David M. Dougherty and Eugene B. Barnes, eds. Le «Galien» de
Cheltenham. Purdue University Monographs in Romance
Languages 7. Amsterdam: John Benjamins B.V., 1981. Pp. xxxvi
Cette publication a fait l'objet de commentaires et comptes-
rendus de longueur et d'opinion très diverses auxquels nous
renvoyons le lecteur.1
Pour pénétrer clairement l'ensemble de la question, celui-ci
serait avisé de consulter d'abord l'étude que Dembowski a fait
paraître il y a quelque temps déjà dans cette revue2
et ensuite, ensuite
seulement, de lire l'introduction de Dougherty et Barnes. Celle-ci,
contrairement à ce qu'en dit Régnier dans son compte-rendu, ne
nous a pas toujours paru très claire.
En ce qui concerne la datation du manuscrit, Gaston Paris ne
déclare pas que celui-ci a été exécuté vers la fin du quinzième
comme l'affirment les éditeurs (p. x), mais seulement au
G. Di Stefano, Le Moyen français,
s.d., p. 114-15; W. W. Kibler,
58 (1983): 1033-35; G. Roques, Zeitschrift für romanische Philologie
100 (1984): 191-93, ont une opinion très défavorable de cette publication. Selon
une autre perspective
, plus nuancée, "l'édition a besoin d'un sérieux
époussetage", C. Régnier, Cahiers de civilisation médiévale ,
27 (1984): 374-77.
D. D. R. Owen, Medium Aevum
52 (1983): 323 et, de façon plus approfondie,
P. F. Dembowski, Romance Philology
38 (1984-85): 537-42, présentent un
jugement plus favorable. Nous remercions l'Institut de Recherche et d'Histoire
des Textes qui nous a communiqué la liste des comptes-rendus.
10 (1983-84): 83-98.
«Le Roman de la geste de Monglane,» Romania
12 (1883): 5.
Miquet / Dougherty & Barnes's Le «Galien» 65
texte comme le fait remarquer Dembowski) était dénoté Gualïen et
resiouir était transcrit resjouïr.
Comme pour toute édition de textes, la critique peut ne pas
être d'accord sur l'opportunité de telle note ou de telle entrée dans le
glossaire. On corrigera les erreurs ou omissions relevées notamment
par Régnier et Dembowski.
Notons quelques erreurs qui se sont glissées dans les notes
aux pages ix-xxxvi (pp. 151 à 156 de l'édition):
•n. 10, ajouter: pp. 69-71
•n. 12, Levy ne parle pas de "1290" mais de la "fin du XIIIe,"
lire 'p. ix' et non 'pp. viii ss. de cet ouvrage.'
•n. 14, il n'y a pas de pages numérotées en chiffres romains,
l'ouvrage commence p. 7.
•n. 15, les Epopées françaises, 2e éd., II, 143: ne comporte pas
la citation en question.
•n. 30, Gautier, II, et non III, 446
•n. 36, "pp. 197-203": ces pages ne renvoient pas aux Notes,
mais plutôt à la Table des noms propres, ou d'ailleurs
il n'est pas fait mention "de l'auteur et du scribe."
Pour les coquilles:
•p. xv: pallié, et non pallié à
•p. xxiv [-(i)er]: XCII et non LCVII; ajouter: CXXXIX
•p. xxv [-on] : XXXI, et non XXI
•p. 155, n. 59: identiquement à, et non que
•p. 158, n. 19 ss.: public, et non publique
•p. 160, n. 1532: capitales, et non capitaux
•p. 161, n. 2907 ss.: pour, et non pour-
•p. 163, n. 4002: vis-à-vis de, et non vis-à-vis
Voici donc une publication assez contestée. Elle nous semble
cependant présenter une introduction intéressante, un texte
suffisamment authentique et suffisamment commenté pour constituer
66 Olifant / Vol. 12, No. 1 / Spring 1987
dans son ensemble une présentation actuellement adéquate d'un texte
encore trop peu connu. Qui ira plus loin?
Carleton University, Ottawa
Ruth H. Webber
Martin de Riquer. Chanson de Roland / Cantar de Roldán y el
Roncesvalles navarro. Biblioteca Filológica I. Barcelona: El Festín
de Esopo, 1983. Pp. 403.
Any new publication by Martín de Riquer, one of Spain's
most erudite scholars, deserves careful scrutiny. He will principally
be remembered by readers of Olifant for his important and extremely
useful Les Chansons de geste françaises (2nd ed., Paris, 1968)
together with La leyenda del graal y temas épicos medievales
(Madrid, 1968). In Riquer's edition of the Chanson de Roland, the
French text occupies the left side of the page accompanied by a
Spanish prose translation on the right, with notes filling the
The text itself is based on the Oxford manuscript with
prudent emendations and lacunae filled in to make it more readily
comprehensible, following the models of Roncaglia and Segre. A
couple of examples involving difficult readings will illustrate the
point. The Bédier text is used for comparison. Lines 603-4:
Ço dist Marsilies: «Qu'en parlereient...
Cunseill n'est proz dunt hume...
Ço dist Marsilies: «Qu'en parlereiens mais?
Cunseill n'est proz dunt hum afiez n'ait.
Webber / Riquer's Cantar de Roldán 67
Esp..., icil fut filz.. .Burdel...
Esperveris, icil fut filz Burel,
celui ocist Engelers de Burdel.
Riquer also indicates the points at which Segre introduces laisses
from other manuscripts, principally V4, to fill out omissions in O.
These passages are then reproduced and translated in Appendix I.
The Spanish translation, which aspires to be as literal as possible,
by moments has almost an epic ring. A number of sub-titles have
been added to help the reader orient himself in the story.
The numerous footnotes are meant to explain the text rather
than to provide linguistic information. For the latter purpose Riquer
refers the reader to the edition of Cortés Vázquez (Salamanca,
1975). The notes include discussions of difficult words and terms,
protagonists and their names, geography and place names, all with
pertinent bibliography. The notes are concise syntheses of the topics
in question, and they include frequent references to comparative
material, Spanish in particular. Among the longer notes, for
example, is one based on "sun nevuld" (216), which presents the
matter of Roland's parentage, and another discussing the Turoldus
problem (4002). The question of the Baligant episode (2609), on the
other hand, is given short shrift.
The introductory material is made up of six brief sections:
historical background, content of the Roland, theories concerning its
origin, prehistory, transmission, and extant texts, followed by a
short statement about this edition and a minimal basic bibliography.
Riquer's presentation is intelligently conceived and well balanced
and, despite its brevity, does not differ greatly in its essentials from
that of Les Chansons de geste françaises. For matters like dating,
which are not taken up either here or in the notes, he refers to his
own earlier book.
68 Olifant / Vol. 12, No. 1 /Spring 1987
Appendix II contains an edition of the Roncesvalles fragment
entitled "El Roncesvalles navarro," which is also accompanied by a
Spanish prose translation. The transcription follows Horrent rather
than Menéndez Pidal in places where the manuscript is illegible. The
notes supply all the information necessary for a full appreciation of
the problems surrounding this important text and its relationship to
the French epic. It is interesting to note that Riquer repeats here his
debatable theory to the effect that Charlemagne's wiping off Oliver's
head means it had been decapitated and that the motif was borrowed
from the earlier Spanish epic, the Siete infantes de Lara. The
decapitation is made explicit in the latter, but not in the
Roncesvalles, and the formulas in question reappear in the
romancero in scenes where there is a dead or dying victim,
suggesting that they had wide oral circulation and were not
necessarily a direct borrowing.
In sum, Riquer's edition of the Chanson de Roland was not
intended to make a new contribution to Roland studies, but rather to
offer in the most usable form possible an introductory text for
Spanish students of the French epic. As such it cannot be surpassed.
Ruth H. Webber
Joan B. Williamson
Norman Daniel. Heroes and Saracens: An Interpretation of the
Chansons de Geste. Edinburgh: Edinburgh University Press, 1984.
Pp. vii + 349. (Distributed in the U.S. by Columbia University
In Heroes and Saracens, Daniel examines the unofficial
attitudes of Christians to Islam and the Arabs, as seen in Old French
epics. Daniel considers that since the official Christian attitude found
its expression in philosophical and political tracts and decrees, the
Williamson / Daniel's Heroes & Saracens 69
views expressed in the epics were those of the poets who made
propaganda for the Western chivalry that paid for them and that
provided their subject matter. Daniel felicitously covers a wide range
of epics, but, curiously in a study of attitudes in fictional writings,
dismisses the second Crusade Cycle as "too fantastic to be
considered close to actual events."
Daniel's point of departure is that fictions are inhabited by
fictional people who may or may not be intended as the real thing.
Perhaps the genesis of this book lies in a point Daniel made in his
Arabs and Medieval Europe (London: Longman, 1975, p. 53) that
no medieval writer seems to have realized that "Saracen" has no
Arab equivalent form. What then, asks Daniel in Heroes and
Saracens, do the poets mean by this term? He divides the book into
three parts. In the first he describes the Saracen of the epic and in the
second he examines the nature of the epic Saracen gods. Part Three
consists of two chapters: corroborative evidence from a limited
number of historical sources to give an idea of how the layman of
the time viewed Islam and the Arabs; and the author's own
conclusions to the book. Part of the excitement this book generates
derives from the contributions that Daniel makes from his viewpoint
as a historian and expert on the medieval Arab world. Therefore, the
historical overview of what the layman thought is overly brief and
does not fully meet the expectations raised by the book on pp. 18-
19. One must hope that Daniel will indeed write the volume that he
recognizes such a topic demands.
While noting that the later epics are of necessity different
from the early ones because they have different time-frames, Daniel
establishes an epic convention of the Saracen and his gods. Taken
together, the epics have created stories which are "homogeneous not
only in idea, mood and feeling, but in the framework and
background within which they are composed" (p. 8). These
elements all "combine to create the characteristic convention of
behavior and situation" (p. 8).
This convention, Daniel shows, has nothing to do with
reality. Saracen society, dress, courtly pastimes, attitudes to war,
women and love, and the like, as portrayed in the epics, do not
70 Olifant / Vol. 12, No. 1 / Spring 1987
correspond to the reality of Islam, but reflect Christian models, with
occasional exaggeration deliberately introduced for comic effect. The
poets are not necessarily ignorant of historical knowledge, but show
a total disregard for veracity in their literary portrayal of the Saracen.
This disregard for historical accuracy is even more marked in
the portrayal of the Saracen gods. Daniel posits that the poets could
not have been ignorant of the basic monotheism of Islam and that the
multiplicity they lent the pagan gods must have been deliberate. The
epic Saracen's idolatry also flies in the face of Moslem prohibition
of images. The epic polytheistic parade of Saracen gods is a literary
convention, concludes Daniel. In support of his assertion Daniel
cites the case of Joinville, who mentions neither idols nor gods and
who gives no sign of ever having heard of the Tervagant
convention. The epic convention of the Saracen gods was arrived at
by the deliberate intent to create a religion as different as possible
from the Christian model.
This is a book that is worth reading. It is also easy to read.
Rather general in its conclusions and with little that is startlingly
new, Daniel's book is written in a style that is easy to follow, and
the focus of its presentation provides for a clearer understanding of
the epic Saracen and his gods.
Joan B. Williamson
Long Island University
Joan B. Williamson
Carla Bozzolo and Ezio Ornato. Pour une histoire du livre manuscrit
au moyen âge: Trois essais de codicologie quantitative. Paris:
Editions du CNRS, 1983. (1st ed., 1980).
Of interest to all medievalists, this book deserves to be
brought to the attention of readers of Olifant. The second edition,
reviewed here, differs from the first only by the addition of further
Williamson / Bozzolo and Ornato's Trois essais 71
notes, a list, and an index of the various charts and diagrams
included, and by the updating of the early bibliography and listing
of manuscripts the authors mention.
Bozzolo and Ornato situate medieval manuscript book
production within its economic, historical and cultural context,
concentrating on two areas of basic research: causes and processes
governing the greater or lesser number of volumes produced; and
causes and processes governing production techniques and the
practices of individual tradesmen. The three essays of the title deal
respectively with the production of the manuscript book in Northern
France; with the make-up of gatherings and problems of imposition,
the procedure that originated in the context of manuscript book
production to become the norm in the production of printed books;
and with folio dimensions, which became increasingly standardized
as the fifteenth century advanced.
It is the first essay, on manuscript production in Northern
France, that is of particular interest to the student of literature. There
is a wealth of detail in this essay. The authors have analysed data
drawn from medieval library inventories, contracts, accountings and
receipts, in terms of numbers and dimensions of book pages and the
number of lines per page. In this way they have established average
prices for manuscript books in Northern France. Decorated and
luxury volumes were eliminated from the study since the quantity of
decoration and its artistic quality vary to such an extent from book to
book that meaningful comparison is impossible.
Since different currencies were used and the values allocated
to these varied constantly, particularly with the severe inflation of
the fourteenth and fifteenth centuries, prices have been stated based
on the value of a royal Chancery notary or secretary's daily
remuneration (with the cost of room and board factored in). A
second-hand book in circulation cost six and a half days'
emoluments, while the commission of a new book required double
this salary. Bozzolo and Ornato note that a biliophile would not own
many books. Assuming that a secretary were to devote all his
emoluments over a twenty-year career to the purchase of books (a
totally improbable situation), he would only accumulate some 500
72 Olifant / Vol. 12, No. 1 / Spring 1987
volumes. In reality, totals were much lower. The highest recorded is
of 198 books belonging to one Nicolas de Blaye. The richest people
did not necessarily have the most books, wealth being translated,
not into the number of books owned, but into the quality of their
The medieval hand-written book was a costly object.
Although paper cost thirteen times less than skins, this does not
mean that a book made out of paper cost thirteen times less than one
using parchment. The price of the materials was a minor part of the
cost of production. The salary of the scribe ranged from one and a
half to five and a half times the price of the parchment required for
the volume. The high cost resulted from the slowness of execution.
In the fifteenth century, for which records are available, the average
day's output for a copyist was three pages. This, of course, explains
the rising tendency in Chancery circles for authors with good scribal
training to save money by penning their own works.
The hand-written book remained usable for several
centuries, with volumes copied in the thirteenth century still being
used in the fifteenth. The hardiness of the product and the reduced
price of the second-hand copy compared with that of a new book
explain why books were handed from possessor to possessor. This
habit tended to limit new book production. Since the total number of
books one owner might possess was limited, book production could
only be augmented by an increase in the number of potential
owners. Thus, while the creation of new works would lead to new
books, demographics, or an increase in the book-owning
population, was the real controlling factor of book production.
Since a book was a costly object and was not seen as a
necessity of life, hard times, such as those of war, pestilence or
economic difficulties, exerted a limiting influence on book
production. Thus Bozzolo and Ornato have noted a reduction of
book production between the ninth and twelfth centuries, followed
first by a moderate then by a marked increase in the thirteenth. Then
came a decrease during the fourteenth century, ending with the
fifteenth when book production was only slightly superior to the
Williamson / Bozzolo and Ornato's Trois essais 73
output of the thirteenth. These fluctuations correspond to changes in
the society that produced the books. While the long-term increase in
production is explained partly by cultural changes when learning
became more important, the hand-written book was immediately
sacrificed in times of crisis.
The intelligently programmed quantitative approach based on
solid documentation makes this text a valuable contribution to the
study of the book in the Middle Ages.
Joan B. Williamson
Long Island University
1988 Kalamazoo Medieval Institute Conference
Call for Papers
Joan B. Williamson will be organizing the 1988 Société Rencesvals
sessions at the Kalamazoo Medieval Institute Conference. Please
send abstracts and proposals to her by October 5, 1987, so that Dr.
Gründler's deadlines may be met. Please contact her early with
abstracts or proposals so that the exact number of sessions required
may be established in due time.
Address: 3, Washington Square Village, 7-1
New York, New York 10012
74 Olifant / Vol. 12, No. 1 / Spring 1987
Patricia E. Black received two "Faculty Development Awards" from
California State University-Chico to pursue research. She is
continuing the study of the Chanson de Guillaume, what close
examination of the heroine's role reveals about the society, literary
values, and the composition of the poem, and comparison of other
works in an effort to define clearly the place of the epic poem in the
development of the roman. She has read papers on the Chanson de
Guillaume at MLA in 1986 and at Kalamazoo in both 1986 and
Catherine Jones, who is currently writing a dissertation on Hervis
de Mes with Douglas Kelly at Wisconsin, presented a paper entitled
"The Mechanism of Dispersion in Late Epic Composition: Hervis de
Mes" at Kalamazoo this spring.
Jean-Louis Picherit has been promoted to full professor at the
University of Wyoming. He is currently working on Le Gab
d'Olivier in French literature. His recent publications include an
article in Le Moyen Age 91 (1985) entitled, "Le Livre de la
Prod'hommie de l'homme et le Livre de Prudence de Christine de
Pisan: chronologie, structure, composition," and "Formes et
Fonctions de la matière proverbiale dans le Songe du vieil pelerin de
Philippe de Mézières," in Actes du Colloque Internationale
Université McGill, La Locution, Montréal, 1984.
Don A. Monson received an NEH Fellowship for College Teachers
and is a Visiting Fellow at Clare Hall, Cambridge. He is writing on
"Andreas Capellanus, Scholasticism and the Courtly Tradition."
Forthcoming publications include "L'idéologie du lai de Lanval" in
Le Moyen Age, and "La surenchère chez Chrétien de Troyes" in
Poétique. His "Jaufre Rudel et l'amour lointain; les origines d'une
légende," has just appeared in Romania 106 (1985): 36-56.