Also in the Variorum Collected Studies Series:

CHARLES BURNETT VARIORUM COLLECTED STUDIES SERIES

Magic and Divination in the Middle Ages
Texts and Techniques in the Islamic and Christian Worlds

MARIE-THERE SE D' ALVERNY (Ed. Charles Burnett)

Pensee medievale en Occident
Theologie, Magie et Autres Textes des XIIe-XIIIe Siecles

MARIE-THERE SE D' ALVERNY (Ed. Charles Burnett)

La connaissance de l'Islam dans l'Occident medieval

MARIE-THERE SE D' ALVERNY (Ed. Charles Burnett)

Etudes sur le symbolisme de la Sagesse et l'iconographie medievale Arabic into Latin in the Middle Ages
MENSO FOLKERTS
The Development of Mathematics in Medieval Europe
The Arabs, Euclid, Regiomontanus

PAUL KUNITZSCH
Stars and Numbers
Astronomy and Mathematics in the Medieval Arab and Western Worlds

RAYMOND MERCIER
Studies on the Transmission of Medieval Mathematical Astronomy

MENSO FOLKERTS
Essays on Early Medieval Mathematics
The Latin Tradition

LUIS GARCiA-BALLESTER (Eds Jon Arrizabalaga,

Montserrat Cabre, Lluis Cifuentes, and Fernando Salmon)
Galen and Galenism
Theory and Medical Practice from Antiquity to the European Renaissance

DONALD R. HILL (Ed. David King)

Studies in Medieval Islamic Technology
From Philo to al-Jazari - from Alexandria to Diyar Bakr

DANIELLE JACQUART
La science medicale occidentale entre deux renaissances (XIIe s.-XVe s.)
 Charles Burnett

Arabic into Latin in the Middle Ages

The Translators and their Intellectual

and Social Context

ASHGATE
VARIORUM
This edition © 2009 by Charles Burnett

Charles Burnett has asserted his moral right under the Copyright, Designs and Patents
Act, 1988, to be identified as the author of this work.

Published in the Variorum Collected Studies Series by

Ashgate Publishing Limited Ashgate Publishing Company

CONTENTS
Wey Court East Suite 420
Union Road 101 Cherry Street
Famham, Surrey Burlington, VT 05401--4405
Preface vii
GU97PT USA
England
Acknowledgements viii

I Ashgate website: http://www.ashgate.com King Ptolemy and A1chandreus the philosopher: the
I
earliest texts on the astrolabe and Arabic astrology at
ISBN 978-0-7546-5943-3 Fleury, Micy and Chartres 329-68
Annals o/Science 55. London, 1998. Addendum, ibid. 57, 187
British Library Cataloguing in Publication Data 2000
Burnett, Charles (Charles S. E)
Arabic into Latin in the Middle Ages : the translators and their intellectual and
social context. II Physics before the Physics: early translations from
- (Variorum collected studies series) Arabic of texts concerning nature in MSS British
1. Arabic language - Translating into Latin - History - To 1500. Library, Additional 22719 and Cotton Galba E IV 53-109
2. Islamic learning and scholarship - Europe - History - To 1500. Medioevo 27_ Padua, 2002
3. Arabic literature - Translations into Latin - History and criticism.
I. Title 11. Series
478'.02927'094'0902--dc22 III Adelard of Bath and the Arabs 89-107
Rencontres de cultures dans la philosophie medievale, eds
ISBN 978-0-7546-5943-3 J. Hamesse and M Fattori (Rencontres de philosophie
medievale 1). Louvain-Ia-Neuve and Cassino: 1nstitut d'etudes
Library of Congress Control Number: 2009904444 medievales, 1990

IV Antioch as a link between Arabic and Latin culture in

the twelfth and thirteenth centuries 1-78
Occident et Proche-Orient: contacts scientifiques au temps
des croisades, eds 1. Draelants, A. Tihon, and B. van den Abeele.
Turnhout: Brepols, 2000

V 'Magister Iohannes Hispalensis et Limiensis' and

Qusta ibn Luqa's De differentia spiritus et animae:
a Portuguese contribution to the arts curriculum? 221-67
VARIORUM COLLECTED STUDIES SERIES CS939
Mediaevalia, Textos e Estudos 7-8. Porto, 1995

VI John of Seville and John of Spain, a mise au point 59-78

c
J,j Mixed Sources
Product group from well-managed
foresu and other controlled sources Printed and bound in Great Britain by
Bulletin de philosophie medievale 44. Turnhout, 2002

FSC =~~::s~:a:~h:C~~~ii2412 TJ International Ltd, Padstow, Cornwall

 vi CONTENTS

VII The coherence of the Arabic-Latin translation

program in Toledo in the twelfth century 249-88
Science in Context 14. Cambridge, 2001
PREFACE
VIII Michael Scot and the transmission of scientific culture
from Toledo to Bologna via the court of Frederick 11
Hohenstaufen 101-26
Micrologus 2. Turnhout, 1994 This collection of articles on the transmission of Arabic learning to Europe will
concentrate on the identity ofthe Latin translators and the context in which they
IX Master Theodore, Frederick lI's philosopher were working. Future collections will deal with the transmission of the works of
225-85
Federico II e le nuove culture, Atti del XXXI Convegno specific Arabic authors, particular subject matters, and translation methods. The
storico internazionale, Todi, 9-12 ottobre 1994. Spoleto, articles are arranged in roughly chronological order of translator, beginning with
1995
the earliest known translations from Arabic at the end of the tenth century (I),
progressing through eleventh-century translations made in Southern Italy (11),
Addenda and Corrigenda 1-5 translators working in Sicily and the Principality of Antioch at the beginning of
the twelfth century (Ill and IV), the first ofthe twelfth-century Iberian translators
Index Manuscriptorum 1-4 (V), the beginnings and development of 'professional' translation activity in
Toledo (VI, VII), and the transfer of this activity from Toledo to Frederick lI's
Index Nominum 1-8 entourage in Italy and Sicily in the thirteenth century (VIII, IX). Most of the
articles include editions of texts that either illustrate the style and character of
the translator or provide the source material for his bio-bibliography. A list of
This volume contains viii + 412 pages addenda and corrigenda comes at the end of the volume, to which reference is
made by means of asterisks in the relevant places in the articles. The volume
concludes with indexes of manuscripts and proper names.

CHARLES BURNETT
London
31 March 2009.
PUBLISHER'S NOTE

The articles in this volume, as in all others in the Variorum Collected Studies Series
ha~e. not bee~ given a new, continuous pagination. In order to avoid confusion, and t~
facllztat~ th~lr use where th~se same studies have been referred to elsewhere, the origi-
nal pagmatzon has been mamtained wherever possible.
Each article has been given a Roman number in order ofappearance, as listed in the
Contents. This number is repeated on each page and is quoted in the index entries.
Asterisks in the ma'!tins are to alert the reader to additional information supplied at
the end of the volume m the Addenda and Corrigenda.

ACKNOWLEDGEMENTS
I would like to thank the following publishers and institutions for permission
to reproduce the articles included in this volume: Taylor & Francis, London
(http://www.informaworld.com) (for article I); 11 Poligrafo casa editrice, Padua
(11); ~repols Publishers, Turnhout (Ill, IV and VI); the Funda<;ao Eng. Antonio de
~lmeld~, Porto (V); Cambridge University Press, Cambridge (VII); the Societa
mtemazlOnale ~er 10 studio del medioevo latino, Certosa del Galluzzo (VIII);
and the FondazlOne Centro italiano di studi sull'alto medioevo, Spoleto (IX).
I am grateful to the National Library of Scotland, Edinburgh, and the Berlin
Staatsbibliothek PreuBischer Kulturbesitz for permission to reproduce additional
illustrations in IV and V.
I
King Ptolemy and Alchandreus the Philosopher: The Earliest Texts on
the Astrolabe and Arabic Astrology at Fleury, Micy and Chartres
Sunimary
This paper reassesses the importance of the Benedictine monasteries of St Benoit *
of Fleury and St Mesmin of Micy (both on the outskirts of Orleans), and the
Cathedral of Chartres for the early diffusion of Arabic learning concerning the
astrolabe, and it relates this diffusion to that of the judicial astrology of
, Alchandreus philosophus' and the astronomical tables of the Preceptum canonis
Ptolomei. Evidence is given for the fact that already, by the turn of the millennium,
the elements were in place for a corpus of a new, mathematically based, practical
science of the stars, consisting of works partly of Arabic and partly of Greek
inspiration. This corpus was progressively revised and inspired, in turn, further
translations from Arabic, until it reached its most mature form in the mid-twelfth
century. Until recently, scholarship has tended to concentrate on the cathedral
schools of North-east France and Lotharingia, and the monastic schools of South
Germany, and to see Gerbert d'Aurillac, the Archbishop of Reims, as the pivotal
figure. While these schools were undoubtedly important in the diffusion of the new
science of the stars in the eleventh century, it is argued that a key role in the initial
stages of the diffusion was played by the interrelated centres of Fleury, Micy and
Chartres at the beginning of the century, and Gerbert may not have contributed
as much as has been believed. Additional sections are devoted to the authorities
in this corpus, 'King Ptolemy' and 'Alchandreus'. A reworking of the Arabic
material on the construction of the astrolabe by Ascelin of Augsburg was copied
into this corpus. The text has previously been known from only one manuscript;
a new edition, from five manuscripts, is provided here, together with a translation
and commentary.
Contents
I: The Early Collection at Fleury and Micy .................................. 330
11: The Early Collection at Chartres.............. .......... ..... ......... ...... 334
Ill: The corpus of works on the science of the stars, c.960-1050............. 338
IV: King Ptolemy ................................................................. 340
V: Alchandreus Philosophus .................................................... 343
VI: Ascelin of Augsburg, Compositio Astrolabii .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 343
An earlier version of this paper was read at the Universitc d'cre at Chartres in July 1996 and was published
(without notes) as 'L' Astronomie aChartres au temps de I'cvCque Fulbert', in Le temps de Fulbert, Actes de
I'Universire d'cre, Chartres (1996), 91-103. I am grateful for the help of Silke Ackermann, David Juste,
Koenraad Van Cleempoel, Paul Kunitzsch, David King, Emmanuel Poulle, Anna Somfai, and Sabina
BlumemOder.
I I
330
I: The Early Collection at Fleury and Micy
Towards the end of the tenth century, in Catalonia, the southemmost region of
the French domain, certain Arabic texts (or, at least, doctrines) on the astrolabe and
astrology became known to the Latins. I At about the same time, Hindu-Arabic
numer~l~ and the game of chess (also of Arabic provenance) appeared in the region;2
and thIS IS also the place and time to which modem scholarship attributes the earliest
known Latin astrolabe (i.e. the 'Destombes astrolabe', Institut du monde arabe,
Paris). This interest coincides with, and cannot be independent from, the renewal of
interest in the science of the stars in Arabic al-Andalus with the school of Maslama
al-Majrili, who drew up a star table in 978. 3 The first Latin texts include (1) crude
Latin versions of Arabic material on the construction and use of the astrolabe (De
mensura astrolapsus h", and Sententie astrolabii J') and (2) elaborate Latin
remaniements of the same material (De mensura astrolabii h' and De utilitatibus
astrolabii J),4 which are accompanied in some manuscripts by a prologue in the same
elaborate style (beginning 'Ad intimas summe philosophie ... ') setting out the
co~tents of an astronomical corpus which contains these works, some of which the
wnter of the prologue says he has translated from Arabic. 5 That Lupitus or Sunifred
an ~rchdeacon in Barcelona, was responsible for the crude translations and/or th~
Latm elaborations, as Millas Vallicrosa claims,6 is disputed. At any rate, both the
crude texts and the remaniements were probably the work of a single Catalan scholar
who knew Arabic.
For this reason, among others, the attribution of De utilitatibus astrolabii to
Gerbert d'Aurillac is unlikely.7 When he was still very young, Gerbert spent 967-70
1 The two original Arabic texts identified so far are (1) a portion of al-Khwiirizmfs treatise on the use
of the astrolabe, of which a section of the third part of the Sententie astrolabii is a literal translation' Paul
Kunitz~ch,.' AI-Kh~iirizmi as ~ sou~ce for the Sententie astrolabii', in From Deferent to Equant: A V~/ume
of S~udles In. the HIStory of SCle~ce In the Ancient and Medieval Near East in Honor of E. S. Kennedy, eds
Davld A. Kmg and George Sa~lba, Annals of the New York Academy of Sciences, vol. 500 (New York,
1987), 227-36; a~d (2) the ?pemng?f chapters 2 and 3 of Pto lemy's Planisphere (in Paris, BNF, lat. 7412,
If. 1~r-v only),. edited and discussed m Paul Kunitzsch, 'Fragments of Ptolemy's Planisphaerium in an early
Latm translatIOn', Centaurus, 36 (1993), 97-101.
2 H. M. Gamer, 'The earliest evidence of chess in Western literature: the Einsiedeln Verses' Speculum
29 (1954), 734-50. ' ,
3 On the De~tombes astrolabe see the exhaustive treatment in the articles collected in Physis, 32 (1995).
For the connectIOns between the school of Maslama and the texts and instruments that became known in
Catalonia, see the authoritative survey by Paul Kunitzsch, 'Les relations scientifiques entre I'Occident et
le ~onde arabe a l'epoque de Gerbert', in Gerbert reuropeen: Actes du colloque If Aurillac, 4-7 juin 1996,
edited by N. Charbonnel and J.-E. lung (Aurillac, 1997), 193-203.
H
4 The compe.ndia for.these texts (h 1', h', J, etc.) are derived from those devised by N. Bubnov, Gerberti
opera mathematlca (Berlm, 1899), xv, and summarized in W. Bergmann, Innovationen im Quadrivium des
10. und 11. Jahrhunderts. Sudholfs Archiv, Beiheft 26 (Stuttgart, 1985),227-8. See also section Ill.
5 J.-M. Millas Vallicrosa, Assaig lfhistoria de les ideesfisiques i matematiques a la Catalunya medieval
(Barcelona, 1931), I, 275: 'nolui fidi interpretis officium precavere sed potius ut in Arabico habetur
simpliciter interpretari '. The prologue' Ad intimas ... ' occurs in Vati~an, Reg.la;. 1661; Munich, Clm 560
and 14689 (all three. with connections to Fleury; Arno Borst, Astrolab und Klosterreform an der
Jahrtaus~ndwende .[Heldelberg, 1989], 68); and London, British Library, Royal 15.B.lX. The first three
manuscnpts contam the texts of stages Band C (as listed in section Ill); the fourth manuscript includes
some. texts of stage B/C attached a~ a late! date to Hermann the Lame's corpus (stage D). In the case of
Mumch, Clm 14689, the prologue Immediately precedes De utilitatibus astrolabii.
6 Millas Vallicrosa (note 5), 195; G. Feliu i Montfort, 'Sunifred, Anomenat L1obet, ardiaca de
Barcelona (finals del segle X)', in II Col.loqui If Historia del Monaquisme Catala Sant Joan de les Abadesses
1970 (Abadia de Poblet, 1972), I, 51-63. ' ,
7 Emmanuel Poulle, 'L'Astronomie de Gerbert', in Gerberto: Scienza, storia e mito. Atti del Gerberti
~y,!,posiu"! (Bobbio 25-27 luglio 1983) (Bobbi~, 1985),.597-617 (615-16), considers the prologue 'Ad
mtlm~s ... to be by Ge~bert, but at the same time demes that Gerbert made any translations from the
ArabiC. Borst (note 5) wntes of' Gerberts Schiiler' as the author of De utilitatibus astrolabii s. v. index and
Paul Kunitzsch (private communication) has now withdrawn his statement in Glossar' der arabi;chen
The Astrolabe and Arabic Astrology in France 331
in the cathedral school at Vich as a student studying the quadrivium; this was before
Maslama had drawn up his star table. When Gerbert went on to become schoolmaster
(scolasticus) and eventually Archbishop of Reims, he did not cut off his connections
with Catalonia. It was from there that he sent the letters in 984 to Lupitus in
Barcelona asking for a work De astrologia,8 and to Bishop Bonfilius in Gerona asking
for a copy of' Joseph Sapiens' (or' Joseph Hispanus ') 'about the multiplication and
division of numbers '.9 But the works that can firmly be attributed to Gerbert are on
arithmetic, geometry (i.e. the abacus), and music, and on sundials and the
construction of a celestial sphere (not an astrolabe) for the teaching of the positions
of the stars and planets,IO and this is confirmed by the glowing account of his teaching
methods in the quadrivium given by his student and biographer, Richer, which again
1I
makes no mention of the astrolabe, but only of various spheres and hemispheres.
One would expect to see the origin of the new astronomical corpus in Catalonia. 12
Some 20 km from Vich, on the stratafrancisca connecting Barcelona with Toulouse,
is the Benedictine abbey of Ripoll. Manuscript no. 225 of the monastery contains
chapters from the two Latin remaniements, De utilitatibus astrolabii and De mensura
astrolabii, the Sententie astrolabii, and several further chapters on the use of the
astrolabe for measuring objects, which are the first chapters of a text known as
Geometria incerti auctoris. This collection (henceforth, the 'Early Collection ')
includes Arabic terms and a bilingual text on the climes. 13 None of these works have
attributions to Gerbert, and a more likely interpretation would be that they are copies
of works produced locally.
FachausdrUcke in der mittelalterlichen europiJischen Astrolabliteratur (Gottingen, 1983), 479, that it is by
Gerbert. From the remark that some stars' are not or are hardly seen in our region', Bergmann (note 4),
152, draws the inference that the text must have been written in a northerly region (i.e. Reims) rather than
in Catalonia, but this evidence does not appear sufficiently conclusive for locating the text.
8 Gerbert, Die Briefsammlung Gerberts von Reims, edited by F. Weigle (Berlin, 1966), letter no. 24, 46-?
'De astrologia' could refer to works on either astronomy or astrology. David Juste has suggested that It
might be the astrological Liber Alchandrei philosophi (on which, see section Ill). Poulle, however, considers
Lupitus' De astrologia to be the texts on the astrolabe: 'La Iitterature astrolabique latine jusqu'au XIII"
siecle', Physis, 32 (1995), 227-37. Gerbert's interest in the science of the stars is witnessed by his attempt
to get a copy of' M. Manlius (v.l. Manilius) de astrologia': letter no. 130 (to Rainard of Bobbio in 988),
in Weigle (ibid.), 157-8.
9 Gerbert, Briefsammlung, letter no. 25, in ibid., 48. Apparently the same work is described as 'De
multiplicatione et divisione numerorum Iibellum a loseph Ispano editum', in Briefsammlung (to the Abbot
of Aurillac in 984), in ibid., letter no. 17,40.
10 The instruments that Gerbert describes are most clearly identified by Poulle (note 7).
11 Richer, apud Bubnov (note 4), 378-81. The only possible reference in Gerbert's authentic work to the
use of an astrolabe is in his Regulae de numerorum abaci rationibus, in Bubnov (ibid.), 8, where the
'geometricus radius secundum inclinationem et erectionem' (for which calculations on the abacus are
useful) is the •alidade'; the same term is used by Constantine, the dedicatee of the Regulae, and (without
'geometricus') in De utilitatibus astrolabii; see Bubnov (ibid.), 123, who also gives this as a reason for
suggesting that it is very probable that the use of an astrolabe was known to Gerbert (p. 109). Poulle,
however (private communication), considers that the reference here could equally well be to a nocturlabe.
The ambiguity of 'radius geometricus' is well illustrated by Adelard of Bath's use of it in. Regule .u,baci:
•Abacus is a "radius geometricus" quia cum ad multa pertineat, maxime per hoc geometnce subtllitates
nobis iIIuminantur'. Here it clearly means a 'geometrical ray' and has nothing to do with a rod.
12 Poulle (note 7), 612, is of the same opinion: '11 est difficile de croire que la science arabe ait pu 6clore
en Lorraine avant d'avoir ete connue en Catalogne'.
13 This collection is what A. van de Vyver describes as 'la premiere forme (ou groupe) du plus ancien
recueil de traites sur l'astrolabe', which he identifies in ten manuscripts; 'Les Premieres Traductions
Latines (xe_xie s.) de traites arabes sur l'astrolabe', in Premier congres international de geographie historique
(Brussels, 1931), n, 266-90 (275-7); see also Millas Vallicrosa (note 5), 175,290-1, pIt. IX. P. Tannery
questions Bubnov's claim of the priority of one arrangement of chapters of the Geometria incerti auctoris
(Bubnov's D) over the other (Bubnov's E, which is found in the manuscripts discussed in this paper); P.
Tannery and A. Clerval, 'Une Correspondance d'ecolatres du lIe siecle', Notices et extraits des mss de la
Bibliotheque national et autres bibliotheques, 36 (1900), 542.
I
332
Arno Borst, however, considers that the Early Collection was originally put
together in the Benedictine abbey of St-Benoit of Fleury, which was on the outskirts
~f Orleans. Its abbot at the time was Abbo of Fleury, whose fame for teaching the
hberal arts was equal to that of Gerbert, and whose extant works show his interest
in the computus and in astronomy. 14 Abbo had been a student at Reims in 972-5 at
the time when Gerbert had recently arrived there from Catalonia. But Abbo and
~erbert were political rivals, the former upholding the rights of monks, the latter, of
blshop~. Abbo had become schoolmaster at St-Benoit of Fleury in 975 and, after a
short tIme as schoolmaster in Ramsey in the East Midlands of England (986-8),
became Abbot of Fleury in 988. He remained there until his death in 1004. Fleury and
Reims were the two most highly sought-after schools in the late tenth century.
The reason why Borst considers that Fleury may have been the place where the
Early Collection was compiled is that three of its texts, with the same variant
readings, were used in a 'Lehrbuch' which arrived at an early date in Reichenau,15
and a copy of the Early Collection was made in St Mesmin de Micy, 16 an abbey also
outside Orleans. It happens that the man who was likely responsible for the Micy
copy, Constantine of Fleury, was in Fleury at the same time as the future Abbot of
Reichenau, Bern of Priim. Bern was Hermann the Lame's mentor, and had been
attracted to Fleury because of the reputation of Abbo. The collection would have
been compiled c.995 in the circle of Constantine l7 and would have reached Ripoll
from Fleury.
An alternative scenario for the contact between Constantine and Bern could be
proposed (see below). But the priority ofFleury to Ripoll is also suggested by the date
assigned to the Ripoll manuscript in recent scholarship, i.e. the early-to-mid-eleventh
century rather than late tenth century, as Millas Vallicrosa had thought. 18 This later
date would correspond with the time of the abbacy of Oliva (1008-46), a cultured
man with a wide acquaintance, under whose abbacy many manuscripts were copied
at Ripoll.19 It was in his time that Arabic notes were added to a manuscript of
Boethius' Arithmetica (presumably by a Mozarab),20 and a native of Barcelona who
had been a monk at Ripoll transferred to Fleury.21 What is undoubted is the close
relation of the Ripoll manuscript with the Fleury manuscripts (whichever way the
:: A. va~ de Vyver, 'Le~ <Euvres inedites d'Abbon de Fleury', Revue Benedictine, 47 (1935), 125-69.
That IS, the ~anuscnpt, unknown to van de Vyver, of which the surviving fragment (now MS
Konstanz, StadtarchlV, Fragmentensammlung, Mappe 2, Umschlag 8 Stiick 7) is edited by Borst (note 5)
I~~ "
16 V~tican, Reg. lat. 598. For the relation between the fragments from Reichenau and the Micy
manuscnpt, see Borst (note 5), 69.
17 Ibid.
18 See especially the judgement of Jean Vezin, as reported in G. Beaujouan, W. M. Stevens, and A. J.
Turner, 'Les Apocryphes mathematiques de Gerbert', in Gerberto: Scienza storia e mito Atti del Gerberti
SY":fosium (Bobbio 2~-27 lug/~o 1983) (Bobbio, ,19~5), 65fH!. ' ,
. On t~e manusc~pts.ofR:lpoll, ~~. Beer, Die Handschriften des Klosters Santa Maria de Ripoll,'
Sitz. d. ~lS. A~. d., WlSS. In Wle~, phll.-hlSt. Klasse, 155/3 (1907) and 158/2 (1908), translated into Spanish
by Bamds y Giol, Los manuscnts del Monastir de Santa Maria de Ripoll', Boletin de la Real Academia
de Buenas Letras, Barcelona, 5 (1909-10), 137-70, 230-78, 299-365, 492-520. The main hand of the
m~nuscript of rhetorical ~nd mus!cal texts - Ripo1l42 - is very similar to the hand that wrote ff.l-64 of
Ripoll.225. Anna Somfal has pomted out to me that Oliva was apparently responsible for an 'edited
collection' of texts on astronomy and the computus, contained in the Ripoll manuscript Vatican Reg lat
123. This collection is similar in form to, and could be seen as complementary to, the Early Collecti~n of
astrolabe texts.
20 Ripoll, MS 168, ff.42r, 62r-v, 91r.
21 Borst (note 5), 75. For letters between Oliva and Gauzlin (Abbo's successor as Abbot of Fleury) see
Paris, BNF, lat. 2858, ff.66v-70r.
I
The Astrolabe and Arabic Astrology in France 333
influence spread). For van de Vyver,22 the Ripoll manuscript has only two
'congeneres' among the ten manuscripts that include the Early Collection: Leiden,
Scaliger 38, which includes a calendar of Abbo of Fleury,23 and the manuscript copied
in Micy, already mentioned. This leads us to take a closer look at the role of
Constantine.
Constantine of Fleury had been brought up in the abbey at Fleury and had the
prospect of becoming abbot there in 987. He was unjustly rejected, and found a refuge
in Reims with Gerbert. From 988 to 996 he was Dean at St-Mesmin de Micy, and
from 1011 to his death in 1020/1 he was abbot of the same monastery. He had a
reputation for his expertise in music. To him Gerbert addressed his principal
mathematical works, including his letter on how to construct a celestial sphere
(Epistola de sphaera). Constantine was Gerbert's intimate friend and confidant, and
could perhaps be termed his literary executor. For the oldest manuscript of Gerbert's
collection of letters and other writings (Leiden, Voss. lat. Q 54) was written by a
certain' Stabilis' in Micy, who, if he is not the same as 'Constantinus' ,24 must have
made the collection under Constantine's supervision. The terms of the dedications
suggest that Constantine had asked Gerbert for further explanations of arithmetical
problems and for an explanation of the construction of a heavenly sphere. He
deferred to Gerbert's arithmetical skills as Gerbert had deferred to Constantine's
musical skills.25 But the reason why Constantine asked Gerbert's advice may probably
have been because he already had the corpus of astrolabe texts which he wanted to
supplement with other mathematical information. There is no evidence that
Constantine inherited these texts from Gerbert.
Evidence, on the contrary, that Constantine was the promoter of interest in the
astrolabe comes from the fact that the same 'Stabilis' asked persistently for an
explanation of the construction of the astrolabe from Ascelin of Augsburg. 26 Ascelin,
too, is not attested as a student of Gerbert, but based his De mensura astrolabii on the
Latin elaboration that Millas Vallicrosa thought was made by Lupitus, i.e. h'. This
22 Van de Vyver (note 13), 275-6. The close relation of these three manuscripts is confirmed by
Berfflann (note 4), 88.
3 Leiden, Scaliger 38, ff.7v-13r. . .
24 I am of the opinion that 'Stabilis' is 'Constantine '. The names mean the same thmg; Ascelin of
Augusburg plays on the meaning of the' Stabilis' at length in the dedication of his text on the astrolabe
to Stabilis in MS Oxford, Corpus Christi College, 283; see also section VI: 'Since, then, I h~ve
considered that you, a beloved man, being of unsullied honesty, by your own name and at the same tl~e
truly by your character and equally by the word of truth [are] "Stable/Constant" ... '. K. F. Wemer, ID
discussing 'Stabilis', 'Zur Uberlieferung der Briefe Gerberts von Aurillac', Deutsches Archiv, 17 (1961),
91-144 (100-01), was apparently unaware of the Oxford manuscript and thus did not use the evidence of
the preface. 'Stabilis' occurs in no archival document, but only as the signature of the compiler of Gerbert's
letter-collection in the Leiden MS and in Ascelin's dedication. In both cases he is a described as a 'monk',
which would correspond to Constantine's position before he became abbot. It is quite possible that
'Stabilis' is Constantine's own nom-de-plume, a situation that has ample parallels in the period, e.g.
Comestor/Manducator and Ocreatus/Hosatus: Charles Bumett, 'Ocreatus" in Vestigia Mathematica,
edited by M. Folkerts and J. Hogendijk (Amsterdam, 1993),69-78. However, for the argument, whether
one is talking about one scholar of the astrolabe in Micy or two does not matter. The important thing is
that the interest was centred at Micy.
2S Gerbert, Brie/sammlung, in Bubnov (note 4), letter no. 92, 6: 'vel in musica perdiscenda vel in his,
quae fiunt ex organis, per Constantinum Floriacensem supplere curabo. Est enim nobilis scolasticus,
adprime eruditus, michique in amicitia conjunctissimus'.
26 See section VI. Paul KUnitzsch, Typen von Stemverzeichnissen in astronomischen Handschriften des
10.-14. Jh. (Wiesbaden, 1966), 27, identifies Ascelin with the well-known Adalbero, Bishop of Laon
(977-1030), who was also known as 'Ascelinus' or •Azzelinus', and had been a friend of Gerbert's at
Reims; Lexikon des Mittelalters, s.v. Adalbero and Riche, Gerbert d' Aurillac: le pape de fan mil (Paris,
' 19~7), 54. However, a connection with Augsburg does not feature in the biography of the Bishop of Laon,
which is well documented.
I I
334
latter work was also the basis for Hermann the Lame's well-known writing on De
mensura astrolabii, which he composed avowedly as a book to precede De utilitatibus
astrolabii to form a new astrolabe corpus which would be completed by another text
of his own composition - De horologio viatorum - on the portable sundial. If K. F.
Werner's arguments that Hermann was sent when 7 years old to the cathedral school
of Augsburg, rather than directly to Reichenau (where he is firmly attested only at the
age of 30), then it is possible that Ascelin was the teacher of Hermann.27 Once again,
Gerbert does not come into the picture, but nor is it necessary to propose St-Benoit
of Fleury as an intermediate stage in the diffusion of astrolabe texts, since Ascelin, not
Bern, becomes the teacher of Hermann, and Constantine is the sole person necessary
to connect all these students of the astrolabe.
That Constantine owed his interest in, and knowledge of, astrolabe texts to his
upbringing in St-Benoit of Fleury, however, is quite possible. For the abbot in this
period, Abbo of Fleury, was skilled in the mathematical arts and appears to have
possessed a copy of Preceptum canonis Ptolomei (as is discussed below). Moreover,
the earliest manuscript of another corpus of texts put together using Arabic sources,
the astrological corpus of Liber Alchandrei philosophi, includes a work by Abbo. 28
The importance of Constantine/Stabilis of Fleury, monk and abbot of Micy, for
the advance of astrolabe studies is shown by another factor, i.e. the evidence of the
knowledge of the Early Collection in Chartres.
11: The Early Collection at Chartres
Probably the first words of Arabic in Latin poetry that a Western audience would
have heard occur in four lines of verse by Fulbert, Bishop of Chartres (1006-28):
Abdebaran Tauro, Geminis Menkeque Rigelque,
Frons et Calbalazet prestant insigne Leoni;
Scorpie, Galbalagrab tua sit, Capricornie, Deneb,
Tu, Batanalhaut, Piscibus es satis una duobus. 29
These verses were not meant to entertain, but rather were an aid to remembering
the fixed stars that were added to the rete of the astrolabe. These stars, all with Arabic
names, were some of the brightest stars in the sky, and enabled one to locate the
celestial coordinates. Fulbert's verses on them have the same function as his verses on
the libra and its parts, and on the computus. But in the case of the astrolabe-verses we
are fortunate to have the notes he took in preparation for writing them. In these the
27 Werner (note 24), 10Cr7.
28 A. van de Vyver, 'Les Plus Anciennes Traductions latines medievales (xe_xie s.) de traites
d'astronomie et d'astrologie', Osiris, 1 (1936),658-89 (677-8): Abbo of Fleury's text' Studiosis astrologiae
primo sciendum est ... ' (written 978) is included in the Liber Alchandrei corpus in Paris, BNF, lat. 17868
(end of tenth century): 'peut-etre parce que cet ensemble de textes a passe par Fleury'. The Liber
Alchandrei corpus also accompanies De utilitatibus astrolabii in Munich, Clm 560, and London, British
Library, Add. MS 17808, and excerpts from it were copied into a miscellany belonging to Adhemar of
Chabannes (Leiden, Voss. Lat. 0 15), who also copied Paris, BNF, lat. 7231, a collection of rhetorical
works almost identical to that in a Fleury manuscript of the turn of the tentHleventh centuries (Paris,
BNF, lat. 7696); Jean Vezin, 'Leofnoth: un scribe anglais it Saint-Benoit-sur-Loire', Codices manuscripti,
3 (1977), 109-20. For the fullest account of the manuscripts and contents of the Alchandrean corpus up
to now, see David Juste, 'Les doctrines astrologiques du Liber Alchandrei', in Occident et Proche Orient
au temps des Croisades: Traductions et contacts scientifiques entre 1000 et 1300, edited by A. Tihon, B. van
den Abeele, and I. Draelants (Louvain-Ia-Neuve, 1998), in press.
29 F. Behrends (ed.), The Letters and Poems of Fulbert of Chartres (Oxford, 1976), xxvii-xxviii, 260-1.
M. McVaugh and F. Behrends, 'Fulbert of Chartres' notes on Arabic astronomy', Manuscripta, 15 (1971),
172-7, see no reason to doubt the authorship, but Borst (note 5), 75, prefers to see a student of Fulbert
as the author.
The Astrolabe and Arabic Astrology in France 335
signs of the zodiac (in Latin) are given in one column, and the presence or absence
of a fixed star in that sign is listed in a parallel column. Moreover, he adds a list of
28 Arabic terms used in connection with the astrolabe and gives their Latin
equivalents or explanations. He has taken these terms and most of the Latin
equivalents from Sententie astrolabii. But some of the translations differ from those
found in the manuscripts in Sententie astrolabii in such a way that the expert on
astrolabe literature, Paul Kunitzsch, would see in Fulbert someone who was
experienced in handling texts on the astrolabe. 3O If so, he is one of the earliest scholars
to show such a skill. The mnemonic poem and preliminary notes on the fixed stars are
taken from the chapter De stellis horarum,31 which is related to Sententie. Both these
texts were in the Early Collection. A clue to how Fulbert became aware of this comes
from Chartres 214; for this is one of the manuscripts of the Early Collection.
The immediate problem is, of course, that this manuscript does not exist any more
for it was destroyed during the Second World War. It is possible that fragments might
be found among the debris that was preserved after the disastrous bomb, but this
paper is based on the assumption that no such fragments exist. Instead, what we have
is a detailed catalogue entry, and descriptions, collations of individual works, and a
few photographs of isolated folios made by specialists who had access to the
manuscript before its destruction. 32 These make it abundantly clear that the
manuscript was very close in its contents to one manuscript contained within Oxford,
Corpus Christi College, 283. This latter manuscript, then, can usually give a good idea
of the contents of the former. 33
In addition to the Early Collection the Chartrian manuscript includes further
works. Two are texts that were sent to the Abbey of Micy. The first is Gerbert's
Epistola de sphera, which also occurs in the manuscript written by 'Stabilis'. 34 The
second is the work of Ascelin of Augsburg, which occurs only in this manuscript, its
sister manuscript in Corpus Christi College, and, without its preface, in Avranches,
235 and British Library, Add. MS 17808. 35 The Micy connection is undoubted, and
the presence of two texts with Micy dedications in Chartres 214 and its sister
manuscript suggests very strongly that the exemplar of the Chartres and Corpus
Christi manuscripts came to Chartres from Micy. The evidence of these manuscripts
corroborates the assumption by Wemer and Borst that Fulbert derived his knowledge
of Sententie astrolabii from Fleury via Micy,36 and makes it unnecessary to propose
30 Kunitzsch (note 7), s.v. FC.
31 Millils Vallicrosa (note 5), 292.
32 These include the description in Catalogue general des manuscrits des bibliotheques pub/iques de
France: Departements Xl, Chartres, edited by H. Omont et al. (Paris, 1890); and Y. Delaporte, Les
manuscrits enlumines de la Bib/iotlJeque de Chartres (Chartres, 1929). A photograph of f.3Or (an astrolabe
plate) is reproduced in van de Vyver (note 13), pit. 11, and as Figure 3 (below); and another folio is
reproduced in H. Michel, 'Les Tubes optiques avant le telescope', Ciel et terre, Bulletin de la Societe beige
tfastronomie, de meteorologie et de physique du globe, 70 (1954),175-84 (177 has an image of a man using
the nocturlabe, whose text is the same as that of Avranches, Bibliotheque municipale, 235, f.32v).
33 The relationship between the two manuscripts has been discussed in Charles.Burnet~, 'The contents
and affiliation of the scientific manuscripts written at, or brought to, Chartres In the time of John of
Salisbury', in The World of John of Salisbury, edited by M. Wilks (Oxford, 1984), 127-60. The next ~losest
manuscript is Avranches, Bibliotheque municipale, 235, which, however, shows a later stage In the
transmission of astrolabe texts since it includes Hermann the Lame's De mensura astrolabii.
34 MS Leiden, Voss. Lat. Q 54. In the dedication in these two manuscripts, 'papa' has ~ added to
Gerbert's name and 'abbot' to that of Constantine, although the letter had probably been wntten many
years before either scholar had these titles. This indicates that these copies or their exemplar must have been
made after 1011 (when Constantine became abbot).
3S See section VI, which also includes fragments from two further manuscripts.
36 Werner (note 24), 103-5; Borst (note 5), 75-6.
I I
336
that he was a student of Gerbert, for which the evidence is meagre. That the
connections between Chartres and Micy were close is further suggested by the fact
that after Fulbert's death, his student Si go composed the verses to accompany the
pictures on the tumulus leaflets which were to be inserted into the cathedral
martyrology, but the pictures were painted by Andreas of MicyY
There is, however, another text, additional to the astrolabe texts, shared by the
Chartres and Corpus Christi College manuscripts, that points again towards Fleury,
i.e. the Preceptum canonis Ptolomei. This is a set of astronomical tables and the rules
for their use based on the Handy Tables of Ptolemy and other Greek texts and put
together in Rome in AD 535. 38 This was the only set of astronomical tables available
in the Latin West before Petrus Alfonsi and Adelard of Bath's translations of the
tables of al-KhwarizmI in the early twelfth century. Its history in the early Middle
Ages is obscure, but the earliest Latin manuscript - British Museum, Harley 2506,
written c.AD 1000 - or its archetype, can, with some certainty, be associated with the
circle of Abbo of Fleury, whether at Ramsey or, as is more likely, at Fleury.39 David
Pingree, the editor of the text, shows that what he calls' the Chartrian manuscripts'
most probably derive from another copy (jJ) of the exemplar of the Harley
manuscript, and to the same family belong the Corpus Christi College manuscript
and Avranches 235. It is quite plausible that the manuscript from which the scribe of
Chartres 214 took the astrolabe material also contained the Preceptum canonis and
that this whole corpus came from Fleury via Micy.
Preceptum canonis was apparently known to the author of De utilitatibus
astrolabii, who refers to it three times in the text,40 and specifically mentions it as a
complement to the astrolabe texts in the prologue accompanying the Latin
remaniements of the astrolabe texts. 41 Thus, the Preceptum can be ranged with the
original texts on which the author of the Latin remaniements based his work. After
all, it would make sense to have a set of astronomical tables alongside the texts on the
construction and use of the astrolabe. 42
The problem with this interpretation of the evidence so far is that neither Chartres
214 nor its sister manuscript from Corpus Christi College are contemporary with
Fulbert. One might object that the Chartres manuscript could have come to the
cathedral city at any time between the early eleventh and late twelfth centuries, and
that it might have arrived from one of a number of different centres. Palaeography
37 Behrends (note 29), xliii.
38 D. Pingree, 'The Preceptum Canonis Ptolomei', in Rencontres de cultures dans la philosophie
mid;evale (Louvain-Ia-Neuve-Cassino, 1990), 355-75; Preceptum canonis Ptolomei, edited by D. Pingree
(Louvain-Ia-Neuve, 1997).
39 Van de Vyver (note 14), 142-4, would see the corrector of the Harley manuscript as 'peut-etre
Abbon " and shows that the Aratea in the manuscript incorporates corrections that a second hand had
added to British Museum, Harley 647; he concludes that Harley 2506 'aura ete apporte en Angleterre par
Abbon, tout comme' Harley 647 'ou bien ete donne posterieurement a Ramsey au cours des echanges de
mss. entre les deux abbayes, au sujet desquels nous sommes informes'.
40 Pingree (note 38), 374: the references are in 1.2 (Bubnov [note 4], 116), XIII.2 (135), and XVIII.l
(139). On each occasion the citation is to 'canones Ptolomaei' and the contexts suggest the Preeeptum.
41 Millas Vallicrosa (note 5), 275: 'Quod opusculum, cuilibet velit complicare libro sive eanonibus
Ptolomei sive Vitruvio ... ' (my italics).
42 Preceptum canon is accompanies De utilitatibus astrolabii in its original form (i.e. not in the
incomplete and interpolated form in which it appears in the Early Collection) in Avranches, 235. The fact
that it also accompanies the Early Collection in Oxford, Corpus Christi College, 283, and Chartres 214
lends weight to Borsfs theory that the Early Collection was put together in Fleury. Another indication of
the travelling together of the Preeeptum Canonis and the Early Collection is contained in Harley 2506,
where the name' Abbo' has been replaced by that of' Bemon', (fr. 3Ov-32), i.e. the same Bern of Priim who
Borst considers the man who took the Lehrbueh (based on the Early Collection) to Reichenau.
The Astrolabe and Arabic Astrology in France 337
is an inexact science, but the twelfth-century date of Chartres 214 (if it was an integral
rather than a composite manuscript) is confirmed by the inclusion in it of one further
work: the translation of al-KhwarizmI tables by Adelard of Bath, made c.1l26.
However, that this must be considered separately from the rest of the manuscript is
suggested, once again, by comparison with the Corpus Christi College manuscript.
For, although the Corpus Christi manuscript also has the tables of al-KhwarizmI (as
is discussed below43), they are in a separate codex from that which contains the
astrolabe texts and Preceptum canon is. The latter codex dates from the late eleventh
century. Thus, I would suggest that the common exemplar of these sister manuscripts
was in existence at least before the end of the eleventh century.
The strong likelihood that MS 214 was in Chartres already at least in the second
quarter of the twelfth century is indicated by the fact that one, and perhaps two, of
its texts were copied into the 'library of the seven liberal arts' (Heptateuchon) of
Fulbert's spiritual successor in the cathedral school, i.e. Thierry of Chartres. 44 The
penultimate text in the second volume of Heptateuchon (which fortunately survives in
photographs) is the canons and tables of Preceptum canonis Ptolomei. Since MS 214
has been destroyed it is no longer possible to compare its text with Thierry's to
establish which was copied from which. But there is another text for which such a
comparison is possible, and that is the astronomical tables of al-KhwarizmI in the
version of Adelard of Bath, which is the last item both in MS 214 and in the
Heptateuchon. Fortunately, MS 214 was used by the editors of this text early this
century, and it is quite clear that Thierry took his copy from MS 214.45
It is likely, too, that the Corpus Christi manuscript was also in Chartres. 46 Here
the story is more complicated because, as mentioned above, the relevant portion of
the manuscript consists of two codices. The second codex consists of al-Khwarizmfs
astronomical tables, but not in the version of Adelard. Rather, it is a version
originally by Petrus Alfonsi, the Arabic-educated Jew from Huesca in Aragon (then
part of the Arabic kingdom of Zaragoza), who converted to Christianity in 1106 and
is afterwards attested in the West Midlands of England. 47 But to the archetype of the
Corpus Christi manuscript some chapters from the Adelard of Bath version had been
added. Then, after the text had been copied, another hand added to the tables values
from two sources: the Adelard of Bath version of the tables·ofal-KhwarizmI, and the
Tables of Toledo. But the corrections from the first source (Adelard of Bath) are so
close to the text in MS 214 that they could have been copied from it. Moreover, it is
likely that the Tables of Toledo were also in Chartres in the mid-twelfth century. For
another destroyed manuscript of that century, Chartres 213, contains a collection of
43 See next paragraph.
44 MSS Chartres, Bibliotheque municipale, 497, 498. A detailed description of the second volume (on
the quadrivium) is in Burnett (note 33),142-3,147. It is curious that Tbierry does not include any astrolabe
text, or the associated material from Geometria meerti auetoris, in Heptateuehon. The agrimensorial
material that is included is, rather, the selection known as Liber podismi, which makes no mention of the
astrolabe, but may be the same as the' podismus' mentioned by Ralph of Liege in a letter to Ragimbold
of Cologne; Tannery (note 13), 531. It is not clear whether this text existed at Chartres at the time of
RalEh's correspondence.
S The relationship between the Latin manuscripts and the variants is in H. Suter, A. Bjembo, and R.
Bestbom, Die astronomisehen Ta/eln des MuI)amme~ ibn Miisii al-Khwllrizmr in der Bearbeitung des
Maslama ibn A!rmed al-Madjfi(r und der lateinisehe Ubersetzung des Athelhard von Bath (Copenhagen,
1914) and an unpublished collation by Joshua Lipton.
46 The argument of the following paragraph is set out in more detail in Burnett (note 33),130-2,146-7.
47 John Tolan, Petrus Alfonsi and His Medieval Readers (Gainesville, 1993); Estudios sobre Pedro
Alfonso de Huesca, edited by Maria JesUs Lacarra (Huesca, 1996); Charles Bumett, 'The works ofPetrus
Alfonsi: questions of authenticity', Medium Aevum, 66 (1997), 42-79.
I I
338
texts that appears, from the description in the catalogue, to be the astrological
writings of Raymond of Marseilles. Raymond is the earliest Latin scholar to use the
Tables of Toledo, which he adapted for the meridian of Marseilles in 1140. 48
My hypothesis, therefore, is that Chartres 214 is a copy of two manuscripts. The
first is a manuscript of the early Latin compilation of texts on the science of the
astrolabe (the Early Collection) to which two works had been added in Micy. This
manuscript could have come to Chartres directly from Micy after Constantine had
become Abbot of Micy (lOll), and could have been used by Fulbert to compose his
astrolabe glossary and poem on the fixed stars. This manuscript would also have been
the exemplar of the first relevant codex of the Corpus Christi manuscript, which was
copied in the late eleventh century and is in a continental hand. The second
manuscript copied into Chartres 214, however, was of a completely different nature,
date, and provenance. It contained a set of astronomical tables translated c.1126 by
Adelard of Bath, a scholar active in the West Midlands. Just as the arrival of the Micy
manuscript in Chartres in the early eleventh century must have been due to the
interest of Bishop Fulbert in the first Arabic texts concerning astronomy, so the
arrival of a manuscript representing the second wave of translations from Arabic in
the second quarter of the twelfth century was due to the desire of another great
Chartrian teacher who actively sought out new works for the teaching of the liberal
arts, i.e. Thierry of Chartres.
The contrast between the two periods is brought out well by the words of
Hermann of Carinthia, the pupil of Thierry, who went to Spain to translate Arabic
mathematical texts, one of which he dedicated to his teacher. He called himself the
'second Hermann " both honouring and implying an advance on the' first Hermann'
(i.e. Hermann the Lame) whose name had become associated with the first corpus of
astronomical texts translated from Arabic. 49 For over a century this corpus had
provided the West with a basic knowledge of practical astronomy and astrology. But
the corpus was by no means static: the constituent texts were progressively revised
and improved, partly, no doubt, as a result of greater familiarity with the use of
astronomical instruments. Above all, they provided Western scholars with a taste for
a mathematically based science of the stars which lead, in the twelfth century, to a
more thorough exploitation of the Arabs' achievements of the field.
Ill: The corpus of works on the science of the stars, c.960-1050
This table draws together the works discussed in this paper, and certain others, to
show in a schematic way the progressive elaborations of a corpus of works on the
astrolabe, astronomy, and astrology from the late tenth to the early eleventh
centuries. For simplicity's sake, associated works translated from Greek in antiquity
or of purely Latin origin (such as those of Hyginus, Aratus, Calcidius, Firmicus
Matemus, Martianus Capella, Pseudo-Hipparchus, Isidore, Bede, Helperic, etc.)
have been omitted.
(A) Crude Latin versions of Arabic material:
De mensura astrolapsus h".
Sententie astrolabii J'.
48 Burnett (note 33), 133-6, 140.
49 Hermann of Carinthia, De essentiis, edited by C. Burnett (Leiden, 1982), 346; the preface to
Hermann's translation of Ptolemy's Planisphere, dedicated to Thierry, is edited in J. L. Heiberg, Ptolemaei
0p'era astronomica minora (Leipzig, 1907), c1xxxiii--clxxxvi; see also Charles Burnett, 'Advertising the new
science of the stars circa 112~50', in Le XIIe siecie, edited by F. Gasparri (Paris, 1995), 147-57 (15~1).
The Astrolabe and Arabic Astrology in France 339
De orologio secundum alchoram J'a.
Translated/written by a Catalan author (Lupitus?).
Parallel to: Liber Alchandrei philosophiso and Preceptum canonis Ptolomei, both works
associated with Abbo of Fleury.
(B) Literary remaniements of the crude texts:
De mensura astrolabii h'.
De utilitatibus astrolabii J (19- or 21-chapter version).
The prologue 'Ad intimas '" '.51
The prologue is written by a translator from Arabic; and therefore it should be
assumed that the same Catalan scholar (Lupitus?) who was responsible for A,
also made these remaniements. 52
Parallel to: the literary remaniements of Liber Alchandrei philosophi: Quicumque nosse
desiderat ... and Proportiones competentes in astrorum industria. 53
De utilitatibus astrolabii is accompanied in two manuscripts (Avranches 235, and
British Library, Add. MS 17808; both 19-chapter versions) by Fulbert's poem on
the libra and its parts. 54 The Alchandrean corpus includes a work of Abbo in Paris,
BNF, lat. 17868. 55
(C) The Early Collection based on texts A and B, and Geometria incerti auctoris. 56
This includes the' 19-chapter' De utilitatibus astrolabii (see p. 331 above).
(D) The new corpus put together by Hermann the Lame:
De mensura astrolabii h (Hermann's own composition).
De utilitatibus astrolabii J (from B above).
De horologio viatorum hv (Hermann's own composition).
so For manuscripts of this text, see van de Vyver (note 28), 66£r7; Juste (note 28); and the old
catalogues of the Austin Friars in York: 'Argaphalon. chaldeusjbreviarium alhandrei' (M. R. James,
Fasciculus J. W. Clarke dicatus [Cambridge, 1909], no. 159, 36), and St Augustine's, Canterbury:
'Breviarium alhandredi sum(mi) astrologi et peritissimi de soia (7) qualibet ignota nollo decrete' (M. R.
James, The Ancient Libraries of Canterbury and Dover [Cambridge, 1895], 332).
SI See note 5. The authors of De utilitatibus astrolabii and the prologue' Ad intimas ... ' both use 'sollers
indagatio' and refer to 'canones Ptolomei'. The authors of De mensura astrolabii and De utilitatibus
astrolabii both use vo/vel/us for rete.
S2 Millas Vallicrosa (note 5) suggested that the author was Lupitus; Borst (note 5), 67-9, endorses this
sugf:estion.
3 For the manuscripts of these texts, see van de Vyver (note 28), 66£r7; and Juste (note 28). The
stylistic and terminological similarity between these remaniements and those of the astrolabe texts needs
exploring; van de Vyver, ibid., 677, points out the similarity between the incipit 'Quicumque nosse
desiderat legem astrorum ... ' and that of the De utilitatibus astrolabii CQuicumque astronomicae discere
peritiam disciplinae ... '), and considers that 'Quicumque nosse '" ' 'n'est peut..etre qu'un remaniement
po~terieur de [Liber Alchandrei phi/osophll tout comme le De utilitatibus astrolabii a refait en un latin
mellleur, les Sententiae Astrolabii' (p. 681); Proportiones competentes and the prologue •Ad intimas ... '
both use 'uranicus' and' intime philosophie'. The problem about associating these remaniements is that the
author of the prologue' Ad intimas ... ' tries to distance the subject he is introducing from astrology.
54 Behrends (note 29), 254-5. There are three further manuscripts of Fulbert's poem.
ss Another work that regularly accompanies these texts is the Liber de abaco of Bernelinus Junior of
Paris (written 999--1(00), which is also in Burgo de Osma 7; Munich, Bayerische Staatsbibliothek, Clm
491; Oxford, Bodleian Library, Auct. F.l.9 (first three books); Leiden, Bibliotheek der Rijksuniversiteit,
~l. 38; London, British Library, Add. MS 17808; and Paris, BNF, n.a.l. 229. Bernelinus considers
himself to be in the tradition of Gerbert.
• S6 Borst (note 5), 69; Borst speaks of drafts (' Vorlagen ') from which was made the textbook
( Lehrbuch') on the astrolabe, a fragment of which survives in MS Konstanz, Stadtarchiv, Fragmenten-
sammlung, Mappe 2, Umschlag 8, Stiick 7. Note that the chapters constituting Geometria ineerti auctoris
have been isolated rather artificially by Bubnov (note 28). Moreover, the extra chapters on measurement *
and on the astrolabe published in MilIas Vallicrosa (note 5),302-8,324-7, are in exactly the same style and
u,se the same terminology as that of Geometria incerti auctoris as edited by Bubnov. The style is extremely
Simple, without any rhetorical flourishes.
I I
340
Several pieces of evidence suggest that the early astrolabe texts accompanied texts
from the Alchandrean corpus:
(1) William of Malmesbury clearly had such a combination in mind when he wrote
in Gesta regum Anglorum, edited by W. Stubbs (London, 1887),11, c.l67, that Gerber~
'vicit scientia Ptholomeum in astrolabio, Alhandreum in astrorum interstitio lulium
Firmicum in fato'. '
(2) In MS Oxford, Bodleian Library, Ashmole 369, ff.77r-85v, Proportiones
competentes in astrorum industria is followed without a break by excerpts from
Hermann the Lame's corpus (De mensura astrolab;;, edited by J. Drecker,
'Hermannus Contractus iiber das Astrolab', /sis, 16 [1931], 20{}-19 [208-10] and De
utilitatibus astrolabii, chs VII-XVI); I owe this information to David Juste.
(3) MS Munich, Bayerische Staatsbibliothek, Clm 458 combines Proportiones
competentes in astrorum industria (ff.I-24r), and material related to Liber Alchandrei
philosophi (ff.24r-35v), with the preface of Ascelin of Augsburg's treatise on the
astrolabe (f.35v-36r) and names the whole text 'Liber ludiciorum Messahalla'.
Although this a fifteenth-century copy, it evidently reflects an early combination.
Once again, I owe this information to Juste.
That such a combination can be specifically assigned to Fleury at the time of
Abbot Abbo is suggested by the eleventh-century manuscript, Munich Bayerische
Staatsbibliothek, Clm 560. 57 One may take each text of this manuscrip{ separately.
(1) De utilitatibus astrolabii (19-chapter text), one of the original fopns (before the
Lehrbuch) of the remaniements of the primitive material (i.e. stage B above).
(2) De alio horologio and De orologio secundum alchoram (J'a), both 'primitive' texts
(stage A above).
(3) Pr~logu~ '~d intimas ... " which seems a prologue to the remaniements (stage B).
(4) JulIus FmnIcus Maternus, Mathesis, bks I and 11. The presence of this work at
Fleury is proved by the inscription of the copy in Paris, BNF, lat. 7311: 'In nomine
patris et filii et spiritus sancti et in honore pii patris Benedicti lulii Firmici Materni
lunioris Matheseos incipit liber primus'. Jean Vezin gives the evidence to show that
this manuscript was copied by the English scribe Leofnoth at St-Benoit-sur-Loire,
Fleury (where the body ofSt Benedict was preserved) in 'Leofnoth: un scribe anglais
it Saint-Benoit-sur-Loire', Codices manuscripti, 3 (1977), 109-20 (p. 118).
(5) .~ib~r ~l~handrei philosophi .. No title. The scribe writes (f.61r): 'Sequentia ad opus
JUlll FlrmICl vel (read: ut) arbltror non pertinent'.
IV: King Ptolemy
That the astrolabe was an invention of Ptolemy is clearly stated in the prologue
of the translator (Prologue' Ad intimas ... ') where we read:
Among all [the Egyptians] Ptolemy especially shone out in this discipline [i.e.
astronomy]. Just as he was more brilliant in its study, so he also handed down
to posterity this explanation [sententia] more clearly. For among the other
notable aids in this art provided by him, he invented a certain instrument which
was both most useful for learners and a mighty miracle for those looking at it.
57 For descriptions, see Bergmann (note 4), 23fr7, B. Bischoff, Die sudwestdeutschen Schreibschulen und
Bibli~tJu:ken in der Karolingerzeit, 2nd edn, 2 vols (Wiesbaden, 1960-80), I, 262; and U. Reichel, Astrologie,
Sort/leg/um, T~aumdeut",!g:.Formen von Weissagung im Mittelalter (Bochum, 1991). According to Borst
(n~te ~), 68, thiS manuscnpt IS one of the four manuscripts of the prologue' Ad intimas ... ' that' go directly
or mdlrectly back to Fleury' (gehen direkt oder indirekt auf Fleury ziiruck).
The Astrolabe and Arabic Astrology in France 341
... For the 'Wazza1cora' was obtained by a divine mind; in Latin it means 'flat
sphere', which also, by another name, is 'the astrolabe [astrolapsus] of
Ptolemy'.58
The author of this prologue does not say that he has translated a text written by
Ptolemy,59 but rather that he faithfully reproduces what he has found in an 'Arabic
source' (' Arabici fontis ... stabilem observem seriem ') and interpreted 'in a simple
way what is in the Arabic' ('ut in Arabico habetur, simpliciter interpretari').60 Nor
does he imply that Ptolemy was a king of Egypt. The attribution of the text to
Ptolemy was, rather, added to the title of the main section of the most primitive
version of the translation/compilation on the uses of the astrolabe (Sententie
astrolabii J' section C) in certain manuscripts: 'Incipiunt capitula libri horologii regis
Ptolomei', Paris, BNF, lat. 11248; 'Incipiunt capitula horologii Ptolomei', Avranches
235, f.30v; and' Horologium regis Ptolomei', Vatican, Reg. lat., 598. Consequently
the literary remaniement of Sententie astrolabii - De utilitatibus astrolabii J - in one
branch of manuscripts (e.g. Paris, BNF, lat. 14065, British Library, Add. MS 17808,
and Avranches 235) is called 'Regulae (ex libris) Ptolomei regis de compositione
astrolapsus '. The rules for the construction of the astrolabe are also attributed to
Ptolemy in the remaniement in Avranches, 235, f.68r, and British Library, Add. MS
17808, f.80r (derived from h"), which is headed 'Incipit compositio astrolapsus
secundum Ptolomeum', and begins' lubet rex Ptolomeus'.
The attribution of the texts to Ptolemy would seem, then, to be the responsibility
of later redactors and scribes, and not to the original translator/compiler of the Latin
text from Arabic sources (i.e. Lupitus?). It was these later redactors and scribes who
added the epithet 'rex'. This they probably did because Isidore of Seville called
Ptolemy a king of Alexandria in Etymologiae, Ill, 26.1: 'De institutoribus eius [se.
astronomiae]. In utraque autem lingua diversorum quidem sunt de astronomia
scripta volumina, inter quos tamen Ptolemaeus rex Alexandriae apud Graecos
praecipuus habetur. Hic etiam et canones instituit, quibus cursus astrorum
inveniatur'. This passage comes from Cassiodorus (and incidentally refers to
Preceptum canon is Ptolomel)' However, Cassiodorus is much better informed on
Ptolemy, and does not call him a king. 61 Whether the misidentification of Pto lemy the
58. Prologue' Ad intimas ... " in Millas Vallicrosa (note 5), 274: 'inter omnes precipue Ptolomeus hac
clarwt di~iplina. Qui sicut studio clarior, ita etiam hanc sententiam luculentius posteris tradidit. Nam inter
~tera .hwus artis insignia ab ipso subministrata adiumenta quoddam instrumentum et utillimum
dlscentlbus et magnum miraculum considerantibus adinvenit .... Est autem Wazzalcora divina mente
compar~ta, quod latine sonat plana spera, que etiam alio nomine astrolapsus Ptolomei'. These words are
ech~ m the first words of Hermann the Lame's description of the construction of the astrolabe: 'In
met~enda igitur subtilissimae inventionis Ptolomei waltalchora, id est plana sphaera, quam astrolabium
VOCitamus ... ', in J. Drecker (ed.), 'Hermannus Contractus iiber das Astrolab', /Sis, 16 (1931), 200-19
(2~>. For the Arabic sources that claim Ptolemy as the inventor of the astrolabe, see D. M. King, 'The
ongm ofthe astrolabe according to the medieval Islamic sources', Journal of the History of Arabic Science,
5 (1!81), 43-83~ reprinted in idem, Islamic Astronomical Instruments (Aldershot, 1987), article Ill.
be The mention of' hanc sententiam' in note 58, if it refers to the translation that follows would rather
, 'the ?~i~on' or 'explanation' of Ptolemy that someone else has written down; comp~re the use of
sententla m Walcher of Malvern's rendition of Petrus Alfonsi's 'Sententia Petri Ebrei, cognomento
Anphus, de dracone, quam dominus Walcerus prior Malvernensis ecclesie in latinam transtulit linguam'
and 0ther examples of 'sententiae' discussed in Burnett (note 47),45-6.
60
Prologus 'Ad intimas ... " in Millas Vallicrosa (note 5), 275. Kunitzsch (private communication)
suggest~ that he may have assumed that Ptolemy invented the astrolabe from his knowledge of the
authe~tlc Planisphere of Ptolemy, of which two fragments of a Latin translation accompany the astrolabe
mat;nal (note I above).
1 Cassiodorus, Institutiones, edited by R. A. B. Mynors (Oxford 1937) bk 11 155-6: 'De astronomia
vero disciplina in utraque lingua diversorum quidem sunt scripta v~lumin~; inte; quos tamen Ptolomeus
apud Graecos praecipuus habetur, qui de hac re duos codices edidit, quorum unum minorem, alterum
I I
342
astronomer with one of the rulers of Egypt is original to Isidore or predates him is,
as yet, unclear. A source of confusion for Isidore and others may have been a passage
in Martianus Capella's De nuptiis Mercurii et Philologiae, which refers to the
'surveyors of king Ptolemy' who established the rules of geometry. 62 This could have
been read as implying that Ptolemy the astronomer was responsible for geometrical
investigations too and was, therefore, one of the Ptolemaic kings. The phrase.
concerning the 'surveyors of king Ptolemy' occurs also in the works of the
agrimensores, known to, and copied alongside, the early astrolabe texts. 63
One can see a development of this legend in the words of Raymond of Marseilles,
written in 1140: ' Astronomers are also wont to argue about which of the twelve kings
of Egypt who were all called" Ptolemy" was the composer of the astrolabe. But, since
two of the kings are said to have been wiser than the rest, it is said by the more
authoritative of astronomers that it was the Ptolemy who was called" the Great" who
composed it'.64 To counter these stories, Gerard of Cremona, when he translated
Ptolemy's magnum opus, the Almagest, reproduced in a preface some phrases from the
Arabic anthologist Abu l-Wafa' al-Mubashshir ibn Fatik's Mukhtgr al-bikam65
pointing out that 'Ptolemy was not one of the kings of Egypt who were called
"Ptolemies", as some people think: but Ptolemy was his name, as if someone was
called "Chosroe" or "Caesar'''.66 However, the feeling that a king Ptolemy was
responsible for some texts on the science of the stars lingered on. For example, in MS
Catania, Biblioteca universitaria 87 (85), of the fifteenth century, a note accompanies
the astrological judgements (Iudicia) attributed to Ptolemy to the effect that 'Ptolemy,
one of the kings of Egypt, composed this text; he is not the same Ptolemy as the one
who composed the Almagest or Tetrabiblos, but another one'.67 Moreover, Ptolemy
the astronomer continued to be portrayed in sculpture and illustrations, as a king.68
maiorem vocavit astronomum. Is enim canones, quibus cursus astrorum inveniantur instituit'. Pingree
(note 38), 355-(i0, has set out the evidence that Cassiodorus is referring to Preceptum Canonis Ptolomei,
a cOfY of which he has in his hands.
6 Martianus Capella, De nuptiis, VI, 595-8: 'Nam a mensoribus regis Ptolomaei, qui totam Aegyptum
tenebat, adiutus ... '. This had been pointed out by Bubnov (note 4), 139, n.2.
63 Bubnov (note 4), 362, lines 7-8.
64 Raymond of Marseilles, Liber cursuum (from the edition in preparation by M.-T. d'Alverny, E.
Poulle, and C. Burnett): 'Solet etiam ad hoc altercatio inter astrologos fieri, cum XII Egypti reges
prefuerint qui omnes Ptolomei vocati in hac scientia valuerint, quis illorum compositor astrolabii extiterit.
Sed, cum duo ex ipsis sapientiores fuisse memorentur, fertur ab astrologorum senioribus Ptolomeum qui
magnus cognominatus est illud composuisse'.
65 This is the text that became famous later on in Europe in Latin and various vernacular translations
(the Bocados de oro, and The Dicts and Sayings of the Philosophers); F. Rosenthal, 'Al-Mubashshir ibn
Fatik', Oriens, 13-14 (1960), 132-58; Paul Kunitzsch, Der Almagest (Wiesbaden, 1974), 98-9. Other
Arabic authorities took care to refute the idea that Ptolemy was one of the kings of Egypt: e.g. ~a'id al-
AndalusI (ed. Bu-'Alwan [Beirut, 1985],88-90) and Ibn al-QiflI (ed. J. Lippert [Leipzig, 1903], 95ff.); I owe
these references to Kunitzsch. None of the sources described by King (note 58) refer to Ptolemy, the
inventor of the astrolabe, as a king.
66 'Ptolemeus vero hic non fuit unus regum Egypti, qui Ptolemei vocati sunt, sicut quidam estimant:
sed Ptolemeus fuit eius nomen: ac si aliquis vocaretur Cosdrohe aut Cesar'; the bibliographical
information in the preface is reprinted from the edition of the Almagest of Venice (1515), in F. BoIl, Studien
Uber Claudius Ptolemiius (Leipzig, 1894), 58.
67 Catania, Biblioteca universitaria, 87 (85), f.89r: 'Hunc tractatum edidit Ptolomeus unus de regibus
Egipti, nec est idem Ptolomeus qui edidit Almagestum vel quadripartitum sed diversus'. The same
distinction is made by Francesco Giuntini, Speculum astronomiae (Lyons, 1581), 17: 'Dicit Ptolemaeus in
primo Quadripartiti. ... Alii autem astronomi, sicut Ptolemaeus rex Aegypti ... ' (a quotation from Iudicia
follows).
68 For example, on the frontispiece of The Elements ofGeometrie ... , translated by H. Billingsley, with
the famous preface to Euclid by John Dee (London, 1570). For other depictions of Ptolemy as a king, see
J. Tezmen-Siegel, Die Darstellungen der septem artes liberales in der Bildenden Kunst als Rezeption der
Lehrplangeschichte (Munich, 1985), 113, n.69, pIts 13, 15,38,55; these are placed in the context of other
The Astrolabe and Arabic Astrology in France 343
V: Alchandreus Philosophus
, Alchandreus Philosophus' is the name given to the author of the first and main
text in the Alchandrean corpus (Liber Alchandrel) in its earliest manuscript: Paris,
BNF, lat. 17868.69 Millas Vallicrosa suggested that the name was a corruption of' al-
Kindi' on the grounds that al-Kindi promoted the use of lunar mansions, of which
the earliest Latin version is in Liber Alchandrei.'° For reasons yet to be explained, al-
Kindi was a particularly renowned figure in al-Andalus, and at least one philosophical
work of his (De quinque essentiis) is known only from its translation into Latin in
Toledo in the late twelfth century. But it seems unlikely that' Alchandreus' is a
corruption of 'al-Kindi', on philological grounds. Moreover, the use of lunar
mansions in Liber Alchandrei is completely different from that of al-Kindi in his letter
on weather forecasting (known in Latin as De mutatione temporum).71 A corruption
of 'Alexander' might seem more plausible. Alexander of Macedon appears as an
authority within the text several times, and a work on onomantic astrology entitled
'Astronomia Alexandri' is mentioned in a list of authorities on the science of the
stars. 72 However, as Juste has pointed out,73 it is illogical that the author of the book
should be the same as one of the authorities he cites, and it is notable that the spelling
of Alexander is quite correct in the citations.
VI: Ascelin of Augsburg, Compositio Astrolabii
The following edition includes all the readings from all five known manuscripts:
Oxford, Corpus Christi College, 283, s. xi, ff.95v-97r ( = Cc; Figures 6-9); Avranches,
Bibliotheque municipale, 235, s. xii (formerly Mont-St-Michel), ff.71v-73v (= A;
Figures 10-14); London, British Library, Add. MS 17808, s. xi, ff.84r-85r (= L;
Figure 15); Cambridge, University Library, Ii.6.5 (formerly Bury St Edmunds), s. xiP,
ff.121r-122r (= C); and Munich, Bayerische Staatsbibliothek, Clm 458, s. xv,
ff.35v-36r (= M). The text, with preface, once occurred on ff.38v-41r of Chartres,
Bibliotheque municipale, 214 (destroyed in the Second World War). In MS A the
work is headed 'Item alio modo quomodo fiat astrolabium' and the same scribe has
added 'compos[itio] Ascelini' in the margin. Presumably from this information
Robert of Torigni, the abbot and librarian of Mont-St-Michel, has described the text
in the list of contents on f.Av as ' Alia compositio secundum Ascelinum '. MS L has
no title. The heading in MS Cc is in striking colours: red for the dedicator and blue
for the dedicatee. 74 MS C only contains chapters 4 and 5 of Ascelin's work, in a
section of the manuscript written in the early 12th century (ff.112-25). The chapters
from Ascelin are components of a compilation on the astrolabe put together from
various sources, none of which is named. It begins with De utilitatibus astrolabii (' Est
crowned scholars in D. N. Hasse, 'King Avicenna: the iconographic consequences of a mistranslation',
Journal of the Warburg and Courtauld Institutes, 60 (1997), pp. 230-43.
69 Also Vatican, lat. 4084, f.7r: 'Exerptum de astrologia Alahandrei philosophy Saracenorum'
(information from Juste).
70 Millas Vallicrosa (note 5), 253.
71 The lunar mansions are the subject of a volume being prepared currently by Silke Ackermann and
myself.
72 That is, Ut testatur Ergaphalau, edited and discussed in Charles Burnett, 'Adelard, Ergaphalau and
the science of the stars', in idem, Magic and Divination in the Middle Ages (Aldershot, 1996), article 11.
73 Juste (note 28).
74 Bergmann (note 4), 223-5, edited Ascelin's text from the Avranches manuscript alone in his
Innovationen im Quadrivium, and discussed the work on pp. 99-105. The same manuscript was used by
Kunitzsch for Ascelin's star-table which belongs to 'type Ill' in Typen von Sternverzeichnissen in
astronomischen Handschriften des zehnten bis vierzehnten Jahrhunderts (Wiesbaden, 1966),27.
I I
344
quidem guazalcora tabula ... '; tI.112r-21r), chs 2, 4-6, 8-10, and 19 (lost folios
between tI.119 and 120 might have included chs 10--19); Ascelin; Bede (1), De cursu
soUs (' Bissextili an no prima ... '; f.122r-4r), chs 1-4; a chapter on the shadow square
(' Componitur in astrolabio '" '; tI.124v-5r) which is also in Millas Vallicrosa (note 5),
324, lines 70tI.; and instructions on inscribing the hour-lines (' Qualiter hore distingui
debeant in astrolabio: Ad horas vero distinguendas ... '; ff.125r-6v), added by a
ditIerent, but probably contemporary, hand. 7s The last item appears to adapt the data
in Ascelin's treatise for the meridian of Bury St Edmunds, which was the home of the
manuscript. MS M gives the preface only.
The following comments may be made concerning the relationship between the
manuscripts:
1. The text and the star-table must be considered separately.
2. MS Cc gives the most complete text, being one of the two extant manuscripts that
include the preface (once in Chartres, 214), and the only one to give the last paragraph
on the shadow square. The authenticity of the last paragraph is guaranteed by its style
and terminology, and especially the correspondence of the syntax of the opening
sentence - 'Quadratum geometricis aptum inventionibus sic mensurandum accipe' -
and the sentence which begins the instructions on inscribing the fixed stars on the
volvel: 'Stellas in predicto volvello inscriptas sic mensurandas accipe'.
3. MS Cc refers to the star-table. This, however, is absent in the manuscript. Since
the star-table is not Ascelin's original composition, but rather the star-table already
drawn up for the earliest astrolabe texts (De mensura astrolabii hI! = Kunitzsch, table
Ill), it could be regarded as of the same status as a model of the astrolabe: Ascelin's
text is to be used as an explanation both for how to make an astrolabe and for how
to use the star-table. Thus the lack of a star-table in MS Cc does not mean that
Ascelin's text is defective in this manuscript.
4. The scribe(s) of MS A knew two manuscripts for Ascelin's text. Originally scribe
A copies a text that is very similar to MS L, as part of a collection of texts shared by
MSS A and L; in the common exemplar the text was probably anonymous. Then
scribe AI (who may be the same scribe) in addition to making corrections adds
alternative readings which correspond to those of MS Cc, and may have come directly
from that text. 76 This new manuscript provided scribe A I with the name of the author
of the text.
5. MS A has the star-table, to which scribe AI has made corrections. In this case,
however, the correction cannot have come from MS Cc, which does not have the star-
table, nor does a manuscript related to MS L seem to have been the original exemplar.
The A I-readings come from a source, which, in one case, is described as being 'in
veteri', i.e. 'in the old [manuscript)'. Both A and AI-readings are found in MS L.
6. The table listing which star-pointers are fixed to which circles on the astrolabe is
found only in MS A. This feature, as has been pointed out by Bergmann,77 is not
found in the earlier astrolabe texts. Its introduction, however, should not necessarily
be attributed to Ascelin, since it does not occur in the two eleventh-century
manuscripts of his work (MSS Cc and L). Following a suggestion of Bergmann, I
propose that the table was written down from observing an actual astrolabe, either
7S For a full description of the manuscript, see M. Folkerts, Die alteste lateinische Schrift aber das
indische Rechnen nach al-!Jwiirizmf. Bayerische Akademie der Wissenschaften, philos.-hist. Klasse, Abh. n.
F., 113 (Munich, 1997),23-4.
76 The only reading of A l that does not correspond to MS Cc is the correction of' decimum octavum'
to 'octavum decimum' in note 150.
77 Bergmann (note 4), 103-5.
The Astrolabe and Arabic Astrology in France 345
by scribe A or a scribe of one of the ancestors of his text. It should not be considered
as part of Ascelin's text. 78
7. Scribe A I also had available a figured diagram for the construction of the astrolabe
very like that in Paris, BNF, lat. 11248, f.26r (Figure 1), illustrating De mensura
astrolapsus h', for he annotates the text by referring to figure' d " which corresponds
to the' d' on that diagram.
8. MS Cc presents a text free from the additions and omissions in MSS A and L. The
only manuscript that might have preserved a text even closer to the Ascelin's original
document is the much-lamented Chartres 214. MS Cc's text is of a high quality, as
befits that of a carefully written treatise, addressed by one humanist scholar to
another. It is this text, therefore, that is presented here, with only occasional necessary
corrections and emendations; the readings from the other manuscripts are given in
the apparatus criticus. The orthography of scribe Cc has been carefully observed; also
the fact that the he tends to write out the numbers in full whereas the scribes of MSS
A and L use roman numerals.
9. MS C is not helpful for establishing the history of the text, since not only does it
contain only two chapters, but also within these two chapters it considerably
abbreviates the text, missing out sections and phrases.
10. MS M, although a late copy, gives the only independent witness to the text of the
preface, now that Chartres 214 has been destroyed. Its text does not derive from that
of MS Cc.
(Preface: MS Cc and M)
ASCELINUS TEUTONICUS CIVIS AUGUST~ CIVITATIS. STABILI AUREN-
LIANENSI MICIACENSI MONACHO SALUTEM. 79
Quantam in administrandis negotiis valentiam80 et in virtutibus ornatum firma
gerat amicicia, multi veterum doctorum et eorum81 primi litterarum monimentis
tradidere, ut quod ipsi vigilanti studio dediti, rationis etiam semitam82 sine deviatione
insistentes, haut temerario iudicio de re tam precipua sensere,83 nostram informandi
gratia et futuram mundi prosequeretur ~tatem. Ergo quid possit amicicia, quanta
etiam fulged o 84 maximarum virtutum irradiata comitatu sese efferat rationis85 ab aula,
a compluribus veterum summ~ auctoritatis viris, maxime autem a Cicerone et
Simaco,86 accepta sententia, statuo totius facultatis me~ operam voluntati amicorum
benigne semper responsuram, non quod omnimodis arbitrerB7 amicici~ decorem
moribus meis ornamento plenum et integrum obtigisse,88 verum89 ut cuius rei estuans
appetentia integram modo 90 quoque habitus nequeo attingere naturam, atIectu tamen
quodam illam amplectar91 vel per imaginariam formam. Quoniam igitur92 te dilectum
78 Since the star-pointer table is adduced by Bergmann as the only piece of evidence for Ascelin's text
being intermediary between De mensura astrolabii h' and Hermann the Lame's De mensura astrolabii, its
function as an intermediary must now be called into question.
79 Ascelinus Stabili Suo, MS M.
80 Quanta ... valentia, MS M.
81 MS M adds 'qui'.
82 semita, MS M.
83 senseme, MS Cc.
84 fulgido, MSS Cc and M.
8S rationis efferat, MS M.
86 Simmaco, MS M.
87 arbitror, MS M.
88 obtegisse, MS M.
89 MS M omits.
90 integrum mo, MS M.
91 amplector, MS M.
 I
I
The Astrolabe and Arabic Astrology in France 347
346
1l8
utpote incontaminat~ honestatis virum, nomine proprio simulque vere moribus parte meridian~ line~ extremum observa[re],117 et ubi interius orientalem tangat
119
lineam pone signum. Quod dum tetigerit circinus iam artior factus uti res exigit,120
pariterque verbo veritatis STABILEM,93 legibus amicici~ non imparem94 existere
consideravi, voluntatis tu~ affectionibus pro posse meo inservire, ratione iubente, circumducas illum, et videbis solstitip21 ~stivalis orbiculum.
animum induxi. Huiusmodi namque animi inductione9S a Tulliana sententia non [2] Primi climatis l22 latitudo l23 .xvi. gradibus l24 terminatur. H~c ~umenda. est
videor recedere, qua96 dignos refert amicicia quibus ipsis inesse perspicitur, propter inferius ab orientali linea de nonaginta partibus et ex altera fronte elUsdem lm~
quod iure diligantur. superius. Spacium quoque quod inter has duas extremitates extract~ .latitu~inis
Simachi etiam rationi97 hoc tenens propositum non possum reclamare, qui, relinquitur primo tribus intervallis distinguitur, secundo unumquodhbet tnum
scribens amico suo de lege amicici~, in capite su~ epistol~ convenienter98 hoc apposuit: interstitiorum in tria, tercio unumquodque secund~ divisionis spacium in duo, quarto
terci~ dimensionis bipertita membra iterum in duo diminuentur. Hac divisionis sic
12S
, Amicorum est honesta petere et honesta concedere'. Te autem, dum honesta petas, 126
Simachi99 sententiam non refellere perspicua docet ratio. Ego vero, ut evidenter re habita ratione, puncta repperientur .xxxvii. et intervalla .xxxvi. qu~ in unam
ipsa demonstrabitur, qu~ rogas uti concedam ad modum facultatis me~, preter quantitatem numeri coadunata, sub singulis acceptis .v. reddunt summam deter-
iudicium laboris laboro, ultra quamlibet lOO negantibus viribus efficientiam voluntate minationis nonaginta. 127 His factis, ducenda est l28 regula ab orientali linea usque ad
aspiro.101 Accipe igitur l02 quod desiderabas opusculum, non integre, sed ingenii mei determinationem latitudinis climatis,129 sive sit h~c latitudo .XVi.130 ut pote primi
pro modulo diligenter 103 elaboratum, ad componendum instrumentum astrolapsus. I04 climatis,131 Sive 132 quilibet alius numerus occurrat, prout climatum133 diversitas extat.
Hinc diligenti consideratione locus line~ longe extra circulum protens~, quem regula
(Text: MSS A, C (in part) Cc and L) predicta transcenderit, notetur. Postea observatur l34 ut a capite orientalis line~ regula
l3S
[1] Componas circulum ~quinoctialem ad arbitrium. Hunc divide in partes non moveatur, sed constanter ibi fixa, inter .xxxvi. reliqua puncta qu~ climatis
l38
subsequuntur136 latitudinem vicatim l37 ducatur, et linea in longum protensa
quatuor, e quibus unam ad te versam partire per tria intersticia quorum
unumquodque continet triginta gradus lOS ut lO6 omnibus simul collectis compleantur quotiens tacta 139 fuerit ab eadem eminente regula cauta notatione imprimatur, hocque
nonaginta, quem unaqu~libet portio quadrifid~ sper~I07 necessario l08 si recte sit tarn diu fiat, donec l40 conscendat regula mutando de .xxxvi. punctis punctum ad
partita 109 debet llO includere. Quo facto, ut circulos terminales ratione possis invenire, ultimum. 141 His ita punctorum rationi ad~quatisnotationibus, perquiratur medietas
sume viciniores orientali line~lll gradus triginta et ex his viginti quatuor circumscribe, inter longissime a se distantes notas. Hoc in loco medietatis positus circinus,142 ut
horum ultimum cauta 112 observatione denota. Dum hoc feceris, accipe regulam et utrasque notas l43 extremas attingere l44 valeat, circumducatur, et fiet primum
impone earn not~ impress~, ita coaptando ut ex altera sui parte septemtrionalis line~ ALMUKANTARAT. 14S Postea inter secundas extremas notas medietati imposito
supersideat extremitati. Regula vero ita posita quocunque in loco orientalis line~ circino et circumducto, secundum fiet ALMUKANTARAT .146 Eodem quoque modo
tangat eminentiam affige notam, quam usque dum attingere possitll3 distrahe pedes
circino, et dum attigerit earn circumvolve et habebis hiemalis solstitii circulum. 117 observare, MSS A, Cc and L.
Interiorem fere eadem indagatione poteris perquirere, eandem quippe predicti numeri 118 tangit, MS A.
viginti quatuor notam 1l4 in altera parte llS sui regula debet obtinere. 1l6 In altera vero 119 autem, MS A.
120 exigerit, MS A.
92 ergo, MS M. 121 solsttii, MS A.
93 dilectum ... STABILEM] dilectum stabilem nomine utpote in certamine honestate virum, MS M. 122 clymatis, MS A.
94 parem, MS M. 123 Scribe Al writes 'latitudo' above 'altitudo'.
124 .xv. gradibus, MSS A and L; scribe Al writes 'vel .xvi. gradibus' in the margin.
95 intentione, MS M.
96 quia, MS M. 125 diminuuntur, MSS A and L.
97 Scribe Cc l corrects from 'ratione'. 126 reperientur, MS A.
98 MS M omits. 127 redditur summa terminationis .xc., MSS A and L.
99 MS M omits. 128 est] MS L omits.
lOO qualibet, MS M. 129 climatis] MSS A and L omit.
101 adspiro, MS M 130 .xv., MSS A and L; scribe Al writes 'vel .xvi.' in the margin.
102 ergo, MS M 131 clymatis, MS A.
103 non integre, sed ingenii mei pro modulo diligenter] non indiligenter, MS Cc. 132 vel, MSS A and L.
104 Scribe A I adds' Compos. Ascelini' in the margin. 133 clymatum, MS A.
105 gradus], MS Cc omits, MS L places before 'triginta'. 134 observetur, MS L.
106 MS A adds 'in'. I3S clymatis, MS A.
107 spher~, MS L 136 subsecuntur, MSS A and L.
108 necessario], MSS A and L omit. 137 meatim, MSS A and L.
109 Scribe A I adds' partita ' in the margin and attempts to correct the faulty word (' partia '?) in the text. 138 H~, MSS A and L.
110 potest, MSS A and L. 139 tracta, MS A.
III Scribe Al adds 'versus .d.' in the margin. 140 quousque, MSS A and L.
112 causa, MS A. 141 ad ultimum] scribe A I adds in the margin, MS L omits.

113 MS L corrects from 'possis'. 142 circus, MS L.

114 MS A gives' .xxiiii. notam predicti numeri' with sigla indicating a change of order, apparently to 143 notas] MSS A and L omit.

that of MS Cc; MS L gives' .xxiiii. predicti numeri notam'. 144 adtingere, MS L.

115 alteram partem, MS A. 145 almucantarat, MSS A and L.
116 optinere, MS L. 146 almucantarat, MSS A and L.
I I

348 The Astrolabe and Arabic Astrology in France 349

per tercias notas se l47 respicientes, tercium, per quartas quartum, per quintas sexto decimo scorpionis; Alhauui in vigesimo quarto sagittarii; Altair in quarto
quintum, per sextas sextum, atque ita per ordinem donec concurrentibus col- decimo capricorni; Delfin179 in vigesimo quinto eiusdem; Alferat in trigesimo aquarii;
lateralibus l48 notis redeat ALMUKANTARAT I49 decimum octavum. lSO Alhadip in 180 duodecimo tauri; Alrif1 81 in vigesimo nono 182 capricorni; Wega in primo
[3] Ita compositis almukantarat,151 spatium quod retro almukantarat l52 vacuum gradu l83 eiusdem; Benenaz l84 in octavo decimo libr~; Calbalagrab 18S in vicesimo
remanet in duodecim per circulos qui dicuntur terminales dividatur, medio indiviso quinto scorpionis ;186 et ad hunc modum reliquis omnibus secundum regularem
manente. Postea circino apposito tam diu moderare ut secunda puncta intimi et numerum prim~ line~, ubi traditio huius mensur~ continetur, stationes tribue, hoc
extimi l53 circuli altero pede circini tangantur, et tunc facias lineam ab extremo circulo quoque diligenter observans l87 ut ad numerum secund~ line~ altitudo l88 sublimetur
usque ad intimum,l54 qui pes dicitur horarum, et habebis prim~ hor~ certum stell~ cuiusque. Quam vero altitudinem l89 ut l90 certa determinatione comprehendere
terminum, et hoc sit communiter dictum de ~teris discriminationibus horarum. possis, sic accipe. Stell~ Alramech l91 posite in libra hoc modo altitudinem
[4] Volvelli lSS quod alii rete,156 alii rotam nuncupant, per subscriptam sententiam comprehende. 192 A loco ipsius stelle193 supradict~ Alramech l94 qui l9S est ut supra
tene compositionem. Ambitus eius amplitudinem metire l57 secundum tropici hiemalis retulimus 196 in vigesimo quarto l97 gradu libr~, computa utrimque .lxv. gradus et ex
quantitatem; semicirculi in eodem positi revolutionem interiorem ~uinoctialis alterutra 198 parte vigesimum l99 quintum gradum200 imprime nota. 201 Quibus notis202
circuli 158 co~qua dimensioni; eiusdem quoque latitudini 159 .xxiiii. l60 maiora intervalla adhibita regula quocumque in loco stellam attingat,203 detrunca eam, et ecce predict~
208
in umbone computata tribue. Zodiaci vero circuli rotationem 161 sic collige. Appone stell~ certam repperistF04 altitudinem. 20s Cuius ad exemplum206 de207 c~teris omnibus
circinum ita coaptans eum ut de terminalibus minorem totum comprehendas orbem, 209
secundum numerum cuique ascriptum universalem hanc tene sententiam.21O
maioris inc1usa tantum summitate. Hoc in loco circinum affixum circumducas et
revertetur zodiacus in quo cas~ signorum et cursus solis per signa162 comprobantur. 163
Sic repertum, zodiacum partire per duodecim l64 intervalla, dans unicuique sex
umbonis maiora spacia, prout ipsorum signorum distenduntur interstitia. Quando (MS Cc only)
autem huius volvelli divisio instituatur, vide ut 165 ALMERII66 circumposito 167 [6] Quadratum geometricis aptum inventionibus sic mensurandum accipe. Ab
umboni 168 in eo loco ubi capita tabularum infiguntur semper sit appositum. Hoc occidentali linea spacium quod continetur usque lineam septentrionalem divide in
quoque caute 169 observato 170 in huiusmodi compositione nunquam te fallet ratio, qua duo. Postea singula de duobus in geminam divisionem iterum diminue. Hinc
duce 171 difficilem vix possis carpere callem. l72 unumquodlibet horum in tria, de tribus quodlibet in bina deinde partire, hoc caute
[5] Stellas in predicto volvello inscriptas 173 sic mensurandas accipe. ALRA-
MECH 174 in 175 vigesimo quarto gradu libr~176 locum 177 sortiatur; ALFECAT178 in 179Delphin, MS A.
180MS A omits.
147 sese, MSS A and L. 181Alriph, MSS A and L.
148 Scribe AI corrects 'lateralibus' to 'collateralibus', lateralibus, MS L. 182Scribe Al corrects' .xxx.' to '.xxix.'.
149 almucantarat, MSS A and L. 183primo gradu] .i., MS A.
ISO Scribe Al has added sig/a above the words to indicate a change in order to 'octavum decimum'. 184Benenab, MS Cc.
151 almucantarat, MSS A and L. 185Calbalagraph, MS A.
152 almucantarat, MSS A and L. 186locum sortiatur ... scorpionis], MS C omits.
153 MS L corrects from 'intimi'. 187observa, MS L.
154 ultimum, MSS A and L. 188Scribe Cl lines through and expunges' altitudo' and writes' latitudo' above.
155 Sol velli, MS C (error in rubrication). MS C begins here. 189Scribe Cl lines through and underlines' altitudinem' and writes' latitudinem' above.
156 MS C corrects from 'recte'. 190 MSS A and L omit.
157 metiri, MS C. 191 Scribe AI corrects from 'aramech'.
158 in eodem ... circuli] MS C omits by saut du merne au meme. 192 altitudinem comprehende] habebis altitudinem, MSS A and L; Stell~ Alramech ... comprehende]
159 latitudinis, MS A. MS C omits.
160 viginti quatuor, MS C. 193 ipsius stell~] MS C omits.
161 rationem, MSS A, C and L. 194 alramehc, MS C.
162 per signa solis, MS C. 195 qu~, MS C.
163 probantur, MS A. 196 ut supra retulimus] MS C omits.
164 .xii., MS C. 197 vigesimo quarto] .xxV., MS C, scribe Cl writes' .iiii.' above.
165 ne, MSS A and L. 198 altera, MSS A and L.
166 MS C adds above: 'id est denticulus'. 199 vicesimum, MSS A and L.
167 circumpositio, MS L, scribe Cl lines through and expunges. 200 et ex alterutra ... gradum] MS C omits.
168 Scribe Cl lines through and expunges. 201 imprime notam, MS A, imprimens notas, MS C.
169 MS A omits. 202 MS C omits.
170 MS A adds and then expunges 'ut'. 203 adtingat, MS L.
171 Scribe Cl corrects from 'dulce'. 204 certam repperisti] MS C omits.
172 Here MS A leaves a space of about two-thirds of the page, perhaps for a diagram. 205 certam repperisti altitudinem] altitudinem certam, MSS A and L.
173 MS C omits. 206 Cuius ad exemplum] Sic, MS C.
174 Alrameth, MS A. 207 MSS A and L omit.
175 Scribe AI adds in margin. 208 MS C omits.
176 gradu Iibn;] Scribe Cl corrects from 'gradualibus'. 209 asscriptum, MS A, adscriptum, MS L.
177 MS C omits from this word to 'scorpionis'. 210 cuique ... sententiam] adscriptum cuique stelle, MS C. MS C ends here.
178 Alfeca, MSS A and L.
I I
350 The Astrolabe and Arabic Astrology in France 351

perpendens, ut inter lineam occidentalem et septentrionalem natura prodeat recti Star-pointer table: MS A
quadrati cuius quartus angulus tribus ceteris ~quus ut equalitas exigit recti quadrati In parvo circulo circa centrum:
terminetur in medio centri. EXPLICIT. Benenaz, Wega, Alriph, Algol, Alhaico, Egregez, Alrucuba, Alhadip.
In ¥quinoctiali circulo:
Star-table: MSS A and L Aldebaran, Algoze, Aldiraan, Maleuxe, Calbalazeda.
In hiemali circulo:
Latitudo Altitudo Signa Sten~211 Calbalagrab, Achimech, Algurab, Alhabor,
Rigel, Gaitoz232 derep, Gaitoz patan, Libdideneb. 233
.xxiiii. .lxv. Libra Alramech In zodiaco circulo:
.xvi. .lxxi. Scorpius Alfeca Alramech, Elfeca, Alhauui, Altair, Oelphin, Alferaz .
.xxiiii. .lvii. Sagittarius Alhauui
.xiiii. •r'"
1111 •212 Capricomus Altair l3
.xxv. .lviiii. Capricomus Delphin214
.xxx. .lxv. Aquarius Alferaz21S Translation
.xii. .Xci. 216 Taurus217 Alhadip Ascelin the German, citizen of the city of Augsburg, to Stabilis of Orleans, monk
.xxviiii. .lxxiii. Capricomus Alriph218 of Micy, greetings!
.i. .lxxii. Capricomus Wega vel Concerning how much effect firm friendship has in getting things done and
winga219 adorning things with virtues, many of the old doctors (and the foremost ofthem) have
.xviii.220 .lxxiiii. Libra Benenaz
.xxv. .xiiii. Scorpius Calbalagrab
transmitted the evidence in the monuments of literature, so that what they, dedicated
caput serpentis221 to night-long study and sticking to the path of reason without deviation, have realised
.viii. .xv . Libra Alchimech without a rash judgement concerning such an important matter, should be sought for
.xviii. .xli. Virgo Algurab the sake of informing our own age and the future ages of the world. Therefore, having
.i. .. 222 Cancer
.XXXII. Alhabor received from very many ancient men of the greatest authority, especially Cicero and
.viii. .xxxviiii. Gemini Rigel
.xx. .xxxvi. Aries Gaitoz patan Symmachus, the statement of what friendship can do, [and] how much brightness,
.xx. .xxxvi. 223 Pisces Gaitoz dereb224 irradiated by the companionship of the greatest virtues, issues from the hall of reason,
.Vii. 22S .xviiii. Aquarius Licdideneb I have decided that the effort of my whole talent should always reply kindly to the
cauda serpentis226 wishes of friends, not because I think that the glory of friendship has fallen to my
.xxvii. .lxi. 227 Taurus Aldebaran
.xviii. 228 character full and complete as an ornament [1], but so that for whatever thing I am
.Iv. Gemini Maleuze229
.xvij.23O .lvi. Cancer AIgoze thirsting with desire but whose possession I am in no way able to obtain the full
.vi. .Hi. Leo Aldiraan nature of, nevertheless by some feeling I might embrace it and have it through an
.xviii. .lxi. Leo Calbalezeda imagined form. Since, then, I have considered that you - a beloved man, being of
.x. .lxxi. Taurus Algol unsullied honesty, by your own name and at the same time truly by your character
.iii. .lxxiiii. Gemini Alhaico (sic)
.xix. .lxxv. and equally by the word of truth being 'Stable/Constant' - are not unequal to the
Leo Alrucuba
.xxvi. .lxxii. Cancer Egregez23I laws of friendship, I have brought my spirit to serve as far as I am able the desires
of your wish, reason dictating this. I do not seem to depart from the statement of
211 stellae, MS L. Cicero in this 'bringing of' the spirit, in which he relates that those are worthy of
212 liii, MS L, MS A before correction. friendship for whom it is perceived to be innate because they are rightly loved.
213 Althair, MS A.
214 Delfin, MS L. Holding to this conviction, I am not able to object to the reasoning of Symmachus
21S Alferat, MS A. who, writing to his friend about the law of friendship, at the beginning of his letter
216 . MS...A after .co?,ect.lOn,
:xc~:,
. .
217 .CXt., MS L and MS A before correction (?) . appropriately places this statement: 'It is the duty of friends to seek honourable
.Xll. lxxm. ScorplUS wntten by scribe AI in the margin.
218 Alrif, MS L. things and to grant them'. Perlucid reason instructs us that you do not go against the
219 Sc~be AI adds 'vel winga' above 'Wega'; uuega vel uniga, MS L. opinion of Symmachus while you seek honourable things. But, as will be shown by
220 Scnbe AI adds' r. xvii' in the margin. the matter itself, in spite of the difficulty of the work, I work hard at what you ask
~~ Scribe AI adds 'caput serpentis' alongside 'Calbalagrab'. that I should give, as far as I am able; I aspire by a wish which exceeds any practical
.xxxi. , MS L, MS A before correction.
: .xxxv., MS L, 'in veteri R.. xxxv.' written by scribe AI in the margin. ability, since strength is lacking. Accept, then, this work that you desire, worked out
derep, MS A. for constructing the instrument of the astrolabe, not completely, but diligently
22S .viii., MS L.
226 Scribe AI adds 'cauda serpentis' alongside 'Licdideneb' according to the small measure of my intelligence.
227 .li., MS L. . 1. You should draw the equinoctial circle as you like. Divide it into four parts.
228 'xl' (sic) written by scribe AI in the margin. Subdivide the one facing you into three intervals, of which each contains 30 degrees,
229 Maleuxe, MS L.
230 '.xii.' written by scribe AI in the margin. 232 Scribe Al corrects from 'Gaizo'.
231 MS L ends here. 233 Scribe AI adds' cauda serpentis' in smaller letters.

I
352
so that the sum of the degrees is 90, which is the number of degrees that each portion
of the quartered sphere should necessarily contain, if it has been correctly subdivided.
Having done this, so that you can find the terminal circles rationally, take the thirty
degrees closest to the oriental line, and from these mark off 24; mark the last of these
with careful observation. When you have done this, take a ruler and put it on the
mark that you have made, moving it so that its other end lies on the end of the
northern line. When the ruler has been placed in this way, in whichever place it
touches the extension of the oriental line put a mark. Stretch the feet of a pair of
compasses so that they can reach this mark, and when they reach it, revolve the
compasses, and you will have the circle of the winter solstice. You can seek the inner
[circle] by almost the same investigation, i.e. the ruler ought to reach the same mark
of the aforesaid number 24 on the other side. But on the other side observe the
extreme of the meridian [southern] line, and, where it touches the oriental line on the
inside, put a mark. When the compasses touch this, having been made narrower as
the situation requires, you should revolve them, and you will see the little circle of the
summer solstice.
. 2. The l~titude of the first clime is terminated in 16 [alias 15] degrees [16 degrees
IS the value In De mensura astrolabii h'; '15 degrees' is Hermann the Lame's value].
This should be taken from the 90 divisions on the lower side of the oriental line, and
on the higher side of the same line on the other side. Also, the space that is left
between these two extremities of the extracted latitude is divided first by three
intervals. Secondly, each of the three intervals is divided into three; thirdly, each
space of the second division into two. Fourthly, each of the halved divisions of the
third measurement again is reduced into two. When this method of division has been
performed in this way, 37 points are found and 36 intervals which, when brought
together into one quantity, when 5 have been given to each of them, return 90 [read
180?] as the sum of the limit. When this has been done, the ruler should be lead from
the oriental line to the determination of the latitude of the clime, whether this latitude
be 16 [alias 15] - i.e. of the first clime - or whether any other number should arise
~ccording to the differences of the climes. With careful observation, the place of th~
lIne extended from here to far outside the circle, which the aforementioned ruler
passes outside of, should be noted. Then, it is observed that the ruler should not be
moved from the beginning of the oriental line, but is fixed there firmly, and moved
in turn through the remaining marks which follow the latitude of the clime and the
line, extended lengthwise, is marked with a careful annotation as many tim~s as it is
touched by the same ruler as it extends [?]. This should be done until the ruler turns
from the 36th mark to the last one. When these annotations of marks have been
accurately equalized, the midpoint between the marks which are furthest away from
each other is sought. In this midpoint the compasses are placed in such a way that
they are able to touch each of the extreme marks; they should be revolved, and the
first almucantar will be made. Afterwards, between the penultimate marks, when the
compasses have been placed on the midpoint and revolved, the second almucantar
will be made; in the same way through the marks third from the ends, the third
[almucantar], through the fourth [from the ends], the fourth, through the fifth, the
fifth, through the sixth, the sixth, and so on until, the marks now being adjacent to
one another, the 18th almucantar arises.
3. When the almucantars have been made in this way, the space which remains
empty behind the almucantars should be divided into twelve on the circles which are
called 'terminal', while the middle [space] remains undivided. Afterwards, when the
I
The Astrolabe and Arabic Astrology in France 353
compasses have been positioned, arrange them so that the marks second from the end
of the innermost and outermost circle are touched by one leg of the compasses, and
then you should make a line from the outermost to the innermost circle, which is
called 'the foot of the hours' [?], and you will have the certain limit of the first hour,
and this rule has a general application to the determination of the other hours.
4. Understand the composition of the volvel- which others call the' rete' [net],
others the wheel- through the following explanation. Measure the size of its
circumference according to the size of the winter tropic. Make the inner circumference
of the semicircle placed inside it equal to the size of the equinoctial circle. To its
breadth [Le. circumference] give the 24 [read 72?] greater divisions measured on the
rim. But understand the turning of the zodiacal circle in this way: position the
compasses, so arranging them that you encompass ~he. whole smalle! orb of the
terminal circles, but only the top of the higher orb IS Included. HaVing fixed the
compasses in this place, you should lead them round and the zodiac will be r~turned,
in which the houses of the signs and the course of the Sun through the SignS are
ascertained. Divide the zodiac, found in this way, into twelve intervals, giving to each
the six larger spaces on the rim, as far as the intervals of the signs themselves are
extended. But when the division of this volvel is to be undertaken, make sure that the
pointer [almerl1 is always placed on the surrounding limb in that position where the
beginnings of the tables are inscribed [?]. When this too has been ~arefully o~served,
calculation will never fail you in a composition of this kind. Under ItS leadership [read
'without its leadership'?] you would hardly be able to pick out a difficult path.
5. Here are the stars inscribed on the volvel, that should be measured in this way.
Alramech should be allotted a place in the 24th degree of Libra; Alfecat in the
16th of Scorpio; Alhauui in the 24th of Sagittarius; Altair in the 14th of Capricorn;
Delphin in the 25th of the same; Alferat in the 30th of Aquarius; Alhadip in the 12th
of Taurus' Alrif in the 29th of Capricorn; Wega in the first degree of the same;
Benenaz i~ the 18th of Libra; Calbalagrab in the 25th of Scorpio. In this way, give
their positions to all the rest according to the allotted number in the first column [of
the table], where the received opinion for this measurement is included, observing t~is
carefully too: that the height of each star is measured according to [the number In]
the second column. In order that you should understand the height by a firm
calculation, do this: understand the height of the star Alramech in Libra in this way.
From the place of the star itself (the aforementioned Alramech), which is, as we have
mentioned above, in the 24th degree of Libra, calculate 65 degrees on both sides and
make a mark on the 25th degree [read' 65th degree '] on each side. Having marked
these, using a ruler, in whatever place it touches [the line of the latitude of] the star,
cut it off, and, behold, you have found the definite height of the aforementioned star!
In accordance with the example of this [star] take this general rule for all the others
according to the number assigned to each one.
6. Here is a quadrant suitable for discoveries made by geometry; it should be
measured in this way. Divide the space which is contained between the western and
northern lines by two. Afterwards, again divide each of the two into two divisions.
Then divide each of these into three, and then each of the three into two, weighing
this up carefully: that between the western and northern lines the nature of a
quadrangle should arise whose fourth angle, being equal to the other three, as the
equality of a quadrangle demands, should finish in the middle of the centre. The end.
 I
I
The Astrolabe and Arabic Astrology in France 355
354

Commentary
Ascelin describes the construction of the astrolabe in an abbreviated way, without
any reference to what is happening in the heavens. When compared with other early
astrolabe literature, what is striking is that he dispenses with illustrative figures. This
makes it very difficult to follow the text unless one knows other texts in which the
stages of construction are illustrated by figures. Ascelin appears to understand the
construction well enough, although there are some errors in the text, which may
be due to copyists. He tries to make the construction easier by limiting the divisions
of angles to divisions by two and three. In the following commentary I am indebted
to the very clear exposition of Hermann the Lame (ed. Drecker [note 58]), but have
used diagrams illustrating Mensura astrolapsus (h'). Hermann's text is not only
clearer than Ascelin's, but also it includes more details of construction and more
Arabic words, and continues with instructions on how to mark the hours, days, etc.
Hermann also uses as his example, his own latitude (48°) and mentions the
construction of several climate plates. Ascelin's text cannot be Hermann's immediate
source. I am much indebted to Paul Kunitzsch for clarification of Ascelin's text.

1. The drawing of the circles of the equinox, and the winter and summer solstices
(Figure 1).
The construction requires ruling a line through F (at 24° from the oriental line
AC) and D (the end of the northern line), which marks the end of the diameter of the
circle of the Winter Solstice at G. The circle of the Summer Solstice is obtained by Figure 1. Paris, BNF, lat. 11248, ff.25v-26r; from Millas Vallicrosa (note 5), pit. 14.
drawing a line from E (at 24° from the oriental line AC on the other side) to D, and
drawing a circle through I.

2. The drawing of the almucantars (Figure 2).

Point I marks the degree of the climate required, counted from C, and point G
marks off the same number of degrees counted from A. The semicircle IG is divided
into 36 parts, by successive divisions by threes and twos. Each of these parts consists
of 5° (Ascelin probably meant to write '180' as the product of 5 and 36, and the
number of degrees between I and G, rather than '90'). Then a line is drawn from C
through I to meet the extension of the northern line at K. Keeping one end of the ruler
always on C, lines are drawn through each of the divisions marked on the semicircle
IG, the last line to be drawn being CG. The first almucantar is obtained by drawing
a circle that passes through the extreme points marked on the northern line, i.e. KL.
The next almucantar is obtained from drawing a circle through the points closest to
K and L, and so on, until altogether 18 almucantars are drawn.

3. The drawing of the hour-lines (Figure 3).

The arcs of the equinoctial and solstitial circles not cut by the almucantars are
each divided into 12 equal parts. The points marking the end of the first of these parts
on each of the circles are joined, using compasses, and the same is done for the second
and subsequent divisions. . ,
Figure 2. Paris, BNF, lat. 11248, ff.26v-27r; from Millils Vallicrosa (note 5), pIt. 15.
4. The construction of the rete (Figure 4).
The rim is divided into 72 divisions (12 divisions each subdivided into six parts).
The rete consists of an outer circle, which is the same size as the circle of the Winter
Solstice; an inner semicircle, which is equal to the equinoctial circle; and the circle of
the zodiac, which is drawn in such a way that it includes the whole of the circle of the
 I
I
356
The Astrolabe and Arabic Astrology in France 357

Figure 3.. Chartres, Bibliotheque municipale, 214, f.30r; from van de Vyver (note 13), pIt. 2.

Summer Solstice, and touches the circle of the Winter Solstice (as in the circle
diameter 01 in Figure 1). 'Eiusdem quoque latitudini ... tribue' seems out of place. It
should mean 'To the breadth [i.e. circumference] of the zodiac circle give the 24 [read
72] greater divisions measured on the rim'. In that case it repeats the instructions in
'Sic repertum, zodiacum partire ... interstitia '. Hermann makes it clear that the
zodiac circle is divided by placing the ruler between the centre of the' terminal circles'
and each of the 72 divisions of the rim, and marking off the points where the ruler
••
Figure 4. Vahcan,
· Rego lat. 598 , f.l20r;lines
fromadded).
Millas Vallicrosa (note 5), pit. 10 (construction
cuts the zodiac circle. This results in the divisions on the zodiac circle being of
different lengths. The pointer (a/men), according to Hermann, is situated between
Sagittarius and Capricorn; it is not clear what Ascelin means by 'capita tabularum'. altitude is discovered by drawmg. a rI~e from the centre. of which
the astrolabe through
is divided the
into 360
star's latitudinal position on the zodIac to t .e oU,t;; ~~.' ions of the rim' in Section
b
5. The marking of the fixed stars on the rete (Figure 4). divisions/degrees. (Note that t~e ~uthor ~en~lons d IVI~umn of the table in both
The latitude of each fixed star is its degree within a sign of the zodiac (more 4 above.) Count the degrees mdIcated m t e secon t cO' and observe where they
properly described as 'longitude '), as indicated in the first column of the Star-table. Its directions from the point where the line me~ts the ~u e~ Tt mthat line cuts the first line
end. Then rule a line between the two endpomts, an were
I I

358 The Astrolabe and Arabic Astrology in France 359

Figure 5. Vatican, Reg. lat. 598, f.120r; from Millas Vallicrosa (note 5), pIt. 11.

is the position of the star. According to Paul Kunitzsch (private communication) this
method of finding the altitude of the star is unique to Paris, BNF, lat. 7412, Ascelin Figure 6. Oxford, Corpus Christi College, 283, f.~5~ (with permission of the Masters and
and Hermann the Lame. Fellows of Corpus Chnstl College).

6. The shadow square

Finally, Ascelin describes how to construct the shadow square on the back of the
astrolabe. Comparison with Figure 5 makes his wording clear.
I I

360 The Astrolabe and Arabic Astrology in France 361

r 11"_ ,.... ........._..r~. "".".............

___.....roa.t~........,'.......... f,........,.....
•1'C'_,.,........... ..-.......... ,Ja.4"....r... ~,.

~1-' "-id .j.m. ~ ....................;;...-.

"'4mkUU .
....................·(ifi ..... ~...... r .. ."...",• ..,

~....r:.......,...-4,.... .......,...".Al , ...

.. .t~.,....,.~~
ti....~......... ~...,.".". . . .,,.......
....... IW

~~··.....r_J........,.....,.,..rtl...... "
........... UItIIIIf,. . . . . . . . . ......,......"" . .'

,. ,.,. .",. ."re.".",.,.. ..

.,......r,.,.......
...........,.Q.f.

~---....,.
1'11 .. ., .......... . .
:i'Lberut.........•

. -__ w .. 6 .. _......
'.
.:'. '.:•,.. .
t),.,........... ....,.......-..r.~ ~
,
. ... :\

P'f"..........".....",......,. . ."i"...
S~. .~Iw ...If'M....rtf...

Figure 7. Oxford, Corpus Christi College, 283, f.96r (with permission of the Masters and Figure 8. Oxford Corpus Christi College, 283, f.96v (with permission of the Masters and
Fellows of Corpus Christi College). , Fellows of Corpus Christi College).
 I'
I I I

362 363
The Astrolabe and Arabic Astrology in France

..
dt ........'dIfft..........
.,.."_-'·-cpo,._
,... ."..
'.t ~___..... 4t...".•.
. . . . . In • ....".,..,... ...... _,,......,.,. ••• "'_
". :\1'r'* Cft._-..,..rah.r4r,__IV...,. .C",
~~~.J~"'''''~'.M'.~~C_''
_rmuml~""t.a;"""'~'~s.~
~,....,.~n.,lIIItuILdMf.......,vr .•"""'"""",AW'
.,r.~~tnMnM.""""'_W"''''",,,,,,,,,,,,,,
IUdr""Jt."ffU~"""'''_''''''''''"",,'''''''''~'
~Gr.,....,..tt«.ft tAtIr~.UtW" .'I!""~"'"
..... Q.... -~ . . . .~4'AUl·l s....rr.~ ........
""~I'U~..",·l1~·",~·"'·,....,·w
. ~A~ut'·.· ........ ~a.""''''::''~.At
fIIrt"·'" p;' .....~.t::.:•. A. . .·•"1Jr_,_~ .m.......
".....7t CQn ••W.t"~·~·."lfJI!lIf_~I!""'.
.
-..."., ,.,..CAfw.,...m=

•. .................
M.. ,.tMI"".,..,;;'" - , ..
......tw.4L"",,~......
...nIt 4..l,....."......II'IIJ.INDft..-
;..,
~~....4...~'~...-~
~~K~. . . . ~~.~.~-J~
~~~~ ••• Mr~~~w.~~~
,,-.....

" ' . . , ......, . " . . . . . . . . . . . \.. . . . . . . . . . ., . . . . . .411

Q;--""- Mi.rtf"1"""""""'-' - - -
,..".,..,r <fIIoJi

__
.A.......... _ ............
. _ _ . . . . . " . . . . . .n1~• •
fIIIf'I1 ............. ~".
~~_~* .~Mr.~.,a..-

Oxford, Corpus Christi College, 283, f.97r (with permission of the Masters and Figure 10. Avranches, Bibliotheque IDunicipale, 235, f.7lv (with permission).
Fellows of Corpus Christi College).
I I

364 The Astrolabe and Arabic Astrology in France 365

Figure 11. Avranches, Bibliotheque municipale, 235, f.72r (with permission).

Figure 12. Avranches, Bibliotheque municipale, 235, f.72v (with permission).

I I
366 367
The Astrolabe and Arabic Astrology in France

"--.-.
'f\..··
.. '

.p'
~.
"fJ
.:IIt,. ....
......... ..
.....
CK'

' .

Figure 13. Avrancbes, Bibliotbeque municipale, 235, f.73r (with permission). Figure 14. Avranches, Bibliotheque municipale, 235, f.73v (with permission).
I
I
368

Addendum to 'King Ptolemy and Alchandreus the Philosopher:

The Earliest Texts on the Astrolabe and Arabic Astrology at Fleury,
Micy and Chartres H

I am extremely grateful to Catherine Jacquemard for drawing my attention to the

fact that a fragment of Ascelin of Augsburg's Compositio Astrolabii can be found in
MS Vat. Reg.lat. 1661, fols 56r-57r, and for sending me photocopies of the relevant
pages. I had missed this witness despite the fact that the manuscript had already been
described by Elisabeth Pellegrin in Les manuscrits classiques latins de la Bibliotheque
Vaticane, Ill, Paris, 1978, pp. 341-4. This manuscript (V) is closely related to
Oxford, Corpus Christi College, 283 (Cc), with which it alone shares the final
paragraph on the shadow square. Also, in common with Cc, it includes no tables.
However, V adds several errors of its own. In general, V uses roman numerals rather
than writing out the numbers in words, and writes 'ci' for 'ti' before vowels. The
following is a complete list of V's readings:
Preface] Vomits.
[1] gradus] CCV omit; compleantur] compleatur V; indagatione] indagationem V; In
altera ... observa] Vomits; artior] arctior V.
[2] climatis] clymatis AV; secund~] V adds supra lineam; diminuentur] diminuuntur
ALV; observatur] observetur LV; subsequuntur] subsecuntur ALV; quotiens]
quoties V; quartum per quintas] Vomits.
[3] moderare] moderate V
[4] hiemalis] hyemalis V; circuli rotationem] rotacionem V; affixum corr. ex affectum
V; circumducas] circumduas V; Sic] Si V; volvelli] uoluelul V; compositione] V
omits; ratio] oratio V.
[5] Delfin] Delphin AV; Benenaz] Benenab CcV; sublimetur] sullimetur V; ipsius
stelle] ipsius stell~ V; nota] notis (?) V; ascriptum] adscriptum LV.
[6] V has exactly the same text as Cc. A space of one third of a page is left blank after
the explicit.

Errata
It would be more accurate to write 'vigesimo quinto scorpionis' rather than
'vicesimo ... ' as the degree of Calbalagrab. Also, the diagram whose presence is
implied by a note in the Avranches manuscript2 occurs on fol. 38v of the same
manuscript; thus Figure 13 = MS Avranches 235, fol. 38v, and Figure 2 = MS
Avranches 235, fol. 39v.
Figure 15. London, British Library, Add. MS 17808, f.85r (with permission).
1 Annals of Science, 55 (1998), 329-68.
2 Ibid., p. 345.
3 Ibid., p. 355.
 11

PHYSICS BEFORE THE PHYSICS: EARLY

TRANSLATIONS FROM ARABIC OF TEXTS
CONCERNING NATURE IN MSS BRITISH LIBRARY,
ADDITIONAL 22719 AND COTTON GALBA E IV*

This article draws attention to the Arabic provenance of certain

texts in two important collections of works on natural science and
medicine written in England in the early to mid-twelfth century:
London, British Library, Additional 22719 and Cotton Galba E IV.
Two of these texts - on the elements, and on metals (with impli-
cations for alchemy), respectively - share the characteristic of
ascribing natural processes to the wisdom of God (sapientia Dei),
and, in the first text, the creative force is most usually described as
Nature herself. The investigation of these texts leads to the sugge-
stion that, contrary to the generally accepted view, there was con-
siderable interest in naturalia (usually referred to with the equiva-
lent term from Greek: "phisica") in Southern Italy already in the
eleventh century. It also questions the common tendency to attri-
bute all translations from Arabic in this period to Constantine the
African who was active in Salerno and Monte Cassino from 1077
and died before 1098.

I. THE MANUSCRIPTS

The relevant details of the two manuscripts on which this arti-

cle centres are as follows. London, British Library, Additional

* I am grateful to A. Colinet, Fritz Zimmermann and members of an Arabic

reading class, David d'Avray, Klaus-Dietrich Fischer, Monica Green, Michael Mc-
Vaugh, Danielle Jacquart, Francis Newton, Peter Lautner, Anna Somfai, Pinella
Travaglia and Nigel Wilson. It was only when this article was in an advanced state
of preparation that the author became aware of the work of H. Grensemann and
U. Weisser (see note 15 below), which confirmed the Arabic origin of the Addi-
tional De elementis. This monograph has enabled the author to dispense with some
of the more detailed arguments for this Arabic origin, and to correct the edition in-
cluded in Appendix I below. Its authors were unaware of the Additional MS.

11
54
22719 (= A) appears to have been copied in the second or third
decade of the twelfth century, l and belonged to the priory of St
Nicholas in Exeter, which was a cell of Battle Abbey.2 It is unlike-
ly, however, that it was written in either of these places; rather, it
shows closer resemblances to manuscripts written at Bath in the
early twelfth century.3 It has been written with great care, but per-
haps by a scholar rather than by a professional scribe. A second
hand has written the last three folios (fols 200V-202V), but follows
the same scribal practices as the first. The manuscript consists of a
copy of the Pantegni of Constantine the African, before which,
within which, and after which, other texts have been added. At
the beginning of the manuscript there is a Christian spell and a
glossary of medical epithets headed « genera effectuum pigmento-
rum (sic)) (( ypnetica: quce somnum vel saporem [sic for "so-
porem"] faciunt [ ... ] »). The ten books of the Theorica of the Pan-
tegni (fols 2V-16IV) are followed without a break by (I) a text on the
elements (fols 16IV-163v), and (2) some recipes including English
words: « Ad lapidem frangendum vel proiciendum [ ... ] species
propter saporem et fac electuarium (fols. 163v-164r) ». Then follow
books I and 11 (fols 164v-192v) and the first half of book IX (fols
193r-200r) of the Practica of the Pantegni. Between books 11 and IX
there are an untitled work on the humours (fo1. 192V: «Sanguis
facit homines boni modi») and a recipe (( Gentiance, Centaurece
maioris et minoris [ ... ] ad modum avellane cum pusca»). After
Book IX there is a series of texts which had been described on fo1.
165r as « que dam phisica »: Costa ben Luca's De physids ligaturis (fols
I. I owe this judgement to Michael Gullick. The MS is no. 349 in R. Gameson,
The Manuscripts ofJ?a,rly.Norman England (ca. 1066-1130), London, British Academy,
1999, p. 94, where It IS listed as being of the first half of the twelfth century.
2: See N.R. Ker, Medieval Libraries of Great Britain, London, The Royal Historical
Soaety, 1964, p. 85. For further information on the manuscript see Constantine the
African and al, ibn al-abbas al-Magusi: the Pantegni' and Related Texts, ed. C. Burnett
and D. Jacquart, Leiden, E. J. Brill, 1994, pp. 342-43.
3. The han~ and scribal practices do not resemble those of manuscripts known to
ha~e bee~ wn~en at Battle Abbey, but rather a group of similar-looking manu-
scnpts wntten 1n Bath. I owe this judgement to Michael Gullick and Tessa Webber.
For illustrations from three pages (fols 2V, 163v and 200V) see Burnett, The Introduction
ofArabic Learning into England, London, The British Library, 1997, pp. 26-28.
11
Physics before the (Physics' 55
200V-20IV), the De metallis (fo1. 20IV), De dbis (ff 20Iv-202r) and De
ponderibus (fo1. 202r-v).
Cotton Galba E IV (= G) is a composite manuscript. The rele-
vant codex (fols 187-244) comes from the Benedictine Abbey of
Bury St Edmunds, as is clearly stated (in a later hand) on the first
page of the twelfth-century portion:
Liber monachorum sancti Edmundi in quo continentur libri .XXIIII. de
medicina, de herbis, pigmentis, oleis et huiusmodi.
Recent palaeographical analysis has placed this manuscript in the
mid-twelfth century.4 It was written in two contemporary hands
(fols 187r-228r, and fols 228r-244v respectively). Both appear to be
professional scribes, who have copied the texts somewhat hastily,
but their copies were then carefully corrected by a contemporary
hand before the bifolia were stitched together.5 The manuscript
begins with a treatise on the division of philosophy (fols 187 r- 189 v :
«Sciendum est quid sit philosophia, quce eius materia, quod offici-
um [ ... ] »).6 The loss of a quire has resulted in the absence of the
end of this text, and of the beginning of the next, which is the
4. E. Parker McLachlan, The Scriptorium ofBury St Edmunds in the Twelfth Century,
New York-London, Garland Publishing, 1986, p. 323. For the contents of the man-
uscript, see also R. Thomson, ~iber Marii de Elementis': The Work of a Hitherto Un-
known Salernitan Master, «Viator», 3 1972, pp. 179-89. The contents are known from
the library catalogue ofJohn Dee, who possessed the manuscript when it still had
the now-lost items: see RJ. Roberts-A.G. Watson,john Dee's Library Catalogue, Lon-
don, The Bibliographical Society, 1990, MI44 and p. 125·
5. See e. Burnett et al., Adelard of Bath: Conversations with his Nephew, Cam-
bridge, Cambridge Univ. Press, 1998, pp. XLIII, XLV and L, where the nature of
the copying is discussed in respect to Adelard's Questiones naturales. A fra~e~t
of an unidentified medical text was copied in pencil across the top of a blfoli-
um before it was stitched (see the top margin of fols 22IV and 220r): «[ ... ]tac-
tum; per naturalem fr(ig}iditatem facit eucrasiam vel discrasiam, per acciden-
talem vero facit constrictionem pororum: similiter sulet (?) ignis aer et terra se-
cundum suas qualitates naturales vel naturales, unde eorum naturales per. natu-
rales, et accidentales per accidentales reperiuntur vel temperentur, vel dlstem-
perantur». . .
6. The extant portion deals mainly with the elements. Hence the text IS edited
in R.e. Dales under the tide Anonymi (De elementis': From a Twelfth-Century Collec-
tion of Scientific Works in British Museum MS Cotton Galba E. IV, «Isis», 56 1965, pp.
174-89·

11
Liber Marii de elementis (fols 190r-200r). Then come the same De ele-
mentis as in MS A (fols 200V-20IV), a translation of Hippocrates'
Book of Airs, Waters and Places (called here «de aere et aquis»; fols
20IV-204v), Alfano's translation of Ne me sius's On the Nature oJMan
(the Premnon physicon; fols 204v-214r), Adelard of Bath's Questiones
naturales (fols 214r-228r), the Physiognomia de tribus auctoribus (fols
228r-233v), and incomplete copies of the De spermate of Pseudo-
Galen (fols 233v-238v) and pseudo-Soranus, Isagoge (Questiones me-
dicinales; fols 238v-244v). The latest work may be the Liber Marii de
elementis, if, as I have argued elsewhere,7 it is by Marius Salerni-
tanus, the teacher of Gottfried of Viterbo. Since Gottfried was
born ca. 1125, Marius was teaching in Salerno ca. 1145-1150, and
possibly earlier. The now lost part of the manuscript once includ-
ed texts by authors which are unattested elsewhere: Euphon,
Adamarius, John the Melancholy, Wiscard and Picot. They may,
therefore, be local writers, for a « Picot medicus» is attested as a
member of the household of the earls of Gloucester in the mid-
twelfth century,8 and a « Guichardus» (= "Wiscard"?) was Ade-
lard's authority on the abacus. 9 If this is so, the «experimenta Ab-
batis» which follow the «experimenta Wiscardi» could be the
medical experiments of Anselm, the Italian abbot of Bury St Ed-
munds from 1121-1148, who may himself have been responsible for
bringing the early texts in the manuscript from Italy.10
7. e. Burnett, The Works of Petrus Alfonsi: Problems of Authentidty, «Medium Ae-
vum », 66 1997, pp. 42-79; see also pp. 76-77 below.
8. He attested Robert, Earl of Gloucester's foundation charter for Margam
Abbey in 1147, and a deed of William, Earl of Gloucester between 1166 and 1183:
see E. Kealey, Medieval Medicus, Baltimore, Johns Hopkins University Press, 1981, p.
142·
9. Adelard of Bath, Regule abad, ed. B. Boncompagni, Intorno ad uno scritto inedito di
Adelardo di Bath intitolato 'Regule abad', «Bollettino di bibliografia e di storia delle
scienze matematiche e fisiche », 14 1881, pp. 1-134 (100).
10. Anselm was the nephew of St Anselm, archbishop of Canterbury, and not
only was born in northern Italy, but also was abbot of the Greek-Latin double
house of St Saba in Rome before coming to England. See R.M. Thomson, The Li-
brary ofBury St Edmunds in the 11th and 12th Centuries, «Speculum »,471972, pp. 6rJ-
45 (63 0 and 634).
11
Physics before the Physics' 57
2. THE DE ELEMENTIS
The text shared by both manuscripts is the De elementis (this text
will henceforth be referred to as the «Additional De elementis»).
This short work was first edited, from MS G alone, by Richard
Dales who identified it as a translation of the chapter on the ele-
ment~ in Nemesius's On the Nature of Man. ll He published the
text and noted that the translator aimed at a literary rather than a
lite:al style, and avoided references to God and the Bi~le. He .also
realised that it had nothing whatsoever in common Wlth the hter-
al translation made from Greek by Alfano, archbishop of Salerno
from 1058 until his death in 1085,12 or the revision of Alfano's text
made in 1165 by Burgundio of Pisa.13 Nevertheless he assume~
that the translation was made from Greek, as did Moreno Moraru,
in his comprehensive analysis of the manuscript tradition of On
the Nature of Man. 14 Recently, however, Hermann Grenseman~
and Ursula Weisser have demonstrated in detail that the AddI-
tional De elementis was translated from Arabic. 15 An Arabic origin
is already suggested by the use of certain terms and phrases, suc~
. 16 and th e p hase
as the verb "inuenire" for "to eXlst", r " pars partt"
11. R.e. Dales, An Unnoticed Translation of the Chapter De elementis from Nemesius'
'De natura hominis' «Mediaevalia et humanistica », 13 1967, pp. 13-19.
12. Alfano of S~erno, Premnon physicon, ed. e. Burkhard, Leipzig, .Teubner, 1917.
13. Nemesius of Emesa, De natura hominis, trans. Burgundio of Pisa, ed. G. Ver-
beke and J.R. Moncho, Leiden, E. J. Brill, 1975· . ., . .'
14. M. Morani, La tradizione manoscritta del 'De natura homlnls dl Nemeslo, Milan,
Pubblicazioni della Universita Cattolica del Sacro Cuore, 1981, p. 61. .
15. H. Grensemann and U. Weisser, Iparchus Minutiensis aJ!as Hipparchus Metapo~­
tinus. Untersuchungen zu einer hochmittelalterlichen lateinischen Ubersetzung von NemeslOs
von Emesa, De natura hominis, Kapitel 5. De elementis, Bonn, Dr. Rudolf Habelt
GMBH, 1997. This monograph includes an edition of the De elementis ~ade from
MS G and Munich, Bayerische Staatsbibliothek, Clm 25353, the ArabiC text of
chapter 5, transcribed from MS Damas~s, zahiriya, 'a~ 2871, and the. Greek
text set in parallel with a German translation of the ArabiC text .and the Latin text.
See 52-58 for evidence that the Latin text is translated from ArabiC rather than from
Greek.
16. In the following notes the numbers in italics refer to the section-numbers. of
the edition of the Additional De elementis in Appendix I below, and Alf. = the eqUlv-
alent in Alfano's translation. «Inuenire» corresponds to the Arabic root W~J-D ~=
"find"): 29 and 41; cfr. udinuentionen 5 = al-mawjudat; cfr. TcX oV'tcX (eXlstentia
Alf.).
II
58
for "each other".17 And this Arabic origin is conf1fl11ed by the cor-
respondences between the text and one of the extant versions of
Nemesius's work in Arabic.
In the Arabic tradition Nemesius's work was translated at least
8 elements (24-27). The author of the De elementis is clearly confu~ed
twice.t An earlier version appears in the company of the Sirr al-
khalfqa of ~seudo-Apollonius, a work probably already extant by
th~ e.arly nInth century.19 In the late ninth century, one of the
pnnClpal translators of technical works from Greek into Arabic
working in Baghdad, Isba,q ibn I:Iunayn (d. 910), made another
translation. Isbaq ibn I:Iunayn's translation appears to be inde-
pendent of the earlier version; it has survived in four manuscripts
and is attributed to Gregory of Nyssa (as was the Greek text used
~y Burgundio). It is this version that is comparable to the Addi-
bonal De elementis.20
The ~~ditional De element!s does not give a literal rendering of
NemesIus s text. Some sectIons have been summarised, others
changed in ways that will be discussed below. That the De elementis
is a translation, rather than a treatise based on a previously-made
Latin translation, is suggested by the fact that, on two occasions,
two attempts at translating the same phrase remain in the text (in
20 and 44). The immediate source of the Latin text cannot be iden-
tified. Nevertheless, it is possible to see evidence of its ultimate
derivation from Isbaq ibn I:Iunayn's translation because of features
that it shares with the Arabic version over against the Greek.
17· "Pars parti" 6) translates ba~u f. .. ] barqan, the usual phrase in Arabic for "each
other", but ba~ lit~rally means "part"; cfr. illtlAwv (adinvicem Al£).
~8. For the versions of the text see Metri Haji-Athanasiou, Le traite de Nemisius
d'Emese 'De Natura Hominis'dans la tradition arabe, 2 vols, Ph.D, Paris, Universite Pan-
theo~-~~rbonne (Pa?s I~, 19~~, vol. 11 pp. XLIX-LVIX; K. Samir SJ., Les versions arabes
de Nemestus de lJoPn$, m Leredrta classica nelle lingue orientali, ed. M. Pavan and U. Coz-
zoli? 1st. della Enciclopedia ltaliana, Rome 1986, pp. 99-152, and Grensemann and
Welsser, Iparchus Minutiensis, pp. 1I?-31.
19· ~s incomplete version has been edited by U. Weisser alongside the Si" al-
khaliqa m Buch uber das Geheimnis der Schiipfung und die Darstellung der Natur (Buch der
l.!rsach:n) von Pseudo-Apollonius von Tyana, Aleppo, Institute for the History of Ara-
biC Saence, 1979, pp. 53?-632.
20. I have u~ed t~e edition provided in Haji-Athanasiou's dissertation [see note
18 ~bove]~ which mcludes the variants of four manuscripts; Grensemann and
Welsser did not use this edition.
11
Physics before the Physics' 59
These include additions,21 and differences in content in respect
to the Greek text. The most significant of the latter occurs in the
account of the Platonic regular solids associated with each of the
at this point; he shows no familiarity with Calcidius's explan~t1on
of the situation.22 What is obvious, however, is that he conSIders
the cubic form associated with the earth to derive from two-di-
mensional figures "composed of four sides" (i.e., rectang~es), and
the solid forms associated with the other elements to denve from
triangles, whereas Plato 23 (and Nemesius following him) makes
the distinction between isosceles triangles (for the cube) and sca-
lene triangles (for the other figures). The distinction in the De ele-
mentis is exactly what we find in the Arabic.24
The translator of the Additional De elementis disguises the fact
that he takes the text from Arabic by avoiding any Arabic tran-
scriptions, and by occasionally substituting Greek words. When
he makes these substitutions, he sometimes chooses the wrong
words.25 Nevertheless, his use of words such as "chaos" (8) and
21. E.g. 3 Et singula a singulis alia plus alia minus accipiunt = wa akhad~ kullu
wa~idin minha min al-akhari miqdaran ma kathfran wa (read akt~ara aw!) aqalla; mstead
of this phrase the Greek merely adds .~~ov xal Tlt'tov (~agts ~t mmu~ ~£). to t~~
previous phrase; 27 et cum aqua pulVls uru~r = wa yatta~rdu (sal. al-turab ) br-I-ma,
the Greek omits this phrase; 46 a quo alia elementa procederent = wa huwa al-
mawlud li-l-'anasir al-ukhar; the Greek omits.
22. Timaeus ~ Calddio translatus commentarioque instructus, ed. J.H. Waszink, Lon-
don, Warburg Institute, 19752 , pp. 321-22.. . . , .
23. Timaeus, p. 54- This passage is beyond the pomt where Caladius s translatton
breaks off, and is notoriously difficult to understand.
24. Compare De elementis 26: «Harum ergo trium ?gurarum. (~f the other three ele-
ments) compositio est triangula, equalem circundans line:u?, chiblcus autem rectam
circundans line am {possibly read: chibicis autem <comp~sltto qua~angul~ e~ualem
circundans lineam}., with the Arabic: kawn al-ka1J mm murab~aat m~tasawryyat al-
a4la' (<< the formation of the cube is from rectangles of equal Sides ~), m co~trast to
the Greek: 'tov 6£ xupov e~ i.o01tAeUpWV 'tp1.yciNwv ("but the cube IS from Isosceles
triangles"); and De elementis 26 «qui eorum (the other three soli~) n~llum ~st e _qua~­
tuor angulis compositum., wit~ the Arabic: innahu laysa sha u: ~mhu br-tarkrb mm
murabba'at "that none of them IS composed from rectangles, m contrast to the
Greek: ~*' au 'tGJJ 'tP1.GJJ 't1. OXTl~wv ei.<; xupov ("none of these three shapes <can
be built up> into a cube"). .. . "
25· 25 "Organorum" is used wrongly for "solids/three-dimensIonal bodies
11
60
"ether" (4), and the recognition of the Arabic phrase "the people
of the portico" as the "Stoici" (3), suggest that the he had some
knowledge of Greek. These words (along with "petra" where Al-
fano uses "lapis", and "organum") were already used in Latin. But
there are two further words which suggest a more direct acquain-
tance with Greek:
I) The spelling "chibicis" 25 for "of a cube", which may reflect
the medieval pronunciation of the Greek XUP1.)dic; as the genitive
of an adjective qualifying a feminine noun, though this form does
not occur in this context in Nemesius.26
. 2) The use of the term "edra" (feminine singular) for (geomet-
ncal) base in 25-2 6. This is the Greek word fapcx, whose primary
meaning is "seat", and which appears in the compounds Ox't&£?
apov ("octahedron") and eixoa&£apov ("eicosahedron"), which are
transliterated by Alfano as "octaedron" and "icosaedron".27 But the
word fapcx is not used on its own in this context in Nemesius. In
the Arabic, however, the compounds are resolved into their con-
stituent words, so that oX't&£apov becomes dhu al-thamanf qawalida
(( possessing eight bases»). Qalida (plural qawalid) comes from the
root meaning "sit", and it is, perhaps, not surprising that a transla-
tor with a knowledge of Greek should substitute the Greek
equivalent fapcx, dropping the aspiration (as his contemporary
Greek speakers would do) but keeping the feminine case (which
the w~rd .loses in the neuter compounds mentioned above). This
comblnanon of an Arabic translation with a Greek veneer should
be borne in mind.
The Latin style of the Additional De eiementis cannot be said to
be elegant or accomplished. Cursus is not employed, and the syntax
is a bit clumsy: the writer tends to use participles instead of simple
verbs (e.g. "sint inherentia" 10 rather than "inherent"), to omit the
antecedents to relative clauses, and to omit forms of "esse" (some-
(<rCepe<4ux-ra, al-ashka~ al-mujassam~); 38 chaos = hawiya; cfr. &puooo<; (abyssum Al£).
The translat~r, followmg .the Arab~c, adds a phrase which, understandably, is not in
the Greek: «Id est chaos m greca lingua» (= hawiya bi-lisan al-yUntiniyfn).
26. See note to the translation, p. 97 below.
27· In Calcidius's discussion of the same passage (ed. Waszink, p. 321), the terms
are "octahedron" and "icosahedron", alongside "pyramis" and "cubus".
II
Physics before the Physics' 61
times these have been supplied in MSS MG, e.g., in 7 and 31). Nev-
ertheless, there are examples of variatio and poetic paraphrase. For
variatio (where different expressions are used to translate the single
term used in the Greek and Arabic texts), see 10 "copulatur",
"similis est", "coniunctus" and "non dissidet" (for auvOOt't£'tCX1.,
y~lu); 15 "fit", "est", "mutatur in", and "eat" (for yiveraz, ¥ira); ~5
"arbitratur", "autumat", "attribuit" and "existimat" (no verb In
Greek; shabbaha); and 44, 45, and 46, "putaret", "dixit", and "in-
tellexerunt" (for AeyWV, qala [twice]/ra'a). For poetic paraphrases of
single words in the Greek and Arabic, see 12 "in petrre rigorem
transit" (for OOtOAtlloihcx1., istalyarat); 20 "quicquid terrre radicitus est
affixum" ('ta q>u'ta, al-nabat). Other characteristics of the writer are
the expression "nulli dubium est", the use of "ut" for "e.g." or "i.e."
and "per hoc" for "because of this", and the words "universaliter",
"ratus", "colligata", "consistere" (for "esse"), "determinare" (rather
than "definire") and "caret" (rather than "indiget").
In regard to contents, it must be stated first of all that the Ara-
bic translation by Ishaq and the Greek text of Nemesius are closer
to each other than either of them is to the Additional De elementis.
The Latin text summarises several passages, and develops an argu-
ment which has the concept "element" more at the centre of at-
tention than it is in Nemesius's text. Having presented in detail
the usual view that the elements are four, and described their
qualities (with references to the opinions of Aristotle, Plato and
the Stoics), the author mentions that Aristotle adds a fifth element
(4). Plato, however,28 considers the heavens to be made out of
earth and fire, and derives the other two elements from these (5).
But certain authorities propose another element - chaos - as the
origin of the four elements 06; the description of "chaos" as an el-
ement does not appear in Nemesius}. The text ends with the
proofs that there must be more than one element, and the argu-
ments again the philosophers who believed that everything de-
rived from one element alone.
What is most striking about the contents of the Additional De
elementis, however, is that, as Dales has already pointed out, the
28. The author seems to be more in favour of Plato's opinion.

11
62
writer avoids almost all mentions of God, and disguises quotations
coming from the Bible, while keeping the attributions to pagan
Greek philosophers; in this he does not follow Isbaq, who, as a
Christian himself, presumably had no qualms about keeping the
Biblical quotations. E.g., in 11 and 20 the translator of the Addi-
tional De elementis substitutes "nature" for the Demiurge of
Nemesius (translated al-bari', a Koranic word for "Creator", by
Isbaq), in 36 he refers to the opinion of the Hebrews as that of
« nonnulli », and in 3?-38 the Bible 29 becomes « quidam codices»
and the name "Moses" is omitted.30 It is difficult to see why a
Christian translator would have made these changes. It is more
plausible to surmise that they had already been introduced into
the Arabic text from which the Additional De elementis derived,31
and originated in a milieu in which a Hellenism was cultivated
that was regarded as independent ofJudaeism, Christianity and Is-
lam, i.e. a philosophical milieu. With this point in mind, we may
turn to the other texts in the Additional manuscript.
29. In Arabic, « the <revelation> delivered by Moses at the beginning of the
Torah».
30. Note also that there is no mention in the Additional De elementis of either
Nemesius, bishop of Emesa, or Gregory of Nyssa, the great Church Father, as au-
thors.
31. Whether the De elementis, in itself, was meant to be a self-contained text is
also unclear. One cross-reference to an earlier chapter (chapter 2 or 3) in Neme-
sius's work remains: see 23. A text with the same title as chapter three of Neme-
sius's work - De unione corporis et animae liber unus - is mentioned as a work of Al-
fano, archbishop of Salerno, in Peter the Deacon's De viris illustribus. Pietro Cap-
paroni {11 TIe Quattuor humoribus corporis humani' di A!{ano 1° Ardvescovo di Salerno
(Sec. XI), Rome, Istituto nazionale medico farmacologico « Serono», 1928, p. 12),
claims quite plausibly that this is an excerpt from the Premnon physicon, but it is
possible that more chapters from an Arabic-Latin version of On the Nature of Man
were circulating in the Middle Ages. This could explain inter alia how Arnau de
Vilanova was apparently quoting Galen's definition of the soul from On the Na-
ture of Man, chapter 2 - «animam esse miscibilium armoniam» - in words
which correspond to neither Alfano's nor Burgundio's translation of the text: see
Arnau de Vilanova, De intentione medicorum, ed. M. McVaugh in Arnaldi de vil-
lanova Opera omnia, v. I, Granada, Seminarium historiae medicae granatensis,
1999, pp. 171-72.
11
physics before the Physics'
3. QUEDAM PHISICA
On folio 165 r of MS A, after the list of chapter-headings for Part
I of the Practica Pantegni, the scribe notes that « certain (~orks on)
natural science follow at the end of this (book) on (medical) prac-
tice» (<< Quedam phisica sequ(un)tur in ultimo huius pra~cre»).
That the word "phisica" is here simply the ~;eek ,:ord equ1valent
to the Latin "naturalia" ("concerning nature) and 1S not me~nt to
imply specifically the medical science, as it ,:as ~o do later 1n the
twelfth century,32 is shown by the texts to which 1t refers. The first
of these is the De physicis ligaturis of Costa ben Luca,. a ork co~­ w:
cerning the reason why magical recipes are effective. 1n ed1- n:
cine.33 The Arabic original of this work has not been 1dentlfied,
but there is no reason to doubt its attribution to the well-known
ninth-century Melkite translator and scientist, Qust a ibn L~a.
The Latin translation was already known to Marbod of Rennes 1n
ca. 1090 ,34 This work is philosophical rather than medical, q~ot­
ing, in addition to Galen, Cleopatra and Diosco~d~s as. medical
authorities, Plato, Socrates, Aristotle, and the « hbn antlquorum
philosophorum ». . '
After this follows the De metallis (edited 1n Appendix II below),
on the origin of the seven metals and thei~ ~ssociation ~th the
seven planets. This has no relevance to med1C1ne at a~l, but 1S a re-
markably early Latin text on mineralogy. Its doctr1ne does .n?t
correspond exactly to that of any other text on metals. The ?ngtn
of metals from vaporous and smokey exhalations trappe~ 1n the
bowels of the earth can be found in Aristotle's Meteorologlca 378a.
32. The compilers of the list of works of Ger~d of Cre~ona (nI 4-n87) used th~
title « physica» to describe the category of medical translattons: see K. S~dh?ff, Die
kurze Vita' und das Verzeichnis der Arbeiten Gerhards von Cremona, ~ Archiv. ~r Ge-
schichte der Medizin», 8 1914, pp. 73-82, and it is significant that, ffi"a re,:-s~~~ of a
translation made in Toledo in the mid-twelfth century, the term medi~ l~ re-
placed by that of"physici": see Abii Ma'sar, On Historical Astr~/ogy (D~ magnrs conJunc-
tionibus), IV 178, eds C. Burnett and K. Yamamoto, 2 vols, Lelden, Bnll, 2000, vol. 11,
p. ;~3'The text is edited by J. Wilcox and J.M. Riddle in Qusfii ibn Luqii's Physical
Liga;ures and the Recognition of the Placebo Effect, «Medieval Encounters», I 1995, pp.
I-50.
34. Ibid., pp. 22-23·
11
But the identification of these exhalations with quicksilver (mer-
cury) and sulphur respectively is a development within the Hel-
lenistic and Arabic alchemical tradition, found in Pseudo-Apollo-
nius's Si" al-khaltqa,35 and Avicenna's kittib al-Shifo',36 and hence in
the Latin texts ofHermann of Carinthia's De essentiis (written 1143)
and the Liber Marii de elementis.37 The account in the Additional De
metallis does not depend on either of these Latin works, and most
probably has a direct Arabic source. Characteristic of its doctrine
are the description of the compressed water (identified with
quicksilver) as "heavy and viscous", and a rigorous classification of
the ingredients of the metals in terms of brightness, clarity, weight
and hardness. Above all, the Additional De metallis assimilates this
doctrine to the association of each of the metals with each of the
planets. The integrity of this assimilation is shown by the fact that
the De metallis gives the rationale of the planet-mineral association
in terms of the same criteria of brightness, clarity, and weight, on-
ly substituting elemental qualities for the criterion of hardness.
While the association of metals with planets goes back to classical
times,38 the assimilation of this association with the Peripatetic
doctrine of the origin of metals is a later development, found in
the Si" al-khaltqa and Hermann's De essentiis. Moreover, the De
metallis mentions "res" ("brass") as the mineral of Venus rather
than "cuprum" ("copper"), which recalls Hermann's statement
that brass ("auricalcum"), too, is attributed to Venus.39 The De es-
35. See P. Travaglia, in Aristoteles Chemicus. 11 IV Libro dei "Meteorologica" nella
tradizione antica e medievale, Atti del seminario di studio italo-francese, Venezia, 1-3
dicembre, 1999.
36. For the Arabic, Latin and an English translation of the relevant text, see EJ.
Holmyard and D.e. Mandeville, Avicennae de congelatione et conglutinatione lapidum,
Paris, P. Guenther, 1927. A new edition of the Latin text is provided in R French,
Teaching Meteorology in Thirteenth-Century Oxford: The Arabic Paraphrase, «Physis», 36
1999, pp. 99-131 (121-29).
37. Marius: On the Elements, ed. Re. Dales, Berkeley-Los Angeles-London, Univ.
of California Press, 1976, pp. 152-55.
38. The sources are most conveniently brought together in M. Berthelot, Intro-
duction a l'etude de la Chimie des anciens et du moyen age, Paris, G. Steinheil, 1889, pp.
73-86.
39. See De essentiis, fols. 75vH-76rA (ed. e. Burnett, Leiden, EJ. Brill, 1982, p.
206): «Unde submoto statim igne, soluta materia vi terre vel aque astricta, sicque
11
Physics before the (Physics'
sentiis was written too late to have had any direct influence on the
De metallis, but the similarities suggest that both works derive from
the same kind of sources, and that these sources were Arabic. The
De metallis was recognised as being relevant to alchemy since, in a
thirteenth-century manuscript written in Italian hands - Bodleian
Library, Digby 121 - it was prefIXed to a copy of Geber's Liber sep-
tuaginta, an important alchemical work translated by Gerard of
Cremona {d. 1187).40
That the De metallis originates from the same Latin milieu as the
Additional De elementis is indicated by the fact that the creative
force in both works is described as «sapientia Dei ». In this short
text there is not so much scope for variatio, but there is a hint of it
in the author's use "est simile" and "similitudinem seruat" in 4 to
break the monotony of the repeated "imitatur" (perhaps all repre-
senting the Arabic shabbaha).
The De metallis is followed by De dbis (edited in Appendix II be-
low) which gives a list of one hundred species of plant cultivated
for food, which can be divided into ten different categories of ten
plants each.41 In medical texts plants are usually classified accord-
ing to their elemental qualities (as in texts with the title De
gradibus) or according to their effects ("constipative", "emetic",
etc.). The classification in this text is more akin to that in encyclo-
pedias of natural science. The origin of the text is unclear. Several
in maternam soliditatem rediens, procedit in formas, queque sui generis - Solis si-
quidem aurum, Lune argentum, Saturni plumbum, Iovis stannum, Martis..ferru~,
Veneris cuprum, Mercurii argentum liquidum, quod pro natura Mercuru cet~~s
instabilius. Est autem contentio de cupro inter Martem et Venerem, Venen m
sortem dato auricalco, quod quoniam pars cupri, quemadmodum et calebs fern,
nobis huiusmodi distributio visa». There is some fluidity in the significance of the
words "cuprum", "res" and "auricalcum" in the Mi~dle Ages.. ..,
40. The text is found on fo1. Ir-V of the manuscnpt, under the otle «Inaplt liber
primus de .VIl. generibus metallorum». The text is immediat~ly followed ?y ~.e
rubric «Incipit liber principiorum divinitatis» w~ch ?egins Wlth th~ usual maplt
for the Liber septuaginta: «Laudes sint deo habeno graoam [ ... ]~. This .early manu-
script of the Liber septuaginta does not seem to have been nooced. Richa:d Hunt
corrects the date to "s. XIII" in his notes to the Quarto Catologue of the Dlgby col-
lection, reprinted 1999, p. 63· .
41. These numbers are not, in fact, exactly correct, since eleven categones. ~e
listed, and some categories contain nine items, others eleven, but whether this IS
because of scribal errors, or the author's inability to count, is unclear.

11
66
* of the names are Greek ("facus", "alica", "attriplex", "lapsana", "xi-
10 [carrata]" and « caulis [ ... ] que Grece dicitur siccocambri »), at
least two names are Arabic: "fistic" = "pistachio", and "musa" =
"banana", while two others that are not in the dictionaries also ap-
pear to be Arabic: "zarur" and "fafelus".42
The last of these «quedam phisica» - De ponderibus - is also en-
cyclopedic in nature, but is the only explicitly medical text; for it
applies the information of the encyclopedias to medicine. It con-
sists of a rearrangement of the material in Isidore, Etymoiogiae, XVI
xxv (De ponderibus) and XXVI (De mensuris), to which it adds an in-
troduction,43 and a final section specifying the medical uses of
Isidore's measurements.44 Although other manuscripts apparently
attri~ut~ the text to Constantine the Mrican,45 there is nothing
Arablc In the text; the use of "denarii uictoriales" as a point of ref-
erence, rather, points to a late Antique provenance.
4. THE TRANSLATIONS FROM ARABIC IN MSS ADDITIONAL 22719
AND COTTON GALBA E IV
In MS A the De eiementis, as we have seen, is accompanied by
other translations from Arabic: the De physids ligaturis of Qusta ibn
42. Other names that are not in the dictionaries are "brucuca"' "mala cocila",
"0linica", "uaIia", "mala citonia", "molongianre", "cambricombos", "terfes", "me-
dica", "dolica", "mauricocus", "portulaca", and "molothi"
4~. MS. A, fo1. 202r. « P~ndera medicinalia signa conati sumus narrare, que a
plensque 19norata sunt, promde errores legentibus faciunt. Quapropter formas eo-
rum et caracteres ut a ueteribus signata sunt subicias, quanuis diversre gentes huic
mensurre pondus uel adiciant ignorantes, uel detrahant nescientes,. (cfr. Isidore,
Etymologiae, XVI XXVII, I).
44. MS A~ fo1. 202V: «Medicinalis autem libra habet uncias .XII., libra habet drag-
mas .XCVI., hbra habet scrupulos .CC.LXXXVIII., libra habet obolos .DLXXVI., libra ha-
bet siliqua~ tria mi.li.a .CCCXII., libra. pensat argenti solidum .xxIIII., libra pensat
grana ordel .XIII. mllia. CCCLXVIlI". Similar sets of equivalents in smaller denomi-
nations are then given for "media libra", "uncia", "semiuncia", "ceratis" "sexta-
rium", "siclus", "amphora", "silicus", "nux qure habet magnitudinem auell~re" and
"nux maioris magnitudinis". For other similar texts see F. Hultsch, Metrologicorum
scriptorum reliquiae, 2 vols, Leipzig, Teubner, 1864.
. 45. ~ee L. Thorndik~-P. Kibre, A Catalogue oflndpits oJMediaeval Soentific Writings
In Latin, London, Mediaeval Academy of America, 1963, p. 1058, where the earliest
manuscript mentioned is Vendome 174 (s. XII).
11
Physics before the Physics'
Luqa, and the Pantegni of Constantine the African, which is based
on the Complete Book on the Art of Medidne of ~i ibn al-'Abbas al-
Majusi, with additions and substitutions from other A~abic .~d
Latin medical works; the De metallis may have an Arablc ongtn,
and the De dbis includes Arabic names of plants.
In MS G, the Additional De elementis precedes a version of Hip-
pocrates, On Airs, Waters, and places. It has been sug~ested: since the
researches of Freudenmann and Diller, that this verSlon (( the
younger Latin version»; called in this article «De aere et aquis»
after its title in MS G) was translated from Arabic.46 After the
publication of the Arabic version of Airs, Waters, ~nd places (mo.st
likely by l:Iunayn ibn Isi)aq, the father of Isi)aq lbn l:Iunayn) In
1969, these suggestions can be confirmed, and it is even possible to
identify which of the two variant versions of the Arabic text was
used by the Latin translator: namely, version 11.47 The Arabic ver-
sions and the Latin· translation differ from the Greek text in that
46. K. Freudenmann, Beitrage zur Uberlieferung der hippokratis~hen Ab.ha~dlu~g nepi
aepwv UOclrWV ,onwv, Diss., Tlibingen 1922, p. n8ff., and H: D~ller, Dre Uberlreferung
der hippokratischen Abhandlung nepi aepwv UOclrWV ,onwv, LeipZig, Teubner, 1932, pp.
105-35 (= H, cited by section number), which develops Freudenmann's ar~ments.
For earlier Latin versions of the text, made directly from Greek, see H. Dlller, and
H. Grensemann, Hippokrates De aeribus aquis loos, interlineare.:4usgabe der spatlateini-
schen Ubersetzung und des Fragments einer hochmittelalterlichen Ubersetzung, Bonn, Dr.
Rudolf Habelt GMBH, 1996.
47. Hippocrates, On Endemic Diseases (Airs, Waters and Places), e~. and tr~s1. by
J.N. Mattock and M.C. Lyons, Cambridge, Heffers, 1969 (ArabiC :rec~cal and
Scientific Texts,s). Mattock and Lyons were unaware of the relatlonship o~ the
Latin translation in the Galba manuscript to the Arabic texts. The only differ-
ences I have noted are that, after fu8, H omits a section (Ar. ed., pp. 4110_4514);
in H64, the phrase «fumus ad cerebrum ascendit et circumtegit il~ud,. does .not
occur in the Arabic text here (109), but can be found in the ArabiC text equiva-
lent to HI2 (pp. 18 12-13), where it is missing in the Latin; in H76 ~ pecora sua ca-
strant et armenta, ideoque nimis quieta sunt bonumque habent alimentum,. does
4
not correspond to the Arabic text, which is difficult to understand here (pp. 121 -
114 8
5); in H 84 the Latin omits «the fruits [ ... ] do not ripen,. (pp. 13 ); in H9 the
Latin omits «because of the moistness [ ... ] bellies,. (pp. 145 15-16) and « because
they continuously ride horses, their desire is weakened,. (p. 1471); moreover, t~e
Latin omits some repetitions and obscure proper names. On .the other ~and, In
H62 the Latin version is more coherent than any of the ArabiC manuscnpts (93-
95). For further examples of the similarities of the Latin version to Arabic ver-
sion 11, see note 63 below.

11
68
they silently incorporate several glosses deriving from Galen's
commentary on the text. Since Arabic version 11, as the editors ar-
gue, is a revision of version I, which is closer to the Greek, it is un-
likely that there existed a Greek text identical to version II.
Another text in MS G is the De spermate of Pseudo-Galen. This
is a wide-ranging text on embryology, and on the influence of var-
ious climates and different planets and configurations of the heav-
ens on the characters of individuals.48 Neither a Greek nor an
Arabic source has been found for this text, but an Arabic origin is
suggested by its alternative title Duodecim portce, in which "porta"
would be a calque on the Arabic word biib meaning primarily
"door", as well as being the usual word for "chapter".49 It is also
significant that De spermate appears to quote Airs~ Waters and places
under the title Hippocrates' Physica,50 and cites the opinions of Eu-
nomios and Theodoros, who are otherwise mentioned only in
Nemesius's On the Nature ofMan; the garbled form of the name of
the former authority ("Emnoinos"), and the lack of verbal corre-
spondence with Alfano's translation of Ne me sius, make it possible
that the author of the De spermate was using a Greek or Arabic ver-
sion of On the Nature ofMan independently from Alfano's transla-
tion.
The De physicis ligaturis, De eiementis, De metal/is, De aere et aquis,
4~. For a study of the contents of the text, and an edition of the Middle English
VefSl?n, .see P~~~ P~ta, flv!ediev~l Embryology in the Vernacular: the Case of'De spermate',
Helsmki, So~e~e Neophilologtque, 1998 (Memoires de la Societe Neophilologi-
que de HelsmlCl, LIn). A.much-needed edition of the Latin text has been planned
by Dr Pahta. Reference IS made to the copies in MS G, MS British Library, Addi-
tional 18210, and as printed in Galen, Opera, Basel 1542, vol. VIII pp. 133-54.
49. The use of "porta" for "chapter" occurs in an acephalous alchemical text of
Arabic derivation in Oxford, Bodleian Library, Digby 119, fols IIlr-1I5r and in Pseu-
~o-Avicenna's De an~ma in arte alchemica. Also, 1tUAll ("porta") is used for "chapter"
m the Greek translanon of Ibn al-Jazzar's Guide for the Traveller (see note 57 below).
50. Compare BL Add. 18210, fo1. 126v, with De aere et aquis, H103, and HIIO-III.
Note, ho~ever, that an unidentified « liber perhi physeos» is cited in the «epistola
Yppocrans de quatuor humoribus» in Paris, Bibliotheque nationale de France,
11219 (s. IX), fols 18v-20v: see E. Wickersheimer, Les manuscrits latins de mededne de
haut moyen age dans les bibliotheques de France, Paris, Editions du centre national de la
r~cherche scientifique, 1966, p. 115; see also P. Kibre, Hippocrates latinus v, «Tradi-
no », 35 1979, p. 277.
11
physics before the Physics'
and De spermate share two distinctive features: (I) they are not
works of practical medical use, but rather deal with nature, either
directly with the subjects of natural science, such as elements,
metals, the development of the embryo, and the causal role of the
planets and the signs of the zodiac, or with the relationship be-
tween natural causation and magic; (2) they all disguise their Ara-
bic sources and interlard their texts with Greek words and Greek
authorities.
Examples of the Greek veneer of the De aere et aquis are the use
of the words "chaos" (for Arabic al-hawii' = "air"), "pleuresim ac
peripleumoniam" (H68), "catarticum", "yreos" (for ''yeros'', the sa-
cred disease; HI3), and "lienteria, dissinteria et epielmos" (H 14).
That these Greek words are used for interpreting an Arabic text is
indicated by the cases in which the wrong Greek word has been
chosen: e.g. "causon" is used for both for xipoOl and x«\JoOl, be-
cause the translator has confused the similar-looking Arabic tran-
scriptions of the two words: qirsus and qusus. The Greek element
in the De spermate is even more evident. While it uses the same
names of diseases as the De aere et aquis {"causon", "dissenteria",
"lienteri{c)a" and "pleuresis")51 it also employs the words «praxis»
(or "braxis"), "phantasia", "petra", as well as favouring the Greek
word "sperma" over the Latin "semen" for the main subject of the
text.52 In one place there is a whole sequence of Greek names:
«armoponiam, metroponiam, id est dolorem ylii, coloptomiam, id
est dolorem matricis ».53 But the direct knowledge of Greek texts
on the part of the author is more questionable. E.g., he cites «Aris-
totle's Physics» for a statement concerning the five parts of the
soul, for which he gives Greek words. This is probably a confused
reference to the same author's De anima 414a33, in which various
powers (rather than "parts") of the soul are listed, but even here the
De spermate gives the wrong Greek words for two of these pow-
51. De spermate, pr. Basel, 149B, ed. Pahta, p. 231, and elsewhere. .
52. This, too, is found in De aere et aquis H82, whose phrase «QUla vero sperma
ab omnibus corporis descendit membris » is reminiscent in form and .v0cabul:ur of
the opening of the De spermate: «Sperma hominis descendit ex omru corpons hu-
more [ ... ]». . . .
53. See ed. Pahta, pp. 234-35, where variants in the Lann manuscnpts are gtven.

11
70
ers.54 This implies a context in which the Greek language was
known better than Greek philosophical texts themselves. It also
shows the author's/translator's preoccupation with situating his
work within natural science, which is similarly manifested in his
quotation of Hippocrates's "physics".
The assumed date of these translations, and this Grecizing ten-
dency, naturally brings to mind the translating activity of Constan-
tine of Africa, whose version of the Pantegni is the main text in MS
A. The style of Constantine is aptly characterised in the title of an
article by Gotthard Strohmaier: «Constantine's pseudo-Classical
terminology and its survival ».55 Among the terms mentioned
above, Constantine uses "pleurisis" and "peripleumonia".56 On
the other hand, as Daremberg pointed out nearly ISO years ago, a
Grecizing style was natural for Latin medicine in a period in
which it was heavily dependent on Greek texts, and any work
written before the early twelfth century is likely to have exhibited
this style. The more ancient Greek tradition, according to Darem-
berg, permitted translators to embellish their Latin with a quanti-
ty of Greek words, expressions and formulas which gave them an
appearance of great erudition, with the consequence that the
Latin medical tradition is full of Greek words, whether in transla-
tions, or in original works. 57
54. See MS G (quoted in ed. Pahta, pp. 220-21; cfr. pr. Basel, col. 14IA): «Aristot.
in physica esse quinque partes anime, phitice, id est, nutrimentum, stitice (restheti-
con, pr. Basel), id est sensus, quinitici catatopon (cineticon catatopon, pr. Basel), id
est uim ad omnia progrediendi, oretice (noeticon, pr. Basel), id est rationem ». The
Latin variants imply the Greek words <j>Ut1.XOV, cxioOt)t1.xov, X\VT)t1.XOV XCX'tIX 'tOOtOV,
0P£Xt1.XOV, VOT)'t\xov. For the first and the last of these De anima 414a33 gives
i}p£1tt1.XOV and 6\cxvoT)t1.XOV respectively.
55. In Constantine the African and 541, ibn al-54bbiis al-Magusl, pp. 90-98. See also G.
Baader, Zur Terminologie des Constantinus Africanus, «Medizinhistorisches Journal », 2
1967, pp. 36-53·
56. These two diseases are described in Pantegni, Theor. x, ch. 10, and are drawn
attention to in the margins of MS A, on fols 157V and 158r.
57. C. Daremberg, Recherches sur un ouvrage qui a pour titre Zad al-Moufafir, en arabe,
Ephodes, en grec, Viatique, en latin, et qui est attribue, dans les textes arabes et grecs, aAbou
Djafar, et dans le texte latin, a Constantin, «Archives des missions scientifiques et lit-
teraires, choix des rapports et instructions »,21851, pp. 490-527 (see 515-17 for sever-
al examples of this tendency).
11
physics before the Physics' 71
5. THE ORIGIN AND DIFFUSION OF THE ANONYMOUS EARLY ARA-
BIC-LATIN TRANSLATIONS
It is remarkable that none of the Arabic-Latin translations men-
tioned above is attributed to Constantine in the earliest ma~u­
scripts; and this lack of attribution is confirmed {with t~e poss1ble
exception of the De spermate}58 by the lists of the tran.slatlons made
by Constantine, drawn up by Peter. the Deacon and 1n .the Chron-
icle of Monte Cassino soon after his death. 59 Constantlne was no-
torious for advertising his own authorship, even of works of
which he was the translator rather than the author, and it :,oul~
seem strange that he did not clai~. any of t~ese translatlons.
Moreover, while the general Greaz1ng sty~e 1S ~ha~ed be~een
these works and the translations of Constantlne, s1gruficant d1ffer-
ences can also be observed. For example, the same passage from
Hippocrates's On Human Nature is q~ote~ in t~e Ad~itional ~e ele-
mentis and the Pantegni, but the word1ng 1S entlrely different.
De elementis, 40 -41
Pantegni, I 4, Paris, BN late 7402, fo1.
3V
Idem Ypocras in eodem: si homo, Sufficit nobis Ypocratis dictum, di-
inquid, de uno fieret elemento, di- centis: si enim homo ex uno tan-
versis nullis subdetur passionibus. . tum consistit, impassibilis est. Non
Quod enim in eodem sibi obviaret enim quod sibi obstet invenitur.
invicem? Item Ypocras: si, inquid, Quod si patitur, ab uno necesse est
diversa non pateretur, diversis me- ut curetur.
dicinis non reficeretur.
58. An alternative title for the De spermate - (Microtegni' -,occurs in both Pet~r,t~e
Deacon and the Chronicle of Monte Cassino (see followmg note); ~he ,posslblli~
that this may refer to the De spermate rather than Galen's Ar:s Parva,ls dl,scussed m
Burnett, The Chapter on the Spirits in the Pantegni of Constantine the African, m Burnett
and Jacquart, Constantine the African, pp. 99-120 (109). ,
59. The lists are conveniently edited in parallel columns m H. Blo~h, ~onte Cas-
sino in the Middle Ages, 3 vols, Cambridge (Ma)-Rome, Harvard Uruverslty Press-
Edizioni di Storia e Letteratura, 1986, I pp. 127-29. ,
60. The fact that there are no translator's prefaces to t~ese works can also be m-
terpreted as being contrary to Constantine's ~s';lal practice, though such prefaces
are often dropped in the course of the transffilss~on, of a ~ext. ,
61. One could object that the quotations, are mdire~ l~ bot~ ~es, b~t, m_fact,
the Arabic text is very similar (for the ArabIC of al-MaJusl, see :Alllbn al-~bbas ;1-
Majiisi, The Complete Medical Art, facsimile of MS A.Y. 6375, Frankfurt, Institute or
the History of Arabic-Islamic Science, 1985, Book I chapter 5, fol. 19r).
II
72
In discussing elements in the same chapter of the Pantegni, "parti-
~ula" is used where the De elementis has "pars" (both translating
JUz), and the phrase "constructio et destructio" where the De ele-
mentis has "generatio et destructio" for yev£cJ1.C; XCI! q>ttop«, kawn (or
tawallud) wa fasad; Alfano uses what became the standard Latin
translation: "generatio et corruptio".
Similarities can be found between the texts mentioned here.
The De physids ligaturis shares with the Additional De elementis a
predi~ection for quoting Greek authorities (see above, p. 63), but
mentlons no Biblical authorities. The De spermate, likewise, cites
Ammonius, Democritus, Porphyry, Socrates, Plato, Aristotle,
Theodoros, Eunomios, and Hippocrates, but no Church Father or
Biblical text. Similarities in style and vocabulary can also be ob-
served between the Additional De elementis (E) and De aere et aquis
(H). Both texts use "complexio" (lli), "ratus" (Hlo) , "uniuer-
saliter" (H80 and 104), "inuenitur" (H4 6) , "fumus" (H64),62 and
variatio and paraphrase are found in De aere et aquis: cfr. Hn "sem-
per ab eis incutitur" for "these (winds) blow continuously"; "pueri
tarde polluuntur" for "boys mature late".
But there are conspicuous differences too. In comparison with
~he other works the De aere et aquis is written in a much more pol-
Ished style, which exhibits elements of prose-rhythm (cursus) and,
at one point, breaks out into verse:
Qui autem habent £lumina in que pluviaIes et alie confluant aque
Hi magis incolumes sunt illis ac meliores;
Qui vero non
Sed potius fontes ibi sunt pigreque paIudes
Sive salinarum £luit unda bibentis in usum
Hos multum varie faciunt distare figure. '
Sunt quoque citrini
Venter eis largus, multum quoque splenis habetur. 63
62. For the generalised use of "fumus" in the Additional De eiementis, the De sper-
mate and th: Pantegni, see p. 93 below (Alfanus uses "vapor" and "fumus").
63· Ed. Dlller, p. 103, H106. This is a reasonably literal translation of the Arabic;
ctr. ed. Mattock and Lyons, p. 158: « If there are running rivers there, which draw
off fro~ them stagnant water, both rain water and other water, the people here are
healthier and better (= version lr, "bulkier" version 1) than those others. If there are
II
Physics before the Physics' 73
This brings to mind the style of Alfano rather than that of Con-
stantine. For he not only used cursus in the preface to his transla-
tion of the Premnon physicon,64 but was also an accomplished poet.
The Additional De elementis, on the other hand, would seem to
be distant from Alfano, since it bears no similarity at all to Alfano's
version of Nemesius's On the Nature of Man. This may imply that
the Additional De elementis was translated by someone who was
unaware of the Premnon physicon. If so, this would suggest that, in
spite of the known contacts between ~fano and. Constantine, one
cannot talk about a single commuruty of medical scholars, em-
bracing Salerno and Monte Cassino, who knew each othe~s'
works. The possibility remains, however, that the translators did
not realise that they were both translating the same work, because
the original texts were either anonymous or misattri~uted.65
Finally, the De physids ligaturis exhibits features which may sug-
gest that the author was familiar with the British Isles, for .he
refers to «Scotia »66 and gives a word in a vernacular form which
. Ient.67
corresponds to the AngIo-Saxon equlVa
The English features of De physids ligaturis lead us naturally to
look again at the earliest context in which these texts appear: the
English MSS A and G. These two manuscripts are very different
from each other. The arrangement of the contents of MS A can
no rivers there, and the people's waters come from springs, marshes and, swamps,
the appearances of the people differ: their complexi~ns ar: yellow (= veTSlon lr, v~­
sion 1 omits "their complexions are yellow"), and their bodies are large, as are their
spleens also •.
64. Alfano's use of cursus in his Vita S, Christinae ~d been n~ted by Tore J~son,
Prose Rhythm in Medieval Latin, Stockholm, AlmqUlst and Wiksell Internattonal,
1964, pp. 68, 108, and 112. . ,
65· For the different misattributions of NemeslUS'S text, see W. Telfer" Cynl of
Jerusalem and Nemesius ofEmesa, London, SCM Press, 1955, Pp.,203-6. Alfano s trans-
lation is cited in later authors as being composed by «Remtgtus.; see I. Brady, Re-
migius-Nemesius, «Francisc. Stud.• , 8 1948, pp. 275-84.
66. For the reference to the land of the Scots as a cold place, presumably corre-
sponding to « Scythia. in the Arabic original, see Burnett, The Introduction ofArabic
Learning into England, p. 28. . .
67· De phisids ligaturis, ed. Wilcox, p. 37: «Alii oculum samboc, Id es~ hyrcoc~~,
radici maioranae samsuci liganu. The "stamboc" is "stanbucca" ("wild goat) m
Anglo-Saxon.

II
74
best be explained as being due to the fact that the scribe had in
front of him a copy of Constantine the African's Pantegni, into the
empty spaces of which he inserted several other useful and!or in-
teresting texts - the choice of text depending on whether they
would fit into the spaces rather than whether they were most ap-
propriate for the context. In other words, it is quite likely that MS
A (or its exemplar) has been compiled from at least two sources: a
manuscript of Constantine's Pantegni, and a manuscript, or manu-
scripts, containing the short texts which are not attributed to Con-
stantine, and at least,one of which is of English provenance since it
includes an English name of a herb. 68 It is to a group of these texts
that the scribe has applied the description «quedam phisica» (fo1.
165r).
~S G, on the other hand, although it may have been hastily
wntten, was planned from the beginning to contain a sequence of
texts, in a rational order, starting with a general introduction to
"philosophia", followed by several works on natural science, and
ending with texts of a more practical aim (how to beget the best-
tempered children, physiognomy, and medicine). Care has been
taken over the rubrics to the texts, and their titles and authors ap-
pear as running heads throughout the volume. Several of the texts
have been "edited", in that they have been meticulously corrected,
and variants have been given from two manuscripts of the same
text. Hence it is only in MS G that the Additional De elementis is
called «a book », and is terminated with an "explicit".69 The au-
thor of the general introduction to "philosophia", who, alas, can-
not yet be identified, but who shows a wide range of Latin learn-
ing and is keen to quote Aristotle's Physics whenever he can,70 may
have been responsible for the collection.
68. MS ~, fo1., 163v, in~ludes a, re~pe with the ingredient « Radix sperewurd qure
uocatur del gratIa »; see IllustratIon In Burnett, The Introduction oJArabic Learning in-
to England, p. 27.
69. In MS Munich, Bayerische Staatsbibliothek, Clm 23535 the words «De ele-
mentis liber» are added as a heading by a recent hand.
70. See Dales, Ylnonymi' TIe elementis', p. 181: «Ut dicit Aristoteles in Phisica sua:
"Phisis ~s,t na~ralis motus alicuius elementi ex se" »; ibid., p. 183: «Notandum est
quod diat Aristoteles in phisica sua ignis esse tres species: carbonem, flammam,
11
physics before the (Physics' 75
The most significant fact about MS G is that, amongst ~ts extant
texts it contains not a single work that can be firmly attnbuted to
Con;tantine;71 its contents, rather, represent the kind of learning
from Southern Italy that Adelard of Bath, who visited Mag~a
Graeda and dedicated one work to the bishop of Syracuse, explolt-
ed in his Questiones naturales, of which the earliest copy is included
in MS G.72 Adelard claims that he is using the «studies of the
Arabs» for his Questiones naturales, but, up to now, scholars have
found it difficult to explain what Arabic sources Adelard could be
alluding to. Now that at least two, and possibly t~ee, of the texts
in MS G (the De elementis, De aere et aquis and, p~sslbly, De sferm~te)
have been shown to be translations from Arablc, Adelard s clalm
begins to make more sense.73 Moreover, Adelard's wo~k is specifi-
cally on questions concerning natural science, and his Southern
Italian sources too, are, more or less explicitly, on the same sub-
ject. Alfano, for a reason which has not yet been explained, not *
only omitted the name of the original author of On .the Nature of
Man (so giving the impression that the wo~k ,:as his ~wn), but
substituted the title Premnon physicon, translanng It as «Snpes natu-
lucem» (= Topics, v, 5, 134b28-29); ibid., p. 184: «quo legitur. in phisica terra est im-
mobilis ita intelligendum est». Note also his references to his own text as concern-
ing « phisica », e.g. ibid. p. 181: « Nun~ de p~sica ~a~d~m est ». That the a~th~r
may not have had direct access to Aristotle s PhySICS IS cl~ed ~y Tho~as Ri~khn
in Die (Physica' und der (Liber de causis' im 12. Jahrhundert, Frelburg 1. Schwelz, Uruver-
sitatsbibliothek, 1995, p. 44· "
71. In the lost portion, which cannot be proved ~o be ~tegral WIth th~ rest of the
manuscript, MS G used to contain two works attnbuted ~ the ~anuscnpt ~o Con-
stantine: (Libergraduum'and (Liberde herbis': see Thomson, Llber Man! de Elementls, p. 183.
72. See Burnett et al., Adelard ofBath, Conversations with his Nephew, pp. XXII-XXV.
For the absence of the influence of Constantine's translations in Adelard's work,
see B. Lawn The Salernitan Questions, Oxford, Clarendon Press, 1963, p. 26.
73. For an'apparent influence on the Questiones naturales of De aere et aquis dir~~ly
or via the De spermate, see Burnett, The Chapter on the Spirits, pp. II~-~. Adelard IS In-
debted to Nemesius for his quotation of Xenocrates's definitIon ~f ~e. so~l
(« Quod et Zenocrates (sic) intelligens, ipsam animam numerum esse diffiniVlt, di-
cens: ''Anima est numerus se movens" »: De eodem et diverso, in Burnett et al., Adelard
of Bath, Conversations with his Nephew, pp. 46-47), and his ~escription ,of the t~ee
cells of the brain (ibid., pp. 70-71) is close to that of Ne~esIUs. ~an,o s transl~tIon
of Nemesius also occurs in MS G, and, although this remaInS his most likely
source, Adelard does not quote Alfano literally in the defInition of the soul, and ~s­
es, rather, the terminology favoured by Constantine in his description of the braIn.

11
ralium », and explaining it as «the trunk of natural things, because,
just as many little branches sprout from one trunk, so very many
channels of the science of naturalia will be supplied abundantly
from this fount of learning ».74
It is in the service of natural science and knowledge of the en-
vironment, too, that one can see even the more medical texts in
MS G. Hippocrates's De aere et aquis is full of discussions of differ-
ent characteristics of different places, and it is not inappropriate
that this work should apparently be called « Hippocrates' Physica»
in citations in the De spermate, another medical text with interests
for the natural philosopher in MS G.75
If the Premnon physicon indicates Alfano's preoccupations, the De
elementis of «Marius» (the second text in MS G) may represent
those of a student of his. For it is very likely that this «Marius» is
Marius Salernitanus, the teacher of Gottfried of Viterbo, whom
Gottfried links with Alfano as holding the same opinions about
natural science:76
Notandum est quod Alfanus Salernitanus Archiepiscopus et Marius
Salernitanus preceptor meus in contrarium sentiunt dictum. Dicunt enim
superiora corpora, firmamentum scilicet, Solem et Lunam et omnia
sydera, non ex elementis constare. Composita subiacent corrupcioni et
dissolucioni, ipsa quoque elementa permutantur et corrumpuntur, ideo-
que ex elementis non constant. Talia enim cuncta sunt hodie quanta et
74. Alfano, Premnon physicon, ed. Burkhard, p. 3: «Eritque ei titulus Premnon
physicon, hoc est Stipes naturalium, quia, sicut ex uno stipite multi ramusculi pul-
lulant, sic ex huius fonte doctrinae plurimi scientiae naturalium rivuli exuber-
abunt ». Note that in another English manuscript of the Premnon physicon - London,
British Library, Harley 3969 - Alfano's text (fols 168r-196r) is followed immediate-
ly by Alberic of Monte Cassino's De dictamine - a typically Casinese product, but
with nothing to do with Constantine. Although Harley 3969 was copied not earli-
er than the mid-thirteenth century, it is probably a copy of a twelfth-century Eng-
lish Benedictine manuscript which included works of William of Malmesbury,
Apuleius, and Priscianus Lydus: see R. Thomson, William of Malmesbury, London,
The Boydell Press, 1987, pp. 198-99.
75. See p. 68 above.
76. MS Wtirzburg, Univ.-Bib1., Cod. M.ch.f.23, fo1. 38r-v, in D. Gottschall, Mar-
ius S~lernit?nus und Go~tfried von Viterbo, «Stidhoffs Archiv», 75 1991, pp. 111-13 (113).
The ldentlty of «Manus» and the other works related to his De elementis are dis-
cussed in Burnett, The Works ofPetrus Alfonsi, pp. 56-61. Note, however, that there is
no evidence in the Liber Marii de elementis of the use of the Premnon physicon,
11
Physics before the (Physics' 77
talia ab inicio mundi fuerunt. Et quamvis ipsa videantur calefacere et in-
frigorare et humectare et desiccare, ipsa in s.e ~e~ue c~da n~que fridiga
neque humida neque sicca sunt et non sunt In lPSiS qU~ltates iHe, sed na-
turales virtutes quibus peragunt actiones suas. Exempli c~usa m~tus ~ale­
facit, sed ipse calidus non est. Aqua frigida et nix calefaaunt .et ipse. In se
calorem naturalem non habent. Ego cum Alfano et cum Mano sentlo.
Note that Alfano, archbishop of Salerno, and Marius of Salerno, my
teacher, think along opposite lines [to the view expressed in the previous para-
graph, that the heavens are made of the four elements]. For they say that the
higher bodies - i.e., the firmament, the S.un, the. Moon an~ all the stars-
do not consist of the elements. CompOSite bodies are subject to corrup-
tion and dissolution; the elements themselves also change and are cor-
rupted. Therefore, they (the higher bodies) do not consist of the ele-
ments. For all such bodies have the same mass today as they also had at
the beginning of the world. Although they may seem (or c~n be seen) to
heat, to cool, to moisten and to dry, they themselves are neither hot, nor
cold, nor wet, nor dry, and in them are not these qualities, but rather nat-
ural powers by which they perform their actions. For example, move-
ment heats, but it is not hot itself. Cold water and snow heat, and t~ey
themselves do not have a natural heat within them. I hold the same Vlew
as Alfano and as Marius.
Given the absence of Constantinian works in MS G, the pres-
ence of the Additional De elementis in both MSS A and G would
suggest that this text (possibly along with some of the ~ther short
texts in MS A) was added to the major text of Constantlne from a
source more similar to that of the texts in MS G.77
6. CONCLUSIONS ANn SUGGESTIONS FOR FURTHER RESEARCH
What this article has tried to show is that, from a detailed analy-
sis of the contents of London, British Library, Additional 22719
and Cotton Galba E IV, it is possible to hypothesise that several
77. The contents of the third manuscript of the Additional De elementis - Mu~ch,
Bayerische Staatsbibliothek, Clm. 23535 -, are also o,riented towards Alfano, sm,ce,
in addition to earlier Greek-Latin translatlons, they mclude the De stomacho, which
Constantine dedicated to Alfano, and a fragment of a translation of Hippocrates'
Airs, Waters and places, which could have been made by Alfano: see Grensemann,
Hippokrates De aeribus aquis locis, pp. 233-35 and 251-54. Only the De stomacho, appar-
ently, mentions the name "Constantinus".
11
tex.ts.were translated from Arabic into Latin, independently of the
a~vIty of Constantine the African, and that many of these were
of Interest to the "philosopher" who investigated natural science,
r~ther than to the doctor. This preoccupation with naturalia is in-
dIcated not only by the existence of many works relevant to natu-
ral science (De eiementis, De metallis, De dbis, De aere et aquis, etc.),
but also by the frequent references to « physica », either in the ti-
tles of the texts translated in this context (De physids iigaturis, Prem-
non physicon), or in the works cited within them, whether the cita-
tion is correct or not {Aristotle's Physica, Hippocrates's Physica).78
The movement to exploit the riches of Arabic culture did not
necessarily start with the translation of a corpus of medical texts,
as a result of the fortuitous arrival of Constantine the African from
Qayrawan in Salerno in, or a little after, 1077.79 Constantine's con-
tribution is undoubtedly very important, and there is little reason
to question the attribution of the medical translations to which he
so carefully attached his name, and which were attributed to him
by Peter the Deacon and the Monte Cassino Chronicle. Also,
~any of these texts came from Qayrawan. Nevertheless, it is plau-
SIble that. some of the original Arabic texts, whether translated by
Constantlne or anonymously, had already been introduced into
Arabic Sicily, and the Byzantine provinces of Southern Italy, be-
fore Constantine came onto the scene, and would have become
78. Note that Adelard too includes a wrong attribution to «Aristotiles in Phisi-
~s» in his Questiones naturales: see Burnett et al., Adelard of Bath, Conversations with
~IS N~h~~ pp. XXII-XXV. One may add here the references to an unidentified « Ga.
m penphislcon» ("Galen ?n physics") cited in the margins of the earliest precisely
datable copy of Constantme's Pantegni - Erfurt, Wissenschaftliche Bibliothek der
Stadt, Ampl?n.. Q. 184, copied in an Italian hand in 1147: fol. 40r, «Ga. in periphisi-
con: precedit UlftuS, qu~ .est potentia, et cura secundo sequitur motus eius [ ... ] »;
f~l. 40~, «[::.] uel nutntma secundum Ga. qui in periphisicon actiones huius ita
disporut, salicet quam cito cibus e uenarum uasis resudatione egressus est, dissemi-
natu~ ad mem~rum natur~iter si~i ~ttrahens ipsum, deinde apponitur ipsi, postea
courutur~ ad ulttmum finaliter asslmtlatur [ ... ] »; fol. 46r, «Ga. in periphisicon con-
stare ueslcam [ ... ] ».
79· For this date, see F. Newton, Constantine the African and Monte Cassino The
Text ofthe ttIsagoge", in Burnett and Jacquart, Constantine the African, pp. 16-47 (1~-20).
For the st,ory: and, a date ca. 1165, see M. McVaugh, Constantine the African, in Dictio-
nary of Saentific BIography, ed. c.c. Gillispie, New York, Charles Scribner's Sons,
1971, III pp. 393-95.
11
Physics before the Physics} 79
known to Latin scholars in a more gradual and less dramatic way
during the course of the eleventh century.80
This could explain the origin of the collection of medical texts
which became known as the Articella, and which, in its earliest
form, contained no work firmly attributed to Constantine.81 Two
of these texts - significantly, by Hippocrates - are thought to have
been translated from the Arabic versions of l:Iunayn ibn Isbiq,82
and the whole collection was prefaced by a handy introduction to
medicine by the same l:Iunayn: the Isagoge. Yet these Hippocratic
works were brought together with direct translations from Greek
(Theophilus's De urinis and Philaretus's De pulsibus), and the Arabic
origin of the Isagoge was so heavily disguised,83 that both early
commentators and more recent scholars thought that the author
80. The use of De physicis ligaturis before 1090 by Marbod (see p. 63 above) pre-
dates the known use of any of Constantine's works.
81. The two lists of Constantine's works, on the other hand, mention a transla-
tion of a commentary (presumably Galen's) on the Aphorisms (<< Expositio Afori-
smi »), and Peter the Deacon mentions « Pronostica », which, in Bloch's view, could
be Galen's commentary on the Prognostics. See Bloch, Monte Cassino, pp. 128-29 and
133. The translator Gerard of Cremona is also credited wi~h translati~ns of bot~ of
these commentaries, and it is not yet clear whether he reVIsed an earlier translatton
of Constantine, or whether two independent translations are involved.
82. For the Arabic origin of the Prognostics see B. Alexanderson, Die Hippokratische
Schrifi Prognostikon: Oberlieferung und Text, Goteborg, Almqvist & Wicksel~ 1963 (~tu­
dia Graeca et Latina Gothoburgensia, 17) pp. 170-73, who shows that the translatton
of the Prognostics that was included in the Articella (beginning: « Omnis ,qui me.dicine
artis studio [ ... ] ») is from the Arabic, but is different from the tr~slatton w~ch ac-
companies Galen's commentary, equally made from the ArabIC, but pOSSIbly by
Gerard of Cremona (see previous note); for that of both the Prognostics and the Apho-
risms see G. Baader, Articella, in Lexikon des Mittelalters, 9 vols and Registerband, Mu-
nich, Artemis, 1980-1999, vol. I pp. 1069-70. Mark Jordan concludes from the evi-
dence that «it is simplest to suppose that they [the Aphorisms and Prognostics] were
translated from the Arabic, perhaps by Constantine or his colleagues »: Medidne as
Sdence in the Early Commentaries on johannitius, «Traditio »,43 1987, pp. 121-44 (126).
83. I:Iunayn's name became "Johannitius", and the work was giv~n the Greek
name for "introduction": eisagoge. Some of the more extreme Greasms, such as
«virtus zotica» for the "animal power" (<< virtus vitalis ») were dropped in later re-
visions of this Isagoge; see D. Jacquart, A l'aube de la renaissan~ medicale des xie-xiit sie-
des: IHIsagoge johannnitii' et son traducteur, «Bibliotheque de l'Ecole des charte~ », 144
1986, pp. 211-40, reprinted in Ead., La sdence medicale ?ccidentale entre deux r~JSSances
(XlIt s,-xvt s.), Aldershot, Variorum, 1997, article I. The Influence of the ArabIC U,rlage
is demonstrated in U. Weisser, Noch einmal zur Isagoge des johannidus: Die Herkunft des
lateinischen Lehrtextes, «Siidhoffs Archiv »,701986, pp. 230-35.
11
80
was Greek. 84 The assumption that Constantine the Mrican was
introduced to Desiderius, the abbot of Monte Cassino, with a
«letter of recommendation» by Alfano, giving his translation of
the Isagoge as a testimony to his competence,85 rests on a misinter-
pretation of a Latin text found in some of the earliest manuscripts
of Constantine's Pantegni;86 the Isagoge is not found in manuscripts
of Constantine's Pantegni except when it is accompanied by the
other texts of the Articella in compendious medical collections. It
is, nevertheless, probable that the Isagoge predates the other texts
which were definitely translated by Constantine, and this may al-
so be the case for the other texts of the early Articella.87
The case is even stronger for works of a more physical or philo-
sophical nature, since no such work is included among the ancient
lists of Constantine's translations. This is, however, an area which
needs considerably more research. All I am proposing here is that,
in doing this research, one should consider the likelihood that the
intellectual society in Southern Italy for much of the eleventh cen-
tury was such that the Greek language was well known, but Greek
texts were often unavailable in their original language. The same
texts, however, were readily at hand in Arabic translation among
the Arabic communities in Sicily and North Mrica, and could be
used faute de mieux to make up for the absence of the Greek texts.
Moreover, the texts that were sought after were not only those rel-
evant to the practice of medicine, but also anything that could elu-
cidate the wonderful workings of Nature, and they prepared the
way for the reception of Aristotle's Physics in the twelfth century.
84. The debate is summarised in Jordan, Medidne as Sdence, pp. 123-28.
85. The story is told in Bloch, Monte Cassino, pp. 100-1.
86. See Bumett, Encounters with Raz'- the Philosopher: Constantine the African, PettUS
Alfonsi et Ramon Marti, in Pensamiento hispano medieval: Homenaje a Horado Santiago-
Otero, ed.J.-M. Soto Rabanos, Madrid, Consejo superior de investigaciones cienti-
fieas, 1998, pp. 973-92 (976). There is a manuscript of the lsagoge in Monte Cassino
(no. 225), written in Beneventan script between 1075 and 1090, but Francis Newton
claims that « the scribe had his training in a different centre [from Monte Cassino] »:
Newton, The Text of the 7.sagoge', pp. 2']-28.
87. The most detailed study of the lsagoge is that by Jacquart (see note 83 above),
in which at least two versions are distinguished; Jacquart inclines towards attribut-
ing the translation to Constantine on the grounds of style and terminology, but the
question of authorship is not yet closed.
11
Physics before the (Physics' 81
APPENDIX
In the following editions, scribal insertions are noted with obli~ue
slashes: \ /; glosses with round brackets: ( ... ); a.e. = before correctIon;
p.e. = after correction; i.m. =in the margin; inv. =transpose; add. = a~d;
om. = omit; "re" represents the e-caudata. For ease of reference sectIon
numbers have been added; in the case of the Additional De elementis
these correspond to the section numbers in Alfano's Premnon physicon
as edited by C. Burkhard, Leipzig, Teubner, 1917·
I. THE ADDITIONAL DE ELEMENTIS
The Manuscripts
The De elementis is extant in three MSS: British Library, Additional
22719 (= A), British Library, Cotton Galba E IV (= G) and Muni~h,
Bayerische Staatsbibliothek, clm. ~3535 ~= M). The t~xt ~as .be~n e~t­
ed by H. Grensemann and U. Welsser m Iparchus ~tnut'en:'s al,as ~,~­
parchus Metapontinus. Untersuchungen zu einer hochm'ttel~l~erltch~ latetnt-
schen Obersetzung von Nemesios von Emesa, De natura hom,nts, Kap,!el5: De
elementis, Bonn 1997. They bas~ their edition on MS M, and l~clude
neither the readings of MS A, nor (in most cases) the alternative or
corrected readings of MS G.
British Library, Additional 22719 (= A). For a desc~ption.of the con-
tents of this manuscript, see pp. 53-55 above. The scnbe wntes .on b~e
lines which have been drawn with a dry point, but also keeps m mmd
two further levels: that of the tops of the minims (the "headline"), and
a medial point between the base line and the he~~~e. The.text is ar-
ticulated according to several different levels of diVlslOn, which can be
listed in descending order of importance: rubricated initial capitals, ei-
1
ther red above, green below, or red on the right and green on the left;
discrete green gallows starting at the headline an~ embracing t~e last
punctuation mark and the initial of the new se~on; ~alf-rubncated
capitals, with a single colour (usually green) shading elther the lower
portion, or the left-hand portion of a letter; a green dot over t~~ first
word of a subsection - green dots are also used to show the pOSltlOn of
I. Letters A and V tend to be filled by one colour only.
11
82
an item in a list; rubricated lower case letters in green; punctus versus;
punctus medius. This degree of articulation of a scientific text is remark-
able, and is the subject of separate study.2
The e-caudata is used correctly for classical "ae" except for one in-
stance of "Hec" for "Haec" (27) and "quedam" in the note "quedam
phisica" on fol. 1651; the ampersand is always used for "et", and + more
commonly than "·e·" for "est". Prepositions are either joined to the
nouns that follow, or separated by a space which is about half that sep-
arating other words. There are no running heads, and litterae nobiliores
do not extend into the margin.
British Library, Cotton, Galba E IV (= G). For the contents of this
manuscript, see pp. 55-56 above. The scribe uses the e-caudata inconsis-
tently and occasionally wrongly (s sanre), and never uses the punctus
versus. All punctus are on the base line. A dotted or broken line is writ-
ten under words to be cancelled, but erasure is more common, and
very frequent. The ampersand is used throughout for "et", and "e." for
"est". The paraph mark is a simple gallows; the same with an "eye" in
it is used as a hiatus mark. Prepositions are usually separate from the
words they follow, and double "ii"s have acute accents on them.
Munich, Bayerische Staatsbibliothek, clm 23535 (= M)3 is a manu-
script of the twelfth century, consisting of an acephalous copy of Phi-
laretus, De pulsibus, a fragment of Hippocrates' Book ofAirs, Waters and
places (possibly in a translation by Alfano of Salerno), Liber elephantie
(on leprosy, without attribution), Constantine the African's Liber sto-
machi (dedicated to Alfano), the Liber de elementis and Liber oculorum
(both without attribution), Liber Democriti, De curatione febrium (incipit:
« quae signa determinationis febris appareant»), Byzantius's De acutis et
longis morbis epitome, Oribasius's Euporiston, id est parabilium libri, the
same author's Expositiones vocum medidnae ordine alphabetico, and Libellus
de ptisana. There is no indication of provenance or date, but the hand
and the scribal practises suggest the early twelfth century.4
2. C. Burnett, The Presentation of a Sdentific Text in the Early Twelfth Century: MS
British Library, Add. 22719, in preparation.
3. For descriptions of the contents of this manuscript see Grensemann, Hip-
pokrates De aeribus aquis focis, pp. 247-48, and Grensemasm and Weisser, Iparchus
Minutiensis, pp. 26-29, and E.W. Goos, Oribasius Latinus. Ad Eunapium Eporiste, liber
prim us, Utrecht, Ph.D, 1989. I am grateful to Klaus-Dietrich Fischer for informa-
tion on this manuscript.
4. Goos, Gribasius Latinus, p. 78, following the advice ofBernhard Bischoff, con-
11
1
I
physics before the Physics'
The scribe uses "tam" or "tm" as an abbreviation for "tamen" (not
"tantum"), "here" for "hae", "e" and "+" for "est", and both the tironian
symbol and the ampersand for "et". He writes "helementum" for "ele-
mentum" on two occasions (first "elementa" in 25, and in 43). His
punctuation is less dense than that of MSS A or G, and consists only of
a simple punctus, an interrogation mark, and a rare punctus elevatus (in the
shape of a small tironian et or "7" at the level of the headline). On the
other hand, he appears to articulate the text by means of (I) variation in
word-spacing, not dividing words which go closely together ("quequi-
dem", "sicaret", "necsciri", "quidsint", "aterra", "infulgure", "aereat"),
but adding extra spaces instead of punctuation between clauses (( nec-
essarium est Terra vero ... », « Nullum tamen elementorum est
qucrquidem sensibiliter», « In aqua et igne diligens lector»), and!or
(2) extending the last letter of a word before a syntactical break.s The e-
caudata is inconsistently used for "ae". He uses no paraphs, but a variety
of hiatus marks. The text has been corrected, and missing phrases have
been added in the margin, in one case twice (Jo «fit aer here ergo tres
qualitates aerem faciunt id est obtusum rarum' motus »).
The Relationship between the Manuscripts
The scribal habits of MS G appear to be younger than those of MS
A. In the case of the De elementis the scribe of MS G appears originally
to have copied the text from a similar (if not the same) manuscript as
MS A, but made several mistakes in copying. Very soon after the copy
was made, MS G's text was extensively corrected, perhaps by compar-
ison with a second manuscript of the De elementis.6 Three significant
corrections correspond to the reading of MS M (see 2, 4 and 31). On
many occasions, corrections have been written over erasures, so that it
siders that the manuscript is from Italy and was written in the second half of the
twelfth century. This seems too late.
5. For the use of these techniques see P. Saenger, Space between Words, Stanfo~d,
Stanford Univ. Press, 1997, pp. 36-37 (hierarchical word blocks) and 61-63 (sp~aal
terminal forms of letters at word-ends). In the second case the letter extenSion,
which is always in the form of an extension of an oblique stroke towards the top
right, where it ends with a stopping of the pen, could be interpreted as a letter plus
a high distinctio or punctus. . .,
6. The hint that there was a second manuscript is the alternattve reading mtro-
duced by "al." (11), but it is possible that alternative readings were already men-
tioned in a single exemplar.
11 11

is no longer possible to read MS G's original text. The readings of MS

1 10 sit frigida et humida
Physics before the Physics)
frigida et humida sit
8S

A agree with those of MS G before correction three times (2,10,31); 11 mirifice elementa elementa mirifice
11 te uertas considerans ea ea considerans
MS A agrees with MS G once after correction (8) and twice with an teditis (miscopyingforte vertis)
alternative reading given by MS G in the margin or above the line (5, 11 non desit similitudo similitudo non desit
11). Most of the other variations between MSS A and G involve syn- 11 conuertibilia sunt ut terra ut terra et ignis conuertibilia [non]
onyms (12 apparet A, palam est G), the inclusion or omission of small et ignis sint
words like "scilicet", "id est" and forms of "esse", the substitution of a 12 ob combustionem nimiam ob nimiam combustionem
12 tunc scilicet scilicet tunc
relative for a demonstrative at the beginning of a sentence (10 Huius potestate est calidum
12 calidum est potestate
A, Cuius G; 11 In his A, In quibus G), and mistakes in one or other of 13 dumtaxat insunt qualitates qualitates actualiter insunt
the manuscripts. actualiter dumtaxat
MS M, on the other hand, exhibits features which appear to be as 18 ignis fumo extincti ignis extincti fumo
old as those of MS A. On at least two occasions it gives a superior text 18 suo in loco in suo loco
18 calidus est est calidus
to both the other MSS (12 "sulphur" for "fulgur" AG; 16 «inferiores aer aqua
21 aqua aer
partes frigidre· superiores uero calidre» - AG reverse the phrases). 23 est destructio alterius rei alterius rei est destructio
However, an attempt seems to have been made to polish up the Latin, 25 habere autumat autumauit habere
by omitting repetitions7 and changing the word order.8 The transposi- 25 elementa tria tria elementa
tions are not noted in the apparatus, but are listed as follows. 27 qualitatem habent habeant qualitatem
27 in aqua inuenimus inuenimus in aqua
MSA MSM 30 qualitatem ignis ignis qualitatem
2 in mundo tocius est est in mundo tocius 30 necessario faciunt faciunt necessario
2 sunt simplicia simplicia sunt 31 ad stabilitatem motus motus ad stabilitatem
2 est elementum elementum est 32 habent alio modo qualitates alio modo qualitates habent
3 profecto qualitates qualitates profecto 35 duo esse esse duo
3 est elementorum elementorum est 41 est unum unum est
5 elementum esse esse elementum 42 non existit ex uno existit non ex uno
6 singulre qualitates qualitates singulre 45 ab eo procedere procederent ab eo
7 differunt qualitatibus qualitatibus differunt
8 copulari inuicem inuicem copulari MS M omits alternate translations,9 but has incorporated two glosses
into the text (see 2 and 20). It is unclear where the latter come from,
but the second one starts off with a phrase which is a more literal
7. See 18 in which MS A gives the text as in the Arabic (<< unus qui procedit de translation of Nemesius's text than the one common to MSS AGM,
aqure fumo, alius vero de ignis fumo extincti procedit») whilst MS M tightens up
the phrase by omitting the repeated words (<< unus est qui de aqure, alius de ignis which suggests that the gloss was made at the time of the translation,
fumo extincti procedit»). if it did not already exist in the original text.
8. That a deliberate effort to change the word order has been made is indicated
by the fact that the changes become less frequent as the text proceeds (and as the
reviser becomes less exigent). In itself this does not tell us whether a text like MSS
AG was changed into that of MS M or vice versa. However, the tendency of MS M
to present a better Latin style suggests that this manuscript gives the revised order:
in the majority of cases MS M's order puts the verb at the end of its phrase, or the
adjective before its noun. Note also what appears to be an incomplete change of
word-order: the repetition of "se habet" in 31; MS A and G before correction have
the words in the second position, and G after correction has transposed the words 9. Except in 26 where MS M alone offers what seem to be two translations of
to the first position. the same verb: "dividitur destituitur".
11
86
LATIN TEXT
1 Physics before the (Physics'
I. <DE ELEMENTIS>
* In the following translations angle brackets « » surround words im-
<I> De elementis: quomodo ad inuicem commiscentur et quomodo
ab inuicem separantur. lO
<2> Elementum in mundo totius est corporis minima pars. Elementa
quattuor sunt: 11 terra, aqua, aer, ignis. Talis enim ordo ab inferioribus ad
superiora, a minus dignis ad digniora, factus uidetur. Hrec quattuor cor-
pora sunt, simplicia quodammodo ad eorum scilicet12 qure constituunt
comparationem. Elementum nichil aliud est nisi eius cuius est elemen-
turn similitudo. Initium autem non est 13 similitudo eorum 14 qure ab eo
procedunt. Elementa tamen ad ea qure ex eis 15 fiunt similia esse necessa-
rium est. <3> Terram uero, aquam, aerem et ignem esse elementa, nulli
dubium est, in quibus profecto qualitates perfectas 16 potestate et actu in-
uenimus. Nullum tamen est elementorum qure quidem 17 sensibiliter
comprehendimus, quin ad se inuicem sint copulata. 18 Sunt enim elemen-
ta commixta pars parti et singula a singulis, alia plus, alia minus, accipiunt.
Verumtamen 19 coniunctorum 20 taliter uniuscuiusque natura seorsum ap-
paret.
'I
10. Incipit liber de elementis G de elementis liber M (recent hand)
11. quattuor sunt] i.e. add. G
12. scilicet] om. GM
13. Initium autem non est Gp.c.: Immo initium non A Immo initium est G a.c.
Initium autem non M
14. similitudo eorum] quia sunt inuisibilia illa elementa add. M
15. his M
16. qualitatem perfectam G
17. qure quidem M: quicquid A quod quidem G
18. sint copulata M: non sit copulatum AG
19. Verumtamen A: Verumptamen GM
20. coniunctorum AG: licet coniuncta M
11
TRANSLATIONS
plied by the Latin text or punctuation; square brackets ([ ]) indicate re-
dundant words in the Latin text.
I. ON THE ELEMENTS
<I> About the elements. How they mix with each other, and how they
separate from each other.
<2> An element in the world is, of the whole body, the smallest part.
The elements are four: earth, water, air, fire. For such an order is seen to be
made from the lower ones to the higher (from the less worthy to the wor-
thier).! These four are bodies, <being> "simple"2 in some way in compari-
son with those things that they constitute. An element is nothing other
than the likeness of the thing of which it is an element. But the starting-
point (origin) is not the likeness of those things which proceed from it.3 El-
ements, however, must be similar to those things that are made from them.
<3> That earth, water, air and fire are elements, is doubtful to no one. In
these, certainly, we find qualities which are perfect both potentially and ac-
tually. However, whatever we perceive with the senses is none of the ele-
ments, unless they are joined to another. For the elements are mixed, part
with part,4 and single <elements> receive from single <elements>, some
more, others less. But the nature of each one of the <elements> joined to-
gether in this way can still be discerned separately.
I. The Greek and Arabic texts have no equivalent to "from the less worthy to the
worthier".
2. The corrector of MS A has changed the punctuation to give this meaning
rather than "these four bodies are simple".
3. The text of MS A and the uncorrected text in MS G give a wrong meaning:
"or rather, <an element is> the starting point, not the likeness [ ... ]". The original
sense is that an element must be distinguished from a cause (cXpxi'! = initium), in that
a cause is d!fferent from the thing caused. The meaning is obscured in the Latin by
the translation of the adjective "similar" as a noun "likeness", probably because of
the ambiguity of the Arabic text. MS M adds: "because elements are invisible".
This is not in the Greek or Arabic and has the nature of a gloss.
4. This is a too-literal translation of the Arabic ba(qu ba(qan (literally: "part with
part") which here means "some with others".

11
88
<4> Elementorum uero singula habent 21 qualitates duas specificas, ter-
ra enim frigida et sicca est, aqua frigida et humida, aer calidus et humidus
secundum naturam sui in se, ignis uero calidus et siccus.zz <5> Ipsre sane
qualitates elementa non sunt. Impossibile enim est 23 ex incorporeis quali-
tatibus corpora fieri. Impossibile etiam est 24 corpus, si caret 25 qualitatum
perfectione, actualiter elementum esse. Quod Si 26 esset, in 27 infinitum
elementa procederent; adinuentiones 28 enim qualitate 29 secundum plus
minusue differunt. Iterum, nec sciri potest quid 30 sint elementa. <6> Ne-
cessarium ergo est ut elementa corpora sint simplicia, qualitatum quoque
perfectionem actualiter habentia, id est calidum et 31 humidum, frigidum
et siccum. Hre enim singulre qualitates corpus omnifariam mutant. In ali-
is uero qualitatibus non similiter. Albedine32 enim corpori superue-
niente,33 non ex toto fit alterati034 quemadmodum calido,35 frigido et ce-
teris.
<7> Ea uero elementa contraria dico qure duabus differunt qualitatibus.
Verbi gratia,36 ut aqua et ignis. Aqua enim frigida et humida, ignis uero
calidus et siccus est. 37 Terra autem aeri opponitur. Est enim terra frigida et
sicca, aer uero calidus et humidus. <8> Quoniamquidem contraria ne que-
unt copulari inuicem, ut copulatio fieret inter terram et aerem, aquam
Natura posuit permisitque aquam duas habere qualitates - id est frigidi-
tatem et humiditatem - ut unio fieret, scilicet 38 per frigiditatem terrre,
et 39 per humiditatem aeri, uniretur. <9> In aqua et igne, diligens lector,
21. singula habent AMa.c.: singulas habent Mp.c. singulum habent Ga.c. singulum
habet Gp.c.
22. secundum naturam sui in se ignis uero calidus et siccus Gp.c.: ignis uero
calidus et siccus secundum naturam sui in se AGa.c.M
23. enim est G: enim A est enim M
24. est] om. A
25. si caret] sicaret Aa.c.M
26. si] non add. A
27. in] om. M
28. adinuentiones AM] adinuentionem G
29. qualitate] uel te uel tes G supra (i.e., G hesitates between ((qualitatem': ((qualitate"
and ((qualitates"
30. quid AM: qu:e G
31. et] om. M
32. Albedine AG: Albedini M
33. superueniente AG: \adlueniente M
34. alteratio AG: alternatio M
35. calido] et add. G
36. Verbi gratia] om. M
37- est] om. A
38. unio fieret scilicet] om. M
39. et] om. M
11
1 Physics before the Physics'
q> Single elements have two specific qualities. For earth is cold and
dry; water, cold and moist; air, hot and moist according to its nature in it-
self.5 But fire is hot and dry. <s> Certainly, the qualities themselves are
not elements. For it is impossible for bodies to be made from incorporeal
qualities. It is also impossible for a body, if it lacks the perfection of qual-
ities, to be actually an element. If it were not <impossible>, elements
would go on for ever. For, existent things6 differ by quality according to
more or less <of the quality>. Again, <if elements consisted of incorpore-
al qualities>, it is not possible to know what the elements are. <6> It is
necessary, then, that elements are simple bodies; also <that they are> ac-
tually having a perfection of qualities, namely, hot and moist, cold and dry
<etc.>. For these single qualities7 change body altogether. But for other
qualities <the case> is not the same. For, when whiteness comes over a
body, a total alteration is not made, as <it is> in hot, cold and the others.
<7> I call those elements «contrary» which differ by two qualities, e.g.,
like water and fire; for water is cold and most, but fire is hot and dry. And
earth is opposite to air; for earth is cold and dry, but air is hot and moist.
<8> Since contraries cannot be joined to each other, for a join to be made
between earth and air, Nature 8 posited water and allowed water to have
two qualities, namely, coldness and moisture, so that a union could be
made: i.e., through coldness it might be joined to earth, and through dry-
ness, to air. <9> In water and fire, careful reader, a similar situation may be
5. The phrase "according to its nature in itself" has been put in the wrong place
in MSS AM, since it applies not to fire, but to air (as is clear from 16 below, and as
is in the Greek and Arabic); MS G corrects the wrong order of the phrases by
means of superscript letters of the alphabet.
6. It is not clear whether the meaning of "adinuentiones", apparently a calque on
Arabic mawjudat, would be clear to a Latin audience. ..
7. Instead of "these single qualities" the meaning should be "these quallt1es
alone".
8. The Greek and Arabic have here « the Creator,..
 ~
II II
I
90 ! Physics before the (Physics' 91
simile perscrutetur. Taliter equidem40 elementa colligata esse manifestum scrutinised.9 It is clear, then, that the elements are bound together in such
est. away.
<10> Huius 41 alligationis perfectio esse notatur, cum singula singulis <10> One may note the existence of perfection in this binding, since
sint inherentia sequentibus se simili qualitate, ut aqua qua:, cum sit frigida single <elements> are sticking to single <elements> which follow them
et humida, per frigiditatem terra:42 copulatur priori,43 dispositis quodam 44 with a similar quality. E.g., water, since it is cold and moist, is joined
ordine ut unum inferius, alterum superius emineat. Item, per humidi- through coldness to earth, being before it (since <they are> arranged in a
tatem aeri aere[m] inferior copulatur.45 Aer uero humiditatis respectu
certain order, so that one <sinks> lower, the other rises higher). Likewise,
aqua: coniunctus, cum igne,46 qui a terra per siccitatem non dissidet,
it is joined to air, being below air. 10 But air, while joined to water in re-
calore participat.47
<n> In his48 ergo uniuersaliter ordinem contemplare circularem. spect to moisture, also shares in heat with fire, which <in turn> agrees
Natura quippe mirabilis mirifice elementa considerauit ut quoquomo- with earth through dryness.
d049 te uertas, considerans ea, seu 50 ab inferioribus ad superiora uel <11> In <taking> all these <factors> together, then, focus your attention
econuerso, non desit similitudo, quod 51 elementorum extrema conuer- on the circular arrangement <which arises>. For wonderful Nature has
tibilia sunt,52 ut 53 terra et ignis. <12> Ablato enim54 ab igne calore,55 terra taken thought for the elements in a wonder-making way, so that wherev-
efficitur. Hoc enim 56 in fulgure apparet,57 quod ignem esse nulli 58 dubi- er you look, while considering them either from the lower to the higher,
urn est. Fulgur enim ob combustionem nimiam in petra: rigorem transit. or in the opposite direction, similarity is not lacking,l1 because <even> the
Vnde omne fulgur aut petream59 aut sulphuream habet naturam. Quid elements at each end can turn into one another, as <do>12 earth and fire.
enim est 60 sulphur61 nisi ignis extinctus ?62 Fulgur tunc scilicet sulphur ef- <12> For, on the removal of heat from fire, earth is produced. This is ob-
ficitur quando actualiter ab igne calor aufertur. Vnde calidum est 63 potes-
vious in a thunderbolt, which no one doubts is fire. For a thunderbolt13
passes into the hardness of stone because of excessive burning. Hen~e
40. equidem AG: enim M every thunderbolt has either a stony or a sulphurous nature. For what IS
41. Huius A: Cuius GM
42. terrre] per humiditatem aeri aere inferior add. AG
sulphur other than extinguished fire? [A thunderbolt, i.e.] sulphurl~ ~s
43. priori M: priori bus A prioris G produced at the time when heat is actually removed from fire. Hence It IS
44. quodam AG: quodammodo M
45. aeri aerem inferior copulatur M: aeri similis est est quoque inferior aere AG
46. igne] etiam add. G 9. The Greek and Arabic spell out how air joins water to fire, but the Lan,n
47. qui a terra [ ... ] participat] cum igne etiam calore participiat (sic) qui a terra writer leaves it to his «careful reader» (who is absent from the Greek and Arabic
\non! dissidet per siccitatem G texts) to work out the details himself. . '
48. his AM: quibus G 10. The corruption in MSS AG possibly arises from the insertlon of text m,the
49. quoquomodo A: quomodo G quo\quo/modo M margin (italicized here) into the wrong place, giving th.e convo~ute~ meanm~:
50. seu A: eis G (probably a corruption if"uel") siue M "water [ ... ] is joined through coldness to earth through mOISture to alT, .betng below ~lT
51. quod AM: scilicet G [ ... ]. Likewise, through moisture it is similar to air; i~ is also below air". The Latm
52. sunt AG: non sint M author seems to be making heavy weather of the simple statement that the ele-
53. ut AM: id est G al. ut G supra ments are linked by their qualities in a certain order. His preoccupation with
54. enim AM: igitur Gp.c. « higher» and « lower », which was already apparent in <2>, is probably influenced
55. Ablato ... calore AG: Ignis enim calore ab eo sublato M by the Arabic text (min asfala ilil 1-(uluWt). ,
56. enim AG: autem M 11. Section 11 up to this point is an addition vis avis the Greek and ArabiC texts.
57. apparet AM: palam est G 12. MS G's reading "i.e." may be more appropriate here than MSS AM's "as".
58. nulli AG: nemini M 13. In the Greek and Arabic "rue" is the subject.
59. petream A: petraream G petrream M 14. It seems that the translator originally mistakenly wrote "thunderbolt" ("~l­
60. est] om. M gur") in this and the previous phrase instead of "sulphur". On the second o~caslOn
61. sulphur M: fulgur AG he (or a later annotator) corrected the text with a gloss "scilicet sulphur" which was
62. extinctus] est add. MAa.c. then incorporated into the text. On the first occasion MS M alone restores the cor-
63. in G rectword.
11
92
tate, siccum in actu.64 <13> Diximus enim quod in elementis dumtaxat in-
sunt qualitates actualiter, in aliis autem non nisi cum afficiuntur circa ele-
menta.
<14> Elementa enim Dei sapientia mutabilia65 fecit, ut nec ipsa
perirent, nec ea qua: ab illis perficiuntur. <15> Quia uidemus, quando ter-
ra liquescit et 66 attenuatur, fit aqua. Aqua uero econuerso, si constringitur
et solidatur, terra est; eadem si calefiat,67 reddatque ex se fumum, conse-
quitur ut aer fiat. Aer quoque continuus et spissus factus, aqua fit;68 si
idem siccatur, mutatur in ignem. Item, si ignis extinguitur et amittit 69 sic-
citatem, aer eat.70 Est enim aer ignis extinctus, calida:que aqua: fumus.
<16> Palam igitur est ex his duobus aerem fieri, ac per hoc calidus aer, in-
frigidatur 71 tamen aqua: terra:que uicinitate, cuius 72 inferiores partes
frigida: superiores uero calida: 73 inueniuntur. <17> Hoc enim 74 iccirco
contingit quoniam cito patitur, quia natura:75 mobilis est.
<18> Intellexit enim Aristotiles aeris modos duos esse: 76 unus est 77 qui
procedit 78 de aqua: fumo,79 alius ueroBO de ignis fumo extincti procedit. 81
Ignisque fumus calidus; aqua: 82 etiam suo in loco calidus est, uerum83
cum de suo loco mutatur paulatim infrigidatur 84 et fit aqua. <19> Et, ne
argueretur ignorantia: falsa:que opinionis, quia superiora loca frigidiora
inueniuntur, aeris duplicem dicit esse naturam.
<20> Vniuersaliter autem omne corpus ex omnibus constituitur ele-
64. siccum in actu AG: actualiter uero siccum M
65. mutabilia AM: mirabilia G
66. et M: ut plumbum etiam A ut plumbum G
67. calefiat AG: calefacta fit M
68. aqua fit A: fiet aqua G aqua fiet M
69. amittit M: amittat A dimittit G
70. aer eat A: aer fiet G, aereat M
71. infrigidatur AG: infrigdatur M
72. cuius] aeris add. G
73. inferiores [ ... ] calidre M: superiores partes calidre inferiores uero frigidre AG
74. enim AG: autem M
75. quia naturae AM: et natura G
76. Intellexit [ ... ] duos esse AG: Aristotiles siquidem duos esse modos aeris in-
tellexit M
77. est] om. G
78. procedit G: precedit A om. M
79. fumo] om. M
80. uero AG: qui M
81. procedit] om. G
82. aqure GM: aqua A
83. uerum AG: sed M
84. infrigidatur AG: infrigdatur M
11
Physics before the (Physics' 93
hot potentially, but dry actually. <13> For we have said that in the ele-
ments alone do the qualities exist actUally; in other things <they exist> on-
ly when they are affected with respect to the elements.
<14> For the Wisdom of God 15 made the elements interchangeable, so
that neither they themselves, nor those things which are perfected by
them, perish. <15> For we see, when earth liquifies 16 <and> is stretched
out, it becomes water. But water, on the contrary, if it is squeezed and so-
lidifies, is earth. The same <water>, if it is heated and gives off vapour
from itself, ends up as air. Air, also, when made continuous and thick, be-
comes water. If the same <air> is dried, it changes into fire. Likewise, if
fire is extinguished and loses its dryness, air comes. 17 For air is <both> ex-
tinguished fire and the vapour from hot water. <16> It is clear, then, that
air comes from these two <elements>, and, because of this, air is hot;18 it
is made cold, however, by the closeness of water and earth: its lower parts
are cold, its higher parts hot. <17> For this happens because it is quickly
affected, since it is of a mobile nature.
<18> Aristotle understood the states of air to be two: one which pro-
ceeds from the vapour of water, the other from the vapour of extin-
guished fire. 19 The smoke of fire is hot; also <the vapour> of water in its
place is hot. But when it moves from its place, it gradually cools down
and becomes water. <19> Lest he (Aristotle) is proved guilty of ignorance
and false judgement since higher places are <generally> found <to be>
colder, he says that the nature of air is double.
<20> Altogether, then, every body is made up of all the elements, both
15. The Greek and Arabic have "the Creator wisely [ ... ]".
16. MSS AG add "as <in the case of> lead (+ also A)", which has no equivalent in
the Greek and Arabic and was probably originally a gloss.
17. If"eat" is not a mistake for "fiet" (MS G) or "exit", it may have been used for
variatio, since it translates the same verb as is translated in the surrounding phrases
by "fit", "est", and "mutatur in".
18. The meaning should be "air is hot through its own nature". .
19. The Latin uses "fumus" for both "vapour" and "smoke", whereas two differ-
ent words are used in both the Greek and the Arabic. This broad use of "fumus" is
also found in the Pantegni, where the « natural spirit arises from the fumus of the
perfect blood»: see Burnett and Jacquart, Constantine the African, p. 108.
11
94
mentis et quicquid terrre radicitus est affixum85 et animalium. 86 Natura
enim quod magis mundificatum est omnium generum corporibus di-
stribuit.87 <21> Nominat uero ista corpora Aristotiles naturalia qure non
ita coniunguntur ut maneant, uerum in naturam88 eorum qure ea consti-
tuunt transeunt et postmodum non separantur. Ipsa uero quattuor - id est
terra, aqua, aer,89 ignis - in singulis non separata inueniuntur. Sed quod fit
ex illis non est quicquam90 ex illis, sicut compositum ex quattuor farma-
Chii S91 unguentum. 92 <22> In unguento 93 enim partes minutre sunt et
compositionis unio 94 est ex sola uicinitate. In corpore uero elementa
commutantur 95 et uniuntur. Iterum, destructis corporibus,96 fit ad ele-
menta transitio. Igitur omnia97 manent, nec fit rerum multiplicatio nec
diminutio. <23> Ideoque dicitur quod unius rei gene ratio est destructio al-
terius rei, et econuerso. Et non solum in anima sicut diximus est huius-
modi consideratio, uerum 98 in corpore quoque. <24> Inquit enim Plato
quod tria elementa mutantur ad inuicem, terra uero99 immobilis
perseuerat. <25> Propter hoc organorum formre elementa similia facit.
Qure organa lineam equalem 100 circumdant. Terra autem 101 in chibicis 102
85. affixum AG: affixus M
86. animalium] Natura traxit sibi quod subtile est et mundum in elementis, ut
subtilem terrre et ignis et aeris et aque facit unum quid, ut in homine aliquo. Ideo
uetitum est siccare aliquam, quia trahis subtilem partem add. M
87. distribuit] pariter omni generi corporum add. G
88. naturam AM: natura G
89. Ipsa uero [ ... ] aer AG: Terra uero aer aqua M
90. quicquam AG: quicquid M
91. farmachiis Ap.c.M: formachiis Aa.c. farmaciis G uel tis G supra
92. unguentum GM: unguenti A
93. unguento AGMp.c.: unguentum Ma.c.
94. unio AG: humor M
95. commutantur AG: mutantur M
96. iterum destructis corporibus AG: destructisque corporibus iterum M
97. omnia] genera add. G
98. uerum] etiam add. G
99. uero] om. M
100. equalem AM: rectam G
101. autem A: enim GM
102. chibicis A: chibicum G chibic M
11
Physics before the (Physics' 95
what is fixed to the earth by its roots, and <the bodies> of animals.20 For
Nature has distributed what is more cleansed to corporeal entities of all
kinds. 21 <21> But Aristotle calls those corporeal entities «natural» which
are not so joined together that they remain <elements>; rather, they pass
into the nature of those things which they constitute, and afterwards are
not separated.22 The four <elements> themselves, i.e., earth, water, air,
and fire, are not found separate in single things. But what is made from
these is not anyone of them, like an ointment composed of four ingredi-
ents.23 <22> For in an ointment there are very small particles and the uni-
ty of the composition consists in their proximity <to each other> alone,
whereas in a corporeal entity the elements are changed and united.24
Again, when bodies are destroyed, a passing over into the elements oc-
curs. Therefore, all things25 remain, and no multiplication 26 or diminu-
tion of things occurs. <23> For this reason it is said that what is the gener-
ation of one thing is the destruction of another thing, and vice versa. This
kind of observation <can be made> not only in <the case of> the soul, as
we have said,27 but in the <case> of the body also. <24> For Plato says that
three elements change into each other, but the earth stays unmoveable.
<25> Because of this he makes the elements similar to the shape of instru-
ments 28 which surround themselves with an equalline.29 For he thinks
20. MS M adds in the text at this point what was probably originally a gloss. The
meaning of this is obscure, but the first phrase is a more literal translation of the
phrase that immediately follows ("Natura traxit [ . " ]" = tile; <puaewe; [ .•• ] eAXOUOT)e;,
al-.tabl'at akhadhti; contrast "Natura [ ... ] distribuit" AGM), which suggests that the
addition is due either to the translator, or was already in the original text.
21. The Latin translator and the scribes had problems with this phrase, which, in
Greek, means "to the generation of all corporeal entitites". MS G adds a second
translation: "pariter omni generi corporum", but in both versions "genus" is erro-
neously used in place of "gene ratio" (= Alf.; cfr. yEveOl.e;, tawallut/).
22. The meaning should be "are not separable".
23. The reading ofMSS GM - "compositum [ ... ] unguentum" ("ointment com-
posed of [ ... ]") - is probably better.
24. The difference is between a mixture and a compound.
25. MS G adds "genera" ("kinds"), taking up "omnia genera" in <20> above.
26. "Augmentum", or "crescentia" (= Alf.), would be a better translation.
27. This is a reference to an earlier passage in Nemesius's text (ch. 2 or 3), not in-
cluded in this excerpt.
28. The sense required is "solid figures". "Organorum" looks as if it may be a
mistake for "ort(h)agonorum", but "rectangles" is equally inappropriate in the con-
text.
29. The second "organa" is redundant, but the Latin writer is obviously not fa-
miliar with geometry. MSS AM's "equal line" recalls Nemesius's reference to tri-
11
formam arbitratur, quia non tantam quantam alia habet mutationem;
aquam 103 uero uiginti edras habere autumat;104 mobiliorem autem for-
mam igni attribuit; aerem quippe edras octo habere existimat,105 quia ue-
locior aqua et106 igne grauior est; ac per hoc elementa tria mutabilia, ter-
ram immobilem iudicauit. <26> Harum107 ergo trium figurarum - id est
ignece et octo 108 atque .XX.109 edrarum - compositio est triangula
equalem 110 circumdans lineam; chibicus 111 autem circumdans rectam 112
lineam. Forma quippe triangula, si diuiditur,113 formam sui similem facit;
cibicum 114 uero impossibile est harum trium unam facere, quia eorum
nullum est e quattuor angulis compositum. <27> Necessarium ergo quod
omnia qualitatem habeant 115 ad inuicem sicut et ipsce figurce. Terra tamen
non prorsus passione caret, sed leuia corpora earn diuidunt et operantur
secundum elementa, nec conuertitur ad illud·per quod diuiditur, uerum
in suam naturam mutatur. 116
Hoc 117 enim in aqua inuenimus: si parum pulueris tollis et ponis in
aquam et moueas 118 earn, dissoluitur et cum aqua puluis unitur; si uero
non moueatur 119 aqua et quiescit, inferiora petit puluis et separatur ab
aqua. Hcec operatio non est conuertibilis, sed mixta separantur. <28>
103. aquam AM: aqua G
104. autumat AG: autumauit M
105. existimat AM: estimat G
106. et AM: etiam G
107. Harum AM: Horum G
108. octo] aeris A supra
109..xx.] aqure A supra
110. equalem AM: rectam G
Ill. chibicus A: cibicus G chibic M
112. circumdans rectam AM: rectam circundans G
113. diuiditur] destituitur add. M
114. cibicum A: chibic GM
115. qualitatem habeant] inv. G
116. mutatur AG: reuertitur M
117. Hoc GM: Hec A
118. moueas AG: mouis M
119. moueatur AG: mouetur M
11
Physics before the Physics) 97
earth <has> the form of a cube,30 because it does not have so great a
change as the other <elements>. But he asserts that water has twenty
bases; and attributes the more mobile form to fire)1 He thinks that air has
eight bases because it is swifter than water and more sluggish than fire;
and through this he has judged that three elements are changeable, <but>
the earth is immobile. <26> The composition of these three shapes - i.e.,
the fiery,32 that of eight, and that of twenty, bases33 - is triangular, sur-
rounding <them> with equallines;34 but the cube surrounds <them> with
a straight line.35 The triangular form, of course, if it is divided, makes a
form similar to itself; but it is impossible for a cube to make one of these
three, because none of them is composed of four angles. <27> It is neces-
sary, therefore, that they all have the quality in respect to each other that
these shapes also have. Earth, however, does not completely 36 lack the
ability to be affected, but light bodies divide it and operate according to
the elements, and <earth> is not changed to that by which it is divided,
but changes <back> into its own nature. 37
For we find this in water: if you take a small amount of dust, put it in
water, and shake it (the water), the dust dissolves and becomes one with
the water. But if the water is not shaken and becomes still, the dust makes
for the lower parts and is separated from the water. This operation is not
convertible, but what are mixed are separated.38 <28> Hence Plato says
angular figures with three equal sides (equilateral), but MS G's reading "rectam
line am" retains one of the Greek and Arabic terms used here, "euthugrammos": ta
mepe4Lata tr::rv euthJyp~~wv OXTI~cXtwv' ("the solids of straight-lined figures"; cfr.
quae rectis constant lineis AI£).
30. The Latin is corrupt. Perhaps one should read "terram (> terra in) chibicis
(from )(u~1.)(iic;) formam <habere> arbitratur"; "form of a cube" would be equiva-
lent to the Arabic al-shakl al-muka"ab ("the cubic form").
31. The parallel Arabic text suggests that "formam ignis (= "pyramid", al-shakl al-
na") igni attribuit" should be read, and that the first "ignis" has dropped out by
haplography.
32. Again, "the fiery shape" is a literal translation of the Arabic for "pyramid".
33. MS A has added above the line "aery" and "watery" respectively, perhaps
confused by the preceding "fiery" for "pyramid".
34. I.e. equilateral triangles.
35. This is corrupt. Perhaps read "chibicis autem <compositio quadrangula>
equalem circundans lineam" ("but the composition of a cube is quadrangular, sur-
rounding <them> with equal lines").
36. The Greek and Arabic omit "completely".
37· The Latin adds "nature".
38. The required translation is: "This operation does not involve conver-
11
Vnde dicit Plato quod terram ignis acuitas 120 dissoluit et mouet 121 in se; in
aqua et aere est similiter. <29> Et diuidit elementa alio modo, dicens sin-
gula elementa tres qualitates habere, ut ignis acuitatem, raritatem habet et
motum, extremum elementum - id est terra - tres habet qualitates igni
contrarias - id est obtusum, spissum et stabile, et istis qualitatibus ignis et
terra sibi oppugnant. Hoc122 non 123 inuenitur in qualitatibus qure sunt se-
cundum complexionem. Dicit 124 quoque, si elementa recipiunt125 de suo-
rum extremitatibus sociorum unam qualitatem, faciunt alia elementa.
<30> Vt si iungas cum duabus qualitatibus ignis126 - id est cum raritate et
motu - unam terrre qualitatem - id est obtusum 127 - fit aer. Hre ergo tres
qualitates aerem faciunt: id est obtusum et128 rarum et motus. 129
Iterum,130 si iungas cum duabus terrre qualitatibus - id est cum obtuso
scilicet131 et spisso - unam qualitatem ignis - id132 est motum - fit aqua.
Obtusum ergo et spissum et motus 133 aqure necessario faciunt formam.
<31> Sicut ergo 134 acutum ad obtusum, ita ignis ad aerem; et sicut rarum
ad spissum, ita aer ad aquam; et sicut ad stabilitatem motus, ita aqua ad
terram se habet.135 Superficiei natura est cuius communis terminus linea
est,136 coniungens extremas lineas; corporum uero minime, sed duo com-
munes termini duas extremitates coniungunt. <32> Dicitur quod elemen-
ta habent alio modo qualitates: terra enim et aqua naturaliter grauitatem
habent, deorsum 137 mobilia; aer et ignis econtrario superiora naturaliter
petunt.
<.33> Stoici quoque quattuor elementa dicunt esse: eorum duo faciunt,
ut aer et ignis, duo quidem patiuntur, ut terra et aqua.
120. terram ignis acuitas AG: terra ignis acuitatem M
121. mouet AM: mouetur G
122. Hoc AM: Hec G
123. non AG: nunc M
124. Dicit GM: Dixit A
125. recipiunt AM: accipiunt G
126. ignis AG: ignem M
127. obtusum] et rarum et motus add. M
128. et] om. M
129. motus AM: motum G
130. Iterum AM: Item G
131. scilicet] om. M
132. id AG: hoc M
133. motus AM: motum G
134. ergo] se habet add. GM
135. se habet] expung. G
136. linea est G: linea A est linea M
137. deorsum G: seorsum AM
11
Physics before the (Physics' 99
that the sharpness of fire dissolves earth, and moves it into itself <i.e.,
fire>. In water and air <the case> is similar. <29> He (Plato) also divides
the elements in another way, saying that single elements have three qual-
ities, e.g., fire has sharpness, thinness, and movement. The <other> ex-
treme element, i.e., earth, has three qualities contrary to fire, namely, the
blunt, the thick, and the stable, and by these qualities fire and earth op-
pose each other.39 This is not found in qualities which are according to
mixing (complexion). He (Plato) also says: if elements receive one quali-
ty from the extremes of their companions,40 they make other elements.
<3 0> E.g., if you join with two qualities of fire - i.e., thinness and move-
ment - one quality of earth - i.e., the blunt - air is made. These three
qualities, therefore, make air: namely, the blunt, the thin and movement.
Likewise, if you join to the two qualities of earth - i.e., the blunt and the
thick - one quality of fire - i.e., movement - water is made. Therefore,
the blunt, the thick and movement necessarily make the form of water.
<3 1> Just as, then, the sharp is to the blunt, so is fue to air; and just as the
thin to the thick, so is air to water; and just as movement to stability, so is
water to earth. The nature of a surface is <that> whose common limit is a
line joining the extreme lines,41 but this is not the case for corporeal enti-
ties; rather, two common limits join two extremes. <.32> It is said that ele-
ments have qualities in another way. For earth and water naturally have
weight, moving downwards, but air and fire, on the other hand, naturally
seek the higher <regions>.
<33> The Stoics also say there are four elements. Two of them are ac-
tive - i.e., air and fire - two are passive - i.e., earth and water.
sionlchange (~ttPOA'" istil}ala), but a sequence of mixing together and separating
again".
39. "Sibi inuicem" would be expected.
40. The sense required is: "If one of the two extreme elements receives one
quality from the other [ ... ]". The use of the word "socius" in this context arises out
of a confusion caused by the fact that the Arabic word ~ahib has the double sense of
"companion" and "the other (of two)", but means "the other" here {the Greek
omits}.
41. The Latin writer appears to think that a definition of a "surface" is involved,
rather than the Platonic idea that a common ratio {wttAoyitt, nasiba} connects two
surfaces, in contrast to three-dimensional bodies, for which two common ratios are
required. The Arabic follows the Greek here.
11
100
<34> Aristotiles autem quintum138 his elementis adiungit - id est
,
!
ether,139 circularem motum habens - et non autumat ex quattuor ele-
mentis crelum constare. Plato autem palam ostendit quia crelum de terra
et igne factum est, <35> dicens: « quod factum est necesse est 140 cor-
poreum esse,141 uisibile, quod tangi potest. Impossibile est quidem 142 uisi-
bile sine igne consistere; rursus,143 quod tangi potest sine solido corpore,
corpusque solidum preter terram. Quapropter corporibus omnibus terra
et ignis principant et impossibile est duo esse quin colligata sint a medio
quodam», et quod coniungit aliud 144 quam ea qure coniunguntur est.
Quod plane sentitur uel 145 sentimus in duobus elementis superius dictis.
<36> Sunt nonnulli qui de existentia creli terrreque dissident, dicentes
<37> quod non de elementis crelum et terra essentiam trahunt, quia in
quibusdam codicibus dicitur: «in principio creata 146 sunt crelum et terra».
<38> Et item dicitur quod ex chaos celum et terra consistunt, rur-
sumque 147 tacetur, ipsumque chaos et creatum et elementum putatur et
pro148 omnibus corporibus preparatum, quasi ab eo omnia essentiam tra-
hant. Hoc enim elementum - id est chaos - in Greca lingua elementum
indefinitum determinatur. <39> Verum quicquid sit 149 horum, sciendum
est tantum 150 ex nichilo creata esse omnia.
<40> Amplius, in eos qui dicunt elementum esse solummodo unum -
id est 151 aut ignem aut terram aut aerem 152 seu aquam - sufficit nobis
Ypocratis dictum dicentis: «Si enim homo ex uno tantum consistit, im-
passibilis est; non enim quod sibi ob stet inuenitur. Quod, si patitur,
necesse est ut ab uno 153 curetur». <41> Quodcunque enim 154 recipit pas-
sionem, duo necessario 155 habet, uidelicet mutationem et sensum. Si
138. quintum G: quantum A qntum M
139. ether AG: corpus Ma.c. ethere M i.m.
140. est Ap.c.GM: esse Aa.c.
141. esse] aut add. M
142. quidem A: om. G quid M
143. rursus AG: rursum M
144. aliud] est add. AG
145. sentitur ud] om. M
146. creata AM: creati G
147. rursumque AG: rursum hoc dictum M
148. pro] ud pre G supra
149. sit AM: sunt G
150. est tantum A: tamen est G est tamen M
151. id est] om. G
152. terram aut aerem GM: aerem aut terram A
153. necesse est ut ab uno MG: ab uno necesse est ut A
154. enim] om. A
155. necessario AG: necessariam M
Physics before the (Physics'
<34> Aristotle, however, adds a fifth <element> to these elements,
101
11
namely, ether, which has a circular motion, and he does not claim that the
heavens are made from the four elements.42 Plato, however, shows that
the heavens are made from earth and fire, <35> saying: «what is made,
must be <something> corporeal <and> visible, which can be touched. It is
impossible that what is visible exists without fire; again, that what can be
touched <exists> without a solid body, and a solid body aside from earth.
Therefore, in all corporeal entities, earth and fire dominate, and it is im-
possible that two things exist without being bound by a certain mean»,
and what joins is different from those things that are joined. This is clear-
ly sensed (or «we clearly sense this») 43 in the two above-mentioned ele-
ments.
<3 6> There are some who disagree concerning the existence44 of the
heavens and the earth, saying <37> that the heavens and the earth do not
draw their essence from the elements, because it is said in certain books:
«In the beginning the heavens and the earth were created». <38> And,
likewise, it is said that the heavens and the earth are from chaos, and no
more is said. And chaos itself is thought to be both created and an ele-
ment, and prepared for 45 all bodies, as if they all draw their essence from
it. This element, i.e., chaos, is defined in the Greek language as an «in-
definite element ». <39> But, whichever of these it is, it should be known
only that all things are created from nothing.
<40> Another point: against those who say that there exists only one
element - i.e., either fire, or earth, or air, or water - sufficient for us are
the words of Hippocrates saying: <41> For whatever is affected, necessar- *
ily has two <things>: namely, change and sensation. If, then, there is <on-
42. The "non" would seem more appropriate within the indirect statement: "he
claims that the heavens are not made [ ... ]". but the negative has the same position
in the Greek and Arabic, and in Alfano's translation.
43. Evidently this is a second translation of the same word, omitted by MS M.
44. "Existentia" is the wrong word here - "creatione" (= Arabic khalq; the Greek
has no equivalent) would be more appropriate - but is consistent with "essentiam"
in next phrase.
45. MS G gives "ud pre" ("or before"), which also gives good sense.
11
11
102

autem elementum est unum, non inuenitur in quid debeat156 mutari; si157
1 Physics before the Physics' 10 3

non mutatur et in una permanet qualitate, non patitur, licet sensibile sit.
Iterum, quod patitur, necesse est ut ab alio passionem recipiat. Et 158 si est
elementum unum, non inuenitur159 alia160 qualitas preter suam a qua161
162
animal patiatur. Quod si animal non habet a quo passionem recipiat,
163
unde patitur? <42> Vbi ergo ratum 164 fecit quod hoc impossibile est,
per 165 ypothesin 166 concessit secundum esse possibile,167 dicens: «Si pati-
tur, necesse est ut ab uno curetur; uerum uidemus quod cura non est ex
uno sed ex multis; homo ergo non existit ex uno». <43> Apparet ergo168
elementa quattuor esse ex ea quidem 169 ratione qua 170 inconuenientire 171
partem probabant.
<44> Thalius enim, cum putaret172 unum esse dumtaxat - id est
aquam - elementum, significare uolens quod reliqua tria ab aqua
procedant,173 uel 174 credens tria reliqua 175 ab aqua procedere,176 dicit
quod graue et spissum terram faciunt, leue aerem, leuius ignem.
<45> Anaximenus quoque, cum dixit aerem tantum 177 esse, uoluit eti-
am monstrare alia ab eo procederep8 <46> Yrcleus 179 uero et Ypar-
156. debeat AG: debet M
157. si AG: et si M
158. Et AG: Quod M
159· inuenitur AG: inuenietur M
160. alia AMp.c., om. GMa.c.
161. qual qualitate add. G
162. animal] om A
163· unde AM: quomodo Gp.c.
164· ratum AM: rarum G
165. per AM: quod G
166. ypothesin G: ypothesim A ipothesim M
167· secundum esse possibile AM: possibile esse G
168. ergo AG: igitur M
ly> one element, there is not found <anything> into which it should
change. If it does not change and remains in the same quality, it is not af-
fected, although it is able to be felt. Again, what is affected, must be af-
fected by another thing. And if there is <only> one element, there is not
found another quality from which an animal may suffer aside from its
own. But if an animal does not have <something> by which it may be af-
fected, what is the source of its suffering? <42> When, therefore, he
(Hippo crates) made it understood that this is impossible, he conceded as
a hypothesis that a second <explanation> is possible, saying: if it suffers, it
is necessary that it is cured by one thing. But we see that a cure does not
come from one thing, but from many. A man, therefore, does not exist
from one thing. <43> That the elements are four, then, is apparent as a re-
sult of the same reasoning by which they proved the side of inconven-
ience.46
<44> For Thales, when he thought that there was one <element> only
- i.e., water - wishing to designate <it as> an element because the other
three proceed from water (or believe that the other three proceed from
water),47 says that the heavy and thick make earth; the light, air; the
lighter, fire. <45> Anaximenes also, when he said that there was air
alone,48 wished, <in his case> too, to show that others proceed from it.
<46> But Heracleitos and Hipparchos of Metapontos understood that it
i:
f
",':.,,',

169· quidem] om. M ~

170. qua AG: qure M I
'I,

171. inconuenientire AM: inconuentione Ga.c. inconuentire Gp.c.

172. cum putaret AM: computaret G
173· procedant AM: procedantur G
174- uel] om. G
175· reliqua AG: alia M
176. significare [ ... ] procedere A: credens tria reliqua ab aqua procedere signifi-
care uolens quod reliqua tria ab aqua procedant G credens tria reliqua ab aqua pro- 46. I.e., by which they tried to prove their wrong theories.
cedere M
47. This seems to be an alternate translation of "volens quod reliqua tria ab aqua
17'J. tantum] elementum add. G
procedant", perhaps originally added in the margin, and hence inserted in different
178. Anaximenus [ ... ] procedere AG: Anaximenus dixit aerem tantum esse ele- places in MSS A and G and omitted in M.
mentum, uoluit etiam ostendere ut alia procederent ab eo M 48. MS G, perhaps for greater clarity, adds "elementum", giving the meaning:
179· Y rcleus AM: Y rdeus G "that air alone was an element".
 11
11

104 1 Physics before the Physics'

was air from which other elements proceeded. This, like the above-men-
105

thus 180 Minutiensis 181 intellexerunt ignem esse a quo alia elementa
procederent, quod ut supradicti 182 sapientes suis rationibus approbarunt. 183
<47> Quicquid tamen dicant, confirmantur184 eadem elementa uicissim
mutabilia esse. Quod cum sit, mutabilia necesse est omnia esse elementa,
quodcumque enim horum accipias,185 alterum ab altero nasci uidebis. 186
tioned' wise men, they approved by their own49 reasoning. <47>
Whichever they say <is the single element>, however, the same elem~n~s
are confirmed to be mutually convertible. And, since this is the case, It lS
necessary that all the elements are convertible. For, whichever of them
you take, you will see that one of them is born from the other.

2. CERTAIN <WORKS ON> NATURAL SCIENCE

2. QUE DAM PHISICA

I) Costa ben Luca, De physicis ligaturis; edited by J. Wilcox andJ.M. Rid- I) The De physids ligaturis is translated in Wilcox and Riddle, pp. 40-5 0.
dle, Qus{a ibn Luqa's Physical Ligatures and the Recognition of the Placebo Effect,
«Medieval Encounters », I 1995, pp. I-50 (31-39).
2) De metallis MS A, fo1. 20IV. 2) <On the metals>
This text is found elsewhere only in Oxford, Bodleian Library, Digby
121 (= D), fols Ir-2T, whose readings have been added in the apparatus. D
omits all the names of the metals inserted interlinearly in MS A.
(No title) 187

<I> Septem sunt genera metalli: primum argentum uiuum, secundum
plumbum, tertium stagnum, quartum aurum, quintum argentum, sex-
turn ferrum, septimum res. <.2> Hrec omnia sic Bunt: quando aqua grauis
et uiscosa est in uisceribus terrre, non habens exitum foras, si prope sit sul-
phur, ilIa aqua, longo tempore excocta sulphureo calore, uertitur in hrec
omnia secundum magis et minus: <3> in primis (supra: mer(curius)) in ar-
gentum uiuum;188 quod si calor sulphuris superare non poterit illud ui-
uum argentum, semper sic permanet. 189 Quod si (supra sol) uis sulphuris
superet et illud uiuum argentum sit190 darum et lucidum et durum et
graue, uertitur in aurum. Quod (supra l(una)) si sit191 darum et lucidum,
<I> There are seven kinds of metal: the first is mercury, the second,
lead, the third, tin, the fourth, gold, the fifth, silver, the sixth, i~on, the
seventh, brass. <.2> All these are made in this way: when water lS ~ea~
and viscous in the bowels of the earth, 50 having no exit to the outslde, If
sulphur is dose by, that water, cooked for a lon.g time by the heat of sul-
phur, is turned into all these <metals> according t? a greater or lesser
<amount>: <3> at first, into quicksilver {Mercury);51 lf the heat of t~e su~­
phur is not able to overcome it, that quicksilver remai~s a~ways. m this
state. If the force of sulphur overcomes it, and that qUlcksllver lS dear,
bright, hard and heavy, it turns into gold (the Sun). If it is dear and bright,

180. Yparthus A: Yparcus G Iparchus M

181. Minutiensis AMp.£:: Minuciensis G Minutientis Ma.£:
182. supradicti AM: supra dixi G
183. approbarunt AG: approbauerunt M
184. confmnantur AG: confirmatur M
185. accipias AM: accipies G
186. uidebis] explicit add. G
187. Incipit liber primus de .VII. generibus metallorum D 49. The meaning "the same" is required. . ,,' " " .
50. It is not dear whether there is a deliberate word-play W1th VlSCOSUS (VlS-
188. uiuum] darum et lucidum add. D
189. sic permanet] siquidem manet D cous") and "viscera" ("bowels" or "viscera"). .'
51. The abbreviations of the names of the planets assoaated W1th each metal are
190. sit] fit D
19I. sit] fit D added by the original scribe above each of the metals.

11
106
minus graue atque durum, fit 192 argentum. Et 193 (supra m(ars)) si sit graue
e.t ~urum,194 non clarum et obscurum, fit ferrum. Quod (supra s(aturnus))
~I SIt ~a~e et molle et ferro obscurius, fit plumbum. Quod (supra i( up-
Iter)) SI SIt clarum et lucidum et molle et plumbo minus graue, sumit na-
turam 195 stagni.196 Quod (u(enus)) si sit siccum et minus graue et luci-
dum, uertitur in es. <4> Hrec autem .vii. metallica similitudinem habent
.vii. planetis, hoc in ordine 197 artifice sapientia Dei 198 distinguente: satur-
~um i~ta~r plumbum, quia utrumque frigidum et obscurum et graue;
lOuem Imltatur stagnum, quia utrumque est clarum, lucidum, humi-
dum; 199 martem imitatur ferrum, quia siccum, graue et obscurum est
utrumque et 200 calidum; solem imitatur aurum, quia est clarius, lucidius,
grauius, et calidius; venerem imitatur es, quia clarum, siccum, minus
graue et lucidum; argentum uiuum mercurio est simile - nam 201 que-
madmodum realiter 202 uidemus quod argentum uiuum attrahitur et in-
cor'~oratur creteris metailicis, ita mercurius attrahitur et superatur creteris
stellis planetarum; argentum lunre similitudinem seruat, quia clarum, lu-
cidum, frigidum, humidum. <5> Inter hos203 .VII. planetas et .VII. metalli-
ca, ut luna et sol digniora sunt creteris planetis, ita204 aurum et argentum
digniora sunt creteris metallis.
3) De dbis, MS A, fols. 20IV-202r. This text has not been identified in any
other manuscript. Abbreviations which could be realised in different ways
(i.e. "6" ="om" or "on"; ne ="hec" or "hrec") have been printed in italics.
DE CIBIS. Centum genera ciborum sunt quibus nutritur corpus
preter animalia.
Pomorum .xxx. preter agrestia.
<I> De his .x. sunt quorum exteriora sumuntur in cibum, interioribus
192. fit] autem add. D
193. Et] Quod D
194. durum] et add. D
195. naturam] et materiam add. D
196. stagni] stangni D
197. hoc in ordine] in hoc ordine D
198. sapientia Dei] inv. D
199. clarum, lucidum, humidum] clarum et lucidum et humidum D
200. et] om. D
201. nam] quia D
202. re aliter] reuera aliter D
203. hos] has D
204. ita] et add. D
11
1 Physics before the Physics' 107
but less heavy and hard, it becomes silver (the Moon). And if it is heavy
and hard, but not clear, and is dull, it becomes iron (Mars). If it is heavy
and soft, and duller than iron, it becomes lead (Saturn). If it is clear,
bright, soft and less heavy than lead, it takes on the nature 52 of tin
Oupiter). If it is dry and less heavy, and is bright, it turns into brass
(Venus).53 q> These seven metals have a likeness to the seven planets,
which the creating wisdom of God distinguishes in this order: lead imi-
tates Saturn, because each of them is cold, 54 dull and heavy; tin imitates
Jupiter, because each is clear, bright and moist; iron imitates Mars, be-
cause each is dry, heavy, dull and hot; gold imitates the Sun, because it
(gold) is clearer, brighter, heavier and hotter <than the rest>; brass imi-
tates Venus, because it (brass) is clear, dry, less heavy, and bright; quicksil-
ver is similar to Mercury - for, just as we see in reality that quicksilver is
attracted and incorporated by the other metals,55 so Mercury is attracted
and overcome by the other stars of the planets; silver preserves a similari-
ty to the Moon, because it is clear, bright, cold and moist. <5> Among
these seven planets and seven metals, just as the Moon and the Sun are
more worthy than the other planets, so gold and silver are more worthy
than the other metals.
3) On Foods. The problems of identifying the names of the natural
foodstuffs here are such that a translation is inappropriate.
52. MS D adds "and material".
53. The criteria are brightness and dullness, heaviness and lightness, and hardness
and softness. These are qualities of the quicksilver. In "Marius", De elementis, too
the quicksilver has different qualities, being bright ("clarus"), thick ("spissus"), and
gross ("grossus"), but the variety of the metal also depends on the colour of the sul-
phur and the length of time of the "cooking"of the two.
54. The author adds another differentiating criterion here, namely, the four qua-
lities.
55. The phrase "incorporatur creteris metallicis" is unclear, ~spe~a1ly since the
author has previously said that all the metals are made from qUlcksllver. Note also
"metallica" for "metalla".
 11
11

108
Physics before the (Physics' 109

repudiatis, ut sunt cerasa, pruna, persica, brucuca, i.e. mala cocila, oliua, 4) De pondenbus (see p. 66 above).
dactili, iuiuba, olinica, et ualia, zarur, nepica.
<2> De quibus proiciuntur exteriora, et interiora comeduntur sunt
hrec: amigdala, nux, auellana, castanea, pinea, fistica, nux indica, mala
4) On weights. See p. 66 above.
punica, glans.
<3> Alia uero .x. tota manducantur, ut hec: ficus, mora, pira, mala cito-
nia, citra, sorbia, iumeis, xilo carrata, uua.
<4> Alia .x. que sunt pinguedo terrre, i.e. melones, citruli, cucumeres,
cucurbitre, citrones magni, musa, cannamel, molongianre, cucumeres ro-
tundi, que dicuntur facus.
<5> Rursus alia .x. sunt quorum radices manducantur, ut cepe, allia, ra-
pa, radices, pastinaca, rafanus, qui dicitur colocas, napones, uel nauuntij,
cambricombos, i.e. radix ca\u!lis, terfes, quoddam rotundum in modo oui
gallinre, nigrum sine foliis, sine radice.
<6> Item, .x. grana sunt seminum de quibus fit farina panis, ut sunt hec:
granum hordei, alica, grana hordeum, si gala, milium, panicum, risi, far,
medica.
<7> Leguminorum (sic) genera sunt .X.: faba, cicer, dolica, lenticula,
fafelus, mauricocus, pisa, lupina, cicercula, /202r/ lenticula rub ea.
<8> .x. genera sunt seminum unde fit oleum et comeduntur: sisamum,
linisemen, cannabi semen, radicis semen, et rape, semen melonis magni, se-
men cucurbitre, semen lentisci, semen lactucre, semen papaueris.
<9> .x. sunt genera foliorum, duo genera caulium, unum genus habet
folia extra explicata, alterum habet intus inclusa, quod grece dicitur sicco-
cambri,20S i.e. inclusus caulis, spinachie, attriplices, bleta, portulaca, lactu-
ca, scariola, molothi.
<10> Item .x. alia folia que et ortolana et agrestia inueniuntur, sicut
feniculum, sparagus, brusci, malua, lapsana, borraginas,206 cardus, herba
citola, cannana, eruca.
<11> Item .x. \alia! folia que sunt uictus etiam sicut condimenta, appi-
urn, petrosilinum, menta, cerifolium, porrus, nasturcium,207 ruta, anetum,
coriandrum, synapis, sisimbrium.

205. or 'sictocambri'.
206. borragenas a.c.
207. nastarcium a.c.
,
I
III
\

ADELARD OF BATH AND THE ARABS

In this paper I would like to explore an enigma: the English scholar *

Adelard of Bath, who was adult by the beginning of the twelfth century
and probably still alive in 1150 A.D.!. The still unrivalled authority
Charles Haskins dubbed him «the frrst. .. to assimilate Arabic science in
the revival of the twelfth century»2, and yet the equally renowned MIle.
d' Alvemy wrote that «one searches vainly for a precise citation of a
single Arabic authority in his works»3. In the Quaestiones naturales
Adelard states on several occasions that what he has written results from
his «Arabic studies» (Arabum studia), and yet Brian Lawn can state that *
«the work contained ... no specifically Arab learning»4. Let us turn
again to what Adelard says about his debt to the Arabs: For seven years
he has scrutinized the Arabum studia to the best of his ability 5. His

1 Throughout this article frequent reference will be made to Adelard of Bath: An

English Scientist and Arabist of the Early Twelfth Century, London, 1987 (Warburg
Institute Surveys and Texts, XIV), which is a collection of papers edited by C.
BURNETI (= Adelard of Bath) ; I am particularly indebted to the insightful article by
Margaret GIBSON on Adelard's biography on p. 7-16 of this volume. Other works
referred to in abbreviated form are Adelard's De eodem et diverso, ed. H. WILLNER,
MUnster i.W., 1903 (Beitrage zur Geschichte der Philosophie des Mittelalters, 4, 1)
(=DED), Quaestiones naturales, ed. M. MOLLER, MUnster LW., 1934 (Beitrage zur
Geschichte der Philosophie des Mittelalters, 31, 2) (= QN), and De opere astrolapsus,
ed. B.G. DICKEY, in Adelard of Bath: an Examination Based on Heretofore Un-
examined Manuscripts, unpublished Ph.D dissertation, University of Toronto, 1982
(= DOA). All these texts of Adelard will be cited by page and line numbers.
2 C.H. HASKINS, Studies in the History of Mediaeval Science, 2nd edition,
Cambridge, Mass., 1927, p. 42.
3 M.T. D' ALVERNY, Abelard et I' astrologie, in Pierre Abelard - Pierre le Vene-
rab
e, C.N.R.S. colloque n° 546, ed. R. LOUIS, J. JOUVET et al., Paris, 1975, p.
639: «11 est probable qu'[Adtlard] ne savait que peu ou pas d'arabe ... On cherche
vainement une citation precise d'un seul savant arabe dans ses oeuvres».
4 The Salernitan Questions, Oxford, 1963, p. 21 ; revised Italian edition I
Quesiti Salernitani, 1969, p. 39.
5 QN, 4, 29-32: Meministi, nepos, septennio iam transacto ... id inter nos
convenisse, ut Arabum studia ego pro posse meo scrutarer.
III
,
I
90
«nephew» has often heard him expounding the opmlOns of the
Saracens6. He argues the cause of the Arabs not his own 7. He follows
the example of the Arab masters who use rational arguments, rather than
blindly following authority8. In his work on the astrolabe (De opere
* astrolapsus) he writes: «What I have learnt from the Arabs I will write
down in Latin»9, and encourages his princely dedicatee (the future King
Henry 11 of England) to be bold enough to wish to understand not only
the things that Latin writings contain, but also the opinions of Arabs
concerning the sphere and the courses and movements of the planets 10.
An exploration into what Adelard meant by «Arabic studies» may lead
us to an understanding of the significance of the meeting of European
and Arabic cultures in the first half of the twelfth century.
-1-
Adelard is a colourful character. We have a picture of him wearing a
bright red cloak over a blue shirt and a blue cap striped with red and
bordered with ermine. He is teaching the elements of the abacus to two
bearded and bored-looking students, huddled in drab coats 11 • This pic-
ture, in a fifteenth-century manuscript, is likely to be pure guess-work
on the part of the illuminator. However, it captures Adelard's perso-
nality very nicely. He himself tells us he wore a green coat and a green
6 QN, 5,2-7 (nepos speaking) : Saracenorum sententias te saepe exponentem
auditor tantum notaverim earumque non paucae satis futiles mihi videantur... quippe
et illos impudice extollis et nostros detractionis modo inscitia invidiose arguis.
7 QN, 5, 14-17: <me> Arabicorum studiorum sensa putet proponere ... causam
Arabicorum, non meam agam.
8 QN, 11,23-24 : Ego enim aliud a magistris Arabicis ratione duce didici, tu
vero aJiud auctoritatis pictura captus capistrwn sequeris; cf. also QN, 1,20: nepos ...
aliquid Arabicorum studiorum novum me proponere exhortatus est; QN, 12, 14-15 :
(nepos speaking) : neque Arabum tuorum auctoritates sequi tutum sit; QN, 37, 8-10
(nepos speaking) : Cum enim nuper a parte orientali venires, qua causa studii diutur-
ne steteras, ego ... lacrimasfudi.
9 DOA, 148, 10: quod Arabice didici, Latine subscribam.
10 DOA, 147,9- 148, 1: non solum ea quae Latinorum scripta continent" intel-
ligendo perlegas, sed et Arabum sententias super spera el circulis stellarumque moli-
bus intelligere velle presumas.
11 Leiden, Bibliotheek der Rijksuniversiteit, Scaliger, 1, f. lr; this picture is
reproduced as the frontispiece to Adelard ofBath.
III
ADELARD OF BATH AND THE ARABS 91
emerald ring. He hunted with a hawk and played the lute, and discussed
medicine and astronomy with wise men on the banks of the Loire and in
the bright sunshine of Salerno. He was fascinated by theories of vision,
and himself directed his gaze most constantly at the stars.
It is not inappropriate, then, to imagine his works as forming a grand
triptych, of the type familiar from medieval altarpieces. On the one side
panel we have the works intended for the entertainment and not-too-
arduous instruction of young noblemen and the members of bishops'
households: the work on the care of hawks (De cura accipitrum), the
little allegory in verse and prose on the seven arts (De eodem et diverso),
the introduction to the abacus, and, in the most prominent position, the
masterly dialogue concerning the causes of things (the Quaestiones
naturales). In the central panel are the raw materials of Arabic mathe-
matics. Most important are the Elements of Euclid in a translation of an
Arabic version, and the Astronomical Tables (Zij) of al-Khwarizmi.
They are flanked by three shorter works: an introduction to astrology
by Abii Ma'shar, some astrological maxims attributed to Ptolemy, and a
work on magic talismans by Thabit ibn Qurra. The right-hand side panel
contains only one work - a work which incorporates the results of the
major Arabic translations just mentioned, and interprets them for a Latin
audience. This work - the De opere astrolapsus - includes, as its first
part, Adelard's conception of the shape and the function of the universe
- a cosmos the operation of which is encapSUled in the «bronze book»
of the astrolabe l2 .
A considerable amount has been written about the chronological
sequence of these works 13 . The facts are complicated in that
(a) The translations are all undated.
(b) The other works can only be dated by their dedications or by internal
references. In the case of the dedications, one must be aware that a
work could have been completed many years before it was dedicated to a
patron; the occasion of the dedication was probably dictated by the
personal circumstances of the author - e.g., he needed a job, - and an
12 For these works see the catalogue of writings of Adelard in Adelard of Bath, p.
163-196.
13 See especially HASKINS, Studies (op. cit., n. 2 above), p. 26-27, 29 and 33-
35, LAWN, The Salernitan Questions (op. cit., n. 4 above), p. 26-30 (Italian ed., p.
45-48), and DICKEY,Adelard of Bath (op. cit., n. 1 above), p. 8-13.
III
92
old work might be dragged out again for this purpose. The dedication of
the astrolabe treatise to the young Henry suggests a date as late as 1149-
50. However, the contents, as I show later, do not preclude a consi-
derably earlier date for its composition. In the case of the internal refe-
rences, the facts are unreliable, for the real Adelard and «Adelard» the
protagonist in the dialogue may be different.
The fmnest indication of at least a relative chronology is given by
cross-references. These establish that the De cura accipitrum was writ-
ten after the Quaestiones naturales, and the De opere astrolapsus follow-
ed De eodem et diverso, and the translations of the Elements and the Zij.
Instead of elaborating convoluted theories about the dating of his
works, as others have done, I would prefer to group them according to
their literary form and manuscript affiliation. The works in the left-hand
panel all feature the «nephew» (nepos), either as the silent recipient (De
eodem et diverso) or the active partner in the dialogue (as in the Quaes-
tiones naturales and the De cura acciptrum)14. The three shorter Arabic
translations occur together in manuscripts already from the mid-twelfth
century, and we may presume they were written close on each other's
heels.
The two side panels of the triptych balance each other. Both include
original works by Adelard, which show his characteristic literary style :
a fluent, idiomatic Latin, rich in anecdotes, wit and quotations from the
Latin Classics. The works in both panels have prefaces dedicating the
texts to noble or royal patrons. The dedication of the astrolabe text to
Henry regis nepos who has just reached (or just passed) the age of
discretion (Le. 16) nicely recalls the persona of the young nepos in the
earlier works. The De opere astrolapsus and the De eodem et diverso
both use the Geometria incerti auctoris - a work from the Latin land-
measuring tradition - and the later work refers the reader to the earlier.
The works on the two side panels contrast with the central group in
that they are addressed to the cultured amateur - they are literary pieces
sugaring the pill of the scientific or quadrivial branch of the seven liberal
arts which were - at least in theory - the backbone of every gentle-
14 I believe that what is usually read as H. suo in the one manuscript of the
Abacus treatise (Leiden. Scaliger I) which contains the dedicatory preface could also
be read asN{epotiJ suo.
III
ADELARD OF BATH AND THE ARABS 93
man's education. As Adelard writes in the preface to the Regulae abaci:
When he and his nephew were dining at the «philosophical table», his
nephew noticed that most people were tucking into the trivial dishes with
relish and leaving aside the quadrivial dishes. So Adelard had to per-
suade his nephew to tackle the unappetizing scientific works by fIrst
presenting him with a Pythagorean hors-d' oeuvre - Le. the «table» or
«course» (mensa has both meanings) of Pythagoras 15 • The «table of
Pythagoras» is, as is well known, the traditional Latin name. for the
counting-board or abacus. The preface to the astrolabe treatIse pro-
pounds the ideal of Plato's philosopher-king I6 . Adelard r~commends
that the King's family should study the liberal arts, and pomts out the
importance of a knowledge of cosmology by saying that, just as the
householder should know the composition of his own house, so the
dweller in the universe should know the reason for the wonderful beauty
of the cosmos, or else be cast out. He gives a liberal education a moral
justification17.
The central pieces in the triptych, however, are of a completely dif-
ferent tone. They lack dedications; no attempt has been made to make
the Latin elegant. They are unimaginative workmanlike translations
from the Arabic. Adelard's name appears only in the titles: Ysagoge
minor per Adelardum Batlwniensem ex Arabico sumpta ; Astronomi-
corum prestigiorum... (liber) per Adelardum Batlwniensem ex Arabico
translatus ; Ezich Elkauresmi per Athelardum Batlwniensem ex Arabico
sumptus ; Institutio artis geometrice ... per Adelardlim Batlwni~ns~m ex
Arabico in Latinum sermonem translata, etc. The only authonal mter-
vention is perhaps typical of Adelard' s wit: when describing a talisman
15 Regulae abaci. MS Scaliger I. f. Ir: Cum inter nonnulla /erc~l~ phil~so­
phicae mensae apposita nobis deXlrorsum solitariis discumbentibus pr~xunl. co~vlvae
de parte secunda tripliciter sumerent. et me de quadrifida lance pauca 0" tuo Insullante
omnia /astidires. quippe quae ab aliis seposita et hactenus intemptata tibi videres •
Pitagoricum antidotum ante praelibasti.
16 DOA. 147.5-7: Ait enim beatas esse res publicas si aut philosophis regende
tradantur aut earum rectores philosophU adhibeantur (from BOETHIUS. De consola-
tione philosophiae. I. pr. iv (ed. R. PEIPER. p. 11. l. 15». .
17 DOA. 148. 1-6: Dicis enim ut in domo habitans quilibet. si materiam elus
vel compositionem. quantitatem et qualitatem. situm et distinctionem ig~ret. ta~i
hospitio dignus non est. Sic qui in aula mundi natus atque educatus est. SI tam mI-
rande pulcritudinis rationem scire negligat post discretionis annos. ea indignus atque.
sifieri posset. eiciendus est.
III
94
for driving scorpions out of a city, he mischievously substitutes Bath
(his own city) for Baghdad l8 . The translations are meant for personal
use, or for the use of his own apprentices. These are no longer young
noblemen or princes, but are people known, in most cases, only as
students and/or authors of mathematical works.
For example, an «Ocreatus» is responsible for adding some proofs to
Adelard's translation of Euclid's Elements l9 , and he composed a little
treatise on an Arabic multiplication table for his master20 . Several
names, including that of Ocreatus, are mentioned in glosses to the
Elements, and echo the light-hearted banter of the school room: «This
proposition of ours is proved true without any help from John» ; «Only
Adelard will be able to understand this problem» ; «Goodbye, Regine-
rus. Whoever does not know how to reply to you should present you
with a white cow ! »21. It is in all probability to this circle of Adelard
with his pupils or apprentices that the reference in a mid- to late- twelfth
century introduction to the arts from the Franciscan Convent in Coventry
applies:
The difficulties of geometry go back to Pythagoras ... Then, after Hippo-
crates and Aristotle, Euclid alone among the Greeks dealt with the whole
subject. Someone called Qusta ibn Liiqa treated the same subject in Ara-
bic ... Boethius among us [Latins] is more a translator and interpreter than
an original writer - which is the case too of certain modem writers who
have become famous in the subject, such as Adelard, John and Williarn22.
18 MS Lyon 328, f. 73v (Adelard of Bath, p. 135, n. 10).
19 M. CLAGETT, The Medieval Latin Translations from the Arabic of the
<Elements> of Euclid, in [sis, 44 (1953), p. 16-42 (see p. 20-21).
20 The introduction to this text is headed Prologus Ocreati in Helceph ad Adelar-
dum Batensem magistrum suum ; see Adelard of Bath, p. 174-175.
21 See M. FOLKERTS, Adelard's Versions of Euclid's <Elements>, in Adelard of
Bath, p. 55-68 (p. 63-64). The Latin expressions, which were often copied from one
manuscript into another, are: Vera ergo probatur esse proposicio nostra absque
lohannis industria .. quod propositum est Adelardi patebit ingenio .. Vale, Reginere,
quicumque nesciret libi sic respondere utinam dmet libi vaccam alham.
22 This accessus has been edited from MS Cambridge, Trinity College, R.15.16,
f. 1-3, in BURNETI, Some Unusual Divisions of Science, in Actes du 8e congres
international de philosophie medievale, ed. S. KNUUTTILA, Acta philosophica
fennica, Helsinki, 1990.
III
1 ADELARD OF BATH AND THE ARABS 95
The translations reflect the chalk-dust (or rather, the wax tablet and
scraper) of the school-room. Not only are they unpolish~ and no~­
literary, but also they change from manuscript to manuscnpt : there IS
no canonical text. In regard to al-Khwarizmi's astronomical tables
several well-known astronomers of the second quarter of the twelfth
century seem to have had a hand in tampering with the text, including
Petrus Alfonsi, Hermann of Carinthia and Robert of Ketton. At each
stage not only was there a revision of contents, sometimes with refer-
ence to an Arabic text, but also the technical vocabulary was changed.
Robert in particular strives to replace Arabic terms by Latin terms -
incidentally inventing in the process the term sinus (our «sine»)23.
The glosses in these early manuscripts of Euclid's Elements and al-
Khwarizmi's astronomical tables show how masters and students were
struggling to understand the Arabic text better, giving Latin, and so~e­
times Greek, equivalents to Arabic technical terms, and even addmg
Arabic grammatical notes, or whole phrases written in Arabic trans-
literated into Roman characters24 .
One may contrast the manuscripts of the QULlestiones naturales which
are rarely glossed, and when glosses do occur they merely give variant
readings, or parsing signs. That the Quaestiones naturales were regard-
ed not so much as a scientific textbook as a literary work and a model of
good Latin style is indicated by the fact that they accompany works like
pseudo-Quintilian's Declamationes undeviginti maiores and works of
Hildebert and Proba Falconia in some manuscripts25 .
The contrast between the side-panels of the triptych, therefore, and
the central picture is that between works written for the general education
of a man of culture and works written for specialists - a contrast well
brought out by Adelard's slightly younger contemporary Hermann of
Carinthia.
23 See R. MERCIER, Astronomical Tables in the Twelfth Century, in Adelard of
Bath, p. 87-119. .. .
24 See my Some Comments on the Translating of Works from ArablC mto Latm
in the Mid-Twelfth Century, in Miscellanea mediaevalia, 17 (1985), p. 161-171 (see
p. 167-168).
25 E.g., Leiden, Bibliotheek der Rijksuniversiteit, Voss. lat. Q.84, and the lost
manuscript from Waltham Abbey (Adelard of Bath, p. 182).
III
96
1
I
In the introduction to his De essentiis, written in 1143, Hermann of
Carinthia addresses his fellow translator, Robert of Ketton, in the
following terms :
You remember, I think, that while we went forth from our inner sanc-
tuaries into the public festival of Minerva, the multitude of people milling
around were gaping at us with open mouths, not valuing us so much as
individuals as admiring the trappings and decorations which long vigils and
our most earnest labour had acquired for us from the depths of the treasuries
of the Arabs. At that time I began to have a very deep sense of pity
concerning those men who were so impressed by these outward appearan-
ces : How much they would value the undergarments, if it were lawful for
them to look at them !26
Hennann is reluctant to reveal the «undergannents», but is eventually
persuaded to do so by the Goddess of Wisdom (Minerva) herself, who
says to him: «Distribute what has been given to you with a liberal hand
- without any hesitation. For our wealth increases when it is given
freely»27.
The import of the highly artificial language of Hermann' s introduction
is that Robert and he have been studying Arabic astronomy in the «inner
sanctuaries» (adyta) of the Ooddess of Wisdom. He has pity on the
general public who admire the trappings of the astronomers without
understanding the basic principles which he and Robert have spent many
sleepless nights working out. In the De essentiis he undertakes to reveal
these underlying principles (which he coyly calls the «undergarments of
the Goddess») to the general public. The De essentiis recasts, in the
form of a work of literature, the works of Abii Ma'shar and al-Battani,
Euclid and Theodosius, which Hennann has been translating.
But is it the case that Adelard in his works for the generally well-
educated gentleman, like Hermann in his De essentiis, is revealing the
secrets of the translations that he has been studying with his professional
pupils?
It is in the popular texts on the side panels of oUf triptych that Adelard
specifically refers to Arabic learning, and he identifies it as his own.
26 De essentiis, 58rD, ed. C. BURNEIT, Leiden, 1982, p. 70.
27 Ibid., 58rH, p. 72.
III
ADELARD OF BATH AND THE ARABS 97
The two texts which make this point most clearly are the Quaestiones
naturales and the De opere astrolapsus. How do these works relate to
the Arabic learning of the translations Adelard was responsible for?
-ll-
The most important references to Arabic in the Quaestiones naturales
have been mentioned at the beginning of this article. No one has known
how seriously to take Adelard. He says (or at least has the «Adelard>~ of
the dialogue saying) that he has been away for seven years studymg
Arabic whilst his «nephew» has been wasting his time on French
education. He dedicates his De eodem et diverso to William, bishop of
Syracu se. He could have known him in Normandy or England since
many of the Norman dignitaries in Sicily were recruited from the «other
branch» of the Norman clan, and kept up close connections with their
Anglo-Norman colleagues28 . But the facts that the «nephew» in the
Quaestiones naturales implies that «Adelard» has seen Mt Etna29 , and
that «Adelard» at least had journeyed as far South as Salemo, where he
discussed medj.cine and natural science with a Greek30, tend to suggest
that his sojourn in Syracuse was real. If so, he might have picked up
Arabic there.
But the curious thing is that
(a) in spite of detailed investigation, in particular by Brian Lawn, no use
of Arabic sources - not even of the heavily disguised Arabic texts
translated by Constantine the African - has been discovered in the
Quaestiones naturales.
(b) The Arabic texts which Adelard translated came not from Sicily but
from Spain : the tables of al-Khwarizmi were in the form revised for the
meridian of Cordova. The book on casting magical talismans is known
in Arabic only in the Ghayat al-8ilim (Picatrix) - a mid-ele~enth
century Spanish Arabic compilation of magic. Moreover, I am relIably
28 For a summary of the connections between England and Sicily in this period
see R.V. TURNER,Les Anglo-Normands et la Sicile, in Etudes Normandes, 3 (1986),
p.38-60.
29 QN, 19, 9-10.
30 DED, 33, 18-34,2.
III
98
infonned, the transliterations of Arabic words in Adelard' s works betray
a Spanish, not a Sicilian-Arabic, pronunciation31 . Although there were
some renowned Arabic scholars in Sicily (such as al-Idrisi), Latin
scholars turned directly to the Greek texts which were available there
(but not of course in Spain) for the recovery of ancient mathematics.
Thus, Euclid's Elements and Ptolemy's Almagest received their first
Latin translations from Greek in Sicily within a few years of Adelard's
lifetime32. One of the translators in Sicily - Eugenius - had recourse
to an Arabic version of Ptolemy's Optics only because he could not fmd
a Greek copy, which he very much regretted 33 . Moreover, as far as we
know, works of astrology and magic were not translated into Latin in
Sicily during this period (while these formed the majority of texts
translated in Spain), and few medical or physical translations were added
to the Constantinian corpus anywhere in the first half of the twelfth
century.
Most telling is the fact that De eodem et diverso, though addressed to
the bishop of Syracuse, does not mention Arabic learning at all, but
only, as one might expect, the superior Greek learning of Magna
Graecia : «What French studies are ignorant of, studies from beyond the
Alps will describe : what you do not learn amongst us Latins, eloquent
Greece will teach you »34 . Why then the constant harping on about
Arabic studies in the Quaestiones naturales? People have suggested
that Adelard is simply employing a literary device. The Arab epitomizes
the exotic: thus the alleged origin of Wolfram von Eschenbach's
Parzifal is an Arabic story by Kyot. The voyages of the «Adelard» of
the dialogue to Cicilia, Tarsus and Antioch in search of Arabic wisdom
are akin to other sages' trips to Egypt or India in search of the prisca
theologia.
31 This has been confinned for me in private communication with P.Sj. van
Koningsveld.
32 See I.E. MURDOCH, Euclides graeco-latinus : a hitherto unknown medieval
Latin translation of the <Elements> made directly from the Greek, in Harvard Studies
in Classical Philology, 71 (1966), p. 249-302.
33 See M.T. D' ALVERNY, Translations and Translators, in Renaissance and Re-
newal in the Twelfth Century, ed. R.L. BENSON and G. CONSTABLE, Oxford, 1982,
p. 421-462 (see p. 435).
34 DED, 32,27-29: Quod enim Gallica studia nesciunt. transalpina reserabunt ;
quod apud Latinos non addisces. Graecill facuntkJ docebit.
III
ADELARD OF BATH AND THE ARABS 99
Nevertheless there might be some truth in Adelard's claim. When
writing the Quaestiones naturales he may already have started to study
Arabic mathematics and astronomy. He mentions that the cause of
eclipses is discussed by many people35 and that «certain people» have
written about the contrary motion of the planets36 • He is already
promising to write about astronomy himself:
Not all philosophers are of the opinion that the higher stars are fixed in the
aplanos. For they say that they turn about in the air in a much lower
course below the firmament. This I will deal with in its own place, if I
live long enough37 .
In another passage in the Quaestiones naturales his nephew reminds
Adelard how they spent a few days with an old sorceress (anus
praestigiosa) with the intention of learning how to perform magical
incantations38 , since Adelard was most «studious of incantations»39.
They may have been preparing to translate the Arabic text of Thabit ibn
Qurra which Adelard calls the Liber praestigiorum and which includes
precise instructions on the prayers to be directed to the spirits, in order to
drive out scorpions, for destroying a city etc. Moreover his sympathy
for astrology (which led him to translate Abli Ma'shar's introduction) is
evident from several passages in the Quaestiones naturales40. That there
are not more parallels between the Quaestiones naturales and the trans-
lated works is doubtless due to the fact that the subject-matter of the
former is not mathematics, astrology or magic, but «physics». Even
35 QN, 62, 27-29 : Neque illud nunc arguo, cur eclipticam patiatur obscurita-
tem ; tractatum etenim hoc a multis est.
36 QN, 64, 16: Hoc enim quod quidam scripserunt. intelligo.
37 QN, 32, 32-36: non omnibus ... philosophis placet ... superiores stellas in
aplano fixas esse. Aiunt enim et eas longe inferiori ambitu infra caelum in aere volvi.
Quod suo in loco. si vitam vixero. diffiniens tractabo.
38 QN, 53,27-30 (nepos speaking) : Cum enim tempore ut scis iam praeterito
anum praestigiosam studio incantationis discendae adissemus. ibique anilibus imbuti
sententiis nescio an sensibus aliquot diebus moraremur ...
39 QN, 54, 1-2 (nepos speaking) : Tu vero more tuo. quoniam incantationibus
studiosus eras.
40 QN, 67, 10-11 : Si eorum [i.e., the planets'] actio inferiorum animalium
mortis vitaeque causa sit ... ; cf. QN. 61, 37-38; cf. DED, 32, 7-10: Hanc [sc.
astronomiam] si quis sibi privatam facere posset. non modo praesentem rerum
inferiorum statum, verum etiam praeteritum vel futurum non diffiteretur. Superiora
quippe ilia divinaque animalill inferiorum naturarum et principium et causae sunt.
III
100
when the subject-matter «rises to the stars» in the final portion of the
Quaestiones naturales4 1, the questions asked are physical: the reason
why the Moon alone of the planets lacks its own light, the .cause of its
duskiness, the reason for falling stars, and the question of whether
planets have souls.
In summary, one might say that when Adelard came to write the
Quaestiones naturales he had already started to translate texts from
Arabic, or, at least, had embarked on the subjects in which he later was
to translate texts. That still does not justify the clear implication of his
words in the Quaestiones naturales, that the subject-matter of that work
has been taken from his «Arabic studies», because it is apparent that no
Arabic book has been used as a source. In his treatise on the astrolabe
similar statements occur concerning Arabic studies. Since this work
manifestly draws on Arabic-Latin translations made by Adelard, we
might expect Adelard's claim to Arabic authority to be more cogent here.
Let us turn to the treatise on the astrolabe.
-111-
In the preface to the treatise on the astrolabe Adelard, as we have
seen, promises that «what he has leamt from the Arabs he will write
down in Latin»42 and he exhorts the young prince Henry to be bold .
enough to wish to understand not only the things that Latin 'writings
contain, but also the opinions of Arabs concerning the sphere and the
courses and movements of the planets43 • Note that in the second phrase
the Latin sources are described as being written down (Latinorum
scripta), whereas the Arabic sources are not described as being «written»
but only referred to as sententiae. The treatise on the astrolabe falls into
two halves: the first half is the cosmology, the second half describes
the parts of the astrolabe and their use. At the beginning of the second
half Adelard once again refers to his authorities. These are his magistri:
41 QN, 61,22 (nepos speaking): Huius itaque induti pennis ascendamus ad
astra.
42 DOA, 148, 10: quod Arabice didici. Latine subscribam.
43 See n. 10 above.
III
ADELARD OF BATH AND THE ARABS 101
«What I have received from my masters should be written down»44.
Throughout the second half there are frequent references to these
«masters» : «The longitudes of the climes were understood by our
masters by means of the difference of the hours of the eclipse»45.
«Concerning their longitude should be added frrst what we have taken
from our masters, then what we have worked out for ourselves,»46 etc.
It becomes clear that these magistri are Arabic speakers when Adelard
writes: «Now we must speak concerning the equation of the places,
which our masters call tezwietelbuiut»47. Is it plausible then, to suggest
that Adelard used Latin writings, but Arabic informants in writing about
the astrolabe ? What do the sources of the work indicate ?
Aside from the Classical Latin prose and poetic texts (including
Macrobius, Cicero, Virgil and Lucan) appropriate to the literary style of
the work, the treatise on the astrolabe used two of the earliest and most
widely-copied Latin texts referring to the astrolabe.
a) The Geometria incerti auctoris, compiled at the end of the tenth or the
beginning of the eleventh century48 - the earliest work of the Latin
land-measuring tradition to include procedures using an astrolabe.
b) The De utilitatibus astrolabii of Gerbert of Aurillac - a literary
reworking of the frrst Latin translation of an Arabic astrolabe text49•
Both these works occur together in two of the manuscripts of
Adelard's translation of the astronomical tables of al-Khwarizmi (MS
Chartres 214 and Oxford, Corpus Christi College, 283). Both works
once occurred alongside Adelard's translation of Abii Ma'shar and
Thabit's work in the mid-twelfth-century manuscript, Avranches 235.
Most significantly, the De utilitatibus astrolabii is found in the Worcester
Cathedral manuscript of the tables of al-Khwanzmi (Oxford, Bodleian
44 DOA, 172, 11-12: De hoc itaque quod a magistris acceperim. ut scientiam
sitienti morem geram. non minus subscribendum est.
45 DOA, 191,6-7: Sciendum itaque quod longitudo climatum non aliter quam
per diversitatem horarum eclipsis a magistris nostris hactenus deprehensa est.
46 DOA, 193,6-7: De eorum longitudine ea prius quae a magistris accepimus,
ilia deinde quae a proprio ingenio habuimus subdenda sunt.
47 DOA, 216, 7-8: Nunc de casarum equatione, quam magistri tezwietelbuiut
vacant, dicendum est.
48 See N. BUBNOV. Gerberti postea Silvestri 11 papae Opera mathematica (972-
}OO3). Berlin. 1899. who discusses and edits the work on p. 310-364.
49 Ed. N. BUBNOV, ibid.• p. 109-147.
III
102
, ADELARD OF BATH AND THE ARABS
Library, Auct. F.1.9) which is closest to Adelard's circle. The trans-
literation of Arabic terms in this copy is remarkably close to that in
Adelard's treatise on the astrolabe. Note especially the representation of
the ain as a c or ch in alcankabut and alchidoda.
To the Worcester Cathedral version of the Zij the treatise on the
astrolabe refers frequently (with such phrases as «This will be demon-
strated by an astrolabe with the help of the Zij»50) and it is the coin-
cidence in phraseology between the two works which is the principal
indication adduced by Bruce Dickey that the Worcester Cathedral version
of the Zij must be assigned to Adelard51 . Almost all the Arabic words in
the treatise on the astrolabe which are not already in De utilitatibus astro-
labii can be found in the Zij52, with one conspicuous ex'ception which I
shall mention shortly.
The other work written in Latin to which Adelard refers is his own
translation of Euclid's Elements: «the fifteen books of the art of geo-
metry which I translated from Arabic into Latin »53 . And this, as far as I
have been able to determine, is the sum total of the written sources of the
treatise on the astrolabe. They include both the Latinorum scripta which
Adelard had inherited from his predecessors, and the new translations -
of al-Khwarizmi and Euclid - which he assumed his reader to have at
hand.
50 DOA, 209,4-5: hoc et auxiliante Ezic astrolabio d~monstrabitur.
51 Adelard of Bath, p. 35-44.
52 These are almustakim (K34, M8),felekburug (K7, M7+4),felektedewir (K8,
M7+31), elmobzote (K61, MI9), elmankiuz (K62, M22), adhel (K61 and K62),
schemelia (K49), genubia (KlO, MI5), ezic (M34), dikaikas (K4, M5), tezwietelbuiut
(K52) ; words already occurring in De utilitatibus astrolabii are : alcotob (K40),
alchidada (KI9), artifa (K17), almukantaraz (K31), alcankabut (Kl), almori (K32a),
almagriba (K24). Elofoc (<<horizon») (K55) occurs elsewhere only in De horologio
secundum aikoram, a text occurring in the same manuscripts as De utilitatibus astro-
labii; alcemut (K44) does not seem to occur in the Zij, or De utilitatibus astrolabii
or any of the texts associated with the latter, and usually appears in the fonn azimuth
in twelfth-century astrolabe texts. [K =item number in P. KUNITZSCH, Glossar der
arabischen Fachausdracke in der miuelalterlichen europtiischen Astrolabliteratur,
1982, p. 455-571 (Nachrichten der Akademie der Wissenschaften in GOttingen, phil.-
hist. Klasse) ; M =item number in Appendix to R. MERCIER, Astronomical Tables
in the Twelfth Century, in Adelard of Bath, p. 116-118].
53 DOA, 193,3-4 : intelligat eam apud Euclidem a quindecim Iibris artis geo-
metrice.quos ex Arabico in Latinum convertimus sermonem esse coniciendam.
103
III
This does not, however, exhaust the sources of the work. There are
anecdotes, descriptions of the meanings of Arabic words (which are not
always explained in the Zij)54, personal asides, etc., which suggest in-
formal conversations. Recently MIle d' Alvemy pointed out a similarity
- in concept, but not in terminology - between the ten-sphere universe
that appears in the cosmological section of the De opere astrolapsus and
the cosmos described by Petrus Alfonsi in his Dialogi 55 . Another pas-
sage in Adelard' s work gives the impression of being based on anec-
dotes:
[The fIrst clime], by the natural orientation of its position, may be said to
be the domus philosophica. For there all seeds spring up spontaneously
and the inhabitants preserve an unblemished rightness of action and truth in
speech in all situations, subservient only to God in the fIrst place, but to
the spirits of the planets in a secondary way. Here all men share every-
thing in common and follow the way of nature and reason only. When
they meet any man of any religion in their common dealings they greet
him with this maxim: «aiekadeb», which means «beware of the beast». If
we follow the Arabs, these people were granted this fatherland, in which,
when all the planets except for Mercury were in their exaltations, and at the
time which the Creator decided on and when the condition of the heavens
was encouraging generation, the fIrst man was bom56 .
This has no direct literary parallel, but once again it is a passage from
Petrus Alfonsi's Dialogi which gives the most similar sense:
54 To the words listed above the following meanings have been added in DOA
only: genubia =dextra, elofok =orizon, tezwietelbuiut =aequatio casarum. DOA
corrects the interpretation in the Zij of mankiis as diminutus to indirectus, adopts the
fonn minutiae instead of minuta for daqB'iq, and retains circulus for falak, and rectus
for al-mustaqim.
SS See PSEUDO-ARISTOTLE, De elementis, in Pseudo-Aristotle in the Middle
Ages: The Theology and Other Texts, London, 1986, p. 63-84 (see p. 65).
56 DOA, 169,9-170,9: Undefit ut naturali sedis positione domus philosophica
esse perhibeatUT. fllic enim et omnia semina sponte proveniunt et indigene tam
morum honestatem quam verborum veritatem modis omnibus illesam custodiunt,
solique Deo principaliter, stellarum vero numinibus secundario obnoxii, in communi
omnia ponentes feliciter degunt solamque natUTe et rationis viam sequentes. Cum
aliquem cuiuslibetlegis virum in commerciis suis vident. proverbio utuntUT tali :
aiekadeb, id est «cave bestiam». Quibus. si Arabes sequimUT, eam patriam habitare
datum est in qua primus homo. omnibus planetis preter MercUTium in regnis suis
existentibus, Creatore volente statuque celi ad generationem applicante, exortus est.
III
104
, ADELARD OF BATH AND THE ARABS
The atmosphere (at Arin in the fJfst clime) is most well-tempered. so that
the periods of spring. summer. autumn and winter are always equal. There
grow aromatic spices of a beautiful colour and honeyed taste. The bodies
of men are not too thin or fat. but well fleshed out. The temperateness of
the weather renders men's bodies consonant with themselves. because they
abound in wisdom and naturaljustice57.
Three additional elements are found in Adelard's story :
a) The worship of the spirits of the planets: this apparently is a
condonement of the procedures illustrated in the Liber praestigiorum of
Thabit, translated by Adelard58 .
b) The Arabic phrase used in greeting - aiekadeb : this is the one
Arabic term that is conspicuously absent from the De utilitatibus
astrolabii and the Zij59.
* c) The statement that all the planets were in their exaltations except
Mercury when Adam was born. This idea does not occur in Adelard' s
literary sources, but it does give a clue to his oral sources. For precisely
this statement occurs written in Arabic (but in Roman letters) in the
margin of the Zij in the Worcester Cathedral manuscript60•
What I am suggesting is that the Arabic stories and the meanings of
Arabic words have in fact been told to Adelard by an Arab speaker. All
the Arabic glosses in the Worcester manuscript and in related manus-
cripts are in transliterated form, and occasionally two Arabic words have
57 Dialogi. in PL. 157. c. 547C-D (concerning Arin which is the city in the fIrst
clime through which the Prime Meridian passes) : probamus... aerem ibi tempe-
ratissimum esse, adeo ut veris, aestatis, autumni et memis semper ibifere tempus sit
aequale, ibi aromaticae species pulchri coloris et melliflui nascuntur saporis. Corpora
quoque hominum non macilenta ibi sunt nimis aut pinguia, sed mediocris succi
discretione decora. Temporis quoque temperies hominum corpora sibi consona reddit
et pecora, quia ineffabili pollent sapientia et 1IlJlurali justitia.
58 Cf. MS Lyon. 328. f. 73r: inscribanturque eis quatuor stellarum anuli ... hac
prece interdicta: '0 stellarum istarum spiritus, promovete hunc .. .' .
59 The Arabic original of this expression would appear to be iyyaka <WB> dabba.
«beware of the beast».
60 MS Bodleian Auct. F. 1.9. f. 127v: Cullel.kauuekib.kumna.fiscerafehu.ille
otari ahine unxi ademu .. i.e. kull al-kawakib qumna fi sharafiha illa 'Ufjrid ~na
unsm'a adamu ; see H. SUTER et al.• Die astronomischen Tafeln des Mul)ammed ibn
Miisa aJ-Khwarizmi. Copenhagen. 1914. p. xxi.
105
III
been confused because of their aural similarity61. The confusion would
not have arisen if Adelard had seen them written down. Adelard could
have jotted down the striking anecdote about the birth of Adam when
working on the Zij with an Arabic speaker, and then incorporated it, in
Latin translation, into his treatise on the astrolabe. In short, there is no
evidence that Adelard read Arabic. Nor did he claim to. What he learnt
from the Arabs he learnt not from scripta, but from teachers, or a
teacher. For this reason the search for Arabic written sources in the
Quaestiones naturales too is futile.
We may even guess who Adelard's teacher was. None other than
Petrus Alfonsi. He was responsible for one form of the tables of al-
Khwarizmi, he collaborated with Walcher, prior of Malvern in the dio-
cese of Worcester on an astronomical text which appears in the Worces-
ter Cathedral manuscript, he was from Huesca in Northern Spain (Ara-
gon)62. His collaboration with Adelard would explain:
a) the similarities between doctrine in the treatise of the astrolabe and
Petrus' Dialogi.
b) the fact that the texts that Adelard translated are Spanish Arabic, and
that the form of transliteration of Arabic words suggest Spanish pro-
nunciation.
c) One may even suggest that Petrus, who is described as a «doctor»,
and who gives physica prominence in his account of the parts of learning
in his Letter to the Peripatetics of France, helped to determine the
subject-matter of the Quaestiones naturales63.
One might, therefore, go on to doubt that Adelard's «translations» are
translations at all, in the proper sense. They may rather be Latinizations
of translations made by Petrus, or, assuming that Adelard could at least
speak and understand Arabic (probably from his sojourn in Sicily),
Petrus may have dictated the Arabic texts for Adelard who would write
61 Thus mankas, «reversed» is incorrectly rendered as diminutum. as if it were
man~ii$ (see n. 54 above).
6 On Petrus Alfonsi see J .M. MlLLAs-VALUCROSA. La aportaciOn astronOmica
de Pedro Alfonso. in Sefarad. 3 (1943). p. 65-105; J.HL. REUTER.Petrus Alfonsi:
An Examination olms Works, their Scientific Content, and their Background. D.
Phil. diss.• Oxford. 1975 and M.T. D' ALVERNY. Pseudo-Aristotle, De elementis. art.
cit. (n. 55 above).
63 See MlLLAs- VALLICROSA. art. cit. (n. 62 above). p. 99.
III
106
down the Latin versions of the words as he heard them (the procedure
apparently used by Gundissalinus and Avendauth in Toledo a few de-
cades later)64. What can be said with reasonable certainty is that neither
the Quaestiones naturales nor the treatise on the astrolabe give evidence
that Adelard read any Arabic text directly - nor even that Adelard was
in a milieu in which Arabic texts were available. For, when the astrolabe
treatise was dedicated to Henry, in 1149-50, it was very much out of
date. By that time the new tables of Toledo, which progressively re-
placed those of al-Khwarizmi, were already known in the South of
France, for in 1145 they were adapted to the meridian of Marseilles by
Raymond of Marseilles. In 1143 Hermann of Carinthia had already
translated the Planisphere of Ptolemy and written a much more sophis-
ticated cosmology than that in the fIrst part of the De opere astrolapsus
(i.e. Hermann's De essentiis). Ptolemy's Almagest had become known
through the epitomes of al-Battani and al-Farghani, both translated in the
1130s. We must imagine Adelard working in the West Country of
England still using texts which Petrus Alfonsi had brought from Spain
with him soon after his conversion to Christianity in 1106 - working in
isolation away from the new currents of learning flowing from the
Arabic world into Christian Spain, the South of France and Italy.
So for the scholar in Worcester or in Bath in the second quarter of the
twelfth century the studia Arabum were something remote, something
from another world 65 . But for Adelard they were more than that.
Throughout his original writings Adelard prides himself on being a
Platonic philosopher. Plato is «meus Plato» or «familiaris meus»66.
Adelard adheres to the Calcidian motto that «for everything there must be
a legitimate cause and reason (ratio )>>67. Geometers establish that ratio at
64 See BURNETI, Some Comments, art. cit. (n. 24 above), p. 165-166.
65 One may contrast the case of Constantine the African : for him as he worked
in Salerno or Montecassino, the Arabs were close and a real threat Hence he chose to
disguise the Arabic origin of his translations; see D. JACQUART, Principales etapes
dans la transmission des teXles de medecine (X/e_x/ve siecle), this volume, p. 251,
sgg.
66 DED, 13,34 and 13,20-21. Cf. QN, 31,20-22 (nepos speaking) : Cum enim
et in philosophicis contemplationibus et in physicis causarum effectibus ethicisque
etiam consultibus Platoni te perutus consentire perceperim ... ; QN, 34, 39 (nepos
speaking) : Plato tuus.
67 Timaeus a Calcidio translatus commentarioque instructus, ed. J.H. WASZINK,
Plato Latinus, IV, London-Leiden, 1975, p. 20. For Adelard's concern with etiology
, ADELARD OF BATH AND THE ARABS 107
III
a practical level- by drawing lines, in cutting out the parts of the astro-
labe, and by making calculations68 . But philosophers pursue a more
noble path - for they discern the truth more clearly through speculative
thought. Throughout the Quaestiones naturales and the treatise on the
astrolabe the philosophers' opinions are contrasted to those of the vulgus
or populus69 . Philosophia is the Goddess to whom Adelard owes alle-
giance in De eodem et diverso, and the utopia in the fIrst clime in the De
opere astrolapsus is the domus philosophica. That this «philosophical
ratio» is Platonic is evident from Adelard's strange claim, in the astro-
labe treatise, that the real universe can only be apprehended intellectually
by an almost angelic intelligence, so that human reason can but aspire to
understanding an image of that world - i.e. the model universe of the
astrolabe7o .
Yet this philosophical ratio is not just Platonic; it is also Arabic.
Jean Jolivet has pointed out several instances in which the «philo-
sopher», the man who uses reason alone to search for the truth, is ap-
parently considered to be an Arab or a Muslim. Such is the philosopher
in Anselm's Cur deus homo and Peter Abelard's Collationes ; Peter the
Venerable appealed to the reason of the Muslims in his Summa contra
Saracenos 71 • In conclusion I would venture to say that Adelard's
conception of Arabum studia was not formed so much by his direct
acquaintance with a limited number of Arabic texts, as by a sub-
conscious belief, nourished - if anything - by lack of direct acquain-
tance with Arabic scholarship, that Arabic learning was rational learning,
and further, that their rationes were Platonic.
see BURNETI, Scientific speculations, in P. DRONKE, A History o/Twelfth-Century
I Western Philosophy, Cambridge, 1988, p. 151-176 (see p. 169-170).
il 68 DOA, 172, 8: the astrolabe is put together geometricarum rationum soller-
tia ; for other references to geometry see DOA, 176, 1-3, 187, 1 and 199, 11.
I
69 Cf. QN, 15, 29-30 and 49, 12-13, and DOA, 203, 6-7 : artificiales quidem
hore populo, naturales vero ut philosophis note sunt, ita populi indignantur. .
70 DOA. 172. 8-11 : quicquid /ere angelica intelligentia in mundo intellectual,
notare debet ac potest, toturn id humana ratio in hoc mundano exemplo ... discernere
possit.
~
. 71 J. JOUVET. L'Islam et la raison tI apres quelques auteurs latins des IXe et XIle
siecles, in L' Art des confins. Melanges offerts aMaurice de Gandillac. ed. A. CAZE-
NAVE and J.-F. LYaTARD, Paris. 1985, p. 153-165.
 IV

Antioch as a Link between Arabic and Latin Culture

in the Twelfth· and Thirteenth Centuries *

Nam qui fuimus Occidentales, nunc

facti sumus Orientales, ... qui fuit
Remensis aut Carnotensis, nunc
efficitur Tyrius vel Antiochenus.
Fulcheri Carnotensis Historia
Hierosolymitana 1095-1127, ed.
H. Hagenmeyer, Heidelberg, 1913,
p.748.

The ancient city of Antioch embraced several cultures and religions.

The population included Greek orthodox, Arabic Melchite (= «Syrian»)
and Jacobite Christians, Muslims and Jews 1 . After its conquest in the
First Crusade in 1098, a Latin Frankish element was added. While Arabic
was the dominant language, Christianity was the dominant religion 2 . This

·1 am most grateful for the help of Silke Ackermann, Luc Deitz, Dag Nikolaus Hasse,
David King, Benjamin Kedar, Paul Kunitzsch, Lucy McGuiness, Michael McVaugh,
Baudouin van den Abeele and Steven Williams for reading sections of this article and
providing materials.
1 See J. PRAWER, Social Classes in the Crusader States: The" Minorities ". in
A History of the Crusades, ed. K.M. SETTON, vol. V, ed. N.P. ZACOUR and H.W. HAZARD,
Madison Wi., 1985, p. 59-115, see p. 59-62, A.N. POLIAK, L'Arabisation de l'Orient
semitique, in Revue des etudes islamiques, 12, 1938, p.35-63, and C. CAHEN, Un
Docurn.ent concernant les Melkites et Ies Latins d'Antioche aux temps des croisades in
Revue des etudes byzantines, 29, 1971, p. 285-92.
2 J. PRAWER, Social Classes ... , p. 65.
IV
would seem to provide ample opportunities for exchange between Arabic
and Latin culture. And yet scholarship has persistently denied, or at least,
conceded little, in respect to such exchange. For example, Charles Homer
Haskins, who still provides the best overview of «Translators in Syria
during the Crusades », states that «the Crusaders were men of action
rather than men of learning, and there was little occasion for western
scholars to seek by long journeys to Syria that which they could find nearer
home in Spain », and his assessment has been followed by most scholars
since3 . While fully admitting that there never developed a systematic
programme for translating Arabic texts in Antioch (or elsewhere in the
Crusader States), as there did in Toledo or in the court of Frederick II
Hohenstaufen, I would still claim that there was a significant cultural
exchange, which up to now has been largely overlooked. In this paper I
would like to explore the nature and contents of this exchange, which takes
a different form from that in the western Mediterranean, but is none the
less far-reaching in its consequences.
I would like to start with a reference not known to Haskins. In the
preface to his translation of Thabit b. Qurra's work on casting talismans
for effecting magic, John of Seville and Limia writes the following words:
This [Arabic] book, then, I, with the help of God's
spirit, obtained from my Master - a book which no Latin
other than a certain Antiochene [A.ntiocenus], who once
obtained a part of it, ever had:'
3 C.H. HASKINS, Studies in the History of Mediaeval Science, 2nd ed., Cambridge,
Ma. 1927, chapter 7 (.. Translators in Syria during the Crusades .. ), see p. 130; followed
by C. CAHEN, La Syrie du nord a npoque des croisades et la principaute franque
d'Antioche, Paris, 1940, p. 569-78 and M.-T. D'ALVERNY, Translations and Translators,
in Renaissance and Renewal in the Twelfth Century, ed. R.L. BENSON and G.
CONSTABLE, Cambridge, Ma., 1982, p.421-62, see p.438-9. A slightly more positive
note is sounded by B.Z. KEDAR in The Subjected Muslims of the Frankish Levant in
Muslims under Latin Rule, 1100-1300, ed. J.M. POWELL, Princeton N.J., 1990, p. 135-74
(reprinted in KEDAR, The Franks in the Levant, 11th to 14th Centuries, Aldershot, 1993,
article XVIII), see p. 174. More positive still is the assessment of R. HIESTAND in Un
centre intellectuel en Syrie du Nord? Notes sur la personnalite d'Aimery d'Antioche, Albert
de Tarse et Rorgo Fretellus, in Le MoyenAge, I, 1994, p. 7-36.
4 This preface is edited in C. BURNE'IT, Magister lohannes Hispalensis et Limiensis
and Qus~ ibn LClqa's «De differentia spiritus et animae» : a Portuguese Contribution to
the Arts Curriculum? in Mediaevalia. Textos e estudos, 7-8 (1995), p. 221-67, see p. 253 :
Hunc ergo librum ab ipso, Dei iuvante spiritu, habui, quem nullus Latinorum preter
quendam Auriocenum (correct to : Antiocenum), qui quondam eius partem habuit,
adeptus fuerat. The identification of this "Antiochene .. with Adelard of Bath was first
made by R. LEMAY in The True Place of Astrology in Medieval Science and Philosophy :
2 3
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
It is very likely that this «Antiochene» is none other than the weU-
known philosopher and translator from Arabic, Adelard of Bath, whose
earlier translation of ThAbit's work is extant under the title Liber
prestigiorum Thebidis secundum Ptolomeum et Hermetem. This has, in
fact, much less material than John's translation. Moreover, we know from
Adelard's own testimony in his original work on Questions on Natural
Science that he was in the principality of Antioch. For, there he relates two
personal experiences: that « a certain old man in Tarsus in Cilicia » taught
him how the web of the sinews of the human body could be detected, by
suspending a cadaver in flowing water5 ; and that he was shaken by an
earthquake when on the bridge in Mamistra in partibus Antiochenis6 .
Tarsus and Mamistra (ancient Mopsuestia) were both in Cilician Armenia,
but they were incorporated into the Crusader principality of Antioch and
remained an integral part of the principality until they were lost to the
Seljuks in 11377 . Adelard must have visited them before this date, since he
considers Mamistra to be in partibus Antiochenis : other evidence for his
biography suggests that he was there in the time of the great earthquake
of 11148 •
Adelard had written, near the beginning of his Questions on Natural
Science, that he had spent seven years abroad. Part of that time was spent
in Sicily, presumably in the household of William, bishop of Syracuse
(attested 1112, 1115 and 1116), to whom he dedicated another work, On
the Same and the Different; from here he visited Salerno and Magna
Graecia. For how much of the seven-year period he was in Antioch is
unclear, but if we are right in identifying him with the Antiocenus
mentioned by John of Seville and Limia, then either he was there long
enough to be identified with the place, or the translation of Thiibit's Liber
prestigiorum was thought to have originated there. In either case, the
possibility arises that Adelard had found the Arabic manuscript ofThiibit's
Towards a Definition, in Astrology, Science and Society, ed. P. CURRY, Woodbridge, 1987,
p. 57-73, see p. 70.
S Questiones naturales, Q16, in Adelard of Bath, Conversations with His Nephew :
On the Same and the Different, Questions on Natural Science, and On Birds, ed.
C. BURNETT, I. RONCA, P. MANTAS ESPANA and B. VAN DEN ABEELE, Cambridge, 1998,
p.122.
6 Questiones naturales, Q50, in Adelard of Bath, Conversations with His Nephew ... ,
p.184.
7 One may note that the three major ecclesiastical centres in the Principality were
Antioch, Tarsus and Mamistra, which were all seats of archbishops.
8 See L. COCHRANE, Adelard of Bath: The First English Scientist, London, 1994,
p. 32-3. William of Tyre singles out Mamistra as being especially affected by the
earthquake in Chronique, 11,23, ed. R.B.C. HUYGENS, 2 vols, Turnhout, 1986, I, p. 529.
IV
1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
work in the Middle East. If this is so, we are bound to ask whether he
found in the same place the Arabic texts of Abo Ma'sar's Abbreviation of
the Introduction to Astrology and Pseudo-Ptolemy's Centiloquium (both of
whose translations accompany that of Thabit's Liber prestigiorum in the
manuscripts, while the first is quoted by Adelard in the Liber
prestigiorum 9 ) and even that of Euclid's Elements, Adelard's most
important translation. This is not implausible. Adelard never mentions
Spain or al-Andalus, and the parts of Sicily and southern Italy that he
visited were predominantly areas of Greek language and scholarship;
Adelard himself mentions them only as a source of his Greek learning1o.
Adelard may have come to the principality of Antioch because of the close
relationship between the Norman princes in Syracuse and in Antioch _
both called Tancred and probably cousinsll.
In Antioch Adelard would have been a near contemporary, if not a
colleague, of the best-known translator who was definitely working there.
This is Stephen of Pisa, who went to Antioch in search of Arabic wisdom.
His importance in the translating-process of the early twelfth century, I
believe, has been underestimated. His role is intimately linked with that of
the relationship between Antioch and Pisa. The Pisans with their ships
helped the Crusaders to conquer the Holy Land. In recognition of their
services in 1108 Tancred, in addition to giving them a large part of the
important seaport of Laodicea (Latakia), presented them with a quarter in
Antioch itself: it included the parish of Saint-Saviour and was in the
centre of the city12.
9 For the first of these texts see ABU MA'SAR, The Abbreviation of the Introduction
to Astrology Together with the Medieval Latin Translation of Adelard of Bath, ed.
C. BURNETT, K. YAMAMOTO and M. YANO, Leiden, 1994. The two extant Arabic
manuscripts of this work are of Middle Eastern origin.
10 ADELARD, De eodem et diverso, in Conversations with His Nephew ..., p. 70. The
astronomical tables of al-Khwlrizml that Adelard translated are in the redaction of
Maslama al-Majrl~l, the tenth-century scholar of C6rdoba, but this work was known in
England to Walcher, prior of Great Malvern, already before Adelard, through the agency
of Petrus Alfonsi, the convert from Judaism who came to England from Huesca, in the
Spanish Muslim kingdom of Saragossa, and Adelard could have used an Arabic
manuscript in England; see C. BURNE'M', The Works of Petrus Alfonsi: Questions of
Authenticity, in Medium lEvum, 66,1997, p. 42-79.
11 See L. NICHOLSON, Tancred, A Study of His Career and Work in their Relation to
the First Crusade and the Establishment of the Latin States in Syria and PalestiM,
Chicago, 1940.
12 G. MOLLER, Documenti sulle relazioni delle citta toscaM coll'oriente cristiano e
coi turchi, Florence, 1879, p.3, no. 1; for further concessions to, and renewals of
4 5
IV
It is becoming increasingly evident that Pisa, aside from being a centre
of commerce, with quarters in cities all round the Mediterranean, was also
one of the key locuses for the translation activity. The Greek-Latin aspect
of this activity is quite well-known. It is in Pisa that Burgundio, the
prolific translator of works by Galen, Aristotle, and the Church Fathers,
spent most of his long life as a judge (he died in 1193)13. His knowledge of
the Greek language and of Greek texts was, at least in part, due to the fact
that the Pisans had a quarter in Constantinople 14 , and that Burgundio
visited the Byzantine capital at least twice during his lifetime. On the first
occasion (1136) he was in the Pisan quarter there in the company of James
of Venice and Moses of Bergamo, as an interpreter for the high-level
discussions between the Western and Eastern church 15. James of Venice's
translations of Aristotle's works complemented those of Burgundio. Pisa
was the home not only for Burgundio, but also for the brothers Hugo
Etherianus and Leo Tuscus, who in their turn translated a substantial
amount of Greek writings into Latin. These brothers, too, spent some time
in Constantinople. But, after Constantinople, the major source of Greek
works was Antioch. Burgundio translated Chrysostom's homilies on
Matthew from a manuscript that Pope Eugene III had obtained on his
behalf from Aimery, patriarch of Antioch 16 • The same patriarch
corresponded with Hugo Etherianus 17 , who presumably got his copy of the
privileges for, the Pisans in Laodicea and Antioch see documents no. 4 (1154) and no. 13
(1170).
13 To the information in the fundamental monograph by P. CLASSEN, Burgundio
von Pisa Sitzungsberichte der Heidelberger Akademie der Wissenschaften, Phil.-hist.
Klasse i974 must now be added those of G. VUILLEMIN-DIEM and M. RASHED, in
Burgu:w.io de Pise et ses manuscrits grecs d'Aristote: Laur.87.7 et Laur.81.18, in
Recherches de T~ologie et Philosophie medievales, 64, 1997, p. 136-98, and F. BOSSIER,
L'elaboration du vocabulaire philosophique chez Burgundio de Pise, in Aux origines du
lexique philosophique europeen, ed. J. HAMESSE, Louvain, 1997, p. 81-116.
14 The Emperor Alexios had given them a special quarter in October, 1111.' on the
southern shore of the Golden Horn, opposite Galata; see W. HEYWOOD, A HlstOry of
Pisa: Eleventh and Twelfth Centuries, Cambridge, 1921, p. 54.
15 L. MINIO-PALUELLO, Iacobus VeMticus Grecus: Canonist and Translator of
Aristotle, in Traditio, 8,1952, p. 265-304, reprinted in id., Opuscula: The Latin Aristotle,
Amsterdam, 1972, p. 189-228, see p. 190.
16 The relevant portion of Burgundio's preface is edited as no. 67 in R. HIE~TAND,
Papsturkunden far Kirchen im Heiligen Lande, Abhandlungen der Akademle der
Wissenschaften in Gottingen, Phil.-Hist. Kl. F. 3, Nr. 136, Gottingen, 1985, p. 201-2.
17 PL 202, cc. 231 and 232; A. DONDAlNE, Hugues Etherien et Le/m Toscan, in
Archives d'histoire doctrinale et litteraire du moyen age, 27, 1952, p.67-134. See also
K. CIGGAAR, Manuscripts as Intermediaries, the Crusader States and Literary C~oss­
Fertilization, in East and West in the Crusader States: Context-Contacts-Confrontatlons,
 IV
Syrian theologian Theodore Abu Qurra from that source. What made the
Pisan-Antioch conduit different from that of the Pisan-Constantinople
one, however, was that it also provided a route for Arabic texts, and this is
where the work of Stephen of Pisa is significant.
Stephen of Pisa made a complete translation of the kitllb al-malakl of
'Ali ibn al-'Abbas al-Majnsi, translating the title rather literally as Regalis
dispositio «( The Royal Arrangement»). The kitrlb al-malaki, also known as
the kitllb kamil a$-$inll'a at-tibblya «( the complete book of the medical
art »), composed in the mid-tenth century, was very popular in the Arabic
world. It consisted of ten books of Theorica (medical theory) and ten books
of Practica (medical practice). At the beginning or end of several of these
20 books Stephen gives dates and details of the circumstances of the
translation and its copying. A typical example may be taken from the
beginning of the eighth book of the Practica :
Here begins the eighth sermo of the second part of the
book of the complete art of medicine which is called the
Royal Arrangement of Hali son of Abbas, disciple of
Abimeher Moysi son of Seyar [i.e. Aba Mlihir Mllsa ibn
Sayyar]; the translation from Arabic into Latin of
Stephen the disciple of philosophy. He wrote the copy
himself and completed it in the year from the passion of
our Lord 1127, on Saturday, November the third, at
Antioch. Thanks be to God, the beginning and end of
things 18.
* Other explicits mention other dates within 1127, and the participation
of the scribes Alduinus and Pancus. The manuscripts are unanimous in
giving Antioch as the place of copying; an independent source describes
this Stephen as the 'nephew of the patriarch of Antioch'. The dates include
ed. K CIGGAAR, A. DAVlDS and H. TEULE, Leuven, 1996, p.131-51, where this example
and further examples of texts being translated from Greek into Latin (at the request of
Aimery), and from Latin or Old French into Arabic, are given; for the latter see also
J. NASRALLAH, Histoire du mouvement litteraire dans l'eglise melchite du Ve au XXe
siecle, I11.1, Louvain-Paris, 1983.
18 HALY FILIUS ABBAS, Liber totius medicine necessaria continens quem
sapientissimus Haly filius Abbas discipulus Abimeher Moysi filii Seiar edidit: regique
inscripsit unde et Regalis dispositionis nomen assumpsit ... Lyons, 1523, f. 261v : Incipit
sermo octavus secunde partis libri completi artis medicine qui dicitur Regalis dispositio
Hali filii Abbas discipuli Abimeher Moysi filii Seyar. Translatio Stephani phylosophie
discipuli de arabico in latinum scripsitque ipse et complevit anno a passione Domini
millesimo centesimo vicesimo .vii. mense Novembris die .iii. feria septima apud
Antiochiam. Deo gratias rerum principio et fini. For further examples see C.H. HASKINS,
Studies in the History of Mediaeval Science ... , p. 133 and Appendix III below.
6
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
the unusual formula « such-and-such a year from the Passion of our Lord»
(rather than from his birth: incarnatione) but it is not necessary to add the
thirty-three years of Chirst's life to the dates indicated, for Stephen's
translation was already known before 1140 to Northungus, a doctor
working in the Cathedral curia of Hildesheim 19.
In the prologue Stephen states that he had to make the tran~lation
because an earlier Latin version was not at all faithful to the ArabIC text,
and seriously defective. We know that this earlier version of 'All ibn al-
'Abbas's work was that of Constantine the African called the Pantegni 20•
Stephen's judgement is correct. Constantine renders the Arabic freely and,
while translating all ten books of the Theorica, translated only parts of
three books of the Practica. Stephen's literal translation conforms to the
norms laid down (albeit later21) by Burgundio of Pisa. It is significant,
however, that a first attempt to continue Constantine's translation h~d
already been made, apparently in Pisa, not long before. Stephe~ mad~ h~~
complete translation. For, in a twelfth-century manuscnpt now 10 Berhn,
half-way through the ninth book of the Practica (the book on s~gery),
there is a note stating that «Constantine the African translated thIS far ...
The rest of book nine of the Practica was translated by the doctor John - a
Saracen who had recently converted to the Christian faith - and Rusticus
of Pisa son of Bella and a physician by profession, on the expedition to
beseige' Majorca23 .,. This was the crusade of the Pisans and Catalans
19 See Appendix III below. The reference to 'Stephen the nephew of the patriarch of
Antioch' occurs in ms London, B.L., Sloane 2426, fol. 8r: .. .Albausoardi Cordubens.is qui
Pantegni composuit, quamvis Constantinus ipsum composuisse mentiatur, t~men lpsum
Constantinus et Stephanus nepos patriarche Antiocensis transt".ler~nt (~s was first
noted by C.H. TALBOT in Stephen of Antioch, Dictionary of SClenti{ic Bwgraphy, e~.
C.C. GILLISPIE, XIII, New York, 1976, pp. 38-9). For Northungus's use of S~phen s
translation see M. WACK, 'All ibn al-'Abblls al-Maglls1 on Love, in Constantme the
A{rican ... (following note), p. 161-206 ; see p. 197-9.
20 See the articles in Constantine the African and ~I ibn al· 'Abblls al-Maglls1 :the
Pantegni and Related Texts, ed. C. BURNETr and D. JACQUART, Leiden, 1994.
21 The translation-method is set out in Burgundio's prefaces to his translations of
Chrysostom's homilies on Matthew (translated 1151) and John (1173) ~ s~e CLAS~E~,
Burgundio von Pisa ... , passim and C. BURNETT, Translating fro",: Ar~blC mto Latm l~
the Middle Ages: Theory, Practice and Criticism, in Editer, tradulre, mterpreter: essalS
de methodologie philosophique, ed. S.G. LOFTS and P.W. ROSEMANN, Louvain-Ia-Neuve,
1997, p. 55-78.
22 Ms Berlin Staatsbibl. Preussischer Kulturbesitz, lat. fol. 74, described as no. 898
in V. ROSE, Verz~ichniss der lateinischen Handschriften der koniglichen Bibliothek zu
Berlin, 11.3, Berlin, 1905, p.1059-65.
23 Ibid., p. 1061 : Hue (Nunc ms) usque Constanti.nus A/fricar:us phi~soph~s ~
nobilis medicus translator fidelissimus huius none partlcule clrurgle practlcf extlterlt.
7

IV
against the Arabs in Majorca, which lasted from 1114 to 1115. The
completion of Practica, Book IX thus took place in a Pisan context, and was
probably known to Stephen before he went to Antioch24 • For we have
another testimony - that of Matheus Ferrarius- who states that
Constantine the African « translated only three books of the Practica
P~ntegni (for <the rest> were destroyed at sea). But Stephen, a certain
Pisan, went to those parts, and, learning that language, translated the
whole of the Practica, which is consequently called the Pantegnian and
Stephenian Practica ,.25. This composite Practica is in fact what we find in
the above-mentioned Berlin manuscript in which Constantine's translation
of Practica Book IX with the addition of the contribution of Rusticus and
John the Saracen is surrounded by Stephen's translation of the rest of the
Practica. Such a patchwork probably did not satisfy Stephen. For he went
on to translate also those books which Constantine had already
translated26 .
Stephen of Pisa appended to his complete translation of the k. al-
malakl a Medicaminum omnium breviarium. This is an alphabetical
catalogue of the Greek materia medica in Dioscorides with their Arabic
equivalents and a few translations and explanations in 'Latin (see Plate 7).
~is is a remarkable work from both a philological and a medical point of
VIew, and was perhaps the most complete medical glossary compiled for a
Western audience up to that date. Stephen concludes modestly:
Dehfnc in expeditione ad obsessionem Maioric'l Iohannes quidam Agarenus quondam, qui
nov,ter. ad fide": Christiane reLigion is venerat, cum Rustico Pisano, Bellf filius ac
pro(ess&~ne rrn:d&CUS .hanc nonam particulam Practice ad finem usque ad principium
dec&m'l m latmam lmguam deo adiuvante transtulerunt. The 'fili" and 'medic" in the
manuscript are probably miscopyings of 'filio' and 'medico', making Rusticus the son of
Bella and physician. There was another translator .. Rusticianus Pis anus » in the late
thirteenth c. ; see CIGGAAR, Manuscripts as Intermediaries ... , p. 147-8.
24 V. ROSE, Verzeichniss ... , p. 1064, suggests that, on the contrary, John the
Saracen already knew Stephen's translation since his reference to Constantine as
translator f&delissimus would seem to be a deliberate repost to Stephen's criticism of
Constantine's lack of fidelity. This is unlikely.
. ~5 A Practica vero Pantegni non nisi tres libros translatavit. Erat namque ab aqua
~&ss,pata ..Stephanon autem quidam Pisanus ad illas (orientales?) partes ivit et linguam
,llam add&Scen~ eam ex to~ transtuLit, quae nunc Practica Pantegni et Stephanonis dicitur
(ms Erfurt, WIssensch. Blbl., Amplon. 0 62, f. 50r); see R. CREUTZ, Die Ehrenrettung
Konstantins von Africa, in Studien und Mitteilungen zur Geschichte des Benediktiner-
Ordens und seiner Zweige, 49, 1931, p. 2&-44, see p. 41.
26 The ninth book of the Practica is entirely missing (except for the title and the
first few chapter headings) in the twelfth-century manuscript of Stephen's translation
Worcester, Cathedral Library, F.40 (see f. 108va). An early use of both the Constantini~
and the Stephenian Pantegni is in the medical compilation of Northungus ; see n. 19
above.
8 9
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
This is what we have found up to now by our own
effort in Syria concerning the interpretation of the names
of medicaments in Greek and Arabic. If posterity finds
any mistakes in it, it should not blame us. For it is not
our work but the labour of others, and we have merely
written it down in the way that we found it27.
Stephen's list became the major source of the most popular list of
equivalent names in materia medica in the Middle Ages: Simon of Genoa's
Synonyma (end of thirteenth century), which was printed in the
Renaissance and was used by the founders of modem Arabic studies in
their lexicographical work28 •
Stephen, however, does not claim to be a doctor, but rather « a disciple
of philosophy ,.29. and in his prologue to the Regalis dispositio he says that
translating works on medicine is only the first and lowliest step in
revealing the « secrets of philosophy,. found amongst the Arabs : starting
from the things on the health of the body he wishes to rise to those that
are relevant to the excellence of the soul 30 • Can other works be attributed
to him?
In a note published in 1950 Richard Hunt printed the text of a
subscription to a twelfth-century manuscript of the Rhetorica ad
Herennium in the Ambrosian Library in Milan (Cod. E. 7 sup.). It reads:
Scribsit (sic) hunc rethoricorum librum [lacuna of four
letters] scriba Stephano thesaurario Antiochie anno a
31
passione domini millesimo centesimo vicesimo prim0 .
27 V. ROSE, Verzeichniss ... , p. 1063 ; see Appendix I, A4.
28 See D. JACQUART, La coexistence du grec et de l'arabe dans le vocabulaire
medical du Latin medieval: l'effort linguistique de Simon de Genes, in Transfert de
vocabulaire dans les sciences, ed. M. GROULT, Paris, 1988, p. 277-90; reprinted in
eadem, La science medicale occidentale entre deux renaissances (XIIe s.-XVe s.),
Aldershot, 1997, article X.
29 Liber regal is (HALY FILIUS ABBAS, Liber totius medicine ... ), f. 5r: Incipit
prologus Stephani philosophie discipuli in Libro medicine qui dicitur Regalis dispositio.
The same appellation is repeated throughout the text (see n. 18 above).
30 Appendix I below, A1l17-18]. Stephen's interest in philosophy is also manifest in
the prologue to the second part of the Regalis dispositio in which he contrasts the
medicus with the physicus (natural scientist); Appendix I, A2(3-4] ; he implies that he
will discuss the principles of natural science elsewhere: see ibid., [10--11].
31 R.W. HUNT, Stephen of Antioch, in Medieval and Renaissance Studies, 6, 1950,
p. 172-3 ; see Plate 8.
IV
The scribe [name missing] wrote this book of rhetoric
for S~ephen the treasurer at Antioch in the year from the
PassIon of the Lord 1121.
. ~unt suggested that this is none other than Stephen of Pisa especiall
m VIew of t.he fact that here, as in the Regalis dispositio, V:e find th~
u~usual datmg formula: « the year from the passion of the Lord ». Hunt
dIscovered that there was a treasurer called Stephen at the Benedictine
~onastery of St Paul, one of the principal religious foundations in Antioch
Just north of the Pisan quarter, who had a house there sometime betwee~
1126 and 113032• It is tempting to identify the two Stephens.
. ~ut that is not all. I~ bo~h thi.s manuscript and in the manuscripts and
edItions . of the Re~all,s dl,sposl,tio, we find a remarkable system of
numer~tIon. ~e Latm letters in their alphabetical order have been used
:~dd~~mal dl~ts, ~n !~e analogy of the Greek alphanumerical notation
e. ArabIC abJad . Now, this unusual feature appears in another
context m the twelfth century: that is, in a cosmological work said to be
translated by « Step hen the Philosopher ~.
~he ~iber Mamonis in astronomia a Stephano philosopho translatus
su~ves. In only one manuscript: Cambrai, Bibliotheque municipale 930
wntten In the twelfth century3", Unfortunately, this is lacking its e~d, s~
that we .no. long~r have the colophon which may have included the
ch~ractens~5c .datmg formula. However, as Haskins and Hunt have both
pomted out '. It does have the peculiar feature of using the Latin alphabet
for numera~s m exac~ly the same way as in the manuscript of the Rhetorica
ad Herenmum and In the Regalis dispositio. In the Liber Mamonis the
numbers go up to 300 : the letters a to i for the units, the letters k to s for
the tens, t for 100, ~ for 200, ~nd x for 300. Moreover, the script and lay-
out of the Cambral manuscnpt is remarkably similar to that of the
32 See Appendix I C below.
33 See Appendix n.
t' C·tH. HA~KlNS, Studies in the History of Mediaeval Science ... , p xii draws
34
atte n lOn 0 a « hber Mamnonis' 1 te tw 11th . ,
Abbe '. . . . . In a a e -century catalogue of the Benedictine
! at Whlt~y.' the. tItle, IS read as .. Liber Marmionis (unidentified),. in the recent
Engltsh Benedtctme L,branes: The Shorter Catalogues,eds R. Sharpe et 1 Lo d
'~:o~' ~:7, no. 33, but th~ possibility remains that it was the same wor: The ~a:~
, up now unexplained, but certainly would have sounded Arabic in the twelfth
century, recalling the common Arabic names Ma'mnn (the name of a caliph) and
Maymnn.
35 C.H. HASKINS, Studies in the History of Mediaeval Science ... , p. 102-3 ;
R. W. HUNT, Stephen of Antioch, p. 172.
10
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
Milanese manuscript36 . The Liber Mamonis is not a translation from
Arabic. Rather, it is a Latin writer's own account of Ptolemaic astronomy.
It refers repeatedly to Latin scholarship, and especially to Macrobius. The
writer knows the Almagest which he calls by its Greek name (Sinthasis =
Syntaxis), and interprets the meaning of other Greek astronomical terms
(e.g., p{arlanselinio quod nos plenilunium dicimus, f. 34r). Nevertheless, he
states that he is following « a certain Arab ~37, and the form of the work is
similar to Arabic cosmologies (kit ab f 1 'l-hay'a). The identification of the
Latin writer with Stephen of Pisa is very likely, since he could handle
Greek. Haskins already considered his authorship a possibility, and a
comparison of the style and vocabulary of the prefaces of the Regalis
dispositio and those of the Liber Mamonis corroborates the hypothesis that
the two Stephens are the same 38 •
36 For a photograph of the Ambrosian ms see F. STEFFENS, Lateinische
Palaographie, Berlin and Leipzig, 1929, no. 83, and Plates 2 and 8 below.
37 Cambrai 930, f. 38r, Preface to Book 4 (Appendix I, B4[2]) : Verum cum in aliis
[libris] Arabem quendam plurimum sec uti sumus, in hoc quoque per multum sequemur
(This passage was cited in C.H. HAsKINS, Studies in the History of Mediaeval Science ... ,
p. 102). For details concerning the relationship between the Dresden Almagest and the
Liber Mamonis see C. BURNETT, The Transmission of Arabic Astronomy via Antioch and *
Pisa in the Second Quarter of the of the Twelfth Century, in H. HOGENDIJK - A. SABRA,
New Perspectives in Islamic Science (forthcoming).
38 The prefaces are edited in Appendix I below. They all display great erudition and
adopt a similar tone of criticism, whether towards detractors from the pursuit of wisdom
(first preface to Regalis dispositio) or towards false philosophers (first preface to Liber
Mamonis) ; both works accuse detractors of clinging too much to what they have already
learnt and not accepting new knowledge (Al[19], B2[6-l0». They both distinguish men
from brute animals by their possession of'animus' (A1[3] , B4[3] ; other authors would use
the term 'ratio' in this context). In terms of phraseology and vocabulary, the following
correspondences are notable: divina ... benignitas (A1[17]), divini muneris ... benignitate
(B1[4)) and divina ... benignitas (B3U]) ; in cunabulis philosophie (A1[12]), in philosophie
cunabulis (Bl[16)); consumere laboris .. .operam (A1[17]), consumpta opera (B1[8]);
pernecessarius (Al[161, B4[3)) ; latinitas (A1[13], A2[111, B1[9], B2[lJ[5], B4[6), referring
to Latin literary culture) ; latina facultas (Altitle, B2[1» ; inertia (A1[19], B1[19], a vice
which is singled out); discendi cupidi (A1[20)), discendi cupiditas (Bl[19)); acutius
intuens deprehendet (Al[lOJ), acutius intuentibus deprehendatur (B2[5»; prima huius
operis et non minima parte absoluta (A2[1)), iam non minima propositi operis parte
absoluta (B3[l]) ; sentire used in the sense of 'to notice' : eius ultimam et maiorem deesse
sensi partem (A1[13», cum eius traditioni contraria sentirent (B2[6]); criticism of
previous writers for having 'vitiated' (depravare) the subject (Al[131, B2[2-4]) ; rimari
studui (A1[13]), rimari studeam (B3[5]) ; the use of esse in the sense of 'be possible'
(Greek influence?) : nullum sit.. .curare (A2(8)), horum alterum ... fuisse dicendum non est
(B2[3]) ; the use of quoque as second word in the sentence to introduce a new point:
A1[9)[10), Bl[3) ; Dei gratia adjacent to a mention of 'us' : A3(1) and B3[5].
11
IV
Moreover, there is a Arabic-Latin translation of the Almagest which
seems to be the very one that Stephen the Philosopher was using or
directly involved in translating. This is the translation of the first four
books of Ptolemy's work in Dresden, Landesbibl., Db. 87, f. 1-71v (ca. 1300
A.D., formerly belonging to Berthold of Moosberg), first discussed by
Heiberg in 191139 . Here, too, the same alphanumerical notation is used,
and we find some unusual astronomical terms which recur in the Liber
Mamonis : e.g., circulus rotunditatis for « epicycle », sexagenaria for minute
(Greek E~l1Koo"'tbv), and Greek transliterations, pansilini and sinodus for
opposition (of the Moon to the Sun) and conjunction (of any two planets
including Moon and Sun) respectively, and the Greek title of the text;
«megali xintaxis ». Also, the Dresden Almagest shares with the Regalis
dispositio the characteristics of using the term « sermo » for the constituent
books, and naming the work and the translator at the end of each of the
books. But instead of a translation of «Stephen the disciple of
philosophy», this Almagest version is described as a philophonia
translatione dictaminis wittomensis (with variants) ebdelmessie 40• The
significance of this phrase is as obscure as that of the name «Liber
Mamonis» ; the first personal name could be a corruption of Wintoniensis,
i.e. « of Winchester »41, but the second is clearly the Arabic name 'Abd al.
39 J.L. HEIBERG, Noch einmal die Mittelalterliche Ptolemaios-Ubersetzung, in
Hermes, 46, 1911, p.207-16 (see p. 215-6). C.H. HAsKINS, Studies, p. 108-9 gives
further information, prints the chapter headings of the first book and the beginning of
Ptolemy's preface, and notes 'a confusing form of numerals: b = b = 2 ete.' (p. 109). I am
grateful to Menso Folkerts for the loan of a microfilm of this manuscript. The scribe of
this manuscript, unfortunately, appears to be totally ignorant of the subject-matter.
40 F. 15v : explicit primus sermo libri mathemathice Ptolomei, qui nominatur megali
xintaxis astronomie translacione dictamine philophonia wittomensis ebdelmessie ; f. 31r :
explicit secundus sermo ... astronomie translacione dictaminis wintomiensis ebdelmessie ;
f. 49v: explicit tercius sermo ... astronomie philophonia translacione dictaminis
wittomensis ebdelmessie ; f. 71r: phylophonia wuttomensis ebdelmessie. explicit quartus
sermo... astronomie translacione dictaminis. One may be tempted to emend these
colophons to translacione wintoniensis dictaminis ebdelmessie: «the translation of the
man from Winchester of the dictation (presumably in a vernacular language) of 'Abd ai-
Maslb», but it is not impossible that there should be an .. 'Abd al-Maslb of Winchester .. :
see C. BURNETT, 'Abd al·Maszl], of Winchester, in L. NAUTA - A. VANDERJAGT (eds.),
Between Demonstration and Imagination: Essays in the History of Science and
Philosophy Presented to John D. North, Leiden, 1999, p. 159-169.
41 The only known scholar from Winchester who could have lived at this time is
Henry of Winchester, who wrote a commentary on the Salernitan text, the ISGgoge of
Iohannitius ; see R. SHARPE, A Handlist of the Latin Writers of Great Britain and Ireland
before 1540, Turnhout, 1997, p. 176.
12
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
Maslh42. The close relationship between this translation of the Almagest
and the Liber Mamonis needs further exploration, but everything indicates
that it too was translated in Antioch, and is likely to be the first attempt in
the Middle Ages to translate the Almagest into Latin43 .
There is a further feature in the Liber Mamonis that was noted by
Richard Lemay, but was overlooked by Haskins and Hunt: i.e., the use of
Hindu-Arabic numerals in their oriental form, in addition to the Latin
alphanumerical notation44 . Lemay did not explore the implications of this,
but I think they are crucial.
The oriental forms of the Hindu-Arabic numerals occur rather rarely in
Latin manuscripts. The numeral forms in current use in Western scripts
nowadays, that are usually referred to as « Arabic », appear. ~ have arise~
in al-Andalus in the tenth century, as a mixture of the ongtnal Sanskrit
42 Although this means « servant of the Messiah ,. it is not exclusively a Christian
Arabic name; it was, for example, the name of the white Mamluk governor of Mosul,
serving Zang'l, the sultan of Aleppo in the mid-twelfth century.
43 Hitherto the earliest Latin translation of the Almagest has been thought to be
that made from Greek in Sicily in ca. 1160, described in C.H. HASKINS, Studies, p. 157-
61. It may be that other scientific works translated from Greek into Latin should be
assigned to the same context, such as the Metaphysica Translatio Anonyma and the
fragmentary Physica Translatio Vaticana (both translated b~ the same translator. before
the end of the twelfth century) which, according to the editor of the former,. give the
appearance of being translated by someone more familiar with Arabic ~han With ~reek
(Metaphysica, trans. William of Moerbeke, ed. G. VUILLEMIN-DI~M, Anstotel.es La~mus,
XXV, 3.1, Leiden, 1995, p. 7) ; the translatorl scribe of the Phystca Translatw Vattcana
(ms Vatican, Reg. lat. 1855, f. 88-94) uses the letters of the alphabet to numJ:>er the
books as does the scribe of the Rhetorica ad Herennium (there are no book tItles or
num~rs in the earliest manuscripts of the Metaphysica Translatio Anonyma):. see
Physica Translatio Vaticana, ed. A. MANSION, Aristoteles Latinus, VII, 2, 2nd ed., Lelden
and New York, 1990, p. 1, and especially p. 22 : ARISTOTILIS. PHISICE. ACROASEO~.
.A (supra: id est .a) EXPLICIT (the gloss is explicable by the fact th~t ,,1 .. IS
represented in the alphanumerical notation by lower-case .. a,. no~ capItal .. A .. ).
Mansion considers this version to be earlier than that of James of Venice. Stephen the
Philosopher pronounces himself as sympathetic t~ the ~eripatet~cs (Appendix I, B3[2})
and his language includes frequent allusions to Aristotelian doc~rInes : ~ee .C. BURNE1'!, *
A Note on the Origins of the Physica Vaticana and Metaphyslca medIa, In Festschrift
F. Bossier (in press).
44 R. LE MAY , De la Scolastique a
l'histoire par le truchement de. la p~ilolo~ie:
itin~raire d'un m~di~viste entre Europeet Islam, in La diffusione delle. sctenze '~lan:uch~
nel medio evo europeo Convegno internazionale dell'Accademia Nazlonale del Ll~cel,
Rome, 1987, p. 399-535, see p. 471-2. Lemay's arguments for the attribution of the Ltber
Mamonis to Hermann of Carinthia are unconvincing.
13
IV
forms and Visigothic forms 45 . They were used, albeit in very restricted
contexts, both in Latin and in Arabic. When Toledo became the main
centre for the transmission of Arabic scientific texts, it was these forms of
the numerals that were adopted, and one manuscript calls a variety of
them figure toletane, presumably because of their association with that
centre46 . The oriental forms of the Hindu-Arabic numerals, on the other
hand, are closer to the original Sanskrit forms, and developed into the
normal forms used nowadays in Arabic. They are called figure indice in the
same Latin manuscript that called the western forms figure toletane. But
in the Latin context they are used almost exclusively in an interrelated
group of astronomical texts copied in the twelfth century. These are all
Latin texts associated with the Jewish polymath Abraham ibn Ezra and a
set of tables, drawn up ca. 1145, precisely for Pisa. They consist of the
Pisan Tables themselves, and, by Abraham, a general work on
astronomical tables written before 1154, instructions for the construction
and use of the astrolabe, and two sets of rules for the use of the Pisan
Tables. An early manuscript that includes a summary of the instructions
for the tables adapted for the meridian of Lucca and refers to the year
« 1160 », also uses the oriental forms. Moreover, this manuscript also
contains an astronomical table with the Latin alphanumerical notation
and a key to its use 47 .
Ibn Ezra was born in Tudela in the Muslim kingdom of Saragossa
between 1089 and 1092 and spent the earlier part of his life in Northern
Spain. But in the early 1140s he started to travel to other parts of
Christian Europe, and the first area in which he is attested in Hebrew
sources is in Tuscany: precisely, in Lucca, where there was an important
Jewish community. Whether Ibn Ezra introduced the oriental forms of the
Hindu-Arabic numerals into Pisa is questionable. At the beginning of his
Hebrew work on arithmetic, Sefer ha-Mispar, he mentions Hindu-Arabic
numerals and substitutes for them the letters of the Hebrew alphabet. Of
the several manuscripts of this text that I have seen, only one gives the
Hindu-Arabic numerals in their oriental form, as an alternative to the
45 R. LEMAY, The Hispanic Origin of Our Present Numeral Forms, in Viator, 8,
1977, p. 435-62.
46 The term figure toletane is used to describe these numberals in ms Munich,
Bayer. Staatsbibl., elm 18927, f. 1r; see Figure la in R. LEMAY, The Hispanic Origin ...
47 Ms London, B.L., Harley 5402. More details concerning these manuscripts and
illustrations of their numeral forms are given in C. BURNETT, The Introduction of Arabic
Learning into England, London, 1997, p.48-53 and Fig. 23-5. For the astronomical
table, see C. BURNETI', The Transmission of Arabic Astronomy ... (n. 37 above).
14
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
western forms which Ibn Ezra could have been familiar with in Spain48. It
is only in the Latin versions of Ibn Ezra's works that the figure indice are
preferred, and this preference must have been due to Ibn Ezra's Latin
collaborator, who was presumably working in Pisa49 . This collaborator is
anonymous. But it is significant that, at the beginning of the Liber
Mamonis, «Stephen the philosopher» says that «we have fulfilled our
promise, having written a treatise on the rules which we had proposed for
the canon of astronomy (i.e. astronomical tables} ... »50. «Stephen's» tables
have not been satisfactorily identified, and it is worth considering whether
either of the sets of instructions for the Pisan tables betrays signs of his
hand51 . In any case, Ibn Ezra's collaborator worked in a context in which
the same Arabic numerals were used as those that Stephen used in the
Liber Mamonis. That context is likely to have been Pisa, and the Arabic
numerals are likely to have come from Antioch.
Once we establish that the Antioch-Pisa connection was a conduit
through which Arabic works arrived in Europe, then the possibility arises
that other works which have not up to now been located in a translating
centre, may have entered in this way. These include a work on weather
forecasting according to the Sapientes Indiae of which the earliest
fragment occurs in the manuscript from the McClean collection of the
Fitzwilliam Museum that contains one set of the rules for the Pisan
tables 52 ; an independent translation of the same work was made in Spain,
48 Two sets of numeral forms are given in the margin of f. 1v of Paris, B.N.F.,
hebreu 1052; in the text of this manuscript, however, and in the copies of Sefer ha-
Mispar in mss Paris, B.N.F., hebreu 1049, 1050 and 1051, only the western forms are
used. I am grateful to Tzvi Langermann for alerting my attention to these manuscripts.
.9 In the case of the astrolabe text we have the explicit statement that a Latin
writer was «following the dictation of Abraham .. : ut ait philosophorum sibi
contemporaneorum Habraham magister noster egregius quo dictante et hanc
dispositionem astrolabii conscripsimus ; see ms B.L., Cotton Vespasian A 11, f. 40v.
50 Ms Cambrai, 930, f. 2r: Quoniam in canonem astronomif quas proposueramus
regularum exsequ<u>to tractatu promissum exsolvimus ... See Appendix I, Bl[l).
51 Lemay has suggested that these rules are those in Florence, Bib!. Naz., Con.
Soppr. J.l1.10, f. 235ra-239va (Regulae in Canonem Astronomiae) ; see R. LEMAY, De la
Scolastique... p.472. The similarities, however, are slight, and the astronomical
terminology is completely different. One must note that none of the Ibn Ezra versions
use Stephen's distinctive term for «minutes .. : sexagenarie. In the astrolabe text, at
least, the usual term minuta is found; Rudolph of Bruges, who probably knew Abraham
in Beziers later in the 1140s, uses the termpunctumlpunctus.
52 Ms Cambridge, Fitzwilliam McClean 165, f. 47v (Incipit: Dixit almgsius:
Sapientes Indif secundum Lunam de pluviis iudicant ... ) ; an incomplete edition of this
text, from Paris, B.N.F., lat. 7316 (not 7326), is included in G. LIBRI, Histoire des sciences
1TUJthe1TUJtiques en Italie, Paris, 1838, I, p. 372-6.
15
IV
which strongly suggests that one should look to another place for the
translation of this text53 • Another work on weather forecasting is
attributed to John of Seville, but includes a section on the degrees of the
zodiac associated with different cities, in which the only cities to be
mentioned are Jerusalem, Rome, Pisa, Lucca, Palermo and 'Affrica', and
Pisa is signalled out as the place where the data has been corrected by
direct experience54 •
The importance of Pisa, then, as a source for both Greek and Arabic
material is considerable. For the latter, as I have shown, we already have
the example in 111411115 of John the Saracen working with Rusticus
Pisanus on completing book nine of Constantine's Practica Pantegni.
Before the middle of the century, the scientific works of Abraham ibn Ezra,
Stephen of Pisa and their collaborators, were known, if not composed
there. The arrival of Arabic texts and even Arabic-speaking scholars in
Pisa must be added to the other Arabic elements that have been observed
in medieval Pisa : the Islamic geometric motifs in the architecture of Pisa
(as well as of Lucca), the Arabic names of Pisa's most populous quarter -
Kinsica - and two of her gates, the Porta Samuel and the Porta Agazir,
and the apparently Arabic names of some of her families 55 • Moreover, both
Antioch and Pisa continued to be important centres in the thirteenth
century.
First, there is the well-known mathematician Leonardo of Pisa, known
* as Fibonacci (ca. 1170-after 1240). If I am right about Pisa being the place
where the figure indice first became used in Europe, it is a nice coincidence
that the greatest mathematician of the Latin Middle Ages, who adopted
wholeheartedly the Hindu-Arabic numerals and the Saracen way of
calculation, comes from Pisa. He learnt his trade in the Pisan quarter of
Bougie in present-day Algeria, rather than in the Crusader States, but he
visited Syria, and it is appropriate to see him as a successor to John the
53 The translation made in Spain is the Liber imbrium (incipit: Superioris
discipline ... ) in Erfurt, Wissensch. Bibl., Amplon. Q 365 (twelfth cent.), f. 44r-50r and
elsewhere, translated by Hugo of Santalla (Tarazona, mid-twelfth century).
54 Ms Paris, B.N.F., lat. 7316A, f. 46r (within Tractatus pluviarum et aeris
mutationis secundum magistrum Iohannem Yspalensem, incipit: Volens aeris scire
naturam ... ) : Astrologi dixerunt quod Ierusalem est Cancri " dicunt alii quod Leonis 9
gradibus.. secundum veritatem autem Aquarius est eius signum, Leo signum Rome.
Dixerunt Pise signum esse Piscem.. experimento autem 2 gradus Aquarii. Lucce vero
Cancrum, signum Palermi primum gradum Leonis, signum Affrice quartus gradus
Leonis.
55 D. HERLIHY, Pisa in the Early Renaissance: A Study of Urban Growth, New
Haven, 1958, p. 33.
16
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
Saracen, Stephen of Antioch and Abraham ibn Ezra. One of his friends was
Theodore of Antioch, the emperor Frederick I1's philosopher, to whom
Leonardo addresses letters at the end of two of his works. It would take too
long to discuss the role of Theodore, the Jacobite Christian from Antioch,
who was educated in Mosul and Baghdad, and served the Seljuk ruler of
Konya, the Armenian regent, and the Holy Roman Emperor. Suffice to say
that one of the mathematical problems that Leonardo dealt with, on
Theodore's request, was the subject of a text by Theodore's teacher in
Mosul, Kamal-ad-Din ibn Ytlnis. Moreover, Theodore gives one of the
earliest testimonies in the West to the knowledge of the full version of
Aristotle's Secret of Secrets, translated by Philip of Tripoli, and it is
significant that Philip of Tripoli found the Arabic manuscript of the text in
Antioch56 •
I hope I have shown that, contrary to the impression one gets reading
Haskins, the level of intellectual exchange between Arabic and Latin
culture in Antioch was high. Stephen of Antioch was undoubtedly the most
impressive of the figures involved in translations from the Arabic there in
the twelfth century, and he was quite unusual among the translators of
the Middle Ages in being able to handle Greek as well as Arabic .. It is
unlikely, however, that he was an isolated case. He worked in a tradition,
which had its roots, on the one hand, in the Arabic-Latin translations
56 On Theodore, see C. BURNE'IT, Magister Theodore, Frederick ]]'s Philosopher, in
Federico ]] e le nuove culture, Atti del XXXI Convegno storico internazionale, Todi, 9-12
ottobre 1994, Spoleto, 1995, p. 225-85. This article can be corrected and completed by the
article ofB.Z. KEDAR and E. KOHLBERG, The Intercultural Career of Theodore of Antioch,
in The Mediterranean Historical Review, 10, 1995 (= Intercultural Contacts in the
Medieval Mediterranean, ed. B. Arbel, London, 1996), p.164-76. Kedar and Kohlberg
give a translation of the key document of Bar Hebraeus that differs in small details from
that provided by me (that Theodore returned to Antioch but did not remain there long
.. because he felt there were deficiencies in his knowledge .. ; that the Sultan of Konya
« considered him to be an eccentric character and so did not make him welcome .. ; and
that Frederick Il .. assigned to Theodore as iqfll' [land granted by tenure] the town
(madlna) of Kamllhll [unidentifud1 .. ), and add that a Basel manuscript presents
Theodore as the translator of Aristotle's De animalibus CAristoteles Latinus, Il, p. 803, no.
1121), that Petrus Hispanus mentions Theodore, «the emperor's physician,., as his
master (Et quia diffinitio est speculum rei, quia declarat esse rei per substantiam, de
morbis tamquam de actionibis quae <sunt> contra naturam, veram diffinitionem dare non
possum ut medicus sensitivus <sed tantum descriptionem et denominationem>, sed
magister meus Theod<or>us medicus imperatoris consentit omnes diffinitiones artis
medicine esse veraces quia quicquid cadit sub sensum verum est quia veritas ipsa est, i.e.
secundum practicam : from Die Ophthalmologie (liber de oculo) des Petrus Hispanus, ed.
and trans. A.M. BERGER, Munich, 1899, p. 4-5), and that Theodore is unlikely to be the
same as the man whom Bar Hebraeus mentions as the teacher of Ya'qQb b. SiqIllb (or
Saqllln) in Jerusalem. See also the articles by Hasse and Williams in this volume.
17
IV IV

ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

made in Salerno, Montecassino and Pisa, and, on the other, in the Stephen of Pisa speaks of the arabica veritas within which one can find
translations of Adelard of Bath, who may have been a colleague of his in nourishment not only for the body, but also for the SOUI 61 .
Antioch. Moreover, Stephen worked with other people; not only his scribes
Alduinus and Pancus, and the mysterious « 'Abd al-Masth of Winchester »,
but also the Greek and Arabic informants who provided him with the
meanings of the materia medica in Dioscorides ; and he expects his readers
also to be able to consult Arabic texts and experts 57 • I believe that more
evidence of cultural exchange in the city will emerge, if we are prepared to
find it58 •
The situation in Antioch was indeed different from that in the Toledo of
Gerard of Cremona and Dominicus Gundissalinus, or in the Sicily of
Frederick 11. One difference was that the Jewish element, which played a
crucial role in these other centres, was not so prominent. But the richly
cosmopolitan and mainly Christian society in Antioch generated its own
brand of intellectual exchange which involved three languages - Greek,
Arabic and Latin. Perhaps the closest parallel is the culture of Roger II's
Palenno in the mid-twelfth century, where George of Antioch was the
« emir of emirs» or chief minister59 • There was certainly a continuity
between the Sicilian and South Italian culture and that of Antioch. But
while Roger's Sicily provided the Latin West principally with translations
of Greek philosophical and scientific works and Arabic sources were
regarded as a « second best », Antioch provided religious texts, and the
Arabic sources received the same approbation as those of the Greek
Fathers. For, in language reminiscent of the philological Bible study that
spurred twelfth-century scholars to go back to the Greek and the HebrewS<>,

57 For consultation of Arabic texts see Stephen's prologue to Regalis dispositio

(Appendix I below, Al[19]) : Rogamus ... ut si quid in his errasse illis videbimur, consulta
Arabum veritate, si poterint, nos arguant : for experts, see his prologue to the second part
of Regalis dispositio (Appendix I below, A2[12]) : In totius operis {iTU! omnium que apud
Diascoridem sunt medicaminum breviarium subdidimus, hinc eorum nomina grece, illinc
arabice habens, ut in cuius veTU!rit hoc opus manus, quod queque res sit aut grecum si
inveTU!rit aut certe arabem sit illi posse consulere.
58 Raymond of Marseilles' reference to the 'Caldei' as sources for astronomical texts,
alongside the 'Arabes', is presumably to Syrians: Arabes et Caldei, nobis Christianis
mirabiliter invidentes, quotienscumque aliquid sese pollicentur daturos, aut veris
quaedam {alsa interserunt, aut omnino {alsa tribuunt (Liber cursuum, written 1141,
Paris, B.N.F., lat. 14704, fol. lllv).
59 See C.H. MASKINS, Studies in the History o{ Mediaeval Science ... , p. 155-93.
It is worth noting here that La Fazienda de ultra mar, a guide to the Holy Land,
60
used the original Hebrew text of the Bible. According to its preface, this work was
composed by archdeacon archdeacon Aimery of Antioch and addressed to Raymond,
archbishop of Toledo 0125-52), thus providing an intriguing link between Antioch and
Toledo at a time when the Arabic-Latin translating enterprise in Toledo was just
beginning. This lead Heistand also to consider that Stephen's activity was not an isolated
case; see ALMERICH, La {azienda de ultra mar, ed. M. LAZAR, Salamanca, 1965, and
R. HIESTAND, Un centre intellectuel en Syrie du Nord? ... , see p. 35. However, the
twelfth-century date of this text has now been brought into question: see B. KEDAR,
Sabre la geTU!sis de la Fazienda de illtra Mar, in ArnJles de historia antiqua y medieval,
t. 28, 1995,p. 131-6.
61 See first passage in n. 51 above; see also Stephen's comment that a chapter on
.. critical days .. (the Moon's influence over the progress of illnesses) in the Pantegni is not
in the arabica veritas, Regalis dispositio (HALY FIUUS ABBAS, Liber totius medicine ... ),
f. 134v : Invenimus autem in latino capitulum quoddam quod arabica non habet veritas,'
post capitulum octavum sermonis decimi qui intueri velle {atigatur a veteri requirat
translatione, quoniam nos eo rum tantum que in arabica erant veritate translatores sumus.

18 19
IV
1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
Appendix I
The Prefaces of Stephanus Philosophus62
Editorial Conventions :
a.c. before correction
p.c. after correction
i.m. in the margin
< > editorial addition
[ ] editorial deletion
Other emendations are signaled by italics. Modern punctuation has
b~en supplied to aid the comprehension of the text, but any sharp
dIvergences from the manuscript punctuation are noted. In the English
translations angle brackets < > enclose words which have been added for
clarity, but only in those cases where there is the possibility of ambiguity,
or where it is suspected that a word or phrase is missing in the Latin.
Parenthetic explanations are placed in square brackets [ ].
A. The prefaces to the two Parts of the Regalia diBpoaitio
Note that two twelfth-century manuscripts appear to have been
produced in the same scriptorium and cover respectively the first and
second Parts of the Regalis dispositio : Leipzig, Univ.bibl., 1131 (= L)63,
62 I am most grateful to Lucy McGuinness and Luc Deitz for suggestions concerning
the Latin edition and the English translation of these prefaces. For further examples of
Stephen's translation of the Regalis disposino see D. JAQUART, Le sens donne par
Constantin l'A{ricain a son C1!uvre: les chapitres introductifs en arabe et en latin, in
Constantine the African and ~1 ibn al-'Abblls al-MakUsI(n. 20 bove), p. 71-89 (83-5, 87-
8).
63 This manuscript was wrongly described by C.H. Haskins (Studies, p.132) as
containing the second Part of the Regalis dispositio, and the date that it contains was
wrongly transcribed by Amdt and Tangl (Schrifttafeln, fourth edition, no. 23) as« 1179 ...
The manuscript comes from the Cistercian monastery of Altzelle. It lacks the last one
and half chapters of the last book of Part I, breaking off in the middle of a folio in Book
10, ch. 10. I am grateful to Bernd Michael of the Staatsbibliothek for details concerning
the Berlin manuscript.
20
IV
and Berlin, Staatsbibl. Preussischer Kulturbesitz, lat. fo1. 74 (= B). Both
manuscripts have been written by several hands, but they have the same
kind of illumination: a foliated initial at the beginning of nearly every
book, and a habit of reversing rubricated initials or turning them through
90°. Their illumination suggests that they were written in the South of
France, and are therefore possibly from Montpellier. Moreover, it is clear
that the archetype of their combined text is the direct ancestor of the text
in InS V (see below) or its archetype. The Berlin manuscript (as noted
above, p. 7) gives Constantine the African's text of Practica, Book IX
(edited from this manuscript by J. Pagel in Archiv fur klinische Chirurgie,
81, 1906, pp. 735-86); the Leipzig manuscript gives extensive parallel
passages from Constantine's Pantegni in the chapters on urines, Theorica,
Book VII, chapters 12-14. These two manuscripts tend to give a reliable
text, but, for the second part, we have the additional twelfth-century
testimony of ms Worcester, Cathedral Library, FAO (= F) which could be *
older than the Berlin manuscript (it provides the e-caudata more
frequently and consistently than any other manuscript), and is very
carefully and elegantly written in a typical Anglo-Norman hand: according
to Roderick Thomson and Michael Gullick, who are cataloguing the
manuscripts of the cathedral, it is a local product from the mid-twelfth
century.
1. Aside from L, there appear to be only four further testimonies to the
first part of the Regalis dispositio. Two manuscripts :
V Vatican, BAV, Urb. lat. 234, s. xv
W Vatican, BAV, Vat.lat. 2429, s. xv
*
And two printed editions:
ve No printed title. On the first verso begins the index of contents
(lncipit tabula omnium librorum halyabatis ... ). Printed by Bemardinus
Ri~ii (Bernardino Rizzo) of Novara, at the expense of Ioannes Dominicus de
Nigra. A fifteenth or sixteenth-century reader has annotated the British
Library copy of this edition (IC.22656) =vel.
ly HALY FILIUS ABBAS, Liber totius medicine necessaria continens
quem sapientissimus HaZy filius Abbas discipulus Abirneher Moysi filii
Seiar edidit: regique inscripsit unde et Regalis dispositionis nomen
assumpsit ... Printed by Iacobus Myt, Lyons, 1523.
The earlier edition is said to have been prepared by Antonius Vitalis
Pyrranensis (f. 186v: 'Antonii Vitalis pyrranensis opera
emend<at>issimus redditus est' ) ; the later edition is said to have been
purged of innumerable faults by Michael de Capella (sig. +lv = Michael's
21
IV IV

1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

dedicatory note): quot quantaque vitia (quibus scatebat) inter legendum vit~ suppeditantur74 per eius provided for human life through its
castigaverim, optimi cuiusque iudicio relinquo. However, it is clear that the administrationem utilitates, cum administration, since he who is
second printing is simply a copy of the first, with some mistakes added. It is qui sapientissimus iudicatus judged most wise encourages us not
is reasonable to assume, therefore, that the manuscript source of the est non solum eius only to follow its administration, but
printed edition was in Venice. administrationem sequendam, also to learn it throughout our whole
The Vatican manuscripts are not necessarily older than the Venice verum ipsam 75 toto quoque vi~ life.
edition. tempore 76 discendam hortetur77.
[3] Animus quippe hominis, quo For the mind of man, by which alone
solo c~teris homo animantibus man excels all other animals, if it is
1) In nomine summi Dei qui, cum In the name of the highest God Who, prestantior est, si non 78 not formed by laws, habits and rules
trinus64 sit personis, unus est although three in persons, is one in huiusmodi fuerit legibus 79, of this kind, seems in this way to
essentia, a quo et ad quem essence, and from Whom and to moribus institutisque instructus, differ only a little from the other
omnia, incipit prologus Stephani Whom are all things : here begins the sic parvam tantum 80 a c~teris brute animals, as Boethius says:
philosophie65 discipuli in libro prologue of Stephen the disciple of brutis differentiam habere « Those who dedicate themselves to
medicine qui dicitur regalis 66 , philosophy to the book of medicine videtur, ut apud Boetium est, eos seizing, to gluttony and self-
quem ab67 arabico in latinam which is called« Royal », which he qui rapinis, viscerationibus81 indulgence, and to idleness, live as
transtulit facultatem. translated68 from Arabic into the luxurieque ac desidie dediti sunt, the lion, the pig and the ass 82 . »
resources of Latin. leonem, suem, asinumque vivere.
[1] «Usque ad canos discendam « Wisdom is to be learnt until one [4) Quoniam igitur h~c clara, illa Since, then, some things are
esse sapientiam » preceptum goes grey »72 is a precept of Solomon, autem absurda83 , et ascendere 84 praiseworthy, others are silly, and to
salomonicum69 est, quem pre whom we acknowledge, on the quidem pulcrum85 , descensus ascend is beautiful, but descent is
ceteris hominibus in arce 70 authority of the Holy Writ, to have autem peior statu est, inniti86 worse than staying in the same place,
sapienti~ enituisse divine pagine shone forth beyond all other men on sapienti~ vestimentorum illis it is a most precious thing to support
testimonio 71 accipimus. the pinnacle of wisdom. gradibus preciosissimum est, qui oneself with those steps of the clothes
[2J Non ergo parvum quid Wisdom, then, is no small matter, nor
sapientia est, parv~ve73 human~ are the advantages small that are 74 supplentur V Lm.
75 Wadds'in'
76 tempore vite W ue ly
77 ortetur W p.c.
64 trinis V
78 vero VWue ly
65 phylosophye W
79 legibus) VW ue ly omit
66 regalis] W ue 1y add 'dispositio'
80 sic parvam tantum) sit, licet parvam tamen LV, sic licet parvam tamen W, sic
67 ex ue iy
parvam non ue Iy
68 It may be more accurate to translate this word more literally as 'transferred', 81 viscantionibus W ue ly
especially since the same verb is used in the phrase for transmitting knowledge in the
preface to the second Part ('scientiam non transferre', A2[ll)). 82 Stephen is paraphrasing the section in Boethius's Consolatio Philosophiae which
describes how men who behave badly lose their human nature and become like the
69 salamonicum ue ly animal whose nature epitomizes their sin: Bk. IV, pr. 3.
70 arte Vue Iy
83 absurda autem V
71 testimonium LV 84 descendere VW
72 Ecclesiasticus 6, 18: usque ad canos inuenies sapientiam. 85 pulchre W
73 parv~vel parve ut VW a.c., nec parve W p.c., non parve ue ly 86 inniti] innuit VW

22 23
IV IV

1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

Yet it is not the same in bodies as it

apud Boetium insigniti dicuntur. of wisdom, which are called [7] Non enim sicut in corporibus,
«distinguished,. by Boethius87 • ita et in anima. Corpora quippe, is in the soul. For bodies, as they are,
[5] Quos si quis renuat, ad If anyone rejects them, he necessarily ut sunt, circa qualitates statum have a « standing still» of qualities if
oppositam illi descendat necesse descends to the ignorance and habent, si ipsa eademque they remain unchanged (l refer to the
est ignorantiam et stulticiam. stupidity which is the opposite of perseverent--perseverantiam absence of change of those things
Habet autem hoc etiam88 gradus, wisdom. For this also has steps, but dico circa ea qu~ de his secundum which can be predicated of them
set precipites quique deiiciant89, they are all steep; they throw down substantiam et quantitatem according to substance and quantity;
non sustollant90 • Nam status 91 and do not provide support; for habent predicari ; omne etenim for every body has both existence and
quidem inter h~c92 ratione staying in one place among these corpus et esse habet et quantum. dimension). These <bodies> then,
utentibus habet impossibilitatem. things is impossible for those using H~c igitur quoniam et voluntate since they lack both will and the
reason. carent et voluntatis operatione, operation of will, while they remain
[6] Sicut enim qu~ est in animatis For, just as the augmentative power cum eadem circa ea94 qu~95 dicta unchanged in respect to what has
corporibus virtus augmentativa in animate bodies has to increase the sunt perseverent, circa qualitates been mentioned, have a .c standing
usque ad prefinitum tempus size of bodies up to a predetermined statum habent nisi alterius vi still» of qualities unless they are
habet augere corpora, statu time and does not stay sti1l 93 , so too commutentur96. moved by the force of something else.
carens, sic et hominis ratio ad the reason of man does not stay still [8] Anima autem non sic97 . Cum But the soul is not thus. For, since
predicta statu caret. in respect to what has been enim et vis volendi et98 vis insit there is also in it a power of willing
mentioned. illi operandi qu~ sunt99 in and a power of operating certain
anima 100 ad sapientiam instruments in the soul for wisdom,
instrumenta quedam lOl , et« velle and since wisdom is cc to will to be
87 Stephen is alluding to the ladder-like steps from n (practice) to e (theory) on the just ,., if this will ceases, it is
quidem esse iustum ,.102 sapientia
clothes of the goddess Philosophia, which Boethiua describes in his vision at the
beginning of the Consolatio Philosophiae, Book I, pr. 1, line 20. sit, si hoc velle cesset, stultitia stupidity.
88 autem hoc etiam] autem etiam hoc W, etiam hoc ue, etiam hic ly est.
[9] Operari quoque iusticiam To do justice also belongs to a wise
89 deiciant VW man, to cease from doing it, to a
sapientis est, cess are stulti ;
90 substollant VWue cessare enim a bene velle primus stupid man ; for to cease from a good
91 V adds p.c., hoc ue est stulticie precepsque 103 gradus. will is the first and precipitate step
92 hoc W, hic ue ly Nullus igitur nobis inter to stupidity. There can be for us,
93 The augmentative power of the virtus naturalis is decribed as the 'nutriens sapientiam stulticiamque potest then, no standing still between
virtua', in Regalis dispositio, Theorica, Book IV, chapters 1 and 2 (ed. L~ons, 1523, f. 43r-
44r): cf. IV.2 , f. 44r: Nutriens virtus... servit... generanti augmentando et crescere
faciendo fetus membra in quantitate, extendit enim ea in longum, latum et altum,
huiusque actio uirtutis est a principio generationis fetus usque dum finis occurrat 94 circa ea) VW ue ly omit
iuuentutis, que est trigintaquinque annorum, et iam tunc ab actionibus suis defreit. The 95 ea qU{l] eademque W
term augmentativa uirtus itself, however, appears to be an insertion into 'All ibn aI-
96 vi commutentur) violentia mutentur V p.c.
'Abbas's text, made originally by Constantine in his translation of the Pantegni, Theorica,
Book IV, chapter 2 (hec in tres dividitur uirtutes: generativam virtutem, nutritiuam, 97 non sic) ue 1 adds i.m. 'Hominibus anime status non est sicut corporis'
pascitiuam. Virtus augmentatiua que connominatur virtus nutritiua generatiue ministrat)
and taken over as a gloss in Stephen's translation of the same passage : una est virtus
98 vis volendi et) VW ue ly omit
generans, secunda est nutriens, tertia pascens... Bec etenim ougmenta nutrientis sunt 99 est W
uirtutis et augmentative (this sentence has no equivalent in Constantine; the last two 100 animal W ve ly add 'et'
words have no equivalent in the Arabic original) _. .Nutriens virtus vera scilicet que
crescere dat generanti seruit uirtuti. Constantine's vocabulary, in turn, was influenced by 101 quedam instrumenta V
that of Alfano's translation of Nemesius's Premnon phis icon ; see chapter 26, ed. 102 istud V a.c.
C. BURKHARD, Leipzig, 1917, p. 107: nutribilis et ougmentativo et generativa, quae sunt
naturales {virtutes}. 103 precesque W

24 25
IV

esse status. wisdom and stupidity.

,
.i
!
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

[13] Impegi itaque in quendam Thus I fixed upon a certain book *

IV

[10] In ipsa quoque sapientia In wisdom itself also he who looks qui apud arabes 116 medicine which among the Arabs is called the
statum non esse acutius intuens more intently will understand that complementum dicitur et regalis « completion of medicine» and

deprehendet ; nam velle quidem there is no standing still; for wisdom librum. De quo si quid haberet « royal ». On investigating whether

esse in omni re secundum is to will to be wise in everything latinitas requirens 117 , eius Latinity had any of this book, I found
hominis posse sapientem according to human possibility, thus ultimam et maiorem deesse sensi that it lacked the latter and greater
sapientia est, sic et discere. also to learn. partem, alteram 118 vero part of it, but that the other part had
[11] Cui rei, quoniam hominis Since the lifetime of man is not interpretis calida depravatam 119 been vitiated through the hot-headed *
vita l04 non sufficit, dum de die in sufficient for this, while he learns fraude l20 . deceit of the interpreter.
diem discitur - cessare enim 105 ab from day to day - for to cease from [14] Nomen etenim auctoris 121 For he had cut out the name of the
hoc cum possis stultitia est - non this when you still have the ability is titulumque subtraxerat, seque author and the title, and put himself
status in ea set provectus l06 stupidity -then not standing still but qui interpres extiterat et as creator of the book-he who had
invenitur. rather forward movement is found in inventorem libri 122 posuit, et suo merely been the interpreter - and
wisdom. nomine titulavit. QU~123 ut had entitled <the book> with his own
facilius posset, et in libri prologo name. To do this more easily, he
[12] Ipse igitur salomonicum l07 Therefore, following the command of et in aliis multa pretermisit missed out many necessary things
sequens mandatum, non solum Solomon, I strove for the sake of pluribus necessaria locis, both in the prologue to the book and
latinam, verum l08 arabicam wisdom to search through not only multorumque ordines in many other places, and, changing
quoque linguam sapienti~ gratia the Latin but also the Arabic commutans, nonnulla aliter the orders of many things, he put
rimari studui, ut quanto language, so that the more I should protulitl24, hoc uno tantum forward some things in the wrong
diversarum scientiam linguarum have the knowledge of different observato : nichil prorsus ex suis way, observing this one thing alone:
nossem 109, tanto id quod in languages, the more expressly I addidit. In quo manifeste nobis that he added nothing at all of his
cunabulis 110 philosophi~ll1 should understand the substance, the innuit ipsum interpretem pocius own. In this he showed himself
rudis 112 aliquando didiceram, dimensions and the quality of what I quam scriptorem fuisse. clearly to us to have been the
quid 113 esset, quantum had once learnt as a mere beginner in interpreter rather than the writerl25.
qualeque 114 , expressius the cradle of philosophy115.
in telligerem.
115 Note that Stephen mentions here the first three categories in Aristotle's
Categories.
116 ad V
104 vita hominis W ve 1y
117 nota versionem (?) Constantin(i) V Lm.
105 est ve ly
118 ve Iy place a paragraph mark before 'alteram'
106 proiectus VW ve ly
119 depravata V
107 salamonicum ve ly
120 fraude1 vel adds 'malitiam interpretis'
108 V adds 'etiam'
121 enim et actoris V
109 noscere ve Ly
122 liberi ve Iy
110 in incunabulis V, ab incunabulis ve Iy
123 Quem W
111 phylosophye W
124pertulit ve Iy
112 nudis VW ve Iy
125 The implications of this passage for what was understood by .. translator,. in the
113 quod Wve Iy Middle Ages are explored in C. BURNETI, Translating from Arabic into Latin in the
114 qualemque VW Middle Ages: Theory, Practice, and Criticism, in S.G. LoFTS - P.W. ROSEMANN (eds.),
1
26 27

I
IV
[15] Quare magis arguendus est
qui, cum alterius librum de
lingua in linguam transferret,
confidentia seu impudentia
nescio qua 126 et illi quod 127
elaboraverat abrogare et sibi
usurpare non erubuit.
[16] Que quoniam nobis minime
placent, liberque iste
pernecessarius hominis est vite
scientiaque conspicuus 128,
aggredimur et ipsi set alia via
librum hunc transferre : auctori
quod suum est, nobis quod
interpretis, ascribentes, ut et
qu~129 desunt in latino
suppleantur, qu~130 transposita,
suis ordinibus <reddantur>131,
qu~ aliter prolata, prout sunt in
arabico transferantur.
[17] His igitur in libris nostri
primum consumere laboris
proposuimus operam, tametsi
alia his prec1ariora lingua habeat
apud se arabica recondita : omnia
scilicet philosophie 132 arcana 133,
quibus deinceps, si divina
Editer, traduire, interpreter: essais de methodologie philosophique, Louvain-la-Neuve,
1997, p. 55-78 (see p. 62-3).
126 quam W
127 quidem V
128 perspicuus L, V expunges, but writes the word in the margin
129 quia W
130 ve ly adds 'autem'
131 ve ly adds 'et'
132 pbylosopbye W
1
133 arcana] ve adds 'in lingua arabica arcana'
Therefore, he should rather be
condemned who, when he translates
the book of another author from one
language into another, by some over-
confidence or impudence has not
blushed either to take away from the
author what he had laboured over, or
to usurp <that material> for himself.
Since this procedure does not please
us at all, and that book is very
necessary for the life of man, and is
outstanding for its learning, we also
approach the task of translating this
book, but following another method:
ascribing to the author what is his,
and to ourselves what belongs to the
interpreter, so that what is lacking in
Latin is supplied, what was placed in
the wrong position is restored to its
proper position, and what was put
forward wrongly is translated as it is
in Arabic.
We have, then, proposed to devote the
effort of our labour first to these
books, although the Arabic language
has, hidden within it, other things
more noble than these: namely, all
the secrets of philosophy, to the
translating of which, afterwards, if
28
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
dederitl 34 benignitas, exercitatum
dabimus transferendis ingenium.
[18] Leviora enim hec 135
preferimus, ut ad difficilia via
nobis sit, et que 136 corporibus
necessaria sunt tempore
preponimus, ut his 137 sanitate
proposita arte medicin~, qu~ ad
animi attinent 138 excellentiam
longe altiora subsequantur.
[19] Rogamus itaque
modemorum improbam et ad ea
qu~ semel didiceruntl 39
defendendum paratam
pertinaciam140 , ut si quid 141 in
his 142 errasse illis videbimur,
consulta 143 arabum veritate, si
poterint, nos arguant, et sancta 144
ea, [qu~ semel didicerunt] aut
propter invidiam quod
vitiosissimum est aut propter
inertiam 145 discendi quod
contemptissimum 146
134 desit V
135 boc W
136 sit, et que] tantum quoque LV, tantum queque LW
137 hiia ve ly
138 attinet V
139 didicerint V a.c.
140 pertinatiam L, pertinentiam VW
141 quia V
142 biis ve ly
143 consulata L
144 scientia V
145 inhertiam ve lyW a.c.
146 contentissimum VW
IV
divine kindness permits, we will
devote our skill <once it has become>
practised.
For we have put these easier subjects
first, so that there is a path for us to
the difficult subjects, and we provide
first what is necessary for bodies, so
that, when healing has been provided
for these by the art of medicine, what
belongs to the excellence of the mind,
being much more lofty, should follow.
Thus in the face of the wicked and
well-prepared tenacity of the
Modems to defend what they have
once learnt, I beg that if we shall
seem to have made mistakes in these
things, they should prove us wrong, if
they can, by consulting the truth of
the Arabs, and when that has been
established, we demand that they
should at least cease from criticizing
us, either because of envy, which is
most vicious, or because oflaziness to
learn, which is considered most
29
IV IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

existimatur l47 , saltem nobis contemptible. Prologus Stephani 154 The prologue of Stephen the disciple
detrahere desinant poscimus 148 . philosophie 155 discipuli in of philosophy to the second part of the
[20] Non enim aut invidis For we do not take pains for the secundam regalis libri artis Royal Book of the art of medicine,
elaboramus aut 149 inertibus l5o , invidious or the lazy, but for those medicine partem 156, que est which is the practice of medicine and
set discendi cupidis et qui de 151 who desire to learn, and who do not medicine 157 0peratio the cure of diseases.
aliorum industria suas non blush to support their reasonings on morborumque 158 curatio.
erubescunt munire rationes 152. the labour of others. [1J Prima huius operis et non Since the first and larger part of this
minima parte absoluta, qu~ est work has been completed, being the
2. The second part of the Regalis dispositio occurs in several scilicet rerum mituralium et non knowledge of natural, non-natural,
manuscripts l53 . In addition to those manuscripts and printed editions used naturalium ac extra naturalium and extra-natural things163
for the first part (V f. 162rb, W f. 86va, ve f. 79va, ly f. 136r) the following scientia 159 circa hominis concerning the subject of man, we set
manuscripts have been used: subiectum, ad eam qu~ huius our hands to that which is the aim of
operis est finis manum this work, i.e. the practice of
Worcester, Cathedral Library, F.40 (= F), s. xii.
mittimus 160, medicine scilicet medicine, in which part both the
Berlin, Staatsbibl. Preussischer Kulturbesitz, lat. fol. 74 (= B), s. xii (for operationem, qua 161 in parte et preservation of health and the
these two mss, see p. 20 above). sanitatis custodia et morborum rejection of diseases and the neutral
neutrique 162 repulsa continetur. state are contained 164.
Paris, B.N.F., lat. 6914 (= P), s. xv. This is a beautifully-written
manuscript of Regalis dispositio, Part 2, Books 1-5 only. At the top of the
first folio is written: Anno domini m 0. cccco. /.xxxv 0. die xii ianuarii
G

Iohannes de bailleul hunc librum incepit, perfi,ciat deus sua clementia. P is

the only manuscript that apparently has changed all the alphanumerical 154 B omits
notation into roman numerals, as well as correcting some of Stephen's non- 155 philosofie V, philosophye W
Classical Latin: 'neutrique' to 'neutrorumque' [1] and 'posse ... fuit' to 156 secundam ... partem) secundam libri artis medicine partem V, secunda regalis
'possibile fuit' [121. libri artis medicine parte W, secunda parte regalis libri artis medicine ve 1y
157 Wve ly add 'complementum et' ; BP omit 'medicine'
Erfurt, Wissensch. Bibl., Amplon. F 250 (= E), s. xiii med.
158 morborum qut! B
BEF provide a very similar text, which appears to be reliable. It is 159 scientia ac extra naturalium V
striking that the readings of the printed editions in Part 2 (but not in
160 V adds above 'mittimus' : 'ponimus' (reading unclear)
Part 1) are very close to those ofW.
161 que W
162 neutrique) neutrorumque P
163 Stephen uses the standard division of the texts of the so-called Articella ; cf.
Iohannitius, Isagoge, beginning : Medicina dividitur in duas partes, scilicet in theoricam
et practicam. Quarum theorica in tria dividitur: scilicet in contemplatione rerum
naturalium et non na'turalium et earum que sunt contra naturam (Articella, Venice, 1502,
147 estimant ut L, existimant ut V, extimatur W sig. A i recto).
148 possimus V 164 This three-part division of practice is also found in the texts of the Articella :
149 V adds p.c. Iohannitius, lsagoge (continuation of quotation in previous note) : .. .ex quibus egritudinis
et sanitatis et neutralitatis et suarum causarum et signi{icationum scientia procedit ;
150 inhertibus Vve ly
Galen, Tegn.i, beginning : Medicina est scientia sanorum, egrorum et neutrorum (Articella,
151 qui de) quidem W Venice, 1502. Sig H viii recto). It is striking that al-Majlls1 (Halyabbas) has a two-part
152 rationes) vel adds 'nota pulcrum dictum' division here, and misses out the «neutral state .. ; cf. Regalis dispositio, Book 1, chapter
4 (Lyons, 1523. f. 9r) : Operatio autem in duas distribuitur partes, in custodia sanitatis in
153 The fullest list is in C.H. HASKINS, Studies ... , p. 131-2. sanis et morborum rnedelam.

30 31
IV
, ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
IV

[2J Dico autem hanc 165 medicin~
finem, hoc est huiuS 166 artis
formam, quoniam qu~ in
precedentibus decursa 167 sunt
omnium ad hanc est
propositum l68 , set diversum l69 .
[3] Nam naturalium quidem
scientia rerum 170 medico ad
liquidum necessaria non est ut l71
ex causis per demonstrationem
eam foveat, et ad ultimum eas
usque 172 prosequatur, set a
phisico 173 qu~ dicta sunt
tantum 174 suscipitl 75
philosophol76, nec est illi viS I77
utrum omnino ita sit 178 curve 179
absolvere. HuiuS 180 namque
subiectum homo cum sit181 ,
medici scilicet quod 182 et
165 hunc Wue ly
166 huiusmodi W
167 in alio, discussa V Lm.
168 esse dicunt propositum Wue ly
169 diversari Wve ly
170 rerum scientia E
171 ut) P adds 'et'
172 eas usque) eas B, eam usque V, usque EPWue ly
173 phisyco ue
I call this the aim of medicine, i.e. the
form of this art, since the intention of
all things which have been mentioned
in the previous books was leading up
to this, but in a different way.
For the knowledge of natural things
is not obviously necessary for a doctor
so that he may develop it from causes
through demonstration, and finally so
that he may follow them to the end,
but he merely takes over what is said
by the natural philosopher, and it is
not in his power to resolve whether it
is so completely, and why. For, since
the subject of this is man, what is
part of the subject of the natural
philosopher certainly concerns the
doctor, but not in the same way as it
phisici 183 pars subiecti concerns him [i.e., the natural
philosophil84 , set non quo et illius philosopher] .
modo l85 •
[4] Nam phisici186 quidem For belonging to the natural
secundum 187 componentia - <philosopher> is <the investigation>
animam 188 scilicetl89 et corpus according to the components, i.e. the
huiusque partes et utriusque l90 soul and the body and its parts, and
potentias atque ad invicem 191 the powers of each and their relations
proportiones, medici secundum to each other, but belonging to the
humorum temperantiam 192 in doctor is <the investigation>
qua sanitas constatl93 , et according to the temperament of the
contra 194 ubi languor195 et humours in which health consists,
morbus. and, on the other hand, where
suffering and illness <subsist> 196 .
[5] Quid itaque homini secundum What happens to man according to
hec dU0 197 fiatl 98 accidentia, these two accidents is for the doctor
medici est rimari, totusque l99 to ponder, and his entire aim is to
183 physici W
184 phylosophy W
185 illius modo] illius moto V a.c., illiusm~ ly
186 philosophi P
187 quidem secundum) quod ue ly
188 anima ly
189 scilicet animam P
190 utrisque V
191 abinvicem BEV
192 temperantia ly
193 consistit P
194 et contra] et econtra B, et contrarie P, econtra Wve ly

174 tantum] B omits, V adds p.c., C8usam W 195 ubi languor] ubi langor BEVWve ly, ut longor P
196 The language here is similar to that of Aristotle's Physica ; 193b ; cf. the versi~n
175 suscipiat V
in the Translatio Anonyma, ed. MANSION (n. 43 above>, p. 26: ...speculandum e~t quid
176 philosopho] phisico P, philosofo V differt mathematicus a phisico... Si namque physici quid est sol aut luna. sc~re ~st,.
177 illi vis) illi ius B, illius VW ve ly, V adds 'ius' i.m. accidentium vero secundum se nichil, est absurdum. Aliter et quia videntur d~entes 'P~~
de natura et figura et luna et sole, et utrum speralis est terra et m.u,:dus aut. non. De. h~s
178 omnino ita sit] ita sit V a.c., alias ita sit V p.c., ita sit omnino Wve ly itaque tractat mathematicus, sed non inquantum est phys~c~ corpons termmus
179 curve) contrarie P unumquodque.
180 Huiusmodi W 197 duo hec EP
181 homo cum sit] cum sit hoc E 198 fiant Pve ly
182 scilicet quod] scilicet quidem BV, secundum quod P, scilicet quod Wve 1y 199 totusque) P adds 'est'

32 33
IV
, ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

opposites they may drive away the

IV

eius finis alterum tueri si sit, preserve the one, ifit is present, and rerum217 , ut per opposita
alterum cum acciderit200 to drive away the other when it .quibus 218 hominis COrpUS219 aut evils that are present by which the
repellere. occurs. intus aut extra afficitur220 et body of man is affected or altered
[6] Set quoniam huic201 etiam But since for this man too it is alteratur que insunt mala from inside or without, and through
quibus fiat 202 medicaminum necessary to know the natures of the pellant221, que bona per consona agreeing <medicines> they may
naturas opus est scire 203 , qu~ sint medications by which this is tueantur. preserve the goods.
herbarum, ciborum achieved, they have necessarily to [8] Porro morborum et ex quibus Moreover, the recognition of illnesses
pigmentorumque vires ad hoc204 seek which strengths of herbs, foods sint222 quibusque significentur and their origins and by what
prone qu~ve205 inobedientes 206 , and colours are amenable, causis cognitio, quasi basis et symptoms they are indicated is, as it
qu~ contra207 , querere necesse unsuitable, or opposed to this, and totius est huius223 sedes artis, were, the basis and seat of the whole
habent, et208 non quascumque not <take> whatever they have, since quoniam nisi hec224 cognita sint, of this art, since, unless these are
habent, quoniam209 morbus et illness and health are contraries, as nullum sit225 eorum nisi forte known, it is impossible to cure
sanitas contraria sunt, are equal temperament of the casu curare. anyone except by chance.
temperiesque210 humorum ac211 humours and their lack of [9] Premissa itaque 226 quasi Thus they [i.e. the contents of Part 1]
intemperies, nec212 nisi per temperament, and they can only be huius sequentis operis are put first like propositions2So for
contraria aut adduci aut expelli introduced or driven out by propositiones sunt, ut illis the whole of this work that follows, so
prevalent213 . contraries. diligenter et227 agnitis et that, when they have been carefully
[7] Hinc214 non 215 naturalium The knowledge of the non-naturals is memorie mandatis null us in228 both understood, and committed to
necessaria illis est216 scientia necessary for them so that through agendo error229 occurrat. memory, no error in practice should
occur.
[10] Set quoniam opus hOC231 ex But since we are translating this
arabica232 transferimus lingua, work from the Arabic language, and
200 accidit P
201 hec P
202 fiant B a.c. P
217 rerum scientia P
203 est scire) est scienti{l BV, scire est PW
218 quibusque F
204 hec ue 1y
219 hominis corpus] corpus hominis E, P adds 'componitur'
205 prone qU{lve] prone queve etiam V a.c., prone que V p.c., proprie que Wve 1y
220 efticitur P
206 obedientes V a.c.
221 pellantur W ue ly, peUat V
207 contrarie P
222 sunt BPWue ly
208 set P
223 huius est V
209 habet, quoniam] habent simul, quia P
224 nisi hec) ubi hec E, nisi hoc V
210 temperiensque B
225 sit] scit P, fit ue ly
211 et P
226 itaque) igitur P
212 nec] non BVWve 1y
227 et) BV omit
213 nec nisi per contraria aut adduci aut expelli prevalent] non nisi... prevalent PV
228 etiam W
a.c., non nisi... prevaleat V p.c., et intemperies non nisi per contraria aut educi aut
expelli prevaleat Wue 1y 229 error agendo P, agendo herror V
214 Huic Wve ly 230 Note the logical language.
215 non] cum P 231 hoc) P omits
216 necessaria illis est] illis est necessaria BV, necessaria illia W 232 harabica B

34 35
 IV
IV

1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

plenam, secundam250 non omnino incomplete, and the second part not
omniaque hic fere 233 posita almost all the names of the
aut certe turbidum quid, quod 251 at a1l 252 , or at least some disordered
medicaminum nomina arabum medications placed here are put
magis obesset scienti~ haberet piece of work which rather gets in the
proferuntur lingua, et234 nos forward in the language of the Arabs,
way of knowledge.
latina235 parum habebamus 236 and we hardly had common Latin latinitas.
[12] Nec vero omnino lectorem But we have not altogether consigned
assueta, prout sunt in arabico ea words <for them>, we put them the reader to error and worry, but, as
proferimus etiam237 qu~ cognita forward as they are in Arabic, errori 253 et sollicitudini
remisimus set254 quod posse was possible for us and as the Orient
nobis 238 sunt nonnunquam, que sometimes even those which are could provide, at the end of the whole
incognita ubique, set ad latine known to us, in all cases where they nobis 255 fuit et Oriens256 habebat,
in totius operis fine omnium work we have added a compendium of
formam declinationis inclinata239. are unknown, but we have declined
them according to the form of Latin que257 apud Diascoridem258 sunt all the medications that are in
declensions. medicaminum 259 breviarium260 Dioscorides, having on this side the
subdidimus, hinc eorum names in Greek, on that side, in
[11] Nisi etenim sic, nulla esset For if we did not use this method,
nobis ad presens transferendi via, there would be no way for us of nomina261 grece, illinc262 arabice Arabic, so that the man into whose
cum qui240 utrasque nosset241 translating at present, since we have habens 263 , ut in cuius venerit264 hands this work comes can ask what
nullum242 haberemus 243 linguas. no one who knows both languages hoc opus manus 265 , quid 266 each thing is, if he finds a Greek, or
queque res sit aut grecum si indeed an Arab270 .
Malui244 igitur245 paulo infirmus <well enough>. I preferred, therefore,
videri quam scientiam non to seem a little infirm than not to invenit267 aut certe268 arabem269 ,
transferre, quoniam 246 minus est transmit knowledge, since it is less of sit illi posse consulere.
pauca quam omnia aliquem a task fors someone to ask about a
interrogare,247 cum primam few things than about everything,
250 BPV add 'quidem'
quidem 248 partem set non 249 when Latinity has the first part
251 quid, quod] quidque V a.c.
252 A reference to the previous translation of Constantine.
233 omniaque hic fereJ omniaque fere hoc E, omniaque fere hic P, omnia fere hic Wve
ly 253 herrori V
234 Wve ly add 'quia' 254 si BW ve ly
235 latina] latini B, latina V, in latina PWve ly 255 posse nobisJ nobis possibile P
236 habebamus parum PWve ly 256 orationes V a.c.
237 et P 257 operis fine omnium] opere fine omniumque E
238 qu~ cognita nobisJ eam qu~ cognita nobis B a.c., que nobis cognita Wve ly 258 dyascoridem P
239 inclinatam BV 259 medicinarum Wve ly
240 cum qui] cumque E 260 breviarum Ely
241 nosceret P, nosset ve, nocet ly 261 nomina eorum P, orum nomina V
242 nullas FV 262 illic BV
243 habemus P 263 habemus W
244 Malo V 264 venit P
245 ergo P 265 manus hoc opus venerit E
246 quam V a.c. 266 quod BVve ly
247 aliquem interrogareJ interrogare aliquem E, aliquam interrogare FP 267 invenerit P
248 primam quidemJ quidem primam E, primam P
268 si invenit aut certe) sive invenerit Wve ly
249 set nonJ non tamen Wve ly 269 arabicum PW

37
36

JI
IV IV

1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

Latin names is not secure 284 , any

[13] Reliqua ver0271 que But as for the rest276 , which are only elucubrati 282 iunctamus 283 ut,
arabica272 sunt tantum 273 , aut 274 in Arabic, I have either put them quoniam Latinorum nobis studious reader who approaches our
propriis protulimus forward with their proper definitions, nominum ad liquidum peritia non work has people he can consult about
diffinitionibus, que quidem when I was able to do so, or have est, qui ad nostrum accesserit unknown <medications>. For both in
potuimus, aut omnino siluimus, been completely silent about them, opus studiosus lector, de Sicily and Salemo, where especially
studium daturi ut si quovis modo intending to make an effort, if in any incognitis quos possit consulere there are students of these matters,
nobis posse gratia dederit divina, way divine grace gives the possibility habeat. Nam et in Sicilia et there are both Greeks and people who
inter greca et arabica nomina to us, to insert the Latin names Salemi, ubi horum maxime know the Arabic language, whom he
inseramus 275 latina. among the Greek and Arabic ones. studiosi sunt, et greci habentur et who wishes can consult285.
[14] Nunc ad propositum Now let us return to our topic. lingue gnarl arabice, quos qui
revertamur. 277 voluerit consulere poterit.
[2] Et greca quidem nomina ut We put forward the Greek names as
3. Preface to Breviarium. The text is taken from ms B, f. 334rv and ms sunt plane exponimus, quippe they are, since we use letters which
F, f. 134vb; the printed editions do not include the Breviarium; this qui 286 eisdem fere in sono utimur are almost the same in sound as
preface has also been transcribed from ms B by Valentin Rose in quibus et illi elementis, arabica those they use ; not thus for the
Verzeichniss (n. 22 above), p. 1062-3. non sic, cum apud illos quedam Arabic names, since among them
sint littere287 quarum 288 omnino there are certain letters to which the
latina extorris est lingua. Latin language is completely alien.
[1] Ad umbilicum, per Dei Having brought to an end, thanks to Scripsimus itaque illa Thus we have written them with the
gratiam, nostre translationis God, the labour of our translation, as vicinioribus quibus potuimus. closest <letters> as we have been able
deducto labore, quod reliquum est remains to do and as we promised at to.
quodque 278 in huius secund~ the beginning of the second part, we [3) lam ergo hinc propositum Now, then, let us begin the proposed
partis polliciti sumus principiis, add a breviary of all medications, incipiamus. work from here.
medicaminum omnium which we toiled over by comparing
breviarium 279 subdimus, quod, the books of Dioscorides written in 4. The Explicit to the Breviarium, B f. 343r, F f. 136v, transcribed
collatis et arabice280 et grece281 Greek and Arabic, and we join it so
V. Rose, p. 1063.
scriptis Diascoridis libris, that, since our competence in the

[1] Hec sunt qu~ in Siria289 ad This is what we have found up to now
270
an Arab.
The implication is that it is more likely that the reader will meet a Greek than presens nostra invenit manus de Iby our own effort in Syria concerning
271 V adds p.c.

272 arabice P, arabicam ue Iy 282 elucubrata FM

273 sunt tantum] tantum sit F, tantum sunt EVa.c.Wue Iy 283 invetamus F, inventamus Rose
284 Stephen has arranged the breviary that follows in three columns, with the Greek
*
274 'aut' before 'arabica sunt tantum' BEFPVWue Iy
names first, in alphabetical order (by the first two letters), the Latin names in the
275 inseremu8 P middle and the Arabic names on the right. See Plate 7. For more than half of the Greek
276 Presumably, the materia medica mentioned by Halyabbas but not by Dioscorides. names Stephen fails to find a Latin equivalent.
277 revertimur VWue Iy 285 This suggests that Stephen's audience is in Italy rather than Antioch.

* 278 quodque Rose, quoque BF 286 que B *

279 breviarum F 287 latere B
280 arabicis Rose 288 quare B
281 grec~ B 289 syria F

38 39
 IV
medicaminum interpretatione
grece et arabice 290 • In quo si quid
erratum posteritas invenerit291,
nobis non inputandum credat.
Neque enim noster est, set
aliorum labor, et sic inventum
posuimus292 . Si autem dederit
Deus et ocium gerendorum
affuerit, utrum que plenius
rimaturi sumus.
[2] Hec interim huic subdidimus
operi, quo sit studiosis ad laboris
emolumentum aliquis finis. Et
nos Dei gratia totum quod de
regali polliciti fueramus
explicuimus.
B. The prefaces to the four Books of ' Stephen the Philosopher's
translation of the Book of Mamon'
This text is found in only one manuscript, Cambrai, Bibliotheque
municipale, 930.
1. Preface to the first Book, f. 2r.
Incipit liber Mamonis in
astronomia a Stephano
philosopho translatus.
[1] Quoniam in canonem
astronomi~ quas proposueramus
regularum exsequ<u>to tractatu
promissum exsolvimus,
secundum hoc opus -licet
290 grec~ et arabic~ MS
* 291 invenit B
* 292 The Vat. Urb. catalogue suggests that ms V, f. 371r, omits the sentence 'Ne que
enim ... posuimus'.
the interpretation <of the names> of
1 ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
medications in Greek and Arabic. If
posterity finds any mistakes in it, it
should not blame us. For it is not our
work but the labour of others, and we
have merely written it down in the
way that we found it. But if God
allows and the leisure for doing it is
available, we shall explore each more
fully.
Meanwhile I have added these
<interpretations> to the work, so that
students should have some
completion of the benefit of their
labour. And, by the grace of God, we
have completed everything that we
had promised concerning the Royal
<Arrangement>.
Here begins the book of Mamon on
astronomy, translated by Stephen the
Philosopher.
Since we have fulfilled our promise,
having written a treatise on the rules
we had proposed for the canon of
astronomy, I approach this second
task -although hard and concealed in
40
arduum et subtilissimo ac
multiplici natur~ celatum
archano - non inconsulta aut
impudenti temeritate, set
frequenti et animi et utilitatis
ammonitione aggredior.
[2] Sit enim licet magnorum
super his gravissimorumque
disputatio philosophorum, tamen
mediocres persepe maxima,
quemadmodum maiores curant
minora.
[3] Illud quoque attendendum est
plurimum quod, cum omnis a Deo
sit sapientia, ea autem verior et
sine scrupulo fallati~ concessa sit
« nemo noverit ».
[4] Unde et qui graves habentur
philosophi sepe errasse293
maximis in rebus, eorundemque
verius et perspicatius alios qui
nec philosophiam adepti essent,
nec ad eam aliquando posse
pertingere existimaverent, de
divini muneris larga benignitate
hausisse noticiam comperimur.
[5] Testes sunt Plato et
Aristotiles quos omnium
liberalium artium fere magistros
habemus. Quorum Plato in
multis a veritate dissonat,
Aristotiles294 mundum non esse a
Deo conditum de nichilo set cum
eo sicut nunc est, tamquam cum
corpore umbram, processisse, et
condidit et argumentis fallacibus
conatur asserere-eo nimirum in
293 MS extra se
294 MS Aristotilis
IV
the most subtle and multiple secret of
nature -not with inadvised and
shameless temerity, but as a result of
the frequent suggestions of the mind
and of utility.
For, although there is concerning
these matters a disputation of great
and most serious philosophers,
nevertheless lesser men very often
look after greater matters just as
greater men look after lesser ones.
That also is most of all to be observed
that, since all wisdom is from God,
that wisdom is truer and is conceded
without a stumbling-block of
falseness <that>« no one knows ».
Hence we find that those
philosophers who are thought to be
serious have often erred in the
greatest matters, while others who
have not obtained philosophy nor
have thought <themselves> to be able
to reach it at any time, have often
drawn forth know ledge of the same *
things more truly and clearly from
the generous kindness of the divine
gift.
Witnesses to this are Plato and
Aristotle, whom we hold to be
masters virtually in all the liberal
arts. Of these, Plato is discordant
with the truth in many things, while
Aristotle has both stated and
attempts to prove with fallacious
arguments that the universe was not
created by God from nothing, but
proceeded with him just as it now is,
like a shadow accompanying a body -
41
IV
loco intellectus et animi et
oculorum privatus offitio, qui
fidelium simplicitati divina
nascitur misericordia.
[6] Idem ipse in hac de qua
proposita est disputatio
questione, cum de celestibus
speris dissereret, octo positis, de
nona non, ut quidam arbitrantur,
consulto tacuit, set se ad eius
noticiam nequaquam pervenisse
manifestum nobis reliquit
testimonium.
[7] Quod nullatenus arroganter
dictum cuipiam videri velim, et
quod tant~ gravitatis et scienti~
et ex eisdem auctoritatis adepte
philosophus ignorasse dicatur me
non latuerit. Nam etsi inter
maximos locum non obtineam, ad
eosdem 295 tamen aspirans,
mediocrium invasi disciplinam.
[8J Habet enim ille sua qui bus
plurima consumpta opera
perpetuitatis dum philosophantes
vixerint nomen adeptus est,
* quorum tamen pluraque a
maioribus, omnia autem a Deo
preter obfuscata falsitatis errore
accepit. Quare nobis quoque, qui
nichil aliis derogamus, si quidem
idem omnium ditissimus Deus
annuat, invideri dedecet, cum ab
eo accepta alios docere quam
295 e08 MS a.c.
; ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
in this statement the intellect, which
is a natural endowment of the
simplicity of the faithful through the
divine mercy, is deprived of the use of
its mind and eyes.
The same philosopher in this
question with which the proposed
disputation is concerned, when he
argued about the celestial spheres,
having posited eight, was silent about
the ninth, not, as some think,
deliberately, but he left clear
evidence to us that he had in no way
arrived at the knowledge of it.
I should not wish to seem to anyone
to have said this arrogantly, and it
should not have escaped my notice
that a philosopher of such
seriousness and knowledge and
authority obtained from <his skill
and knowledge> should be said to
have been ignorant. For, although I
do not obtain a place among the
greatest, nevertheless, aspiring to
them, I have approached the learning
of scholars of the middle rank.
For he [Aristotle] holds his own
<opinions>, thanks to which, after
devoting an immense amount of labor
<to them>, he won eternal renown
whilst there were still men alive who
philosophized. However, he has taken
the majority of these opinions from
his predecessors, but all from God,
except those clouded over by the error
of falseness. Therefore for us too, who
do not deny to others their due, if the
same God, most rich in everything,
42
ignavie silentia296 tegere
malumus.
[9] H~c autem ideo, quia nisi
tanta foret obtrectantium
multitudo, feraciores 297 habuisset
latinitas auctores fertiliorque
apud nos philosophi~ seges
pullularet. Cum etenim plurimi
essent exercitus detrahentium,
pauci qui benigne susciperent,
pauciores certe artium scriptores
magis exterrebantur multitudinis
immanitate quam adunarentur
aliquorum benigno studio.
[10) Unde factum est ut, que fere
plenitudinem posset habere
artium, nunc ceteris gentibus
Europa videatur humilior, quippe
qu~ quos educat contra fontem
scienti~ sepius oblatrantes sentit
sibi ipsis rebelles, nunc h~c nunc
illa nunquam consona
ruminantes.
[11] Qu~ res tantum attulerit
litteralis scienti~ odium ut a
quibus summe venerari debuerat,
rerum p<ublicarum> rectoribus
summe odiretur. Qua ex re illud
quoque malum ortum non dubito,
quoniam cum equitatis
observand~ causa reges ceterique
bonorum ordines qua et leges
296 MS silenti{!
297 MS ferociores
298 For this phrase compare Cicero, De officiis, 3, 5 : nullus {eracior in philosophia
locus est.. .quam de officiis.
IV
grants it, it is shameful to be
grudging, when we prefer to teach
*
others what we have received from
Him rather than to bury it under the
silence of inactivity.
<We do> this because, if the number
of critics had not been so great,
Latinity would have had more
productive authors and the crop of
philosophy298 would sprout more
fertilely amongst us. For, since the
armies of the critics were so
numerous, <and> those who received
<philosophy> kindly were few, the
<even> fewer writers of the arts were
more terrified by the size of the
majority than united by the kind
support of a few.
Hence it happened that Europe, who
could have had virtually the complete
fullness of the arts, now seems to be
lower than the other nations, since
she experiences that those whom she
brings up by the fount of knowledge
are more often back-biting rebels
against themselves, now scheming
these things, now those, never
achieving harmony.
This condition brought such hatred
for the knowledge of letters that it
was supremely hated by the leaders
of the Republic, by whom it should
have been supremely venerated. I do
not doubt that that evil has also
arisen from this condition, because,
when kings and the other ranks of
the good <exist> for the sake of
43
IV
constitute sunt-eius autem
quedam pars naturalis, quedam
sit posita, quarum alterius
omnem fere scientiam, alterius
maiorem partem litterali plus 299
exercitio adipiscamur - ignorata
semita300 veritatis, qui reges esse
debuerant, nomen tantum
retinentes tirannos se operibus
exibebant.
[12] Ad consuetudinum enim
inconsultam observantiam
* delapsi, eadem propemodum lege
iudicant, qua qui induxerunt in
usum If. 2vl tiranni - 301his in
consuetudinibus et eorum qui eas
peperunt observantia[m] magis
avaritie ardor exprimi quam
hominum societatis communis
utilitas, ut ita dicam, somn<i>ari
potest.
[13] Earum qui adepti sunt
noticiam apud improbos,
sapientis nomen impudenter
arrogant, dumque eas ubicumque
possunt irritant, iuste se
iudicasse falso gloriantur.
Quapropter veritatis inscii, pars
avaritie laborant, pars sui profusi
inconsiderate sua dilapidant,
atque, ut habeant que sociis sui
furoris largiantur, alios falso
iudici<o> condempnant, aliis vi
sua palam extorquent,
299 MS plus
300 MS semi~
301 No punctuation in the manuscript.
. 302 The word 'dilapidare' is rare in Classical
lme 898), but is common in civil and canon law, in the sense of« squander...'
observing the justice by which laws
too are set up - of this <justice> part
is natural, part conventional, of
which we obtain virtually all the
knowledge of the one, and the major
part of that of the other more by the
exercise of reading and writing -
having neglected the path of truth,
those who ought to have been kings,
retaining the title alone, showed
themselves by their deeds to be
tyrants.
For, having lapsed into an inadvised
observance of customs, they judge
almost by the same law by which the
tyrants <did> who had brought <the
customs> into use -in these customs
and the observance of those who have
engendered them, the ardour of
avarice can be expressed rather than
the common utility of the society of
men can be dreamt of (so to speak).
Those who have obtained a
knowledge of these customs among
bad men, arrogate shamelessly the
name ofa« wiseman,., and, while
they cause havoc wherever they can
with these customs, they falsely boast
that they have judged justly.
Therefore, ignorant of truth, some
strive for greed, others, lavish with
their own goods, squander them302
without thought, and, so that they
should have what they may donate to
the companions of their madness,
Latin (it occurs in Terence Phormia
, 306
44
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
existimantes se in suos liberales
videri, si eos quacumque
locupletent ratione.
[14] Qua in re duplex malum esse
prospicio, quoniam et hii qui
eripiunt aliis quo alios
gratificent, rapine vitio societatis
humane artissimum vinculum
detrahunt, et hii ipsi quibus
conferunt, aliena turpiter
occupant. Qu~ quantum a iusticie
liminibus arceantur facile 303
cuiusvis sane ponderantis omnia
nec se ipsum nimium amantis
animus deliberat.
[15] Nee hoc quidem mirum si is
qui, adolescentia, que ~tas vitia
maxime suadet, turpiter
transacta, turpitudinis usum
quippe incorreptus iuventuti
intulerit, firmus iam his que
consuevit, ad naturalem german~
que iustici~ veritatem non
revertitur.
[16] Qui cum a pueritia in
philosophie cunabulis enutritus,
factus iuvenis, puer sibi ipse
videatur - tanta est huius
virtutis cautio. Quanto enim quis
plura de ea306 novit, tanto plures
difficilioresque occurrunt306
303 MS adds p.c.
304 For« germana iustitia .. see Cicero, De officiis, 3, 69.
306 Reading unclear; perhaps 'eo' a.c.
occurrunt] MS p.c.
IV
condemn some with a false judgement
<and> openly extort by force from
others what belongs to them,
thinking that they seem to be
generous to their own people, if they
enrich them in any way.
In this matter I see the evil to be
twofold: since both these who snatch
from some so that they might please
others, by the vice of plunder tear
apart the most strong bond of human
society, and these to whom they give
the gifts themselves basely possess'
what belongs to others. How much
this is divorced from the portals of
justice, the mind of anyone who
sensibly weighs up everything and
does not love himself too much can
easily ponder upon.
Nor is this surprising if he who has
passed his youth in a base fashion -
youth being the age which most of all
encourages vices -uncorrected, has
brought his practice of baseness into
his manhood, <and,> now firmly set
in those things which he has become *
accustomed to, does not return to the
truth which is natural and belongs to
real justice30".
He who has been nurtured from
childhood in the cradle of philosophy,
when he becomes a man, should seem
to himself to be a boy - so much is the
caution of his virtue. For, the more
anyone knows concerning it, the more
numerous and the more difficult are
45
IV
IV

ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

questiones, quarum
scrupulositate non animus
inquirentis deterretur, set
discendi cupiditas quedam cum
iocunditate augetur.
[17] Set de his hactenus. Alterius
enim sunt negotii. Verumtamen
locus hic paululum attingendus
fuit, ut detractionis venenum
quantas pariat incommoditates
manifestum sit, atque sic corrupti
mores detrahere quiescant, sui a
bonis et simplicibus caveantur.
the questions that occur, by whose
pitfalls the mind of the enquirer is
not deterred, but a certain desire for
learning with pleasure is increased.
But that is enough about these
matters. For they belong to another
subject. Yet this topic deserved to be
touched upon a little, so that it may
be clear how great the inconveniences
are that the poison of criticism gives
rise to, and thus corrupt characters
may cease to criticise, and the good
and the simple may take precautions
for their own characters.
in sua Sinthasi disponuntur
circuli in309 speris etiam quo
modo possint inveniri,
laborantibus in hac arte via
teratur.
[21] In quo - nichil enim
perfectum michi vel cuiquam ad
explicandum concessum
arbitror - si quid pretermissum
superflueve positum fuerit
sapientium arbitrio corrigendum
relinquo.
2. Preface to the second Book, f. 15r.
struggling in this art, the way can be
trodden to how those circles which
are arranged in their spheres by
Ptolemy in his Almagest can be
found.
In this - for I think that nothing is
granted to be perfectly explained,
either by me or by anyone else - I
leave whatever is omitted or
redundant to be corrected by the
judgement of wise men.

[18] Quoniam autem in canonis
regulis multa tetigimus, que in
hoc opere explicari desiderant,
promissum preterire consilium
non fuit, ut quod illic dubietatis
scrupulus fastidium generaverit,
huiuS307 operis benefitio sopiatur.
[19] Atque hec est ratio que me
maxime ad hoc opus coegit, ne
autem anxium308 lectorem a
studio repulsum iri paterer
nostratumque utilitati quoad
posse consulerem, neve quod
pollicitus fueram aut ignorasse
aut inertia neglexisse arguerer.
[20] Placet igitur celestium
sperarum circulos, numerum,
ordinem quo veri us potero
quantumque humana patitur
ratio, aperire, ut, qui a Ptholomeo
307 hoc MS
308 MS anexium
But since in the rules for the canon
we have touched upon many things
which need to be explained in this
work, it has not been our desire to
eschew our promise, so that any
annoyance that a stumbling-block of
doubt has generated there, might be
appeased by the benefit of this work.
And this is the reason which
compelled me especially to this task,
lest I allow the worried reader to be
driven away from study and <so
that> I should consider as far as
possible the benefit to our people, or
lest I should be accused either of
having been ignorant of what I
promised or of having neglected it out
of laziness.
I have decided, therefore, to reveal
the circles of the celestial spheres,
their number and order, in as true a
way as I can and as far as human
reason permits, so that, for those
[1] In astronomie michi suscepta In my task of undertaking a
disputatione laboranti, de qua disputation concerning astronomy, on
pauca certe habet latinitas which Latinity has few <facts> for
eorumque310 pleraque erroris certain and most of those obscured by
obfuscata caligine, obici fortassis the darkness of error, my attitude
animus <a> doctis poterit could perhaps be criticised by the
arrogans in invidia quod in learned as being arrogant out of envy,
Macrobium inter philosophantes because on so many occasions I attack
non mediocrem totiens acrius rather harshly Macrobius, who is not
invehar, eoque amplius quod of the middle rank among those who
usque ad hee tempora omni philosophise; all the more so because
caruerit obtrectationis livore. up to this time he has lacked every
bite of criticism311 .
[2] Quibus vellem satis esset mea I would wish that my intention
cognita voluntas intelligantque should be well enough known to these
me latine tradere facultati men, and they should understand
nostratum incognita auribus that I hand over to the resources of
309 MS In (as if beginning a sentence)
310 eorum MS a.c.
311 For a comparable criticism by a translator from Arabic of those who rely on
Macrobius, see Petrus Alfonsi's Epistola ad Peripateticos Francie, ed. J. TOLAN in Petrus
Alfonsi and his Medieval Readers, Gainesville etc. 1993, p. 167: Alii vero post lectionem
Macrobii et ceterorum qui in hac arte laborasse videntur, satisfecisse sibi et artis huius
notitiam plenius se esse consecutos presumunt. Porro, cum ab eis eorum ratio (qui se scire
dicunt) exigitur, in ostensionis argurnentatione deficiunt et in auctores totam vim sue
probationis refundunt.

46 47
IV
archana, que cum frequentibus
vigiliis diutumis cogitationum
recessibus exquisita
comparaverim, quorum
Macrobium aut inscium fuisse
video aut intellecta perversa
depravasse expositione.
[3] Horum alterum cum ad filium
suum, quem« sapientia sua
sapientiorem fieri» vellet,
scriberet fuisse dicendum non 312
est; nemo enim dilectum sciens
perverse instruit. Non igitur
intellecta veraciter depravasse,
set non intellexisse potius et
ignorasse iudicandum est.
Quamobrem non michi in huius
artis peritia philosoph0 313 , set
cum inscio contencio est.
[4] Quod si et intellexisse et
intellecta commode et desiderato
nature ordine exposuisse dicetur
- neque enim ipsos assumpturos
membrum tercium existimo
intellecta depravasse - legant
prius nostra quam distrahant,
atque tum demum cum nos If.
15vl in sententie nostre
subsidium attulimus rationum
cum Macrobianis in unum
tamquam certaminis discrimen
collatarum, equa lance314
partiantur conflictum, agentis
scilicet et defendentis seu testis
312 Ms adds p.c.
313 pho MS
314 lancee MS a.c.
Latin secrets that are unknown to the
ears of our people, which I have
obtained having sought them out by
frequently staying up all night and
retreating into contemplation during
the day. These secrets I see that
Macrobius either was ignorant of or,
if he understood them, he vitiated
them with a skewed exposition.
Of these alternatives, since he wrote
to his son whom he wished « to
become wiser by his own wisdom »,
the second should be said not to have
been possible; for no one knowingly
instructs wrongly he whom he loves.
Therefore one should judge not that
he vitiated what he had understood
truly, but rather that he did not
understand and was ignorant. For
this reason the contest for me in the
skill of this art is not with a
philosopher, but with an ignoramus.
But if he is said to have both
understood and to have explained
what he understood in an appropriate
way and that required by nature - for
I think that they will not accept the
third alternative (?) : that he vitiated
what he understood - let them read
our words before they criticize them
and then at last when we, for the
support of our opinion, have
introduced the decision of, as it were,
a contest of arguments with
Macrobius brought together in one
place, they should divide the fight
impartially, i.e., by settling what has
48
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
<c>omissa, per sancta
iuridicaria315 sedentes constantia,
omni vacuum passione parem
confligentibus assensum
prebeant enodatisque meis
Macrobiique sententiis quodque
in sui patrocinium intellectus
uterque asserit, quem 316 de statu
deiectum aut forte[m] inferiorem
animadvertunt317, ipsum aut
inscium horum aut
intellectorum318 depravatorem
iudicent, ac tamen demum cuius
sit secte amicitior3 19 assensus sub
veritatis indagine.
[5) In Macrobium igitur nostra
iccirco maior est animadversio
quoniam apud nostratum
opinionem ceteris ipsum
copiosiorem in astronomia et
sentio et relatum per
quamplurimos est. Quamobrem,
cum is artem teneret ceterisque
amplius peccasse acutius
intuentibus deprehendatur, ewn
iustius reprehensionis lima
corripimus. Qui cum alios
precellere credatur, inveterat[i]o
iam diu errore latinitas sequaces
implicuit, atque hec quidem
precipua causa in ipsum
nominatim conquerendi michi
est.
315 iniudicaria MS
316 quam MS
317 animadvertit MS
318 MS adds p.c.
319 amiticior MS
IV
been comitted by the plaintiff and the
defendent or witness through sacred
judicial constants (?) ; they should
provide a support for those fighting
that is fair and without any
favouritism, and having unraveled
my opinions and those of Macrobius
and what each <of us> claim in the
support of what we understand, they
should judge him whom they notice
has been thrown off his feet or is by
chance the lesser one, either to be
ignorant of these things or a
corruptor of what he has understood
and, finally, <should judge> to whose
side assent is more compatible under
the scrutiny of truth.
We refer, then, to Macrobi us more
<than any other author> because
both I experience that, and it is said
by very many people that, in the
opinion of our people he has more to
say on astronomy than the rest.
Therefore, if he had control of the art
and is found by those looking more
acutely to have sinned more than the
rest, we more justly correct him with
the file of criticism. Since he is
believed to have excelled others, now
for a long time Latinity has involved
his successors in an inveterate error,
and this is the principal reason for
my complaint against him by name.
49

IV
[6] Secunda autem, quod cum
eius traditioni contraria
sentire<nt> que 320 diuturnam
adepta fidem, ea tamquam
validissimis niteretur
propugnaculis321 , eis vero et
novitas et novitatis sepius comes
nocet invidia, rationibus
quamplurimis confirmari
desiderabant, ut, si quidem ipsis
edax livor fidei derogaret,
quandoque fida rationum
remearetur custodia, eosdemque
detractantes si forte aurem
accommodent, aut ad eorum que
dicimus inclinent assensum, aut
erubescentes acrioribus
urgeant<ur> stimulis, cum.
animadvertunt322 se ubertati
nostri presidii nulla posse ratione
resistere.
[7J Nam quod et consuetudo
eorum sit323 virorum 324, ut vel
vitiis quoque ad optinendam
quamvis falso iusticie sedem
opitulentur apud hominum
opinionem, et nova licet virtutis
opinione et honestatis
pulcherrimo signata sint
cirographo, tamquam iniquitatis
plena veneno popularis absorbeat
320 sentirent que] sentireque MS
321 propugnaculisJ propinquavit in MS
322 animadvertit MS
323 fit MS
324 virium MS
~
I
But the second reason is that, when
they perceive tliings contrary to his
tradition (which, having obtained a
long-lasting credence, relies on that
<credence> as on the most strong
supports), and, on the other hand,
when novelty and, even more
frequently, novelty's companion,
envy, stands in the way of these
<people>, they desired <the new
teaching> to be confirmed by as many
arguments as possible so that, if
gnawing envy were to harm their
trustworthiness, at least the
trustworthy security of rational
arguments should speak in their
favour, and, if by chance the same
critics were to lend an ear, they
should either assent to what we say,
or be urged with sharper goads and
blush with shame when they notice
that they cannot resist by any kind of
argument the abundance of our
support.
For, that it is the custom of those
men that they should even by vices
help any false person to obtain a seat
of justice in the opinion of men, and
that novelties, although they are
sealed by the sanction of virtue and
the most beautiful writing of honesty,
the mind of the people sucks in as if
full of the poison of iniquity, both
many carefully and fully documented
50
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
animus, cum multa
philosophorum accurate
copioseque dicta factaque tum
Solonis quiddam 325 eximium
precipuus nobis testis esse
poterit.
[8] Is enim cum 326 Atheniensibus
sepius 327 interpellantibus ut
legum aliqua sanciret instituta
quibus eorum res publica
populique mores et privatorum
status regerentur, rennueret,
tandem optimatum bonorumque
flexus precibus se quod
postulabant facturum pollicitus
est si iurarent ipsi omnisque
Athenarum populus se decennio
quicquid ipse servaret servaturos.
[9] Quibus postulata
concedentibus quippe quos in
Solonem nulla iniusticie argueret
suspitio, quem et sectatorem
veritatis et iusticie non otiosum
censorem persenserant, leges
promulgat, scriptas tradit, ipse
urbe egressus post decennium
redi[i]t ; cuius in adventu civitas
plurimum letata causas more
accuratissime inquirit, dat
Solon [em] 328 ne deleret
mutatumve iri sineret
quodpiam 329 in his que ipse
exquisitissime sub equilibertatis
ratione descripserat, nec eum hec
325 quendam MS
326 'cum' added.
327 sepe a.c.
328 salonem MS
329 quempiam MS
IV
deeds of philosophers and a
remarkable incident concerning Solon
can be a special witness for us.
For he, when the Athenians were
frequently urging him to sanction
some institutes of laws by which their
Republic, the lives of the people, and
the positions of the subjects could be
regulated, refused; at length, won
over by the prayers of the great and
the good, he promised that he would
do what they demanded, provided
that they and all the people of Athens
swore that they would for ten years
abide by whatever he himself abided
by.
When they agreed to these conditions,
since they harboured no suspicion of
injustice in Solon whom they
considered to be both a seeker after
truth and a tireless advocate of
justice, he promulgates the laws and
commits them to writing; leaving the
city, he returns after ten years; on
his return the city, rejoicing greatly,
asks for the precise explanation for
<the imposition of> this time period;
Solon gives <the answer> that it is
lest be destroyed or allowed to be
changed any law among those that
he himself had written under the *
reasoning of the most subtle
51
IV
fefellit ratio.
[10] Nam populus inconsueta
primo graviter ferens, detrahebat what they were unaccustomed to with
his que non noverat. At ubi
consuevit, cupidissimo
amplexatus animo, et illa[m]
summa fovit diligentia et
auctorem plurima honestavit
glana. 19itur, cedente novitate,
evanuit invidia, quia in
amarissimis educatis etsi mel
quia novum amarum, consueta
autem dulcia licet amara sint
videantur, ubi quis desuescens
malum paulo amplius bonum
afi'ectaveri[n]t, animus in If. 16r/ seems to be bitter, and their
contraria vertitur.
[11] Insurgat igitur quantumlibet Let envy, however more just it is than
inter cetera iustior vitio invidia
que cuius est ipsum continuis
excoquit doloribus nec patitur in
quiete securum. Illius autem in
quem sit aut vix aut <n>umquam However, it scarcely or never darkens
obfuscat bona. Nichil enim nobis
ad perpetuitatem offundet
caliginis quoniam, ut speramus,
cedente novitate, cedent etiam
invidentes < >330 iustiorque
censura sequentium, quibus
nostrum neuter minus erit
cognitus quicquid livor infuderit
maledictorum absterget.
[12] Set ne totum in prohemiis
utilitatis parum conferentibus
330 Lacuna of about 12 letters.
deliberation; nor did this reasoning
fail him.
For the people, at first tolerating
difficulty, criticised those things
which they did not know. But when
they had become accustomed <to
them>, embracing them with the
most desirous spirit, they both
supported those <laws> with the
utmost diligence and honoured their
author with the greatest glory.
Therefore, when novelty passed away,
envy vanished, because, for those
brought up on most bitter <foods>,
although honey, because it is new,
customary <foods>, although they are
bitter, seem sweet, when anyone gets
unused to the evil and desires the
good a little more, his mind turns into
the opposite <affections>.
vice in other respects, rise up, which
cooks with continuous griefs
whatever belongs to it, nor allows
<itself> to be secure in peace.
the good things of the man against
whom it is directed. For no darkness
will enwrap us for ever, since, as we
hope, when the novelty goes away,
the envious too will depart, and the
more just criticism of those who
follow, for whom neither of us will be
known less <than the other>, will
wipe away whatever the spite of bad
words has sullied us with.
But, lest we spend the whole time on
prefatorial matters which provide
52
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
tempus consumamus, que dicta
sunt suffitiant, funem ab anchora
solventes demus vela ventis,
procedatque divina annuente
gratia propositi operis
desiderabilis et continua
disputatio.
3. Preface to the third Book, f. 26v.
[1] <I>am non minima propositi
operis parte absoluta, robustiores
ad ea que sequentur exurgimus,
ut, quem divina annuit
benignitas, exequendi operis
transcursum nostra non
preterlabatur sine utilitate
negligentia.
[2] Neque enim Epicuri 331
aliquando dogma audivimus, set
Peripatetice potius accedimus
claritati que, quamvis ad
virtutem [claritatem] multiplici
conexa332 rationis cathena
affectat333, primo quidem
scintillulis interlucentibus, set ex
ipsius infantia defectus
patientibus, animum cogitantis
transverberat. Post autem
pleniori instantia refer<ens>334,
circumfusis hominum vulgis
lucem cogit infundere.
[3] Qua in re mirum quoddam de
humani pravitate intellectus
occurrit qui<a>, hominum parte
feliciori ad scintillas veritatis
331 Epicur. MS
332 conex. MS
333 affect. MS
334 refer. MS
IV
little benefit, let what has been said
be sufficient, and, releasing the rope
from the anchor, let us present the
sails to the winds ; with the favour of
divine grace let the desirable and
continuous disputation of the planned
work proceed.
Now that not the smallest part of the
proposed work has been completed,
we rise more vigorously to those
matters which follow, so that we
should not pass by the cursory
treatment of the work to be done,
which divine kindness has approved,
uselessly because of our negligence.
For we have not at any time studied *
the doctrine of Epicurus, but rather
we have approached the clarity of the
Peripatetics which, although it
aspires to virtue interconnected
through a many-linked chain of
reason, at first by flashes which are
intermittent, although they suffer
defects because of their infancy, it
shakes the mind of the thinker. But
afterwards, reflecting with a fuller
steadiness, it forces the light to pour
over the crowds of surrounding men.
In this matter a surprising thing *
happens concerning the depravity of
the human intellect, because, when
the more fortunate part of men run
53
IV
usquam visas tota velocitate sui
ingenii currente, ut ex scintillis
ignem possint iudicii 335 excutere,
pars quedam visu exterretur et
tamquam ad fornacem fabri
ignivomam pavet ex ignavia
accedere; pars vero ita modicum
prius aspernatur lumen, et quasi
humile[m] ac cito defitiens
contempnit ; nonnulli vero ne in
lucem sibi ipsi prodeant et audire
fugiunt et videre. If. 27rl
[4] Unde fit ut, primis ad
lucifluum calorem veritatis
accedentibus, secundi quidem
ignavia torpentes, ignavia
iaceant ; tercii autem clariora
* putantes ea que manu attractant,
ferali vivant immanitate aut
spurca336 desidia ; quarti autem,
ne sui ipsius iudices eorum fiant
conscientie, tenebrarum malunt
latibula quam in lucem prodire.
[5] Nobis Dei gratia trium nichil
instat, hominis enim intelligens
me habere celsitudinem, esse in
imis refugio, humanam ab
horrore modestiam gravissimum
facinus 337 animadvertens, mei
ipsius in male contractis sepe
* iudiciaria sum correptor[um]
maiestate. Ex quo fit ut, otium
contempnens, secreta nature
frequenti et multa rimari
studeam investigatione.
335 iudic. MS
336 spurta MS
337 facimus MS, followed by 'ad' expunged.
,
I
I
I
with the whole swiftness of their
intelligence to the flickers of truth,
whenever they are seen, so that they
can strike the fire of judgement out of
the flickers, another part is
frightened by the sight and, out of
inactivity, fears to approach, as it
were, a blacksmith's furnace belching
flames; yet another part first depises
such a small amount oflight and
holds it in contempt as being lowly
and quickly failing; but some, lest
they should go forward into the light
for themselves (?) have avoided
hearing and seeing.
Hence it happens that, when the first
go forward into the light-bringing
heat of truth, the second kind, torpid
from inactivity, lie low, but the third
kind, thinking that what they handle
is too bright, live in beastly excess or
unclean sloth; but the fourth kind,
lest their consciences become their
judges, prefer the hiding-places of
darkness rather than to go forward
into the light.
Thanks to God, none of the <last>
three apply to us. For, understanding
that I have the highness of man, that
<God> is a refuge in the depths,
holding human modesty resulting
from fear as the most grave oifense, I
am often a corrector of myself in bad
handlings of things, by a judicial
mastership. From this it results that,
despising rest, I strive to search out
the secrets of nature with a frequent
and manifold investigation.
54
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
[6] Set, quemadmodum ait
Tullius, temeritate G. I. Cesaris
omnia suo tempore confusa in
Romani imperii universitate
fuisse, tametsi postea propagante
altius diffusiusque Christi
verissimo imperio radices sue
composite sub Constantini
proximeque sequentium
imperatorum temporibus fuerint,
nobis tamen sicut et tunc
multoque durius contigit.
[7] Omnia enim apud nos
quorundam temeritate, avaritia,
superbia, ignavia, negligentia,
falsis denique traditionibus
permixta sunt, recessit pudor,
fugit munditia, deest fides,
sacramenta diffiuxerunt.
[8) Set de his alias. Verum hec
pauca diximus ut significaremus
nostro tempore fugatis virtutibus
vitiorum tumultus invaluisse.
[9] Quare multas esse
improborum cum constet turbas,
proborumque pauci numero
magni re habentur, decet
invigilare ad confutandas diaboli
ingenii nequitias.
[10) Solus enim, pro pudor,
humanus animus deviat et a
natura degenerans invenitur cum
post occasum angeli omnia
naturam et indicti concordiam
ordinis tueantur, hoc lapides,
metalla, plante, bruta animantia,
hec omnia celestia corpora de
quibus intendimus perpetua
338 Source not yet found.
IV
But, as Cicero says338, when, by the
rashness of G. I. Caesar all was
confused in the entire Roman empire
in his time, although afterwards,
when, the most true empire of Christ
spreading more deeply and widely, its
roots were established in the times of
Constantine and the emperors who
immediately succeeded him,
nevertheless it happened to us, just
as it happened then, but much more
severely.
For all things among us, by the
rashness of certain people, are mixed
with greed, pride, idleness,
negligence, and false traditions ;
shame has left, purity has fled, trust
is lacking, the sacraments have
disappeared.
But about these things elsewhere!
Yet I have said these few things to
indicate that in our time the virtues
have been put to flight and a tumult
of vices has grown strong.
Therefore, since it is agreed that the
hosts of evil men are many, and of
good men, the number is considered
small but their value great, it is
fitting to keep awake in order to
confute the wickednesses of the wiles
of the Devil.
For, alas, the human mind only
deviates and is found to be
degenerate from nature, since after
the fall of the angel all things
preserve <their> nature and the
harmony of the order imposed on
them: e.g., stones, minerals, plants,
brute animals - all these celestial
bodies about which we aim <to write>
55

IV
servant immobilitate.
[11] Que res quomodo, cur, et a
quibus facta sit, aliud tempus
disserendum. Nunc de propositis.
[12J Hic igitur liber ordinem
sperarum et de Sole et Luna
circulos et speras, eclipsis
utriusque rationes continebit.
Verum artiori studio hec et que
postea sequentur prosequemur.
[13] In duobus primis ideo
diffusius pervagavi ut lectoris
intellectum admitterem. Nunc
vero fundamento operis posito,
per politam 340 volumus partem
surgere, quod et ipsa intuentium
oculos magis capiat, et structure
nobilitate diutius apud se animos
audientium conversari compellat.
4. Preface to the fourth Book, f. 38r.
[1] Quartus hic laboris nostri
decursus de .e. planetarum speris
et circulis et octava denique nona
spera disserens, transcurso maris
alto, fune[re] anchore portus
tranquillo attinget.
[2J Verum cum in aliis arabem
quendam plurimum secuti
sumus, in hoc quoque per
multum sequemur, licet quedam
de sperarum numero et
339 Literally: .. immobility .. , presumably as the opposite of .. mobilitas .. which
suggests .. changeability ...
340 politum MS
keep to a perpetual regularity.339
How, why and by what it was done,
should be discussed at another time.
Now to the subject-matter!
This book, then, will contain the
order of the spheres and the circles
and spheres concerning the Sun and
Moon, and the reasons for the eclipse
of them both. But we will pursue
these matters and what follows with
a closer study.
In the first two <books> I have
wandered through the subject more
diffusely, so that I might draw in the
intellect of the reader. But now,
having placed the foundation of the
work, we wish to rise through a
refined part, because it both captures
by itself the eyes of those beholding
it, and, by the nobility of its
structure, compels the ears of the
audience to turn to it for a longer
period.
This fourth voyage of our labour,
discussing the spheres and circles of
the five planets and finally the eighth
sphere, having crossed the high seas,
will reach the harbour with a quiet
rope of the anchor.
But since in other <books> we
followed for the most part a certain
Arab, in this also we will follow
<him> through much, although we
have found certain things concerning
56
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
rotunditatum invenerimus et de
circulis quidem et inclinationibus
planetarum vera perstrinxit a
quibus sperarum numerus
dissonat. Hoc autem suis in locis
aperte monstrabitur.
[3] Ceterum cum in
precedentibus ingenii vigor
velocissimus pernecessarius
fuerit et studii exercitium341 , hoc
tamen in labore ultimo precipue
invigilandum est propter nimiam
rerum subtilitatem que ratione
tantum animi qua differt homo a
belua percipiantur.
[4] Verumtamen If. 38vl ne
pulcherrime philosophie huius
simulachrum diutius sua lateat
subtilitate neve inquirentium
multimodas quasi fantasticum
eludat intentiones, quam aperto
poterimus omnia stilo
persolvemus, eotenus tamen ne
brevitatis modum possit
excedere.
[5] Non enim parva apud latinos
diutius inquievit questio :
quonam modo erraticorum .e.
globi, quorum natura342 indictus
cursus in orientem est, fiant
retrogradi et ab oriente
relabantur in oeciduas partes?
[6] Et hec343 quidem, ut verum
fateamur, questio digna est et
proponi et solvi, set a nemine
tamen eorum absoluta. Nec hoc
341 exerticium MS
342 MS adds p.c.
343 hoc MS
IV
the number of spheres and epicycles,
and he has touched upon the truths
about the circles and the obliquities
of the planets with which the number
of spheres is dissonant. This will be
shown clearly in the proper places.
But, although in the preceding
<books> a most swift liveliness of
intelligence and the application of
study were very necessary, in this
last labour one must be especially
vigilant because of the excessive
subtlety of the matters which are
perceived only by the reason of the
mind, which differentiates a man
from a beast.
But lest the image of this most
beautiful philosophy should lie
hidden any longer in its subtlety, or
lest as it were, a fantastic kind of
thi~g should elude the manifold aims
of the enquirers, we will solve
everything in as clear a style as we
can on the condition, however, that it
doe~ not exceed the bounds of brevity.
For among the Latins a not small
question has for a rather long time
caused disquiet: how the orbs of the
five planets, whose course is assigned
by nature to the east, become
retrograde and flow back from the
east to the western parts.
This question, to tell the truth, is
worthy both of proposing and of
solving, but has been completely
solved by none of them. Nor do we
57
IV IV

ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

mirum ducimus, cum occulta sit think this surprising, when the propinquior [est] pluribus the Sun, the closer ought to
res et geometricalibus exquisita matter is hidden and sought out and tamquam cui a proximo retrograde for more <days>, as being
et ap<p>robata argumentis proved by the arguments of geometry prevalente vis inferretur, that to which force is applied from an
quorum latinitas inscia in of which Latinity is ignorant, and paucioribus qui remotior <est>, overwhelming neighbour; the one
divulgato diu multumque consequently has wallowed around retrogradi deberet. which is more remote ought to
volutatur errore. for a long time in great and retrograde for less <days>.
widespread error. [10] Amplius : si ab opposito sui Moreover, if the violence of the rays
(7J Cum enim If. 39rl astutiores For when the more astute of them radiorum Solis violentia repellens of the Sun's driving them back from
horum proposite rei veritatem could not find the truth of the illos facit retrogrados, Veneris et its [the Sun's] opposition makes them
nulla possent invenire ratione, proposed matter by any reasoning, Mercurii nulla est retrogradatio ; retrograde, there can be no
* fictum quoddam et violenter Solis they attached a certain fiction - and numquam enim ad oppositum retrogradation for Venus and
radii<s> concesserunt, dicentes violently so -to the Sun's rays, saying possunt ascendere. Mercury; for they can never attain to
eo rum maiori impulsu that the planets become retrograde by opposition345 •
retrogrados fieri planetas, quasi the greater impulse of <the Sun's [11] Quod si retrahi etiam a But, if they should be said also to be
possint amplius Solis radii quam rays>, as if the Sun's rays have more sinodo dicantur, fient et reliqui dragged back from conjunction, the
ipsarum in quibus volvuntur power than the eternal courses of the eadem ratione secundo other <planets> too should be made
sempitemi cursus sperarum. spheres themselves in which they retrogradi, nee hii duo Solem to retrograde a second time by the
revolve344 • aliquando precurrerent set same reasoning, and these two would
[8J Verum id quam frivolum sit But it will be obvious how silly this is, citiores ad Solem accederent, not at any time go in front of the Sun,
faciIi patebit argumento. Satumi, by an easy argument. The violentia radiorum repercussi but the sooner they approached the
Iovis ac Martis retrogradatio in retrogradation of Saturn, Jupiter and retrocederent; sic illis semper Sun, they would retrograde, repulsed
opposito fit Solis, Veneris et Mars happens when they are in sepiusque repulsis nunquam cum by the violence of the rays; thus,
Mercurii in sinodo, et Saturni opposition to the Sun, that of Venus Sole sinodare liceret. when they are always and frequently
retrogradatio longius ab opposito and Mercury in their conjunction, and driven back, they would never be
Solis secundum sui tarditatem the retrogradation of Saturn begins allowed to be in conjunction with the
cursus incipitur et pluribus fit further away from the opposition of Sun.
diebus, paucioribus Martis, love the Sun, in keeping with the slowness [12] Quare tamdiu inveterata Therefore, let this so long inveterate
medium eorum sicut medius est of its own course, and happens in cum magistro errore cadat opinion fall with its teacher because
tenente. Set Mars Soli more days, that of Mars in less days, sententia, et quod de his of its error, and let us see in this little
propinquior, minus Iuppiter, Jupiter having the medium <number certissime iudicandum sit, in work of ours what should be judged
remotior Saturnus. of days>, just as it is the middle <of nostro videamus opusculo. about these matters with greatest
the three>. But Mars is closer to the certainty346.
Sun, Jupiter less <close>, Saturn
furthest away.
[9J Quod si retrogradatio eorum Yet, if their retrogradation is said to
vi radio rum Solis fieri dicatur, be caused by the force of the rays of

344 Cf. William of Conches, Philosophia Mundi, 2.33b-35, taken up in Dragmatieon,

IV, 4.1 (ed. Honca, p. 87) : Dieunt igitur quidam Solem esse attractivae naturae,' cum isti 345 Venus is never more than ca. 47 degrees from the Sun, Mercury not more than
planetae praeeedunt Solem, si propinqui sunt, attrahit ad se, ut adamas vel magnetes
ca. 20 degrees.
ferrum ; sin autem remoti, eogit illos stare, donee transierit. For the classical source, cf
Macrobius, Commentary, 1.20.5. 346 The end of the preface is not clearly marked in the manuscript.

58 59
IV
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

C) The reference to 'Stephen, the treasurer of the church of St

Paul' in the Cartulary of the Chapter of the Holy Sepulchre, AppendixII
Jerusalem347 :
The Latin Aphanumerical and Mixed Numerical Notation
De jure enim ~cclesiastico, tota For the whole process of ownership in the Works Associated with Stephen the Philosopher
illa possessio processerat, had been conducted according to
presertim cum earn dominus ecclesiastical law, especially since in Antioch348
sanct~ memori~ Bemardus lord Bemard, the patriarch of sacred
patriarcha oHm pro memory, had once given [the garden]
commutatione domus 8tephani, in exchange for the house of 8tephen,
thesaurarii ~cc1esi~ 8ancti Pauli, the treasurer of the church of 8t Paul
dedisset, datam ~tiam autentica and had also confirmed the donation ' The alphanumerical notation in the Rhetorica ad Herennium copied for *
sui privilegii sanctione quod apud
se habebant, cum domni august~
with the authentic sanction of his
privilege, which they had with him,
J 8tephen (Milan, Bibl. Ambrosiana, Cod. E. 7 sup.), the Regalis dispositio,
the Liber Mamonis, the astronomical table in mss London, B.L., Harley
recordationis Boamundi junioris with the agreement of the lord 5402 and Pommersfelden 66, and the translation of the Almagest in ms
convenientia confirmasset. Bohemund the younger, of blessed Dresden, Landesbibl., Db. 87, employs lower case letters of the Latin
memory. alphabet (supplemented by two Greek letters) as follows:
1 a 10 k 100 t
2 b 20 200 u
3 c 30 m 300 x
4 d 40 n 400 y
5 e 50 0 (500 z)
6 f 60 p (600 9)
7 g 70 q (700 <1»349

8 h 80 r
9 90 s
This alphanumerical notation can be distinguished from two other
kinds of Latin alphanumerical notation used in a decimal system, found
occasionally in the Middle Ages :
1) Those that are transliterations of the Greek, Hebrew or Arabic
alphanumerical notations and therefore do not follow the order of the

347 Le Cartulaire du chapitre du Saint·Sepulcre de Jerusalem, ed. G. BRESC-
BAUTIER, Paris, Libraire orientaliste Paul Geuthner, 1984, p. 180.
348 I am very grateful to David King, Fritz Saaby-Pedersen and Benno van DaIen for
advice on the numeral systems described in this Appendix.
349 This table corresponds (up to 400) with that in MS Harley 5402, fol. 16v; see
C. BURNE'IT, The Transmission of Arabic Astronomy ... , n. 37 above. The numbers in
brackets occur only in the Dresden Almagest. Variation and insecurity in respect to
annotating the higher hundreds is a characteristic of this system, as will become clear.

60 61

IV
Latin alphabet: e.g., the numerals on the tenth-century «Destombes»
astrolabe belonging to the Musee du monde arabe, Paris (quasi-
transliterations of Western Arabic forms; of tenth-century Catalan
provenance)350, and the lists of letter-number equivalents given in mss
Cambridge, Fitzwilliam McClean 165, f. 47v and London, Wellcome " 21
f. 4r (both related to the Alchandrean corpus and both of twelfth-century
Anglo-Norman provenance) which appear to be transliterations from
Hebrew (with « 0» for « 60» as on the astrolabe of the Musee du monde
arabe). Such a Latin alphanumerical notation has the advantage of
perpetuating the ancient Semitic alphabetical order common to the other
three systems, by which, for example, « i » is always « 10 », « k » is always
« 20 », etc.
2) Those in which the first nine letters of the alphabet are used with
place value as equivalent to the nine Hindu-Arabic digits, with the
addition of the sign 0 as the equivalent of zero. This system appears in the
foliation of two manuscripts written probably in the late thirteenth
century in Flanders 351 . The compound numbers are written with the lower
values on the left; hence A = 1, B =2, 1 = 9, 0A = 10, AA = 11, 0B = 20, 01
= 90, 00A = 100, A0A = 101 etc. It is exactly parallel to Abraham ibn
Ezra's procedure in using the first nine letters of the Hebrew alphabet as
the nine digits in his Sefer ha-Mispar (see n. 48 above), but these Latin
and Hebrew systems seem to have been arrived at independently352.
360 See P. KUNITZSCH, Letters in Geometrical Diagrams: Greek - Arabic - Latin, in
Zeitschrift fur Geschichte der arabisch-islamischen Wissenschaften, t. 7, 199112, p. 1-20,
and D.A. KING, The Earliest Known European Astrolabe in the Light of Other Early
Astrolabes, in Physis, 32, 1995, p. 359-404, see p. 371-2.
351 See L. GILlSSEN, Curieux Foliotage d'un manuscrit de droit civil: la somma
dkzon (Bruxelles 9251 et 9252), in Studia Gratiana, 19, eds I. FORCHIELLI and
A.M. STICKLER, Rome, 1976, p. 303-11.
352 Other number-letter equivalents which do not tit into a decimal system are found
in onomantic texts: see H. SIGERIST, The «Sphere Of Life And Death» in Early
Mediaeval Manuscripts, in Bulletin of the History of Medicine, 11, 1942, p.292-303 and
C. BURNETT, The Eadwine Palter and the Western Tradition of the Onomancy in Pseudo-
Aristotle's Secret of Secrets, in id. Magic and Divination in the Middle Ages, Aldershot,
1996, article Ill. When the Latin alphabet is used to list items and divisions of text, to
indicate the correct order of words wrongly copied, in mnemonics for numerals in
computistic texts, etc. this cannot, of course, be described as a number system, although
for the first nine letters this use is identical to that of the Latin alphanumerical notation.
Examples ofsuch a use are given in Gilissen, op. cit., p. 310-1. For further discussions of
alphanumerical and other systems of numerical notation see D.A KING, Ciphers,
(forthcoming).
62
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
The occurrence of this alphanumerical notation in a small number of
twelfth-century texts suggests that these texts come from the same
circle353 . The forms used in the Dresden Almagest (see Plate 1) probably
represent an early attempt to use the alphanumerical notation. The
original intention of the translator has been obscured by the corrupt
nature of this late copy, in which there are consistent errors: .In compound
numbers the 10s and 10,000s are invariably missing, except l~ the case of
10 and 10,000 themselves, which are represented by « k .». !hls.error may
suggest that the 10s and 10,000s were in~nded. to be d~stIngUlshe~ from
the hundreds and units by being written ID a different Ink or rubncated,
and were never filled in on the copy used by the scribe of the Dresden ms.
Sections of the text use roman numerals only (f. 6r-19r and 32r-35v) and
sometimes mixed forms are used; e.g., f. 35v : « ccc. et e dies et d. sex. et
secunde» = 365 14' 48" days. The numbers representing thousands .and
above are either written out in full (f. 51r: « sex milia et z et ~ dies»
=6,585) or by starting from the beginning of t~e alphabet agrun, an~
adding « milia» (f. 68r: « c milium et t et d parclum» = 3,144 degrees,
f. 64v : « x. milia et a milia et ~ et c dies» =311,783 days).
The scribes of the works to which Stephen's name i.s attach~d show
more confidence and consistency in using the alphanumencal notation:
A) In Milan, Bibl. Ambrosiana, Cod. E. 7 su~. (Rhetorica ~
Herennium) the alphanumerical notations have been wntten by the mam
hand which has also added the serial numbers of the figures of speech and
thou~ht in the margins354 . (There are no puncta betwe.e~ ~ompound
numbers). For example, on f. 28v, the scribe has added Expl"cr.t l,,?er :c. ad
erennium. Incipit .d., to indicate the end of Book 3 and the begInmng of
353 Almost the same notation appears to have been reinvented, as a ~ore logical
onomantic alphabet, by Comelius AGRIPPA, De occulta philosophia (1533 versl~n); see!
20 (ed. V. PERRONE COMPAGNI, Leiden, 1992, p. 307); hence ms Erlangen, Umv. B., 8
<Irm. 1127), f. 2r (discussed in G. FRIEDLEIN, Zahlzeichen und das eleme'!'tare Rechne;,
der Griechen und Romer, Erlangen, 1869, p. 20 and Table 4), and Athanasl; ~~~'
Oedipi Aegyptiaci Tomus Secundus, pars altera, Rome 1653, ~. 488 ( . . d'
. . umverselle
H&Bto,re . ' des ch,'ffres, P ans,
. 1981 , p. 29) . The same eqwvalents of letters
.
anh'
numbers are used for numbers up to 500 ; but the annotations for 600 to 900 are J, v, 1
and hu respectively. .
354 The use of this numeral system had already been pointed out by R. SABBADINI ID
Spogli Ambrosiani Latini in Studi italiani de {ilologia classica, 11, Flor~nc~, 1~3,
p. 165-388 (see 272-6). Sabbadini was unaware of any other text or manuscnpt ID which
such a numeral system was used.
63
IV
355
Book 4 • On f. 34r, the scribe adds the heading «De dignitatibus
, ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
verborum que sunt .ne. » to Rhetorica ad Herennium , IV, xiii, and the
numbers from cc•• a: » (1) to« .ne. »(45) appear in the margins on f. 34r-43r
(see PI~te ~). Slmllarly, on f. 43r, he adds the heading «De sententiarum
exornatIombus que sunt .ki. » to Rhetorica ad Herennium , IV.xxxv, and
the n~m~e~ « .a. » (1) to « .ki. » (11) appear in the margins of f. 43r-51r.
The m~lcatIons of the explicits and incipits of the books and the
n~~benng of the parts of speech and thought does not seem to have been
ongmal to the text of the Rhetorica ad Herennium 356 and may be the idea
of the scribe or his patron.
B) In the Regalis dispositio the alphanumerical notations appear
~hroug~out the text and the marginal additions, for the measurements of
m~.edlents (see Plate 3). They appear in all the manuscripts and printed
edItIons that I have used, with the exception of ms P (see Appendix I B)
but are not used universally in anyone manuscript or edition. Roma~
numerals frequently take their place, but sometimes it is the roman
numeral .that appears in the earliest manuscript and the alphabetical
numeral m the p~nted edition, sometimes vice versa 357• All the examples I
have seen are of smgle numbers, ranging from « a » (1) to « n» (40).
C) The Liber Mamonis uses three kinds of numerals in the body of the
text (there are no annotations) :
1) Roman numerals are used for single digits and certain other low
compound numbers.
2) Alphanumerical notation is used in the same context as Roman
numerals, ~ut also for most compound numbers up to three decimal places.
Th.ey may mterchange with roman numerals : e.g., f. 8r: Dividitur etiam
CUl.usvis q.uanti~a.ti~ c~rculus om,!is in .z.d. [24] partes quarum queque .xv.
gradus pnme dw~s~oms complectl.tur. More examples are given by Haskins
Studies, p. 102-3. The component letters are each usually followed by ~
punctum and are often rubricated (e.g. on fol. 22v).
355 ~other ex~ple of a lower case letter being used to indicate the book number,
app~ars m the Phy"ca Translatio Vaticana, which is another translation which should
pOSSIbly be attached to the same intellectual circle as the works mentioned here . see n.
~~~ I
356 It does not appear in modem editions of the Rhetorica ad Herennium .
. 357 The use of alphanumerical notation in the Regalis dispositio was first noticed by
Mlchael McVA~GH, to whom I am most grateful: see his edition of Guigo de Caulhiaco
(Guy de C~auhac), Inventarium sive Chirurgia Magna, Leiden, 2 vols, 1997,11, p. 69 and
84. In quot~g from t~e Regalis dispositio Guy substitutes Arabic numerals for Stephen's
alphanumencal notatIon.
64
IV
3) Numbers over three decimal places are represented by figure indice
(Hindu-Arabic numerals in their Oriental forms; see Plate 4)358.
D) The astronomical table in mss B.L., Harley 5402, f.16r, and
Pommersfelden, 66, f. 84r uses roman numerals as one set of coordinates,
and the alphanumerical notation elsewhere. In the Harley ms the zero is
confusingly represented by « t », which is the common symbol for zero in
astronomical works written both in roman and in Hindu-Arabic numerals.
Some letters have been written in small versions of capitals, perhaps to
avoid confusion with the figure indice 359•
The use of alphanumerical notation was inspired by the examples of
both the Greek and the Arabic alphanumerical systems. For the mixed
alphanumerical and Hindu-Arabic system also there are parallels in both
Arabic and Greek, though in the case of Greek the manuscript example is
at least one hundred years later than the example in the Liber Mamonis.
1) It appears to have been common practice among Arabic astronomers
to use Hindu numerals when numbers reach the higher hundreds or
exceed three decimal points 360 • In Arabic astronomical tables (Zljes) and
related works, the abjad is used for all numbers up to 360. Hindu
numerals, on the other hand, are used in many cases where the numbers
do (or in principle could) reach the higher hundreds or exceed 1000. This
includes nearly all year numbers in calendar conversion and mean motion
tables, as well as calculations concerning chronology, but also other
calculations such as determinations of sines. There are also cases of mixed
systems, in which the numerals shift to Hindu numerals for the higher
hundreds or for numbers above 1000, such as the year values and other
high numbers in al-Blrllnl's Ql1nlln al-Matudi (1030 A.DJ in ms Oxford,
Bodley Or. 516 (see Plate 6), and the tangent tables in the Zlj of I:Iabash
(ca. 850 A.D.) in ms Istanbul, Yeni Cemi 78412, f. 228r (l3 th century) and in
the ZIj of Jamal aI-Din Abo 'l-Qasim ibn Mabftl~, known as al-BaghdadI
(date uncertain), in Paris, B.N.F., ar. 2486 (copied 1285 A.D.), f. 227r-229r.
2) In ms Vat. gr. 211 (which includes late XIIIth c. tables of Gregory
Chioniades), the year values in the chronological tables, where numbers
358 Since there are no high numbers in the Regalis dispositio or Rhetorica ad
Herennium it is not possible to say whether the scribes of these texts would have used
fr.gure indice, if the opportunity had arisen.
359 See C. BURNE'IT, The Transmission of Arabic Astronomy ... (n. 37 above).
360 I am very grateful to Benno van Dalen for providing the information in this
paragraph.
65
IV
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

soon e~ceed.1000, are written in Hindu-Arabic numerals. All other values

are wntten In Greek alphanumerical notation361• Appendix ill
O~~ final p~int : Use of alphanumerical notation (the abjad) to number The Dating System in the Works Associated with
defin~tIons (as In the copy of the Rhetorica ad Herennium) is also found in
ArabIC manusc~pts. A particularly relevant manuscript of this kind is one Stephen the Philosopher in Antioch
of t~e translatIons of Greek logical texts which has been annotated by a
LatIn scholar (s.ee PI.ate 6). ~e provenance of this manuscript is unknown,
but, on the baSIS of Its ArabIC and Latin script, it is not impossible that it
comes from the same milieu as the Rhetorica ad Herennium manuscript.
The only place mentioned in the texts associated with Stephen the
Philosopher is Antioch. The dates, however, present more of a problem.
Both in the manuscripts and printed editions of the Regalis dispositio and
in the colophon of the Ambrosian Rhetonca ad Herrenium, the date of
copying is given in the form« a passione Domini (Salvatoris) ». In regard to
the Regalis dispositio these dates are of the copying of the text and not of
the translation; the Rhetorica ad Herennium manuscript was copied for
Stephen. The only external evidence we have for the dates of Stephen's
activity is the procuring of his house in Antioch between 1126 and 1130
(see Appendix I C above) and the use of his translation of at least part of
the Regalis dispositio in Heidelberg before 1140 (see n. 19 above).
The dating system discussed here is referred to in H. E. MAYER, Varia
Antiochena: Studien zum KreuzfahrerfUrstentum Antiochia im 12. und
frUhen 13. Jahrhundert, Monumenta Germaniae Historica, Studien und
Texte, Bd. 6, Hannover, 1993, p. 41. He gives the example from the
Ambrosian manuscript (an no a passione domini millesimo centesimo
vicesimo primo) and takes it for granted that one has to add 33 years to
1121, making 1154. However, A. GIRY, in Manuel de diplomatique, new
ed., Paris, 1925, p. 90, gives examples from the eleventh century of a
passione Domini referring to the era of the Incarnation, and dating from
the Passion may simply indicate a year beginning from Easter, which is
common in Antioch (Mayer, Ibid.). The addition of weekdays to the
calendar dates is not decisive here, since most of them are incompatible
with either dating.
That 33 years should be added to the dates «a passione Domini!
Salvatoris» is suggested by the different dates appearing in two copies of
the same book of the Regalis dispositio, one in years «a passione
Salvatoris », the other in years « ab incamatione Salvatoris » :
Mss WV : Finitur sermo quintus prime partis libri
completi artis medicine ... scriptus novembris die vicesima
361 See The Astronomical Works of Gregory Chioniades I The ZZj al- :4la'l, ed. octava feria secunda anno a passione Salvatoris millesimo
D. PINGREE, Amsterdam, 1986, p. 15, 17,20-21,35 and 44. "

66 67
 i
IV
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

Alduinus and Pancus in other manuscripts: The ~16.9 date in the Leipzig
.c. vicesimo septimo Alduini manu, expletus manu Panci
manuscript of the Regalis dispositio would Simply Indlca.te th~t at l~ast o~e
vi° diebus existente mense aprilis .mO. CO. xxvii 362 .
later scribe followed the example of Stephen and hiS scnbes ID their
Ms L, f. 95r : Finit sermo quintus prime partis libri practice of dating.
completi artis medicine ... scriptus .xvi. Kalendas Martii
feria sexta anno ab incamatione Salvatoris .mO. CO. lxo.
viiiiO .
The two scribes are obviously different men, but if we assume that they
belong to the same circle, the separation of the two dates (42 years)
suggests that one should add 33 years to the years «a passione
Salvatoris ».
However, the same argument can also yield the opposite conclusion,
when applied to the case of the colophon to the second part of the Regalis
dispositio. For here too there is a variation between the expressions «a
passi one Domini» and « ab incamatione Domini », but this time the year
mentioned is the same :
Mss P, Basel, D.ii.18 ete (See HASKINS, Studies,
p. 134): ... scriptusque eius manu Antiochie a passione
Domini millesimo centesimo vicesimo (v.l. 1107) septimo
mense ianuario vicesimo septimo die feria quarta.
Ms D, f. 334r and F, f. 134vb: Hic liber translatio
Stephani philosophi~ discipuli de arabico in latinum
scriptusque (scriptus F) eius manu antiochi~ ab
incamatione domini millesimo centesimo vicesimo CB
omits) septimo, mensi ianuarii, vicesima septima die,
feriaque quarta, deo gratias rerum principio et fini, a quo
principia, per quem vectus, et omnum est scientiarum
finis.
Here, only one scribe is mentioned: Stephen himself. Ms F is probably
the oldest manuscripts of the Regalis dispositio, and it is significant that it
gives the name of the scribe, i.e. Stephen, only at the end of the text. The
most economic explanation for the differences in date is that Stephen did
complete his translation and wrote out it out by the 27th January, 1127
A.D. Assuming the year began at Easter, this would be equivalent to 27 th
January, 1128 by our reckoning, and would postdate all the other dates in
« 1127» mentioned in the colophons to individual books copied by Stephen,

362 C.H. HAsKINS, Studies, p.133. MS V's «mo.coco. xxvii.° .. seems to be a later
scribe's corruption of this date, but Alduinus and Pancus were clearly copying a
manuscript very similar to L (see p. 20 above>. The editions do not have this passage.
69
68
IV
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

Plate 1 Dresden, LandesbibL, Db. 87, f. Mv.

Plate 2. Milan, Bibl. Ambrosiana, Cod. E. 7 sup., f. 36v.
A pU88,e from the translation of Ptolemy's Almage.t, Book IV, ch .7, showin, the
use of alphanumerical notation. In Hnes 13-14 the number 311,783 is written as '][ miHs. Rumples of alphanumerical notation Hating the kinds of ftpre. of speech
et a miH. et ~ et c'. (Ad Bereraraillm, IV. mii-m) ; here numbers 10 to 13.

70
71
 IV
, ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
IV

Plate 3. Berlin, Staat.bibI. Preus8ischer Kulturbesitz, lat. fol. 74, f.28r.
* A page from the earliest manuacript of the Re6td;' d;'pt»itw (Part U, Book I, ch.
17) .boMn, the use of alphanumerical notation. In line 6 the values fe' (= 5) and Id' (. 4)
can be seen; in line 10, 'lE' C. 10) and in line 11, 'I' C. 20).
Plate 4. Cambrai, Bibliotheque municipaie, 930, f. 27v-28r.
Note the use of two kinds of numerala : (1) alphanumerical notation: duplum a
terra u.que .okm id e.t .b. [.2) .. .Novie•. Ld. [= 24] ...milk .eptilt6eltti .Lh. [= 1728) ;
(2) oriental numerala : redd"ltt 46666 de.

72 73
 I
IV
Plate 5. Ms {)][ford. BOOI. Libr., Or. 516, fol. 27v.
This shows the use of Hindu-Arabic numerals for hi,her numbers
in aI-DlrUnl's QanDn al-Maf'DdI.
(by oerminioD of the Bodleian Library)
74
IV
ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE
"
Plate 6. Ms Istanbul, Topkapi Sarayi, Ahmet III 3362, f. 21v.
The be,tnnin, of aD Arabic tran.lation of Aristotle'. Categories, with Latin
annotations in a hand similar to that of the annotator of
Milan, BibL Ambl'08iana, Cod. E. 7 sup., f. 36v.
The Latin annotations are definitions from Boethius' (or the .. Composite .. ) version of
the same text: AL-MUTl'AFAQA] equivoca sunt quorum solum nomen commune e.t et
secundum illud nomen ratio subatantle elf diversa ; AL-MUTAWATIYA] univoca .unt
quorum nomen commune eat et secundum nomen ratio sub.tantie illi". e.t eadem; AIr
MUSIITAQQA1 denominaliva dicuntur quaecumque ab al;quo solo differentia CG8U
.ecundum illud nomen IaGbent appellationem.
Note that the definitions have been numbered in the ma.rtin with alphabetical (abiad)
numerals, from aUfto wau.
75
IV
IV

ANTIOCH AS A LINK BETWEEN ARABIC AND LATIN CULTURE

Plate 7. Berlin, Staatsbibl. Preusai8cher Kulturbesitz, late tol. 74, f. 335r.

A page from the Breviarium of Stephen of Antioch in which he has placed the
Greek name8 of the Materia medica from Dioscoride8 in the tirst column, the Arabic
name8 in the third column, and has tried, as far as he can, to find the Latin
Plate 8. Milan, Bibl. Ambrosiana, Cod. E. 7 sup., f. 52r.
equivalents, which he has put in the middle column.
The colophon of the copy of the Rhetorica ad Herennium, mentioning
'Stephanu8 Thesaurarius Antiochie'.

76 77
IV -, v

«Magister Iohannes Hispalensis et Limiensis» *

and Qusta ibn Luqi's De differentia spiritus et animae:
a Portuguese Contribution to the Arts Curriculum?

The only text of the Libri naturales studied in the European

universities which was generally recognized not to be by Aristotle,
was the De differentia spiritus et animae of Qusta ibn Liiqa. This was
translated, according to the earliest manuscript, by «Iohannes
Hispalensis et Limiensis». This article investigates the identity of this
«lohannes», who could have been from the region of Lima (Limia)
in Portugal, and considers the early history of the De differentia up
to the time of its inclusion among the Libri naturales. Two of the
manuscripts of the De differentia discussed are from the British Is-
les, and it would seem appropriate for a Britisher honouring a distin-
guished Portuguese professor, to show how these manuscripts shed
light on a Portuguese scholar of an earlier century. 1

The Libri naturales were the texts on Aristotelian natural philo-

sophy prescribed to be read within the Arts Faculties of the European
universities. The earliest collection, known to modem scholars as the
Corpus vetustius, was put together in the first decades of the thirteenth
Plate 9. The astronomical table iD ms London, British L., Barley 5402, f. 16r.
(by permission of the British Library)
century from translations from Greek and Arabic, and provided Ro-

I I am grateful for the help of Michael Evans. Emma Gannage. Marilyn

Nicoud, Judith Wilcox, Roger Wright. and especially Oag Nikolaus Hasse.
78
v v
bert Grosseteste, Petrus Hispanus Portugalensis, Roger Bacon and
Albertus Magnus with their know ledge of the texts of Aristotle's natu-
ral science. Towards the end of the thirteenth century this collection
was superseded by the Corpus recentius, in which Arabic-Latin
translations of Greek works were replaced by William of Moerbeke's
translations directly from the Greek. The Libri naturales consisted both
of Aristotle's genuine works - Physics, De caelo, De gene ratione
et corruptione, Meteora, De anima and the Parva naturalia - and
works which were believed to be by Aristotle, such as the De plantis
of Nicolaus of Damascus, the De mineralibus, which was three chap-
ters from Avicenna's Shifii~ and the De causis, in reality a cento of
texts from Proclus's Elements of Theology. 2 One text, however, was
included, even though it was generally recognized not to be by Aris-
totle at all: Qusta ibn Luqa's De differentia spiritus et animae. 3
The De differentia is divided into three parts, preceded by a brief
prologue. The first part (lines 33-293) concerns the spirit, which is
a subtle body diffused from the heart through the blood vessels (as
2 For a convenient account of the constituents of the Corpus vetustius and Cor-
pus recentius see B. G. Doo, «Aristoteles Latinus», in The Cambridge History of
Later Medieval Philosophy, eds N. KRETZMANN, A. KENNY and J. PINBORG, Cambridge,
1982, pp. 45-79.
3 The edition used is that of Judith C. WILCOX in The Transmission and Influ-
ence of Qusta ibn Luqa's 'On the Difference between Spirit and the Soul', Ph. D.,
City University of New York, 1985, and the references are to the line numbers of
this edition. In preparing her edition Wilcox has taken into account all the MSS
known to her, and so replaces the inadequate edition of C. S. BARACH (Excerpta e
libro Alfred; Anglici de motu cordis,' item Costa-Ben-Lucae de differentia animae
et spiritus, Bibliotheca philosophorum mediae aetatis, 2, Innsbruck, 1878) which is
based on only three MSS. She also rightly calls into question the existence of two
versions along the lines proposed by the editors of Aristoteles Latinus. Codices (2
vols, Rome and Cambridge, 1939-55; see I, p. 197): the 'Translatio Hispalensis' and
a
'Recensio anonyma'. Instead, she discerns different two versions: 'John of Seville's
version' which is found in the majority of the manuscripts and shows several 'revi-
sions' of an original translation; and 'Hermann of Carinthia's version' which is found
complete in one manuscript and in three further fragments. 'Hermann of Carinthia's
version' is an abbreviation, with some changes in terminology and phraseology, of
'John of Seville's version', but is unlikely to have anything to do with Hermann of
Carinthia; see Appendix 11.2 below, p. 266 [46].
222 [2]
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
the «vital spirit»), giving rise to the «animal spirit» in the brain which
operates sensation, cogitation and understanding, and memory, and
spreads to the rest of the body through the nerves. This section is lar-
gely medical in sources and content. The second part (lines 294-
-519) concerns the soul and is structured round the definitions of
Plato and Aristotle respectively (This is discussed in detail below).
The third part (lines 520-75) summarises the differences between the
spirit and the soul.
That the De differentia was regarded as integral to the Libri natu-
rales is indicated by the fact that Adam of Buckfield in Oxford in
the 1240s comments on it along with almost all the other texts of the
Corpus vetustius, 4 and that the statutes of the Arts Faculty of the Uni-
versity of Paris in 1255 prescribe its study alongside the other Libri
naturales. 5 Moreover the overwhelming majority of manuscripts that
contain the work are manuscripts of the Corpus vetustius or recen-
tius. Over 150 such manuscripts are listed in Aristoteles Latinus. Codi-
ces and in Wilcox's edition.
That the work was recognized not to be by Aristotle is clear from
the citations of Qusta's name by Alfred of Shareshill and Albertus
Magnus. /\ Qusta's authorship is indicated by the rubrics that appear
4 C. H. LoHR, «Medieval Latin Aristotle Commentaries: Authors A-F», Tradi-
tio, 23 (1967) pp. 313-414; see p. 323, no. 22, where six manuscripts of Adam's
commentary are listed.
5 H. DENIFLE, Chartular;um Universitatis Parisiensis, 4 vols, Paris, 1891-9, I,
p. 278 (no. 246): «librum de differentia spiritus et ani me in duabus septimanis».
6 Albertus Magnus, De anima, 1.2,13, ed. C. STROICK, Munster, 1968, p. 52.46
('Constabenluce'); Alfred of Shareshill (Sareshel), De motu cordis, c. 10, ed. C.
BAEUMKER, Beitriige zur Geschichte der Philosophie des Mittelalters, 23.1-2, M~nster,
1923, p. 40: «Costa quoque (v. l. Constantinusque) Leuce filius in libro quem de
differentia animae et spiritus edidit duos tantum esse ostendit, vitalem scilicet et ani ma-
lem, et a vitali animalem originem sumere docet.» Note that in the Corpus vetustius
and recentius no attribution is given, but that the work was recognized not to be by
Aristotle is indicated by a gloss in certain manuscripts of the Corpus which reads:
«Aristotiles in libro de sompno et vita in fine primi docet quod .iii. tal ami sunt in
corde. Similiter in fine primi de animalibus dicit quod .iii. sunt ventriculi in corde,
et ita autor iste [i.e., the author of De differentia] contrariatur Aristotili, quam contra-
rietatem inter medicos et Aristotilem recitat Aluredus in libro de motu cordis et
determinat docens quomodo sunt .iii. secundum Aristotilem et quomodo sunt ji. se-
[3] 223
 v v

THE PLATE
1 «MAGISTER IOHANNES HISPALENSIS ET LlMIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRffUS ET ANIMAE

in many of the manuscripts. 7 One family of manuscripts has the fol-

lowing heading:

Incipit Iiber differentie inter anirnam et spiritum quem Consta

ben Luce cuidam arnico suo scriptori cuiusdam regis edidit et
Iohannes Hispalensis ex Arabico in Latinum Raimundo Toletano
archiepiscopo transtuIit. g

This attribution tells us that the De differentia spiritus et animae

Edinburgh, National Library of Scotland, MS Advocates 18.6.11, fol. 108r (reproduced with permission).
was translated by «John of Seville» for «Raimundus» who must be
Raymond de La Sauvetat, archbishop of Toledo from 1125 to 1152.
In the earliest manuscript of the text, however - a manuscript which
probably provides an unrevised version of the translation - the
translator's name is given in a fuller form, in the colophon to the
text:
Perfectus est Iiber Costaben Luce in spiritus et anime differentia
interpretatus a Iohanne Hispalensi et Limiensi. Sit Iaus Deo per
infinita secula. AMEN. 9

The manuscript with this colophon is Edinburgh, National Library

Edinburgh, National Library of Scotland, MS Advocates 18.6.11, fol. 82r (reproduced with permission).
of Scotland, Advocates 18.6.11, which belonged to a doctor called
«magister Herbertus». 10 Herbertus owned several books, mostly of a
medical nature, amongst which the late-eleventh-century Arabic-Latin
translations of Constantine the African are prominent. He donated his
books to the cathedral of Durham in the third quarter of the twelfth

cundum medicos qui medium talamum et sinistrum reputant pro eodem» (London,
British Library, MSS Royal 12.0.11, fol. 358v, and Harley 3487, fol. 202r).
1 A convenient list of the MSS and their ascriptions is given in M. ALONSO ALONSO,

«Traducciones del Arabe al Latin por Juan Hispano (Ibn Dawud»), Al-Andalus, 17
(1952) pp. 129-51 (136-9).
g The oldest representative of this family is, according to Wilcox, Brussels,

Bibliotheque royale, 2772-89, of the 12th century; Wilcox uses this as her base ma-
nuscript.
9 See Plate.

10 M.-T. D' ALVERNY, «Conclusion», in Pseudo-Aristotle. The Secret of Secrets:

Cashel (Tipperary), OPA Bolton Library, Medieval MS 1, p. 105 (reproduced with permission). Sources and Influences, eds W. F. RYAN and C. B. SCHMIlT, London, 1983, pp. 132-
-40 (135-6).

224 [4] [5] 225

v v
century. 11 The De differentia has been added to the manuscript in a
1 «MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRlTUS ET ANIMAE
tiny hand, written with a thin pen, and apparently filling in a gap left
by previous scribes. 12 The Edinburgh manuscript then could have
been written as early as the mid-twelfth century, and may be expected
to be authoritative. The epithet «Limiensis» is written clearly. Who
then was this «Magister lohannes Hispalensis et Limiensis»?
The same form of name (sometimes with the additional or alter-
nate information that the work was translated «in Limia»), and a si-
milar form of colophon are found in four other translations from
Arabic: 13
1) Masha'allah, De rebus eclipsium (or In radicibus revolutio-
num), which gives as the explicit in Paris, BN, 16204, p. 391b: «Et
perfectus est MessehalIa translatus a lohanne Hispalensi in Limia
(without diacritical points on the minims) ex Arabico in Latinum».
2) cUmar ibn al-Farrukhan al-Tabari, Liber universus; Oxford,
Bodleian Library, Digby 194, fol. 127v: «Perfectus est liber univer-
sus Aomar Benigan Tyberiadis cum laude dei et eius auxilio quem
transtulit magister lohannes Hispalensis atque Lunensis de Arabico
in Latinum». 14
11 The list of his donations in Durham Cathedral, MS B.lV.24, fol. 2r, is pu-
blished in Catalogi veteres librorum ecclesiae cathedralis Dunelm., Surtees Society,
7, London, 1838, pp. 7-8. How the manuscript went from Durham to Edinburgh is
unclear.
12 D' ALVERNY (<<Conclusion», p. 135) describes the hand as having «une appa-
rence meridionale avec des caracteristiques anglaises: reuvre d'un voyageur insulaire
en sejour dans des contrees mediterraneennes?» The twelfth-century list of contents
at the beginning of the manuscript does not mention the De differentia, and corres-
ponds to the description in the list of Herbert's donation: «Liber febrium Ysaac qui
dicitur Liber Constantini de febribus. Liber simplicis medicine». Although this should
make us cautious, it should not rule out the likelihood that the De differentia was
added before the manuscript entered the cathedral library, for other texts in the
manuscript are omitted in the descriptions, and the hand of the De differentia is defini-
tely of an early date.
13 The following list supplements, with examples sometimes taken from diffe-
rent, but authoritative, manuscripts, M. ALoNso ALONSO, «Juan Sevillano: sus obras
propias y sus traducciones», Al-Andalus, 18 (1953) pp. 17-49 and L. THoRNDIKE, «John
of Seville», Speculum, 34 (1959) pp. 20-38.
14 D. PINGREE, «The 'Liber Universus' of 'Umar ibn al-Farrukhan al-Tabari»,
[6]
226
3) Thabit ibn Qurra, De imaginibus; Paris, BN, lat. 16204, p.
539: «Finit liber <imaginum> Thesbith bencorath translatus a lohanne
Hispalensi atque Luniensi in Lunia (no marks on the first four minims
of each of these names) ex Arabico in Latinum».
4) AI-Farghani, Rudimenta; Erfurt, Amplonian Q. 351 (s. xii2):
«perf<ectus> lib<er> Alfr<a~ani> in sc. ast. et rad. mot. cel. interp.
a loh. Ispalensi Ilimia [= in Limia?] atque Limensi et expletus est die
XXIIII [die] V mensis lunaris anni Arabum DXXIX.XI die mensis
Marcii era MCLXXIII [ =March 11, 1135]» (variants: «Interpretatus
in Luna a ... » [Paris, BN, 7377B and Florence, Bib1. Naz. Centrale,
Conv. soppr. J.II.10]; «interpretatus in Limia a ... » [Paris, BN, 14704,
fo1. 248r]). *
In all the above cases, «Limiensis» and «Limia» seem to be the
correct readings, sometimes corrupted into «LunensislLuna», and often
without marks on the minims - hence indicating hesitation on the
part of the scribes. Limia is the region which straddles the present-
-day northernmost province of Portugal (Minho) and Spanish district
of Orense (Galicia), through which the river Limia (Spanish)lLima
(Portuguese) flows. The river Lima on the Portuguese side of the bor-
der flows through the heart of the original county of «Portucale»,
which lay between the rivers Minho and Douro. The main town on
its banks - Ponte do Lima (sometimes called «Limia» itself 15) -
is some thirty kilometres from Braga, the ancient ecclesiastical capi-
tal of Portugal, which in turn was only a few kilometres from the
first secular capital of the kingdom of Portugal- Guimadies (capital
1128-43).
The counties of «Portucale» and Coimbra had been part of the
kingdom of Le6n, but had been given by Alfonso VI to his natural
daughter, Teresa, and her husband Henri of Burgundy as their heredi-
tary county. After Henri's death in 1112, his widow was proclaimed
Journal/or the History 0/ Arabic Science, 1 (1977) pp. 7-12. Pingree identifies and
translates the work, and shows that it is not part of 'Omar' on nativities, which prece-
des it in many manuscripts.
IS According to GRAESSE, Orbis Latinus, new ed. by H. and S.-C. PLECHL,
Braunschweig, 1972, s,v. Pons Limicorum.
[7] 227
v
queen by the «Portuguese». 16 This challenge to Leonese sovereignty
, «MAGISTER IOHANNES HISPALENSIS ET LlMIENSIS» AND QUSTA tBN LUQ~"S DE DIFFERENTIA SPIRfTUS ET ANIMAE
was carried further by Henri and Teresa's son, Afonso Henriques,
who became the first «King of Portugal» in 1128. The battle for supe-
riority between the nascent state of Portugal and Lean-Castile can be
seen from the time when Henri and his step-brother-in-law (and cousin)
Raymond of Amous attempted to divide the spoils of Toledo between
them, and Maurice Bourdin, archbishop of Braga, sought to oust the
archbishop of Toledo; he later arrogated the papacy to himself (as anti-
-Pope Gregory VIII from 11l8-21). Alfonso VII, king of Lean and
Castile (<<the Emperor») was fighting Afonso Henriques over control
of Portugal, and, between 1139 and 1152, was promoting Raymond
de La Sauvetat's claim for the primacy of the Toledan see in the face
of the rival claims of Braga. 17 That the Lima valley was regarded
as a region is clear from the frequent references to «Limia» or the
«castles in Limia» in the struggles between the Emperor and Afonso
Henriques. 18 It is in this context that the activity of «Iohannes His-
palensis et Limiensis» should be viewed.
One of the texts attributed to this John - the De imaginibus of
Thabit ibn Qurra - is accompanied by a preface in several manus-
cripts. The manuscripts with the preface are all of the fourteenth cen-
tury or later, and Thorndike, who printed the text from a corrupt ma-
nuscript in Milan, was sceptical about its authenticity. 19 The main
16 Chronica Adefonsi Imperatoris (written 1147), I. 73, in Chronica Hispana
Saeculi XII, eds A. FALQUE, 1. GIL and A. MAYA, I, Turnhout, 1990, p. 184: «mortuo
autem comite Enrico Portugalenses vocaverunt earn [Teresam] reginam».
I7The information in this paragraph has been taken mainly from P. DAVID, «An-
nales Portugalenses Veteres», Revista Portuguesa de Historia, 3 (1945), pp. 81-128,
D. W. LOMAX, The Reconquest of Spain, New York, 1978, and P. LINEHAN, History
and the Historians of Medieval Spain, Oxford, 1993 (especially pp. 269-70 and pp.
328-9).
18 See Chronica Adefonsi Imperatoris , I. 73-7, pp. 184-6. On p. 186 'Ferdinandus
Iohannis' is described as being 'princeps Limie'; on p. 243 he is 'dux Limie'.
19 L. THORNDIKE, «Traditional Medieval Tracts Concerning Engraved Astrolo-
gical Images», in Melanges Auguste Peizer, Louvain, 1947, pp. 217-74 (231-3). The
manuscripts containing the preface are: Erfurt, Amplonian, Folio 380, s.xiv, fols 139v-
-140r; ibid., Quarto 189, s. xiv, fols 68r et seqq.; Milan, Ambrosiana, A 183 inf.,
s.xiv, fols 73v-74r; Paris, BN 7282, s.xiv (but S.xv according to Thorndike), fol. 29r
(information from F. 1. CARMODY, Arabic Astronomical and Astrological Sciences in
228 [8]
v
ground for his scepticism was that the author of the preface describes
himself as seeking «Hispanae partes» for books on astronomy and
talismans. This, Thorndike wrote, «hardly fits John of Seville, who
was of Spain to begin with ... Perhaps the preface of some other trans-
lator has become attached to John of Seville's text.» Perhaps as a
consequence of this negative judgement and the corrupt nature of the
text printed by Thorndike, the data to be inferred from the preface
have not been considered by other scholars investigating the iden-
tity of John of Seville.
When the preface is read in a more reliable manuscript, it does
have a ring of authenticity. 20 The zealous search for a book and its
discovery in an «armarium» have parallels in the prefaces of other
translators of the time. 21 So also does the justification of its subject-
-matter in the light of Christianity. 22 More telling is the fact that the
author was aware of a partial translation of the same text by «qui-
dam Auriocenus». This partial translation exists: it is the Liber presti- *
giorum Thebidis secundum Ptholomeum et Hermetem of Adelard of
Bath, who could, indeed, be the «quidam Anglicanus» or «Angligena»
implied in the corrupt form «Auriocenus», and would have comple-
ted his translation before «Iohannes Hispalensis et Limiensis». A la-
ter forger would hardly have been aware of this other translation. 23
Latin Translation, Berkeley and Los Angeles, 1955, pp. 126-7). The De imaginibus
is found in two versions, one derivative from the other; both versions are edited by
F. J. CARMODY in The Astronomical Works of Thabit b. Qurra, Berkeley and Los An-
geles 1960; Carmody, however, does not edit the preface and provides only scant
information concerning the attributions in the manuscripts.
20 A new edition and translation of the preface is given in Appendix I below.
21 Cf. Hugo of Santalla's preface to his translation of the commentary of Ibn
al-Muthannii' on the tables of al-Khwiirizmi (before 1151): «huius commenti...quod
super eiusdem auctoris opus edictum in Rotensi armario et inter secretiora bibliotece
penetralia tua insaciabilis filosophandi aviditas meruit repperiri»; see C. H. HASKINS,
Studies in the History of Mediaeval Science, 2nd ed., New York, 1927, p. 73.
22Cf. the preface to the translation of the Liber trium iudicum edited in C. BURNETI,
«A Group of Arabic-Latin translators Working in Northern Spain in the mid-Twelfth
Century», Journal of the Royal Asiatic Society, year 1977, pp. 62-108 (80-9).
23 Richard LEMAY, «The True Place of Astrology in Medieval Science and
Philosophy: Towards a Definition», in Astrology, Science and Society, ed. P. CURRY,
Woodbridge, 1987, pp. 57-73 (70), interprets 'Auriocenus' as 'Antiochenus', but still
[9] 229

v
The fact that the preface (along with the attribution to «Iohannes
Hispalensis et Limiensis») is found in manuscripts of both versions
of the text published by Carmody also argues in favour of its authen-
ticity. It might have accompanied the text before it split into two
versions. The existence of such a text is implied by a citation of the
opening of the De imaginihus in a late twelfth-century manuscript of
the revision of Adelard of Bath's tables of al-Khwanzmi by Robert
of Chester. Here we read: «Nota: Dixit Ar(istoteles) in tractatu se-
cundi libri sui, nulla radix est sapientie apud eum qui caruerit
astronomia, nec lumen est geometrie cum vacua est astronomia». 24
It must be admitted that the Latin style of the preface to De imagi-
nibus is more elegant than that of other texts of John, and that of
the De imaginihus itself, which might argue for a revision by a La-
tin stylist. 2S However, it is not uncommon for prefaces to be in a diffe-
rent style from the texts they introduce.
believes that the reference is to Adelard, who visited Antioch. The Liber prestigio-
rum will appear in a collection of Adelard's scientific works edited by myself. It
is interesting to note that already in the 1260s, Albertus Magnus was apparently not
aware that Adelard's and John's translations were of the same Arabic text; for in
the Speculum astronomiae he places Adelard's version among the illicit necromantic
books, but advertises John's version as one of only two texts on talismans that are
not execrable (see edition and translation of Speculum astronomiae in P. ZAMBELLI,
The Speculum Astronomiae and its Enigma, Dordrecht, Boston and London, 1992,
pp. 242-3 and 248-9). It is true that John omits the necromantic elements (i.e., the
prayers to spirits), perhaps out of the same religious scrupulousness that is manifest
in this preface.
24 Madrid, Biblioteca nacional, 10016, fol. 3r. The text is a mixture of the two
versions that Carmody edits (see n. 19 above), which suggests that it may predate
the separation of the tradition into two versions. All the manuscripts of the De imagi-
nibus are of the fourteenth century or later, except Florence, Laur. Plut. 30, cod. 29,
which is described by THoRNDIKE (<<Traditional Medieval Tracts Concerning Engra-
ved Astrological Images», p. 235) as being of the thirteenth century, and also inclu-
des John's translation of the Secret of Secrets.
25 The most conspicuous difference is that, in the preface to the De imaginibus,
the verbs tend to be at the end of their clauses, whereas in the preface to the Secret
of Secrets (edited in Appendix I below), they are where they would be in a Romance
vernacular. Also, the Secret of Secrets preface betrays some slackness in Latin gram-
mar in its use of the infinitive for a purpose clause ('Egressus sum diligenter que-
rere'), of an adjective with a noun in the genitive, instead of in agreement ('a nullo
230 [10]
v
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS FT ANIMAE
The preface ends with a clear statement that the author of the
translation is «Iohannes Hispalensis atque Limiensis» and the place
of composition is «in Limia». The author tells us that he had travelled
into «Hispanae partes» in search of the text that he was looking for.
This involved much hardship amongst people who were not of his
own religion (i.e., presumably he was among Muslims), but, in the
end he found what he wanted in the hands of a «magister» who posses-
sed a library that included at least one book in Arabic. It is unclear
whether the search and the «master» were both within Muslim regions,
and whether «Hispanae partes» is the place of both events. But, in
a way, this is irrelevant,. for «Hispania» in the early twelfth century
meant as much the areas under Muslim domination as those in the
hands of the Christians. It is possible that «Hispanae partes» could
mean the border regions, which would apply to Portugal inasmuch
as the country south of Coimbra was still in Muslim hands (Lisbon
was not recaptured until 1147). But the implication of the words, if
we take the preface as authentic, is that John came from outside the
Iberian peninsula, as did counts Henri of Burgundy and Raymond of
Amous, archbishops Maurice Bourdin and Raymond de La Sauvetat
'and many of their retinue. Seville could even have been the Muslim
area in which John stayed, and his sojourn there could have been the
reason why he became known as «of Seville». 26
interpretum'), and of 'in+abl.' where 'in+acc.' would be expected ('locum ... in quo
introivi').
26 Compare Daniel of Morley's reference to the translator Gerard of Cremona
as 'Girardus Tholetanus' in his Philosophia (l175x87), Mittellateinisches lahrbuch,
14 (1979), p. 244. Richard Lemay's proposal - that 'John of Seville and Limia'
may have been related to Sisnando Davidiz who had been captured from Christian
territory as a boy, became vizier of Muslim Seville, escaped to the north to help
Fernando, king of Le6n, capture Coimbra, and was made count ofCoimbra - a position
he held from 1063 until his death in 1091 - is based on the contention that 'John
of Seville' is 'John David' (the distinguished mathematician who received dedica-
tions from Plato of Tivoli and Rudolph of Bruges), and 'Avendauth' (= 'Davidiz'
in the Romance language); R. LEMAY, «Dans l'Espagne du xii c siecle: les traductions
de I'arabe au latin», Annales. Economies. Societes. Civilisation, 18 (1963), pp. 639-
-65 (648-50) and ID., «De la scolastique ~ I'histoire par le truchement de la philologie:
itineraire d'un medieviste entre Europe et Islam», in La diffusione delle scienze isla-
miche nel medio evo europeo, Convegno internazionale promosso daU' Accademia
[ 11]
231
v v
«Hispania» appears in another preface of «Iohannes Hispalensis».
In the same Edinburgh manuscript that gives the fullest and most
authentic form of the name of the translator of the De differentia, there
is the earliest extant copy of the translation of the short version of
Pseudo-Aristotle's Secret of Secrets. This text and the De differentia
(and no other texts) have been added in the same small hand, which
fills in gaps in the manuscript. 27 The translation of the Secret of Se-
crets is here attributed to «Iohannes Y spalensis». It might be worrying
that the addition «et Limiensis» (or «in Limia») is missing, 28 but, on
the other hand, the full version of the name tends to occur only in
colophons, and the colophon to the Secret of Secrets is absent. 29 What
is more significant is the title John gives to his dedicatee:
Domine T. gratia dei Hispaniarum regine, Ioh(anne)s Yspalensi(s)
sal(utem)
This «T» appears in other manuscripts as «Tarasia», 30 and scho-
lars have had little hesitation in identifying her with Teresa, who, as
we have seen, was proclaimed «queen» by the Portuguese after the
nazionale dei Lincei, Rome, 1987, pp. 399-535 (410-8). As Lemay himself points
out, Marie-Therese d'Alverny had shown that the names 'Johannes' and 'Avendauth'
are never brought together in the manuscripts (M.-T. 0' ALVERNY, «Avendauth?» in
Homenaje a MilMs- Vallicrosa, I, Barcelona, 1954, pp. 19-43). Whether 'John David' ,
sometimes called 'of Toledo' is 'John of Seville (and Limia)' remains unproven.
27 See p. 226 [6] above.
28 0' ALVERNY (<<Conclusion», p. 135) is wrong in implying that the epithet
'Limiensis' is found in the Secretum secretorum text also. See Plate.
29 One might add that if, as I suspect, the scribe copied the Secret of Secrets
and the De differentia from the same source, he (or his source) might have felt it
unnecessary to repeaPthe full name of the translator. The variation 'Hispalensis'l
/,Yspalensis' need cause no anxiety, especially since the scribe shows traits of an
(Iberian?) uncertainty of whether to include oh's or not: see Appendix below for
examples in the preface to the Secret ofSecrets, and the spellings 'onoret' and 'ypocra-
tis' in the opening of the text of De differentia.
30 S. 1. WILLlAMS, The Scholarly Career of the Pseudo-Aristotelian Secretum
secretorum in the Thirteenth and Early Fourteenth Century, Ph. D., Northwestern
University, Evanston, Illinois, 1991, pp. 12-13 and 183. The earliest reference to
'Tharesia' being the dedicatee is in Thomas of Cantimpre, De natura rerum (ca.
1230), ed. H. BOEsE, Berlin, 1973, p. 77 with n. 24.
232 [12]
«MAGISTER IOHANNES HISPALENSIS ET LlMIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRrruS ET ANlMAE
death of her husband. For there was no other «Spanish» queen with
this name (or even with the initial «T») during this period. But, to
address someone as the «queen of the Spains» is still contentious at
this time, when the royal families of Le6n-Castile and Aragon-Navarre
were also seeking supremacy in the Iberian peninsula. 31 It suggests
that the dedicatee is pandering to the pretentions to power of the for-
mer countess Teresa, at the very time when the archbishop of Bra-
ga was proclaiming himself pope. We learn from the preface to the
Secret of Secrets, in the Edinburgh manuscript, that John and Queen
Teresa had been talking about «utilitas corporis», 32 but that he was
not a doctor himself. He describes his method of translation, in which
he errs on the side of literalness, because he is not sure of his compe-
tence in the subject-matter.
Another medical fragment may be attributable to this «John». For,
in a Vienna manuscript of the fourteenth or fifteenth century, a cure
for the disease of the kidneys is said to have been sent from John
of Seville to «Pope Gregory». This could be Maurice Bourdin him-
self, who, as we have seen, set himself up as pope Gregory VIII. The
recipe is astrological, suggesting either (a) the carving of the figure
of a lion on a sigil of pure gold when the Sun is in Leo and the other
planets are in specified places, and binding that sigil to the loins oppo-
site the kidneys, or (b) stamping the image in frankincense and giving
it to the patient to drink. This recipe would appear to be taken from
the text on talismans attributed to Hermes, which it is very likely that
John would have been interested in. 33
31 From the account in the Chronica Adefonsi Imperatoris (n. 16 above) it is
quite possible that the Portuguese too were calling Teresa 'queen of the Spains'. For
the expression 'of the Spains' (which is often interchangeable with 'of Spain' and
'of the Spanish'), see 1. A. MARAvALL, El concepto de Espaiia en la Edad Media,
2nd ed., Madrid, 1964, pp. 63-5 and 41t.
32 THORNDIKE, <dohn of Seville», p. 25, interprets this as 'human physiology'.
The preface is edited and translated in Appendix I.
33 The recipe is found in Vienna, 6sterreichische Nationalbibliothek, 5311,
fo\' 41 vb: 'Mirabilis cura contra malum calculi vellapidis vel contra malum yliorum
Hermetis, quam misit magister Johannes Ispalensis Gregorio Pape patienti id malum' .
Thorndike's suggestion ('John of Seville', p. 27) that the anti-pope Gregory VIII is
the recipient is turned into a certainty by LEMA Y (<<Dans l'Espagne du XIIc siecle:
[ 13] 233
v v
One more piece of evidence must be considered in reconstruc-
ting what is known about «lohannes Hispalensis et Limiensis». This
is to be found in one early manuscript of the De differentia which
has not been taken into account by anyone who has written on the
text of the De differentia, and is not included either in Aristoteles La-
tinus, or in Wilcox's list of manuscripts. The manuscript is Medieval
MS no. 1 of the GPA Bolton Library of Cashel, Tipperary. It consists
of several booklets, probably put together in the abbey of Tewkesbury
(near Gloucester) in the late twelfth or early thirteenth century. 34 The
De differentia shares a booklet with a text on Arabic arithmetic by
a pupil of Adelard of Bath called Ocreatus, and a fragment of a
commentary on Porphyry's Isagoge. Since the Helcep sarracenicum
of Ocreatus occurs in only one other, inferior, manuscript and the
commentary appears to be unique, there is reason to believe that these
copies of the two texts are close in date to their composition.
The text in the Cashel manuscript is almost exactly that of the
twelfth-century Brussels manuscript that Wilcox chose as her base
manuscript. It omits the same passages in the prologue and inclu-
des the full account of the history of the work in its rubric. But here
it differs in one important respect from the Brussels manuscript. It
reads:
Incipit liber differentie inter animam et spiritum. Constabe Luce
cuidam amico suo scriptori cuiusdam regis eddit (sic) et Ioh(anne)s
ispoliisis (sic) episcopus ex Arabico in Latinum Raimundo Tole-
tano archiepiscopo transtulit. 35
les traductions de l'arabe au latin», p. 652). The 'lion sigil' was well-known in the
Middle Ages; it was referred to by Pietro d' Abano and Arnold of Villanova, and
appears (a) in the text of Hermes's De imaginibus incorporated into the Latin trans-
lation of the magical handbook Picatrix (ed. D. PINGREE, London, 1986, p. 82), and
(b) in the Liber imaginum signorum Hermetis cum additionibus magistri Arnaldi
de Villanova, in MS Vatican, lat. 4082, fols 213v-214r; see D. PINGREE, «The Diffu-
sion of Arabic Magical Texts in Western Europe», in La diffusione delle scienze
islamiche nel medio evo europeo, Convegno internazionale promosso dall' Accade-
mia nazionale dei Lincei, Rome, 1987, pp. 57-102 (91).
304 This provenance is indicated by the calendar in one of the booklets, which
shows the closest parallels to that of Tewkesbury Abbey (OSB). I owe this informa-
tion to Professor Nigel Morgan.
3~ See Plate.
[14]
234
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
No other manuscript of the De differentia (on the evidence of Aristo-
teles Latinus. Codices and Wilcox's dissertation) mentions that this
John was a bishop. There are, however, at least two manuscripts of *
cUmar's Liber universus attributed to «Iohannes Hispalensis et
Limiensis» which do call him a bishop:
1) British Library, Harley 3731, fo!' 81v: «perfectus est univer-
sus liber hacmar beniganu (?) tiberiadis cum laude Dei et eius auxi-
lio quem transtulit magister Iohannes Hispalensis et Lunensis epc (=
episcopus) ex Arabico in Latinum».
2) Erfurt, Amplonian, Q. 365, fo!' 119: «quem transtulit magis-
ter Iohannes Hyspalensis atque Lunensis episcopus ex Arabico in
Latinum». 36
That two completely different works should designate John as a
bishop means that we have to take the epithet seriously. Is it possi-
ble that «Iohannes Hispalensis et Limienses» was a bishop? Approa-
ching this question from a completely different (i.e., Arabic) direc-
tion, and without the knowledge of the Latin manuscripts, P. Sj. Van
Koningsveld made exactly the same suggestion. For he noticed that
«the only report on commenting activities in Arabic concerning Latin
texts» occurred in the History of Spain of Rodrigo Jimenez who des-
cribeda certain Mozarabic bishop of Seville in the following terms:
Et in isto medio fuit apud Hispalim gloriosus et sanctissimus Ioan-
nes Episcopus, qui ab Arabibus Caeit almatran vocabatur, et magna
scientia in lingua Arabica c1aruit, multis miraculorum operatio-
nibus gloriosus effulsit, qui etiam sacras scripturas catholicis expo-
sitionibus dec1aravit, quas in formationem posterorum Arabice
conscriptas reliquit. 37
«Between [those two dates] there was in Seville the glorious and
most saintly bishop John, who was called Sayyid al-matran [the Ara-
36 ALONso, «.Juan Sevillano», pp. 41-2. This must be the manuscript referred to
erroneously as 'of Omar on nativities' that Thorndike mentions as adding 'epyscopus'
to 'Iohannes Hyspalensis atque Lunensis': see 'John of Seville', p. 23 and n. 14 above,
37 P. Sj. V AN KONINGSVELD, The Latin-Arabic Glossary of the Leiden Univer-
sity Library, Leiden, 1977, pp. 51-2.
[15] 235

v
bic honorific title for «Metropolitan»] and was well-known for his
great knowledge in the Arabic tongue; he shone out, glorious for
working many miracles. He also explained the sacred scriptures with
orthodox commentaries, which he left written in Arabic for the
education of posterity».
Van Koningsveld identified this bishop with the archbishop of
Seville at the time of the submission of the city to the Almohads, in
1148, who fled to Talavera and died there. 38 He would be one of
the bishops that Ibn cAbdun in his description of Seville in the early
twelfth century complained about (in our only Arabic reference to
Christian translations in Spain), when he warned Muslims that they
«should not sell to the Jews or Christians books concerning scien-
ce ... [because] they translate them and attribute them to their co-
-religionists and their bishops». 39 It is difficult, however, to equate
the «Iohannes Hispalensis et Limiensis» who may have come to
«Hispanae partes» from elsewhere with a Mozarabic bishop· who
presumably was born and brought up in an area of Muslim domina-
tion. The epithet «episcopus» (if it represents a function and not a
surname 40) remains a mystery.
«Magister Iohannes Hispalensis et Limiensis» therefore emerges
as a translator to whom at least five medical and astrological works
are attributed. To confirm that the same author was responsible for
all these translations would need further investigation of the texts
38 Richard Lemay did not consider this possibility, but suggested rather that the
translator John of Seville began to call himself 'John David' (or 'Avendauth') after
the bishop had fled to Christian Spain, in order to distinguish himself from the bishop
'John of Seville'; see his «Dans l'Espagne du XIIc siecle: les traductions de I'arabe
au latin», p. 660, n. 1. The suggestion that John of Seville was the bishop of Segovia
(1149-52) who succeeded Raymond as archbishop of Toledo, and held the see from
1152 to 1166, proposed by ALONSO (<<Notas sobre los traductores toledanos Domingo
Gundisalvo y Juan Hispano», AI-Andalus, 8, 1943, pp. ] 55-88 [174-7]) and aired
again by Lemay (ibid.), must now be viewed in the light of our knowledge that arch-
bishop John, like his two predecessors in Toledo, was a Frenchman - Jean de
Castelmoron-sur-Lot.
39E. L~VI-PROVEN<:AL, Seville musulmane au debut du xiie siecle, Paris, 1947,
section 206 (p. 128).
40 A 'Richard Bishop' was a teacher of John of Salisbury (see p. 250 [30] below).
236 [16]
v
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIW.E
themselves, which in turn would require producing reliable editions
of the Latin texts and their Arabic originals (where they exist) and
the compilation of Arabic-Latin glosses. 41 A few provisional obser-
vations, however, may be made. In the preface to the Secret of Se-
crets, John states that he was discussing «utilitas corporiS» with
the Queen of the Spains. The exact meaning of this phrase is unclear,
but it is striking that amongst the sources listed at the beginning of
the De differentia, Galen' s book De usu partium is referred to as «in
utilitate membrorum». 42 Thus we. see the use of the same term -
«utilitas» - and the possible knowledge of the same text, on the part
of the translators of both the De differentia and the Secret of Secrets.
One can add that the Secret of Secrets, like the De differentia often
41 Of all the texts concerned only the De differentia has received an edition that
reaches a sufficient standard (that of Wilcox). Wilcox has compared the text with
the Arabic MS Gotha 1158 as edited by G. GABRIELI - «La risalah di Qus~a b. Liiqa.
'sulla differenza tra 10 spirito e l'anima'», Rendiconti della Reale Accademia dei Lin-
ceL Classe di scienze morali, storiche e fliologiche, Ser. 5, 19 (1910), pp. 622-55;
another manuscript, from the Khalidi Library in Jerusalem, was edited by L. CHEIKHO
in Al-Machriq, 14 (1911), pp. 94-109; a third, Istanbul, Ahmed IlI, 3447 by H. Z.
OLKEN, Ibn Sina Risaleler, 2, Istanbul, 1953, pp. 84-100. All these editions were to
be replaced by a new edition from all the manuscripts by Andre d' Alvemy, which
was left unfinished at his death: see M. T. D' ALVERNY, «Les traductions d' Aristote
et ses commentateurs», Revue de synthese, 3rd series, 49-52 (1968), pp. 125-44 (142),
where two readings from this edition are given: '~iyagha' ('jewelry') for '~ina'a'
('skill', line 380) and the root 'b.d.n.' for 'Phaedo' (line 9). The editions of Thabit's
De imaginibus (n. 19 above) and al-Farghani's Rudimenta (De scientia astrorum,
Berkeley, California, 1943) by F. 1. CARMODY need revising. The editions of the short
version of the Secret of Secrets by SUCHIER (in Denkmiiler Provenzialischer Litera-
tur und Sprache, Halle, 1883) and J. BRINKMANN (Die apocryphen Gesundheitsregeln
des Aristoteles fUr Alexander den Grossen in der Uebersetzung des Johann von Tole-
do, Leipzig, ]914) need replacing, and A. R. BADAWI'S edition of the Arabic text
in Fontes graecae doctrinarum politicarum Islamicarum, Islamica 15, Cairo, 1954,
needs supplementing in the light of the considerable amount of new information we
now have concerning the short and long versions in Arabic (see the studies in Pseudo-
-Aristotle. The Secret of Secrets. Sources and Influences, eds W. F. RYAN and C.
B. SCHMITI, London, 1982). The works of Masha'allah and 'Umar ibn al-Farrukhan
al-Tabari have not been edited in modem editions.
• 42 The early translation from the Arabic was called De iuvamentis mem-
brorum.
237
[17]

v
accompanied Aristotle's Libri naturales, and received a scholastic
commentary from Roger Bacon. 43
Then, there are the seeming additions from Aristotle's De ani-
ma in both the De differentia and Thabit's De imaginibus. Wilcox
points out that the second of the two definitions of the soul attribu-
ted to Aristotle in the Latin text of the De differentia is omitted in
Gabrieli's Arabic text. This is the one introduced by «In libro autem
Aristotelis quem fecit de ani ma, talis est diffinitio», and Wilcox claims
that since this definition is followed by the words «Redeamus ad opus»
and, indeed, by a return to the Arabic text, the definition must be an
addition by the translator. 44 In fact, the situation is not as clear-cut
as this, since the first definition in the Latin text, which is the same
as the single definition given at the beginning of the second part of
the De differentia, does not correspond to any definition in the three
Arabic editions, nor to Aristotle (the discordant words are «agentis
et» in place of «naturalis instrumentalis» 45). The second definition
in the Latin text, however, combines two definitions of Aristotle, which
are kept separate in all three editions of the Arabic text of the De
differentia. The Latin text reads: «anima est perfectio corporis natu-
ralis instrumental is viventis potentialiter» (<<the soul is the perfection
of a natural body which has organs and has life potentially»). Aristotle
had written in his De anima, that «the soul is the first perfection
(<<entelekheia») of a natural body which has life potentially» (412a27-
-8) and that «if one has to give a definition which applies to all souls
universally, it would be the first perfection of a natural body which
has organs» (412b5-6). These words of Aristotle are accurately re-
produced at the beginning of the second part of the Arabic text of
the De differentia, where we read: «Aristotle defines the soul as the
4~ See WILLlAMS, «The Scholarly Career», pp. 146-52, giving a list of 'Manus-
cripts [of John of Seville's version] of Scholarly Provenance'.
44 WILCOX, The Transmission, p. 236, note 23, referring to lines 404-9. The text
is reproduced below in Appendix 11.
..~ It is difficult to see whether 'agentis' is an alternative translation for 'I)ayy'
(living) - hence giving the doublet 'agentis et viventis' and explaining the presence
of 'et' - or meant to be an equivalent for 'fabjCj' (,natural') or "ali' ('having organs').
The first explanation is more likely, given John's propensity for using doublets (see
Appendix 11 below).
238
v
1 «MAGISTER IOHANNES HISPALENSIS ET L1MlENSlS» AND QUSTA lBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
«first 46 perfection ofthe natural body possessing organs» or, in another
definition different from the first, he writes 'the definition of the soul
is that it is has life potentially'». That this is Qusta's original text,
and that John's version represents a distortion of it, is indicated by
the fact that, at the very end of the discussion of Aristotle's definition
of the soul Qusta claims that «having organs» and «having life po-
tentially» are synonymous 47 and that the two definitions are in fact
one. John's text repeats this statement (lines 476-80 and 486-7) which
no longer makes sense since in his two definitions the differentiating
feature is not between «instrumentalis» («having organs») and «viven-
tis potential iter» (<<having life potentially»), but rather between «agen-
tis» and «naturalis instrumentalis». 48 John's combination of the two
definitions is not, then, in the Arabic manuscripts of the De differentia
that have been edited, and could (as the words «redeamus ad opus»
suggest), be his addition. But he did not take the combined defini-
tion directly from the De anima, but could rather have been following
both Arabic and Latin authors who had already combined the two defi-
nitions in exactly the same way as John does. 49
The addition of a reference to the De anima (even though it is
inaccurate) occurs also, in a similar way, at the beginning of the De
imaginibus. We have the beginning of one version of the Arabic text
in the Picatrix. It reads as follows:
46'First' is missing in Gabrieli's and Ulken's editions, but present in Cheikho's.
47This statement (equivalent to the Latin lines 486-7) is in Cheikho's and Olken's
editions.
48 The confusion here is manifest also in the obscure wording of John's text:
'Quod autem videtur corrumpere primam particulam et in loco eius ponere 'viven-
tis potentialiter', non multum discordat ab alia, quia interpretatio utrarumque diffi-
nitionum est una' (lines 476-80).
49 Cf. Qus~'s own translation of a Greek doxography: Aetius Arabus, Placita
philosophorum, IV,2,6, ed. H. DAIBER, Wiesbaden, ]980, pp. 190-1: 'Aristotle belie-
ved that the soul was the first perfection of a natural body having organs and living
potentially; by the word 'perfection' he means the thing which actually ('ftcl an ') is';
Calcidius's rendering of Aristotle's definition is «prima perfectio corporis naturalis
organici possibilitate vitam habentis»: Plato, Timaeus, a Calcidio translatus com-
mentarioque instructus, eds P. J. JENSEN and J, H. WASZINK, London, 1962, c. ]22
(p. 235.8-9).
[19] 239
[ 18]

v
«Thabit ibn Qurra said in the book he composed on talismans: 'The
1 «MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRrruS ET ANIW.E
noblest part of the science of the stars is the science of talismans' ,
and he adds: 'No body has 1ife which lacks spirit'». 50
Adelard's translation gives only: «Whoever is skilled in geometry
and philosophy, but has no experience of the science of the stars, is
useless. For, of all the arts, the science of the stars is most excellent
in fact and most useful because of the effect of talismans». 51 John's
translation appears to take up the first sentence in his Arabic origi-
nal and amplify it by a reference, precisely, to Aristotle's De anima:
«Thabit said that Aristotle said that he who has read philosophy
and geometry and every science, but is ignorant of the science of
the stars, is unable to do anything, because the science of talismans
is more worthy than geometry and higher than philosophy. 2 The
Philosopher (Aristotle) said in the second treatise of his book [i.e.,
the De animal that, just as there is no movement for a body that
lacks a soul, nor life to an animate body except through food which
is digested and suited to the body's nature, 52 so there is no light
of wisdom when the science of the stars has been left out. 3 And
just as the spirit cannot live except by the food which is suited
to the body's nature, so there is no root of wisdom for him who
lacks philosophy (or the science of the stars), nor is there the light
of geometry when he lacks the science of the stars; and the height
and summit of the science of the stars is the science of talis-
mans.» S3
50 Picatrix, I.v.36 (ed. PINGREE, p. 23): « ... Thebit ben Corat in libro quem com-
posuit De ymaginibus, qui sic ait: sciencia ymaginum est nobilior pars astronomie.
Et subdit: corpus caret vita deficiente spiritu».
51 MS Lyon, Bibliotheque Municipale, 328, fol. 70r: «Quicumque geometria atque
philosophia peritus, astronomie expers fuerit, ociosus est. Est enim astronomia omnium
artium et re excellentissima et prestigiorum effectu commodissima».
s2This statement has only a general equivalent in the first chapter of the second
book of Aristotle's De anima.
5l De imaginibus, 1-2, ed. CARMODY, p. 180: «Dixit Thebit Bencorah: Dixit Aris-
toteles: Qui philosophiam et geometriam omnemque scientiam legerit et ab astrono-
mia vacuus fuerit, erit occupatus et vacuus, quia dignior geometria et altior philoso-
phia est imaginum scientia. 2 Et iam dixit Philosophus (v.1. + Aristoteles) in secundo
tractatu sui libri quia sicut non est motus corpori anima carenti nec vita animato cor-
240 [20]
v
In the Picatrix the meaning of «body without spirit» is explained
in necromantic terms: talismans which are not fit for receiving the
spirits of the planets are similar to dead bodies in which there is no
«spirit». 54 John would seem to equate the soul of the philosophers
with the «spirits» of the necromancers and alchemists. As has been
suggested above, 55 John may have deliberately left out the necroman-
tic elements of Thabit' s treatise. If he was also adding what he thought
were the words of Aristotle, 56 then he proves not to be as literal a
scholar as «John of Seville» has always been held to be.
John's translation of the Secret of Secrets would presumably have
been made between 1112, when Teresa took over the rulership of the
county of Portugal from her deceased husband, and 1128, when she
was imprisoned by her son. His translation of the De differentia shows
a different allegiance, since it is dedicated to Raymond de La Sau-
vetat, archbishop of Toledo from 1125-52, and, as we shall see be-
low, probably dates to before 1143. Three other translations were ma-
de «in Limia» including the translation of al-FarghanI's Rudimenta,
which is precisely dated to 11 March, 1135. John had discussed the
«utilitas corporis» and may have advised the anti-Pope Gregory VIII
on a cure for kidney-stones, but, in the preface to the Secret of Secrets,
pori nisi per cibum quo diriguntur (v.1. digeruntur) et aptantur eius nature, ita non
est lumen sapientie cum astronomia evacuata fuerit. 3 Et quemadmodum spiritus non
poterit subsistere nisi per cibum quo aptantur nature corporis, ita non est radix scien-
tie apud eum qui philosophia caruerit (v,I. apud eum qui caruerit astronomia) nec
est lumen geometrie cum vacua fuerit astronomia; sublimitas autem et altitudo astro-
nomie est imaginum scientia,» (Carmody's Version I, with variant readings from
Version 1).
54 Picatrix, I.v.36 (ed, Pingree, p. 23): «Et hoc dixit (Thebit) propter ymagines
que non fiunt temporibus congruis et opportunis, que non erant apte ad recipiendum
spiritus planetarum et tunc erunt similes corporibus mortuis in quibus non est spiritus.
Et quando fiunt congruis temporibus debitis et opportunis, recipiunt spiritus et infu-
si ones virium planetarum et erunt similes corporibus viventibus ex quibus postea sequ-
untur mirabiles effectus.»
55 See n. 23 above.
56The one other possible addition to Qus~a's text made by John is the referen-
ces to Empedocles's De anima (see n. 68 below). Wilcox refers to several other addi-
tions vis-a-vis Gabrieli's Arabic text, but these almost invariably are present in
Cheikho's and Ulken's texts.
241
[21]
v
he denies that he is a doctor. In the preface to the De imaginibus (if
we take this to be genuine) he implies, rather, that his main interest
is in the science of the stars, for which he had sought a book or books
in «Hispanae partes», an area which included Muslim territory. Whether
the large number of astrological texts attributed to «John of SevilIe»
(without the further epithet «and Limia») are also translations by
«Iohannes Hispalensis et Limiensis» remains to be investigated. 57 It
appears, however, that none of these other texts refer to their trans-
lator as «episcopus» or imply any connection with Portugal. Nor did
«John of Seville» tackle any further texts on medical matters.
* a substance, in that it can receive contrarIes. Gundlssahnus substItu-
As we have seen, the De differentia was addressed to Raymond,
archbishop of Toledo. The fact that the text was in Toledo is testified
by its use by an archdeacon of Segovia resident in Toledo in the time
of Raymond's successor, John, archbishop from 1152 until 1166. 51!
This archdeacon, Dominicus Gundissalinus, had collaborated with a
certain «A vendauth» in translating the De anima portion of the Shifii)
of Avicenna, which the two translators dedicated to archbishop John.
Some time after this translation, Gundissalinus wrote an original work
on the soul in which he made his starting point and main source the
De anima of A vicenna. 59 However, in addition to A vicenna, he used,
without acknowledgement, the De differentia.
51 The starting-point of this investigation should be a thorough and scientific
investigation of the language and style of the translations, preferably based on good
editions furnished with Arabic-Latin glossaries. Laurenzo Minio-Paluello's detective
work on the identity of the translators of Aristotle's works from Greek into Latin
in the Middle Ages provides an excellent model.
58 Another Toledan connection could be provided by the use of Qusta's work
in the Dialogus of Petrus Alfonsi who could have been in Toledo in the second quar-
ter of the twelfth century. That Petrus was indebted to John's translation rather than
the Arabic original is argued in BURNElT, «The Works of Petrus Alfonsi: Questions
of Authenticity», in Pedro Alfonso, ed. J.-M. LAcARRA, Zaragoza, 1996.
59]. T. M UCKLE, «The Treatise De Anima of Dominicus Gundissalinus», Mediae-
val Studies, 2 (1940), pp. 23-103. The parallels between Gundissalinus's text and
the De differentia have been pointed out by Muckle and Wilcox (op. cit., pp. 102-
-3), but are set forth in full in Appendix 11 below.
242 [22]
v
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS F:T ANIMAE
Gundissalinus seems only to use the second part of Qusta's work
_ that on the soul - but draws material from almost the whole of
this part. It is clear that the text he knows is the more common version
(called by Wilcox the «John of Seville» version), rather than the
abbreviated version. However, Gundissalinus does not simply excerpt
from this version; rather, he adapts the material to his other sources,
and fits it into the framework of his own treatise.
The bases for the discussion of «What is the soul?» for both Qusta
and Gundissalinus (chapter 2) are the definitions of Plato and Aris-
totle each word of which is explained in turn. In the case of Plato's
defi~ition - that the soul is an incorporeal substance moving the bo-
dy _ Gundissalinus, following Qusta, f~rst show~ ho~ the sou~ is
tes «contraria» for Qusta's «opposita», and adds to the moral oppo-
sites of Qusta (virtues and vices), aesthetic and intellectual opposites
(joy and sadness, and knowledge and ignorance). Then he shows that
the soul is not body, taking three out of Qusta's five arguments and
reversing the order of the questions of whether t~e. s.oul can be. inani-
mate or animate body. The last words in the defImtton (<<movlOg the
body») Gundissalinus has already discussed in a separate section of
his previous chapter (<<How the soul moves the body»), using Qusta's
arguments. In the latter section Avicenna' s discussion of ~het~er the
soul moves is the starting point, 60 and this has led GundIssahnus to
distort Qusta's arguments. For Qusta explores the different ways in
which anything can be moved by a non-mover, as .a preliminary .to
discovering how body can be moved by the non-movlOg soul; Gun~Is­
salinus makes the soul the subject of the different ways of movlOg
and thus confuses the issue.
In exploring Aristotle's definition Gundissalinus omits ~ll men-
tions of the troublesome first definition in the Latin text (hne 310:
«Anima est perfectio corporis agentis et viventis potentialiter») and
only gives the second definition: that soul is the perfection of the natu-
ral body which has organs and has life potentially. However, Gun-
60 «Si autem anima movetur dum movet, tunc aliquo sex motuum movetur ... »
(ed. MUCKLE, p. 33, lines 31 et seqq.): this passage comes from Avicenna, De alli-
ma, 1.2.
243
[23]
v ;
I
I
dissalinus adds «first» to «perfection». This may be because of Qusta's
later specification (repeated by Gundissalinus) that by «perfection»
is meant «first» rather than «second» perfection, but may also reflect
Calcidius's reporting of Aristotle's definition. 61 When it comes to
explaining what kind of «natural body» is involved, Qusta differen-
tiates between simple bodies and composite bodies, and defines sim-
ple bodies as one of the four elements; these are not ensouled. Gun-
dissalinus, aware of a fifth element, perhaps through knowing the
pseudo-Avicennan De caelo et mundo, adds to the simple bodies
«another which is not an element or made from the elements, such
as a celestial body», but then has to qualify this statement by saying
that Plato believes that these bodies are ensouled. Finally, Gundissali-
nus tries to make sense of John's confused account of the equivalence
of «having life potentially» and «possessing organs» (lines 480-7) by
dropping the reference to «two definitions» and simplifying the
argument.
The fact that such prominence is given to the De differentia in
Gundissalinus's De anima is remarkable. For Plato and Aristotle's
definitions of the soul he could have made more use of Calcidius' s
commentary on the Timaeus, or of Nemesius' s Premnon physicon, both
of which were widely read in the twelfth century. 62 These texts he
seems to have neglected. The De differentia may have been chosen
instead because it so clear and logical: the arguments are developed
step by step and there is nothing superfluous. Moreover, Qusta was
well served by his translator. But another reason why he turned to
l'il Plato, Timaeus, a Calcidio translatus commentarioque instructus, c. 122 (p.
235.8-9): «prima perfectio corporis naturalis organici possibilitate vitam habentis».
The definition including 'first perfection' (but without 'having life potentially') is
also in Avicenna, I.v, in Avicenna Latinus. Liber de anima, ed. S. VAN RIET, 2 vols,
Louvain and Leiden, 1968-72, I, p. 80.1: «perfectio prima corporis naturalis instru-
mentalis». For a convenient list of definitions of the soul available in Latin in the
twelfth and early thirteenth century see D. A. CALLUS, «The Treatise of John Blund
On the Soul», in Autour d'Aristote, offert a A. Mansion, Louvain, 1955, pp. 490-
-1, to which may be added the definition of Isaac, De definitionibus, ed. 1. T. MUCKlE,
in Archives d'histoire doctrinale et litteraire du moyen age, 11 (1937-8), p. 312.
62Plato, Timaeus, c. 126 (p. 241.8-9) and c. 122 (p. 235.8-9); Nemesius, Prem-
non physicon, ed. C. BURKHARD, Leipzig, 1917, c. 2, p. 24.
[24]
244
v
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN lUQA"S DE DIFFERENTIA SPIRITUS ET ANIMAE
the De differentia may have been that the text had already been used
by a scholar in Spain whose work was well-known to Gundissalinus.
This scholar is Hermann of Carinthia, who completed his cos-
mological masterpiece, the De essentiis, at Beziers in 1143. In this
work, Hermann, like Gundissalinus, gives Plato and Aristotle's defi-
nitions of the soul as they were reported in the De differentia. 63 That
the source was Qusta in John's translation is indicated not only by
the nearly exact reproduction of the definitions (Hermann replaces
«potentialiter» by «potentia»), but also by his indication of the source
of Plato's definition: «Plato .. .in Cadone». Here we see the perpe-
tuation of John's reading of «~ » (<<f») as « J » (<<q») - a con-
fusion that can arise very easily in the maghribi (Western) form of
Arabic script. It appears, however, that Hermann added information
from the De differentia after the first draft of the De essentiis, because
the whole section reproduced in Appendix 11 below is not in the earliest
version of Hermann's text. 64
Hermann adapts Qusta's definitions to his own thought on the
soul in a much more radical way than does Gundissalinus. He labels
Plato's statement as a «definition», and Aristotle's as a «description»,
in line with his assertion earlier in the De essentiis that «a definition
applies to species and genera; a description only to individuals». 65
Aristotle's statement can only be a description because its object is
not a substance (and therefore cannot be a species), but rather a power
( «v irtu s» ) of the anima mundi. Plato's statement, on the other hand,
is a definition, but only of the human soul which alone is a substance
and incorporeal.
I'iJThat the definitions of the soul in the De essentiis come from Qus~'s work
was noted by M. ALONSO (<<Traducciones de Juan Hispano», p. 139) andT. SllVERSTEIN,
«Hermann of Carinthia and Greek», in Medioevo e Rinascimento: Studi in onore di
Bruno Nardi, Florence, 1955, pp. 688-92. The parallels are given in Appendix 11 below.
I'i4 This version is a twelfth-century fragment of the text in the private collec-
tion of Marvin COlKER who describes it in «A Newly Discovered Manuscript of
Hermann of Carinthia's De essentiis», in Revue d'histoire des textes, 18 (1988), pp.
213-28.
I'iS De essentiis, 62vE, ed. C. BURNETT, Leiden and Cologne, 1982, p. 108. The
difference between 'definition' and 'description' is spelt out in Isaac, De definitio-
nibus. ed. MUCKLE. p. 300 and passim.
[25]
245
 .1
v - ,
There is another connection between the De essentiis and De
differentia which has, up to now, not been noticed. Just before the
section which draws on the De differentia, Hermann writes:
Nec mirum quod ipse altissimus omnium Auctor proprie dignitati
conformare voluit. Unde nec ex aliena materia, nec laborante
repugnantium nexu, sed quemadmodum Platoni visum est et a
Pantocle presertim enodatum est, ne semel natum, umquam occidat
et nisi proprio arbitrio degeneret, post hoc exercitium ad origina-
lem dignitatem rediturum. 66
This passage only makes sense if «Pantocle» is regarded as a
corruption of «Empedocles», and the passage is translated:
«Nor is it surprising that the very Author of all things wished it
[the soul] to conform to its proper worth. Hence it is not from alien
material nor from a straining bond of warring elements, but as Plato
thought and especially as was explained by Empedocles _ <such
that>, once born, it should never die, and unless it degenerated
by its own will, after its struggle in this world, it should return
to its original worth.»
This doctrine is not Qusta' s, but must come from some work
referring to Empedoc1es's opinions extant in Muslim Spain, such as
The Book of the Five Essences, the Aim of the Wiseman (translated
in the late thirteenth century as Picatrix), or, most likely, the Theologia
Aristotelis. 67 In John's translation of the De differentia, however,
66 Ibid., 71 vD, p. 174.
67 The first two Empedoclean sources are diSCUSSed by A. NAGY in «Di alcu-
ni scriUi attributi ad Empedocle», Rendiconti delta Reale Accademia dei Lincei.
Classe di scienze morali, storiche e filologiche, Ser. 5, 10, pp. 307-20 and 325-44,
and M. AStN PALACIOS, Ibn Masa"a y su escuela, Madrid, 1914, and summarised
~y WILCOX, ~he Transmission, pp. 64-76. For the Theologia Aristotelis _ i.e., por-
tIOns of Platmus, Enneads IV and V in Arabic translation - see in particular 1.30-
-34: «Empedocles says that the souls were in the high and sublime place, and when
they erred they fell into this world ... (31) God ... call[ed] men ... to go back to their
own Original world, the high and sublime ... (34) The sublime and divine Plato has
described the soul and ... has mentioned in many places how the soul descends and
enters this world and that she will surely return to her own world, the true, the first
246
[26]
v
«MAGISTER IOHANNES HISPALENSIS ET LlMIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMllE
«Empedoc1es on the soul» is added to the end of the list of authorities
given by Qusta at the beginning of the text. 68 This addition could
have been made either in the maghribi manuscript used by John, or
by the translator himself. In any case, the reference to the authority
of Empedoc1es in both the De essentiis and the De differentia is an
indication of the lively interest in the Presocratic Greek philoso-
pher in the Iberian peninsula in the twelfth century.
The name, too, in both authors, shows a similar process of
distortion, especially when one looks at the earliest manuscript of the
De differentia, that of Edinburgh. Here one finds the form «btnbtJi
in anima» in the list of authorities at the beginning of the text (fo1.
l04v). However, at the second reference to Empedocles (fo1. l06r;
at the beginning of the section on the soul), the name is spelt out more
fully: ~. In this manuscript, therefore, we catch in the act, as
it were, the assimilation of the letters «ch> to «d», which can happen
very easily in a Latin manuscript. Almost all the other manuscripts
of the De differentia show this second «d». 69 A form «Bendecliz»
could be the original translator's transliteration of an Arabic form
«Bandaklis» or «Banduklis» (<<e» and «0» [= Arabic «u»] easily being
confused in Latin script). The usual Arabic form of the name is
world» (translation from the Arabic by G. Lewis in Plotinus, Opera, n, eds P. HENRY
and H.-R. SCHWYZER, Paris and Brussels, 1959, pp. 227-9). I do not intend here to
go into the thorny problem of what was genuine Empedocles and what was Pseudo-
-Empedocles. The whole question has been reopened in a magisterial way by Peter
KINGSLEY (Ancient Philosophy, Mystery, and Magic: Empedocles and Pythagorean
Tradition, Oxford, 1995), to whose advice I am indebted. Suffice to say that genuine
Empedoclean theories were included in the so-called pseudo-Empedoclean works.
68 See Appendix 11 below. Empedocles appears in none of the Arabic editions
of the De differentia nor in the earliest manuscript of the 'vulgate' Latin text (the
Brussels MS used by Wilcox as her base manuscript).
69 NAGY, «Di a1cuni scritti», p. 315, gives the variants 'benededinis, bendedis,
bendedes, bendedis, bendedis, bendediis, bOdidis, b' nedediz, bn dicendis' and 'bideudis';
and only 'bendeclarum' and 'bendeclinis' with the 'cl'. Wilcox chose the reading
'benededis' (The Transmission, pp. 143 and 167). Aristoteles Latinus. Codices, I, p.
197 gives 'audidis' (Vat. Urb. Lat. 206, fol. 335r) and 'anbemdeclus' (Paris, BN,
lat. 6325, fols. I 67v). Wilcox's alleged 'Hermann of Carinthia' version omits al\
mention of Empedocles, perhaps because the name was unintelligible (hence, too,
this version substitutes or omits most Arabic terms).
[27] 247

v
«Anbaduklis», but, since the ductus of the «n» and «b» is exactly the
same in Arabic and alephs tend to be dropped or put in the wrong
places, a misreading or miscopying as «Banduklis» is not unlikely.
Hermann's «Pantocle» implies a nominative «Pantocles» which shows
the same transposion of «n» and «b/p», and the same ending as the
form in the De differentia. As for the change of voiced consonants
into unvoiced, this might have been influenced by Hermann' s attempt
to restore a Greek root in an Arabic word which he knows has an
ultimately Greek origin, as he commonly does. 70 There is no «p» in
Arabic, and Hermann may have guessed (wrongly, as it turns out) that
the root «pan/pant-» (= «all») lay behind the proper name he saw before
him. On the other hand, another text probably translated from Ara-
bic in Hermann's time, also gives a «pan» as the first syllable of
Empedocles' name, though its final syllable is different. 71
The use of the De differentia by Hermann of Carinthia proves
that it was already translated by 1143 (assuming always that this is
the date of the revised version of the De essentiis). How had Hermann
acquired a copy? We know his movements only between 1138 and
1143. He is not known to have visited Toledo, but he was working
somewhere in the valley of the Ebro with his colleague Robert of Ket-
ton, when Peter the Venerable, abbot of Cluny, met them and persua-
ded them to translate some texts on the Muslim religion, including
the Qur'an. One of these texts Hermann translated at Le6n. He was
not, therefore, far from the Limia region.
Hermann could, too, have been the conduit through which the De
differentia reached the archdeacon of Segovia. For Gundissalinus is
the only person known to have made use of Hermann's De essentiis,
and he might therefore have had a privileged access to Hermann' s
books. It is clear that neither Hermann nor Gundissalinus took the
quotations from the De differentia from each other's works; they quote
70 Examples are 'telesmatici' from Arabic '[a~~ab] aHilasmat' ('masters of talis-
mans'; originally 't£A£O'j.1a'ttlcol), and 'genezia' replacing Arabic 'mawaJid'
(,[astrological] nativities'; y£ve9Ata).
71 This is the Turba philosophorum (the relevant section does not survive in Ara-
bic) which includes 'Pandolfus' among its philosophers. This is clearly Empedocles,
but doctrines on the soul are not included in the extant Arabic or Latin portions of
this text; see KINGSLEY, Ancient Philosophy, chapter 5.
248 [28]
v
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DlFFERENfIA SPIRnvS ET ANIMAE
different phrases. However, as far as can be discerned, the ~ext they
used was the same, 72 and both used it without citing the ArabIc author
or the translator.
Hermann's «magister» was Thierry ofChartes who is well-known
for having brought together the most comprehensive collection of t~xts
on the Liberal Arts of his time in his Heptateuchon, probably complIed
in the 1130s in Paris. Hermann of Carinthia dedicated one of his
translations from Arabic (that of Ptolemy's Planisphere) to Thierry,
and recommended other Arabic-Latin translations in his preface. n
The Heptateuchon included two of the new translations - a version
of Euclid's Elements possibly by Robert of Ketton, 74 and the astrono-
mical tables of al-Khwarizmi in a revision probably made by Her-
mann himself. Moreover, the Heptateuchon is the earliest manus-
cript to include a text of the «new Aristotle»: Boethius's translation
of Aristotle's Topics.
The debate continues as to how, when and by whom the «new
Aristotle» was introduced into Europe. Several of Aristotle's logical
texts (the Logica nova) and Libri naturales had been translated in
the second quarter of the twelfth century by James of Venice (Poste-
rior Analytics. Physics, De anima. De memoria, De intelligentia and
the Metaphysics) and an anonymous translator (De gene ratione et
corruptione and parts of the Nicomachean Ethics [Ethica vetus and
nova]). 75 The earliest direct evidence for this is an entry that Robert
of Torigny, abbot of Mont Saint-Michel from 1154 until 1186, added
72 One distinctive feature shared by Hermann and Gundissalinus against the
common version published by Wilcox is the transposition of the words 'movens cor-
pus' in Plato's definition. . .
73 Hermann of Carinthia, De essentUs, pp. 4-10, and C. BURNETI, «Advertlsmg
the New Science of the Stars, c. 1120-50», in Le Xlle siecle: Mutations et renou-
veau en France dans la premiere moitie du XIIe siecle, ed. F. Gasparri, Paris, 1994.
pp. 147-58 (150-0.
74 Robert of Chester's (7) Redaction of Euclid's Elements, the so-called Ade-
lard n Version, eds H. L. L. BUSARD and M. FOLKERTS, 2 vols, Basel, Boston and
Berlin, 1992, I, pp. 22-31. 'Of Ketton' is the more correct designation of the 'Robert'
associated with Hermann.
7~ L. MINIO-PALUELLO, Opuscula: the Latin Aristotle, Amsterdam, 1972, pp. 189-
-228.
249
[29]
v v
to his Chronicle between 1157 and 1169, which mentions that James
of Venice translated and «commentatus est» (i.e., perhaps «revised»)
several of the texts of the Logica nova, including the Topics. Moreover,
in two of the manuscripts in Mont Saint-Michel are the earliest copies
of the De gene ratione et corruptione, Ethica vetus and nova, Meta-
physics, Physics, De intelligentia, De anima, De longitudine et brevi-
tate vitae and De memoria (now Avranches, Bibliotheque Municipale,
221 and 232). Probably in 1167 John of Salisbury asked his former
teacher, Richard Bishop, for copies of Aristotle's works that he had
in his possession. Richard Bishop had taught John in Paris but was
archdeacon of Coutances from 1163 to 1170, and bishop of A vranches
from 1170 until his death in 1181. He was a neighbour and friend
of the abbot of Mont Saint-Michel; so he could have been responsible
for bringing copies of the «new Aristotle» from Paris to Mont Saint-
Michel. 76
In one of the two Mont Saint-Michel manuscripts - Avranches,
232 - the De differentia is also included. 77 It is in the same fascicle
76 The story outlined in this paragraph is told by F. Bossier and J. Brams in
the introduction to their edition of the Physica, translatio vetus (Aristoteles Latinus,
VII. I , Leiden, 1990, pp. xxi-xxiii). They base their conclusions on the research of
Minio-Paluello, but end with the suggestion that Richard Bishop had passed his
manuscripts to his friend Robert ofTorigny. This suggestion also occurs in the thorough
study of the manuscripts of Aristotle's works from Mont Saint-Michel of Coloman
VIOLA, S. J.: «Aristote au Mont Saint-Michel», in MilLenaire monastique du Mont
Saint-Michel, 2: Vie Montoise et rayonnement intellectuel du Mont Saint-Michel
ed. R. Foreville, Paris, 1967, pp. 289-312; see also M.-T. D'ALVERNY, «Les nouveau~
apports dans les domaines de la science et de la pens~ au temps de Philippe Au-
guste: La philosophie», in La France de Philippe Auguste, Colloques CNRS, Paris,
1982, pp. 863-80 (870-3).
17 For a detailed description of this MS see Yii~anna ibn Masawayh (Jean Me-
sue), Le livre des axiomes medicaux (Aphorism i), eds D. JACQUART and G. TROUPEAU,
Geneva, 1980, pp. 48-52. Different scholars have proposed different dates for the
fascicle that includes the De differentia and Ethica nova (e.g., D' ALVERNY, «Nouveaux
e
apports», p. 871 [fin du xii siecle], Wilcox, p. 138 [late twelfth or early thirteenth
cent.], Aristoteles Latinus. Codices, no. 408 [s.xiii ift ], Jacquart and Troupeau, p. 52
[so xiii]). In any case, there is no other twelfth-century copy of the Ethica nova.
Rene Gauthier's argument that the Ethica nova was translated by a different person
(Michael Scot?) from the Ethica vetus and De generatione et corruptione (Ethica
250
[30]
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LljQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
as the Ethica nova and is written in the same hand. This fascicle was
bound together with the De generatione et corruptione, Ethica vetus,
Physics, De memoria, De longitudine et brevitate vitae and Metaphy-
sica vetustissima, at an early date. Here we see, for the first time, the
De differentia in the company of Aristotle's Libri naturales. Might
this text have been recommended to Thierry of Chartres by Hermann
of Carinthia, and have arrived at Mont Saint-Michel from Paris in the
company of texts of the «new Aristotle»? The editors of the indivi-
dual texts of Aristoteles Latinus have recognised that the copies of
the texts of Aristotle in the two Mont Saint-Michel manuscripts are,
in most cases, the earliest witnesses to the versions which became the
vulgate. 711 It is either from them or their near relations that the manus-
cript tradition of Aristotle's Libri naturales descends. Wi1cox has
shown that the Mont Saint-Michel copy of the De differentia is a revi-
sed version in respect to the Edinburgh copy and that it too has a res-
pectable progeny. 79 In the manuscripts that may be earlier than this
copy, the De differentia accompanies medical texts (the Edinburgh
manuscript), miscellaneous scientific texts (the Cashel manuscript),
and theological texts (Brussels, Bibliotheque royale, 2772-89 110). We
may be justified then, in seeing the combination in A vranches 232
as the beginning of the association of the De differentia with the Libri
naturales - an association which was to prove lasting.
Nicomachea, Praefatio, Aristoteles Latinus, 26.1-3, Leiden and Brussels, 1974, pp.
cxxxviii-cxlvii) has been attacked by Joanna Judicka who, in her edition of De
generatione et corruptione (Aristoteles Latinus, 9.1, Leiden, 1986, pp. xxxiv-xxxviii),
defends Minio-Paluello's claim that one translator was responsible for all three texts.
Judicka suggests that, in the case of the De generatione et corruptione, Avranches
232 is 'sans doute legerement posterieur a [Oxford, Bodleian Library, Selden supra
24]'; its text is a revised version in respect to that of the Selden MS, in much the
same way as the text of De differentia is revised in respect to the Edinburgh manuscript
version.
78 This is the case at least for the Physics, Metaphysics and the Ethica vetus.
79Wilcox, The Transmission, p. 127. For examples of the revisions see Appen-
dix 11 below.
HO J. VAN DE GHEYN, Catalogue des manuscrits de la Bibliotheque royale de Bel-
gique, 2, Brussels, 1902, pp. 310-1 (with works by Fulbert of Chartres, St Jerome,
Drogo, Anselm and Hugo of St. Victor; this catalogue dates the manuscript to the
[31 ] 251
v v
Appendix I
The Prefaces of «Magister Iohannes Hispalensis (et Limiensis)>>
In the following editions modem conventions of punctuation have
been followed. However, the orthography of the texts has been retai-
ned. Square brackets indicate words in the manuscript which appear
to be redundant; angle brackets indicate editorial additions.
1. The preface to Thibit ibn Qurra, De imaginibus (Paris, Biblio-
theque Nationale, lat. 7282, fol. 29r).
Cum, ceteris astronomie libris perlectis, veluti cursu<u>m pla-
netarum aliisque que ad hanc artem pertinere videbantur, nichil horum
ob quorum intentionem Hispanas petieram partes adeptus fuissem, tanto
tedio per aliquot dies affectus, tabui ut, sopita desperacione quod inter
huius scientie peritos iam sciolus habebar, tanti frustra laboris inchoati
sollicitudinem [sollicitudinum] abiecerim. Videbam me namque in hac
diutissime ante elaborasse, preterea de propriis non pauca in ea, cum
nichil aliunde lucri facerem, expendisse, presertim cum gentes inter
efferas constitutus, procul a fidei domesticis tocius propter Deum,
expers consilii degerem. Quid animi haberem non est meum vestram
instruere prudenciam.
Hac igitur tanta me sollicitudine male pertractum, magister intuens
atque quid haberem sollicite querens, audito mentis mee langore sub-
risisse visus est. Demum librorum suorum volumina perquirens, non
magni corporis librum Arabico sermone con scriptum ex armario suo,
in quo libri eius non pauci continebantur, protulit. De quo, cum que
contineret attenta mente indesinenter percunctarer, se in hec verba
resolvit: «Ne credideris, karissime, omnes qui in hac scientia student
ad eius usque intema pertingere posse. Sunt enim nonnulli qui adeo
planetarum cursus copulacionesque eorum norunt, preterea signorum
demumque ferme tocius celi machinam mente contemplantur, ut nil
13th century). Munich, Bayerische Staatsbibliothek, 18917, listed by Wilcox as being
of the 12th century (p. 310), is described as being of the 14th and 15th centuries
in Aristoteles Latinus. Codices, no. 1071.
252 [32]
«MAGISTER IOHANNES HlSPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRrruS ET ANIMAE
sibi de hac arte deesse putant, set, ut verum fatear, tanto ab ea distant
intervallo quant9 qui numquam ex eo gustaverit quicquam. Sapientes
namque nostri eorum subtilius considerantes naturam planetarum atque
significationem nunc in bono, nunc in malo esse videntes, ex sui
capacitatis ingenii inmensitate quedam quasi argumenta atque exem-
plaria excerpsere eaque totius huius artis summam posuere (MS
posuisse) - librum scilicet componentes quem ymaginum esse dixe-
runt, in quo qual iter prodesse aut etiam obesse sibi vel alii possint,
sicut in sequentibus declarabitur, habetur. Unde et quidam sub religio-
nis obtentu hanc scienciam velud ignominiosam diiudicant, non atten-
dentes Deum hanc pocius servis suis adaptacionem terrarum suarum
et ad vindictam malefactorum, ad laudem vero iustorum, eum in suis
mirabile pre ceteris agnoscendum largitum fuisse. Sed forte michi quis
ex aliis obiciat affirmans non velle Dei fore quisquis malum operatur.
Ad quod ego: 'An nescis securim ad incidenda ligna factam? Numquid
si quis cum ea hominem mente perversa occiderit, ob hanc causam
securis usus quasi calumpniatus reprobandus sive abiciendus est? Non
ita est, inquam,' sed quia adversariorum questiunculis sufficienter
responsum est, de aliis agamus.»
Hunc ergo librum ab ipso, Dei iuvante spiritu, habui, quem nullus
Latinorum preter quendam Auriocenum, qui quondam eius partem
habuit, adeptus fuerat. Si quis ergo huic dans operam hunc scire voluerit,
modo de omnibus studeat hunc librum, videlicet ymaginum, habere.
Nam per eum si ibi providus fuerit, ad totius huius doctrine summam
proculdubio pertinget. Adhibe ergo animum, quisquis es, atque mente
sollicita revolve que in ipso legenda sunt, atque planetarum signifi-
caciones tarn in bono quam in malo sollerter considera. Finit prologus.
Liber ymaginum incipit Thebit ben Cora a lohanne Hispalensi
atque Limensi in Limia ex Arabico in Latinum translatus.
Translation:
Having read through the other books of astronomy, e.g. of the courses
of the planets and others which seemed to be relevant to this art, and when
I had obtained none of the things for the sake of which I had sought Hispanae
partes, for several days I lay wasting away and affected by such aversion
that, sedating my desperation on being now thought to be halting among the
[33] 253
v v
experts of this science, I threw from me the worry of such labour undertaken
in vain. For I saw that I had laboured in this for too long a time, and, moreover,
that I had expended not a little of my own resources in this - since I was
making no profit from elsewhere - especial1y when, placed amongst wild
races, I was living without help far from the domestic <comforts of people
having> complete faith towards God. It is not necessary for me to te11 you,
wise as you are, what state of mind I was in.
The Master, seeing me in a bad way because of such worry and solicitously
enquiring what was wrong, when he had heard about the weariness of my
soul, seemed to smile. Finally, looking through the volumes of his books,
he produced a small book written in Arabic from his bookcase, in which
several books of his were enclosed. When I was demanding from him
incessantly with an eager mind what it contained, he started to say this about
the book:
«Do not believe, dearest friend, that al1 those who study this science
can arrive at the heart (esoteric knowledge) of it. For there are some who
know the courses and conjunctions of planets and also of the signs, and
contemplate in their minds the machine of almost the whole sky, to such
an extent that they think nothing is lacking to them of this art. But, to confess
the truth, their distance from it is as great as that of him who has never tasted
anything of it. For our wise men, considering the nature and significations
of the planets more subtly, seeing them now to be for the good, now for
evil, from the hugeness of the capacity of their intelligence have excerpted
certain, as it were, proofs and examples, and have made them the sum of
this whole art - composing a book which they have caned 'on talismans'.
In this is contained how the talismans can help or hinder the wise men or
another person (as wi11 be shown in what follows). Hence also certain people
under the pretence of religion, judge this science to be, as it were, shameful,
not noticing that God has bestowed this utility of His world rather on His
servants, and, for the punishment of i11-doers and the praise of the just, He
has made the art to be acknowledged amongst His works as wonderful beyond
all others. But perhaps someone may object saying among other things that
it is not the wi11 of God that anyone should operate evil. To which I reply:
'Do you not know that an axe is made for splitting wood? Surely, if anyone
ki11s a man with the axe, with a wicked mind, the use of an axe should not
for this reason be blamed or rejected as if condemned? This is not the case,
I say'. But because I have replied sufficiently to these petty objections of
our adversaries, let us turn to other things.»
This book, then, I obtained from him, with the help of God's Spirit-
a book which no Latin other than a certain Auriocenus, who once obtained
254 [34]
«MAGISTER IOHANNES HISPALENSIS ET lIMIENSIS» AND QUSTA ION LUQA'S DE DIFFERENTIA SPIRrruS ET ANIMAE
a part of it, ever had. If anyone, therefore, paying attention to this, wishes
to know it, he should only make every effort to have (of a]] things) this book,
i.e., a book on talismans. For through it, if he pays careful attention, he will
without doubt obtain the sum of a11 this teaching. Pay attention then, whoever
you are, and with a careful mind, think about what should be read in it, and
consider diligently the significations of the planets both for the good and
for evil. The prologue ends.
The book of talismans of Thiibit ibn Qurra, translated from Arabic into
Latin in Limia by John of Seville and Limia, begins.
2. The oldest version of the preface to Secret of Secrets (Edinburgh,
Adv. 18-6-11, fol. 82r [= E]): KI
Domine .T. gratia dei Hispaniarum 82 regine, Iohannes Yspalen-
sis salutem. Cum hutilitate corporis olim 83 tractaremus et a me qua-
si K4 essem medicus vestra nobilitas quereret ut brevem libellum
<facerem> 85 de observatione diete vel de continentia corporis, id est
qualiter se deberent continere qui sanitatem corporis cupiunt observare,
accidit ut mee menti cogitanti vestre iussioni obedire huius rei exemplar
et Aristotilis philosofi Alexandro edite 86 repente occurreret, quod
HI Other editions of this preface, from later manuscripts, appear in Opera hactenus
inedita Rogeri Baconi. ed. R. STEELE, 5, Oxford, 1920, pp. xvii-xviii (from British
Library, MS Add. 26770 = R), Brinkmann (n. 41 above), and H. SUCHIER, DenkmiiLer
provenzalischer Literatur und Sprache, Halle, 1883, p. 472. I have consulted British
Library, Burney 360 (= B) as a representative of the later tradition of the text and
have checked R. The texts of the preface in later manuscripts differ considerably
from each other and from the early version in the Edinburgh MS. I mention variants
from Rand B only when they are significant or might help correct the Edinburgh
manuscript.
82 Hispanorum B.
83 Cum de utilitate corporis hominis B, Cum de utilitate corporis oHm R. Note
the spelling of 'hutilitate' indicating that the scribe is uncertain about where to put
'h's; see 'hedificaverat' and 'exibui' below.
114 cum B, ac si R.
8S facerem, add. Steele (not in R).
H6 et AristotiIis ... edite] E is difficult to construe, RB give a more understandable
text: 'ab Aristotile philosofo Alexandro editum'.
[35] 255
v v
excerpsi de Iibro quod Arabice vocatur aracelas, 87 id est secretum
secretorum, quem fecit, sicut predixi, Aristotiles philosofus Alexan-
dro regi magno, de dispositione regni, 88 in quo continentur multa re-
gibus utili a, Quem quidam 119 interpres, iussu imperatoris sui cum mul-
to labore quesivit. De cuius inventione sic ait:
«Egressus sum diligenter querere quod mihi preceptum est ab
imperatore et non cessavi soIJicite circuire loca vel templa in qui-
bus suspicabar philosofos sua abscondisse opera vel in quibus
commendaverunt suas doctrinas, donec pervenirem ad quoddam aJta-
re quod sibi hedificaverat Hermes, in quo Sol venerabatur a quibus-
dam. Ibique inveni quendam senem prudentem et religiosum scientia
et doctrina seu moribus ornatum. Huic adhesi et ei cum summa
reverentia placere studui, et amabilem me illi exibui et verbis dulcissi-
mis eum linivi, quousque secretum locum michi detegetur in quo
introivi. 90 Et sic auxiliante Deo et fortuna imperatoris invento quod
mic hi preceptum fuerat et quod diu quesieram, cum gaudio < ... >
translatum est.»
Ex quo ego presens opus tantum in Latinum transtuli, 91 non ex
toto litteraturam sequens - quod a nuIJo interpretum posse perfici
/82v/ arbitror - sed, iuxta posse meum, in quibusdam sensum, in
quibusdam etiam sensum et litteraturam 92 secutus sum. Nec mirum
H7 This corruption of the Arabic 'sirr al~asrar' is difficult to explain; B gives
'cyrecesar', R 'tirosesar'.
HH regiminis B, regiminum R.
K9 The abbreviation in E suggests 'quidem' which is also the reading of R.
90B and R substitute a longer passage for 'in quo introivi': «in quo inveni plura
philosophorum scripta et secreta inter que hunc Iibrum aureis litteris inveni» (B).
However, this appears to be a later addition rather than an omission in E, since it
does not occur in Philip of Tripoli's long version of the Secret of Secrets (ed. Steele,
Opera hactenus inedita, p. 39) or in the Arabic preface from which this story is taken
(ibid., p. 177).
91cum gaudio ... transtuli] Rand B give a fuller text, of which the return home
and the information concerning which languages the translation was made from and
to have parallels in Philip's version and the Arabic; cf. B: «reversus sum cum gaudio,
portans mecum desiderium meum. Post hec ab eodem a Greco in Arabicum trans-
latum, transtuli in Latinum presens opus». There seems to be a lacuna in E after 'cum
gaudio'.
92 It is possible that the second 'sensum' should be omitted, with Rand B.
256 [36]
«MAGISTER IOHANNES HISPALENSIS ET LlMIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
si imperitia mea hoc egi 93 cum pene omnes sapientes qui fuere in-
terpretes ita egisse noscuntur. Nam diversitas translationum indicat
quod nullus valet sequi semper litteraturam. Ego autem in omnibus
magis litteraturam secutus sum ne longius a veritatis tramite recede-
rem. Nemo ergo me in aliquo diliquisse miretur aut culpet, dum co-
ram omnibus confiteor me totius scientie pati inopiam. Possideat ergo,
iubente 94 Deo, nobilitas vestra 95 cum magna fortuna corporis inco-
lumitatem, et sciat se habere in hoc opere magnum profectum si
ascultaverit consilium. Michi autem pro labore a domino donetur 96
in futuro premium. 97
Translation:
Lady .T. by the grace of God, queen of the Spains, John of Seville gives
greetings!
When we were once discussing the uses of [the parts of] the body and
Your Nobility was asking from me, as if I were a doctor, that I should compose
a short booklet on the observation of a regimen (dieta), or about the continence
of the body, i.e., how those should discipline themselves (continere) who
wish to preserve the health of their bodies, it happened that there suddenly
occurred to my mind, which was thinking about obeying your command, an
example of this thing - being an edition of Aristotle the philosopher to
Alexander. I excerpted this from the book which in Arabic is called «Aracelas»,
i.e. the Secret of Secrets, which, as I have said, Aristotle the philosopher
made for Alexander, the great king, concerning the disposition ofthe kingdom,
in which many things useful for kings are contained. This book a certain
translator, by the command of his emperor, sought with much effort. About
its discovery he says this:
93 egit RB.
94It is possible that 'iuvante' is the right reading, confused with 'iubente' by
a Spanish-speaking scribe.
9~ vestra B, una E.
i
96 donech E.
97 et sciat.. .premium] not in RB. Note the similarity, however, between this
ending and the ending of the De differentia: «Auferat a te deus omnem tristiciam ... et
det tibi fortunam in hoc et in futuro seculo».
[37] 257
 'I
v I
«I went out to seek diligently what I had been ordered by the emperor,
and I did not cease careful1y to go round places or temples in which I thought
philosophers had hidden their works, or (men) in whom they had entrusted
their teachings, until I arrived at a certain altar which Hermes had built for
himself, in which the Sun was venerated by certain people. There I found
a certain wise and religious old man, adorned with knowledge and teaching
or good conduct. I stuck to this man and strove to please him with the greatest
reverence and made myself loved by him, and flattered him with the sweetest
words until he revealed to me a secret place, into which I entered. Thus with
the help of God and the good fortune of the emperor, having found what
had been commanded of me and what I had sought for a long time, with
joy <... the book> was translated.»
From this book I translated the present work only into Latin, not fol1owing
the letter entirely, which I think no translator can do perfectly, but, to the
extent of my ability, I have followed the sense in certain cases, the [sense
and the] letter in others. Nor is it surprising if through my inexperience I
have done this, since almost all wise men who have been interpreters are
known to have acted in this way. For the differences between translations
indicates that no one is able to follow the letter always. As for me, I have
rather fol1owed the letter in all cases lest I might depart from the path of
truth by any extent. For no one should wonder at or blame me if I make
mistakes in anything, since I confess before aU men that I suffer a lack of
competence in every branch of knowledge. Let Your Nobility possess, by
God's command, health of body with good fortune, and let her know that
she has in this work great profit, if she listens to the advice. To me, however,
may a reward from God be given in future for my efforts.
258 [38]
v
«MAOIS'reR IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQ~:S DE DIFFERENTIA SPIRrruS ET ANIMAE
Appendix n
In the following paral1els between Qusta ibn Luqa's De differentia spi-
ritus et animae and Gundissalinus's De anima and Hermann of Carinthia's
De essentiis, italics indicate the use of the same words, and bold type-face
significant differences in terminology. The line-numbers of Judith Wilcox's
edition of De differentia are given; for Gundissalinus the page and line num-
ber of Muckle's edition, and for Hermann the folio division in Burnett's edi-
tion, are provided. The discussions of each of the ways in which a thing is
moved, and of each of the words in the definitions of Plato and Aristotle
are numbered, and further proofs (introduced by «item») are indicated by
letters of the alphabet. Comparisons with the Arabic texts of the De differen-
tia, as edited by Gabrieli, Cheikho and Ulken, are made in the footnotes,
and comments on the «unrevised» version of the De differentia complete the
Appendix. For a discussion of the differences between these passages in De
differentia and the corresponding passages in Hermann and Gundissalinus
see above pp. 242-5 [22-5].
The editions:
G. GABRIELI, «La risalah di Qusta b. Luqa, 'sulla differenza tra 10 spirito e
J'anima'», in Rendiconti della Reale Accademia dei Lincei. Classe di scienze
morali, storiche e filologiche, Ser. 5, 19 (1910) pp. 622-55 (with Italian
introduction and translation).
L. CHEIKHO, «RisaIa fi'l-farq bain al-riiQ wa'l-nafs», in AI-Machriq, 14 (1911)
pp. 94-109 (in Arabic).
H. Z. ULKEN, Ibn Sina Risaleler, 2, Istanbul, 1953, pp. 84-100 (Turkish and
Arabic).
Judith C. WILCOX, The Transmission and Influence of Qusta ibn Luqa's 'On
the Difference between Spirit and the Soul', Ph. D., City University of New
York, 1985 (Latin, Hebrew and English translations).
J. T. MUCKLE, «The Treatise De Anima of Dominicus Gundissalinus», in
Mediaeval Studies, 2 (1940) pp. 23-103.
Hermann of Carinthia, De essentiis, ed. C. BURNETI, Leiden and Cologne,
1982.
[39] 259
v v
«MAGISTER IOHANNES HISPALENSIS ET LlMIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRrruS ET ANIMAE

1. Gundissalinus's use of the De differentia De differentia Gundissalinus

De differentia
(368) Dicamusque quod omne quod
movetur aut movetur per motu m sui
moventis, quemadmodum movetur
plaustrum per (370) motum bovum,
vel movetur cum id quod movet eum
non rnoveatur, quia quod movet ali-
quid aut movet et movetur vel movet
et non movetur. Et hoc fit quatuor
modis: quia (1) aut erit per desiderium
eius a quo movetur, quemadmodum
movetur amator ad eum quem amat,
(2) aut per (375) odium aut fugam sive
terrorem, 98 quemadmodum movetur
inimicus ab inimico suo vel contra
eum; (3) aut per acturn 99 naturalem,
quemadmodum movetur lapis a
pondere cum pondus per semetipsum
sit immobile; (4) vel quia id quod
movet est occasio principalis vel cau-
sa lOO illius quod (380) movetur,
quemadmodum rnagisteriurn est cau-
sa motionis motus rnagistri cum
magisterium non moveatur per mo-
turn magistri. Sic anima movet corpus
et ipsa non movetur per motum eius.
Anima igitur est causa motionis ani-
malium per voluntatem et opus atque
mutationem et (385) ipsa non movetur
aLiquo modo motion is corporis eo quod
ipsa sit incorporea.
98 'fugam sive terrorem' is a doublet, translating the single Arabic word 'al-
-tanaft' ('mutual incompatibility') in Gabrieli's text, 'al-munlifir' ('avoidance') in
Cheikho's, and 'al-tanlifura' ('avoidance') in HUlken's.
Y9 Arabic 'fiT ('act').
100 'occasio principalis vel causa' translates a single Arabic word 'sabab' ('cau-
Gundissalinus
(p. 33.30) Omne enim quod movet
aliud, aut quiescit dum movet, aut
movetur dum movet. ... [the inter-
vening argument comes from
Avicenna] ...
(p. 36.14) Quicquid autem a re
quiescente movetur, uno quinque
modorum movetur, quoniam move-
tur anima (1) aut per desiderium eius
quod appetit (2) aut per odium eius
quod respuit (3) aut per terrorem eius
quod refugit (4) aut per vim natu-
ralem sursum vel deorsum tendit, ut
lapis deorsum, ignis sursum, vel
sicut ferrum movetur ab adamante,
(5) aut quia res movens rei motae
causa principalis existit, sicut
scientia movet artificern, ipsa ta-
men non moveatur. Hoc ergo quinto
modo anima movet animalia, quia
est causa motus animalium per desi-
derium et per voluntatem et per opus
atque mutationem; ipsa tamen non
movetur aliquo modo motionis
corporum.
(307) Dicamus itaque quod Plato
philosophus diffinivit animam sic:
«Anima, inquit, est substantia
incorporea movens corpus' ... [Aris-
totle's definition; see p. 265 [45]
below] ... (313) (la) Ostendamus-
que quod anima sit substantia, et
dicamus quia quicquid recipit op-
posita cum sit unum (315) numero
et immutabile in sua essentia est
substantia: sed anima recipit virtu-
tes ac vitia cum sit una numero. Ut
anima Platonis, que inmutabilis est
in sua essentia, recipit virtutes ac
vitia, que sunt opposita; anima igitur
recipit opposita cum sit una numero
et (320) inmutabilis in sua essentia,
et ita est substantia ... (b) ...
(325) (2) Nunc ostendamus quod
anima sit incorporea et dicamus: (a)
uniuscuiusque corporis qualitates
sunt perceptibiles (MS B adds
«sensu»), 101 et cui us qualitates non
percipiuntur a corporeo sensu in-
corporeum est. Qualitates autem
anime sunt virtutes ac vitia, que sunt
insensibiles; (330) anima ergo est
incorporea. (b) Et item, omne corpus
subiacet omnibus sensibus aut
quibusdam ex eis, sed anima non
subiacet omnibus sensibus neque
quibusdam ex eis; anima igitur non
est corpus.
(37.8) Plato animam sic definit
dicens: «Anima est substantia in-
corporea corpus movens». (1) Quod
autem anima sit substantia sic
probatur: quicquid recipit contra-
ria, cum sit unum et idem numero,
substantia est. Sed anima, manens
una et eadem numero, recipit con-
traria quae sunt virtutes et vitia, gau-
dium et tristitia, scientia et igno-
rantia. Ergo anima substantia est.. ..
[the next section from Avicenna]
(39.38) (2) Item quod anima non sit
corpus sic probatur. (a) Omne cor-
pus habet qualitates perceptibiles
aliquo sensu; sed qualitates animae
non sunt perceptibiles aliquo sen-
su; ergo anima non est corpus.
(39) (b) Item omne corpus subiacet
omnibus sensibus vel aliquibus; sed
anima nullis subiacet sensibus;
ergo anima non est corpus.

se') in Gabrieli's text, or two words 'sabab biidi" ('principal cause') in Cheikho's
and Ulken's. 101 Arabic 'maQsiisat'.

260 [40] [41] 261


v I!
De differentia
(c) [tem, omne corpus aut animatum
est aut inanimatum, et si anima est
corpus (335) aut est animata aut ina-
nimata; et impossibile est ut anima sit
inanimata si est corpus, quia incon-
veniens est ut anima sit inanimata. Et
si dixerimus quod anima sit animata,
reiterabitur nobis sermo de anima
anime utrum sit corpus vel non,. et
(340) ascendet hoc ad infinitum. Non
est ergo anima corpus .. .. (d) (e) ...
(366) (3) Et quia iam patet quod anima
sit substantia incorporea, nunc expo-
namus quibus modis movet corpus (the
description of the 4 modes of move-
ment by an unmoved thing follows, as
above, p. 260 [40])
(404) Aristotiles philosophus ita diffi-
nivit (405) animam: ut diceret quod esset
perfectio corporis agentis et viventis
potentialiter. In Iibro autem Aristotelis
quem fecit de ani ma, talis est diffinitio:
«Anima est perfectio corporis naturalis
instrumentalis viventis potentialiter.»
Redeamus ad opus. 102 '" (425) Dicamus
ergo quod perfectio duobus modis dica-
tur: est enim perfectio prima, et est secun-
da. Prima namque perfectio in homine
est sapiencia atque magisteria. Secunda
vero perfectio in homine est studere in
his que novit ex magisteriis et (430)
scientiis. Verbi gratia: medicus dicitur
perfectio prima propter scientiam medi-
cine; cum vero ceperit operari quod scit,
dicitur perfectio secunda. Anima igitur
est perfectio prima, quia qui dormit etsi
careat sensu tempore dormitionis, est ei
tamen anima (435) sensibilis. Et omnis
species atque perfectio est species atque
perfectio alicuius rei. Anima ergo est
species et perfectio corporis.
For the Arabic text here see pp. 238-9 [18-9] above.
102
«Prima perfectio est. . .incidere ensi» from Avicenna, De anima, 1.1., ed. S.
103
Van Riet, I, p. 27.
262
"',,,
Gundissalinus
(c) Item omne corpus aut est animatum
aut inanimatum. Ergo si anima est cor-
pus, aut est inanimata aut animata. Si
autem anima est animatum corpus - sed
omne animatum corpus aJiqua anima est
animatum - tunc anima est animata ab
alia anima; et similiter ilia eadem quaes-
tio de ilia alia ani ma, et sic in infinitum.
Anima ergo non est animatum corpus;
sed nee est corpus inanimatum, quia quod
non vivit aliud vivificare non potest; ani-
ma autem corpus vivificat; ergo anima
nullo modo est corpus sed est substantia.
Ergo est anima substantia incorporea.
Quod autem sit movens corpus iam su-
perius demonstratum est. Vera est igitur
definitio animae secundum PI atonem ,
quod anima est substantia incorporea
corpus movens.
(40.13) Aristoteles autem sic defi-
nivit animam, dicens: <<Anima est
prima perfectio corporis naturalis,
instrumentalis, viventis potentialiter.
Perfectio autem alia est prima, alia
secunda. Prima perfectio est per
quam species fit species in effectu,
ut figura ensi; secunda perfectio est
ut aliquid eorum quae consequuntur
speciem rei aut ex actionibus eius aut
ex passionibus eius, sicut est inci-
dere ensi. 103 Prima perfectio est
scientia medicinae in cognitione,
secunda perfectio est medicina in
operatione. Anima ergo est perfectio
prima quia statim cum unitur copori
fit in eo plena potentia vivendi vel
sentiendi, et sic per animamperjicitur
species corporis animati quod ante
animam erat in sola potentia. . ..
[42]
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
De differentia
Modi autem corporum sunt duo, quia
sunt quedam corpora in qui bus est
species naturalis, ut animalia et
arbores, ignis (440) et aqua et quicquid
habet motum naturalem in semetipso.
Et sunt quedam quibus adquiritur
species per magisterium ut ostium et
scamnum ... (448) Modi quoque
corporis naturalis sunt duo, quia sunt
quedam simplicia, quedam vero (450)
composita. Simplicia veTO ut ignis, aer,
aqua, terra; composita ut animalia et
arbores. Anima autem non est species
simplicis corporis sed compositi
naturalis, quia quicquid habet animam,
est animatum, id est vivit, et quicquid
vivit est convertible vel dissolubile
(455) et necesse est ei cibus quo possit
recuperare quod dissolutum est ab eo,
et qui auxilietur eius vegetationi dans
ei incrementum.
Cibus quoque indiget diversis ins-
trumentis ex quibus sunt quedam ei
necessaria ut deferant eum ad cor-
pus reficiendum (460) eumque
currere ac penetrare faciant, utguttur
et vene in animalibus, torusque
(truncus quoque MS A) ac rami in
arboribus. Et quedam sunt necessaria
refecturo corpori ut eiciant ab eo
queque superjlua, ut sunt in ani-
malibus pori sive exitus resine in
arboribus. Multiplicantur quoque
(465) instrumenta in animalibus
propter magnitudinem perfection is
eorum et multitudinem operum'
eorum.
[43]
v
Gundissalinus
(40.26) Corpus autem aliud natu-
rale est ut arbor, aliud artificiale ut
scamnum quod animatum esse non
potest. Ut ergo removeatur artifi-
ciale apposuit naturale. Sed naturale
aliud est simplex, aliud compositum.
Simplex autem aliud est quodlibet
elementorum quatuor, aliud nec ele-
mentum nec elementatum ut quod-
libet caeleste corpus. Sed nullum
elementorum est animatum nec ali-
quod caelestium corporum secun-
dum Aristotelem, licet aliter vide-
atur Platoni.
(40.33) Nullum ergo simplex cor-
pus est animatum; sed nee omne
compositum naturale ut lapis. Ideo
apposuit instrumentalis, id est haben-
tis instrumenta quibus iuvatur ad vi-
tam. Instrumentorum autem alia sunt
necessaria ad (44) recipiendum
nutrimentum, alia ad supeiflua eici-
endum. Necessaria autem ad reci-
piendum nutrimentum in vegetabili-
bus sunt truncus, radix et rami et alia
huiusmodi; in sensibilibus autem
guttur et venae et alia huiusmodi.
necessaria veTO ad supeiflua expel-
lendum in utrisque sunt pori, sed in
sensibilibus etiam alia. Multipli-
ciora enim sunt instrumenta sen-
sibilium quam vegetabilium propter
multipliciores eorum. . ..
263
v v
De differentia
(479) ... interpretatio utrarumque
(480) diffinitionum est una. Dicens
enim viventis potentialiter noluit per
hoc inte11igi quod corpus ita esset in
sua essentia antequam esset anima;
post hec (?) recepisset animam per
hoc quod possibile fuit sibi vivere;
sed vo]uit intelligi per hoc quod dixit
(485) «potentialiteT» quod esset ei
instrumentum cui esset possibile uti
actibus vite. Idem ergo significat
'instrumentalis' quod 'viventis
potentialiter' .
264
Gundissalinus
(43.11) Dicens ergo viventis
potentialiter tantum mud ostendit
corpus per animam perfici, quod ante
animam habuit potentiam vivificari,
quod pene idem sonat instrumenta]e,
scilicet cui possibile est uti actibus
vitae.
[44]
«MAGISTER IOHANNES HISPALENSIS ET L1M1ENSIS» AND QUSTA lBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE
2. Hermann of Carinthia's use of De differentia
De differentia
(9) et ecce scribo tibi quedam collec-
tiva que excerpsi de Hbro (10) Plato-
nis qui vocatur Cadon et eius libro
qui vocatur Tymeus et ex libris Aris-
totelis philosophi et Theofra(s)ti ac
Benededis in animam; ex libro quo-
que Galieni quem fecit de concordia
quarumdam sententiarum gloriosis-
simi Ypocratis atque Platonis et ex
libro (15) eiusdem Galieni quem fe-
cit in opere cirurgie et in utilitate
membrorum ... (307) Dicamus ita-
que quod Plato philosophus diffinivit
animam sic: «Anima» inquit «est
substantia incorporea movens cor-
pus». Aristotiles vero in diffinitione
anime (310) ait sic: «Anima est per-
fectio corporis agentis et viventis
potentialiter'... (404) Aristotiles
philosophus ita diffinivit (405)
animam: ut diceret quod esset per-
fectio corporis agentis et viventis
potentialiter. In Libro autem Aris-
totelis quem fecit de anima, talis est
diffinitio: «Anima est peifectio cor-
poris naturalis instrumentalis vi-
ventis potentialiter.» Redeamus ad
opus ... (337) inconveniens est ut
anima sit inanimata ... (473) Et hee
diffinitio [Aristotilis] est universa-
lis .. .. (480) Dicens enim viventis po-
tentialiter ... (484) voluit intelligi
per hoc quod dixit (485) «potentiali-
teT» quod esset ei instrumentum cui
esse possibile uti actibus vite.
[45]
Hermann, De essentiis
(71 yE) Recte quidem quale Plato dif-
fin it, Aristotiles describit. Plato qui-
dem in Cadone, «Anima est» inquit
«substantia incorporea corpus
movens.» Aristotiles vero in libro De
anima sic: «Anima est» ait «peifectio
.
corporis naturalis instrumentalis
potentia viventis.» Et alibi: «Anima
est peifectio corporis agentis et
viventis potentia». Videtur itaque
diffinitio quidem (F) magis propria
tertio generi, seu quia solum hoc
incorporeum dicimus, cum in ger-
mine sive animali bruto nichil
supersit ultra tri plic em illum spiritum
quo vivit, spirat, sentit, quem corpus
esse secundum originis rationem sci-
mus, seu quia neutrum illorum subs-
tantia sit, si (read sed?) potius dica-
mus ea duo animandi genera virtu-
tes anime mundi, quemadmodum vi-
sum est eis qui si ea corpora esse con-
cedant, superesse sibi putant alias
item eis corporibus animas requi-
rendas. Nee enim inanimata dici (G)
eonsonum est. Descriptio [se.
Aristotilis] vero universalis. Perficit
enim anima corpus potentia vivens
dum vitam aetu ministrat (<<vivens»
inquam «potentia» cuius dispositio
vite animeque actionibus parata).
265
v v
«MAGISTER IOHANNES HISPALENSIS ET L1MIENSIS» AND QUSTA IBN LUQA'S DE DIFFERENTIA SPIRITUS ET ANIMAE

Hermann has swapped the order of the two definitions of Aristotle, but
made sure that the reference to the De anima is still attached to the correct
definition. The abbreviated version (called by Wilcox «Hermann of Carin-
thia's version») has mistakenly assigned the first of the two definitions to
Aristotle, perhaps because of unclear punctuation in its archetype. This, and
the fact that the abbreviated version makes no mention of «Cado» are argu-
ments against an attribution to Hermann. The equivalent passages in this
abbreviated version are as follows:
spiritum et animam») and in the preface to the Secret of Secrets (<<secre-
tum locum .. .in quo introivi»; an accusative would be expected in good Latin),
and (2) the fact the Edinburgh manuscript has no dedication, and may there-
fore represent a version of the text that predates the copy dedicated to Ray-
mon de La Sauvetat.

(204) Anima, inquit Plato, est substantia incorporea corpus mo-

vens. Aristoteles autem sic: Anima, inquit, est perfectio cor-
poris agentis et viventis potential iter. ... (281) Anima inquit
[Aristoteles] est perfectio corporis agentis et viventis poten-
tialiter in Libro suo de anima; et aliter: anima est perfectio cor-
poris naturalis instrumenti viventis potentialiter.... (225) et
ridicu]um est si dixeris earn [inJanimatum ... (the sentence about
Aristotle's definition being universal is omitted) ... (331) Igitur
«potentialiter» ., .sed voluit intelligi quod esset instrumentum cui
esset possibile uti actibus vite.

Indications that the text in Edinburgh, Advocates 18.6.11 is unrevised

in the above passages are:
1) the common placing of the copula before its complement: e.g., «est
incorporeum» (328); «est anima» (334); «sit anima» (336, 337, 338 and 341),
etc. This follows the Arabic order «in kanat aI-nafs ... » (vel sim.).
2) The use of «et» where other manuscripts have «atque» (e.g., 430).
Note that this manuscript cans the author «Iohannes Hispa]ensis et Limien-
sis» while most other manuscripts of other texts of his call him «Iohannes
Hispalensis atque Limiensis».

However, some tenns and passages are closer to the Arabic in the revi-
sed version:
1) «Insensibiles» (325) for the Edinburgh MS's «invisibiles»: here
«insensibiles» is a better translation for «la maQsusat».
2) In line 450 «ignis, aer, aqua, terra» is exactly what is found in the
Arabic; the Edinburgh manuscript has «est aer et cetera elementa».

One may notice also (l) the indiscriminate use of «in + ab!.» which
appears in the title of the text in the colophon of the Edinburgh manuscript
(<<in spiritus et anime differentia»; the regular version is «de differentia inter

[47] 267
266 [46]
 VI

JOHN OF SEVILLE AND JOHN OF SPAIN: A MISE AU POINT1

The controversy concerning the identity of 'Iohannes Hispalensis', a

translator of mainly astronomical and astrological texts from Arabic into
Latin in the twelfth century, and whether he is the same man as
'Avendauth', 'Iohannes Hispanus', 'magister Iohannes' and 'Iohannes David',
has long engaged scholars, and no single solution has won general
acceptance. 2 To arrive at such a solution it is necessary to take into account
not only attributions (on which most arguments up to now have relied), but
also the style and terminology of the texts, and the company they keep in
the manuscripts. The editions of translations attributed to 'Iohannes
Hispalensis' and the other Johns, which have recently been completed or are
currently in process,3 should facilitate this research. In this article, I should
like to suggest some provisional distinctions that delineate two corpora of
texts, the first, those of'lohannes Hispalensis' (John of Seville), the second,
those of '(magister) lohannes (Hispanus)' (Master John of Spain), as a
starting point for further research.

1. John of Seville

The following translations from Arabic are attributed to 'Iohannes

Hispalensis' in their manuscripts:4

1 I am grateful for the help of Paulo Alberto, Alexander Fidora, Anne-Marie Vlasschaert,
Steven Williams and David Juste in the preparation of this article.
2 The following earlier studies remain useful: M. ALoNSO ALONSO, "Juan Sevillano: sus
obras propias y sus traducciones", in AJ-Andalus 18 (1953), pp. 17-49; M.-T. D'ALVERNY,
"Avendauth?", in Homenaje a Milltis-Vallicrosa, 2 vols, Barcelona, 1954-6, I, pp. 19-43; L.
THORNDIKE, "John of Seville", in Speculum 34 (1959), pp. 20-38. See also S. J. WILLIAMS,
The Secret of Secrets: The Scholarly Career of a Pseudo-Aristotelian Text in the Latin Middle
Ages, Ann Arbor, 2003, pp. 31-59, and Serafin VEGAS GoNzALEZ, La Escuela de Traductores
de Toledo en la Historia del Pensamiento, Toledo, 1998, pp. 51-65.
3 The relevant editions are mentioned in the lists of texts that follow.

4 Those which include 'et LimiensislLimensislLunensis' are marked with an 'L'.

Addressees and dates are given where they exist. For a detailed account of the texts
specifically attributed to 'Iohannes Hispalensis et Limiensis' see BURNETT, "'Magister
Iohannes Hispalensis et Limiensis' and Qus18 ibn Liiqa's De differentia spiritus et animae: a
Portuguese Contribution to the Arts Curriculum?", in Mediaevalia, Text08 e Estudos 7-8,
Porto 1995, pp. 221-67. For the appearance of the same topographic epithets, but in reverse
order, see the beginning of a text on spiritual magic in MS Venice, Biblioteca Nazionale
Marciana, lat. XIV.174 (4606), s. xiv, fol. 23v: "In nomine Domini. Incipit commemoratio
ystoriarum et mirabilium operis Veneris a Toz Greco inventa, a Iohanne vero Iuliensi atque
Yspalensi ex Hebrayco in Latinum translata." While there is no record of John of Seville
translating from Hebrew, it is plausible that he was involved in translating magical texts in
VI
60
1) Pseudo-Aristotle, Secret of Secrets, addressed to Tarasia, the Queen of
the Portuguese from 1112-285
2) Qusta ibn Liiqa (Costa ben Luce) , De differentia spiritus et animae,
addressed to Ra~ond de La Sauvetat, Archbishop of Toledo 1125-52 La
3) Masha'allah (Messehalla), De rebus eclipsium L
4) <Umar ibn al-Farrukhan at-Tabari (Omar), De nativitatibus L
5) Thabit ibn Qurra (Thebit), De imaginibus L7
6) Abii Ma<shar (Albumasar), Liber introductorii maioris ad scientiam
iudiciorum astrorum (1133?) L8
7) al-FarghanI (Alpharganus), Liber in scientia astrorum (11 March 1135)
L Limiensi) ex Arabico in Latinum".13
8) al-QabI~I (Alcabitius), Introductorius9
9) Masha'allah (Messehalla), De interrogationibus (De receptione
planetarum)
10) A text on the construction of the astrolabe, beginning "Astrologicae
speculationis exercitium"lO
11) A text on the use of the astrolabe, beginning "Primum capitulum in
inventione" Lll
addition to Thebit's De imaginibus. However, the study of the complex situation of the
translation and transmission of magical texts is still in its infancy.
li Provisional edition by H. SUCHIER, Denkmiiler Provenzalischer Literatur und Sprache
Halle, 1883, pp. 473-80.
6 A. revised version of .Ju~th WILCOX's edition of Qusti ibn Liiqa's De differentia spiritus
et ~1l;,mae ("The TranslDlSslon and Influence of Qusta ibn Luqa's 'On the Difference between
Spmt and the Soul"', Ph.D., City University of New York, 1985) is in the press.
7 Francis CARMODY edits two forms (I and J) of the Latin text in his The Astronomical
Works of ThfJbit b. Qurra, Berkeley and Los Angeles, 1960, pp.180-97.
~ E~ted. b>: ~chard LEMAY in his Abii MacAar al-Balbi, Liber introductorii maioris ad
sC£entwm}ud£cwrum astrorum, 9 vols, Naples, 1995-6 (= Lemay), V.
9 Edition in preparation by the present author.
10 Partial edition in J. M. MILLAs VALLICROSA, Las traduccwnes orientales en los
manuscritos de la Biblio.teca Catedral de Toledo, Madrid, 1942, p. 316 (half-way down the
page? to p. 321. No ArabiC author is named. This text is 'NN' in P. KUNlTZSCH, Glossar der
arab£schen Fachausdru.cke in der mittelalterlichen europaischen Astrolabliteratur
Nachrichten der Akademie der Wissenschaften in G(Sttingen I. Philologisch-historisch~
Kla~se, . Jahrg~ 1982, Nr. 11, G(Sttingen, 1983, pp. 459-571 (see pp. 496-7). The
attnbuttons of this text and item 11 to John of Seville are found in MS Pommersfelden
Schlossbibliothek 66, fols 106ra and 134ra, and in the Speculum astronomiae: see P:
ZAMBEW, The Speculum Astronomiae and its Enigma Dordrecht/Boston/London 1992 p
218. " ,.
11. Partial edition in MILLAs VALLICROSA, Las traducciones, pp. 261-84. The text is
ascnbed to ~aslam~ ~-~rili in Latin, but in reality corresponds to an Arabic text by Ibn
~-~~r. This text IS SJ' ID KUNlTZSCH, Glossar der arabischen FachausdrUcke, pp. 486-7.
This IS presumably the text mentioned in the Biblionomia of Richard of Fournival (before
VI
John of Seville and John of Spain 61
From their subject matter, and the similarity of their colophons, the
following anonymous translations may also belong to the corpus of
translations of John of Seville:
12) Abii Ma<shar, Liber experimentorum
13) Abii Ma'shar, Flores
14) Abii Ma<shar, De magnis coniunctionibus 12
The fullest information concerning the authorship of these translations
usually occurs in the respective colophons, which follow a regular pattern:
"Perfectus est <title and author of Arabic text> cum/sub laude Dei et eius
auxilio/adiutorio, translatuS/interpretatus a Iohanne Hispalensi (etJatque
Only one of these works (item 7) includes an unambiguous date of
composition: "Perfectus est liber Alfragani in scientia astrorum et radicibus
motuum celi, interpretatus a Iohanni Hispalensi in Limia atque Limensi, et
expletus est die .xxiiii..v. mensis lunaris anni Arabum .dxxix., .xi. die
mensis Martii era .mclxxiii."14 The Christian date is that of the Spanish era,
which is 38 years ahead of the annus domini, and thus, in this case, is
equivalent to A.D. 1135. Two further works have dates, in one or other of
their numerous manuscripts, which are more difficult to interpret: namely,
item 6: "Perfectus est Liber introductorii maioris in magisterio scientie
astrorum editione Abimaser et interpretatione Iohannis Yspalensis ex
Arabico in Latinum sub laude Dei et eius auxilio anno .71., scriptus est liber
iste anno domini nostri Ihesu Christi .1171. mense Aprili",15 and item 8:
"Perfectus introductorius liber Alcabisii ad magisterium iudiciorum
'
astrorum, octavo die mensis Ianuarii tertie indicionis annis domini perfectis
.1181. Explicit Deo gratias".16 In both cases an anno domini date is given
1260) as "Johannis Hyspalensis atque Linensis liber de opere astrolabii secundum
Mascelamach" (L. DELISLE, Le cabinet des MSS de la Bibliotheque nationale, 11, Paris, 1874,
p. 527), which, however, is now missing from the manuscript (Paris, BNF, lat. 16652): see R.
H. RoUSE, "Manuscripts Belonging to Richard of Foumival", in Revue d'histoire des tutes 3
(1974), pp. 253-69 (at p. 263).
12 Edited by the present author in the second volume of K. YAMAMOTO and C. BURNETT,
Abo Ma ~r on Historical Astrology: The Book of Religions and Dynasties (On the Great
Conjunctions), 2 vols, Leiden, 2000. For the Toledan Tables and Abii 'Ali on nativities, see
below.
13 For examples of the colophons see BURNETI, "'Magister lohannes Hispalensis et
Limiensis"', pp. 225-7 and LEMAY, IV, pp. 307-9.
14 This is an edited version of the variants, a selection of which is given in LEMAY, IV, pp.
121-2.
15 This form of the colophon occurs in MS British Library, Harley, 3631. fol. 57v (here
'.1171.' is read incorrectly as '1170' by Lemay), and two related MSS: see LEMAY, IV, pp. 72,
119-21, 164-5 and VI, pp. 677-8. The more usual form of the explicit follows the formula of
other texts: "Perfectus est liber ... astrorum translatus a Iohanne Hispalensi".
18 This date occurs only in Vatican, Barberini 236, fol. 54r, whose version of the text
includes features (such as the distribution of the letters of the Greek alphabet among the

VI
62
rather than the Spanish era, but in item 6, Moritz Steinschneider argued
that '1171' should be interpreted as if it were the Spanish era giving
'113~'.~7 One would n?t expect a Spanish era date to be reported as ~ anno
do,!"m~ .date, and RIchard Lemay, more plausibly, interprets the more
emgmatIc 'anno .71.' ('anno .a1.' in a second manuscript) as the era date
<11>71, Le: 1133. Although the repetition of the date in the colophon implie~
that two dIfferent epochs are at issue, the interpretation of 'anno .71.' as
1133 cannot be accepted uncritically.
. !l!e colop~on of !te~ 7, together with item 3's "a Iohanne Hispalensi in
LImIa and Item 5 s a Iohanne Hispalensi atque Limiensi in Limia"
suggests that Iohannes either came from a certain "Seville in Limia"
~ade. his t~anslation ~n "Limia"; in both cases 'Limia' is probably to be
Ide?tdied ';Ith the regIOn of the Lima valley in Portugal-an identification
which receives some support from the fact that item 1 is dedicated to the
Queen of Portugal. Nevertheless, in many cases, the word interpreted here
as 'limia' or 'limiensis', can also be read 'luna' or 'lunensis" thus the
possibility is not entirely ruled out that John was associated with a locality
called 'Luna'.18 What must be noted is that no translation is stated to have
been made in Toledo.
.Th~ only connection of a text with Toledo is the dedication of item 2. A
dedication to ~chbishop Raymond, however, need not necessarily imply
that John was m Toledo. He could have been looking for a patron, especially
after the Queen of Portugal's defeat by her son in 1128. There is no clear
?ocu~ent~ evidence of him in Toledo,19 nor has his name been positively
Identified m any other Spanish document.
The peri~d .of activity of this 'Iohannes Hispalensis (et Limiensis)', as
~an be seen, I~ 1~ the 1120s and early 1130s. With the possible exception of
Item 10, no ongmal works are included in this list of texts. 20 In no case is a
collaborator mentioned, which suggests that John of Seville was a native
signs of. the zodiac) that appear to be older than those in any of the other ca. 200
man~~pts of the text, but it is difficult to reconcile the indictio number with the anno
domml date. Moreover, the Barberini manuscript does not name the translator.
17 M. STEINSCHNEIDER, Die hebraeischen Uebersetzungen des Mittelalters Berlin 1893 p
568. ' , ,.
~8 This iD;terpretation is favoured by Richard LEMAY, who refers to several places with
thiS name 10 Aragon and LOOn: Abu Ma'shar and Latin Aristotelianism in the Twelfth
Century, Beirut, 1962, pp. 13-14.
19 A 'JO?anni Sibili' ~s given land, along with a group of other people of apparently Arabic
(Mozara~lc) descent, 10 Campo Rey near Toledo on 2 May 1146 (F. J. HERNANDEZ, Los
C!a~ulanos de Toledo, 2nd ed., Madrid, 1996, p. 57, no. 54) and is still living there in 1182
(Ibid., p ..188, no. 199); there is nothing that indicates that this 'John' was involved in
scholarship. Later, a 'Iohan de Sephila' attestates several documents between 1192 and
1226 as a canon of Toledo and 'scriptor archiepiscopi' (ibid., p. 698, s.v. 'Juan de Sefila'). (I
owe these references to Steven Williams and Alexander Fidora).
20 The. case of the one original work mentioned in scholarly literature, the Epitome totius
aBtrologwe of 1142, is dealt with in section 6 below.
l
VI
John of Seville and John of Spain 63
speaker of Arabic. In some manuscripts of items 2 and 4 he is called
'episcopus'.21 Only occasionally, in the rubrics to item 4, is he referred to as
'magister'; elsewhere there is no hint that he has been educated in the Latin
Schools.
2. (Master) John (of Spain)
A rather different picture emerges concerning a group of texts associated
with the names 'magister Iohannes', 'Iohannes Hispanus' and 'magister
Iohannes Hispanus'. Here we have an example of collaboration: with
Dominicus Gundisalvi, who is attested, as Archdeacon of Cuellar, in
0; documents from Toledo Cathedral from 1162-1181.22
a) al-Ghazzali (Algazel), Liber Theorice philosophie (Maq~id al-falsafa).23
b) Ibn Gabirol (Avicebron), Fons Vitae. 24
In the first instance the name is given as 'magister Iohannes'; in the
second, the manuscripts give 'Iohannes Hispanus/Hispanis'. There seems no
reason to doubt that the same man collaborated on both texts, although a
detailed analysis of the vocabulary and terminology of the two texts has not
yet been made .
21
m
See BURNE'IT, "'Magister Iohannes Hispalensis et Limiensis , pp. 234-6. To the
evidence adduced here can be added the incipit of item 2 in MS Tortosa, Archivo de la
Catedral, 80: "Incipit liber de differentia que est inter spiritum et animam quem Consta Be
Luca pg... .li (pagano?) cuidam regi edidit, quem Iohannes Ispalensis Episcopo (for '-us'?)
domino suo Raimundo Toletane sedis archiepiscopo ex arabico in latinum transtulit": see J.
MARTtNEZ GAzQUEZ and A. MARANINI, 'La recensio unica del De differentia inter spiritum et
animam de Costa Ben Luqa', in Faventia 19 (1997), pp. 115-29 (see p. 122).
22 See J. F. RIVERO RECIO, "Nuevos datos sobre los traductores Gundisalvo y Juan
Hispano", in Al-Andalus 31 (1966), pp. 267-80 and C. BURNE'IT, "Magister Iohannes
Hispanus: towards the Identity of a Toledan Translator", in Comprendre et Maitriser la
nature au moyen age: Melanges d'histoire des sciences offerts cl Guy Beaujouan, ed. D.
JACQUART, Geneva, 1994, pp.425-36. A further document from Burgos mentions a
'Dominicus Gonsalvi' (without a rank) involved in 1190 in a dispute (?) between the bishops
of Palencia and Segovia; since the subject of the dispute is the church of Santa Marfa de
Cu~llar, it is very likely that this is the same Gundisalvi: see D. MANSILLA, "La
documentaci6n pontificia del Archivo de la Catedral de Burgos", in Hispania Sacra 1 (1948),
pp. 141-63, n. 40; I owe this reference to Alexander Fidora. Adeline RUCQUOI'S arguments
for differentiating between a translator 'Dominicus Gundisalvi' and an original writer
'Gundisalvus' ("Gundisalvus ou Dominicus Gundisalvi", in Bulletin de philosophie rnedievale
41 (1999), pp. 85-106) have been effectively refuted, in my view, by Alexander FIOORA and
Marfa JesUs SOTO BRUNA, in '"Gundisalvus' ou 'Dominicus Gundisalvi'? Algunas
observaciones sobre un reciente articulo de Adeline Rucquoi", in Estudios ecleswticos 76
(Universidad Pontificia Comillas, 2001), pp. 467-73. The original writings show such an
intimate use of the translations and of further Arabic texts, that their author must at least
have been Dominicus's alter ego.
23 Ed. J. T. MUCKLE, Algazel's Metaphysics: A Mediaeval Translation, Toronto, 1933.
24 Ed. C. BAEUMKER, in Beitrii.ge zur Geschichte der Philosophie des Mittelalters, I,
Munster, 1892-5.
 l
VI
VI
John of Seville and John of Spain 65
64

The collaboration of Master John (of Spain) with Dominicus Gundisalvi
would suggest that another text that was known to the Archdeacon, and is
attributed to 'magister lohannes', belongs to the same 'Iohannes':
c) al-Khwarizmi (Algorismi25 ), Liber de pratica arismetice. 26
This is the most elaborate of all the twelfth-century Latin versions
deriving from al-Khwarizmi's text on calculating with Hindu-Arabic
numerals. It is a Latin composition rather than a direct translation from
Arabic, and exploits other Latin texts, such as the Helcep Sarracenicum of
Ocreatus, the pupil of Adelard of Bath, and an algorism which was also the
source of a Liber pulveris. Andre Allard, who has edited several twelfth-
century algorisms and studied their mutual relationship, has drawn
attention (p. xx) to Dominicus Gundisalvi's possible knowledge of this text:
this is the explicit reference to a 'Liber Algorismi' in Gundisalvi's De
divisione philosophiae, in which the title27 and the order of procedures
matches that of the Liber Algorismi de pratica arismetice: 28
De divisione philosophiae, p. 91 Liber Algorismi de pratica
arismetice, p. 62-3
Arithmetica alia est practica ... Ex Oportet prenoscere ... qualiter fiant
materia vero accidit ei aggregari et aggregationes et diminutiones,
disgregari, multiplicari et dividi et multiplicationes et divisiones,
huiusmodi que docentur in libro insuper etiam suarum radicum
Algorismi. adinventiones.
In an untitled text explaining the differences between the various
astronomical tables current in his time (which I shall call for convenience De
differentiis tabularum), a 'Iohannes Ispanus' promises to write about the
Hindu-Arabic numerals of al-Khwarizmi:
25 The variants of the Latin form of the name are 'Alchorismi' 'Alghoarismi' and
'Algorismi'. '
26 Ed. A. ALLARD, MuJ:tammad ibn Muss al-Khwsrizmi, Le Calcul [ndien (Algorismus).
Paris and Namur, 1992, pp. 62-224.
27 In the Liber Algorismi 'Algorismi' should be interpreted as the name of the author ('al-
Khwiri~mI'~, and 'de ~rati~ ~~metice' as the subject; in Gundisalvi, since the second part
of the tItle IS absent, Algonsml could be either the name of the author or the name of the
subject-matter: we may see here a transition to 'algorismus' becoming the technical term for
Indian arithmetic. Other early algorisms use different terms for the topic: in Helcep
S~rracenicu~ the Arabic term for 'calculation' (~isab) is simply transliterated, whereas the
L,ber pulvens :efers to th~ Arabic name for the Indian numerals: ghubarC or 'dust' figures;
more common IS the term numeruslnumeri Indorum'.
28 ~UNDISSALINl!S [G~DISALVI), De divisione philosophiae, ed. L. BAUR (Beitr~ge zur
Geschichte der Phllosophle des Mittelalters 4), Munster, 1903. Similarities in terminology
are underlined.
... maxime illud genus numeri quod fit per litteras Indorum quo
Sarraceni maxime utuntur quod probus Alchoarismi mirabili
ratione invenit, de quibus, nos auxiliante Deo, vel in fine huius
operis vel in alio loco, si Deus voluerit et vita comes fuerit,
aliquantulum pro modulo imperitie nostre tractare decrevimus.
... especially that kind of numeral which is written with the letters
of the Indians and used above all by the Saracens: the good al-
Khwarizmi, with remarkable intelligence, discovered this kind and
we, with the help of God, have decided to deal with these numerals
a little, according to the measure of our inexperience, either at the
end of this work or in another place, if God wills and we live long
enough.
It seems very likely that the author is referring to the Liber Algorismi de
pratica arismetice, especially since the latter is an original treatment of the
subject and not a translation. A little later in the De differentiis ta~ularu,!"
the author gives his own name, and a reference to another text whlch, thls
time, he says he has translated from Arabic:
per quos <computare possis> ordinatos cursus in libro quem ego
lohannes Ispanus interpres existens de Arabico in Latino transtuli.
... through which you can compute the regular courses <of the
planets> in the book which I, lohannes Ispanus, being the
interpreter, translated from Arabic into Latin.
There are several references in the De differentiis tabularum to these
tables and their canons. We can be reasonably sure that they were drawn up
for Toledo, but we cannot, as yet, determine whether they are one of the four
twelfth-century versions currently known. 29 We may, therefore, a.dd two
more items to this group of texts associated with Master John (of Spam):
d) De differentiis tabularum. 30
e) Liber de cursibus (not identified).
29 Three versions have been edited by F. S. PEDERSEN, The Toledan Tables, 4 vols
(continuous pagination), Copenhagen 2002; the fourth version is ~h~t presumed: to be the
basis of Raymond of Marseilles' adaptation of the tables to the mend~an of!darseIlles. None
of these versions is attributed to a Latin author. Pedersen, pp. 194-5 10vestlgates the nat~re
of the tables referred to several times in "Qui ad astronomie scientiam desiderat perve~"
(De differentiis tabularum), and concludes (p. 195): "It is fairly certain that the :John' m
question was, according to himself, the translator of a set of ~bles that contained ~e
Toledan mean motion tables. It is also possible, though less certa1O, that he had something
to do with some of the canons treated here."
30 Edited by J. M. MILLAs VALLICROSA in "Una obra astron6mica desconocida de J~hannes
Avendaut Hispanus", in Osins 1 (1936), pp. 451-75, and reprinted in idem, Estudws sobre
historia de la ciencia espalioZa I (Barcelona, 1949), pp. 263-88.
VI
66
According to one of its two manuscripts, the De differentiis tabularum
was written because two Englishmen, called Gauco and William, requested
and paid for it ("rogatu et ope duorum Angligenarum, Gauconis scilicet et
Willelmi"), and Charles Haskins remarked that a 'William Stafford' was
archdeacon of Madrid in 1154, when he attestated a charter at Toledo. 31 A
date in the vicinity of 1154 is not incompatible with that of the composition
of the text.
The best manuscript of the Liber Algorismi de pratica arismetice is
Paris, BNF, lat. 15461. This belongs to a group of three large manuscripts
written by the same hand, which give texts written or translated in Toledo.
The other two-Paris, BNF, lat. 9335 and Vatican, Ross. lat. 579-give the
most authoritative versions of the mathematical translations made by
Gerard ofCremona (1114-87), the doyen of the translators at Toledo. 32 Paris,
BNF,lat. 15461, on the other hand, does not contain any text attributable to
Gerard, but its Toledan credentials are ensured by the fact that Liber
Algorismi de pratica arismetice is followed by a calendar relevant to Toledo
in particular because of its inclusion of the Toledan saint Eugenius. Since it
includes the feast of the "transfer (of the relics) of St. Eugenius" it must
postdate 1156. 33 This calendar is accompanied by a computus referring to
the "present years" of 114334 and 1159. 35 Following the calendar and
computus is the final work in the manuscript, which is untitled here, but
which, from another manuscript we know to be the Liber mahameleth. 36
31 C. H. HAsKINs, Studies in the History of Medieval Science, 2nd ed., Cambridge, Ma.,
1927, p. 127, referring to a charter printed by Fita in Bolet{n de la Academia de la Historia
viii. 63, 188~.'fh:i
s , and a second charter attestated by William at Toledo in the same year:
are summarised In HERNAN'DEZ, Los Cartularios, nos 97 and 101.
32 See BURNETr, "The Coherence of the Arabic-Latin Translation Program in Toledo in the
Twelfth Century", in Science in Context 14 (2001), pp. 249-88, which includes a critical
edition of the list of translations drawn up by Gerard's students (socii).
33 See fols. 15r and 16r. On February 12th is noted the "translatio s. Eugenii ep(iscop}i et
m(artyr)i", while his "day" is November 15 ("Eugenii ep(iscop)i et m(artyr)i"). An arm of
Saint Eugenius, the legendary pupil of Dionysius the Areopagite and first bishop of Toledo,
was brought to Toledo on 12 February, 1156: see J. F. RIVERO RECIO, San Eugenio de Toledo
y su culto, Toledo, 1963, pp. 58-60, 181-3. This Toledan connection of the calendar was
pointed o~t by M.-T. D'ALVERNY in "Translations and Translators", in Renaissance and
Renewal £n the Twelfth Century, eds. R. L. BENSON and G. CONSTABLE, Cambridge, Ma.,
19.82, . pp. 421-62, reprinted in eadem, La Transmission des textes philosophiques et
scumtifiques au moyen dge, Aldershot, 1994, article II (see p. 459).
34 Fol. 19v: "A nativitate Christi usque in presens tempus, quando epacte tres currunt,
sunt anni .1143."
36 Fol. 21v: "A nativitate vero Christi usque in presens tempus quando epacta nulla currit
sunt anni .1159." 1159 is also the starting date in a table ofindictiones on fol. 22v. Note that
nowhere in the calendar or computus text is the Spanish era used.
36 MS Padua, Biblioteca capitolare, D.42 entitles the text "Incipit liber mahameleth de
numeris". This title is confirmed by an internal reference to the book: Paris, BNF, lat.
15461, fol. 28ra (ed. VLASSCHAERT, p. 25): "deinde accedendum est ad hunc librum
mahamelet".
VI
John of Seville and John of Spain 67
None of the four manuscripts of the text give any author, and the date of
composition can only be conjectured from the mention of certain units of
currency-"morabitini melequini" and "morabitini baetis"-which are
attested in the mid-twelfth century.37 'Mahameleth' is a transcription of the
Arabic word 'mu(amalat', a technical term for "arithmetical procedures
applicable to matters of everyday life / business arithmetic". The text is
based on several Arabic sources, rather than a literal translation of any of
the them. A unifying element in Paris, BNF, lat. 15461 is calculation, and
this is epitomised in the table added at the end of the manuscript (fol. 5Ov),
which repeats on a large scale the tables of Indian numerals occurring
within the Liber mahameleth on fol. 26v, and with the Liber Algorismi on
fol. 9r.3S
The compilatory nature of the Liber mahameleth reminds one of the
nature of the Liber Algorismi de practica arismetice: both are substantial
Latin compositions concerning subjects for which the Latins had no
adequate terms: 'mu(amalat' and '<the arithmetic of> al-Khwarizml',
respectively. There are other similarities shared by the two texts:
1) Both use examples with 'morabitini' and 'modii'. The coin named after
the Almoravids was particularly common in Barcelona, where it was
introduced between 1117 and 1135 (it was not introduced into Castile
until ca. 1180).39
2) The title of the Liber Algorismi in Paris, BNF, lat. 15461 (P)-Liber
Alchorismi de practica arismetice-recalls phrases at the beginning of
Liber mahameleth, in which 'arimethica ... practica' is the subject.40
37 J. SESIANO, "Le Liber mahameleth, un traite matMmatique latin compose. au XII~me
si~cle en Espagne", in Histoire des mathimatiques arabes: actes du premJ.er coll~ue
international sur l'histoire des mathimatiques arabes (Alger 1,2,3 decembre 1986), AlgIers,
1988, pp. 69-98 (see pp. 71 and 95, n. 3) identifies the 'morabitini baetis' with coinage fro~
Baeza which could be situated. in the mid-12th century; F. MATEU Y LLoPIS, Glosano
Hispanico de Numismatica, Barcelona, 1946, includes (p. 143) the 'morabitinos melechinos'
(attested in Barcelona between 1143 and 1153).
38 It should be noted that this table is written in a later hand than that of the manus~pt,
and serves preeminently as an indication of the most significant content .of the m~U8~pt:
the calculation by Indian numerals which is a theme common to the L£ber Algonsm£ and
the Liber mahameleth. That the table has been copied directly from the latter .work,
however, is clear from the occurrence of the numbering of the columns (= the notae In ~e
Liber mahameleth), which is reminiscent of abacus tables rather than ~ose accompan~
the algorism; indeed, the table on fol. 50v is titled "Tabula abaCl de opere practtco
numerorum".
39 ALLARD, Le Calcul [ndien, pp. xvi-xvii.
40 Anne-Marie Vlasschaert has also pointed. out to me (in a private communication) the
verbal correspondence of the discussion of the establishment of names for ~ finite number of
numerals in the Liber mahameleth (Paris, BNF, lat. 15461, fol. 26rb) and In the seco~~ P~ *
of the Liber Algorismi (ibid., fols 12vb-13ra), but since this second part (whose ~~on IS
being prepared by Andre Allard) gives the impression of being a miscellany of additions to
the first part, it is difficult to know what significance to place on this correspondence.
VI
VI
68
John of Seville and John of Spain 69

But the most significant indication is their common relationship to the

num~rus in se attenditur dicityr numerus per §e at~nditur. dicitur
chapter on 'arithmetic' in Gundisalvi's De divisione philosophiae. This text
as is well known, is in itself an intelligent compilation of Arabic and Lati~ theorica vel sgeculativa; gua vero theorica vel sgecul~tivil: qua vero in
sources, being to a large extent an expansion of a Latin translation of al- consideratur in materia dicitur materia. dicitur gractica vel activ~.
Farabi's On the Classification of the Sciences. The reference to the Liber gractica vel activa. Ac per hoc Et quoniam artis arimethice <est>
Algorismi in this text has already been quoted. The relation of Gundisalvi's theorica nullum numerum ex utroque modo de numero tractare,
work to the Liber mahameleth is even closer; for most of the opening of the numeris sed ex solis unitatibus ideo arimethica alia est theorica,
Liber mahameleth (after the first paragraph) corresponds verbatim with compositum esse asserit ... alia practica. 43
Gundisalvi's words: 41 <4> Partes vero gractice <4> Practice autem species multe
De divisione philosophiae, pp. 90-3 Liber mahameleth, principali ter due sunt, scilicet sunt, quoniam alia est s~ientia
Paris, BNF, scientia coniungendi numeros et cQniungendi numeros, alia
15461, fo1. 26ra
scientia disiungendi. Sed scienti~ disiungendi, alia est scientia
<1> Numeru§. ergQ de hiis est gue <1> Numerl,l.S ergo est de his gue negotiandi, alia est scientia occulta
coniungendi alia ~st scientia
utrQgue modo considerantur. in se l,l.trogue modo considerantl,l.r. per numeros inveniendi, et multe
in se aggregandi. alia dugljilndi. alia
scilicet et in materia. cum motu et scilicet et in materia.
Num~rus multiglicimdi; scientia vero alie. Illa autem que docet numeros
sine motu. Numerus enim in se
autem in se consideratur cum eiys disiungendi alijil est diminuendi. jillia ~oniyngere alia est awregandi. alia
consideratur cum eius essentia et naturJl vel prQPrieta§ tantym per §e
dividendi. Scientia vero rjildice§ duglandi. alia multigli~~ndi; que
nronrietas tantum att~nditur. sicut attengitur. sicut quod est par vel numerorum inveniendi sub utrague vero disiungere, alia est diminuendi,
cum accipitur abstractus ab omni impilr et cetera hyiusmodi alia mediandi,44 ali~ dividendi.
sensato ... In materia consideratur qu~ continetur. quoniam numeri radix
docentur in jilrimethi~a NicQmacm. utroqu~ modo invenitur. scilicet Scientia autem radices numerorum
cum attenditur ut aliquid eo
myltigli~imdo et dividendo ... inveniendi sub utraque continetur.
numeretur, quemadmodum utuntur
eo quoniam rildi~ utroqu~ modQ
in commerciis et in invenitYr. s~ili~et coniungendo et
negotiationibus secularibus. 42 Unde
disiungendo, quia myltiplicando et
secundum hoc alia accidunt ei ex se, minuendo.
alia ex commixtione materie. Ex se
enim accidit ei guod est 12.ar vel <5> Species vero practice sunt <5> Scientia vero negotiandi: alia
im12.ar. superfluus vel diminutus fi diversitates negotiationum, in est vendendi et em~ndi. alia ~st
cetera huiusmodi. que assignantur quarum unaquaque ars tota mutuandi ~t ~ccommodandi. ali~ est
in arithmetica Ni~homachi. exercetur, quarum alia est scientia conducendi et locandi. alia
vendendi ~t ~m~ndi. alia mutuimdi exp~ndendi et conservandi. et multe
<2> Ex ma~ria vero accidit ei <2>
In milteria vero consideratur et ~~~ommodimdi. ~lia e§t alie de quibus in sequentibus
aggregari et disgregagri, cum prout est <in> subiecto condy~endi et lo~andi. alia est tractabitur. Scientia vero per
multiplicari et dividi et huiusmodi attenditur, ut tres vel quattuor,
~xpendendi et conservandi; alia numeros occulta inveniendi et est in
que docentur in libro Algorismi. secundum quod ad multiplicandum predictis speciebus negotiandi et est
scientia est prQfunditates seu
et dividendum et cetera huiusmodi latitudines vel altitudines vel in inveniendo pondere rerum vel
humanis usibus suifragatur, quod quelibet alia spatia rerum profundita~ vel capacitate ex
docetur in arimethica AlcQrizmi et inveniendi. De quibus omnibus cognita earum longitudine vel
Mahameleth. sufficienter tractatur in libro qui latitudine vel econverso.
<3> IlIa vero consideratio au~ <3> IlIa autem arabice 'mahamelech' vocatur.
consideratio aua

41 Al-Farabi's words are in italics. Correspondence with the Liber mahameleth is

underlined. Significant divergences are put in bold.
42 Note that the Arabic equivalent of "in commerciis et in negotiationibus secularibus" is
"fi l-mu'amalat. al-suqi!a wa l-mu'amalat al-madiniyat" ("in commerce in the market place
and commerce In the Clty"): ed. A. Gonzalez Palencia, in al-Farabi, Catalogo de las ciencias,
2nd ed., Madrid, 1953, p. 54.
43 Cf. Domingo GUNDISALVO, De scientiis lGundisalvi's version of al-Farabi's On the
Classification of the &iences), ed. M. ALoNSO ALoNSO, Madrid and Granada, 1954,. p. 85:
"Ideo Arithmetica alia est practica, alia est theorica" (On p. 18 Alonso compares thIS text
with Gerard's translation of the same passage: "Quarum una est scientia numeri activa est
altera numeri speculativa").
44 'alia mediandi' occurs in Gundsalvi's De scientiis, ed. Alonso, p. 87.
VI
70
It is likely that the Liber mahameleth is prior to the De divisione
philosophiae, because in the former 'mahameleth' remains a technical term
and 'algorismi' is still used as the name of the Arabic author (according to
the most natural interpretation of the passage), whereas in the De divisione
philosophiae both words only appear as parts of titles of books-these books,
I would suggest, being the very books included in Paris, BNF, lat. 15461. For
sections <4> and <5> have no equivalent in al-Farabi's text, but have been
added by Gundisalvi not only here, but also in his version of al-Farabi's On
the Classification of the Sciences, where the editor, while ignorant of the
Latin Liber mahameleth, has intuitively suggested that the additional
passages "appear to have been taken from the treatise called in Arabic 'k. al-
c Postulates of the First and the Fifth Book of Eucl1.d,53 a ~ork. hItherto
mu amalat'''.45 On the other hand, the Liber mahameleth includes some
phrases from al-Farabi that are also in the'De divisione philosophiae. The
impression that one gets, therefore, is that the Liber mahameleth is written
either by Gundisalvi or by a close associate of his.46 If the latter, we are
inclined to conclude that another text should be added to Master John (of
Spain)'s authorship:
f) Liber mahameleth. 47
One of the principal sources of the Liber mahameleth is Euclid's
Elements, of which the second book in particular is exploited. Anne-Marie
Vlasschaert noticed that some of the formulaic phrases in the Liber
mahameleth corresponded with those used in MS Vatican, Reg. lat. 1268.48
This manuscript contains several versions of the Elements; the one in which
the correspondences can be found is a hitherto unedited text,49 whose
material is described by H. L. L. Busard as being borrowed from the versions
of the Elements known as 'Adelard Version 11', 'Hermann of Carinthia's
Version' and 'Gerard of Cremona's Version'.5o The knowledge by the
45GUNDlSALVI, De scientiis, ed. ALoNSO, p. 87: "en arabe no hay nada de esto; parece
tomado del tratado llamado en arabe K. al-mu'amalat". Alonso discusses this addition on pp.
21-3.
One may add to this the fact that the places mentioned in the Liber mahameleth are,
46
precisely, Toledo and Segovia (cf. SESIANO, "Le Liber mahameleth", p. 71): Segovia was the
diocese to which Gundisalvi's archdeaconate belonged.
47 An edition of the Liber mahameleth has been promised by Jacques Sesiano (see his "Un
recueil du XIlIe siecle de problemes mathematiques", in Sciamus 1 (2000), pp. 71-132 [po
130)). Another edition is included in the unpublished thesis of Anne-Marie VLASSCHAERT,
"Le 'Liber mahameleth'. Edition critique, traduction et commentaires", Universite
catholique de Louvain, annee academique 2002-2003.
48 A.-M. VLASSCHAERT, "Le cLiber mahameleth"', IV, p. 259. The phrases in question are
"verbi gratia" (introducing the example), "quod sic probatur" (introducing the proof), and "et
hoc est quod monstrare voluimus" (the conclusion of the proof).
49 This version occurs on fols 72r-113v and includes only Books 5 and 6, Book 10 and the
definitions and propositions 1-4 of Book 11 followed without a break by the rest of the
Elements in Gerard's Version.
50 H. L. L. BUSARD, The First Latin Translation of Euclid's Elements Commonly Ascribed
to Adelard of Bath, Toronto, 1983, pp. 11-12.
VI
John of Seville and John of Spain 71
author/compiler of the version thought to be by Hermann of Carinthia
(which survives in only one manuscript), and of that attributed to Gerard
(whose best copy is in the sister manuscript of Paris, ~NF,. lat. 1~461:
Vatican, Ross. lat. 579), might point, in particular, to Gundlsalvl, wh~ IS t~e
sole witness to another text by Hermann. 5l The use by the compIler, lD
addition to the Gerard Version of the Elements, of Gerard's translation of a
commentary to Book 10 by MuQammad ibn cAbdalbaqi,52 which is r~ely
quoted elsewhere, would also suggest that this Vatican version w~ compll~d
by someone very close to Gerard. Moreover, the compiler/author lDclu~ed lD
his text a large portion of al-Farabi's Explanation o( Jhe Problems. m the
unknown in Latin. Gundisalvi relied on al-Farabi's The Class1.ficat1.On of the
Sciences for his own De divisione philosophiae, and includes in the latter
passages from Arabic texts that are otherwise unknown i~ La.tin tr~slatio~,
such as a portion of the Shifa/ of Avicenna on De conven~nt1.a e.t d1.fferen.t1.a
scientiarum. 54 Thus one might be tempted to attribute this Vatican verSIon
of the Elements to Gundisalvi, or-again-his collaborator, Master John (of
Spain). But one must be cautious. For the form~laic phrase~ referred to. b~
Vlasschaert are also distinctive of the 'Gerard VerSIon', and the Gerard VersI~n
in turn appears to have been 'edited' by another author, whe~ compared ~lth
other translations securely attributable to Gerard; none of Its manuscnpts
include an attribution. 55 So one has also to ask who this 'editor' may have been. 56
Thus we have seven texts, all of which (except possibly for the lost
astronomical tables with their canons) are original compositions or
compilations, and most of which have some connection with Dominicus
61 Gundisalvi took over substantial portions of Hermann's De essentiis ("?thout
acknowledgement) in his De processione mundi: see BURNETT, "The Blend o~ La~n and *
Arabic Sources in the Metaphysics of Adelard of Bath, Hermann of Cannthia, and
Gundisalvus" (in the press).
62 H. L. L. BUSARD, The First Latin Translation, pp. 11 and 407-8.
53 For the extant Arabic and Hebrew versions of this text see Gad F'REUD~~, "La
philosophie de la geom~trie d'al-Firibi: Son commentaire sur le d~but du .pre~ll1er li~ et le
debut du cinquieme livre des Elements d'Euclide", in Jerusalem Studr,es. m . Arab&C a~
Islam 11 (1988), pp. 104-219. I am very grateful to Rildiger Arnzen for suggesting that this
text may be the source.
64 See also A. FIOORA, "Nota sobre Domingo Gundisalvo y el Ar~toteles Arab~", in Al-
Qan.ora 23 (2002), pp. 201-8, where it is argued that GundisalVI consulted directly an
Arabic text of Aristotle's Physics.
66 See A. A. BJORNBO, "Gerhard von Cremonas Obersetzung v?n Alkhwarizmis Algebra
und von Euklids Elementen" in Bibliotheca mathematica, 3rd senes, 6, 1905, pp. 242-5 (for
the original grounds for the' attribution) and P. KUNITZSCH, "Findings in Some Te:as ~f
Euclid's Elements (Mediaeval Transmission, Arabo-Latin)", in Mathemata: Fe~tBchnft fUr
Helmuth Gericke, eds. M. FOLKERTS and U. LINDGREN, pp. 115-28 where eVIdence for a
reworking of a translation by Gerard is given.
66 The Gerard and Vatican Versions of the Elements and the Latin tr~lation. of al- *
Farabi's Explanation of the Problems in Euclid, are the subject of a forthcommg article by
the present author.
VI
72
Gundis~l~i. !i0
precise dates are given, but their associations suggest
comp~sItIon In the 1150s (if not the late 1140s) and afterwards, including
the time of the ~ar~er o~ Gundisalvi, the earliest record of which falls in
1162. ~e a~socI~tIon WIth Gundisalvi has inclined Rivera (followed by
myselO, to Ide~tIfy Master John (of Spain), the author of these texts, with
the John of SpaIn who had succeeded Gundisalvi (perhaps directly) to the
archdeacon~te of Cuellar (Colarensis) by 5 March 1193. There is a rich
documentation conc~ming this John, who was also dean of the cathedral,
went on to become bIshop of Albarracin-Segorbe in 1213, and died in 1215.
3. John of Seville =Master John (of 'Spain)?
Now, it ma~ be ~empting to .combine into one person a prolific translator
for whom the hlstoncal record IS tenuous (i.e. John of Seville) with
homonym, 'John of Spain', for whom a more ample docume~tary
ce~tred on Toledo cathedral, can be provided. But if the mathematician and
:e:a::
~hIlosop~e~, Master ~ohn (of Spain), is the archdeacon of Cuellar who died
In ~215, It IS hardly lIkely that he can be John of Seville, who was already
active by 1128 at the latest. If the identification with the archdeacon is not
made, however, we may be inclined to see two stages in the career of one
man: a st~ge including a date before 1128, and 1135, in which translations
from ArabIC were m~de, apparently 'in Limia'; and a stage starting in the
late 114?s a~ the earhest, in which original works were written. In this case,
the attnbutIons of two of these works (items c and d) in one manuscript
e~ch, to John of Seville, may have some truth. One m~y be dealing with a
SIngle scholar whose activity extended from the late 1120s to the 1160s. This
sch~lar would have both a series of translations and several original works
~ his .name, and some points of attachment to figures well-known from the
hlstoncal record: Teresa (Tarasia) Queen of the Spains Ra d In fact, a 'Master John David of Toledo' had acquired an almost
Ar hb' h f ' , ymon ,
C IS op 0 Toledo, and Dominicus Gundisalvi, Archdeacon at the
~athe~r~l. T? preserve a connection with the cathedral, instead of
Iden~Ifying. hIm with John of Spain, Archdeacon of Cuellar, one might
conSIder hi~ to be th~ 'Johannes astronomicus' whose obit on the 3
Sept,ember IS reco:ded In t~e cathedral obituary, 58 and/or the 'Johannes
magister scolarum whose SIgnature occurs in two acts of the cathedral
57 See n. 22 above.
&8 See R.. GoN~~Z RUIZ, Hombres y libros de Toledo, vol. I, 1086-1300, Madrid, 1997
153 ~fernng to B1bhoteca Capitular, MS 42-30, fol. 105v). The obituary of course does ~:i
~eco th~ y?ar. of this John's death, but the fact that he is refe~ed to simply as
astr?nOmlCUs , wlth~ut ~y ~ther indication of his office, suggests that he must have had a
conSIderable reputatIon 10 th1s capacity.
VI
John of Seville and John of Spain 73
chapter of Toledo of 1174, beside the signatures of 'Dominicus Colarensis
archidiaconus' and 'Girardus dictus magister' (i.e. Gerard of Cremona).59
One point in favour of the identification of John of Seville with Master
John (of Spain) is that Gundisalvi made use of one text translated by John of
Seville-namely, the De differentia spiritus et animae of Qusta ibn Liiqa-in
his own treatise on the soul;60 he is one of the first scholars to use Qus!a's
text, and he exploits it for his original work just as he exploits the
translations of other Arabic texts in which he was involved, such as al-
Farabi's On the Classification of the Sciences and Ibn Sma's On the SOUl. 61
Another point is that, although Gundisalvi was based at Toledo cathedral,
his archdeaconate belonged to the cathedral of Segovia, and it was at
Segovia that Raymond, the dedicatee of the De differentia spiritus et animae,
was bishop before being elevated to the archbishopric of Toledo.
Also, it may not be true that the activity of John of Seville belongs to a
different period from that of Master John (of Spain). We simply do not know
the dates of most of the translations of John of Seville. Moreover, there is a
translation of an astrological text which could suggest that John Seville was
still active as a translator in the second' half of the twelfth century: namely,
Abii <Ali's De nativitatibus, whose colophon conforms to the formula
characteristic of the translations of John of Seville, and adds the date of the
translation: July, 1152 or 1153.62 The translator is named as 'Iohannes
Toletanus', and this could suggest that John of Seville, having earlier been
in 'Limia', had settled in Toledo by 1153. The evidence of this work needs to
be looked at carefully.
4. John of Seville + Master John (of Spain) =John Davld (of Toledo)?
legendary status by the early thirteenth century, since he was credited to
have represented "all the masters of Toledo" in predicting a general
59 This is the proposal favoured by Marie-Therese D'ALVERNY: see "Algazel dans l'Occident
latin", in Academie du royaume du Maroc, session de novembre 1985, Rabat, 1986, pp. 125-
46, reprinted in La Transmission des textes philosophiques, article VII, see p. 5.
m
60 See BURNETl', "'Magister Iohannes Hispalensis et Limiensis , pp. 260-4.
61 The other early user of Qusta's text was Hermann of Carinthia, whose writings, as we
have seen, Gundsalvi knew.
62 Oxford, Laud. Mise. 594, fol. 106: "Perfectus est Liber nativitatis mense Iulii anno ab
incarnatione domini millesimo .cliii., cum laude Dei et eius auxilio"; the translator appears
in the title on fol. 94r: "Albohali de nativitatibus tr. a Iohanne Toletano". The colophon in
London, British Library, Royal 12 C XVIII, fol. 10r gives 1152 as the date, and adds
"translatus a magistro Iohanne Toletano". For other manuscripts of this work see L.
THoRNDlKE, A History of Magic and Experimental Science, II, New York, 1923, pp. 75-6, and
idem and P. KIBRE, A Catalogue of lncipits of Mediaeval Scientific Writings in Latin, revised
edition, London, 1963, cols. 784-5. The text was printed as Albohali Arabis astrologi
antiquissimi ac clarissimi De iudiciis nativitatum, ed. Ioachim HELLER, Nuremberg, 1546.
The incipit is "Dixit Albohali, Iste est liber in quo exposui omnes significatores ... "
VI
74
conjunction of the planets in Libra 1229. 63 That this reputation arose from
the existence of a real 'John David' is shown by the fact that we find, already
in the 1130s or early 1140s, evidence for the celebrity of such a figure. For
Plato of Tivoli, working in Barcelona in this period, dedicated a translation
of an Arabic text on the astrolabe to 'his most serene friend, John David,
most skilled in the four disciplines of mathematics ... most zealous in the
science of the stars-nay rather in every science committed to script'. To the
same John David, Rudolph of Bruges, the pupil of Hermann of Carinthia,
dedicated another text on the astrolabe, which he wrote in Beziers in 1144.64
If we identify the personality John of Seville + Master John (of Spain), with
this John David (who is otherwise unknown), we can provide the
contemporary reputation that his works would seem to deserve. But caution
is necessary. 65
5. John of Seville + Master John (of Spain) + John Davld (of Toledo) ~
Avendauth
In the past, scholars have identified John of SevillelMaster John (of
Spain) with Avendauth,66 the only other collaborator of Gundisalvi whose
63 A gloss, written in ca. 1200 on fol. 112r of MS British Library, Add. 16606, begins:
"Omnibus ad quos presens scriptum pervenerit, magister lohannes David Toletanus et
omnes eiusdem loci magistri salutem, gratiam Sancti Spiritus et solamen ... nos cum
omnibus magistratibus atque consodalibus nostris peritisque astrologis Toletanis ... ":
D'ALVERNY, "Avendauth?" p. 31. The original prediction was made in 1179 for a conjunction
in 1186: see H. GRAUERT, Meister Johann von Toledo, Munich, 1901 (at p. 168), and G. DE
CALLATAY, "La grande conjonction de 1186", in Occident et Proehe·Orient: Contacts
scientifiques au temps des Croisades, eds I. DRAELANTS, A. TIHON and B. VAN DEN ABEELE,
Turnhout 2000, pp. 369-84.
64 The two dedications are printed and discussed in full in D'ALVERNY, "Avendauth?" pp.
29-32. Both texts have since been edited: R. LoRCH, G. BREY, S. KIRSCHNER, and C.
SCHONER, "lbn a~-~affiirs Traktat iiber das Astrolab in der Ubersetzung von Plato von
Tivoli", in Cosmographica et Geographica. Festschrift far Heribert M. Nobis zum 70.
Geburtstag, ed. B. FRITSCHER and G. BREY, Munich, 1994, 2 vols, I, pp. 125-80, and R.
LoRCH, "The Treatise on the Astrolabe by Rudolfof Bruges", in Between Demonstration and
Imagination: Essays in the History of Science and Philosophy Presented to John D. North,
eds L. NAUTA and A. VANDERJAGT, Leiden, 1999, pp. 55-101. Plato addresses his patron as
"suo serenissimo amico lohanni David in quatuor matheseos disciplinis peritissimo .. .in
astronomia, immo in omni literarum scientia studiosissimo", while Rudolph writes
"dilectissimo domino suo lohanni Dd." The two texts follow each other in MS Oxford,
Bodleian, Digby 51, a manuscript written under the supervision of a scholar who could
write Arabic, and who brought together texts on mathematics composed, for the most part,
in the mid-twelfth century in northeast Spain and southern France; this twelfth-century
scholar could well be a key player in our story.
65 It would, for example, seem odd for Plato to dedicate a new version of Ibn as-Saffiir's
treatise on the astrolabe to someone who had made his own version of the same te~' (see n.
11 above).
66 The identification was made by Moritz STEINSCHNEIDER, Die hebraeischen
Uebersetzungen des Mittelalters, Berlin, 1893, pp. 981-4, and followed, inter alios, by MILLAs
VALLICROSA (see n. 30 above).
VI
John of Seville and John of Spain 75
name is recorded: he introduced Ibn Sina (Avicenna)'s Shifa/ to a high-
ranking official, and translated the book On the Soul from it, with the help
of Gundisalvi. The identification rested on the fact that 'Avendauth' is a
transcription of the Hebrew for 'son of David' (ibn Daiid) and the supposition
that the dedication of the translation of On the Soul referred to the
translator as 'Iohannes Avendauth'. Marie-Therese d'Alverny, after a
comprehensive survey of the manuscripts of the translation, established that
the original reading of the dedication was "Reverentissimo Toletanae sedis
archiepiscopo et Yspaniarum primati Iohanni Avendehut Israelita
philosophus, gratum debitae servitutis obsequium", and that 'Iohanni' is
John, Archbishop of Toledo from 1152 to 1166.67 She also proposed that
Avendauth was the well-known Jewish scholar, Abraham ibn Daiid, who
fled from C6rdoba to Toledo in ca. 1160 and composed a text, heavily
indebted to Avicenna, called On the Sublime Faith. Since the time of her
article, this latter identification has found further support among those who
have studied the works of Abraham ibn Daiid. 68 Master John (of Spain) was
not involved in the translation of the Shifl and the characteristics of the
translations that he did collaborate on differ from those of the translations of
Avendauth. 69
6. John of Seville ~ John of Seville?
It has been taken for granted that John of Seville, aside from being a
translator of astrological texts, also wrote an original summary of the
doctrines and techniques of astrology which was printed in Niimberg, in
1548, under the title Epitome totius astrologiae. The manuscripts of this
work (of which there are over 30, and which refer to the work by the names
of its various books: namely an Ysagoge followed by four partes) agree in
attributing the text to John of Seville. The author of the critical survey of
texts on the science of the stars of the mid-thirteenth century, the Speculum
astronomiae, refers to each of the parts of the text as being by John of
Seville.70 Moreover, at certain points within the text, the 'present date' is
67 D'ALVERNY, 'Avendauth?', p. 32.
68 E.g. G. D. COHEN, A Critical Edition with a Translation and Notes of the Book of the
Tradition by Abraham ibn Daud, London, 1969, and T. A. M. FONTAINE, In Defence of
Judaism: Abraham lbn Daud: Sources and Structures of ha-Emunah ha-Ramah, Assen and
Maastricht, 1990, pp. 262-3.
69 D'ALVERNY, "Avendauth?", p. 41; cf. her conclusion: "Le mysterieux 'magister lohannes'
etait peut-~tre un clerc toledan, moins expert en arabe qu'lbn Daud, mais en revanche pl1;U'
proche de Gundisalvi par son education tant sacree que profane." Cf. also S. VAN RIET, m
her introduction to Avendauth and Gundisalvi's translation of Ibn Slni's De anima, I-Ill,
Leiden, 1972, p. 99*, n. 26: "Bien que l'archidiacre Dominique soit nomme comme l'un des
traducteurs dans l'incipit [d'Algazel, Metaphysique] .. .le vocabulaire de cette traduction
latine d'Algazel, m~me la ou i1 s'agit en arabe des m~mes mots, ne parait pas identique a
celui du De anima" (examples follow).
70 ZAMBELLI, The Speculum Astronomiae and its Enigma, pp. 218, 226, 230, 234 and 236.
VI
76
given as being an no domini 1142, falling nicely into the period of John's
activity, and this date is confirmed by the values given to the longitudes of
the fixed stars for that time.
Yet, there are serious problems in ascribing this text to John of Seville: 71
(a) no translation made by John is explicitly cited, but rather, texts by
Ptolemy, Dorotheus and Hermes; (b) the astrological terminology differs
radically from that used by the translator, especially in regard to the names
of stars and constellations, for which the Epitome finds Latin equivalents; (c)
the topographical allusions are not to Toledo or any other locality in Spain,
but rather to Italian towns, and especially to Pisa where the author implies
he has made direct observations of the stars; (d) the text is remarkably close
to the original astrological works of Abraham ibn Ezra, who was in the
vicinity of Pisa, and drew up instructions for astronomical tables for Pisa in
the 1140s and 50s. From similarities of style and terminology, a work on the
astrolabe attributed to a 'John' can be associated with the same author as
that of the Epitome. Thus we have two works that cannot easily be fitted
into the (Euvre of John of Seville, the translator:
a) John of Seville, Epitome totius astrologiae, written in 1142.
(J) a work on the construction of an astrolabe beginning: 'Dixit Iohannes:
Cum volueris facere astrolabium ... '72
7. Conclusions
A few observations may be made on the basis of the data in this article:
1) The scientific activity in Toledo cannot be considered in isolation from
that of other centres. Toledo was as much the place to which Latin
translations from Arabic and original Latin texts were brought, as the
locus from which they were diffused. The evidence for a translator called
'John of Seville' at Toledo is weak and his early career was probably
spent in northern Portugal; Gundisalvi knew, and probably brought to.
Toledo, works by translators active in northeast Spain (e.g. Hermann of
Carinthia) and Latin works by Thierry of Chartres and his school; the
works of Master John (of Spain) show the knowledge of texts by Adelard
of Bath and his pupil Ocreatus, as well as translations made in
northeast Spain and coinage distinctive of Barcelona (the morabatini).
2) One must be prepared for a considerable difference in style, and perhaps
vocabulary, between translations, on the one hand, and original works,
on the other, written by the same author. The divergences between the
71 These problems are being dealt with in detail in a separate study by me. The relation of
the text with the astrological texts of Abraham ibn Ezra in Hebrew is being investigated by
Renate Smithuis.
72 Partial edition by MILLAs VALLICROSA, Las traducciones orientales, pp. 322-7 = 'JU'in
KUNITZSCH, Glossar der arabischen Fachausdrilcke (n. 10 above), p. 488.
VI
John of Seville and John of Spain 77
Epitome totius astrologiae and the translations of John of Seville are
probably too substantial to be accounted for as the responsibility of a
single scholar, but the presence of "good" Latin in the original works of
Master John (of Spain)73 need not rule him out as the man responsible
for the awkward literal translations of astrological texts.
3) One has also to recognise that authorship is a fluid concept. The literal
sense of an Arabic text may have been interpreted by one man (who may
remain anonymous) and put into good Latin and/or polished up
scientifically by another (who may put his name to the translation). On
the other hand, yet a third person may write an original work, based on
translations commissioned or made by himself (as Gundisalvi appears to
have done). Also, a more paraphrastic translation may be revised to
make it more literal, as William of Moerbeke was to do, in his revisions
of his own translations from Greek in the late thirteenth century.
4) Early manuscripts of texts translated from Arabic tend to show
considerable variation from each other: e.g., terminology is changed,
alternative translations are given, or the Latin style is improved. It is
often difficult to decide which is the original translation, and whether
the changes are due to the translator himself or to other known, or
unknown, scholars.74 The later dates attached to individual manuscripts
of John of Seville's translations-1171 for item 6 and 1181 for item 8
respectively-may be dates of revision, rather than composition. 75 It is
evident that the 'Johns' worked in an atmosphere where recourse to the
Arabic originals was always possible, and where there was lively debate
concerning the contents of the texts. Thus many small changes could
have been introduced into copies, or as marginal notes.
5) Attributions in manuscripts must be analysed very carefully for
evidence that they are integral to the text. For, frequently a title and
ascription has been added by a later hand; the ascription in a colophon is
often more reliable, but may still be the invention of the scribe of the
text. Ideally, the evidence of the incipits, explicits and colophons of every
For example, in the Liber algorismi.
73
For a detailed analysis of one such translation see BURNETT, "The Strategy of Revision
74
in the Arabic-Latin Translations from Toledo: The Case of Abii Ma'shar's On the Great
Conjunctions", in Les Traducteurs au travail: leurs manuscrits et leurs methodes, ed. J.
HAMESSE, Turnhout, 2001, pp. 51-114. The case of al-Qabi~rs Introductorius, whose edition
is in progress, is considerably more complex, owing to the number of manuscripts that
preserve early versions of the text.
75 LEMAY (IV, pp. 120, 218-75), claims that 1171 is the date of the revision of the
translation of Abii Ma'shar's Liber introductorii maioris, which he attributes to Gerard of
Cremona. A similar process of revision can be discerned in the translation of Abii Ma'shar's
De magnis coniunctionibus, but the involvement of Gerard in both cases can only be
determined by comparing the distinctive elements of the revision with Gerard's own
translations, which has not yet been done.
VI

78
VII

copy of a text should be taken into consideration, so that the stages in

the genesis of common (mis)attributions can be documented. 76
The hunt for the identity of John of Seville and Master John (of Spain)
can only bring us to a certain point, beyond which further progress cannot be
made, even when the manuscripts containing the works attributed to these
two men have been studied, critical editions have been completed, and the
language employed in them has been analysed. But this quarry should not
be the sole aim of the quest. For much of the pleasure lies in the chase itself
which reaps its own rewards. Through making detailed descriptions of
manuscripts, critical editions and comprehensive glossaries, one can discern
common features in terminology, vocabulary and style, see how one work
draws material from another, and witness the struggles of scholars to make
sense of difficult subject matter. Piece by piece one can build up a picture of
the intellectual and social culture of the societies in which Arabic science
was being rediscovered in the West.
Addendum (July, 2003):
In a survey of over 200 Latin manuscripts of John of Seville's
translation of Alcabitius's Introductorius (to be published in Al-Qabisi
[Alcabitius), Introduction to Astrology, the Arabic and Latin texts, with ~
English translation, edited and translated by Charles Burnett, Keiji
Yamamoto, and Michio Yano, London, 2003), the majority of the
manuscripts attribute the translation to John of Seville (lohannes
[h]ispalensisl[h)yspalensis); one manuscript alone (Paris, Bibliotheque
nationale de France, nouv. acq. lat. 1893, fo1. 109rb), 'atque lunense'. Twenty
manuscripts, however, substitute '(h)ispan(i)ensisl(h)yspan{i)ensis' for
'hispalensis'. The form '(h)ispanus' is never found.
A calendar date, but no year, is included in the colophon of one
manuscript of CUmar ibn al-Farrukhan at-Tabari (Omar), De nativitatibus:
Florence, Biblioteca nazionale centrale, con. soppr. J.II.10, s. xiiiex, fol. 140r:
perfectus est universus liber Omar Benfargan Tyberiadis de nativitatibus et
perfecta est eius translatio nona die mensis iu(lie), feria 6 cum laude dei et
eius auxilio amen.
76 A good model of what is needed when recording titles and attributions is provided by
Agostino PARAVICINI BAGLIANI in Le Speculum Astronomiae, une enigme?, Micrologus'
Library 6, Tumhout, 2001.
The Coherence of the Arabic-Latin Translation Program in
Toledo in the Twelfth Century*
Argument
This article reassesses the reasons why Toledo achieved prominence as a center for Arabic-
Latin translation in the second half of the twelfth century, and suggests that the two principal
translators, Gerard of Cremona and Dominicus Gundissalinus, concentrated on different areas
of knowledge. Moreover, Gerard appears to have followed a clear program in the works that
he translated. This is revealed especially in the Vita and the "commemoration of his books"
drawn up by his students after his death. A new edition of the Vita, Commemoratio libronlm and
Euiogium, based on all the manuscripts, concludes the article.
Toledo is justifiably famous as the principal center for the translation of Arabic
scientific and philosophical texts into Latin. Several factors contributed to its
preeminent position. One was the linguistic mix of its population. When Alfonso VI
of Castile captured Toledo from the Arabs in 1085, the city capitulated without
bloodshed and its inhabitants were allowed to stay and to keep their possessions and
privileges; Alfonso declared himself"the king of the two religions." Nevertheless, we
are told, most of the Islamic elite emigrated, while the common people converted to
Christianity in great numbers (Rubiera Mata 1991, 75-91). The Jews in the city
stayed put, though they were subject to periodic pogroms. However, the most
significant element in the population was the Mozarabs, the "Arabized" Christians
who had preserved the liturgy of the Visigothic church and whose numbers were
augmented by the Islamic converts. The bulk of the population, therefore, spoke both
Arabic and a Romance dialect, and Arabic was the language of religion and culture.
A second factor was Toledo's preeminence as a center of scientific learning even
before the capitulation to Alfonso. In Islamic Spain, after the breakup of the caliphate
in 1031, the kingdom of Toledo under the Banii Niin was rivaled as a center of
learning only by Saragossa under the Banii Hiid. The cadi of Toledo, Sacid al-
Andalusi (1029-70), wrote a history of science (The Categories of the Nations), and also
patronized scientific research, most notably that by az-Zarqilliih, who compiled
*I am very grateful for the help of Danielle Jacquart, Richard Lorch, Peter Linehan, and Patricia
Stirnemann.
VII
250
astronomical tables and other practical works on the science of the stars (Jacquart and
Micheau 1990, 135-37). The departure of the Islamic elite may have prevented this
scientific tradition from developing. It is likely, however, that scientific expertise and,
even more so, books, remained among Arabic scholars in Toledo. One indication of
this is that the translator Gerard of Cremona at the beginning of his career (and
therefore perhaps already in the late 1130s) was attracted to Toledo because he knew
that he would find there Ptolemy's Almagest (in Arabic). Another indication is that,
some time before 1140, az-Zarqilluh's tables were rendered into Latin as The Toledan
Tables. The drawing up of astronomical tables for a particular place was often
associated with an important political event, and the coronation of Alfonso VII as
Emperor in 1135 would have presented a suitable occasion for a new version of the
tables. 1
That it is not until the 1130s that we have any evidence in Toledo of an interest in
translating Arabic texts into Latin is not surprising, for translations can only be made
if there is an interested audience who do not know the original language of the texts.
The only such audience in Toledo in the decades immediately after its conquest were
the Cluniac clergy brought in by the French archbishops, Bernard of Sedirac
(1086-1125) and his successor Raymond de La Sauvetat (1125-52), and their
primary interest was in reforming the Church rather than in advancing scientific
learning. Moreover, they were hostile to the indigenous Mozarabic parties who could
have introduced them to Arabic science. Nevertheless, it is in the context of the
Cathedral, as the only influential segment ofToledan society who did not understand
Arabic, where one must look for the beginning of the translation movement in
Toledo.
The first evidence of this is a translation of Qus~a ibn Luqa's On the difference between
the soul and the spirit, by John of Seville and Limia, dedicated to Raymond de La
Sauvetat. John also dedicated a translation of the regimen of health from "Aristotle's"
advice to princes (Secretum secretorum) to a person of importance, Queen Teresa, the
natural daughter of Alfonso VI and first ruler of the kingdom of Portugal. John was
perhaps casting around for patronage; it would have been natural to turn to Queen
Teresa because he himself seems to have originated from Portugal where he
completed other translations. 2 His dedication of a text to Raymond represents an
attempt to find favor in another quarter. The text was wisely chosen. It is short and
easy to read, and treats of a subject that is relevant to theology; for it puts into context
the prominence given to "spirits" in the new medical learning taken from Arabic texts
,in Italy, which posed a threat to Christian doctrines on the immortal soul.
1The date and authorship of the Latin version of the Toledan tables remains unknown. A possible use of the
Toledan tables in Aragon in February 1106 is discussed in North 1995. However, the form in which they were
known at this date is not clear. The earliest clear use of the Latin tables is by Raymond of Marseilles who
adapts them to the meridian of Marseilles in 1141. Note also that the author of the Almagestum parvum, which
appears to belong to the mid-twelfth century, speaks of the tables as "very recently composed" (see Lorch
1995, V, 410: "et super hoc arzacel tabulas motuum toleti novissime composuit").
2 For his biography, and translations made "in Limia" (a region in Northern Portugal), see Burnett 1995a.
VII
The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 251
Whether this dedication ensured John a place among the cathedral clergy is not
known. The earliest substantial use of the translation itself was made shortly afterwards
by an archdeacon in the cathedral, Gundissalinus (to whom we shall return), but
John's dedication remains an isolated testimony to an Arabic-Latin translation
destined for Toledo (aside from the astronomical tables) in the first half of the twelfth
3
century. Even this translation may not have been made in Toledo, and it dates to a
period in which translations were being made in other parts of Spain, and in
particular in the valley of the Ebro where the remnants of the kingdom of the Banu
Hud were still in power (see Burnett 1977 and 1992,1041-44).
The situation changes towards 1150, when Toledo becomes the principal center for
translations. Various factors may account for this. First, in 1140 the last of the Banu
Hud, Jacfar Alpnad III Sayf al-Dawla, whose library had been used by the translators
of the valley of the Ebro, exchanged his property in Rueda Jal6n (on a tributary of
the Ebro) for part of the city ofToledo. 4 We know that the royal library of the Banu
Hud was particularly rich in works on mathematics, astronomy, astrology, and magic,
and the texts on geometry that Gerard of Cremona chose to translate correspond to
5
those used by one of the kings of the dynasty in the late eleventh century. Secondly,
the rise of the intolerant Islamic regime of the Almohads in North Africa and their
spread to Islamic Spain in 1147 forced Arabic Christians (Mozarabs) and Jews to
emigrate, and Toledo was the natural place for them to choose as their new home.
Moreover, under archbishop John of Castellmoron (1152-66), there was much more
communication between the Mozarabic community and the Frankish clergy than
during the first decades after the conquest. Thirdly, the continual arrival of Frankish
clergy (which included nationals from several European countries), and the
development of a Frankish quarter in Toledo ensured that translations from Arabic
both would receive an audience locally and could easily be conveyed abroad.
Among the exiles from the Almohads was the Jewish philosopher, Abraham ibn
Daud. He had fled from Cordoba, and had settled inToledo by 1160, where he wrote
works in Arabic and Hebrew on philosophy and astronomy and the history of the
Jews in Spain. He is, in all likelihood, the "Avendeuch Israhe1ita" who wrote a letter,
in poor Latin, addressed to some important person, advertising the fact that he
intended to translate the Shifii', the philosophical encyclopedia written by Avicenna
(d. 1037); he added a specimen of his translation to his letter (see Birkenmajer 1970,
3 The work was also known at an early date in Salernitan circles; see Jacquart 1988, 426.
4 According to the Arabic historian Ibn al-'Abbar, he was given "half the city of Toledo"; see Encyclopedia of
Islam, s.v. Hiid. Gonwez Palencia considers Sayfad-Dawla (Zafadola) to be one of the very few Muslim elite
who remained in Toledo: see Gonwez Palencia 1926-30, I, 151-53.
5 The sources of the comprehensive book on geometry, al-Istikmal, written by Yiisuf al-Mu'taman ibn Hiid,
king of Saragossa from 1081 to 1085, include Euclid's Elements and Data, the De spheris ofTheodosius,
Menelaus, the Conics of Apollonius, Archimedes' On the Sphere and Cylinder, Eutocius' commentary on that
work, Thabit ibn Qurra's treatise on amicable numbers and Ibn al-Haytharn's Optics; cf. Gerard's translations
in "geometria;' Appendix I below, nos 4, 5, 8 and 16. Gerard's version ofTheodosius' De spheris belongs to
the same family as that used by al-Mu'taman; see Lorch 1996, 165, 172.
 VII
252
, The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century
95-100). It is probable that the important person was Archbishop John, and that this
"letter of introduction" achieved its purpose. For, the same Avendauth secured the
help of the archdeacon who knew John of Seville and Limia's work Dominicus
Gundissalinus, and togethe~ they translated a complete book of the Sh.lfcr - the one
on the soul. Avendauth dedicated the translation to Archbishop John in the following
terms:
To Jo~n, th~ most reverend archbishop ofToledo and primate of Spain, Avendauth, the
Israehte phIlosopher, gives hommage, recognizing the debt that is due to him. . ..
Th.erefore I have attempted to put into effect your order, Lord, to translate the book of
AVlcenna t~e p~ilosopher concerning the soul. ... Thus you have the book, translated
from ArabiC, With me taking the lead and rendering each word in the vernacular
lan~age, .an? archdeacon Dominicus turning the words into Latin. (D'Alverny 1989,
repnnted m Idem 1994, article Ill, 195)
This dedication d.escri~es w?at became a common practice: that a Mozarab or Jew,
who was not profiCient m Latm, made an intermediate translation of a text into the
vernacular language (whether the colloquial Arabic of Toledo or the local Romance
di~ect), which a cleric educated in the Latin schools transferred into good Latin
(Villanueva .1996, 23-34) . I~ also implies that the project of translating Avicenna's
work . wa~, If not comffilsslOned, at least supported by the archbishop himself.
~undissalmus, who presumably came from Old Castile and may have been educated
m ~he Fren~h schools, was particularly interested in psychology and cosmology,
6
~hlch. l:d him to translate further Arabic texts on these subjects, and to use these in
hiS ongmal works. Avendauth's name does not appear again as a collaborator, but
rather that of "Iohannes Hispanus." Some scholars have made one person out of these
two collaborators, but it is possible that "Iohannes Hispanus" was a Mozarab and the
same . as .the "Iohannes Hispanus" who was dean of Toledo, who succeeded
Gundissalmus as archdeacon ofCuellar, and died in 1215. 7
No dedication exists associated with any translation of the greatest of the Toledan
tran~lators: Gera:d of Cremona (1114-87), to whom over seventy translations are
asc.nbed, m subJe~ts ranging from mathematics, through medicine to Aristotelian
philosophy. There IS, however, evidence that he was a member of the clergy of the
cathedral. He attestated three documents, in 1157,1174, and 1176, as a canon of the
cathe~~, and a~er ~s death in 1187 his students or colleagues (sod') wrote a
eulogistic poem m which they called him "the glory of the clergy" (gloria elen), and
stated that, although he was born in Cremona, he lived and died in Toledo. 8 The later
documents (of 1174 and 1176) append to Gerard's name the words" dictus magister"
6 See below, p. 264.
: The identity of'Iohannes Hispanus' is explored in Burnett 1994.
See p. 256 and Appendix I below.
VII
253
"called the Master." The implication of this sobriquet is not clear, but it is probably
a term of approbation like calling Aristotle "the Philosopher" or Averroes "the
Commentator"; it is certainly not an official title for a teacher in a school or
college. 9
The patronage of an archbishop and the participation of an archdeacon continued
in the next generation of translators. Mark of Toledo and Michael Scot were both
canons of the cathedral (as Gerard had been) at the turn of the twelfth to the
thirteenth century. Michael accompanied the archbishop, Rodrigo Jimenez
(1208-47) to the Fourth Lateran Council in Rome in 1215. He continued the
translations of Aristotle's works begun by Gerard, and added those of commentaries
by Averroes (d. 1198), Gerard's near contemporary in Cordoba. Mark's main interest
was in medicine, but on the request of Mauritius, an archdeacon of Toledo, he also
translated the Coran and the profession of faith of the founder of the Almohad
movement, Ibn Tumart. Michael Scot left his canonship in Toledo some time before
1229 (Hernandez 1995, 68), and ended his life working for Frederick 11
Hohenstaufen in Sicily. But Hermann the German continued the translation of
Averroes' commentaries, one of which he completed in Toledo on June 3, 1240, "in
the chapel of Saint Trinity." This has recently been identified with the monastery of
St. Trinity, in the Frankish quarter next to the Cathedral, which had been founded
a little after 1195 specifically for rescuing Christian captives in Islamic territory. Since
the Brothers were taught Arabic so that they could negotiate with Islamic authorities,
it is quite likely that Hermann found linguistic help there (Gonzilez Ruiz 1996.
51-64, and Gonzilez Ruiz 1997, 586-602).
At least until the departure of Michael Scot for Italy, then, the translation activity
was associated with the cathedral rather than with any other institution in Toledan
society. Until 1180 (the date of the death of Cerebrun of Poitiers) the archbishops
were French, and the cathedral chapter remained predominantly Frankish until the
early thirteenth century (see Hernandez 1996). Some members of the local
community participated in the translations. We are told of a Mozarab called
"Galippus," who helped Gerard translate Ptolemy's Almagest, and of a Jew "Abuteus"
who helped Michael Scot translate a text on cosmology. Nevertheless, the direction
of the translation enterprise remained preeminently in the hands of foreigners, and
was an export commodity, rather than one for the local community, who, for the
most part, could not read Latin. Gerard of Cremona himself probably kept in contact
with Italian centers; one report states that his books were returned to Cremona after
his death, and to three manuscripts of a translation of a work on the calendar,
probably made by him, is added a horoscope cast in Cremona on 23 March 1191. 10
The earliest collection of his translations is an Italian manuscript written in the late
9This is also the interpretation in Ricklin 1995, 81.
10MSS Cambrai 168/163, fol. 103v, Vatican, Reg. Lat. 1285 and Vienna, Ostereichische Nationalbibliothek,
5463 (the work is the Liber erarum, a short text based on Hebrew chronology): see Burnett and Yamamoto
2000, n, xxiii.
VII
254
twelfth century.ll Scholars came from several countries to Toledo to seek out texts and
copy manuscripts. A "Thaddeus" came from Hungary and copied a manuscript of
Gerard of Cremona's translation of the Almagest in 1175;12 a Frenchman (possibly
Roger de Fournival, the court astrologer of King Philippe Auguste) copied the same
text, using local - i.e. Toledan - parchment. 13
Typical, perhaps, is the itinerary of the Englishman, Daniel of Morley, who relates
that, disappointed in the kind of studies that were being pursued in Paris, and hearing
"that the doctrine of the Arabs, which was devoted almost entirely to the quadrivium,
was all the fashion in Toledo in those days," went there and both disputed with Gerard
of Cremona about the validity of astrology, and learned "the doctrine of the Arabs"
from Gerard's assitant, Galippus, "in lingua Tholetana" (i.e., in the local Romance
dialect). He probably did not stay there long, however, but rather, on his own
testimony, brought books back with him to England. 14 The predominance of this
"export market" for the translations explains, and is explained by, the fact that no
university developed in Toledo itself. There was not sufficient local interest or
clientele for a large number of students and teachers to form themselves into a
corporate university body, as was happening in Paris, Bologna, and Oxford. Most of
those who were interested in Arabic learning had their roots elsewhere and wished to
benefit the countries or centers from which they originated. The program for
translation was, to a large extent, determined by what was required in the newly
burgeoning European universities, which were outside Spain.
What was this program? We get some idea of it from the report of Gerard's students
(SOciI) who, after his death in 1187, drew up a list of his works, accompanied by a brief
account of his life, summarised in a poem, and attached this to his last translation, the
Tegni of Galen with the commentary of cAll ibn Ric;iwan. The text may be translated
as follows:
Just as a lit candle should not be put in a secret place or under a bushel, but must be raised
up on a candlestick,15 so the glowing deeds of good men should not be left unspoken of,
as if buried under silence and neglect, but should be presented to the ears of the people
of today (modernt), since they open the door of virtue to those coming afterwards, and
the examples of the ancients, worthily commemorated, as it were instil an ideal image of
life into the eyes of those now living. Lest, then, master Gerard of Cremona lie hidden
11 Paris, BNF, lat. 9335; see d'Alverny 1982, 458-59, reprinted in idem 1994, article 11. For the dating of the
manuscript see Lorch 1995, article 11, 71.
12 MS Florence, Biblioteca Medicea Laurenziana, 89, sup. 45; see Kunitsch 1986-91, 1,16.
13 I owe the last detail to Patricia Stirnemann; the manuscript is Paris, BNF, lat. 14738.
14 Daniel ofMorley, Philosophia, ed. G. Maurach, MittellateinischesJahrbuch, 14, 1979,204-05 (at p. 212); the
relevant passages are reproduced and discussed in Burnett 1995b.
15 Luke 11,33: "Nemo lucernam accendit et in abscondito ponit neque sub modio sed supra candelabrum."
The following translation owes much to McVaugh (in Grant 1974, 35) and translates the text edited on pp.
275-6 below.
VII
The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 255
under the darkness of silence, lest he lose the favour of the renown that he has merited,
lest through presumptuous theft an alien heading be affixed to the books translated by
him - especially since he himself inscribed none of them with his name - all the works
translated by him, as much those on dialectic as those on geometry, as much those on
astronomy as those on philosophy, as much also those on medicine as those on other
sciences, have been listed very carefully by his students (sociI) at the end of this Tegni,
translated by him last (or most recently) - imitating Galen in commemorating his own
books at the end of the same work - so that if anyone who is an admirer of their aims
is looking for one of his works, through this list he might find it more quickly and
become more confident about it. For although Gerard spurned the glory of fame,
although he fled fawning praises and the empty pomp of this world, although he refused
to allow his name to be spread around by clutching at clouds and vanities, nevertheless
the aroma of the fruit of his works, diffused through the centuries, announces and
declares his goodness.
Although he flourished also with temporal goods, his mind was not elated or
depressed by the abundance or absence of those goods, but in a manly way faced good
and bad turns of fortune alike, and always remained in the same state of constancy. An
,enemy to the desires of the flesh, he adhered to spiritual values only; he laboured to
benefit all present and future generations, not unmindful of those words of Ptole my: "Do
even better when you approach the end oflife.,,16
Although from his very cradle he had been educated in the lap of philosophy and had
arrived at the knowledge of each part of it according to the study of the Latins (Latinorum
studium), nevertheless, because of his love for the Almagest, which he did not find at all
amongst the Latins, he made his way to Toledo, where, seeing an abundance of books in
Arabic on every subject ifacultas) and, pitying the poverty he had experienced among the
Latins concerning these subjects, out of his desire to translate, he thoroughly learnt the
Arabic language, and in this way, trustworthy in each - i.e., the subject-matter (scientia)
and the language (as A~mad in his letter On Ratio and Proportion says, "It is necessary that
the interpreter, in addition to the excellence which he has acquired from the knowledge
of the languages from which and into which he translates, should also have knowledge
of the subject (ars) which he translates"),17 in the manner of a prudent man who, walking
through green meadows, weaves a crown from flowers - not from all of them, but from
the more beautiful - he read through the writings of the Arabs (scriptura Arabica) , from
which he did not cease until the end of his life to transmit to Latinity, as if to a beloved
heir, in as plain and intelligible way as was possible for him, books of many subjects
16 This is one of the "sayings of Ptole my" from the section of Abii 1-Wafa' al-Mubashshir ibn Fatik's Mukhtar
al-bikam devoted to Ptolemy; one may compare the quaint English translation by Scrope in Biihler 1941, 224:
"The nerer that thou arte dethe pe more pou shuldiste travaile to do wele". The whole of the biography and
a selection of the sayings of Ptolemy appears in the preface of Gerard's translation of the Almagest; see
Kunitzsch 1974, 98-99.
17 The context and the wording of this quotation as it occurs in the copy of ~d's De proportione et
proportionalitate in MS Paris, BN 9335, fo1. 95vb is the following: "Possibile enim est ut verba hic translata
'proportionis minutionem' in Greco sint significantia, sed in linguam Arabicam non sunt in suo loco translata.
Locutionum namque ordo in duabus linguis est inequalis. Hec autem est habitudo eius qui non perfecte
transfert. Oportet enim ut interpres preter excellentiam quam adeptus est ex noticia lingue de qua et in quam
transfert, artis quam transfert scientiam habeat." For a discussion of this passage see Burnett 1999.
VII
256
1
!
ifacultates) - whatever he esteemed as the most choice. He went the way of all flesh in
the seventy-third year of his life, in the year of our lord Jesus Christ 1187.
These are the names of the books that he translated.
The list follows, after which there are eight lines of verse: 18
Gerard, fount, light and glory of our clergy,
author of good counsel, hope and consolation of the poor,
was an enemy to fleshly desire, but praised spiritual values.
His brightness was that of the inner man.
The deeds of the man preserve his life as long as scholarship flourishes.
The books which he translated adorn his living fame.
Cremona boasts that she has given birth to this sans pareil.
He lived at Toledo. Toledo returned him to the stars.
The socii not only knew the titles of the works translated by Gerard, but also had
a good knowledge of their subject-matter. For, in the Vita, they quote from two of
them: Ptolemy's Almagest and Apmad ibn Y lisuf's On Ratio and Proportion. 19 Moreover,
they add. certain details about the texts in the list of works that they append: that
Gerard did not translate the second book of the Pseudo-Aristotelian work De causis
proprietatum et elementorum because he did not find a complete text in Arabic; and that
he only translated the first three books of Aristotle's Meteora because he certainly
would have known (as the socii knew) that the fourth book had already been
translated. They added notes on ar-RazI, az-ZahrawI and Ibn SIna, in the first case
mentioning the book al-1:lawf, which Gerard had not translated. 20
They also had some idea about the place in scholarship of Gerard's translations. For
t?ey have classi~ed the works according to dialectic, geometry, astronomy, philosophy
(I.e. natural phdosophy and metaphysics), and medicine, with some miscellaneous
texts at the end.
18 The verses are in rhyming hexameters. Lemay claims that the Vita, Commemoratio librorum and Eulogium were
not ~i~en at the same time and by the same people (see Lemay 1978, 173-74); his arguments are not
convIncIng.
19 T~~se quotati~ns are s~gnifica~t since the first is from a biography of Pto lemy which emphasizes his moral
qualitIes and which proVIdes a kind of model for this VJ'ta of his successor Gerard' the second is from a work
which points out the necessity of the study oflogic for mathematics. ' the subject of his De eodem et diverso (Willner 1903; Burnett et al. 1998). He himself
20 .I~ the cas~ o~Abii :l-.~as~ az-Zahciwi they were aware that the Surgery was only one part of his vast work
K,tab at-t~rif l,-man aJlza an at-tasnif, but they gave to this work part of the name of the author himself
("azaugui" < "azaragui" = az-Zahciwi). This is confirmed by the explicit of the translation (cited by Leclerc
1876, 423, from Paris, BNF, lat. 7127): "Hunc librum transtulit Magister Gerardus Cremonensis in Toleto de
arabico in latinum, et est tric~sima particula libri Azaragui quem composuit Albucasim." According to Michael
Scot,. the Canon too was sald to be known by the name of its author, Avicenna: Liber introductorius, MS
Mumch, clm 10268 fol. 19r. Al-ljawf (Continens) was not translated until the second half of the thirteenth
century by Faraj ben Silem in Sicily.
VII
The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 257
This classification is not haphazard. The socii have chosen first the subjects of the
seven liberal arts that provided the framework for traditional education in the secular
sciences among the Latins. The fact that these seven arts, which were the parts of
"philosophia," were the (supposed) curriculum of Classical Antiquity (and especially
Greek Antiquity) is significant. The Latins were aware from Boethius and Martianus
Capella and other authors of late antiquity that a complete education (enkyklios
paideia) consisted in the arts oflanguage - grammar, rhetoric, and dialectic - and the
arts of things - arithmetic, geometry, music, ,and astronomy. In all these arts the Latins
oflate antiquity had Greek models, and attempted to build up a body of texts in their
own language that would substitute for those Greek models. The arts oflanguage, or
trivium, were well provided for by the works of Donatus, Priscian, Cicero, Quintilian
- and Boethius. Boethius (ca. 480-524/5) succeeded in transmitting to the Latins
several translations of, and commentaries on, Aristotle's works on dialectic. But he
also intended to translate or adapt the basic Greek texts on each of the subjects of the
quadrivium into Latin. He got as far as translating Nicomachus' Introduction to
Arithmetic and writing a textbook on music based on the work of the same author and
further texts by Euclid and Ptolemy. It appears that he started to translate Euclid's
Elements but did not get further than the fifth book, and only translated the proof of
the first theorem. Finally, Cassiodorus attributes to him a translation of a work by
Ptolemy, but, if he did make such a translation, nothing remains of it (see Pingree
1981,155-61).
These, then, are the Latinorum studia that Gerard was brought up on from his
cradle. According to his socii Gerard was aware of the gaps in the Latinorum studia -
Latinorum penuria ("the poverty of the Latins") - just as his fellow translator,
Burgundio of Pisa, was. 21 There was no need to translate anything on grammar or
rhetoric, theoretical arithmetic or music, because the Latins were well supplied with
textbooks on these subjects. The main gaps were the remaining parts of rhetoric and
dialectic, geometry, and astronomy. The textbooks for these were known both
through their being mentioned by Boethius and Cassiodorus, and, in the case of
Aristotelian rhetoric and dialectic, in al-FarabI (Wt; shall come back to this). One can
see from the beginning of the translating movement in the twelfth century that it was
the aim of the translators to fill in these gaps.
For example, Adelard of Bath, in the early years of the century wrote a book in
which he outlined the subject-matter of the seven liberal arts, emphasizing their
mutual dependence and how they are embraced under the term "philosophia": this is
translated Euclid's Elements and some texts on astrology and astronomy. A manuscript
The phrase "penuria apud Latinos" is used in Burgundio of Pisa's prologue to his translation (from Greek)
21
of the commentary of St John Chrysostom on the Gospel ofStJohn; see Classen 1974, 84. For the Latins'
inadequacy in geometry in particular see the statement of Stephen the Philosopher in the prologue to the
fourth book of his Liber Mamonis: "et ap < p > robata argumentis quorum latinitas inscia in divulgato diu
multumque volutatur errore"; see Burnett 2000,58.
VII
258
written by English and Norman hands in ca. 1140 (now Oxford, Trinity College, 47)
adds Adelard's translation of Euclid's Elements to Boethius' translations of Aristotle's
rhetorical and logical texts, and Boethius' texts on arithmetic and music, to make an
"up-to-date" textbook of the liberal arts.
The most thoroughgoing example of this process before Gerard of Cremona,
however, is the Heptateuchon ofThierry of Chartres, the two-volume library of texts
on the liberal arts, with an introduction concerning the importance of marrying
science with philology (drawing on Martianus Capella), which was probably put
together in Chartres in the early 1140s, and which provides a nice counterpoise to the
sculptures of the seven arts on the Portail Royal of Chartres Cathedral, dating from
the same period. 22 Inserted into the relevant sections of the Heptateuchon are a
redaction of Adelard's translation of Euclid's Elements, and Adelard's translation of the
astronomical tables of al-Khwarizmi Moreover, probably after the completion of the
Heptateuchon, another translator, Hermann of Carinthia, who addresses Thierry as his
master, recommends to him further works on geometry and astronomy that he and
his collaborator Robert of Ketton had been translating from Arabic. 23
This is the context, too, in which the new translations from Greek, made by James
ofVenice, must be viewed (Minio-Paluello 1972, 189-228). He was probably slightly
older than Gerard. The first notice we have concerning him is his presence as an
interpreter for discussions between the Eastern and Western churches in Con-
stantinople in 1136. His translating activity is mentioned in the entry in the
Chronicle of Robert ofTorigni added between 1157 and 1169 which reads that
"James, the clerk ofVenice, translated from Greek into Latin and commented upon
some books of Aristotle: i.e., the Topics, and the Prior and Posterior Analytics." In other
words, he was completing the arts of rhetoric and dialectic with new translations and
commentaries. 24
It is against this background that one must look at the first three categories of
Gerard's translations, as listed by his socii. These are three of the seven liberal arts, in
their canonical order: logic (dialectic), geometry, and astronomy. Gerard obviously
did not see the need to translate anything from Arabic on grammar or rhetoric, and
the three texts listed on logic all relate to the Posterior Analytics (including a translation
of the work itself), this being a text especially relevant to the demonstrative argument
used in the sciences. Gerard probably did not know James' translation, which is first
mentioned by Robert ofTorigni (as we have seen), and by John of Salisbury writing
in 1159; the priority of the two versions is debatable. The socii's geometry list begins
with the standard textbook on the subject - Euclid's Elements. The only version that
Gerard is likely to have known is the redaction attributed to Adelard of Bath, which
is also represented in the Heptateuchon: this redaction was copied either entirely
without proofs, or with brief" directions for proof" replacing the proofs and their
22 See Jeauneau 1995, and, for the quadrivial texts in the Heptateuchon, Burnett 1984.
23 Hermann's preface to his translation of Ptole my's Planisphere is in Heiberg 1907, c1xxxiii-clxxxvi.
24 For his translation of works on natural science see below, pp. 259-60.
VII
1
!
The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 259
accompanying labeled figures. Gerard's own translation25 reproduced the proofs in
full. For the rest, Gerard expanded not only the number of texts on geometry, but
also the range of geometry itself by including algebra; moreover, he added
mathematical arts previously unknown to the Latins: perspective and statics. 26 The
texts chosen by Gerard are those of Greek authors - Euclid, Theodosius, Archimedes,
"Mileus" (= Menelaus), "Tideus" (= Diocles) - commentaries on the Elements by
Arabic authors, and some original Arabic texts on the same topics, and on topics that
were unknown to the Greeks, such as algebra.
The last of the seven liberal arts is astronomy, and this forms the next section in the
socii's list. One might expect the Almagest of Ptolemy to come first - after all it was
for this that Gerard came to Toledo, according to the Vita - but it is preceded by one
work: the Rudiments of al-Farghani 27 In fact this is an easy introduction to the
subjects covered in the Almagest, and either could have been tackled before the great
work by Gerard himself, or might have been intended by him to be read first by his
students. The work of al-FarghanI proved very popular, and we know that Michael
Scot used al-Fargharu, but never graduated as far as the Almagest. The other texts
under the heading of astronomy comprise, once again, works by Greek and Arabic
authors, including the Spanish Muslims, Ibn Mucadh ofJaen (d. 1093) and Jabir ibn
Afia~ of Seville (H. ca. 1150). Astrology is conspicuous by its absence, a point we shall
come back to later.
With the next category - philosophy - we leave the seven liberal arts behind and
come to a completely different area of study, Aristotelian natural philosophy and
metaphysics. The very fact that the word "philosophia" has been transferred from the
seven liberal arts to natural science and metaphysics is significant. Unlike in the case
of the subjects of the seven liberal arts there was no pre-existing Latin educational
program into which these subjects could fit. Nor would Gerard even have found a list
of the textbooks for these subjects in Latin sources. Yet it is clear from the works that
he translated, and, in particular, from the order in which they are listed, that Gerard
knew the canonical order ofAristotle's works on natural science. This canonical order
had been established in Alexandria in the late Classical period, and was transmitted,
with the works themselves, both to the Islamic world, and to Byzantium. It is from
the latter source thatJames of Venice and Burgundio of Pisa (d. 1193), both of whom
were together in Constantinople in the famous 1136 meeting of the Eastern and
Western Churches, must have derived their knowledge of some ofAristotle's texts on
natural philosophy and metaphysics. For, between them, they translated the Physics,
De generatione et corruptione, the De anima, part of the Parva naturalia, and the
25 Or revision, see pp. 267-8 below.
26 Perspective immediately follows geometry in al-Faclbi's On the Classification of the Sciences (see below,
p. 260), while statics follows astronomy (the same order is found in Gundissalinus' De divisione philosophiae).
27 The same text had been translated by John of Seville and Limia in Limia in 1135. Gerard appears to have
known John's translation, the phrasing of which he sometimes follows; for examples of his revision of John's
translations see below, pp. 268-9.
VII
260
Metaphysics. 28 Whether they had the intention of creating a complete Latin corpus of
Aristotle's natural science and metaphysics, however, is less clear. Burgundio, at least
(who translated the De generatione et corruptione), was partly motivated by consideration
of the relevance of natural science to medicine, a motivation which may also have
induced their contemporary in Sicily, Henricus Aristippus, to translate the fourth
book of the Meteora, which, like the De generatione et corruptione, is about the mixing
of the elements.
Gerard's program seems more clear-cut than that of his contemporary Greek-Latin
translators. And it is so, in large measure, no doubt because of his knowledge of the
Arabic philosophers' divisions of sciences along Aristotelian principles. 29 This is
manifest
. in
32
several texts, including ones by Qusta . ,
ibn Liiqa 30 al-KindI, 31 and
AVlcenna. But a source immediately at hand is a work listed at the end (bar one) of
the texts of philosophy translated by Gerard: On the Classification of the Sdences of al-
Farabi 33 Al-Farabl's work not only provid~d a template for the subjects to be covered
in a course of "philosophy" in the Aristotelian sense, but also supplied a checklist of
textbooks to be used for that course. For, if one turns to the section on natural
science, one finds that al-FarabI divides it into eight parts or "enquiries" (fu~ii~), and
for each enquiry he specifies which text or section of a text by Aristotle (or in the
Aristotelian tradition) covers that enquiry. 34
Thus al-Farabl's first three enquiries are covered by Aristotle's Physics, De caelo and
De generatione et corruptione respectively. These are listed in this order by Gerard's sodi,
with a pseudo-Aristotelian text, De causis proprietatum et elementorum quatuor, inserted
quite naturally between the De caelo and the De generatione et corruptione; for it covers
both consideration of different parts of the earth (in fact, it is the most "geographical"
of the Aristotelian corpus) and the elements themselves. The next enquiry, according
to ~- FarabI, "is concerning the principles of actions and passions and those things
whIch are proper to the elements alone, without considering what is composed from
the elements" and is covered by the first three books of the Meteora of Aristotle. This
is the next work on the socii's list. The sodi add that "Gerard did not translate the
fourth book, because he surely found that it had already been translated." The fourth
book of the Meteora had, indeed, been translated by Henricus Aristippus in Sicily, and
28 See Vuillemin-Diem and Rashed 1997. That Burgundio and James coordinated their activity is still to be
determined.
29 For the Arabic situation see Jolivet 1996.
30 See Daiber 1990.
31 See Guido and Walzer 1940.
32 Michot 1980; French translation by Rabi'a Mimoune in Jolivet and Rashed eds. 1984, 143-51; Latin
translation by AndreaAlpago in AvicennCf' philosophi prCf'ciarissimi ac medicorum principis, Compendium de anima, De
mahad ... , Aphorismi de anima, De dijfinitionibus et qUCf'sitis, De divisione scientiarum, Venice, 1546, fols 139v-
145v.
33 Al-Farabi had also been important in Islamic Saragossa, since his commentaries on Aristode formed the
basis of those ofIbn Baija (Avempace, d. 1139).
J.4 See Appendix 11.
VII
The Coherence of the Arabic-lAtin Translation Program in Toledo in the Twelfth Century 261
that translation must have been known in Toledo at least by the time of the sodi. But
one could propose another reason for Gerard's non-translation of the book: simply
that he did not get that far in al-Farabl's list. For the fourth book of the Meteora is,
in fact, described by al-FarabI as the textbook for the fifth enquiry of natural science.
It is at this point that Gerard stopped, whether being prevented by his death in 1187,
or for some other reason. He had, however, made sure that, for the first four parts of
the program in natural science, he had provided some Greek and Arabic discussions
of Aristotle's works: Alexander of Aphrodisias' small treatises on time and "that
augment and increase occur in form not in matter," al-Farabl's commentary on
Aristotle's Physics, and al-Kindl's On the five essences. That Gerard was not uninterested
in the remaining parts of natural science is shown by the fact that he also translated
some texts relevant to the parva naturalia (belonging to al-Farabl's "eighth enquiry"):
Alexander's On the Senses, and al-Kindi"'s On Sleep and Vision. But one indication that
Gerard was following a program (and specifically al-Farabl's program) of translating
Aristotle's works on natural science, is that the same program was continued by a
successor ofGerard's at Toledo - in fact, most likely by one of the very sodi who wrote
the Vita.
This was the Englishman, Alfred of Shareshill. He translated textbooks for al-
Farabl's next two enquiries of natural science: the sixth, on minerals, and the seventh,
on plants. Finding no work on minerals by Aristotle himself, he translated the
chapters on minerals in the Shifo) of Avicenna (to which we shall return). He was
probably responsible for adding Aristippus' translation of the fourth book of the
Meteora to Gerard's translation of the first three, and tacking the chapters on minerals
onto the end of the text. For this composite Meteora is described in its colophon as:
The book of Meteora of which the supreme philosopher, master Gerard the Lombard,
translated the first three books from Arabic into Latin, but Henricus Aristippus translated
the fourth from Greek into Latin. The last three chapters were translated by Alfred the
Englishman of Shareshill from Arabic into Latin. (MS Oxford, Selden supra 24, fol.
109r.)
For the botany, Alfred translated a work De plantis, which was in reality composed
by Nicholas of Damascus but included much ofAristotle's lost work on plants. Alfred
wrote glosses to the whole of the composite text of the Meteora as well as to the De
plantis. The strongest indication that Alfred had al-Farabl's list in his mind when
choosing to translate works on minerals and plants is in his first gloss to the Meteora,
which reads:
The tide of the book is etc .... It must be noted that al-Farabi in his book On the sciences,
the chapter on the natural sciences, says: "The fourth enquiry is concerning the
principles of actions and passions and those things which are only the first elements,
 VII
262
without considering what is composed from the elements, and it is in the first three
books of the book Meteora.,,35
Al~EirabI's eighth,. and last, enquiry is "concerning what is common to the species of
an~mals,~nd what IS proper to each of the species" and is the subject of the "book of
ammals and the book of the soul and the books which are after them until the end
36
of the libri naturales. It is curious that no Toledan translation of Aristotle's De anima
appea~s to ha".e been made; and evenJames of Venice's Greek-Latin translation of the
work IS ?rst cIted ~nly a~ter the turn of the thirteenth century. This may be because
the sectIOn of AVIce~n~ s Shifii' d.evoted to the soul was translated in Toledo by
~vend~uth and DommIcus Gundissalinus, as we have seen, and became popular
Immediately. As for the "book of animals," al-EirabI had in mind the 15-book Arabic
De animalibus w~ich combine~ three books on the subject by Aristotle. Although
Alfred of Shareshtll refers to .thIs work in his original writings, he does not appear to
have embarked on the formIdable task of translating it. It fell to another Britisher at
Toledo to complete this task: i.e., Michael Scot, who was a canon ofToledo cathedral
by 1215, and completed his translation of the De animalibus there before 1220.
. ~ut, t~ return to Gerard. Preceding the four works on natural science in the socii's
lIst IS a smgle ~ork on m~taphys.i~s: Liber Aristotilis de expositione bonitatis pure, a work
better know? m t~e Latm tradItIon as De causis by "Aristotle" (see Ricklin 1995,
69-121). AnstotelIan m:taphysics and natural science naturally accompanied each
ot~er. In the condemnatIOns of the new science in Paris, in 1215, it is the works of
Anst~tle on natural science and metaphysics and the commentaries on them that are
mentIOned. It is q~ite logical t~a~ metaphysics, dealing with first principles, should
p~ecede na~ural SCIence, and thIS IS the order that Avicenna adopted in his Danesh-
~ameh~ whIch was ~onsequen~!y followed by Algazel in his Maq~id al-falasiJa , "the
mtentIons of the phil~sophers, translated by Gundissalinus and Johannes Hispanus in
To~edo at the same tIme as Gerard was working there. However, Avicenna in his
Shifii', an~ al-FarabI "rise up," as it were, to metaphysics, as "scientia divina," after
natural sC1~nce. In the Catalogue of the Sciences, Gerard would have found that the
whole tOpIC was dealt with in Aristotle's book de metaphysicis. 37
It appears that this work was not available in Arabic in Toledo. Instead, Gerard
turned to the De causis. This text is, in reality, a Neoplatonic compilation based largely
35 Alfred ofSareshel's Commentary on the Metheora ifAristotle, ed. J. K. Otte Leiden 1988 p 37' "T't I tali.
Lib A' l' h'l h' ' , , . . 1 U US s.
er mtote IS p I ~SOP.I s~pientis in factura impressionum superiorum que sunt in alto et inferius, tractatus primus.
N~tandum Alfar:ablUs In libro n,e sClenciis capitulo de naturalibus, ait: 'Quarta inquisitio est de principiis
actIon~m et ,pa~~lOnu~ et que prIma sunt elementa solurn sine compositis ab eis, et est in primis < tribus >
tractatIbus Libn Impresslonum superiorum"'.
36 The ,phrase "and the b?oks whic~ are after them until the end of the libri naturales," referring to the parva
ndadt~r~lla, does not occur In the ArabIC as edited by Gonz:ilez Palencia 1932, and could be the Latin translator's
a ItlOn.
37 Gonz:ilez Palencia 1932, 163: "in libro suo de metaphysicis."
, The Coherence of the Arahic-lAtin Translation Program in Toledo in the Twelfth Century
VII
263
on the Elements of Theology of Proclus. It is not mentioned at all by al-FarabI, and of
the mere five Arabic writers known to have used the work (or a derivative text), three
are from al-Andalus, one being Moses Ben Jacob ibn Ezra of Granada, who died not
long after 1135, another being Solomon ibn Gabirol, the Jewish philosopher and poet
who was born in Malaga in 1021, lived in Saragossa, and died in Valencia in 1058,
whose Fons vitae was translated by Gundissalinus and Iohannes Hispanus (Kraye et al.
1987,41; and Schlanger 1968, 73-76).
It must be remembered that Aristotle was always regarded as being "obscure" and
needed elucidation. Gerard of Cremona, as we have seen, translated some texts that
were regarded as helpful for understanding Aristotle's works - by Alexander of
Aphrodisias, al-KindI and al-Farabl. Alfred of Shareshill composed commentaries on
the Meteora and De plantis, perhaps using glosses (e.g., by Alexander) that were already
in the Arabic manuscripts. The culmination of this process, however, is found in the
later work of Michael Scot, who, after translating the De animalibus and moving to
Italy, made use of the recent work ofAverroes to provide commentaries for the other
texts on natural science: the Physics, the De Caeio, the De anima, and the
Metaphysics. 38
It was this combination of texts of Aristotle, translated from Greek or Arabic, and
the commentaries of Averroes and (in the case of the Meteora and De plantis) Alfred
of Shareshill, which became the textbooks in natural philosophy and first philosophy
(i.e. metaphysics) in the universities from the second quarter of the thirteenth century
onwards.
The ShifcP of Avicenna has already been mentioned quite frequently in passing.
This large work, as is well known, is Avicenna's encyclopedia of philosophy in the
Aristotelian tradition. It is divided into four units, each called ajumla, or "collection":
on logic, on natural science, on mathematics, and on metaphysics. Avicenna, like al-
FarabI, divides natural science into eight parts, though not quite in the same order:
the De anima comes sixth, before plants and animals; separate parts are given to the
soul and to animals; and combinations of the elements and minerals are put together
in one part. He devotes to each of these parts a single book of the "natural science"
collection (jumla) of the Shifii'. The Shifii' is not a commentary on Aristotle's works,
but provides Avicenna's own philosophy on the same topics as those covered by
Aristotle, with the addition of the mathematical sciences, which Aristotle did not
write about. The translation of the Shifii' in Toledo (and, later, elsewhere) can be seen
as running parallel to that of Aristotle and his commentators. The books of the Shifii'
on the same topics as Aristotle's books were sometimes translated in addition to those
of Aristotle, sometimes in substitution for Aristotle's. Gerard of Cremona, however,
apparently played no part in translating the Shifii'. Instead, this was superintended by
his colleague, Dominicus Gundissalinus.
38 Of these commentaries, only the De caelo is clearly attributed to Michael; the attribution of the De anima
occurs in one manuscript only, the others are unattributed. Nevertheless, on the grounds of style and date it
is likely that Michael was responsible for them all.
VII
264
In contrast to the "Master" Gerard, Gundissalinus did not teach, but was actively
engaged in ecclesiastical administration. He was an archdeacon of Segovia cathedral
who was resident in Toledo, the metropolitan diocese, and is mentioned in numerous
documents from the cathedral until 1181. His interest in the Aristotelian division of
sciences is manifest in the fact that he too made a version of al-F:irabI's On the
Classification of the Sciences, and used this as a framework for his own De divisione
philosophiae. In this, and in the several original works that he wrote, he shows that he
has been well educated in the LAtinorum studia, and, in particular, in the works of
scholars associated with Chartres, including Thierry of Chartres and the translator
Hermann of Carinthia. 39 But he did not show any interest, either in his original
works or in the works he chose to translate, in the texts ofAristode on natural science
and metaphysics. Instead, he translated the philosophical texts that were being read by
Islamic, and especially, Jewish, scholars educated in Islamic Spain - and one such
Jewish scholar, as we have already seen,4O may have introduced him to these texts:
Abraham ibn Daiid. 41
Between them Gundissalinus and Avendauth translated the jumla of the Shifo> the
metaphysics ("first philosophy"), a part of the logic, and the individual book on the
soul. The beginning of the physics - the first book of the jumla on natural science -
was also translated at this time, probably under the supervision of Gundissalinus,
though no name is attached to the translation. 42 That Gundissalinus apparendy had
the whole Shifo> at hand is evident from the fact that in his De divisione philosophiae
he cites another passage from Avicenna's logic ("on the subalternation of the
sciences"), which does not appear in Latin elsewhere. A hundred years later, another
translator, Juan Gonsalvo of Burgos, continued the translation of the physics section
from the very point where the twelfth-century translator had broken off (in fact, in
mid-sentence), and translated several further books of the jumla on natural science. It
is presumably from Avendauth's Arabic manuscript that Alfred of Shareshill made his
translation of the chapters on mineralogy (and possibly of a chapter on flooding).
Michael Scot translated the section on animals (the last book of the natural science
jumla) which he dedicated to Frederick 11 Hohenstaufen, his patron in Sicily. But the
39 For correspondences in Gundissalinus' works with texts connected with Chartres and Paris, see Burnett
1990. That Gundissalinus owed his system of accessus to each of the sciences in his De divisione philosophiae to
Thier~ is argued by Fredborg 1988, 16-20; he is the only author known to have used the cosmological
materIal from the De essentiis of Hermann of Carinthia, a pupil ofThierry of Chartres.
40 See pp. 251-2 above.
41 The identity of' Avendeuchl Avendauth Israhelita' with Abraham ibn D:iiid was first suggested in d' Alverny
1954. It would seem to be confirmed by the pervading influence of Avicenna in Ibn D:iiid's writings: see
Cohen 1967, xxiv: "Above all, although he [Ibn D:iiid] never acknowledges the fact, he seems to have
absorbed thoroughly the writings of Ibn Sin:i and to have appropriated the Aristotelian thought which the
great Arab philosopher had expounded in his commentaries." The willingness of Avendauth to collaborate
with a Christian scholar also fits the character of Ibn D:iiid who, unusually for a Jewish scholar, wrote about
the history of Rome and the beginnings of Christianity (ibid., xxvii-xxviii).
42 For details of the translations of the Shifit, see d' Alverny 1993, article IV. All these texts, with the exception
of the Logic, have been edited by van Riet 1968-92.
, The Coherence <if the Arabic-Latin Translation Program in Toledo in the Twelfth Century
VII
265
work was not necessarily translated there, since it was copied alongside Michael's
translation of Aristode's De animalibus, made, as we have seen, in Toledo.
The Arabic texts in which Gundissalinus' collaborator is "Iohannes Hispanus" are
also distinctive of the Hebrew academic community: one is the Maq~id al{aliisifa of
Algazel (al-Ghazzali), which was much used by Abraham ibn Daiid; the other is. the
Pons Vitae or "fount oflife" written in Arabic by the Jewish philosopher Ibn Gablrol.
It does not seem unfeasible that these two works should have been brought to the
attention of Gundissalinus and the Archbishop of Toledo by Abraham ibn Daiid
alongside the Shifo> of Avicenna. Another work that Abraham could have brought is
the Liber de causis, which, as we have seen, was known to Ibn Gabirol; for it was called
in its earliest manuscript (Oxford, Bodleian Library, Selden supra 24), the
"Metaphysica Avendauth", and its author is named by Albertus Magnus as "David
Iudaeus."43
Thus, in the field of Aristotelian philosophy in Toledo, we see a remarkably rich
mixture in the mid- to late-twelfth century: on the one hand, the original texts of
Aristode and accompanying works by Alexander of Aphrodisias, al-Kindi and al-
FarabI; on the other hand, works of Avicenna and Algazel which did, in fact, more
accurately represent the reading-matter of the Jewish and Arabic scholars of the tim~.
The outbreak of an interest in the works ofAristode himself among a group ofArabiC
scholars in Cordoba in the late twelfth century is an isolated phenomenon that had
momentous repercussions in the West, through the translations of Averr~es'
commentaries and al-Bi~riijI's Aristotelian astronomy (see Sabra 1984), but whlCh
failed to affect, to any noticeable degree, the general predominance of Avicenna
among Arabic philosophers, and ofAlgazel among the theologians. Averroes (d. 1198)
was working in Cordoba at the same time as Gerard was working in Toledo and both
scholars were interested in the same subjects. But whether this Cordoban
"Aristotelian revolt" in the court of the Almohads (which, it must be remembered,
was entirely Islamic, since the Jews had been expelled) had repercussions in the Toledo
ofGerard ofCremona's time is difficult to tell. It must be noted, however, that Gerard
had access to Arabic texts in the Aristotelian tradition which had ceased to be read
elsewhere in the Islamic world, including treatises of al-Kindi which have been
preserved in Arabic only thanks to a chance interest on the part of the Theos~phists
ofIsfahan in the seventeenth century (see Endress 1994, 175). And even they did not
44
rescue al-Kindl's text On the five essences which Gerard translated.
Nevertheless, Aristotelian philosophy was not Gerard's main interest, nor did his
translations in this field have such a large influence as those in other fields. For his
43"Ibn D:iiid" means "son of David:' The De causis is the first of Gerard's philosophical translations to be
known outside Spain, and travels with Gundissalinus' translations; see Burnett 1997, 6~. So~e s~holars, su~h
as Adriaan Pattin, suggest that Gundissalinus played some part in its translation: see diSCUSSIOn m Taylor (m
Kraye et al. 1987).
44 The Arabic text of De quinque essentiis, which appears among the Rasail ai-Kind; edited by Abii Ri<,ia
(Cairo, 1953, 11, 8-34), is a modern translation of the Latin text.
 VII
266
1 The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century
~rabic-Latin translations of Aristotle's works were eventually replaced by translations
dIrectly from Greek - first, by those of James ofVenice and Burgundio of Pisa, and
then, towards the end of the thirteenth century, by those ofWilliam of Moerbeke _
an~ his translations of Arabic commentaries and other accompanying works were
eclipsed by the great commentaries of Averroes.
In the field of mathematics (which we have already surveyed) Gerard's translations
had. a much ron ore lasting effect on Western scholarship. But surpassing even this
achIevement, In terms of quantity and effect, were Gerard's translations of medical
texts, which are the next category in the list of the socii. This field has been covered
very expe~tl! by Danielle Jacquart, to whom I am much indebted. 45 As in philosophy,
so In medICIne, she sees that the "Toledan enterprise of translation" evidenced not a
?aphazar~ affair, but a project in the true sense of the word Gacquart 1992, 60 and
In CardaIllac 1991, 177-91). This is clear from the list of texts of the socii. First come
nine texts of Galen. Although Galen was known to the Latins as the greatest of the
Greek doctors, very few of his writings had been translated into Latin before Gerard's
time. ?alen was regarded as being as much a philosopher as a doctor, and as being too
complIcated for the requirements of the ordinary physician. Constantine the African
Gerard's principal predecessor as a translator of medical texts from Arabic into Latin'
knew of the list of 16 works which had been selected from amongst Galen's vas~
output for the teaching of medicine in Alexandria, but he translated only one of these
(the Megategni or Methodus medendr). Gerard, on the other hand, translated at least five
more texts on this list. The choice of the remaining texts seems to have been made
in accordance with his interests (and that of his contemporaries) in element-theory,
the temperaments and therapeutic method Gacquart 1992, 58). The next two items
on the socii's list also treat these philosophical aspects of medicine: Isaac Israeli's On the
Elements and On the Description of Things and their Definitions. The other texts in this
section are, for the most part, substantial texts on medicine by the Arab successors to
Galen, ar-Razl (represented by three texts), two natives of al-Andalus - az-Zahrawi
and Ibn al-Wafid - and, above all, Avicenna, whose Canon medicinae in Gerard's
translatio~, became the principal comprehensive text for medical traini;g in Europe,
and remaIned on the curriculum into the eighteenth century. Appropriately, this
work comes at the end of the section, followed only by the Tegni (or Ars parva) of
Galen, to the end of which the list itself has been appended.
Considering its importance, one should pay more attention to the medical section
of th.e socii's list, but I have neither the time nor the competence to do this. Moreover,
the list does not end here, and we must briefly consider the last items.
. ~he~e are two groups of three works, the first on alchemy,46 the second on
dIVInatIOn. The relationship of alchemy both to natural science and to medicine is
obvious. Al~he~y is listed amongst the divisions of natural science by Gundissalinus,
who took hIS lIst from an anonymous translation of an anonymous Arabic text called
45 Jacquart's articles are conveniendy collected in Jacquart 1997.
46 On the identification of the works on alchemy translated by Gerard, see Halleux 1996, 891.-92.
VII
267
"On the rise of the sciences" (De ortu scientiarum).47 Avicenna included an attack on
alchemy in his Shifti) , but elsewhere seems to approve of it (see Anawati 1996,
875-79).
The last group of works concerns divination: first, divination by means of figures
drawn randomly on the sand or on paper (i.e., geomancy),48then divination according
to a system of questions which are related to answers derived by a process of random
calculation,49 and, finally, a divinatory technique based on the Moon's position in the
zodiac each month. 50
The very last item is the most local of the texts, for it is a calendar put together
from an Arabic calendar arranged according to the risings and settings of the anwif (or
"lunar mansions") and a Christian liturgical calendar, for the Arabic-speaking
Christians of Cordoba. 51
The socii were justly proud of the achievements of their master. Modern
scholarship has attributed even more translations to Gerard (see especially Lemay
1978, 175, nn. 58a-d, 183 and 187f.), but what has also become clear is that Gerard
was not always translating de novo. This has been demonstrated in the case of the
translation of ar-Razi's Uher Almansoris, of which two versions exist. Danielle Jacquart
has suggested that the first version was made by an as yet unidentified translator,
whereas the second shows the application of a more rigorous word-for-word
equivalence and, in particular, of Gerard's terminology. 52 Other works may exhibit
Dominicus Gundissalinus, De divisione philosophiae, ed. L. Baur, Beitriige zur Geschichte der Philosophie des
47
Mittelalters, 4.2-3, Munster, 1903, p. 20: "scientia naturalis universalis est quia octo sciencie sub ea
continentur: scilicet sciencia de medicina, sciencia de iudiciis (corrected from Baur's 'indiciis'), sciencia de
nigromantia secundum physicam, sciencia de ymaginibus, sciencia de agricultura, sciencia de navigacione,
sciencia de speculis, sciencia de alquimia, que est sciencia de conversione rerum in alias species, et hec octo
sunt species naturalis sciencie"; from De ortu saentiarum, ed. C. Baeumker 1918, 20.
48 For the texts on geomancy attributed to Gerard see Charmasson 1980,111-19,129-39. The text WIth the
.
incipit "Estimaverunt Indi," mentioned in one manuscript of the Commemoratio librorum, is also attributed to
Hugo of Santall a, whereas a text with the incipit"Si quis per artem geomanticam," which is more consistendy
attributed to Gerard, already seems to have been known in Hereford in c. 1195-97; see Burnett 1995c.
49 A summary of the history of this text is given (alongside that of Gerard's other astronomical translations) in
Kunitzsch 1992, 79-80.
541 The tide "Alfeal (i.e. Arabic al1a'l, "omen, fortune") secundum motum lune" is found in Paris, BNF, lat.
9335, fo1. 140r (see p. 281 below), at the beginning of a table of "accidentia," or "happenings" when the
Moon is in each of the signs of the zodiac. A preliminary table indicates which sign of the zodiac the Moon
is in on each day of each Latin month. The two tables are introduced with the instru~ti~n: "Capitulu~
cognitionis mansionis Lune: Scias quid preteriit de mense arabico (in margin: id est l~na~I, Id est quota ~nt
Luna) et accipe illud in linea que est super tabulam, et extrahe ipsum ad signa que sunt In linea que opporutur
mensi Latino in quo tu es, et scies tunc ubi mansio est Lune ex signis per illud, si deus voluerit." "
51 For the description of this text as a "sacerdoai mar(tyro)logium" compare the use of the word "martyrologium
for a computus written in Spain in 1055 A.D. in MS Tortosa Cathedral, no. 10: "Incipit martyrologium de
circulo anni"; see Martinez Gazquez and G6mez Pallares 1994.414.
52 Jacquart 1997, article VIII "Note sur la traduction latine du Kitiib al-mansiiri de Rhazes."
 VII
VII
268
1 The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century 269

the sam~ phenomena. 53 Such a process of revision is particularly obvious in the case
of a subject-matter not mentioned at all by the socii: i.e., astrology. consultation of the Arabic original, and his hypothesis that the reviser was Gerard of
Cremona is quite plausible. 57 The Great Conjunctions has been subjected to a similar
The absence of astr?logy fro~ the list of the socii is especially striking, considering
revision, and comparison of the terminology suggests that the same two scholars w.ere
that the. only eye-wlt~ess eVIdence we have of Gerard's teaching portrays him
involved. 58 No translator or reviser is named in the manuscripts, but to manuscnpts
e~~oundmg an astrologlCal text. This is by Daniel of Morley, the Englishman who
of the revision (including Paris, BNF, lat. 16204) detailed notes on variant Arabic and
VIsIted Toledo early in Gerard's career. He would have us believe that he heard Gerard
Latin readings, interpretations of obscure passages, and co~ents on the mathe~~~­
of Cremona lecturing on astrology, and held a disputation in which Gerard defended
ical calculations have been added. Among these comments IS reference to a word .In
astrology and Daniel raised objections (see Burnett 1995b). The fact that Daniel
the Toledan dialect.,,59 This suggests that, while the place of activity ofJohn of Seville
makes Gerard's authorities Firmicus Maternus and the versions of Abii Macshar's
and Limia remains unclear,60 the revision of this text, and presumably of other texts
introduction to astrology made by Adelard of Bath and Hermann of Carinthia all
in this astrological corpus, took place in Toledo.
t~ree being works already well known in England, makes one a little suspicious about We have seen in Toledo, then, that there was a clear division between the
his account. Nevertheless: it would be a little surprising for Gerard to deliberately
translating activities of Gerard of Cremona and Dominicus Gundissalinus. The one
neglect astrology when It was regarded as being of equal scientific cogency as
medicine, and when most astronomers were also astrologers. favored the authentic works of the Greeks and their Arabic commentators, the other
favored Avicenna's philosophical approach to philosophy and the reading-~tte~ of
In fact, it s~ems that a large corpus of Latin astrological texts of Arabic origin was
contemporary Jewish scholars. Considering that both Gerard and Gundissalinus
put togethe~ In Toledo. The evidence for this is the common terminology and the
worked within the precincts of Toledo cathedral, it is hard to believe that they were
co~prehenslVeness of the astrological corpus best represented by the manuscript
unaware of each other's work, or inimical to it (the key text here is al-Farabi's On the
Pans, BNF, lat. 16204, a manuscript copied for Richard of Fournival, the son of the
Classification of the Sciences: in this case they certainly shared each other's findings).61
astrolog~r of the French royal court, Roger. This manuscript contains, in order, Another large subject-area, astrology, had been dealt with by anot~er scholar, John of
astrol~gl~~ t:xts,~y A?~, Macshar, :"orks on horoscopes and weather forecasting, texts Seville and Limia (the translations of his that have dates were wntten between 1133
by Masha allah, JerglS, Sahl, CAll al-clmrani (Haly), Thabit b. Qurra and Pseudo-
and 1145), but its texts were being studied and revised by Gerard and his colle~gues.
Ptol~my. ~4 The largest of_these texts - the Great Introduction to Astrology and the Great Thus while the internal coherence and rationale of the translating enterpnse of
ConJunctIons, both by Abu Macshar - are comparable in their bulk to Euclid's Elements
Gera;d is demonstrated in the Commemoratio librorum by his socii, one can see strategies
and Ptolemy's Almagest. The texts in the Paris manuscript provide between them a
and a sense of order also at the level of the translating activity in twelfth-century
curriculum in astrological science as complete and coherent as do Gerard's
Toledo as a whole, whether one looks at it chronologically - from Gerard, through
translations for the scien~e of medicine. While it is not possible at this stage in our
Alfred of Shareshill to Michael Scot (and eventually Hermann the German) - or
research to assert categoncally that all these texts were collected in Toledo,55 the bulk
synchronically, with Gerard and Gundissalinus s~aring res.po~sibilities be~een
of them exhibit a homogeneity that suggests a single enterprise. That enterprise is
themselves, and building on the work ofJohn of SeVllle and Llffila. What remaInS to
probably due to John of Seville and Limia, whose name as translator is attached to the
be explained is the driving force behind this translation enterpri.se. I h~ve s~~gested
first text, Abii Macshar's Great Introduction. 56 However, Richard Lemay has
above that the intellectual motivation came from the burgeorung uruversltles and
demonstrated that the Great Introduction was thoroughly revised, with further
other intellectual centers outside Toledo. But that is an insufficient explanation for the
question of who organised the production and who paid for it. The Vita of Gerard
53 For example, Galen's On the Temperaments (no. 47 below) and the translation of Euclid's Elements attributed

to Ger~rd (no. 45 below) ~hich has b~en observed to be not in Gerard's style. His revision of a previous
translation of th: Elemen~s IS ~ alternative explanation to that given by the editor: that a subsequent reviser
n~l99~. _
couched Gerard s translation m a more elegant Latin style: see Busard ed. 1984; the extent of the knowledge
of the Greek tradition manifest in this version is striking. 58 See Burnett andYamamoto 2000, and Burnett 2001. A feature common to the revision of both texts ofAbu
Ma'shar is the substitution of "generatio" for "effectus;' "impressio" for "vestigium," "continuatio" for
54On the original contents of this manuscript and its connection with Richard of Fournival, see Pingree
1987, esp. 84-87, 100--{)2. "coniunctio," and "dispositio" for "esse." . ,
~5 The work of 'All al-'ImranI on elections, at least, was "interpreted in Barcelona by Abraham [bar Hiyya]
m 1134."
59 Paris, BNF, 16204, p. 246: "et in Toleto dicitur maluero." Lemay had already pomted out the Importance

of these comments, but did not realize that they accompany only the revised version: Lemay 1962, 1~.
56 I was over cauti~us in he~itati.ng to identify "John of Seville and Limia" with "John of Seville" in Burnett
60 As pointed out above (pp. 251-2), the only place of translation mentioned in any of John of Seville and

Limia's works is Limia itself.

1994, 242, especIally consIdermg that two manuscripts of the Great Introduction name the translator as
''Jo?an~es H.yspalensis/Hyspanensis ex Luna" (paris, Bibliotheque de l'Universite, lat. 640, and Cambridge,
UmversIty LIbrary, Kk..1.i).
61 Other texts on which they probably both worked (for there are two versions of ~~m) were .~-~ndi's De
intellectu (though this is not mentioned in the Commemoratio librorum) and lsaac Israeli s De dejimtlombus, both
texts being keys for the understanding of the other works they were interested in.

VII
270
suggests that he was a rich man, and a later legend claimed that he was paid by the
king of Castile,62 but documentary evidence is lacking. What is beyond doubt is the
scale and importance of the enterprise, which has no match in the history of western
culture.
References
D'Alverny, Marie-Therese. 1954. "Avendauth?" Homenaje Millas Vallicrosa, 19-43. Barcelona: CSIC.
D'Alverny, Marie-Therese. 1982. "Translations and Translators." In Renaissance and Renewal in the Twelfth
Century, edited by R. L. Benson and G. Constable, 421-62. Oxford: Clarendon Press.
D' Alverny, M.- T. 1989. "Les traductions adeux interpretes, d' arabe en langue vernaculaire et de langue
vernaculaire en latin." In Traduction et traducteurs au moyen age. Colloques internationaux du CNRS,
IRHT 26-28 mai 1986, 193-206. Paris: CNRS.
D' Alverny, M.- T. 1993. Avicenne en Occident. Paris: Lirairie philosophique J. Vrin.
D' Alverny, M.-T. 1994. La transmission des textes philosophiques et scientifiques au Moyen Age. Aldershot:
Variorum.
Anawati, George C. 1996. "Arabic Alchemy." In Encyclopeia of the History ofArabic Science, 3 vols. Edited
by Roshdi Rashed, vol. Ill, 853-85. London and New York: Routledge.
[Anonymous] 1710-40. Giornale de letterati d'Italia. Venice.
Baeumker, Clemens, ed. 1918. De ortu scientiarum. Beitrage zur Geschichte der Philosophie des
Mittelalters 19. Miinster: Aschendorff.
Birkenmajer, Aleksander. 1970. Etudes d'histoire des sciences et de la philosophie du moyen age. Studia
Copernicana 1. Wroctaw, Warsaw and Cracow: Ossolineum.
Bj0rnbo, Axel A. 1902. "Uber zwei mathematische Handschriften aus dem vierzehnten Jahrhundert."
Bibliotheca Mathematica 3.3:63-75
Boncompagni, Baldassare. 1851. Della vita e delle opere di Gherardo Cremonese traduttore del secolo duodecimo
e di Gherardo da Sabbionetta astronomo del secolo decimoterzo. Atti dell'Accadernia Pontificia de' Nuovi
Lincei rv. Rome.
\
Biihler, Curt E, ed. 1941. The Diets and Sayings of the Philosophers: The Translations by Stephen Scrope,
William Worcester and an Anonymous Translator. Early English Text Society 211. Oxford: Oxford
University Press.
Burnett, Charles. 1977. "A Group of Arabic-Latin Translators Working in Northern Spain in the Mid-
Twelfth Century." Journal of the Royal Asiatic Society year 1977:62-108.
Burnett, C. 1984. "The Contents and Affiliation of the Scientific Manuscripts Written at, or Brought
to, Chartres in the Time ofJohn of Salisbury." In The ~rld ofJohn of Salisbury, edited by M. Wilks,
127-59. Oxford: Blackwell.
Burnett, C. 1990. "A New Source for Dominicus Gundissalinus' Account of the Science of the Stars?"
Annals of Science 47:361-74.
Burnett, C. 1992. "The Translating Activity in Medieval Spain." In The Legacy of Muslim Spain, edited
by S. K. Jayyusi, 1036-58. Handbuch der Orientalistik, 12. Leiden: E. J. Brill.
Burnett, C. 1994. "Magister Iohannes Hispanus: Towards the Identity of a Toledan Translator." In
Comprendre et mattriser la nature au moyen age. Melanges d'histoire des sciences cfferts a Guy Beaujouan,
425-36. Geneva: Librairie Droz.
62 MS Chantilly, 322, fo1. 45v: "libri [ongeomancy] quem magister Gerardus de Cremona, magnus medicus in
phisica, transtulit de arabico in latinum, habens expensas a rege Castelle"; quoted in Charmasson 1980.
VII
The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century 271
Burnett, C. 1995a. "'Magister Iohannes Hispalensis et Limiensis' and Qusta ibn Liiqa's De differentia
spiritus et animae: a Portuguese Contribution to the Arts Curriculum?" Mediaevalia. Textos e estudos
7-8:221-67.
Burnett, C. 1995b. "The Institutional Context of Arabic-Latin Translations of the Middle Age~: A
Reassessment of the 'School of Toledo' ." In The Vocabulary of Teaching and Research between the Middle
Ages and Renaissance, edited by o. Weijers, 214-35. Turnhout: Br~pols.. . "
tt C 1995c . "Mathematics and Astronomy in Hereford and Its RegIOn m the Twelfth dB"
Burne,.
Century.
h
In Medieval Art, Architecture and Archaeology at HereJord, edited by D. Whitehead, 50-59. Lee s: ntIs
Archaeological Association. .. .
Burnett, C. 1997. The Introduction ofArabic Learning into England. London: BntIsh L1br~ry. .
Burnett, C. et al. 1998. Adelard of Bath, Conversations with his Nephew. Cambndge: Cambndge
University Press.
Burnett, C. 1999. "Dialectic and Mathematics According to Ahmad ibn Yiisuf: A Mod~l for Gerar~ of
Cremona's Programme of Translation and Teaching?" In Langage, sciences, philosophle au XIIe slecle.
Edited by J. Biard, 83-92. Paris: Librairie philosophique J. ~rin.. .
Burnett, C. 2000. "Antioch as a Link between Arabic and Latm Culture l~ the Twe.lfth and ThIrteenth
Centuries." Occident et Proche-Orient: Contacts scientifiques au Temps des CrOlsades. EdIted by 1. Draelants,
A. Tihon, and B. van den Abeele, 1-78. Turnhout: Brepols.
Burnett, C. 2001. "The Strategy of Revision in the Arabic-Latin Translations from Toledo: T?e Cas~ of
Abii Ma'shar's On the Great Conjunctions." In Translators at ~rk: Their Methods and ManUScripts, edited
by]. Hamesse, Louvain-la-Neuve: FIDEM (in press). ..
Burnett, C. and KeijiYamamoto, eds. 2000. Abii Ma'sar, On HistoricalAstrology: The Book of ReligIOns and
Dynasties (On the Great Conjunctions), 2 vols. Leiden: Brill. .' . ,
Busard, Huburtus L. L., ed. 1984. The Latin Translation of the ArabiC Version of Euclid s Elements Commonly
Ascribed to Gerard of Cremona. Leiden: E. J. Brill.
Cardaillac, Louis, ed. 1991. ToIMe, XIIe-XIIle siecles. Paris: Autrement. f • ,. f .f
Charmasson, Therese. 1980. Recherches sur une technique divinatoire: la geomanCle dans 1oCCldent medlevale.
Geneva and Paris: Librairie Droz. . .
Classen, Peter. 1974. Burgundio von Pisa. Sitzungsberichte der Heidelberger Akademle der WIssenschaf-
ten, phil.-hist. Klasse, 4.
Cohen, Gerson D. 1967. Ibn Daad SeJer Ha-Qabbalah. London: Routledge ~nd Kegan P7~1.. . ..
Daiber, Hans. 1990. "Qosta, ibn Liiqa (9 Jh.) iiber die Einteilung der WIssenschaften. Zeltschrift fur
Geschichte der Arabisch-Islamischen Wissenschaften 6:93-129.
Dunlop, Derrick M. 1971. "Hiidids." In Encyclopedia of Islam, new ed. III:542-3. Leiden and London:
E. J. Brill and Luzac.
Endress, Gerhard. 1994. "Al-Kindi iiber die Wiedererinnerung der Seele." Oriens 34: 174-221. .
Fredborg, Margareta. 1988. The Latin Rhetorical Commentaries ofThierry of Chartres. Toronto: PontIfical
Institute of Mediaeval Studies. .
Gonzalez Palencia, Angel. 1926-30. Los mozarabes de Toledo en los siglos xii y xiii. 4 vols. Madnd.
Gonzalez Palencia, Angel. 1932. Alfarabi. Catlilogo de las ciencias. Madrid: CSIC.
Gonza!vez Ruiz, R. 1996. "El traductor Hermann el Aleman." In El Escuela de Traductores de Toledo,
edited by E M. Villanueva,.51-64. Toledo: Diputacian provincial de Toledo. . ., ,
Gonzalvez Ruiz, Raman. 1997. Hombres y libros de Toledo (1086-1300). Madnd: FondaclOn Ramon
Areces. . . P
Grant Edward ed. 1974. A Source Book in Medieval Science. Cambridge: Harvard Umvemty ress.
Guid~, M. and Richard Walzer. 1940. "Uno scritto introduttivo. allo studio d~ Aristot~le:' Memorie della
R. Accademia Nazionale dei Lincei. Classe di scienze morali, stonche e fllologlche. Sene VI, vol. VI, fasc.
V. Rome. 1 d' .r h H'
Halleux, Robert. 1996. "The Reception of Arabic Alchemy in the West." In Encyc ope la 0 t e IStOry
ofArabic Science. 3 vols. Edited by R. Rashed, vol. 3:886-;0~. ~on.do~; Routled~e.
Hernandez, Fernando J. 1991. "La cathedrale, instrument d assnrulatIon. In Cardatllac 1991.

VII
272
Hernandez, Fernando J. 1995. "La Fundaci6n del estudio de Alcala de Henares." En la Espafia Medieval
18:61-83.
Jacqu~rt, Danie~e. 1988. "Aristotelian Thought in Salerno." In A History of Twelfth-Century ~stern
Philosophy, edited by P. Dronke, 407-29. Cambridge: Cambridge University Press.
Jacquart, Dan~elle. 1992. "Les traductions medic ales de Gerard de Cremone." In Pizzamiglio 1992.
Jacqu~rt, Damelle. 1997. LA sdence medicale ocddentale entre deux renaissances (XIIe s.-xVe s.). Aldershot:
Vanorum.
Jacquart, D. and Fran~oise Micheau. 1990. LA medecine arabe et l'ocddent medieval. Paris: Maisonneuve et
Larose.
Jeauneau, E. 1995. L'Age d'or des ecoles de Chartres. Chartres: Editions Houvet.
Jolivet, Jean, and Roshdi Rashed, eds. 1984. Etudes sur Avicenne. Paris: Les Belles Lettres.
Jolivet, J. 1996. "Classifications of the Sciences." In Encyclopedia of the History ofArabic Sdence, edited by
R. Rashed, vol. 3:1008-25.
Kraye,.]., Ryan, W E and C. Schmitt, eds. 1987. Pseudo-Aristotle in the Middle Ages. London: Warburg
InstItute.
Kun~tzsch, Paul. 1974. Der Almagest. Gottingen: Vandenhoeck und Ruprecht.
Kumtzsch, Paul, ed. 1986-91. Ptolemy. Der Sternkatalog des Almagest. 3 vols. Wiesbaden: Harrassowitz
Verlag.
Kunitzsch, Paul. 1992. "Gerard's Translations of Astronomical Texts, especially the Almagest." In
Pizzamiglio 1992,71-84.
Leclerc, Lucien. 1876. Histoire de la medecine arabe. Paris: E. Leroux.
Lemay, Richard. 1962. Abu Ma'shar and LAtin Aristotelianism in the Twelfth Century. Beirut: American
University of Beirut.
Lemay, R. 1978. "Gerard of Cremona." In Dictionary of Scientific Biography, edited by C. L. Gillispie,
Supplement I: 173-92. New York: Scribners.
Lemay, Richard, ed. 1996-7. Abii Ma <shar al-BalIJi. 1996-97. Uber Introductorii maioris ad sdentiam
judidorum astrorum. 9 vols. Naples: Istituto universitario orientale.
Lorc~, Richard. 1995. "Some Remarks on the Almagestum parvum." In idem,Arabic Mathematical Sdences,
artIcle 5. Aldershot: Variorum.
Lorch, R: ~ 996. "The Transmission ofTheodosius' Sphaerica." In Mathematische Probleme im Mittelalter:
der latelnlsche und arabische Sprachbereich, edited by M. Folkerts, 159-83. Wolfenbiitteler Mittelalter-
Studien 10. Weisbaden: Harrassowitz Verlag.
Martinez Gazquez, J. and J. G6mez Pal1ards. 1994. "Manuscritos cientificos latinos de la Catedral de
Tortosa." Hispania Sacra 46:413-24.
M~urach, Gregor, ed. 1979. "Daniel ofMorley, Philosophia." Mittellateinischesjahrbuch 14:204-50.
MIchot, Jean. 1980. "Les sciences physiques et metaphysiques selon la Risiilah fi aqsiim al-<ulum
d'Avicenne." Bulletin de Philosophie medievale 22:62-73.
Minio-Paluello, Lorenzo. 1972. Opuscula: the LAtin Aristotle. Amsterdam: Hakkert.
North, John D. 1995. "'Aragonensis' and the Toledan Material in Trinity MS 0.8.34." Cahiers du moyen
age grec et la tin 65:59-61.
~tte, James K., ed. 1988. Alfred of Sareshel, Commentary on the Metheora ofAristotle. Leiden: Brill
Pmg~ee, David. 1981. "Boethius' Geometry and Astronomy." In Boethius: His Ufe, Thought and Influence.
Edited by M. Gibson. Oxford: Blackwell.
Pin~ree, ~avid: 1987. "Th~ Diffusion of Arabic Magical Texts in Western Europe." In LA diffusione delle
sClenze Islamlche nel medlO evo europeo. Edited by B. Scarcia Amoretti, 57-102. Rome: Accademia
nazionale dei Lincei.
Pizzamiglio, Pierluigi, ed. 1992. Gerardo da Cremona. Annali della Biblioteca statale e libreria civica di
Cremona XLI. Cremona: Biblioteca statale e libreria civica.
Ravai~son, Felix. 1849. Catalogue general des manuscrits des bibliotheques publiques des dq,artements. Vol. I.
Pans: PIon.
VII
The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 273
Ricklin, T. 1995. Die 'Physica' und der 'Uber de causis' im 12. jahrhundert. Freibourg: Universitatsverlag
Freiburg Schweiz.
Rubiera Mata, M. J. 1991. "Les premiers Mores convertis ou les premices de la tolerance." In Cardaillac
1991.
Sabra, Abdelhamid I. 1984. "The Aristotelian Revolt against Ptolemaic Astronomy: Averroes and al-
Bitriiji." In Transformation and Tradition in the Sdences: Essays in Honor of 1. Bernard Cohen, edited by E.
Mendelsohn, 133-53. Cambridge: Cambridge University Press.
Sarton, George. 1950. Introduction to the History of Science, Vol. 11.1. Baltimore: Williams and Wilkins.
Schlanger, Jacques. 1968. LA philosophie de Salomon Ibn Gabirol. Leiden: E. J. Brill.
Sezgin, Fuat. 1974. Geschichte der arabischen Schrifttums. Vol. 5. Leiden: E. J. Brill.
Steinschneider, Moritz. 1904. Die europiiischen Obersetzungen aus dem arabischen bis Mitte des 17.
jahrhunderts. Vienna: Carl Gerold's Sohn.
Sudhoff, Karl. 1914. "Die kurze Vita und das Verzeichnis der Arbeiten Gerhards von Cremona." Archiv
fUr Geschichte der Medizin 8:73-82.
Van Riet, Simone. 1968-92. Avicenna LAtinus. Louvain and Leiden: Editions orientalistes E. J. Brill.
Villanueva, E M. 1996. "In lingua tholetana." In idem, ed., El Escuela de Traductores de Toledo, 23-34.
Toledo: Diputaci6n provincial de Toledo.
Vuillemin-Diem Gudrun and Marwan Rashed. 1997. "Burgundio de Pise et ses manuscrits grecs
d'Aristote: Laur. 87.7 et Laur. 81.8." Recherches de theologie et philosophie medievale 64:136-98.
Willner, Hans, ed. 1903. Adelard of Bath: De eodem et diverso, Beitriige zur Geschichte der Philosophie des
Mittelalters, edited by C. Baeumker, IV.l. Miinster: Aschendorff.
Wiistenfeld, Ferdinand. 1877. Die Obersetzungen Arabischer ~rke in das LAteinische seit dem XI.
Jahrhundert. Abhandlungen der Koniglichen Gesellschaft der Wissenschaften zu Gottingen, 21.
Gottingen: Dieterich 'sche Verlags-Buchhandlung.
Appendix I
A Critical Edition of the Vita, Commemoratio librorum and Eulogium of Gerard of
Cremona
The Vita, Commemoratio librorum and Eulogium have been transmitted in the
following manuscripts:
• A: Erfurt, Wissenschaftliche Bibliothek der Stadt, Amplon. Fol. 266a, s. xiie, fo1.
126v
• B: Oxford, Bodleian Library, Ashmole 357, s. xiv, fo1. 57r (list of astronomical
works only)
• L: Leipzig, Universitatsbibliothek, 4° 1119, s. xiii!, fo1. 39r
• La: Laon, Bibliotheque municipale, 413, s. xiv (Italian), fo1. 100v (Commemoratio
librorum only)
• M: Leipzig, Universitatsbibliothek, 4° 1148, s. xiv\ fols 233v-234v
• 0: Oxford, All Souls, 68, s. xiii, fo1. 111r (Commemoratio librorum only)
• P: Paris, Bibliotheque nationale de France, lat. 14390 (from StVictor), s. xiv, fols
223r-v
 VII
•
" VII
274
• V: Citta del Vaticano, Biblioteca Apostolica, Vat. lat. 2392, s.xiii, fols 97v-98r
• W: Citta del Vaticano, Biblioteca Apostolica, Vat. lat. 2393, s. xv, fol. 100v
In ~ddition, most of ~he Vita is quoted by Francesco Pipino (d. 1316) in his
Chronrcon, ed. Muraton, Rerum Italicarum scrip to res ab anno aerae Christianae
quingentes.imo ad millesimu~q~ingentesimum, Milan, 1723-51, IX, col. 600-01 (= Pi).
* . The Vita, ~ommemoratlO hbrorum and Eulogium have been edited in whole or part
In the followmg works:
1) [Anonymous] 1710-40, X, 286-89 (excerpts from MSS VW); cited Boncom-
pagni 1851.
2) Ravaisson 1849, 218-19 (MS La).
3~ ~oncompagni 1851 (MSSVPi and readings from B, 0 and W); this includes a
facSImIle oN.
4) Lecler~ 1,876, Il, 402-07 (French translation of Vita and transcription of
CommemoratlO lzbrorum and Eulogium, all from MS P; occasional mention of readings
fromVLaO).
5) Wiistenfeld 1877, 56-77 (MSS LM with occasional readings from VP).
6) Sudhoff 1914 (MSS ALMV; L is used as the base manuscript).
~odern studies and translations are based on these editions (especially 3, 5 and 6),
a?d mclude the valuable English translation and commentary of Michael McVaugh
(In Gra~t 1974, 35-38), and the annotations on the list of translations by George
Sarton (~~ Sarton 1950, 338-44), and Richard Lemay (in Lemay 1978).1 None of
~hes~ edItIOns, however, include readings from all the known manuscripts, and none
IS ~a~Isfactory from a philological point of view. 2The following is a first attempt at an
edItIOn of the Vita, Commemoratio librorum and Eulogium, paying particular attention to
the philological aspects of the text. 3 MSS V and P are clearly the best manuscripts'
readings and, orthog~aphy 0~V4 have been preferred where both manuscripts giv~
equally plaUSIble :ers~ons. P IS the only manuscript to give all the section headings to
;~e C:0":,memor~tl.o hbrorum., after each of which he adds "R" (presumably for
rubnca ). AdditIOnal readmgs taken from other manuscripts or added by the editor
are placed in angle brackets: < >. In parallel columns to the list of Gerard's works
are pla~ed, the titles of t~e mathematical texts as they are given in the earliest and most
authontatIve ~anuscnpt: Pa~is, Bibliotheque nationale, lat. 9335 (late twelfth
century), descn~ed b~ A. A. BJ~rnbo (Bj0rnbo 1902), and checked personally.
All the te~ts, m thIS m~nuscnpt are included in the commemoratio librorum except
Tractatus Eucl,d,s de sp~cuz's (fols 82r-82v), Liber de aspectibus Euclidis (fols 88v-92r),
two shor~ works on tngonometry - Liber Saydi Abuothmi (fols 125v-126r) and Liber
~de:am~tl ,(fols 126r-126v) - and a work on arithmetical puzzles: Liber augmenti et
d,mmutlonrs, ~oca~us numeratio divination is, ex eo quod sapientes Indi posuerunt, quem
Abraha,"'. comp,zavlt ~t secundum librum qui Indorum dictus est composuit (fols 126v-133v).
In addItIOn, th,e scnbe has added on fol. 28v the order in which certain geometrical
and astronomIcal texts should be read, according to "Iohanicius" (presumably
,
The Coherence oJ the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 275
f::Iunayn ibn Isl:Iaq, the major translator of works from Greek and Syriac into Arabic),
using the same format as that of the compilers of the commemoration librorum (whose
numbers are added after the relevant texts):
Ordo qui est post librum Euclidis (4) secundum quod invenitur in scriptis Iohanicii:
Euclidis de aspectibus tractatus unus; Theodosii de speris tractatus tres (5); Autolici de
spera mota tractatus unus (30); Euclidis de apparentibus tractatus unus; Theodosii de locis
habitabilibus tractatus unus (26); Autholici de ortu et occasu duo tractatus; Theodosii de
die et nocte duo tractatus; Esculei de ascensionibus tractatus unus (27); Arsodochii de
elongationibus planetarum et earum (sic) magnitudinibus tractatus unus.
Modern identifications of the works are given in brackets. 5
< Vita; MSS ALMVPPi >
Sicut lucerna relucens in abscondito non est ponenda neque sub modio, sed6
supra7candelabrum locanda, sic nec splendida facta8 bonorum, velut sub pigra9
taciturnitate 10 sepulta, sunt reticenda,l1 sed auribus modernorum presentanda, cum
virtutis ianuam 12 sequentibus 13 aperiant 14 et antiquo rum exempla 15 quasi vite
ymaginem oculis presentium digna commemoratione 16 insinuent. Ne igitur17
magister Gerardus 18 Cremonensis sub taciturnitatis tenebris 19 1ateat,20 ne fame gratiam
quam meruit, amittat,21 ne pe~2 presumptuosam rapinam libris ab ipS023 translatis 24
titulus infigatu~5 alienus,26 presertim cum nulli eorum nomen suum inscripsisset,
cuncta opera ab eodem translata, tarn de 27 dyaletica28 quam de geometria, tarn de 29
astrologia30 quam de phylosophia,31 tarn etiam32 de physica quam de aliis scientiis,33 in
fine huius Tegni,34 novissime ab eo translati - immitando Galienum de 35 com-
memoratione suorum librorum36 in fine eiusdem - per sOCiOS37 ipsiuS38 diligentissime
fuerunt connumerata, ut, si aliquis intentionum39 ipsorum amator de eis aliquid
optaverit,40 per hanc inscriptionem citius inveniat et de eo securior41 fiat. Licet42 enim
fame gloriam spreverit, licet favorabiles laudes et vanas 43 seculi pompas 44 fugerit, licet
nomen suum nubes et inania captand0 45 nollet46 dilatari,47 fructus tamen operum eius
per secula redolens probitatem48 ipsiuS49 enuntiat atque declarat.
Is50 etiam cum bonis floreret temporalibus, bonorum tamen51 affiuentia vel absentia
eius animum nec extulit nec depressit, sed viriliter duplicem oc < c > ursum 52
fortune 53 patiens,54 semper in 55 eodem statu constantie permanebat;56 carnis desideriis
inimicando,57 solis58 spiritualibus adherebat,59 cunctis etiam60 presentibus atque futuris
prodesse laborabat,61 non immemor62 illius Ptolomei: 63 'cum fini 64 appropinquas,
bonum cum augmento operare.'
Et cum ab 65 istis66 infantie 67 cunabulis68 in gremiis phylosophie 69 educatus esset et
ad cuiuslibet partis70 ipsius71 notitiam secundum72 Latinorum studium pervenisset,
amore tamen73 Almagesti, quem74 apud Latinos minime reperit/ 5 Toletum76
per < r > exit, ubi librorum77 cuiuslibee8 facultatis habundantiam79 in Arabic 0 cernens
et80 Latinorum penurie de ipsis quam 81 noverat miserans,82 amore transferendi
VII VII

276 The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 277

linguam < e > didicit83 Arabicam, et SiC 84 de utroque -. de85 scientia videlicet86 et < 10> 126 LiberTrium Fratrum tractatus 55r Verba filiorum Moysi filii Sekir, id
ydiomate - confisus,87 quemadmodum Hametus88 in epistula sua De proportione et .i. est Maumeti, Hameti, Hasen (The Banii
proportionalitate89 refert: 90 'Oportet ut91 interpres92 preter excellentiam93 quam Miisa, On Geometry)
adeptus est ex notitia lingue de qua et94 in quam95 transfert, artis96 quam transfert97 < 11 > Liber Ameti127 de proportione 64r Epistola Ameti filii losephi de
scientiam habeat,' more prudentis, qui 98 virida99 prata perlustrans, coronam de et proportionalitate tractatus .i. proportione et proportionalitate
floribus - non de omnibus, sed de pulcrioribus - connectit,l00 scripturam101 revolvit (Al).mad b. Y iisuf, On Ratio and
Arabicam, de qua plurium 102 facultatum libros quoscunque 103 valuit 104 elegantiores 105 Proportion)
92v Abbacus (lbn cAbd al-BaqI, The
Latinitati tamquam 106 dilecte heredi, planius ac intelligibilius quo ei possibile fuit, < 12> Liber ludei super decimum
usque ad finem vite sue 107 transmittere non cessavit. Viam autem universe carnis Euclidis tractatus .i. Book of the Jew on the Tenth Book of
Euclid)
ingressus est anno vite sue.lxxiii.o, in anno domini nostri lhesu Christi.mclxxxvii.o 108 129
< 13 > Liber Alchoarismi 128 de 110v Liber Maumeti filii Moysi
l30 131
iebra et almucabula tractatus .i. Alchoarismi de algebra et almuchabala
(al-Khwarizmi, Algebra)
< Commemoratio librorum; MSS A (until 23) B (21-32) Lla (33-68) 0 (4-71) < 14> Liber de practica132 geometrie 116v Liber in quo terrarum
LMPV> tractatus .i. corporumque continentur
mensurationes Abhabuchri, qui
dicebatur Heus, translatus a magistro
Hec vero 109 sunt nomina librorum quos transtulit: 110 Girardo Cremonensi in Toleto de
Arabico in Latinum, abreviatus (Abii
Bakr, On Terrestrial Measurements)133
De dialetica 111
< 15 > 134 Liber Anaritii 135 super (An-Nayrizi, Comm. on Euclid's
Paris, BNF, lat. 9335 and identification
Euclidem136 Elements)
< 1 > Liber analeticorum 112 (Aristotle, Posterior Analytics) < 16> 137 Liber datorum Euclidis (Euclid, Data)
posteriorum Aristotilis tractatus .ii.
tractatus .i.
< 2 > Liber commentarii Themistii 113 (Themistius, Comm. on Aristotle's Post. < 17 > Liber Tidei 138 de speculo 84r Sermo de eo quod homo in speculo
super posteriores analeticos 114 tractatus Anal.) tractatus .i . < videt > et in eo quod non est
.i. 115 speculum et de causis illius, quem
< 3 > 116 Liber Alfarabii de silogismo (al-Farabi, On the Syllogism) collegit ea ex libris antiquorum Tideus
filius Theodori a Ruegoiu (?) medicus
De geometria 117 (Diocles, On Burning Mirrors)
< 4 > Liber Euclidis 118 tractatus .xv. (Euclid, Elements) < 18> 139 Liber Alchindi 140 de 75r Liber lacob Alkindi de causis
< 5 > LiberTheodosii 119 de speris 1r Liber Theodosii de speris aspectibus tractatus .i. diversitatum aspectus et dandis
tractatus .iii.120 (Theodosius, Spherics) demonstrationibus geometricis super eas
< 6 > Liber Archimedis 121 tractatus .i. 122 28v Liber Arsamithis de mensura circuli (al-Kindi, On Optics)
(Archimedes, On the Measurement of the < 19> Liber divisionum141 tractatus .i. (Euclid, Book of Divisions or lbn cAbd
Circle) al-Baqi, On the Division of Suifaces?) 142
< 7 > Liber de arcubus similibus 30r Epistola Abuiafar Ameti filii losephi < 20 > Liber carastonis 143 tractatus .i. (Thabit b. Qurra, Book of the Roman *
tractatus .i. de arcubus similibus (Al).mad b. Y iisuf, Balance)
On Similar Arcs)
< 8 > Liber Milei l23 tractatus .iii. 32v Liber Milei de figuris spericis De astrologia144
(Menelaus, On Spherical Figures) < 21> Liber alfagrani l45 continens (al-Farghani, Rudiments)
< 9 > Liber Thebit 124 de figura alkata 125 (Thabit b. Qurra, On the Sector-Figure) capitula .XXX. 146
tractatus .i. < 22 > Liber Almagesti tractatus .xiii. (Ptolemy, Almagest)
VII
278
< 23 > Liber introductorius 147
Ptolomei 148 ad artem spericam 149
< 24 > Liber Iebri 150 tractatus .viiii.
< 25 > Liber Messehala 151 de orbe
tractatus .i.
< 26 > Liber Theodosii 152 de locis
habitabilibus tractatus .i.
< 27 > Liber Esculegii153 tractatus .i.
< 28 > LiberThebith 154 de expositione
nominum Almagesti tractatus .i. 155
< 29 > Liber Thebit 156 de motu
accessionis et recessionis 157 tractatus i.
< 30 > Liber Autolici 158 de spera mota
tractatus .i.159
< 31 > Liber tabularum Iahen 160 cum
regulis suis.
< 32 > Liber de crepusculis tractatus
.i. 161
De phylosophia 162
< 33> 163 Liber Aristotilis de 164
expositione 165 bonitatis pure. 166
< 34 > Liber Aristotilis de naturali
auditu tractatus .viii.
< 35 > Liber Aristotilis 167 celi et mundi
tractatus quatuor. 168
< 36 > Liber Aristotilis de causis
proprietatum et 169 elementorum 170
quatuor 171 tractatus primus; 172 tractatum
autem secundum non transtulit eo
quod 173 non invenit eum 174 in Arabico 175
nisi de fine eius parum. 176 177
< 37 > Liber Aristotilis de generatione
et corruptione
(Geminus of Rhodes, Introduction to the
Phenomena)
Oabir b. Aflal;t, On the Flowers from the
Almagest)
(Masha'aliah, On the Orb)
25r Liber Theodosii de locis in quibus
morantur homines (Theodosius, On
Habitable Places)
22r Liber Esculei de ascensionibus
(Hypsicles, On the Rising of the Signs)
23v Liber quem edidit Tebit filius Chore
de his que indigent expositione
ante quam legatur Almagesti (Thabit b.
Qurra, On the Exposition ofTerms in the
Almagest)
141 r Tractatus patris Asen Thebit filii
Core in motu accessionis et recessionis
(Thabit b. Qurra, On the Forward and
Backward Motion of the Stars)
19r Liber Autoloci de spera mota
(Autolycus, On the Moving Sphere)
(Ibn Mucadh, Tables ofJaen)
(Ibn Mucadh, On the Dawn)
(Pseudo-Aristotle, De causis)
(Aristotle, Physics)
(Aristotle, On the Heavens)
(Pseudo-Aristotle, On the Causes of the
Properties and the Four Elements)
(Aristotle, On Generation and Corruption)
VII
The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 279
< 38 > Liber Aristotilis methaurorum178 (Aristotle, Meteorology, Books I-Ill)
... 179
tractatus .m.; quartum autem non
180
181
transtulit eo quod sane invenit eum
translatum.182 183
< 39 > 184 Tractatus unus Alexandri 185 (Alexander of Aphrodisias, On Time, On
Affrodisii 186 de tempore et alius de sensu the Senses, and That Augment and Increase
et alius de eo quod augmentum et Occur in Form, not in Matter)
incrementum fiunt 187 in forma et non in
yle.
< 40 > Distinctio 188 Alfarabii 189 super (al-Farabi, Comm. on Aristotle's Physics)
librum Aristotilis de naturali auditu.
< 41> 190 Liber Alkindi de 191 quinque 31 v Liber de quinque essentiis quem
essentiis. Iacob Alchildus (sic) filius Y saac
compilavit de dictis Aristotilis (al-Kindi,
On the Five Essences)
< 42 > Liber Alfarabii 192 de scientiis. 143v Liber Alfarabii de scientiis
translatus a magistro Girardo
Cremonensi in Toleto de Arabico in
Latinum (al-Farabi, On the Classification
of the Sciences)
< 43 > Liber Iacob 193 Alkindi 194 de (al-Kindi, On Sleep and Vision)
.. 195
sompno et vlSlone.
De fisica 196
< 44 > 197 Liber Galieni de elementis 198 (Galen, On the Elements)
tractatus .i. 199
< 45 > Expositiones2OO Galieni super (Galen, Comm. on Hippocrates' Treatment
lib rum Y pocratis de regimine acutarum ofAcute Diseases)
202
egritudinum201 tractatus .iiii.
(Pseudo-Galen, Secrets of Medicine)
< 46 > Liber de secretis Galieni
tractatus .i.
< 47 > Liber Galieni de (Galen, On the Temperaments)
complexionibus tractatus .iii.203
< 48 > Liber Galieni de malitia (Gal en, On the Evils of an Unbalanced
complexionis diverse 204 tractatus .i. Temperament)
< 49 > Liber Galieni de simplici (Galen, On Simple Medicines)
medicina tractatus .v.
< 50 > 205 Liber Galieni206 de creticis (Galen, On Critical Days)
diebus tractatus .iii.
< 51 > Liber Galieni de crisi tractatus (Galen, On Crises)
... 207
.111.
 VII VII
280 The Coherence of the Arabic-LAtin Translation Program in Toledo in the Twelfth Century 281

< 52 > Liber Galieni de expositione
libri Ypocratis in pronosticatione 208
tractatus .iii.
< 53 > Liber veritatis209 Ypocratis
tractatus .i.210
< 54 > Liber Y sac 211 de elementis
tractatus .iii.
< 55 > Liber Y sac212 de 213 descriptione
rerum et diffinitionibus earum214 et de
differentia inter descriptionem et
diffinitionem215 tractatus .i. 216
< 56 > Liber Abubecri217 Rasis qui
dicitur Almansorius218 tractatus.x. 219
< 57 > Liber divisionum220
continens.cliiii. capitula cum quibusdam
confectionibus eiusdem.
< 58 > Liber Abubecri221 Rasis
introductorius in medicina parvus.
< 59 > Pars222 libri Albenguesim223
medicinarum simplicium224 et ciborum.
< 60 > Breviarius225 Iohannis
Sarapionis 226 tractatus .vii. 227
< 61 > 228 Liber Azaragui229 de
cirurgia230 tractatus .iii.
< 62 > 231 Liber Iacob232 Alkindi233 de
gradibus tractatus .i.
< 63 > 234 Canon Aviceni 235 tractatus
.V.
236
< 64> Tegni237 Galieni cum
expositione Ali Abrodoan238
De alchimia239
< 65 > Liber divinitatis240 d e241 .lxx.242
< 66 > Liber de aluminibus243 et salibus
(Galen, Comm. on Hippocrates'
Prognostics)
(Pseudo-Hippo crates, Book of the Truth)
(Isaac Israeli, On the Elements)
(Isaac Israeli, On the Description ofThings
and their Definitions)
(ar-RazI, The Book ofAlmansor)
(ar-RazI, The Book of Divisions)
(ar-RazI, Short Introduction to Medicine)
(Ibn al-Wafid, Book of Simple Medicines
and Foods)
(Yal)ya b. Sarafyun, Breviary)
(Abu-I-Qasim az-ZahrawI, Surgery)
135r Liber Iacob Alkindi phylosophi de
gradibus (al-KindI, On Degrees of
Compound Medicines)
(Avicenna, Canon)
(Galen, Tegni, with the Comm. of cAll
ibn RiQwan)
Oabir b. I::Iayyan, Book of Divinity)
(Pseudo-RazI, OnAlumens and Salts)
< 70 > Liber de accidentibus alfel. 252 140r Alfeal secundum motum lune
(Divination according to the Movement of
the Moon)
< 71 > Liber Anohe253 et est tamquam 151 v Liber Anoe. In hoc libro est
sacerdocii mar(tyro)logium254 t .. xiii. 255 rememoratio anni et horarum eius et
reditionum anoe in horis suis et
temporis plantationum et modorum
agriculturarum et rectificationum
corporum et repositionum fructuum,
Harib filii Zeid episcopi quem
composuit Mustansir imperatori eAn'b
b. Sacd, Calendar)
Rasis Abubecri 256 fecit alhaugue57 et almansorium et divisiones. 258
Albucasim259 fecit azaugui 260 et261 eius cirurgiam,262 cuius cirurgiam263 transtulit
Magister Gerardus. 264
Aviceni Alboali265 fecit canonem.
< Eulogium; MSS LMPV>
Gerardus nostri fons, lux et gloria268 cleri,
266 267
auctor69 consilii, spes et solamen egeni,
voto 270 carnali fuit hostis 271 spirituali
applaudens,272 hominis splendor fuit interioris.
Facta viri vitam studio florente 273 perhennanr74
viventem famam 275 libri quos transtulit ornant. 276
Hunc sine consimili277 genuisse Cremona superbit,
Tolete78 vixit, Toletum279 reddidit astris. 280
Deo gratias. 281
Notes to Appendix I
I See also the discussion in Steinschneider 1904,16-32, in Ricklin 1995,71-76, and in Pizzamiglio ed. 1992,
3-7.
2 The text has suffered from the fact that Leclerc andWiistenfeld were bitter enemies, and that the manuscript-

< 67 > Liber luminis luminum
De geomantia244
* < 68 > Liber45 geomantie de artibus
divinatoriis 246 < qui incipit:
estimaverunt Indi. > 247
< 69 > 249 Liber alfadhol,250 id est dharab
de bachi251
(pseudo-RazI, The Light of Lights) readings of the most recent edition (that ofSudhofl) cannot be relied upon.
3 I am grateful to the librarians of the University Library in Leipzig, the Amplonian Library in ErfUrt, and the
Vatican Library, for sending me photographs of their respective manuscripts. MSS B, La and W have not been
consulted personally; their readings have been taken from the secondary works mentioned above.
(Geomancy) 248 • The orthography of V has not been followed where the scribe inserts an extra 'h', writes 'ngn' for 'gn'
('dingna', 'Tengni') or 'et' for 't' ('tolectum') or 'tt' (,mict-'). These spellings may reflect the scribe's
vernacular.
S These agree with the identifications in McVaugh (in Grant 1974), except where mentioned otherwise.
(Alfadhol, Book of Lots) 6 P om.
7 super LMP.
VII
282
,
!
The Coherence of the Arabic-Latin 'Hanslation Program in Toledo in the Twelfth Century
VII
283

8 famaM. 60 etiam A om.

9 A adds 'bonorum'. 61 laborat A.
10 tacitur morte M.
62 inmemo A, inmemor P.
11 reticeda L, recitenda M, retinenda P. 63 Tholomei MP; L adds in margin 'morale verbum'.
12 iamiam P. 64filiiM.
13 sequentibus L om. 65 L om.
14 apereant M, apperiant P.
66 ipsis ALMPPi.
15 exemplo M. 67 in facie L, insince (?) M.
16 commemoratione A] cum memoratione LMV: 68 cunabilis M.
17 ergo MP. 69 ph(ylosoph)ie AMPiV, phisice L, ph'ye P.
18 G. etardus A, Girardus LMP. 70 artis Pi.
19 tenebri L. 71 illius M, Pi om.
20 luceat L before co"ection, lateat L cifter comction. 72 si M.
21 admictat v: 73 tantum AL.
22 per L om. 74 quod M.
23 eo MP. 75 repirit LM, reperiit APV:
24 translatatis L. 76 tolectum v:
25 infingatur L. 77 in librorum copiam M, libros Pi.
26 abenus M. 78 cuiusque AL.
27 L om. 79 Pi om.
28 dialetica ALP. 80 A om.
29 LP om. 81 quas M.
30 astronomia M. 82 miserat L.
31 phya A, phisolophia L, phica et M, phylosophya v: 83 didicit v:
32 etiam AL om. 84 si L.
33 tarn de dyaletica ... scientiis]first excerpt in Pipino's Chronicon. 85 utroque de] utraque L.
34 TengniV:
86 scilicet P.
35inM. 87 confixus L.
36 librorum suorum AMP.
88 Hametus L om., hannerus A.
37 sotiosv. 89 proportionato M.
38 ipsius M om. 90 referri A.
39 intencionem L. 91 M om.
40 optavit M. 92 interpretes v:
41 de eo securior] de certior M, de eo certior P. 93 excellencia L.
42 The passage from this word until the end of the Vita is quoted by Pipino, with the exception of the words 94 AL om.
"quemadmodum Hametus ... revolvitArabicam." 95 quaV:
43 novas Pi. 96 artes M.
44 pompas L om.
97 artis quam transfert] P om.
45 in mania captanda M. 98 quia LM.
46 noluerit Pi (perhaps co"ectiy). 99 viridia P.
47 dilari LM. 100 conectit L.
48 probitataton proprietate M. 101 A om.
49 eius P. 102 de qua plurimum ALp, de quamplurium Pi.
50 quo M. 103 libros quoscumque] quosque P.
51 cum M. 104 voluit MPPi, L adds 'quam'.
52 cursumM. 105 elonganciores A.
53 duplice fortune occursum P. 106 numquam M, tanquam P.
54 patiens fortune AM. 107 ALMP om.
55 inA om. 108 mc1xxxiiii W.
56 permanebis M. 109 ALaOP om.
57 inmutando M. 110 La adds 'Magister Girardus Cremonensis'; 0 adds 'Magister Girardus Cremonensis in Toleto'.
58 solumM. 111 De dialetica V] de dialectica p, ALMO om.
59 detrahent (or detraherit) A. 112 amaleticorum LMP.
VII VII
284 The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century 285

113 themesii A, ithemistii L, temistii M, Temetistii P. 165 compositione La.

114 analencorum P. 166 pure bonitatis M.
115 primus P.
167 PV om.
116 L omits 3.
168 tractatus quatuor] La om.
117 De geometria] P only.
169 LaV om.
118 etuclidis A.
170 helementorum V.
119 TeodosiiV.
171 LaV om., tractatus quatuor P.
120 tertius P.
172 tractatus primus] LP om., primus tractatus 0.
121 Archimedis A, archimenidis LMP, archinenidis 0, archimendis V. 173 quia L.
122 primus P.
174 P om.
123 milci M, misey P.
175 arabicum M.
124 thobit L, tebith M.
176 parum in ipsius fine M.
125 arthacata A, albeata L, alchata MP.
177 tractatum autem ... parum] P om.
126 M om. 10-12.
178 meteororum L, metheororum LaM, metaurorum 0, metaherorum P.
127 Hameti LMo.
179 quantum L, P om.
128 argorismi A, acharismid. L, argorisim M, algorismi OP. 180 P om.
129 L om.
181 eo quod] quia P.
130 gebra M.
182 invenit eum sane translatum M, sane translatum invenit P.
131 almucabala AO, ali\mil cabala L, altacabila M, almicubila P. 183 quartum autem ... translatum] La om.
132 pratia P, pratica OV.
184 La omits 39.
133 For discussion of the identity ofAbii Bakr, see Sezgin 1974, 389-91. 185 aserande M.
134 M omits 15-16. 186 afrodisii LMP, anfrodisii 0.
135 avaricii L.
187 et incrementum fiunt] fuit L.
136 A adds 'tractatus'; OP adds'tractatus.i.'. 188 distinctio LM om.
137 A transposes 16 and 17; P omits 16. 189 Affarabii LM, alfharabii 0.
138 thidei MP.
190 MLa omit 41.
139 A omits 18-20.

140 askimii M.
191 Alkindi de] alchinididus L, A1chindi de
192 alfharabii 0, A1pharabii P.
° P.

141 demonstrationum MP. 193 M om., iacobi P.

142 Sezgin 1974,387-88.

143 tarastonis M; °
omits 20.
194 alchindi LOp, A1chini La, alkemii M.
195 sompno in visione L, sopno et visione V.
144 De astrologia VP] Isti sunt libri ast(rolog)ie quos transtulit Gerardus Cremonensis de Arabico in Latinum 196 De fisica V] LMO om., de physica P.
B,ALMOom. 197 La swaps 44 and 45.
145 affagani L, alfragan M, alfragani P. 198 esis M.
146 xxx.i. M,.xx. V. 199 ii. LaM.
147 introductoriis L.
200 Expositionem La.
148 tholomei Ap, talemei M, podomiei 0.
201 acutarum egritudinum] acutorum egritudinum Lp, egritudinum acutarum M, acutorum La.
149 A ends. 202 iii LM, 3 P. Early manuscripts of this work consist of only three books; apparently the fourth book was translated later;
ISO rebi M. the relative roles of Constantine the African and Gerard are unclear.
151 messobala L, messanala M, messzhala 0. 203 tractatus.ii M, La om., tractatus 3 P.
152 todosii L; 0 omits 26.
204 laP om., diverso o.
153 esclilegii M, Cusculei P.
205 La swaps 50 and 51; 0 omits 50.
154 thebith post nominum L, thelith M, thebiti P, thelith M, thebit 0, tebit V. 206 From here onwards La omits 'Liber Galieni'.
155 Almagesti tractatus.i. M om. 207 i. La.
156 thebith L, thebth M.
208 pronostic. M, prognostic. 0.
157 translationis P. 209 virtutis M.
158 autolosicide M, actolici P. 210 iii. LaP.
159 tractatus.i.] P om. 211 ysaac LMOP.
160 et ahen L, jaberi P. 212 isaac L, ysaac MOP.
161 B finishes here. 213 M om.

162 De phylosophia] LMO om., de phylosophya P, de phylosophyia (sic) V. 214 et diffinitionibus harum L, La om., et de diffinitionibus earum 0.
163 La starts here. 215 diffinitionem et descriptionem OV.
164 MO om. 216 i.] L om., primus P.
VII VII
286
The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century 287

217 erubecri L, albubatri M, albubecri P, abubetri V.

269 actor vw.
218 almasorius LP, almansor La.
219 i. P. 270 foto M in rubrication but 'voto' in instructions for rubrication.
271 hostis MV.
220 V adds 'almansoris'.
272 Ad plaudens P.
221 abubetri L, albubetri M, albubecri P, V om.
222 Ars La. 273 fiorento M, fiorere P.
274 perhennat M.
223 abenguefiti LV, albenguesui La, abenguefeti 0, albenguenfin M. 275 vincentem famam M, viventem formam L.
224 simplitium V.
276 hornant L.
225 breviaarius M.
m consilii L, consilio W.
226 sarapionis iohannis L, iohannis serapionis MLaP. 278 tolecti V.
227 xv. 0, 8 P.
279 toledum M, tolectum V.
228 La omits 61.
280 artis L.
229 arazugni L, azaragrii M, azarugui Op, azatugui V. 281 Deo gratias] V only.
230 cyrugia M, cirugia V.
231 V exchanges 62 and 63.
232 V om., iacobi P.
233 achindi LLap, alkimi M.
234 La omits 63. Appendix 11
235 civconn M.
236 v.ZI L, ZI
237 Tengni V.
° (possibly '21'). The Eight Parts of Physics and Their Books
238 aliacro doan L, Haly Abrodahan La, hali abrodohan M,
239 Haly abrodoan
. OP. Al-FarabI, On the Classification of the Sciences, trans. Gerard of Cremona, In
De alchimia] P only. Gonzcilez Palencia 1932,161-63:
240 diminuitatis M.
241 L adds a lacuna, P om. Et dividitur scientia naturalis in octo partes magnas.
242lxxx. M.
(1) Quarum prima est inquisitio de eo in quo communicant corpora naturalis
243liminibus L, alluminibus La, alm'bus M, luminibus V.
244 De geornantia] P only. omnia, simplicia eorum et composita ex principiis et accidentibus consequentibus ilia
245 LM add 'de'. principia. Et hoc totum est in Auditu naturali.
246 de artibus divinatricibus La, de artibus divinantibus M, LOV om. (2) Et secunda est inquisitio de corporibus simplicibus an sint inventa et si sunt
247 qui incipit: estirnaverunt Indi] M only. inventa tunc que corpora sunt et quantus sit eorum numerus. Et hec est consideratio
248 See p. 267 above.
in mundo quid est et que sint partes eius et quot sint et quod ipse sunt in summa tres
249 La omits 69-71 and'Rasis Abubecri ... transtulit Magister Gerardus'.
250 alfadoch L, alfadolum M, alfodochi P. aut quinque. Et hoc est in consideratione in celo et discretione eius a reliquis partibus
251 id est dharab de bachi] V supra, id est de arab de bachi L (tharab M), id est de arab de biachi 0, de arabachi mundi et quod materia que est in eo est una. Et est in parte prima tractatus primi libri
P, i.e.) perhaps 4arab dhahabi= 'golden bough'. Celi et mundi ... usque ad finem tractatus primi libri Celi et mundi ... est in principio
252 alfeth L, alfa M, alfhel 0, alfeb (or alfeh) V. tractatus secundi libri Celi et mundi usque circiter duas tertias eius ... et hoc est iliud
253 anoche L, arite M, anhoe 0, anoe P.
254 mar(tyro)1oglUm . ] mar'1' quod consideratur in fine tractatus secundi et tercii et quarti libri Celi et mundi.
eglUm L, quatuor togiii M, P co"ects 'inar-' to 'marlogium'.
255 et est.. .. xiii.] OV om., ... t .. i. M. (3) Et tertia est inquisitio de generatione corporum naturalium et eorum
256 albutrati M, albubetri P, abubetri V. corruptione sive commutatione.... Et hoc est in libro De generatione et corruptione.
257 alhangui LM, alhagui P. (4) Et quarta est inquisitio de principiis accidentium et passionum que propria
258 L adds'lbi'.
259 albugafi L, albucasin M.
sunt elementis solum sine compositis ab eis. Et est in primis tribus tractatibus libri
260 azauguri L, titangin M, azaugin (with an abbreviation mark on the :g') P. Impressionum superiorum.
261 L om. (5) Et quinta est consideratio in corporibus compositis ab elementis et quod ex eis
262 cuius cyririgiam M, cuius cirugiam V, quam P. sunt similium partium et ex eis que sunt diversarum partium ... hoc est in tractatu
263 cyrurgiam M, cirugiam V.
quarto libri Impressionum superiorum.
264 Giraldus L, Girardus M, OV om.
265 Aviceni Avolai La, Avicenni aboali M, Viceni alboai P. (6) Et sexta, et est in libro Mineralium, est consideratio in eo in quo communicant
266 Girardus ML. corpora composita similium partium ...
267 uten M. (7) Et septima, et est in libro Plantarum, est consideratio in eo in quo communicant
268 gloria MV, regula LW. species plantarum ...
VII

288
VIII

(8) Et octava, et est in libro Animalium et libro Anime et qui sunt post utrosque
usque ad postremum Librorum naturalium, est consideratio in eo in quo communicant
species animalium et quod propriatur omni speciei eorum. Et est pars secunda
speculationis in compositis diversarum partium.
MICHAEL SCOT AND THE TRANSMISSION
OF SCIENTIFIC CULTURE FROM TOLEDO TO BOLOGNA
VIA THE COURT OF FREDERICK 11 HOHENSTAUFEN 1

Michael Scot is well-known in history and legend as the court

astrologer of Frederick 11. Yet, in spite of several important studies
in recent years, his life and his works are still shrouded in mysteryl.
His earlier career in Spain, his contacts with other Italian cities, and
the early reception of his works are less researched 3. One problem
has been to reconcile the person of Michael Scot the translator of
Aristotle, Avicenna, and Averroes, with Michael Scot the astrologer
and author of the three books of the Liber introductorius: i. e., the Liber
quatuor distinctionum, Liber particuiaris, and Liber physionomie 4 • In this
article explorations of the Spanish context of Michael's career and the
Italian transmission of his writings lead to raising some fundamental
questions about the authorship of the Liber introductorius.

I. I am grateful to the advise of several people, including Silke Aekennann,

Stefano Caroti, Jose Manuel Fradejas Rueda, Kristen Lippincott, Nige1 Palmer and
Johannes Thomann.
2. Since C. H. Haskins's pioneering article in his Studies in the History of Medie-
val Science, 2nd ed., Cambridge, Mass. 1927, 272-98, the following are some of the
most important studies: L. Thorndike, Michael Scot, London 1965; L. Minio-
Paluello, «Miehae1 Seot», in The Dictiot,ary of Scientific Biography, ed. C. C. Gillispie,
XI, New York 1974,361-65; R. Manselli, «La Corte di Federieo 11 e Miehe1e Seo-
to», in L'Averroismo it, Italia, Rome 1979, 63-80; P. Morpurgo, «Fonti di Miehe1e
Seoto», Accademia t,azionale dei Lincei, Rendicotlti de/la classe di scietlze morali, storiche
efilologiche, serie 8, 38 (1983), 59-71; Id., «Note in margine a un poemetto astrolog-
ieo presente nei eodici del Liber particularis di Miehele Seoto», Pluteus, 2 (1983), 5-
14; G. M. Edwards, The Liber Introductorius of Michael Scot, Ph. D. Diss University
of Southern California, 1978; Id., «The Two Redaetions of Miehael Seot's Liber
IntrodllctorillS», Traditio, 41 (1985), 329-40; see also the articles by S. Caroti, D. Jae-
quart and S. Williams in this volume.
3. For the first topic see the brief article of C. H. Haskins, «Miehae1 Seot in
Spain», in Homenaje a Bonilla y Sat' Martft" Madrid 1927-1929, 129-34.
4. It is quite clear from the proemillm that Miehael regarded all three parts (libn)
of his work as the Liber introd'lctorills, i. e., 'the book introducing astrology', to
VIII
Michael Scot and Toledo
Michael Scot first appears on the scene in Toledo. It was there that
he completed his translations from Arabic of al-Bi~riijl's De motibus
caelorum in 121 7, and Aristotle's De animalibus some time before 1220.
Here, too, he may have composed a medical recipe for a certain
'Philip' as he lay sick in the city of Cordoba I. In 1215 Michael had ac-
companied Rodrigo, archbishop of Toledo, to Rome for the Fourth
Latera~ ~ouncil, at which Rodrigo was demanding support for the
recogmtIOn of Toledo as the metropolitan See of Spain. Several schol-
ars have drawn attention to the importance of this record i. In Rome
he might hav~ met Etienne ofProvins, who would later be charged by
the church wIth the task of assessing the orthodoxy and regulating the
~hi~h he added a general preface (proemium) see ms. Miinchen, Bayerische Staats-
bIbhoth~k, ~l~ .10268 (hen~.eforth M), fol. Ir (beginning of proemiwlI): «Incipit
pro.hemlU~ hbn mtroductorn .. :»~.M, fol. 19V (end of proem ill m): «Volumus lib rum
toClUS ~.rtIS collect~m, pro. nov~clls .scolaribus incipere ordinate qui merito dici
potest mtroductorlUs.. ~IC. e~ml hber constat ex tribus libris: primus quidem
constat ex ~uatt~or dlstl?ctIOmb~s; secundus liber est simplex, et ipsum librum
appell~mus partlcularem ... ; terclUS vero liber dicitur 'physionomie' ... Et tunc
est epll~gus, ve~ud ante. omne~ libros de prenominatis prohemium ... Explicit
pr?heml~m tO~I~S ~pe~ls et pnncipali~er primi libri. Nunc incipit liber primus
edi[ c:]tus m 4 dlstmCtlombus quarum pnma hec est». This division is clear from the
rubnc.s t~ ~ach d!stitlCtio; see M, fol. I 18rb: «Explicit secunda distinctio primi libri,
nunc mCIpIt terCIa». Scholars have misleadingly given the title Liber introductorills to
the first part only.
* I. Thorndike, Michael Scot, 121. An ambiguous testimony is given by the Jewish
scholar Juda b. Salomon ha-Cohen who was born in Toledo ca. 121 5 and corre-
sponded with Frederick 11 ca. 12 33. He says that when he was still in Spain (i e
p~esumably ~oledo) «the philosopher of the Emperor» addressed some questions· t~
hIm conc~rnmg geometry and astronomy. This could suggest that either Michael
S~~t was m T~ledo at the time, or (as is more likely) he sent the questions from
~ICIly; see .C. Slrat, «Les traducteurs juifs cl la cour des rois de Sicile et de Naples»,
m Traducttott et traducteurs au Moyett Age, Colloque international du CNRS, Paris
19 89, 169-9 1 (175)·
. 2.. The doc~ment w~ch includes Michael's name (magister Michael Scotus) is ed-
Ited m~. F. R.Ive~ RecIO, «Personajes hispanos asistentes en 12 15 al IV Concilio
de Letra~», Htspama Sacra, 4 (1951), 335-38. It is referred to by M.-T. d'Alverny,
«TranslatIOns and Translators», in Rettaissattce attd Renewal in the Twelfth Century
ed .. R. L. Benson an~ G. Consta~le, Oxford 1982, 421-62 (456); P. Morpurgo:
«MIchele Scoto: tra SCIenza dell'amma e astrologia», Clio, 19 (1983), 441-50 (44 2),
n·.5; Edwards, «The Two Redactions», 329, n. I, and A. Paravicini Bagliani, «La
sCIenza. arab~ ne~a Roma del duetento: prospettive di ricerca», in La dufusiotte
delle ~ct~nze Islattltche nel M.edio ~vo europeo, Rome 1987, 103-66 (16 5). See also
S. WIllla~ls, «The Early Clrcu~atlOn of the Pseudo-Aristotelian Secret of Secrets in
the West. the Papal and Impenal Courts», Microloglls, 2 (1994), 12 7-44.
102
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
teaching of Aristotle's works on natural science; to him Michael was
to dedicate his translation of Averroes's Large Commentary on the De
caelo et mundo. Michaelmight also have met other scholars working in
the Papal Court, and would have attracted the papal support which
furthered his career in the I220S. What has not been pointed out,
however, is the significance of the fact that he should be in the retinue
of Rodrigo in the first place, alongside several other dignitaries of the
Cathedral of Toledo.
The archbishops of Toledo had patronized scholarship, and in par-
ticular, the translation of texts from Arabic, since the mid-twelfth cen-
tury. John of Seville, the first major translator of scientific works in
Spain, had dedicated the translation of a medical work to Raymond,
archbishop from 1 126 to I I 5 I. Raymond's successor, John (I 15 1-66)
saw the translating activity brought within the precincts of the Cathe-
dral. For the prologue and several parts of Avicenna's philosophical
encyclopedia, the Shifa', were put into good Latin by the archdeacon,
Dominicus Gundissalinus, and dedicated (where dedications survive)
to John. For understanding the Arabic text Gundissalinus at first had
the help of a Jewish philosopher called Avendauth, and later of a ma-
gister Johannes (Hispanus) who is probably the same as the man of that
name who was Gundissalinus's successor as archdeacon of Cuellar I. In
documents of I 174 and I 176 from the Cathedral the name Girardus dic-
tus magister occurs. Here we may be in the presence of the greatest of
the twelfth-century Arabic-Latin translators, Gerard of Cremona, who
translated the Almagest in I 175 I. Gerard was not only responsible for
his own translations, but also used, and may have revised, earlier ones
by John of Seville 3. In turn, Gerard's translation of al-rarabl's On the
Division of the Sciences was revised by Gundissalinus. This translating
activity must have continued after the death of archbishop John in
1166, for Gundissalinus is still signing documents in 1181 and Gerard
died in 1187, leaving pupils (socii) who drew up a long list of his works.
I. See J. F. Rivercl, «Nuevos datos sobre los traductores Gundisalvo y Juan
Hispano», Al-Attdallls, 31 (1966),267-80.
2. For the full documentation of the charters of Toledo Cathedral see the fine
edition of F. J. Hernandez, Los cartlllarios de Toledo, Madrid 1985, from which
I much of the infonnation here has been taken; see nos. 165 and 174.
3. This is the suggestion of R. Lemay in his «Fautes et contresens dans les tra-
J
ductions arabo-Iatines medievales: l'Introductorillm ill astrottomial1l d'Abou Mashar
~
de Balkh», ReVile de syuthese, 89 (1968), 101-23 (102-3).
i
10 3
I

VIII
«Hispanus» eventually became bishop of Segorbe and, according to
one document, accompanied archbishop Rodrigo to the Fourth
Lateran Council in 121 5I.
Thus (if we are right in our identification) Johannes Hispanus and
perhaps some of the socii of Gerard of Cremona were alive when
Rodrigo became archbishop in 1208, and we may presume that he
encouraged their interest in Arabic. For he is none other than the
Rodrigo Jimenez de Rada who wrote, or supervised the writing of,
the Historia Gothica and the Historia Arabum, both of which drew heav-
ily on Arabic sources 1. It is not surprising, then, that he should have
had in his entourage a person such as Michael Scot, who, in the words
of Pope Gregory IX was «skilled in Hebrew and Arabic» 3. A year after
Michael translated al-BhrujI's astronomical text, Salio, a canon of
Padua, translated a text on geomancy from Hebrew into Latin in
Toled0 4 • Moreover, it was a canon of Toledo Cathedral called Mark
who translated the Koran and the profession of faith of the Mahdi of
the Almohads, the ruling Arabic dynasty in Spain of the time. Mark
testifies as a deacon in the Cathedral from 1191, and a canon from 1198
to 1216, and dedicated his translations of Islamic religious texts to
Rodrigo, and to archdeacon Mauritius who suggested the project. He
also translated medical works s.
One might, then, see Michael Scot as a successor to Gundissalinus,
Gerard and Johannes Hispanus and a colleague of Salio and Mark. In
the charters of Toledo Cathedral a magister scholarum whose initial is
(cM» signs a document in 1208 6• Is it fanciful to believe that this could
I. Rivera, «Nuevos datos», 279. Rivera doubts the reliability of this document
from Zurich, on the grounds that Hispanus was in Toledo at the end of the year
for he died there on I I Dec. 1215.
2. See Historia de reb'ls Hispanie, ed. J. F. Valverde, Corpus Christianorum,
Continuatio mediaevalis 72, Turnhout 1987, 11: «[Rodrigo] conoda la lengua
arabe. De las cronicas musulmanas que debio de utilizar solo hemos podido local-
izar la del Moro Rasis de la que existia un codice en la Biblioteca de la Catedral de
Toledo».
3. Haskins, Stlldies, 272 .
4. See L. Thorndike, «A Third Translation by SaliO», SpeclIltlm, 32 (1957), 116-
17 (117): «Completus est liber Salcharie (or, Falcharie) Albassarith translatus a mag-
istro Salione de ebrayco in latinum Tholeti iio die exeunte Februario mOccoxviiio
anno Christi».
5. M.-T. d'Alverny and G. Vajda, «Marc de Tolede, Traducteur d'Ibn
Tiimart», AI-Andallls, 16 (1951), 99-140, and 17 (1952), 259-307.
6. One may note that the previously attested magister scholanml most frequently
104
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
be Michael Scot himself, whose interest in music (one of the functions
of the magister scholarum being to train the choir) is well attested in the
Liber introductorius? One may add that, since Michael had died by 12 36,
he could easily have arrived as a young man in Toledo by 1200, or
even earlier.
When we turn to Michael's translations it is clear that he is the in-
tellectual successor to Gundissalinus, Gerard and Johannes Hispanus.
In his translation of al-Bi~rujl he saved himself the trouble of translat-
ing the extensive quotations of Ptole my's Almagest by simply substitut-
ing the relevant passages from Gerard's translation I. In the case of the
Shiia' of Avicenna one could say that he took up the work of translat-
ing from the point where Avendauth and Gundissalinus had left off.
One can almost see an «Avicenna Latinus» project centred on Toledo
Cathedral, which had begun with the translation of the prologue of
the Shifa' by Avendauth in the time of Archbishop John and contin-
ued with the translations of the On the soul, Metaphysics and part of the
Logic by Avendauth and Gundissalinus. To this Avicenna project
Michael Scot added that of translating the works of the «Cordovan
Aristotelians» Averroes and al-Bi~ruJi, both of whom had died in the
very last years of the twelfth century 1. He also used Gundissalinus's
adaptation of al-Farabl's On the division of sciences for a division of
sciences of his own 3. In the Liber introductorius he quotes the translation
of Avicenbrol's Fons vitae made jointly by Gundissalinus and Johannes
Hispanus 4, and cites several propositions from the Liber de causis whose
translation was attributed to Avendauth in the earliest manuscripts s.
signs only with his initial; J. magister scolanml. Moreover, references to magistri are
not frequent in the charters of the Cathedral in this period.
I. Al-Bi~riiJi, De motiblls caeionml, ed. F. J. Carmody, Berkeley and Los Angeles
'
1952, 20-21. . . .
2. For this Aristotelian movement amongst the Arab and Jewish mteIlectuals m
Cordoba in the latter half of the twelfth century see A. I. Sabra, «The Andalusian
Revolt against Ptolemaic Astronomy: Averroes and al-Bi~riiJi», in Traniformation
atld Traditiotl itl the Sciences: Essays itl Honor of 1. Bernard Cohetl, ed. E. Mendelsohn,
New York 1984, 133-54·
3. The surviving extracts from this division of sciences are edited by L. Baur as
an appendix to his edition of Dominicus Gundissalinus, De divisiotle philosophiae,
Munster 190 3 (Beitrage zur Geschichte der Philosophie des Mittelalters, IV, 2-3),
39 8-400 .
4. Morpurgo, «Michele Scoto: tra scienza ~e?'anima e astr~logia», 45 0: . .
5. Michael refers to the first five proposltlOns of the Llber de callSIS m hiS
proemium (ms. M, fol. Ira). For the attribution to Avendauth see ms. Oxford,
10 5

VIII
As to Michael's contemporaries at Toledo, Salio of Padua's works
travelled with those ofMichael (as we shall see), and Mark of Toledo's
texts on pulses complemented Michael's on urines I • The continuation
of an «Avicenna» project and, perhaps, even the existence of an
«Averroes» project in Toledo before 1215 is suggested by the fact that
in the University of Paris in that year there was a prohibition against
the reading of Aristotle's works on metaphysics and natural science,
which included not only the summae of these works but also the writ-
ings of «Mauritius Hispanus». It has been suggested that this could
have been archdeacon Mauritius, who knew Arabic literature suffi-
ciently well to introduce Mark of Toledo to Islamic texts, and who
could have had a similar role as a promoter of translations in the early
thirteenth century as Gundissalinus had had in the late twelfth 2.
One might say that Michael Scot could have been introduced to the
works of Gundissalinus and his collaborators in Italy or France, and
above all in his native Great Britain which maintained a close relation-
ship with Spain and where several early manuscripts of the translations
survive. But it is worth noting that the earliest manuscript of the Fons
Vitae of Avicenbrol is still in the Cathedral library at Toledo 3 and that
the earliest manuscript of several of Gundissalinus's original works and
translations (including a portion of the Metaphysics of Avicenna) was
written by Spanish hands in the early thirteenth century4. Moreover
when we turn to the other sources of Mic ha el's original works we find
Bodleian Library, Selden Supra 24: «Explicit Metaphysica Avendauth»; see
d'Alverny, «Avendauth?», in Hometlaje a Millils Vallicrosa, I, Barcelona 1954,19-
43 (23)·
I. For Mark's translation of two Galenic texts on pulses see L. Thorndike and
P. Kibre, A Cataloglle of Incipits of Mediaeval Scientific Writings in LAtin, London
19~3, cols 303 and 1012. Michael's work on urines is printed in A. H. Querfeld,
MlChael ScottllS IInd seine Schrift De secretis natllrae, Diss., Leipzig 1919, 51-59. It is
closely related to the «Short Salernitan Urine-Treatise» discussed in G. Keil, «Der
'Kurze Harntraktat' des Breslauer 'Codex Salernitanus' und seine Sippe», Diss.,
Bonn 1969. This relationship was pointed out in R. Reisert, Der siebenkammerige
Utems, Hannover 1986 (Wiirzburger medizin-historische Forschungen, 39), 55.
2. See M.-T. d'Alverny, «Deux traductions latines du Coran au Moyen Age»,
Archives d'histoire doctrinale et litteraire dll Moyen Age, 16 (1947), 69- 131 (129-30).
3. Toledo Cathedral, ms. 95-21; see]. M. Millas Vallicrosa, LAs tradllcciones ori-
etltales ett los manllscritos de la Biblioteca Catedral de Toledo, Madrid 1942, 34.
4. On this manuscript - Vatican, Ott. Lat. 2186 - see M.-T. d'Alverny, «Les
traductions d'Avicenne. Quelques resultats d'une enquete», in Actes d/l Ve COttgres
ittternatiottal d'arabisants et d'islamisants, Bruxelles 1970, 154, and Avicenna, Philo-
sophia prima I-IV, ed. S. Van Riet, Louvain 1977, 125*.
106
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
that they too exist in manuscripts which could have been in Toledo in
Michael's time, or are copies of such manuscripts.
I have taken as a test case the sources of the third distinctio of the
Liber quatuor distinctionum as it is found in the Munich ~anuscript of
the (<long version» I. This distinctio is devoted to the practIce of «horary
astrology» - i. e., to answering questions concerning hidden or future
events (interrogationes) or to advising on the best time to embark on
particular activities (electiones) - these being determined b! the «figure
of the heavens» or «horoscope» at the time of the questIOn, and the
time of the inception of the activity respectively. This is in distinction
J to «genethlialogical astrology» (nativitates) which determines the char-
acter of the newly born and the events of its life from the horoscope
at the time of birth.
Michael Scot begins with an oft-cited description of the good
astrologer2. This is followed by (a) the use of the astrolabe, (~) the
procedure to follow when a client asks a question, (c~ the meanmg of
certain astrological terms, (d) an example of a partIcular case, (e) a
simple way of predicting from the «hour» of the question, with.o~t the
use of an astrolabe or tables, (f) descriptions of the characteristics of
the planets, and (g) significations of the planets as «lords of the year».
I. I. e., Miinchen, Bayerische Staatsbibliothek, Clm 102~8 (ms. M~. F?r t?e
two principle versions see L. Thorndike, «Manuscripts of Mlchael Scot s Ltber tn-
trodllctorillS», in Didascaliae: Swdies itt honor of Attselm M. :tlbareda, Ne.w Y?r~ 1?6.1,
42 5- 28 and Edwards, «The Two Redactions». The headmg of the third dtsttncttO m
the long version gives a misleading idea of its conten~s (M, f~l. ~ I.~rb): «Nunc
incipit tercia [distinctio] in qua continentu.r mul~a cap~tula de luditllS .~ulta~m
rerum secundum quod diversi auctores hulUs .artls sentlUnt..Et de nOtl~la a:tl~m
arismetrice, geometrie, cyromancie, geomanCle, et prophec~e». In .Pa~s, Blblio-
theque nationale, nouvelles ac.quisit~ons lat~nes 1401, the se~tlon .which IS devoted
to judgements and «interr?gatl0ns» IS de.scn?ed as «the con~muatlO.n o~ t~e secon~
distinctio» (Fol. 85 v: dnciplt no tu la doctnnalls et .est proherru~m hulUs libn s~que~
tis de interrogationibus et iuditiis ad interrogatlones petentlum et est co~tmuatlo
secunde distinctionis») whereas the third distinctio is headed: «super creatlOne cor-
poris et signis ac plane~is» (fo1. loor). Moreov~r t~ere is litde ~verlap betwe.en the
material in the Paris and the Munich manuscnpts m these sectlOn~. T~ese discr~p­
ancies indicate that considerable caution must be exercised in att.nbutmg any~hing
in either of these versions to Michael Scot, and further w~rnmg notes WIll be
sounded in the following section of this article. However, until the texts of the two
versions have been studied in detail I provisionally refer to the author of the ma-
terial in the Munich manuscript as «Michael Scot». .
2. See Morpurgo, «Note in margine», 9-Il, and Thorndike, Michael Scot,
9 2-95.
10 7
VIII

The final, longest, section (h) deals with interrogations and elections
, MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE

astrological lots and finding hidden things - was compiled in the

VIII

mid-twelfth century by Hermann of Carinthia I • Michael also uses one

on specific subjects.
of the very earliest Arabic-influenced astrological texts known in the
At various stages in his work Michael gives lists of authorities I.
West - the letter of Argafalau to Alexander. This was incorporated
However, one must be careful, because he often quotes an authority
into the Liber Alchandrei (Matllematica Alchandret), of which the earliest
at second or third hand and omits to mention his immediate source.
manuscript is of the late tenth century, and from which Michael takes
Moreover, he sometimes bolsters an authority by adding the names of
other authors. There is no alternative but to compare his text to that several chapters. .
In short, Michael's sources are mainly texts translated In Toledo
of the astrological texts themselves. This article marks a first attempt
and to a lesser extent, Barcelona 2. Of course, most of these texts had
at this task 2 • The results corroborate C. Haskins's statement that
a wide diffusion, and were known, for example, from Parisian manu-
Michael's sources are for the most part twelfth-century translations
scripts to the author of the Speculum Astronomiae in the mid-thirteenth
from Arabic into Latin made in Spain 3. These include the texts on the
century3. But it is worth noting that Michael's authorities correspond
construction and the use of the astrolabe attributed to John of Seville,
very largely to those in two manuscripts from the library of the
and his translations of Thabit ibn Qurrah's De imaginibus 4 • Abu Ma'
Cathedral of Toledo, now Madrid, Biblioteca nacional, 1000 9 and
shar's Introductorium maius (<<The Greater Introduction to Astrology»),
1 53 4. These manuscripts include the Toledan Tables, the. treatises
which he cites by book and chapter, and Sahl ibn Bishr's De interroga- 00
concerning the construction and the use of the astrolabe attnbuted to
tionibus which may be a translation ofJohn. Michael makes greater use
John of Seville, Haly's Electiones, the introductions of al-Qab1!i1 and
of a shorter introduction to astrology by al-Qabl~l (Alchabitius), also
al-Farghanl, the capitula Regis Almansoris and the lett~r of Argaphalau
translated by John. Likewise, for astronomy, he makes much use of the
to Alexander. These manuscript cannot be the dIrect sources for
shorter text of al-Farghanl, in Gerard of Cremona's translation, rather
Michael since they were copied later in the thirteenth century. How-
than the Almagest which he knows, but cites only occasionally. He
ever they were copied by, or for, Alvaro of Toledo (/loruit 12 67 until
uses three translations produced in Barcelona by Plato of Tivoli and , f . S I
after 1286), one of the leading Spanish intellectuals 0 the tIme. n
Abraham bar I:Iiyya - those of the capitula Regis Almansoris of al-Razl,
MS 100 53 Alvaro included Michael's translation of ,al-Bi~rujl, an~
Ptolemy's Centiloquium, and the Electiones of cAn al-cImranl (known in
added in his own hand his commentary to Averroes s De substantta
Latin as «Haly»). A more unusual work that he uses is the De indagatione
orbis which is one of the works of Averroes said to have been trans-
cordis - an astrological collection from several authorities, including
lated by Michael. It is this manuscript of al-Bi~rujl which ~ves the
Antiochus, Masha'allah, Dorotheus, cUmar ibn al-Farrukhan al-Taba-
most detailed information concerning the date and authorshIp of the
fi, an obscure «(Zimus», and «(Rasic» (Rashlq ibn cAbdallah al-Ha~ib) s.
text, including the information that Michael used the services of a
The De indagatione cordis - on the principles of making judg;ments,
Jew called «(Abuteus levita» .
.1.. M, f~l. I25ra: «.... sententiam 6 philosophorum, scilicet Rasic, Messehalla,
DnvI, Basglco, Hermetls et Eninii»; M, fol. I28ra: «Zael tolletanus ... Haly Sarra- I. The text is edited in S. M. Low Beer, «Hermann. ?f Carinth~a: the 'Liber
cenu~, Messehall~h Indus, ~.bumassar Grecus, Mahumar Syrus, Drivius Persye,
imbrium', the 'Fatidica', and the 'De indagatione cordls», unpublished Ph. D.
Emp10us Tabbanus, AlcablclUs et Ptholomeus rex Egipti ... »; see Thorndike, diss., City University, New York 1979, 261-3 80; s~e als? Burnett, «He~ann of
Carinthia's Attitude towards his Arabic Sources», m L Hom",e et SOtl IItlwers all
Michael Scot, 124-26.
2: ~he results of this investigation will be more fully set out elsewhere. For a Moyetl Age, ed. C. Wenin, Louvain-Ia-Neuve 1~86, I, 307- 22 (30~-IO).
2. Note that Plato ofTivoli, although based m Barcelona, dedIcated one work
prelmunary sOtldage of sources see Appendix infra.
3. Haskins, Studies, 28 4- 85. to John David of Toledo; see M.-T. d'Alvemy, «Avendaut~?, 29· .
3. Alberto Magno, SpeCl41llnl astrotlomiae, ed. P. ~a~belh. e~ al., P.lsa 1977·
4. This has been noted bv N. Daniel, The Arabs itl Mediaeval Europe, London 4. See the detailed descriptions of this manuscnpt m Mlllas Valllcrosa, Las tra-
1975, 28 5,
d'lCciotles, 166-202. . I
5. On Rashlq see F. Sezgin, Geschichte des arabischetl Schrifttllms, VII, Leiden 5. Millas Vallicrosa, Las trad'lCciotles, 34-35, and Alv~ro de Toledo, CometJtano a
1979, 197-98. I have not come across citations of this astrologer in Latin outside the
De Substantia Orbis de Averroes, ed. M. Alonso, Madnd 1941.
De itldagatione cordis and the Liber itltrod'lCtoriIlS.
10 9
108

VIII
. In the pe~son of Alvaro, then, one has someone who not only was
mterested, lIke Michael, in both astrology (he also translated the Libro
c~mplido of cAlI ibn abl'l-rijal from Castilian into Latin) and Aristote-
lIan cosmology - who, in fact, saw no real incongruity between these
subjects - but who also had access to Michael's works and the manu-
scrip~s that Michael used. He would have had that access in Toledo.
For Alvaro dedicated his commentary on the De substantia orbis to the
* a~chbishop of Toledo, Gonzalo Garda Gudiel. In 1273, whilst he was
stIll Dean of !oledo, Garda Gudiel drew up an inventory of his
books. T~ese ~ncluded «all the commentaries of Averroes, except a
~ew ... wntten m the hand of the translator» - i. e., in Garda's opinion,
m the hand of Michael Scot himself But that is not all. Another man-
uscript is described as «seven quaternia of the book De animalibus writ-
ten in the hand of the translator». As we have seen, Michael translated
Aristotle's De animalibus in Toledo, before 1220. Of the other sources
kn~wn to Mic~ael ~n Archbishop Garda's library one may mention
AVlcenna, Malmomdes, al-Farghanl and Theodosius. That Garda
picked up ~he~e texts i.n Spain is suggested by the fact that they include
a manuscnpt m Arabic. Moreover, one may contrast this list to an-
other one made by the same archbishop in 1280 after he had moved to
Italy - by which time he had been able to include some of the latest
translations of William of Moerbeke. By this date Garda had also ac-
quired Michael's translations of al-BitruJi and Averroes's commentary
on Aristotle's De caelo et mundo l •
To conclude this section: I have tried to show that Michael can be
seen as a continuer of the activity of the twelfth-century translators in
Toledo, and like them, may have had a particularly close association
with the Cathedral. He would have become aware of the Arabic
Aristotle, Avicenna, Averroes and al-BitruJl in Spain, and could have
mad~ consid~~able progress in translating works of these philosophers
an~ m compIlmg his own introduction to astrology by 1220, the date
aSSigned to his move to Italy. This is indicated by the facts that (a) his
sources were available in Toledan manuscripts, and (b) his translations
rema~ned in Toledan libraries. One cannot rule out the possibility that
he might have come across several of his astrological and philosophical
sources elsewhere. This is especially possible in the case of Averroes
, I. M. Alonso, «Bibliotecas m~d~evales. de 10s Arzobispos de Toledo», Razotl Y
Fe, 41 (1941), 295-309 (303-6); Mlllas VallIcrosa, us tradllCciotles, 17.
110
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
whose works (unlike those of Avicenna) were studied by Jewish phi-
losophers such as Jacob Anatolio who «was bound to» Michael Scot for
a while in Naples in the early 1230SI. But I do think there is a case for
re-considering Michael's debt to Toledo.
Michael Scot and Bologna
However, there is a problem. We do not know in exactly what
form Michael originally wrote what has become known as the Liber in-
troductorius. The earliest manuscripts to contain any part of the work
are of the second half of the thirteenth century2. Of these, the one
manuscript to contain the Liber quatuor distinctionum gives it in a short
form that Glenn Edwards has recently argued descends from an abbre-
viated form of the original work, of which the «long version» (itself in-
complete) is a more accurate representative 3• The earliest manuscript
of the latter was copied in Bologna in 13 20 ; it is on this manuscript -
Munich 10268 (our MS M) - that this study of the Liber quatuor distinc-
tionum is based. The Bologna connection is significant. Towards the
end of the thirteenth century there was in that city a magister Bartholo-
maeus Parmensis. In modern scholarship this man has been generally
dismissed as an ignorant bore «<un verboso imbecille»)4. It has long
been recognized, however, that he had something to do with the «long
version» of the Liber introductorius, because, in the Munich manuscript
I. See C. Sirat, «Les traducteurs juifs». Anatolio also translated al-Fargham and
Ptolemy's Almagest into Hebrew. ... .
2. These mss. are Paris, Bibliotheque nationale, nouvelles acqulSltlOnS latllles,
1401 ( 79 A. D.), which contains a short form of the Liber qllatllor di~tinctionllm ~ol­
lowed by the Uber particularis; Oxford,. Bodleia.n Lib~ry, Cano~. MIsc ..555, whIch
12
contains the Uber particlllaris and the Llber physIOtIOnu~; and possIb~y ~e~lllgrad, Ms.
Lat. F. v. IX, I, which contains the section of the Llber qllatllor d,stltlctlOnllm on De
signis et imagitlibllS ceU.
3. Edwards, «The Two Redactions». . . ., .
4. B. Nardi, quoting Pierre Duhem, III «Bartolomeo da Parma», III D,zlotJano
biografico de.~li Italiatli, VI, Rome 1964, 747-5 0 (750). T~e works .0fBart~010mew,
many of which cover similar subjects, need to ~e studIed ~nd dl£fer~ntIated from
each other see L. Thomdike, A History of MagIC atld Expenmmtal SCletlCe, II, New
York 19 3: 835-3 8, and Th. Charmasson, Recherches sllr line techniqlle divitlatoire: la
geomatlcie dam l'Occidetlt medieval, Geneve-Paris 1980, 14 1-55. The full~st account
2
remains that ofE. Narducci in his I primi dlle libri del tractatus sphaerae dl Bartolomeo
da Parma, Rome 18 8 5, 1-31·
III
VIII
1 MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
there is a marginal note against a horoscope which states that: «This
judgement has been compiled by magister Bartholomew of Parma and
added to this work as a mark of great respect (multo amore), and it was
for Lord Henricus de Flisco on the first Sunday of May, 128 7» I.
What has not been realized is the extent to which Bartholomew's
work corresponds to that of the long version of the Liber quatuor dis-
tinctionum, and to the Liber particularis. The only works by Bartholom-
ew edited up to now are the prologue to his Breviloquium and the first
two books of his Tractatus sphaerae 2 • Several passages from the latter
work correspond word for word with passages in the Liber particularis
and Liber quatuor distinctionum. Bartholomew tells the same anecdote
about the rusticus who stages a coup d'hat by posing impossible questions
to a king3 and reproduces the same table of climates 4. Moreover, he
reports a conversation between Michael Scot and Frederick 11 in which
Michael demonstrates infinity by drawing a circle on the ground whilst
Frederick is not looking, and then asks him where he started the
circles. The prologue to the Breviloquium is very close to Michael's
description of the division of the science of the stars 6• In other works
of Bartholomew which survive only in manuscript the similarities are
even more striking. Two manuscripts from Vienna and one from
the Hospital of Kues contain a corpus of Bartholomew's astrological
writings 7 . Of these, the Liber de occultatis is a slightly abbreviated version
of the letter from Argaphalau to Alexander in the Liber quatuor distinc-
I. M, fol. 125va: «Iudicium compilatum a magistro Bartholomeo Parmensi et
huic operi additum grandi amore, et fuit pro domino Henrico de Flisco mc-
clxxxvii de menseMaii prima domini ca introitus sui», see Edwards, «The'Two
Redactions», 337.
2. Ed. Narducci, I primi dlle libn·.
3· Compare Tractatlls sphaerae, ed. Narducci, 123-24 with ms. M, fols 122Vb-
123ra.
4· Compare Tractatlls sphaerae, 121-22 and ms. M, fol. 12Irb. Bartholomew and
ms. M agree against al-Farghani, the ultimate source.
s· Tractatlls sphaerae, 12 4- 2 5.
6. N?te in particular, ~he definitions of astronomia and astrologia and the division
of the sCience of the stars mto spen'ca, gnomonica, horoscopica,fablllosa, sllperstitiosa and
ymagitlaria, ed. Nard~cci, I primi dlle libri, 32-35, ms. M, fols 16va- 17ra .
7· Mss. Vienna, Ost~rre~c~ische Nationalbibliothek, codd. 312 4 and 543 8; ms.
Bernkastel-Kues, Hospltalblbhothek, 209, described in Verzeichnis der Handschrift-
etl-Sa~~mlllng . des Hospitals ZII Clles, ed. J. Marx, Trier 190 5, 19 8- 201, and A.
Krchnak, «Die Herkunft der astronomischen Handschriften und Instrumente des
Nikolaus von Kues», Mitteillmgetl IInd Forschllngsbeitriige des Ctlsatllls-Gesellschajt, 3
(19 63), 109-81 (145-48).
I12
VIII
tionum I , whilst the Significationes naturales planetarum extracte de libro con-
siliorum magistri Bartholomei Parmensis has an obvious relationship to one
set of descriptions of the characteristics of the planets in the Liber quat-
uor distinctionum 2. A work on astrological judgements, which, though
not attributed, is most probably the Consilia of Bartholomew itself,
turns out to be a highly elaborated version of the judgements attributed
variously to Aristotle and Ptolemy 3, with extra chapters which corre-
spond to some of the judgements in the last part of distinctio 3 of the
Liber quatuor distinctionum4, whilst Bartholomew's De electionibus corre-
sponds to other chapters in the same section of Michael' s work s.
But what is striking is the nature of the correspondences. They are
most common in the rubrics and in the opening sentence or sentences
of a passage. For example notula doctrinalis is commonly found as a sec-
tion heading in Liber quatuor distinctionum and Bartholomew's Tractatus
sphaerae. The latter work has a long title to the section on solar and lu-
nar haloes (called alibett) which is almost the same as the title found in
Liber particularis; in this case, however, the contents of the respective
chapters do not correspond closely 6 • In describing the characteristics
of the planets both Bartholomew and Michael begin with the same,
rather unusual, phrase: [Such and such a planet] «is a planet and not a
star» 7. And in another description of the planets they both begin:
I. Compare Liber de occllltatis in ms. Vienna, 3124, fo1. 198v-199v with ms. M,
fols 127rb-127vb (ultimately from Liber Alchandrei, c. 6S). The Liber de oCCIIltatis was
«compiled» (compilal/ls) by Bartholomew ofParma in 1280. An excerpt further on
in the same ms. (fol. 209v) is headed «Argafilam astronomus» and corresponds to M
fo1. 126rb (ultimately from Liber Alchandrei, c. 68).
2. Compare ms. Vienna, 3124, fols 202r-203v with ~:s. M:, fols. 126va-127rb ...
3. I refer to the close relationship between the .IlId/.Cla Anstotelts. and th~ IlIdlCla
Ptolemei in «The Contents and Affiliation of the SCientific Manuscnpts Wntten at,
or Brought to, Chartres in the Time ofJohn of Salisbury~, in The World ofJohn .of
Salisbllry, ed. M. Wilks, Oxford 1984, 127-60 (134); editIOns of both texts are m
preparation.
4. Compare ms. Kues 209, fol. 12V (c. 31) with ms. M, fo1. 14Ira, and ms. Kues
209, fo1. 2 Ir (chap. 78) with ms. M. fol. 13 8va. .
5. Compare Bartholomew's De electiotlib'IS, ms. Kues 209, fols 46v-54v and VI-
enna, 5438, fols II7va-128rb, with ms. M, fols 135v-I37r. . .
6. Compare Tractatlls sphaerae, ed. Narducci, I frin~i dlle ~ibri, 158: «~e alhbetl,
id est circulo qui quandoque generatllr circa Solem m die et cIrca Lunam m nocte»,
and Liber partiClliaris, c. 61, Oxford, BodIeian, Canon. Misc. 555, ~01. ~ova: ~De
alibeti, id est de circulo qui quandoque apparet in aere circa Solem m die et circa
Lunam in nocte" (the differences are marked by italics).
7. «... est planeta et non stella ... ,.; ms. Vienna, 3124, fol. 204r-205r (t?e ope?-.
ings of the paragraphs of the text entitled Sigtlificationes platletartlm cllm jilentlt donmu
113

VIII
[Such and such a planet] «is so-called by the Ancients because it is a
God» I. The chapters on elections begin in both authors: «(When you
are asked on what day or on what hour of the day it is good to ... »2.
Also, the sources cited by Bartholomew are very similar to those in
the Liber introductorius, including Haly's Electiones, al-Bitriijl, and al-
Qablfi. One could go on.
All these factors strongly suggest to me that Bartholomew did not
merely copy from Michael Scot extensively, but that he was also re-
sponsible for editing the text of the Liber quatuor distinctionum and Liber
particularis at some stage 3. Whether one of the shorter versions of the
Liber quatuor distinctionum would then be closer to the text before
Bartholomew interfered with it remains to be investigated 4. All I can
say at present is that the language and style of the version of the Liber
quatuor distinctionum in the Munich manuscript is strikingly similar to
that we find in the works of Bartholomew. Moreover, the fact that
we have no manuscripts of the Liber introductorius which predate
Bartholomew's floruit makes it all the more necessary to excercise
caution in differentiating the contributions of Michael from those of
Bartholomew.
The entanglement of the two authors is evident, too, from the
manuscripts of Bartholomew's works, which include excerpts from
Michael's work. One of these MSS, Vienna 3124, gives several passages
on the science of the stars of which the first is attributed to «(Michael
Scot the astrologer» 5, but at least one passage is from Bartholomew's
anni mlmdi it' qllalibet revolllcione et hoc seCllndllm Bartholomellm de Parma) and ms. M,
fo1s I29rb-I30ra .
. I: «.... dicitur ab antiquis quod est deus ... »; the openings of the paragraphs of
Stgnificattones naturales planetamm extracte de libro consilionml (see p. I 13, n. 2) and ms.
M, fo1s I26Va-I27rb.
2. «Cum interrogatus fueris quo die vel qua hora diei bonum sit ... »; ms. Kues
209, fols 50r-pr and ms. Vienna 5438, fols 123ra-b; ms. M, fols 135va-I36vb,
I38rb, 139ra, 139Vb and I40va.
3. T.his possibilty is briefly explored, and rejected by in Edwards, «The Two
RedactlOns», 338. One may note that the rubric notula doctrinalis does not occur in
other works of Michael Scot, such as the commentary on the Sphere of Sacrobosco
(ed. L. Thorndike, The Sphere of Sacrobosco and Its Commentators, Chicago 1949,
247-342) or the Alchemy (ed. H. S. Thomson, Osiris, 5 [1948], 529-59).
. 4. A .cursory examination of Paris, Bibliotheque nationale, nouvelles acquisi-
tions latmes, 1401 confirms that while tlotllla doctrinalis is a common chapter head-
ing (see fols 98r, 104r, etc.), the other parallels adduced here do not occur.
~. Fol.. 2?6r: «Capitulum de hiis que generaliter significantur in partibus duo-
declm cell Slve donubus secundum magistrum Micalem Scotum astrologunu.
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
Breviloquium I. This is also one of the two manuscripts of Salio's trans-
lation of AbIT Bakr's De nativitatibus, and is the only manuscript to give
two translations of Salio together. As we have seen, Salio was in To-
ledo at the same time as Michael and it is probably Abii Bakr's name
that appears in the list of authorities at the beginning of the Liber
particularis 2 • Thus MS Vienna, 3124 brings together works of Michael,
Salio, and Bartholomew. The Kues manuscript of Bartholomew's
works has also been said to include a passage from the Liber quatuor
distinctionum3. In fact, the passage in question is attributed to «(William
Scot»4. An abbreviation for «(Michaeh> could easily be expanded as
«(William», and the peculiar forms of the constellation names (particu-
larly demon meridianus) are those of Michael. However, the passage
cannot be found in Michael's works; nor is it likely that it should,
because it mentions King Manfred, who ascended the throne of Sicily
some twenty years after Michael's death. However, this text nicely
leads us on (or rather, back) to Sicily.
Michael Scot and Palermo
The chapter attributed to «William Scot» is on past and future con-
junctions, and discusses in chronological order, major historical events
associated with «(great conjunctions» (i. e., conjunctions of Saturn and
Jupiter), starting from the Flood and going through the first circumci-
sion, the reception of the Mosaic law, the first emperor, Christ's na-
tivity, etc. The two most recent great conjunctions are described as
follows: «(Another [conjunction] was in the time of King Manfred, un-
der the [celestial] image of Agitator, when flagellation broke out in
Lombardy. Another was under the image of the «(Southern Demon»
I. Fo1. 2IOV-21 Ir (<<Gloriosus et eternus Deus ... ») = Breviloqllillm, ms. Oxford,
Hertford College, 4, fol. 74v. On fo1. 205V another unattributed excerpt «On free
will» includes an «exemplum de Theofilo in provincia Scicilie». .
2. Ms. Wellcome 507, fol. 5r: «habubachet»; Oxford, Canon. MISC. 555, fo1. If:
«habubacher». D. Jacquart has suggested that Abii Bakr's De tlativit~tibll~ i~ a source
for the Liber physiotlomie which immediately follows the Liber partlClllans In several
mss. (see her article in this volume). .
3. See Edwards, «The Liber ItltrodllctorillS» , XIX, and Id., «The Two RedactlOnS»,
33 2 , n.I2. . . . .., 1
4. Ms. Kues 209, fol. 76v: «Dicta Wllhelml SOCtl de comunctlOmbus p aneta-
rum transitis et futuris».
VIII

(demon meridian us) because of which those acts of devotion and flagel-
,
i
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
VIII

connection with that city I. These are avenues to explore. However,

lations remained, the empire was empty (?), and the people experi-
up to now we have little evidence of what he did in these cities. With
enced exaltation in many places» I. After this the predictions refer to
the exception of his translation of Averroes's commentary on the De
the future. This short text, then, was evidently written soon after the
caelo et mundo dedicated to Etienne of Provins, none of Michael's
death of Manfred (1266) and refers to the spread of the flagellant
works is dedicated to anyone other than the Emperor Frederick H,
movement which originated in Umbria in 1259-602. Another untitled
and it is only as the astrologer of Frederick H that we know much
text in the same manuscript notes that «the Sun signifies emperors and
about Michael's Italian sojourn.
kings '" and all places in which royal majesty remains in control». Four
When entering Frederick's service, Michael had already translated a
examples of such places are given: Rome, Constantinople, Tripoli and
considerable number of works, and may indeed have added a dedica-
- regnum Siciliae 3• However, when the same text comes to describe the
tion (such as that of Avicenna's De animalibus) to a text which he had
significance of Mercury, we read that the planet signifies all the arts
translated several years earlier. What appears to be the particular fruit
(they are listed) and all places of study and students, like schools and
of his years with Frederick is his original work. The previous section
the rooms (camere) of scholars, and the places named are «Alexandria
of this article has pointed out the difficulties in knowing exactly what
on the [Mediterranean] sea, Venice, and the whole of Lombardy» 4. So
this original work consisted of and great caution must be exercised in
by the time that this text was written scholarship was firmly located in
accrediting anything in the Liber introductorius to Michael. However,
the North of Italy. Had Michael Scot contributed to this scholarship?
taken on its own merits, the Liber introductorius displays great inven-
Michael either spent periods of time or made visits to Rome
tiveness, both in its use of sources and in the addition of original ma-
and several cities of North Italy. He was in Rome with Archbishop
terial. To this we may now turn.
Rodrigo of Toledo for at least part of the Fourth Lateran Council
The Liber introductorius is probably the first work to fully adapt the
which lasted from 1215 to 12 I 7 5, and letters were written on his behalf
Arabic-Latin translations of astrological works (which belong to an
by Pope Honorius 11 in 1224 and 1225, and by Pope Gregory IX in
122 7. In 1220 and perhaps again in 1231 he was in Bologna 6• The fact Islamic milieu) to the Christian and Imperial society for which it was
compiled. Thus when drawing from Sahl ibn Bishr's work on interro-
that he uses the Pisan calendar and, in 1228, received the dedication
gations and cAli al-cImranl's on elections, the chapters on when to cir-
of the revised edition of Leonard of Pisa's Liber de abaco suggests a
cumcise a child are omitted, but chapters are added on whether some-
one who has died will go (or rather should go) to paradise or hell,
I .. Ms. Kues 20 9, fo1. 76v: «Alia fuit tempore Manfredi regis sub ymagine Agi-
tatons ~uand~ ~act~ est verberatio horninum in Lombardia. Alia fuit sub ymagine whether someone has been baptised recently, whether a cleric should
~em~rus mend~aru propter quam remanserunt ille devociones et percussiones et receive a benefice, and whether a man in religious orders should
Impenum vacav!t at populus per loca exaltationem percepit». change his order2. There are two chapters on acquiring a bishopric,
2. I am grateful to Nigel Palmer for pointing out that these sentences refer to
the flagellant mo~e.me?t; see J. Henderson, «The FlagelIant Movement and Flag- and finally, there is a chapter which begins: «When you are asked by
ellant Confraterrutles In Central Italy, 1260-1400», Studies it, Church History, 15 someone ... whether the hoped-for emperor, pope', king, legate, patri-
(197 8), 147-60.
arch, bishop or abbot, who has been elected or called to that honour,
3· Ms. Kues 209, fo1. 79 r: «Solis sunt imperatores, reges '" de locis significat should arrive in this year or not ... »3.
Ro.mam, Constanti~opolin~, .Tripolim, regnum Sicilie, omnem locum in quo regia
malestas perseverat In dOmInlO». The narrative is lightened by the introduction of anecdotes. One of
.4·!'As. Kues ~09, fo1. 79~: «Mercurius est significator omnium disciplinarum these concerns a sceptic who tested the astrologers judgement by
cUlUslibet facultatts '" De 10ClS mundi tenet Alexandriam in mari, Venetiam, totam
Lumbardiam et multa loca vicina, omnia loca studencium ut sunt scole, camere
scolarum ... ». I. Haskins, Studies, 275 and 290.

5· Edwards, «The Two Redactions», 329, n. I, points out that there is a refer-
ence to the Fourth Lateran Council in the Liber itlfroductorius.
6. Williams, «The Early Circulation», 12 7-44.
116
2. Ms. M, fols 13ub, 132va, 133vb and 135rb respectively.
3. Ms. M, fo1. 144r: «Cum interrogatus fueris ab ~liquo ... que~ente utrum sper-
atus imperator vel papa vel rex aut legatus vel patn~rcha vel eplscopUS vel abbas
electus in honorem vel vocatus hoc anno debeat verure vel non ... ».
il7
 1
VIII
\
cutting his toe-nails when the Ascendent was Gemini, and the Moon
was in Pisces, waning, joined with the Head of the Dragon and
combust with .the Sun. As he was cutting the toes of his larger, right,
foot: and pounng scorn on the astrologer's art, someone struck his pig,
makmg the poor animal rush into the house and overturn the bench
on which he was sitting. He promptly dropped his nail-cutter onto his
foot and lost three toes I.
One indication of a Sicilian or Southern Italian context for the
composition of the Liber introductorius is its references to Maimonides.
In Hebrew texts of the period there are several accounts of debates on
subjects in the Guide to the Perplexed involving both Michael Scot and
Frederick lP, and one manuscript attributes both the standard Medi-
eval Latin translation of the Guide and a discussion of the parables, pre-
cepts and commandments in the Guide to Michael3. What has not
bee.n po~nted out, so far as I know, is a quotation apparently from
Malmomdes {«Raby Moyses») in the Liber quatuor distinctionum: «Raby
Moyses, the great philosopher, said in a certain book that each of the
spheres of the seven planets has a thickness of 500 years, that is, as
much distance as a man can walk on a straight path which is clean
(mundam) and suitable for walking in 500 years, walking every day in
a normal manner, which is taken to amount to 4,74 0 miliani [i. e,
4,74 0 ,000 miles?]. Likewise he says in another place that thou and bohu
are signifying the emptiness of the celestial space'»4. However, Bar-
t~olomew of Parma also refers to Maimonides in a very similar way in
hIS Breviloquium; this again suggests contamination between the texts
of the two Latin authors s.
I. Ms. M, fol. 13 6r .
2. See Sirat, «Les traducteursjuifs», 169-91.
3· Haskins, Studies, 282.
4· M~. ,M, fo1. 54 r: «Dixit Raby Moyses magnus ph(ilosophu)s in quodam libro
~uod quthbet pla~etarum 7 habet inspissitudinem viam quingentorum annorum _
Id est tantum spatIum quantum posset aliquis homo ire per viam planam et mun-
dan~ atque congruam ad eundum in quingentis annis, non cessando ire omni die
debIto modo natur~ q~od sU1:nitur numero 4740 millianorum. Item dicit alibi quod
thou e~ bohu sunt sIg.mficantIa vacuitates spatii celestis». I have not identified these
q~otatIons, but c.. Suat has suggested to me that these doctrines of Maimonides
mIght have been pIcked up orally.
5· Ms. Ox~ord, Hertford Colleg~ 4, fol. 80v: «Dixit Rabimoses philosophus in
quo?am s~o hbro quod corpus tOtlUS terre est rotundum ut pila ludi et equalis
undIque hcet montes et alpes sint in ea, tamen sunt tot valles et caverne quod ad
118
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
Another work that Michael may have come across in Frederick's
court is the Liber novem iudicum. It is clear that he has not read the work
because the list of nine astrologers which are said to be the «the nine
judges» in the Liber quatuor distinctionum does not correspond with the
list in the real «Book of the Nine Judges» I. Nor have I found any cor-
respondences in the third distinctio with the Liber novem iudicum 1 • The
latter, which was probably put together by Hermann of Carinthia and
Hugo of Santalla, or compiled from their translations, in the mid- to
late twelfth century, appears to have reached Frederick's court during
the time of Michael Scot, or a little later, for the explicit of one man-
uscript reads: «Here ends the book of the Nine Judges, which the Sul-
tan ofBabilon sent to the Emperor Frederick at the same time that the
Great Khalif sent magister Theodore to the same Emperor Frederick» 3.
Whilst this explicit cannot be taken at face value, it is reasonable to as-
sume that this Theodore is Theodore of Antioch, who succeeded
Michael Scot as court astrologer, and continued Michael's work on
translating Averroes 4.
It may be significant that hardly any direct use of any of Michael
Scot's translations is evident in the Liber introductorius. Averroes and
Avicenna appear as names in the Liber quatuor distinctionum 5, but no
equalitatem sui compensantur, et est eius grossicies in rotundum, id est in circuitu
sui, 6500 miliarii».
I. Ms. M, fol. I28ra: «Nota quod Zael Tolletanus ... iudicia multa pro question-
ibus hominum compillavit. Similiter et Haly Sarracenus, Messehallah Indus, Albu-
massar Crecus, Mahumar Syrus, Druvius Persye (i. e., Dorotheus), Empm 9 Tab-
barius, Alcabicius et Ptholomeus Rex Egipti et cetera, quorum 9 phyloso~hy
librum composuerunt multe veritatis qui dicitur 'liber 9 iudicum'»: The «mne
judges» of the Liber tlovem iudiCllm, which exists in sev~ral manuscrIpts and ~as
printed by Peter Liechtenstein in Venice in 1509, are Anstotl~, Albumasar, Al~n­
dius, Aomar (CUmar ibn al-Farrukhan al-Tabafi) , Abenalhalat (Ibn al-Khayya~),
Dorotheus, Jergis, Mesehalla and Zahel (Sahl ibn Bishr). . . .
2. Note that in the list of authorities in ms. M, fo1. Ivb, the lzber tlovem ItldlCllt1l
does not occur, but amongst the books which adorn the margins of the same folio,
«Novem iudicum» appears between «Ally albeiaiel» (probably cAll ibn abll-rijal)
and «Claviculla Salomonis». Other works which post-date Michael Scot, such as
«Cuido Bonati» also appear amongst these books in the margin. . . .
3. Ms. London, British Library, Royal 12 C.VIII, fo1. 78r: «LIber novem lUdI-
cum, quem missit soldanus Babilonie ~mpe~tori Fed~rico te~npore quo et magnus
Chalif misit magistrum Theodorum eIdem ImperatorI Fede~co».
4. Haskins, Studies, 246-48. Theodore added a translatIon of the preface to
Michael's translation of Averroes's commentary on Aristotle's Physics; see Thorn-
dike, Michael Scot, 25. . . .
5. Ms. M, fo1. 19r: «averoys» is listed as an authority of astrology; ,b,d., «aVlcenna»
119
VIII
passages, so far as I have noticed, are cited from them. Michael seems
to have been particularly interested in Arabic texts on zoology. When
one turns to the chapter on animals in the Liber particularis one expects,
from the first sentence, that he is going to classify animals by their
common differences, as do Aristotle and Avicenna. Michael writes:
«An animal, in the strict sense, is «every living being», such as any
beast. One must know that there is a very great difference between an-
imals in shape, in purpose (officium) and in faculty (virtus)) I. However,
instead of describing those animals which have two legs and those
which have four, or those which have fins and those which have lungs
as in the Historia animalium or Abbreviatio Avicenne de animalibus (both
of which Michael had translated), the Liber particularis immediately
places before our eyes «an animal which gives off an aroma of such
sweetness that it attracts many things to itself, since the aroma is dif-
fused over a distance of four miles»2. This is followed by a list of fan-
tastic animals, which, if it bears any relation to Aristotelian zoology,
can only be regarded as a parody of it. The next chapter of the Liber
particularis is no more serious. It purports to describe «the parts of ani-
mals», which have many functions, «some to help, and some to harm».
But in no way do we get a summary of Aristotle's De partibus animalium
(translated by Michael). Instead we are told that «the drop which is in
the middle of each testicle of the beaver kills the man who imbibes it.
The brain of a cat and an ass infatuates the man who eats them», etc ... 3.
One might say that Frederick 11 refused to follow Aristotle blindly, in
stating in his De venatione cum avibus that he preferred to rely on expe-
rience and his own investigations, since Aristotle had had little or no
practice in falconry4. Nevertheless it is strange that Aristotle should be
entirely absent in the Liber particularis. Richard Lemay has pointed out
is given as an example of the author's name being used on its own (without a title)
to refer to a book.
I. .Lib~r .partiCIIlaris, ~. 11 5, ms. Oxford, Canon. Mise. 555, fol. 53 2a: «Animal
propne dlCltur omne Vlvens ut omnis bestia. Seiendo quod maxima differentia est
In genere animalium, figura, officio et virtute».
2. Ibi~.: «Animal est euius odor est tante suavitatis quod ex eo attrahit multa ad
se eum dlffundatur proeul per 4 miliaria».
3. Liber p~rticl~laris, e. 11 6, De membris animalillm, Canon. Mise. 555, fol. 52 v:
«Membra arumahum sunt multum virtuosa in multis, quorum quedam iuvant alteri
et quedam noeent. Verbi gratia gutta que est in medio euiuslibet testieuli eastorei
occidit edentem. Cerebrum gatte et asini edentem infatuat ... ».
4. See Haskins, Stlldies, 312.
120
, MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
VIII
a conspicuous use, instead, of the Liber Nemroth which probably orig-
inated in the late tenth century; this needs further investigation I •
More questions have been raised in this paper than have been an-
swered. Scholarly work on Michael Scot is in its infancy. Proper edi-
tions of the various parts of the Liber introductorius should prepare the
way for detailed investigation of the work's sources, and comparisons
with closely related works such as those of Bartholomew of Parma.
These studies in turn should shed light not only on the multi-faceted
character of Frederick I1's astrologer, but also on the process of Lati-
nization and Christianization of Arabic astrology, on the society of
Frederick's court, on the history oflearning in Spain and in Italy in the
thirteenth century - and much more.
APPENDIX
Some of the Astronomical and Astrological Sources
of the third distinctio of the first book of tile Liber introductorius
in MS Munich, Bayerische Staatsbibliothek, elm. 10268 (M)
Abbreviations
L MS Oxford, Laud. Mise. 594
Ma MS Madrid, Biblioteca nacional, 10009
P MS Paris, Bibliotheque nationale, lat. 16204
Pb MS Paris, Bibliotheque nationale, lat. 17 868
1. M, fol. I2ora: Dicit enim, scilicet primo verbo, Ptholomeus in Centilo.-
quio: Astrologus numquam debet rem dicere alteri sp~cialiter :el~t est affir-
mando sic et sic erit ... Item dicit Ptholomeus: Sunt erum multI qUI dant vera
iudicia fortunate quorum nec significationes ... ex parte infundentis dona gra-
tiarum et virtutum.
Pseudo-Ptolemy, Centiloquium, verba 1 and 2, with 'HaIy's' commentary,
in Plato ofTivoli's translation (= Madrid, Biblioteca nacional, 100 15, fo1. 20r).
2. M, fol. I2ovb: Astrolabium est necessarium instrumentum ... vel dicitur
astrolabium alio nomine, videlicet astrolapsus, id est astrorum lapsus quia per
lapsum.
I. R. Lemay, «De la seolastique a l'histoire pa~ le tru~he~ent de la philo~ogie:
itineraire d'un medieviste entre Europe et Islam», In La dftjilSlone de/la SCletlze Islam-
iche tIel Medio Evo ellropeo, Convegno internazionale dall'Aeeademia nazionale dei
Lineei, Rome 1987, 399-535 (51 8- 26).
121
VIII
John of Seville (?), Construction of the Astrolabe, Ma, fol. 18rb-va, ed.
Millas Vallicrosa, Las traducciones (p. 106, n. 3 supra), 316, lines 14- 24.
3· M, fols I2ovb- I2 Ira: Almucantarath sunt quedam linee ... assafac, id est
lumen vel rubor quod vel qui videtur in aere ante ortum Solis et post
occasum.
This is largely the same asJohn of Seville (?), Use of the Astrolabe, Madrid,
Biblioteca nacional, 10053, fol. 112va, ed. Millas Vallicrosa, ibid., 263.16-19,
264.4-8,264.22-26,267.12-14.
4· M, fol. I2 Ira-b: Stellarum fixarum nomina sunt multa secundum quod
ipse stelle sunt multe ... denebalgedi sub hac figura.
The star-list from the same source as Ma, fol. 23 r.
5· M, fol. I2Irb: Climatum vero latitudo secundum Alfagranum diligenter
approbatum. Hec est tabula per gradus, minuta et horas.
The reference is to al-Farghanl's climate-hours, from Elementa astronomica,
ed. R. Campani, 11 ILibro dell'aggregazione delle stelle', Citta di Castello 19 10 ,
c.8.
6. M, fol. 122va: Verum est quod secundum modum illorum qui Kalen-
darium composuerunt Sol 15 Kalendarum sequentis mensis in quo hec noticia
scribitur intrat signum ...
Helperic, Liber de computo, Patrologia latina, 137, cols 17-4 8, see col. 23:
Igitur zodiaco ipsi eo loco tribuentes initium, quo Solem mox factum constat
esse locatum, id est, ab eo signo quod xv. kal. Aprilis intrare dicitur.
7· M, fol. I23va: Nota quod sex sunt illa que possunt leviter decipere
quemlibet astrologum, id est, ipsum deducere in errorem. Primo si instru-
mentum eius sit falsum ... sexto si fortune fuerint equales et mali sint in pro-
batione.
Masha'allah, De interpretationibus, printed in Messehalae libri tres, ed. I.
Heller, Niimberg 1549, fol. 44V: Scito quod astrologus potest errare quatuor
modis '" Michael gives Masha'allah's four ways, and inserts two further ones.
8. M, fol. 124ra: Dicit enim Almansor in suo libello quod status omnium
bonorum temporalitate mutatur a malo in bonum et a bono in malum cum
signa mutantur a regimine universali per menses et figuras planetarum in
signis ...
Capitula Regis Almansoris, Ma, fol. 108rb: Status omnium bonorum muta-
tur de bono ad malum vel de malo ad bonum comutantur signa et figure plan-
etarum que significant eos de ascensione in descensione et econverso.
9· M, fol. I24rb: Al(mansor) quod numquam erit pauper qui habuerit
Iovem directum et in bono aspectu pro significatore sue nativitatis.
Capitula Regis Almansoris, Ma, fol. 106rb: Numquam erit pauper et inops
cuius nativitatis dominus fuerit Iuppiter.
10. M, fols I24rb-I24va: Et cum interrogatus fueris de aliqua re que debeat
proficere alteri vel meliorari ... putaverit se adeptam fuisse.
122
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
SaW in Bishr, De interrogationibus, Ma, 125va, P, 446a-b: Si autem interro-
gatus fueris de aliqua re que debet proficere et meliorari ... putaverit se earn
adeptam fuisse.
11. M, fol. 124va: Anahubaraath est quedam scientia novenarie ... gradus
Aquarii Iovis. . .. . . . ..
AI-Qabl~l (Alcabitius), Isagoge, book IV (Alcabrtll ad .magtstenum tudtCtorum
astrorum Isagoge, commentario loannis Saxonii delcarata, Pans 1521, fol. 24V).
12. M, fol. I24va (continuation of no. 11): Et de hiis latius dictu~ est in 1.8
differentia tractatus libri Albumassar qui dicit sic: unumquodque signum di-
vidatur in 12 partes propter iuditium experimendum quarum queque est 2
graduum et dimidium alterius etc.
Michael is referring to Abii Macshar, Maius introductorium, book 5, chapter
18, but confuses the terms 'novene' (described in book 5, c. 17) and 'duode-
narii' (described in c. 18).
13. M, fol. 124va: Adorungen est divisio ascendentis per tres partes ...
AI-Qabl~l, Isagoge, book IV (fol. 25r).
14. M, fol. I24vb: Dominus orbis, id est significationis domini orbis ... et
sic per ordinem.
AI-Qabl~l, Isagoge, book IV (fol. 25r-v).
15. M, fol. I24vb: Alburze dicitur dominatio septem planetarum in horis ...
Al-Qabl~l, Isagoge, book IV (fol. 26r-v), on 'albuzic'.
16. M, fol. I25ra: Albuiz est inventio partis fortune ... aliis partibus rerum,
ut patet in ultima parte Alcapicii.
AI-Qabl~l, Isagoge, book IV, fol. 26v (albuzic). Micha~l has e.rroneously
made two astronomical terms out of the single term albuzJc descnbed by al-
Qabl~l.
17. M, fol. I25ra: Iudiciorum astronomie omni~o b.ipertita es~ se?tenti~,
una per nativitatem nati, et altera per consilium saplentls et. persplcac~s cogt-
tationis hominis, quod totum sciri potest natura et arte vel vlgo~e alten~s ta~­
turn. Natura ut divina gratia illustrante, qua dona gratiarum dlversa dlve.~ls
noscuntur exiberi et in eis probabiliter constituta, unde quidam sunt grOSSl III
scientia litterarum, alii in alia doctrina preter scripturam et sic de reliq~is.
Omnes quidem phyQosophi) sunt in concordia quod signa celi sunt .12. et CI~­
gunt ipsum in circuitu velut cinctura hominem vestitum. De qUlbus Anes
dicitur esse primum etc. . .,
This is the same material as that found in Hermann of Cannthla s De
indagatione cordis, ed. S. M. Low Beer (n. 3~ ab~ve), 27~-1, and. M~ L, fol.
145va: Astronomie iudiciorum omnimodo blpartlta ~s~. via: u~a s~qu~dem e~t
questionum, genezie sive etiam annalium, altera consilll et cogttattorus hOffil-
num ... cuius generis primum Arietis natura est ...
18. M, fol. I2 5rb: Rasich dicit: Omnis planeta in dignitate .fortis. e~t, in
contrario debilis, directus audax, retrogradus quasi nullius valons, qUl Slmu-
12 3
VIII
latur ho.mini quiescenti. Unde sublimi augmentum boni influunt et addunt.
Depressl vero, subtrahunt ut fons desiccatus.
See Hermann, De indagatione cordis, ed. Low-Beer, 274, MS L, fol. 146ra.
19· M, fo~. ~25rb: Druviu~ dicit: tollantur gradus ab ascendente usque ad
gradum do~ml hore re~n~n~ls et co~e~tis, dividantur per 9 tantum vel per 12
vel per duahtatem. e~ dimtdiam umcUlque signorum dandam et ubi finitur
compo.tus horum, lbl est pars fortune et dux locatus in loco sententiam pre-
ostendit.
See Hermann, De indagatione cordis, ed. Low-Beer, 275- 6, MS L, fol. 146rb.
. 2~. M, ~o~. 125r~: ~asgund dicit et Mauhmar: in eligendo partes testifica-
~loms dom.ml .questloms. I 8 fo~i~u.din.es dentur domino ascendentis, 13 dom-
100 exaltatloms, ~~ domtno tnplicltatls, 6 domino termini et 3 domino faciei.
~ac tam~n conditIone quod planete sint in cardinibus si has debeant fortitu-
~mes optm.ere. Sed si extra sint ab eis diminuuntur et datur sibi bisse tantum
Id est duahtas premi.sse partis tertia remota. Messehallah dicit quod plane~
tarum qua?tum plus m ascendente optinuerit dignitates fortitudinum ille cen-
s~tur d.ommus ~uestio~s et. h~b~re ducatum rei, cum in domo sint 5 fortitu-
dmes, m exaltatlOne 4, m tnphcltate 3, in termino 2 et dimidia in facie I
~ee He~ann, De in1agatio.ne ~ordi~, ed. Low-Beer, 27 6, MS'L, fol. 146rb:
CUlUS~O~l vero est par~lU~ distnbutlO: Orientis dominus 18 vendicat, domi-
nus ~rmclpatu~ ~3 dum m ~pso ~adu sit ~~entalis, dominus trigoni primus 10,
dO~l~US terml~l 6. Hac discretlOne adhiblta id observandum est quoniam in
cardlmbus partlUm numerus integer est, extra cardines, diminutus ad hunc
mod~m. r:>um vero in 3° fuerint aut 9° aut 5° aut in xi° bissem retinent. In
averSlS 4,. m 2°, 6°, 8°, .12°, trientem. Hiis diligenter prospectis, cui plures
concessennt pa~es mehus.que constiterit, is nimirum obtinebit ducatum.
Messeh~la ~ero mter o~.~.a ~ro~idendum est, inquid, quis in oriente piu res
d
habe~t digmta~es. I?onllctlll sIqUldem sunt 5, principatus 4, termini 3, primi
ommorum tngom 2, decani unus.
2 I. M, ~ol. 12 5va-b: Signa celi mobilia sunt 4, scilicet Aries ... Signa fixa
sunt 4 ... SIgna communia sunt 4 ...
These opening sentences of, and the occasional sentence within each
paragraph are from Al-Qabl~l, Isagoge, book I (fols 3r- v). ' ducciones, 332), and P, 509b-poa: Dixit Alchaiat astrologus: considerandus est
22. M, fol. 126~a: A~trologus postquam famosus est in populo civili fre-
que~ter solet questlOnan de multis causis et diversis rebus ' .. si Pisces bene sibi
contmget.
C~ose to ~iber Alc~andrei, c. 68, MS Pb, fol. I IV: Si Aries se elevat, quod
quent
. proficlet sed
. prospera VI magna et labore suorum ... Si Pisces , l'n 0 mm. quo d
b'Itur.
quesIent
23· M, ~ol. 126va: Plato qui 23 annis docuit grandem Aristotilem prius-
quam monretur compillavit hoc opus ... ad manus meas. ' Haly, Electiones, 11, chap. 10, diff. I, Ma, fol. 37Va, P, 532b.
Clo.se to. Liber ~lchandre~, c. ~4~ MS Pb, fol. I I r: Dicis te in opere quodam
Pla~om~ P~I~osophI, quod mscnbItur summa operis universi ... sed sancietur
ventas lUdICIO tuo.
VIII
MICHAEL SCOT AND THE TRANSMISSION OF SCIENTIFIC CULTURE
24. M, fol. 127rb-127Vb: Si aliquis venerit ad te in 3 partes causa postulandi
de aliquo pro se vel pro altero in hora Saturni ... uni colori eorum quos
diximus.
Close to Liber Alchandrei, c. 65, MS Pb, fol. IIr: Ideoque si quis ad te venit
causa huiusmodi aliquid querendi, quod tu dicis valde tamen mirabile est,
adhuc non interrogatus iuxta horae ipsius in qua advenit planetam tacite
respondebis humane cogitacioni, insuper etiam sub manu aliquid abscon-
derit, non visu quod sit, dicere poteris ... (fol. IIV) In inicio vero horae Sa-
turni si venerit, lib rum apportat, in quo aliquid a te queret, in manu vitrum
aut aliquid viride ... (fol. IIr) si in fine hore ipsius (sc. Lunae), de re perdita,
in manu aliquid rubeum.
25. M, fol. I 27Vb-128ra: Non est hora determinata planete in qua noviter
intromittere et aliquid laxare ad intromittendum. Et super hoc dixerunt multi
quod si quis aliquid inceperit sub hora Saturni cadet in angustiam, contribu-
lationem facile, nec bene prosperabitur ... Sub hora Lune et ipsa sic crescens,
convenienter prosperabitur et econtrario.
Close to Liber Alchandrei, c. 67, MS Pb, fol. I IV: His indubitanter iuxta
rationem tuam evenientibus ... Si in hora Saturni evenerit, respondebis illum
cadere in angustias, perdere sua, plangere ami cum mortuum. Si in hora
Lune immo illa crescente, ille de hoc quod querit prosperabitur; si vero
decrescente, ille si in iudicio contendet, iudicabitur vel vincetur ...
26. M, fol. 13 Iva: Dicunt enim in concordia Messehalla, Dorotheus,
Alchindus, et Alcorayth, quod in omni questione duo principaliter attendun-
tur, scilicet dominus ascendentis et Luna ubique sit. Horum vero quem in-
venimus fortiorem dicamus esse significatorem initii cui usque rei de qua et
pro qua motus est quem ipse querens. Albumassar enim dicit et est verum
quod quinque sunt que significant finem uniuscuiusque rei: Primum est
dominus quarte domus; secundum est dominus domus Lune sive Luna; ter-
tium est planeta cui Luna iungitur prius [ut] a signo in quo ipsa est; quartum
est dominus domus partis fortune; quintum est dominus quarti signi procul a
Luna in quo ipsa morari probatur.
See Haly, Electiones, I, diff. 2, Ma, fol. 24Vb (ed. Millas Vallicrosa, Las tra-
ascendentis dominus et Luna et quem eorum fortiorem inveneris, faciemus
significatorem initii ... De fine dixit Albuma~ar: Quinque significant finem
rerum: primus horum est dominus quarte domus; secundus, dominus domus
Lune; tertius planeta cui iungetur Luna ad ultimum a signo in quo est; quartus
dominus domus partis fortune; quintus, quartum a signo in quo est Luna.
27. M, fol. I J5rb: Cum interrogatus fueris que ars danda sit puero vel
puelle ... Et si doctrina danda fuerit scriptura sit ...
28. M, fol. IJ5rb: Et si doctrina fuerit cantor vel instrumentorum so ni-
tus ... lyre celi.
Haly, Electiones, 11, chap. 10, diff. 2, Ma, ibid., P, 532b-533a. This para-
12 5
VIII

IX
graph and Michael's
..reworking of it are edited in Burnett, «T' .
. al' b' . eona e pratIca
mUSIC I ara e m SIcIha e nell'Italia meridionale in eta' norm
N E:IT. .. anna e sveva»
uove wemertd" 11 (1990), 79-89 (86- 87). '
29· M, fo1. .1 35 va : In eruditione natandi et navigandi ... 5 fortitudinum
HaIy, Electlones, 11, chap. 11, diff. 2, Ma, fo1. 37Vb. .
0
. 3 . M, fo1. 1.35rb-137rb: Most of these paragraphs draw from HaIy's Elec-
tlones. Th~ relatIon. of M to HaIy can be demonstrated in a passage on sowin
and pIantm~ trees In which the words from Haly are in italics: M, fo1. 136~ Master Theodore, Frederick II's Philosopher 1 *
(Haly, ~lect~ones, 11, ~hap. 5, diff. 4, Ma, fo1. 36ra, P, 530a-b): Et nota uod
p~o semmatlOn~ mehus est quod Luna sit in mobili, et ascendens etiam sit~o-
bile. Pro plantatlone arborum sit ascendens fixum signum et Lun . fi .
Aq' b' a In IXO ut In
. uano propte~ orutatem si~, et i~ Tauro propter dignitatem quam dicitur
m. eo hab.er~, sltque Saturnus dlrectus In sequenti angulo ascendentis et ibi habeat
allquam. dl?mtatem vel ad minus bonum testimonium vel sit in ascendente. Iu iter
aut~m SIt '~ ~spectu ascendentis Iaudabiliter domini eius absque vitio, ho~ est Theodore the Philosopher is a figure who appears quite fre-
IupIter aspIcIat Saturnum bono aspectu et loco in quo sit ei bonum testimonium S d quently in the entourage of the emperor Frederick 11 (1194-1250).
cavendum est a presentia Martis. Hoc est ab eius coniunctione vel aspectu 'ali~
eo ~uod Mars semper impedit Iovem et Saturnum quantum bone. Quod si There is awareness that he was an intermediary between Frederick
fuent observatum, Deo voIente, semen perficiet muItum et pIantatio facta. and Muslim intellectuals and that he was greatly respected by
Frederick 2. No study, however, has been devoted especially to
Theodore, who has tended to be eclipsed, in scholarship and public
imagination, by the court astrologer, Michael Scot. This article
brings together what we know concerning Theodore and his works.
It restores to him a place within an academic community which
was not confined by religious or political boundaries, but em-
braced the whole Mediterranean basin and the Middle East, and it
explores the significance of the role of 'philosopher' in such a
community.
The Latin sources 3 reveal that in the autumn of 1238, at the

1 I am very grateful for the help of Baudouin van den Abeele, Silke

Ackermann, Roberto Casazza, Alejandro Coroleu, Emma Gannage, Martin-

Dietrich Glessgen, Dag Nikolaus Hasse, Peter Jackson, Jeremy Johns, Nicolai
Rubinstein and Clare Woods.
2 The fullest and most accurate accounts remain those of M. AMARI in Storia
dei Musulmani di Sicilia, Florence, 1854-72, Ill, pp. 711-3, and J. L. A.
HUILLARD-BREHOLLES, in Historia diplomatica Friderici Secundi, [= H. B.] 6
vols, Paris, 1859-61 (especially I, pp. dxxix-dxxxi), on which depend more
recent summaries of Theodore' s career such as those of C. H. HASKINS in Studies
in the History of Mediaeval Science, 2nd edition, Cambridge, Mass., 1927, pp.
246-8 and D. ABULAFIA in Frederick II: A Medieval Emperor, London, 1988, p.
126
263-4.
3 See Testimonia 1-7 and Texts 1 and 3 below.

IX
226
siege of Brescia, Theodore is engaged in philosophical disputes
with the Dominican friars. The chronicle of Rolandino mentions
that, in 1239 at Padua, he casts a horoscope for the Emperor. In the
following year he is mentioned in the imperial documents in con-
nection with drafting a letter in Arabic to the king of Tunis. At the
same time he compounds syrups and violet-sugar for the Emperor
and his household, and sends a box of violet-sugar (with an ac-
companying letter) to Pier della Vigna. In 1240-41 at the siege of
Faenza the Emperor is correcting the Latin version of an Arabic
work on falconry known as M oamin, which Theodore had trans-
lated for him. In a document of 9 March 1243 the vineyard of
« domini Theodori imperialis. philosophi » in Messina is mentioned.
In November 1250, Frederick 11 grants to the nephew of his Marshall
the domains which «Master Theodore our philosopher held so
long as he lived ». He must, therefore, have died before this date.
The most substantial of his writings is the translation of the
Moamin, whose manuscripts present one or other of two different
prologues. Besides this it is probable that he translated another
Arabic falconry treatise which appears only in manuscripts of the
Moamin and is written in the same style - the Ghatrif4. One manu-
script of a Latin translation of Averroes's Proemium to his Large
Commentary on Aristotle's Physics states that «This is the
proemium to the commentary of Averroes on the book of Physics
of Aristotle, which master Theodore translated on the request of
the scholars who were in Padua » 5. At some stage in his career he
addressed a letter on regimen to the Emperor 6. Sometime after
1225, the renowned mathematician, Leonard of Pis a (Fibonacci),
appended to two of his works - the Flos and the Liber quadrato-
rum - letters to Theodore, one solving a problem the latter had
raised.
See B. VAN DEN ABEELE, La fauconerie au moyen age, Paris, 1994, p. 29.
4
Erfurt, Amplonian, Fol. 352, fol. 104v: «Istum (sic!) est prohemium com-
5
menti A. super libro Physicorum Aristotilis, quod transtulit magister Theodorus ...
rogatu scolarium qui erant Padue »; Aristoteles Latinus, Codices, I, p. 104.
6 A new edition, and translation, is given below, Text 2.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 227
Theodore is called Frederick's astrologer only in ~he chr?nicle
of Rolandino 7. The other sources are almost una~lmous In de-
scribing Theodore as « philosophus» or, more precIsely, the Em-
peror's « philosophus ». In the imp~rial documents of 1239-4.0 t~e
Emperor refers to him as «Theodorus phiIosophus (et) ~ldehs
noster» 8 an d « phl'losophus noster» 9. Theodore. refers '
to hImself
10
as «Theodorus philosophus» in his letter to PIer della Vlg~a ,
and the superscript of his Letter on Regimen reads: «Eplstola
Theodori philosophi ad imperatorem Fridericum» .11. L~o~ard of
Pisa addresses Theodore as «domine Theodore, lmpenahs aule
summe philosophe » 12.
If these were all the documents we had for the life of Theodore
I we would have only patchy glimpses of the man (partly because of
the absence of imperial documents except for the years 1239-40)
and would have no idea of his origins. !here a~e, however, two
I
I further sources, which considerably amphfy ~he pIcture, but whose
implications have not been sufficiently conSIdered up to no,:. The
first is a reference in the introduction to the Frenc~ « catechIsm of
medieval science» called the Book of Sidrach, WhICh .talks of « ~~
home de Antioche, qui out a noun Theodore le phl~OSophe» ,
who, as one of the Emperor's courtiers, obtained by ?nbery a copy
of the book and sent it to « patriarche Obert de Anhoch~ ». There
7 On the basis of this sole testimony Huillard-Br~holles (n. 2 above): 1.1, P:
dxxxi states that «toutefois, le principal emplol de Theodor~ etalt cel~l
d'astr~logue en titre, ou du moins, de tous les astrologu~s que.Fre~enc entreten~lt
:=.. • Theodore est le seul qui soit designe nommatlvement par ~s
is son service, h· h h ·t from another chroDl-
chroniqueurs contemporains »; a passage w lC e Cl es . h f the
cler, Oodi (MURATORI, VII.83, new ed. VIII.213), does not glV~ t e nam~ 0
astrologer. Haskins would seem to press the evidence too ~ar m suggestmg that
Theodore succeeded Michael Scot as court astrologer (Studles, p. 247).
8 See Testimonia 3 and 6.
9 Testimonium 6.
10 Text 1. ·bl b d as « phisici »
11 Here the abbreviation for « philosophi » could POSSI Y e rea .
Testimonium 8.
12
Testimonium 12; cf. British Library, MS Harley 4486, o.
13
V:
;.«
f: I 72· Th odre
1- R
de Antioche »; both references are cited in H. L. D .. ARD , Cata ogue 0 d 0-
mances in the Department of Manuscripts in the Brztlsh Museum, I, Lon on,
1883, p. 905.

IX
228
was an Albert, patriarch of Antioch from 1226-46, and the thir-
teenth-century date of the Book of Sidrach 14 leads us to expect
that its author knew his facts, although these facts are embedded
in a largely fictitious tale of a text being passed through history
from a legendary King Boctus to ail institute that achieved legen-
dary status - the «school of Toledo ».
. For the second testimony we are on much surer ground, for it
IS that of the well-known Syriac bishop and writer, Barhebraeus
(Gregorius Abulfaragius, 1225/6-86), who provides the following
account in his Mukhfa~ar fa:Jrikh al-duwal (<< Short history of the
dynasties») 15:
Among (the Christian scholars of our time) was the hakim Thadhurl
of ~ntioch, belo~ging t~ the Jacobite sect. He mastered' the Syriac and
LatID languages ID AntlOch, and he became educated there in some of
the sciences of the Ancients. Then he emigrated to Mosul and studied
under Kamal aI-DIn ibn Yunus the works of al-Farabl and Ibn Slna and
solved Euclid and the Almagest. Then he returned to Antioch bu~ did
not ~rolong his stay there because he realised an incapacity in'learning
by himself. So he returned a second time to Ibn Yunus and matured the
knowledge that he had not firmly gained before. He then went to Bagh-
dad and perfected the science of medicine, and mastered its subtleties
and hunted down its peculiarities. Then he went to the Sultan cAla' al-
OIn to serve him, but he found him strange and did not become close
to him. So he travelled to Armenia and served Qustantln the father of
king l:iatim, but he did not enjoy their company, s~ he went with an
ambassador, who was there, to the Emperor, king of the Franks. He
14 S~e B. ~INS,
«Le Livre de Sidrac - Stand der Forschung und neue
Ergeb~lsse », m Wissensliteratur im Mittelalter und in der /rahen Neuzeit, ed.
H. BRUNNER and N. R. WOLF, Wiesbaden, 1993, pp. 37-52. The sources of the
Book of Sidrach up to now identified are entirely Western, and include the
« Iohannes Hispalensis et Limiensis » translation of the regimen from the Secret
of secrets of Pseudo-Aristotle (see n. 17 below).
15 For the Arabic text see Testimonia 13 and 14. A German translation of the
first passage is given in H. SUTER, Beitriige zur Geschichte der Mathematik bei
den Griechen und Arabern, Abhandlungen zur Qeschichte der Naturwissenschaf-
ten und der Medizin, 4, Erlangen, 1922, pp. 7-8. See also «Ibn al-cIbri » in
Encyclopedia of Islam, 2nd ed., and, for the first edition and a Latin translation
of the~~ passages, Historia compendiosa dynastiarum authore Gregorio Abul-
Phara}lO, ed. E. Pocock, Oxford, 1672, I, p. 521 (Arabic) and 11, p. 341 (Latin).
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 229
obtained from him great favours and found benefit for himself through
him, and (the Emperor) bestowed on him as a fief a city, just as it was,
with its territories (or bestowed on him a fief in Messina<... 1> with
its territories)16. But when he had secured his position and much wealth,
he yearned for his homeland and family, but (the Emperor) did not allow
him to go. So he stayed until the opportunity gave him the possibility,
through the departure of the king on one of his expeditions to the we-
stern regions. Then he packed his bags, put together his possessions and
boarded a ship, which he had prepared for his escape. He travelled on
the sea with those of his servants who were with him, seeking the land
of Acre. But on their journey a wind blew and threw them ashore at a
town in which the king had anchored. When this was announced to
Thadhur1, he took some poison which he had with him and died, from
shame rather than fear, because the king would not have allowed someone
like him to be killed.
A few pages earlier Barhebraeus had mentioned that
In the year 626 (AD 1226) there died YaCqub ibn ~aqlan the Christian *
Melkite doctor from Jerusalem. He was born in Jerusalem the Sublime
and learnt there some philosophy under Thadhurl, the philosopher
(<< failasuf») from Antioch, who will be mentioned after this date.
One would expect these accounts to be reliable, because
Barhebraeus himself was a Jacobite whose family moved to Antioch
in 1244; he studied there and, after a few years in various places,
he returned to the neighbouring city of Aleppo in 1253 where he
passed the rest of his life as Metropolitan of the Syriac church.
Thadhuri is an acceptable Arabic transliteration of « Theodorus »,
and the information here confirms that of the Latin documents that
Theodore was referred to as «the philosopher» (<< failasuf », for
which, on one occasion, the Arabic equivalent «~akim» is given),
16 There seems to be a corruption here. The fourth form of the verb q-l- C is
the technical term for bestowing a fief, but usually takes two direct objects; «bi-
madina » should mean « in a city» Gust as «bi-hi » means « in Antioch » in the
second sentence). On the other hand, the phrase is tantalizingly close to «bi-
massina » (= « in Messina »); in this case the following word or words would be
a corruption of the fief itself. I owe this explanation to Jeremy Johns. The one
variant provided by Pocock is « bi-madinatihi » (<< in his city»).
 IX
230
and that he served the Emperor, from whom he received property,
perhaps in Messina. But they give us much further information, all
of which appears to be reliable.
First, he was born in Antioch. Throughout this period Antioch
was a Christian principality, ruled by the Frankish princes
Bohemond IV (1201-16 and 1219-33) and Bohemond V (1233-
52). Hence it is not surprising that Theodore acquired a mastery of
Latin here. Some time before 1230 Philip of Tripoli found a copy
of the Secret of Secrets attributed to Aristotle in Antioch, and trans-
lated it. The earliest known citations of his translation are in the
works of Frederick II's astrologer, Michael Scot, and the prologue
to Theodore's Letter on Regimen documents the arrival of the text
into Frederick's hands 17. Antioch's association with the West was
very close. An Antiochene origin could account both for Theodore's
knowledge of Latin and for his opportunity to come to the notice
of the Emperor.
But the next sentence in Barhebraeus's biography shows how
Theodore took advantage of the very best that was available in
Islamic culture. Kamal aI-Din ibn Yiinus (1156--1242) was, by all
accounts, the most learned and sought-after teacher in the Islamic
world of his generation 18. He was an expert not only in Arabic
philology, Islamic law and the Koran, but also in the sciences of
the Ancients (i.e., the Greeks), and especially in mathematics. He
had established a school in a mosque in Mosul to which even Jews
and Christians came to hear his interpretations of the Torah and the
Gospels. Amongst his illustrious pupils was Naslr . aI-Din al-Tusi
. ,
the great astronomer and Hellenist (1201-74) 19. It was to Kamal
17 S. J. WILLIAMS, «The Early Circulation of the Pseudo-Aristotelian Secret
of Secrets in the West », Microiogus, 2, 1994, pp. 127-44 (137-8). Williams
shows that it is the new translation of Philip of Tripoli that is at issue, and not
the older « Iohannes Hispalensis et Limiensis» translation of the section of the
Se~~et, of Secre~s on ~egimen. He concludes with some caution (p. 138) that
Phlhp s translatIon arrIved at Frederick's court ca. 1228-35.
18 SUTER, Beitriige (n. 15 above, pp. 1-7), conveniently assembles and trans-
lates the accounts of Kamal aI-Din ibn Yunus in the Arabic biobibliographers.
19 For the most up-to-date biography of al-Tiisi see F. J. RAOEP, Na~ir ai-Din
al-Tiisl's Memoir on Astronomy, 2 vols, New York etc., 1993, I, pp. 3-15. Al-
Tusi may have studied with Kamal aI-Din in the late 1220s and early 1230s. The
IX
MASTER THEODORE, FREDERICK U'S PHILOSOPHER 231
aI-Din that the Ayyubid Sultan, al-Malik al-Kamil (reigned 1218-
38), passed on the questions on mathematics that Frederick 11 had
sent him 20. The only works that survive of his are on mathematics 21
and include a treatise which treats exactly the same problem as the
one that Theodore. asked Leonard of Pis a to solve for him 22. But
it was not only mathematics that Theodore learnt from him; he
also studied the works of the philosophers al-Farabi and Ibn Sina;
which is significant, as we shall see. Barhebraeus's narrative im-
plies that Theodore appreciated the superior academic climate of
Mosul in comparison to Antioch where he felt intellectually iso-
lated.
The next stage in Theodore's education took him to Baghdad.
We are not told who his teachers were there. There was much
scholarly interchange between Mosul and Baghdad, and Kamal al-
Din had himself studied in Baghdad. Theodore specialised in medi-
cine there and completed the acquisition of skills which would
make him useful to rulers.
The first of these rulers was the « Sultan (Ala) aI-Din ». This
could have been the Grand Master of the Isma (ilis or Assassins,
term « Hellenist », as used by Ragep and other scholars, indicates a scholar who
was « heir to the intellectual traditions of late Greek antiquity that had begun to
be «naturalized» within an Islamic context» (ibid., p. 4); it equally applies to
Ibn Yiinus.
20 SUTER, Beitriige, p. 3, following al-Qazwini (1203-83). Ibn Abi U~aibi(a
(after 1194--1270) states that Frederick sent an envoy directly to Kamal aI-Din,
and that the ruler of Mosul bade him to do something about his usual bedraggled
appearance in order to impress the foreignor (ibid., p. 6).
21 C. BROCl(ELMANN, Geschichte der arabischen Literatur, Supplementband I,
Leiden, 1937, p. 859, to which may be added another copy of the letter on the
theorem that the sum of the squares of two odd numbers is not a perfect square
in MS Cairo, Egyptian National Library, DR 703,4 (information from David
~~. .
22 This had been pointed out by Suter in his Nachtrag to Die Mathematr.ker
und Astronomen der Araber und ihre Werke, Abhandlungen zur Geschichte der
mathematischen Wissenschaften, 14, 1902. The correspondence is noted by Roshdi
Rashed who emphasises the importance of seeing the achievement of Leonard of
Pisa within the context of the Arabic mathematical tradition in his «Fibonacci
et les mathematiques arabes », Microiogus, 2, 1994, pp. 145-60; I would wish !o
see Theodore' s intellectual orientation as parallel to that of Leonard, and ID
respect to the Arabic philosophical tradition.
 IX
232
'Ala' aI-Din MUQammad III (ruled 1221-55), whom Nasir aI-Din
~l- Tusi .later served, benefitting from the holdings of hi~ famous
lIbrary In Alamut. But it is more likely to have been the Sultan of
the Seljuks of Rum, 'Ala' aI-Din Kay-Qubadh I (ruled 1219-37),
not only because he was called a « Sultan» rather than a « Grand
Master », but also because the kingdom of RUm, with its capital at
Konya ~Iconium) was closer to Antioch - and to Europe. Moreo-
ver, dunng Kay-Qubadh's reign the kingdom reached its apogee 23.
Kay-Qubadh expanded his territory, rebuilt his capital and com-
pleted the Great Mosque which can still be seen today. According
to the principal chronicler of the dynasty, Ibn Bibi, Kay-Qubadh
was versed in all branches of knowledge. He invited to his court
the father of the mystic Jalal aI-Din al-Rumi, and the sultan al-
Malik al-Kamil sent him a leading jurist to be the head of his
ruqaha:1. It would not be surprising if Theodore found his court
attractive, especially since legend had it that the bones of Plato
were preserved in a monastery in the capital. However, things did
not work out well, and Theodore next went to the court of the
Sultan's neighbour - and enemy - the Armenian « Qustantin the
father of king l:Iatim ». . .
Heloum I (Hay ton in the Western sources) was king of Cilician
Armenia from 1226 until 1270. His father, Constantine of Lampron
lived un~il 1261, but had been regent between the death of Kin~
Leon 11 In 1221 and the accession of King Heloum, and was the
pow~r beh~n.d the throne 24. For, in 1225, Leon's daughter, Isabel,
marned Phlhp, the son of Bohemond rv, and, in reaction to Philip's
attemp~s to bring into Cilicia Antiochene customs and subject the
ArmenIan or.th~dox. church to Rome, Constantine, it appears, ar-
ranged for hIS lmpnsonment and death. Constantine then married
23 C CAHEN, Pre-Ottoman Turkey, London, 1968, pp. 124, 201, 208-9 and
258. Theodore's presence in the Seljuk court does not seem to have been noticed
before.
24 IDEM, La Syrie du Nord, Paris, 1940, pp. 631-5 and S. DER NERSESSIAN
« The Kingd~m of C~lician Armenia », in K. M. SEITON, A History of the Cru~
sades, 11, Phlladelphla, 1962, pp. 630-61 (651-2). Again, Theodore does not
appear to ~e m~~tioned in the histories of Cilician Armenia, in spite of the fact
that Suter IdentifIes Hetoum correctly (BeitrlJge, n. 15 above).
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 233
Isabel to his own young son Hetoum and set him up as king in
1226. This enraged the Prince of Antioch so much that he formed
an alliance with 'Ala' aI-Din Kay-Qubadh in order to invade Ar-
menia. This plan was unsuccessful, but the Sultan invaded the
country in 1233 and imposed tribute, and bad relationships continued
between Antioch and Cilician Armenia. Whether Theodore was
affected by these conflicts we do not know; in any case, he did not
stay in Armenia. The wording of Barhebraeus's account does not
necessarily imply that he was in Constantine's service after the
accession of Hetoum.
No embassy from Frederick 11 to the king of Armenia is re-
corded, though it is quite likely that the Emperor should have sent
one in preparation for, or during, his crusade to the Holy Land in
the late 1220s. Did Frederick «poach» the philosopher from
Constantine of Lampron, or did Theodore himself seize an oppor-
tunity for becoming a courtier of a much more powerful master
than Constantine? One manuscript of the Liber novem iudicum
says that the « great Caliph» sent Master Theodore to the Emperor
Frederick » 25. The Latin writer is obviously confused about the
difference between eastern potentates, but it is possible that
Frederick could have obtained Theodore from Constantine as part
of a diplomatic deal. Whatever was the case, the next stage in the
biography of Theodore must be pieced together from the Latin
sources which we have mentioned above.
These sources are silent ·about the death of Theodore. There are
no grounds for including him amongst the astrologers killed out-
25 London, BL, Royal 12 O.VIII (written in Italy in the 14th century), fol. 1:
«Liber novem iudicum quem missit Soldanus Babilonie Imperatori Federico
tempore quo et magnus Chalif misit magistrum Theodorum eidem Imperatori
Federico ». The Liber novem iudicum as it is known in Latin manuscripts cannot
have been sent at this date from an eastern potentate since several, if not all, the
individual astrological texts which make up the nine judges were translated from
Arabic in Spain by Hermann of Carinthia and Hugo of Santalla in the mid-
twelfth century. Michael Scot refers to a slightly different set of astrological
authorities as the «liber 9 iudicum », but these, too, he seems to know from
Latin rather than Arabic sources; see my «Michael Scot and the Transmission
of Scientific Culture from Toledo to Bologna via the Court of Frederick 11
Hohenstaufen », Micrologus, 2, 1994, pp. 101-26 (119).

IX
234
side the walls of Pavia in 1248 26 • On the other hand Barhebraeus's
account of Theodore's homesickness and attempt to escape back to
Antioch accords well with events earlier in his career - when he
interrupted his studies in MosuI to return to Antioch, and when he
found employment successively in two states which were close to
his home country. He was making for Acre because that was the
principal port on the seaboard of the eastern Mediterranean. His
unfortunate encounter with Frederick 11 could have taken place
anywhere 27.
To put precise dates to the stages of his career is difficult.
Theodore is in Frederick's service by 1238 (and possibly already
by 1233, if he is the philosopher with whom Juda b. Salomon
corresponded 28). He must have arrived in Frederick's court after
1219 ('Ala) aI-Din Kay-Qubadh's accession), and probably after
1221 (Constantine's accession to the regency) or even 1226 (if
Barhebraeus's wording implies Hetoum was already king). Only
by identifying the embassy sent to Armenia by Frederick 11 could
one discover when Theodore left the East. At some stage of his
career (and before 1226) he would appear to have been in Jerusa-
lem, where he taught the otherwise unknown Ya qiib ibn ~aqIan 29.
C
However, even if the chronology is unclear, the mere fact that
Theodore took advantage of the best teaching in the Middle East,
and was in turn, patronised by two of the most important rulers of
the area, invites us to look for the influence of this background in
his writings.
26HASKINS, Studies (n. 2 above), p. 247, n. 24.
27Suter's scepticism about the story of Thadhfiri's death is based on his
misapprehension that Frederick 11 must have been in a town on the Syrian coast
at the time; see his Die Mathematiker und Astronomen der Araber und ihre
Werke, Abhandlungen zur Geschichte der mathematischen Wissenschaften 10
1900, no. 345, p. 137, note d.
28 See below pp. 235-6.
• 29 Bar~ebrae~s's ref~rence to Theodore's teaching in Jerusalem may lie be-
hmd MOritz Stemschnelder's statement that Theodore «solI schon «um 1184»
in Jerusalem gebUiht »: Die europiiischen Ubersetzungen aus dem Arabischen
bis Mitte des 17. Jahrhunderts, Sitzungsberichte der Kais. Akademie der Wis-
senschaften .n Wien, phil.-hist. Klasse, 149, Vienna, 1904, no. 116.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 235
11
First there are his Arabic writings. Unfortunately none of these
has been identified so far 30. His knowledge of the language made
him useful for the Emperor, as one can see from the extant in-
structions for him to draft a letter destined for the Emir of Tunis
(Testimonia 4 and 5). He. also apparently wrote in Arabic about
his own interests. For we have evidence of a correspondence
concerning geometry between the Jewish scholar, Juda b. Salomon
ha-Cohen and « the philosopher» of the emperor Frederick. Juda
writes:
In my youth, when I was still in Spain, the philosopher of the Emperor
addressed some questions on geometry to me. I replied when I was 18
years old and still with my father... in the city of Toledo. But the matter
had some consequences, for it lead me to go to Tuscany, and there I
translated from Arabic into Hebrew those questions ... When the solu-
tions were shown to the emperor Frederick he greatly rejoiced at the
replies I had given to the man who pretended to be a philosopher in his
presence. There were between us many discussions on numerous sub-
jects with numerous questions and replies [these were made in Arabic] ...
and this continued for ten years 31.
It is likely that this «philosopher of the Emperor» was
Theodore, because the correspondence was in Arabic and took place
exactly between those years, 1233-43, when Theodore could have
been in the Emperor's service, and no one else during that period
is consistently called «the Emperor's philosopher ». Juda speaks
30 A reference to the opinion of « the philosopher Theodorus (Diyfidfirfis)>>
concerning the Milky Way appears in an anonymous text in Berlin, MS Ahlwardt,
1174, p. 96ff, and is reported in E. Wiedemann, «Beitrage zur Geschichte der
Naturwissenschaften, LXXIV: Uber der Milchstrasse bei den Arabern» (1926-
' , 7), reprinted in Au/siitze zur arabischen Wissenscha!tsgeschichte, 2, 1970, p.
670 .
31 C. SIRAT, « Les traducteurs juifs a la cour des rois de Sicile et de Naples »,
in Traduction et traducteurs au moyen age, ed. G. CONTAMINE, Paris, 1989, pp.
168-91 (175), and, more fully, in BAD., « Juda b. Salomon ha-Cohen, philosophe,
astronome et peut-etre kabbaliste de la premiere moitie du XIIIe siecle », Italia,
2, 1977, pp. 39-61.
 IX
236
rather disparagingly about this philosopher, but this could be be-
cause of an excess of youthful arrogance. Theodore, as we have
seen, studied Euclid with Kamal aI-DIn ibn Yiinus, and he ad-
dressed an arithmetical question to Leonard of Pisa (Testimonium
9). Leonard addressed a new method of calculating arithmetical
problems (in the dedication of an appendix to his Flos) to Theodore
so that he could correct and put it in order 32. This implies that
Leona~d respected Theodore's judgement in arithmetic, as befitting
a pupIl of Ibn Yiinus.
Theodore would have received or interpreted Arabic letters sent
to ~he imper~al cour~. He must have acquired a considerable repu-
tatIon for. thIS, for hIS name is attached, as a recipient, to a forged
letter, ~hlCh appears to have been deliberately circulated as propa-
ganda In favour of a crusade lead by Frederick 11 against the
Mongols 33. This letter is addressed to « his wise friend most dear
Theo~or~, the p~ilosopher of the most unconquerable Caesar» by
« Al~lndl, t~e pne~t and philosopher of the Caliph of Baghdad» 34
and It was clTculatIng before 1246-8, i.e., probably within the life-
time of Theodore.
.More information might be expected from the extant writings
attnbuted to Theodore. These are all in Latin. The letter to Pier
della Vigna (if complete) is very shott, and shows no distinctive
traits; it was evidently drafted with the help of an imperial notary.
The Letter on Regimen, on the other hand, has a more indi-
vidual character. Unfortunately the one manuscript that survives
has been hastily copied and shows some signs of disorder. The
Letter i~ both succinct and witty, including several plays on words
formed In the same way (2, 5, 15 and 19). It treats in succession
food, drink, « accidents of the soul », sleep and sexual intercourse;
32 Testimonium 8. Theodore is different from the «intimate friend» here
who asked for an easier way to solve certain problems in the Liber abbaci
33. This letter, which occurs in the same manuscript as Theodore's Le~ter on
Regimen (see pp. 266-70 below), is edited and discussed in my «An Apocryphal
Letter from t~e Arabic Philosopher al-Kindi to Theodore, Frederick II's Astrolo-
ger, Concermng Gog and Magog, the Enclosed Nations and the Scourge of the
Mongols », Viator, 15, 1984, pp. 151-67.
34 «~rudenti vi~o amico suo karissimo Theodoro invictissimi Cesaris filoso-
fo, Alkmdus Alcahf de Baldac sacerdos et filosophus, salutem in Omnipotente ».
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 237
it may be incomplete, since one would expect some words on the
bath, on purging and on exercise 35. A leitmotif is the theory that
good health is maintained by the possession of a good « appetite ».
Two other regimens were addressed to Frederick 11: (a) Adam of
Cremona's Capitula... de regimine et via itineris et fine peregri-
nancium (<< Chapters ... on the regimen, the journeying and the aim
of those going on pilgrimage») written before Frederick's depar-
ture for the Holy Land in 1227/8 36 , and (b) Petrus Hispanus's
Epistola super regimen sanitatis 37. These draw heavily on
Avicenna's Canon and the old translation of the Secret of Secrets
respectively. Theodore's work does not correspond verbally to any
of these texts, and in some places gives different advice. Frederick
seems to have been aware of these discrepancies. For example, the
Secret of Secrets recommends rest after meals; Theodore says that
one should rather do some light exercise and not rest until at least
two hours have elapsed (29). Frederick is said to have had two
men disembowelled in order to show the respective effects of sleep
and exercise, probably to discover which authority was more
reliable 38. There are no Arabic words and apparently no specifically
Arabic doctrine in Theodore's work; rather a Greek word - discrasia
35 These are the « non-naturals » in Galenic medicine, classified by Constantine
the African, in Pantegni, Practica, 1, c. 18 (MS British Library, Add. 22719,
fol. 170v) as « aer, exercitium, balnea, cibi et potus, somnus et vigiliae, coitus,
corporis mundificatio » and « accidentia anime ».
36 Edited by F. HONGER, Aertzliche Verhaltungsmassregeln aUf dem Heerzug
ins heilige Land fUr Kaiser Friedrich /I, Leipzig, diss., 1913. This text cites
Aristotle's Meteora and «Avicenna »: it depends closely on the regimen in
Avicenna's Canon medicinae, Book 1, Fen 3, Doctrina 2.
37 British Library, Harley 5218, fols lr-3r, headed «Epistola m. Pe. Hyspn.
missa ad Imperatorem Fridericum super regimen sanitatis », and beginning « Suo
domino pre cunctis mortalibus metuendo Friderico Romanorum imperatori fontis
Pegasei liquore instructo magister Pe. Hyspanus eius alumpnus senex artis medi-
cine professor seipsum et opus conservande recipe sanitatis. Sanitas est tempera-
mentum custodiens ... » The work, like the Secret of Secrets, gives instructions
on what to do from the time of getting up onwards, and cites «Aristoteles ad
Alexandrum magnum imperatorem ». An edition, and discussion of the authenti-
city of the preface is given in M. H. Da Rocha Pereira, Obras medicas de Pedro
Hispano, Coimbra, 1973, pp. 427-93. Contrary to the assertion of Haskins
(Studies, p. 257), there is no indication here that this Petrus Hispanus is a
student of Theodore.
38 HASKINS, Studies, p. 262, referring to Salimbene's chronicle.
 IX
238
- is used (23), which points to the terminology of the School of
Salerno and the translations of Constantine the Mrican 39.
The use of Constantinian terminology is also found in the Short
Prologue to Theodore's translation of M oamin, where reference is
made to lovesickness «qui dicitur grece «hereos» (<< which is
called «hereos » [i.e., eros] in Greek») 40. This Short Prologue is
the ~ne that Occurs in the majority of the manuscripts, and it em-
phasIses the usefulness of falconry as a therapeutic exercise. The
~oamin itsel~ is pre-eminently a medical work, treating of the J.i
dIseases of buds and of the dogs used in the chase.
There is, then, nothing in the medical writings that immedi-
a~ely. strikes one as coming from Theodore's experience of medi-
cIne In Baghdad or elsewhere in the Middle East, rather than as an
elaboration of the medical information already available in the
West. Nor is there anything distinctly Arabic in the style or
v?~abularly of these original texts 41. We must consider the possi-
bIlIty that Theodore was not solely responsible for the redaction of
these texts. Indeed, we have the explicit statement that the trans-
lation of Moamin was revised by the Emperor, and in the Short
Prologue there are lists of hunting instruments and animals which
correspond with those in Frederick's own De arte venandi cum
39 F~ederick suppo~ted the School of Salerno. The Pantegni of Constantine
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 239
avibus 42. These lists do not occur in the Long Prologue. Here,
rather, Theodore refers to himself in the first person (32) (he is
only indicated in the third person as the tran.sl~tor. in the S~ort
Prologue) and gives Frederick's full and offICIal tItle - a tItle
matching exactly the form of his name in documents from the
imperial chancery. It would seem then, that the Lo~g Prologue
accompanied the copy of M oamin that Theodore dedIcated to the
I
Emperor, and that the Short Prologue represents a later revision,
by (or sponsored by) the Emperor himself as a preparation for a
:,
more general diffusion of the work. If this is so, one might expect
to find the authentic voice of Theodore in the Long Prologue.
This Long Prologue is found in three closely related manu-
scripts. It includes the same discussion of falconry's beneficial
effects on the health as that found in the Short Prologue. However,
it adds before this a long exposition on the nature of pleasure
(<< delectatio »), which cites Aristotle's De anima, Nicomachean
Ethics, and Rhetorica 43.
The first thing to be considered is whether this extra section is
a direct translation of material in a preface in the Arabic original
of the Moamin. Unfortunately, this original has not been identified.
The Arabic text that is closest in its contents is the Kitab I)awari
al-.tayr(<< the book on birds of prey») by al-~hitrifibn ?udama a.l-
Ghassani 44, portions of which also appear In the LatIn Ghatrif.

the Afncan was used 10 the Prim us liber of Michael Scot's Liber introductorius'
see D. JACQUART, «La physiognomonie a l'epoque de Frederic 11 », Micrologus:
42 See Short Prologue, 10-12 = De arte venandi cum .avi~us, ed. C .. A.
2, 1994, pp. 19-37 (24). For the parallels in Theodore's work see the notes to the
text edited below. Willemsen, Leipzig, 1942, p. 4. Martin-Dietrich Glessgen I~ h~s forthcommg
40 For « h~reos» as the term for lovesickness in the Middle Ages see M.
edition of the Latin and Italian translations of Moamin has mdlcated that the
W ~CK, Lo~eslckness in the Middle Ages: The Viaticum and its Commentaries,
same zoological terms are used in the Moamin a.nd the De ~rte, .and that the De
~hllacdel~~la, 1990. Wack turns her attention to the terminology for lovesickness arte venandi has been planned with the Moam1O already 10 mmd. The man~­
scripts do not show an unrevised and a revised version of the text of the Moamzn
10 « Ah Ibn al-cAbbas al-Magusi and Constantine on Love» in ~lf ibn al-
itself; my claim would be that the revised version co~p~etely super~ede~ the
~Abbas al-Magusi, eds C. BURNETT and D. JACQUART, Leiden, 1994, pp. 161-202,
~ut, of all the instances of the use of the term eros/ereos/heros/hereos that she unrevised version, and all we have left of Theodore' s ongmal translation IS the
hsts, n~ne states that the term is Greek. In Constantine' s De melancholia (a
dedication referred to here as the Long Prologue.
43 HASKINS, Studies, p. 318, was the first to draw attention ~o this prologu~
translatIOn of a work by Is~aq ibn CImran) horseriding is recommended as a cure
for melancholy (ed. K. Garbers, Hamburg, 1977, pp. 124-5). and the unexpectedness of these citations: « ... the known verSiOns of t~e Nl-
• 41 In the trans~ation of Moamin on the other hand Arabic terms for species of
comachean Ethics and Rhetoric, made in the thirteenth century, have not hitherto
been connected with Sicily».
buds, some medical substances, and illnesses, are retained' but even so the
Latin text is idiomatic and easy to understand; see VAN DEN AB~ELE:La fauco~erie
44 A facsimile of a manuscript of this text, with a preface by Detlef Moller,

(n. 4 above), pp. 28-9. has appeared in the Publications of the Institute for the History of Arabic-
Islamic Science, ed. F. Sezgin, series C, vol. 25, Frankfurt, 1986.

IX
240
The short preface to this text merely gives the history of the book.
However, another Arabic text which was more similar to the origi-
nal of the Moamin was translated into Spanish in the court of
Alfo?so el Sabio, king of Le6n and Castile (ruled 1256-84), under
the title Libro de los animales que cazan 45. This text includes a
longer and more theoretical prefatory chapter, that emphasises the
nobility of hunting and the fact that hunting is the most appropriate
diversion for kings, and remarks that «knowing how to rule is a
great ~art of philosophy» 46. The closest parallel here to the argu-
ment In the Long Prologue is the description of the two « activi-
ties» (<< mester ») of kings: (1) overcoming their enemies and con-
trolling what they have to defend; this is their necessary activity;
J
I
(2) the hunt, which is a diversion from, and an imitation of, their
first activity; this is their voluntary activity 47. However, the starting
point here is not the definition and types of pleasure and there is
no mention of Aristotelian philosophy in this preface. On the other
hand, pleasure is the starting-point in the Short Prologue, and is a
preoccupation in Frederick's own De arte venandi 48. It would seem,
then, that the Long Prologue is of oriental inspiration, but that
Theodore accommodated the oriental mat~rial to the common pat-
tern of Latin prefaces that emphasise both the pleasure and the
45 [Mu1}ammad ibn cAbd Allah ibn cUmar al-Bayzar], Libro de Los animales
que cazan, ed. J. M. FRADEJAS RUEDA, Madrid, 1987.
:: Ib~d., p. 10: « saber gove~nar es una grand partida de filosofia ».
. IbId., pp. 10--11: «E qUI estas dos cosas pudiere complir conviene quel
dIgan rey, porque 10 ven~e por fuer~a e por apoderamiento que aya sobre las
c.o~as ques le quieren defender ... e esto puedelo aver por armas e por maneras de
hdIar e por esto es costumbre de los reyes de usar maneras de lidiar por tal que
sean reyes. ~ las cosas v~rdaderas an unas semejan~as en 10 demas que semejan
maneras de Juego. E la hd es mester verdadero que conviene a los reyes por tal
de apoderars~, e por ta~ de aprender maestria de lid, usan unas cosas en juego
que son semeJan~as de hd, e es maestria de ca~a en que ay an depuerto e rembran~a
del mester por que ellos regnan e por aver gozo de ven~er, ca el mester primero
~erdadero fazenle porque non 10 pueden escusar, mas el segundo, que es de
Juego, fazenlo por su voluntad, e este mester que es de ca~a faziendoles pro en
deportandose en ello es cosa muy semejante al mester verdadero que es lidiar e
ven~er ».
48 See n. 50.
IX
MASTER THEODORE, FREDERICK n's PHILOSOPHER
241
benefit of the subject matter 49; he also added a philosophical di-
mension. Aristotle is mentioned by Frederick 11, too, as someone
whom he has followed but found wanting 50. Theodore is more
indebted to the Peripatetic philosopher, but his citations are not
unproblematic.
The most obvious use is that made of the Ethics, whose tenth
book is on pleasure. The sentence immediately preceding the
reference to the Ethics is a quotation from Book X. We may com-
pare the Long Prologue with Grossetestes's «pure» version of the
Ethics:
Ethics
Long Prologue
10 Et unaqueque istarum [opera-
(1175b27) Secundum unamqua-
tionum] coniungitur delectationi et
mque enim operacionem propria
tend it ad finem proprium.
delectacio est.
(1174b32-3) Perficit autem opera-
cionem delectacio... ut superve-
niens quidam finis 51.
The problem is, that Grosseteste's version of the Ethics was
not made until 1245-6. We do, however, have versions of Books
11_III (<< Ethica vetus » or « translatio antiquissima ») and of Book
I (<< Ethica nova» or « translatio antiquior ») and citations from a
translation of the rest of the Ethics which appear to belong to the
« translatio antiquior ». Moreover, Grosseteste based his own ver-
49 This rhetorical devise, which derives from Horace Ars poetica, vv. 333-4,
apparently is not a feature of Arabic prefaces; see P. FREIMARK, Das Vorwort als
literarische Form in des arabischen Literatur, Diss., Munster i. W., 1967.
so De arte venandi, ed. WILLEMSEN, p. 1: «in scribendo etiam Aristotilem, ubi
oportuit, secuti sumus. In pluribus enim, sicut experientia didicimus, maxi~e in
naturis quarundam avium, discrepare a veritate videtur. Propter hoc non seq~Imur
principem philosophorum in omnibus, raro namque aut nunquam venatlOnes
avium exercuit, sed nos semper dileximus et exercuimus ». .
51 «For to each activity there is a proper pleasure »; « Pleasure compl~tes the
activity ... as an end which supervenes »; translations by W. D. Ross rev.Ised by
J. O. URMSON, in The Complete Works of Aristotle, ed. J. Barnes, 2 vols, Princeton,
NJ,198~ILpp. 1858and185~

IX
242
* sion of the Ethics on these older versions 52. Among the few (and
earliest) citations from later books of the Ethics are some in the
Liber introductorius of Michael Scot, who refers to Aristotle's text
as the «Ethica Nicomachia» 53. On the basis of these quotations
and the testimony of a thirteenth-century manuscript that « transtulit
magister Michael Scotus de Greco eloquio in Latinum », Rene
Gauthier has suggested that Michael Scot himself was responsible
for the translatio antiquior 54. The title used by Michael in his
Liber introductorius is almost the same as the one we find in the
Long Prologue (<< Nicomachia »), and is different, for example,
from that later adopted by Grosseteste (<< Aristotilis moralia ad
Nicomachum »), and the terminology of the Long Prologue is also
compatible with this translation of the Ethics. But the passages
J
I
I
from Book X alluded to by Theodore have not been found in any
other source; they are the earliest citations of these passages in
Latin.
The allusion to the Rhetoric is not so obvious. Theodore may
have had in mind 1362a19-21, which, in the oldest translation
(<< translatio anonyma»), reads: « deliberant vero non de fine, sed
de hiis que sunt ad finem, hec enim sunt honesta secundum
ope rationes ». This « translatio anonyma », however, survives com-
plete in only two manuscripts, the earliest of the mid-thirteenth
century. Versions were made by Hermann the German shortly be-
fore 1256 and William of Moerbeke later in the thirteenth century
- both too late to have been used by Theodore 55.
The reference to the De anima is even more oblique. The ver-
sion of Aristotle's text current from the mid-twelfth century on-
wards - that of J ames of Venice - does not talk in terms of a
52 R. A. GAUTHIER in Aristoteles Latinus, XXVI.I-3, fasciculus primus, Ethica
Nicomachea, prae/atio, Leiden and Brussels, 1974, esp. pp. clxxxvii--cxcv.
53 These are brought together and analysed by GAUTHIER, ibid., p. cxxxvi.
54 GAUTHIER, ibid., pp. cxlii--cxlvii. Joanna Judycka has questioned this in her
edition of the De generatione et corruptione (Aristoteles Latinus, IX.l, Leiden,
1986, p. xxxviii) and thinks rather that the early translations of Books 11 and Ill,
and that of Book I are by the same person.
55 See Aristoteles Latinus XXXI.I-2, Rhetorica, ed. B. SCHNEIDER, Leiden,
1978, praefatio.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 243
« perfectio » of an entity (as the translation of Aristotle~s « entele-
kheia »), but rather of an « actus» 56. The term « perfectlo ): occu~s
in some late Classical sources 57, but i~ particul~r1y p~oml~e~t. In
Latin translations of Arabic texts that dISCUSS ~nstotle s def~nltlOn
of the soul, being the translation of the ArabIc word .« kamal » (~
« perfection»). Thus it is the word used in the tr~nslat1?ns of Q~sta
ibn Luqa's De differentia spiritus et animae, AVlcenna s De anlm~,
and Averroes's Great Commentary on the De anima. The latter IS
attributed to Michael Scot in one of the 57 manuscripts that con-
tain it 58. The Long Prologue, however, seems less indebted ~o
Aristotle's De anima than to Avicenna's text. This is apparent In
the echoes of distinctly Avicennan doctrines in the statements that
« there is a proper perfection, according to th.at which ~N ature) has
granted its preparations to receive from the vlftue flowIng from the
active intellect (2») and that there are «ten senses)~ (9 ~d 24;
i.e., the five external senses and the internal senses WhICh A~l~enna
distinguished into five also). The assertion that « the ~epetlt1on of
the activity has been weakening the instruments (3) » l~ also found
in Avicenna's De anima, but not in Aristotle's. Yet tn all these
cases Theodore's terminology does not match that of the only known
Latin translation of Avicenna's De anima - that made by Avendauth
56 Aristotle's «universally applicable» definition of the soul,. that it is « the
first perfection of a natural body which has organ~)~ (412b?-.6) IS translated by
James of Venice as «actus primus corporis phYSlCl orgamcl» (apud Albertus
Magnus, De anima, ed. C. STROICK, Miinster i. W., 1968, p. 67). . .
57 For example, in Calcidius; see Plato, Timaeus, a CalcldlO translatus
commentarioque instructus, eds P. J. JENSEN a~d J. H. W~SZINK: Lon~o.n: 1962, c.
122 (p. 235.8-9). A convenient list of the verSIons of Ar~sto!le s definItion of t~e
soul current in Latin before the early thirteenth century IS g~ve~ by D. CALLU,S In
«The Treatise of John Blund on the Soul », in Autour d 'Arzstote, offert a A.
Mansion, Louvain, 1955, pp. 490-1. .. .
58 P . BN lat 14385 fol. 133r: «Incipit commentum hbn de amma
ans, ,. , [ . '1 t M· h I
Aristotelis philosophy quem commentatus est Averroes in greco s~c. e . lC a~
Scotus transtulit in latinum »; ibid., fol. 16Ov: « Explicit liber.de a.mma.ArlstoZ.hS
commentatus ab Avenrost et a magistro Michaele Scoto POSltUS In latIno. ». so
Paris, BN, late 16156 includes the statement «Michael .Scottus super .hbros d~
caelo et mundo et de anima »; see Averrois Cordubensls ~ommentarlum mag
num in Aristotelis De anima, ed. F. S. CRAWFORD, Cambndge, Mass., 1953, p.
xiii.
 IX
244
and .Gundissalinus
. in Toledo between 1152 and 1166 59, as th.& I
I oWIng companson will show:
Long Prologue
2 propria inest perfectio, secundum
id quod preparationes suas dedit eis
recipere de virtute fluente ex in-
tellectu agente
3 et fuit iteratio operationis debili-
tans instrumenta
. Note, in particular, how «praepar~tio », «fluere» and « debi-
l~tare » are .used in the Long Prologue, where the Toledan transla-
tIon ~f ~vIcenna's De anima gives «aptare », «emanare» and
« fatIgan» respectively.
e.1O-
Avicenna, De anima
cf. Avicenna, Y.5, ed. Van Riet , 11,
p. 127.48-50: Cogitationes enim et
considerationes motus sunt aptan-
tes animam ad recipiendum ema-
nationem.
and ibid., V.6, ed. Van Riet, 11, p.
.
150.64-5: quia, cum voluerit , co-
nlUngetur (anima) intelligentiae a
qua emanat in earn forma intellecta.
Avicenna, V.2, ed. Van Riet, 11, p.
97.12: Item quod hoc probat suffi-
cienter, hoc est quod virtutibus ap-
prehendentibus per instrumenta ac-
cidit ex perseverantia operis fati-
gari: instrumenta enim fatigantur
ex iugi motu et destruitur eorum
complexio quae est eis substantialis
et naturalis.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 245
commentaries, only that on De caelo is unambiguously attributed
to Michael. Yet those on the De anima, Physica and Metaphysics
also seem to have been made in the late 1220s and in the same
milieu 60; this is likely to be the Sicilian kingdom of Frederick 11,
where William of Luna (at Naples) was translating the Middle
Commentaries on the Isagoge, Categories, and De interpretatione
at the same time 61. The association of the texts in manuscripts and
the consistency in vocabulary would also suggest a single transla-
tor, and the attribution of the De anima commentary in one manu-
script might suggest that this is Michael. But the same vocabulary
is used in the Long Prologue 62, and, as we have seen, The~dore
himself is given the credit of translating one prologue to the Great
Commentary on the Physics; his is, in fact, the only name associated
with the translation of any part of that work 63. One must consider
the possibility, therefore, that Theodore at the very least was part
of a team which, aside from translating Averroes's works, was
engaged in completing a translation of the Nicomachean Ethics 64
and the Rhetorica. Theodore would have had excellent qualifica-
tions for joining this team, because of his knowledge of Arabic and
his training in philosophy.
But that he might not have been capable of producing these
translations on his own can be seen from the insecurity in the Latin
of the Schools that the Long Prologue betrays. The terminology
may be that of the Latin versions of Averroes, but the syntax and
60 R. A. GAUTHIER, «Notes sur les debuts (1225-40) du premier «Averroisme »,

In inve~tigating Theodore's citations of the Nicomachean Ethics,
the Rhetorzca and the De anima, therefore, one can conclude that
he doe~ not m~ke any obvious use of a translation earlier than
transl~tIons attnbuted to Michael Scot. But these attributions of
Revue des sciences philosophiques et theoiogiques, 66, 1982, pp. 321-74 (333-
4).
61 R HISSETTE, «Guillaume de Luna - Jacob Anatoli - Jacob Mantinus »,
Bulletin de philosophie medievaie, 32, 1990, pp. 142-58 (147).
62 Aside from «perfectio » mentioned above, one could mention « ope ratio »

the Nl~omachean Ethics and Averroes's Great Commentary on the
De anlma, themselves are not firm. Of the translations of Averroes 's
~d9 Ed. S. VAN RIET, Avicenna Latinus, Liber de Anima 2 vols Louvain and
L et en, 1968-72. ' ,
and « ens» which are used in the De anima commentary; compare ed. Crawford,
p. 387 (= Aristotle, De anima, 429a23-4), «non est in actu aliquod entium
antequam intelligat », with James of Venice's translation (apud Albertus Magnus,
ed. C. STROICK, p. 178): « hoc nihil actu horum quae sunt ante intelligere ».
63 This commentary was already known to Grosseteste in 1228-32; see
GAUTHIER, « Les debuts» (n. 52 above).
64 One problem that applies equally to Michael and Theodore is that the
« translatio antiquior » of this work appears to have been made from Greek.
 IX
246
some of the words chosen may indicate someone whose Latin is
influenced by the Arabic vernacular 65. Examples are the paratactic
construction of the sentences and typical Arabic formulae (such as
comparatives in the place of superlatives, and adjectives followed
by genitives), and the wrong use of Latin words: «reliquus» for
« alter» (perhaps Arabic «'akhar») 66 and «victoria» where a
psychological state or a vice would be expected.
In the Long Prologue, therefore, Theodore appears to speak in
his own voice and to use material which he would have known in
Arabic. He may have mentioned the Nicomachean Ethics, Rhetorica
and De anima, because their recent translation into Latin had been
encouraged by Frederick himself; ethics and rhetoric were also
particularly relevant for a ruler. But the similarities between the
passages in the Long Prologue and a text he does not mention,
Avicenna's De anima - might be due to Theodore's training in
philosophy many years earlier, when he studied Ibn SIna alongside
al-FarabI under Kamal aI-DIn ibn Ylinus in Mosul.
Unfortunately, no text on philosophy by Ibn Ylinus has sur-
vived. What still has to be discovered is a parallel to the doctrine
in the Long Prologue that there are two pleasures in each activity,
one coming from the exercise of the skill itself, the other coming
from within the practitioner, by which he obtains relaxation. This
differentiation appears to turn on the practical application of the
art and the theoretical imagining or rehearsing of that art. This
doctrine cannot be found in the De anima, Nicomachean Ethics or
Rhetorica, and since it is only fully explained after the reference
to Aristotle's works, the author too implies that it comes from
another source - or is his own invention. For it is clear that Theodore
6S Much caution should be used in assessing such an influence and differenti-
ating between influences of different vernaculars. Many of the features listed
here and below (p. 281) are found in other Medieval Latin texts, and it is only
the aggregate of such features that might be seen to provide a profile that
matches one vernacular rather than another.
66 A more idiomatic usage is found in the Letter on Regimen: « Una ... altera »
6. Note too that there are some variations in terminology between the Long
Prologue and the Short Prologue: «comestio» 7 and 9 and «commesatio» 3;
« coitus» 9 and « venerea delectatio» 3; «operatio » 3 etc., and « actio» 7.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 247
developed the argument concerning pl~asure. in t?e Long Prologue
specifically to underline his own relahonshlp with t~e Em~eror..
As we have seen, this argument draws on Anstotle s N,-
comachean Ethics Book X. Aristotle had mentioned the pleas~res
of the musician, the' mathematician, the geometer and the architect
- i.e. all people exercising an intellectual skill. The author of t~e
Long Prologue substitutes for these the most imp~rtant o~ders In
Frederick's kingdom: the king, the priest, t~e soldier, t?e Judge -
and the philosopher. The pleasures of the pnest, the soldl~r 'an.d the
judge are described only briefly (if at all); that of the ~Ing IS, of
course the subject of the prologue, as it was the subject .of ~he
Short Prologue. But the philosopher is given first place. HIs. f~rst
pleasure - i.e., that deriving from his art - turns out to be prachslng
demonstrative logic, speaking the truth and showing when .oth:r
people are not speaking the truth 67. His second .pleasure, which ~s
rather that of relaxation, is speculating on phYSICS. But the culmi-
nation of his rational activity is
16 in the management of rulership and in the drawing-up (<< inventio »)
of ·~ivil laws, whose perfecting needs the help of the ruling P?wer. For
philosophy needs a ruling power which its virtuous state befIts.
What could be a more appropriate statement for Frederick's
philosopher to make to his master: and what. could be a more
prominent place to express this senhment than In the preface to a
work on the subject that engaged Frederick's ~ttention. more than
any other? Theodore at the same time emphasises the Importance
of philosophy, and shows how he depends on the supp~rt of the
Emperor. And, moreover, he shows that the role of the phIlosopher
in a successfully governed empire is just as. impo~tant and profes-
sional as that of the priest, judge and soldier - Indeed,. more so.
For we have here an echo of the Platonic ideal of the PhIlosopher-
King.
67 These activities are roughly those t h at one Iearn s from Aristotle's Poste-
rior Analytics, Topics and Sophistici elenchi respectively.
 IX
IX
248
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 249
III fJakim = «fisicus, medicus, sapiens» 72. In Latin sources, the
As the Empe ' h'l scholars that I have come across who are referred to by their con-
role It w rlor s p I osopher, Theodore performed a specific temporaries or call themselves «philosophi » all either originated
. as a ro e not shared by an on 1 .
apparent exception is John of Paler~o :,~:~ I~the court. ~he in an Islamic educational milieu, or are involved with translating
when addressing his Liber quadratorum .onard of Plsa, texts from Arabic into Latin. These include
as «vester philosophus» but in the I'mpto. F1reddenck 11, refers to 1. «Avendauth israelita philosophus », Gundissalinus's collabo-
P I ' ena ocuments John f rator on the translation of Avicenna's De anima 73; it has been sug-
a er~o a~pears as « notarius (et fidelis noster) » in th 0
texts In whIch Th o d . e same con- gested that he was the Jewish scholar, Abraham ibn Dafid, who
The role w d'f: ore IS called «philosophus et fidelis noster» 68. was born and educated in Cordova, and fled to Toledo in about
who' as I erent from that of the Christian, Michael Scot 1148; he was the author of aJ- caqida aJ-rafjcalEmunah Ramah (<< the
IS never called «the E ' . ,
Michael S t 'f h . m~eror s PhIlosopher» 69; instead, sublime faith») 74.
co, I e receIves a tItle beyond that of . 2. Gerard of Cremona, the most prolific Arabic-Latin translator
appears 1 « magIster »
. as «astro ogus» or «astrologus domini Fred " R ' of the twelfth century (d. 1187); he is referred to as «summus
Imperatoris » 70 Theodore's r I encI ome
* had it~ preced~nts and paraI~e~s~~~:::f:r:;~~!~ilOSOPher rather philosophus » in one of the earliest manuscripts to contain a work
of his 75.
It IS clear from Barhebraeus's account that . h'l
(faiJasiif) is equivalent to the Arabic h '.k- ~< P .1 osophus » 3. Philip of Tripoli, the translator of the Secret of Secrets, who
sical Arabic dictionary of Lane is de~i~: ijaklm, In the. Clas- describes himself in the preface to this work as the « philosopher »
modern lan u . ' as «a sage, In the of Guido de Vere, the bishop of Tripoli 76.
and th g age, a phIlosopher, and particularly a physician» 71 What were the characteristics of the .{1akim or philosopher? He
centur; ::~c~~~~ o~ meanings is implied in the mid-thirteenth~ was engaged in secular learning - that of the Ancients, and pre-
In g ossary known as the Vocabulista in Arabico: eminently of the « Philosopher» himself, Aristotle. To this learning
he added the sciences of medicine and astrology, but in both cases,
placed more emphasis on the theoretical aspects - on the books -
68 H.B., V. 2, p. 726: «cum magistro J h d rather than the practical aspects. This learning may be described as
nostro; ibid., 745, «H Abbatem ~ anne e Panormo notario et fideli
nostros ». . et notanum Johannem de Panormo fideles « Hellenistic» and involved the pursuit of knowledge for its own
6~ The nearest we get to this is Leonard f . , '.
begmning of the Liber abbacI' as d . o. PIsa s addressmg Mlchael at the
. « omme ml magist M' h
phtlosophe» (Scritti di Leonardo Pisano d er IC ael Scotte, summe
1857-62, I, p. 1), but here «philoso hus '. e . B. BONCOMPAGNI, 2 vols, Rome, 72 Vocabulista in Arabico, ed. C. SCHIAPARELLI, Florence, 1871, p. 88.
is clear when one compares this !ith ;~~s ~ term of approbation, not a title, as 73 He heads the dedication to his translation of Avicenna's De anima: «Rev-
«reverende pater domine Theodore im ~rd~ Leonard uses for Theodore: erentissimo Toletanae sedis archiepiscopo et Yspaniarum primati Iohanni,
(Testimonium 8 below) Thus Haski '. . penahs aule summe phylosophe» Avendehut israelita philosophus, gratum debitae servitutis obsequium »; AVICENNA,
245-6): «Two of Frederick's. ns IS 10 error whe h . ( . Liber de anima, ed. S. Van Riet, I, p. 103*.
co t' n e wntes ,Studies, pp.
«philosopher » ... Michael Scot uMr lerts sTeehm to have borne the official title of 74 M. T. D'ALVERNY, «Avendauth? », in Homenaje a Millas-VallicrosQ., I,
70 •. ... as er eodore ... ». Barcelona, 1954, pp. 19-43.
The latter tItle appears in the rubri .
Liber introductorius (e.g., the opening of th c~ ~ the v~nous. parts of Michael's 7S See the colophon to Aristotle's Meteora in Oxford, Bodleian, Selden supra

Studies, p. 291). As «astrolo us» he a e I er partlcuiarzs printed in HASKINs, 24 (late 12th century), fol. 114r: «completus est liber Metheororum cuius tres
on fol. 206r of MS Vienna 3~24 ( lpears as the. author of a text beginning primos libros transtulit magister Gerardus Lombardus, summus philosophus ».
p.114). , s e e URNETr,« Mlchael Scot» [no 25 above], 76 MS Marburg, Universitatsbibliothek, 9, fol. 65r: « Domino suo excellentissi-
71 E W L .
. . ANE, An Arabic-English Lexicon, 1863 (reprinted 1968) s
mo ... Guidoni venerabili de Valencia civitatis Tripol. gloriose pontifici philoso- *
, .v. phus suo rum minimus clericorum ... »
 IX
250
sake 77. In the Islamic world secular learning did have a place in
the madrasa (a school attached to a mosque), as can be seen in the
case of Ibn Yiinus, who made the madrasa of the Zain aI-Din
mosque in Mosul a kind of college. However, it was not as well
~ntegrated i?to an institutional context as were the artes (which
Included phIlosophy) in Western schools and universities. There-
fore, the pakfm was more 1i~ely to find support under the direct
patronage of a ruler whom he served on a personal basis, rather
than within a university or a religious order, as he might have done
in the West. He was expected to accompany the ruler, and to ad-
vise hi~ on medical a~d astrological matters. The ruler might prize
t~e phIlosopher for his learning, and jealously guard him, or use
him as a valuable trading commodity. This is what we find in the
cas~s of 'Umar Suhrawardl (1145-1234) who served the Caliph al-
Na~lr (1180-1225) and 'Abd al-La!lfal-Baghdadl (1162-1231) who
was at the court of Bahramshah in Erzinjan. Na~lr ai-Din al-Tiisl
was successively the pakfm to two Assassin rulers and the~ to
Hiilegii, the Mongol Khan. Hiilegii would not travel without Tiisl'S
sanctio?,. and he. ma~e him a wazii: Such a philosopher probably
was Phlhp of Tnpoh to Bishop Guido de Vere. Such a philosopher
certainly was Theodore. As we have seen, he served three separate
rulers, and his case may be regarded as typical.
. He i~ attested as accompanying the Emperor on his campaigns
In Bre~cla and Padua (Testimonia 1 and 2). On the latter he gave
Fredenck 11 some astrological advice; his Letter on Regimen advices
his master, in a rather peremptory tone, what he should and should
no~ do to keep him in good health. Above all, he practised secular
phIlosophy. The most significant indication of this is his confron-
tation with the the Dominican Inquisitor, Roland of Cremona. The
account ~Testimonium 1) makes it sound as if secular philosophy
was considered to be a threat to the church just like another heresy;
Roland belongs to the group of highly educated Dominicans who
sought greater knowledge of Aristotle and his Arabic commenta-
77 See RAGEP, Na~Ir aI-Din al-Tiisi's Memoir on Astronomy (n 19 above) pp
4,8-9. . ,.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 251
tors to root out heresy using its own weapons 78. Theodore was
greatly valued by Frederick 11, who was loath to part with him. But
Theodore does not seem to have returned the affection. Both the
story in the prologue to The Book of Sidrach and Barhebraeus's
account tell of his attempt to leave the Emperor's presence stealthily.
In the latter account he kills himself, using the poison which, in
his capacity as a doctor, he carried with him.
Whether the ideal of the co-operation of philosopher and ruler
in governing the country and drawing up the laws, as set out in the
Long Prologue to the Moamin, represents something special to
Frederick's court, remains to be investigated 79, as does the evi-
dence that the ideal was put into practice in the Empire. The preface
to Frederick's law code, the Liber Augustalis, does have a
philosophical tone and shows some similarity in concepts to the
Long Prologue 80. But since it was written in 1231, it could predate
78 This is the argument of O. O. STROUMSA, in «Anti-Cathar Polemics and the
Liber de duobus principiis », in Religionsgespriiche im Mittelalter, ed. B. LEWIS
and F. NlEwOHNER, Wolfenbiitteler Mittelalter-Studien 4, Wiesbaden, 1992, pp.
169-83, who discusses the use of new translations from Arabic and Hebrew in
combatting the Cathar heresy. Roland had a reputation for his knowledge of
philosophy, which is supported by the wide range of his quotations from the
philosophers (especially Avicenna) in his writings; in the early 1230s he preached
against heresy in Toulouse, and later in the same decade he was appointed by
Pope Oregory IX to be Inquisitor for Northern Italy; see E. FILTHAUT, Roland
von Cremona O. P., Vechta i. 0., 1936, p. 11 (<< Rolandus ... cuius fama celebris
et excellens in philosophicis habebatur »), and'pp. 23-5.
79 Such an investigation should take in both the Western and the Arabic
tradition of the ideal state being ruled by a Philosopher-King, or at least by a
king and a philosopher in close co-operation.
80 H.B., IV.l, p. 3: «Post mundi machinam Providentia divina formatam et
primordial em materiam nature melioris [conditionis] officio in rerum effigies
distributam, qui facienda previderat facta considerans et considerata commendans,
a globo circuli lunaris inferius hominem creaturarum dignissimam creaturam ad
imaginem propriam effigiemque formatam, quem paulo minus minuerat ab
Angelis, consilio perpenso disposuit preponere ceteris creaturis ».
Here we have Providence (<< Providentia » for Theodore' s « Sollicitudo »), busy
Nature (<< nature melioris officio »; «1 sollicitudo naturae gubernans omnia»),
the realm below the Moon (<< a globo circuli lunaris inferius »; «1 quae intus
speram Lunae content a sunt »), and the dignity of man (<< hominem creaturarum
dignissimam creaturam »; « 6 homo est speculatio entium »).
IX
252
Theodore's arrival in Frederick's court, and it owes its language to
the notary Pier della Vigna.
That Frederick should have his personal philosopher, after the
model of Islamic rulers and other potentates of the Middle East is
not surprising when one considers what close intellectual contacts
he had with these rulers. We have already seen how he sent letters
to the Ayyubid Sultan, al-Malik al-Kamil, and perhaps, directly to
Kamal aI-Din ibn Yfmus, concerning medicine, philosophy and
mathematics. Seven optical questions that he sent to al-Malik al-
Kamil were passed on to Shihab aI-DIn AJ:.unad Ibn Idrisi al-Qarafi
(d. 1283-6) whose answers are extant 81. Frederick used the occa-
sion of the negotiations with the Egyptian Sultan over the transfer
of Jerusalem in 1229 to pose queries on difficult philosophical,
geometric and mathematical points, which the Sultan passed on to
Shaikh 'Alam aI-DIn Qaisar and other scholars at his court 82. Ac-
cording to the Arabic chronicles Frederick continued to correspond
. with al-Malik al-Kamil and his two successors. The last of these,
al-Malik al-$aliQ, sent Siraj aI-DIn UrmawI to him, and this phi-
losopher spent some time as the Emperor's honoured guest and
wrote a book on logic for him 83. The same Siraj aI-DIn UrmawI
(d. 1283) had been a student in Mosul, and later went to the court
of the Sultan of Konya, where Theodore had been 84. Two Arabic
letters of Frederick 11 are extant, though they do not deal with
philosophy 85. The story of Frederick's relations with the Emir of
81 E. WIEDEMANN, «Fragen aus dem Gebiet der Naturwissenschaften, gestellt
von Friedrich 11, dem Hohenstaufen », Archiv fUr Kulturgeschichte, 11, 1914,
pp. 483-5; Wiedemann gives translations of the questions posed by the Emperor,
which are found in an optical work of al-Qarafi.
82 F. GABRIELI, Arab Historians of the Crusades, translated by E. J. Costello,
London, 1969,p.270.
83 Ibid., pp. 276-80. Several logical texts of his are still extant, and were very
popular in the thirteenth century and afterwards; see BROCKELMANN, Supplement-
band, I, p. 848 and D. GUTAS, «Aspects of Literary Form and Genre in Arabic
Logical Works », in Glosses and Commentaries on Aristotelian Logical Texts,
ed. C. BURNETT, London, 1993, pp. 29-76 (61).
84 CAHEN, Pre-Ottoman Turkey, pp. 254-5.
8S GABRIELI, Arab Historians, pp. 280-3. The letters were written on 23 Au-
gust, 1229.
; IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 253
Tunis in the preface to the Book of Sidrach (Testimonium 12),
though legendary, could be paralleled in fact. The most famous
correspondence, however, is that which is detailed in the preface
to Ibn Sab'In's Sicilian Questions (written between 1238 and 1242).
This tells us that the Emperor sent a questionnaire of four ques-
tions concerning metaphysics and the soul to the Middle East,
Egypt, Syria, Iraq, Dariib and the Yemen. When he failed to get. a
satisfactory reply from these sources he then sent envoys to Tunis,
with a similar lack of success. Eventually he applied to the Almohad
caliph, Abu Mu1).ammad al-Rashld (1232-42), who passed the ques-
tions on to the governor of Ceuta, telling him to search for the man
he regarded as the most intelligent philosopher in his kingdom.
This was Abu Mu1).ammad 'Abd al-I:Iaqq ibn Sab'In (1217-69/71),
a Peripatetic and a Sufi, who used in his answers Aristotle's Meta-
physics, Averroes's Great Commentary on the De anima and Alex-
ander of Aphrodisias. These answers survive as the Sicilian Ques-
tions , which include the full text of the questions as posed by
Frederick 86.
This extensive correspondence must be seen as complementary
to the cultural exchange that was going on within Frederick's court
and the Two Sicilies, where works of Aristotle, Avicenna and
Averroes were being translated from Arabic into Latin and Hebrew,
where Maimonides' Guide to the Perplexed was known, and
probably translated into Latin, and where Jacob Anatoli, Michael
Scot and the Emperor engaged in philosophical discussions 87.
Moreover, the liveliness of this philosophical atmosphere,. which
provided the universities of the West with many of the ba.slc ~exts
for their Arts Faculties, must be seen in the context of the hvehness
of debate and the simultaneous existence of several great Hellenistic
scholars in the Mediterranean basin as a whole. For in the East,
following the tradition of Fakhr aI-DIn al-RazI (d. 1209), were
Kamal aI-DIn ibn Yfmus (d. 1242), Athlr aI-DIn al-AbharI (d. 1265),
86 M AMARI translates the preface in «Questions philosophiques adressees
aux savants musulmans par l'empereur Frederic 11 », Journal Asiatique, ~th
series, I, pp. 240-59. The Arabic text is edited by S. YALTKAYA, Ibn Sab ID,
Correspondance philosophique, Paris, 1943 (with a preface by H. CORBIN).
87 See SIRAT, « Les traducteurs juifs » (n. 31 above).
IX
254
Na~ir ai-Din al-Tusi (d. 1274), Siraj ai-Din Urmawi (d. 1283) and
Barhebraeus (d. 1286). These scholars brought the Hellenistic tra-
dition to its culmination 88, and Avicenna was particularly favoured
among them 89, while in the West Ibn Sabcin was renowned as the
« last of the Peripatetics », and was acquainted with the works of
Averroes 90. Although Theodore of Antioch could not aspire to the
fame of these scholars, his career is comparable to theirs, and
shows how political and linguistic boundaries did not prevent ideas
from travelling, and how the development of Hellenistic scholar-
ship in the Mediterranean in Latin and Arabic in the mid-thirteenth
century can be considered as a single intellectual movement 91.
88 Thirteenth-century Islamic Hellenism seems to be a field still waiting to be
explored. Dimitri Gutas has pointed out the importance of the original works
and commentaries on logic by these authors in «Aspects of Literary Form» (n.
83 above), pp. 61-2. Their works remained standard for several centuries after-
wards and represent the last flowering of Hellenism in Islam. Important hints on
the significance of their work, and on the readiness of scholars to travel in seach
of the best teachers, are given in Ragep, Na~lr al-Din al-Tiisi's Memoir on
Astronomy (n. 19 above), pp. 8 and 9.
89 GUTAS, «Aspects of Literary Form », pp. 56-62; Ibn Yunus, as we have
seen, lectured on Avicenna, and Barhebraeus translated two texts of Avicenna
into Syriac.
90 Encyclopedia of Islam, 2nd edition, s.v. Ibn Sabcin. Ibn Sabcin traced his
spiritual descent from Hermes, Socrates and Plato; CORBIN, apud Yaltkaya (n. 86
above), p. vi.
91 One should also, of course, include Hebrew (as manifested in the activities
in the court of Frederick 11 and elsewhere) and Greek (especially considering the
lively intellectual climate of the kingdom of Nicaea), but a discussion of this
belongs to another study.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 255
TESTIMONIAl
H. B. = J. L. A. Huillard-Breholles, Historia diplomatica Friderici
Secundi, 6 vols, Paris, 1859-6l.
1. Autumn, 1238. Humbert de Romans, apud E. Filthaut, Roland
von Cremona und die Anfiinge der Scholastik im Predigerorden,
Vechta i. 0., 1936, p. 27.
Fr. Rolandus, vir magnus in philosophicis et theologia, cum semel
existens Cremonae, et a fratribus quibusdam venientibus de exercitu
domini Frederici, quem tenebat tunc ante Brixiam, audisset quod
philosophus ejus multumconfudisset eos de. sua p~i~oso~h~a de q~a
nescierant respondere, succensus zelo honons ordiniS, diXit: Statim
sternite mihi asinum. Erat enim podagricus, et pedes ire non poterat.
Quod cum factum fuisset, intrans in exercitum super asinum cum
quibusdam fratribus, incoepit quaerere ubi esset ille philosophus. Quod
audientes multi magni qui eum noverant et honorabant, convenerunt, et
convocato illo philosopho, dixit: Ut scias tu, magister T?e.odore, q.u~d
ordo Fratrum Praedicatorum habet philosophos, ecce do ttbl coram ISttS
optionem ut vel opponas de quacumque p.hil.osop~ia v~lueris,. vel
respondeas mihi. Qui cum elegisset opponere In Il~a dlsput~ttone,. s~c de
eo gloriose triumphavit quod ad magnam glonam cesstt ordiniS et
honorem, et habuit exinde eum philosophus ille in magno honore.
Father Roland, a great man in the philosophical subjects 2 and
theology, once at Cremona heard from certain brothers coming f~om the
army which Lord Frederick was then keeping ~utsi?e B~escla, that
(Frederick's) philosopher had confounded them With hiS phIlosophy, ~o
which they did not know how to reply. Fired with a zeal to defend hiS
Order, he said: « Prepare an ass for me at once» ! (For he suffered from
gout and was not able to walk). When this had been done, he went on
1 I am very grateful to Jeremy Johns for advice concerning the place names
and the technical terms in the documents printed below.
2 I.e., perhaps, the «philosophia prima, philosophia moralis» and
« philosophia naturalis », of the university arts faculty.
 IX
256
his ass into the midst of the army with some of the brothers and began
to enquire where that philosopher was. When they heard this, many high
people who knew and respected him, formed a gathering. He then said
to the philosopher, who had been called: « So that you may know, Master
Theodore, that the Order of Preachers has its philosophers too, I offer
to you before all these present the option of either putting forward
whatever philosophical question you wish, or replying to mine ». When
he chose to take the lead in the disputation, he (Roland) defeated him
in such a glorious way that it redounded to the great glory and honour
of the Order, and that philosopher held him (Roland) in great honour
from that day forth.
2. 1 May, 1239, Rolandino of Padua, apud Muratori, VIII, 228
(new edition, VIII. 1, 66).
Et cum carrocio paduano, circa finem eiusdem mensis, duxit
exercitum ipsum ad Castrum franchum, locum Tarvisinorum. Et horam
mocionis elegit per consilium magistri Theodori, sui astrologi, qui stetit
cum astrolabio sursum in turri communis expectans, ut dicebatur, quod
ascenderet prima facies vel horoscopus Leonis, cum diceret Iovem esse
in illo. Set cum per astrolabium hoc videre non posset, tempore nubibus
obumbrato, si licit urn est dicere, tunc fuit in sua electione deceptus, quia
nec erat Iupiter in Leone, nec Leo tunc ascendebat, set Virgo. Et ita,
cum Scorpio tunc esset domus itineris, quia tercia, innuebat quod exercitus
vel imperator, ad cuius peticionem fiebat, offendi deberet in fine; quasi
cauda Scorpionis hoc faceret, que dicitur infidelis et venenata.
At about the end of the same month with the Paduan carriage 3 he
lead the army itself to Castelfranco in Treviso. He elected the time of
his withdrawal following the advice of Master Theodore, his astrologer,
who stood with an astrolabe on top of the city tower expecting, as was
s~id, that the first decan of Leo would ascend (or be the horoscope),
SlDce he said that Jupiter was in it. But since he could not see this by
means of an astrolabe, the time being obscured by clouds (if one can say
that), he was on that occasion deceived in his election 4, because Jupiter
was not in Leo, nor was Leo ascending at that time, but rather Virgo.
Thus, since Scorpio was at that time the house of journeys, i.e., the third
3 The carriage to which the standards were fixed.
4 «Elegere» and «electio» refer to catarchic astrology (the branch of as-
trology which enables you to choose the right time to start an activity), of which
the textbooks in Latin are entitled De electionibus.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 257
house 5, it hinted that the army or emperor on whose request (the elec-
tion) was made would ultimately meet an obstacle, as if this were caused
by the sting of the scorpion which is called faithless and poisonous.
3. 15 Dec. 1239. Frederick II's instruction to Henry the abbot,
concerning sending an imperial boat to bring Theodore back
home (H. B., V. 2, p. 556).
... Cum magister Theodorus filosophus fidelis noster in regnum de
nostra licentia revertatur, fidelitati tue precipiendo mandamus quatenus
vackettam Iscle coopertam curie nostre que nuper venit cum galeis, et
est apud Pisas, eidem pro reditu suo statim bene paratam et munitam
debeas assignare.
Since Master Theodore the Philosopher, our subject, is returning to
our kingdom with our permission, may I instruct your honour and bid
you to assign to him immediately the covered boat of Is~hia b~long~ng
to our court, which recently arrived with the galleys and IS outsIde Plsa,
having prepared and furnished it well for his return.
4. 10 Feb. 1240. Theodore is instructed to write a letter in Arabic
to the Emir of Tunis (H. B., V. 2, p. 727).
Facte sunt littere per G. de Tocco ad magistrum Theodorum ut scri-
bat regi in littera arabica super negotio supradicto et missa est ei carta
sigillata et non scripta [sigillo].
The letter has been written by G. de Tocco to Master Theodore
(instructing him) to write in Arabic about the above-mentioned business,
and a sheet of parchment has been sent to him, with a seal but without
anything written on it.
5. 10 Feb. 1240. The instructions in detail (H. B., V. 2, p. 745).
De mandato facto per magistrum P. de Vinea scripsit G. de Tocco:
Fredericus, etc., magistro Theodoro, etc. Ecce mittimus ad regem Tunisi
R Abbatem et notarium Johannem de Panormo fideles nostros vel si
forte idem nota~ius Johannes adversa valetudine prepeditus non posset,
R. de Amicis justitiarius Sicilie ultra flumen Salsum fidelis noster
S I.e., the third division of 30 degrees into which the eclipt~c circle is ~i­
vided (starting from the ascendant point) at the time of the « election »; the thud
house indicates brothers and short journeys.

IX
258
ordinabit .alium juxta quod sibi litteris nostris mandavimus, mittendum
cum pred!cto H. quem cognoverit opportunum. Cum igitur ipsos nuntios
nostros htteras habere deceat ad eumdem regem, mittimus discretioni
tue cartam sigillatam et non scriptam, mandantes ut in lingua arabica ex
part~ nos~ra scribas eidem regi qualiter mittimus ad eum predictos duos
nuntIos fldeles nostros per quos super hiis que inter nos et eumdem
regem tractanda sunt plene sibi nostram patefacimus voluntatem· unde
cred~.t eis secure in hiis que ex parte nostri culminis dixerint ta~quam
nuntns ad eum per nos tram celsitudinem destinatis. Intellecto vero negotio
~er H. Abbatem pro quo ipsos mittimus, conformes te negotio et formes
htteras secundum quod honori nostro et qualitati negotii videris expedire.
. On the ins!ruction of Master Pier della Vigna, G. de Tocco has
wntten: Fredenck, etc., to Master Theodore, etc. We are sending to the
king of Tunis Henry the Abbot and the notary John of Palermo our
subjects, or, if by chance John the notary cannot go because of bad
health, R. de Amicis, the justiCiary of Sicily beyond the Salso our
subject, will instruct some other person that he thinks suitable to b: sent
~it~ Henry, according to what we have ordered in our letter. Since it is
fItting tha~ these envoys of ours should take a letter to this king, we send
to !our WIS~ s~lf a sh.eet of parchment with a seal but without anything
~ntten on ~t, Instructing that you should write to the king in our name
~n the ArabIc language, that we are sending these two envoys, our sub-
Ject~, through ~hom .we can fully reveal our wishes concerning the
subJect~ of the dIScussIons between us and the king. As a result of (what
y~u wnte) h~ should have complete faith in the envoys sent by our
HIghness to hIm concerning what they say in the name of our Loftiness.
Whe? Henry the Abbot has explained to you the matter on which I am
sendmg them, you should fit yourself to the matter and construct the
letter according to what you think befits our Honour and the nature of
the matter.
6. 12 Feb. 1240. Frederick 11 asks Richard de Pulcaro to provide
the necessary supplies for Theodore to prepare syrups and vio-
let-sugar for himself and his court (H. B., V. 2, pp. 750-1).
. M~ndante domino imperatore per magistrum Roggerium de Camera
scnpslt R. de Salerno: Fredericus, etc., Riccardo de Pulcaro, etc. Ecce
quod magistro Theodoro philosopho et fideli nostro damus nostris litteris
in ~andatis ut de sciropis et zuccaro violaceo tarn ad opus nostrum
f~clat. quam ad. opus came re nostre in ea quantitate sicut sua circumspectio
VI dent expedue. Propter quod fidelitati tue precipiendo mandamus
quatenus ad requisitionem dicti philosophi nostri zuccarum et omnia
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER
259
alia necessaria pro zuccaro et sciropis predictis faciendis studeas sine
defectu quolibet exhibere eidem de pecunia curie nostre que est per
manus tuas.
On the order of the Emperor through Master Roger de Camera, R.
de Salerno has written: Frederick, etc. to Richard de Pulcaro, etc. Note
that we are giving to Master Theodore the Philosopher, our subject,
instructions by letter that he should make syrups and violet-sugar both
for our requirements and for those of our chamber, in whatever quantity
his discretion should think appropriate. Because of this we order our
faithful subject to make every effort to provide for him the sugar and all
the other necessities for the (violet) sugar and syrups, using the money
from our court that is in your hands.
7. 12 Feb. 1240. The instructions to Theodore himself (loc. cit.).
Item eodem die de eodem mandato scripsit idem magistro Theodoro
ut sciropos et zuccarum violaceum tarn ad opus domini quam et pro
camera faciat in ea quantitate quam viderit expedire et expensas recipiat
a Riccardo de Pulcaro et significet domino statum suum.
On the same day concerning the same instructions the same man
wrote to Master Theodore that he should make syrups and violet-sugar
both for the requirements of our Lord and for those of our chamber, in
whatever quantity he should think appropriate, and that he should re-
ceive his expenses from Richard de Pulcaro, and he should render an
account to his Lord.
8. After 1225. Letter of Leonard of Pisa to Theodore, following the
Flos (ed. B. Boncompagni in Scritti di Leonardo Pisano, 2
vols, Rome, 1857-62, 11, p. 247).
Assiduis rogaminibus et postulationibus a quodam mihi amicissimo
invitatus ut modum sibi componere m solvendi subscriptas avium et
similiu~ questiones; quia ipse tanquam noviter in hoc magisterio
educatus fortiora pabula in libro meo numeri apposita pavescebat, lac
sibi, vel~t noviter genito filio, suavitatis preparans, ut robustior 6 effectus
capere valeat artiora, presentem sibi modum in~eni, ~er quem non s?lum
similes questiones solvuntur, verum et omnes dlver.sItates con~olamlnu~
monetarum. Et quia ipsum in ilIa scientia prestantlorem et utIlem elegI,
6 robustius ed.

IX
260
vobis, reverende pater domine Theodore, imperialis aule summe
phylosophe, mittendum decrevi, ut ipso perlecto, que utilia sunt, vestre
celsitudinis probitas, resecatis superfluis, reconservet.
Invited by the persistent demands and prayers of a certain intimate
friend of mine to compose for him a way of solving the following
problems concerning birds and similar ones, because he himself, as if
newly introduced to this art, feared the stronger food set out in my book
on Number, preparing the milk of sweetness for him, as if for a newborn
born son, so that, having been made stronger, he might be able to un-
derstand more difficult things, I found the present way for him, by
which not only similar problems are solved, but also all the different
kinds of problems involving money. And because I have chosen this
(way) as being more outstanding and useful in that science, I have
decided to send it to" you, reverend father Theodore, supreme philoso-
pher of the imperial court, so that, having read it through, the goodness
of your Highness should"preserve what is useful, having cut out what is
redundant.
9. Leonard of Pisa, Questio mihi proposita a magistro Theodoro
domini imperatoris phylosopho, following the Liber quadrato-
rum (ed. BO!lcompagni, ibid., 11, p. 279):
Volo invenire tres numeros qui insimul aggregati cum quadrato pri-
mi numeri faciant quadratum numerum super quem quadratum, si addatur
quadratus secundi, egrediatur inde quadratus numerus; cum quo quadra-
to addito quadrato tertii similiter quadratus numerus inde proveniat.
I wish to find three numbers which, when added together with the
square of the first number make a square number; from which, if the
square of the second number is added, there results another square
number; from which, when the square of the third number is added,
once again a square number results.
10. 9 March, 1243. A document from Messina mentioning
Theodore's property (D. Ciccarelli, I1 Tabulario di S. Maria di
Malfino, I, Messina, Biblioteca dell' Archivio Storico Messine-
se, 11, Messina, 1986, 1, no. 13, pp. 29-30).
In nomine Domini Amen. Anno eiusdem incarnacionis millesimo
ducentesimo quadragesimo secundo, nono die mensis martii prima
indictione, imperante domino nostro Friderico Dei gratia excellentis-
simo Romanorum imperatore semper augusto, Jerusalem et Sicilie rege,
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 261
anno eius imperii vicesimo tercio, regni Ierusalem octavo decimo et
regni Sicilie quadragesimo quinto. Feliciter, amen. Coram. nobis n~bili
magistro Guidone de Columpnulis iudice Messane, notano et testIbus
subnotatis ad hoc specialiter rogatis Phevronia venerabilis abbatissa
monasterii Sancti lohannis de Prodromo monialium grecarum Messane
et conventus eiusdem sponte dederunt et concesserunt in cabellam
Russomanno de Milacio civi Messane totam et integram vine am
monasterii eorum sitam in territorio Messane in contrata dicta de Paleariis,
iuxta vineam archiepiscopatus Messane et secus vineam domini Theodori
imperialis philosophi et secus flomariam de Cam~ariis ~inc usque a.d
annos septem completos scilicet septem completas vIndemIas pro tarenIS
trecentis auri ad generale pondus Regni...
... the whole vineyard of their monastery situated in the territory of
Messina in the district of Paleariae 7, next to the vineyard of the
archbish~pric of Messina, and following (the boundary of) the vineyard
of Theodore, the imperial philosopher, and following (the boundary of)
the channel 8 of the Cammarii. ..
11. November, 1250. Frederick gives the property of the late
Theodore, his Philosopher to the nephew of his Stablemaster
(F. Schneider, «NeueDokumente vornehmlich aus Siiditalien »,
Quellen und Forschungen aus italienischen Archiven und
Bibliotheken, 16, 1914, pp. 1-54 [51-2]).
Fr. Dei gratia Romanorum imperator semper augustus., lerusalem ~t
Sicilie rex. Augetur erga excellentiam nostram nostrorum fIdes et devotIo
subiectorum, cum benemeritis fidelibus nostris congruis beneficiorum
largitionibus providemus. Per presens igitur ~rivilegium not~m fieri
volumus universis fidelibus nostris tarn presentIbus quam futuns, quod
nos attendentes puram fidem et devotionem sinceram, quam Fulco Rufus,
nepos Petri de Calabria marescalle nostre magistri, fidelis" noster, e.r~a
excellentiam nostram habet, considerantes quoque grata et accepta servIcIa
7 «Pagliara» is a common place name, meaning « haystack » ~n Ital!an, but,
more usually a rude, thatched hut in the Sicilian dialect. There IS a vIllage .of
Pagliara about 20 km south of Messina. The place name also appear~ as a famIly
name, e.g., of Gualterius de Paleariis, and of Gentilis de Paleans, count of
Monopello in the Abruzzi. .. .
8 «Flomaria» or «flumaria» is a channel or canal, usually artlflclal, used
for irrigation or milling, etc.

IX
262
que culmini nostro semper exhibuit et exhibere maiora poterit in futurum,
terram sancte Cristine et casale Prancanice, que fuerunt quondam magistri
Theodori philosofi, ad manus curie nostre rationabiliter devoluta cum
hominibus, iusticiis, pertinentiis et omnibus rationibus suis, prout ea
dictus magister Theodorus plenius et melius tenuit, quoad vixit, videli-
cet que de demanio in demanium et que de servicio in servicium, eidem
Fulconi et heredibus suis in perpetuum de gratia nostra et ex certa scientia
duximus concedenda, ita tarn en, quod tarn ipse quam heredes ipsius ea
a nobis et heredibus nostris teneant inmediate et eciam recognoscant ac
de eis debita et consueta servicia prestare tenentur, viventes inde iure
Francorum, videlicet quod maior natu exclusis aliis minoribus fratribus
et coheredibus succedat in eis, inter eos nullo tempore dividendis. Ad
huius autem nostre concessionis memoriam et stabilem firmitatem presens
privilegium per Nicolaum de Brundusio notarium et fidelem nostrum
scribi et maiestatis nostre sigillo iussimus communiri.
Data Fogie anno dominice incarnationis millesimo ducentesimo quin-
quagesimo, mense novembris none indictionis, imperante domino nostro
Fr. Dei gratia invictissimo Romanorum imperatore semper augusto,
Ierusalem et Sicilie rege, anno imperii eius tricesimo secundo, regni
Ierusalem vicesimo octavo, regni vero Sicilie quinquagesimo primo,
feliciter amen.
Emperor Frederick, by God's grace always august king of Jerusalem
and Sicily: The faith and devotion of our subjects is increased towards
our Excellency when we provide for our well-deserving subjects corre-
sponding largesse of benefits. Therefore, through the present privilege
I wish it to be known to all our subjects, both present and future, that
we, taking note of the pure faith and sincere devotion which Fuleo
Rufus, the nephew of Peter of Calabria, our Stablemaster, and our subject,
holds towards our Excellency, and considering also the welcome and
acceptable services that he has always paid to our Highness, and that he
will be able to show greater ( services) in future, have decided to give
the estate of Santa Cristina and the village 9 of Prancanica 10, which
9 For «terra» as «estate» and «casale» as «village », see H. BRESC, Un
monde mediterraneen. Economie et societe en Sicile 1300-1450, 2 vols, Accade-
mia di scienze, lettere e arti di Palermo!Ecole Fran~aise de Rome, Rome, 1986,
I, pp. 7-13, and ID., «La casa rurale nella Sicilia medievale: massaria, cas ale e
«terra », Archaeologia medievale, 2, 1975, pp. 428--32.
10 No place of this name is mentioned in V. AMICO, Dizionario topografico
della Sicilia, ed. and trans. G. DJ MARzo, 2 vols, Palermo, 1855-6, or in other
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER
263
were once of Master Theodore the Philosopher, having devolved into
the hands of our court justly, with its population, customs, appertena~c~:,
and all rights, just as the aforesaid Master Theodore had held th~m, ~ ~
and well as long as he lived - i.e., whatever belonged to' ~he ~ Impena
ld
demesne belongs to the demesne, and whatever was held In. fief shou
be held ~n fief, to the same Fuleo and his heirs in perpetuity, ...
12. From the preface to The Book of Sidrach (British Library,
Harley MS 1121, s. xiv, fols 1v-2r).
A res un grant tens celi que fust el tens del Emperour Fredric an
n~r de Tunes (Times MS passim) estoit tenu mout sage~ home et
::~gon<d>oit a quanque hom li demandoit. Dunt les messag~rs 1 E~~~rour
s'a~erveillerent mout de si grant science dont ele pout ventre E SI q I~ lur
fist entendant q'en son tresorie aveit un livere que le roy d'Es~a~ne
avoit mande <ses> 12rl ancestres, et les messagers le c~nterent. a ~-
Dunt l'Emperour fust mout talentifs d'aver cel hvere: SI man a
perour. . ur de Tunes priant q'il le mandast le hvere, et le
un messager au seign . t
. de Tunes li manda a dire q'il ly envoiast un clerc qUI seu.
selgn~r. 1 tl'n L'Emperour li manda un frere menour de Palerne qUI
saraZlnOIS e a . . I Ere
out non frere Roger. Celi le translata en latIn et le porta ~ mperou
il out grant ioye de li, si le tint mout cher. En la eo~rt 1 Emper~ur out
ioche qui out a noun Theodre le phtlosophe, qUI mout
un ho m de Ante , '1"1 sa mout
estoit ame del Emperour. Quant il oist parler de cet Ivere, I .pen,. .
coment il le pout aver e tant don a et promist al cham~erleln q tl pnst
l'ensan laire (-arie MS) de lui et le fist priveement escnvere que nul nel
't ~ res un tens Theodre le philosophe le manda en present au
s:~;ila~ch: Obert (Olert MS) de Antioche. Le patriarch~ le usa tote sa
~ie et il avoit un clerk oue li, qui avoit a noun Johan Pleres de Lyouns.
Ce~tui le contre escrit et aloit a l' escole a Tolete e le porta avesque luy,
et ensi revint a Tolete.
Mter a long time he who was the Lord of Tunis in the. time of the
Emperor Frederick was held to be a very wise men and rephed to what-
. kn t J emy Johns G ROHLFs, Dizio-
doc~ments. of th~ 12th-13t~ cel~tu~~~lia °o:~n~al:~ Centro di' St~di Filologici et
nano storlco del cognoml ne a . s Pracanica as a common surname
Linguistici Siciliani, Palermo, 1984, p. 152, g~ve ommon surname in Reggio
. Mil d (p 150) Placamca as a c
at Messma and azzo an d : from one of several villages in the toe of Italy
Calabria; these names may enve
called « Placanica ».
 IX
IX
MASTER THEODORE, FREDERICK n's PHILOSOPHER
265
264
ever question he was asked. Hence the envoys of the Emperor marvelled
much at such great knowledge, wondering where it could come from.
He informed them that there was in his treasury a book that the king of
Spain had sent to his ancestors, and the envoys told this to the Emperor.
Because of this, the Emperor was very eager to have this book and sent
an envoy to the Lord of Tunis beseaching him to give him the book, and
the Lord of Tunis sent a message to him saying that he should send a 14. Ibid., p. 253.
clerk who knew Arabic and Latin. The Emperor sent him a friar minor
of Palermo who was called Roger. He translated it into Latin and brou-
ght it to the Emperor who enjoyed it greatly, and held it very dear. In
the court of the Emperor there was a man of Antioch who had the name
'Theodore the Philosopher', who was much loved by the Emperor. When
he heard people speak of this book he thought very hard how he might
have it, and gave and promised so much to the Chamberlain that he took
the book for him and he got it copied secretly, so that no one found out.
A little later, Theodore the Philosopher gave it as a present to the pa-
triarch Albert of Antioch. The Patriarch used it all his life. He had a
clerk with him who was called John Pier de Lyons. This man transcribed (For the English translation see above, pp. 228-9).
it and went to the School of Toledo, bringing it with him. So it came
back to Toledo.

13. Barhebraeus, Mukhfa~ar fa :Jrikh al-duwal, ed. Antun $alii).anI,

Beirut, 1898 (reprinted 1958), p. 273.

" • 1 ..... 1'1

~fi'
- • ~.
-
~u
'I
-J
_,11 W
", 1'1 f'>-' '" ,', , .
I~' ~~
•• .",. •
'I <:'"U:..'.ll
.,,",
~
'," ;,,;._-)~I.:
-
',. /",
J:'-''->-;
\ _
~_ ••
• ..,

.",J..ul
~.
JLf ~ i--!" . ~J.\
I
\..iII'';-'' w.."
JI .~~
J'.:
J. '-'~j)\ \~ ..,ek·"""'"... ~
-
l- I~. ~L:::..:~ I. .". •

;,,''''''''\:11''1-(~ : 'I "-I-L '-'-" '-''",..,,'., ,i.,.:',~1

1"'_ \ '. . ,
....... UQJ _.,~ • ~')
~----.
,~..\...J..tJ
~.
. L:>-J I.
'-' .
.-' ~~ "-".-,,:,.: - _. '.;
~.

',.1\ JI ~\; 0... ~W

''''''':' . . , /~
~I, ~
ll. ,_
.... _ "1.
~('.\ ~ ,
. . . . aJ
":' ~.
~I .... ~
....
; -i,
--:' 0..> ~

- o~~1
-l..-:.;J -,'~I \~ ~·z~ ~\~,~~
. ..L;J JI ,..J.~\. ~ ~.... \~\ t.. ~ ~.-v,~•..........
.-JI.I
,~j
JI .., E
...k
•
. ~). '<I.J.A:-.t.; ~~• -....,.
,-, . • \ - ' . , /
·,.I..J.!I~;j..;. ':"lhLJ\ ~. ~~,~...:. J -'."

.!J~ 0lS' J.,..-) \....

,.... ) ....i
,
1"1:"':";' ~ ).
I' J , : i J
16- .,!lU\
,
LI .. ...k.:.k...j ~...L>-J
• ..... \ ~
•... ,')'1
J

"" If ~..lC. ~~ 'j~ ~ .J .J.>;-JJ )'~I c.:.... Jl:..t r ~~...:!I..!Jl.. )J~~jIJ

JI ,tit; ~'::4 .j ,j~y' {~ ~~ o~ JI Jl:.;,1 .Jt.. ~~ ~b- ~ ~ . Ulr4
,v- J ";1.1,1 • ~ ~ :11 ~')I\J JI .;1.";' ..Q.,u. ~ ~I :~ .. ~ ~...:ll ~<"I ,jl
L' J \ .:r-'" • J J ~ '."':' \..:-J..r • ~

.... ...L>- ~,.A ~ ;... c: .r-:!I J )L.. J ~~..; ~·~I ..J.j ,jl$"" ~:.....:- ~J .J~.... I

IX
266
TEXTS
In t~e
following editions square brackets indicate redundant
wor~s In the manuscripts; pointed brackets, editorial additions.
Sectlon-numbe~s (usually corresponding to sentences) have been
adde~ for ease In cross-referencing; « ae » has been printed where
the dIphthong is indicated in at least one manuscript. Texts 3 4
and 5 have been established jointly by Baudouin van den Ab~el~
and the author. . Epistola Theodori phi(losoph)i ad imperatorem Fridericum.
1. LE~ER of Theodore to Pier della Vigna, Paris, Bibliotheque
natlonale, lat. 13059, fol. 39r, no. LXXIII printed in H B V.
2 751' , . ., .
, p. .
. Prudentissimo hominum suo tanquam domino domino magistro P de
ymea Theodorus ~hilosophus salutem et in multa commendatione' se
Ipsum. ~ua?tum trI~udium mihi esset si vera de vestro statu desiderabili
e.xpectatlOnlbus mels aliquis nunciaret, nec pagina, puto, capere nec
hngua ref~r~e valer~t, etc. Et ecce in nostri memoriam de zucaro violaco
plenam plxld~m ~lltto .vobis, ut alicui me non tradatis neglectui qui
nollem a vobIs ahquo Intersticio telluris abjungi.
To the most ~ise of me~, as to his Lord, to Lord Pier della Vigna
Theodore t?e phIlosopher gIves his greeting and himself with much
commendatIon. How much jubilation I would have if someone were to
a~nounce true. things < meeting> my expectations concerning your de-
sued state, neIther would a page (I think) contain, nor would a tongue
be ~ble to relate, etc. Behold, so that you do not forget us, I send to you
a. l~ttle box full of violet su~ar, so that you do not consign me to ob-
hVlOn - I who would not wIsh to be separated from you by any space
of land.
2. THE Letter on Regimen.
. The following edition has been made from the only known manu-
SCrIpt, Marburg, Universitatsbibliothek, MS 9, s. xiv in, fols 98v-99r 1.
1 I am very grateful to the librarian, Dr. U. Bredehorn, for promptly sending
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 267
This manuscript comprises two codices. The second one (fols 64-166)
consists entirely of texts associated with Frederick or his son Manfred,
and the two Popes of their time, Gregory IX and Innocent IV. Among
these texts is the Secret of Secrets, the letter from al-Kindi to Theodore
(see p. 236 above), and Adam of Cremona's regimen for the pilgrimage
to the Holy Land (see p. 237 above), the latter two occurring only in this
manuscript. One peculiarity due to the. author or a later scribe is a
confusion of intervocalic « b » and «v» (11 and 34), which is a feature
shared with the Spanish vernacular. The previous edition by Karl Sudhoff
in Archiv fUr Geschichte der Medizin, 9, 1915, pp. 1-9, has been helpful
in interpreting obscure readings in the hastily-written manuscript.
1 Celsitudo vestra precepit ut de conservanda sanitate certas scriberem
vobis regulas, sed citius ad manus vestras venit scriptum antiquissimum
in secretis Aristotelis, quod ad Alexandrum imperatorem per epistolam
inquirentem de sanitate corporis edoceri transmiserat, in quo summatim
quidquid super hoc requiritis, invenitur. Est autem tale:
2 Ut autem salutem conserves et vitam, conserva que mando ut te
ipsa conservent.
<Cibi>
3 Cum comedis, famem depelle, non appetitum, et si plura ingesseris,
plurimum 2 tibi non maneat appetitus. 4 Cibos autem non propter sapores,
sed sapores propter cibos sumas. 5 Et istud serva, ut numquam obruas
appetitum; hoc enim est via 3 salutis. 6 Cum enim in quatuor virtutibus
totum esse servetur - in appetitiva videlicet, retentiva, digestiva et
expulsiva - una servatur ex altera. 7 Expellitur enim bene quod bene
digeritur, bene digeritur quod bene retinetur, et bene retinetur quod bene
appetitur. 8 Bonus igitur appetitus totius boni est initium. 9 Sic econverso
[quod] si quod male appetitur [immittitur]\ male recipitur, male digeritur
me copies of these folios. A full description of the contents of this manuscript is
given in my «An Apocryphal Letter from the Arabic Philosopher al-Kindi to
Theodore, Frederick II's Astrologer, Concerning Gog and Magog, the Enclosed
Nations and the Scourge of the Mongols», Viator, 15, 1984, pp. 151-67 (161-2).
2 « plus» would be expected.
3 Corrected from « vinum ».
4 The correct reading may be « ... si quod immittitur male appetitur ... »

IX
268
et male expell.itur, tunc 5 erit destructio. 10 Sic igitur conservaberis i
modo I sumendl;
.conservaberis
. etiam in modo assumptl· . 11 P rosplcles
.. n
ergo, Ice~ conserva~ens 6 appetitum in modo et modicitate sumendi, ne
destruas Illum per d~a que .sumis, que quidem, si noxia fuerint, licet
pauca, . ~e~t~uent et ~lfU~ qUI~em modicum totum corpus corrum it. 12
A nOXllS IgItur. abslInebls; vltabis cruda, salsa, acumin<os>a c!rupta
dura,. nervosa., Immatura, stiptica, indigesta, indigestibilia, m~lancolica'
colenca, lubnca, cenosa, paludosa. 13 Fructibus ut medi/98vb/camenti~
utere. 14 ~d me~sam regem non voluptas sed voluntas inclinet. 15 Te
reg~s et a~los, SI te conscilium regat. 16 Post cibum deambulabis in
pedlbus t~IS per amena et delectabilia loca. 17 Nec statim ad laborem
nec ad qUletem accedes. 18 Medium am at natura.
<Potus>
~9 Vinum tibi temperabis et te vino, quod non erit acetosum
turbldum, no~um, acerbum, nigrum, grossum, sed digestum, desecatum:
au~eum, od<:>r~ferum et v~tustum. 20 Bibe cum sitis et cum cibo. 21 Si
falI~atus fuens, non statIm bibas. 22 Et si necesse fuerit, vinum aqua
cahda temper~tum .s~~es; sic quando que dum estuabis aut sudore
laxabun!ur pon, a fr~gldl~ potationibus erit continentia neque tunc vento
ex.ponans. 23 AperlIs enlm calore poris ventus intrabit et fiet discrasia
entque causa apostematum et malorum coagulationum.
<Accidentia anime>
24 Tristitiam pellas et iram; hec autem corpus aut animum necant.
<Somnus et viligiae>
. . 25 Sompnus moderatus et vigilia non diuturna. 26 Dormiendum erit
lIbl cum natura ~etit, et vigilandum cum ratio docet, ne cedas vitio. 27
Sed natura paucIs c?ntenta 7 est, vitium vero plurimis. 28 Meridianos
fug: s.o~pnos; .serolInos vero, noctuales et ma[ti]tudinales amplectere
~t SI fl:n potent, num~uam sol te vide at dormientem. 29 Post aliment~
ImmedIate non dor~la~, sed ~umpto cibo post duas horas differas
sompnum, et tunc, SI dIes fuent, unius tantum hore dormitionem as-
5 tc MS.
6 conservaberis MS.
7 contempta MS.
IX
MASTER THEODORE, FREDERICK II' S PHILOSOPHER 269
sumes. 30 Et cave ne post sompnum mox potum assumas. 31 Ieiunus
numquam potes; hoc enim pot ius melancolicis convenit. 32 Nam dum
sit 8 prima digestio, super epar dor/99ra/mias; dum sit secunda, super
splen et iterum sompnus est sumendus. 33 Etiam ratio super hiis
manifestat<ur>, quoniam in stoma,cho cibus ut in olla decoquitur et
mediante calore per ebullitionem purum ab impuro discernitur. 34 Erit
igitur causa: olla super ignem, ut bona fiat decoctio et discretio, facta
decoctione, reponatur olla ab igne et agravetur 9 <quo>que epar, ut
membris distribuat nutrimentum. 35 Iacebit igitur super sinistram et
exinde redibit ad dextram, ut membrum carnosum et spongiosum, si
nimis excrasserit, reddat - i.e., decrescat superfluitas - et epati quasi
iam fatigato subministret. 36 Sic erit cum levitate et salute surgendum.
<Potus>
37 Providebis ut, dum fit digestio, a potu maxime frigido abstineas;
minueret enim fervor per quem habet fieri ebullatio et transmutatio. 38
Vinum vincas aqua, non tarn quantitate quam potentia. 39 Quod si bibere
ieiuno stomacho assuescas, cardiace 10 passionis erit causa et paralisilIl
inducet (40 Paralisis est infirmitas per quam amittitur sensus et motus
in membro propter humiditatem ibi collectam), 41 visumque minuet et
oculos efficit rubicundos, cerebrum et mentem tardabit.
<Coitus>
42 A coy tu abstinebis nisi cum 11 natura petierit, et tunc eius neces-
sitati satisfaciendo, non vitio subiacebis, sed necesse erit ad plus semel
in ebdomada. (43 In die tantum semel prandendum est; in mense semel
minuendum ad plus, si natura valuerit; in anno semel purgandum.) 44
Moderatus taliter coytus moderate calefacet corpus, superfluitates siccabit,
cerebrum purgabit et renes et totum corpus levius efficiet. /99rb/ 45 At
contra immoderatus immoderate calefacit corpus et desiccat humidum et
infrigidat, datque frigiditati et siccitati dominium, vires enervat, oculis
caliginem prestat multorumque longorum morborum est causa et
destructio vite. 46 Abstinentes vero corpora edificant et raro luxuriantes
8 sitet MS.
9 agrabetur MS.
10 MS above: id est doloris in stomacho.
11 nisi cum] non enim MS (?).
 IX
270
prospici~s senescentes. 47 Erit ergo huius res ieiuno stomacho
p~eno .- I.e.: non continuo post cibum nec ante sompnum sed iam' f:~:
dl~~st1one In epate, postq~am singula membra proprium' sortita fuerint
nu nmentu~. ~8 Tunc enlm est corpus et membra validiora; nam ue
Plus.;ppeht qUIa hoc est proprium. 49 Tempus enim et ratio super ~oc
mam ~sta sunt. 50 Membra enim crossiora videmus ieiunis uam lenis
et fortIOr est homo ieiunus quam satur 51 I ' . d" q P
f . db" . elUnlum IClmus non famem'
ames enlm e. tl.e~ h~mlnes facit et fames est inanitio tam membroru~
lua; st.omachl;. lelUnlum enim stomachi tantum, non membrorum 52
u ~ dlffe.r~nham: satur stomachus ieiunis membris; satur a~tem
mem rum leluno .stomacho. 53 Erit igitur hoc facere cum plena sunt
me.mbra; tu. nc enlm purgari est illud opus nature, CUI'US
reh I est recipe re,
nere, e Igere et repellere quod fuerit expellendum.
Letter of Theodore the Philosopher to the Emperor Frederick
1 Your Highness ordered that I should write down for 0 .
rules for pres~~ing health, but, before that, there came into y o:;~:~~:
: ~:ry ';ld wnhng on the secrets of Aristotle, which he had iransmitted
h~alt: ~p:ro~ Alex.ander. ask~ng through a letter to be taught about the
on this °SU~j:ct~~~;~e7t:~)hiSISs!~~~d put briefly whatever you require
so t;a;Ot:rese~eh your health
and your life, preserve my orders to you
ey mlg t preserve you.
Foods:
3 When ~ou eat, drive away hunger, not appetite, and if you digest
::'~~id ;o~ ~tlld not have an ap~etite any more (reading plus). 4 You
5 Ke : e . o~ nO.t because of Its flavour, but flavour because of food
th' .e~h thIs In mInd, that you should never destroy the appetite' fo;
IS IS e way of health 14. 6 For, since the whole being is preserv~d in
12 Advic t .
. e pu ID a proverbial form is noted in italics. Some arallels from
COdnsAta~tIDe, Pantegni, Practica I (from MS London, British LibraryP Add 22719)
an vlcenna Ca M, d' ' . " ,
Book 1 F
13 '
3 D no? e lcma.e, Vemce, 1507 (reprinted, Hildesheim, 1964)
en , octnna 2, are gIven.
puns !~s:;~ in
2 :heodofre puns on two meanings of «conservare», so here he
14 meamngs 0 «servare».
. ~mpare Canon, c. ? (h2vA): «Oportet... ne aliquis comedat
desldenum ... Et preterea mhIl est deterius quam fastidium ». nisi post
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 271
four virtues - i.e., the appetitive, the retentive, the digestive and the
expulsive - one is preserved as a result of the other. 7 For that is
expelled well which is digested well, that is digested well which is
retained well, and that is retained well f~r which there is a good appe-
tite. 8 A good appetite, then, is the beginning of the whole good. 9 Thus,
, on the other hand, if some (food) is badly « appetited », badly received,
badly digested and badly expelled, then there will be destruction. 10
Thus, then, you will be preserved in the way you take in food. You will
also be preserved in the manner of food that you take in. 11 You should
make sure, then, that, even if you preserve your appetite in the way you
take in food and in your moderation in doing so, you do not destroy the
appetite by those foods that you take in, which, if they are harmful, will
destroy it, even in small quantities, and a little amount of poison corrupts
12. the whole body. 12 You will abstain, then, from harmful things; you will
avoid raw, salty, sour, rotten, tough, sinewy, unripe, astringent, undigested,
undigestible, melancholic, choleric, greasy, ashy and muddy foods. 13
Use fruit as medicines. 14 Not indulgence but the will should guide the
king to his table. 15 You will rule yourself and others, if good advice
rules you 15. 16 After eating you will walk around on foot through pleasant
and delightful places. 17 You will not immediately go to work nor to
rest. 18 Nature loves a medium.
' Drinks:
19 You will temper the wine for yourself, and moderate yourself
with the wine 16. The wine should not be acid, cloudy, new, bitter, dark
and thick, but digested, dry, golden, fragrant and old. 20 Drink when
you are thirsty and with food. 21 If you are tired, you should not drink
immediately. 22 And if necessary, you will take wine tempered with hot
water 17; thus, whenever you are roasting or the pores are opened by
sweat, there will be avoidance of cold drinks and you should not be
exposed to the wind at that time. 23 For, when the pores are opened by
Another pun: on « rex/regere ».
'
15
16 Another pun of exactly the same type: «Tempering wine» presumably
means diluting it with water; «tempering yourself» means being moderate in
partaking of wine. Compare Pantegni, Practica, 1.18 (fo!. 176r): «vinum tem-
peratum bibant ».
17 Compare Canon, c. 8 (h4vb): «post exercitium bibat prius vinum aqua
calida temperatum ».

IX
272
th~ heat, the wind will enter and there will be a bad complexion, and it
will be the cause of spots and bad coagulations.
Accidents of the soul:
24 You should drive away sadness and anger; for these can kill the
body and the mind 18.
Sleep and wakefulness:
25 Sleep (should be) moderate and (one should) not stay awake for
long periods. 26 You should sleep when nature bids, and you should stay
awake when reason tells you to 19, lest you should yield to vice. 27
Nature is content with little, vice with excesses. 28 Avoid sleep at mid-
d.ay; but embrace evening, night-time and morning sleep 20, and, if pos-
sible, never let the Sun see you sleeping 21. 29 After eating you should
not sleep immediately, but, having taken food, you should defer sleep
for two hours, and then, if it is still daylight, you should take a nap of
only one hour. 30 Beware lest after sleep you take drink immediately.
31 You should never drink on an empty stomach; for this rather is the
custom of melancholics. 32 While the first digestion is taking place, you
should sleep on (the side of) the liver; during the second, on (the side
of) the spleen sleep should be taken [a second time] 22. 33 The reason for
this is clear, since the food in the stomach is cooked down as in a
cooking-pot, and through the mediation of heat, the pure is separated
from. the impure by boiling. 34 This will be the cause, then: the pot over
the fue, so that there should be a good cooking down and separating;
when the cooking has been done, the pot is removed from the fire and
the liver is also squeezed, so that nourishment might be distribut~d to
the parts of the body. 35 He will lie, therefore, on his left side, and from
18 ~mparePantegni, Practica, 1.10 (fol. 170v): «Oportet etiam quaedam
~xe~c~tIa non esse. assuefacta, i.e., tristiciam, iram, angustiam, cogitationem,
mVldlam. Hec emm corporis complexionem mutant et calorem naturalem
defectant ».
19 This seems to be only a weak reference to Christian vigils if at all
20 ' •
I have not found a parallel for these three periods of sleep.
:: Compare Canon, c. 9 (h5vb): «in die. quoque dormire est malum».
Compare Canon, c. 9 (h5vb): « ... melIor est ut super dextrum incipiat latum
et deinde ad sinistrum revolvatur. Quod si eius inceptio super ventrem fiat
auxilium fert satis magnum in digerendo propterea quod calorem reunet naturale~
et ipsum comprehendit quare augetur ».
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 273
that position he will return to his right side, so that the fleshy and
spongy part, if it has become too fat, may return - i.e., its exce~s may
diminish - and it may serve the liver which is by now tired (and In need
of help). 36 Thus getting up will be performed with (a feeling of) light-
ness and health.
Drinks 23:
37 You will ensure that, while digestion is taking place, you abstain
from drinking - especially cold drink; for the heat through which the
boiling and transforming (of the food) has its being, wou~d be di~inishe?
38 You should subdue wine with water, not so much In quantity as In
power. 39 But if you should become accustomed to drinking on an
empty stomach 24, that will be a cause of disease of the stoma~h 2S and
will bring on paralysis (40 Paralysis is a disease through which sense
and movement in a part of the body is lost because of the moisture that
has collected there), 41 and it will diminish sight and make the eyes red,
and will slow down the brain and the mind.
Sexual intercourse:
42 You will abstain from intercourse unless when nature bids, and
then, by satisfying necessity, you will not lie open to vice; but it will be
necessary not more than once a week. (43 One should dine only once
a day; one should be bled not more than once a month, if nature is
strong enough; one should be purged once a year.) 44 Intercourse
moderated in such a way will moderately warm the body, dry up the
superfluities, purge the brain and make the kidneys and the whole body
lighter. 45 On the other hand immoderate intercourse immoderately warms
the body and dries up the moist, and cools, and gives predominance to
the cold and the dry, weakens the strength, brings darkness over the
eyes, and is the cause of many long diseases and the destruct~on of
life 26. 46. Those who abstain, build up their bodies, and you Will see
23 This section should accompany 19-23.
24 Theodore takes up again the argument of 31.
2S «Cardiace passionis» is glossed as «pain in the stom~ch ~). R. J. Long
points out the correspondence of « cardia.» and. «os stomac~l» In «.~lfred ~f
Sareshel's Commentary on the Pseudo-ArIstotelIan De plantls: A CntIcal Edi-
tion», Mediaeval Studies, 47, 1985, pp. 125-67 (132).
26 Compare Pantegni, Practica, 1.16 (fol. 174v): «Decet tamen non multum
IX
274
those who indulge rarely, reach old age. 47 The performance of this
thing 27 will be on an empty stomach, not a full one - i.e., not immediately
after food nor before sleep, but when the digestion in the liver has taken
place, after each part of the body has received its proper nourishment 28.
48 For then the body and its parts are stronger; for it has an appetite for
more, because this is proper to it. 49 The time and the reason for this
are clear. 50 For we see the parts of the body are fatter in hungry people
than in full people, and a man with an empty stomach is stronger than
a full man. 51 I say «having an empty stomach », not «hunger », for
hunger makes men weak and hunger is the emptiness as much of the
parts of the body as of the stomach; «having an empty stomach» is
(emptiness) of the stomach only, not of the parts of the body. 52 Hear
the difference: full stomach with empty limbs; full limbs with an empty
stomach. 53 To do this, therefore, will be when the limbs are full; for
then to be purged is the need of nature, to whom belongs receiving,
retaining, choosing and rejecting what has to be expelled.
3. SHORT PROLOGUE to M oamin.
Edited from Paris, Bibliotheque nationale, 7019 (s. xiv 2) = A, and
Bologna, Biblioteca universitaria 153 (s. xiv) = O. Some readings from
Frederick 11, De arte venandi cum avibus, ed. C. A. Willemsen, Leipzig,
1942, and the late 15th-century Italian translation of Iammarco Cinico,
edited by M.D. Glessgen (in press), are provided.
1 Liber Moamyn falconarii de scientia venandi per ayes et
quadrupedes ut solatium habeatur 29.
frequentari (coitum). Vim enim dissolvit, calorem naturalem defectat, pectori,
pulmoni, stomacho et epati nocet, totum corpus infrigidat ac desiccat, inerciam
generat et senectutem, et senium festinat ».
27 This appears to be a euphemism.
28 This is consistent with what is said in 34. Compare Pantegni, Practica, 1.3
(fol. 166v): « Caveant etiam exercitia post cibum quod si fuerit aliqua necessitas
convenit tempus expectari quoad usque cibus digest us omnino de stomacho
descendat, ut facillime [ad] epate digeratur »; and ibid., 1.19 (fol. 176r): « Coitum
vero non multum sed rarissime operentur et post digestio'nem ».
29 Incipit liber Moamyn de curis egritudinum avium rapidarum A, Incipit
liber Moamini falconerii translatus de arabico in latinum per magistrum
Theodorum phisicum domini Federici Romanorum imperatoris. Et correptus est
per ipsum imperatorem tempore obsidionis Faventie (MSS Rome, Biblioteca
IX
MASTER THEODORE, FREDERICK U'S PHILOSOPHER 275
2 Reges pluribus delectationibus gaud~n~.. 3 Aliis enim 3)•• victoria
plus placet, aliis regimen populorum pe~. eXlbltlonem ~egu~~ alns plac~!
plus magnorum ope rum constructio, ah~s co'?messatt~, .alns ven~r~~
delectatio aliis thesaurorum congregatto, alns beneflclorum eXlbltto,
aliis ludo;um et festivitatum ordinatio, aliis plus virtutum et scientiarum
acquisitio, aliis venatio, aliis alia. 4 Ex .his 32 omnibu.s v;natio videtur
magis conveniens regibus et magis propna. 5 Fere. enlm. omnes reges
ceterique magnates hanc pre ceteris appetunt, nec l~mento: 6. p~r earn
enim magna tarn corpori quam anime comparatur utthtas. 7 QUIa In aere
puriori vivunt, omnes sensus 34 delectantur, ?mnia memb~a. suas ~eddunt
perfectius actiones, dampnosa v~luptas ~en~ns ceteraq~e ~ltta re.~ltt~n~ur,
tristitia solitudinis 3S, melancohca cogltatto, amor nlmlUS qUI dlcttur
grece hereos 37, immo omnis desperatio 38 per hanc exulant et fugantur.
8 Habet namque ambulationem, equitationem suavem, fortem 39 et
mediocrem sed vicissim. 9 Et exercitium tale si leve sit et moderatum,
calorem naturalem confortat, superfluitates expellit, grata quiete mem-
bra recreat, cibi generat appetitum.
10 Venatores vero 40 predam quam venantur acquirunt 41 mediantibus
variis instrumentis que sunt arcus cum torculis 42, baliste et ~rcus cu~
sagittis, mucrones, lancee, venabula, iac~la, pe~ice, .fosse: r4~tta, laquel,
viscus alieque decipule, canes, leopardl, gattl malmonlS . 11 It.em,
lutrie 44 furce 4S bidentes et tridentes hami, sagene, rete cum qUlbus
venam~r pisces ~. 12 Aves etiam viventes de rapina sunt nobilius et
Angelica, 1461; New Haven, Beinecke Library, 103; Milan, Biblioteca Trivul-
ziana, 695 and a private collection).
30 A omits.
31 veneris A.
32 A omits.
33 autem O.
34 0 adds « et ».
3S sollicitudinis A, sollicitudine Italian.
36 et A.
37 herteos O.
38 despectatio A.
39 fortiter O.
40 quoque O.
41 A omits.
42 circulis A, torcule Italian.
43 maimone A.
44 luteie 0, lutrie De arte venandi.
4S furecti De arte venandi (= « ferrets»), furche Italian (= « forks»).
46 pudes A.

IX
276
mirabilius instrumentum venandi; per eas enim alacritatem, gaudium et
delitias habet homo.
13 Hoc considerans, Moamin 47 falconarius composuit hunc librum
in arabicum de venatione, et divisit in quatuor tractatus, 14 quem magister
Theodorus philosophus mandato Cesaris transtulit in latinum. 15 Primus
tract~t~s continet theoricam venationis que fit per aves rapidas; secundus
medlclnas occultarum infirmitatum; tertius curas manifestarum
infirmitatum; quartus naturam et medicamen quadrupedum cum quibus
venamur. 16 Explicit prologus 48.
1 The book of Moamyn concerning the science of hunting with birds
and four-footed animals, to achieve solace.
. 2 Kings enjoy very many pleasures. 3 For some victory (in battle)
g~v~s most pleasure, for others the ruling of the people through the
gIvmg of laws, ~or others the construction of great works pleases most,
for others, feastmg, for others the pleasures of love-making, for others,
the heaping-together of riches, for others the giving of benefits, for
others the organisation of games and festivals, for others rather the
acquisition of virtues and sciences, for others, other things. 4 Out of all
these, hunting seems to be most appropriate to kings and most proper.
5 For almost all kings and other magnates seek this before all other
pleasures - and not unworthily: 6 for through it a great usefulness is
obtained for both the body and the soul. 7 Because they live in the purer
air,. all the senses are delighted, all the parts of the body perform their
actIons more perfectly, the damnable indulgence in love and other faults
are sent away, the unhappiness of being alone (or worry), melancholic
thoughts, excessive love (which is called in Greek « hereos ») - in fact
all despair is exiled and put to flight through this. 8 For it includes
walking, and gentle, energetic and moderate horseriding in turn. 9 Such
exercise: ~f it is light and moderate, strengthens the natural heat 1 expels
superflUltIes, recreates the parts of the body with welcome relaxation,
and generates an appetite for food.
10 But hunters acquire the prey which they hunt by using various
instruments, which are crossbows, catapults, and bows and arrows,
swords, lances, harpoons, spears, snares, pits, nets, traps, bird-lime and
other deceits, dogs, cheetahs and lynxes 49. 11 Also, otters, ferrets, two-
47 Moamyn O.
48 Explicit prologus] A omits.
49 « G a tt us malmoDls
. . »
generally means « ape» but since apes are never used
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 277
pronged and three-pronged forks, seines and nets with which we hunt
fish. 12 Birds which also live off prey are the noblest and most remarkable
instrument of hunting; for through them a man has happiness, joy and
delight.
13 Considering this, Moamyn the falconer composed this book in
Arabic concerning hunting, and divided it into four treatises. 14 This,
Master Theodore the Philosopher, on the command of the Caesar, trans-
lated into Latin. 15 The first treatise contains the theory of the hunting
that is done with birds of prey; the second the medicines of hidden
diseases, the third the cures of visible diseases, the fourth, the nature
and medication of the four-footed animals with which we hunt. 16 The
prologue ends.
4. LoNG PROLOGUE to Moamin.
The following edition is based on the three manuscripts that include
the Long Prologue, Valencia, Biblioteca Universitaria, 601 (GC 402) ~s.
xv) =A 1, Vatican, Biblioteca apostolica Vaticana, Reg. lat. ~446 (s. XIV)
= J, and Nantes, Musee Dobree, 19 (s. xv Med) = Z. WhIle there are
relatively few variations between the manuscripts, no manuscript can be
shown to have been copied from any other.
1 Sollicitudo naturae gubernans omnia quae intus speram Lunae
contenta sunt, qua ~ didicerunt sapientes sapientiae regulas et 51 ~uas
provecti ornaverunt operationes, dare intendens omnibus entIbus
naturalibus fine m sue perfectionis breviori et rectiori via - 2 et 52 fuerunt
perfectiones diverse secundum diversitatem specierum; nam unicuique
entium propria inest perfectio, secundum id quod prepara~iones ~u.as
dedit eis recipere de virtute fluente ex intellectu agente, et Imposslbtle
est devenire ad perfectionem propriam sine diversa operatione prodeunte
ex virtutibus inventis in proprio instrumento sui, ut videtur in membris
animalium et plantarum, 3 et fuit iteratio operationis debilitans
instrumenta, ex quorum debilitate debilitas sequitur 53 virtutum et
in hunting and 'linces' are mentioned in this context in De arte venandi (p. 4),
perhaps lynxes are intended here too.
. 50 quam Al, quia J.
51 Alomits.
52 que J.
53 insequitur Al.
IX
278
negligentia [non] complendi 54 operation em vel operationes producentes
ad finem et perfectionem propriam - 4 adiunxit 55 unicuique operationum
corporalium delectationem propriam ante et post operationem illam, ut
allevietur virtutibus labor operandi per studium delectationis 56
(quemadmodum sapiens rex qui dat militibus divitias et honores, et
concedit eis uti delectatione), 5 ut non tantummodo propter hoc eis fiat
laboris oblivio, sed etiam mortis, que, licet sit finis universalis nature,
tamen contraria 57 est nature particulari. 6 Et delectatio hec 58 manifesta
est in operatione animalis, sed manifestissima est in corpore humano 59
et specie sua; nam homo est speculatio entium et ideo 60 vocatur minor
mundus. 7 Verbi gratia: comestio, potatio, auditio, visio et unaqueque
operationis humanae delectationi 61 inest, et ablata delectatione
intermittitur 62 operatio, licet noverit dimittens quod in dimissione operis
illius fiat ultima corruptio, ut videtur in eo qui 63, amiss a delectatione
comedendi, amittit comestionem, et, amiss a delectatione potandi, amittit
et potationem, et sic de ceteris operationibus. 8 Et humana operatio
bipartita est: est enim 64 que dam communis omni speciei, alia propria. 9
Communis est ut comestio, potatio, coitus, superfluitates expellere et
sentire per decem sensus et comprehendere per intellectum, et superbia
et victoria, invidia 65, avaritia et operationes quarum principium est per
naturam et perfectio per voluntatem 66, et cetere operationes. 10 Et
unaqueque istarum coniungitur delectationi et tendit ad finem proprium,
ut in libro De anima, et Nicomachia et in Rethorica declaratum est.
11 Proprie quidem operationes sunt ut philosophare, regnare,
sacerdotizare, militare, iudicare et huiusmodi operationum propriarum
quae inveniuntur in 67 aliquo individuorum 68 humane speciei. 12 Et
54 comprehendi Z.
55 This word is the beginning of a new paragraph in Al JZ.
56 operacionis J.
57 contrarius AIZ.
58 hic J.
59 suo J.
60 imo Z.
61 dilectationem J.
62 interimitur J.
63 quia J.
64 est enim] et enim AIZ.
65 et victoria, invidia] victoria J.
66 AIZ add: et ille qua perfectio est et principium per naturam per voluntatem.
67 AIJ omit.
68 J adds: ut.
1 MASTER THEODORE, FREDERICK 11' S PHILOSOPHER
quilibet operans huiusmodi ope ratione 69 habet duas delectationes, unam
pertinentem ad artem, aliam ad artificem per quam compatietur laborem
279
IX
venientem ex operatione. (13 Diffinitio autem delectationis est 70 com-
. prehensio perfectionis convenientis virtuti comprehendenti 71.) 14 Verbi
gratia: delectatio philosophi pertinens proprie arti sue est comprehendere
quod ab intellectu suo non evadat quicquid potest humana natura
comprehendere per nobilius argumentum et diffinitionem quae vocatur
demonstratio, et quando loquetur veritatem dicet, et mentientem sciet 72
arguere. 15 Reliqua autem delectatio quae appropriatur ei est imaginare
quod est in motu suo et in quiete et in dispositione. 16 Consequitur id
quod rapuerat virtus rationalis, maxime in regimine dominationis 73 et in
inventione civilium legum, cuius perfectio indiget auxilio dominii 74;
nam philosophia indiget dominio cui indecet 75 virtuositatem suam. 17
Delectatio autem pertinens arti sacerdotis satis nota est. 18 Delectatio
vero militis pertinens 76 arti sue <est> imaginare 77 inimicos vincere,
quod consequitur populi defensio et non tollerantia iniuriae. 19 Reliqua
delectatio ad ipsum pertinens est imaginare victoriam in ludo
instrumentorum belli et in armis 78. 20 Et universaliter omnis artifex has
duas habet 79 delectationes usque ad textores, et absurdum quidem esset
si ars 80 regalis, cum sit nobilior aliis generibus 81, privetur ilIa re quae
in omnibus artibus ex necessitate invenitur. 21 Hoc 82 etenim esset contra
naturam et contradicens ei quod testificatur ens; proprietates 83 vero non
transmutantur.
22 Delectatio autem regis pertinens regno suo est imaginare quod
preceptum suum sit transitivum 84 in ovile suum secundum intentionem
69 in huiusmodi operatione J, huiusmodi operationem Z.
70 et Z.
71 J adds: virtuti.
72 sciat AIZ.
73 donationis J.
74 domini AIZ.
75 indecent (changed to, or from, « indocent ») A I.
76 arti sacerdotis ... pertinens] J om ..
77 imaginari J.
78 read « inermis » 1
79 duas habet] has habet dellai (1) duo Z.
80 has AIZ.
81 gentibus A I JZ.
82 Hec AIJ.
83 proprietas J.
84 transituum A I.
 j
IX
280
suam, et 85 quod diligatur et quod 86 timeatur a suis hominibus et a
c?nvicinis .sui~ prout potest. 23 Reliqua delectatio ei pertinens est spe-
CI~S venation IS secundum suam diversitatem, cuius nobilior pars est
aVlcare aves. 24 Avicare enim est exercitium conveniens non solummodo
omnibus membris, sed etiam omnibus sensibus manifestis et absconsis.
25 .Qu~d conv,:nit ~mnibus membris manifestum est quia indiget
equltahone suavI, forti et mediocri, et unaqueque earum non multum fiat
et leviter tollerandum; exercitium autem huiusmodi carminat 87 calorem
n~t.uralem et e~p~llit superfluitatem et preparat 88 membra ad acceptionem
clbl. 26 Et niSI esset verborum prolixitas, monstraremus quomodo
delectatur singulus sensuum 89 in ipsa arte.
27 Huic etiam 90 coniungitur [delectatio,victoria 91,] utilitas. 28 Hec
siq~idem. docebit equitare et bene pati esse super equum et morigerabit
et ddatablt spem, et dabit appetitum victorie et dabit sollertiam intellectui
et universaliter puIcrius solatium est et nobilior 92 ludus quod possi~
habere natura humana 93, et mirabilius est. 29 Reges autem si habeant
delectationes alias, ut comestiones, potationes, coitus et cetera, hoc non
habent in quantum reges, sed inquantum homines; inquantum enim sunt
reges, non habent propriam delectationem nisi venationem 94.
30 Considerans autem dominus noster serenissimus Imperator 95
Fredericus Hus 96 semper augustus, Iherusalem et Sicilie rex, istius
delectationis nobiIitatem 97 imperatoribus et regibus appropriandam
dumtaxat, et videns antecessores suos et contemporaneos reges in
delectatione a naturaIi veritate appropriata sibi et exhibita non sollicitos
esse, sed potius somnolentos, 31 servorum sui Iimitis minimo imperavit
praesentem librum faIconarii transferre de arabico in latinum, ut eorum
85 que J.
86 J omits.
87 c 2 arminat J.
88 preparant Al.
89 sensum J.
90 autem delectationi J.
91 victorie J.
92 J adds « est ».
93 humana natura J.
94 venatione A 1, venationes Z.
95 Imperator serenissimus Z.
96 Fredericus] Z omits.
97 nobilitate Al.
IX
MASTER THEODORE, FREDERICK U'S PHILOSOPHER 281
sit recordatio quae sapientum sollertia adinvenit per experimentum et
principium inveniendorum in posterum. 32 Ego igitur cum obedientia et
devotione debita domini mei dignum preoccupavi preceptum, presens
opus tractatu quaternario dividendo, 33 primo in theoricam huius artis,
secundo in medicinas occultarum infirmitatum, tertio in curis 98
manifestarum infirmitatum, quarto in medicamen rapinorum quadrupe-
dum.
34 Explicit prologus. Incipit tractatus primus de theorica huius artis 99.
Comments on the Latinity of the preface.
It seems that sections 1-5 are all one sentence of which « adiunxit» is
the main verb (This explains why « adiunxit » is given such prominence
in the manuscripts where it is written with a capital A at the beginning
of a new paragraph). Both this sentence and the rest of the article show
a paratactic structure (characterised by sentences and phrases beginning
with « et ») which is Semitic rather than Latin. The construction « non ...
nisi» (29) is also distinctly Arabic. The word order is also not Classical
Latin, in that verbs do not come at the end of their phrases. «Esse» +
a participle is used where one would expect a finite verb (e.g., 3 fuit
debilitans). Sometimes the syntax seems to be completely awry (2
secundum id quod preparationes suas; in 7 one would expect
« Comestioni, potationi... delectatio inest »; 16 cui indecet virtuositatem
suam). Adjectives are regularly followed by a genitive rather than by a
noun in agreement (2 unicuique entium; 7 unaqueque operationis; 11
aliquo individuorum; 26 singulus sensuum). Comparative forms are used
instead of superlatives (14, 23, 28) - another Arabic feature. Demonstra-
tive pronouns are used with the sense of a simple definite article (26 in
ipsa arte). The wrong Latin word is used (15, 19, 23, «reliquus» for
« alter »; 15, 18, 19, 22 « imaginare » for « imaginari ») and « victoria»
(9) is used in a strange sense. I have taken into account these « errors»
in making the following translation.
1 The providence of nature, governing all things which are con-
tained within the sphere of the Moon, by which wise men have learnt
the rules of wisdom and, having advanced in learning, have elaborated
their own activities, intended to enable all natural entities to reach the
aim of their perfection by a shorter and more direct path. 2 Their per-
fections differed according to the difference of species; for to each
98 curas Al.
99 Explicit. .. art is] J omits.
 IX
282
entity there is a proper perfection, according to that which (Nature) has
gra.nte~ its preparations to receive from the virtue flowing from the
a~tIve Intellect. It is impossible for it to arrive at its proper perfection
wIthout the different activities coming from the virtues found in its own
instrumen~, ~s is seen in the parts of animals and plants. 3 The repetition
of the actIvIty, however, has been weakening the instruments, and as a
resu!t of their weakening there follows a weakening of the virtues, and
a fadure to. complete the activity or activities leading to its proper end
and per~e~t~on. 4 (So nature's providence) added to each of the corpo-
real actI:ltIes a proper pleasure before and after being active, so that,
for the vIrtues, the labour of the activity might be alleviated through the
pursuit of a pleasure (just as a wise king, who gives his soldiers riches
a~d honours, also allows them to enjoy pleasure), 5 so that because of
thIs there might result the forgetting not only of labour, but also of
death, which, although it is the end of all nature, nevertheless is con-
tra.ry to a particular nature. 6 This pleasure is clear in the activity of an
anImal, but most clear in the human body and its species. For man is the
speculation of entities, and therefore he is called the «microcosm ». 7
E.g., for eating, drinking, hearing, seeing and each of the human activi-
tie~ ~he~e ~xists a pleasure, and when the pleasure is taken away, the
actIvIty IS mterrupted (v. I. , destroyed), although he who neglects it, knows
that, in ne.gl~cting th~t activity, the final corruption (i.e., death) super-
venes. ThIS IS seen In the man who, lacking the pleasure in eating,
forgoes eating, and, lacking the pleasure in drinking, forgoes drinking
also, and the same with the other activities. 8 Human activity is of two
p~rts, for one is common to the whole species, another proper (to indi-
vlduals). 9 The common one is eating, drinking, sexual intercourse and
expelling superfluities, sensing through the ten senses 100 and under-
stan~ing through the intellect, and pride, sense of achievement, envy,
avance and the activities whose principle is from nature but whose
perfection is through will, and the other activities. 10 Each one of these
is joi~ed to a pleasure, and tends towards its proper end, as is made
clear m the De anima, the Nicomachean Ethics, and the Rhetorica.
. 11 But proper activities are, for example, «to philosophise », «to
reIgn », «to be a priest », «to be a soldier », «to be a judge », and
proper activities of this sort, that are found in some individual of the
human species. 12 Whoever is active in this kind of activity, has two
~.e., the f~ve external senses and five internal senses described by Avicenna
100
(e.g., ID De amma, 1.5), to which reference is made below in 24.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER 283
pleasures: one pertaining to the art, another to the practitioner through
which he comforts the labour coming from the activity. (13 The defini-
tion of pleasure is «the grasping hold of the perfection appropriate to
the virtue which is doing the grasping ».) 14 For example, the pleasure
of the philosopher, pertaining to his proper art, is to understand that
whatever human nature can understand does not escape from his intel-
lect, through the most noble argument and definition, which is called
demonstration, and that, whenever he speaks, he will tell the truth, and
that he will know how to prove that someone is lying. 15 The other
pleasure, which pertains to himself, is to imagine what a thing is in its
movement, its rest and its disposition. 16 There follows that which the
rational virtue had seized - especially in the management of rulership
and in the drawing-up of civil laws - whose perfecting needs the help
of the ruling power; for philosophy needs a ruling power which its
virtuous state befits. 17 The pleasure pertaining to the art of the priest
is well enough known. 18 But the pleasure of the soldier, pertaining to
his art, is to imagine defeating his enemies, which results in the defence
of the people and the intolerance of injustice. 19 The other pleasure,
which pertains to himself, is to imagine victory in the game of instru-
ments. of war 101 and in arms (or even when unarmed ?). 20 In general
every practitioner down to humble weavers, has these two pleasures,
and it would be absurd if the regal art, when it is more noble than all
other kinds, were deprived of that thing which must of necessity be
found in all arts. 21 For this would be contrary to nature and contradict
that which reality testifies (but properties are not transferred).
22 The pleasure of the king, then, pertaining to his ruling, is to
imagine that his command is affecting his flock according to his inten-
tion, and that he is loved and feared by his people and his neighbours
(respectively) as far as possible. 23 The other pleasure, pertaining to
himself, is the kinds of hunting according to their diversity, of which the
noblest part is to hunt with birds 102. 24 Hunting with birds is an exercise
which suits not only all the parts of the body, but also all the senses,
both the visible and the hidden ones. 25 That it suits all the parts of the
body is clear because it requires gentle, energetic and moderate horse-
riding, and each of these should not be done ( too) much and should be
borne lightly. Exercise of this kind works wonders on 103 the natural heat
101 A reference to chess.
The verb « avicare » is not attested elsewhere.
102
103 The use of the word « carminare » in this context seems to be an innova-
tion on the part of the author. It should mean either «card» (as in« carding

IX
284
and expels superfluities and prepares the parts of the body for receiving
food. 26 If it did not demand too many words, we would show how each
of the senses takes pleasure in this art.
27 To this is also joined usefulness. 28 For this art will teach one
horsemanship and good comportment on a horse, and will satisfy one
and increase one's hope, and it will give an appetite for achievement
and keenness to the intellect, and altogether it is the most beautiful
solace and most noble game that human nature can have, as well as
being most marvellous. 29 If kings have other pleasures, such as eating,
drinking, sexual intercourse etc., these they do not have inasmuch as
they are kings, but inasmuch as they are men. Inasmuch as they are
kings, they do not have any proper pleasure other than hunting.
30 Our Lord, the most serene Emperor Frederick 11, always august
king of Jerusalem and Sicily, considering that the nobility of this pleasure
alone should be made their own by emperors and kings, and seeing that
the kings who were his predecessors and his contemporaries were not
anxious about the pleasure which was their own and shown to be so by
natural truth, but on the contrary were negligent, 31 has ordered the
least of the servants within the bounds of his kingdom to translate the
present book of falconry from Arabic into Latin, so that there may be
a record of those things which the expertise of wise men has discovered
by experience, and a basis for discovering things in future. 32 I, then,
with due obedience and devotion, have hastened to fulfil the worthy
order of my Lord, by dividing the present work into four treatises, 33
the first on the theory of this art, the second on the medicines for hidden
diseases, the third on the cures of visible diseases, the fourth on the
medication of four-footed hunting animals.
IX
MASTER THEODORE, FREDERICK 11' S PHILOSOPHER
285
2 Primo de prohemio. Capitulum primum.
3 Dixit Gatriph Persicus: Multi Persarum et Grecor~m s~pientes in
arte avium rapacium multos libros scripserunt super hoc, In qUlbus mult:
dispendiose (read-sa ?), inutilia et non necessaria compreh~nderunt ..
Sed libros et dicta eorum transcurrentes ~n. p~esent.i .vo.lu~l~e brevlter
compilavi et etiam ea que necessaria et utlha Invenl In 111ls In hoc meo
libro recitare volui.
1 Here begins the treatise on birds: concerning the training of bi~ds
and the medications of the same. This book is translated from PersIan
into Latin.
2 First about the prologue. Chapter one.
3 Gatriph the Persian said: many of the Persians and <:Tre~ks ~ho
were wise in the art of birds of prey wrote many books on thIs, .In whIch
they included many dispendable, useless and unnecessary thIngs. But
reading through their books and sayings, in the present ~ook. I have
compiled (them) briefly and also have wish~d to rehearse In thIs book
of mine what I found necessary and useful In those.
could have been Theodore. The work is translated from Arabic (see pp. 226.an.d
239 above). The mention of Persian is probably due to the fact that Ghatnf IS
regarded as a Persian falconer.

34 The prologue ends. The first treatise on the theory of this art
begins.

5. THE OPENING of THE LATIN Ghatrif.

Edited from Milan, Biblioteca Trivulziana 695.

1 Incipit tractatus avium de doctrina eorum et de medicaminibus

infirmitatum eorumdem qui liber est translatus de persico 104 in latinum.

wool») in which case it may have the metaphorical sense of stirring up the
natural heat; or (more probably) «charm », i.e., bring about magically.
104 This is manifestly wrong and has lead scholars to deny that the translator
 ADDENDA AND CORRIGENDA

I. King Ptolemy and Alchandreus the philosopher: the earliest texts on the
astrolabe and Arabic astrology at Fleury, Micy and Chartres

Paul Kunitzsch has added a useful supplement to this article in 'A Note
on Ascelinus' Table of Astrolabe Stars', Annals of Science 57 (2000),
pp. 181-5. Important recent additions to the study of the early astrolabe
corpus and the Alchandreana are: Catherine Jacquemard, Olivier Desbordes,
Alain Hairie, Du quadrant vetustior a l'horologium viatorum d'Hermann
de Reichenau: etude du manuscrit Vaticano, BAVOtt. lat. 1631,/ 16-17v,
in Kentron, 23, 2007, pp. 79-124; David Juste, Les Alchandreana Primitifs:
Etude sur les plus anciens traites astrologiques la tins d'origine arabe
(Xe siecle), Leiden, 2007, pp. 1-9; and Arianna Borrelli, Aspects of the
Astrolabe: 'architectonica ratio' in Tenth- and Eleventh-Century Europe,
Stuttgart, 2008.
p. 339, n. 56. The reference should be to Bubnov (note 4 above) and to pages
310-65 ofBubnov's work.

11. Physics before the Physics: early translations from Arabic of texts
concerning nature in MSS British Library, Additional 22719 and Cotton
Galba E IV

p. 66. The plants can be identified as follows: alica, spelt; attriplex, orach;
lapsana, charlock; fafelus (i.e. faselus), bean; mala citonia (i.e. kidonia),
quinces; medica, clover; dolica (i.e. dolichus), kidney bean; portulaca,
purslain; molothi (i.e. moloche), mallow (I am grateful for the advice of
Klaus-Dietrich Fischer). The text De ponderibus is also in Madrid, Biblioteca
de la Universidad, 119 (formerly 116-Z-31), fols 127v-130.
p. 75. The scribe or annotator of an early thirteenth-century manuscript of the
work, Avranches, BM, 221, was aware that Alfano was the translator rather
than author of the work, fol. 113r: 'Nemesius episcopus grece fecit librum
quem vocavit prennon phisicon, id est stipes naturalium. Hunc transtulit
N. Alfanus archiepiscopus Salemi'.
p. 86. Further manuscripts of the text of De elementis are Paris, BNF,
lat. 544, fols 67r-v, Prague, University Library, XII.F.26, fols 46v-48r, and
Madrid, Biblioteca de la Universidad, 119, fols 131 v-134r. For the readings
of these manuscripts see C. Bumett, 'Verba Ypocratis preponderanda
 2 ADDENDA AND CORRIGENDA
omnium generum metallis. Hippocrates On the Nature of Man in Salerno
and Montecassino, with an Edition of the Chapter on the Elements in
the Pan tegn i' , in La Scuola Medica Salernitana. Gli autori e I testi, eds
D. Jacquart andA. Paravicini Bagliani, Florence, 2007, pp. 59-92.
p. 101. A passage from the English translation is missing in section <41> after
'Hippoc~ates saying': 'F?r, if man consists of one (thing) only, he is incapable
?f.suffenng; for nothIng IS found that can stand against him. But, ifhe suffers,
It IS necessary that he is cured by one (thing).'
Ill. Adelard of Bath and the Arabs
For the most up-to-date account of Adelard of Bath's life and works see now
C. Burnett, 'Bath, Adelard of', in Oxford Dictionary ofNational Biography,
Oxford, 2004, vol. 4, pp. 339-41. It would be more accurate to translate
'Ara~um ~~~i,a' (p. 89) in this context as 'studies of the Arabs' and 'quod
Arablce dldlCl (p. 90) as 'what I have learnt in Arabic' .
p. 104. Another reference to the same account of the birth ofAdam appears as a
note to Hermann of Carinthia's De occultis, in MS Boston, Public Library,
148~, fol: 20~: 'Se~undum eosdem (Caldeos) etiam, omnibus planetis in
regms SUlS eXlstentIbus excepto Bucarih, natus est Adam in terra Arim ex
Digete et Vulcanis femine'.
IV. Antioch as a link between Arabic and Latin culture in the twelfth and
thirteenth centuries
p. 6. N?te also ~hat MS Mantua, Biblioteca Civica, A.IY.17, fo1. 164rb, gives
the InfOrmatIOn: Nov. 28, feria secunda, 1127, written by Alduinus.
p. 11, n: 37. The article 'The Transmission of Arabic Astronomy' is now
pubhshed In T~~ Enterprise of Science in Islam: New Perspectives, eds
Jan P. HogendlJk and Abdelhamid I. Sabra, Cambridge, Mass., 2003,
pp. 23-51.
p. 13, n. 43. The article 'A Note on the Origins of the Physica Vaticana and
Metaphysica media' is now published in Tradition et traduction: les textes
philosophiques et scientifiques grecs au moyen age latin. Hommage Cl
Fernand Bossier, eds R. Beyers et aI., Leuven, 1999, pp. 59-69.
p. 1~. On Fibonacci see now C. Burnett, 'Fibonacci's "Method of the Indians'"
In Bollettino di Storia delle scienze matematiche, year 23, 2003 [published
2005], pp. 87-97.
p. 21. For a description of the Worcester manuscript see now R.M. Thomson
and M. Gullick, A Descriptive Catalogue of the Medieval Manuscripts in
Worcester Cathedral Library, Cambridge, 2001, pp. 25-6.
ADDENDA AND CORRIGENDA 3
p. 21. Another manuscript is Mantua, Biblioteca Civica, A.IV.17 (consisting only
of Theorica, Parts I-V; information from Monica Green). Another copy of the
first Part of the Regalis dispositio was once in Richard de Foumival 's library:
see L. Delisle, Le Cabinet des manuscrits de la Bibiotheque national, Paris,
1874, vol. 11, p. 534, item 152. The first printed edition (ve) was published
in Venice in 1492.
p. 27. For the expression 'impegi in' see Jerome's preface to Daniel (Biblia
sacra, vol. 16, Rome, 1981, p. 5): 'impegi novissime in Danihelem, et tanto
taedio affectus sum ... '
p. 27. 'calida' should be interpreted as 'callida' ('clever'): I owe this suggestion
to Klaus-Dietrich Fischer.
pp. 38-40. Note the readings of V (Urb.lat. 234): n. 278 quoque BFV; n. 283
invenit V; n. 286 que BV; n. 291 invenit BV; n. 292 V includes the sentence
'Neque enim ... posuimus , .
Note corrections: p. 41 [4] existimaverent> existimarent; p. 42 [8] pluraque >
pleraque; p. 43 [8] silentia> silentio; p. 44 [12] propemodum > populum; p.
45 [15] germane que> germaneque; p. 51 [9] equilibertatis > equilibritatis; p.
53 [2] the scribe has left several words incomplete, which Richard Lemay (in
an unpublished typescript) has completed as follows: Epicur. > Epicureum;
conex. > conexit; affect. > affectum; refer. > referta [3] (p. 54) iudic. >
iudicum; p. 54 [4] attractant, emend to 'attractent'; p. 54 [5] correptorum,
Lemay emends to 'correptus'; p. 58 [7] violenter> violentum (Lemay).
p. 61. For other examples of alphanumerical notation, all within onomantic
contexts and always up to z = 500, see Burgo de Osma, Archivo de la
Catedral, 7, s. xi-xii, fol. 104v (D. Juste, Les A1chandreana Primitifs, p.
681), Cambridge, Trinity College, R.7.23, s. xiv, p. 212, and Cambridge,
University Library, li.I.13, s. xiii, fol. 202r. The fact the equivalents end
at z = 500 shows that, in these instances, the numerals are being used as a
equivalents to letters of the alphabet rather than vice versa.
p. 72. In line 7 the value 'e' (= 5) can be seen.
V. 'Magister lohannes Hispalensis et Limiensis' and Qustl ibn LOql's De
differentia spiritus et animae: a Portuguese contribution to the arts
curriculum?
'Magister' is a little misleading in the title, since John of Seville very rarely
appears as a master. There is no need to make a sharp distinction between
'John of Seville and Limia' and a simple' John of Seville', as implied in
this article.
p. 227. Other texts attributed to 'John of Seville and Limia' are (1) the 'Book
of Venus' in Venice, Biblioteca nazionale Marciana, lat. XIY.174 (4606),
 ADDENDA AND CORRIGENDA 5
4 ADDENDA AND CORRIGENDA

p. 277. <20> Liber carastonis, is followed (fo1. 85v) by a fragment of a translation

fo1. 22v: see VI p. 59n and Paolo Lucentini and Vittoria Perrone Compagni,
of the Arabic text of the introduction to Apollonius, Conics: M. Clagett,
I testi e I codici di Ermete nel Medioevo, Florence, 2001, p.p. 84--6 (item
Archimedes in the Middle Ages, vo1. IV, part 1, Philadelphia, 1980, p. 3.
37); (~) the translation of Albumasar's Great Introduction ot Astrology in
p. 280. Danielle Jacquart suggests to me that <68> should be split into two
two thIrteenth-century manuscripts: see VII p. 268n; (3) the translation of
Alcabicius's Introduction to Astrology in one fourteenth century manuscript: works: Liber geomantie, and De artibus divinatoriis.
see VI p. 78.
VIII. Michael Scot and the transmission of scientific culture from Toledo
p. 229. 'Auriocenus' should be read as 'Antiocenus' ('of Antioch'): see IV
pp. 2-3. The text, and the differences between the versions by Adelard to Bologna via the court of Frederick 11 Hohenstaufen
and John of Seville and Limia are discussed in detail in Charles Burnett
p. 102, n.1. I argue in IX pp. 235-6, that 'the philosopher of the Emperor' is
'Iabit ibn Qurra the J:Iarranian on Talismans and the Spirits of the Planets':
La Coronica, 36, 2007, pp. 13-40. See also Vittoria Perrone Compagni, more likely to be Theodore of Antioch.
p. 110. The correct name ofthe archbishop is Gonzalo Perez, the name 'Garcia'
'Studiosus incantationibus: Adelardo di Bath, Ermete e Thabit', Giornale
and 'Gudiel' have been erroneously attached to him: see F. 1. Hemandez and
critico dellafilosofia italiana, 80-82, 2001, pp. 36-61.
P. Linehan, The Mozarabic Cardinal, Florence, 2004, pp. 422-4.
p. 235. The attribution to a 'bishop John' also appears in MS Tortosa, Archivo
de la Catedral (Biblioteca Catedralicia), 80, fo1. 156r: see VI p. 63n.
IX. Master Theodore, Frederick I1's philosopher
VI. John of Seville and John of Spain, a mise au point
This article appeared almost simultaneously with B.Z. Kedar and
E. Kohlberg, 'The Intercultural Career ofTheodore ofAntioch' , in Medieval
p. 67, n. 40. This 'secondpart'hasnowbeenpublished in C. Burnett, Ji-WeiZhao
Mediterranean, ed. B. Arbel, London, 1996, pp. 164--76 (see article IV, n.
and K. Lampe, 'The Toledan Regule (Liber Alchorismi, part 11): A Twelfth-
Century Arithmetical Miscellany', Sciamus, 8, 2007, pp. 141-231. 56 above).
p. 229. E. Kohlberg and B. Kedar, in 'A Melkite P~ysic~an in Fr~nkis?
p. 71, n. 51. The article 'The Blend of Latin and Arabic Sources' is now
Jerusalem and Ayyubid Damascus: Muwaffaq al-Dm Ya qub b. SIqlab ,
published in Metaphysics in the Twelfth Century: On the Relationship among
Asian and African Studies, 22, 1988, pp. 113-26 (at pp. 116-7), consider
Philosophy, Science and Theology, eds M. Lutz-Bachmann, A. Fidora and
that Barhebraeus's identification of Theodore with Ibn Saqlab's teacher is
A. Niederberger, Turnhout, 2004, pp. 41-65.
unlikely; al-Qifti talks only of 'the Antioch philosopher', and says he died
p. 71, n. 56. See now C. Burnett, 'Euclid and al-Farabi in MS Vatican, Reg.
Lat. 1268', in Words, Texts and Concepts Cruising the Mediterranean Sea, in 1184/5.
p. 242. The older translations ofthe Ethics I, 11 and III and of De generatione et
Orientalia Lovaniensia Analecta 139, eds R. Arnzen and J. Thielmann,
corruptione are now regarded as being the responsibility of one translat~r,
Leuven, Paris and Dudley, Mass., 2004, pp. 411-36.
namely Burgundio of Pisa: see A. Bossier, 'L'elaboration du vocabulaIre
philosophique chez Burgundio de Pise', in Aux origines du lexique
VII. The coherence of the Arabic-Latin translation program in Toledo in
philosophique europeen. L'influence de la latinitas, Louvain-la-Neuve,
the twelfth century
1997, pp. 81-116. . .
p. 248. For another example of 'philosophus' = 'hakim' see the IllustratIon ~f
p. 274 Another manuscript of the Commemoratio librorum and Eulogium is
Petrus de Ebulo, in MS Bern, Burgerbib1. cod. 12011, fo1. 97r where there IS
Prague, Narodni Knihovna Ceske Republiky (University Library), XIII.
an 'achim' of King William 11 of Sicily consulting an astrolabe: reproduced
F.26 (Truhlar 2364), s. xiii-xiv, fols 45v-46r. This manuscript omits item
in Petrus de Ebulo, Liber ad honorem Augusti, eds T. Kolzer and M. Stahli,
12, places item 20 under 'astrologia', reads 'aliis' for 'alius' in item 39, and
'diffinitionibus' for 'confectionibus' in item 57, and gives the title 'De cirugia' Sigmaringen, 1994, p. 43.
p. 249, n. 76. 'philosophus' is probably a mistake for 'Philippus'.
(sic) to the last three items (Rasis, Albucasim and Avicena).
..
;

'!If

INDEX MANUSCRIPTORUM

All the manuscripts mentioned in this volume are listed in alphabetical order of the
name of the place in which they are now located. As guides to the names of the Western
and Islamic libraries I have used respectively P.O. Kristeller, Latin Manuscript Books
Before 1600, 4th ed., revised and enlarged by Sigrid Kramer, Munich, 1993, and
F. Sezgin, Geschichte des arabischen Schrifttums, VI, Leiden, 1978. Corrections (corr.)
or additions (add) to the text (in the section Corrigenda and Addenda at the end of
the volume) are indexed as relevant after the page number of the article concerned.
BM = Bibliotheque municipale; UB = Universitatsbibliothek.

Avranches, BM Cambridge
221: 11 75 (add.); V 250 Fitzwilliam Museum
232: V 250, 251 McClean 165: IV 15n, 62
235: I 335n, 336n, 339,341,343-51, Trinity College
363-7; I (add.) 187; III 101 R.7.23: IV 61 (add.)
R.15.16: III 94n
Barcelona, Arxiu de la Corona d' Arag6 University Library
Ripo1l42: I 332n li.1.13: IV 61 (add.)
Ripo1l168: I 332n !i.VI.5: 1343-9
Ripo1l225: I 331, 332n Kk.l.l: VII 268n
Basel, Offentliche Bibliothek der Canterbury
Universitiit St Augustine: I 339n
D.ii.18: IV 65 Cashel, GPA Bolton Library
Berlin, Staatsbibliothek PreuBischer Med. MS 1: V 224, 234, 251
Kulturbesitz Catania, Biblioteca Universitaria
lat. Fol. 74 (Rose 898): IV 7n, 21, 87: I 342n
30-40,68,72,76 Chartres, BM
Bern, Burgerbibl. 213: I 337
cod. 12011: IX 245 214: I 335,336n, 337,338,343,356;
Bernkastel-Kues, Bibliothek des Hospitals III 101
209: VIII 112n, 113n, 115n, 116n 497: I 337
Bologna, Biblioteca Universitaria 498: 1337
153: IX 274-6
Boston, Public Library Damascus, Zahiriya
1488: III 104 (add.) 'amm 2871: 11 57n
Brussels, Bibliotheque royale Dresden, Landesbibliothek
2772-89: V 225n, 251 Db. 87: IV 12,61,63, 70
Burgo de Osma, Archivo Biblioteca de la Durham, Cathedral Library
Santa Iglesia Catedral B.IY.24: V 226n
7: I 339n; IV 61 (add.)
Edinburgh, National Library of Scotland
Cambrai,BM Advocates 18-6-11: V 224, 225, 255-7,
168/163: VII 253n 266-7
430: IV 40-59,73 Erfurt, Wissenschaftliche Bibliothek der Stadt
930: IV 10, 11n, 15n Amplon. F 250: IV 30-38
 INDEX MANUSCRIPTORUM 3
2 INDEX MANUSCRIPTORUM

Oxford lat. 14385: IX 243n

Erfurt, Wissenschaftliche Bibliothek der 343-50,368 lat. 14390: VII 273, 275-87
All Souls College
Stadt (cont.) Add. 18210: 11 68n lat. 14704: IV 18n; V 227
68: VII 273
Amplon. F 266a: VII 273, 275-87 Add. 22719: IIpassim; IX 237n, 270n tat. 14738: VII 254n
Bodleian Library
Amplon. F 352: IX 226 Add. 26770: V 255-7 lat. 15461: VI 66-71
Ashmole 357: VII 273, 275-87
Amplon. F 380: V 228n Burney 360: V 255-7 lat. 16156: IX 243n
Ashmole 369: I 340
Amplon. 0 62: IV 8n Cotton Galba E IV: 11 passim lat. 16204: V 226,227; VII 269;
Auct. F.1.9: I 339n; III 101-2, 104n
Amplon. Q 184: 11 78n Cotton Vespasian A 11: IV 15n VIII 121, 125-6
Canon.Misc. 555: VIII I11n, 113n,
Amplon. Q 189: V 228n Harley 647: I 336n lat. 16652: VI 61n
115n, 120
Amplon. Q 351: V 227 Harley 1121: IX 263 tat. 17868: I 334n, 339, 343;
Digby 51: VI 74n
Amplon. Q 365: IV 16n; V 235 Harley 2506: I 336n VIII 121, 124-5
Digby 119: 11 68n
Erlangen, UB Harley 3487: V 225n n.a.l. 229: I 339n
Digby 121: 11 65, 104-7
844: IV 63n Harley 3631: VI 61n n.a.l. 1893: VI 78n
Digby 194: V 226
Harley 3731: V 235 n.a.l. 1401: VIII 107n, 111n, 114n
Laud. Misc. 594: VI 73n; VIII 121
Florence Harley 4486: IX 227 Pommersfelden, Schlossbibliothek
Or. 516: IV 65, 74
Biblioteca Medicea Laurenziana Harley 5218: IX 237n 66: IV 61, 6
Selden supra 24: V 251n; VII 261;
Plut. 30, cod. 29: V 230n Harley 5402: IV 14n, 61, 65, 78 Prague, Narodni Knihovna Ceske
VIII 106n; IX 249n
89, sup. 45: VII 254n Royal 12.C.XVIII: VI 73n Republiky (University Library)
Corpus Christi College
Biblioteca Nazionale Centrale Royal 12.G.II: V 225n XII.F.26: 11 86 (add.)
283:1333n,335,336n,343-50,
Cony. Soppr. J.II.10: IV 15n; V 227; Royal 12.G.VIII: VIII 119n; IX 233 XIII.F.26 (Truhhir 2364):
359-62, I (Addendum) 187;
VI78n Sloane 2426: IV 7n VII 274 (add.)
III 101
Wellcome Library
Hertford College
Gotha, Forschungs- und Landesbibliothek 21: IV 62 Ripoll: see Barcelona
4: VIII 115n, 118n
1158: V 237n Lyon,BM
Trinity College
328: III 94n, 104n; V 240n St Petersburg, Public Library
47: VII 258
Istanbul Lat.F. v.lX.l: VIII 111 n
University Library Madrid
Padua, Biblioteca Capitolare
A 6375: 11 71n Biblioteca de la Universidad Toledo, Biblioteca Capitular
D.42: VI 66n
Topkapi Sarayi, Ahmet III 119: 11 66 (add.), 86 (add.) MS 42-30: VI 72n
Paris
3362: IV 75 Biblioteca Nacional MS 95-21; VIII 106n
Bibliotheque de l'Universite
3447: V 237n 10009: VIII 109, 121-6 Tortosa, Archivo de la Catedral
lat. 640: VII 268n
Yeni Cami 10015: VIII 121, 123-4 10: VII 267n
Bibliotheque nationale de France,
784: IV 65 10016: V 230n 80: V 235 (con:); VI 63n
ar. 2486: IV 65
10053: VIII 109, 122
bebreu 1049: IV 15n
Jerusalem, Khalidi Library: V 237n Mantua, Biblioteca Civica Valencia, Biblioteca universitaria
bebreu 1050: IV 15n
A. IV. 17: IV 6 601: IX 277-81
bebreu 1051: IV 15n
Konstanz, Stadtarchiv Marburg, UB Vatican City, Biblioteca apostolica
bebreu 1052: IV 15n
Frag., Mappe 2, Umschlag 8, St. 7: 9: IX 249n, 266-70 Vaticana
lat. 544: 11 86 (add.)
I 332n, 339n Milan Barberini 236: VI 61n
lat. 2858: I 332n
Biblioteca Ambrosiana gr. 211: IV 65-6
lat. 6325: V 247n
Laon,BM A 183 inf.: V 228n lat. 2392: VII 274, 275-87
lat. 6914: IV 30,64,68
413: VII 273 E 7 sup.: IV 9, Iln, 61, 63-4, 71, 77 lat. 2393: VII 274
tat. 7019: IX 274-6
Leiden, Bibliothek der Rijksuniversiteit Biblioteca Trivulziana lat. 2429: IV 21, 22-38,67
lat. 7127: VII 256n
Scaliger 1: III 90n, 93n 695: IX 284-5 lat. 4084: I 343n; V 234n
lat. 7231: I 334n
Scaliger 38: I 339n Munich, Bayerische Staatsbibliothek Ott. Lat. 2186: VIII 106n
lat. 7282: V 228n, 252-3
Voss. Lat. 0 15: I 334n Clm 458: I 340, 343-6 Reg. lat. 598: I 332n, 357, 358
lat. 7311: I 340
Voss. Lat. Q 54: I 333, 335n Clm 560: I 330n, 334n, 340 Reg.lat. 1268: VI 70
lat. 7316: IV 15n
Voss. Lat. Q 58: III 95n Clm 10268: VII 256n; VIII 102n, 105n, Reg. lat. 1285: VII 253n
lat. 7316A: IV 16n
Leipzig, UB 107, 108n, 111, 112, 113n, 114n, Reg.lat. 1446: IX 274-81
lat. 7377B: V 227
401119: VII 273 117n, 118n, 119n, 121-6 Reg.lat. 1661: I 33On;
tat. 7412: I 330n
1131: IV 20-30 Clm 14689: I 330n I (Addendum) 187
lat. 7696: I 334n
40 1148: VII 273, 275-87 Clm. 18917: V 252n Reg.lat. 1855: IV 13n
lat. 9335: VI 66; VII 254n, 255n,
Leningrad: see St Petersburg Clm 18927: IV 14n Ross.lat. 579: VI 66, 71
267n,274
London Clm 25353: 11 57n, 74n, 77n, 82-104 Urb. lat. 206: V 247n
lat. 11219: 11 68n
British Library Urb.lat. 234: IV 21-38, 38-40 (add.),
lat. 11248: 1341,355
Add. 16606: VI 74n Nantes, Musee Dubree 67,68n
lat. 14065: I 341
Add. 17808: I 334n, 335,339,341, 19: IX 277-81
4 INDEX MANUSCRIPTORUM

Vendome,BM cod. 5463: VII 253n

174: 11 66n
Venice, Biblioteca naziona1e Marciana
lat. XIV. 174 (4606): VI 59n
Vienna, Osterreichische Nationa1bibliothek
cod. 3124: VIII 112n, 113n, 114-5;
IX 248n
cod. 5311: V 233n
cod. 5438: VIII 112n, 114n
Worcester, Cathedral Library
F.40: IV 8n, 21 (add.), 30-40
Wiirzburg, UB INDEX NOMINUM
Cod. M.ch.f.23: 11 76n
York, Austin Friars
no. 159: I 339n
This word index includes all proper names of premodern persons, of places, of peoples,
dynasties and sects. In cases where different spellings ofthe proper name are found, the
lemma is generally under the English form (i.e., 'John of' rather than 'Iohannes de' or
'Jean de'). In Arabic names 'al-' is disregarded for alphabetical purposes. This index also
includes titles or incipits of anonymous texts. An 'n' following a page number indicates
that the reference can be found in a note on that page. Corrections (cart:) or additions
(add.) to the text (in the section Addenda and Corrigenda at the end of the volume) are
indexed as relevant after the page number of the article concerned.

Abbo of Fleury: I 332, 333, 334, 336, 339, 340
'Abd al-Latifal-Baghdadi: IX 250
'Abd al-Masih of Winchester: IV 12-13, 18
Abelard: see Peter Abelard
Abraham bar Hiyya: VII 268n; VIII 108
Abraham Ibn Daud: VI 75; VII 251, 264,
265; IX 249
see also Avendauth
Abraham Ibn Ezra: IV 14-15, 16, 17,62,76
Abu 'Ali: VI 73
Abu Bakr: VII 277, 284; VIII 115
Abu I-Qasim: see al-Zahrawi
Abu Ma'shar: 11 63n; III 91, 96, 99, 101;
IV 4; VI 60--61, 77n; VII 268;
VIII 108, 119n, 123, 125
Abu Mahir Musa ibn Sayyar: IV 6
Abu Muhammad al-Rashid, Almohad
caliph: IX 253
Abuteus levita: VII 253; VIII 109
Acre: IX 229,234
Adalbero, bishop of Laon: I 333n
Adam, first man: III 104 (add.)
Adam of Buckfield: V 223
Adam ofCremona: IX 237,267
Adamarius: 11 56
Additional De elementis: 11 57-62, 71-2,
73,74,86-105
Adelard of Bath: I 331n, 336, 337, 338;
11 56, 75, 78n; Ill; IV 2n, 3-4, 18;
V 229, 230, 234; VI 76; VII 257, 268
Adhemar ofChabannes: I 334n
Aetius Arabus: V 239n
Affrica: IV 16
Afonso Henriques, king of Portugal: V 228
Agrippa: IV 63n
Ahmad ibn Yusuf: VII 255, 276, 277
Aimery, patriarch of Antioch: IV 5, 18n
'Ala' aI-Din Kay-Qubadh I, Seljuk sultan:
IX 228, 232, 233
'Ala' aI-Din Muhammad Ill, Isma'ili Grand
Master: IX 228, 231-2
'Alam aI-Din Qaisar: IX 252
Alamut: IX 232
Albarracin-Segorbo: VI 72
Alberic of Monte Cassino: 11 76n
Albert, patriarch of Antioch: IX 227-8, 263--4
Albertus Magnus: V 223, 230n
see also Speculum astronomiae
Albumasar: see Abu Ma'shar
Albumassar Grecus: VIII 119n
Alcabicius: VIII 119n
see also al-Qabisi
Alchandreana: IV 62
Alchandreus Philosophus: I 343
see also Liber Alchandrei Philosophi
Alduinus: IV 6 (add.), 18,68-9
Aleppo: IX 229
Alexander of Aphrodisias: VII 261, 263,
265, 279; IX 253
Alexander of Macedon: 1343
Alexandria: I 341; VII 259; VIII 116
Alexios, emperor: IV 5n
Alfadhol: VII 280
Alfano, archbishop ofSalerno: 11 57, 62n,
73, 75 (add.), 76-7, 80, 82;
IV24n
Alfonso VI, king of Castile and Le6n: V 227;
VII 249, 250
Alfonso VII, king of Castile and Le6n:
V 228; VII 250
 2 INDEX NOMINUM
Alfred of Shareshill: V 223; VII 261, 263,
264, 269; IX 273
AIgazel: see al-Ghazali
Alhandreus: 1340
'Ali al-'Irnrani: VII 268; VIII 108, 109,
114, 117, 125
'Ali ibn al-'Abbas: see al-Majusi
'Ali ibnAbi'I-Rijal: VIII 110, 119n
'Ali ibn Ridwan: VII 254, 280
Almersius: IV 15n
Almohads: V 236; VII 251, 265; VIII 104
Altzelle, Cistercian monastery: IV 200
Alvaro of Toledo: VIII 109, 110
Ammonius: 11 72
Andrea Alpago: VII 2600
Andreas of Micy: I 336
Anselm, abbot of Bury St Edmunds: 11 56
Anselm, archbishop of Canterbury: 11 56n
Antioch: III 98; IV; V 229 (con:); IX 228,
229,230,234
Antiochus: VIII 108
Antonius Vitalis Pyrranensis: IV 21
Apollonius ofPerga: VII 251n, 277 (add.)
pseudo-Apollonius ofTyana: 11 58, 64
Apuleius: 11 76n
Arab: IV 56
Arabs: III 89-90 (corr.), 100-101, 103,
107; IV 36
Aratea: I 336n
Archimedes: VII 251n, 259, 276
Argaphalau: VIII 112, 113n
'Arib ibn Sa' d: VII 281
Arin: III 104
Aristotle: 11 61, 72; III 94; IV 5, 41-42;
V 221, 238, 249; VII 257, 258, 260,
265,276,278; IX 249, 253
Categories: IV 27n
De anima: V 238, 240, 243, 245;
IX 239, 242-3
De animalibus: IV 17n; VIII 102, 110, 120
Metaphysics: IV 13n
Meteorologica: 11 63, 64n, 78; IX 237n
Nicomachean Ethics: IX 239, 241-2,
246,247
Physics: 11 74, 75n; IV 13n, 33n;
IX 226
Rhetorica: IX 239, 242, 246
pseudo-Aristotle
De causis: VII 278
Iudicia: VIII 113, 119n
Secret of Secrets: IV 17; V 232,237,
255-8; VI 60; IX 228n, 230, 237, 267
Armenia: IX 228
Arnau de Vilanova: 11 62n; V 234n
Arsodochius: VII 275
Articella: 11 79-80; IV 31 n
Ascelin of Augsburg: I 333, 334, 335, 340,
343-66; I (Addendum) 187
Assassins: IX 231
Astrologicae speculation is exercitium: VI 60
Athenians: IV 51
Athir aI-Din al-Abhari: IX 253
Augsburg: I 351
Autolycus: VII 275, 278
Avendauth: V 236n, 242; VI 74-5; VII 252,
264; VIII 103, 105n; IX 243, 249;
see also Abraham Ibn Daud
Avendeuch Israhelita: VII 251
Averroes: VII 253, 263; VIII 103, 106, 109,
110, 117, 119-20; IX 243, 254
Avicebrol, Avicebron: see Ibn Gabirol
Avicenna: VII 256, 260; VIII 11 0, 117,
119-20; IX 251n, 254
Canon Medicinae: VII 266, 280, 281;
IX 237
Daneshnameh: VII 262
al-Shifa': VI 71; VII 251, 265; VIII 103,
105, 106
De anima: V 242, 244n; VI 73, 75;
VII 252, 262; IX 243-4, 246,
249, 270-74, 282n
De mineralibus: 11 64; V 222; VII 261
see also Ibn Sina
Baeza: VI 67n
Baghdad: III 94; IX 228, 231
Baghdad, caliph of: IX 236
Bahramshah of Erzinjan: IX 250
Banu Hud, rulers of Saragossa: VII 249, 251
Banu Musa: VII 277
Banu Nun, rulers of Toledo: VII 249
Barcelona: VI 67n, 74, 76; VII 268n;
VIII 108-9
Barhebraeus: IV 17n; IX 228-31 (corr.),
234n, 248, 254, 264-5
Bartholomew of Parma: VIII 111-4, 118
Basgund: VIII 124
Bath: 11 54
al-Battani: III 96
Battle Abbey: 11 54
Bede: I 344
Bern ofPrum: 1332,334
Bernard de Sedirac, archbishop of Toledo:
VII 250
Bernelinus Junior: I 339n
Bernon: I 336n
Berthold of Moosberg: IV 12
Bible: 11 62
Billingsley, H.: I 342n
al-Bitruji: VII 265; VIII 102, 105, 109,
110,114
Boctus, king: IX 228
INDEX NOMINUM
Boethius: I 332; III 94; IV 23, 24; V 249;
VII 257, 258
Bohemond IV, ruler of Antioch: IX 230, 232
Bohemond V, ruler of Antioch: IX 230
Bologna: VIII 116
Bonfilius, bishop of Gerona: I 331
BookofSidrach: IX 227-8,251,253,263
Bougie: IV 16
Braga: V 227,233
Brescia: IX 226, 255
Burgundio of Pisa: 11 57, 58, 62n; IV 5,
7; VII 257, 259, 260, 266; IX 242
(corr.)
Bury St Edmunds, abbey: 1344; 11 55
Byzantius: 11 82
Calcidius: 11 59, 60n; V 239n, 244;
IX 243n
Capitula regis Almansoris: VIII 122
Cassiodorus: I 341; VII 257
Castelfranco: IX 256
Catalonia: I 331
Cathars: IX 251 n
Cerebrun ofPoitiers, archbishop of Toledo:
VII 253
Chartres: I 336; VII 258, 264
Christ: IV 55
Chrysostom: IV 5
Church Fathers: IV 5
Cicero: 1351; III 101; IV 43n, 45n, 55;
VII 257
Cicilia: III 98
Clavicula Salomonis: VIII 119n
Cleopatra: 11 63
Coimbra: V 227,231
ComestorlManducator: see Petrus
Constantine, Holy Roman emperor: IV 55
Constantine the African: 11 53, 54, 66, 67,
70,71,74,75,77,78,79,80;
III 97, 98, 106n; IV 7-8, 16,21, 24n;
V 225; VII 266, 285; IX 237n, 238,
270-74
Constantine of Fleury: I 332, 334, 338
Constantine of Lampron: IX 228,232-3
Constantinople: IV 5; VIII 116
C6rdoba: III 97; VII 251, 253, 265
Costa ben Luca: see Qusta ibn Luqa
Coventry: III 94
Cremona: VII 253, 256, 281
Crusaders: IV 1-4
Cue lIar: VI 72
Cum volueris facere astrolabium: VI 76
Daniel ofMorley: V 231n; VII 254, 268, 280
De artibus divinatoriis: VII 280 (add.)
De cibis: II 65-6, 67, 68-9, 106-8
3
De difJerentiis tabularum: VI 65-6
De indagatione cordis: VIII 108
De metallis: 11 63-5, 67, 68-9, 104-7
De ortu scientiarum: VII 267
De physicis ligaturis: 11 54-5, 72, 73
De ponderibus: 11 66 (add.)
Dee, John: I 342n; 11 55n
Democritus: 11 72, 82
Desiderius, abbot of Monte Cassino: 11 80
Destombes astrolabe: I 330; IV 62
Diocles: VII 277
Dionysius the Areopagite: VI 66n
Dioscorides: 11 63; IV 8, 18,37,38-9
Dominicus Gundissalinus/Gundisalvi:
III 106; IV 18; V 242-5,248,
259-64; VI 63-73, 76;
VII 251, 252, 259n, 262-5, 266,
267n, 269; VIII 103, 104, 105,106;
IX 244
Donatus: VII 257
Dorotheus: VIII 108, 119n, 125
Dresden Almagest: IV 12-13,61,63
Druvius: VIII 124
Druvius Persye: VIII 119n
Durham: V 225
Ebro, valley of the: V 248, VII 251
Empedocles: V 241n, 247
England: III 97
Epicurus: IV 53
Epitome totius astrologiae: VI 75-6, 77
Escu1eus: VII 275, 278
Etienne of Provins: VIII 102, 117
Etna: III 97
Euclid: I 342n; III 91, 94-5, 98, 102;
IV 4; V 249; VI 70-71; VII 251n,
257,258,259,268,275,276,277
Eugene Ill, pope: IV 5
Eugenius, emir of Sicily: III 98
Eugenius, Saint: VI 66
Eunomios: 11 68, 72
Euphon: 11 56
Europe: IV 43
Eutocius: VII 251n
Faenza: IX 226
Fakhr aI-Din al-Razi: see al-Razi
al-Farabi: VI 68, 70, 71, 73; VII 257, 259n,
260,261,262,263,264,265,276,
279,287; VIII 103, 105; IX 228, 231
al-Farghani: V 227, 237n, 241; VI 60;
VII 259, 277; VIII 108, 109, 111n,
112n,122
Fernando, king of Le6n: V 231n
Fibonacci: see Leonard of Pisa
Firmicus Maternus: I 340; VII 268
 4 INDEX NOMINUM
INDEX NOMINUM 5

Fleury, abbey of St Benoit de: I 332, 334, Hellenism: 11 62

336,340
Helperic of Auxerre: VIII 122
Francesco Pipino: VII 274, 282
Henricus Aristippus: VII 260, 261
Frederick 11, Holy Roman Emperor: IV 17, Henricus de Flisco: VIII 112
18; VII 253; VIII; IX
Henry 11, king of England: III 92, 100, 106
Fulbert, bishop of Chartres: I 334, 335, 336, Henry the abbot: IX 257-8
338,339
Henry of Burgundy: V 227-8,231
Fulcher of Chartres: IV 1
Henry of Winchester: IV 12n
Fulco Rufus: IX 261-2
Herbertus, doctor: V 225
Hermann of Carinthia: I 338; 11 64;
G. de Tucco: IX 257
Galen: 11 63, 68, 78n, 79n; IV 5, 31n;
III 95-6, 104 (add), 106; IV 13n;
V 222n, 245,248-9,251,259-67;
V 237; VII 255,266,279,280
VI 70-71, 73n, 76; VII 258, 264,
pseudo-Galen, De spermate: 11 68, 69-70,
71, 72, 75, 76 268; VIII 109, 119, 123-4; IX 233n
Galippus: VII 253, 254 Hermann the German: VII 253, 269; IX 242
Gatriph: IX 284-5 Hermann the Lame: I 330n, 332, 334, 335n,
see also al-Ghitrif 338,339, 340, 341n, 352,354,356
Hermes: V 233, 256; IX 254n
Gauco: VI 66
Geber, alchemist: 11 65 Hetoum I, king of Armenia: IX 232, 234
Hildebert of Lavardin: III 95
Geminus of Rhodes: VII 278
Hippocrates: 11 67, 68-9, 70, 71, 72, 76,
Geometria ineerti auetoris: 1331, 339n;
III 101 77n, 78, 79, 82; III 94; VII 279
pseudo-Hippocrates: VII 280
George of Antioch: IV 18
Honorius Ill, pope: VIII 116
Gerard of Cremona: I 342; 11 63n, 65, 79n;
Hosatus: see Ocreatus
IV 18; V 231n; VI 66, 70-71, 73,
Huesca: 1337; IV 4n
77n; VII; VIII 103, 104, 105; IX 249
Hugo Etherianus: IV 5
Gerbert of Aurillac: I 330, 331n, 332, 333,
Hugo of Santall a: IV 16n; VII 267n;
335,336,336,340; III 101
VIII 119; IX 233n
Ghayat al-Hakim: see Pieatrix
Hiilegii, Mongol khan: IX 250
al-Ghazali, AIgazel: VI 63; VII 262, 265
Humbert de Romans: IX 255
al-Ghitrif: IX 239-40
Hunayn ibn Ishaq: 11 67, 79; VII 275
see also Gatriph
see also lohannitius
Giuntini, Francesco: I 342n
Hypsicles: VII 278
Golden Horn: IV 5n
Gonzalo Perez 'Gudiel', archbishop of
lammarco Cinico: IX 274
Toledo: VIII 110 (eorr.)
Ibn 'Abd al-Baqi: VII 277
Gottfried ofViterbo: 11 56, 76-7
Ibn 'Abdun: V 236
Greeks: IV 39
Ibn Abi U saybi' a: IX 231 n
Gregory VIII, anti-pope: V 228, 233, 241
Ibn al-'Abbar: VII 251n
Gregory IX, pope: IX 251n, 267, VIII 104,
Ibn al-Haytham: VII 251n
116
Ibn al-Khayyat: VIII 119n
Gregory ofNyssa: 11 58, 62n
Ibn al-Qifti: I 342n
Grosseteste: see Robert Grosseteste
Ibn al-Saffar: VI 74n
Guichardus: 11 56
Ibn al-Wafid: VII 266, 280
Guido Bonatti: VIII 119n
Ibn Bajja: VII 260n
Guido de Vere, bishop of Tripoli: IX 250
Ibn Bibi: IX 232
Guimaraes: V 227
Ibn Ezra: see Abraham Ibn Ezra and Moses
Gundissalinus, Gundissalvi: see Dominicus
Ben Jacob ibn Ezra
Gundissalinus
Ibn Gabirol, Avicebron: VI 63; VII 265;
VIII 105, 106
Haly: see 'Ali al-'Imrani
Ibn Mu'adh: VII 259, 278
Haly Sarracenus: VIII 119n
Ibn Rushd: see Averroes
Hametus: VII 276
Ibn Sab'in: IX 253, 254
Hayton: see Hetoum
Ibn Sina: IX 228, 231
Hebrews: 11 62
see also Avicenna
Ibn Tumart: VII 253
Ibn Yunus: see Kamal aI-Din
al-Idrlsi: 11198
Innocent IV, pope: IX 267
lohan de Sephila: VI 62n
lohannes Hispalensis et Limiensis: see John
of Seville and Limia
lohannes Hispanus: VI; VII 263, 265;
VIII 103-4
lohannes Toletanus: VI 73
lohannitius: 11 79n; IV 31n; VII 274,275
Isaac Israeli: V 244n; VII 266, 269n, 280
Isabel, queen of Armenia: IX 232-3
Ischia: IX 257
Isfahan: VII 265
Ishaq ibn Hunayn: 11 58,61,62
Ishaq ibn 'Imran: IX 238n
Isidore of Seville: 1341,342; 11 66
Isma'ilis: IX 231
Jabir ibn Aflah: VII 259
Jabir ibn Hayyan: see Geber
Jacob AnatolilAnatolio: VIII 111; IX 253
Ja'far Ahmad III Sayfal-Dawla Ibn Hud:
VII 251
Jalal aI-Din al-Rumi: IX 232
James of Venice: IV 5, 13n; V 249-50;
VII 258, 259, 262, 266; IX 242-3
Jergis: VII 268; V1II119n
Jerome: IV 27 (add)
Jerusalem: IV 16; IX 252
Jews: IV 18
Johannes: see lohannes and John
Johannes, astronomicus: VI 72
Johannes, magister scolarum: VI 72-3
Johannes de Bailleul: IV 30
Johannes Sarapion: VII 280
Johanni Sibili: VI 62n
Johannitius: see lohannitius
John, bishop of Seville: V 235
John, mathematician: III 94
John David: V 231n, 236n; VIII 109n
John David of Toledo: VI 73-4
John of Castelmoron-sur-Lot, archbishop of
Toledo: V 236n, 242; VI 75;
VII 251, 252; VIII 103
John ofPalermo: IX 248, 257-8
John of Sacrobosco: VIII 114n
John of Salisbury: V 250; VII 258
John of Seville: IV 16; VI; VII 252;
VIII 103, 108, 109, 122
John of Seville and Limia: IV 2-3;
V (eorr.); VII 250-51, 259n, 268,
269; IX230n
John of Spain: see Iohannes Hispanus
John Pier de Lyons: IX 263-4
John the Melancholy: 11 56
John the Saracen: IV 7-8, 16, 17
Joseph Hispanus: I 331
Joseph Sapiens: I 331
Juan Gonsalvo of Burgos: VII 264
Juda b. Salomon ha-Cohen ibn Matkah:
VIII 102n; IX 234,235
Julius Caesar: IV 55
Kamaha: IV 17n
Kamal aI-Din ibn Yunus: IV 17; IX 228,
230-31,236,246,250,252,253
Kay-Qubadh I, Seljuk sultan: IV 17n;
IX 232,234
al-Khayyat: VIII 125
al-Khwarizmi: I 330n, 337; III 91, 95,
101-5, 106; IV 4n; V 230, 249;
VI 64-70; VII 258, 277
al-Kindi: 1343; VII 260,261,263,265,
269n, 277, 279, 280; VIII 119n, 125;
IX 236, 267
Kinsica: IV 16
Konya: IX 232, 252
Kyot: III 98
Laodicea: IV 5n
Latinity: IV 43,47,49,58
Latins: IV 57
Leo Tuscus: IV 5
Leofnoth, scribe: I 340
Le6n: V 248
Leon 11, king of Armenia: IX 232
Leonard of Pisa, Fibonacci: IV 16 (add);
VIII 116; IX 226, 227, 231, 236,
248,259-60
Liber Alehandrei philosophi: 1334,339,
340; VIII 109, 113n, 124-5
Liber AlehorismilAlgorismi: VI 64-70 (add)
Liber Anoe: VII 281
Liber Augustalis: IX 251
Liber de aecidentibus alfel: VII 281
Liber de aluminibus et salibus: VII 280
Liber de eausis: VIII 105
Liber divinitatis: VII 280
Liber elephantie: 11 82
Liber geomantie: VII 280 (add)
Liber luminis luminum: VII 280
Liber mahameleth: VI 66-70
Liber Mamonis: IV 10-13, 15, 40-59,64,67
Liber Marii de elementis: 11 56, 64
Liber Nemroth: VIII 121
Liber novem iudieum: VIII 119; IX 233
Liber podismi: I 337n
Liber pulveris: VI 64
Lima: V227
Limia: V 227; VI 62
 6 INDEX NOMINUM
Lisbon: V 231
Loire: III 91
Lombardy: VIII 116
Lucan: III 101
Lucca: IV 14, 16
Lupitus: 1330,331,339,341
al-Ma'mun, caliph: IV IOn
Macrobius: III 101; IV 11,47-9, 57n
Magna Graecia: IV 3
Mahumar Syrus: VIII 119n
Maimonides: VIII 110, 118; IX 253
Majorca: IV 7-8
al-Majusi: IV 6-10,11, 19n, 20-40, 64, 67-9
Malik al-Kamil, Ayyubid sultan: IX 231,
232,252
Malik al-Salih, Ayyubid sultan: IX 252
Mamistra: IV 3
Manfred, king of Sicily: VIII 115; IX 267
Manilius: 1331n
Marbod of Rennes: 11 63
Marius: 11 56, 76-7
see also Liber Mar;;
Marius Salemitanus: 11 56
Mark of Toledo: VII 253; VIII 104, 106
Marseilles: 1338; VI 65n; VII 250n
Martianus Capella: 1342; VII 257, 258
Masha'allah: 1340; V 226, 237n; VI 60;
VII 268, 278; VIII 108, 119n, 122, 125
Maslama al-Majriti: 1330,331; IV 4n
Matheus Ferrarius: IV 8
Mauhmar: VIII 124
Maurice Bourdin: V 228,231,233
Mauritius, archdeacon of Toledo: VII 253;
VIII 104
Mauritius Hispanus: VIII 106
Menelaus of Alexandria: VII 251n, 259, 276
Messahalla: see Masha'allah
Messehallah Indus: VIII 119n
Messina: IX 226, 229-30, 260-61
Michael de Capella: IV 21
Michael Scot: VII 253, 256n, 259, 262, 263,
264, 269; VIII; IX 225, 227n, 230,
233n, 238n, 242, 244-5,248,253
Micy, abbey of St Mesmin de: I 330, 332,
333,334,335,336,338,351
Mileus: see Menelaus
Minerva: III 96
Moamin: IX 226, 238-40, 251, 274-84
Mongols: IX 236
Mont Saint-Michel, abbey: V 250,251
Monte Cassino, abbey: 11 53, 71, 73, 78,
80; IV 18
Montpellier: IV 21
Moses, the prophet: 11 62
Moses Ben Jacob Ibn Ezra: VII 263
7
INDEX NOMINUM
Moses of Bergamo: IV 5 al-Razi, Abu Bakr Muhammad ibn Zakariya:
Pococke, Edward: IX 228n
Mosul: IX 228, 230-31, 234, 246 Porphyry: 11 72; V 234 VII 256, 266,267,280,281
al-Mubashshir ibn Fatik: I 342; VII 255n Capitula Regis Almansoris: VIII 108, 109
Porta Agazir: IV 16
Muhammad ibn 'Abdalhaqi: VI 71 Porta Samuel: IV 16 pseudo-Razi: VII 280
al-Razi, Fakhr aI-Din: IX 253
Porto do Lima: V 227
al-Nasir, caliph: IX 250 Portucale: V 227 Reginerus: III 94
Nasir aI-Din a1-Tusi: see a1-Tusi, Nasir aI-Din Portugal: V 227-8; VI 62, 76 Reichenau: I 332, 334
al-Nayrizi: VII 277 Reims: I 332
Praeceptum canonis Ptolomei: I 334, 336,
Nemesius: 11 56,57-62,68, 75 (add); V 244 Remigius: 11 73n
337,339,341,342n
Nicholas of Damascus: V 222; VII 261 Rhetorica ad Herennium: IV 9,61,63-4,66
Primum capitulum in inventione: VI 60
Nicomachus: VII 257 Richard Bishop: V 236n, 250
Priscian: VII 257
Normandy: III 97 Priscianus Lydus: 11 76n Richard of Foumival: IV 21 (add); VI 60n;
Northungus ofHildesheim: IV 7, 8n VII 268
Proba Fa1conia: III 95
Richard de Pulcaro: IX 258-9
Proclus: V 222; VII 263
Ocreatus, Hosatus: I 333n; III 94; V 234; Richer: I 331
Ptolemy
VI 64, 76 Almagest: III 98, 106; IV 12-13,47; Ripoll, abbey: I 331, 332, 333
Oliva, abbot: I 332 Robert, earl of Gloucester: 11 56n
VI1250,256,257,259,275,277;
Oribasius: 11 82 Robert Grosseteste: IX 241, 245n
VIII 105, 111n
Orient: IV 37 Robert of Ketton: III 95-6; V 248-9; VII 258
Handy Tables: I 336
Planisphere: I 330n; III 106 . Robert ofTorigni: 1343; V 249; VII 258
Padua: IX 226 Rodrigo Jimenez, archbishop of Toledo:
pseudo-Ptolemy
Pagliara: IX 261 Centiloquium: IV 4; VII 268; V 235; VII 253; VIII 102-3, 104, 116
Palencia: VI 63n Roger 11, king of Sicily: IV 18
VIII 108, 121
Palermo: IV 16, 18 Roger Bacon: V 238
Iudicia: VIII 113
Pancus: IV 6, 18,68-9 Roger de Camera: IX 258-9
See also Praeceptum canonis Ptolomei
Pandolfus: V 248n Roger of Foumival: VII 254, 268
Ptolemy, king of Egypt: 1340-42
Pantoele: V 246, 248 Roger of Palermo: IX 263-4
Pythagoras: III 93,94
Paris: V 223,249,251 Roland ofCremona: IX 250-51, 255-6
Pavia: IX 234 Rolandino ofPadua: IX 227, 256
al-Qabisi: VI 60, 77n, 78; VIII 108, 109,
Peripatetics: IV 53 Rome: IV 16; VIII 102, 116
114, 123-4
Peter Abelard: III 107 Rudolph of Bruges: IV 15n; VI 74
see also A1cabicius
Peter of Calabria: IX 261-2 al-Qarafi, Shihab aI-Din: IX 252 Rueda Jal6n: VII 251
Peter the Deacon: 11 62n, 71, 78, 79n Rusticanus Pisanus: IV 8n
al-Qazwini: IX 231n
Peter the Venerable: III 107; V 248 Rusticus of Pisa, son of Bella: IV 7-8, 16
Quintilian: VII 257
Petrus Alfonsi: I 336, 337, 341n; III 95, pseudo-Quintilian: III 95
103-4,105; IV 4n, 47n; V 242n Sacrobosco: see John of Sacrobosco
Qur'an: V 248
Petrus ComestorlManducator: I 333n Sahl ibn Bishr: VII 268; VIII 108, 117,
Qusta ibn Luqa: 11 54, 63, 66-7, 104; III 94;
Petrus de Ebulo: IX 248 (add) 119n, 123
V; VI 60, 73; VII 250, 260; IX 243
Petrus Hispanus: IV 17n; IX 237 Sa'id al-Andalusi: I 342n; VII 249
Qustantin of Armenia: see Constantine of
Philaretus: 11 79, 82 St Nicholas, Exeter, priory: 11 54
Lampron
Philip of Antioch: IX 232 St Paul, Antioch, abbey: IV 10, 58
Phi lip of Cord6ba: VIII 102 St Trinity, Toledan monastery: VII 253
R. de Amicis: IX 257-8
Philip of Tripoli: IV 17; IX 230, 249, 250, Salemo: 11 73; III 91, 97; IV 3, 18,39;
R. de Salemo: IX 258-9
256n IX 238n
Ragimbo1d of Cologne: I 337n
Philippe Auguste, king of France: VII 254 Salimbene: IX 237n
Rainard of Bobbio: 1331n
Picatrix, Ghayat al-Hakim: III 97; V 239, Salio ofPadua: VIII 104, 106, 115
Ralph of Liege: I 337n
240n, 241 Ramsey: I 332, 336 Sapientes Indiae: IV 15
Picot: 11 56 Saracens: VI 65
Rashiq ibn 'Abdallah al-Hasib: VIII 108,
Pier della Vigna: IX 226, 227, 236, 252, Saragossa: IV 14; VII 251n
123-4
257-8,266 Raymond of Amous: V 228, 231 Sayyid al-Matran: V 235
Pietro d'Abano: V 234n Scotia: 11 73
Raymond de La Sauvetat, archbishop of
Pisa: IV 4, 14-15, 16, 18; VI 76; IX 257 Secret of secrets: see pseudo-Aristotle,
Toledo: IV 18n; V 225, 228, 231,
Plato: 11 59, 61, 63, 72; III 93, 106-7; Secret of Secrets
241,242,267; VI 72, 73; VII 250;
IV 41; V 223, 243, 244, 245; IX 232, Segorbe: VIII 104
VIII 103
254n Segovia: V 248; VI 63n, 70n; VII 264
Raymond of Marseilles: I 338, 342; IV 18n;
Plato of Tivoli: VI 74; VIII 108, 109n, 121 Seljuks of Rum: IX 232
VI 65n; VII 250n
 8 INDEX NOMINUM
Seville: V 231, 235
Sicily: 11 78; III 98; IV 39; VIII 116
Sigo: I 336
Simon of Genoa: IV 9
Sisnando Davidiz: V 231n
Socrates: 11 63, 72; IX 254n
Soldanus Babilonie: IX 233n
Solomon: IV 22
Solon: IV 51-2
Southern Italy: 11 53,80
Speculum astronomiae: VI 60n; VIII 109
Stabilis of Micy: 1333,335,344-5,351
Stephen of Pisa, Stephen the Philosopher:
IV; VII 257n
Stoics: 11 60, 61
Suhrawardi, 'Umar: IX 250
Sultan of Babilon: VIII 119
Sunifred, archdeacon of Barcelona: 1330
Symmachus: I 351
Syracuse: 11 75; III 97; IV 3-4
Syria: IV 39
Talavera: V 236
Tancred, prince of Antioch: IV 4
Tarazona: IV 16n
Tarsus: III 98; IV 3
Terence: IV 44n
Teresa, queen of Portugal: V 227-8,232-3,
241; VI 72; VII 250
Tewkesbury: V 234
Thabit ibn Qurra: III 99, 101, 104; IV 2-4;
V 227,228, 237n, 240, 252-5; VI 60;
VII 251n, 268, 276, 277; VIII 108
Thaddeus of Hungary: VII 254
Themistius: VII 276
Theodore Abu Qurra: IV 6
Theodore ofAntioch: IV 17; VIII 119; IX
Theodoros: 11 68, 72
Theodosius of Bithynia: III 96; VII 251n,
259,275,278; VIII 110
Theofilus in provincia Sicilie: VIII 115n
Theophilus: 11 79
Thierry of Chartres: 1337,338; V 249, 251;
VI 76; VII 258,264
Thomas of Cantimpre: V 232n
Tideus: VII 259, 277
Toledan Regule: VI 67n (add.)
Toledan Tables: I 337; VIII 109
Toledo: III 106; IV 14, 18, 19n; V 242,248;
VI 62, 66, 70n, 76; VII 249, 251,
256,261,268,269,275,281;
VIII; IX 235
Toledo, school of: IX 228, 263-4
Torah: IX 230
Toulouse: IX 251n
Tripoli: VIII 116
Tudela: IV 14
Tunis, emir of: IX 235, 257-8
aI-Tusi, Nasir aI-Din: IX 230, 232, 250, 254
'Umar ibn aI-Farrukhan aI-Tabari: V 226,
237n; VI 60, 77n; VIII 108, 119n
Umbria: VIII 116
aI-Urmawi, Siraj aI-Din: IX 252, 254
Venice: VIII 116
Vich: I 331
Virgil: III 101
Visigothic: IV 14
Vitruvius: I 336n
Walcher, prior of Greater Malvern: I 341 n;
III 105; IV 4n
Whitby, Benedictine priory: IV IOn
William, Angligena: VI 66
William, bishop ofSyracuse: III 97; IV 3
William, mathematician: III 94
William of Conches: IV 57n
William of Luna: IX 245
William of Malmesbury: 1340; 11 76n
William of Moerbeke: VI 77; VII 266;
VIII 110; IX 242
William Scot: VIII 115
William Stafford: VI 66
William of Tyre: IV 3n
Wiscard: 11 56
Wolfram von Eschenbach: III 98
Xenocrates: 11 75n
Ya'qub ibn Siqlab/Saqlan: IV 17n;
IX 229 (con:), 234
Yahya ibn Sarafyun: VII 280
YusufaI-Mu'taman Ibn Hud, king of
Saragossa: VII 251 n
Zael: see SahI ibn Bishr
ZaeI Tolletanus: VIII 119n
aI-Zahrawi, Abu I-Qasim: VII 256, 266,
280,281
Zain aI-Din mosque, Mosul: IX 250
Zangi, sultan of Aleppo: IV 13n
al-Zarqalluh: VII 249, 250
Zimus: VIII 108-9