  of 
the  Husband  with  Two  Wives.  It  was  unquestionably  popular 
in  the  twelfth  and  thirteenth  centuries,  but  it  seems  necessary 
to  distinguish  between  romances  such  as  Horn,  Bevis,  and  Guy, 
in  which  the  second  lady  is  simply  an  embarrassing  reward  and 
the  hero's  relation  with  her  entirely  devoid  of  emotional  inter- 
est, and  those  in  which  it  involves  him  in  a  psychological  con- 
flict and  is  introduced  for  precisely  that  purpose.  It  is  difficult 
therefore  to  accept  Matzke's  belief  (p.  234)  that  this  whole 
group  of  stories  not  only  "  belong  to  a  region  in  which  the 
Celtic  substratum  could  exert  decisive  influence,"  but  that  they 
"  rest  upon  a  more  primitive  habit  of  society  in  which  a  wife 
could  be  pushed  aside  for  a  new  and  more  favorite  rival." 
Rather  do  such  romances  as  Eliduc  or  llle  et  Galeron  present  a 
special  twelfth-century  adaptation  of  a  test  of  loyalty  in  a  lover 
or  a  husband.  To  amour  courtois  the  rights  of  love  were 
even  more  sacred  than  those  of  marriage,  and  for  this  reason 
an  episode  setting  forth  the  emotional  conflict  of  a  man  who 
was  both  husband  and  lover  made  an  especial  appeal  to  such 
writers  as  Thomas  in  his  Tristan,  or  to  Marie  de  France  in 
Eliduc.  In  her  Lai  del  Fraisne,  however,  there  is  no  real  con- 
flict, for  the  lover  gives  way  with  complete  docility  to  his  vas- 
sals' demand  that  he  marry  a  proper  wife  and  beget  a  proper 
heir.  However  delicately  revealed  in  this  version  or  brutally 
in  the  ballads,  the  essential  situation,  the  discarding  of  a  mis- 
tress for  a  wealthy  and  legitimate  wife,  is  not  to  be  referred  to 
any  one  race  or  time  or  creed. 

The  gentle  service  rendered  by  Fraisne  or  by  Fair  Annie  at 
her  lover's  wedding-feast  recalls  the  self-sacrifice  of  that  mediae- 
val synonym  for  all  patience,  Griselda.9     Whatever  the  origin 

7  Cf .  Matzke,  "  The  Source  and  Composition  of  llle  et  Galeron"  Mod. 
Phil.,  1907,  iv,  471-88;  Cowper,  "The  Sources  of  llle  et  Galeron"  Mod.  Phil. 
1922,  xx,  35-44- 

8  Cf.  Schoepperle,  1,  158-77;  n,  524-28. 

9  For  references  concerning  Chaucer's  version  of  her  story  in  the  Clerk's 


LAY  LE   FREINE  299 

of  her  legend,  it  seems  impossible  to  connect  it  directly  with 
Marie's  Lai.  The  story  of  Griselda  is  essentially  an  exemplum 
on  patience;  it  tells  of  the  cruel  succession  of  marital  trials 
endured  by  a  peasant  girl  married  to  a  great  lord.  His  appar- 
ent marriage  to  another  is  in  reality  her  supreme  test,  and  when 
she  has  endured  that  too  with  the  patience  that  only  a  Chaucer 
in  the  Middle  Ages  seems  to  have  suspected  was  "  importable," 
she  was  properly  rewarded.  Her  trials  and  her  reward  make  a 
definite  structural  sequence.  Fraisne's  story,  on  the  contrary, 
is  pure  romance ;  it  tells  of  a  girl  losing  birthright,  home,  family, 
friends,  love  itself,  yet  marvellously  regaining  them  all.  It  is 
a  love  adventure  only  saved  from  tragedy  by  the  law  in  popular 
fiction,  mediaeval  or  otherwise,  of  the  "  happy  ending." 


BIBLIOGRAPHY 

Texts:  (i)  Auchinleck  MS.  ed.  Weber,  1810, 1,  357-71;  H.  Varnhagen, 
Anglia,  in,  415-23  (1880);  Trans.  E.  Rickert,  Romances  of  Love,  1907, 
p.  47  ff. 

Studies:    Cf.  Wells,  Manual,  p.  783. 
Child,  F.    English  and  Scottish  Ballads,  11,  63-83  (1886). 
Foulet,  L.     "  Marie  de  France  et  les  Lais  Bretons,"  Zts.  f.  rotn.  Phil. 

xxix,  19-56,  292-322   (1905). 
Guillaume,  G.     "The  Lay  le  Freine,"  MLN.  xxxvi,  458-64   (1921). 
Holthausen,  F.    Lay  le  Freine,  v.  91,  Anglia,  xm,  360  (1890-91). 
Kohler,  R.     See   Warnke.     V er gleichende  Anmerkungen,   pp.   lxi-xcviii 

(1901). 
Laurin,  A.    Essay  on  Language  of  Lay  le  Freine.    Diss.  Upsala,  1869. 
Matzke,  J.    "  The  Legend  of  the  Husband  with  Two  Wives,"  Mod.  Phil. 

v,  211-39  (1907). 
Marie  de  France,  Lai  del  Fraisne,  ed.  Warnke,  Die  Lais  der  Marie,  Bibl. 


Tale  see  Hammond,  Chaucer  Manual,  p.  304;  Wells,  Manual,  pp.  726-28. 
In  the  main  Chaucer  derived  his  story  from  the  Latin  version  made  by 
Petrarch  in  1373  from  Boccaccio's  Decameron,  Tenth  Day,  Tenth  Tale.  A.  C. 
Lee,  The  Decameron,  Its  Sources  and  Analogues,  1909,  pp.  348-56,  men- 
tioned among  other  early  versions  that  by  Sercambi  (Novelle,  ed.  Renier, 
1889,  No.  108,  p.  401)  and  the  anonymous  French  Mystere  de  Griseldis, 
Marquise  de  Saluces,  par  personnages,  1395  (restaure  par  Ch.  Gailly  de 
Taurines  et  Leonel  de  la  Tourasse,  Paris,  1910).  Cf.  Monacis,  La  Novella 
di  Griselda  secundo  la  lezione  de  un  manuscritto  non  ancora  illustrato  del 
Decamerone,  Perugia,  1902;  Literaturblatt,  xxiv,  117-19;  R.  Schuster, 
Griseldis  in  der  frz.  Literature.  Diss.  Tubingen,  1909. 


300 


MEDIEVAL   ROMANCE    IN    ENGLAND 


Norman,  m,  1885,  1901,  pp.  56-67;  trans.  F.  Luquiens,  Four  Lays 

of  Marie,  N.  Y.,   1903;   E.  Mason,  French  Mediaeval  Romances, 

Everymans  Library,   191 1. 
Rickert,  E.    Seven  Lais  of  Marie  de  France.    Lond.,  1901. 
Schoepperle,  G.     Tristan  and  Isolt,   1913,  vol.   1,   158-77,   The  Second 

Isolt;  vol.  11,  525-28,  The  Problem  of  the  Second  Isolt. 
Warnke.     See  Marie  de  France. 
Zupitza,  J.    "Zum  Lay  le  Freine,"  Eng.  Stud,  x,  41-48  (1886). 


SIR   DEGARE 

Versions.  The  romance  of  Sir  Degare  is  preserved  in  five 
manuscripts  and  in  three  sixteenth-century  editions,  the  relation- 
ship of  which  has  not  been  determined.  The  earliest,  the  Auch- 
inleck  manuscript,  contains  the  most  complete  version  of  the 
story  and  represents  an  original  probably  composed  early  in  the 
fourteenth  century  in  the  South  Midland  dialect.  The  name  of 
the  hero,  carefully  explained  (1.  229)  to  mean  something  that 
"almost  lost  it  is"  (v.  214),  suggests  the  French  word  esgare 
and  the  possibility  of  a  French  source.  Whether  the  Middle 
English  version  is  a  translation  of  this  lost  Lai  d'Esgare  or 
merely  a  clever  imitation  of  the  Lai  style,  the  incidents  are  cer- 
tainly those  typical  "  aventures  Whereof  Britouns  made  her 
layes."  The  style  is  simple,  brief,  yet  picturesque;  the  rime 
the  familiar  short-riming  couplet.  In  date,  form,  and  context, 
Degare  belongs  with  the  other  Middle  English  versions  of  such 
lays  as  Orjeo  or  the  translations  of  Marie  de  France's  Lanval x 
and  her  Lai  del  Fraisne.  From  the  Middle  English  redaction  of 
this  last  poem,  the  Degare  poet  even  borrowed  definite  ideas  and 
phrases.  But  Degare,  Freine,  and  Orjeo,  must  all  have  been  of 
approximately  the  same  date  since  in  not  one  does  the  language 
antedate  the  fourteenth  century  and  since  they  were  all  copied 
in  the  Auchinleck  manuscript  by  the  same  scribe,2  probably 
between  1330  and  1340. 

Origin.  Within  the  comparatively  brief  compass  of  nine  hun- 
dred and  ninety-three  lines  the  story  of  Degare  manages  to 

1  The  long  ME.  version  of  Sir  Launfal  was  preceded,  according  to 
Kittredge  {American  Jour,  of  Phil.  1889,  x,  5)  by  an  earlier  version  in 
Middle  English.  From  this  original,  x,  was  derived  the  extant  short  ver- 
sion now  represented  by  a  poem  of  535  verses  in  MS.  Rawlinson  C  86 
(sixteenth  century).  This  text  is  much  nearer  to  Marie's  Lanval  than  is 
the  long  version,  and  must  represent  rather  closely  the  lost  original  version 
in  Middle  English.  This  must  have  been  made  at  about  the  same  time,  in 
the  same  style  and  verse,  as  the  other  translations  of  "  Breton  "  lays. 

2  See  Muriel  Carr,  "  Notes  on  a  Middle  English  Scribe's  Methods," 
Univ.  of  Wisconsin,  Studies  in  Language  and  Lit.,  11.  p.  152   (1918). 

301 


302  MEDIAEVAL   ROMANCE    IN   ENGLAND 

combine  an  astonishing  number  of  folk-lore  and  romance  motifs. 
The  name  of  the  hero,  like  that  of  the  maiden  Freine,  signifies 
the  ultimately  popular  character  of  the  tale.  The  taste  for  such 
cognomens  reappears  even  in  the  later  more  composite  romances 
of  the  type  of  Emare  or  Degrevant. 

The  first  part  of  the  story  is  devoted  to  an  account  of  the 
hero's  parents.  His  mother  is  long  kept  from  marriage  with 
the  kings  and  princes  who  seek  her  hand  by  an  over-devoted 
father  who  overthrows  every  suitor  in  a  royal  tournament.  The 
tournament  is  said  to  be  held  on  the  anniversary  of  the  death 
of  the  king's  wife.  Though  shortened  and  rationalized,  the  sit- 
uation evidently  comes  from  the  large  group  of  stories  in  which 
a  king  seeks  to  marry  his  own  daughter.  In  the  folk-tales  the 
desire  generally  rises  because  she  alone  can  fulfill  some  special 
condition  which  the  dead  wife  had  made  the  king  swear  to 
observe.  In  English  romances  such  as  Apollonius  or  Emare  the 
character  of  the  Incestuous  Father  is  more  frankly  recognized.8 
The  Princess  in  Degare  is  made  at  last  to  meet  a  lover  when 
she  goes  wandering  in  the  forest.  He  is  a  magnificent-looking 
stranger  who  asserts  that  he  has  long  loved  her,  ravishes  her, 
and  on  leaving  her,  prophesies  the  birth  of  a  child  for  whom 
he  leaves  his  own  pointless  sword.  Whether  the  Fairy  Wooer 
was  originally,  as  has  been  argued,4  the  Angel  in  Joachim's 
garden,  or  whether  he  belonged  to  the  lineage  of  splendid  Other- 
world  beings  who  appear  in  Celtic  legend,  there  is  little  ques- 
tion that  in  this  particular  instance  he  was  inspired  by  the 
account  of  Tydorel's  father  in  the  "  Breton  lay  "  of  that  name. 

The  description  of  the  secret  birth  of  this  love-child,  of  the 
Maiden  Messenger  who  at  the  princess's  command  carries  him 
through  a  moonlit  night  to  the  door  of  a  hermitage,  the  poet  of 
Degare  borrowed  definitely  from  the  English  version  of  Marie's 
Lai  del  Fraisne.  In  the  French  version  nothing  was  said  of  the 
moonlight,  but  the  English  translator  added  this  effective  touch 
which  in  turn  the  Degare  poet  was  to  borrow.5    The  latter  tells 

3  That  the  poet  had  this  in  mind  is  shown  by  the  Princess's  uncalled  for 
comment  concerning  her  child:  "  Every  man  wolde  it  in  euery  stede/That 
my  father  on  me  it  wan."     See  Emare  here,  note  6. 

4  See  Gowther  here,  note  3. 

5  There  are  likewise  passages  of  verbal  imitation:  cf.  Freine,  lines  85, 
145,  149,  189,  197,  with  those  in  Degare,  lines  179,  217  ff.,  239,  245. 


SIR   DEGARE 


303 


in  detail  of  the  gifts  left  with  the  child  but  in  this  makes  a 
curious  departure.  In  addition  to  the  usual  gold  and  silver,  a 
pair  of  fairy  gloves,  which  his  mother  has  received  from  her  un- 
known lover,  is  left  with  Degare  and  a  written  command  to  the 
effect  that  he  is  to  wed  no  lady  whose  hands  the  gloves  will 
not  fit.  A  parallel  to  this  is  hard  to  find,  since  gloves  are  a 
somewhat  too  sophisticated  article  of  dress  for  folk-lore  to  make 
common  use  of,  but  there  is  at  least  one  parallel  in  the  fifteenth- 
century  Catalan  version  of  La  Fille  sans  Mains  (Rom.  xxx,  520), 
in  which  the  dying  wife  of  the  Emperor  Contasti  begs  him  to 
marry  no  one  less  beautiful  than  she,  nor  one  whom  her  gloves 
will  not  fit.  This  request  motivates  the  episode  of  the  father's 
insistence  on  marriage  with  his  own  daughter.  Stories  of  this 
type  were,  however,  so  widely  diffused  6  and  the  Catalan  text 
itself  is  so  late  a  composite,  that  there  is  no  improbability  in 
supposing  that  some  much  earlier  version,  using  this  particular 
feature,  may  have  caught  the  attention  of  the  original  author 
of  the  Lai  d'Esgare.  In  the  Middle  English  version  the  gloves 
having  been  thus  introduced,  serve  the  purpose  of  Recognition 
Tokens  when  years  later,  the  young  Degare  overthrows  his 
grandfather  in  the  suitors'  tournament,  receives  his  Mother,  un- 
touched of  course  by  Time,  as  his  prize,  and  only  remembers, 
after  the  marriage  ceremony  has  been  performed,  to  try  on  her 
white  hands  the  gloves  of  fate.  The  happy  solution  of  this 
(Edipus-like  situation  in  Degare  or  in  Eglamour,  where  simi- 
larly a  mother  recognizes  her  son  in  her  new-made  husband,  is 
typical  of  the  care-free  naivete  of  romance  when  untouched  by 
ecclesiastical  influence. 

Degare's  own  history,  before  this  reunion  with  his  mother, 
is  briefly  told.  His  bringing  up  in  the  woodland  hermitage  and 
his  setting  forth  armed  only  with  a  rough  oak  sapling,  suggest 
the  beginning  of  the  Perceval  story  of  the  Forest-Reared  Youth. 
Degare's  rescue  of  an  old  knight  from  a  dragon,  his  refusal  of 
the  knight's  wealth  and  daughter  are  mere  commonplaces  of 
knightly  adventure.  Equally  well-known  are  Degare's  combats 
with  his  relatives:  he  fights  with  his  grandfather  and,  over- 
throwing him,  wins  his  own  mother;   later  he  fights  with  his 

6  Miss  Cox,  Cinderella,  Folk-Lore  Society,  pp.  52-79,  enumerated  sixty- 
six  tales. 


304  MEDIEVAL   ROMANCE    IN    ENGLAND 

father,  who  appears  as  a  wandering  knight-errant  and  recognizes 
his  own  sword  in  the  hands  of  his  son.  In  Degare's  quest  for 
his  father  occurs  an  adventure  which,  unlike  these  others,  sug- 
gests a  definite  literary  source. 

Accompanied  by  one  follower,  Degare  goes  through  the  forest 
until  he  comes  to  a  lonely  castle.  The  drawbridge  is  down,  the 
gate  open,  but  no  living  being  appears.  Degare  enters  the  hall 
of  the  castle,  in  which  a  great  hearth  fire  is  burning,  seats  him- 
self on  a  dais,  and  is  presently  served,  first  by  three  maidens 
in  hunting  guise  who  bring  him  venison,  then  by  a  yellow-faced 
dwarf  clad  in  a  furred  green  surcoat,  who  spreads  the  table  and 
lights  many  torches.  A  lady  and  fifteen  maidens,  clad  in  red 
and  green,  enter  and  eat  with  the  hero  in  silence.  Later  the 
lady  plays  sweet  music  7  to  him,  and  despite  his  amorous  desires 
he  falls  into  a  deep  slumber  on  which,  next  day,  she  gently 
rallies  him.  When  he  questions  why  she  has  so  many  women 
and  no  men,  she  tells  him  of  the  furious  suitor  who  has  ravaged 
her  land  and  killed  all  her  men  except  the  dwarf.  Needless  to 
say,  Degare  promptly  kills  this  giant,  and  when  the  lady  offers 
herself  and  her  possessions  to  the  youth,  he  promises  to  return 
and  accept  her  bounty  within  a  twelvemonth. 

All  this  seems  to  have  some  definite  connection  with  Giglain, 
Gawain's  son,  about  whom,  near  the  beginning  of  the  thirteenth 
century,  Renaud  de  Beaujeu  wove  the  elaborate  romance,  Le  Bel 
Inconnu  (ed.  Hippeau,  i860).  Renaud's  source  seems  also  to 
have  provided  material  for  the  Middle  English  poem  sometimes 
ascribed  to  Thomas  Chestre,  Libeaus  Desconus  (ed.  Kaluza, 
1890).8  In  both  versions  the  hero  similarly  comes  to  the  castle 
of  a  lady  possessed  or  persecuted  by  a  militant  suitor.  In 
Renaud's  courtly  version  she  is  described  as  skilled  in  the  Seven 
Arts,  "  la  pucele  as  blances  mains"  (1.  1925),  and  is  recogniz- 
able, despite  Renaud's  rationalizing  tendencies,  as  the  Fairy 
Mistress  of  a  Bower  of  Bliss.9    Her  magic  powers  are  described 

7  A  parallel  to  this  exists  in  the  ancient  Irish  tale  of  the  inram  Maelduin. 
On  the  Island  of  Women,  magic,  sleep-inducing  music  prevents  the  com- 
panions from  entering  the  fairy  island.  Cf.  Revue  Celt,  tx,  489;  Brown, 
Harvard  Studies,  vni,  75. 

8  For  the  history  of  the  conflicting  views  on  the  relationship  of  Renaud's 
poem  and  that  in  Middle  English  see  Schofield,  Harvard  Studies,  1895,  iv, 
59  ff.    Schofield  believed  the  two  poems  had  a  common  source. 

9  Cf.  Schofield,  ibid.,  pp.  36,  129,  197,  for  indications  of  Renaud's  bor- 
rowings from  Chretien's  Erec,  especially  in  the  description  of  the  lie  d'Or. 


SIR   DEGARE  305 

by  Renaud  as  amusing  illusory  arts  10  which  she  practices  upon 
her  lover  in  teasing  punishment  for  seeming  lack  of  devotion. 
He  is  made  to  think  himself  in  dire  peril  and  later  has  to  en- 
dure the  laughing  jests  of  the  lady  over  his  previous  terror.  In 
Sir  Degare  the  lady's  fairy  music  enchants  and  inhibits  the 
hero,  and  her  words,  though  more  sedate  and  brief,  have  the 
same  jocose  quality.  Of  this  there  is  nothing  in  the  Libeaus 
Desconus,  for  here  the  lady  is  a  sorceress,  an  evil  dame  dy  amour, 
ungraced  by  jest  in  her  relations  with  Gingelein.  But  if  in 
these  respects  Degare  is  closer  to  the  French  poem,  the  more 
prosaic  description  of  the  castle  and  the  special  reference  to  the 
great  fire  burning  in  the  hall,  seem  closer  to  the  text  of  Libeaus. 
The  explanation  may  lie  in  the  knowledge  possessed  by  the 
author  of  the  French  Lai  d'Esgare  of  the  lost  common  source 
of  Le  Bel  Inconnu  and  of  Libeaus,  or  in  the  knowledge  pos- 
sessed by  the  author  of  the  Middle  English  Degare  of  these  two 
extant  texts. 

BIBLIOGRAPHY 

Texts:  (i)  Advocates  Library,  Edin.,  Auchinleck  MS.  W  41,  ed. 
Laing,  Abbotsford  Club,  1849;  desc.  Kolbing,  Eng.  Stud,  vn,  178-91; 
(2)  Cambridge  Univ.  Libr.  Ff  11,  38  (about  602  verses);  (3)  Bodleian, 
Selden  C  39  (about  352  verses),  printed  by  John  King,  1560;  (4)  Percy 
Folio  MS.,  ed.  Hales  and  Furnivall,  1867-69,  111,  16  ff.;  (5)  Duke  of 
Sutherland's  MS.,  now  Egerton  2862,  desc.  Eng.  Stud,  vn,  192-93;  (6) 
undated  edition  by  Copland,  reprinted  by  Utterson,  Early  Popular 
Poetry,  181 7,  1,  113. 

Studies:    Cf.  Wells,  Manual,  p.  784. 
Furnivall,  F.    See  Texts,  hi. 

Kaluza,  M.    Libeaus  Desconus,  Leipzig,  1890,  p.  cliv. 
Schofield,  W.     English  Literature  —  to  Chaucer,  pp.  186-87. 

10  The   magician   in    Chaucer's  Franklin's    Tale   practices  similar   illusory 
arts.    See  Partinopeus  here,  note  9. 


SIR   DEGREVANT 

Versions.  Two  slightly  imperfect  Middle  English  manu- 
scripts contain  the  only  known  version  of  Degrevant.  Both 
manuscripts  were  written  in  the  fifteenth  century,  and  were  de- 
rived from  the  same  source,  though  this,  according  to  Finster- 
buch  (p.  74),  was  not  the  original  text  of  the  poem.  The 
Thornton  manuscript,  the  earlier  of  the  two,  was  written 
before  1430,  probably  not  long  after  the  composition  of  the 
poem.  The  lateness  of  its  date  and  the  thoroughly  English  char- 
acter of  the  setting  make  it  unlikely  that  there  was  ever  an 
antecedent  French  version  of  the  tale.  The  author  seems  to 
have  been  a  minstrel ;  he  prays  for  special  blessing  on  those 
who  love  "  gamen  and  glee  "  ;  he  praises  his  hero  who  loves  to 
have  "  mynstralles  in  haulle,"  who  of  "  gyfte  was  never  gnede;" 
and  he  takes  pains  to  point  out  that  Degrevant's  secret  tryst 
is  betrayed  by  a  forester,  not  a  minstrel,  for  "  mynstrals  are 
ay  curtayse."  His  allusions  show  that  he  was  familiar  with 
romance  stories.  From  the  phrase,  "  me  were  lever  than  al  5e 
golde  in  Ryne  "  (v.  541),  it  would  appear  that  he  knew  some- 
thing of  the  Nibelungenlied.  His  hero,  he  says,  was  a  "  Knyghte 
of  de  Table  Rownde,/As  it  es  made  in  Mappamonde," 1  and 
was  well  known  to  King  Arthur,  dame  Gaynore,  Perceval,  and 
Gawayne.  The  connection  with  Arthurian  romance  is,  however, 
purely  artificial,  and  aside  from  the  similarity  of  name  there 
is  no  reason  to  regard  Degrevant  as  the  villainous  Agravain  of 
Arthurian  legend. 

The  dialect  of  the  poem  is  northern  and  contains  a  number 

1  Halliwell  suggested  that  the  English  writer,  translating  from  an  Anglo- 
Norman  text,  might  have  mistaken  d'^Egrivauns  for  the  entire  name. 
Rickert,  p.  xlix,  noted  that  in  the  list  of  Arthur's  knights  given  by  the 
chronicler,  John  Harding  (1368-1460),  is  the  name  Degrevant.  Harding 
was  describing  the  great  Round  Table  which  still  hangs  in  the  palace  hall  at 
Winchester.  (See  description  by  Smirke,  "  The  Hall  and  the  Round  Table 
at  Winchester,"  Proceedings  of  Archaeol.  Institute  of  Great  Britain,  1846.) 
The  romancer  must  have  had  some  such  chart  in  mind,  an  Arthurian 
"  Mappamonde." 

306 


SIR    DEGREVANT  307 

of  curious  North  Country  words.  Like  Sir  Perceval  of  Galles 
the  poem  is  fashioned  in  sixteen-line  stanzas  riming  aaabcccb- 
dddefffe.  The  structure  of  the  triplet  and  the  tail-rime  verses 
is  explained  by  Luick  and  Finsterbuch  by  reference  to  that  in 
the  half  lines  of  older  alliterative  poetry.2  The  devices  in 
Degrevant  for  linking  stanzas  together  by  repetition  of  impor- 
tant words  or  by  repetition  of  the  whole  or  part  of  the  last  line 
or  last  two  lines  of  one  stanza  in  the  introductory  lines  of  the 
next  appear  in  Sir  Percyvelle,  in  the  Aunters  of  Arthur,  and  the 
Avowynge  of  Arthur  (Medary,  p.  255).  In  these  last  two  ro- 
mances, as  also  in  Degrevant,  the  end  of  the  poem  is  linked  to 
the  beginning  by  the  repetition  of  the  introductory  four  lines. 
Brown  {Rom.  Rev.  vn,  275)  suggested  that  Welsh  alliterative 
poetry  of  the  thirteenth  and  fourteenth  centuries  may  have  in- 
fluenced the  authors  of  these  romances  in  this  matter  of  stanza 
linking.  In  his  opinion  the  poem  originated  in  or  near  counties 
which  were  a  part  of  the  old  Welsh  border.  There  is  additional 
confirmation  for  this  theory  in  the  Welsh  form  Gaynor  for  the 
name  of  Arthur's  Queen,  in  the  allusions  to  Westwale  (Thorn- 
ton MS.,  v.  151 1 )  and  to  Degrevant  and  his  friends  as  "  wylde 
men  of  the  west"  (1.  1367),  and  finally  to  a  possible  Welsh 
betrothal  custom.3  The  only  English  place  name  in  the  ro- 
mance is  in  v.  1401  to  "  towelles  of  Alsame  "  (Eylyssham),  the 
town  of  Alysham  in  Norfolk,  where  the  linen  industry  was  estab- 
lished in  the  fourteenth  century. 

The   value   of   the   romance   has   been   variously   estimated. 
Halliwell  (p.  xxiii)  thought  its  descriptive  notices  of  early  cos- 

2  Biilbring,  "Avowynge,"  Morsbach's  Studien,  Bd.  L  (1913),  argued  that 
the  triplet  verses  should  be  read  with  four,  the  tail-rime  verses  with  three 
beats  (Hebungen) ;  Luick,  Anglia,  xxxix,  269,  believed  that  all  the  verses 
should  be  read  with  two  beats.  He  thought  that  the  triplet  verses  had  the 
structure  of  the  first  half  line  of  the  alliterative  long  line,  and  that  the  tail-rime 
verses  had  that  of  the  second  half  line.  With  this  Finsterbuch,  after  his 
elaborate  study   of  Perceval  and  Degrevant,  in   general  agreed. 

3  The  reference  is  to  the  habit  of  night  courtship,  or  courting  on  a  bed, 
which  is  called  cnocio  or  streicio  in  Wales,  jenstern  in  Germany,  kilt  in 
Switzerland,  questing  in  Holland,  and  bundling  in  old  and  New  England. 
Cf.  Halliwell,  note  to  1.  1544;  Douce,  Illus.  of  Shakespeare,  1,  113;  C. 
Masson,  Journeys  in  Balochistan,  Lond.,  1842,  in,  287,  tells  of  the  custom 
among  the  Afghan  tribes.  Cf.  J.  Rhys  and  D.  B.  Jones,  The  Welsh  People, 
Lond.,  1906,  pp.  583;  Rhys,  Arthurian  Legend,  p.  175;  Baskerville,  "English 
Songs  on  the  Night  Visit,"  PMLA.  xxxvi  (1921). 


3o8  MEDIEVAL   ROMANCE    IN   ENGLAND 

tume  and  architecture  of  peculiar  interest ;  Luick  (p.  v)  found 
its  literary  quality  negligible;  William  Morris  and  his  painter 
friends,  when  they  were  adorning  the  House  Beautiful  at  Upton, 
Kent,  chose  to  paint  on  its  walls  scenes  from  this  picturesque 
tale  (Mackail,  Life  of  Morris,  i,  158).  As  a  story  of  hunting 
raids  and  swift  reprisals,  of  gallant  tournament  and  moonlit 
wooing,  it  has  indubitable  variety.  To  a  degree  unusual  in 
Middle  English  poetry  the  author  lingers  over  the  description 
of  beautiful  and  luxurious  things.  Like  the  Pearl  poet,  he  pauses 
to  describe  the  dress  of  the  maiden,  her  pearl-fretted  violet  robe, 
her  gleaming  ribbons,  as  she  comes  in  the  early  morning  by 
the  rose  bushes  where  her  lover  waits.  Like  an  early  Keats,  he 
pictures  a  midnight  feast  where  on  ivory  boards  and  in  golden 
cups,  is  "  na  dayntese  to  dere/Na  spyces  to  spare,"  whilst  the 
lady  harps  "  notes  ful  swet  "  to  the  lover  beside  her.  The  poet 
makes  rich  and  lovely  the  chamber  of  love;  the  roof  is  inlaid 
with  "  besauntes  "  and  painted  with  scenes  from  the  Apocalypse 
of  St.  John,  the  Epistles  of  Paul,  and  the  "  Parabylles  "  of 
Solomon.  The  corbels  are  golden  archangels,  "  ffyfthy  made  of 
0  molde  " ;  the  four  "  gospellers,"  Austin,  Gregory,  Jerome,  and 
Ambrose,  stand  on  four  pillars,  and  the  walls  are  painted  with 
knights  of  many  lands  and  kings  enthroned,  Charlemagne  and 
Godfrey  of  Bouillon  and  "  Arthure  de  Bretayne."  4  The  azure- 
colored  bed  is  embroidered  with  gay  jewel  work  and  popinjays 
of  green ;  it  has  sendal  covered  pillows  "  wroght  in  Westwale  "  ; 
its  curtain  run  on  red-golden  rings.  In  all  this  there  is  cer- 
tainly elaboration  enough,  but  it  is  far  too  zest  ful  to  hint  of 

4  Cf.  J.  C.  Wall,  Mediaeval  Wall  Paintings,  Lond.,  1914,  p.  109.  Wall's 
description,  pp.  44-8,  of  the  twelfth  century  painting  of  saints,  apostles,  and 
the  Apocalypse  in  Kempley  Church,  Gloucestershire,  is  of  special  interest  in 
connection  with  the  Degrevant  passage.  Many  romances  refer  to  painted 
walls.  Cf .  O.  Sohring,  "  Werke  bildender  Kunst  in  alt.  frz.  Epen."  Rom.  Forsch. 
xn  (1900);  also  the  Roman  de  La  Rose,  the  Lady  of  the  Fountain  in 
the  Mabingion,  the  French  prose  Lancelot,  where  Morgan  shows  Arthur  the 
paintings  of  Lancelot's  adventures,  etc.  In  addition  to  the  peculiarly  detailed 
account  of  the  painted  walls,  the  reference  in  Degrevant  to  Alysham  in 
Norfolk  leads  one  to  suspect  the  poet  of  familiarity  with  a  region  which 
was  famous  in  the  fourteenth  century  for  the  painted  roofs  and  screens 
produced  by  its  native  craftsmen.  Cf.  C.  E.  Keyser,  A  List  of  Buildings  in 
Great  Britain  having  Mural  Decoration,  Lond.,  1883  (Index,  Norfolk).  Cf. 
W.  O.  Sypherd,  Studies  in  Chaucer's  Hous  of  Fame,  Chaucer  Soc,  1907, 
pp.  83-86. 


SIR    DEGREVANT  309 

decadence.  The  author's  description  is  as  graphic  in  kind  as 
is  his  power  of  characterization.  Conventional  as  that  is  in 
some  ways,  it  escapes  again  and  again  into  piquancy.  Degre- 
vant  may  be  love-vanquished  at  first  sight  of  Melydore,  but  he 
keeps  his  wits  and  his  vigor;  and  the  maiden  herself,  though  a 
"  pervenke  of  pryse,"  is  also  a  bit  of  a  shrew  who  can  bid  her 
maid  entertain  her  guest  in  "  twenty  deuelle  way."  Her  father, 
indeed,  she  makes  sweat  with  rage  when  he  is  chased  by  Degre- 
vant  within  his  own  castle  doors  and  is  bullied  there  by  his 
lively  daughter.  He  agrees  perforce  to  her  marriage  with  his 
foe:   "  Hit  is  as  dou  wylle;/I  cane  say  na  more." 

Origin.  The  opening  episodes  of  the  romance  tell  of  the 
attack  made  on  the  hunting  preserves  of  Degrevant  by  his 
powerful  neighbor,  of  Degrevant's  hasty  return  from  the  Holy 
Land,  of  his  challenge,  of  his  foe's  second  attack,  and  the  battle 
in  which  Degrevant  and  his  men  hunt  their  opponents  like  deer 
through  the  fen.  All  this  has  been  likened  by  Dr.  Rickert  (p. 
xlvii)  to  the  hunting  raid  and  the  battle  of  Otterburn  (1388), 
celebrated  in  the  two  famous  ballads  of  the  Battle  of  Otter- 
burn  and  the  Hunting  of  the  Cheviot.  Realistic  as  are  the 
ballads  and  the  romance,  the  latter  has  nothing  of  the  race  feud 
of  the  Percy  and  the  Douglas.  It  is  altogether  local  and  per- 
sonal in  tone.  The  author  seems  to  have  a  definite  region  in 
mind,  a  countryside  of  forest  and  fen,  of  glades  and  stream,  of 
fell  and  "  ling,"  yet  near  the  sea,  for  the  tide  fills  the  moat  of 
the  castle  of  Degrevant's  enemy,  and  Degrevant's  rival  comes 
by  sea  from  France.  These  details  might  suggest  portions  of 
the  sea  coast  of  Cumberland  or  Westmoreland,  if  the  reference 
to  Degrevant  as  riding  overnight  out  of  Westwale  is  not  to  be 
taken  literally.  As  for  the  defiant  hunting  and  subsequent 
battle,  —  Nessler  (Geschichte  der  Ballade  Chevy  Chase,  Berlin, 
191 1),  has  pointed  out  that  similar  episodes  are  to  be  found  in 
Geoffrey  of  Monmouth's  Historia  (1,  c.  12),  in  Guy  of  Warwick 
(v.  6714),  in  Degare,  Triamore,  —  in  other  words,  that  this  bit 
of  typical  narrative  antedated  the  ballads  and  presumably  also 
the  battle  of  Otterburn  itself. 

Degrevant  has  the  closest  analogies  in  plot  to  two  Middle 
English  romances,  Eger  and  Grime  and  the  Erie   of  Tolous. 


3io  MEDIAEVAL   ROMANCE   IN   ENGLAND 

Like  the  first,  Degrevant  presents  a  vivid  and  unusual  setting, 
a  wild  country  where  the  dun  deer  run  in  dales  neighbored  by 
the  sea.  Certain  scenes,  such  as  the  first  meeting  of  the  lovers 
in  the  garden,  the  lavish  richness  of  the  lady's  room  where  the 
hero  is  cared  for,  and  the  breaking  of  the  parks 5  and  the  killing 
of  his  enemy's  deer  by  the  angry  hero,  suggest  some  actual  re- 
lationship between  the  two  stories.  In  both  may  be  noticed 
likewise  the  tendency  toward  elaborate  description. 

With  the  Erie  of  Tolous,  Degrevant  has  even  more  evident  con- 
tacts. Both  poems  are  found  in  the  Thornton  manuscript  and 
begin  with  very  similar  prayers  to  the  Trinity.  In  each  the 
hero  is  attacked,  in  Degrevant  in  a  hunting  raid,  in  the  Erie 
through  invasion  of  his  land  by  a  foe  whom  presently  he  is  to 
put  utterly  to  rout.  Becoming  interested  in  the  daughter  {Deg- 
revant), in  the  wife  {Erie)  of  his  foe,  the  hero  makes  a  secret 
journey  to  her  home,  accompanied  by  only  one  man,  his  faith- 
ful squire  {Degrevant) ,  a  treacherous  captive  {Erie).  The  wife 
of  the  hero's  enemy  reproaches  him  for  his  cruelty  to  the  young 
man.  The  hero  makes  shift  to  see  the  lady  in  a  garden  {Degre- 
vant), in  a  chapel  {Erie).  On  departing  from  her  home  the  hero 
is  treacherously  ambushed,  but  after  a  terrific  combat  fights  his 
way  clear.  The  ambush  is  planned  by  a  forester  {Degrevant), 
by  the  captive  {Erie).  Later  on  the  hero  enters  his  enemy's 
land  in  order  to  participate  in  a  great  tournament  {Degrevant), 
in  an  ordeal  by  battle  {Erie).  After  his  victory  peace  is  made 
and  the  hero  marries  the  daughter  {Degrevant),  the  widow 
{Erie)  of  his  former  foe.  From  these  resemblances  it  is  hardly 
possible  to  doubt  that  the  author  of  Degrevant  made  use  of  the 
extant  Middle  English  version  of  the  Erie  of  Tolous. 

5  In  Degrevant  the  word  park  is  used  in  the  legal  sense  of  a  land  en- 
closure which  was  expressly  intended  for  deer.  Cf.  Turner,  Select  Pleas  of 
the  Forest,  Selden  Soc,  Lond.,  iqoi,  p.  cxv,  who  noted  that  the  word  is 
still  often  found  as  a  field  name  in  the  west  of  England.  Degrevant,  1.  107, 
refers  to  the  breaking  of  the  parks;  1.  143,  to  making  them  " commoune "  ; 
1.  145,  to  re-enclosing  them.  Chase  meant  a  private  forest  and  warren 
(Degrevant,  1.  1771),  the  land  over  which  the  exclusive  right  of  hunting  ex- 
tended  (Turner,  pp.  cix,  cxxiii). 


SIR    DEGREVANT  311 


BIBLIOGRAPHY 

Texts:  (i)  C,  Camb.  Univ.  Libr.  Ff  1,  6,  ed.  Halliwell,  Thornton 
Romances,  pp.  177-256;  (2)  L,  Thornton  MS.  Lincoln  Cath.  A,  1,  17, 
excerpts  by  Halliwell,  ibid,  notes;  stanzas  1-6,  both  MSS.  by  Schleich, 
Eng.  Stud,  xii,  140-2  (1888-9).  Both  MSS.  entire,  ed.  K.  Luick,  Sir 
Degrevantj  Wiener  Beit.  z.  eng.  Phil.  Bd.  xlvii  (191 7).  Trans.  Rickert, 
Romances  of  Love,  p.  107. 

Studies:   Cf.  Wells,  Manual,  p.  785. 
Finsterbuch,  F.     "  Der  Versbau  der  mitteleng.  Dichtungen  Sir  Perceval 

of  Galles  (pp.   1-174)   und  Sir  DegrevantP     Wiener  Beit.  z.  eng. 

Phil.  xlix.    Wien,  1920.  rev.  Liter aturbl. 
Luick.     See  Texts. 
Medary,  M.  P.    "  Stanza-Linking  in  Middle  English  Verse,"  Rom.  Rev. 

vii,  243-70  (1916),  Degrevant,  pp.  255-6. 
Brown,  L.     "  On  the  Origin  of  Stanza-Linking  in  English  Alliterative 

Verse,"  Rom.  Rev.  vn,  271-283   (1916). 
Rickert.    See  Texts. 


THE   HISTORY   OF   SIR   EGER,   SIR   GRIME, 
AND    SIR    GRAYSTEELE 

Versions.  The  history  of  Sir  Eger  can  be  retraced  in  Eng- 
land only  to  the  end  of  the  fifteenth  century.  Of  the  copies  * 
once  current  only  two  have  thus  far  been  found.  The  best  and 
oldest,  a  poem  of  1474  lines,  roughly  written  in  short  riming 
couplets,  and  divided,  as  it  almost  certainly  was  not  in  earlier 
versions,  into  six  parts  or  cantos,  is  in  Bishop  Percy's  Folio 
manuscript  (P).  The  second  text,  a  rambling,  somewhat  inco- 
herent version  of  double  the  length  and  half  the  effectiveness 
of  P,  is  known  only  in  the  Aberdeen  print  (L)  of  1711  and  its 
modern  reprints.  In  vocabulary  and  orthography,  however, 
Reichel  (p.  4  and  n.  to  v.  1006)  thought  this  older  than  the 
Folio  version.  In  each  version  the  language  is  so  modernized 
and  corrupted  that  it  is  doubtful  if  even  a  much  more  careful 
study  than  has  yet  been  given  to  it  could  determine  the  home 
of  the  original  poem2  or  indicate  with  exactness  the  period  of 
its  composition.  The  extant  versions  seem  to  be  independent 
derivatives  of  this  lost  original,  but  removed  from  it  by  a  num- 
ber of  lost  intermediary  texts  (Reichel,  p.  15). 

The  earliest  known  allusion  to  the  romance  comes  in  1497, 
when  the  Treasurer's  Accounts  of  James  IV  of  Scotland  state 
that  "  twa  fithelaris  sang  Gray  Steil "  to  the  king  at  Stirling 
(Hales,  1,  342).  It  was  not  even  then  a  new  song,  to  judge 
from  the  affiliations  of  the  story  with  older  romance.  From  the 
sixteenth  century  numerous  references  to  Eger  are  recorded  by 
both  Laing  and  Hales  (p.  343)  ;  in  the  Complaynt  of  Scotland, 
1549,  it  is  listed  with  other  known  romances;  the  name  Gray- 
steele  was  used  at  least  three  times  as  a  sobriquet  for  well- 

1  Bishop  Percy  described  a  copy  of  the  romance  which  he  possessed  in 
1800.     Cf .  Furnivall,  p.  342 ;  Complaynt  of  Scotland,  EETSES.  xvn,  p.  lxxix. 

2  Hales  remarked  (p.  342):  "The  language  is  unquestionably  Scottish"; 
Reichel,  p.  10,  referred  to  the  "  mittelscottischen "  original  text.  Beyond 
these  mere  assertions  no  one  seems  to  have  gone. 

312 


THE   HISTORY   OF   SIR   EGER  3^ 

known  personages  in  the  sixteenth  century;  and  several  writers 
of  note  allude  to  the  romance,  —  John  Taylor,  the  Water  Poet 
(1623),  especially  comments  on  its  popularity  in  Scotland,  and 
couples  it  in  this  connection  with  Sir  Degare.  Editions  of  the 
romance  were  printed  in  1599,  1602,  1606,  1687  (Rickert,  p. 
182). 

The  romance  tells  of  the  sworn-brotherhood  between  two 
noble  knights,  Eger  and  Grime;  of  the  defeat  that  Eger  en- 
dures at  the  hands  of  Graysteele,  a  champion  who  challenges 
all  comers  in  his  land ;  of  the  pride  of  Winglayne,  beloved 
of  Eger,  who  will  have  none  for  husband  but  an  unconquered 
knight ;  of  Eger's  fear  to  lose  her  when  he  has  been  defeated  by 
Graysteele,  of  Grime's  battle  in  the  guise  of  Eger  with  Gray- 
steele, and  of  Graysteele's  death;  of  the  fame  of  the  exploit; 
of  the  humility  of  Winglayne  before  the  supposed  champion 
and  of  her  marriage  with  Eger.  Grime  himself,  more  truly  the 
hero  of  the  story  than  Eger,  meanwhile  wins  for  himself  the 
lovely  Lady  Loospaine  (v.  1407;  L  version,  Lillias),  who  has 
cared  for  Eger  after  his  defeat,  and  later  nobly  welcomed  and 
aided  Grime.  The  plot  has  no  special  distinction,  but  no  one 
who  has  commented  on  the  romance  has  failed  to  feel  the  nota- 
ble charm  of  its  style,  vividly  pictorial  as  it  is,  quaintly  hu- 
morous, terse  or  tender  at  will,  and  with  a  power  of  characteri- 
zation which  justifies  Bishop  Percy's  verdict  that  this  is  one  of 
the  best  of  the  ancient  epic  tales  in  the  Folio  (Hales,  p.  353). 
A  single  passage,  interesting,  too,  from  the  point  of  origins,  may 
serve  as  illustration. 

Eger  returns  from  his  conflict  with  Graysteele,  battered  in 
body  but  still  more  battered  in  spirit.  He  sits  on  his  bed  and 
tells  his  trouble  to  Grime;  he  is  boyish,  petulant,  plaintive,  and 
most  frankly  bewildered  at  his  own  discomfiture.  He  has  had 
the  best  of  weapons,  the  best  steed,  the  utmost  confidence  in 
himself,  the  best  possible  spirit  for  the  adventure.  But  Gray- 
steele has  defeated  him,  and  for  a  final  mark  of  ignominy,  has 
cut  off  his  little  finger.  The  account  most  happily  characterizes 
the  impetuous  youth  and  brings  out  the  contrast  between  him 
and  the  silent  stronger  Grime,  in  truth  "  a  dogged,  canny  Scot," 
who  vainly  tries  to  comfort  him.  Besides  this  the  passage  is 
especially  rich  in  the  pictorial  detail  which  characterizes  the 


314  MEDIEVAL   ROMANCE    IN    ENGLAND 

romance.  Aptly  it  describes  the  "  Forbidden  Land  "  kept  by 
Graysteele,  his  "  fresh  iland  "  with  its  towered  castles,  and  the 
mighty  knight  himself  in  all  his  red  magnificence.  It  tells  of 
the  castle  to  which  on  his  homeward  journey  the  suffering  Eger 
comes  one  moonlight  night,  of  the  little  arbor  he  enters,  of  the 
lovely  Lady  Loospaine,  "  red  as  rose  in  rain  "  who  cares  for  his 
wounds  and  later  comforts  him  with  her  sweet  singing  as  he  lies 
in  her  rich  chamber.  The  luxury  of  her  abode,  here  and  else- 
where in  the  romance,  is  repeatedly  emphasized. 

Origin.  The  passage  just  summarized  indicates  the  basic 
type  to  which  part  at  least  of  the  story  of  Eger  belongs.  How- 
ever much  it  has  been  rationalized  by  a  story-teller  more  inter- 
ested in  personality  and  in  scenes  of  fact  rather  than  of  fancy, 
it  belongs  with  the  group  of  Fairy  Mistress  stories.  Loospaine's 
beautiful  Otherworld  abode  lies  beyond  a  river  in  the  "  Forbid- 
den Land  " ;  its  entrance  is  guarded  by  the  gigantic  Graysteele ; 
the  lady  herself,  of  superlative  beaut}^  and  healing  powers,3  is 
recognizable  as  fee,  and  the  main  business  of  the  tale  is  with 
the  combat  through  which  she  is  won  by  a  mortal  knight.  Scho- 
field's  brief  comment  (p.  232)  that  "  at  bottom  this  seems  to 
be  a  story  of  the  Iwain  type,"  is  more  than  confirmed  by  a 
closer  comparison.  Iwain,  written  about  11 70  by  Chretien  de 
Troyes,  was  translated  in  the  fourteenth  century  into  Middle 
English,  and  from  this  version  (vv.  425-26)  the  Eger  poet  bor- 
rowed at  least  one  couplet  (vv.  119-20)  and  many  incidents  of 
his  narrative.  The  ill-fame  of  Graysteele,  the  secretly  under- 
taken exploit  of  Eger  to  find  him,  Eger's  overthrow,  his  account 
of  it  to  Grime,  Grime's  present  departure,  also  in  secret,  to  at- 
tempt the  same  adventure,  his  victory,  and  marriage  with  the 
widowed  Lady  Loospaine,  all  closely  parallel  the  first  part  of 
Iwain.4     Thus  Colgrevance  in  Arthur's  court  tells  of  his  wan- 

3  Loospaine  gives  to  Eger  a  magic  potion  which  at  once  restores  his 
strength,  but  the  effect  of  it  is  lost  when  he  returns  to  his  own  domain. 
In  the  Old  Irish  Serglige  Conculaind  and  the  Old  French  Iwain  the  fairy- 
heroine  gives  the  hero  a  potion  which  restores  his  lost  wits.  Cf.  A-  C.  L. 
Brown,  Iwain,  Harvard  Studies  vm,  34-40. 

4  Ed.  by  W.  Foerster,  3rd  ed.  Halle,  1006.  For  the  Middle  English  ver- 
sion, cf.  Schleich's  ed.  Oppeln,  1887;  Billings,  Guide,  pp.  156-60;  Wells, 
Manual,  65-67,  771. 


THE   HISTORY    OF    SIR   EGER  3!5 

derings  in  the  strange  forest  (Chretien's  Broceliande),  of  his 
fair  welcome  at  a  castle,  of  his  encounter  with  a  Giant  Herds- 
man who  directs  him  to  a  Perilous  Well,  of  his  defeat  there  at 
the  hands  of  a  knight  (Chretien's  Esclados  the  Red).  Thus 
Iwain,  having  heard  the  tale,  goes  secretly  away  to  try  his 
fate,  meets  and  kills  Esclados,  and  ultimately  weds  Laudine, 
the  Red  Knight's  widow.  Although  Eger  is  much  condensed, 
these  structural  likenesses  in  the  narrative  indicate  an  essential 
dependence  on  the  older  story.  The  important  change  in  Eger 
which  makes  Graysteele  the  enemy  of  Loospaine,  the  murderer 
of  her  brother  and  her  one-day  husband,  instead  of  her  husband 
and  protector  as  was  Esclados  in  Chretien's  tale,  may  be  ac- 
counted for,  at  least  according  to  Brown's  theory  in  connection 
with  Iwain,  as  representing  a  tendency  by  which  in  primitive 
tales  the  servant  of  a  fee,  "  originally  only  a  creature  of  the 
fee,  sent  out  by  her  to  test  the  hero's  valor,"  takes  on  the  guise 
of  a  suitor  or  a  husband  whom  it  becomes  necessary  to  over- 
throw before  she  can  be  won  by  a  mortal  lover.5  Brown  (Iwain, 
p.  50)  has  indicated  the  probable  confusion  that  took  place  even 
in  Chretien's  Celtic  sources  between  the  fee  story  proper,  and 
the  story  of  a  giant  and  his  unwilling  captive.  There  is  ancient 
enough  authority,  therefore,  for  the  hostility  between  Loospaine 
and  Graysteele. 

In  addition  to  the  structural  similarity  to  Iwain,  Eger  has 
still  some  other  possibly  Celtic  connections.  One  of  these  is 
the  strange  attribute  ascribed  to  Graysteele.  In  preparing 
Grime  for  his  combat  Loospaine  tells  him,  as  "  no  woman  alive 
knoweth  so  well  "  as  she,  that  Graysteele's  power  increases  by 
a  man's  strength  with  every  hour  from  midnight  until  noon, 
and  wanes  correspondingly  in  the  afternoon.     There  is  obvious 

5  Cf.  Brown,  ibid.,  ch.  iv,  The  Combat  Motive.  As  regards  Eger,  the  L 
version,  which  ends  with  the  marriage  of  Eger  to  Loospaine  after  the 
death  of  Grime,  represents,  perhaps,  the  influence  of  a  more  primitive  ver- 
sion than  P.  In  the  simple  form  of  the  story  there  would  be  but  one  hero 
and  one  heroine,  as  in  the  ballads  of  Sir  Cawline  or  Sir  Lionel,  and  it  would 
be  he  who  won  the  lady  by  his  own  exploit.  In  this  case  Eger  would 
originally  have  won  Loospaine  for  himself,  as  in  L  he  ultimately  does.  His 
present  unsatisfactory  part  in  the  romance  may,  therefore,  be  due  to  the 
introduction  of  the  Sworn-Brotherhood  motif,  and  the  consequent  doubling 
of  the  hero's  role.  On  the  other  hand  the  two  heroes  may  both  have  be- 
longed to  the  original  story,  if  that  were,  as  is  suggested  below,  a  variant 
of  the  Nibelungenlied. 


3i6  MEDIAEVAL   ROMANCE    IN    ENGLAND 

analogy  here  with  Gawain's  increase  of  strength  from  morn 
to  noon,  a  trait  which  has  been  held  to  connect  Gawain 
with  a  Celtic  solar  myth.6  Of  less  primitive  character,  though 
in  entire  accord  with  traditional  descriptions  of  the  magnificent 
beings  of  the  Celtic  Otherworld  as  they  are  rationalized  in  medi- 
aeval romances,  is  the  emphasis  on  Graysteele's  splendor  and 
might.  Like  Esclados  in  Chretien's  Iwain,  like  Valerin,  Guin- 
evere's Otherworld  lover  in  Ulrich  von  Zatzikhoven's  Lanzelet 
(ed.  K.  Halm,  Frankfurt,  1845,  *•  4972  ff-)>  or  Gasozein  in  Hein- 
rich  von  dem  Turlin's  Diu  Crone,7  Graysteele  is  a  warrior  of 
sumptuous  appearance,8  a  noble  and  heroic  figure,  yet  a  foe  to 
the  lady  of  the  story.  Like  the  mysterious  knight,  who  in  the 
Old  French  Lai  de  VEspine  kept  the  ford  on  the  vigil  of  St. 
John,  Graysteele  defends  a  "  riding  place  "  across  his  river,  and 
fights  with  the  true  hero  of  the  story.  Of  their  adventure  a  lai 
might  well  have  been  made,  even  as  the  French  poet  says  the 
"  Bretons  "  made  one  concerning  the  fight  at  the  Ford  of  the 
Thorn.  Finally,  in  the  Eger  poet's  description  of  Loospaine  her- 
self, there  is  a  detail  for  which  there  seems  no  antecedent  save  in 
Celtic  tradition.  The  lady  is  said  (vv.  619-21)  to  have  between 
her  eyes  a  curious  pin  spot  of  white  and  red,  and,  when  he  looks 
upon  it,  poor  Grime  forgets  all  other  things.  For  Loospaine's 
lovely  blemish  the  only  notable  parallel  is  the  famous  and  fatal 
"  love  spot  "  which  made  the  Irish  hero  Diarmaid's  beauty  irre- 
sistible to  the  women  who  beheld  it.9 

These  traditional  Celtic  elements,  if  such  they  be,  offer  an 
interesting  contrast  to  the  affiliations  of  the  Eger  story  with  so 
complete  a  product  of  Anglo-Saxon  and  Teutonic  imagination  as 
Beowulf.  Grendel's  strength  of  thirty  men,  his  character  as  a 
demon  of  the  fens,  the  arm  which  Beowulf  tears  from  him,  and 

6  Cf.  J.  Weston,  The  Legend  of  Sir  Gawain,  Lond.,  1897,  pp.  12;  G. 
Paris,  Hist.  Lit.  xxx. 

?  Ed.  by  G.  H.  Scholl,  Tubingen,  1852,  vv.  3699;  cf.  K.  G.  T.  Webster, 
"  Arthur  and  Charlemagne,"  Eng.  Stud,  xxxvi,  341 ;  Schoepperle,  Tristan,  11, 

535- 

8  See  Orfeo  here,  note  7.  Hale's  idea  (p.  351)  that  "the  brilliant  opu- 
lence of  Graysteele's  appearance  points  to  an  Oriental  origin,"  is  unsupported 
by  any  evidence. 

9  For  full  bibliography  concerning  the  ancient  texts  of  the  Diarmaid 
story  and  of  the  modern  Irish  and  Scotch  oral  versions,  see  Miss  Schoepperle's 
Tristan,  11,  399,  n.  2,  401-2. 


THE   HISTORY   OF   SIR   EGER  31 7 

exhibits  as  a  trophy  in  Hrothgar's  hall,  are  dimly  recalled  in 
the  mighty  stature  and  prowess  of  Graysteele,  in  the  detailed 
account  of  his  great  hand  which  Grime  cuts  off,  and  gives,  in 
token  of  her  enemy's  death,  to  Loospaine,  and  of  her  exhibition 
of  it  in  her  father's  hall  before  all  the  nobles.  So  also  in  the 
ballad  of  Sir  Cawline  (Child,  No.  61),  which  in  its  first  adven- 
ture seems  derived  from  the  same  source  as  Eger,  the  Eldrige 
King  who  haunts  the  moors  at  night,  seeking  whom  he  may 
destroy,  who  is  wounded  and  flees  away,  leaving  his  hand  to 
Cawline,  is  to  some  degree  reminiscent  of  Grendel.10  Perhaps 
the  transition  from  the  epic  monster  to  the  strange  warrior  of 
the  ballad  may  be  dimly  traced  through  that  story  of  Gervase 
of  Tilbury  to  which  Scott  (Minstrelsy  of  the  Scottish  Border, 
ed.  Henderson,  Edin.,  1902,  11,  319)  first  called  attention.  It 
told  of  a  ghostly  warrior  at  Ely  who  could  be  summoned  on 
moonlit  nights  at  a  certain  entrenchment  by  any  challenger. 
In  this  and  in  the  current  folk-tale,  also  referred  to  by  Scott 
(Marmion,  n.  4),  concerning  the  bloody  spirit  who  haunted  a 
forest  in  the  northern  Highlands,  insisting  on  battle  with  all 
whom  he  met,  the  Warrior  tale  has  no  hint  of  any  romantic 
interest.  In  comparison  with  Eger,  therefore,  it  is  of  interest 
only  as  showing  the  distribution  and  continuance  of  a  story 
from  which  certain  primitive  details  might  have  been  absorbed 
by  the  romance. 

The  possible  mixture  of  Celtic  and  Teutonic  elements  in  the 
account  of  Graysteele  and  Grendel,  is  further  suggested  by  the 
presence  in  Eger  of  certain  ideas  that  have  been  held  to  be 
characteristically  Teutonic.  Among  these  Dr.  Rickert  (p.  xxiii) 
mentioned:  the  sworn-brotherhood  of  the  heroes,  the  defence  of 
a  ford  or  pass,  the  use  of  the  cut-off  fingers  as  evidence  of 
death,  the  naming  of  the  mysterious  sword  Erkyin  or  Edgeking, 
brought  from  beyond  "  the  Greekes  sea,"  and  so  terrible  that 
"  no  man  durst  abyde  the  winde  "  of  it  before  his  face.11    To 

10  Sir  Cawline  is  perishing  with  love  for  a  Princess;  she  bids  him  kill 
"  the  eldridge  king "  ;  the  strange  champion  comes.  "  By  an  aukeward 
stroke "  (v.  1029)  Cawline  cuts  off  his  hand  and  brings  it  back  to  the 
Princess.  So  also  "  with  an  arkward  stroke "  (v.  1029)  Grime  strikes 
Graysteele  on  the  knee. 

11  It  should  be  noted  that  although  these  features  do  appear  in  Teutonic 
story,  they  appear  also  in  Irish,  Norse,  and  French  epic.     The  Defense  of  a 


3i8  MEDLEVAL   ROMANCE    IN   ENGLAND 

these  may  be  added  the  important  episode  of  the  ruse  which  is 
practiced  on  the  proud  Winglayne.12  Grime  wins  her  for  his 
friend  Eger  no  less  surely  than  Sigurd  won  Brunhilde  for 
Gunther,  although  the  details  of  the  story  differ  widely.  In 
the  P  version  of  Eger  the  ruse  is  never  revealed,  but  in  L,  Wing- 
layne, who  has  gone  with  Eger  to  visit  his  newly  wedded  friend, 
leaves  her  husband  in  furious  anger  when  she  discovers  the 
truth.  One  may  admit  with  entire  justice  the  stupidity  of  L; 
but  in  the  unmotivated  death  of  Grahame  (Grime),  who  falls 
sick  and  dies  shortly  after  his  marriage,  in  the  separation  of 
Winglayne  and  Eger,  is  it  not  possible  to  catch  some  faint  re- 
membrance, more  clearly  preserved  than  in  P,  of  that  most 
famous  story  in  which  the  hero  who  had  won  the  Valkyrie  only 
to  give  her  up,  and  himself  to  wed  another,  ultimately  paid  for 
his  deceit  with  his  life?  English  romance13  elsewhere  shows 
the  influence  of  the  famous  legend,  and  late  though  the  L  ver- 
sion is,  it  is  not  impossible  that  its  lost  original  owed  to  the 
older  romance  not  only  the  ruse  of  Grime,  but  its  tragic  sequel. 
Eger,  as  it  stands,  seems  then  to  show  the  combination  of  a 
simple  folk-tale  in  which  a  hero  wins  his  bride  by  killing  a 
superhuman  creature  with  a  Celtic  Fairy  Mistress  story  modi- 
fied by  French  romancers,  plus  the  story  of  a  Valkyrie-like 
heroine  and  the  ruse  by  which  she  is  married  to  a  conquered 
man.  Besides  this  mixture  of  racial  stories,  of  human  and 
supernatural  elements,  certain  historic  and  realistic  features 
claim  attention.  Investigation  along  these  lines  has  been  carried 
no  further  than  the  brief  pioneer  suggestions  of  Dr.  Rickert  (p. 
xxiii).  The  exact  description  (v.  101)  in  the  poem  of  Gray- 
steele's  land,  a  tract  lying  along  a  river  which  soon  empties  into 


Ford  is  an  especially  frequent  theme  in  Old  Irish  epic.  Cf.  Die  altirische 
Heldensage  Tain  bo  Cualnge,  ed.  Windisch,  Leipzig,  1908,  passim.  For  the 
Friendship  and  Sworn-Brotherhood  themes  see  note  7  under  Amis  here  and 
note  3  under  Athelston. 

12  On  the  Valkyrie  nature  of  Brunhilde  see  V.  Gildersleeve,  Mod.  PhU. 
vi,  343-75  (1908-09).  It  is  the  ruse  practised  on  Winglayne  which  connects 
her  with  Brunhilde  and  which  differentiates  her  from  the  many  heroines  of 
folk  story  who  are  in  love  with  a  giant  or  a  monster  and  who  set  their 
mortal  lovers  apparently  impossible  tasks  in  order  to  be  rid  of  them.  Cf. 
Gerould,  The  Grateful  Dead,  ch.  11,  "  The  Lady  and  the  Monster." 

13  See  Sir  Degrevant  here;  also  under  the  bibliography  of  Beves  of 
Hampton  the  articles  by  Hibbard  and  Brockstedt. 


THE   HISTORY    OF    SIR   EGER  3Ig 

the  sea,  of  the  seven  cities  by  the  sea  (v.  935),  of  the  two  fords 
that  cross  the  river,  of  the  island  on  which  all  comers  have  to 
encounter  Graysteele,  suggested  to  her  that  strip  of  "  Debatable 
Land  "  along  the  Solway,  between  the  Esk  and  the  Sark,  which 
was  the  scene  of  so  many  conflicts  between  the  English  and 
Scottish  borderers  in  the  fifteenth  and  sixteenth  centuries.  In 
this  district  one  of  the  principal  clans  was  the  Grahams,  and 
to  their  name  the  hero  Grime  (Grahame  in  the  L  version)  seems 
to  owe  his  own.  Grime  is  said  (v.  21)  to  be  of  Garwicke, 
which  Scott  identified  with  Carrick  in  Ayrshire.  In  the  six- 
teenth century  Grime's  fight  with  Graysteele  was  localized,  ac- 
cording to  the  reference  in  Sir  David  Lyndsay's  Interlude  of 
the  Auld  Man  and  his  Wife,  "  necht  half  a  myle  beyond  Kin- 
neill,"  a  name  Dr.  Rickert  identified  with  that  of  a  stream  in 
Dumfriesshire.  The  fact  that  the  ballad  Sir  Lionel  (Child,  No. 
18),  which  is  at  least  slightly  related  to  Eger,  is  localized  by 
the  Esk,  is  again  an  indication  of  the  presence  in  some  form  of 
the  Graysteele  legend  in  the  Solway  district.  Further  investi- 
gation of  these  tempting  clues  is  greatly  to  be  desired,  even 
though  they  lead  but  into  a  "  Land  of  Doubt,"  —  a  name  that 
in  the  1711  edition  of  the  poem  was  piously  substituted  for  the 
"  Forbidden  Country  "  of  old  romance. 


BIBLIOGRAPHY 

Texts:  (i)  Percy  Folio  MS.  (1474  lines),  ed.  Hales  and  Furnivall, 
1,  340-400,  Lond.,  1868;  (2)  Eger  and  Grime  (2861  lines),  Aberdeen, 
1 71 1,  repr.  D.  Laing,  Early  Metrical  Tales,  pp.  1-96,  Edin.,  1826; 
1899:  abstract  with  quotations,  Ellis,  Specimens,  pp.  546-567  (1848). 
Trans,  from  the  Folio    E.  Rickert,  Romances  of  Friendship,  pp.  137  ff. 

Studies  : 
Hales  and  Furnivall.    See  Texts,  pp.  341-54. 

Reichel,  G.  "  Studien  zu  der  schottischen  Romanze:  The  History  of 
Sir  Eger,  Sir  Grime,  and  Sir  Graysteele.,f  Eng.  Stud,  xrx,  1-66 
(1894).    A  comparison  of  the  versions,  notes  on  the  text,  glossary. 


TABLE  OF  ABBREVIATIONS  AND  REFERENCES 

Italics  here  indicate  the  abbreviated  form  of  reference  used  throughout  this 
book.  This  list  of  references  is  supplementary  to  the  special  bibliographies  for 
each  romance.  It  includes  all  general  collections  of  Middle  English  romance 
texts  of  whatsoever  date;  also  the  general  histories  of  mediaeval  literature  printed 
between  1900-1923  and  the  special  studies  which  cover  several  romances  or  deal 
with  themes  recurrent  in  Continental  and  Middle  English  romance.  See  Index 
for  classification  by  subject. 

Amer.  Jour,  of  Phil.    American  Journal  of  Philology.     Baltimore,  1880- 

Anglia.    Zeitschrift  fur  englische  Philologie.     Halle,  1877- 

Anglia  Bbl.    Beiblatt  zur  Anglia.    Halle,  1890- 

Archiv  fur  das  Studium  der  neueren   Sprachen  und  Literaturen,  ed.  Herrig. 

Braunschweig,  1849- 
Arnold,  F.  C.    Das  Kind  in  der  deutschen  Litteratur  des  XI-XV  Jahrhunderts. 

Diss.     Greifswald,  1905 
Aron,  A.    Traces  of  Matriarchy  in  Germanic  Hero  Lore.    Univ.  of  Wisconsin 

Studies  in  Language  and  Literature,  IX.  1920 
Ashton,  J.    Romances  of  Chivalry.     London,  1890 
Ausgaben  und   Abhandlungen   aus   dem   Gebiete   der    romanischen   Philologie. 

Marburg,  1881- 

Baake,  W.    Die  Verwendung  des  Traummotivs  in  der  englischen  Dichtung  bis 

auf  Chaucer.    Diss.    Halle,  1906 
Baldwin,  C.    An  Introduction  to  English  Mediaeval  Literature.    New  York,  19 14. 
Baskerville,  C.  R.    Early  Romantic  Plays  in  England.    Mod.  Phil.,  1916,  XIV, 

229-51;  467-512 
Becker,  P.  A.    Grundriss  der  altfrz.  Literatur.    Heidelberg,  1907 
Bedier,  J.    Les  Legendes  Epiques.    Paris,  1908-13,  1914-21 
Beszard,  L.    Les  larmes  dans  l'epopee  jusqu'a  la  fin  du  Xlle  siecle.    Zeitschrift 

fur  rom.  Philologie,  1903,  XXVII 
Billings,  A.  H.    A  Guide  to  Middle  English  Metrical  Romances.    N.  Y.,  1901, 

1905 
Bolte,  J.  and  G.  Polfvka.    Anmerkungen  zu  den  Kinder-  und  Hausmarchen  der 

B ruder  Grimm.    Leipzig,  19 13-18 
Bonilla  y  San  Martin,  A.    Libros  de  Caballerias.    Madrid,  1907 
Booker,  J.  M.    A  Middle  English  Bibliography  to  1907.    Heidelberg,  191 2 
Brandl,  A.    Spielmannsverhaltnisse  in  fruhmittelengl.    Zeit.  Kon.-preuss.    Akad. 

der  Wissenschaft  Sitzungsbericht.     Berlin,  1910,  XL,  873-92 
British  Museum  Catalogue  of  Additional  Manuscripts.     London,  1906-10.     B. 

M.  Catalogue  of  Romances,  H.  L.  D.  Ward,  vols.  I,  II,  1883-93;   J-  Herbert, 

vol.  Ill,  London,  1910 
Brown,  C.  A.    A  Register  of  Middle  English  Religious  Verse.    Oxford,  1916-20 

Cambridge  History  of  English  Literature,  vols.  I,  II.    Cambridge,  1907-08 
CCCbg.    Corpus  Christi  College,  Cambridge 

321 


322      TABLE    OF   ABBREVIATIONS   AND    REFERENCES 

Chambers,  E.  K.    The  Medieval  Stage.    Oxford,  1902 

Chauvin,  V.    Bibliographie  des  Ouvrages  Arabes.    Leipzig,  Liege,  189  2-1909 

Child,  F.  J.    The  English  and  Scottish  Ballads.    5  vols.    Boston,  1898 

Clephan,  R.  C.     The  Tournament.     London,  1919 

Comfort,  W.  W.  Character  Types  in  the  Old  French  Chansons  de  Geste.  Pub- 
lications Modern  Language  Association,  1906,  XXI,  279-434 

Crane,  R.  The  Vogue  of  Mediaeval  Chivalric  Romance  during  the  English  Re- 
naissance.    Diss.     Menasha,  19 19 

Creek,  H.  Character  in  the  " Matter  of  England"  Romances.  Diss.  Urbana, 
1911.     Cf.  JEGP.  x,  429-53  (1911) 

Cripps-Day,  F.  H.     The  History  of  the  Tournament  in  England  and  in  France. 

London,  19 18 
Critchlow,  F.    On  the  Forms  of  Betrothal  and  Wedding  Ceremonies  in  the  Old 

French  Romans  d'Aventure.    Diss.     Chicago,  1906 
Curry,  W.  C.     The  Middle  English  Ideal  of  Personal  Beauty  in  the  Metrical 

Romances,  Legends,  Chronicles.    Diss.     Baltimore,  1916 

Deutschbein,  M.    Studien  zur  Sagengeschichte  Englands.    Cothen,  1906 

Easter,  B.  De  la  Warr.  A  Study  of  Magic  Elements  in  the  Romans  d'Aventure. 
Diss.     Baltimore,  1906 

EETS.    Early  English  Text  Society,  Original  Series.    London,  1864- 

EETSES.    Early  English  Text  Society,  Extra  Series,  1867- 

Edwardes,  M.  Summary  of  the  Literatures  of  Modern  Europe  to  1400.  London, 
1907 

Ellis,  G.  Specimens  of  Early  English  Metrical  Romances.  London,  1805;  re- 
vised by  Halliwell,  1848 

Eng.  Stud.    Englische  Studien.    Heilbronn,  Leipzig,  1877- 

Esdaile,  A.  J.  A  List  of  English  Tales  and  Prose  Romances  printed  before  1740. 
London,  191 2 

Faral,  E.  Les  Jongleurs  en  France  au  Moyen  Age.  Paris,  1910;  Sources  Latines 
des  Contes  et  Romans  Courtois  du  Moyen  Age.    Paris,  1913 

Farnsworth,  W.  O.  Uncle  and  Nephew  in  the  Old  French  Chansons  de  Geste. 
N.  Y.,  1913 

Fellinger,  F.  Schwangerschaft  und  Geburt  in  der  altfrz.  Literatur.  Gottingen, 
1907;  Das  Kind  in  der  altfrz.  Literatur.    Gottingen,  1908 

Foerster,  W-  Der  Feuertod  als  Strafe  in  der  altfrz.  erzahlenden  Dichtung. 
Studien  zur  engl.    Phil.  L.  Halle,  1913 

Folk  Lore  Society  Publications.    London,  1868- 

Frahm,  W.  Das  Meer  und  die  Seefahrt  in  der  altfrz.  Literatur.  Diss.  Gottin- 
gen, 1914 

Fundenberg,  G.    Feudal  France  in  the  French  Epic.    Princeton,  1918 

Galpin,  S.    "Cortois"  and  "Vilain,"  A  Study  of  the  Distinctions  Made  by  French 

and  Provencal  Poets  of  the  12th,  13th,  and  14th  Centuries.    New  Haven,  1905 
Garnett,  R.,  and  E.  Gosse.    An  Illustrated  History  of  English  Literature.    London, 

1903-04 
Gautier,  L.    Les  fipopees  Francaises.     2  ed.    Paris,  1878-97;   Bibliographie  des 

Chansons  de  Geste.     Paris,  1897 
Geissler,  O.      Religion  und  Aberglaube  in  den  mittelengl.  Versromanzen.     Diss. 

Halle,  1908 


TABLE  OF  ABBREVIATIONS  AND  REFERENCES   323 

Gerould,  G.  H.  Saints  Legends.  Boston,  1916 
Germania.  Stuttgart,  1856-58;  Wien,  1859-92 
Goerke,  G.     Ueber  Tierverwandlungen  in  franz.     Dichtung  und  Sage.     Diss. 

Konigsberg,  1904 
Groeber,  H.    Grundriss  der  romanischen  Literatur.    Strassburg,  1902 
Gross,  C.    Sources  and  Literature  of  English  History.    N.  Y.,  191 5 
Grossmann,  W.  Fruhmittelengl.    Zeugnisse  iiber  Minstrels.    Diss.    Brandenburg, 

1906 
Grundtvig,  S.    Danmarks  Gamle  Folkeviser.    Copenhagen,  1853-90 
Gummere,  F.    The  Sister's  Son.    An  English  Miscellany.    Oxford,  1901 
Guyer,  F.     The  Influence  of  Ovid  on  Crestien  de  Troyes.     Romanic  Review, 

1921,  XII,  97-134;  216-47 

Hagen,  F.  H.  von  der,     Gesammtabenteuer.     3  vols.    Stuttgart  u.  Tubingen,  1850 
Hallauer,  M.  Das  wunderbare  Elemente  in  den  Chansons  de  Geste.      Diss. 

Basle,  1918 
Halliwell-Philipps,  J.  O.    The  Thornton  Romances.    Camden  Society.    London, 

1844 
Halpersohn,  R.     tjber  die  Einleitung  im  altfrz.    Kunstepos.    Diss.    Heidelberg, 

1911 
Hammond,  E.    Chaucer,  A  Bibliographical  Manual.    N.  Y.,  1908 
Hartland,  E.  S.     Primitive  Paternity:    Myth  of  Supernatural  Birth.     London, 

1909-10 
Hartshorne,  C.  H.    Ancient  Metrical  Tales.    London,  1829 

Hazlitt,  W.  C.    Remains  of  Early  Popular  Poetry  of  England.     London,  1864-66 
Herbert,  J.    See  British  Museum  Catalogue 
Heyl,  K.     Die  Theorie  der  Minne  in  den  altesten  Minneromanen  Frankreichs. 

Diss.     Kiel,  191 1.     Marburger  Beitrage  z.  rom.  Phil.  IV. 
Huebner,  W.     Die   Frage   in   einigen   mittelengl.    Versromanen.     Diss.     Kiel, 

1910 

Jacobius,  H.  Die  Erziehung  des  Edelfrauleins  im  alten  Frankreich  nach  Dich- 
tungen  des  XII,  XIII,  und  XIV  Jahrhunderts.  Beihft.  Zts.  fur  roman. 
Philologie,  XVI.    Halle,  1908 

Jahresbericht  iiber  die  Erscheinungen  auf  dem  Gebiete  der  germanischen  Philologie. 
Berlin,  1879,  Leipzig,  1883- 

JEGP.    Journal  of  English  and  Germanic  Philology.    Illinois,  1905- 

Kahle,  R.    Der  Klerus  im  mittelengl.  Versroman.    Diss.     Strassburg,  1906 

Ker,  W.    Mediaeval  English  Literature.    London,  191 2 

Koehler,  R.    Kleinere  Schriften.    Berlin,  1898-1900 

Koerting,  G.    Grundriss  der  Geschichte  der  engl.  Literatur  von  ihren  Anfangen 

bis  zur  Gegenwart.    5th  ed.  Minister,  1910 
Kramer,  P.    Das  Meer  in  der  altfrz.  Literatur.    Diss.     Giessen,  1919 
Kuehn,  O.    Erwahnung  u.  Schulderung  von  korperlichen  Krankheiten  in  altfrz. 

Dichtung.     Diss.     Breslau,   1902;    Medizinisches  aus  der  altfrz.  Dichtung. 

Abhandlungen  z.  Gesch.  der  Medizin,  Hft.  VIII,  Breslau,  1904 
Kurtz,  B.    Studies  in  the  Marvelous.     Univ.  of  California,  Berkeley,  1910 

Lalande  de  Calan,  C.  de.  La  Bretagne  dans  les  Romans  d'Aventures,  Revue 
de  Bretagne,  1903,  XXX,  XXXI;  L'Element  Celtique  dans  les  Romans 
d'Aventures.    Saint-Brieuc,  191 1 


324      TABLE   OF  ABBREVIATIONS   AND   REFERENCES 

Landau,  L.  Hebrew- German  Romances  and  Tales  and  Their  Relation  to  the 
Romantic  Literature  of  the  Middle  Ages.    Leipzig,  19 12 

Lane,  F.  Ueber  Krankenbehandlung  und  Heilkunde  in  der  alten  Liter,  des  alten 
Frankreich.     Diss.     Gottingen,  1904 

Langlois,  C.  La  Societe  francaise  au  XIHe  Siecle  d'apres  dix  Romans  d'Aventure. 
3e  ed.    Paris,  191 1 

Lawrence,  W.  W.  Mediaeval  Story  and  the  Beginnings  of  the  Social  Ideals  of 
English-Speaking  People.    N.  Y.,  191 1 

Leach,  H.  G.    Angevin  Britain  and  Scandinavia.    Cambridge,  1922 

Leibecke,  O.  Der  verabrete  Zweikampf  in  der  altfrz.  Litteratur.  Diss.  Gottin- 
gen, 1905 

Lot-Borodine,  M.  La  Femme  et  PAmour  au  XHe  Siecle  d'apres  les  poemes  de 
Chretien  de  Troyes.    Paris,  1909;  Le  Roman  Idyllique  au  Moyen  Age.    Paris, 

1913 
Luetjens,   A.     Der  Zwerg  in   der    deutschen  Heldendichtung  des  Mittelalters. 

Breslau,  1911.     Germanistische   Abhandlungen,  38 
Luft,  F.     tlber  die  Verletzbarkeit   der  Ehre  in  der  altfrz.  Chansons  de  Geste. 

Berlin,  1907-08 
Liter aturblatt  fur  germanische  u.  romanische  Philologie.    Leipzig,  1880- 

Magnus,  H.      Der  Aberglauben  in  der  Medicin.      Abhandlungen  zur  Geschichte 

der  Medicin,  VI.     Breslau,  1903 
Mason,  E.    French  Mediaeval  Romances.    Everyman  Library.    N.  Y.,  191 1 
Massing,  E.    Die  Geistlichkeit  im  altfrz.  Volksepos.    Darmstadt,  1904 
Menendez  y  Pelayo,  M.    Origenes  de  la  Novela  Nueva.     Biblioteca  de  Autores 

espanoles.    Madrid,  1907 
Merk,  C.  J.    Anschauungen  iiber  die  Lehre  und  das  Leben  der  Kirche  im  altfrz. 

Heldenepos.    Halle,  19 14 
Meyer,  F.    Jugendziehung  im  mittelalter  —  nach  den  altfrz.  Artus-  und  Aben- 

teuerromanen.     Solingen,  1896 
MLN.    Modern  Language  Notes.    Baltimore,  1886- 
MLR.    Modern  Language  Review.    Cambridge,  England,  1905- 
Morris,  R.    Specimens  of  Early  English.    Part  I,  Oxford,  1887;   Part  II,  Morris 

and  Skeat,  4th  ed.,  Oxford,  1898 
Mod.  Phil.    Modern  Philology.    Chicago,  1903- 
Mott,  L.    The  System  of  Courtly  Love.    London,  1904 

Mueller,  O.     Turnier  und  Kampf  in  den  altfrz.  Artusromanen.     Erfurt,  1907 
Mueller,  R.    Die  Zahl  3  in  Sage,  Dichtung,  u.  Kunst.    Teschen,  1903 

Neumann,  E.    Der  Soldner  (soudoyer)  im  mittelalter  nach  den  frz.  u.  proven- 

zalischen  Heldenepen.    Diss.     Marburg,  1905 
Noack,  G.     Sagenhistorische  Untersuchungen  zu  den  Gesta  Herwardi.     Diss. 

Halle,  1 9 14 
Nutt,  A.    The  Influence  of  Celtic  upon  Mediaeval  Romance.    London,  1904 

Ogle,  M.    Classical  Literary  Tradition  in  Early  German  and  Romance  Literature. 

MLN.    1912,  XXVII,  233-42;   Amer.  Jour,  of  Phil,  1913,  XXIV,  125  ff. 
Oschinsky,  H.    Der  Ritter  unterwegs  u.  die  Prlege  der  Gastfreundschaft  im  alten 

Frankreich.    Halle,  1900 

Palaestra,  Untersuchungen  und  Texte.   Leipzig,  Berlin,  1898- 
Panzer,  F.    Hilde-Gudrun.    Halle,  1901 


TABLE  OF  ABBREVIATIONS  AND  REFERENCES   325 

Paris,  Gaston.    Poemes  et  Legendes  du  Moyen  Age.    Paris,  1900 

Paul,  H.     Grundriss  der  germanischen  philologie,  3  vols.    Strassburg,  1891-1900 

Paul  und  Braune.  Beitrdge  zur  Geschichte  der  deutschen  Sprache  und  Literatur. 
Halle,  1874- 

Percy  Society.    Publications  of  the  Percy  Society.    London,  1840-52 

Petit  de  Julleville,  L.    Les  Mysteres.    Paris,  1880 

PFMS.  The  Percy  Folio  Manuscript,  ed.  Furnivall  and  Hales,  4  vols.  London, 
1867-69 

PMLA.  Publications  of  the  Modern  Language  Association  of  America,  Balti- 
more, 1884-1901,  Cambridge,  1902- 

Potter,  M.    Four  Essays.    The  Horse  as  an  Epic  Character.    London,  191 7 

Prelle  de  la  Nieppe,  E.  Les  Costumes  chevaleresques  et  les  armes  offensives  des 
XII,  XIII,  XIV  siecles.    Annales  de  la  Soc.  Archeol.  de  Bruxelles,  1900 

Puckett,  H.  W.  The  Fay,  Particularly  The  Fairy  Mistress,  In  Middle  High 
German.    Modern  Philology,  1918,  XVI,  297-313 

Quellen  und  Forschungen  zur  Sprach-  und  Culturgeschichte  der  germanischen 
Volker.    Strassburg,  1874- 

Reich,  O.     Beitrage  zur  Kenntnis  des  Bauernlebens  im  alten  Frankreich  auf 

Grund  der  Zeitgenossischen  Literatur.    Diss.     Gottingen,  1909 
Remmpis,  M.     Die  Vorstellungen  von  Deutschland  im  altfrz.  Heldenepos  und 

Romanen  und  ihren  Quellen.     Zeitsch.  fur  roman.  Phil.  191 1,  XXXIV 
Revue  Celt.    Revue  Celtique.    Paris,  1870- 
Rickert,  E.    Romances  of  Friendship,  vol.  I;  Romances  of  Love,  vol.  II.    London, 

1908 
Ritson,  J.    AEMR.    Ancient  English  Metrical  Romances.    3rd  ed.    London,  1802 
Robson,  J.     Three   Early   English  Metrical   Romances.      Camden  Soc.   Lond. 

1842 
Rom.    Romania.    Paris,  1872- 
Rom.  Rev.    Romanic  Review.    New  York,  19 10- 
Rom.  Forsch.    Romanische  Forschungen.     Erlangen,  1883- 

Sallentien,  V.    Handel  u.  Verkehr  in  der  altfrz.  Literatur.    Roman.  Forsch.  19 12, 

XXXI,  1-154 
Savage,  E.    Old  English  Libraries.    London,  191 1 
SATF.    Societe  des  anciens  textes  francais.    Paris,  1875- 
Schepp,  F.    Altfrz.  Sprichworter  und  Sentenzen  aus  den  hofischen  Kunstepen. 

Diss.     Greifswald,  1905 
Schevill,  R.    Ovid  and  the  Renascence  in  Spain.    Univ.  of  California;   Berkeley, 

Publications  in  Modern  Philology,  19 13,  IV,  1-268 
Schmidt,  Fr.     Das  Reiten  und  Fahren  in  der  altfrz.  Litteratur.    Diss.     Gottin- 
gen, 19 14 
Schoepperle,  G.     Tristan  and  Isolt,  A  Study  of  the  Sources  of  the  Romance. 

P'rankfurt  a.  M.,  London,  19 13 
Schofield,  W.    English  Literature  from  the  Norman  Conquest  to  Chaucer.    N.  Y., 

1906;  Chivalry  in  English  Literature.  London,  191 2 
Schroetter,  W.  Ovid  und  die  Troubadours.  Halle,  1908 
Schubert,  C.    Der   Pflegesohn    (Nourri)   im  frz.   Heldenepos.     Diss.     Marburg, 

1906 
Siefkin,  O.    Das  geduldige  Weib.  in  der  englischen  Literatur  bis  auf  Shakspere. 

I,  Der  Konstanzetypus.    Diss.     Leipzig,  1903 


326      TABLE    OF   ABBREVIATIONS   AND    REFERENCES 

Soechtig,  O.     Zur  Technik  altengl.  Spielsmannsepen.     Leipzig,  1903 

Soehring,  O.    Werke  bildender  Kunst  in  altfrz.  Epen.    Erlangen,  1900 

Spence,  L.    A  Dictionary  of  Mediaeval  Romance  and  Romance  Writers.    Lond. 

1913 
Studies  in  Philology,  University  of  North  Carolina,  N.  C,  1906- 

Studien  zur  Vergleichenden  Literaturgcschichte.    Berlin,  1901-09 

Suchier,  H.  und  Birsch-Hirschfeld.   Geschichte  der  frz.  Litteratur  von  den  altcstcn 

Zeiten.     Leipzig.  1900-13 

Thorns,  W.  J.    A  Collection  of  Early  Prose  Romances.    Revised  ed.    London, 

1907 
Tilley,  A.    Medieval  France.    Cambridge,  1922 

Voltmer,  B.  Die  mittelenglische  Terminologie  der  rittcrlichen  Vcrwandtschafts 
u.  Standesverhaltnisse  nach  den  hoiischen  Epen  u.  Romanzen  dcs  13  u.  14 
Jahrhunderts.    Diss.     Kiel,  191 1 

Voretzsch,  C.    Einfiihrung  in  das  Studium  der  altfrz.  Literatur.     Halle,  1905,  13 

Walker,  T.    Die  altfrz.  Dichtungen  vom  Helden  im  Kloster.    Tubingen,  19 10 

Ward,  H.    See  British  Museum  Catalogue 

Weber,  H.  Metrical  Romances  of  the  XIII,  XIV,  and  XV  Centuries.  3  vols. 
Edinburgh,  18 10 

Webster,  K.  G.  The  Twelfth  Century  Tourney.  Kittredge  Anniversary  Papers. 
Boston,  19 1 3 

Wells,  J.  E.  A  Manual  of  Writings  in  Middle  English.  1050-1400.  New  Haven, 
1916;  Supplement,  1919 

Weston,  J.  Chief  Middle  English  Poets.  Boston,  1914;  Three  Days'  Tourna- 
ment.    London,  1902 

Wilmotte,  M.  L'E  volution  du  Roman  francais  aux  Environs  de  n 50.  Acad. 
Royale  de  Belgique,  Bulletin,  1903;  De  l'origine  du  roman  en  France  la 
traduction  antique  et  les  elements  Chretiens  du  roman.     Paris,  1923 

Witter,  E.  Das  burgerliche  Leben  im  mittelenglischen  Versroman.  Diss.  Kiel, 
1912 

Wohlgemuth,  F.  Riesen  und  Zwerge  in  den  altfrz.  erzahlenden  Dichtung.  Diss. 
Tubingen,  1906 

Wood,  M.  M.    The  Spirit  of  Protest  in  Old  French  Literature.    N.  Y.,  191 7 

Wiilker,  R.    Geschichte  der  englischen  Literatur.    Leipzig,  1906-07 

Zadi,  E.    Der  Trobadour  Jaufre  Rudel  und  das  Motiv  der  Fernliebe  in  der  Welt- 

literatur.     Diss.     Greifswald,  1920 
Zeitschrift  fur  deutsches  Altertum  und  deutsche  Litteratur.    Berlin,  1841- 
Zeitschrift  fiir  franzosische  Sprache  und  Litteratur.  Oppeln,  Leipzig,  1879-91, 

1891- 
Zeitschrift  fiir  romanische  Philologie     Halle,  1877- 
Zeitschrift  fiir  vergleichende  Literaturgcschichte.    Berlin,  1887- 


INDEX   OF   MATTERS   AND   LITERATURE 


The  Index  lists  alphabetically  all  the  references  in  this  book  to  classical  and 
mediaeval  authors,  to  motifs  and  themes  in  mediaeval  literature,  to  historical 
English  personages,  and  to  Middle  English  romances.  All  anonymous  foreign 
literature  or  literature  of  doubtful  attribution  is  indexed  under  the  language  in 
which  it  is  written.  For  modern  references  here  cited  by  the  author's  name  only, 
see  the  preceding  Table  of  Abbreviations  and  References. 


Aalof,  101,  102 

Abduction  of  child,  8,  140;  by  werwolf, 
216;  of  heroine  by  suitor,  236;  of 
mortal  queen  by  fairy  lover,  198 

Abingdon,  monastery,  133 

Abode,  supernatural.    Cf.  Fairy  Castle 

Abraham  and  Isaac  legend,  74 

Accusation,  false:  of  murdering  a  child, 
18;  of  dishonoring  a  princess,  83, 
119,  137,  236;  of  treachery  to  king, 
143.     Cf.  Adultery 

Adam  de  la  Hale,  167 

Adder  springs  from  cross,  243 

Ademar  de  Chabannes,  152 

Adenes  le  Roi,  Bert  aus  grans  pies, 
31  n.  14;   Cleomades,  192  n.  20 

Adopted  son.     See  Schubert 

Adultery,  woman  falsely  accused  of, 
17,  242  n.  5,  285;  because  of  bearing 
twins,  295 ;  Germanic  laws  for  punish- 
ment of,  20.  Cf .  Innocent  persecuted 
wife 

iElfric,  6 

JEUsl,  king  of  Deira,  30 

Aimon  de  Varennes,  Florimont,  200  n.  1 

Alberic  de  Trois  Fontaines,  283,  287, 

293 
Alexander,  King,  77,  136;    Alisaunder, 

132  n.  11, 147, 177;  Buik  of  Alexander, 

i35  n.  13 
Alfred,  King,  Boethius,  196 
Alfwynn,  daughter    of    ^Ethelflaed    of 

Mercia,  no 
Alliteration  in  Middle  English  romance, 

85,  215  n.  3,307 
Alms  given  by  wife  to  unrecognized 

husband,  10 
Alphabet  of  Tales,  69;  Latin  version,  13 


Amadas,  73-78 

Amadas  and  Idoine.    Cf.  French  litera- 
ture, Amadas 
Ambiguous  Oath,  ^y  n.  3 
Ambroise's  L'estorie  de  la  guerre  sainte, 

iSo,  153 

Ambush,  119,  310 

Amis  and  Amiloun,  46,  65-72,  191  n.  17 

Amoryus  and  Cleopes,  192  n.19 

Angels,  54,  119,  169,  243 

Angevin  Britain.     See  Leach 

Animals  carry  off  children,  270,  277, 
280.  Cf.  Children;  St.  Eustache. 
Faithful  animals,  180;  dog,  201  n.  3, 
259  n.  18,  286  (cf.  Lion);  Grateful 
animals,  55,  136  n.  17.  Helpful  ani- 
mals, 219.     Magic  animals,  234 

Anlaf  Cuaran,  109,  no;  Anlaf  Guth- 
frithson,  109 

Anna  and  Joachim,  legend  of,  54,  302 

Antoninus,  Archbishop  of  Florence,  60 

Antwerp  legend,  247 

Apocalypse  of  St.  John,  scenes  painted 
from,  308 

Apollonius  of  Tyre,  26  n.  6,  29,  94,  164- 
173,  192,  265,  302 

Apuleius,  171,  190,  205-206  n.  13.  See 
Partonope 

Arab  tales,  80,  189;  version  of  Seven 
Sages,  175,  179 

Arabian  Nights,  8,  17,  190  n.  14  (Thou- 
sand and  One  Nights),  62  n.  8 

Armenian  version  of  Seven  Sages,  178; 
tales,  120 

Arthur  and  Merlin,  147,  177 

Arthur,  King,  allusions  to,  26  n.  6,  92, 
105,  no,  219  n.  9,  241,  265,  306, 
308 


327 


328 


INDEX 


Art.  See  Sohring.  Special  legends 
represented  in  ivories,  misericords, 
illuminations,  tapestries:  Beves,  115 
n.  1;  Guy,  136  n.  17;  Chatelaine  de 
Vergi,  256  n.  12;  Chevalier  au  Cygne, 
242  n.  4;  Constance  Legend,  33  n.  18 
(la  Fille  sans  mains) ;  La  belle  Helene 
de  Constantinople,  27  n.  9;  Macaire, 
287;  St.  Eustache,  6  n.  2.  Cf. 
Chertsey  Tiles;  Wall  Paintings 

Arundel,  town  named  for  horse,  117 

Athelstan,  King,  128,  132,  140,  144 

Athelston,  133,  143-146 

Attack  on  guest  or  messenger,  135 

Authorship  of  romances,  theories  of 
single  and  communal,  1 24 

Automata,  180  n.  12,  192  n.  20 

Avowynge  of  Arthur,  307 

Awntyrs  of  Arthur,  307 

Baldwin  V  and  VI,  Counts  of  Flanders 
and  Hainault,  215 

Ballads,  British:  Aldingar,  18,  37,  243, 
285  n.  5;  Cawline,  276  n.  3,  315  n.  5, 
317;  Chevy  Chase,  288,  309;  Fair 
Annie,  297;  Ritchie  Boy,  88;  Lady 
Diamond,  88,  256;  Orfeo,  195-196; 
Otterburn,  309;  Robin  Hood,  156,  158, 
160;  Geste,  159-160;  R.  H.  Newly 
Revived,  160;  R.  H.  and  Gandelyn, 
156,  160;   Young  Beichan,  89 

Ballad  style  imitated  in  romances,  144, 
158,  160,  195,  243,  293 

Bandello,  37,  40 

Baptism,  119,  201,  279  n.  1.  Cf. 
Rites;  Transformation 

Bareleta,  Gabriel,  13 

Bargain  contract,  77 

Basile's  Penlamerone,  28 

Basket  of  flowers  conceals  youth,  189 

Baumholtz,  Albrecht,  13 

Beard,  red,  292  n.  3.    Cf.  Superstitions 

Beardless  man  unlucky,  292  n.  3 

Beauchamp,  Richard,  Earl  of  Warwick, 

55 
Beaumanoir,  Philippe  de  Remi,  Sire  de. 

Jehan  et  Blonde,  94;   La  Manekine, 

26  n.  8,  27,  31-32 
Beauty,  human,  216.     See  Curry 
Becket,  Thomas,  144 
Bed,  magnificent,  308 
Bee-hives  thrown  into  city,  152 


Beggar    as    messenger,    98;     beggar 

woman  accused  by  noble  lady,  242 
Beleaguered  cities,  ladies  rescued  from, 

221,  225 
Bellefo rest's  Histoires  Tragiques,  167 
Bellerophon  legend,  122-123 
Bells  rung  without  hands,  16  n.  4;   243 
Benoit  de  St.  More,  Roman  de  Troie, 

15,  205  n.  9,  207,  226,  228 
Beowulf,  24  n.  3,  30,  89,  316 
Bernard,  Count  of  Toulouse,  38,  42 
Berte,  mother  of  Charlemagne,  legend 

of,  31  n.  14;    184  n.  3;    206  n.  13; 

292  n.  4 
Betrothals.     Cf.  Bundling;  Rites 
Beuter,  Anton,  35 
Beves  of  Hampton,  7,  10,  29,  47  n.  5; 

69  n.   8;    93,   115-126,   132  n.   n; 

134,  135  n.  15;    137,  141,  153,  160; 

191  n.  16;  222  n.  14;  280  n.  2;  287, 

298 
Biblical  names,  232  n.  2 
Birck,  Johannes,  16 
Bird  gives  warning,  7 
Birth  in  forest,  119.     See  Supernatural 
Birth  of  children.    See  Fellinger 
Black  Bull  of  Norroway,  folk-tale,  234 

n.  4 
Blois,  Counts  of,  250 
Blood  restores  life,  70;    blood-stained 

clothes  as  proof  of  death,  119.     Cf. 

Brotherhood 
Blows  Shared,  folk-tale,  80  n.  3 
Boat  drawn  by  swan,  243,  248 
Boccaccio,    Filocolo,    186,    188    n.    9; 

Decameron  (IV,  9),  253;  (IV,  1),  256- 

257,260;  (IV,  5),  266;  (X,8),69n.  7 
Boethius,  196 
Bohemian    versions   of   romances:    of 

Robert  of  Sicily,  58;    of  Seven  Sages, 

178 
Bower  of  Bliss,  304 
Braban  Silvius,  legend  of,  247 
Brabant  legends,  250,  251 
Bremis,  John,  102  n.  8 
Breton  ballads,  4,  8,  9  n.  5;   lays,  36, 

41,  51,  53,  105,  197  n.  3,  198-199, 

259,  294,  301,  316 
Bride  abandoned  on  wedding  day,  129, 

137;  substituted  bride,  292  n.  3,  (Ar- 

fert);    bride- winning,  89 
Bromyard,  John,  80  n.  3 


INDEX 


329 


Brothel,  girl  sold  for,  167,  170 
Brotherhood,  blood-,  sworn-,  145  n.  3, 

313,31511.  5;  317,  318  n.  11 
Brunanburgh,  Battle  of,  108  n.  4,   109, 

132  n.  11 
Brunhilde  (Brynhilde),  69  n.  8,  318 
Brutus  legend,  123 
Buddhistic  legends,  6 
Buffets  exchanged,  152 
Biihel,  Hans  von,  27,  31 
Bulgarian  version  of  Solomon  legend, 

62 
Bundling,  betrothal  custom,  307  n.  3 
Butcher  of  Paris,  270 
Byzantine  romance,  168,  189,  229 

Caesar,  Julius,  in  legend,  247 

Calais,  burghers  of  Calais  and  Ed- 
ward III,  135 

Cannibalism  of  Richard  I,  1 51-15  2 

Canute  in  England,  108-109 

Capricious  lady,  226,  228 

Captive,  heroine  held,  188  n.  11 

Carbonell,  Miguel,  35,  38  n.  5 

Carolingian  legends,  16,  31,  56,  68  n.  6, 
184  n.  3.    Cf.  Charlemagne 

Castel  Pilgrim,  152 

Catalan  tales  and  romances,  35.  Cf. 
Spanish  Literature 

Catalogue  lists  in  romances,  263,  290 
n.  1 

Catalogues  of  mediaeval  libraries,  129, 
255  n.  9,  263  n.  1,  284.    See  Savage 

Catskin,  29 

Caxton,  103,  166 

Celtic  fee,  208  n.  17,  220;  Celtic  influ- 
ence, 42,  53,  197,  207,  298,  302;  see 
Lalande  de  Calan;  Nutt.  Celtic 
origins:  of  Guirun,  259;  of  Swan 
Knight,  248.  Celtic  Otherworld,  193, 
197  n.  4, 198,  316.  Celtic  solar  myth, 
315.  Celtic  tales  and  romances, 
Children  of  Lir,  248;  Morraha,  218- 
219;  Serglige  Conculaind,  248;  Toch- 
marc  Etain,  197,  248.  Cf.  Breton 
lays;  Chretien;   Mabinogion 

C6sar  de  Nostredame,  35 

Chains,  talismanic,  241 

Champion,  boy  or  dwarf  acts  as,  37, 
242-243.    Cf.  Ballads,  Aldingar 

Chansons  de  Geste,  16  n.  3,  65,  84,  184- 
185,  242,  267,  283,  286 


Character  types,  71  n.  10.  See  Creek; 
Comfort 

Charlemagne,  68,  283,  287,  308 

Chaste  Duchess,  Queen,  Wife,  38  n.  6, 
39,  42,  285 

Chaste  wife  wooed  by  brother-in-law, 
12 

Chastity  miraculously  preserved,  8,  15, 
119.     Cf.  Well 

Chdtelain  de  Couci.  Cf.  Knight  of 
Courtesy 

Chaucer's  Adam,  279;  Hous  of  Fame, 
205  n.  9;  Pyramus,  192  n.  19; 
Canterbury  Tales,  156:  Clerk's  Tale, 
298  n.  9,  299;  Franklin's  Tale,  205 
n.  9;  Man  of  Law's  Tale,  24;  Monk's 
Tale,  63;  Nun's  Priest's  Tale,  63; 
Sir  T  ho  pas,  97, 127,  263.  Chaucerian 
influence,  203 

Cheating  of  minors,  158 

Chertsey  tiles,  151,  154 

Chess  game,  135  n.  15,  137,  198 

Chestre,  Thomas,  79  n.  1,  269,  304 

Chettle,  Henry,  Sir  Placidas,  7 

Chevalere  Assigne,  239-252 

Childbirth.     See  Fellinger 

Child  in  mediaeval  literature.  See 
Arnold;  Fellinger;  Schubert.  Cf. 
Accusation;  Champion;  Child  vowed 
to  the  Devil,  52-53,  77  n.  6 

Childlessness,  vows  to  prevent,  52 

Children  linked  by  name  and  circum- 
stances of  birth,  191;  in  love  with 
each  other,  184;  seven  at  a  birth, 
240,  242;  stolen  by  animals,  8,  27 
n.  9,  270,  277  n.  4;  turned  into 
swans,  180,  240  n.  2 

Chivalric  life,  212,  216.    See  Langlois 

Chivalry.    See  Schofield;  Voltmer 

Choice  of  Woe,  7 

Chretien  de  Troyes,  80;  Charette,  197, 
228;  Cliges,  214  n.  2,  227;  Erec,  234, 
304  n.  10;  Iwain,  136  n.  17,  181, 
207,  211,  222,  228,  314-315.  See 
Guyer;  Lot-Borodine 

Chronicles,  100 

Church,  The.  Cf.  Clergy;  Pietistic 
elements  in  romance;  Religion;  Rites. 
See  Massing;    Merk 

Cicero,  74 

Cinderella,  Male,  54,  291 

Circe.    Cf.  Enchantress 


330 


INDEX 


Classical  allusions,  200  n.  1;    mediae- 

valized     classical     romances,     207; 

heroes,  136.     Cf.  Greek,  Latin 
Classical   Influence   and   Tradition   in 

Mediaeval   Literature.     Cf.  Benoit; 

French   literature:    Roman   d 'Eneas, 

Roman  de  Thebes.     See  Ogle 
Cleges,  79-80 

Clergy  in  romance.    See  Kahle 
Cleves,  legends  of  House  of,  246,  251 

n.  11 
Cloister,  Hero  in.    See  Walker 
Colbrand,  Danish  giant,  127,  132,  136; 

Canticam  Colbrandi,  131 
Colliers  and  quarry,  216 
Combat,  island,  280  n.  2;  judicial,  132 

n.  ii,  35,  37,  7i,  "4,  184  n.  2,  244; 

single,  108  n.  4,  236,  277  n.  6;    of 

father  and  son,  141,  227  n.  5,  277, 

281,  288;   of  man  and  dog,  287;  of 

man  and  woman,  168;  "of  relatives, 

227,  303.  See  Leibecke 
Comic  elements  in  romance,  119,  270. 

See  Theodor 
Companions,  helpful,  291 
Complaint  of  Scotland,  235  n.  4,  312 
Confession   to   inanimate   object,    292 

n.  4;    confession  to  priest  heard  by 

lover  or  husband,  41  n.  12 
Confidante,  85,  188  n.  11,  221,  228,  237 
Constance  legend,  23,  276 
Constantinople,  233.    Cf.  French  Liter- 
ature, La  belle  Helene 
Copland,  Robert,  translator,  166,  239; 

William,  printer,  4,  127  n.   1,   239, 

263,  284 
Cornwall,  91-92 
Corpse  held  for  debt,  73,  76;  mutilated 

and  substituted  for  that   of   living 

man,  17,  265.    Cf.  Embalming 
Costume,   interest   in,    307-308.      See 

Prelle  de  la  Nieppe 
Counsellors,  false,  174 
Counsels,  Three,  for  travel,  292  n.  3 
Court  of  law,  proceedings,  162 
Courtly  love,  202,  209,  228,  234,  298. 

Cf.    Ovid    here.      See    Heyl;     Lot- 

Borodine;  Mott 
Cowherd  rears  a  prince,  216 
Crescentia  legend,  12,  19-20,  23  n.  2, 

167  n.  4.    Cf.  Florence  of  Rome 
Croyland  Abbey,  51 


Crusaders'  tales,  40,  46,  179,  193 
Crusades,   cycle   of   romances   dealing 

with,  244;  Crusading  spirit,  137 
Cupid  and   Psyche  legend,   171,   190, 

205-206,    209-210;     folk-tales,    206 

n.  12.     Cf.  Apuleius 
Cup,  marvellously  decorated,  187 
Cursor  Mundi,  4 
Cut-off  finger,  313;    hand.     Cf.   Girl 

without  hands 

Danes  in  England,  89,  108 

Danish  ballads:  Ravengaard,  18;  Sak- 
arias,  4,  7,  8,  9;  legends,  Hrolfssaga, 
107  n.  3,  123;  versions  of  Apollonius, 
165,  168;  Der  Seelen  Trost,  13;  Floris, 
189;  Octavian,  267;  Partonope,  204, 
208  n.  16;  Seven  Sages,  178 

David,  151;  and  Uriah,  122 

Death,  fabricated  evidence  of,  31  n.  14 

Defence  of  a  child,  219 

Degare,  277,  288,  301-305,  309,  313 

Degrevant,  302,  306-311 

Demon  birth  of  Richard  I,  150 

Demon  must  disappear  when  asked 
name,  249  n.  9 

Denis  Pyramus,  200 

Desclot,  Bernard,  35,  41 

Devices  on  shield,  281.  Cf.  Heraldic 
devices 

Diabolic  human  being,  54 

Didier  of  Lombardy,  68 

Diederic  von  Assenede,  186 

Diez  de  Games,  Guitierre,  27 

Disenchantment  of  werwolf,  218 

Disguise  motif,  93,  237;  disguise  as 
pilgrim,  93  n.  9,  152;  as  minstrel, 
93,  119,  198;  maiden  as  minstrel, 
118;  disguise  as  beggar,  88,  239;  as 
palmer,  119,  135  n.  15,  151;  as 
merchant,  190;  as  animals,  214,  221; 
as  fool,  227-228 

Disinheiritance,  156,  161 

Dog  as  gift,  236;  hero  transformed  into 
a  dog,  235  n.  4.    Cf.  Animals 

Dol,  Archbishopric  of,  294,  296 

Dolopathos,  240,  248.  See  Latin  versions 
of  Seven  Sages 

Doon  de  Mayence,  117,  245 

Dragons  and  dragon  fights,  118,  124, 
128,  136,  257,  303 

Drama.    See  Chambers 


INDEX 


331 


Dramatic  versions  in  England  of  Middle 
English  romances:  Robert  of  Sicily, 
61;  Eglamour,  275.    See  Baskerville 

Dreams,  prophetic,  15,  84,  136,  153; 
of  attacking  animals,  39  n.  8,  47,  71, 
112  n.  8;  of  arms  stretched  out  over 
many  lands,  112,  222.    See  Baake 

Drug  causes  forgetfulness  of  love,  210; 
drugging  of  guard,  119 

Dubbing  of  hero,  92 

Dugdale,  Antiquities  131;  Baronage, 
127;  Monasticon,  93  n.  9 

Duplication  of  climax,  90 

Dutch  literature:  versions  of  Beves, 
118;  Bandello's  Duchessa  de  Savoia, 
37;  Chatelaine  de  Couci,  255;  Cornicke 
van  Brabant,  247;  Gesta  Romanorum, 
165;  Macaire,  284;  Miracle  deVierge, 
13;  Partonope,  202  n.  4;  Robert  of 
Sicily,  58;  Robert  le  Diable,  50;  Der 
Seelen  Trost,  13;  Seven  Sages,  176, 
178;  Vincent  of  Beauvais'  Spec. 
Hist.,  13;  Walwein,  75,  218  n.  7 

Dwarf,  227,  228,  304;  as  pretended 
lover,  285.  See  Lutjens;  Wohlge- 
muth 

Eadburga,  daughter  of  Offa  II,  30  n.  13 
Eastern    elements    and    influences    in 

western  romance:    Beves,  123;    Guy, 

136;    Reinbrun,  140;    Richard,  150; 

Floris,  189-190;  Octavian,  272;  Gen- 

erides,  232 
Eaten  Heart,  Legend  of,  253  n.  1 
Education   of   hero,   92,   97,    174;    of 

heroine,  133,  140,  166  n.  4,  304.    Cf. 

Seven  Arts.    See  Jacobius;  Meyer 
Edward  the  Confessor,  133;    Edward 

HI,  135 

Eger  and  Grime,  266,  276  n.  3,  293,  309- 

310;  312-319 
Eglamour,  8,  231  n.  1,  274-278,  280, 

281,  284,  303 
Eldridge    King,    317  n.    10;    Knight, 

276  n.  3.    Cf.  Sword 
Elias,  legend  of  prophet,  248 
Elizabethan  chroniclers,  131;   vogue  of 

mediaeval   romance   in   Elizabethan 

times,  116,  127.     See  Crane 
Elopement  of  lovers,  119,  221,  237 
Ely  ghost  legend,  317 
Emaret  23-34,  187,  276,  292  n.  4,  302 


Embalming  of  dead  body,  264,  265-266 
Embrace,  fatal,  of  poison  maiden,  77 
Embroidery  representing  famous  love 

stories,  28  n.  10,  187 
Emir's  Tower,  190,  193 
Emma,  Queen,  legend  of,  37  n.  3,  144 
Enchantress,  218,  220. 
Enemy  remounted  by  foe,  135 
Enfance  motif,  101  n.  7 
Eninkel,  Jansen,  27 
Epic  elements  in  romance,  84 
Erie  of  Tolous,  35-44, 184  n.  2,  271  n.  13, 

285  n.  5,  309-310 
Ernst  of  Swabia,  1 20 
Esther,  Book  of,  190;  Queen  Esther,  46 
Etienne  de  Bourbon,  13,  50 
Evolution  of  romance.    See  Wilmotte 
Executioners,  compassionate,  166  n.  4, 

241 
Exempla,  13,  32,  50,  130 
Exile  and  return  of  hero,  theme,  89, 

in  n.  6;    exile  of  lover,  119,  205; 

of  queen,  285 
Exposure  of  child  or  woman,  in  forest, 

24,  269,  302;   on  sea,  23,  26,  30,  92, 

276,  280;  in  tree,  297 

Fabliau,  41  n.  12,  80  n.  3,  180 

Faiel  (Fayet),  chateau,  261 

Fairy  animal,  6  n.  1,  207,  208;  arts,  205 
n.  9;  boat,  208  n.  16;  castle,  (cf. 
magic  house),  141,  195,  206  n.  11, 
211  (He  d'Or),  304;  gifts,  209  n.  19; 
cf.  Gloves;  fairy  knight,  141;  fairy 
land,  197  n.  4;  fairy  lover,  302,  206 
n.  12;  fairy  messenger,  208  n.  16; 
fairy  mistress  types:  Celtic  fee,  181, 
209;  devil,  205  n.  10;  unable  to  be- 
hold Sacrament,  148;  see  Puckett; 
Lamia  type,  151;  swan-maiden,  240; 
Valkyrie  type,  314;  Offended  F6e, 
210,  211;  Proud  Fee,  209,  304,  315; 
fairy  music,  304  n.  8,  305;  fairy  tale 
elements  in  romance,  243,  248.  Cf. 
Magic 

"Faithful  John"  folk-tale,  70;  faithful 
squire,  237  n.  7 

Family  reunited  after  life-long  sepa- 
ration, 271 

Father,  cruel,  276,  280.  Cf.  Combat; 
Incest 

Fauchet,  Claude,  256  n.  10,  261  n.  19 


332 


INDEX 


Fazio,  Bartolomeo,  28,  32 

Feast  broken  by  tumult,  119,  121 

Feudalism.  Cf.  Combat;  Brotherhood. 
See  Fundenberg 

Fidelity,  test  of,  265 

Fiorentino,  77  Pecorone,  28,  32 

Firdausi,  King's  Book,  123 

Fire,  death  by.    See  Foerster 

Fish  swallows  ring,  32  n.  17 

Fisherman  rears  prince,  106 

Fitzwarin,  Fulke,  156  n.  1 

Flame,  supernatural,  indicates  true 
prince,  in 

Fleck,  Konrad,  185 

Florence  of  Rome,  12-22,  265 

Flores  Historiarum,  46,  47 

Floris  and  Blauncheflur,  28  n.  10,  134, 
184-194,  206  n.  13,  229 

Fold  plaited  in  tunic,  234.  Cf.  Recog- 
nition Tokens 

Forbidden  Land,  314 

Fordun's  Chronica,  100 

Foreign  king,  stay  at  court  of,  135  n.  15, 
I07>  233;  training  of  youth  at  court 
of,  92,  121 

Forest,  birth  of  child  in,  287;  hunt  in, 
119;  life  of  lovers  in,  221  n.  13 

Foresters,  cowardly,  288,  306 

Forest-reared  youth,  244,  303 

Ford  defended,  316,  317,  318  n.  n 

Forgery.    Cf.  Letters  substituted 

Forth-putting  woman,  71, 122,  208  n.  18 

Frame  work,  stories  in  a,  174-175  n.  3, 
178  n.  9 

Frederick  II,  Emperor,  58 

French  literature:  Aimeri  de  Narbonne, 
153;  Aiol,  267;  Alexandre,  192  n.  20; 
Aliscans,  n  1;  Alixandre,  Roy  de 
Hongrie,  27;  Amadas  et  Idoine,  28 
n.  10,  73  n.  1;  Amis  et  Amilun,  65- 
66;  Apoloines,  166  n.  4,  167;  Au- 
cassin  et  Nicolette,  84,  118,  191,  206 
n.  13,  216,  221  n.  13;  Athis,  69; 
Aye  &  Avignon,  15  n.  2,  167;  Bau- 
doum  de  Sebourc,  41  n.  12;  Beatrix, 
242;  La  belle  Helene  de  Constantinople, 
7,  8  n.  3,  27  n.  9,  31;  Boeve  de 
Haumlone,  15  n.  2,  115-116,  117; 
Brut  (en  prose),  104;  Chanson 
dyAntioche,  152,  244,  245;  Charles  le 
Chauve,  15  n.  2;  Chatelaine  de  Vergi, 
256;    Les  Chetifs,  244;    Chevalier  au 


Cygne,  206  n.  13,  241-242,  244-245; 
Comte   Claros,   41    n.    12;     Contesse 
d'Anjou,  27,  32;    Croniques  de  Nor- 
mandie,  50;    Desire,  209;    Li  Dis  de 
VEmpereur  Constant,  29,  122;  Ditdes 
III    Chanoines,    56;    Bit    des   III 
Potnmes,  66;  Dit  du  Buffet,  80;  Doon 
Alemanz,  285  n.  5;   Elie  de  St.  Gille, 
267;  versions  of  St.  Eustache,  6  n.  2; 
Florence   de    Rome,    14;     Florent   et 
Octavian,  118,  267;  Floovent,  123, 135, 
270  n.  n;   Floire  et  Blanchcflor,  184, 
185,  187-189;    Galeran  de  Bretagne, 
124,  294,  297;    Gui  de  Warwic,  128 
n.    2,    129,    136,    141;     Guillaume 
d'Angleterre,    3,    7,    8,    9,    10,    123; 
Guillaume  de  Palerne,  214,  215,  218- 
220;    Guingamor,  207-209;    Histoire 
de  Palanus,  36,  40;  Vhomme  qui  fist 
sa  femme  con} esse,  41  n.   12;    Huon 
de  Bordeaux,  192  n.  20;  Ignaure,  254; 
Ipomedon,  224-229;    Isomberte,  242; 
Jehan  de  Paris,   94;    Jourdains  de 
Blaivies,  65  n.  1,  67,  121,  167,  168; 
Lais  (see  Marie  de  France);   Lai  du 
Corn,  37  n.  3;    Lai  de  VEspine,  198, 
316;    Lai  de  Guirun,  254,  257,  259; 
Lai  d'Orphey,  192  n.  20,  196;    Lai 
d'Haveloc,  103-105,  in;   Lancelot  en 
prose,  226,  308  n.  4;    Linaure,  254, 
257;   Lion  de  Bourges,  75,  112;   Ma- 
caire,   283-284,   285  n.   5;    Marques 
de  Rome,  178  n.  9;   Merlin,  52  n.  3; 
Miracle  de  la  Marquise  de  la  Gaudine, 
36,   38,  40,  41;    Miracle  de  Nostre 
Dame  d'Amis  et  A  mile,  65;    Miracle 
de  la  Vierge,  12,  13;   Miracle  du  Roi 
Thierry,    268;    Moniage   Guillaume, 
138;  Ogier  le  Danois,  8,  68  n.  5,  113, 
135  n.  15,  138,  241;    Octavian,  267- 
269;    Olivier  de  Castille,  67,  69  n.  7, 
76;  Orson  de  Beauvais,  15  n.  2;  P arise 
et  Vienne,  118;  Partenopeus  de  Blois, 
200,  202,  203,  208  n.  16;   Pelerinage 
de  Charlemagne,  192  n.   20;    Perles- 
vaux,  258  n.  14;   Pierre  de  Provence, 
9;    Poeme  Morale,  176;    Ponlhus  et 
Sidoyne,  87,  88;    Prothesilaus,   224; 
Prevot  d'Aquilee,   69  n.   8;    Renard 
Contrefait,  218  n.  7:  Richars  li  Biaus, 
75,    112;    French   version    (lost)    of 
Richard  Coeur  de  Lion,  147;    Robert 


INDEX 


333 


le  Diable,  49,  50;  Robert  de  Sesile, 
58-59;  Roman  d'Eneas,  192  n.  20, 
21,  207,  228,  229  n.  6;  Roman  de  la 
Poire,  69  n.  8;  Roman  de  Thebes, 
207,  226,  228;  La  Royalle  Couronne 
des  Roys  d' Aries,  35;  Les  Sept  Sages, 
66,  176,  177  n.  7,  178;  Tirant  lo 
Blanch,  129,  130;  Tristan  et  Iseult 
(see  Thomas),  85,  87  n.  5,  90  n.  7, 
124,  141,  166  n.  4,  198,  221  n.  13, 
227,  236,  292  n.  4;  Tydorel,  53,  54; 
Vies  des  Peres,  13;  Violier  des  His- 
toires  Romaines,  165 ;  Voeux  du  Paon, 
135  n.  13;  Waldef,  87  n.  5,  101,  102 

Friend,  false,  92;  true  friend  serves  as 
squire,  governor,  tutor,  237  n.  7 

Friendship  theme  in  romance,  69,  134; 
story  of  Two  Friends,  69 

Froissart,  Jean,  255 

Froudmont  of  Tegernsee,  196 

Fulgentius,  African  compiler,  206  n.  13 

Gaimar,  Estorie  des  Engles,  91,  93,  103- 
104,  108-109 

Gallows,  man  freed  from,  becomes  an 
enemy,  19  n.  12 

Gamelyn,  144,  156-163 

Game-skinning,  a  special  art,  227 

Garden,  place  for  secret  interviews  of 
lovers,  266,  310;  garden  of  Emir,  187 

Gautier  d'Arras,  I  lie  et  Galeron,  124, 
298;  Gautier  de  Coincy,  13,  21 

Gawain,  101  n.  7,  228,  306,  316 

Geis,  Irish  form  of  prohibition,  209  n.  20 

Generides,  153,  231-238 

Generosity,  regal,  208  n.  19 

Gerard  of  Cornwall,  1 30-1 31 

German  literature:  version  of  Apol- 
lonius,  168;  Buch  der  Liebe,  87-88, 
267;  Diu  Kiinigun  von  Frankreich, 
284;  versions  of  Floris,  185-186, 
Graf  Rudolph,  118,  121;  Graf  von 
Savoien,  3,  7,  8,  9;  Die  Gute  Frau, 
3,  7,  8,  9,  10,  123;  Herzog  Ernst,  121; 
Kaiser  chronik,  12,  18-20,  25  n.  5; 
Karl  Mainet,  121;  version  of  Lion  de 
Bourges,  75;  Mai  und  Beaflor,  27, 
32;  Nibelungenlied,  118,  318;  version 
of  Olivier  de  Castile,  67;  Orendel,  167; 
of  Partonope,  202;  of  Ponthus,  87; 
Peter  von  Staufenberg,  205  n.  10; 
Ritter  Galmien,  36;   Rittertriuwe,  75; 


version  of  Robert  le  Diable,  49,  50; 
Sachsische  Weltchronik,  12;  version 
of  St.  Eustache,  6  n.  2;  of  St.  Hubert, 
4;  Der  Schwanritter,  246;  Der  Seelen 
Trost,  13,  66;  versions  of  Seven 
Sages,  178;  Siegfriedlied,  118;  Von 
dem  Br  ember  gers  end  und  Tod,  257; 
Wilhelm  von  Wenden,  3,  7,  8,  9 

Germany  in  romance  literature.  See 
Remppis 

Gervase  of  Tilbury,  317 

Gesta  Romanorum,  3,  9,  15,  16,  21,  59, 
61,  69  n.  7,  80,  94,  130,  134,  I36, 
164,  165,  168,  177  n.  8,  219,  260,  284 

Ghost  knight,  74;  warrior,  317 

Giant  baptized,  119;  acts  as  champion, 
136,  270,  277;  as  herdsman,  315. 
See  Wohlgemuth 

Gifts  offered,  to  hero,  98,  99  n.  5.  Cf. 
Fairy  Gifts;  Minstrels 

Giraldus  Cambrensis,  151,  286 

Girl  without  hands,  32-3 

Glass  mountain,  235  n.  4 

Gloves,  fairy,  303 

Godfrey  of  Bouillon,  244,  245,  249,  308; 
Godfrey  of  Viterbo,  68  n.  5,  164,  166, 
168 

Gog  and  Magog,  167 

Golden  hair  (Goldenermarchen),  53,  55, 
in,  112,  292 

Goose  girl,  princess  serves  as,  292 

Goufier  de  Lastours,  40 

Governor  or  tutor  of  prince,  237.  Cf. 
Squire 

Gower,  John,  24,  63,  165,  168,  192,  202; 
Confessio  Amantis,  Portuguese  ver- 
sion, 168;  B diodes,  231 

Gowther  {Robert  the  Devil),  49-57,  226 

Grahams,  clan  of,  319 

Grateful  Dead,  73,  75 

Grave  guarded  by  dog,  286.  Cf.  Ani- 
mals, faithful 

Great  Fool,  folk-tale  of,  243 

Greed  Requited,  80 

Greek,  knowledge  of  in  Middle  Ages, 
248;  Legends  and  Myths:  Atreus, 
258;  Bellerophon,  122-123;  Cupid 
and  Psyche,  206;  Helios,  248;  Or- 
pheus and  Eurydice,  196,  197; 
Progne,  258;  Tantalus,  258;  Ro- 
mances: AntheiaandHabrocomes,  169, 
171;    Chaereas   and   Callirrhoe,   170; 


334 


INDEX 


Versions  of  Legends  and  Romances: 
Apollonius,  169;  Floris  and  Blauncke- 
flur,  188  n.  10;  Seven  Sages,  174, 
179;     St.  Eustache  (Placidas),  5,  6 

Gregory  of  Tours,  16  n.  4,  20 

Grendel,  317 

Grey,  Sir  Thomas,  104 

Grimsby,  local  legends  of,  108 

Griselda  theme,  298,  299 

Guardians,  false,  103,  171  n.  8 

Guelphs,  legend  of  origin  of  family 
name,  296  n.  3 

Gui  de  Bazoches,  249 

Gui  de  la  Buigne,  287  n.  7 

Guillem  de  Cabestaing,  253  n.  2,  257, 
259,  260 

Guisbert,  King  of  France,  legend  of, 
56,  63 

Guiscard,  Robert,  52 

Guinevere,  abduction  of,  198  n.  7; 
Gaynor,  231  n.  1,  306,  307 

Gundeberg,  legend  of,  37,  42 

Gunhild,  legend  of,  37,  42 

Guthrum,  King  of  East  Anglia,  no; 
Guthrum's  Dream,  136  n.  16 

Guy  of  Warwick,  10,  93,  104,  108  n.  4, 
116,  127-139,  140,  141,  144,  265, 
271  n.  13,  280  n.  2,  298,  309 

Hair  given  as  love  token,  256  n.  10,  258 

Hamlet  legend,  112,  122-123 

Hand  exhibited  as  trophy  of  victory, 

317,  318;    cut  from  travellers,  247; 

cut  off  and  swallowed  by  a  fish  and 

miraculously  preserved,  32  n.  17 
Hardyng,  John,  131,  306  n.  1 
Harems  in  romance,  189 
Hartmann  van  Aue,  70  n.  9,  71 
Havelok  the  Dane,  103-114,   144,  157, 

158,  222,  228 
Head  kept  in  pot  of  basil,  266;  severed 

head  shown  as  proof  of  death,  253 

n.  2,  254,  257 
Healing  powers,   314.     Cf.   Medicine; 

Physician.    See  Kiihn;  Lane 
Heart,  Legend  of  Eaten,  Heart  eaten 

by  lover,   253,   255;    sent  to  lover, 

255;    symbolic  virtue  of  heart,  259 

n.  17,  260 
Hebrew-German  romances.  See  Landau 
Hebrew  version  of  Seven  Sages,   174, 

175,  179 


Heinrich  von  dem  Tiirlin,  Diu  Crdne, 
192  n.  20,  316 

Heinrich  von  Neustadt,  167,  171 
n.  9 

Heir  missing,  140;  true  heir  rec- 
ognized, 112;  Heir  of Linne,  160 

Hennen  von  Merchtenen,  247 

Henry  III,  151,  161  n.  5 

Henryson,  Robert,  196  n.  2 

Heraldic  devices,  250,  258  n.  14,  276 

Herbert,  translator  of  Dolopathos,   241 

Here  ward  the  Wake,  legend  of,  92,  93 
See  Noack 

Hermit  as  protector,  242,  243,  302; 
as  miracle  worker,  24,  32 

Herold,  J.,  13 

Herrand  von  Wildonie,  59 

Higden,  Ranulf,  144 

Hildegard,  legend  of,  16,  17 

Hind  Horn,  88 

Hind  pursued,  6  n.  1,  234 

Hoccleve,  16 

Holkot,  Robert,  74 

Home-returning  hero,  92 

Honor.    See  Luft 

Horn-blowing  proves  true  heir,  112 

Horn  Childe,  87,  89,  97-102 

Horn,  King,  83-96,  121,  134,  137,  152 
n.  5,  237,  265,  298 

Horse,  magic,  153,  154,  271;  horse 
race,  119;  horse  recognizes  master, 
119,  123,  222  n.  14;  horse  stolen  by 
prince,  121;  horses  of  strange  color, 
55,  226;  killed  shamefully  in  single 
combat,  288.    See  Potter;  Schmidt 

Hospitality  given  to  murderer  of  host's 
son,  135.    See  Oschinsky 

Hue  de  Rotelande,  84,  211,  224 

Humbert  de  Romans,  13 

Humorous  elements  in  romance,  216, 
244,  274,  287,  295 

Humphrey  de  Bohun,  215 

Hungarian  versions  of  romances:  Apol- 
lonius, 165;  Robert  of  Sicily,  58 

Hungary  in  romance,  21,  264,  287 

Hunt,  Leigh,  A  Jar  of  Honey,  61 

Hunts  and  raids,  135,  232,  308,  309; 
hunting  preferred  to  tourney,  226,  293 

Husband  forces  wife  to  eat  lover's 
heart,  253,  254  n.  4;  Legend  of 
Husband  with  Two  Wives,  90  n.  7, 
124,  298 


INDEX 


335 


Icelandic  literature.  Cf.  Norse.  See 
Leach 

Idols  overthrown  by  Christian  hero, 
119,  124,  153;   beaten,  47,  135  n.  5 

Idyllic  romances,  184,  187,  222.  See 
Lot-Borodine 

Illuminations,  MSS.  of  Beves,  115  n.  1; 
of  Guy,  131  n.  10,  135  n.  14;  of 
Richard,  151 

Incest  theme,  169  n.  5.    See  (Edipus 

Incestuous  Father,  23,  24,  25  n.  6,  29, 
32,  170,  242  n.  3,  302 

Incognito,  hero,  293 

Indian  literature:  Mahabharata,  232; 
tales,  254,  259 

Inheritance  regained  from  foes,  160 

Innocent  persecuted  wife,  29,  35.  Cf. 
Emare,  Florence  of  Rome,  Erie  of 
Tolous 

Introduction,  The,  in  romance.  Cf. 
Minstrels 

Invisibility,  magic  art,  205 

Ipomedon,  211,  224-230 

Irish  tales  and  romances:  Children  of 
Lir,  248;  Diarmid  and  Grainne,  198, 
316  n.  9;  Imram  Hui  Corra,  53; 
Imram  Maelduin,  304  n.  8;  orraha, 
218,  219;  Serglige  Conculaind,  248; 
Tochmarc  Etain,  197,  248;  versions 
of  legends  and  romances:  Beves,  117; 
Guy  of  Warwick,  129;  St.  Eustache, 
6  n.  2 

Iron  man,  291 

Isumbras,  3-11,  123,  269,  277 

Italian  literature:  Apollonio,  168; 
Buovo  d'Antona,  118,  123;  Cantare 
de  lo  Bel  Gherardino,  202  n.  6; 
Cento  Novelle  Antiche,  254;  versions 
of  Constance-Emare,  28;  Crescentia, 
18  n.  n;  Fiorio  e  Biancifiore,  188; 
Historia  de  la  Regina  Oliva,  28; 
Istoria  di  Santa  Guglielma,  14; 
Libro  di  Fioravante,  268,  272;  Novella 
delta  Figlia  del  Re  di  Dacia,  28; 
Novella  di  Messer  Dianese,  75;  ver- 
sion of  Olivier  de  Castile,  67;  Orlando 
Furioso,  211  n.  25;  Rappresentazione 
di  Santa  Uliva,  28;  Reali  di  Francia, 
118,  120,  268;  version  of  Robert  of 
Sicily,  58;  of  St.  Eustache,  6  n.  2; 
of  Seven  Sages,  176,  177  n.  7;  of 
Tirant  lo  Blanch,  130;   of  Ogier,  277 


n.  5;   of  Vincent  of  Beauvais'  Spec. 
Hist.,  13 
Ivories,  S3)  242  n-  4 

Jack  the  Giant  Killer,  76 

Jacob  van  Maerlant,  66,  247,  250 

Jacques  de  la  Hague,  50 

Jacques  de  Vitry,  Exempla,  32, 126 

Jaloux  type,  234 

Jean  d'Arras,  Melusine,  151,  207  n.  14 

Jean  de  Conde,  Li  Dis  dou  Magnificat, 

59 
Jean  de  Garlande,  Stella  Maris,  13 
Jean  de  Nesles,  257  n.  14 
Jensen,  Jeppe,  Den  Kydske  Dronning, 

36 
Jewel  gives  light,  193,  204;    jewels  in 

romance,  141,  187 
Jewish  version  of  Beves,  118 
Johannes  de  Alta  Silva,  Dolo pathos,  176, 

180,  240,  241,  248 
Johannes  Junior,  13,  77  n.  8,  177  n.  7 
John  of  Cella,  23;  John  of  Damascus,  5 
Johnson,  Richard,  Seven  Champions  of 

Christendom,  124 
Jonah  legend,  137 

Jongleurs,  tales  of,  68,  241.    See  Faral 
Juan  de  Timoneda,  Patranuela,  168 
Judas  Iscariot  theme,  169  n.  5;  Judas's 

red  hair,  292  n.  3 
Judith,  Book  of,  60;    Empress  Judith, 

legend  of,  38,  39,  42 
Jury  packed  and  bribed,  162 

Kalidasa,  Indian  poet,  232 

Keats,  308 

King  Deposed,  59,  62,  63;    incognito, 

152;  Konig  im  Bade,  59 
King  of  Tars,  15,  45-48 
Kingsmark,  sign  of  royal  birth,  112 
King's  son  injured  or  killed  by  hero, 

119,  121,  135  n.  15 
Kiss  of  disenchantment,  142 
Knife  thrown  at  messenger,  119,  135 

n.  15,  153 
Knight  of  Courtesy  (Chdlelain  de  Couci), 

253-262,  266 
Knight  of  the  Lion  tales,  271  n.  13; 

Chevalier  au  Lyon,  40  n.  10 
Knight  of  the  Swan  legend,  180,  218, 

239-252 
Knighton,  Henry,  Chronicon,  104,  109 


33*> 


INDEX 


Konrad  von  Wiirzburg,  66,  67  n.  3, 

202  n.  5,  246,  255 
Kurzman,  Andreas,  66 

Lacock  Abbey,  93  n.  9 

Lady  and  the  Monster,  318  n.  12 

Lamia  type  in  Melusine,  207  n.  14 

Lamprecht,  167 

Lancelot,  116;  Lanzelel,  226 

Laneham,  Robert,  263  n.  1 

Latin  literature:  De  Arthur 0  Rege 
Britanniae  et  Rege  Gorlagon  lycan- 
thropo,  55  n.  8,  180,  218,  260  n.  18; 
Canticum  Colbrondi,  131;  Colump- 
narium,  27;  Eulogium  Historiarum, 
104;  Flores  Historiarum,  46;  Gesta 
Apollonii,  164,  170;  Gesta  Her ewardi, 
83;  Historia  Meriadoci,  107;  His- 
toria  Regis  Waldef,  102  n.  8;  Historia 
Septem  Sapientum,  177  n.  7,  8; 
Itinerarium  regis  Ricardi,  150;  Mir- 
aculum  (Crescentia),  13;  Philopertus, 
36,  38  n.  6;  S.  Eustachius,  6;  Ser- 
mones  Parati,  255;  Tractat  de  dubii 
Nominibus,  169;  Vitae  Duorum 
Of  arum,  23-24;  Vita  Sanctorum 
Amici  et  Amelii,  66,  68;  Ystoria 
Regis  Franchorum,  28.  See  also 
Exempla,  Gesta  Romanorum;  Legenda 
A  urea 

Laundress  as  character,  235 

Layamon,  52  n.  3,  93 

Lays.  Cf.  French  literature  and  Marie 
de  France 

Learning  of  women,  240.  Cf.  Education 

Legenda  Aurea,  7,  32 

Leprosy  as  a  punishment,  71 

Letter  of  death  (Uriasbrief),  119,  122, 
135  n.  15,  137;  forged  or  substituted, 
23,  24,  25,  26,  29,  242 

Liaison  of  princess  with  vassal,  71, 
264;  of  prince  with  girl,  184 

Libeaus  Desconus.  Cf.  Chestre;  French 
literature,  Le  Bel  Inconnu 

Libraries,  mediaeval.  See  Savage.  Cf. 
Catalogues 

Life  or  death  dependent  upon  story 
telling,  179 

Life  tree,  218 

Likeness  in  name  and  body,  69,  191 
n.  17 

Lillo,  George,  Marina,  166 


Lincolnshire,  105,  108,  149,  154 

Lion,  faithful,  271;  grateful,  136  n.  17; 
fight  with  lion  of  Richard  I,  151; 
unable  to  attack  a  virgin  princess, 
280  n.  2 

Lists  in  romances,  312.     Cf.  Catalogues 

Livy,  112 

Lodge  built  in  woods,  236 

Lodge,  Thomas,  51;  Rosalynde,  158 

Lohengrin,  101  n.  7,  209,  245,  246 

London  in  romance,  140,  143;  riot  in, 
116 

Longfellow,  Tales  of  a  Wayside  Inn, 
61 

Lope  de  Vega,  67 

Lord  of  Lome  and  the  False  Steward, 
290 

Loup-garou,  217 

Louvet,  Jean,  130 

Love  in  Absence  (love  caused  by  hear- 
say), 47,  208  n.  18.    See  Zadi 

Lover,  pretended,  269,  285  n.  5;  super- 
natural, 302.     Cf.  Fairy  Lover 

Love-sick  knight,  276  n.  3,  309;  Love 
sickness,   133,  228 

Love-spot,  316 

Lusignan,  legend  of.    Cf.  Melusine 

Lydgate,  John,  131 

Lyndsay,  David,  319 

Lyric  passages  in  romance,  147 

Mabinogion,  176  n.  6,  198 

Madness    due    to    love-sickness,    209, 

211  n.  23 
Magic  arts,  204,  217,  248,  304,  305 
Magician,  62,  305  n.  11 
Magic    in    romance,    192-193.       See 

Easter;    Hallauer;     Kurtz.      Magic 

house,  206  n.  12;    cf.  Fairy  Castle. 

Magic  hunt,  234;  phial,  67;  pillow, 

234;    potion,  314  n.  3.     Cf.  Ring, 

Ship,  Shirt 
Magnanimity  of  hero,  allows  enemy  to 

refresh  himself,  136 
Maiden  messenger,  302.  Cf.  Confidante 
Maillart,  Jehan,  27 
Maket,  Jakemon,  255,  260;    Chdtelain 

de  Couci,  255,  257,  261,  266 
Maldon,  Battle  of,  99 
Malory,  Sir  Thomas,  55,  80,  198  n.  7, 

218  n.  7,  227 
Mandeville's  Travels,  193 


INDEX 


337 


Manekin,  242;  cf.  Emare;  Girl  without 

hands 
Manning,  Robert,  of  Brunne,  103,  105, 

128 
Mantles  held  for  ransom,  270 
Man  tried  by  Fate,  Job  theme,  3,  5,  8, 

9;     Man    with    Two    Wives.      Cf. 

Husband 
Manuel,  Don  Juan,  59 
Mappamonde,  306 
Map,  Walter,  224,  226  n.  3 
Marie  de  France,  294  n.  1;   Lais,  105, 

200;  Bisclavret,  217,  218-220;  Eliduc, 

124,  237,  298;   Lai  del  Fraisne,  295, 

298,  302;  Guigemar,  152,  207-208, 
234;  Guingamor,  207-209;  Lanval, 
207-209,  301;  Melton,  208,  218-220; 
Yonec,  52  n.  3,  197,  205  n.  10,  208 

Marriage:  forced,  45,  92,  103,  no, 
119;  with  a  god,  191,  206;  mar- 
riage in  name  only  (Sheinehe),  8 
n.  4,  9,  69,  119,  122;  marriage  pre- 
tended as  a  test  of  loyalty,  299; 
marriage  of  widow  to  husband's 
murderer,  119;  marriage  of  mother 
and  sort,  276,  303 

Marvelous,  The.  Cf.  Magic.  See 
Hallauer 

Massinger,  Fatal  Dowry,  76 

Matriarchy.  Cf.  Sister's  Son.  See 
Aron 

Mayence  in  legend,  245 

Medicine  in  Old  French  Poetry.  See 
Kiihn 

Melusine  legend,  151,  206  n.  14,  209 

Mercenary  in  mediaeval  literature.  See 
Neumann 

Merchants.  Cf.  Disguise;  Trade  and 
Commerce 

Merlin,  52  n.  3 

Mesalliance,  190,  191;  princess  and 
page,  221;   feudal  lord  and  peasant, 

299.  Cf.  Liaison 

Messengers,  145.  Cf.  Maiden  mes- 
sengers 

Metempsychosis,  62 

Metham,  John,  192 

Mi61ot,  Jean,  13 

Minstrels.  See  Brandl;  Grossmann. 
Gifts  to  minstrels,  216,  274,  279; 
Minstrels  at  monasteries,  146  n.  4,  5. 
Features  of  minstrel  style,  46,  106, 


143  n.  1,  157,  195,  306.  See  Halper- 
sohn 

Miracle  of  tombs  that  move,  68;  of 
speech  restored  to  dumb  princess,  55; 
of  fruit  in  winter,  79  n.  2;  Miracles 
of  Virgin,  19-20,  36,  53,  65.    Cf.  Bells 

Miraculous  Conversion,  6 

Misbegotten  or  misformed  child,  45, 
48  m  6 

Mistress,  discarded,  serves  bride  at 
wedding  feast,  298 

Mohammedanism,  mediaeval  concep- 
tions of,  47 

Montargis,  Dog  of,  283-284,  286-287 

Morris,  William,  308 

Mortara,  Italy,  68 

Morte  Arthur  e,  15,  225 

Moslem  Princess,  Enamoured,  123 

Mother-in-law,  cruel,  240,  242,  206  n.  12 

Moulton,  Thomas  de,  148,  149  n.  3 

Murderers,  hired,  31  n.  14 

Mutilation  of  dead  body,  254,  264 

Mysteres.    See  Petit  de  Julie ville 

Name,  lover  unable  to  say  whole  of 
sweetheart's  name,  211  n.  25,  229 
n.  6;  popular  names,  51,  106; 
names  indicative  of  experience,  301; 
similarity  of  names,  90  n.  7,  191  n. 
17,  297-298.    Cf.  Demon;  Taboo 

Nature,  feeling  for,  216  n.  5 

Nebuchadnezzar,  legend  of,  56,  61,  63 

Necromancy,  205  n.  9.    Cf.  Magic  arts 

Night  visit  or  night  courtship,  307 

Norfolk,  162 

Normandy,  49,  51 

Norse  literature:  Legends;  Alaflekks- 
saga,  218  n.  7;  AnlocH,  123;  Gudrun, 
258;  Karlamagnussaga,  37  n.  3,  245, 
247;  Njalssaga,  162;  Olaf  Tryggva- 
son  saga,  100,  in;  Volsungasaga, 
217.  Versions  of  mediaeval  romances : 
Amis,  65  n.  2,  66;  Chevalier  au  Cygne, 
247.  Floris,  189;  Octavian,  267; 
Partonope,  201  n.  3,  204,  208  n.  16; 
Ponthus  and  Sidoyne,  88;  Robert  of 
Sicily,  58;  Seven  Sages,  178 

Nose  cut  off,  152 

Octavian,  7,  39  n.  8,  179,  267-273,  274, 
277,  280  n.  4,  284,  286  n.  6,  288,  295 
(Edipus  theme,  277,  303 


338 


INDEX 


Offa  legend,  23,  24  n.  3,  29,  30 

Offspring,  monstrous,  23 

Old  age  of  hero,  138.    Cf.  Cloister 

Oliva,  legend  of,  37  n.  3 

Orchard,  scene  of  supernatural  wooing, 

53 

Ordeal  by  battle  (cf.  Combat,  judicial), 
35,  37;  by  fire,  143;  of  fiery  plough- 
shares, 145 

Orderic  Vitalis,  123 

Orfeo,  192  n.  20,  195-199,  294,  301 

Oriental  literature.  Cf.  Eastern  ele- 
ments in  romance 

Orthodoxy  of  supernatural  lover  proved, 
205  n.  10 

Otfrid,  85 

Other  world  abode,  197,  314.  Cf.  Fairy 
castle 

Ottokar  von  Horneck,  45 

Otto  the  Great,  120 

Outlaw  hero,  160-161  n.  5 

Ovid,  170,  191,  196.  See  Guyer. 
Schevill;  Schrotter 

Painted  walls,  151;    in  romances,  308 

n.  4 
Painter's  Palace  of  Pleasure,  37,  256 
Palermo,  216,  221,  222 
Paradise,  Earthly,  193 
Paris,  Matthew,  46,  250 
Parks   broken   in    hunting   raid,    310 

n.  5 
Parlement  of  Thre  Ages,  231 
Parliaments,     early     mention     of     in 

romances,  105 
Parrot,  speaking,  259 
Partenay,  city  of,  207  n.  14 
Partonope  (Partenopeus)  of  Blots,  132  n. 

11,  152,  190  n.  15,  200-213,  226>  228, 

305  n.  11 
Patience,  test  of.    Cf.  Griselda 
Patronage,  108,  149,  203,  215 
Paul  the  Deacon,  37,  136 
Paym,  Roberto,  168 
Pearl,  308 
Peasant  life.     Cf.  Vilain.     See  Reich; 

Witter 
Peele's  Old  Wives  Tale,  76 
Pembrokeshire,  121 
Penance,  chivalric,  135  n.  13,  153;   cf. 

Calais;  Religious  penance,  living  with 

dogs,  54-55,  56 


Perceval  legend,  153,  243  n.  6;  Perceval, 
306,  307 

Perilous  Well,  315 

Persia  in  romance,  233 

Persian  legends,  232-233 

Physical  peculiarity  cause  of  name, 
296  n.  5;  resemblances,  297 

Physician,  woman  as,  12,  18,  21 

Pierot  du  Ries,  117 

Pierre  de  Langtoft,  103,  no,  151 

Pierre  d'Esrey,  239,  248 

Piers  Plowman,  160 

Pietistic  elements  in  romance,  60,  243, 
279 

Pilgrimage  vowed  for  sake  of  child,  285 

Pilgrim  routes,  to  Compostella,  185, 
191;  to  Italy,  68;  London  to  Canter- 
bury, 143 

Place  names  in  Middle  English,  89,  91, 
92,  98,  100;  in  Old  French,  120  n.  3 

Placidas  (St.  Eustache),  6 

Plutarch,  286 

Poison  Maiden,  77 

Poli,  Stephen,  58 

Polykrates  legend,  32  n.  17 

Ponthus  and  Sidoine,  87 

Porter,  surly,  98,  119,  140,  160,  195 

Potiphar's  wife,  218  n.  7,  233  n.  1 

Prayers  in  romance,  279,  310 

Priestess,  wife  of  Apollonius  serves  as, 
169 

Prince  or  Princess,  False,  substituted 
for  true,  292 

Princess  kept  unwedded  by  her  father, 
302;  as  prize  of  tournament,  226, 
229;    Proud  Princess,  129,  133,  313 

Prisoners,  grateful,  291 

Prison  pit,  hero  escapes  from,  119,  124, 
135  n.  15,  233;  heroine  condemned 
to,  243 

Prohibition  against  looking  at  sweet- 
heart, 205;  against  speaking  of  lady, 
209  n.  20,  256  n.  12;  broken,  191, 
206  n.  12,  209.    Cf.  Counsels 

Promise,  rash,  198,  302 

Prose  romances  in  England,  127  n.  1, 
216  n.  4 

Provencal  literature:  Daurel  et  Beton, 
118;  Philomena,  113,  166  n.  4 

Proverbs  in  Old  French  literature,  see 
Schepp;  in  Middle  Eng.  romance, 
106,  275 


INDEX 


339 


Proxy,  lady  won  by,  318 
Pucci,  Antonio,  169,  227 
Purity,  63 

Pursuit,  vain,  119,  135  n.  15,  237 
Pyramus  and  Thisbe,  191  n.  19 

Quests  of  lovers,  190-191,  232 

Ralph  of  Diceto,  150 

Rank,  terms  of.    See  Voltmer 

Raouf  de  Boun,  Petit  Brut  d'Angleterre, 

103,  104,  109 
Raoul  le  Tourtier,  65,  67 
Recognition  tokens,  10,  71,  232,  297, 

303.    Cf.  Fold  in  tunic;  Ring 
Red,  unlucky  color,  292  n.  3 
Regino  of  Priim,  37  n.  3,  121 
Reinbrun,  129,  140-142 
Reinmann  von  Brennenberg,  257 
Religion  and  superstition  in  romance. 

See  Geissler 
Renart,    Jean,    214,    Lai   de   V  Ombre, 

L'Escoufle,    9,    Guillaume    de    Dole, 

258  n.  15 
Renaud  de  Beaujeu,  Le  Bel  Inconnu, 

142,  205  n.  9,  209,  2ii,  227,  304-305 
Reunion  and  recognition  of  long-sepa- 
rated families,  9,  221,  288 
Richard  Coeur  de  Lion,  47  n.   5,   135 

n.  13,  144,  147-155,  177,  226,  258; 

Richard  I,  28  n.  10 
Richard  le   Pelerin,  244.    Cf.    French 

literature,  Chanson  d'Antioche 
Richard  of  Cirencester,  144 
Richard  of  Devizes,  144,  150 
Riddles  in  romance,  93,  99,  171  n.  8 
Ring,  magic,  92,  98,  153,  188  n.  n, 

218,  220,  275;  as  recognition  token, 

88,  237 
Rishanger,  William,  46 
Rites   of   Church:    cf.   Baptism;    for 

betrothal  rites  see  Critchlow 
Robert  of  Cesile  (Sicily),  50,  58-64 
Robert  of  Gloucester,  147 
Robert  of  Jumieges,  145 
Robert  the  Devil,  63,  226.    Cf.  Gowther 
Rodrigo  de  Herrera,  58 
Roemoldt,  John,  60 
Roger  of  Hoveden,  104 
Roger  of  Toeni,  249 
Romances  of  Antiquity,  228.    Cf.  Ovid 

here.    See  Ogle 


Romance  words  in  Middle  English, 
86  n.  4 

Rome  as  setting  of  story,  179 

Rome,  Book  of,  272,  274,  280  n.  4 

Rosenblut,  Hans,  13 

Roswall  and  Lillian,  226,  265,  290-293 

Roumanian  version  of  Beves,  118 

Round  Table,  306  n.  1 

Rous  Roll,  131  n.  9 

Rowe,  Nicholas,  Fair  Penitent,  76 

Rowlande  and  Otuel,  132  n.  11 

Rowlands's  Famous  Historie  of  Guy  of 
Warwicke,  127,  132 

Roxburghe,  fortress,  105 

Rudborne,  Thomas,  131 

Rudel,  47,  226  n.  1.  Cf.  Love  in 
absence 

Ruses  of  lovers,  258  n.  16 

Russian  versions  of  mediaeval  ro- 
mances: Apollonius,  165;  Beves,  115, 
118;  Seven  Sages,  178 

Sachs,  Hans,  13,  37,  58,  267,  284 

Saint  Albans,  founding  of,  24 

Saints:  Ambrose,  286;  Albano,  56; 
Alexis,  27  n.  9,  55  n.  8,  137  n.  18; 
Anthony,  281;  Cuthbert,  11 1  n.  5; 
Elizabeth,  21;  Eustache,  3,  5,  6,  7, 
9,  10,  11,  27  n.  9,  270,  272,  277, 
280  n.  2;  George,  124,  152,  153; 
Giovanni  Boccardoro,  56;  Guglielma, 
13,  14;  Guillaume  de  Gellone,  38 
n.  7;  Guthlac,  51;  Helena,  9,  29; 
Hubert,  4,  6  n.  1;  James,  157; 
Martin  of  Tours,  27  n.  9,  169; 
Nicholas  of  Bari,  78  n.  8,  281;  Peter, 

Saints  Legends.    See  Gerould 

Saint   Swithin's    Church,   Winchester, 

131 
Saladin,  151,  155 
Samson,  151 
Saracen  in  fiction,  47,  54,   119,   136, 

137,  153,   154,   190,   193,  200,  236, 

270.    Cf.  Moslem  Princess 
Saxo  Grammaticus,  122 
Scandinavia:    Scandinavian  names  in 

England,  92  n.  8.    See  Leach 
Schmidt,  Rudolph,  60 
Schumann,  Valentine,  58 
Scott,  Sir  Walter,  290,  317,  319 
Scullion,  prince  serves  as,  1 1 1 


340 


INDEX 


Sea,  exposure  on,  24-25,  30.  See 
Frahm;  Kramer 

Sebilla,  legend  of,  39  n.  8,  283,  285,  287 

Seneca  the  Rhetorician,  123;  Seneca 
the  Dramatist,  170 

Seneschal,  false,  40,  285;   jealous,  71 

Separation  of  family,  123;  of  husband 
and  wife,  167;  of  lovers,  170,  190, 
191 

Sercambi,  59  n.  5,  254,  299  n.  9 

Serpent,  man  or  woman  changed  into, 
142 

Servants,  invisible,  205,  206  n.  12 

Seven  Arts,  205  n.  9,  304 

Seven  Champions  of  Christendom,  1 24 

Seven  Sages,  174-183,  192  n.  20,  201 
n.  3,  218  n.  7,  219,  233  n.  3,  240  n.  2. 
One  of  Seven  Sages,  234 

Shakespeare's  Pericles,  166;  As  You 
Like  It,  158 

Sheriff,  false,  162 

Ship,  magic,  148,  152,  208  n.  16;  ship- 
wreck, 74,  166  n.  4,  167 

Shirt,  magic,  235  n.  5 

Sicily,  28  n.  10,  59,  220,  221  n.  12 

Sidney,  Sir  Philip,  275 

Simeon  of  Durham,  99 

Sindibad,  philosopher,  174 

Sisters,  false,  206  n.  12 

Sister's  Son.  See  Farnsworth;  Gummere 

Siward  of  Ardern,  133 

Slave,  hero  sold  as,  111,  119,  140,  141 

Sleeping,  dangerous,  under  certain 
trees,  197  n.  5 

Social  life  in  the  thirteenth  century, 
257.  See  Langlois.  Social  Protest, 
see  Wood.  Social  Ideals,  see  Law- 
rence 

Sohrab  and  Rustum,  227  n.  5 

Solomon  legend,  62  n.  8,  63,  171  n.  10 

Songs  combined  with  romance  nar- 
rative, 258  n.  15 

Spanish  literature:  version  of  Alpha- 
betum  narrationum,  13;  (Catalan 
tales:  Historia  del  Rey  de  Hungria, 
27;  Istoria  de  la  Filla  de  V Emperador 
Contasti,  28;)  El  Cavalier 0  Cifar,  3, 
7,  8;  El  Conde  de  Barcelona,  35;  El 
Conde  Lucanor,  59;  version  of  Con- 
stance legend,  27;  of  Erie  of  Tolous, 
43;  of  Florence  de  Rome,  14;  Flores 
y  Blancaflor,  188;  La  Gran  Conquista 


de  Ultramer,  242;  Libro  de  Apolonio, 

168;  Libro  de  los  engannos,  174,  178; 

La  Linda  Melisenda,  67;    version  of 

Macaire,  283;  Oliva,  37  n.  3;  version 

of  Olivier  de  Castile,  67;   Partinuples 

de  Bles,   204;    version  of  Robert  le. 

Diable,  50;    of  Robert  of  Sicily,  58; 

of  St.  Eustache,  6  n.  2;    Tirant  lo 

Blanch,  129 
Speculum  Vitae,  128 
Speech,  of  birds  and  beasts,  70;    mi- 
raculously restored,  55 
Spendthrift  Knight  theme,  73,  79 
Spy  betrays  lovers,  266.   Cf.  Seneschal; 

Steward 
Squire,  faithful,  211  n.  24,  237  n.  7,  310; 

hero  serves  as,  226 
Squyr  of  Lowe  Degre,  263-266 
Stag   with  crucifix,   5;     messenger,   6 

n.  1 
Stakes,  dead  men  set  up  on,  113 
Stanza-linking,  307 
Stepmother,  false,  cruel,  jealous,  174, 

219,  233,  236,  237,  248 
Steward,    false,    236,    264,    283,    291; 

jealous,  236;  faithful,  195 
Stock  phrases  in  Middle  English  ro- 
mance, 97,  116,  143  n.  i,  276 
Strength  increases  up  to  noon,  315 
Strieker,  Der,  59 
Suicide  attempted  or  meditated,   184 

n.  2,  192  n.  19,  201,  253,  256  n.  12, 

260 
Suitor,  barbarous,  cruel,  haughty,  119, 

227,   236,   304;    fated,   171;    lowly, 

265,  275 
Sun  God,  248 
Supernatural  being  as  lover  and  father, 

52>  535    see  Hartland;    as  warrior, 

316 
Superstitions,  popular.     Cf.  Counsels; 

Twins.    See  Geissler;  Magnus 
Swan  as  heraldic  device,  250 
Swan  children,  legend  of,  240,  241-242, 

244-248,  249-251;   Swan  guide,  244; 

Swan  Knight,  legend  of,   241,   243, 

250;  Swan  maiden,  29,  31,  208  n.  17, 

240 
Swans,  children  turned  into,  247-248; 

lovers  (gods)  turned  into,  248 
Swedish  literature:  Hertig  Frbjdenborg, 

257;    Legendarium,  75;    Versions  of 


INDEX 


341 


tales  and  romances:  Apollonius,  165; 

Der  Seelen   Trost,   13;    Floris,   189; 

Seven  Sages,  176,  178 
Sword,  Eldrige,  276  n.  3;    magic,  141; 

pointless  sword  left  for  unborn  son, 

302 
Sword    of    Chastity,    69    n.    8,    236; 

Sword  bridge,  197  n.  4,  136;   Sword 

of  fate,  276  n.  3 
Sworn    brother   or    friend,    101.      Cf. 

Brotherhood 
Symposius,  Enigmata,  171 

Taboo,  perils  of  breaking,  70,  206,  209 

n.  20,  244-245 
Talbot,  William,  93  n.  9 
Talisman,  congenital,  218 
Tapestries,  of  Beves,  115  n.  1;  of  Guy, 

132 
Tasks,  impossible,   77  n.   5,   191,   206 

n.  12,  275,  280 
Taylor,  John,  313 
Tears,    cannot    be    washed    out,    234; 

tears  awaken  hero,  234.    See  Beszard 
Templar  tales,  46 
Test  for  suitable  wife,  fitting  of  gloves, 

303 

Thebes.  Cf.  French  literature,  Ro- 
mande  T. 

Thomas,  author  of  Horn  et  Rimenhild, 
83-84,  88,  91,  99,  101;  author  of 
Tristan,  83  n.  2,  254 

Thomas  of  Walsingham,  46 

Three:  cf.  Counsels;  Wishes.  See 
Miiller 

Thurkill  of  Warwick,  133 

Tobit,  Book  of,  74,  77 

Toeni,  family  of  Roger  of,  249,  250 

Tombs,  false,  192 

Torrent  of  Portyngale,  8,  132  n.  11, 
279-281 

Tortures,  20,  240 

Tournament,  211  n.  26;  Father  over- 
throws daughter's  suitors  at  tourna- 
ment, 302;  Princess  prize  of  tourna- 
ment, 276,  288,  303;  Prizes  at 
tournament,  134;  Three  Days'  Tour- 
nament, 55,  152,  225,  226  n.  2,  229; 
148,  227,  293.  See  Cripps-Day; 
Clephan;  O.  Mueller;  Weston 

Tower  of  Maidens,  189,  190,  193; 
Tower  prison,  20 


Trade  and  Commerce.    Cf.  Merchants. 

See  Sallentien 
Transformation     of     human     beings: 

change  of  color  and  form,  45,  47; 

children  turned  into  swans,  241,  248. 

Cf.  Dog;   Serpent.     See  Goerke 
Travellers  in  Orient,  Italians,  260 
Treasure  stolen  by  a  bird,  9,  277 
Trial  by  Combat.    Cf.  Combat,  judicial, 

132  n.  11 
Triamour,  39  n.  8,  283-289,  309 
Tristrem,  Sir,  97 
Trivet,  Nicholas,  24,  25,  26  n.  7,  30, 

3i,47 
Trojan  legend,  207.     Cf.  Benoit 
Troy  Book,  132  n.  11 
Tuberville's  Tragical  Tales,  256 
Twelve-line,  tail-rime  stanza  in  Middle 

English,  128  n.  4 
Twine's  Patterne  of  Paine  full  Adventures, 

165 
Twins,    considered    sign    of    adultery, 

269,  295 
Two  Friends,  folk-tale.    Cf.  Friendship 

theme,  69,  76 

Ulrich  von  Eschenbach,  3;   Ulrich  von 

Zatzikhoven,  226,  316 
"Undo  your  door,"  264  n.  3 
Unlucky  types  of  people,  292  n.  3 
Usurper  overthrown,  233;    tries  to  kill 

true  heir,  121 
Uther,    father    of    King    Arthur,    53; 

Uther  and  Igerne,  80 

Valerius  Maximus,  74 

Valkyrie  type  of  heroine,  122,  318  n.  12 

Vassals  demand  that  their  lord  marry 

298 
Venantius,  Bishop  of  Poitiers,  169 
Viking  heroes,  90,  91;    sagas,  92  n.  8, 

108  n.  4,  120,  124 
Vilain.  See  Galpin 
Vincent  of  Beauvais,  7, 13,  66,  206  n.  13, 

,24.7 
Virgil,  2>d>  n.  18;    instructor  of  prince, 

176;  maker  of  magic  objects,  180 
n.  12;  Georgics,  196;  Virgilian  influ- 
ence in  romance,  170,  196 
Virgin  Mary,  53,  54;  Cult  of,  20; 
Miracles  of,  12,  13,  19,  20,  36: 
Visions  of,  268 


342 


INDEX 


Virgin  princess  and  lions,  280  n.  2 
Virginity  protected  by  magic,  15  n.  2, 

119 
Vow  not  to  cut  beard,  169;   chivalric 

vows,  135  n.  13 

Wade,  116 

Waldef,  101,  102  n.  8,  in  n.  6 

Walls.     Cf.  Painted  Walls 

Walter  de  Hemingburgh,  151 

Walter  of  Exeter,  131 

Warfare,  methods  of,  154 

Warwick,  133 

Water  of  Life,  77  n.  5 

Wayland  Smith,  208  n.  17.    Cf.  Weland 

Weland,  281  n.  5,  98 

Well  of  Chastity,  98  n.  3;  of  Youth,  193 

Welsh  alliterative  poetry,  307 

Welsh  tales  and  versions  of  romances; 

Amis,  66;  Beves,  116;    Seven  Sages, 

176.     Cf.  Mabinogion 
Werwolf,  180,  214,  217  n.  6,  218,  219  n.  9 
"West   Midland"    of    the    Romances, 

215  n.  3 
Westminster  Abbey,  46,  145 
Wickram,  Georg,  37 
Widow  or  Matron  of  Ephesus,  181 
Wife,     mistakes     her     husband,     69; 

Chaste,  39,  42;  Cruel,  71;  Faithless, 

218,  219,  233;    Innocent  persecuted, 

cf.  Erie  of  Tolous,  Emare,  Florence. 

See  Siefkin 


Wife's  Complaint,  Anglo-Saxon  poem, 
21,  23  n.  2,  31 

Wilde,  Sebastian,  178,  267 

Wild  man,  291 

Wilkins's  Pericles,  166 

William,  Archbishop  of  Tyre,  249 

William  Longespee,  93  n.  9 

William  of  Malmesbury,  37,  144 

William  of  Orange,  38,  138 

William  of  Palerne,  112,  214-23,  229, 
237,  265 

Winchester,  132,  146,  196 

Wirnt  von  Gravenberg,  227 

Wish  child,  52,  55 

Wishes,  Three,  80  n.  3 

Witch  and  daughter,  235  n.  4 

Wolfram  von  Eschenbach,  246 

Wolf's  head,  162 

Woman  as  wooer,  71,  152.  Cf.  Forth- 
putting  woman.  Woman  ransomed, 
77;  Woman  tried  by  fate,  8  n.  3 

Wonder,  paraphernalia  of,  in  Celtic 
romance,  228 

Wynkyn  de  Worde,  51,  87,  127  n.  1, 
148,  166,  178,  215,  223,  229,  239,  263 

Yolande,  Countess  of  Flanders,  215,  220 
Yorkshire,  25,  99 

Youth,  fountain  of,  77  n.  5.      Cf.  Well 
Ywain,  136.     Cf.  Chretien 

Zenophon  of  Ephesus,  171 


I.  AUTHOR'S  NOTE 

Thirty-five  years  have  elapsed  since  the  publication  of  Medi- 
eval Romance  in  England,  a  work  now  long  out  of  print  but 
still,  according  to  the  present  publisher,  in  frequent  demand. 
It  is  regularly  cited  in  current  bibliographies. 

Scholars  who  reviewed  the  work,  some  of  whose  reviews  are 
listed  below,  agreed  in  affirming  its  special  value  for  the  study 
of  the  sources  and  development  of  the  thirty-nine  non-cyclic 
romances  discussed  in  it.  With  some  minor  exceptions,  these 
discussions  have  stood  the  test  of  time. 

Since  1924,  however,  the  bibliography  of  the  scholarship  in 
this  field  has  greatly  increased.  It  was  planned  at  first  that  full 
bibliographical  lists  of  publications  concerned  with  these  partic- 
ular romances  should  form  a  supplement  to  the  new  edition. 
These  lists  would  employ  the  same  form  as  that  used  in  the  first 
edition,  and,  being  added  to  them,  would  constitute  a  complete 
guide  to  the  literature  of  the  subject  up  to  the  year  1959. 

Reasons  of  health  obliged  the  author  to  modify  this  plan,  but 
she  has  sought  to  accomplish  the  same  purpose  by  a  somewhat 
novel  method,  that  is,  by  giving  specific  references  to  the  five 
standard  bibliographies  of  medieval  literature  listed  below.  (See 
III).  This  method  results  in  a  comprehensive  and  co-ordinated 
bibliographical  index  for  each  romance.  For  the  invaluable  nine 
supplements  to  Wells'  Manual  of  Writings  in  Middle  English, 
which  up  to  now  have  lacked  a  general  index  for  additions  to 
material  originally  included  in  the  Manual,  the  new  edition  of 
Medieval  Romance  provides  one  for  the  non-cyclic  romances 
with  which  it  is  concerned.  It  performs  the  same  function  for 
the  same  group  in  the  annual  bibliographies  of  Middle  English 
in  the  Publications  of  the  Modern  Language  Association. 

The  references  noted  infra  for  each  romance  indicate  the 

343 


precise  volume  and/ or  page  of  the  general  bibliographies  in- 
dexed. This  mode  of  reference  is  followed  invariably  and  so 
avoids  the  many  variant  systems  of  numbering,  systems  that  at 
times  are  both  cumbersome  and  confusing.  Though  the  general 
bibliographies  (III)  inevitably  repeat  information,  they  often 
add  to  it  by  critical  notes,  by  mention  of  book  reviews,  by  cross 
references,  and  by  corrections  of  earlier  citations. 

The  romances  themselves,  grouped  in  other  works  under 
diverse  headings,  are  here  arranged  in  simple  alphabetical  order 
according  to  the  personal  name  of  the  hero  or  heroine.  Titles  of 
rank,  such  as  King  or  Sir,  which  appear  so  inconsistently  in 
romance  titles  before  the  names  of  some  heroes  (King  Horn, 
Sir  Orfeo),  and  not  of  others,  here  follow  the  personal  name. 
So,  likewise,  do  such  descriptive  terms  as  Le  Bone  for  Florence 
of  Rome,  or  Lai  (Lay)  for  such  poems  as  Lai  le  Freine  or  the 
Lay  of  Havelok. 

The  system  of  bibliographical  reference  here  employed  does, 
it  is  true,  leave  the  actual  compilation  of  bibliographical  infor- 
mation for  any  one  of  these  romances  to  the  reader,  but  he  can 
be  assured  that  if  he  follows  up  all  the  sources  here  indicated, 
his  final  citations  will  rest  on  five  of  the  most  widely  used, 
expert,  and  persistent  bibliographical  efforts  of  our  time. 


344 


II.  REVIEWS  OF  MEDIEVAL  ROMANCE  IN  ENGLAND 

Modern  Language  Notes,  XLI  (1926),  406f.,  by  Kemp  Malone. 

Journal  of  English  and  Germanic  Philology,  XXV  (1926),  105-114,  by 
Howard  Patch. 

Modern  Philology,  XXIV  (1926-27),  122-24,  by  John  M.  Manly. 

Folk-Lore,  XXXVI  (1925),  291-93,  by  M.  Gaster. 

Zeitschrift  fur  romanische  Philologie,  XLVI   (1926),  500,  by  Alfons 
Hilka. 

Beiblatt  zur  Anglia,  XXXVI  (1925),  332-336,  by  Gustav  Binz. 

Saturday  Review  of  Literature,  Dec.  27,  1924,  1,  419.  Anon. 

Year's  Work  in  English  Studies,  V  (1926),  91  f.,  by  E.  V.  Gordon. 

Modern  Language  Review,  XX  (1925),  339,  by  Cyril  Brett. 


345 


III.  GENERAL  BIBLIOGRAPHIES 


CBEL.  Cambridge  Bibliography  of  English  Literature,  I  (600-1660 
a.d.),  ed.  by  F.  W.  Bateson,  New  York,  Cambridge  (England), 
1941  (data  to  1936);  CBEL,  V,  Supplement,  a.d.  600-1900 
(data  to  1955),  ed.  by  George  Watson,  Cambridge  (England). 

MAN.  Manual  of  Writings  in  Middle  English,  1050-1400,  by  John 
E.  Wells,  New  Haven,  London,  Oxford,  1916;  Sixth  Printing, 
1937;  Supplement  (S)  I  (1919);  II  (1923);  III  (1926);  IV 
(1929);  V  (1932);  VI  (1935);  VII  (1938);  VIII  (1941),  cf.  p. 
1657  for  issues  of  the  Manual  and  its  Supplements;  IX  (1951), 
with  data  to  1945,  ed.  by  Beatrice  Daw  Brown,  Eleanor  K. 
Heningham,  Francis  Lee  Utley.  A  thorough  revision  of  the 
Manual  and  its  Supplements  is  to  be  published.  All  issues  have 
been  published  at  New  Haven,  Conn,  for  the  Connecticut 
Academy  of  Arts  and  Sciences,  by  the  Yale  University  Press. 

PMLA.  Publications  of  the  Modern  Language  Association  of  America. 
Its  "American  Bibliography"  or  "Annual  Bibliography"  under 
the  heading  "English  Language  and  Literature"  has  annually 
devoted  a  section  to  Middle  English.  This  began  with  vol. 
XLII  (No.  1.  1927).  "Bibliography  for  1958"  appears  in  vol. 
LXXIV  (No.  2,  1959). 

R  &  O  Renwick,  W.  L.  and  Harold  Orton,  The  Beginnings  of  Eng- 
lish Literature  to  Skelton,  1509.  Revised  2nd  edition,  London, 
1952.  Selective  bibliographies  and  notes  on  the  Middle  English 
romances. 

Bossuat  Robert  Bossuat,  Manuel  Bibliographique  de  la  Litterature 
Francaise  du  Moyen  Age,  Melun,  1951;  Supplement  (S)  for 
1949-53.  Paris,  1955. 


346 


IV.  INDEX  OF  NON-CYCLIC  ROMANCES  AND 
BIBLIOGRAPHICAL  REFERENCES  (1926-1959) 

Amadace  (das),  Sir.  CBEL,  I,  154-55;  V,  116.  MAN,  p.  787;  S,  I, 
1006;  S,  II,  1107;  S,  III,  1210;  S,  IV,  1302;  S,  V,  1390;  S,  VI, 
1491;  S,  VII,  1608;  S,  VIII,  1706.  R  &  O,  pp.  387-88. 

Amis  and  Amiloun.  CBEL,  I,  154;  S,  V,  116.  MAN,  p,  787;  S,  II, 
1107;  S,  III,  1210;  S,  VI,  1491;  S,  VII,  1608;  S,  VIII,  1706; 
S,  IX,  1899.  PMLA,  LII  (I),  1234.  R  &  O,  p.  387.  Bossuat, 
pp.  22-24;  S,  p.  22. 

Apollonius  of  Tyre.  CBEL,  I,  94f.  MAN,  p.  784;  S,  I,  1006;  S,  II, 
1106;  S,  III,  1209;  S,  V,  1389;  S,  VI,  1491;  S,  VIII,  1705; 
S,  IX,  1898.  R  &  O,  pp.  383f.  Bossuat,  pp.  119f.  PMLA, 
LXXIII(2),  150. 

Athelston.  CBEL,  I,  150f;  V,  115f.  MAN,  p.  766;  S,  II,  1101;  S,  III, 
1205;  S,  IV,  1296;  S,  V,  1383;  S,  VI,  1485;  S,  VII,  1601. 
R  &  O,  p.  358.  PMLA,  XLV  (1),  21. 

Beues  of  Hamtoun.  CBEL,  I,  150.  MAN,  p.  765f;  S,  I,  1003;  S,  II, 
1101;  S,  III,  1205;  S,  V,  1382;  S,  VIII,  1700;  S,  IX,  1892. 
R  &  O,  p.  357f.  Bossuat,  pp.  30-32;  S,  p.  23.  PMLA,  XLIV 
(1),  13  (under  Dickson). 

Breton  Lais.  CBEL,  I,  151;  V,  116.  MAN,  p.  783;  S,  I,  1006;  S,  II, 
1106;  S,  III,  1209;  S,  IV,  1301;  S,  VI,  1490;  S,  VII,  1607; 
S,  VIII,  1704.  R  &  O,  p.  381.  Bossuat,  pp.  144-46. 

Chevalere  Assigne.  CBEL,  I,  146;  V,  115.  MAN,  p.  777;  S,  II,  1104; 
S,  III,  1208;  S,  IV,  1299;  S,  V,  1387;  S,  VI,  1489;  S,  VII, 
1605;  S,  VIII,  1703.  R  &  O,  p.  374.  Bossuat,  pp.  86-88;  S, 
p.  34  (Chevalier  au  Cygne  et  de  Godfrey  de  Bouillon). 

Cleges  (Sir).  CBEL,  I,  158;  V,  116.  MAN,  p.  787;  S,  III,  1210;  IV, 
1302;  V,  1390;  VI,  1492;  VIII,  1706.  R  &  O,  p.  388.  PMLA, 
LIII,  (S)  1233. 

Degare  {Sir).  CBEL,  I,  153;  V,  116.  MAN,  p.  784;  S,  II,  1106;  S, 
III,  1209;  S,  IV,  1302;  S,  V,  1389;  S,  VI,  1490;  S,  VII,  1607; 

347 


S,  VIII,  1705.  R  &  O,  p.  382f.  PMLA,  XLVI  (S),  1347;  L  (S), 
1250;LXXI(2),  132. 

Degrevant  (Sir).  CBEL,  I,  158;  V,  116.  MAN,  p.  785;  S,  I,  1006;  S, 
II,  1107;  S,  III,  1209;  S,  VI,  1491.  R  &  O,  p.  384. 

Earl  of  Toulous.  CBEL,  I,  153.  MAN,  p.  784;  S,  I,  1006;  S,  II,  1107; 
S,  III,  1209;  S,  VI,  1490.  R  &  O,  p.  383. 

Eger  (Sir),  Sir  Grime  and  Greysteel.  CBEL,  I,  160;  V,  117.  MAN, 
S,  V,  1382  (No.  762);  S,  VI,  1484;  S,  VII,  1600.  PMLA, 
XLVI  (1),  1346;  XLVIII  (S),  1311;  LII  (S),  1237. 

Eglamour  of  Artois  (Sir).  CBEL,  I,  157.  MAN,  p.  781;  S,  I,  1005; 
S,  III,  1208;  S,  IV,  1301;  S,  VI,  1490.  R  &  O,  p.  379. 

Emare.  CBEL,  I,  152.  MAN,  p.  783;  S,  III,  1209;  S,  V,  1389;  S,  VI, 
1490;  S,  VIII,  1705.  R  &  O,  p.  382. 

Florence  of  Rome  (Le  Bone).  CBEL,  I,  158;  MAN,  p.  782;  S,  II, 
1106;  S,  III,  1209;  S,  IV,  1301;  S,  VI,  1490.  R  &  O,  p.  380. 
Bossuat,  p.  127. 

Floris  and  Blaunchefiur .  CBEL,  I,  153f.  MAN,  p.  785;  S,  I,  1006; 
S,  II,  1106;  S,  III,  1209;  S,  IV,  1302;  S,  V,  1389;  S,  VIII,  607; 
S,  IX,  1898.  R  &  O,  p.  384.  Bossuat,  pp.  125-27;  S,  p.  39. 

Freine,  Lai  le.  CBEL,  I,  151f.  MAN,  p.  783;  S,  II,  1106;  S,  III,  1209; 
S,  IV,  1301;  S,  V,  1389;  S,  VIII,  1705.  R  &  O,  p.  381.  Bossuat, 
p.  143;  S,  p.  37. 

Gamely n,  Tale  of.  CBEL,  I,  151;  V,  116.  MAN,  p.  766;  S,  III, 
1205;  S,  IV,  1296;  S,  V,  1383;  S,  VI,  1485;  S,  VIII,  1700; 
S,  IX,  1892.  R  &  O,  p.  359.  PMLA,  LXVII  (3),  25. 

Generydes.  CBEL,  I,  159.  MAN,  p.  785;  S,  III,  1209;  S,  VI,  1491. 
1392-95. 
R  &  O,  p.  384f. 

Gowther  (Sir).  CBEL,  I,  153;  V,  116.  MAN,  p.  784;  S,  I,  1006;  S,  II, 
1106;  III,  1209;  S,  V,  1389;  S,  VI,  1490;  S,  VII,  1607;  S, 
VIII,  1705;  S,  IX,  1898.  R  &  O,  p.  383.  Bossuat,  pp.  131f; 
p.  145  (No.  1542). 

Guy  of  Warwick.  CBEL,  I,  149;  V,  115.  MAN,  p.  764f.;  S,  I,  1003; 
S,  II,  1101;  S,  III,  1204;  S,  IV,  1296;  S,  V,  1382;  S,  VI,  1484; 
S,  VII,  1601;  S,  VIII,  1699;  S,  IX,  1892.  R  &  O,  p.  355f. 
Bossuat,  p.  128f.  PMLA,  XLVI  (S),  1346. 

348 


Havelok,  Lay  of.  CBEL,  I,  148f.;  V,  115.  MAN,  p.  763;  S,  I,  1003; 
S,  II,  1100;  S,  III,  1204;  S,  IV,  1296;  S,  V,  1382;  S,  VI,  1484; 
S,  VII,  1600;  S,  VIII,  1699;  S,  IX,  1892.  R  &  O,  p.  354f. 
Bossuat,  p.  53  (Cf.  No.  523);  S,  p.  145  (No.  1544).  PMLA, 
LXXII  (2),  194. 

Horn  (King).  CBEL,  I,  147;  V,  115.  MAN,  p.  762f;  S,  I,  1003;  S,  II, 
1100;  S,  III,  1204;  S,  IV,  1295;  S,  V,  1382;  S,  VI,  1484;  S, 
VII,  1600;  S,  VIII,  1699;  S,  IX,  1892.  R  &  O,  p.  352f.  Bossuat, 
p.  53.  PMLA,  XLV  (1),  24;  XLVI  (S),  1346;  LXXIII  (2), 
152. 

Horn  Childe  and  Maiden  Rimnild.  CBEL,  I,  148.  MAN,  p.  763;  S,  II, 
1100;  S,  III,  1204;  S,  IV,  1296;  S,  VI,  1484.  R  &  O,  p.  353. 
Bossuat,  p.  53  (No.  524). 

Ipomadon.  CBEL,  I,  155.  MAN,  p.  785;  S,  I,   1006;  S,  II,   1107;  S, 

III,  1210;  S,  IV,  1302;  S,  V,  1389;  S,  VI,  1491;  S,  VII,  1607; 
S,  VIII,  1705;  S.  IX,  1898.  R  &  O,  p.  385.  Bossuat,  p.  11  Of. 

humbras  (Sir).  CBEL,  I,  156f.  MAN,  p.  781;  S,  I,  1005;  S,  III,  1208; 
S,  IV,  1301;  S,  VIII,  1704,  R  &  O,  p.  379.  PMLA,  XLVIII, 
(S),  1311. 

King  of  Tars.  CBEL,  I,  154;  V,  116.  MAN,  p.  782;  S,  II,  1106;  S,  III, 
1209;  S,  VIII,  1704.  IX,  1898.  R  &  O,  p.  380.  PMLA,  LVI 
(S),  1226;  LVIII,  (S,  Pt.  2),  1205;  LXIV  (2),  21. 

Knight  of  Curtesy.  CBEL,  I,  160.  MAN,  p.  787;  S,  I,  1006;  S,  III, 
1210;  S,  V,  1390.  R  &  O,  p.  387.  Bossuat,  p.  lllf.  (No.  1161- 
76). 

Lai  or  Lay.  See  Freine,  Havelok. 

Miscellaneous  Romances.  CBEL,  I,  153-160;  V,  116f.  MAN,  784-788. 

Non-Cyclic  Romances.  CBEL,  I,  147-160;  V,  115-16. 

Octovian.  CBEL,  I,  156.  MAN,  p.  782;  S,  II,  1106;  S,  III,  1208;  S, 

IV,  1301;  S,  V,  1388;  S,  VI,  1490;  S,  VII,  1607.  R  &  O,  p. 
379f.;  Bossuat,  p.  39.  PMLA,  LXX  (2),  128. 

Orfeo  (Sir).  CBEL,  I,  15 If;  V,  116.  MAN,  p.  783;  S,  I,  1006;  S,  II, 
1106;  S,  III,  1209;  S,  IV,  1301;  S,  V,  1389,  S,  VII,  1607;  S, 
VIII,  1705.  R  &  O,  p.  381.  PMLA,  LI,  (S),  1225;  LXXIV  (2), 
121. 

Parthenope  of  Blois.  CBEL,  I,  159;  V,  116f.  MAN,  p.  785;  S,  II, 
1107;  S,  III,  1209;  S,  IV,  1302;  S,  VI,  1491;  S,  VIII,  1705. 
R  &  O,  p.  385.  Bossuat,  p.  130f;  S,  p.  39.  PMLA,  XLIII  (1), 
14;  LX  (S,  2),  1206;  LXI  (S),  1289;  LXIII  (S,  2),  34. 

349 


Reinbrun.  MAN,  S,  IX,  1782,  1892  (under  Guy  of  Warwick) 

Richard  Coer  de  Lyon.  CBEL,  I,  150;  V,  115.  MAN,  p.  786;  S,  I, 
1006;  S,  II,  1107;  S,  III,  1210;  S,  IV,  1302;  S,  V,  1389;  S,  VI, 
1491;  S,  VII,  1608;  S,  VIII,  1706;  S,  IX,  1898.  R  &  O,  p.  386. 
PMLA,  LXI  (S),  1237;  LXXI  (2),  132. 

Roberd  of  Cisyle.  CBEL,  I,  157;  V,  116.  MAN,  p.  788;  S,  I,  1006; 
S,  III,  1210;  S,  V,  1390;  S,  VI,  1492;  S,  IX,  1899.  R  &  O, 
p.  388. 

Roswall  and  Lillian.  CBEL,  I,  160. 

Seven  Sages  of  Rome.  CBEL,  I,  155;  V,  116.  MAN,  p.  792:  S,  I, 
1007;  S,II,  1109;  S,  III,  1211;  S,  IV,  1302;  S,  V,  1391;  S,  VI, 
1492;  S,  VII,  1609;  S,  VIII,  1707.  R  &  O,  p.  399f.  Bossuat, 
p.  132f.;  S,  p.  40  {Sept  Sages  de  Rome).  PMLA,  LII  (S),  1238. 

Sir.  SeeAmadace,  Beues,  Cleges,  Degare,  Degrevant,  Eger,  Eglamour, 
Gowther,  lsumbras,  Orfeo,  Torrent,  Triamour. 

Squyr  of  Lowe  Degre.  CBEL,  I,  159f.  MAN,  p.  786;  S,  III,  1210; 
S,  V,  1389;  S,  VI,  1491.  R  &  O,  p.  385. 

Torrent  of  Portyngale  (Sir).  CBEL,  I,  159.  MAN,  p.  782;  S,  II,  1105; 
S,  III,  1208;  S,  VI,  1490.  R  &  O,  p.  379. 

Triamour  (Sir).  CBEL,  I,  159.  MAN,  p.  782;  S,  III,  1208;  S,  V,  1389; 
S,  VI,  1490;  S,  VIII,  1704.  R  &  O,  p.  380. 

William  of  Palerne.  CBEL,  I,  156.  MAN,  p.  765;  S,  I,  1003;  S,  II, 
1101;  S,  III,  1204f.;  S,  IV,  1296;  S,  V,  1382;  S,  VI,  1484; 
S,  VII,  1601;  S,  VIII,  1699;  S,  IX,  1783.  PMLA,  XLII  (1), 
16;  XLIII,  16.  Bossuat,  p.  129f.  S,  p.  39. 


350 


Dat£  Due 

d 

M»..»Oue-  - 

Returned 

Due 

Returned 

MAY  1  3  m 

8 

i 

- 

- 



■ 

1 

1 

i 

i 

1 

* 

c 

■ 

. 

D- 

SZl.03 


3  1BLB  D4ia^  &B57 


I  V b-\  u*&Jr        III 


k±>'  \ 


Date   Due 


KET 





— 

— 


— 


— 


j^Wi:  mm  ■       ^ .  "■