02 Zipser Medical Byzantine Books

EIKASMOS
Quaderni Bolognesi di Filologia Classica
Studi Online, 2
MEDICAL BOOKS
IN THE BYZANTINE WORLD
EDITED BY
BARBARA ZIPSER
BOLOGNA 2013
Medical books in the

Byzantine world
edited by
Barbara Zipser
Bologna 2013
o
Eikasms Online II
ISSN 2282-2178
In memoriam
David Bennett
Table of Contents
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
List of figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
List of abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii
1. Prefatory note: the uses of medical manuscripts
Peregrine Horden (RHUL and Oxford) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Byzantine medicine, genres, and the ravages of time
Vivian Nutton (UCL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3. Disease and where to treat it: a Byzantine vade mecum
Dionysios Stathakopoulos (KCL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4. Two Latin Pre-Salernitan medical manuals, the Liber passionalis and
the Tereoperica (Ps. Petroncellus)
Klaus-Dietrich Fischer (Mainz) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5. The fate of a Greek medical handbook in the Medieval West: the Introduction, or the Physician ascribed to Galen
Caroline Petit (ICS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
6. Aristotle and the Caliphs Dream. Aspects of medical translations
David Bennett (formerly NHS and RHUL) . . . . . . . . . . . . . . . . . . . . . . . . . 79
7. Syriac plant names in a fifteenth century Greek glossary (From the
Wellcome Library Books and Manuscripts)
Nikolaj Serikoff (Wellcome Library). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97
8. The Reception of Galens Art of medicine in the Syriac Book of medicines
Siam Bhayro (Exeter) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
9. Medieval hospital formularies: Byzantium and Islam compared
Peregrine Horden (RHUL and Oxford) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
10. Cancerous cells, Neanderthal DNA and the tradition of Byzantine medicine. Textual criticism in philology and genomics
Florian Markowetz (Cancer Research UK Cambridge and University of
Cambridge) and Barbara Zipser (RHUL) . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Acknowledgements
This volume originates from a conference on Byzantine Medical Manuals in
Context, held in central London on the 19th of September 2009. The conference, which was jointly organized by Peregrine Horden and Barbara Zipser,
formed part of a three year project on Byzantine Medical Manuals: Construction and Use based at the Department of History at Royal Holloway University
of London.
First and foremost we would like to thank the Wellcome Trust for funding
the project (grant no. 039752) and the conference (089306), and the RHUL
Department of History for providing a supplementary grant for the conference.
The volume was copy edited by Barbara Packard. Peregrine Horden helped
with various proofs. The Wellcome Library kindly allowed us to publish a
photograph of MSL 60. Caroline Buckley provided two illustrations.
We are honoured to be published by Eikasmos, and also extend our
thanks to the anonymous peer reviewers and the editors. Last but not least
we would like to thank speakers, chairs and audience at the conference, and
all those who discussed the topic with us in other settings.
Barbara Zipser, RHUL
xi
List of figures
Cover illustration: Caroline Buckley
Figure 1: Wellcome Library, London, MSL 60, f. 71v . . . . . . . . . . . . . . . . . . . 121
Figure 2: Illustration: Caroline Buckley . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
xii
List of abbreviations
ANRW Aufstieg und Niedergang der romischen Welt
APAW Abh. d. K
onigl. Preu. Akademie d. Wiss.
BECh Bibliothèque de lEcole

des Chartes
BMCRev Bryn Mawr Classical Review
BRL Bulletin of the John Rylands University of Manchester Library
ByzZ Byzantinische Zeitschrift
DOP Dumbarton Oaks Papers
G&R Greece and Rome
HR History of Religions
JOByz
Jahrbuch der Osterreichischen
Byzantinistik
JHS Journal of Hellenic Studies
JRA Journal of Roman Archaeology
JAOS Journal of the American Oriental Society
JWI Journal of the Warburg and Courtauld Institutes
M&S Medicina e Storia
MedSec Medicina nei Secoli
P&P Past and Present
PAA . . .
PG Patrologia Graeca
REByz Revue des Etudes Byzantines
RSO Rivista degli Studi Orientali
StudMed Studi Medievali
WO Die Welt des Orients
Prefatory note: the uses of medical manuscripts
Much of this collection of articles concerns medical texts that seem eminently practical or useful.1 But what does either of those adjectives mean?
How can historians of medieval Greek medicine, usually working on writings
that now lack contextual evidence of origins and application, establish criteria of usefulness or practicality? Those who study medical remedies of the
pre-modern period are often asked by lay, non-specialist audiences in particular did they work? The question usually presupposes a certain kind of
effectiveness as the yardstick: that of modern, laboratory-based biomedicine
with its high levels of pain relief on which yardstick medieval remedies generally fall short, proving neutral at best.2 In the same way, the question put
(in effect) to scholars of medieval medical manuscripts were they used?
presupposes a certain vision of the texts Sitz im Leben as providing the sole
criterion.3 In effect, if the text did not sit in the consulting room, and if it
was not frequently in the doctors hand, at least between patients and perhaps
during a consultation, it was not practical or useful. What follows is a short
statement of the obvious contrary position. Just as there are many kinds of
effect and effectiveness that can be ascribed to remedies,4 so there are many
kinds of usefulness with respect to medical texts. In each case we need to
1
The conference from which it derives was indeed conceived as a collective

study of Byzantine medieval manuals, or iatrosophia (for the definition of which see
Nuttons contribution in this volume). In what follows I am grateful throughout for
advice from Barbara Zipser and Nigel Wilson.
2
See e.g. B. Brennessel-M.D.C. Drout-R. Gravel, A Reassessment of the Efficacy of Anglo-Saxon Medicine, Cambridge 2006, 183-195.
3
To borrow a term from Biblical form criticism.
4
E. Hsu, Medical Anthropology, Material Culture, and New Directions in
Medical Archaeology, in P.A. Baker-G. Carr (edd.), Practitioners, Practices and Patients: New Approaches to Medical Archaeology and Anthropology, Oxford 2002, 1-15,
HORDEN
separate out the different kinds to arrive at a suitable typology, and we need
to try to arrive at criteria for each kind ideally, in the case of texts, related
to aspects of the manuscripts in which we find them.
The starting point must be that the production of a manuscript either
to create a text or texts for the first time or to copy an existing exemplar had
some perceived value in the Byzantine world. Nigel Wilson has shown that, in
the middle Byzantine period, the cost of a manuscript of some 400 folios was
the equivalent of several months salary for a low-ranking civil servant. That
cost included the production of the writing materials as well as the copyists
skill and time.5 We need not worry here about the exact period or the exact
salary. The point is the order of magnitude. In simple terms of price, the
modern equivalent to producing or commissioning a manuscript was, shall
we say, buying a small car. It was not the equivalent of buying a modern
hardback volume. The best evidence of the preciousness of parchment or
vellum is its re-use the palimpsest. Witness, most strikingly for us today, the
codex containing three new texts of Archimedes.6 Thus, a medical manual
such as that of John the Physician, 90 folios including pinax in one quarto
manuscript, might have cost our civil servant one months salary.7 Even such a
iatrosophion, when copied on vellum, required a number of hides from sheep or
goats (hardly more than four folia per animal)8 and would have taken several
months to write out. And when various types of paper became generally
available as a substitute writing material from the eleventh to twelfth century
onwards, it was not, at least to begin with, very much cheaper in the Byzantine
empire.9 So it can be taken as axiomatic in this context that no text was
suggests that we substitute was it successful? (in meeting patients expectations)
for did it work? (in relieving symptoms or pain) as our main question.
5
N. Wilson, Books and Readers in Byzantium, in Byzantine Books and Bookmen, Washington DC 1975, 3f., with C. Mango, The Availability of Books in the
Byzantine Empire, A.D. 750-850, ibid. 38f.
6
The Archimedes Palimpsest, ed. R. Netz et al., I-II, Cambridge 2011.
7
B. Zipser, John the Physicians Therapeutics. A medical Handbook in vernacular Greek, Leiden-Boston 2009, 18. This estimate is based on manuscript M, the
only complete witness of the text.
8
Wilson, Books and Readers cit. 2.
9
R.S. Bagnall, Early Christian Books in Egypt, Princeton-Oxford 2009, ch. 3,
for the costs of papyrus rolls and codices; J. Bloom, Paper before Print: The History
and Impact of Paper in the Islamic World, New Haven-London 2001.
Prefatory Note: The Uses of Medical Manuscripts
written out, no copy made, automatically without some calculation, however

rapid, however subconscious, of costs and benefits: the cost of the material,
the benefit of the text or texts that it might already bear (in the case of
a palimpsest), the benefit of the text or texts that might be written on it.
With the Archimedes manuscript, the value of the writing material, the lack
of alternative sources of parchment, and the Palestinian scribe-priests need for
a copy of a prayer book clearly trumped any need felt to preserve mathematics
or philosophy from pagan antiquity. That is why, more generally, we find the
filling up of blank pages or half pages in manuscripts: such spaces were a waste
of a rare and costly medium.
Of course, we always have to bear in mind that we are generalising about
what survives, not what once existed about which we can only conjecture.10
Crossings out, or other signs of editing, are unusual in the codices that have
come down to us because for the most part only fair copies, finished products
and their derivatives have been preserved.11 (At least in the medical sphere,
not even the ancient and late antique papyri have revealed anything much
that could be identified as rough draft or working copy.)12 We presume that
authors or compilers, or those re-working some earlier material, roughed out
their texts on papyrus off-cuts, slates, ostraca, or individual palimpsest sheets
of parchment. These were ephemeral: they have not come down to us. Nor, we
may guess, have the large majority of the more polished manuscripts that existed in the Byzantine period. (Think, for a non-literary parallel, of the 50,000
or so lead seals that have survived without the documents they once authenticated.) That attrition strengthens the point that what was deliberately kept,
as distinct from what was accidentally lost, or censored, or destroyed in violent
circumstances (what one might call the Name of the Rose scenario) was what
had some perceived value. We should take it as a working hypothesis that
no writing was kept unthinkingly. If for example Byzantine scribes stopped
copying Galens longer and more theoretical works in the ninth to tenth centuries, so that they survive if at all only in Arabic, that was because these
substantial treatises were no longer thought worth keeping, given the expense
of the parchment and of the copyists time that such works would require,
10
See V. Nutton, Ancient Medicine, Abingdon-New York 2004, ch. 1.

For a rare example of crossing out see B. Zipser, Deleted Text in a
Manuscript: Galen On the eye and the Marc. gr. 276, Galenos III (2009) 107112.
12
R.S. Bagnall, Everyday Writing in the Graeco-Roman East, Berkeley 2011.
11
HORDEN
their unwieldy size, the perceived redundancy of some of their information, or

the novel availability of more compact alternatives.13
Therefore, if we are trying to set out the possible reasons for creating or
keeping a manuscript of medical writings in Byzantium, the simple preservation of the text for its own sake, or merely out of respect for the standing of its
presumed author, is perhaps the least likely even if that author was Galen.
Pure antiquarianism as an explanation should be automatically suspect.
After r e s p e c t in ascending order of usefulness might come l u x u r y or d i s p l a y. Under this heading we could group those manuscripts
commissioned by a patron to show off his (nearly always his) learning, or
prepared by an author, compiler or scribe to attract or retain the interest of
a patron. The copies of Galen in Greek that end up in Italy after the fall
of Constantinople exemplify this category.14 The usefulness here is symbolic
of wealth, of taste, of learning. We should not underestimate the role of
display copies in the Greek medical manuscript tradition.
Then, s c h o l a r s h i p. Medical manuscripts had their uses for scholars
living in the Middle Ages that had nothing to do with medicine. These range
from the literary the text in the manuscript as a model of good style to
what in modern terms would be science, that is, information about the world:
about the environment and the ways humanity is made up (reproduction,
generation, anatomy, physiology). Perhaps a common mistake in hoping to
establish the Sitz im Leben of many medical manuscripts has been to confuse
the scientific and the medical; to assume that the usefulness of medical texts
lay only in therapy.
This is not of course to diminish some t h e r a p e u t i c application as
a substantial category. But the place of a manuscript a text or a collection
of texts in a doctors practice can be just as variable as its non-medical uses.
We must remain alert to the whole range of possibilities, rather than draw
conclusions too hastily.
What are these possibilities? The first is broadly educational to do
with medical training, the making of a medical career (if only part-time). Does
the codex in question suggest a school, an individual teacher, self-instruction?
13
V. Nutton, Galen in Byzantium, in M. Gr

unbart-E. Kislinger-A. MuthesiusD. Stathakopoulos (edd.), Material Culture and Well-Being in Byzantium, Vienna
2007, 174; see also N. Wilson, Aspects of the Transmission of Galen, in G. Cavallo
(ed.), Le strade del testo, Bari 1987, 47-64.
14
Nutton, Galen in Byzantium cit. 175.
Prefatory Note: The Uses of Medical Manuscripts
Does it suggest display here too self-validation, image-making? What does

its size tell us about where it might have been kept and opened? Could it
literally have been a vade mecum or would it have stayed on the shelf for only
occasional reference? What can we infer from its contents about the possible
circumstances of its application? Or from the way the text is laid out, and the
ease with which it would have been possible to find ones way about it?
If that is the range of possibilities, then the task becomes one of trying
to map features of surviving manuscripts on to it not in any hope of precision
but simply to establish a working framework that can be constantly revised as
new evidence is considered. How much theory, of humours or complexions, do
we take to be more redolent of medical training than of clinical practice? Are
commentaries on standard texts a sign of an educational syllabus?15 What
can we infer from the apparent absence of such commentaries? What size of
manuscript is portable on foot, by pack animal? Quarto size or smaller is
genuinely portable; but how would one preserve the precious book from water
damage? How many such small volumes could one easily transport? What
signs of discolouration from handling, even what density of fingerprinting, do
particular folia display?16 What, if anything, a priori (simply as starting point
for investigation) might we infer from the presence of non-medical texts in a
medical manuscript? Are there prefaces that yield any clues about intended
use? What kind of anthology of abbreviated or excerpted material do we find in
the manuscripts and are there any criteria by which we might distinguish their
compilers purposes? What is the significance of linguistic reworking from
classical to demotic ?17 What finding aids are there (indexes, headings,
spacing out, marginal marks of significant passages) and how comprehensive
and how effective would they have been in facilitating ready reference in the
consulting room, the iatreion, or at the bedside? Serious comparative work
has begun on a number of the fronts bound up in that agenda; some of it is
to be found in the chapters below. Yet much remains to be done before we
15
The second version of John the Physician is in effect a commentary on the

earlier one (Zipser, John the Physician cit. 38-41), made as an aid to translation, not
education. Contrast the material mentioned by Nutton in his contribution to this
volume at nn. 5-7.
16
See for example K.M. Rudy, Dirty Books: Quantifying Patterns of Use in
Medieval Manuscripts Using a Densitometer, Journal of Historians of Netherlandish
Art II 1f. on-line (accessed 22 June 2012).
17
Zipser, John the Physician cit. 28-30.
HORDEN
can pronounce with any confidence about when, in what circumstances, a text
was useful.18
Department of History
RHUL, Egham, TW20 0EX, UK
Peregrine Horden
[email protected]
18
G. Cavallo, Il libro come oggetto duso nel mondo Bizantino,
JOByz XXXI/2 (1981) 395-423, has only brief comments on medical texts,
at p. 409.
Byzantine medicine, genres, and the ravages of time
Byzantine medicine has never enjoyed an enthusiastic press. Owsei Temkins formulation, tradition and empiricism, although accurate, cannot compare with the praise lavished upon the much earlier Galen and, still more,
upon Hippocrates, the father of medicine.1 Although it was adopted by
John Scarborough as one of the guiding threads in his introduction to what
still remains the only collection of essays dedicated to the medicine of this period, its vagueness does not encourage one to go further, since it could describe
almost any medical system.2 There is also a worrying dichotomy between the
views of scholars on the medicine in the Early Byzantine period, defined for
convenience as ending with the conquest of Alexandria in 642, and on what
followed. Much work has been done on the first period, which has become a focal point for recent research. By contrast, very little is known about medicine
in Middle and Late Byzantium, which fits only with difficulty into a narrative
of medical progress. A few new technical terms represent a meagre harvest,
and even if pious Christians are credited with the invention of the hospital, the
extent of that contribution, and its development within Byzantium alone, are
both amply contested.3 It is true, as Stephanos Geroulanos has argued, that
all the basic principles of modern surgery can be found in writings preserved in
Byzantine manuscripts, and that some of the recommendations included there
1
O. Temkin, Byzantine medicine: tradition and empiricism, DOP XVI

(1962) 97-115; cited from the republication in O. Temkin, The double Face of Janus
and other Essays in the History of Medicine, Baltimore-London 1977, 202-222.
2
J. Scarborough, Symposium on Byzantine medicine, DOP XXXVIII
(1984) iv-xvi.
3
P. Horden, How medicalised were Byzantine hospitals?, M&S X (2006)
45-74 (repr. in P. Horden, Hospitals and Healing from Antiquity to the Later Middle
Ages, Aldershot 2008, I, 45-74).
NUTTON
can only have come from surgeons who had put them into practice.4 But, at
the same time, much of what is found in Paul of Aegina in the sixth century
derives from surgeons of the time of Galen, four hundred years previously, if
not from Hellenistic Alexandria half a millennium earlier. Preservation and
continuity are both good things, but they cannot by themselves show that
the practical advice, some of it clearly the result of experience, that could be
found in a later book was actually followed.
Even if due allowance is made for the possible practical skills of Late
Byzantine doctors and surgeons, they are often presented in ways that are
calculated to deter all but the brave. Temkins comments are not untypical.
Greek medical manuscripts are replete with shorter or longer texts,
badly composed, badly marked as to beginning or end and often transmitted anonymously or under pseudonyms [. . . ] We encounter the work [of John the Archiatros] in many manuscripts,
usually in such disarray that one cannot help feeling deep sympathy with its future critical editor.5
When the works of learned men are described so unflatteringly, while at the
same time they are contrasted with a populace that preferred at times to put
its faith in a variety of superstitions and deride doctors for choosing medicines
and dietetic regimens over incantations and purifications, one may indeed contemplate dismissing the medical world of Late Byzantium as a degenerate descendant of a once brilliant family. Compared with the Islamic world, the late
medieval West and, still more, the hellenising Renaissance of the sixteenth
century, all of which were stimulated by the new reception of classical Greek
medical ideas, Byzantium might seem to lie in torpor, and the post-Byzantine
world still more so. Familiarity with its classical heritage had long since lost
the power to stimulate new ideas. Almost all that was left was to transmit the
old notions in a convenient form.
But while this verdict does have some truth in it, it is also based on a considerable ignorance of what material does survive from later Byzantium, much
of it accessible only in manuscript, and on an equally skewed perception of the
4
S. Geroulanos, Byzantine surgery, in M. Gr

unbart-E. Kislinger-A. Muthesius-D. Stathakopoulos (edd.), Material Culture and Well-being in Byzantium (4001453), Vienna 2007, 129-134.
5
Temkin, o.c. 219f.
medical literature of earlier Antiquity. Medical historians frequent concern

with the onward march of progress often diverts attention from the context of
medical discovery (or lack of discovery). Medical education, it should not be
forgotten, was largely geared to the production of effective healers, to whom
the theoretical concerns of Galen were not always as interesting or as relevant
as they might seem to modern scholars. Very few European and American
medical schools, down even into the twentieth century, were progressive in the
sense of having a research imperative, and many saw little reason to introduce major changes into their curriculum for decades, if their graduates were
regarded as competent and effective by their patients.
But it should also be admitted that what survives of later Byzantine
medicine differs considerably in one respect from that in Western Europe and
the Muslim world. Although there are a variety of teaching institutions in
the Byzantine world, ranging from schools to hospitals, and although much of
learned Byzantine medical writing depends on a confrontation with the traditional texts of Hippocrates and Galen, commentaries on these authors are few
and far between. Sibylle Ihm, in her list of commentaries on ancient medical
authors, cites only one commentary on or summary of a Galenic or Hippocratic text by a Byzantine author who lived after the tenth century.6 This
may be simply because texts like this have not been published in a modern
edition, as Ihms remark that she has seen some unedited and unstudied commentaries suggests.7 The ancient writings were certainly copied and recopied,
and the lecture was always the main means of instruction, but it remains
striking that few, if any, of these manuscripts contain the type of marginalia
derived from lectures that are so abundant in Western manuscripts from the
thirteenth century onwards. There are no comments, no points singled out
for further discussion, no exposition of the views of the lecturer, and no investigation of wider matters to give an appropriate context or to develop new
ideas. It may be also significant that Ihms sole example of a later commentary
was produced in the mid-sixteenth century by a doctor who had studied and
spent many years in Northern Italy before returning to teach at the Patriarchal
School in Constantinople.8 Another fragment of late Byzantine teaching, what
6
S. Ihm, Clavis Commentariorum der antiken medizinischen Texte, LeidenBoston-Cologne 2002, 223-258.
7
Ibid. 27.
8
Ibid. 234f. no. 288. See now P. Demont, Le Tub. Mb 23 et quelques medecins
grecs de Chios en relation avec le patriarcat de Constantinople dans les annees 1560-
10
NUTTON
appears to be the introduction to a commentary on Galens Art of medicine,

also reveals Italian influences on the part of its anonymous author, who is
probably to be identified with John Argyropoulos, who taught medicine at
the Kral Hospital in the years immediately before the fall of Constantinople.9
In place of lecture notes, we find the iatrosophion, a somewhat vague
term that can encompass almost anything from a series of prescriptions to
a medical compendium. It is primarily a collection of therapies, sometimes,
like one of the versions of John the archiatros, written in a vernacular form
of Greek, sometimes associated with a hospital, sometimes appealing to the
great names of Antiquity, sometimes containing chants and charms, sometimes
in the form of an exposition, sometimes in that of questions and answers.10
Antonio Garzya, almost in desperation, describes it as a mixed genre in which
different styles are mingled, if not entangled together.11
Such handbooks, common in the period of Turkish rule, have a long
history. They can be traced back through Theophanes Chrysobalantes and
1580, in S. David-E. Gely (edd.), Troika. Parcours antiques. Melanges offerts à
Michel Woronoff, I, Besancon 2007, 323-330; Le commentaire des c. 5 et 6 du traite
hippocratique des Humeurs dans le Tub. Mb 23 (suite et fin: fol. 41v-57v), in V.
Boudon-Millot-A. Garzya-J. Jouanna-A. Roselli (edd.), Storia della tradizione e edizione dei medici greci. Atti del VI Colloquio internazionale, Paris 12-14 aprile 2008,
Naples 2010, 353-412. For my purposes it does not matter whether Crusius learned
man from Chios is to be identified with Hermodoros Lestarchos or not, ibid. 354-357,
since both individuals spent much time in Italy.
9
A.M. Ieraci Bio, Giovanni Argiropoulo e un inedito commento anonimo a
Galeno (Ars med. 1, 1a-b7) nel Vat. Gr. 285, in Boudon-Millot cit. 271-290. The
source of the manuscript goes back to the circle of doctors and students at the Kral
hospital around 1450. It remains to be seen whether this so far unique late Byzantine
commentary on the Ars represents a chance survivor of a long and continuous tradition
of exegesis of this text, or the (re-) introduction into Constantinople of Italian methods
of commentary.
10
B. Zipser, John the Physicians Therapeutics. A medical Handbook in vernacular Greek, Leiden-Boston 2009.
11
A. Garzya, Pour ledition des iatrosophia demotiques, in A. Garzya-J.
Jouanna (edd.), Trasmissione e ecdotica dei testi medici greci. Atti del IV Convegno internazionalle, Parigi 17-19 maggio 2001, Naples 2003, 165-171. See also A.
Touwaide, Byzantine hospital manuals (iatrosophia) as a source for the study of therapeutics, in B.S. Bowers (ed.), The medieval Hospital and medical Practice, Aldershot
2007, 147-173: 149-157; and Bennett in this volume.
11
Paul of Nicaea to the Greek authors who lie at the basis of the early medieval
Latin versions of the Liber Byzantii and the so-called Liber tertius of Galen,
and still further back to Paul of Aegina in the early seventh century, Alexander of Tralles a century earlier, and to Oribasius in the second half of fourth
century.12 Whether Galens compressed Art of medicine should be counted
as an early iatrosophion is doubtful, but his Method of healing, for Glaucon,
would certainly fall under this heading, as would the Introduction to Medicine,
preserved among the Galenic Corpus, despite their larger theoretical content.
Other similar tracts survive only in the form of quotations or of fragments.
Galen himself praised the , Helpful advice, of Athenaeus of Attaleia,
who probably lived in the early years of the Roman Empire, as the best general
treatise by a modern author, although its length, in at least 30 books, might
seem too great for an iatrosophion.13 The same may go for another fragmentary work of some fame, the long treatise, perhaps entitled The layman, by
Rufus of Ephesus, which provides useful advice on a huge variety of conditions
for those who did not wish or who were unable to gain access to a doctor. It
was excerpted in Late Antiquity and, even more so, by the Arabs.14 Its overall
message, that it was in the interests of everyone to know at least the basics of
medicine in case no expert help was at hand, was repeated by Latin writers
in Late Antiquity. They, of course, had a Latin model at their disposal in the
writings of Cornelius Celsus, who lived around the middle of the first century
AD. His On medicine formed part of an encyclopaedia of Artes (Arts and Sciences), and was intended to introduce the best of earlier (and largely Greek)
medicine to cultivated Roman readers who might at times themselves be re
J.A.M. Sonderkamp, Untersuchungen zur Uberlieferung
der Schriften des
Theophanes Chrysobalantes (sog. Theophanes Nonnos), Bonn 1987; Paolo di Nicea,
Manuale medico, ed. A.M. Ieraci Bio, Naples 1996; K.-D. Fischer, Der Liber Byzantii,
ein unver
offentlichtes griechisches therapeutisches Handbuch in lateinischer Ubersetzung, in C. Deroux (ed.), Maladie et maladies dans les textes latins antiques et
medievaux, Brussels 1998, 276-294; idem Galeni qui fertur Ad Glauconem Liber tertius ad fidem codicis Vindocinensis 109, in I. Garofalo-A. Roselli (edd.), Galenismo e
medicina tardoantica. Fonti greche, latine e arabe, Naples 2003, 285-346.
13
Gal. De elem. I 6 (I 457 K
uhn).
14
An edition of the fragments of this treatise is a desideratum. They appear
under different headings in the standard edition of Rufus by Daremberg and Ruelle,
but much more is available in Arabic, see M. Ullmann, Die arabischen Uberlieferung
der Schriften von Rufus von Ephesos, ANRW II/37.2 (1994) 1293-1349.
12
12
NUTTON
quired to intervene directly or through a doctor to treat the ills of their familia.
Like the sections of the work on oratory or agriculture, it made no claims to
novelty, although that did not prevent Celsus from occasional comments that
reveal his own acquaintance with diseases and treatment.15
A similar description might well be applied to the so-called Michigan
medical codex, which contains the remains of 13 folios of recipes and prescriptions and which, although copied in the fourth century, seems to have been
composed at least two centuries before that. Its owner added in his own hand
in the margin a further twenty recipes, thus showing that this was a living collection intended for immediate practical use.16 Similar personal additions can
be found in another papyrus (whose fragments are split between Strasbourg
and Manchester), but this time they appear to have been added at different
times.17
Handbooks of this sort, focussing on recipes and practical advice, are
thus not confined to Byzantium.18 They can be found in Classical Antiquity,
where they cover almost the whole stylistic spectrum. At one end, some writers preferred to use verse to make their technical message more palatable to a
learned but non-technical audience who could appreciate the technical sophistication of a Servilius Damocrates, a Greek doctor in early imperial Rome, or
the Hellenistic poet Nicander of Colophon.19 The medical poems, alas now
lost, by Heraclitus of Rhodiapolis (Southern Turkey) earned him honours from
a variety of cultural institutions in the Greek world of the Roman Empire as
15
G. Sabbah-P. Mudry, La medicine de Celse. Aspects historiques, scien

tifiques et litteraires, Saint-Etienne
1994; C. Schulze, Aulus Cornelius Celsus Arzt
oder Laie? Autor, Konzept und Addressaten der De medicina libri octo, Trier 1999.
16
L.C. Youtie, P. Michigan XVII. The Michigan medical Codex (P.Mich. 758
= P.Mich. inv. 21), Atlanta 1996.
17
M.H. Marganne, Lophthalmologie dans lEgypte

greco-romaine daprès les
papyrus litteraries grecs, Leiden 1994, 133-146.
18
For another example, the Introduction ascribed to Galen, see the discussion
by Caroline Petit in this volume.
19
For Nicander, see J.M. Jacques, Nicandre, Oeuvres, Paris 2002; idem,
Medecine et poesie : Nicandre de Colophon et ses poèmes iologiques, in J. Jouanna-J.
Leclant (edd.), La medecine grecque antique. Cahiers de la Villa, Paris 2004, 109124; for Damocrates, see E. Samama, Les medecins dans le monde grec. Sources
epigraphiques sur la naissance dun corps medical, Geneva 2003, 362 n. 85. An edition of his poems is being prepared by Sabine Vogt.
13
well as the soubriquet of the Homer of medical poetry.20 They used standard
medical and pharmacological sources. The only surviving poem in Latin from
the Early Roman Empire, that of Q. Serenus, has been thought to represent
a reworking of information from Celsus.21 While these poems, like those of
the Hadrianic doctor and litterateur, Marcellus of Side, display the technical
virtuosity of their authors, some proponents also claimed a practical advantage.22 Writing in verse reduced the (considerable) risk of mistakes in copying,
for the rules of metre imposed a check on the intrusion of alien material or the
accidental alteration of numbers and proportions of ingredients. That did not,
however, prevent later scribes from writing out poems as if they were prose.23
But these elegant compositions stand at one end of a spectrum of medical writing.24 They are meant to be appreciated for their literary quality as
much as, if not more than, their medical efficacy. They demand an equally
learned and literate audience who would appreciate them. Similarly, Galens
blend of medicine and philosophy, although far from unique to him, depends
on both patients and practitioners believing that this type of theoretical training is essential for the would-be doctor, and, equally importantly, it requires
institutions, as well as individuals, engaged in teaching others to follow the
subtleties of Galens argumentation. At the very least, it demands a cultured
urban, if not a metropolitan, setting, a situation not everywhere to be found
in the Late Byzantine Empire. The Galenic model of medical education did
continue to be followed, and his writings to be copied, but it was far from
ubiquitous.
20
Samama, o.c. 396f.

J.H. Phillips, The Liber medicinalis Quinti Sereni, in P. Mudry-J. Pigeaud
(edd.), Les ecoles medicales `
a Rome, Geneva 1991, 179-186.
22
V. Nutton, Ancient Medicine, London 2004, 211.
23
Gal. De comp. med. sec. gen. VII 8 (XIV 988 K
uhn). Leipzig, UB 1127, ff.
87r-97r and 1220, ff. 56r-77r contain the poems of Serenus written out as prose.
24
The variety of medical writing and the fragile process of transmission are
emphasized by Nutton, o.c. 1-6, 265-271, and by M.-H. Marganne, Le livre medical dans le monde greco-romain, Liège 2004. For the medical texts on papyri, see
I. Andorlini, Prescriptions and practice in Greek medical papyri from Egypt, in H.
Froschauer-C. R
omer (edd.), Zwischen Magie und Wissenschaft. Arzte
und Heilkunst
in den Papyri aus Agypten, Vienna 2007, 23-34.

21
14
NUTTON
However, while Galen often spent his time expounding theoretical questions of doubtful relevance to everyday medical practice, his drug books illustrate neatly the very different ways in which recipe collections were assembled
and organised. His four surviving large-scale works deal with simples and with
three differently arranged collections of recipes one listing compound drugs
by their types, one by the regions of the body they affected, and the third
devoted solely to antidotes. A fourth tract, dealing with remedies that were
easy to procure, Euporista, is lost except for quotations in Syriac and Arabic,
and the treatise of the same name that goes under his name in Greek is a later
pastiche. These are literary works, clearly organised and at times incorporating much personal and expository material. But at the same time they are all
dependent on the writings of earlier pharmacologists, often copied out verbatim by Galens assistants, to which he subsequently added other recipes as he
found them. (The fact that later copies would have incorporated his additions
makes any attempt at dating problematical.)25 These earlier sources in their
turn depend on a variety of sources and authors, not all of them identified as
doctors or pharmacologists.
How Galen went about collecting his recipes is described at length in his
treatise :26
I was convinced that I had in my possession more remarkable drug
recipes than anyone in the whole of the Roman world, some put in
my way by chance, others that I had added myself. Fate put both
sorts in my path in two ways. Firstly, there was a rich man back
home who wanted so much to acquire a knowledge of significant
drugs that he was prepared to buy some recipes for over a hundred
gold pieces. He succeeded so well that he could purchase recipes
that were highly regarded not only by all the modern doctors in
Asia, but also by the ancients. The recipes for all of these drugs
were carefully preserved in two folded parchment volumes,27 which
25
S. Vogt, Drugs and pharmacology, in R.J. Hankinson (ed.), The Cambridge

Companion to Galen, Cambridge 2008, 304-322.
26
Gal. 31-37: 11-13 ed. Jouanna. For an explanation of the text
behind this translation, see my publication on this work in P.N. Singer (ed.), Galen,
Psychological writings, Cambridge 2011.
27
Although the words could mean two leather folders (into which the doctor
could place slips of papyrus), it is more likely that this is an early example of a
15
one of his heirs, a very close friend of mine, voluntarily handed

over to me without being asked. This was my first piece of luck
in acquiring an abundance of remedies, and now read of the second. When I first came to Rome in my thirty-third year [162 AD],
I found a fellow-citizen and schoolmate of mine called Teuthras
already living in the city. He had obtained the parchments belonging to a doctor called Eumenes,28 who was himself also from
Pergamum and was a particular connoisseur of many drugs among
all doctors. These recipes had been collected in one place from
all over the world during his travels before he settled in Rome
until his death. Teuthras, who died in the first visitation of the
plague, left them to me a little after what I have said was my first
arrival in Rome. If someone had a remarkable drug, I could get
hold of it without difficulty by drawing on these collections and
offering two or three similar ones in exchange. Not only were all
these parchments destroyed in the fire and I still thought this
parchment (charta pergamena) book in codex form, used, like merchants and lawyers
notebooks, for private reference purposes, and increasingly for wider circulation, M.C.
Nicholls, Parchment codices in a text of Galen, G&R LVII (2010) 378-386. At De
comp. med. sec. loc. I 1 (XIV 423 K
uhn), Galen mentions receiving a recipe for early
baldness from a friend, Claudianus, who had found it in a folded parchment after
the death of its owner. Claudianus had used it himself on patients and thought highly
of it. The parallels between the two incidents suggest that Claudianus was Galens
anonymous friend.
28
The sole manuscript of this treatise has Eumenes, but no doctor called
Eumenes is recorded in Galen, whereas he cites a Eudemus of Pergamum, a skilled
practitioner of this branch of medicine, several times in connection with remedies,
Methodus Medendi VI 6 (X 454 K
uhn); De comp. med. sec. loc. IX 5 (XIII 291 K
uhn)
and Ant. II 14 (XIV 185 K
uhn), the last being a metrical reworking of a rare antidote.
The name Eumenes could easily have slipped in through a reminiscence of Eumenes,
the name of two famous rulers of Pergamum. But Galens comments at Methodus
Medendi VI 6 (X 454 K
uhn) imply that he had known Eudemus in Pergamum before
both of them left Asia Minor, i.e. around 150, before Galen left for Corinth and
Alexandria, or, roughly ten years later, in the interval between Galens return from
Alexandria and his first move to Rome in 162. The latter date would be more suited
to Galens description of himself as a practitioner, but would permit only a short time
to elapse between Eudemus move to Rome and his death. Although emendation is
tempting, this tight chronology advises caution.
16
NUTTON
was no great loss but so also was my treatise on the composition

of drugs, which I had prepared with great precision and where I
described how one might make up the most important drugs; only
my prescriptions for a few drugs were preserved because they had
been earlier given to others.
This remarkable description applies equally to the way in which prescription
manuals were compiled and added to in Galens day and in later Byzantium.
It was a continuous process as new recipes were added in the margins or at the
ends of manuscripts. Until the advent of printing fixed a particular form, these
iatrosophia were preeminently fluid in character. Even afterwards, manuals of
this sort, whether in East or West, might go through several reprintings and
reissues, sometimes involving the addition of new material and sometimes not.
This paper has argued that a concentration on the major authors to
survive from Classical Antiquity gives a misleading impression of the variety of
genres of medical writing that existed then and obscures the continuity between
the age of Galen and that of Late Byzantium. This impression is made even
more erroneous if attention is paid only to the great names of the classical
period, and still more if, as classicists have often done, pseudonymous and
anonymous texts are largely disregarded. Two examples suffice to demonstrate
the survival of different genres through the centuries. Texts that take the
form of questions and answers, erotapokrisis, can be found in both papyri and
manuscripts, covering surgery as well as general therapeutics.29 They have
a didactic purpose, although Napthali Lewis suggestion that they bore some
relation to a formal examination for practice, although attractive, cannot be
substantiated on the present evidence.30 This was a genre that passed into the
world of Islam, for one of the most influential of the early textbooks was the
Questions and answers of Hunain ibn Ishaq, fl. 860. This treatise, however,
was redacted a little later into a more usual expository form, and, translated
into Latin as the Isagoge of Iohannitius it became a standard introductory text
in the later medieval Western universities.31
29
A.E. Hanson, Text and context in papyrus catechisms on afflictions of the

head, in A. Garzya-J. Jouanna, o.c. 199-217.
30
N. Lewis, Exemption of physicians from liturgies, Bulletin of the American
Papyrological Society II (1965) 87-89.
31
D. Jacquart, A laube de la renaissance medicale des XIe-XIIe siècles: l Isagoge Johannitii et son traducteur, BECh CXLIV (1986) 209-240 (repr. in eadem,
17
Secondly, students of Latin medical writings from Late Antiquity are

also familiar with a group of short tracts praising the medicinal value of the
vulture, the badger, and the herba vettonica, all of which are credited with a
wide range of curative properties. Arsenio Ferraces Rodriguez has recently also
edited another text of similar date, listing those of the peony.32 But there is a
much earlier Greek example, albeit one that survives today only through the
medium of a word-for-word Latin translation.33 The pseudo-Galenic treatise
on the centaury is a rare survival of a treatise by a doctor sympathetic to
Methodism, who enjoyed a career similar to that of Galen himself around 180
AD. The unknown author came to Rome from Asia Minor, where his brother
Papias, also a doctor, still lived, and was instructed by one of the leading
doctors in the city, from whom he heard of this remarkable panacea. The
text is short, and may have owed its fortunate survival to the fact that it was
written during Galens lifetime and may have been found among his papers,
but it is a reminder of a genre that has almost vanished from our sources.
Seen from the perspective of the wordy, theoretical Galen, or even the
writers of the massive compilations of extracts, Oribasius, Aetius and Paul,
Byzantine medicine is a disappointment. The confusion over the meaning of
iatrosophion, as well as the miscellaneous therapies that can be found there,
only draws attention to the haphazard growth and development of this type
of literature. But a wider focus suggests two reasons why this downgrading of
later Byzantine medicine is unwise. Firstly, what survives intact from Antiquity is the result of a long process of selection and recopying, and fragments in
manuscript or on papyrus, as well as quotations, reveal a much greater variety
La science medicale occidentale entre deux renaissances (XIe s.-XVe s.), Aldershot
1997, I, 209-240).
32
A. Ferraces Rodrguez, Magia y terapia: edici
on, traducci
on y comentario
de un fragmento tardoantiquo sobre la peonia, in A. Ferraces Rodrguez (ed.), Fitozooterapia antigua y altomedieval: textos y doctrinas, A Coru
na 2009, 147-170, with
references to similar writings on plants. For animals, R. Mohler, Epistula de vulture,
Untersuchungen zu einer organotherapeutischen Drogenmonographie des Fr
uhmittelalters, Pattensen 1990.
33
V. Nutton, De virtute centaureae: a neglected Methodist text?, in D.
Langslow-B. Maire (edd.), Body, Disease and Treatment in a changing World. Latin
Texts and Contexts in ancient and medieval Medicine, Lausanne 2010, 213-222.
18
NUTTON
of genres in the early Roman Empire than could be presumed simply from a
glance at the major works of Galen.34 Time has been a great eraser.
Secondly, Late Byzantine therapeutic texts are, on their own terms, far
from dully repetitive. They are living texts, bringing together remedies that
were believed to work, some even centuries old, and adding new material to
the stock either within the text or in the margins. These marginal additions in
turn might then be taken into the text when it was recopied. Analysing and
appreciating such practical texts is much harder, and less often attempted,
than examining more theoretical writings. Distinguishing one therapy from
another is far from easy when both come with the same label proven by
experience, but it is only by the minute comparison of details, rather than in
a search for any new overarching theory, that one will in the end come to a
judicious assessment of Late Byzantine medicine.
225, Sandpit Lane
St Albans AL4 0BT, UK
34
Vivian Nutton
[email protected]
For iatromathematics, derided by Galen, see I. Andorlini, Un anonimo del

genere degli Iatromathematika, Garzya-Jouanna (edd.), o.c. 1-23. Magical charms and
spells co-exist with herbal or dietary remedies in the works of respectable authors
such as Marcellus of Side and Julius Africanus, and it is clear from the attack on such
remedies by both Galen and, earlier, Scribonius Largus that not every doctor accepted
their strictures. The whole question of borrowing from non-Greek sources, less visible
in Classical Antiquity than in late Byzantine medical writings, as discussed in the
contributions by Caroline Petit and David Bennett in this volume, requires much
further study.
Disease and where to treat it: a Byzantine vade mecum
At an unknown time in Late Antiquity, probably in the late fifth-earlysixth century a man must have spent considerable time carving the following
text on a column in the West facade of the Parthenon (by then a Christian
church):
Holy Mary full of grace command the one who is having intercourse with my bride to fall ill with hernia and let me be a physician so that I may find the opportunity to cut off his member
(rhombos).1
Mary sends the disease, but a doctor is called to treat it. This little vignette
suggests a number of features regarding the concepts of disease and their possible treatments in the Byzantine millennium. Plurality and the parallel existence and importance of seemingly opposing structures, both conceptual (and,
as I will discuss below, actual ones buildings as well) are some of the key
aspects of the following overview. Boundaries between high and lowbrow, sacred and secular are constantly put to question and require us to consider more
inclusive categories.2
The question of what constitutes disease and the reasons that cause it
is naturally one of great antiquity. Concepts and definitions received from
1
A.K. Orlandos-L. Branouses, , Athens 1973,

Nr 9, 5. See A. Kaldellis, The Christian Parthenon. Classicism and Pilgrimage in
Byzantine Athens, Cambridge 2009, 78, who offers a slightly different interpretation
of the inscription: he sees rhombos as the bandage worn by hernia patients.
2
The following is meant as a tour dhorizon. As a result footnotes are by no
means exhaustive and only indicate some fundamental and/or recent works on the
topics at hand.
20
STATHAKOPOULOS
Ancient Greece and the early Church Fathers remained valid throughout the
Byzantine era; innovation in this field if at all came in the form of combination.
Disease aetiology can be divided in two broad categories: natural and
supernatural. According to the first category, health and disease are manifestations of processes within the body: the former is a result of the balance of the
four humours, substances that occur naturally in the body (blood, bile, black
bile and phlegm) and disease is a manifestation of their imbalance. Health, the
natural state of the body, can be restored with the help of exogenous actions,
but disease is not understood or described as having any metaphysical aetiology or existence. This is a concept that ultimately goes back to the authors of
the Hippocratic Corpus.3 As elaborated on and expanded by Galen it became
the dominant medical concept at the outset of Antiquity.4 Its prevalence in
Late Antique and Byzantine secular medical texts is expectedly absolute; it
would be pointless to attempt to document it. However, the popularity and
authority of this concept was not limited to medical authors as we can see
by looking at a non-medical text, the twelfth century satire Timarion. Its
protagonist is befallen by a violent fever followed by an inflammation of the
liver and the most appalling dysentery, causing him to vomit up his elemental
bile along with pure blood. Exhausted, he falls asleep, only to be snatched by
two demonic spirits who whisper over his head:
This is the man who lost the fourth of his constituent elements by
vomiting up all his bile. He cannot be allowed to go on living on
the strength of the remaining three. Asclepius and Hippocrates
have said as much in the decree they wrote down and posted up in
Hades whereby no man, even if his body be in good shape, shall
go on living if he has been deprived of one of his four elements.5
V. Nutton, Ancient Medicine, London 2004, 53-102.

O. Temkin, Galenism: Rise and Decline of a Medical Philosophy, Ithaca
1973; V. Nutton, Galen in Byzantium, in M. Gr
unbart-E. Kislinger-A. Muthesius-D.
Stathakopoulos (edd.), Material Culture and Well-Being in Byzantium, Vienna 2007,
151-156.
5
R. Romano, Pseudo-Luciano, Timarione, Napoli 1974, ch. 13, vv. 357-363.
Translation: B. Baldwin, Timarion, Detroit 1984, 51.
4
Disease and where to treat it
21
This type of disease aetiology carried very positive connotations for

the longest time; the fact that it is usually termed rational6 by historians
of medicine suggests that alternatives to it are irrational and has somewhat
skewed our understanding and appreciation of it in its favour. And yet, the
most ancient ideas about disease and perhaps the most popular throughout
the pre-modern world at that perceive it as having supernatural causes: a
supernatural being sends a disease and is (often exclusively) capable of lifting
it. This belief is common in, to name two of the most fundamental texts in
Byzantium, both Homer and the Old Testament, but can be found in older,
non-Greek texts, Egyptian as well as Babylonian.7 The supernatural agent
of disease can be (a) god, or a demon and the trigger for it is predominately
human transgression against which the disease is sent as punishment, chastisement or purification of the pollution that the transgression has caused.8
In the case of magically induced diseases, a demonic force is often manipulated into causing disease through the medium of a spell.9 This religious view
of disease has understandably little to do with medicine as such physicians
would be useless and only appeasing the supernatural being responsible for
it (or in the case of magic countering the demonic manipulation) would help.
Prayers, offerings, and in some cases exorcism were the means to counter disease. Nowhere is this clearer than in instances of epidemics causing massive
mortality. Here is what the sixth-century author John Malalas reports on the
outbreak of the first wave of the Justinianic Plague:
The Lord God saw that mans transgressions had multiplied and
he caused the overthrow of man on the earth, leading to his destruction in all cities and lands. [...] Gods compassion (eusplachnia) lasted at Byzantion for two months.10
6
J. Longrigg, Greek Rational Medicine: Philosophy and Medicine from Alcmaeon to the Alexandrians, London 1993.
7
W. von Siebenthal, Krankheit als Folge der S
unde, Hannover 1950; R.
Parker, Miasma: Pollution and Purification in Early Greek Religion, Oxford 1996,
235-256.
8
For a lexicological exploration see A. Willi, and : etymological and
sociocultural observations on the concept of disease and divine (dis)favour in ancient
Greece, JHS CXXVIII (2008) 153-171.
9
G. Vican, Art, Medicine, and Magic in Early Byzantium, DOP XXXVIII
(1984) 65-86.
10
John Malalas. Chronographia, ed. I. Thurn, Berlin-New York 2000, XVIII
22
STATHAKOPOULOS
In the cases of demonic-induced diseases (whether the demons were conjured

by a human or not) we can equally observe similar traits, although the system
may not be as straightforward and clear. Demonic forces can plague humans
with possession and disease simply as a result of their hatred for them; such
instances of disease can only be countered through rituals and in a Christian
context only through the agency of God, whether in person (as in the case
of Christs exorcisms in the New Testament)11 or through the invocation of
a sacred name or person. The case of the female demon Gyllou or Gellou or
Abyzou (which derives from the Sumerian abzu, the primeval sea of chaos)
is a telling example. This is what the widely popular Testament of Solomon
(dated to 1-3rd centuries, but whose sources are much more ancient) records
regarding the demon:
There came before me [Solomon] one who had the shape of a
woman but she possessed as one of her traits the form of one with
dishevelled hair. I said to her, Who are you?. . . She replied, Obyzouth. I do not rest at night, but travel around all the world visiting women and, divining the hour [when they give birth], I search
[for them] and strangle their newborn infants. . . I, Solomon, said
to her, Tell me, evil spirit, by what angel are you thwarted? She
said to me, By the angel Raphael; and when women give birth,
write my name on a piece of papyrus and I shall flee from them
to the other world.12
The belief in Gyllou spans a period from Sappho until the fifteenth century
and well beyond.13
92 (407); translation: The Chronicle of John Malalas, trans. E. Jeffreys-M. JeffreysR. Scott et al., Melbourne 1986, 286f.; cf. D. Stathakopoulos, Making Use of the
Plague: Readings in Sixth Century History, in K. Belke-E. Kislinger-A. K
ulzer-M.A.
Stassinopoulou (edd.), Byzantina Mediterranea. Festschrift f
ur Johannes Koder zum
65. Geburtstag, Vienna-Cologne-Weimar 2007, 633-639.
11
The language of the miracles evokes older, mostly Hellenistic exorcisms with
the preponderance of the verb (to rebuke), see Ev. Marc. 1,25: (
, ); 4,39 (
, , ); 9,25 ( ).
12
J. Spier, Medieval Byzantine Magical Amulets and Their Tradition,
JWI LVI (1993) 35.
13
Ibid. 25-35.
23
To divide ideas on the perception of disease so neatly and rigidly into two
opposing camps may appeal to the taxonomist in us, but in reality boundaries
were always blurred. What resulted was not, however, a conceptual or medical
syncretism, but the parallel existence of seemingly conflicting concepts. This
may well have been connected to practice: a patient would seek healing from
a secular physician professing a natural understanding of disease and offer
prayers to a divinity asking to be rid of his/her affliction. John Duffy calls
this a lack of contradiction between piety and practicality.14
At the same time, it is equally a question of sources and more specifically
of genre. In hagiography, for instance, disease is on the heels of sin the
causality is never ambiguous. One example from the life of St Theodore of
Sykeon will suffice: a certain priest called Paul was brought to the holy man
with his right hip dislocated and suffering great pain. No medicine had helped
up to that point. Theodore bade him to return to his country and make peace
with the man he had angered. Paul initially denied this, but at Theodores
insistence he confessed that he had quarrelled with his abbot (sin of pride)
and was ultimately cured.15
In hagiography, classical aetiologies of disease were acknowledged, if only
to be refuted in favour of a divine origin. This is what the sixth century
archbishop of Thessalonica John writes regarding a recent outbreak of plague:
Neither babies, nor women, nor the flower of youth, nor men of
arm-bearing and city-service age were spared from the disease:
only the elderly escaped. God had desired thus so that no one
would be able to claim that the epidemic had been a natural
phenomenon caused by the corruption of the air (dyskrasia), and
not a divine punishment.16
Liturgical texts are also unambiguous about disease: here are some examples
from Late Byzantine prayers against disease and plagues: a tempest of sins
14
J. Duy, Byzantine medicine in the sixth and seventh centuries: aspects of

teaching and practice, DOP XXXVIII (1984) 25.
15
Vie de Theodore de Sykeon (= Vita S. Theodori Sykeota), ed. A.J. Festugière, Subsidia Hagiographica XLVIII (1970) I, Mir 81.
16
P. Lemerle, Les plus anciens recueils des miracles de Saint Demetrius et la
penetration des Slavs dans les Balkans, I, Paris 1978, 33, 77.
24
STATHAKOPOULOS
has pushed me into the depths of disease and constant pains, like storms,
drive me forth or in another prayer These [diseases and plagues] are the
salaries of our wickedness, this is the appropriate result of our evil and impure
actions.17
We can assume then that while some genres favoured monocausal aetiologies of disease (secular medical texts, or hagiography to mention the proponents of the two opposite poles of the spectrum), other types of text allow
for a more inclusive approach. It is to such cases that I would like to turn to
at this point.
It has been recently argued that the New Testament displays such tendencies. Ferngren argues persuasively against current orthodoxy, which sees
disease in the New Testament as largely ascribed to demonic actions, that the
general underlying basis in the text is that disease has natural causes, or that
it at least is neither a result of sin nor of demonic presence.18 When asked
whose sins caused a man to be born blind (Ev. Io. 9,1-7), Jesus answers: Neither hath this man sinned, nor his parents: but that the works of God should
be made manifest in him.
The early Christian period, however, was not characterized by the same
conceptual openness or complexity. Perhaps as a result of the historical struggles of Christianity and its subsequent political and social domination or the
focus on the charismatic holy men19 healers and exorcists par excellence
at least from the fourth century onwards we witness not so much the revival
of the concept of illness as a result of sin, but its definite domination.
Yet this dominance created problems as well. Christian thinkers were
faced with new questions as they elaborated on matters of faith. How could
evil or disease and especially plagues be justified with the notion of the good
and kind God? Basil of Caesarea devoted a whole homily to this topic aptly
titled That God is not the cause of evil.20 It is a complex text that vehemently
denies that God is responsible for evil a direct result of human sin. It is very
interesting that Basil reverts quite often to medical metaphors to illustrate
his point: As the physician must cut, burn or even mutilate an affected part
17
J. Goar, Euchologion sive rituale graecorum, Venice 1730, reprint Graz 1960,
543, 554.
18
G.B. Ferngren, Medicine and Health Care in Early Christianity, Baltimore

2009, 43-51.
19
Ibid. 59-63.
20
Bas. Quod Deus non est auctor malorum PG XXXI 329-354.
25
to prevent the disease from spreading to the rest of the body, so God inflicts
disease as a paideusis (education), disease being the lesser evil compared to
sin and its results, eternal damnation.
The persistence of questions on theodicy and by implication on the nature of disease can be traced by looking at the pastoral genre of Questions and
Answers. One of the earliest and most influential of these texts, the Quaestiones of Anastasius of Sinai is fairly detailed on the subject. Question 28
concerns basically the question of random death:21 why do certain evil men or
pagans enjoy a long life while pious Christians die young in sum, the faithful
question the fair judgement of God. Anastasius begins by elaborately stating
that the topic is difficult and that one should rather not probe into such matters, then he provides a long answer (in 26 chapters and some 9 printed pages)
which, in my mind, can be seen as an effort to harmonize Greek ideas about
the natural world and the patristic tradition.
According to Anastasius, God created heaven and earth, the sea and all
visible creation out of four elements fire, water, air and earth from which
both the humans and animals were also made. The nature of these bodies
was given to these four elements to govern, lead and administrate as generals
or horse riders. As such all earthly bodies, plants, animals, all animate and
inanimate beings are led and made and transformed and given life or destroyed
by the mixture of the elements (5). The elements thus control the birth,
growth, disease (6) and death of the body (17). The passage is replete with
humoral terminology: natural life is governed by the good mixture (eukrasia)
of the elements (7); infants die because they are warm beings (thermotera
onta) and, spring brings an abundance of blood (an element connected to that
season since Antiquity) (22). But God has given the knowledge of medicine
to humans and has prepared all the herbs and all sorts of therapeutic things,
with which physicians in many cases with Gods providence save humans
from death (15).22
21
Edition: Anastasii Sinaitae, Queastiones et Responsiones, edd. M. RichardJ. Munitiz, Turnhout-Leuven 2006, 42-48; See also J.A. Munitiz, The Predetermination of Death: The Contribution of Anastasios of Sinai and Nikephoros Blemmydes
to a Perennial Byzantine Problem, DOP LV (2001) 9-20. On Anastasius medical ideas see M.-H. Congourdeau, Medecine et theologie chez Anastase le Sinate,
medecin, moine et didascale, in V. Boudon-Millot-B. Pouderon (edd.), Les Pères de
lEglise face `
a la science medicale de leur temps, Paris 2005, 287-298.
22
These questions naturally preoccupied Byzantine thinkers throughout the
26
STATHAKOPOULOS
The issue of plagues is touched upon, but a different question (66) is

specifically devoted to it: Is it impossible to save oneself from a plague by
fleeing, is the question. Again, Anastasius stresses the mysterious nature of
such occurrences (mysterion bathy) and proceeds: if the plague is sent by God,
it cannot be understood nor escaped; if, however, the mortality ensues as a
result of (polluted) air, which often occurs at the beginning of spring especially
in densely populated, wet and woody landscapes, it can be escaped through
flight from the afflicted area. No-one should think that such a plague ensues
merely from the air without Gods agency, for the winds and all creation obey
Gods command.23
Such statements are, as perhaps expected, not entirely new: they are
in fact quite evocative of stoic thought. Epictetus, for example, wrote the
following:
All things obey and serve the Cosmos [in other words, God], both
earth and sea, and sun, and the other stars, and the plants and
animals of earth; obedient to it also is our body, both in sickness
and in health, when the Cosmos wishes, both in youth and in old
age, and when passing through all the other changes.24
Similar instances of complexity can be found in secular medical texts suggesting if not the acceptance of alternative ideas on disease aetiology, at least the
recognition that such ideas existed and were widely popular. Two of the most
well-known medical authors of Late Antiquity, Aetius of Amida and Alexander
of Tralleis include magical recipes in their work.25 The latter at least is quite
apologetic about such inclusions: There are persons incapable of adhering to
a strict regimen or of tolerating drugs, and they therefore compel us to use
entire period of the Empire. It will suffice here to point to H. Beck, Vorsehung
und Vorherbestimmung in der theologischen Literatur der Byzantiner, Rome 1937,
especially 219-253 and the edition of Nikephoros Blemmydes, Gegen die Vorherbes
timmung der Todesstunde, Einleitung, Text, Ubersetzung
und Kommentar von W.
Lackner, Athens-Leiden 1985, xliii-xciv.
23
Anastasios, o.c. 118f.
24
Epict. fr. 3: Epictetus, with an English transl. by W.A. Oldfather, II,
Cambridge, Mass.-London 1928, 442-445.
25
Aetius of Amida. Libri Medicinales I-VIII, ed. A. Olivieri, CMG VIII/1,
Berlin 1935, especially in book II.
27
occult remedies and amulets in gout.26 In another passage he writes that

when digging for the dangerously poisonous plant hyoscyamus (henbane), the
action has to be done before dawn, when the moon is in the sign of Aquarius
or Pisces while chanting I conjure you by the great name Iaoth, Sabaoth, the
God who fixed the earth and stopped the sea and so on.27 If anything, such
instances suggest a parallel world of supernatural causes of disease that could
not be cured by rational means.28
In the liturgy of St Basil the brethren ask of God to grant winds of
good temperance the term used eukratous evokes the balance of the humours (eukrasia) that is crucial for health in the Hippocratic system. Without
wanting to overstretch this point, in my mind it is quite clear that ideas about
disease in the Byzantine world suggest the co-existence of secular pre-Christian
and Christian ideas especially after at the latest the sixth century when
physicians no longer were equated with paganism after the Justinianic purge
of 545/6.29
It follows that the plurality of ideas about disease and the healers that
represented them would mean a plurality of places in which such a healing
would be sought. Class and financial means certainly played a crucial part
in the choice of where to seek healing, but again a combination of ideas and
practices emerges as the dominant pattern.
Whoever could afford it, would always seek a physician this is by far
the most widespread practice. In most cases this would mean to be visited
by the physician at ones bedside and being tended for the duration of ones
illness by him or, in the case of wealthier and socially more important individuals, such as emperors and their kin by a, at times heterogeneous, team
26
Alexander von Tralles. Therapeutika, ed. T. Puschmann, I-II, Vienna 1878.

Translation: O. Temkin, Hippocrates in a World of Pagans and Christians, BaltimoreLondon 1991, 233.
27
Alexander II 585 Puschmann; Temkin, o.c. 234.
28
For further passages see: Puschmann II 319.4 ff., 473.30 ff. and Temkin, o.c.
232-236.
29
John of Ephesus in Pseudo-Dionysius of Tel-Mahre. Chronicle, Part III,
trans. with notes and introduction by W. Witakowski, Liverpool 1996, 71; see also
M. Whitby, John of Ephesus and the Pagans, in M. Salamon (ed.), Paganism in the
Later Roman Empire and in Byzantium, Cracow 1991, 111-131.
28
STATHAKOPOULOS
of physicians.30 There is no need, I think to refer to such examples; they

are abundant throughout the Byzantine period. It may only be important to
stress that the patient could distinguish between the physicians summoned to
attend him/her and that such gradations are important social markers, both
for patients and professionals. For example, when the Venetian merchant
Giacomo Badoer was suffering from a skin disease in 1436 he had a Greek
doctor visit him (the miedego Panardo or Panarido, probably John Panaretos, who was possibly working at a hospital, the Kral Xenon at the time) and
also another, Latin doctor, maistro Lucha, was also consulted and was paid
more than the Byzantine practitioner.31 When, however, a little later in the
year Badoers own apprentice, Antonio Bragadin, fell ill, only a barber was
summoned to bleed him not the distinguished physicians he had called for
himself.32 Renowned physicians, such as John Zacharias in the late thirteenth
century, certainly made house calls33 and we can expect that they constituted
an important part of a physicians income.34
If a specific case made this necessary or possible, a patient could seek
a physician at the latters home this does, however, suggest a difference
in social status.35 The physicians home is connected to the question of the
iatreion (or its Latin counterpart, the taberna medica), which could designate
just that or possibly also a clinic or a physicians private practice: patients
30
On physician at the bedside of emperors: see A. Kazhdan, The image of the

medical doctor in Byzantine literature of the 10th to 12th centuries, DOP XXXVIII
(1984) 43-51; D. Bennett, Medical practice and manuscripts in Byzantium, Social
History of Medicine XXXIII (2000) 279-291.
31
Il Libro dei conti di Giacomo Badoer (Constantinopoli, 1436-1140), edd. U.
Dorini-T. Bertelè, Rome 1956, 82, 376.
32
Il libro dei conti, o.c. 52; on Badoer and physicians see K.-P. Matschke,
Gesellschaft und Krankheit: Bewohner und Besucher Konstantinopels w
ahrend und
nach der Pest von 1436, Vizantijski Vremennik LV (1998) 48-53.
33
Zacharias House calls: John Zacharias, De Urinis in Physici et Medici
Graeci Minores, ed. J.L. Ideler, II, Berlin 1843, 50-52, 62f., 92f., 154-156, 162f.,
165-167, 181-183, 186f. See A. Hohlweg, Johannes Aktuarios: Leben - Bildung und
Ausbildung - De Methodo Medendi, ByzZ LXXVI (1983) 302-321.
34
T.S. Miller, The birth of the Hospital in the Byzantine Empire, Baltimore,
2
1997 , 149.
35
On this question I rely on G. Harig, Zum Problem Krankenhaus in der
Antike, Klio LIII (1971) 179-195: 179-188.
29
could stay there overnight or for a short period of time after their treatment.36
But in the Byzantine period the situation is not quite as clear. Certainly in
the case of John Zacharias, the iatreion he mentions seems to designate a clinic
for ambulant treatment not his own practice.37
There is admittedly very little evidence for the last possible type of
healing associated with a private physician: treatment by correspondence as
encountered in a short treatise on gout from the Palaiologan period sent to
a patient.38 This was a very popular method in the modern period, but the
Byzantine case may be something different: a patient, who having consulted
his physician asked for additional information, which was then sent in the form
of a letter.
The place for seeking healing in Byzantium that has received most attention by scholars is the hospital. And again within this discussion the lions
share of the information we have (and on which a lot of the reconstructions
are based) is taken up by one text, the Typikon of the Pantokrator monastery
in Constantinople.39 This is not the place to report the long debates over this
question; I will only try to present some data on the specific topic of patients
seeking healing in hospitals. First I must stress the enormous discrepancies
between institutions that are called hospitals (xenones or nosokomeia) and
those we can safely assume were medicalised philanthropic institutions that
provided healing. P. Horden has scrutinized the evidence: very few from the
large numbers of hospitals that populate the bibliography would seem to withstand that test.40 Silence is of course not an argument against the presence of
36
Scholars await with great anticipation the publication of the findings of

such an iatreion, the so-called Domus del chirurgo in Rimini; see R.P.J. Jackson,
The Domus del chirurgo at Rimini: an interim account of the medical assemblage,
JRA XVI (2003) 312-321.
37
Zacharias, De Urinis IV 12,22 (as in n. 32), Ideler o.c. II 95f.; Hohlweg o.c.
309.
38
G. Schmalzbauer, Medizinish-di
atetisches u
ber die Podagra aus sp
atyzan
tinischer Zeit, Jahrbuch der Osterreichischen
Byzantinistik XXIII (1974) 229-243.
39
Typikon edition: P. Gautier, Le typikon du Christ Sauveur Pantocrator,
REByz XXXII (1974) 1-125; Translation: Byzantine Monastic Foundation Documents (thereafter BMFD): A Complete Translation of the Surviving Founders Typika
and Testaments, edd. J.Ph. Thomas-A. Constantinides Hero, Washington, DC 2001,
725-781, with bibliography; see also Miller, o.c. chapters 2 and 8.
40
P. Horden, How Medicalised were Byzantine Hospitals?, M&S X (2005)
30
STATHAKOPOULOS
doctors, nor, however, one in its favour. Second, cases of patients treated in
hospitals are very rare; we almost never get a glimpse of patients being actually treated in hospitals: the Pantokrator Typikon designates what should be
done; mentions in hagiography can be rightly questioned, as their objectives
lay elsewhere and not with the documentation of what would be their competition. For those institutions over which we stand on firmer ground we can
generally say that they were few, small in size and very often short-lived. For
the late Byzantine period at least we can safely assume that such institutions
were emergency hospitals: patients were meant to spend a short time therein
(brachy kairou) and then return home cured and healthy.41 So, who sought
out healing in hospitals? Judging from the Pantokrator Typikon the poor,
and perhaps, as we can reconstruct from some scattered mentions, some professionals with connections to the monasteries to which these hospitals were
annexed;42 probably not those who could afford private physicians.
The largest category regarding places of healing concerns healing shrines
and churches. Mentions to them are expectedly abundant in hagiographic
texts, as the accounts of miraculous healing were both a testimony to a shrines
popularity and at the same time a great means of augmenting it. The anargyroi miraculous healers had clear advantages over secular physicians: their
healing was open to all, and in most cases instant and painless. As such their
target audience knew no social boundaries. The burial sites of certain holy
men (and some few women)43 became places of pilgrimage for those seeking
cure of (usually) chronic diseases, often after the patients had tried their luck
(always unsuccessfully) with secular physicians: it will suffice to mention the
collection of miracles of S. Artemius in the seventh century or those celebrating the healings at Zoodochos Pege at Constantinople in the late Byzantine
45-74.
41
This is clearly the case in the xenon of Panteleemon in Constantineople.

Edition: Actes de Lavra III, De 1329 a 1500, edition diplomatique par P. Lemerle
et al., Archives de lAthos X, Paris 1979, doc. 123, 20-26. The same has been
attributed to perhaps the last of the important xenones in Constantinople, the Kral
from the fourteenth century onward; U.B. Birchler-Argyros, Die Quellen zum KralSpital in Konstantinopel, Gesnerus ILV (1988) 419-444: 424.
42
See Miller, o.c. xxi-xxii, 147-152, 165f.
43
G. Vikan, Art, Medicine, and Magic in Early Byzantium, DOP XXXVIII
(1984) 65-86: 66 on St Thecla.
31
period.44 Patients were healed through incubation or immersion in water, but

predominately through their prayers and repentance. Such practices are typologically not new; they hark back to, say, healings at Asklepieia,45 to name just
one example and they enjoyed a long history of popularity in some instances
up to this day if we only think of Lourdes or Tenos in Greece.
But perhaps all the above constituted only a fraction and even a small
one at that of the possible methods and places of healing to which the
majority of the population turned to. These are not places of healing as such
but objects or methods and means, which people used on their own probably
at their own home: spells, amulets and other means of white magic, as well
as folk remedies or even remedies lifted from medical books.46 To chart this,
we need to link texts and artefacts: for example texts of ritual power (mostly
from Egyptian papyri) and normative texts that chart out the disapproval and
prohibition of magical practises (a process that can be observed up to the end
of Byzantine empire) as well as objects (primarily, through not exclusively,
amulets and rings) that evoke such practises.
The following example can illustrate this. Ammianus Marcellinus mentions that a young man was seen in the bath touching alternately with the
fingers of either hand first the marble (of the wall or perhaps the floor) and
then his breast, and to count the seven vowels (of the Greek alphabet) as a
remedy for stomach ache.47 This can be corroborated by a spell to be written
44
See A.-M. Talbot, Pilgrimage to Healing Shrines: The Evidence of Miracle

Accounts, DOP LXI (2002) 153-173. The shrine is still in existence today as a place
of pilgrimage for healing, albeit not without a possible interruption in the eighteenth
century. On Artemius see: The miracles of St. Artemios, by V.S. Crisafulli-J.W.
Nesbitt, Leiden-New York 1997.
45
From the vast literature see: E.J. and L. Edelstein, Asclepius: A Collection
and Interpretation of the Testimonies, I-II, Baltimore 1945; A. Petsalis-Diomidis,
The Body in Space: Visual Dynamics in Graeco-Roman Healing Pilgrimage, in J.
Elsner-I. Rutherford (edd.), Pilgrimage in Graeco-Roman and Early Christian Antiquity. Seeing the Gods, Oxford 2005, 183-218; I. Csepregi, The miracles of Saints
Cosmas and Damian: Characteristics of Dream Healing, Annual of Medieval Studies
at CEU VIII (2002) 89-121.
46
See Bennett in this volume.
47
Amm. Marc. XXIX 2,28; Ammianus Marcellinus, with an English transl.
by J.C. Rolfe, III, Cambridge, Mass.-London 1939, 233; see M.Th. Fogen, Balsamon
on magic: from Roman secular law to Byzantine canon law, in H. Maguire (ed.),
32
STATHAKOPOULOS
for an amulet against stomach ache and headache from the seventh century
found in a Coptic manuscript.48
Ive already mentioned the magical recipes included in the works of
Aetius and Alexander, which include amulets but such occurrences are of
course not limited to the Late Antique world. Michael Psellos, who dabbled
into esoteric and magical texts throughout his career, mentions in passing the
manufacture of apotropaic figurines by Chaldaeans for the purpose of warding off diseases. But he refuses to divulge the method by which the various
substances are to be mixed and the figurines made. He openly voices his concern that, if they lacked discretion, his students or readers might pick up the
method, put it to use, and then, in the event of trouble, he would be held
responsible.49
A similar case can be found among the writings of Michael Italikos. As
he was known to possess a large collection of spells and incantations, including
some for the relief of swellings and tumors, some friends consulted him about
the recourse to a sorcerer for the treatment of an incurable malignant ulcer.
He tried to dissuade them, but later wrote back to say he had found an ancient
remedy which he will not write down, but will deliver to his friend orally when
they next meet.50 It is all the more remarkable then, when Italikos later
presents a court physician friend of his with an amulet that can ward off the
plague.51
But not all of these self-help remedies were necessarily magical or outlawed.
Byzantines regularly visited hot springs for curative reasons for example the imperial mother and son couple Eirene and Constantine VI in
Prussa/Bursa in 797.52 Or they used drugs that they could have obtained
from a physician or another person with medical knowledge or, although
Byzantine Magic, Washington, DC 1996, 99-115: 111f.
48
Ancient Christian Magic. Coptic Texts of Ritual Power, edd. M. Meyer-R.
Smith, New York 1994, 87, lines 126-32, with bibliography.
49
See J. Duffy, Reactions of Two Byzantine Intellectuals to the Theory and
Practice of Magic, in Maguire, Byzantine Magic cit. 83-97.
50
Duffy, ibid.
51
K.-H. Leven, Gelehrter Aberglauben, abergl
aubische Gelehrte? Michael Italikos, ein Amulett gegen die Pest und die Medizin in Byzanz, Das Mittelalter X
(2005) 33-63: 56-63.
52
Theophanes. Chronographia, ed. C. De Boor, Leipzig 1883, 471.
33
this is not stated as such that they could have put together following a
medical recipe. Such cases can be found in the sources, ranging, for example,
from the lives of early ascetics, who used any knowledge of drugs among them
to help themselves (raw eggs against consumption, or vinegar for an aching
spleen)53 to the well-known fifteenth century intellectual (and hypochondriac)
John Chortasmenos, who consumes a potion prepared by specialists54 or
Giacomo Badoer, whom we have seen above summoning physicians, and who
records in his account book that he often bought drugs from Greek shops or
occasionally used syrups on his own.55
This short and eclectic overview has hopefully highlighted the presence
of different concepts and practices in understanding and treating disease. As
the Byzantines themselves apparently saw no contradiction in seeking solace
from illness from both physicians and holy men, we should also adopt more
inclusive categories. From self-help to lofty trained physicians proficient in
Galenism and from hospitals to healing shrines, diseases, whether brought on
by an imbalance of the humours, God, or Gylou were understood and treated
in a variety of ways.
Centre for Hellenic Studies
KCL, London WC2R 2LS, UK
53
Dionysios Stathakopoulos
[email protected]
H.J. Magoulias, The Lives of Saints as Sources of Data for the History of
Byzantine Medicine in the Sixth and Seventh Centuries, ByzZ LVII (1964) 127-150:
145.
54
H. Hunger, Allzumenschliches aus dem Privatleben eines Byzantiners. Tagebuchnotizen des Hypochonders Johannes Chortasmenos, in P. Wirth (ed.), Polychronion, Festschrift Franz D
olger zum 75. Geburtstag, Heidelberg 1966, 244-252: 245,
lines 25f.
55
Il Libro dei conti (as in n. 31), 280, 376.
Two Latin Pre-Salernitan medical manuals, the Liber passionalis

and the Tereoperica (Ps. Petroncellus)
I. Should1 you wish to produce a medical book, or any book for that
matter, various choices present themselves. You might want to write it yourself, and start from scratch; or perhaps you prefer to use material produced by
others, and then adapt it in a number of different ways; or you could also combine both methods by producing a book based on a copy already in circulation,
but linking and arranging what you have found in a novel way.
The medical books which are the focus of our conference seem to belong
in the category I mentioned last; but even here, there are more ways than
one of going about your business. Your task will be much easier if you are
just compiling a collection of recipes; after all, there is not much difference
between recipes of the culinary and of the medical variety. Galens two major
collections of medical recipes were arranged either by kind ( ) or by
place ( ), and of course excerpts from both entered medieval Latin
recipe collections.
Another type, different from these two Galenic examples, could be an
innovation introduced by Roman writers. I at least cannot think of any Greek
book, or work, in this category I will sketch briefly. Here, it is the main ingredient or, to give it its technical name, the chief and sometimes only simple
or single drug that is the key for arranging the recipes or curae. The most
important and most widely disseminated work of this type was the Herbarius
1
This article forms part of my research undertaken in the context of the

project Textos medicos latinos de los siglos VI-XI: transmisi
on, edici
on crtica y estudio lexico of the Spanish Ministerio de Educacion y Ciencia-FEDER (HUM200613605-C02-01), directed by Arsenio Ferraces Rodrguez, Universidade da Coru
na, Departamento de Filoloxa Espa
nola e Latina, Facultade de Filoloxa, A Coru
na.
36
FISCHER
Apulei, the main item in a string of various shorter and longer treatises.2 Its
totally spurious association with the second-century Latin writer Apuleius of
Madaura, author of a long novel known in English as The Golden Ass, but,
more importantly for our purpose today, of a work called De magia, where
Apuleius main aim was denying any personal knowledge and practice of the
magical arts. Nevertheless, the existence of De magia must have been the
reason why a book listing recipes based on medicinal plants was attributed to
him. Running to roughly 130 chapters and detailing the medical uses of as
many plants, it forms the central core of a collection of treatises all arranged
on the same principle, i.e. of listing recipes by one specific simple drug. This
collection (its German name is Herbariencorpus) further comprises two shortish treatises on specific simples, one on the plant betony (De herba Vettonica)
and the other on the beaver (De taxone), as well as a book on the medical
properties of parts taken from other animals (including man), attributed to an
otherwise unknown Sextus Placidus Papyriensis, 32 chapters in all. It is occasionally followed by a selection from Dioscorides famous account on drugs in
five books ( ) in Latin translation, under the title De herbis
femininis.3 On the whole, I see the Elder Plinys Natural History rather than
Dioscorides as the ultimate source, or perhaps better inspiration, for this type
of medical manual, as could be demonstrated if we compare the entries for
particular plants in both works in detail.
All printed in Antonii Musae de herba Vettonica liber, Pseudoapulei Herbarius, Anonymi de taxone liber, Sexti Placiti liber medicinae ex animalibus etc., edd.
E. Howald-H.E. Sigerist, CML IV, Lipsiae-Berolini 1927. For Placitus, Ferraces
Rodrguez suggests (very convincingly) Placidus. Parts of this corpus are currently
being edited again.
3
Arsenio Ferraces Rodrguez prepared a critical edition for the PhD thesis
at the end of 2011 (her chapter and paragraph numbering and divisions differ from
mine and De Renzis) to replace that by Heinrich F. Kastner, Pseudo-Dioscoridis De
herbis femininis, Hermes XXXI (1896) 578-636 and XXXII (1897) 160. For a fuller
bibliography on this and other Latin texts, compare the entries in Bibliographie des
textes medicaux latins. Antiquite et haut moyen
age, sous la dir. de G. SabbahP.-P.
CorsettiK.-D. Fischer, pref. de M.D. Grmek, Saint-Etienne

1987 [but really published
in 1988], and my Bibliographie des textes medicaux latins. Antiquite et haut moyen
age. Premier supplement 1986-1999, Saint-Etienne

2000.
Two Latin Pre-Salernitan medical manuals
37
But this is not the topic that I wish to address here. Rather, I will
now turn to discussing medical manuals which were distinct from all those
mentioned earlier by offering, in addition to recipes, an account of the signs
and symptoms of a given disease, before they outline therapy and set forth
various recipes and cures. The well-known prototype of this genre in Greek
would be Oribasius in his (Collectiones medicae), of whose
seventy (or seventy-two) books only a fraction has survived, although the
abbreviated versions deriving from them, the Synopsis in nine books for his son
Eustathius, likewise a doctor, and the four books of Euporista for Oribasiuss
friend Eunapius, the famous fourth-century orator, are extant both in Greek
and in at least two Latin translations.
Oribasiuss modus operandi was twofold. He may (a) use a complete
chapter4 from one of his source texts, or (b) compose a chapter himself by
excerpting a running text, as we can see for instance in his account of dietetics
for children up to the age of fourteen (coll. med. lib. inc. 35). All of that chapter
was taken from Galens De sanitate tuenda I 10-12,5 but Oribasius produced
a new Galenic text by the process of cutting and pasting. His procedure
becomes a little more intricate in the following ch. 36, which again relies on
Galen alone.6 The next two chapters, on dietetics for children (lib. inc. 37f.),
are attributed to Mnesitheus and to Rufus of Ephesus, respectively. But since
the original writings of both doctors are no longer extant, we are unable to
discern whether Oribasius resorted once more to an extensive cut-and-paste
job, as he did with Galen, or whether he lifted these chapters as complete
units from the works of Mnesitheus and Rufus, still current and available to
him in the fourth century AD.7
4
The important question, at what point a division into chapters was introduced in medical texts, cannot be considered here.
5
Oribasius had, after all, compiled an anthology (now lost) from Galen alone
before he started work on the Collectiones.
6
The different chapter headings, either author (in the genitive) plus title, or
from such-and-such an author plus title, may indicate different techniques, but I am
unaware of any pertinent studies addressing this question.
7
There is a study on Oribasius technique when dealing with Rufus work
on the diseases of the kidneys and the bladder by Alexander Sideras, Aetius und
Oribasius. Ihre gemeinsamen Exzerpte aus der Schrift des Rufus von Ephesos Uber

die Nieren- und Blasenkrankheiten und ihr Abh
angigkeitsverh
altnis, ByzZ LXVII
(1974) 110-130; according to him, both depend on another writer, Philagrius.
38
FISCHER
But then, Oribasius was not the first writer to produce a medical anthology, and certainly not the last. I base my assertion, i.e. that Oribasius
had forerunners about whom we know nothing, on a work that is rarely cited
by medical historians, because it is, strictly speaking, veterinary rather than
medical in content. The fourth-century Latin version of this work, preserved
in two manuscripts from the second half of the fifteenth century, is commonly
referred to as Mulomedicina Chironis,8 and there we find indeed evidence for
cutting and pasting very much along the lines seen in Oribasius. I consider it
rather unlikely that the veterinary compiler of the Mulomedicina Chironis was
the inventor of this procedure; rather, as we can observe more often than once,
veterinary medicine took its lead, when it aspired to higher things and scientific prestige, from human medicine. Nevertheless, we shall probably never
know which medical anthologies provided that model, because they have not
reached us; this may well be because other, later anthologies, were considered
more useful and of greater practical value.
Two of the Latin translations of Oribasius9 contain additional material
not included in the Greek, and this was in all likelihood excerpted from Latin
sources. When it comes from authors who had written in Greek, we must
assume that these excerpts were taken from Latin translations of their works,
rather than translated on the spot for inclusion in the Latin Oribasius. Among
authors represented there who had written their works in Latin, we find Celsus,
Theodorus Priscianus, and Caelius Aurelianus. Paragraphs and chapters from
their works were obviously incorporated in those places where the need was felt
for a more extensive and detailed treatment not provided by Oribasius, since we
are, after all, looking here at the shorter versions of the complete Collectiones,
i.e. the Synopsis and the Euporista. I shall not dwell on the intriguing and,
as far as I am aware, hitherto unexplored topic of Greek sources present in
Latin translation in the Oribasius Latinus, apart from just mentioning the
8
Cf. my Bemerkungen zu den Autorennamen und zum Aufbau der Mulomedicina Chironis und anderer medizinischer Sammelwerke, in V. Ortoleva-M.R. Petringa
(edd.), La veterinaria antica e medievale (testi greci, latini, arabi e romanzi). Atti
del II Convegno internazionale, Catania 3-5 ottobre 2007, Lugano 2009, 109-117.
9
Mainly in the Aa version, but there are also additions in the La version.
The only complete edition can be found in volumes 5 and 6 of uvres dOribase,
Paris 1876; for details and partial editions see the Bibliographie des textes medicaux
latins (see n. 2).
39
overlap between some passages in the so-called Eclogae of Oribasius, found in

additions to the Latin Synopsis.
The best-documented case of Greek authors surviving in Latin translations of other treatises are the long excerpts on diseases of the stomach,
the bowels and the spleen occurring in the Latin translation of Alexander of
Tralles. They could derive from more extensive Latin versions of the works of
the two Greek physicians Philumenus and Philagrius. At least this seems to
be the common opinion, which may be modified as David Langslow progresses
with his study of the Latin text of Alexander of Tralles of which he will, for
the first time, provide a critical edition.
II. With Alexander of Tralles, who lived in the sixth century and practised, as a Greek physician, mostly in Rome, we have moved from late antiquity
to the early Middle Ages. The Latin medical texts, or compilations, or collections, that I am about to discuss could have been in circulation by this time,
i.e. in the second half of the sixth century, but there is neither internal nor
external evidence to help us with dating them.
A salient feature of the transmission of medical texts in the early Middle
Ages, given currency by Gerhard Baader, is referred to in German as Korpus
uberlieferung, which means that a number of (usually shorter) tracts occur
in the same order and arrangement again and again in several manuscripts.10
Augusto Beccaria had described an old canon of Hippocratic and Galenic
works, which can serve as a good example. I want to mention, because we will
return to this collection later, the series of six books consisting of Aurelius,
Esculapius, two books of Galen, Ad Glauconem, followed by the spurious third
book or Liber tertius, and chapters from book II of Theodorus Priscianus, to
which sometimes Alexander of Tralles De podagra was appended. This arrangement probably came about by chance rather than by intention. It must
be called a conglomerate rather than a collection, and nobody has investigated
so far why some chapters from Theodorus Priscianus book II were included
and others from the very same source were left out.
Although this collection was fairly successful, to judge by the number
of surviving manuscripts, it was not the only type current in the early Middle
Ages, which for the present purpose I will define as extending to the year 1000.
10
Unlike an anthology, they were transmitted and copied together by chance

and not by conscious choice.
40
FISCHER
When we focus specifically on medical manuscripts, the early Middle Ages begin more or less with the creation of the Carolingian minuscule just after 780,
because only very few medical manuscripts written before that date11 survive,
and most of these (Oribasius being the exception)12 are just limited to small
scraps, perhaps a few stray leaves.
III. The two works I will now discuss in some detail both have their
earliest manuscript witnesses in the ninth century. They are the Liber passionalis and the Tereoperica. The earliest manuscript, no longer complete,
of the Liber passionalis is Berlin, Staatsbibliothek Preuischer Kulturbesitz,
Phillipp. 1790.13 It was written in France in the first half of the ninth century,
and it may not be a coincidence that the earliest manuscript of the Tereoperica,
Paris, Bibl. Nat., lat. 11219,14 is likewise of French origin; it was written just a
little later, around the middle of the ninth century, perhaps at Saint-Denis, and
came to be owned by the abbey of Echternach, today on the Luxembourg side
of a little stream that marks the border between Germany and Luxembourg.
Both Tereoperica and Liber passionalis are rather bewildering titles that
should be explained. Tereoperica is real genuine Greek; it represents the badly
mangled form of Therapeutica. A Liber passionalis was usually one containing
the lives of martyred saints; in our case, however, passio is not passion or
martyrdom, but simply another word for disease. I have always found Tereoperica rather unwieldy and prefer making reference to this work by the name
of its spurious Salernitan author Petroncellus, to whom it was attributed as
Practica Petroncelli in the only edition published to date, that by Salvatore
11
They are listed and described in the volumes of E.A. Lowes Codices Latini
Antiquiores (= CLA, the manuscripts are numbered consecutively), I-XI and Supplement, Oxford 1934-1971, where the chief interest is palaeographic.
12
Par. lat. 10233, CLA 592, end of seventh century.
13
Whenever feasible, I use this manuscript. Compare also my study Die
Quellen des Liber passionalis, in Arsenio Ferraces Rodrguez (ed.), Tradici
on griega
y textos medicos latinos en el perodo presalernitano. Actas del VIII Coloquio internacional Textos Medicos Latinos Antiguos, A Coru
na, 2-4 septiembre 2004, A
Coru
na 2007, 105-125, with further bibliography on this unpublished text.
14
My quotations from Tereoperica/Petroncellus follow my own transcription
of this manuscript (now accessible online at Gallica) rather than De Renzis text;
I have also introduced subdivisions of De Renzis chapters. Laura Lopez Figueroa,
Universidade de Santiago de Compostela, is preparing a new edition of the Latin text,
and Danielle Maion, Universit`
a di Udine, of the medieval English translation.
41
de Renzi in his Collectio Salernitana (IV, Bologna 1967). De Renzi used a

much later Paris manuscript (Bibl. Nat., lat. 14025) which contains not one
but two books of Petroncellus; the earlier, ninth-century Paris manuscript lat.
11219 transmits only de Renzis book 1 of Petroncellus. This was later to be
translated into middle English as Peri didaxeon.15
The Liber passionalis relies on one particular text as a backbone for its
structure. This text was, as I argued in a paper delivered in Oxford more
than ten years ago,16 the book of Esculapius on chronic diseases, because the
Liber passionalis usually17 follows the sequence of chapters in Esculapius and
fleshes out this skeleton with material taken from other sources.18 The book
of Esculapius covers, in forty-six chapters, the same material as Caelius Aurelianus in his Chronic Diseases, so much so that Esculapius was believed to be
a condensed version of Caelius Aurelianus five books. Rather, their common
source is Soranuss treatise on chronic diseases, in the case of Esculapius enriched with excerpts from other works, among which we can only identify De
speciali significatione diaeticarum passionum by Caelius Aurelianus.
But to return to the Liber passionalis. For quite substantial portions of
15
The problem presented by the existence of a second book of Petroncellus

in some later manuscripts and its authenticity cannot be discussed here. What De
Renzi calls the third book of Petroncellus (Collectio Salernitana IV, 290f.) is
the surviving fragment of another Latin translation of the Synopsis of Oribasius, of
which I am preparing an edition (Pseudo-Democritus). The edition by Max Loweneck, Peri didaxeon. Eine Sammlung von Rezepten in englischer Sprache aus dem
11./12. Jahrhundert, nach einer Handschrift des Britischen Museums [Harley 6258b]
herausgegeben, Erlangen 1896, is the most recent one available; for ch. 36 and 37,
see pp. 24-27 (Latin and English on facing pages). As far as I know, nobody has
given any thought to the rationale behind the selections from Petroncellus that were,
presumably, made by the translator, or the person responsible for his Latin exemplar.
On Cassius Felix and the Tereoperica, see Anne Fraisse, Cassius Felix, De medic
ina. Edition
critique, traduction, recherches philologiques et historiques, Paris 2002,
LXXVf.
16
Dr Monks Medical Digest, Social History of Medicine XIII (2000) 239251.
17
I consider deviations from this as mainly due to mechanical problems in the
transmission of the Liber passionalis.
18
Cf. my article Die Quellen des Liber passionalis, 105-125 (see note 12). This
supersedes Roses analysis in Die Meermann-Handschriften des Sir Thomas Phillipps,
Berlin 1893, 367-369.
42
FISCHER
its text, I have been unable to identify the source used by the excerptor. Even
Valentin Rose was content in such cases with just referring to a hypothetical
source he calls Aa, shorthand perhaps for Auctor antiquus or antiquissimus
or alter -; but we should not despair. Our knowledge of early medieval medical texts continues to grow through first editions of works vaguely known but
never printed before. On the Greek side, examples are the Byzantine medical
manual attributed to an otherwise unknown Paul of Nicaea,19 or the Anonymus Parisinus,20 important especially because of its doxographical content;
and while Barbara Zipsers edition of John the Physician is now well known to
our readers,21 the Prontuario medico attributed to Demetrios Pepagomenos,22
short but belonging to the same genre, or the Manuale della salute of Ailios Promotos,23 might not even have been heard of by everybody. In Latin,
there is the pseudogalenic Liber tertius,24 and some day in the not too distant
future, I hope to publish an edition of the Liber Byzantii. The last two are
Latin translations of Greek originals not extant in manuscripts, or at least not
identified so far.
In a number of cases, Paul of Nicaea and the Liber Byzantii provide
parallel material for the Liber passionalis; and we also notice some overlapping
with Petroncellus. At present, we are rather at a loss whether these parallels
should be seen as indicative of direct borrowing either way or are just indicative
19
Paolo di Nicaea. Manuale medico, testo edito per la prima volta, con introduzione, apparato critico, traduzione e note a c. di Anna Maria Ieraci Bio, Napoli
1996.
20
Anonymi medici De morbis acutis et chroniis, ed. with comm. by Ivan Garofalo, transl. into English by Brian Fuchs, Leiden 1997.
21
Barbara Zipser, John the Physicians Therapeutics. A medical Handbook in
vernacular Greek, Leiden-Boston 2009. Cf. also Ioanna Oikonomu-Agorastu, Kritische
Erstausgabe des Rezeptbuchs des Cod. Par. gr. 2316, f. 348v-374v, PhD thesis Koln
1981, Thessaloniki 1982, who lists some more editions of iatrosophia.
22
Demetrio Pepagomeno. Prontuario medico, a.c. di Maria Capone Ciollaro,
Napoli 2003.
23
Elio Promoto Alessandrino. Manuale della salute, a.c. di Daria Crismani,
Alessandria 2002.
24
Galeni qui fertur ad Glauconem. Liber tertius ad fidem codicis Vindocinensis 109, editionem curavit Klaus-Dietrich Fischer, in I. Garofalo-A. Roselli (edd.),
Galenismo e Medicina tardoantica. Fonti greche, latine e arabe. Atti del Seminario
Internazionale di Siena 2002, Napoli 2003, 283-346.
43
of the use of one or several common sources, sources which, sadly, still elude
us. One source, however, that we are able to identify with confidence are the
Chronic Diseases of Caelius Aurelianus. Since this work has not survived in
any complete manuscript only a few folios from the manuscript on which
the Renaissance edition was based are still extant , such excerpts acquire far
greater importance and value than they would otherwise possess, since even
in their mangled and debased state they may contain some clues which then
allow us to correct Caelius Aurelianus text.
For those portions of the Liber passionalis where a source remains unidentified, the conclusion must be that these represent fragments of medical works
from antiquity, in most cases I would suggest Greek works not preserved elsewhere, and it is certainly surprising that the study of these fragments has not
attracted more interest from scholars.
The structure of the Liber passionalis is usually simpler and more straightforward than that of Petroncellus. The Liber passionalis may be seen as a
simpler kind of patchwork rather than a tapestry, if I may borrow a simile.
I would now like to examine one of the more difficult sections, chapter 56
of the Liber passionalis, which deals with an acute disease, or rather several,
under the title De spasmis et tetanicis, On spasm and tetanus. The purpose is
to (a) demonstrate the pattern of borrowings mentioned earlier, (b) to make
you aware of the precarious state of the transmitted Latin text, which seriously
interferes with our understanding and, as a consequence, the identification of
underlying sources, and (c) to introduce you to those fragments of medical
writings not preserved elsewhere. I have chosen a medical condition covered
both in the Liber passionalis and in Petroncellus.
Since Esculapius treatise, as I said earlier, dealt only with chronic diseases, all material for this chapter of the Liber passionalis must needs have
come from somewhere else. As we observe, Theodorus Priscianus and Caelius
Aurelianus were the main contributors for differential diagnosis of the condition, with Theodorus Priscianus providing not just the description or definition
of spasm at the beginning of the chapter, but also a fair chunk of the therapy found in the second half of the chapter. Theodorus Priscianus refers to
tetanus, with its subspecies emprosthotonus and opisthotonus, as spasmus, as
does Aurelius. Other authors differentiate25 and may make a distinction be25
Leo Syn. 2,17f. (Anecdota medica Graeca, e Codicibus mss. expromsit F.Z.
44
FISCHER
tween spasm and tetanus with its subspecies as a sequela of trauma (surgical
interventions are expressly mentioned here),26 or arising without trauma.
We will now look at the two definitions of spasmus that we need for
the beginning of the chapter in the Liber passionalis. Aurelius (20 p. 721
Dbg./Gloss. med. p. 82,8-12 Heiberg) defines spasm as follows:
Spasmus est tensio atque contractio partium corporis cum ualde
acuta strictione atque dolore et a planta usque ceruicem recta tensio propter ualidam flegmonen. Hi ceruicem uix mouent, et si
loqui uoluerint, ceruicis nerui illis dolebunt.
Spasm is a tension and contraction27 of parts of the body with
a very acute drawing together and pain, and a straight tension
from the feet to the neck because of a strong inflammation. Such
patients hardly move their neck, and whenever they want to say
something, they feel a pain in the muscles of their neck.
Theodorus Priscianus (II 30): Spasmus appellatus est ex tensione
uel protractione neruorum et uicinorum musculorum ceruicis quam
maxime inpatienti dolore compatientis, ut etiam ad respondendum
os facile aperire non possint. nam sub hoc titulo spasmi uariae
sunt huius passionis figurae. spasmus per omne corpus extensionem rigidam temporalem quidem habet, sub qua ita tenduntur
ut sub uno schemate iacentes tensione molesta et inflexibili teneantur.
Spasm has its name from a tension and contraction28 of the sinews
and neighbouring muscles of the neck with unbearable pain of the
sufferer, so much so that the patient can only open his mouth with
difficulty to give an answer. Moreover, spasm is the general name
Ermerins, Leiden 1840) deals first with spasms and thereafter with the three kinds of
tetanic spasm, i.e. tetanus, emprosthotonus, and opisthotonus.
26
See also Anon. Par. 7,3,22, see n. 20; Archigen. ap. Aet. VI 39 p. 181,12
Olivieri.
27
Compare Scrib. Larg. 255, on opisthotonus, where we meet the same tensio
et contractio neruorum.
28
Theodorus Priscianus seems to be the only Latin writer to use protractio
rather than contractio, so one would be tempted to correct his text rather than assume
a meaning not attested elsewhere.
45
of the condition which manifests itself in various ways. Spasm

is a stiffening but temporary extension in the whole body, and
patients suffer spasms in such a way that they lie in one position
and are held in disagreeable and unbendable tension.
When examining the first section in the Liber passionalis, it becomes evident that its compiler will have borrowed the first six words of his text from
Aurelius before continuing with Theodorus Priscianus. We are also acutely
aware that we have great trouble to make sense of the words in our two
manuscripts of the Liber passionalis, so much so that we will now turn to
Petroncellus. We notice that both works present the same text with only
insignificant variations, and these will most likely prove to be accidents of
transmission. The decisive argument for linking these two texts is not that
both start with Aurelius and shortly after continue with Theodorus Priscianus.
If we keep reading, we see that the parallels continue right to the end of 56,4 of
the Liber passionalis29 and 36,4 of Petroncellus. Even more important is the
fact that both share a lacuna which starts after contractio habet (Lib. pass.
56,2/Petronc. 36,2), and both continue with atque. Likewise, they share a
portion of the recommendations for therapy, Lib. pass. 56,10-12 and Petronc.
36,5f.
To account for this strong resemblance, we may think of three different
scenarios: A copies B, B copies A, or A and B both draw on the same source C.
Copying from each other seems less likely in the present case, because then we
would expect to see more material common to both works in the many other
chapters of their text. This observation therefore bolsters the hypothesis that
both the Liber passionalis and Petroncellus drew here on a common ancestor,
where the lacuna I mentioned (Lib. pass. 56,3f., Petronc. 36,3f.) was present.
This is also a fact to be taken into account when discussing possible dates
for these three works, i.e. the Liber passionalis, the Tereoperica, and their
(hypothetical) common ancestor.
I wish to exploit the text before and after the lacuna still further. We
read at its end that this disease received its name from the heart. Here, a
section in Isidore of Sevilles Etymologies (Orig. IV 6,11) will come to our
aid:30
29
Chapter numbers correspond to the manuscripts (there are a few variations

between St. Gall and Berlin). I have divided the chapters further for ease of reference.
30
Discussed earlier by Arsenio Ferraces Rodrguez, Aspectos lexicos del Libro
46
FISCHER
Spasmus Latine contractio subita partium aut neruorum cum dolore uehementi. Quam passionem a corde nominatam dixerunt,
qui[!] in nobis principatum uigoris habet. Fit autem duobus modis,
aut ex repletione, aut ex inanitione.31
Spasm, in Latin a sudden contraction of parts or muscles accompanied by violent pain. They claim that it received its name from
the heart, being the chief centre of strength in our bodies. It
(spasm) is produced in two ways, either by (excessive) fullness, or
by (excessive) emptiness.
If you fail to see the connection between spasmus with the word for heart, be
it Greek or Latin, I must confess that I am equally at a loss. Nevertheless,
I remembered having read something similar before, in Caelius Aurelianus
(Diaet. pass. 39):32
Unde nomen cardiaca passio33 cepit? a corde quod Graeci
appellant. hoc enim grauius ex ipsa passione uexari multi adseuerant, alii uero quod sit letalis passio, a corde nominatam dixerunt,
quod in nobis principatum uigoris habet.
Where did the cardiac disease get its name from? From the heart,
which the Greeks call kardia, because many authorities assert that
it is affected by this condition rather badly, while others say that
as a fatal condition it received its name from the heart, which is
the chief source of strength in our bodies.
We might wonder if somebody got Isidores file cards mixed up, which, however, need not have been the case,34 since the same assertion appears in the
Glossarium Ansileubi (Gloss. med. 82,13-19 Heiberg):
IV de las Etimologas en manoscritos medicos altomedievales, Apendice: A vueltas
con el espasmo, in Arsenio Ferraces Rodrguez (ed.), Isidorus medicus. Isidoro de
Sevilla y los textos de medicina, A Coru
na 2005, 95-127, at 97-99 and 124-127.
31
Fitinanitione derives from the commentary Lat-A on the Hippocratic aphorisms, Aph. 2,26. For and cf. also Orib. Syn. VIII 15,1 (with Gal.
Trem. palp. 8. (VII 640f. K
uhn).), Paul. Aeg. III 19 (166,21 and 24 Heiberg).
32
Cf. Cael. Aur. Cel. II 30,62.
33
Cardiacae passionis is the reading of the only manuscript, Aug. CXX, and
printed by Rose, Anecdota cit. II 229.
34
As argued by Ferraces Rodrguez, Aspectos cit. (see n. 29) 124.
47
Spasmus: est tensio atque contractio partium aut neruorum cum

dolore uehementi atque acuto. Alii, quod sit letalis passio, a corde
nominatam dixerunt, quod in nobis principatum uigoris habet.
Hippocrates autem dicit, si [in] spasmo febris superueniat, signum
esse salutis, si autem super febres spasmus, malum.
Spasm is the tension and contraction of parts or muscles with a
sharp strong pain. Other authorities say that is was named after
the heart, because it was a fatal condition, because the heart is
the chief source of strength in our bodies. Hippocrates, however,
says that fever arising after a spasm is a sign of improvement, but
spasm occurring after fevers, a bad sign.
The Glossarium Ansileubi or Liber glossarum, to be dated no later than the
eighth century, drew on a great variety of works for its glosses or definitions,
and the manuscripts that were used for this project were often superior in
quality to those available to us for the complete work. In many cases,35 the
Glossarium indicates its source or author by name. This, however, is frequently
different from the name of the author we know; e.g. the excerpt on spasmus
from Aurelius is introduced as Galeni,36 other quotes are said to be taken ex
libris medicinalibus, or come from pandecti37 medici; both leave us completely
in the dark and do not correspond to the title of any compilation known at
present. The quote I just cited comes without an indication of the source, but
you will have spotted the similarity of the first words with Aurelius. This is,
as it turns out, no mere coincidence, because this passage in the Glossarium
repeats the information found in Caelius Aurelianus (Diaet. pass. 32), and we
know that both Aurelius and Caelius Aurelianus based their accounts on the
lost Greek works of the second-century Methodist doctor Soranus:
Ex quibus adprehendis spasmo laborantem? ex non uoluntaria tensione atque conductione partium uel neruorum, cum dolore uehementi atque acuto et pulsu paruo.
How do you recognize that a patient suffers from a spasm? From
the involuntary tension and contraction of the parts or the mus35
36
37
At least for the medical material edited separately by Heiberg.

He takes pride of place.
The correct Latin plural is pandectae .
48
FISCHER
cles, and at the same time strong and sharp pain and a small
pulse.
And now we have, at long last, identified the text after the lacuna in both the
Liber passionalis and Petroncellus as deriving from Diaet. pass.,38 with the
additional bonus of exonerating Isidore. Obviously, Isidores enigmatic entry
on spasm was taken from a text strongly resembling Gloss. med. p. 82,13-19,
which included the quotation from Hippocrates Aph. 2,26,39 a quote present
in the Liber passionalis, in Petroncellus,40 and likewise in Aurelius, but paraphrased in Theodorus Priscianus without the name of Hippocrates.41 As a
consequence, we will conclude that this introductory section on spasm in the
Liber passionalis and in Petroncellus was in existence before Isidore, who died
in 636, finished compiling his Etymologies. Unfortunately, this is of no great
help with dating either the Liber passionalis or Petroncellus, but it provides
further evidence for the compilation of medical (and veterinary)42 books in
Latin. Should we look for additional corroboration beyond the Libri medicinales and the Pandecti medici used as source for the Glossarium Ansileubi
mentioned above, then we can find it in Cassiodorus (Inst. I 31,2), who recommended it to his monks at Vivarium, the monastery in Southern Italy he
founded in 552, if they had no access to Greek learning (si uobis non fuerit
Graecarum litterarum facundia), anonymum quendam, qui ex diuersis auctoribus probatur esse collectus.
Now we return to our analysis of ch. 56 of the Liber passionalis. After a
short distinction of the three kinds of spasm, tetanus of the neck, opisthotonus
38
The origin of the error must be sought in an imperfect manuscript where

Cael. Aur. Diaet. pass. 32 had been merged with the second half of Diaet. pass. 39.
39
The wording is different from the so-called Ravenna version of the Latin
translation of the Aphorisms.
40
Same text, but placed before the first section on therapy in the Liber passionalis (56,9) and after it in Petroncellus (36,7).
41
Theod. Prisc. II 30 haec accidentia corporibus frequenter sine febribus euenire consuerunt. quae quibus superuenerint causam soluunt. superuenientes uero
febribus uitae semper periculum minitantur. The aphorism is also quoted in Paul.
Nic. 125,22-24 Ieraci Bio, and in Trattato delle cure, p. 276 Giacosa (Piero Giacosa,
Magistri Salernitani nondum editi. Catalogo ragionato della esposizione di storia della
medicina aperta in Torino nel 1898, Torino 1901), in a garbled form: quod super si
uenerit (sc. febris), spasmum soluit, ut ypocras ait: spasmum in febrem etc.
42
Vegetius, mulomedicina.
49
and emprosthotonus (56,5), the description of patients suffering from these

varieties is quoted in full (as far as we can tell) from Cael. Aur. Diaet. pass.
33-35 (= 56,6-8), for which the compiler must have had access to this work. If
so, why did he not take the opportunity to fill the gap between Lib. pass. 56,3
and 56,4 and enter the complete text of Diaet. pass. 32 and 39? The compiler
or redactor of Petroncellus would have felt no need for these passages here in
ch. 36 because he was going to deal with these varieties in the following chapter
(37), which is based largely on Cassius Felix ch. 38, and therefore he can omit
at this point Lib. pass. 56,5-8 without any loss of information. Indeed, he will
quote Cael. Aur. Diaet. pass. 33 at Petronc. 37,3 and 35 at Petronc. 37,5.
In ch. 36, Petroncellus also locates the Hippocratic aphorism 2,2643 after the
section on therapy, which is again common to Petroncellus (36,5f.) and the
Liber passionalis (56,10-12) and now follows.
The very wording of this aphorism deserves some attention. It is different
from both the so-called Ravenna version, and from the so-called translatio
antiqua (a version produced perhaps towards the end of the eleventh century);
the former reads in M
uller-Rohlfsens first critical edition:
Febrem super spasmum fieri melius quam spasmum super febrem,
which is a fairly literal rendering of the Greek text,44 translated by Jones as
follows:
It is better for a fever to supervene on a convulsion than a convulsion on a fever.
If we compare the Liber passionalis (56,9),45 the Latin there sounds more like
a paraphrase:
43
De Renzi cit. 215 refers to Coaqu. sent. 156 e 157 rather than aphorisms.
Certainly, Coac. 156 and 157 in Littres edition deal with headaches and cannot be
meant here.
44
, . (Hippocrates,
with an English translation by W.H.S. Jones, IV, London-Cambridge, Mass. 1931,
114)
45
The second half is missing in Sang. 752, therefore I quote from Phillipp.
1790.
50
FISCHER
Yppocrates autem dicit quoniam si spasmum febris obuenerit,46

signum est salutis; si autem febrae spasmus, malum est.
Hippocrates, however, says that if fever arises after a spasm, it is
a sign of improvement; if, on the other hand, spasm after a fever,
it is bad.
Petroncellus is hardly different (36,7):
Ippocras auctor dixit si in pasmum febris obuenerit, signum salutis
est; si autem super febrem spasmus euenit, malum est.
An interesting difference between both texts is Yppocrates autem and Ippocras
auctor. While autem, or rather an abbreviation for this word, could perhaps be
misread as auctor, this is less likely to have happened the other way round, and
this leads to the conclusion that the Liber passionalis cannot, in this instance,
derive from Petroncellus. Again, autem and not auctor is the reading of the
Glossae medicinales (p. 82,17 Hbg.). It is badly placed or rather not required
by the context in the Liber passionalis, and in my opinion shows the imperfect
manner of excerpting.
Autem is furthermore attested by Aurelius (20):47
yppocras autem dicit quoniam si spasmo febris superueniat, signum esse
salutis; quod si spasmus febri[s] superueniat, contra periculi signum est.
Similarly, it may be significant that both Aurelius and Liber passionalis introduce the declarative sentence after dicit with the same conjunction
quoniam, rather than quod, quia or choosing the classical a.c.i. construction.
Let me proceed to one final point in our comparison, the section on
therapy that is shared by the Liber passionalis (56,10-12) and Petroncellus
(36,5f.). In both works, the Latin text is in such a bad state that it is not easy
to follow the recommendation for therapy. The mention of causa constricta48
must indicate that the source for this passage is Methodist. This limits our
search to Caelius Aurelianus and Aurelius, neither of whom provides a close
parallel.
46
Obuenire for superuenire probably had its origin in the faulty reading
subuen-; FEBRISUBUENERIT in scriptio continua was then changed to febris
obuenerit.
47
Replaced by enim in Gariopontus 5,2 (citing the Basel 1536 ed.).
48
Perhaps to be corrected in constrictiua.
51
Therapy starts with fomentations, anointing with Sicyonian oil, and cataplasms, in order to bring about relaxation. The Latin adjective laxatibilis,
attested twice by both witnesses, is not known from anywhere else,49 which
suggests to me that the passage is indeed a fragment taken from an unknown
Methodist work in Latin translation. Also recommended is a rinsing of the
mouth (collutiones oris),50 with warm water, as we learn from Cael. Aur. Cel.
III 8,77.51 Moreover, defecation is to be encouraged with an enema containing oil in which rue and castoreum have been boiled, and more fomentations
are applied, for which roasted salt immersed in hot water is specified,52 and
likewise warming ointments, among them one prepared from the plaster made
with plant juices ( ), and with oils made from seeds with relaxing and
warming properties.
The next four paragraphs, Lib. pass. 56,12-15, were incorporated by the
compiler of Petroncellus in his next chapter, 37 (37,8; 37,11; 37,14; 37,19). It
is surprising to see that the same excerpts from Theodorus Priscianus that
are drawn on twice in Petroncellus 36 and 37 show some variation in their
phrasing. All this makes it abundantly clear how much more work will be
needed before we have sufficiently analysed and understood what went on in
these medical compilations, how they are related, and what additional material
for the study of ancient and medieval medicine they may yield.
Geschichte, Theorie, Ethik d. Medizin
Am Pulverturm 13, D 55131 Mainz
49
Klaus-Dietrich Fischer
[email protected]
It is still found in the late manuscript that de Renzi prints in his edition of
Petroncellus, cit. 215.
50
The ThlL s.v. collutio cites Aurelius (as Cael. Aur. Acut. exc.) for the
spelling colluitio found in both our sources.
51
He uses <os> colluere aqua calida in this passage, but the noun collutio is
not rare. Rinsing the mouth is not found in Aurelius.
52
Paul. Nic. 125,59 says that the sacks containing the salt should be immersed
in hot water, which seems more reasonable.
Appendix:
Liber passionalis (Berol. Phillipp. 1790; Sang. 752)
Petroncellus (Par. lat. 11219; de Renzi, Coll. Sal. IV)
LVII de spasmis et taetanicis dicit [Aurel. 20; Theod. Prisc. II

30 p. 127,11-15]
56,1 spasmus est tensio atque contractationem parci neruorum
uel uicinorum musculorum ceruicis cum dolore uehementi quam
maxime inpatienti dolore cum patientur ut etiam in respondendum os facile aperire non possunt. Nam sub hoc titulo sparmis
uariae sunt huis passiones figurae.
[Theod. Prisc. II 30 p. 127,15-128,1] 56,2 Spasmus per omne
corpus tensionem frigida temporalem quidem habet ita tenduntur ut sub uno scismate iacentis tensionem molesta et inflexa
simile teneatur aut uirum pronus contractio habet <. . . >
[Cael. Aur. Diaet. pass. 32] 56,3 < . . . > atque acuti ex
pulei paruo
[Cael. Aur. Diaet. pass. 39] 56,4 < . . . > Ali[a] uero
quod si<t> [il] litalis passio a[d] corde nominata
dixerunt quod nobis praecibato uigoris addebit
56,5 Titanice uero passionis sunt tres ceruices titanus opystotonus inprostotonus;
XXXVIII Ad <s>pasmos uel ceruicis dolorem

36,1 Spasmus est tensio et contractio partium neruorum
uel uicinarum musculorum. ceruices cum dolore uehementi
maxime contigit patiente dolore os facile aperire non possunt
sub hec titulum uarie sunt passiones
36,2 Spasmus enim per omne corpus frigidam habet temperantiam. Ita adtendunt ut sub uno scemate iacentis tensionem
molestam et inflexam simile teneatur Aut uerum prunus contractio habet <. . . >
36,3 < . . . > et quae acuta ex pulso paruo
36,4 Alia uero si talis passio ad corde nominata dixerunt quod
nobis principato figuris addebent
53
[Cael. Aur. Diaet. pass. 33] 56,6 Quomodo titonicum

adpraehendis ex gutture in longitudinem extensum
aut inflexibile positione clura cum dolore
uehementi et oluria musculorum qui buccis insunt
robore uultus densio conpraehensione ut mutus
aperire non possunt contradictionem neruorum
impetus ciborum
[Cael. Aur. Diaet. pass. 34] 56,7 ex quibus opistotonicum
adprehendis ex supinis aptum hoc est contractum
a posteriore parte conducta ceruice ad dorsum
cum nimia tensione atque uehementi dolore dorsi
et clunum surarum quoque contracto et manum et
digitorum
[Cael. Aur. Diaet. pass. 35] 56,8 ex quibus enprostotonicum adprehendis ex conductione gutturis in
anterioris partis effecta atque mento pectore
adiuncto tensione illorum et precordiorum
continuam ictum de certo et digitorum inplexibile
flexionem qu supradicta est passionis si ex
uulnere fuerint effecti in periculo in sanitate
difficilis iudicantur.
37,3 < . . . > bucca et gutture inflectibile po<si>tione dura ut

mutus os aperire non possunt ex contradictione neruorum.
37,5 Enprostotonum sic intellegis ex conductione gutturis in

anterioris parte effecta. Ad quam mento pectore adiuncto tensione illiorum et praecordiarum contigit si < . . . > in supra
dicta passione uulnera euenirent effecta periculosam et insanabilem sanitatem demonstrat
54
56,9 Yppocrates autem dicit [aph. 2.26] quoniam si <per>

spasmum febris obuenerit signum est salutis si autem super
febrae spasmus malum est
56,10 De cura dicit quomodo hi curantur horum storum
ergo uaporationem opponendum laxatibilis adiutoriis quoniam
causa constrict<iu>a est nimio et unguere sicciono oleo et
cataplasmatis utimus laxatibilis et colu[i]tionis oris
[cf. Paul. Nic. 125, 55?] 56,11 et post c<ol>luitione probis
<s>cybellarum prouocationem in c<list>ere oleo rutatio cum
castorio et uaporationis adhibendum maxime ex salibus uaporationem utimur ex salibus tostis in sacello inmissi calid// feruenti locis extensis uel contractis opponendum est uel locis dolentibus p//tio opponimus medicamen diacilon et olei semilibus
laxaturiis et calidis
56,12 contra autem contractionem ceruicis hoc est tetanus tensionem et in rectum uel opistotonum contractus ad spina dorsi
si uiris uel aetas permiserit et sine febrae fuerit fleotomia adhibitus
56,13 Quod si improstotonicus fuerit et mentum pectore infigat
deinter scapulas cucurbitare cum escarificationem fieri oportebat et tapsie sucum unguentum extensarum partium
<s>pasmum febris obuenerit signum salutis est Si autem super

febrem spasmus euenit malum est
36,5 Cura spasmus Id est uapora-tionis laxatibilis adiutoriis ubi
causa constricta est nimis et ungere oleo sicionio et cataplasma
laxatibilis et coluitionis auris et dabis
36,6 preter coluitionem clisteriabis oleo rutatio cum casturio
propter scibellarum prouocationem et uaporationes ex salibus
tussis in sacellis missi calidis feruenti locis dolentibus opponentis et unguere debes fortiter cum ipsa uaporatione ex unguentis
calidis in his locis ponemus medicamen deaquilon et oleo laxatiuo et calido
37,8 Si aetas uel uires permiserint et febris non est fleotomandi
sunt de uena mediana. et si[i]n minus praeualuerint inter scapulas sanguinem detrahendum est per scarificationes
37,11 si enprostotonicus fuerit et mentum pectore infricatus
inter scapulas cocurbitare cum excalefactione fieri oportet et
tapsie sucum extensura perunguis
55
56,14 et castorium datum cum mulsa et modicum piperis mixtum

[cf. Theod. Prisc. II 31 p. 128,16-129,4] 56,15 Cybos autem
damus digestibilis sucus apula buccellas canditas in aqua calida
muturatae etiam frequent<er> mellegratum infusum fouendi
erunt
56,16 Et si adiutoriis supradictis releuatis etiam oris coperit inpedimentu cadere maluas sicoctas et minuto speciculis anetino
iuscello coctus plenam iam confidentiam suam ministrabo frequenter etiam cum membribus pticmata ex inomelle temperato
lactis his negandus copia
[Theod. Prisc. II 31 p. 129,4-8] 56,17 Omnibus uero tensionibus post fomenta calastica ex oleo sitionio unctione adhibebis
et acoba calidiora et cataplasmata calastica loca diutius mitigabo post clisteris uero calastico expositis exquibalis aut uti
nos conuenit
[Theod. Prisc. II 31 p. 129,10-13] 56,18 Nam et carbonibus
proximis calentibus lectus et strami<n>a calidiora procurabis
Cyrotaria uel acoba ex oleo cyprino his frequenter adhibenda
sunt et adipem leonis aut galbano cummixtis par[s] beneficium
procurauerunt
37,14 Castorium cum mulsa modice piper potui da

37,19 talem cibum accipiat degestibilem. Sucus apola bucellas
candidas in aqua calida maturate frequenter accipiat.
36,14 Item post fumenta unguis loca dolentia et acoba calidiora adhibebis 37,15 unguento acobi calidiora et cataplasma
calastice loca dolentia medicabis
56
[Theod. Prisc. II 31 p. 129,13-130,5] 56,19 dabis etiam interea potionis huiusmodi castorio puluerem cum melle gratro similiter dabis aliquoties piperis intermisces dabis siluo
cum piper et mellegrato similiter dabis diuriticis potionis dabo
sopure < . . . > earum beneficiis consueuerunt
36,15 dabis etiam inter eas potiones castorie puluis cum mell.
grato aliquotiens dabis silfio cum piper et melgrato. Aliquotiens
dabis diuriticas potiones 37,16 Puluera ex casturio cum mell.
grato dabis aliquotiens. silfio cum piper et mell. grato dabis
Similiter diuriticam potionem dabis
Key to fonts and colours:

Bittersweet: Change of sequence
Emerald: Theod. Prisc.
RubineRed: Cael. Aur.
Slanted: Same as Theod. Prisc.
Typewriter:
Same as Cael. Aur.
The fate of a Greek medical handbook in the medieval West: the

Introduction, or the Physician ascribed to Galen
Along with genuine Galenic works, several pseudo-Galenic treatises

made it through the Middle Ages as reference books for the purpose of teaching medicine.1 They were in fact deemed inauthentic much later, during the
Renaissance;2 in late antiquity and the Middle Ages, by contrast, they were
usually perceived as genuine works by Galen and formed the basis of medical learning. Among them, the Introduction, or the Physician, ascribed to
Galen,3 had a most interesting fate. The aim of this paper is to examine several moments in the history of this text, in order to highlight the malleable
nature of ancient medical texts in the hands of medieval readers, especially
the Byzantines.
1
Medieval doctors, in order to teach medical students, used the Galenic material. As a result, some of his works were updated to adapt to the increasing
knowledge (e.g. remedies, operations) of the Byzantines. Naturally, all the material
ascribed to Galen was subject to such changes. By contrast, Hippocratic works were
transmitted with limited changes and adaptations for teaching purposes: in fact, Hippocrates texts were much revered, but of little use. Rather, Hippocrates opinions
were studied through Galens texts. Among the pseudo-Galenic works that were popular in the Middle Ages is the collection of medical definitions called Definitiones
Medicae (XIX 346-462 K
uhn), about which see J. Kollesch, Untersuchungen zu den
pseudogalenischen Definitiones Medicae, Berlin 1973.
2
The editio princeps of Galens works in Greek, the 1525 Aldine is the first
collection to set apart the notha biblia (in volume IV). The editors did not make
any statement about the reasons for their selection, but the issue became an object
of interest among scholars, though without any systematic method of discrimination:
Jacobus Sylvius (Jacques Dubois)s work, Ordo et ordinis ratio in legendis Hippocratis
et Galeni libris (1538) is a landmark in the discussion of authenticity.
3
Introductio sive medicus (XIV 674-797 K
uhn), about which now see my
edition (Galien. Oeuvres, III. Introduction; Medecin, Paris 2009).
58
PETIT
Let us state immediately that this undermines the common assumption

that the key medical works to be read, and thus transmitted through time,
were the sixteen Galenic works of the Alexandrian Canon. In fact, as the
history of a text like the Introduction, or the Physician shows, the material
available to doctors was more diverse than this limited collection of works, and
the perception of which medical texts should count for something was different from ours. Thus, the West apparently displayed distinctive interests in
medical books; in the East, the physicians of Alexandria, and later prominent
translators such as Hunain ibn Ishaq imposed a rather different view of the
corpus, based partly on authenticity, partly on the contents of the available
treatises and, definitely, the so-called Alexandrian canon had more relevance
there than in the West.
The so-called Galenic corpus4 is everything but definite in fact, the
very term of corpus could be deemed inappropriate, since Galen himself did
not control the publication and the diffusion of his works, and, in turn, many
works not by him were transmitted under his name. Galen also thought that
some of his works were lost after the fire,5 such as On Prognosis, which was in
fact transmitted to us. Therefore, it seems more adequate to talk of an open
tradition, a basic collection to which elements were successively added, each
with a specific textual transmission and a chaotic fate.
The pseudo-Galenic Introduction, or the Physician is a medical handbook of disputed date and authorship.6 It could be contemporary with Galen
or of later provenance; the terminus ante quem is a partial late antique Latin
4
About which see V. Boudon-Millots general presentation in her introduction

to Galien. Oeuvres, I, Paris 2007.
5
As we learn from his commentary to Hippocrates Epidemics VI (edited
by Pfaff in CMG V/10,2,2, p. 495,2-12), Galen believed that his work titled On
Prognosis, written in 178 AD, was lost, after his own copy was burnt in the great
fire of 192 AD in Rome, with most of his other works, his library, and his precious
stock of medicines and instruments (about the extent of his loss now see the newly
rediscovered On the Avoidance of Grief). As Galen fails to mention On Prognosis in
the catalogue of his own works (On my Own Books) that he wrote at the end of the
century, it is likely, as V. Nutton convincingly argues in his 1979 edition (Galen On
Prognosis CMG V/8,1, pp. 49-51), that he had not by that time recovered a copy
of the treatise and believed it to be lost forever.
6
On this matter, see my edition xxxvis. and my forthcoming article What
does Pseudo-Galen tell us that Galen does not? Medical schools in the Roman Empire,
The fate of a Greek medical handbook in the medieval West
59
translation of which I shall say more in a moment. The Introduction is one of

the very few ancient texts, together with the Definitiones Medicae (spurious as
well, and equally popular), that provide us with an introduction to medicine
dealing with virtually every aspect of medicine.7 The text begins with eight
chapters on the history, the nature and the parts of medicine that sum up most
of the assumptions of the time on the origins and the merits of the art, with
some elements of mythology that embellish the history, and happen to be common place in ancient encomiums of arts: medicine, according to Pseudo-Galen,
is an art, not a science, but it is a noble art, that was discovered and practised
by gods and heroes before it was brought to perfection by Hippocrates. The
text then proceeds with a relatively brief discussion of physiology and anatomy
(chapter 9-12), before turning to more practical aspects of medicine, about the
names, descriptions and treatments of various diseases, including eye diseases
(chapter 16) and general as well as bone surgery (chapter 17-20).
The text is remarkable not only for its comprehensiveness, but also for a
variety of features, linguistic and historical, and for its distinctive elements of
doxography. It was transmitted rather early on under the name of Galen: all
forty Greek manuscripts display the epigraphe: Galenou Iatros: Eisagoge, and,
although the oldest manuscript is not earlier than the twelfth century AD, the
epigraphe must be older, for the text appears in two separate families of Greek
manuscripts that hint at a much earlier (perhaps ninth century) archetype.
Moreover, a Latin manuscript of the tenth century AD transmits a small bit
of the text (in Latin) under the name of Galen.8 It is thus likely that the
text was considered authentic at an early date. In Galens time, however,
things may have been different: at the beginning of On my own books, Galen
mentions a book with the same epigraphe (thus circulating under his name),
about which two friends argue at the books market in Rome.9 One of them,
whom Galen describes as thoroughly educated, dismisses the book as being a
fake, on grounds of striking differences in style with genuine Galenic texts. If
the story is true, it may well be that the Introduction, or the Physician that
in P. Adamson-J. Wilberding (edd.), Galen and Philosophy, to appear as a BICS
Supplement.
7
One of the key absent topics being gynaecology.
8
The Latin manuscript is Vindobonensis lat. 68, about which see below.
9
There is a discussion about the exact form of the epigraphe in Galens text:
only two manuscripts have Galenos Iatros, corrected by Hermann Schone into Galenou
Iatros.
60
PETIT
we possess is in fact the forgery that circulated in Galens own lifetime with
the same epigraphe. At any rate, for Galen and some of his educated fellow
citizens, it was a fake but it may well have been considered authentic in
other circles; that would explain its later incorporation in the collection(s) of
Galens works.
The Introduction, or the Physician remained virtually unknown in the
East: the oriental tradition has nothing to tell us about this text, although
there is a faint possibility that it was part of the Alexandrian material used
by the Arabs.10 The book, by contrast, has a long history of being read in
the West. Due to the usual gaps in our understanding of the transmission of
ancient Greek texts, the evidence I have gathered may look a bit patchy, as it
takes us straight from late antiquity to Constantinople in the late Palaeologan
period and then in turn to medieval and Renaissance Italy. But each of those
highlights gives us an insight into the life of medical texts through places and
time; in each of the following cases, we find evidence that an ancient medical
text was copied, used, translated or adapted with transformations of varying
extent that reveal the plasticity of ancient medical knowledge in the Middle
Ages.
10
See P. Pormann, Jean le Grammairien et le De Sectis dans la litterature

medicale dAlexandrie, in I. Garofalo-A. Roselli (edd.), Galenismo e medicina tardoantica: fonti greche, latine e arabe, Napoli 2003, 240; The Alexandrian summary
(Jawami) of Galens On the sects for beginners: commentary or abridgment?, in P.
Adamson et al. (edd.), Philosophy, Science and Exegesis in Greek, Arabic and Latin
Commentaries, London 2004, 26. Some Arabic manuscripts are listed in F. Sezgin,
Geschichte des Arabischen Schrifttums. Band III. Medizin, Pharmazie, Zoologie und
Tierheilkunde, Leiden 1970, 139, n. 154; and Nachtrag I, 1974, 408, but only one could
be checked (Dar-al-kutub tibb. 1103): it contains in fact Hunain ibn Ishaqs Questions
and answers on medicine, and, from what the catalogues have to say (for it was impossible to make contact with the relevant libraries), it is likely that the two manuscripts
of Rampur and Hayderabad have the same text and not an Arabic translation of
the pseudo-Galenic work. Earlier on, Diels in Hermann Diels, Die Handschriften der
griechischen Arzte,
1905 (based on Ackermann) mentioned yet another manuscript,
but the enigma of the unusual shelfmark (Norfolk 3383) has not been resolved yet.
As suggested by A. Touwaide, this collection could be related to a gift of the Duke of
Norfolk to Gresham College, London (A. Touwaide, Byzantine Medical Manuscripts:
Towards a New Catalogue, ByzZ CI, 2008, 206).
61
A late antique (?) adaptation in Latin

First, we have evidence of an early Latin translation of part of the
Introduction, or the Physician, with no mention of authorship (neither for the
original, nor for the translation). This translation concerns chapters 16 to
20 of the original Greek text and appears in an otherwise well-known Latin
manuscript of the ninth century (Augiensis 120): this manuscript has preserved
almost everything that we know about surgery in the early Middle Ages. The
translation of chapters 17-20 of our pseudo-Galenic text was painstakingly
edited by Daniel de Moulin in 1964; unfortunately, it could be argued that the
edition is not of a highly satisfactory standard; besides, De Moulin failed to
recognise chapter 16 in the same section of the manuscript, thus providing an
incomplete text.11 Another manuscript (Vindobonensis lat. 68) of the tenth
century bears traces of the same translation, but it has only the very beginning
of chapter 16, and adds little to what we know through the main manuscript.
Klaus-Dietrich Fischer identified both the Vindobonensis manuscript and the
presence of chapter 16 in the Augiensis.12 Naturally, the facts are intriguing
and raise a number of questions: who wrote this translation, where and when?
How did this section end up in this manuscript? Was the rest of the text
translated as well, and, if so, was it by the same person or at least in the same
milieu? Could this translation be related to the so-called Ravenna school?
Most of these questions might well remain unanswered. Some facts, however,
are indisputable: the Latin version of the text is not only a translation, but also
bears traces of some sort of adaptation for a new audience; the Latin is poor (it
must be partly the responsibility of the translator, partly that of the copyist);
the translator shows some awareness of a potential audience, thus suggesting
that the Latin text may have had some teaching purpose; the comparison with
our extant Greek manuscripts demonstrates that the Latin version is earlier
than the archetype of the Greek manuscripts. Finally, bits and pieces of the
11
Daniel de Moulin, De heelkunde en de vroege middeleeuwen, Leiden 1964;

the edition, with translation and notes appear at pp. 53-82.
12
I owe Cloudy Fischer not only the information, but also his own transcription
of chapter 16; in addition, Arsenio Ferraces Rodrguez has shared with me his draft
new edition of the Latin text. I used their material extensively for my edition of the
Greek text.
62
PETIT
text appear in late antique Latin medical texts, thus suggesting that a full
translation circulated at some point.13
For our purpose here, the main issues are those related with the reading and potential use of the text for teaching purposes the problem of the
date, however, makes this investigation frustrating, as the only certain fact is
the date of the manuscript (ninth century), which means that the translation
and/or its subsequent rewriting could be relatively late, too. The date problem is thus double: when was the text translated? Is the text in its current
shape (that is, rewritten) of the same period or later, and if later, could it be
thoroughly medieval? The assumed date of the translation is around the fifth
century according to Klaus-Dietrich Fischer, who, however, rightly points out
the inherent difficulty to such enquiries. Moreover, the late antique and early
medieval material is scanty, and subsequently our knowledge of medicine in
that period is far from extensive. If the translation and/or its rewriting were
clearly datable from late antiquity, then this could point to the so-called school
of Ravenna, in which several translations and commentaries of Galens works
were made. But in recent years, scholars have grown more cautious about
ascribing every late Latin translation to this milieu: indeed, the evidence for
it is usually inconclusive or, as in the case of the pseudo-Galenic Introductio,
merely absent.14
Among the many changes made in the Latin version of the text, the first
striking difference lies in a number of additions to the text, combined with
suppressions of other sections. While some bits and pieces may have disappeared because of the usual problems of textual transmission, the adapter of
13
I have discussed the differences between the Greek original and the Latin
version in my article LIntroductio sive medicus du Pseudo-Galien dans le haut moyen
age latin: problèmes dedition poses par la tradition indirecte, in Arsenio FerracesRodrguez (ed.), Tradici
on griega y textos medicos latinos en el periodo presalernitano.
Actas del VIII Colloquio Internacional Textos Medicos Antiguos (A Coru
na, 2-4
Septiembre 2004), Universidade da Coru
na 2007, 249-270. Of course, the Latin
translation of the pseudo-Galenic Introductio is but one among quite a few ancient
translations (of Hippocrates, Soranos, Oribasios and of course Galen) studied by other
scholars. Some traits are to be found in several translations, but the Latin of pseudoGalens version is too disturbed to allow any stylistic comparison.
14
On this matter see the enlightening piece by M.E. Vazquez B
ujan, Problemas
generales de las antiguas traducciones medicas latinas, StudMed XXV (1984) 641680.
63
the text clearly made some conscious interventions, such as removing remarks
that were not directly useful, and adding sentences meant to clarify a number
of issues thus adding definitions of diseases (skin diseases, scrotal hernias),
glosses of Greek terms, of which the mere transliteration could be obscure,
and meta-textual remarks. Examples of rewriting also exist, such as changing the third-person indicative verbs into imperative forms, which suggests an
adaptation of the text in a more practical fashion, for students.15 In the case
of treating dislocations of the hip, the Latin has squarely moved away from
the Greek, which deems all forms of hip dislocations incurable, and inserted
instead a healing procedure that necessarily comes from another (unidentified)
source.16 Thus, it appears that the text was re-used in a different context, by
Latin speakers, for practical purposes and perhaps teaching.
At any rate, this episode in the textual history of Pseudo-Galens Introductio suggests that is was popular rather early on in the West among
physicians and perhaps used in medical teaching. Whatever the place this
translation was made, the Greek original had already been transferred to the
Latin-speaking part of the world, at the cost of minor changes directed at a
new readership. The early Latin translation did not really have any influence on the later journey of the text, but it hints at some form of continuous
teaching use from antiquity onwards; the fate of the Greek manuscripts in
Constantinople seems to confirm this tendency.
The Greek manuscripts of the Introductio and Constantinople

Before coming to the use of the Introductio in Byzantine teaching in
Constantinople, I intend to outline the key facts of its textual transmission in
Greek.
The pseudo-Galenic Introductio has survived, either in full or in part,
in forty Greek manuscripts, dating back from the twelfth to the eighteenth
15
It is tempting to connect this feature with the numerous imperative forms

that occur in Galens practical opus on anatomy, the Anatomical Procedures intended
for students aiming at performing dissections and vivisections (by contrast with the
more intellectual On the usefulness of the parts of the body).
16
See my 2007 article (cited n. 12) for extensive discussion of the passages and
comparison with other texts, 261-267.
64
PETIT
century.17 The large majority are fifteenth to sixteenth century items; they can
be divided into two groups: the bulk of the manuscripts, about three quarters
of the total number, descend in a more or less direct way from a twelfth
century manuscript that is well-known among editors of Galenic treatises,
the Vaticanus gr. 1845 (= V). Apparently, no copy of the pseudo-Galenic
Introductio as preserved in V was made before the fourteenth century, and
the large majority of copies were made later, in the mid- and late fifteenth
century; to these, one should add the subsequent copies made in the early
Renaissance, in part for editorial purposes, such as the 1525 editio princeps
of Galens complete works in Greek. I call this family of manuscripts the A
family.
The second, smaller group of manuscripts (family B) has no such prestigious ancestor; the oldest manuscript is from the mid-fifteenth century (Urbinas
gr. 67 = U). More interestingly, this family is in turn divided into two groups
each of which transmits only part of the text. If one divides the text according
to the sections outlined in the manuscripts (three sections, which I call a, b
and c), it appears that there is a neat break corresponding to these sections
in other words, it seems that the breaks correspond to mechanical accidents
in the textual transmission. Presumably, the loss of quires is responsible for
the fact that part of family B has sections a and b, while the other part of
family B has sections a and c. Section a serves as a basis for comparing both
sets of manuscripts it is obvious that they all belong to the same family, a
family that shows a number of differences with V and its copies.18
While family A is not relevant to this discussion, family B on the other
hand has several intriguing features. First of all, the initial pinax (a table of
contents preceding the main text) looks different from the one that we find
in V (and its copies, which barely alter the original, albeit for a few copy
mistakes): it is shorter, and yet it has additional words.19 After the first two,
identical lines, the family B manuscripts have the following words:
17
I am summarizing here the data presented and analysed in my edition

lxxxi-cvi.
18
Regarding the classification of the manuscripts of family B against V, see
my edition xciv-cvi.
19
I append the complete pinax of each family in the appendix.
65
.
, , .
This book is excellent, especially for beginners; thus it should be
read right after the Ars Medica.
The summary goes on briefly with the first few chapter headings, then
breaks off and adds this slightly rhetorical exhortation: well, let us start from
chapter One (or: from the beginning).
Thus a mysterious reader added his judgement on the text, which he
deemed worth reading, and indeed excellent; he added: especially for beginners. When and where could such an addition have been inserted? For it
is tempting to see it as an addition rather than an original feature that would
have been lost in V. Indeed, family B is distinctive in that it has a number of
other additions or alterations that could hardly have featured in the original.
At any rate, the hypothesis of an omission in V is not supported by palaeography. Conversely, however, the relatively recent date of the manuscripts of
family B makes it difficult to be conclusive on this matter. The oldest one is
a mid-fifteenth century manuscript, and we have to be content with this late
terminus: how early can be this passage? Well, the mention of the Ars medica
with its common medieval title (he tekhne etoi ten horiken) does not give us
a precise indication, nor does the language or the style of this very short bit.
We need to look at the whole text and its complete (as we know it) textual
tradition in order to gather a few more elements. I shall come back to the
analysis of the B family text as a whole.
At any rate, the purpose of this commentator (in a weak sense) was
presumably medical teaching. Now, while the Ars medica has a continuous
history of teaching use in the Latin West (it features in the Articella), the Introductio on the contrary disappears from our records in late medieval medical
training. Hence, it is likely that this judgement, whether or not it was followed
by immediate strong student interest, had no immediate posterity beyond the
Greek Byzantine world.20
Several elements hint at a starting point in Constantinople. If we consider the origin of our manuscripts (of both families), they all come from
20
It was nevertheless rediscovered in the Renaissance for its qualities as a book

for beginners.
66
PETIT
Constantinople (rather than Southern Italy, for instance); hence first hand
knowledge of the Introductio was most likely restricted to Constantinople.
While the context in which V, the twelfth-century and oldest manuscript, was
written is very obscure, the bulk of our fifteenth-century copies point towards
one and the same scholarly milieu: that of the Xenon of the Kral, where a
considerable scriptorium, acquainted with medicine, worked, learned and (presumably) taught. Those were the pupils of John Argyropoulos that carried on
with their activities even after he left Constantinople, and after the fall of the
city in 1453.21 One of the key actors was Demetrios Angelos, who is known
to us through numerous copies of manuscripts and some notes. Another important character was Michel Apostolis, who, along with many other Galenic
texts, copied one of the key manuscripts of family B of the Introductio, the
Scorialensis Sigma II, 11 and succeeded Argyropoulos at the Xenon of the Kral
in 1452. Demetrios Angelos had copied the Introductio several times, and had
access to manuscripts of both families (interestingly, no contamination can be
established Demetrios did not use any B family manuscript to correct the A
family items, although the latter have obvious lacunae that could have been
filled with the help of the former). The textual transmission of the Introductio
is closely linked to Constantinoples scribes and scholars.
John Argyropoulos is not particularly well-known for his medical knowledge (and, potentially, teaching), but rather (rightly) perceived as a philosopher, who played a major role in spreading Greek and the humanities in Renaissance Italy, through his various trips and later, during his exile after the
fall of Constantinople. This is all perfectly right. The fact, however, that his
students copied a number of Galenic works, and some of them several times,
21
I am passing a bit rapidly on a subject that has been carefully and painstakingly studied by Brigitte Mondrain in a number of articles, and followed up by others.
See for instance B. Mondrain, Jean Argyropoulos professeur `
a Constantinople et ses
auditeurs medecins, dAndronic Eparque `
a Demetrios Angelos, in C. Scholz-G. Makris
(edd.), Polypleuros nous. Miscellanea f
ur Peter Schreiner zu seinem 60. Geburtstag,
Munich-Leipzig 2000, 223-250. B. Mondrain, Comment etait lu Galien `
a Byzance dans
la première moitie du XVe s. ?, in A. Garzya-J. Jouanna (edd.), Trasmissione e ecdotica dei testi medici greci. Actes du IIIe colloque international sur lecdotique des
textes medicaux grecs, Paris-Naples 2003, 361-384. B. Mondrain, Demetrios Angelos
et la medecine: contribution nouvelle au dossier, in V. Boudon-Millot-A. Garzya-J.
Jouanna-A. Roselli (edd.), Storia della tradizione e edizione dei medici greci. Atti
del VI Colloquio Internazionale Paris 12-14 Aprile 2008, Naples 2010, 293-322.
67
with annotations of medical content, suggests a strong interest in medicine.

Not all texts seem to have interested Johns group: rather, they showed some
interest for theoretical medicine (issues of epistemology and method), as displayed in the Ars medica, in the Introductio, and in Galens Simple Drugs
books I-V (in contrast to books VI-XI, which were perhaps neglected for their
mere practical use).22 But there may be some reasonable distance between
reading medical texts (especially those of theoretical content) and practising
as well as teaching medicine. Rather, I think that the interest shown by Argyropoulos and his friends in some theoretical, classificatory Galenic texts is
to be connected with their commitment to Aristotelian scholarship. In terms
of concept, logical articulation, theoretical tools and vocabulary, few Galenic
texts display more Aristotelian features in a medical account than books I-V
of Simple Drugs: it is thus difficult not to see a link between Argyropoulos pet
topic, Aristotle, and the peripatetic feel of Galens texts copied by his pupils.
But this path, of course, should be explored in more depth in Argyropoulos
preserved works, in order to assess the importance of Galen in his intellectual
constructions.
Going back to Pseudo-Galen: one clue makes Argyropoulos and friends
truly strong candidates for the anonymous comment in the Introductios pinax:
as a matter of fact, several manuscripts produced by this circle have the Introductio following the Ars medica just like the anonymous commentator stated
it should be in his second sentence of the prologue.23 If we accept this hypothesis as plausible, then we should consider that the circle of Argyropoulos (with
or without him) was trying to build their own collection of Galenic works for
the purpose of study and, perhaps, of teaching.
22
I suggested this in an article on the textual transmission of Galens XI

books On Simple Drugs: Theorie et pratique: connaissance et diffusion du traite
des Simples de Galien au Moyen Age, in Arsenio Ferraces Rodrguez (ed.), IIIe
Seminario internacional (Fito-zooterapia antigua y altomedieval: textos y doctrinas),
La Coru
na, 21-22 Oct. 2005, La Coru
na 2009, esp. 93-95. While Demetrios and his
colleagues duly copied and at times annotated books I-V (which emphasize method
and principles), they did not bother with copying books VI-XI (which contain the
catalogue of simple medicines). In the Urbinas gr. 67, a key witness for the text,
books VI-XI were simply added in the form of an old manuscript of the fourteenth
century.
23
Those manuscripts have close relationships with one another; they are Marcianus gr. App. V 9; Vaticanus gr. 285 and Parisinus gr. 2271.
68
PETIT
It may well be that the real instigator of this programme was in fact not
Argyropoulos himself, whose celebrity may misleadingly overshadow his colleagues to our eyes, but his successor in Constantinople. Indeed, Argyropoulos
left Constantinople in 1453, but the manuscripts that have the Ars medica and
the Introductio in a sequel are all later than that probably from the third
quarter of the century. The most studied one, the Marcianus gr. App. V 9
(in two volumes), containing an impressive Galenic collection, in great part
copied by Demetrios Angelos, is known to have remained in Demetrios possession (thus, in Constantinople) until 1473: an autograph note by Demetrios
states this very clearly, as observed by Brigitte Mondrain.24 The other two
are also likely to be later than 1453, especially the Vaticanus gr. 285, which is
considered to be from the very end of the fifteenth century and perhaps a copy
of Parisinus gr. 2271 (made by Demetrios): Agallon Moschos, another pupil of
Argyropoulos, could have been the scribe of this manuscript (again, according
to Brigitte Mondrain). Our evidence thus points towards a later date than
1453. In this case, it might be that either Michel Apostolis (who, as already
stated, succeeded Argyropoulos at the Xenon of the Kral in 1452) or Demetrios
Angelos, or both, bore the responsibility for this medical programme.
In turn, however, it could be argued that the archetype25 of the abovementioned manuscripts is the key to the mystery, and that manuscript could
well have been old enough to have been made in Argyropoulos time, or even
before.
Since the bigger picture is so blurred, it might be useful to turn to
the exact contents and differences between the two families of manuscripts.
The facts, once again, are intriguing. I shall first explain about the most
striking differences (number and order of chapters), before turning to smaller
(in appearance) details such as additional words.
The rewriting of the pinax in the family B manuscripts is not the only
distinctive feature of this group of manuscripts: they also display two additional chapters, inserted in the body of the text. The first one is a section on
leprosy, the second one a chapter on purgatives.
24
B. Mondrain, Comment etait lu Galien `

a Byzance dans la première moitie
du XVe s. ?, in A. Garzya-J. Jouanna (edd.), Trasmissione e ecdotica, Actes du IIIe
colloque international sur lecdotique des textes medicaux grecs, Paris-Naples 2003,
284.
25
I call this archetype z. It is not possible to attribute a precise date to its
completion. See my partial stemma of family A in my edition xciv.
69
Pseudo-Galen on leprosy (elephantiasis)

I have discussed this issue at some length in previous work; it is a difficult
question both for historians of medicine and philologists, for the terminology of
leprosy, its symptoms (current and previous, in ancient times and in the Middle
Ages), the alleged causes of the disease and its treatment form a nebulous
body of texts, not all being accessible to scholars, especially Arabic and Greek
medieval medical works.26 I have argued in favour of a Byzantine interpolation
in the pseudo-Galenic treatise for three reasons, based on textual criticism,
literary analysis, and terminology and medical historical problems.27 I want
to make it clear, however, that additional elements of explanation might chip
in at some point and change the picture.
The original section on leprosy (in Greek elephantiasis: 13, 43) follows
the standard pattern of the presentation of diseases in the whole chapter 13
(name of the disease, cause, treatment); in the family B manuscripts, however, it was completed with an additional development on six forms of leprosy, classified according to four animal forms (the elephant/elephantiasis; the
lion/leontiasis; the snake/ophiasis; the fox/alopecy) and two other categories,
called lobe (mutilation) and lepra. Discussing each of these terms in the context
of ancient and medieval pathology would take up a whole article.28 Indeed,
26
See my edition lxix-lxxvi.

Firstly, the passage does not fit within the context: it brings in redundant
remarks on such diseases as alopecy and lepra which appear anyway in the body
of the original Greek chapter and it does not correspond to the standard pattern
operating in the text (and especially in chapter 13). Secondly, the manuscripts that
preserve this section are all late medieval items (the earliest being of the mid-fifteenth
century), and they have undergone a number of changes that make such supplementary
passages look suspicious. Finally, the wording of the passage brings in some technical
terms and ideas that are barely or not at all known in the rest of our ancient sources.
28
Such ideas on animal forms of leprosy spread rapidly from the Islamic
world to Byzantium (see below the discussion of Al-Jazzars Viaticum), and to Western Europe via several Latin translations from the Arabic, such as Abu-l-Qasims Liber
Practicae and the De elephancia ascribed to Constantinus Africanus. I have consulted
the translation of the former by P. Ricci, 1519: Liber Theoricae necnon Practicae Alsaharavii in prisco Arabum medicorum conventu facile principis: qui vulgo Acararius
dicitur: jam summa diligentia et cura depromptus in lucem, f. cxxxv: Quando vero
generantur ex corruptione ciborum in quatuor modos dividitur, uno modo quae generatur ex dominio colerae rubeae et appellatur leonina, secundus modus est quae gen27
70
PETIT
each of them, with their accompanying definitions (based on appearance: for

each form of leprosy makes parts of the body look like animal parts; and
the last two also correspond to conspicuous transformations of the body and
the skin), deserves extensive discussion. There is in fact some uncertainty
about the origin of each of them, and we do not necessarily know when, and
in what context, each term was used leontiasis, for example, is attested as
an ancient medical term (Rufus of Ephesus, apud Oribasius Coll. Med. XLV,
28). Its pertinence in the study of leprosy is demonstrated by the fact that
the concept of leonine facies to describe the tumefaction of the face in some
lepers is still used in modern times. It is, however, an excessively rare term in
ancient Greek and, if both texts of Rufus and Pseudo-Galen echo one another
through a few rare terms, they do not give the same account at all of the
disease: Rufus explains that leontiasis is the name given to an initial stage of
leprosy, while Pseudo-Galen gives a static typology of the existing kinds of the
disease. The progressive deliquescence of the body in leprosy was probably
observed at an early date: Rufus source is the Hellenistic physician Strato of
the third century BC, and Celsus, though ignoring any specific terms labelling
the stages of the disease, shows clear awareness of the visible progress of the
disease.29 Galen himself is of little help when it comes to the description of
the symptoms (it is likely that, in his time, the understanding of leprosy had
eratur ex dominio colerae nigrae et appellatur elephantia, tercius est quae generatur
ex dominio flegmatis et appellatur serpentina, quartus modus est quae generatur ex
dominio sanguinis et appellatur vulpina. About the De elephancia, which is thought
to be inspired by Al-Jazzars lost treatise on leprosy and conveys roughly the same
text, see A.I. Martin Ferreiras edition (Valladolid 1996) and A.I. Martin Ferreira-E.
Montero Cartelle, Le De elephancia de Constantin lAfricain et ses rapports avec le
Pantegni, in Ch. Burnett-D. Jacquart (edd.), Constantine the African and ,Ali ibn
al-Abbas al-Majusi. The Pantegni and related texts, Leiden 1994, 233-246. The De elephancia uses the same terminology, with slight variations such as alopicia for vulpina,
and tyria (= tiria) for serpentina, albeit with the same meaning. On the discussion of
similar ideas and terms in Salerno, see A.M. Ieraci Bio, La relazione greca inedita di
una quaestio salernitana, in V. Boudon-Millot-A. Guardasole-C. Magdelaine (edd.),
La science medicale antique. Nouveaux regards. Etudes

reunies en lhonneur de

Jacques Jouanna , Paris 2007, 249-274.
29
Cels. III 25. Celsus passes rapidly over this disease, which, he says, is barely
known in Italy. Indeed, it was usually assumed that leprosy flourished in Egypt.
71
not really advanced since the early days). At any rate, the key to the problem
lies in the global description of the disease rather than in individual terms.
Why should Pseudo-Galen ignore the facts that earlier physicians knew
very well, while displaying such a puzzling description of the forms of leprosy,
as though he would tell us of several species of cabbage? Pseudo-Galen explicitly refers to the Ancients as the source of such names for the forms of
leprosy; but several ones are either not attested as such in antiquity (ophiasis
and alopecy, both being separate skin diseases in ancient texts)30 or attested
at a relatively late date: the term lobe (mutilation; surviving in modern
Greek loba), and the matching passive perfect participle lelobemenos (mutilated) were used as euphemistic/apotropaic denominations of the disease and
of the ill in Roman Egypt around the fourth century AD (the facts hint only
at the participle, though).31
My theory is that the passage is in fact inspired by much later influences:
as it happens, a fourfold presentation of leprosy, according to animal shapes
(precisely the elephant, the lion, the snake and the fox) appears in a popular
medieval Arabic work, Ibn al-Jazzars Viaticum (Zad al-musafir), book VII.32
In Al-Jazzars theory, each of the animal-inspired forms corresponds to one
of the four predominant humours. Of course, it is striking that that humoral
element should be absent from Pseudo-Galens account if al-Jazzar were to be
the direct source of our controversial passage. And the passage lists six forms
of leprosy, not four. But the popularity of that work in Byzantium is even
more striking, since the ninth century text was translated as early as the tenth
century into Greek; besides, the translation appears in some fifty manuscripts,
several of which date back to the tenth and eleventh centuries. Thus the Viaticum enjoyed considerable success in the Greek medieval world, including
Constantinople. It may therefore have influenced a number of Greek medical
30
They appear later in chapter 17 of the Introductio; the author specifies that
he is talking about diseases that he has not mentioned earlier in the book (well, that
is a clue against the authenticity of the passage on elephantiasis).
31
Jean Gascou, Lelephantiasis en Egypte greco-romaine. Faits, representations, institutions, in F. Baratte (ed.), Melanges Jean-Pierre Sodini, Paris 2005,
261-285.
32
Al-Jazzar wrote a monograph on leprosy, but that treatise is lost. The
Viaticum was also extremely popular and influential in the Latin-speaking world,
especially Spain. I owe Peter Pormann the information about, and the translation of
the relevant section in the Viaticum.
72
PETIT
works. Last but not least, there is at least one manuscript that displays both
the Viaticum and the pseudo-Galenic Introductio: the Lavrae Omega 72 (family A) of the fifteenth century. That manuscript does not show evidence of
interaction; but the fact that both texts appear in the same book corroborates
the possibility that both were read by the same people, and potentially used to
annotate, if not correct one another at some point in history.33 Given their respective popularity and accessibility, one can reasonably speculate that both
texts were part of the background of medical scholars of Palaeologan Constantinople. The content of the section on leprosy, however, does not provide
us with much information about the identity of the person who appropriated
the pseudo-Galenic text, nor does it help us to settle on a more accurate date
of intervention; but it gives us an insight of how wide-ranging and diverse the
Byzantine medical culture was in the Palaeologan period, and how problematic our conception of the early history of leprosy is. As is already known from
studies on other ancient works (such as Galens and Dioscorides pharmacological works), medieval interventions in texts seem to reflect cross reading
of medical works either transmitted in Greek from antiquity, or produced in
the East and translated into Greek in the Middle Ages, or indeed written by
Byzantine scholars. The rest of the specific features that we find in the family
B manuscripts further confirms this.
Indeed, an additional chapter appears after what is now chapter 15 (peri
pharmakeias) in the text (in our modern editions, the additional chapter features before that one): it is a chapter on the preparation of the best purgative
(a heading that barely does justice to the content of the chapter, which deals
with many categories of medicines), and its situation (textual history, vocabulary and problems of overall consistency with the rest of the book) makes
it exactly parallel to the leprosy section. I shall pass more rapidly over this
case. Apart from the main issue regarding textual transmission, which is of
course common with the section on leprosy, two main problems, in my view,
suggest that this chapter was not part of the original text: firstly, the chapter
displays some evidence of first-person discourse which contrasts sharply with
33
It was a common habit in Byzantium to mix ancient and modern works

in the manuscripts, on the basis of theme and usefulness. Several manuscripts of
the pseudo-Galenic Introductio also display some medieval Greek medical works, for
example the Parisinus gr. 2153 (fifteenth century), the Mutinensis gr. 213 (sixteenth
century) and the Vaticanus gr. 292 (fourteenth century).
73
the plain third-person account of the rest of the book;34 secondly, a few technical terms, such as sandalwood (santala) and camphor (kaphoura), hint at
medieval medical knowledge rather than ancient material.35
Yet another set of variant readings of smaller scale deserve attention in
the family B manuscripts. As a preliminary remark, I should state straightaway that these specific features appear in a fourteenth century Latin translation once ascribed to Niccolò da Reggio,36 which gives us a terminus later
than that of the above studied additional sections; such variant readings can
only be prior to the time of the translation, and there is no question here that
34
The text refers to two books of fairly common title (On simple drugs and
On the best way to compound drugs) in the first person.
35
Neither sandalwood nor camphor is attested in our texts before the Middle
Ages, in works by Symeon Seth and Nicolas Myrepsos, according to G. Helmreich,
Handschriftliche Studien zu Galen III, Program des k. humanistischen Gymnasiums
in Ansbach 1913/1914 (1914) 25. In fact, the history of those products is not
very clear. The Arabs most likely brought them both to the West from the East
through various routes (via Constantinople and via North Africa and Spain). Both
words appear in Simon of Genoas Clavis (see the edition of the text on the Simon
Online website under www.simonofgenoa.org) and in quite a few other medieval Greek
and Latin texts. But camphor, in particular, may have been known earlier than is
commonly assumed, according to R.A. Donkin, Dragons Brain Perfume: an Historical
Geography of Camphor, Leiden 1999, chapter 4. The evidence from late antique texts
is fragile and questionable, so that all suggestions that camphor and sandalwood
were known in the West in pre-Islamic times are inconclusive. Camphor, like several
expensive oriental products, was a privilege remedy of the Western elite and not
widely used by physicians before the fourteenth century. Its properties were known
and praised, especially as a means to fight the Black Death, but every single variety
of camphor was too costly to use on a common basis. Sandalwood may have been
known before the Arab conquest, but, similarly, evidence of its spread across the West
is mostly of medieval date (R.A. Donkin, Between East and West. The Moluccas and
the Traffic in Spices Up to the Arrival of Europeans, Philadelphia 2003, chapter 4).
36
About the authorship problem of this translation, see Galen On Prognosis,
edited and translated by V. Nutton, CMG V/8,1, Berlin 1979; on the Latin translation generally my edition cix-cxv for further literature. There is a possibility that
we find other sections of what may have been a complete translation of the Introductio; indeed, some quotations, notably one from chapter 13, appear in Latin medieval
texts. See Galen. On My Own Opinions, ed. V. Nutton, CMG V/3,2, Berlin 1999;
and my edition cxii and nn. 164 and 165.
74
PETIT
they could have been made at the time of John Argyropoulos. But there are
limitations to this comparison between the Greek original (of the family B)
and the Latin translation; since the Latin translation has no matching pinax,
and since it only concerns chapters 1-10 of the Introductio, it is impossible
to develop a full-scale comparison between the family B manuscripts and the
Latin translation. The date and origin of the elements that I have just studied
are to remain beyond our grasp, or rather, they are to remain datable of say
the fifteenth century at the latest.
The Latin translation was clearly made on a Greek model, and dates
back to an earlier archetype of the family B Greek manuscripts than that of
our fifteen-century manuscripts: indeed, the Latin translation has some specific additional features of family B, but it is also exempt from omissions of
the family B Greek manuscripts; besides, it is at times in agreement with
manuscript V (the family A prototype) against them. Finally, it bears traces
of at least one reading that has not survived in Greek apart from a marginal
note in a sixteenth-century edition.37 Among the additional features appearing
in both the family B manuscripts and the medieval Latin translation, some,
I believe, merely correspond to mistakes in the mainstream tradition as represented by manuscript V (omissions); others are more controversial, such as
the insertion of a gloss, poorly transmitted in the family B manuscripts and
reflected in the Latin (Introductio 9,5). Whether or not such features should
be connected with the above-mentioned more significant additions is unlikely,
but the hypothesis cannot be ruled out. In the latter case, John Argyropoulos
and his friends would have less than nothing to do with the adaptation of the
Greek text; rather, the modifications should be attributed to a much earlier
scholar. Indeed, it is even possible that the Latin translation was made after a
Greek exemplar earlier than V itself, thus from the early or high Middle Ages.
Conclusion
Surprisingly, the pseudo-Galenic Introductio owes much of its look in
the K
uhn edition to the Byzantine scholars who read, copied and annotated
it in Constantinople in the late Middle Ages. This is in great part due to the
work of Rene Chartier, who, in the early seventeenth century, discovered part
37
Here I am summarizing the data presented in my edition cxiii-cxv. The

Renaissance edition that has the reading otherwise lost in Greek is the 1556 Juntine
edition curated by Agostino Gadaldini.
75
of the family B manuscripts and inserted the additional chapters into the text
inherited from Renaissance erudition. Then K
uhn had only to reproduce the
resulting Greek and Latin texts. But Chartier was not the only one to harvest
the benefits of the family B: as early as the 1550s, Agostino Gadaldini and
his colleagues gained some knowledge of part of the Greek manuscripts that
remained hidden (or unused) from the Aldine editors, and they also used some
of the alternative evidence at hand in the Latin medieval tradition.38 Hence
the Constantinople milieu that edited the text, be it that of Argyropoulos
and his pupils or not, played a crucial role in the formation of the text as it
has been known, used and perceived since the nineteenth century. The very
additions dating back to the Byzantine period were assumed to be part of the
original text and were commented upon as such; in turn, those passages, with
their unusual contents, laid the ground for further misunderstandings, just like
many other texts of troubled textual transmission, such as the pseudo-Galenic
Definitiones Medicae.39 The most remarkable discussion arose from Mirko D.
Grmek about the passage on leprosy in chapter 13: far from suspecting the
38
C. Petit, Les manuscrits medicaux de Modène et la tradition de lIntroductio

sive medicus pseudo-galenique, in A. Roselli et al., V Convegno Internazionale
Trasmissione e ecdotica dei testi medici greci, Napoli, 1-2 Oct. 2004, Napoli 2005,
167-187.
39
Jutta Kollesch has shown that the text of the Medical Definitions underwent
a multiple interpolating and rewriting process that started early in the Middle Ages
with Monk Meletius, and ended with Rene Chartiers own manipulations. In the
manuscript Palatinus gr. 297, the text of the Medical Definitions was rearranged and
interwoven with extracts from Meletius De natura hominis (ninth century) in order
to produce a new, up-to-date medical work that could be useful to the readers of the
time. But a considerable part of the Greek manuscripts that transmitted the text is
in fact interpolated. Later on, Rene Chartier in turn was concerned with offering the
most useful collection to medical students in Paris, and altered the text considerably
by adding around seventy definitions of various origins. The very form of a collection
of definitions is of course particularly vulnerable to such changes, as the flow of the text
needs not be broken. It is also one of the most popular text-types in medical literature.
About the successive transformations of the text, see J. Kollesch, Untersuchungen zu
den pseudogalenischen Definitiones Medicae, Berlin 1973, 143; and Rene Chartier als
Herausgeber der Werke Galens, in J. Burian-L. Vidman (edd.), Antiquitas GraecoRomana. Acta congressus internationalis habiti Brunae diebus 12-16 mensis Aprilis
MCMLXVI, Prague 1968, 525-530; Rene Chartier, Herausgeber und F
alscher der
Werke Galens, Klio XLVIII (1967) 183-198.
76
PETIT
dubious nature of the section, Grmek commented upon the originality of the
pseudo-Galenic account in the context of (poor) ancient medical knowledge on
leprosy.40 If, as Grmek suggests, the pseudo-Galenic text has contributed to
confuse the conceptual distinctions between elephantiasis (leprosy) and lepra
(a benign skin disease) in the West throughout ages, it is in fact because
the passage under examination did not belong to the original but was added
at some point by some Constantinople scholar who was well acquainted with
Islamic scholarship on the subject. Thus, the Byzantines readings and textual
manipulations of ancient texts echo in modern times in the most unexpected
ways.
Medical texts for beginners like the Introductio, and the Definitiones
Medicae, became influential under the name of Galen, while conveying somewhat different ideas than those of the prestigious physician. This bothered
neither Greek medieval physicians, nor their Latin counterparts from late antiquity till the end of the fifteenth century at least. It is likely than the
criterion of usefulness was much stronger than that of authenticity. For even
in the case of major (authentic) Galenic works, textual manipulations were
common practice: the altered and augmented recipes of Galens treatises On
compound medicines as transmitted in medieval manuscripts testify to it; so
does the Vaticanus gr. 284, offering a mixed text of Galens Simple Drugs and
of Dioscorides Materia Medica. The categories that divide medical texts for
us now were at the time blurred, to the extent that one could mix together
in the same book, or even in the same text, several bits from various authors,
ancient and not, Christian and Islamic.
East and West of the Adriatic, the pseudo-Galenic Introductio has raised
interest and incurred changes to the text, with the purpose of appropriation
as well as adaptation to new audiences. It is, in many ways, representative of
the fate of Galenic works beyond antiquity, and this should stimulate further
investigation into the Byzantine reception of Galenic material.
Univ. of Warwick, Dept. of Class. &
Anc. Hist., Coventry CV4 7AL, UK
40
Caroline Petit
[email protected]
M.D. Grmek, Les maladies `

a laube de la civilisation occidentale. Recherches
sur la realite pathologique dans le monde grec prehistorique, archaque et classique,
Paris 1983, 250.
77
Appendix:
Pinax of the family A manuscripts (as appearing in the prototype V = Vaticanus gr. 1845)

,

...

,

.
Pinax of the family B manuscripts (as appearing in manuscript U = Urbinas
gr. 67)
+

, ,
,
, .
.
Aristotle and the caliphs dream. Aspects of medical translations
The caliph al-Mamun welcomed Greek medicine [. . . ](and) believed that it had obvious and practical advantages.1
Scientific tradition needs more than the mere copying of texts.2
That drugs are the hands of the gods is a familiar aphorism.3 Wholly
mortal was the transmission of Byzantine medical and scientific texts to the
east and their subsequent passage westwards. The history of these events
spans five centuries and is familiar enough not to need here anything but a
brief factual summary of its principal aspects. There follows this summary
an outline of the last sixty folios of a fifteenth century codex containing, inter alia, Byzantine translations of Persian antidotaria. The first theme, the
major Graeco-Arabic translation movement and the second, the subsequent
Persian influence in Byzantine medicine, are then examined in the context of
the translator and his medium, the manuscript.
The single-mindedness of an early Abassid caliph in the Islamic East, alMamun, instigator of the Herculean labour of copying and translating GraecoByzantine medical and scientific texts has served medical historians well. In
contrast, the later and less well-known Byzantine reception of remedies from
Persia is usually seen as of less significance. Gutas calls the Graeco-Arabic
translation movement a social phenomenon and dismisses theories that it was
based on scholarly zeal or the wisdom and single mindedness of enlightened
rulers, if only because of its complexity and deep-rooted nature.4 From the
1
S. Vryonis, The Medical Unity of the Mediterranean World in Antiquity and

the Middle Ages, Crete 1991, 27. The caliph ruled from 813-833.
2
R. Browning, Byzantine Scholarship, P&P XXVIII (July 1964) 3-20.
3
Attributed to Herophilos of Alexandria.
4
D. Gutas, Greek Thought, Arabic Culture, London-New York 1998, 3-6.
80
BENNETT
time of the Byzantine writer Symeon Seth to the fall of Constantinople, the
Persian strain in Byzantine medicine is less well understood.
The Manuscript: Par. gr. 2194
The last sixty folios of codex Par. gr. 2194 (ff. 400v464v), a fifteenth
century manuscript on paper, are the foundations of the discussion that follows
in parts I and II below. These folios include translations into Greek of Persian remedies, including that of Constantine Meliteniotes whose single work is
briefly discussed here; almost nothing is known of his life, and even his dates
are uncertain, perhaps mid-fourteenth century. The same folios also contain
two Byzantine hospital remedy texts copied in a poor and almost illegible
hand, perhaps hastily scrawled for personal use. These two texts are respectively headed Dynameron xenonikon dia peiras and Xenonika. The Xenonika
has several references to a hitherto unrecorded Mauraganos Hospital and its
unnamed aktouarios. Of the two texts, however, the fons et origo of this paper
is the first, the Dynameron. Throughout its seventy-six remedies, a number
of drug ingredients, almost certainly of eastern origin, are included in recipes,
suggesting an early exemplar of this emerging eastern influence in Byzantine
medicine and pharmacy.
Part I below considers aspects of the Graeco-Arabic translation movement, and Part II Persian influences in late Byzantine medicine and pharmaceutics.
I. Aristotle and the caliph
The story goes that the ninth century Caliph Al-Mamun dreamed that
Aristotle appeared to him and told him that Greek reasoning and Islamic
revelation were not incompatible. The next morning, the caliph gave orders
that all of Aristotles works be translated into Arabic. Whether or not there
is a grain of truth in this anecdote, there seems little doubt that Hellenised
Syrian Christians, who some five hundred years earlier had made their own
translations from Greek scientific and philosophical works, were a source for
the caliphs translators. Scholars of the movement have estimated that it
continued for a century and a half to two centuries. Clearly it was not a
sharply defined movement that ended on an appointed date. It took place
at a caliphs pleasure and on his initiative, and continued for many years
Aristotle and the caliphs dream
81
after the close of his reign. By the twelfth century many Greek scientific and
philosophical works that otherwise would be lost reached Europe, by way of
Islamic Spain and consequently were preserved for later ages.
To put the Graeco-Arabic translation movement in a wider perspective
calls for an initial reflection on Greek medicine, its inheritance and development in Byzantium, and its subsequent transmission to the Arab peoples,
chiefly through the instigation of the Caliph al-Mamun. At this time, the
preparation of medicines was the privilege of physicians; a separate discipline
of pharmacology did not yet exist. That was to come in the eleventh century at
the time of Avicenna who is regarded as having separated the art of medicine
from the skills of compounding drugs, thus earning the sobriquet of the father
of modern pharmacology.
The Arabic language was to become the lingua franca of the Arabic
world, but Arab technical advances were yet to be made. Amongst neighbouring lands, knowledge of and skills in medicines were generally most advanced in
Persia at the time of its Arabic conquest in the mid-seventh century.5 Translations of Greek medical works had been made into Pahlavi, or middle Persian,
and their pharmacopeia was based on Greek herbals. The language used by
the Persians for their treatises was none the less Arabic in which equivalent
technical terms necessary in translation were often lacking. In consequence,
transliteration of source words, here Greek, was often adopted, sometimes
being filtered through Syriac translations, but not always with comprehensible results.6 Ullmann observes that many Greek words were translated into
Arabic from Syriac and not directly from the Greek.7
The Sassanid empire came to an end in 644.

For useful examples, see H.D. Isaacs, Medieval Judaeo-Arabic medicine as
described in the Cairo Geniza, Journal of the Royal Society of Medicine LXXXIII
(November 1990) 735.
7
Ullmann, Islamic Medicine, Edinburgh 1978, 15f.
6
82
BENNETT
The Graeco-Arabic translation movement

The Iliad and Odyssey are held up as examples of works handed down
by word of mouth until the Athenians committed them to the page, a small
task compared with that of translating and transcribing the principal GraecoByzantine medical and pharmaceutical texts into the Arabic tongue, on the
orders of the caliph. Customarily referred to as the Graeco-Arabic translation
movement, it occupied numerous scribes and translators for some 150 years or
more. The scope of the work and the will that drove it are remarkable. Many
Greek and Byzantine scientific works might have been lost to later ages if it
had not been for their Arabic preservation in translation. The total number
of Greek works translated is estimated to be some 149, including among that
number most of the works of Aristotle, Hippocrates, Galen, Euclid, Ptolemy,
Dioscorides, Oribasius, and Paul of Aegina.8
The term movement suggests a neat and tidy process that began and
ended within a given timescale, but that is too simple a view. The task took
place against a background of problematic military and diplomatic relations
between Byzantium and the Arab Empire. It could have been an undertaking that never happened if the story is to be believed of the response of the
Ummayad Caliph ,Omar (regn. 634-644) to his field commander at the time
of the Arab invasion of Alexandria.9 Asked what to do with the library at
Alexandria, ,Omar is said to have replied in terms that authorised the destruction of books that accorded with the Book of Allah, which alone was
sufficient for the people; for those books that did not so accord, destruction
was also good enough for them. One hundred years later Caliph al-Mamun,
who, unlike ,Omar, was an ,Abbasid of Persian stock with an appreciation of
the value of knowledge and scholarship, advanced the scholarly translation of
Greek scientific and medical works with the energy and perseverance alluded
to earlier.
By the thirteenth century, the movements momentum had dwindled.
In the middle of that century, however, two major events took place within
a few years of each other, the first affecting the remaining centuries of the
Byzantine imperium whose seat had been in Nicaea since the Latin occupation
8
Said Faiq, Culture and the Medieval Arab Translator, Perspectives VIII/2
(2000) 91. Other works were translated from Hindi, Persian, Syrian and Sanskrit.
9
With acknowledgements to Dr. Richard Stoneman who made reference to
this apparent paradox in his Presidential Address to the Classical Association in 2010.
83
of Constantinople and its territories; the second, the destruction of Baghdad

and the Arabic ,Abbasid dynasty in 1258 at the hands of the Mongols who
sacked and razed the city with great loss of life.
As for Byzantium, in the early morning of 25 July 1261, Emperor Michael
VIIIs forces took possession of Constantinople without striking a blow. It was
to prove the effective end of the Latin occupation. The Empire of Nicaea had
returned to its home. Byzantium survived, a ghost of its former self, until 1453
when the Ottoman Mehmed in turn razed Constantinople, so ending over a
thousand years of the Byzantine Empire. The fate of the population of the
Queen of Cities can only be imagined; among those who had the resources,
many scholars made their way to Italy and shared their scholarly traditions.
It has been suggested that following the destruction of Baghdad Arab
scholars may have fled Baghdad, and come to Byzantium, but this is impossible to prove. An exodus of this kind might have been of common benefit, for
despite Byzantiums now reduced military strength, its creative faculties and
scholarship flourished, not least those of the practice of, and developments in,
medicine. Baghdads destruction and Constantinoples recapture took place
so close together that Touwaide has suggested that the subsequent presence
of, for example, Arabic physicians in the Byzantine lands was owed to the
fall of Baghdad.10 He calls in evidence those instances in Byzantine medical manuscripts of this period in which there is an indication, albeit minor,
of Arabic collaboration or influence, for example in bilingual lexica of plant
names. Whether or not this conjecture can be substantiated, the fall of Baghdad effectively ended the Graeco-Arabic translation movement.
It was a time in which numerous remedy texts were produced, usually of
no known provenance, but found chiefly in medical manuscripts, and occasionally elsewhere, even in the midst of a theological codex. They undoubtedly had
utility, not least for those distant from a Byzantine metropolis without easy
recourse to doctors. Many remedies are practical even if their medical logic
is now unclear; others are patently nonsensical or magical. There were also
major remedy texts, such as that of the twelfth century Nicolaos Myrepsos,11
10
A. Touwaide, More than the Sex of Angels, History of Science Society

Newsletter XXXVII/2 (April 2008) on-line.
11
On Myrepsos, see Anna Maria Ieraci Bio, Testi ginecologici tra Oriente ed
Occidente: 1. Metrodora ed il Dynameron di Nicola Mirepso, in Danielle Jacquart-A.
Paravicini Bagliani (edd.), La Scuola medica salernitana. Gli autori e i testi, Firenze
2007, 283-314.
84
BENNETT
which recorded some 2676 remedies from near and far, and were to gain sufficient credence to become the principal pharmaceutical code of the Parisian
medical faculty until 1651 (see II).
Linguistic Difficulties
Values and scholarship are of little use if the work of the copyists and
translators is too easily subject to scribal error and carelessness, or even distortion and inaccuracy. In the early years of the translation movement first drafts
in Arabic needed revisions interpreting both sense and underlying meaning.
In mitigation, rendering Greek texts into a tongue of another linguistic family
was an undertaking more easily prone to scribal error, problems of sense and
of translation of the names of ingredients.12 Differing syntax was perhaps the
chief difficulty; solutions may have lain in paraphrase and summaries. Often
in the early stages, transliteration of botanic names was the translators only
resort, especially in the matter of equivalent plant names. But these difficulties were not insuperable as a translator of ability and renown, Hunayn ibn
Ishaq (808-873) also known in Latin as Johannitius, was to demonstrate. An
Arab Nestorian scholar physician, he mastered four languages, Arabic, Syriac,
Greek and Persian. It was said that his ability was such that his translations
combined accuracy and sense in such a way as to convey meaning in his translations of Greek medical and scientific works. Tadhari ibn Basil Akhi Istifan,
a working colleague of Hunayn, translated Dioscorides De materia medica but
Hunayn was obliged to revise Istifans translation because of the unsatisfactory
nature of the transliterations of Greek words.
In short, translation offers every opportunity for error. Poorly understood terminology, lack of an adequate medical vocabulary, false or erroneous
translations of plant names or lack of equivalent names in Arabic lands opened
the way to confuse and corrupt the original text. Linguistic problems associated with translation included conceptual complications; consequently, the
relative merits of literal or free translation were disputed. Where accuracy
was not easily attainable, it was open to a translator to add explanations or
additions.
12
M
ullers theory of a family relationship between the Indo-European and
Semitic languages (of which Arabic was then a minor branch) has long been supplanted.
85
As medical texts require clarity, their translator needs not only an understanding of diagnostics but also of pharmacognosy. While this appears to
impose todays standards on another age and on the limits of its knowledge,
a translator of medical texts needs an element of medical learning. Thus,
understanding the terminology of medicine and drugs becomes a necessary
requirement for the translator from a different language.13 A reading of Constantine Meliteniotes Greek translation of Persian remedies shows, as Kousis
observes, that a large number of medical plants is cited with their Persian,
Arabic or Syriac name written in Greek characters (see II).14 He adds that
transcription in Greek characters brings about the alteration of words for ingredients and makes their rendering difficult today. None the less, the outcome
in practice was the creation of an Arabic medical and pharmaceutical terminology that subsequently was, in turn, to loan not a few of its words to the
European family of languages.
The history of translations of Dioscorides De materia medica makes
a useful background to the foregoing observations on the art of translation.
Among the earliest attested translations into Arabic of the ninth century was
one from the uncial Greek text of Dioscorides.15 Early translation into Arabic
of his De materia medica was to stimulate pharmaceutical research in the Islamic world. The gift of an illustrated codex of the Dioscorides manuscript by
Constantine VII to the Caliph ,Abd al-Rahman was valuable for its illustrations that furthered translation into Arabic, and contributed to establishing a
common recognition of plants and their names.16
13
P.E. PormannE. Savage-Smith, Medieval Islamic Medicine, Edinburgh
2007, 31f.
14
A.P. Kousis, Quelques considerations sur les traductions en grec des oeuvres
medicales orientales et principalement sur les deux manuscrits de la traduction dun
traite persan par Constantin Melitiniotis, PAA XIV (1939) 205-220.
15
D. Gutas, o.c. 182. See also passim J.M. Riddle, Dioscorides on Pharmacy
and Medicine, Austin 1985, xxv and the admirable translation of his De materia
medica by Lily Y. Beck (see below).
16
Translations of so influential a text as the De materia medica did not cease
with its Arabic reception. Editions in Latin, Spanish, Dutch, French and Italian are
known in the fifteenth and sixteenth centuries; in 1655 Goodyer translated the text
into English. The most recent and valuable translation into English is that of Lily Y.
Beck, Dioscorides De materia medica, Hildesheim 2005.
86
BENNETT
The Manuscript as Custodian of Knowledge

However accustomed peoples have been to disease in every age and country, as well as to promised cures, the upper hand has rested with disease. All
too often in earlier medicine the source of a disease has been hidden; as the
unknown writer of The art in the Hippocratic corpus observed:
Some, but only a few diseases have their seat where they can be
seen: others, and they are many, have a seat where they cannot
be perceived.17
The Byzantine physician was therefore faced in his time with more than one
dilemma, the nature of the presenting symptoms, what course of treatment to
embark upon and what drugs might be helpful. Some drugs were recognised as
valuable if the diagnoses were broadly correct and this herb or those ingredients
had worked in similar circumstances in past experience. In later centuries some
medicines might have anything up to a hundred ingredients in an attempt to
meet every recognisable aspect of a patients presenting symptoms, but recipes
of that kind were impractical in daily life.18 Diagnosis in turn was dependent
on the doctors prognosis; if, in simplest terms, he saw that recovery was
impossible (the patently hopeless case) he was unlikely to try any treatment
and might cease attendance.
With time the number of potential medicaments and ingredients had
increased in the continual search for cures. For example, it has been claimed
of Symeon Seth (fl. around 1050), philosopher and scientist, that sixty percent
of the drugs referred to in his texts are not to be found in Galen.19 Arabic
medicine, in turn, had access to numerous ingredients, not always familiar
to Byzantium. Small markers of Arabic influence in Byzantine manuscripts
of these centuries are in evidence; for example, the use of transliterated Arabic words and the introduction of glossaries in Byzantine manuscripts.20 In the
17
Hippocr. De arte 9; translation quoted from: Hippocrates, II, with an English transl. by W.H.S. Jones, London-New York 1923, 206f.
18
See J.M. Riddle, Theory and Practice in Medieval Medicine, Viator V
(1974) 157-184. Remedies are none the less amenable to explanation, ingredient by
ingredient. See also J. Scarborough, review of V. Nutton, Ancient Medicine, LondonNew York 2004, BMCRev (2005.07.74) on-line.
19
T.S. Miller, The Birth of the Hospital in the Byzantine Empire, Baltimore
1997, 165 n. 157.
20
See for instance Serikoffs contribution in this volume.
87
tenth- or eleventh-century Vat. gr. 284 containing texts from Galen, Dioscorides,
pseudo-Dioscorides and Philumenos, there are folios with coloured illustrations
of noxious herbs and animal life, against which are written their names in Arabic script. Glosses or marginal notes also occur, as in the eleventh or twelfth
century Vat. gr. 300 that has a marginal note at f. 273r that reads:
The unseen pores (are) in Arabic ettou chalchoul; others (are
called) elmesem.21
Probably just as useful to one owner of Vat. gr. 298 (dated between 1385-1389)
was the , a glossary of almost forty Greek medical words translated
into Arabic (f. 590v).22
At the same time, these fragments suggest receptiveness to knowledge.
The knowledge of pharmacy and medicine in Byzantium has, in the past, been
portrayed as static. That view is now obsolete, as Nutton makes clear.23 Aristotle Eutychiades summarises changes in drug therapy in Byzantium over the
centuries in terms, first of the introduction of new substances and then of new
uses of known substances; to these he adds, quoting Aetios, the modification
of the composition of traditional preparations.24 There follow these changes
the addition of a now increasingly developed Arabic pharmacology fortified by
the more advanced knowledge of Persia and Syria. At this time, technical
translation from Greek into Arabic and contrariwise presented not a few difficulties, particularly in plant names (see above). Nor were all their names either
adequately identified or rendered; some were simply transcribed or had more
than one name. It takes, however, more than fragments of correspondence
and anecdotes of this kind to portray aspects of transmission of knowledge in
a language other than the indigenous one.
21
, . . .
Elsewhere in this codex is a remedy described as
.
23
V. Nutton, From Galen to Alexander, Aspects of Medicine and Medical
Practice in Late Antiquity, DOP XXXVIII (1984) 1f.
24
A.C. Eutychiades, , Athens 1983,
257f.: 258 (d) in respect of the modification of preparations. For a discussion of pharmacodynamics in Hippocrates, see J. Stannard, Hippocratic Pharmacology, Bulletin
of the History of Medicine XXXV (1961) 497-518: 510-514.
22
88
BENNETT
Towards a Summary
The separation of the disciplines of medicine and pharmacy is now taken
for granted; it was not always so. The writings of the Arabic doctors from the
eleventh century disclosed a conceptual understanding of the role of pharmacy
in medicine. The effective separation of the two disciplines is usually attributed
to the time of Avicenna (980-1063); pharmacy made advances within Arabic
medical science as both disciplines came to be recognised further afield. For
example, Abram Sarakenos was aktouarios or court physician at the Mangana
Hospital in Constantinople, arguably in the last century and a half before the
Latin conquest of the city.25 It is possible to suppose he was from the Near
East (Abram equating to Ibrahim in Arabic).26 Abram Sarakenos, however,
brought with him remedies from his homeland; one of them is given in Vat. gr.
299. Each ingredient of this remedy is transliterated from Arabic into Greek
letters and a Greek translation is then added, preceded by that is to say, thus
rendering the text effectively bi-lingual and of use to Arabic and Greek speaker
alike. The contents begin: Alilekch, that is to say, Arabic balsam; (add) two
measures of indigo. . . 27 The number of extant remedy lists, anonymous or
bearing the authors name, are numerous; equally numerous must be the lost
texts. If a census of Byzantine remedy texts could be undertaken, a continuous
repetition of particular remedies, usually with minor variations, would become
evident. The same supposition is likely to hold for Arab medicine. Ullmann
observes that Arabic bibliographers recognise more than a hundred authors
who wrote about materia medica. But only a few of these works are original
independent achievements.28 In other words, there were repetition, variation
and signs of individual preferences, contributions, perhaps, to the dynamic of
25
Vat. gr. 299, ff. 373f. There appears to be no evidence that the Mangana
Xenon survived the Latin occupation.
26
In the same manuscript, Stephen wrote to John Chale (most probably
Khaled) about stomach, spleen and liver affections. Khaled, a pre-Islamic Arabic
name has a meaning of immortal.
27
Vat. gr. 299 f. 374r, l. 22, in chapter (641). .
() . .
. .
. .
. .
. . . .
28
Ullmann, o.c. 103.
89
pharmacy until overtaken by the pharmaceutics of bio-medicine.

Arabic pharmacy effectively came into being under Umayyad rule in
the guise of alchemy. The Prophet is often quoted as saying that for every
malady, he has appointed a suitable remedy. Under the subsequent ,Abbasid
caliphs, initially under Caliph al-Mansur (754-775), founder of Baghdad, and
later Al-Mamun, began the acquisition of knowledge of sciences from other
cultures. It was in the Arab lands that pharmacology came to be seen as
a separate profession from medicine that itself developed from the study of
materia medica. Little by little the emphasis changed from alchemy (as now
understood) to the pursuit of pharmacy through Arab study of Greek and
eastern sources. It was underpinned at this stage by translations from Greek
medicine, first made into Syriac, then translated into Arabic; a third phase
was the revision of the earlier translations of Dioscorides.29 The Persians saw
themselves as stewards of knowledge that combined, in the case of medicine
and pharmacology, knowledge from not only Greek, but also Hindu and Islamic
practice. If stewards they were, they combined not only a thirst for knowledge,
but an accompanying search for means of healing. Had Aristotle appeared to
the Syrians in a dream, he might have reminded them of his observation that
All human beings by nature desire to know.30 Happily, all Aristotles works,
save one, were to be translated in the movement.
Revisiting the Par. gr. 2194
In the last centuries of their shrinking imperium, the Byzantine reception
of Persian remedy texts is something of an enigma. Little has been written
about it; its extent is traceable in a few extant Byzantine texts including the
one text of Symeon Seth in the eleventh century (the Syntagma) and one of
Nikolaos Myrepsos (the Dynameron) in the following century. The four minor remedy texts from Persia in the Par. gr. 2194 is fortuitous. Explanations
of this Persian transmission may be several, for example as a kind of cosmopolitanism, a search for new medical knowledge, the pursuit of new cures,
travellers remedies from afar. Yet, Greek medical works had been translated
29
Touwaide in J.W. Meri, Medieval Islamic Civilisation: an Encyclopaedia, I-II, New York 2006, I 607-609. He also sets out an analysis of this process in Lintegration de la Pharmacologie Grecque dans le Monde Arabe. Une Vue
dEnsemble, MedSec VII/1 (1995) 159-189.
30
Arist. Metaph. 980a 21.
90
BENNETT
into Pahlavi in earlier centuries; perhaps Persian medicine now had something
to offer Byzantium.31 Persia was strategically located between Byzantium
and India and links between these lands existed before Islam became the regional influence and conqueror of Persia. The presence of several Persian texts
in a single Byzantine codex suggests that other texts of this kind had been
exported to Byzantium, some now lost, some with no hint of their Eastern
origin. Others may have succumbed to loss and destruction of manuscripts in
the following centuries.32
II. Persia and Byzantium: Translators and Borrowers
A few centuries after the Graeco-Arabic translation movement made
so great a contribution to Arabic medical lore, remedies of Persian origin
were imported into the Byzantine pharmacopoeia. In earlier centuries the
Byzantine Empire was in constant conflict with the rise of Islam, yet at the
same time Islam valued Byzantine learning, including that in the medical
sciences. Browne, in his lectures on Arabian medicine in 1921, observes that
medical doctrine was only in a very small degree the product of the Arabian
mind and continues, the general history of medicine can only be studied in
connection with the general history of Islam.33 Fifty years later, Ullmann
remarks on the quantity of Arabic pharmaceutical compilations by more than
a hundred eastern authors.34 His observation is not necessarily a criticism; it
is equally arguable that extensive dissemination of the new knowledge gave
rise to a desire to spread this acquisition. Persia, and before her, Syria, had
also been bound up in the transmission of Greek medical works to the Arabic
lands in Syriac, an Aramaic language or dialect, both written and spoken until
the seventh to eighth centuries. That al-Razi was Persian is often overlooked
in the study of Arabic medicine.
In the later medieval centuries of Byzantium, Hunger has suggested,
pharmacology became the main focus of therapeutics.35 Of the more familiar,
31
P. Prioreschi, A History of Medicine, IV, Omaha 2001, 211 and Ullmann,

o.c. 16-19.
32
Par. gr. 2194, 2-4 and 8.
33
E.G. Browne, Arabian Medicine, Cambridge 1921, 2 and 4.
34
See n. 29.
35
H. Hunger, Die hochsprachliche profane Literatur der Byzantiner, I-II, Munich 1978, 308. Of his Syntagma de alimentorum facultatibus, nearly 87% of the
91
if less studied writers, Symeon Seth was an eleventh century Jewish Byzantine
doctor from Antioch whom Temkin has described as the great Orientalist of
Byzantine medicine.36 He wrote, among other works, On the properties of
foods, using not only Greek, but also Persian, Arabic and Indian sources.37
In the thirteenth century, Nikolaos Myrepsos (fl. mid to late thirteenth
century) in his great compilation of 2676 recipes for remedies, the Dynameron,
recorded recipes, not only those accessible to him in Greek medical texts from
Galen onwards, but also from Italian, Persian and Arabic lands. For this thirteenth century court physician of Emperor Ioannes III Dukas Batatzes, Myrepsos, preparer of ointments, is an apt epithet. His work has been compared
with the shorter Dynameron of Nikolaos Salernitatus, despite the Salernitans
fewer recipes (150) and its compilation two centuries earlier. Nicolaos is also
said to have been influenced by Synesios, an eleventh century physician who
introduced new types of drugs, including heiligma and zoulapia as well as ingredients including camphor and Arabic kokkidion. A study of the Persian and
Arabic remedies in the largest section of the Dynameron, sub voce A (for Antidotoi) discloses three recipes attributed to the Persians, one to the Saracens
and one described as Arabike.38
John aktouarios (1275 to around 1328) was the last practising physician
of note in Byzantium and chief physician at the court of Andronikos II Palaiologos. He wrote De methodo medendi, subdivided into six books; he studied
different forms of preparations, introduced new drugs and also wrote a work
on urology (this latter work is said to be a translation of Avicennas treatise
improved by John).39 So brief a summary does little justice to him either for
foodstuffs he records may be found in the works of Arabic physicians.
36
O. Temkin, The Double Face of Janus and other Essays, Baltimore 1977,
214.
37
See Hunger, o.c. 303.
38
Of these five remedies, recipe 66 is among a cluster of five (63-66) under
the broad classification he dia saturion. (that is, using the herb satyrion of which
Dioscorides says (III 128) it must be drunk in black harsh wine for tetanic recurvation
or lockjaw; and they say it is aphrodisiac). The second, recipe 301, is for all debility,
headache, strokes and the sensation of ants running up and down the body. The
Saracen recipe (recipe 221) is a Tryphera for excessive melancholy, every affection of
the head and the stomach and abdomen.
39
See A. Hohlweg, John Actuarius De Methodo Medendi. On the New Edition, DOP XXXVIII (1984) 121-133: 131 n. 81. Some have held that he lived in
92
BENNETT
his medical achievements or those in pharmacology.40 He wrote, however, as a

doctor for doctors, reflecting his practical experience at that time, as Hohlweg
notes.41 Whether or how far he was influenced by Persian medicine is a matter
for further research, but if his work on urology were founded on Avicenna it
would be safe to assume a Persian influence.
Other medical writers and translators of this period included Georgios
Choniates whose name is associated with the Antidotoi, translated from Persian into Greek, and Georgios Chrysokokkos of whose work some fragments
remain. Some seventy years ago Kousis discussed the Antidotoi of Constantine
Meliteniotes, a native of Melitene and near contemporary of Chrysokokkos and
Choniates.42 Whether he was a member of the well known Byzantine Meliteniotes family is uncertain, in concert with a doubt about the attribution of
the translation of fifty-three prescriptions in codex Par. gr. 2194. A flavour
of these Persian remedies is evident in the examples below. All have fairly
simple titles but claim to be preparations for complex or unusual symptoms,
for example:
Recipe 46 Matzoun, the phalasipha: Phlegm disperser, appetite
stimulant, an effective discutient; it is useful for forgetfulness,
stimulates the elderly and renews and arouses the libido; it fortifies the kidneys. 13 ingr.
The final remedy closes, as many such do, with what may only be termed a
tour de force on paper if not in practice. This example has only five more
ingredients than the first but casts its net more widely:
Recipe 53 Antidote: Beneficial for sufferers from epilepsy and
stroke, for sciatic indurations () of the sinews, for (sufferers) from trembling, for pains in the shoulder joints, for sufferers
the eleventh century, but if he translated a treatise of Avicenna it is more likely that
he lived in the last years of the thirteenth and first years of the fourteenth century
between 1275 and 1328.
40
See Eftychiades, cit. 297 for fuller details.
41
Hohlweg, o.c. 132
42
A.P. Kousis, Quelques considerations, passim. For the two manuscripts,
see H. Diels, Die Handschriften der antiken Arzte.

Griechische Abteilung,
APAW (1905/1906), Berlin 1906, 24 and Suppl. 1, 47.
93
from gout; also for throbbing and rheums of the teeth; for sufferers from melancholia, first signs of cataracts, those diseased in the
spleen and with pains in the ribs; also for stomach pains that come
from () fennel juice; effective for chest pains; also for splenic
jaundice sufferers; a purge for kidneys and bladder, stimulant of
the libido; and for chronic diseases; for gnawing pains in the intestines; beneficial also if introduced by a clyster and also for wild
beast bites. 18 ingr.
Space does not allow a reading of the ingredients of the recipes in Constantines
Greek translation from the Persian, but, as Kousis observes, a large number
of medical plants is cited with their Persian, Arab or Syriac name written
in Greek characters. He adds that transcription in Greek characters brings
about the alteration of words for ingredients and makes their rendering difficult
today, although some cited in Dioscorides De materia medica are more easily
recognisable.
Byzantium as Intermediary
Hunger wrote at the close of his chapter on Medicine in his Handbuch
der Altertumswissenschaft that:
. . . the Byzantines of the middle and late centuries played, time
and again, the part of intermediaries (Vermittler) between Oriental and Occidental medicine, and in their turn, received new
impulses from both East and West. During the late period, this
is particularly true of pharmacology which from the thirteenth
century onwards came increasingly to the fore.43
This leads the reader to assume that before the thirteenth century there were
indeed merits that Byzantium saw in Persian medicine that may be glimpsed
in surviving Byzantine medicine. Hunger discussed Seth in some detail, from
his works on dietetics to his Refutation of Galen. More importantly, Seth
introduced into Byzantine medicine Arab and Indian materia medica, but his
Arab sources remain unknown.
Persian pharmacology had a long history, both pre-Islamic and subsequently; at the time Hunger speaks of, Persian medicine had consolidated its
43
H. Hunger, o.c. 315.
94
BENNETT
foundations, not least in Rhazes in the early tenth century and Avicenna in the
early eleventh century.44 Rhazes (865-925), physician and scholar remains, in
the words of Browne, probably the greatest and most original of all the Muslim physicians, and one of the most prolific as an author.45 Persian medicine
before the Arab conquest of 642 has been described as an honourable profession, the doctor a counsellor of kings.46 One of its earliest notable writers
on drugs was ibn Sahl (obit. 869) whose Dispensatory was first compiled in
the ninth century and in the following centuries appeared in several recensions
one of which was compiled in the mid-eleventh century by the physicians of
the Adudi hospital in Baghdad.47 The last three centuries of the early Middle Ages saw both Persian and Byzantine medical writers placing on record
materia medica and remedy recipes of general practicality and orderliness.
In summary, there is a commonality in the practice of humoral medicine
(the heritage of Galen), and the pharmacological principles embodied therein.
Arabic medicine was to make advances in pharmacy, particularly in the use of
chemical compounds, and in clinical pharmacology, that is, the effect of drugs
on the body. In short, Byzantium was Hungers intermediary, albeit that in
turn Byzantium was also amenable to reception of Persian medical lore.
44
Three volumes of Rhazes medical encyclopaedia were devoted to pharmacology. Books II and V of Avicennas Canon were similarly on this topic.
45
Browne, o.c. 44.
46
C. Elgood, Persian Science, ch. 11, iv, in A.J. Arberry (ed.), The Legacy
of Persia, Oxford 1953, 310.
47
O. Kahl, Sabur ibn Sahls Dispensatory in the Recension of the Adudi Hospital, Leiden-Boston 2009; a companion volume is Kahls The Dispensatory of Ibn
at-Tilmid, Leiden-Boston 2007.
95
Without the Manuscript. . .

If there is no one prince to whom this prosaic envoi might be addressed,
the early Abassid caliphs, al-Mansur and al-Mamun, have credit for the initiation and impetus they gave to the Graeco-Arabic translation movement;
the renowned translator Hunayn ibn-Ishaq was their prime instrument for this
end. The movements subsequent influence has an historical importance beside
which the later Byzantine reception of Persian medicine and pharmacy over a
far shorter period is of lesser consequence, but indicative of a certain vitality
in these disciplines in the last Byzantine centuries. A Byzantine openness to
new knowledge and, so far as the nature of that knowledge allowed, means of
healing at so late a stage of the empire.
Temkins observation that Byzantine medicine [. . . ] represents the formation as well as the continuation of a tradition, broken and unbroken comes
to mind. It is arguable that the Arabic world inherited that tradition and
passed it on. Its agent was the social and political history of the Abassids
against a background of wider historical events, including signs of decline of
the Byzantine Empire and, in the Arab lands, conquest of neighbouring peoples and the establishment of the Caliphates.
In many ways translators have the leading roles in the translation movement as arbiters of meaning, knowledgeable in the subject of the text under
translation and sensitive to the nuances of the language being translated. Hunayn ibn Ishaq was reputed for the skill and numerous translations he undertook for the caliph; he fulfilled, it seems, these ideals. He set a standard for
those around him who were engaged in the same task.48 A modern necessity for
piecing together evidence of a topic of this kind is the survival of manuscripts.
The Byzantines were known for the copying of their inheritance of the Greek
masters; much is now lost with the passage of time. Manuscripts remained
subject to destruction, especially in war (witness the burning of great and
small libraries of the past), in intemperate climates and not least in a failure
to realise the significance of their contents (the palimpsest alone has seen the
loss of valuable material). A glance through Diels Die Handschriften der an
tiken Arzte
often singles out the losses where only one copy of a text survives
under a writers name.49 Happily, medieval Islam was a pre-eminent source of
48
49
ByzZ
See Pormann-Savage-Smith, Medieval Islamic Medicine cit. 25-33.

See A. Touwaide, Byzantine medical manuscripts: Toward a new catalogue,
CI (2008) 199208.
96
BENNETT
transmission through their translations from the Greek of works that might
otherwise be lost to later centuries where the Greek original failed to survive.
A summation of this paper might be Pormanns words, that medieval Islamic
medicine was not simply a conduit for Greek ideas but a venue for innovation
and change to which we might add and presentation.
Formerly NHS and RHUL
David Bennett
Syriac plant names in a fifteenth century Greek glossary (from

the Wellcome Library Books and Manuscripts)
The Syriac glossary

The Wellcome Library is one of the worlds most important resources
for the history of medicine and related subjects. Among its holdings is a substantial collection of early printed books and manuscripts in many languages
of East and West. The majority of Greek and Byzantine manuscripts were
acquired in 1984,1 when the manuscript holdings of the Medical Society of
London were transferred to the Wellcome Library. The 169 new acquisitions
included a fifteenth century book written in Greek,2 which comprised a collection of 43 diverse works of medical character. The work represented here
is a list of drugs accompanied by their allegedly Syriac counterparts. It covers only one folio (71v) and is described by the cataloguer, W.R. Dawson, as
3 The
A botanical list in four columns, headed Syriac Names of Plants .
list, still unedited, comprises interesting Graeco-Arabic parallels and names of
plants, written in vernacular Greek. It is hoped that the annotated edition of
this list offered here will contribute to many fields of medical, historical and
linguistic studies. Among these one can name the historical botany of what in
the fifteenth century was the Ottoman Empire and the history of the Greek
language. The late Greek materia medica, which comprised quite a significant
1
Dawson, Manuscripta Medica 68-72. The sale has been made possible
through the generous grant of the Wellcome Trust and good offices of E.J. Freeman
and H.J.M. Symons.
For full bibliographical information, please see the bibliography.
2
Shelved at WMS 60.
3
Dawson, Manuscripta Medica 70.
98
Syriac plant names in a fifteenth century Greek glossary
number of the graeco-oriental glossaries has been an object of intensive study

for the last two centuries. The studies by J. Boissonade, and later by A. Delatte,4 and M. Thomson5 and by L. Tartaglia,6 have amassed a great deal of
the plant-names used by Greeks in the late Middle Ages. The foundations
for the difficult task of correct identifications of often mysterious names were
already laid in 1866 by B Langkavel, who supplied the plant-names with their
Linnaean classification.7
Byzantine Glossaries of Oriental materia medica
Late Byzantine Greek glossaries of Eastern (Arabic and Syriac) materia medica display the level of linguistic perception of Byzantines by their
neighbours and enemies, the Arabs and vice versa. The already mentioned
J. Boissonade, M. Thomson and others tended to see transliterations in the
Saracenic names of the plants written in Greek characters, corresponding
to real Arabic plant names. Their efforts, however, often yielded insufficient
results, due mainly to the fact that the Greek transliterations of the Arabic
plant names were in fact originally Greek names, which were adopted by the
Arabs and later transliterated back into Greek. This was shown in 1999 by the
author of these lines.8 These Saracenic names often took their origin from the
lemmas, which introduced materia medica in various Arabic medical manuals
and dictionaries. The original words (in Arabic) were superscribed by their
Greek translations or explanations, which later were converted into independent word lists. This explains the origin of such strange remedies like
4
5
6
7
8
Delatte, Lexique, Delatte, Anecdota

Thomson, Textes grecs inedits.
Tartaglia, Il lessico medico.
Langkavel, Botanik
Serikov, Saracinskii leksikon
SERIKOFF
99
, which being read in the opposite (namely Arabic) direction yielded the

JK
P ,
, i.e. the word for word translation of the Arabic PB@
9
i.e. , etc.
The titles of such word-lists often show the intrinsic ways of how the
scientific Greek words made their way into Arabic and became domesticized
there. There were generally two classes of Greek words, called by the Arabs
themselves r
um (i.e. of later, mostly Byzantine origin) and y
un
an (i.e. old
Greek words, which came to Arabic through Syriac and Pehlevi). They often
existed on the parallel basis within one and the same work. People often did
realize that the y
un
an words came via Syriac language; therefore the titles
like 10 indicated the way these words penetrated into
Arabic rather than their origin.
Principles of edition
The present edition is diplomatic. The author of the present lines believes that the so-called critical editions produce one virtual, non-existent text
instead of the many which did really exist.11 Therefore all the accents and
spelling features have been retained. The botanical species were supplied with
their scientific classification on the basis of the recent works by A. Dietrich.12
The parallel readings were taken from my Dictionary of Greek Foreign and
Loan Words in Arabic.13
Serikov, Saracinskii leksikon 95 no. 5.

For the slightly awkward title Saracenic Dictionary
one may suggest a conjecture: the could be replaced by or
.
11
Cf. the corrections applied by L. Tartaglia in Tartaglia, Il lessico medico
esp. 549.
12
Dtr., Diosc. Erkl.
13
Comprises more than 50,000 lemmata; is currently being set up.
10
100
Bibliography
Abd
un Umda [Rab.] fol. = Muh.ammad b. Ah.mad b. Abd
un an-Naha at-Tugb
Kit
ab Umdat at.-t.abb f marifat an-nabat li-kull labb (Manuscript Gayangos XL).
Cited after Dtr., Diosc. Erkl.
Boissonade, Anecdota = J.F. Boissonade, Anecdota Graeca e codicibus regiis, Parisiis
1830.
Diosc. [Arab.] = La materia Medica de Diosc
orides, II, ed. C.E. Dubler-E. Teres,
Tetu
an-Barcelona 1952-1957.
Diosc. [Graec.] = Dioscurides. De materia medica, ed. M. Wellmann, Berlin 19071914.
Dawson, Manuscripta Medica = W.R. Dawson, Manuscripta Medica. A Descriptive
catalogue of the Manuscripts in the Library of the Medical Society of London, London
1932.
Delatte, Anecdota = A. Delatte, Anecdota Atheniensia et alia, Paris 1939, 279-393.
Delatte, Lexique = A. Delatte, Le lexique de botanique du parisinus graecus 2419,
Serta Leodiensia. Ad celebrandam patriae libertatem iam centessimum annum recuperatam composuerunt philologi Leodienses, Liège-Paris 1930, 59-101.
Diosc. Erkl. = Die Dioskurides-Erkl
arung des Ibn al-Bait.
ar: ein Beitrag zur arabischen Pflanzensynonymik des Mittelalters, arabischer Text nebst kommentierter
deutscher Ubersetzung,
ed. A. Dietrich, Gottingen 1991.
Dtr. = Dioscurides triumphans: ein anonymer arabischer Kommentar (Ende 12.
Jahrh. n. Chr.) zur Materia medica, arabischer Text nebst kommentierter deutscher
Ubersetzung,
I-II, ed. A. Dietrich, Gottingen 1988.
Ghaleb, Encyclopedie = Ghaleb Edouar al-Maws
uah f ul
um al-Tabah Encyclopedie
des sciences de la nature, Beirut 1988.
Langkavel, Botanik = B. Langkavel, Botanik der spaeteren Griechen vom dritten bis
zum dreizehnten Jahrhunderte, Berlin 1866.
Lexikon ton Sarakinon = The numbers of the lemmata in the Greek
published in Thomson, Textes grecs inedits, 139-168.
Lonitzer, Naturalis Historiae opus novum = A. Lonitzer, Naturalis historiae opus
novum: in quo tractatur de natura et viribus arborum, fruticum, herbarum, animantiumque terrestrium, volatilium aquatilium; item, gemmarum, metallorum, succorumque concretorum, Francofurti 1551-1565.
SERIKOFF
101
Serikov, Saracinskii leksikon = .. .

( "-" XV -
.) LVIII (1999) 84-103.
Serikoff, R
um and Y
un
an = N. Serikoff, R
um and Y
un
an: Towards the Understanding of the Greek language in the Medieval Muslim World, in K. Ciggaar (ed.), East
and West in the Crusader States. Context-Contacts-Confrontations, Louvain 1996,
169-194.
Tartaglia, Il lessico medico = L. Tartaglia, Il lessico medico del codice Lond. Med. Soc.
60, in A. Garzya-J. Jouanna (edd.), I testi medici greci: tradizione e ecdotica. Atti
del III. Convegno internazionale, Napoli, 15-18 ottobre 1997. Les textes medicaux
grecs, Napoli 1999, 547-557.
Thomson, Textes grecs inedits = Textes grecs inedits relatifs aux plantes, ed. M.H.
Thomson, Paris 1955.
Text
:
: :1
: :2
: []:3
: :4
: :5
: :6
[] : 7
[] : :8
[] : :9
[] : :10
: :11
: :12
: :13
: []:14
: :15
: :16
: :17
: :18
: :19
: :20
: :21
: :22
: :23
: :24
: :25
SERIKOFF
: :26
: :27
[]: :28
: 29
: :30
: :31
: :32
: :33
: :34
: :35
[] : :36
[] : :37
: :38
: :39
[] : :40
: :41
: :42
[] : :43
: :44
: :45
: :46
: :47
: :48
: :49
: :50
: :51
: :52
103
104
: :53
: :54
: :55
: :56
..... : :57
: :58
: :59
: :60
: :61
[] : :62
[] : :63
: :64
:
65
Notes
1
The is a misspelt word. The correct Greek form, which corresponds the

Jg or I

Jg, the devils dung would be *. However, in the
correct Arabic I
manuscript the final has been wrongly rendered as the , cf. no. 51. This plant is
.' B@ (Ghaleb, Encyclopedie no. 2470), which is probably
also known in Arabic as @Ym
the Ferula assafoetida. However, the plant has not been exactly identified until the
present day, cf. Dtr. III 75f.
2
The is a mushroom, which grows on trees, the Larix europaea or Larix
SERIKOFF
105
decidua L. and others. The Greek form recorded here is a transliteration of

K
PA (cf. e.g. Abdun Umda [Rab.] fol. 25a,21-27), which in turn was
the Arabic
a transliteration, almost certainly via a Syriac medium from the Greek .
3
literally means Armenian soil and corresponds to the

GAP@The
. A similar lemma is found in Lexikon ton Sarakinon no. 2:
[ AD

, where the last is, no doubt, [] , a white clay, cf. AJ@

J

K. @ AYg@ A
K AJ
J
A K
J
]
@ (Diosc.
Q
Q J
Q kB@
[Arab.] 440,15 [V no. 131]); (Diosc.

[Graec.] III 105.13 [V no. 156]). The expression [], however, is a

transliteration of a different species, the Armenian stone, in Arabic GAPB@ Qj.m '@ (IB
II 12,9).
The is a Greek translation of , which is transliterated

. Ik
into Greek from the Arabic AK
. the balsam tree. The word
has been recorded in Greek sources (Langkavel, Botanik 13,1) and also with a reference to Simon of Genoas, a physician to Pope Nicolas IV, multilingual dictionary as
accessed on www.simonofgenoa.org. The constructions with (a translation for
tree) are recorded in the Greek glossaries of the Arabic materia medica:
the I
. k
(Lexikon ton Sarakinon no. 1, cf. Serikov, Saracinskii leksikon
94) and elsewhere: (Langkavel, Botanik 218,4, (Langkavel,
Botanik 43,1), (Langkavel, Botanik 58,2), etc.
The is not the pumpkin, Cucurbita pepo, but, judging from the
Dtr. IV 167) wild colocynthe (Citrullus colocynthis
Arabic translation ( Jk
[L.]). Two words and are in fact the transliterations of the Arabic
.
Jk
; the is a transliteration
Jk
written, however, without a dot above the nun: Jk
. The
of the same word Jk
n
un was subsequently elongated and the whole word was corrected into g .
6
The is a Greek transliteration of
The (Convulvulus scammonia L.) is in Arabic
Q YK
Y
(Chynan-
106
tum indicum) according to E. Ghaleb (Ghaleb Encyclopedie no. 26080). Due to the

similarity of pronunciation of the Arabic and in the vernacular dialect of Syria
the word Q scent has been understood by the translator as
translated back into Greek accordingly: tear ().
Q
drop and
Correct: . This plant is still difficult to identify (Dtr. I 15).

The Arabic glossaries interpret it as QK
P Y@ I
. an odoriferous reed. An identical
interpretation is found in the Greek glossaries of the Arabic materia medica:
(Lexikon ton Sarakinon no. 143). The interpretation

is obviously a transcription of the Arabic hQ @ k literally guard
against inflammation.
(Santalum album,

K. @ Y J
The is the translation of the Arabic J
white Sandalwood (Ghaleb, Encyclopedie no. 16543, Langkavel, Botanik 186.3). It
corresponds to the first part of this lemma.
8
(Adenanthera), Red
The is translation of the Arabic Qg @ Y J
sandalwood tree (Ghaleb, Encyclopedie no. 16544). It corresponds to the first part of
this lemma.
9
, which is probably
The is translation of the Arabic Q @ Y J
Santalum, sandalwood (Ghaleb, Encyclopedie no. 16542). It corresponds to the first
part of this lemma.
10
. , the beavers
The Greek is a translation of the Arabic QYK. YJk
testiculi (Diosc. Erkl. II 24, Dtr. II 22). The Arab translators often confused two
similarly looking Greek words, , a secretion found near the hinder parts of
the beaver (Dtr. II 22) and , the beaver (Castor fiber ) himself.
11
12
The generic Greek means an ear of a corn, which corresponds to the

generic Arabic J. . However, from a botanical point of view this translation is less

kA @ most probably means the stachys

III 100) and J. is applied to the nard Nardostachys
correct: in the Arabic terminology
Stachys Germanica L. (Dtr.

jatamansi DC (Dtr. I 6), cf. Langkavel, Botanik 177,1.
13
The Greek is a composite of rush and flower (Dtr.

I 14). This is a rush identified as Cymbopogon schoenanthus L. This word was used to
translate the enigmatic , which is obviously also a composite word. The

is the Arabic ZA g stipa, feather grass (Stipa tenacissima L.). The
SERIKOFF
107
(pronounces in colloquial Greek as [demek ]) is a garbled Arabic transliteration of the

Greek shrub: @P @X @X. It could be also the Ottoman
X
in a sense called
14
The word as being applied to a plant could not be found. The Arabic
Q@ (Ghaleb Encyclopedie no. 9505) or
original suggests either translation of k@

KQA K (Dtr. IV 178 n. 7), both of which mean polypody (Polypodium
vulgare L.). However, this is merely a suggestion. Cf. here a similarly sounding
word , which is explained by B. Langkavel as ferula opopanax (Langkavel,
Botanik 129,34) and no. 15.
15
This glossa is also recorded in the Greek glossaries of the Arabic materia medica:
(Lexikon ton Sarakinon no. 355). The Greek (cf. here
the variant readings cited at Langkavel, Botanik 129,34) is a transliteration of the
. or Q
m
. '@ Qm., the opopanax (Dtr. III 47).
Arabic Q
Ag
16
This glossa is also recorded in the Greek glossaries of the Arabic materia medica:
, (Lexikon ton Sarakinon no. 48). This allows us to
suggest for an enigmatic a conjecture (confirmed by Langkavel, Botanik 11,1):

. This is Q P XCK. Sempecarpus anacardium L. (Ghaleb, Encyclopedie no.
3885). The word in this meaning, however, did not originally exist in the
Greek language. It had been invented by the Arabs and came to the Greek language
from Arabic, cf. Serikoff, R
um and Y
un
an 183, notes 46, 47 and other occurencies
(Langkavel, Botanik 11,1, 179,1).
17
The is not the plant Ferula terulago L. (Langkavel, Botanik 129,34)

but the resin of the tree Dorema ammoniacum Don. Its Arabic equivalent is i or
q (Dtr. III 72, 79), of which is a true transliteration. The second part of the
lemma corresponds to the first.
18
The Greek is a transliteration of the Arabic K. AJ., a kind of pepper Piper

cuebeba L. (Dtr. III 44 n. 5) as it comes from the explanation by E. Ghaleb: K. AJ.

J. @ K. A J. (Ghaleb, Encyclopedie no. 23150). The mysterious Greek
therefore should be read either as , which invites a conjecture:
108
[ ] or even simpler: and translates: koupepe this is

kapapa.

K. @ white
The Greek is a direct translation of the Arabic J
pepper (cf. Ghaleb, Encyclopedie no. 21009). This type of composite noun is also
attested elsewhere, e.g. (Langkavel, Botanik 48,2).
19
20
The is the long pepper, mentioned by Dioscurides (Dtr.

III 134, cf. Langkavel, Botanik 144,6). The is the transliteration of the

with the same meaning (cf. ibid.).
P@Y@
21
This lemma is recorded in the Greek glossaries of the Arabic materia medica, e.g.
Lexikon ton Sarakinon no. 1: (cf. Serikov, Saracinskii leksikon
94). This is aloe, Aloexylon agalochum Lour. (cf. Serikov, Saracinskii leksikon n. 56,

Langkavel, Botanik 29.5), which, however, has been incorrectly translated as Y J
, namely the white Sandalwood (Ghaleb, Encyclopedie no. 16543). Cf.
(no. 8).
22
The is explained as , which in Arabic is XP@ XAK.. This

word generally means a thistle, a closer identification is more difficult (Dtr. III 14).
The enigmatic Greek is obviously an explanation based upon the
root XP rose and the traditional Arabic explanation: XP@ hP (Dtr. III 14 note),
which can be interpreted as a sweet flower and a flower.
23
The common knot grass, Polygonum aviculare L., is here followed by

beginning a description of its properties. The Greek , probably a
invites a conjecture: . AJ
@, i.e. dry a property as applied to a remedy.
24
The , mint (Mentha L.) is explained via tourpeth, a resin of the convolvulaceous plant (Dtr. IV 121). Normally this explanation is applied to the
or , Globularia alypum, cf. no. 29.
25
The is a transliteration of the Arabic
Qk wild rue (Langkavel, Botanik

16,2). It was traditionally used to explain the word hemlock: J
K
Q @ @

YJ.@ Qm '@ K
(Diosc. Erkl. IV 71).

26
. nettle
The Greek is a corrupted transliteration of the Arabic Pm.' B@ P QK
SERIKOFF
109
seed, which appears in the form as it appeared in other Greek glossaries of the Arabic materia medica: ,(Lexikon ton
Sarakinon no. 267, cf. Serikov, Saracinskii leksikon 97 n. 56).
27
Sagapenum () is a resin from the tree ferula Scowitziana DC, Var.,

J.. The Arabic iJ
J.@ is likewise a
which is a transliteration from the Persian J
.
transliteration of the same Persian word. The is obviously a transliteration
from the Arabic, cf. also a lemma: (Lexikon ton Sarakinon no.
159).
28
According to B. Langkavel (Langkavel, Botanik 218,2) the means cardamon (cf. nos 37, 59); he also lists the words and , which can be
considered as synonyms to the , a counterpart for the . The exact
etymology of both and remains to me unknown. Cf. here no. 49.
29
The Greek is a later form of the , Globularia alypon (Dtr. IV 159).

It is traditionally explained as tourpeth. The lemma also appeared in other Greek
glossaries of the Arabic materia medica: (Lexikon ton Sarakinon
no. 357), cf. no. 24.
30
A plant, called (lycium is still awaiting a narrower identification (Diosc.

k m, namely collyrium of the
Erkl. I 102). Traditionally it was explained as B
tribe Khawl
an (Diosc. Erkl. I 102). This explanation is reflected in the transliterated
word . I failed to find the explanation of the as Medicago sativa
L.
K , possibly a rye grass (Dtr. IV 38),

J

which has been identified as a genus as , Arabic thorn. However a narrower
31
The looks like a corrupted
identification remains a desideratum.

32
The Greek explanation: looks like a transcription of the collo

quial (Syriac) Arabic K. AJ. . H
. KQ k , i.e. [this is] the carob tree, not the pepper. However, one cannot say this for sure because the exact meaning of the word
still remains obscure. Grammatically, however, the form is
J ) is recorded
. J
possible, because the similar construction ( KCK
110
elsewhere (Diosc. [Graec.] I 103,19 (I no. 109) Diosc. [Arab.] 105,20 I no. 116), cf.
(Langkavel, Botanik 151.9).
33
Both words and are rather difficult to interpret. Similar lemmas,
however, do occur as no. 60 below: : and also
(Lexikon ton Sarakinon no. 43); (Lexikon ton
Sarakinon no. 63); (Lexikon ton Sarakinon no. 84);
(Lexikon ton Sarakinon no. 260, Langkavel, Botanik 236,1).
From these lemmas it becomes obvious that the Greek (the name for the
water-plant Alisma plantago aquatica) as well as and are synonyms.
The similarly written , , is obviously a transcription of the
@ X@ , the synonym of the Arabic name for the Alisma plantago aquatica, PAQ
Q @
@Q@ (Ghaleb, Encyclopedie no. 26494). The enigmatic , , etc., however,
still remains unexplained.
34
Both words and mean Acorus calamus L., a perennial marsh-plant,

in Arabic: h. (Dtr. I 2). The lemma is found in other glossaries of the Arabic
materia medica: (Lexikon ton Sarakinon no. 52);
(Lexikon ton Sarakinon no. 247).
35
The word is the Greek transliteration of the Arabic translation ( ) of

the (Piper nigrum L.), black pepper (Dtr. II 143).
36

AgB@ X, the
The word [] is a direct translation of the Arabic K
dragons blood, a resin of a certain art of the Lileaceae (Dtr. IV 79 n. 3). The
X .

AgB@
is the Greek transliteration of the Arabic K
37
The (or cf. no. 38 ) is the Greek transliteration of the Arabic

A , Elettaria cardamum maior, Wilder (Dtr. III 70 n. 6), which was obviously

written as A. The [] is a translation of the (Q@). This is the white
cardamon. B. Langkavel (Langkavel, Botanik 29,14) identified it as Emblica officinalis
Gaertn. Cf. no. 49.
38
The comprises two words,

black cardamon.
A and , cf. no. 37 and means
SERIKOFF
111
39
The Arabic equivalent for the myrobalan, written as remains unidentified,

however, following B. Langkavel (Langkavel, Botanik 29,14) one might suggest that
might be a garbled form of .

The [] is the direct translation of the A

P @ , the sparrows tongue, the fruits of the ashwood, Fraxinus excelsior L.,
40
(Diosc. Erkl. III 125, Dtr. III 122 n. 5). B. Langkavel (Langkavel, Botanik 189,2)
identified it as Polygonum aviculare.
41
The Greek is a name applied to a number of plants (Diosc. Erkl. I 57).

Traditionally it was translated by the Arabic , here hellenized as . The
lemma is found in other glossaries of the Arabic materia medica:
(Lexikon ton Sarakinon no. 201), cf. Langkavel, Botanik 230.1.
42
The word , i.e. galagal, Alpina officinarum explains the same word,
. J
k and pronounced in colloquial (Syriac?) Arabic
written down in Arabic as Aj
or even Turkish. A similar lemma found in another lexicon:
(Lexikon ton Sarakinon no. 410) just confirms this suggestion. The Greek
k . The word also looks like a transcription of the
. J
is no doubt Aj
Aj.J
k , cf. Langkavel,
Botanik 218,1: . The word is pure Greek
and means of no fruits.
43
The word [] is a direct translation of the ,

aloe-tree (Dtr. III 23).
Q an
Q@
.
44
This lemma provides a Greek equivalent for the Arabic , Q , murrh,

Commifora abyssinica (Dtr. III 27).
45
Maker () is the muscat flower. It was normally translated into Arabic as

AJ
. . (Dtr. I 50, Diosc. Erkl. I 83), which is also reflected in this lemma. Similar
lemmata occur in other Arabic glossaries: (Lexikon ton Sarakinon

no. 61); (Lexikon ton Sarakinon no. 64); (Lexikon
ton Sarakinon no. 202); (Lexikon ton Sarakinon no. 272). Cf. Serikov,
Saracinskii leksikon 96.
46

The Arabic Q transliterated into Greek as is the cortex of cinnamon

Cortex cinnamoni Cassiae (Dtr. I 45 n. 6).
112
47
For the cf. the lemma: (Lexikon

ton Sarakinon no. 175). That is probably why our text should also read , cf. (Langkavel, Botanik 58,2). The is the Carophyllus
aromaticus L., a carnation. It allows us to see in the enigmatic the big
carnation, or the hyacinth. In fact the pronounced in colloquial Greek
as [dzuchindi ] is a transcription of the Italian giacinto.
. @ I k the balsam
The word is the Greek transliteration of AJ
.
seed (probably that of the Commiphora opobalsamum [L.] ). The Greek translation
. k [h.abb] seed has been heard as
was made from the oral source, where the Arabic I
- from fruit. The word is recorded: Langkavel, Botanik
13,1 and means Amyris.
48
49
The is the granum paradisi, the fifteenth century Italian name

for the Malagueta pepper (Capsicum frutescens var. malagueta, Solanaceae). The

explanation looks like a garbled Greek transliteration of the Arabic
* K * X.
However, the actual word, which has resulted in the enigmatic might

well be the cardamon (in Arabic A) (cf. nos 28, 37, 49): A *
K X as it comes from the reference by B. Langkavel (Langkavel, Botanik

218.5).
50
The , common polypody (Poloypodium vulgare L., Var., was traditionally translated into Arabic with its Persian counterpart l'
.
AJ.., which literally means
of many feet (Dtr. IV 178). The Arabic equivalent written in Greek ( ) is
obviously a garbled transliteration of l'
.
AJ..; cf. also Serikov, Saracinskii leksikon 96.
51
All words are variants of the Arabic word P A, which is Camphora officinalis
(Dtr. III 24 note 6). The narrower indications of the camphor mentioned here
, and are to be found at Langkavel,
Botanik 187,2. B. Langkavel identified it as Persea Camfora and listed other forms:
, , and especially . The words
SERIKOFF
113
and might be a garbled form of . About the graphical

errors of and see no. 1.
52
The is the cassier or needle bush (Vachellia farnesiana or Acacia

farnesiana), in Arabic Q.J PAJ
k (Ghaleb, Encyclopedie no. 9610), as it also becomes
evident from its Greek transcription. This plant has been known in Europe since the
sixteenth century.
53
Apart from the present word-list the lemma is mentioned in another

late Greek source: (Lexikon ton Sarakinon no. 252). The
is , a sweet myrrh (Opopanax Chironium L. or the resin
produced from its root (Dtr. III 47). However, the Greek explanation is nothing else
Q, the Indian date, the tamarind (Tamarindus Indica). Oxyphoenicum
but Y J
explained as Thamarindi is to be found in the botanical directory by the German Enlightenment writer Adam Lonitzer (1528-1586): Oxyphoenix. Lat: Palma Sylvestris.
Wilder Dactelbaum Fructus ejus Tamarindi Oxyphoenica seu Tamarindi Actuario
teste humectant et refrigerant in abscessu secundo. (Lonitzer, Naturalis Historiae
opus novum Glossarium s.v.) B. Langkavel identified as Ferula opoponax
L. (Langkavel, Botanik 129,34).
. The
The is the carnation Dianthus caryophyllus in Arabic KQ
second part of the lemma, is the direct transliteration of the Arabic
form (Dtr. III 44 n. 7). Cf. similar lemma: (Lexikon ton
Sarakinon no. 146).
54
55
The is , ginger, Zingiber officinalis, Roscoe, which in Arabic is

. The second part of the lemma, is the direct
written in a form J
J. m.' Q@
transliteration of the Arabic form (Dtr. II 144). Similar lemmata occur in other
texts: Lexikon ton Sarakinon no. 112.
56
The Greek word indicates a kind of chrysanthemum Anacyclus pyrethrum

L.. Arab botanists translated it with a Berber word AgQ Q A (Dtr. III 69), which
garbled version (*Q QKA ) is transliterated here as , cf. the form
114
(Langkavel, Botanik 174,19). The might be a wrongly understood Berber

@

KQ
J A K as it comes in the Arabic Dioscurides: AgQQ A@
name IY
A Q

. Q @ (Diosc. [Arab.] III 70).
JAK H. Q @ @ K
IY
57
The first word is illegible. The Greek looks like H

. Cg., ipomoea (genus
Convolvulaceae), however, this still cannot be said for certain.
58
The enigmatic is a stem-for-stem translation of the , i.e.

and , P = laurus (Dtr. I 45), the low laurel. The then

must be PB@
written as PB@
). The

PB@
QK. J is
(without a dot ober
reconstruction is legitimate, since similar composite

recorded elsewhere (Dtr. III 150).
59
The Greek means cardamon Du Cange, Glossarium Graecitatis s.v.,

Langkavel, Botanik 218,5, cf. no. 37. A similar lemma is found in another source: (Lexikon ton Sarakinon no. 135). B. Langkavel (Langkavel,
Botanik 218,2) indicated that for the word its synonym was . Cf. no.
28.
60
Cf. no. 33.
Q
A (Dtr. III 162),
The , curcuma is translated into Arabic as @
which in this case is transcribed into Greek as , cf. Langkavel, Botanik 110,3.
A similar lemma is found elsewhere: (Lexikon ton Sarakinon no.
387).
61
62
The [] is allegedly a species of , also called by the Arabs

as the dogs tongue (Dtr. IV 90). The in colloquial Greek pronounced as
[endikal ] (= indicon) is obviously a garbled Greek transliteration from the Arabic
K@ (), which as an example has been discovered by A. Dietrich in az-Zahraws
Y
work (Dtr. I 9 n. 7) in the form as follows:
63
JK@ .
The [] is a rare mineral composed of ammonium chloride, it forms

colorless to white to yellow-brown crystals. Its modern translation in Arabic is i
PXA @ . The interpretation given here and equally that found in
SERIKOFF
115
another source: (Lexikon ton Sarakinon no.

198) are still in need of interpretation.
64
(Dtr.
Arabic translation of the thyme (probably Saturea calamintha L.) is A Ag
III 36), which entirely corresponds to the as found in the lemma.
65
This is , a kind of earth used to seal. The transliterated
indicates a different kind of earth, namely the earth of Samos. In our case it should
be , the correct form is or AA (IB III 109,23). The Arabic
.
substrate in our case should possibly look like kA
Wellcome Library, 183 Euston Road
London NW1 2BE, UK; Inst. of Orient. Stud., Russ. Acad. Moscow
Nikolai Serikoff
[email protected]
116
Word list
19 19
The following pages contain a word

list of the Greek and Arabic vocabulary found in the dictionary. The
numbers refer to the notes and to the
main text as well. The sign indicates that the commented word occurs only in the footnotes.
42
19
19
37 , 10 10, 43
43
51 51
41 41
28 28
36 [] 36
26
42 42
[] 3 [] 3
3
[] 7 [] 7
1
40 40
4
50 50
41
41
16
34
4 4
[] 63
34
[] 43
3 3
33
29
3 3
29
42
17
42
[] 63 [] 63
16
41 41
13 13
65 65
22
[] 40
SERIKOFF
117
22
59
22
59
49
[] 7
24
60
22 22
63 63
18
56
38
[] 36
48
60 60
54
60
7 7
62
52
7 7
63 63
11
23
51
33
51
33
51
[] 9 [] 9
51
42 42
51
55
51
55
51
55
51
11
59
42
59
64
[] 28
[] 28 [] 28, 62
[] 62
17
59
[] 37
38
118
27
41
46
44
39
16
10 [] 10
36
35 35
39
49
25
15
42
47
18
21, 4
54
46
8 [] 8
65
65
53
30
15
40
34
30
56
45
56
20
53
47 47
49 49
52 52
22
63
26
61
26
61
33
63
35
47
23
41
50
SERIKOFF
119
56
[] 40
27
27
65
65 65
52 52
65 65
21
13
53
10
65
[] 14
[] 9
20
[] 10
29
[] 8
14
43
55
28
58
28
24
28
32
32
44
46 46
31
35 35
31
35
12
19
26 26
32
45
49 49
14
[] 62
12
31
50
13
62
40
57
120
58
5
7
48
63 63
25
32
[] 3
30
61
52
47
47 , 53 53
64
57
SERIKOFF
Figure 1: Wellcome Library London, MSL 60, f. 71v.
121
The Reception of Galens Art of medicine in the Syriac Book of

medicines
I. Introduction
It is a commonplace that a signicant amount of Greek medical literature from Late Antiquity was translated into Arabic through an intervening
Syriac version.1 In view of this, it is surprising how little research of value has
been done on Syriac medical literature, something that will become apparent
through the course of this paper.2 The value of the Syriac medical corpus,
however, goes beyond its role as the intervening stage between the Greek and
Arabic versions of ancient texts it also aords us a rare glimpse into the
mechanics of reception, that is, how the traditional Mesopotamian medical
system absorbed the new science that came from the West, rst accompanying Christianity as it spread eastwards,3 and, subsequently, with the ocial
sanction of the Abbasid caliphate in Baghdad.4
1
E.g. M.W. Dols, Syriac into Arabic: The Transmission of Greek Medicine,
ARAM I (1989) 45-52; R. Le Coz, Les mdecins nestoriens au Moyen ge: Les
matres des Arabes, Paris 2004.
2
For a previous discussion of this neglect, see S. Bhayro, Syriac Medical Terminology: Sergius and Galens Pharmacopia, Aramaic Studies III (2005) 147-165:
149-152.
3
For a brief summary of the historical circumstances of the eastward spread
of Greco-Roman medicine with Christianity, see P.E. Pormann-E. Savage-Smith, Medieval Islamic Medicine, Edinburgh 2007, 17f.
4
Again, for a brief summary of the Abbasid-sponsored translation movement,
see Pormann-Savage-Smith, o.c. 24-29. For a more detailed and very useful treatment,
which is a little dated in respect of the Syriac sources, see H.D. Isaacs, Arabic Medical
Literature, in M.J.L. Young-J.D. Latham-R.B. Serjeant (edd.), Religion, Learning and
Science in the Abbasid Period, Cambridge 1990, 342-363.
124
Siam Bhayro
The Syriac Book of medicines is perhaps the best demonstration of the

mechanics of reception. In this paper, we shall consider an indicative example:
the reception of the section of Galens Art of medicine that discusses the
physical properties of the head and the corresponding qualities of the brain
(VI 2-12), juxtaposed with traditional herbal remedies for headaches that
probably originate from Mesopotamia. We shall also see that a recent scholarly
treatment of the reception of Art of medicine VI 2-12 in the Syriac Book of
medicines is deeply awed. But rst, some words about Syriac and the Syriac
translations are in order.
II. Syriac
Manfred Ullman wrote that, after the Near East had been more
and more christianized, Koine-Greek lost in signicance in this area as lingua franca, while the native languages, Aramaic in Syria and Iraq, Coptic in
Egypt, and Pahlavi in Persia, ourished again.5 Syriac is perhaps the most
prestigious of the Aramaic dialects that ourished at this time, both in terms
of the scope of its literature and the extent of its inuence. It began as the
dialect of Edessa and its environs, around the area of Urfa in what is today
south-eastern Turkey, with the earliest dated inscription coming from the year
6 CE, but it quickly spread eastwards as the language of eastern Christianity.6 The following three centuries witnessed the translation of the Bible into
Syriac, along with some early Christian writings. From the fourth century
CE, writers such as Aphrahat and Ephrem ushered in a golden age of Syriac
literature that ended only with the spread of Islam in the middle of the seventh century. From this point, Syriac gradually ceased to be the Christian
vernacular, being replaced by Arabic, although it continued to be used as a
liturgical language and a language of learning akin to Latin in early-modern
Europe.7
In this paper, however, we are more concerned with Syriac translation
5
M. Ullmann, Islamic Medicine, Edinburgh 1978, 7.

S. Brock, A Brief Outline of Syriac Literature, Kottayam 1997, 7. For an introduction to the various Aramaic dialects of the time, see S. Moscati, An Introduction
to the Comparative Grammar of the Semitic Languages: Phonology and Morphology,
Wiesbaden 1964, 11f. For the earliest dated inscription, see J. F. Healey, Aramaic
Inscriptions & Documents of the Roman Period, Oxford 2009, 223-225.
7
For more detail, see Brock, o.c. 8-87.
6
The Reception of Galen's Art of medicine
125
activity, especially the translation of scientic works, rather than the Syriac
language or its literature in general. In order to appreciate the proper context
of the Syriac Book of medicines, it is imperative that we understand the distinct
phases of Syriac translation activity, of which, broadly speaking, there are
ve:8
Phase One up to the third century CE: This period witnessed
the rst translations of Biblical and early Christian literature from
Hebrew and Greek into Syriac. While the subject matter was not
scientic, the experience gained, particularly in translating from
Greek into Syriac not a simple task, given how dierent the
two languages are would prove to be invaluable.
Phase Two sixth century CE: This was the century of the rst
Syriac scientic translation movement, in which whole corpora
were systematically translated from Greek, e.g. Sergius of Resh
,Ainas translations of Galens works. Although disparaged by
later writers, these translations were certainly reader orientated.9
Phase Three seventh century CE: This period is characterised
by more rigid, less reader orientated translations, as well as revisions of older translations. There seems to be nothing of interest
to medical historians pertaining to this phase.
Phase Four late eighth and ninth centuries CE: Under Abbasid
patronage, another stage of systematic translation activity took
place, this time from Greek into both Syriac and Arabic.10
Phase Five tenth to thirteenth centuries CE: This is the age
of the great scientic encyclopaedic compilations, in which the
priority was bringing order to the vast corpora and arranging
them in a practical way.11
8
More or less following Brock, o.c. 120-123.

See Bhayro, Syriac Medical Terminology cit. 152-157.
10
For a discussion of the relationship between the earlier, sixth-century, and
later, ninth-century, translation movements, see S. Bhayro-S.P. Brock, The Syriac
Galen Palimpsest and the Role of Syriac in the Transmission of Greek Medicine in the
Orient, BRL, themed issue: R. David (ed.), Ancient Medical and Healing Systems
in the Near East, Greece and Rome, forthcoming.
11
This mirrors the trend in Arabic medical literature see Isaacs, o.c. 347f.
9
126
Siam Bhayro
III. Budges Acquisition of the Syriac Book of medicines
In 1894, Ernest Wallis Budge, on one of his many trips to Mesopotamia

to secure antiquities for the British Museum, found himself in Alqosh, a village
with a rich Christian heritage located approximately thirty kilometres north
of Mosul, in what is now the predominately Kurdish area of northern Iraq.
He encountered a learned resident of Mosul who was in possession of a small
library of manuscripts. One manuscript that caught his eye was a twelfthcentury medical compendium. From its size and binding, Budge concluded
that it had once been part of a monastic library. Despite missing several quires,
enough remained for it to cause great excitement to Budge, who referred
to it as the most important and most perfect Syrian treatise on Anatomy,
Pathology, and Therapeutics known to us.12 If we are to believe Budges
account, the owner guarded his manuscripts jealously. Budge was, therefore,
forced to pay a professional scribe, at his own expense no less, to make a
copy of this manuscript. Thus, regarding the value of this nineteenth-century
copy as a textual witness to the Syriac medical tradition, we are left at the
mercy of this scribes competence and the necessarily hurried nature of the
transaction.13
As a twelfth-century medical compendium, therefore, the Syriac Book
of medicines belongs to the nal phase described above, which means two
things that are relevant for our present purposes. First, the Syriac Book of
medicines is not a translation, but an encyclopaedia that contains excerpts of
much earlier translations, thus allowing ample opportunity for various kinds
of textual corruptions to creep in, either in the transmission of the earlier
translations or in the process of editing, and often abridging, them into the
compendium. Second, its content is drawn from a variety of sources that have
been edited together for the sake of convenience rather than to preserve the
integrity of the various source texts.
Budge described the compendium as consisting of three distinct works.
The rst was the most scientic, containing transcribed medical lectures,
each followed by relevant prescriptions. The second was astrological, and
concerned with omens and spells. The third contained folk prescriptions of
12
E.A.W. Budge, The Syriac Book of Medicines: Syrian Anatomy, Pathology

and Therapeutics in the Early Middle Ages, I-II, London 1913, I xiiif.
13
For Budges description of how he acquired the manuscript, see Budge, o.c.
I xxxvii-xli.
127
the most extraordinary nature. For Budge, each section was distinct, with the
rst having been penned by the most educated scribe, while the second and
third by other, more superstitious and ignorant, scribes.14
Regarding the rst, more scientic, section, Budge wrote: These Lectures were translated from Greek into Syriac by a Syrian physician, who was
probably a Nestorian, and who was well acquainted with Greek and Syriac;
and he may well have been attached to one of the great Medical Schools, which
existed at Edessa (Urfa) and mid (Diarbekr), and Nisibis, in the early centuries of the Christian era. The style of the Syriac is uent and good, and
exhibits everywhere the touch of a master hand.15 Budge is correct in that
his manuscript does indeed contain much Greek science in Syriac translation.
Furthermore, it is indeed likely to be a Nestorian scholarly text. But the way
in which the Greek science has been received within the text, with its careful
ordering of earlier known medical material in abridged form, coupled with the
wealth of non-Greco-Roman medical lore, suggests that this is not a translation of a Greek medical work or series of lectures into Syriac. Rather, it is
a compendium based on a combination of Greco-Roman and Mesopotamian
sources. Furthermore, Budges attempt to divorce the latter two sections from
the rst one was seriously misguided.16
IV. Greek science in the Syriac Book of medicines
That much Greco-Roman medical lore is contained in this text is beyond

doubt. Indeed, between 1926 and 1946, Schleifer produced a series of ten
articles in which he analysed the Galenic passages contained in the Syriac Book
of medicines, with reference to Khns edition of Galens works.17 Schleifers
analysis goes as far as the end of the rst section (ending on page 440 of
Budges edition), demonstrating that a signicant portion of the Syriac text
of the rst, more scientic, section is derived from Galen, although often
in an abridged form. The theoretical basis of the system of medicine in the
14
Budge, o.c. I v-xi.

Budge, o.c. I v.
16
For a recent critique of Budge, which is useful despite probably going a little
too far, see A.H. Becker, Doctoring the Past in the Present: E.A. Wallis Budge, the
Discourse on Magic, and the Colonization of Iraq, HR XLIV (2005) 175-215.
17
See the Appendix below.
15
128
Siam Bhayro
rst section of this manuscript is, therefore, clearly Galenic. But how is this
Galenic tradition received?
Let us consider one example that relates to the anatomy and physiology
of the head and brain:18
[A]
[B]
[C]
[D]
18
The Syriac text is taken from Budge, o.c. I 36, with the following changes:
[E] line 2, for , following J. Schleifer, Zum Syrischen Medizinbuch,

Zeitschrift fr Semitistik und verwandte Gebiete IV (1926) 70-122: 108; [F] line 4,
for
. The English translation is my own, although that of
Budge, o.c. II 33f., has been taken into account.
129
[E]

:
:

:
.
.
:
[F]
.

.
.

:

[A] Concerning the natural framing of the head and concerning the signs of
the natural constitutions of the brain.
[B] It is right, therefore, that before everything, for we shall write concerning
the cure of headaches, that we should nevertheless speak a little concerning
its natural constitutions, for this also is of necessity required of us.
[C] A small head, therefore, is a sign of the bad construction of the brain. But
neither is that which is large by any means excellent. For if it is that by the
rmness of the power of the inner part that it attains for itself much matter
(and) the head becomes thus (i.e. large), it is a sign of evil.
[D] But it is right to distinguish all the (dierent) kinds of heads.
[E] If they are constituted (properly), they signify virtue, and if, moreover,
they are spherical and their veins are rm, and their openings are thin, and
their eyes are set rmly and piercing and keen, they indicate virtue. But those
that slope towards their front, and their veins are weak, and those, moreover,
that are longer than normal measure, they are, for the most part, a sign of
evil. Therefore, just as concerning large heads, so also these we interpret
concerning the brain;19 but sometimes, nevertheless, of these a very few are
19
I.e., just as we interpreted the external signs regarding the size of the head
as being indicative of the constitution of the brain, so we also interpret these distinguishing features just mentioned as indicative of the constitution of the brain.
130
Siam Bhayro
excellent, as (with) the one who the formative power of the inner part was
strong.
[F] The temperate brain, therefore, its mental impulse is moderate, and the
abundance of its motion is not great, and it is not easily harmed by an external accident. For those who are such as these, in their infancy their hair is
reddish, and in their youth it is redder, and in their manhood even more so,
while it is between curly and straight, and they do not bald easily.
As we shall see, this short portion of text contains several distinct components (title, editorial insertions, translations, abridgements) and demonstrates several interesting features (textual corruption, enforced changes, editorial choice and the age-old tension between theory and practice).
In terms of its components, part [A] is the title of this section, and
was the invention, presumably, of the scribe who compiled the overall section
concerning headaches. Also belonging to the hand of this editor is part [B], in
which the reader is warned that, before proceeding to the instructions for how
to cure headaches (the part the reader is most likely to need), some theory
should be learned. This tension between theory and practice is nothing new
indeed, it is a constant feature in the transmission of Galen. According to
Vivian Nutton, this was already the case in Galens lifetime, due, in part, to
the intimidating size and complexity of the Galenic corpus.20 Furthermore,
Andrew Cunningham discusses how this continued in later Alexandrian, Arab
and medieval European contexts.21 In the earliest stage of the transmission of
the Galenic corpus from Greek into Syriac, in the sixth century, this tension
is very clear. For example, in Sergius of Resh ,Ainas introduction to his
translation of the sixth book of Galens Simples, Sergius makes a heartfelt
plea to his student Theodore not to neglect to master the theoretical basis
the rules that pertain

of the art of medicine
in general to the art before considering its practical remedies
20
V. Nutton, From Galen to Alexander, Aspects of Medicine and Medical Practice in Late Antiquity, DOP XXXVIII (1984) 1-14 [reprinted in V. Nutton, From
Democedes to Harvey: Studies in the History of Medicine, London 1988, X, 1-14]: 3f.
21
A. Cunningham, The Theory/Practice Division of Medicine: Two LateAlexandrian Legacies, in T. Ogawa (ed.), History of Traditional Medicine. Proceedings
of the 1st and 2nd International Symposia on the Comparative History of Medicine
- East and West. 1st Symposium: October 22-28, 1976, 2nd Symposium: October
23-29, 1977, Susuno-shi, Shizuoka, Japan, Tokyo 1986, 303-324.
131
the instruction that is in detail.22 The compiler of the

Syriac Book of medicines thus maintains this agenda, stating, in part [B], that
a sound theoretical basis is a necessity.
In order to establish this theoretical basis, the editor inserts, in part [C],
a quotation from an earlier Syriac translation of Galens Art of medicine, VI 2.
The passage from which this translation was drawn probably read something
like this:23

, ,
,
A small head is the specic sign of a poor state of the brain. A
large head, however, is not necessarily a sign of a good state; but
if it has become so through the strength of its residual power,
producing matter both useful and abundant, it is a good sign;
but if solely by a plethora of matter, it is not good.
Comparing the Greek and Syriac versions, it is clear that something is
amiss:24
22
This passage is found in the British Library manuscript BL Add 14,661

1v:1-2v:13, previously edited with a German translation in A. Merx, Proben der
syrischen Uebersetzung von Galenus Schrift ber die einfachen Heilmittel, Zeitschrift
der Deutschen Morgenlndischen Gesellschaft XXXIX (1885) 237-305: 244-248. See
also the section on Theory and Practice in Bhayro-Brock, o.c., in which this passage
is re-edited with an English translation.
23
For the text with a French translation, see Galien, II. Exhortation la
Mdecine, Art mdical, ed. V. Boudon, Paris 2000, 288. The English translation is
based on Galen. Selected Works, transl. by P.N. Singer, Oxford 1997, 352.
24
In the following table, the Greek and Syriac versions are given with their
respective English translations. In the English translation of the Greek, the text in
bold shows what is not represented in the Syriac version. In the columns containing
the Syriac text and its English translation, the text in [bold] shows what should be
present.
For if it is that by the

rmness of the power of
the inner part
A large head, however, is

not necessarily a sign of
a good state;
but if it has become so
through the strength of
its residual power,
producing matter both
useful and abundant,
it is a good sign;
but if solely by a
plethora of matter,
it is not good.
[it is a sign] of evil.
[but if it is an
abundance of matter
alone,]
it is a sign [of virtue]
that it attains for itself

much matter (and) the
head becomes thus (i.e.
large),
A small head, therefore,

is a sign of the bad
construction of the
brain.
But neither is that
which is large by any
means excellent.
A small head is the

specic sign of a poor
state of the brain.
[.
]
]

[
[
]

133
It is clear that, at some stage of copying, a phrase has been omitted

by homoioteleuton, in this case due to the probable repetition of
it is a sign of in the original Syriac translation. This was already

suggested almost one hundred years ago by Schleifer.25 In view of this, the
treatment of the Syriac evidence in a recent critical edition of Galens Art of
medicine represents a regression in scholarship, something that is indicative of
the neglect of the Syriac sources mentioned in the introduction to this paper.
The treatment in question reads as follows:26
Dans certains cas, on constate que le texte syriaque est en dsaccord la fois avec le texte des manuscrits grecs et avec la traduction arabe de Hunain. Cest le cas du premier passage (c. 6,2) o
il est question de la forme de la tte et du pronostic que lon peut
en faire. Le grec comme larabe distinguent deux cas : si la tte
est petite, cest mauvais signe ; mais si elle est grande, ce nest
pas forcment bon signe. Sur ce point, la traduction syriaque est
en accord avec lensemble de la tradition. Mais les divergences
sobservent, lors de la distinction tablie par Galien partir de ce
dernier cas []. La traduction syriaque ignore cette prcision et
conclut loppos []. Cest l aboutir un non-sens.
While it is true that the Syriac text, as it now stands, is nonsensical,
the point is that, as we have seen, it is not the case that the Syriac text is
ignoring the precision of the Greek version and concluding the opposite. The
Syriac text has simply omitted a crucial clause in the process of transmission
for a very obvious reason. While this may seem like a subtle distinction of
little importance, the full signicance of this point will become apparent when
we consider the nature of the abridgements in parts [E] and [F].
Having inserted this except from Galens Art of medicine (part [C]), our
scribe moves to summarise the details. But rst, he is careful to distinguish
between the direct quotation (part [C]) and the following abridgement (part
[E]) by inserting an editorial notice part [D].
What follows, in part [E], is not a direct translation nor even a simple
abridgement of the following portion of the Art of medicine. Instead, we
have an abridgement in which the core theme has been changed from mental
25
26
Schleifer, o.c. 107f.

From Boudon, o.c. 232f.
134
Siam Bhayro
prowess to morality. Thus, while Galens Art of medicine VI 3-10 is concerned

with the head as an indicator of the brains intelligence, mental aptitude and
powers of memory, the corresponding section of the Syriac Book of medicines
is concerned with the head as an indicator of the mind as the seat of morality
and virtue. Thus, for example, the Syriac text has
virtue and
evil, as opposed to the quick-wittedness and

slowness of thought found in Galens Art of medicine VI 10.
Rather than referring to this simply as an abridgement, I would suggest
adopting the phrase thematic abridgement. It is not just that the scribe has
distilled Galens text he has also evidently deemed it to be of little value
and very much in need of updating according to the beliefs of his readers.
Indeed, we can read part [D] as the scribe drawing the readers attention to
the updated agenda. Thus part [E] reects more the changing context in
which the text was transmitted, rather than the abilities of the translators
and editors.27
Having rewritten Galens Art of medicine VI 3-10, the scribe arrives at
VI 11f., which ts in well enough with his new agenda to require less drastic
modication (part [F]):28
27
In addition to this thematic abridgement, it is possible that the scribe has

incorporated a quotation from another Galenic work. Thus
could correspond to
from Galens On Hippocrates Epidemics. Once again, this was
already suggested by Schleifer, o.c. 108. If this is the case, this insertion may have
been provoked by the context, which is describing facial features, although further
investigation may bring another motivation to light.
28
For the Greek text with a French translation, see Boudon, o.c. 291f. The
English translation is based on Singer, o.c. 353.
But it seems to me that two kinds of

indication yet remain of those that I
initially undertook to discuss one
concerning the natural activities and the
other concerning external events. There
is a common explanation concerning both
of them:
If the brain is temperate in respect of the
four qualities,
it will be moderate concerning all of
which we spoke,
and concerning the secretions, those
cleared out through the palate, the ears,
(and) the nostrils, it will also be
moderate,
and hardly damaged by all of the
external events, which are heat, and cold,
and dry, and wet.
The hair of such people is reddish as
babies,
and with children blondish,
and reaching maturity it becomes blonde
they are, I suppose, in between truly
curly and straight,
and neither do they easily become bald.

,
, ,

,
,
,

, ,
, ,
,
,
and they do not bald easily.
while it is between curly and straight,
and in their manhood even more so,
and in their youth it is redder,
For those who are such as these, in their

infancy their hair is reddish,
and it is not easily harmed by an

external accident.
and the abundance of its motion is not

great,
its mental impulse is moderate,
The temperate brain, therefore,
136
Siam Bhayro
Again, we can see that the text is certainly abridged, particularly at

the start, but the immediate theme remains unchanged, albeit it is now in
a dierent context on account of the thematic changes made in part [E]. In
abridging the text, the scribe removes all of Galens references to himself in
the rst person. Also deemed surplus to requirements is the initial section,
which has been made irrelevant by the way the preceding text has been edited.
One interesting dierence between the versions relates to the colour of
the hair, which involves a change from red to blonde in the Greek, but an
intensifying of red in the Syriac. In reference to this point, Boudon states
that la traduction syriaque simplie donc le texte grec,29 which ts in with
her overall line of argument regarding the Syriac Book of medicines. It seems
likely, however, that this is not so much a simplication but an acclimatisation,
that is, a reference to a persons hair becoming blonde as they reach maturity
would make little sense in Mesopotamia.
Thus part [F] represents a good epitome of Galens Art of medicine VI
11f. The epitome in this twelfth century text was probably derived from an
existing Syriac translation, made perhaps in the ninth century, so it represents
a valuable witness to the Syriac medical terminology that underlies the later
Arabic translations.
In view of our observations regarding parts [E] and [F], we must again
take issue with the treatment of the Syriac Book of medicines in Boudons
study, which states:30
Dans dautres cas, la traduction syriaque ne donne quune version
abrge de passages plus dvelopps dans le texte grec et arabe
[]. Le texte syriaque omet donc dlibrment les nuances et
les prcisions prsentes dans le texte grec pour aller plus vite
lessential et sen tenir aux seules conclusions.
Again, it is not the case that the Syriac text just abridges its source
text, deliberately omitting details in order to get to the conclusions quickly.
This seems to accuse the Syriac translator of adopting a lazy approach to
transmission, which runs counter to the sophisticated manner in which the
Galenic source is received in the Syriac text. The scribe has rewritten the
source text to suit his own agenda, in the process simplifying it as well, and
29
30
Boudon, o.c. 233.

Boudon, o.c. 233f.
137
has abridged the sections that suit his purposes with accuracy and in a way
that would be sensible to his readers.
Furthermore, to refer to the Syriac text of the Book of medicines as a
translation really misses the point and invites people to assess it on the wrong
basis. This results in the distinct possibility that an important text will be
relegated to the margins of scholarly endeavour because it is considered of
little value. Indeed, this is clearly the case in Boudons study, as she then
states:31
Quoi quil en soit, le caractre ncessairement tronqu de ces citations syriaques rend dicile leur utilisation pour ltablissement
du texte grec quelles ne permettent pas de sensiblement amliorer.
Aussi navons-nous pas retenu ce tmoignage dans lapparat critique.
This results in an indefensible situation in which Gerard of Cremonas
Latin text is incorporated into Boudons critical apparatus, but the Syriac of
the Book of medicines is not. In other words, a Latin translation based on an
Arabic intermediary, which itself is probably based on a Syriac intermediary,
is used, while a Syriac abridgement is rejected.32
It is clear that a wholesale analysis of the mechanics of Galens reception in the Syriac Book of medicines is a strong desideratum. Fortunately,
Schleifers articles have laid the foundation for this. That this work has not
been advanced in the ensuing century is another sad testimony to the neglect
experienced by the Syriac sources.
V. Practical Remedies in the Syriac Book of medicines
The incorporation of Galens Art of medicine VI 2-12, discussed above,

occurs in the third chapter of the Book of medicines, which runs from folios
1b-30a and is concerned with diseases of the head.33 The rst part, folios
1b-23b, is mostly devoted to theoretical matters, and contains much that is
derived from Galen.34 The second part, folios 23b-30a, is more practical and
31
32
33
34
Boudon, o.c. 234.

See, for example, Boudon, o.c. 267.
See Budge, o.c. I xli-li, 1-61 and II 1-65.
Schleifer, o.c. 75-113.
138
Siam Bhayro
presents a series of medicinal recipes.35 The ingredients vary, but include many
elements that suggest a Mesopotamian rather than a Greco-Roman context
for the recipes. Consider, for example, this list of ingredients for a recipe
for headaches, which is indicative of the kind of recipe encountered in this
section:36
.
.
.
.
.
.
[a]
- fat of jackal or juice/sap of fennel: this phrase is ambiguous because both elements possess multiple meanings.37 The former interpretation was given by Budge,38 and would appear to suggest that this is
not a Greco-Roman recipe. The latter interpretation, which is supported by
ingredients [b] and [e], could leave the origin open to question. The context of its use in the treatment of head complaints, however, resonates much
more with contemporary Arabic prescriptions,39 rather than the earlier Greek
sources, so its use here probably reects an oriental rather than Greco-Roman
context.
[b]
- This is a loan from Greek ,40 which probably refers
to (the resin of) Persian fennel.41 Despite its name, it is described in GrecoRoman sources as growing in Syria,42 which probably includes upper Mesopotamia. Its use for head complaints accords with our observations for substance
[a].
[c]
- pure mumia: The precise meaning of

is unclear,
but it probably refers to a type of asphalt or bitumen that probably originates
35
Indeed, Schleifer gives no notes for these pages.

From Budge, o.c. I 56.
37
See M. Sokolo, A Syriac Lexicon: A Translation from the Latin, Correction, Expansion, and Update of C. Brockelmanns Lexicon Syriacum, Winona LakePiscataway 2009, 451f. and 584.
38
Budge, o.c. II 58.
39
See, for example, E. Lev-Z. Amar, Practical Materia Medica of the Medieval
Eastern Mediterranean according to the Cairo Genizah, Leiden 2008, 166-168.
40
Merx, o.c. 295.
41
See Sokolo, o.c. 966, s.v.
.
42
See Lev-Amar, o.c. 171.
36
139
from India, Iran and Iraq.43 Thus Syriac

and Arabic both possibly derive from an Iranian term (compare Middle Persian mm wax < Old
Iranian *mauma impure liquid).44
[d]
- Syrian rue: Greek Syrian rue is probably a loan from
an oriental original that is realised in Syriac as
and
.45 In his

discussion of the rue, Sergius tell us:
And others call this wild pygn -, and again others -armela -, and the Syrians
name it b - .46 The rst name listed by Sergius for this plant derives from
the Greek rue, while the second is probably the Semitic form that
underlies both Syriac
and Arabic , as well as Greek and
the plants scientic name Peganum harmala.47 The third name, which is the
one used in our recipe, is said by Sergius to be the proper Syriac name for the
plant.
[e]
- juice of : Budge interpreted this as fat made

48
from horns, but
is now known to derive from Cyrene (compare
Greek ),49 the north African city that was a major exporter of the
plant . In his discussion of juice, Sergius states:

.
.
That
which comes from Cyrene we name thus (i.e. after Cyrene), and (also) that
which is from Media and that which is from Syria.50 In other words, the term
had been applied to other regional varieties of in the same
way that Greek has , etc.51 In his list
43
See Sokolo, o.c. 725-762, s.v.

and
; Lev-Amar, o.c. 453f.
See D. Durkin-Meisterernst, Dictionary of Manichaean Middle Persian and
Parthian, Turnhout 2004, 233; D.N. MacKenzie, Manis buhragn, Bulletin of the
School of Oriental and African Studies XLII (1979) 500-534: 531.
45
See Sokolo, o.c. 194f.
46
For the text, see Merx, o.c. 282.
47
See Lev-Amar, o.c. 505f. See also C.C. Townsend, E. Guest, Flora of Iraq
Volume Four Part One: Cornaceae to Rubiaceae, Baghdad 1980, 302-304.
48
Budge, o.c. II 58.
49
See Sokolo, o.c. 1366.
50
For the text, see Merx, o.c. 291.
51
See H.G. Liddell-R. Scott, A Greek-English Lexicon, 9th ed., Oxford 1940,
1599.
44
140
Siam Bhayro
of plant names, Sergius equated

< with
the plant asafoetida (compare Arabic ), which is the Median variety,

so it is possible that this is the variety being referred to in the recipe.52 In
any event, the term being employed is a native Syriac invention based on the
toponym famed for exporting the substance rather than the Greek loan word
for the substance itself.
[f ]

- round and long pepper: This is a common
oriental term for a very common ingredient.
[g]
- gum ammoniac: This is derived from early Middle Persian
*(w)uaq, which was loaned into Syriac and then into Arabic as , and is
preferred in the recipe over the Greek loan word
< .53
[h]
- secretion of beaver: This is a loan from Greek ,
which was also loaned into Arabic as . Its use for treating head complaints, however, is indicative of contemporary Arabic medical texts rather
than the earlier Greek sources.54
[i]
- spurge: At rst sight,
is clearly a loan from Greek
, but it has an uncharacteristic way of transliterating the initial
Greek -, for which we would expect -. Indeed, this is the form Sergius
gives:
.55 It appears that the Syriac form is immediately derived
from the Arabic , suggesting that the term had, by the twelfth century,
become orientalised.56
[j]
- urine of camel: This is certainly not a common
ingredient in Greco-Roman medicine.
Thus terms [a], [c], [d], [e], [f ] and [g] are of clear oriental derivation,
with clear Greek-derived alternatives rejected in the cases of [d], [e] and [g].
Moreover, term [a], taking the more likely explanation, reects an oriental use
of the substance. Terms [b] and [h] are Greek loan words, but the use of the
substances is oriental. Term [i] is a Greek loan word, but through an Arabic
52
53
54
55
56
See
See
See
See
See
Merx, o.c. 296; see also Lev-Amar, o.c. 339f.

Sokolo, o.c. 23 and 54; Lev-Amar, o.c. 333f.
Sokolo, o.c. 1387; Lev-Amar, o.c. 354f.
Merx, o.c. 257.
Lev-Amar, o.c. 487f.
141
intermediary in what was probably its orientalised form. Furthermore, substances [a] (following Budge) and [j] are not typical of Greco-Roman medicine,
and substance [c] was probably unknown in the west. It appears, therefore,
that this is a Mesopotamian rather than a Greco-Roman medicinal recipe.
This is a feature that is repeated throughout the rst scientic section
of the Book of medicines. In successive chapters, the Galenic material has been
edited together with dozens of medicinal recipes that appear to belong to the
traditional, indigenous medicine of Mesopotamia. It seems, therefore, that the
Mesopotamian medical compendia57 did not simply abridge and reorganise
Hellenistic medical lore, but also incorporated those aspects of indigenous
medicine that were still thought to be ecacious.58
VI. Conclusion
The Syriac Book of medicines is an invaluable witness to the mechanics of reception. It shows how Mesopotamian herbals persisted in the face of
the arrival of Greco-Roman science. It also shows how this new science was
received and amended to suit its new cultural, material and religious context.
Since it became known to western scholarship, the Syriac Book of medicines
has suered, initially on account of the prevailing prejudices of the early twentieth century, which sought to contrast the irrational, mystical orient with the
scientic occident, but more recently on account of its neglect or dismissal as
an unreliable source, even when more questionable sources are not dismissed.
Although it is not a translation, in many cases, the Syriac Book of medicines
will be our only source for the Syriac intermediary between the Greek and
Arabic medical terminology, so it certainly merits its place at the centre of
our investigations.
57
I say compendia because their production continued well into the nineteenth
century, e.g. manuscript Syriaque 325 see R.J.H. Gottheil, Contributions to Syriac
Folk-Medicine, JAOS XX (1899) 186-205; see also Bhayro, o.c. 150f.
58
It is interesting that, in an attempt to infuse these medicinal recipes with
the authority of Greco-Roman science, some of these prescriptions were ascribed to
gures such as Galen and Ptolemy. The subject of medical pseudepigraphy in the
Syriac Book of medicines, however, merits a separate, detailed study.
142
Dept. of Theology and Religion

Univ. of Exeter, Amory Building,
Rennes Dr, Exeter EX4 4RJ, UK
Siam Bhayro
S B
[email protected]
VII. Appendix: Schleifers articles on the Syriac Book of medicines
J. Schleifer, Zum Syrischen Medizinbuch, Zeitschrift fr Semitistik und verwandte Gebiete IV (1926) 70-122.
Zum Syrischen Medizinbuch, Zeitschrift fr Semitistik und verwandte Gebiete IV (1926) 161-195.
Zum Syrischen Medizinbuch, Zeitschrift fr Semitistik und verwandte Gebiete V (1927) 195-237.
Zum Syrischen Medizinbuch, Zeitschrift fr Semitistik und verwandte Gebiete VI (1928) 154-177.
Zum Syrischen Medizinbuch, Zeitschrift fr Semitistik und verwandte Gebiete VI (1928) 275-299.
Zum Syrischen Medizinbuch. II Der Therapeutische Teil, RSO XVIII
(1940) 341-372.
Zum Syrischen Medizinbuch. II Der Therapeutische Teil, RSO XX (1942/
1943) 1-32.
1943) 163-210.
1943) 383-398.
Zum Syrischen Medizinbuch. II Der Therapeutische Teil, RSO XXI (1946)
157-182.
VIII. Bibliography
A.H. Becker, Doctoring the Past in the Present: E.A. Wallis Budge, the Discourse on Magic, and the Colonization of Iraq, HR XLIV (2005) 175-215.
S. Bhayro, Syriac Medical Terminology: Sergius and Galens Pharmacopia,
Aramaic Studies III (2005) 147-165.
143
S. Bhayro-S.P. Brock, The Syriac Galen Palimpsest and the Role of Syriac in
the Transmission of Greek Medicine in the Orient, BRL, themed issue: R.
David (ed.), Ancient Medical and Healing Systems in the Near East, Greece and
Rome, forthcoming.
Galien, II. Exhortation la Mdecine, Art mdical, ed. V. Boudon, Paris 2000.
S. Brock, A Brief Outline of Syriac Literature, Kottayam 1997.

E.A.W. Budge, The Syriac Book of Medicines: Syrian Anatomy, Pathology and
Therapeutics in the Early Middle Ages, I-II, London 1913.
A. Cunningham, The Theory/Practice Division of Medicine: Two Late-Alexandrian Legacies, in T. Ogawa (ed.), History of Traditional Medicine. Proceedings of the 1st and 2nd International Symposia on the Comparative History
of Medicine - East and West. 1st Symposium: October 22-28, 1976, 2nd
Symposium: October 23-29, 1977, Susuno-shi, Shizuoka, Japan, Tokyo 1986,
303-324.
M.W. Dols, Syriac into Arabic: The Transmission of Greek Medicine, ARAM
I (1989) 45-52.
D. Durkin-Meisterernst, Dictionary of Manichaean Middle Persian and Parthian,
Turnhout 2004.
R.J.H. Gottheil, Contributions to Syriac Folk-Medicine, JAOS XX (1899)
186-205.
J.F. Healey, Aramaic Inscriptions & Documents of the Roman Period, Oxford
2009.
H.D. Isaacs, Arabic Medical Literature, in M.J.L. Young-J.D. Latham-R.B. Serjeant (edd.), Religion, Learning and Science in the Abbasid Period, Cambridge
1990, 342-363.
R. Le Coz, Les mdecins nestoriens au Moyen ge: Les matres des Arabes,
Paris 2004.
E. Lev-Z. Amar, Practical Materia Medica of the Medieval Eastern Mediterranean according to the Cairo Genizah, Leiden 2008.
H.G. Liddell-R. Scott, A Greek-English Lexicon, 9th ed., Oxford 1940.
D.N. MacKenzie, Manis buhragn, Bulletin of the School of Oriental and
African Studies XLII (1979) 500-534.
144
Siam Bhayro
A. Merx, Proben der syrischen Uebersetzung von Galenus Schrift ber die einfachen Heilmittel, ZDMG XXXIX (1885) 237-305.
S. Moscati, An Introduction to the Comparative Grammar of the Semitic Languages: Phonology and Morphology, Wiesbaden 1964.
V. Nutton, From Galen to Alexander, Aspects of Medicine and Medical Practice
in Late Antiquity, DOP XXXVIII (1984) 1-14 [reprinted in V. Nutton, From
Democedes to Harvey: Studies in the History of Medicine, X, London 1988, 114].
P.E. Pormann-E. Savage-Smith, Medieval Islamic Medicine, Edinburgh 2007.
J. Schleifer, Zum Syrischen Medizinbuch, Zeitschrift fr Semitistik und verwandte Gebiete IV (1926) 70-122.
Galen. Selected Works, transl. by P.N. Singer, Oxford 1997.
M. Sokolo, A Syriac Lexicon: A Translation from the Latin, Correction, Expansion, and Update of C. Brockelmanns Lexicon Syriacum, Winona LakePiscataway 2009.
C.C. Townsend-E. Guest, Flora of Iraq Volume Four Part One: Cornaceae to
Rubiaceae, Baghdad 1980.
M. Ullmann, Islamic Medicine, Edinburgh 1978.
Medieval hospital formularies: Byzantium and Islam compared
I. The novel availability of editions or transcriptions of medical writings associated with hospitals in the medieval Middle East makes possible
something that has not been attempted before: comparison of Byzantine and
Islamic examples of such material.1
This comparison of what we can label hospital iatrosophia2 is important for two reasons. First, it makes good long neglect. If iatrosophia in
general have largely been ignored, those relating to hospital practice have,
until quite recently, remained wholly unknown. Second, the comparison adds
a fresh dimension to the debate about hospital medicalization in the Middle
Ages. This is a more intricate task.
Great claims have by turns been made for the quasi-modernity of both
Byzantine and medieval Islamic hospitals. Philanthropic social welfare and
medical assistance institutions [in Byzantium] [. . .] were in every respect perfect and nearly similar to present day institutions of this kind [. . .] they were
the first fully equipped European hospitals.3 No serious historian would
write in such terms now. Yet the medical superiority of Byzantine hospitals,
1
For the definition and outline history of the hospitals in question, see P.
Horden, The earliest hospitals in Byzantium, western Europe, and Islam, Journal
of Interdisciplinary History XXXV (2005) 361-389; Horden, How medicalised were
Byzantine hospitals?, M&S X (2006) 45-74; P.E. Pormann, Islamic hospitals in
the time of al-Muqtadir, in John Nawas (ed.), Abbasid Studies, II, Leuven 2010, 337381. See also D. Stathakopoulos in this collection. What follows could not have been
written without the unstinting assistance and advice of the late David Bennett.
2
For the definition of iatrosophia see Nutton in this volume.
3
G.C. Pournaropoulos, Hospital and social welfare institutions in the medieval Greek Empire (Byzantium), in XVIIe Congrès international dhistoire de la
medicine, I, Athens 1960, 378, quoted by D. Constantelos, Byzantine philanthropy
and social welfare, New Rochelle, N.Y. 19912 , 118.
146
HORDEN
at least over those to be found in contemporary western Europe, has been

stoutly maintained by Timothy Miller in a number of publications. Xenones
[or nosokomeia hospitals] evolved [he writes, in 2008] as the principal centres of Byzantine medicine because of their link to the archiatroi, the ancient
chief physicians of the polis. And he adds: that members of the ruling elite
sought treatment at Byzantine hospitals demonstrates how different these institutions were from the H
otel-Dieu in Paris or Saint Bartholomew in Lon4

don . Even such modified eulogy of Byzantine charitable establishments has
been questioned. Yet this way of describing medieval hospitals lives on in
effect asserting the superiority of one culture over another by reference to the
degree to which its hospitals were in some sense medicalized.
It lives on, indeed has long flourished, in the realm of scholarship about
early Islamic hospitals. The emergence of the prototype of the modern hospital in medieval Islam: the title says it all.5 That is from 1980. The latest,
and also the most learned and authoritative exponent of such a view has been
Peter E. Pormann. With him, the comparandum is Byzantium, not western
Europe it is always the poorer neighbour. And the contrast drawn is between
(implicitly) backward Christian health care in Byzantium and forward-looking,
secular or at least multi-faith establishments in Islam. In these hospitals, as
in Millers version of Byzantium, elite patients were treated by elite physicians.
And the density of hospital provision in tenth-century Baghdad is ultimately
validated by estimating its statistical superiority to that of modern Pakistan:6
It is therefore evident that the medieval Islamic hospital was a
4
T. Miller, Charitable Institutions, in Elizabeth Jeffreys et al. (edd.), The

Oxford Handbook of Byzantine Studies, Oxford, 2008, 627, reflecting the overall argument of Miller, The Birth of the Hospital in the Byzantine Empire, Baltimore 19972 .
5
A. Sayili, The Emergence of the Prototype of the Modern Hospital in Medieval Islam, T
urk Tarih Kurumu Basimevi Belleten XLIV (1980) 279-286, cited by
Leigh Chipman, The World of Pharmacy and Pharmacists in Mamluk Cairo, LeidenBoston 2010, 135.
6
P.E. Pormann-E. Savage-Smith, Medieval Islamic Medicine, Edinburgh
2007, 101, with italics added; echoing Pormann, Al-Muqtadir cit. and other papers by
him (which is why I attribute the view to Pormann solely rather than to both authors).
For the comparison of proportional hospital provision in tenth-century Baghdad with
that of the whole of Pakistan in 1998 see Pormann and Savage-Smith cit. 110. According to the on-line World health statistics 2010 Pakistan did indeed average 6 hospital
beds per 10,000 of population in 2000-2009, placing it at very near the bottom in
Medieval hospital formularies
147
more elaborate institution with a wider range of functions than

the earlier poor and sick relief facilities offered by some Christian monasteries and hospices. The care for the insane in hospitals was unprecedented and an important part of even the earliest Islamic hospitals. Although early Islamic hospitals drew on
Christian models, there are some things which make them unique.
First, unlike their Christian counterparts, the medicine practised
there was secular in character, insofar as it was based on the principles of humoral pathology rather than religion. Treatment in
Christian institutions before the eighth century would customarily begin with a confession, and be carried out by monks or priests
that is to say, by Christians for Christians through Christian ritual (although of course not limited to it). In contrast, there were
no mosques or places for religious ritual associated with Islamic
hospitals, though in a few instances mosques were attached at a
later date. In Islamic hospitals from the tenth century onwards
(as far as we know), Muslim, Jewish, Christian, and even pagan
doctors worked together, treating not only patients of their own
community, but also those of other communities. Secondly, elite
medical theory and practice came together in Islamic hospitals in
a way that they had not done so before. This is illustrated by
the fact that some of the best physicians of the time worked in
hospitals and produced there some of their most advanced and
innovative research. Thirdly, over time hospitals became centres
for the teaching of medicine, in which students were encouraged
to obtain their medical education. Finally, hospitals were part of
wider public health efforts promoted by the ruling elite.
Again, therefore, the hospital is presented as a centre of medical excellence
of research, teaching, and practice informed by theory. It is a medical-history
equivalent of the translatio imperii: from Constantinople to Baghdad. The
translatio may turn out to be entirely justified, but its basis needs testing
and on as many fronts as the evidence permits.
There are various ways in which the accuracy of the Byzantine-Islamic
comparison could be appraised. But some obvious methodological caveats
terms of world rankings. This statistic seems to tell us more about Pakistan than
about Baghdad.
148
HORDEN
must be heeded first. Is like being compared with like or does the starkness
of the presumed contrast between a Christian and a secular ethos vitiate further discussion? Is what is being compared the generality of hospitals in each
culture, which is what the rhetoric of both Miller and Pormann implies? Or
are really only a few metropolitan hospitals in question, as the specific supporting examples (such as Baghdad) suggest? Still more to the point, on what
yardstick is medicalization being measured the mere presence or absence of
doctors; the degree of authority they wielded; whether they were courtiers;
the number of beds at their disposal; what sort of medicine they were therefore likely to have dispensed; or how that medicines efficacy would have been
assessed in the modern laboratory?7
On a practical level, answers to these methodological questions must be
sought using contemporary evidence, or as near to contemporary as possible.
This is not the place to review that evidence, which (very much in Pormanns
favour, let it be stressed) does include some first-hand, contemporary descriptive texts of elite physicians active in hospitals, as well as later, retrospective
narratives in the same vein; prescriptive (deontological) exhortations to learn
medicine in hospitals; and medical case books that record experiments with
treatments that in some ways might be seen to anticipate modern clinical
trials.8
What should be mentioned here is a wider trend within the scholarship
of early Islamic medicine, that is, medicine from the ninth century onwards.
This trend is to point up the contrast, not between Islam and some other culture, but within Islamic medicine itself a contrast between the sophisticated
procedures and treatments advocated on the theoretical and technical medical
literature of the period and the much simpler responses to the same medical
problems recorded in the case histories.9 The contrast is more than a matter
of the doctors selecting a narrow range of treatments from the large arsenal in
principle available to them.10 It suggests a deliberate avoidance of the sophis7
On that last-mentioned front see e.g. Leigh Chipman, How effective were
cough remedies known to medieval Egyptians?, Korot XVI (2002) 135-157.
8
Pormann, Medical methodology and hospital practice: the case of fourth/tenth-century Baghdad, in P. Adamson (ed.), In the age of al-Farabi: Arabic philosophy in the fourth/tenth century, London 2008, 95-118.
9
See the contributions of Emilie Savage-Smith and Cristina Alvarez-Mill

an

to Social History of Medicine XIII (2000) 293-306, 307-321.
10
Pace Pormann, Al-Muqtadir cit. 364.
149
tication paraded in those writings intended to attract and hold the attention
of potential patrons. It seems almost to concede that these writings were only
(to recall the gloomy tones of Manfred Ullmann) lifeless theory [. . . ] mere
book knowledge.11 Perhaps the practice of elite medicine in hospitals was
different from the theory.
The possible implications of that for the comparison of Byzantine and
Islamic medical literature deserve further exploration elsewhere. For the
moment it is simply worth bearing them in mind as we come to the one type
of evidence that has not so far been pressed into service when juxtaposing the
hospitals of the two empires: the Greek hospital iatrosophia and their Arabic
equivalents, the hospital aqrabadhinat.12
II. The first task is to establish the corpora involved. On the Byzantine
side, David Bennett has surveyed the relevant manuscripts, in a discussion that
supersedes all others in both scope and thoroughness.13 I am indebted to him
for permission to summarize and continue disseminating his main findings.
There are five or six texts in question depending on how one counts a
text that has at some point been divided into two by its copyists.
1. Prescriptions and classifications [of fever?] of the great hospitals,
of the kind that doctors prescribe from experience for healing, especially for
patients in the hospitals. That is the translated title of one version of a
compilation of treatments (parts of which, including the title, variously appear
in at least four other manuscripts). The compilation is divided under sixteen
very miscellaneous headings and dates from approximately 1050. It is found
in the fourteenth-century Vat. gr. 292. Three other manuscripts (nos 2, 4,
and 5 below) also preserve these prescriptions and classifications to varying
extents, but sometimes without the titular ascription to hospitals.
2. Vat. gr. 299 is an anthology of medical writings dating from the later
fourteenth century. Within a long concluding medical compilation (of around
11
M. Ullmann, Islamic medicine, Edinburgh 1978, 106.

B. Lewin, Akrabadhin, or karabadhin, in EI2 (the term derives from the
Greek graphidion, short treatise or register, via the Syriac).
13
D.C. Bennett, Xenonika: medical texts associated with Xenones in the late
Byzantine period, University of London PhD thesis 2003, together with his contribution to this collection, and his forthcoming monograph, Xenonika, which he kindly
discussed with me. See also Miller, Birth of the hospital cit. ch. 9. The summary that
follows revises and corrects that in my paper, How medicalised? cit. 60-62.
12
150
HORDEN
180,000 words) it contains five remedies ascribed to three named physicians of

the Mangana hospital, founded in the mid-eleventh century,14 and one other
remedy ascribed to a named, but otherwise unknown doctor, for whom no institutional affiliation is given. The three hospital physicians are: (a) Stephanos,
archiatros (chief physician) and aktouarios (court physician);15 (b) Abram the
Saracen, aktouarios and basilikos (imperial) archiatros; and (c) Theodore, iatros (physician) at the Mangana. There are six other passages ascribed only
to the Mangana hospital, with no physician named. These are dispersed over
about a half of the compilation but form only a tiny proportion of the whole.
A further six passages in the same remedy collection correspond to parts of the
collection in Vat. gr. 292 (no. 1, above) in which they are said to be derived
from the great hospitals.
3. The fifteenth-century Par. gr. 2194 includes eight remedies ascribed to
Michael, aktouarios of the otherwise undocumented Mauraganos (recte Maurianos?) hospital. These six remedies are found in a text headed, in a hand that
differs from that of the copyist, dynameron xenonikon dia peiras (On the potency of hospital prescriptions found by experience). That text is succeeded
by another similar brief collection entitled, even more simply, xenonika (from
the main Greek term translatable as hospital, xenon). So far as is currently
known, none of the hospital-related material found here survives in any other
manuscript.
4. The Vind. med. gr. 48, from the late thirteenth century, has a text
attributed in its title to Romanos, kouboukleisios (an imperial honorific) of the
Great Church (Hagia Sophia) and protomenutes (chief physician) of the imperial Myrelaion Hospital in Constantinople.16 Fragments of this text survive
in only two other manuscripts. The title kouboukleisios disappeared after the
tenth century, and the Myrelaion hospital was re-founded by the Emperor Ro-
14
Miller, Birth of the hospital cit. 149f., with P. Lemerle, Cinq etudes sur le
XIe siècle byzantin, Paris 1977, 273-283.
15
On the changing meaning of aktouarios see Oxford Dictionary of Byzantium,
I-III, Oxford 1991, s.v.
16
On the protomenutes see U. Criscuolo, Pour le texte du medecin Romanos,
in A. Garzya (ed.), Storia e ecdoctica dei testi medici greci. Histoire et ecdotique
des textes medicaux grecs. Actes du colloque international Paris 24-26 mai 1994,
Naples 1996, 113-131: 114.
151
manus Lecapenus in the mid-tenth century.17 Romanos kouboukleisios cannot

be dated any more precisely than that.
5. The text attributed to Romanos is actually only the first half of a
much longer work. Its second half survives separately, in the Laur. plut. 75,19,
under a different authors name, as the Apotherapeutike of one Theophilos, in
which the material is said to be drawn from hospital books (xenonikon biblon).
Apotherapeutike is an odd term, but its sense of remedy list is clear enough.18
Both these two parts Romanoss and Theophiloss contain passages
similar to parts of Vat. gr. 299 (no. 2 above) where the hospital treatments
are attributed to the Mangana hospital.
6. Manuscript Laur. plut. 7,19, of the thirteenth to fourteenth centuries,
is a collection mainly of theological works. Like nine other manuscripts, it
contains a text (mostly, but not always, the same text) with the title: Therapeutic medical treatments by various doctors set in order according to the
defined procedure of the xenon. This is a short piece of some 2,750 words.
In none of its versions does it live up to the orderliness implied in its title. It
includes abbreviated versions of remedies recorded in four other manuscripts
under the name of John archiatros,19 in one other manuscript under that of
Galen, and in a sixth, under both names. More on the Therapeutic medicines
below.
Overall, then, we have five or six texts, known to us from eighteen
manuscripts, that have hospital connections made explicit in their titles or
their contents. To them can be added two manuscripts (Par. gr. 2315 and
2510) that were copied for hospitals, one manuscript (Scorialensis Y.III.14)
dedicated to a hospital by its scribe George (all three of these from the fourteenth century), and perhaps three or four others that may have been owned
by a hospital in the late Byzantine period, including such luxury products
as the Niketas codex and the rebound Vienna Dioscorides.20 Some fifteenth17
Miller, Birth of the hospital cit. 113f.

On the text see also A.P. Kousis, The Apotherapeutic [sic] of Theophilos
according to the Laurentian codex, plut. 75, 19, PAA XIX (1944) 35-45; Anna
Maria Ieraci Bio, Sur une attribuee `
a Theophile, in A. Garzya, Storia
e ecdotica dei testi medici greci, Naples 1996, 191-205.
19
See further B. Zipser, John the Physicians Therapeutics, Leiden-Boston
2009.
20
Bennett, o.c. appendix V, 440f., listing also post-Byzantine manuscripts
ascribed to xenones.
18
152
HORDEN
century humanist feats of Galenic collecting and scholarship may have had
a hospital setting in the Kral Xenon in Constantinople, but they are as far
removed from patients immediate clinical needs as they are from the style of
the iatrosophion.21
III. As for the Islamic side: there was a genre of medical formularies (or
pharmacopoeias, or dispensatories), written collections of compound remedies
the aqrabadhin. Within that body of writings we find at least three works
explicitly redacted for use in the bimaristan (hospital).
1. Sabur ibn Sahl (d. 869) was a Nestorian Christian physician from
Gondeshapur in Iran, where he may have held a post in the citys famed local
hospital.22 He became a court physician to the Abbasid Caliph al-Mutawakkil
and his successors. His formulary survives in three different versions of varying
length, known to philologists appropriately as the short, middle and long versions. It is noteworthy in passing that the unique surviving manuscript of the
short one (Berol. or. oct. 1839, copied around 900) shows signs of extensive
use: thumb imprints on the cover, reading marks of wax on several leaves,
and numbering of quires and folios as finding aids perhaps added early in the
manuscripts life.23 Overall these different versions survive in five manuscripts
dating from the tenth to the seventeenth century or even later. It is not clear
which version most faithfully reflects Saburs original, perhaps written in Syriac. Nor can it even be known whether any of the surviving redactions were
made during his lifetime. In the mid-eleventh century, someone produced a
fourth version: a revised, rearranged, abridged text explicitly entitled: The
formulary of Sabur according to the ,Adudi hospital, a synopsis of Saburs formulary on the composition of drugs, [in] sixteen chapters (one chapter fewer
than the small version but drawing on both small and large redactions). The
21
Ibid. 37, 62f., 81f., 124; Brigitte Mondrain, Jean Argyropoulos professeur
`
a Constantinople et ses auditeurs medecins, dAndronic Eparque
`
a Demetrios Angelos, in C. Scholz-G. Makris (edd.), POLYPLEUROS NOUS: Festschrift f
ur Peter
Schreiner, Munich 2000, 223-249. See also Nutton and Petit in this volume.
22
For what follows see M. Ullmann, Die Medizin in Islam, Leiden 1970, 300f.;
Sabur ibn Sahl. Dispensatorium parvum, ed. by O. Kahl, Leiden-New York 1994; Sabur
ibn Sahl. The Small Dispensatory, transl. by O. Kahl, Leiden-Boston 2003; O. Kahl,
Sabur ibn Sahls Dispensatory in the Recension of the ,Adudi Hospital, Leiden-Boston
2008.
23
Kahl, Small Dispensatory cit. xi.
153
,Adudi was the hospital founded in Baghdad on the west bank of the Tigris
by the Buyid ruler ,Adud al-Daula. It opened in the year of his death, 982.
The version made for it is represented in a unique manuscript, dated 1341,
the work of a copyist not entirely at ease with what was presumably being
dictated to him: Munich Staatsbibliothek arab. 808,2. I shall come back to
Sabur and this hospital text below.
Some circumstantial evidence suggests that the hospital formulary under
his name became, if not standard, then widespread in the major hospitals
of Baghdad, perhaps of the caliphate. The court and hospital physician alKaskari who worked in tenth-century Baghdad wrote in the prefatory remarks
to his medical compendium:
In each chapter I have limited myself to a description of the
pastes [. . .] pills, pastilles, liquids, collyria [. . .] dressings, and
other things in Saburs Dispensatory to which people resorted
in the hospitals. I compared this with the manuscripts in the
hospitals dealing with this.24
In his Fihrist the bio-bibliographer Ibn al-Nadim (d. 995-998) corroborates
the information that Saburs work was used in hospitals as well as apothecary
shops.25
2. Ibn al-Tilmidh (b. c. 1065 in Baghdad, d. 1154 or 1165) was another
Christian doctor who worked in the ,Adudi hospital.26 He rose to be chief
physician there as well as court doctor to the Caliph al-Muqtadi,. Written
in the 1140s, his formulary became, according to the later biographical dictionaries, the default pharmacopoeia in the hospitals of Baghdad, replacing
that of Sabur.27 This formulary survives in two recensions, long (20 chapters)
and short (13). A version of the short text seems also to have been adopted
as a hospital formulary. The main, long, version comes to us in at least five
manuscripts ranging in date from c. 1200 to c. 1600. According to the subtitles
of some of these it was compiled from a number of [other] formularies.
24
Pormann, Al-Muqtadir cit. 346; Ullmann, Medizin cit. 300.

Ed. G. Fl
ugel, I-II, Leipzig 1871-1872, II 297.
26
The Dispensatory of Ibn at-Tilmid, ed. by O. Kahl, Leiden 2007; Ullmann,
Medizin cit. 306; Chipman, World of Pharmacy cit. 31f.
27
Kahl, Ibn at-Tilmid cit. 5.
25
154
HORDEN
3. Finally in this roster of authors of hospital aqrabadhinat comes the

(Karaite) Jewish apothecary Ibn Abi al-Bayan (d. c. 1236).28 He too was a
court physician, and treated Saladins successor al-,Adil. He was director of
the Nasiri hospital in Cairo and composed his Dustur al-bimaristani for use
in it. It survives in some twelve manuscripts, which represent two different
versions, dating from around 1200 to 1600, and was perhaps based on earlier
compilations already maintained in the Cairo hospital.29
IV. With the cast of texts assembled, we can turn to the business of
comparison.
Even a brief review of the material in Arabic immediately suggests differences from the Byzantine corpus. First, at the simple level of the progress
of scholarship, on the Islamic side there are editions and translations at our
disposal, as well as a significant if small secondary literature (which is one reason why I have given less detail of Arabic than of Greek manuscripts). Even if
some of the editions are very recent, this discursive literature has been building
up over a number of years. That state of affairs contrasts with the virtually
solitary work of David Bennett.30 To some extent the contrast reflects the far
greater degree of philological and historical attention given to Islamic medicine
in general, by comparison with the neglect from which Byzantine medicine has
quite generally suffered. Many of the major Greek texts languish unedited,
not to mention the iatrosophia. There are some studies of individual authors
and writings but no modern synthesis.31
28
Chipman, World of Pharmacy cit. 38-41; Ullmann, Medizin cit. 309.

Chipman, World of Pharmacy cit. 39; Ullmann, Medizin cit. 309.
30
See Bennett, o.c. 34f., for previous work; also his contribution to Social
History of Medicine XIII (2000) 279-291. See also A. Touwaide, Byzantine hospital
manuals (iatrosophia) as a source for the study of therapeutics, in B.S. Bowers (ed.),
The medieval Hospital and medical Practice, Aldershot 2007, 148-173.
31
A Byzantine equivalent to Pormann and Savage-Smith is almost inconceivable in the present state of our knowledge; see meanwhile DOP XXXVIII (1984),
Symposium on Byzantine medicine.
29
155
The contrast owes far more, however, to the fact that the manuscript
traditions of the Islamic works in question are r e l a t i v e l y easy to establish, at least in broad outline. For all the individual variation of particular
witnesses, they fall into distinct recensions each with its own stemma. We
find named authors, with biographies known even if only in stark outline. The
authors are often active at court. They also hold high rank in specific hospitals
that have their own documented histories, two in Baghdad, one (in the case of
Ibn Abi al-Bayan) in Cairo. Their writings are specifically redacted for use in
these hospitals. They display a firm editorial hand. They cite their sources,
but they also rework them. There are some vignettes in narrative evidence
of the actual use of their work. Each authors text seems to achieve market
dominance in its region. That dominance lasts for centuries in the cases of
Sabur ibn Sahl and Ibn al-Tilmidh. Ibn al-Bayan was surpassed in popularity rather more quickly, by the Jewish apothecary al-Kuhin al-,Attar. But
then the latters Minhaj al-dukkan (Management of the [pharmacists] shop)
of about 1260 continued in use for four hundred years, until the 1660s.32 The
status of hospital medicine within the whole genre of the pharmacopoeia is
attested by the way al-Kuhin al-,Attar explicitly removed hospital references
when transplanting material from his principal sources, which included Sabur
and Ibn al-Tilmidh.33 He was trying, almost against the odds, to replace the
perspective of a hospital doctor with that of an apothecary.
Nothing like this can be said of the Byzantine corpus. Its authors or
compilers are anonymous, pseudonymous, or at best obscure. Their careers,
such as they may have been, are lost to us. Though some parts of the corpus
purport to have a specific hospital association (e.g. with the Mangana), the
actual hospitals in which the texts may have been used are unknowable. The
textual history of the material is immensely complex. Writings combine and
recombine, changing title and ascribed authorship. If the Islamic material
presents a stiff but ultimately surmountable challenge, that from Byzantium
is an editors nightmare. Overall, little can be surmised about where or when
the texts were deployed. There is certainly no clear succession of dominant
writings.
The Islamic corpus is probably larger in its page extent than the Byzantine one. I estimate that, in his thesis, David Bennett is discussing something
32
33
Chipman, World of Pharmacy cit. 1.

Ibid. 19, 32f.
156
HORDEN
of the order of 200 to 230 pages of manuscript material, including some overlaps
in content and the whole of the Romanos/Theophilos text, but not counting
wholesale duplication. (Of course this takes no account of variation in script
and folio size. My point is simply that the aggregate is nearer to 200 than to
10 or to 800 pages.) I also estimate very crudely that this is a slightly smaller
aggregate than the three main authors writing in Arabic for hospitals combined. While there are occasional signs that other hospital formularies than
the works of those three were created and preserved in Islamic hospitals, perhaps anonymously,34 it seems likely none the less that those works surviving
in Arabic are most of the iceberg, not just the tip.
That iceberg, to continue the metaphor, is quite substantial in relation
to the ice shelf from which it has become calved. Ullmanns survey of the
medieval literature of pharmacy composed in Arabic35 (that is, not counting
translations of Galen or Dioscorides) allows us to situate the work of our three
hospital writers within a larger corpus of only some twenty-three authors,
who wrote either free standing works or included major sections on complex
remedies within larger treatises (as did Ibn Sina with the Canon). I do not
know of a comparable survey of the Greek manuscript material, but I doubt
that the Bennett corpus would loom anything like as large within it.
To approach the question of relative proportion in a different way: the
scale of the Byzantine material should be set against the estimated aggregate
of 2,200 medical manuscripts surviving in European libraries (although that includes some post-Byzantine material).36 The numbers of hospital manuscripts
could of course be inflated a little. The Byzantine iceberg may well have been
deeper than the Islamic one. Many manuscripts that once existed will have
succumbed to ordinary wear and tear, let alone the Fourth Crusade or the Ottoman onslaught. Hospital material can survive without its title. And more
hospital texts doubtless remain to be discovered, hiding behind misleading or
inadequate catalogue entries. Yet there are limits to the number of hypothetical manuscripts that can plausibly be added. For material can gain as well as
34
Ibid. 39f.
Ullmann, Medizin cit. ch. 15.
36
A. Touwaide, The Corpus of Greek Medical Manuscripts: A Computerised
Inventory and Catalogue, in W.M. Stevens, Bibliographic Access to Medieval and Renaissance Manuscripts: A Survey of Computerised Data Bases and Information Services, New York 1992, 75-92; Touwaide, Byzantine medical manuscripts: toward a
new catalogue, ByzZ CI (2008) 199-208.
35
157
lose its xenon ascription in the unpredictable course of copying and re-copying.
However we exercise the imagination, the number of hospital manuscripts that
were produced in Byzantium must remain a very small proportion a fraction
of one per cent of the entirety of medical writing. We must envisage a tiny
corpus as far as the Byzantine Middle Ages are concerned.37
On the Islamic front, again I do not know of any estimate of the total
surviving bulk of the manuscripts to set alongside that for the medieval Greek
corpus. Ullmanns second index to his survey of Die Medizin in Islam lists
over 1,000 Arabic and Persian book titles, including translations. While some
are no longer extant, it is a fair estimate that many of those that are, perhaps
the majority, survive in more than one manuscript. Simply as a percentage
of the total, then, the Arabic hospital texts are probably as insignificant in
relative bulk as the Greek.
Any such statistic fails, however, to take account of the relative prestige
of the two corpora. The Arabic texts may represent only a sub-genre of the
type of writing called aqrabadhin, but that genre is more distinct than that
of the iatrosophion. A few Greek hospital texts were given the sobriquet
xenonika, but hardly to the extent that implied the recognition of a new kind
of medical writing. The medical world of the great hospitals in Baghdad or
Cairo is a world in which court and hospital practice seem perfectly compatible
for elite physicians, whatever medicine they actually dispensed in the hospitals.
We have no such evidence for regular links between Byzantine hospital doctors
and the imperial court.38
On the other hand, the medical standing of Byzantine hospitals should
not be underestimated. Consider the sheer longevity of the tradition of hospital writing. What survive for us are mostly later medieval copies of ninthto-eleventh-century texts. And some of the xenonika, like the Arabic hospital
formularies, continued to be copied into the sixteenth century. Given the cost
of the materials and the skills required for the making of the least pretentious codex, this longevity is some tribute to the perceived value of hospital
remedies. Considerable stature must have attached to these remedies and
37
Bennett, o.c. 441.

The closest parallel would be the doctors attached to the Pantokator hospital
according to its typikon or foundation charter, who were not supposed to abandon
their hospital duties in order to treat provincial aristocrats, however close to the
emperor they might be: ed. P. Gautier, REByz XLV (1987) 107, lines 1305-1308,
with Horden, How medicalised? cit. 54.
38
158
HORDEN
treatments. This is a medical world in which texts mutate with each copying.
A title, if there is one, becomes an assertion of value rather than a certificate
of authenticity. Witness the remedies which are now attributed to a hospital,
now to John archiatros, now to Galen. What matters in the present context
is not which (if any) of those ascriptions is the right one. Nor is it whether a
given remedy generally originated, or was used, in a hospital. What is significant, rather, is that, at some stage in the remedys manuscript career, someone
sponsor or copyist thought that the hospital ascription was an appropriate
measure of value. A hospital remedy is as good (so the manuscripts imply) as
one supplied by Galen. A hospital archiatros is as good an authority as any of
the other possible names that might be attached to a treatment. And this is
so even in the later medieval period when there were fewer Byzantine hospitals
and it is far from clear that even the great ones continued to function on the
same scale after the Latin conquest ended.39 By the same token, hospital texts
when they are labelled as such keep very good company in the medical
anthologies that have preserved them. They can be found associated with all
the big names from Hippocrates to John aktouarios, one of the last of the
stellar Byzantine physicians.
V. So far the texts have been viewed from the outside: their profiles
within the total pool of manuscripts, within established genres, within their
social setting in hospital and court. The inside, the contents, need scrutiny.
Some common features first. It should already be evident that there is
nothing special about hospital medicine in either Byzantium or Islam. These
texts can be redactions of medical writings for general use. That is precisely
because the hospital, whether Byzantine xenon or Islamic bimaristan, is not
a place for great medical innovation, even though, as Pormann has suggested
for early Islam, it may be a place of experiment, of limited, un-controlled
treatment trials. (On the Islamic side, at least, as noted earlier, a difference
recognized at the time was not between hospital and non-hospital but between
the approaches to dispensing medicine of a physician and an apothecary.)40
39
Miller, Birth of the hospital cit. xvi-xviii, produces evidence, some of it

perplexing or oblique, of doctors active in late Byzantine hospitals in the capital. For a
thorough review see D. Stathakopoulos, Stiftungen von Spit
alern in sp
atbyzantinischer
Zeit (1261-1453), in M. Bogolte (ed.), Stiftungen in Christentum, Judentum und Islam
vor der Moderne, Berlin 2005, 147-157.
40
Chipman, World of pharmacy cit. 18f.
159
Not until the nineteenth century, after all, will hospitals anywhere begin to
provide a type of medicine that could not be had outside the clinic. Again, as
a common feature it should be remarked that both bodies of information about
drugs and remedies, the Greek and the Arabic, depend to a considerable extent
upon a classical Greek past, in which the names of Galen and Dioscorides loom
large. Finally, on neither side of the religious division do the texts, because of
their practical emphasis, generally include background theory of humours or
complexions, semiology, or reference to the condition of the patient. This is a
relatively simple kind of medical writing.
Differences as well as commonalities must be registered, however. To
bring them out I shall focus the comparison on just two texts. They are the
short Greek piece (c. 2,750 words) headed Therapeutic medical treatments by
various doctors set in order according to the defined procedure of the xenon,
T for brevity, and the hospital redaction of Sabur (S), which bears the title
(following Kahls translation), The formulary of Sabur according to the copy
of the ,Adudi hospital, [being] a synopsis of Saburs dispensatory on the composition of drugs, [in] sixteen chapters. This recension (to recall) survives
uniquely in a thirteenth-century manuscript but was dated by Degen and Ullmann to the mid-eleventh century (presumably before the flood that partially
destroyed the hospital in 1045).41 T survives in ten manuscripts dating from
the thirteenth to the sixteenth centuries but Bennett, following Litavrin, is inclined to date its compilation to around the mid-eleventh century also.42 The
two texts are thus more or less contemporary.
The most immediate difference between them lies in their organisation,
which is characteristic of their respective medical cultures and the hospital
texts of each. T is arranged in the traditional head to foot order, at least for
its first 46 remedies, after which a miscellaneous supplement seems to follow:43
41
R. Degen-M. Ullmann, Zum Dispensatorium des Sabur ibn Sahl, WO VII
(1973/1974) 241-258; flood: Pormann and Savage-Smith, o.c. 98; the hospital was not
rebuilt for another 23 years.
42
Bennett, o.c. ch. 6, with App. IV; for dating of the compilation as represented in the earliest manuscript see G. Litavrin, Malade et medecin `
a Byzance,
XIe-XIVe siècles: remarques sur le cod. plut. VII 19 de la Bibliothèque de Lorenzo
de Medici `
a Florence, in Evelyne Patlagean (ed.), Maladie et societe `
a Byzance, Paris
1993, 97-101: 98.
43
Adapted from Bennett, o.c. table 6.3 (p. 249).
160
Head
HORDEN
29
30
1 For a sharp pain in the head
2 For a pain on one side of the head or 31
32
face
3 For a pain in the head and one side of 33
34
the head
4 For a feverishly hot head
5 Decoction for use in cases of giddiness
6 Head lotion
Eyes
7 An eye remedy
8 For trachoma of an eye
9 For cases of itching affecting the eyes
10 For a discharge from the eyes
Nose
11 For a nosebleed
12 For dryness of the nose
Mouth
13
14
15
16
17
A mouthwash
For ulceration of the mouth
Toothpaste
For haemoptysis
For pain in the tongue
Intestines
18 A laxative epithem
19 Affections of the bowels
20 For induration of the intestines
21 Affections of the liver, and pleurisy
22 For the liver
23 Poultices for sufferers from pleurisy
24 For a liver affection
25 For stomach pain
26 A plaster [strictly an epithem] for
stomach pain and every affection arising from there
27 For affections of the spleen
28 An epithem for the spleen
About embrocations for the spleen

For all internal pain and poisoning
Kidney affections
A fomentation for kidney affections
Diuretics
An enema for sciatica
Skin lesions
35 For sores on the privy parts
36 For cases of an external sore on the
private parts
37 For every kind of induration
38 Counter-irritant plasters
39 An ointment for injuries
40 A medicinal ointment
41 For an excrescence of the flesh
42 A method of fomentation of the flesh
43 For a haemorrhage
Feet
44 For swollen feet
45 For gout
46 For a callosity
[end of head to foot organization]
Fever
47 For shivering fits
Emetics
48 Emetics
Inflammations
49 For external haemorrhoids
50 For buboes and the plague
Purges
51 Galens purgative pills
52 Purgative for those suffering from the
dropsy
53 Prescription for purging of phlegm
Ears
54 For the mucous discharge which some
call glyky, whatever its origin (mastoiditis)

55 Ear remedies for blockage of the ears
Throat and lungs
56 Affections of the uvula; for cases of the
enlargement of the uvula
57 Gargles
58 Cough remedies
59 The Linctus
60 Remedies for shortness of breath
161
63 For those who do not keep down their

food
64 A desiccative compound efficacious for
a flux from the head
Miscellaneous lesions
65 For a scald from hot water
66 For pruritis
St Gregorys Salt
67 A salt prepared by Gregory the Theologian

Skin lesion
61 For a suppuration (spreading) from Antidotes
head to hands or feet or any other limb
68 On the great decoction of the xenon
Digestion
69 The great decoction of Athanasios
62 A dry powder for the stomach to be 70 Antidotal drugs
taken with wine
S by contrast, in common with most aqrabadhinat, is arranged predominantly by kind of drug:44

1
2
3
4
5
6
7
8
9
Pastilles
Lohochs
Beverages and robs
Oils
Cataplasms
Enemas
Powders
Collyria
Liniments
10
11
12
13
14
15
16
Stomachics
Hierata
Decoctions and pills
Preserves
Preparing and testing theriac
Treating teeth and gums
Uses and properties of animal parts
Addenda: 49 compounds
Note that only ch. 15 on teeth and gums is topically headed. Note
also that madness is not represented, despite the emphasis on its treatment
in many medieval Islamic hospitals.45 We should also register the unsystem44
45
112-135.
Adapted from Kahl, Recension of the ,Adudi Hospital cit. 9.

M. Dols, Majnun: The madman in medieval Islamic society, Oxford 1992,
162
HORDEN
atic addenda, comparable in that respect to T, but here (rather oddly for a
hospital) mostly concerned with sexual hygiene or cosmetics.46
The structure of the remedies also differs between the two texts. Remedies in T have a short purposive title followed by a list of possible ingredients,
mostly un-quantified, and with few directions for administration. T is a piece
of writing that, if really to be used clinically, allows for, indeed depends upon,
the experience of the doctor in deciding quantity and ingredients according
to availability and need. In contrast, as Kahl summarizes, S proceeds fairly
regularly by name and/or category of drug, range of application, ingredients
with doses, instructions for combining ingredients, and directions for use. One
could speculate that this would have worked better in a larger hospital with
a staff of assistants, not necessarily medically trained, who might dispense
remedies under a physicians supervision. When it opened, the ,Adudi hospital reportedly had 25 doctors, including oculists, surgeons and bone-setters.
The Byzantine texts, on the other hand, are more redolent of small hospitals
that were each essentially one-person outfits.47
The number of remedies marks a further difference between T and S:
over 100 in T where a single title may encompass a number of remedies
but 292 in S, albeit a text at least four times as long.48 Gastrointestinal
problems predominate in both, not surprisingly: 21%, Kahl estimates, in S;
25% in T, according to Bennett. Eye diseases rate surprisingly few mentions:
4.4% (Kahl), 5.7% (Bennett, using the fullest manuscript witness to T). Some
ingredients are common to both: barley, celery, cinnamon, ginger, gum Arabic,
honey, Indian spikenard, mastic, pepper, pomegranate, rose saffron, vinegar,
water, and wine. But the wider range in S reflects the openness of the Islamic
world to inter alia Indian and Chinese imports.49 S employs 411 simples
while T uses 212 (including some ingredients in alternative guises, e.g. arsenic
and orpiment). Most of Ss preparations are compounds. (Remedy 9 in ch.
1 of S invokes 17 ingredients for a pastille). By contrast the majority (80) of
46
Kahls defence (Recension of the ,Adudi Hospital cit. at n. 350) of this supplements basic integrity with the main compilation, contra Degen and Ullmann (Zum
Dispensatorium des Sabur ibn Sahl) takes no account of that oddity.
47
Pormann-Savage-Smith, o.c. 98.
48
I am especially grateful to David Bennett here for his help in compiling the
statistics of T.
49
Kahl, Recension of the ,Adudi hospital cit. 10f.
163
the remedies in T contain between one and four ingredients; 23 recipes have
between 5 and 8; and only the remaining 6 have more than 8 ingredients.
In T, according to Bennett, 74% of the ingredients are vegetable, compared to 67.4% in S, which as well as vegetable and mineral remedies has its
special section (ch. 16) on medicine from animal parts. With that in mind,
compare their respective remedies for buboes:
Take fermented dough three parts; borax, salt, pigeon droppings,
and cock droppings one part in each. [This is] ground, strained,
kneaded with olive oil, and used [applied?].
From the outset, take a drink of natural brine with warm (water);
when it has begun to take effect, boil a measure of lentils with
water until it has been well infused, put the juice of the lentils
into a linen cloth, and strain it, squeezing it well. [Now] put [the
decoction] in another vessel and add to it honey and Indian nard
and boil again. Give it as a drink to the patient since it expels
every unhealthy state of the humours. Another remedy: there is
a large-leafed plant that grows by rivers; give its well-steeped root
as a drink, and boil as well as its [edible residue to eat].
Prejudice might suggest that the (still palatable to modern tastes) herbal/vegetable remedy is Islamic, the (stereotypically medieval) animal-based one
Byzantine; the longer text Arabic, the shorter Greek. As it happens, the
converse is true.50
VI. Comparison of this sort could be extended indefinitely. The Greek
text, for example, has no equivalent of the Arabic section (redacted for a
supposedly secular hospital) of hierata, remedies manifesting divine power.
Yet, in general, no one who wants to assert the medical superiority of early
Islamic hospitals over early Byzantine ones superiority defined in terms of
access to the full range of Galenic medical learning is going to be shaken
by what the foregoing juxtaposition (not just the pair of bubo remedies but
the whole exercise of comparison) has suggested. Still, the Greek material has
much to be said in its favour, too: it is clinically practical. We omit a whole
dimension of medieval hospital history, and a whole dimension of the history
50
Ibid. 63, 165; Bennett, o.c. App. IV, 63f.
164
HORDEN
of iatrosophia too, if we do not attend to it, and seek to place it in the widest
comparative context.51
51
Peregrine Horden
[email protected]
A final-forward looking comparison: differences of scale and sophistication

between Greek and Arabic material pale when both are juxtaposed to the 1,035 detailed recipes in the vernacular Ricettario of the great hospital of Santa Maria Nuova,
Florence, compiled in 1515. See J. Henderson, The Renaissance hospital: healing the
body and healing the soul, New Haven-London 2006, 297-335.
Cancerous cells, Neanderthal DNA and the tradition of Byzantine

medicine. Textual criticism in philology and genomics
Genome analysis and stemmatics may at first sight seem very different
fields of research. Upon a closer look, however, a number of similarities become
apparent. For instance, both disciplines are engaged in reading and interpreting text, looking for structural patterns and similar content; moreover, they
catalogue and classify content and aim to reconstruct lineages. They use very
similar techniques, comparing segments of genome and written text to identify
insertions, deletions, modifications and errors. Despite the undeniably similar methodology, genome analysis is traditionally regarded as an exact science
yielding clear results through a computerized comparison of material, whereas
stemmatics is often associated with philological interpretation. However, as
will become apparent in this paper, both phylogenetics and stemmatics are in
fact faced with very similar challenges. There might be obvious differences in
terminology and the data used, but in the following we will show that genome
analysis and stemmatics share a surprising number of key ideas in analyzing
their respective texts. These similarities will make it possible for us to highlight several examples of how both disciplines can profit from collaborating
with each other.1
The general outline of this article was jointly drawn up by Florian

Markowetz and Barbara Zipser. The section on philology was written by BZ, and
the following section on genetics by FM. BZs work draws on research carried out
during a Wellcome Trust funded project on Construction and Use of Byzantine Medical Texts.
166
MARKOWETZ/ZIPSER
Philology and textual criticism

From the philological side, we are going to look at three topics. First,
identical manuscript copies of the same text, second, a medical manual that
was subject to several consecutive revisions, and finally an editor who improved
the quality of a text. These findings are being presented here because they are
incompatible with the standard methodology of stemmatics.
To begin with, we would like to introduce stemmatics to those members
of our audience who are not from the philological field. In antiquity and the
Middle Ages, books were passed on in handwritten form, one book being copied
from the other. Only a very small, and often arbitrary selection of the total
number of books that were once in existence survives to this day, providing
us with random samples of their history. Most of the texts contained in these
manuscripts survive in more than one copy. With the help of stemmatics it
is possible to reconstruct the history of a given text. For this, samples from
all extant copies of a text are collated, i.e. compared to each other. If two
manuscripts contain the same mistake, they have the same source or one is the
source of the other. The same process is then repeated to identify subgroups.
In the end, the researcher will draw a stemma, or family tree, of the lineage.
In the majority of cases, it is rather straightforward to establish the
history of a text. Sometimes however, the findings are inconclusive, and it
appears that a manuscript contains mainly readings from one group of the
transmission, and some selected variants from another. In these instances, the
scribe used one copy as a basis for the main text, and another (or several)
to correct readings that appeared corrupt. This scenario is commonly termed
contamination.
The process of establishing a stemma might be best illustrated with an
analysis of the early history of Leo medicus Synopsis, a therapeutic manual
written by a professor for a former student.2 It can be proven beyond any
doubt that all extant copies of the text descend from an early tenth century
original, the Par. suppl. gr. 446, which I call S. It is a very expensive and
precious book. The pages consist of high quality parchment, the words are
2
Leo says in the preface of the text (see the first lines of 89 Ermerins) that
he wrote the book for his former student George. There is no further evidence on
the identity of either Leo or George. For an edition of the text see Anecdota medica
Graeca, e codicibus MSS. Expromsit F.Z. Ermerins, Lugduni Batavorum 1840 and
Anecdota nova, ed. J.F. Boissonade, Paris 1844, 366-370.
Textual criticism in philology and genomics
167
very neatly written and a number of chapter headings are highlighted in bright
colours. Yet some fundamental flaw must have occurred in the production of
the codex: the ink used for these ornaments later proved to be corrosive, which
led to holes in the parchment, which in turn deleted text on the other side of
the respective page. All other manuscript copies have gaps in the text where
S is damaged, which proves that it is their source. It is not often the case that
we are able to determine the source of the entire transmission of a text with
such certainty.3
The first topic to be discussed in this paper are two of the manuscripts
descending from S, the fifteenth century Par. gr. 2671 (P), and the sixteenth
century Moscow GIM Sinod. gr. 441 (M). Compared to each other, these
witnesses contain an almost identical text. This was shown in a collation of
pages 129,10-29 to 147,2 of the Ermerins edition.4 This is highly unusual.
Under any other circumstances, one would expect a passage of this length
to contain a plethora of mistakes. Therefore, to validate these results even
further, random samples from other parts of the text were collated, which
yielded the same result.5
Clearly, one of these manuscripts must have been written by an extremely diligent scribe who did not make any mistakes, and had it not been
for yet another case of physical damage, we would not be able to reconstruct
the lineage: in P, a sheet of paper is torn out,6 and M has a corresponding
gap in the text.7 This proves that P is the source of M.
To a textual critic whose sole interest is the reconstruction of a lost
original, a flawless copy such as M might seem irrelevant, as it would not
For a general discussion of the transmission see B. Zipser, Uberlegungen zum Text der des Leo medicus, in A. Hornung-C. Jackel-W.
Schubert (edd.), Studia Humanitatis ac Litterarum Trifolio Heidelbergensi dedicata.
Festschrift f
ur Eckhard Christmann, Wilfried Edelmaier und Rudolf Kettemann,
Frankfurt a.M. 2004, 393-399.
4
The Ermerins edition presents Greek text and a Latin translation on facing
pages. Therefore, just under nine pages of text in an average layout were collated.
5
BZ eventually stopped collating due to constraints of time as M was only
accessible on a microfilm at the IRHT in Paris.
6
This page would have been between f. 215 and 216 in the modern numbering.
7
Ermerins edition is based on P, and contains the lacuna caused by the
missing sheet of paper on page 299,28. This gap can be found on the bottom of f. 7r
in M. See Zipser, Uberlegungen

cit. 396 n. 10.
3
168
MARKOWETZ/ZIPSER
have any bearing on the quality of the text. It is, however, of importance
for all those researchers who use findings derived from stemmatics, which is a
purely abstract analysis of the text, and link them to actual historical events
involving the carriers of this text, namely the books. For instance, if the owner
of a book is known, or if it appears very unlikely that a specific book ever left
a certain location, the exact dependencies between manuscripts become very
relevant indeed. Dismissing the possibility of a flawless copy can easily lead
to wrong conclusions.
The second topic to be discussed is the history of a medical work that
evolved in a series of consecutive revisions, Theophanes Chrysobalantes De
curatione.8 Even though this work was extremely popular in the Byzantine era,
as the unusually large number of manuscript copies suggest, we do not have any
direct information on the setting in which the book came into existence other
than the little the text itself reveals. According to its title, it was commissioned
by Emperor Constantine VII Porphyrogennetos. This suggests that it dates to
the early or mid tenth century and was somehow connected to the Byzantine
court. In the first few lines of the preface, Theophanes addresses the emperor
and describes the rationale behind his work:
, ,
[. . .]
I gave my best that the epitome of the overview of medical theories which your holiness, the emperor by gods grace, had ordered,
covered the entire medical art as far as it was possible while being
concise and clear at the same time [. . .]
From these lines, it appears that the original title of the work was epitome of the overview of medical theories, which is as puzzling in the Greek
original as it is in English translation. Is the term epitome used to describe a
8
For an edition of the text see: Theophanis Nonni Epitome de curatione

morborum Graece ac Latine, ope codicum manuscriptorum recensuit notasque adiecit
I.O.S. Bernard, Gothae 1794-1795. A description of the manuscript witnesses can be
found in: J.A.M. Sonderkamp, Untersuchungen zur Uberlieferung

der Schriften des
Theophanes Chrysobalantes (sog. Theophanes Nonnos), Bonn 1987.
169
genre, or does this imply that Theophanes epitomized an existing work which
was called overview of the medical theories?9
We lack any evidence for such a hypothetical predecessor text Theophanes could have summarized. We do, on the other hand, know of at least
two texts, one of earlier10 and one of later date,11 which are similar to De
curatione, but the exact dependencies between these three remain opaque.12
This proves that there were once more texts in existence than are known to us
today, and that the history of the text starts earlier than the tenth century,
even though we are unable to reconstruct the picture in its entirety.
Some parts of said picture are on the other hand much clearer. De
curatione proved to be very popular, and it soon developed into versions;13
the same applied to one of the related texts, John the Physicians Therapeutics. These developments required extensive philological efforts together with
a profound knowledge of medicine, as well as access to high quality sources.
Evidently, to the scholars involved in these projects, the main objective was to
update and emend texts, rather than preserving them in their original form.
9
The wording of this preface reminds of the corresponding part of Oribasius collectiones medicae. Here, Oribasius also refers to his work as both
overview and epitome. However, these terms are used in equal right to describe his treatise. They do not, for instance, designate an epitome of the overview.
For the remainder of the text, Oribasius is not the main source of De curatione.
10
This is Leos Synopsis discussed above. See Zipser, Uberlegungen

cit. on
the chronology and B. Zipser, Zu Aufbau und Quellen der , in T.
F
ogen (ed.), Antike Fachtexte. Ancient technical texts, Berlin-New York 2005, 107115, especially n. 4, along with Sonderkamp, o.c. 135 and 153f.
11
A therapeutic manual attributed to a certain John the Physician coincides
partially with De curatione; see B. Zipser, John the Physicians Therapeutics. A
medical Handbook in vernacular Greek, Leiden-Boston 2009, 10f., and in particular n.
23.
12
These three treatises share a larger amount of content. To complicate matters, some of this content can be traced back to other sources.
13
This is already evident from Bernards late eighteenth century edition, even
though it is mainly based on only five manuscripts, a fraction of what is known today.
(Readers should be cautioned that sigla are not used consistently.) For instance,
based on all the manuscripts available today, chapter 74 Bernard is transmitted in
three manuscripts, the Monac. gr. 362, the Laur. ant. 101 and the Bucarest Bibl. Acad.
R.S. Rom. gr. 617. The differences in the arrangement of the chapters is evident from
the pinax reproduced in Bernard XXII.
170
MARKOWETZ/ZIPSER
In the context of medieval medicine, this is not an uncommon occurrence.14

Sometimes even translations from foreign languages were added to supplement
existing corpora.15
The obvious problem for a stemmatic analysis is here that we are unable
to determine a point of origin. Our starting point is a tenth century work,
which is at least in part based on an earlier text. Moreover, we are unable
to explain the exact dependencies between De curatione and a later text. All
evidence points to the fact that we are dealing with a number of very similar
texts which were developed by editors.
This leads us to the third and last topic to be discussed in this section,
an editor who improved the quality of a text. At some point between the
last decade of the thirteenth century and the fifteenth century, an unknown
person went through one of the medical manuals discussed above, John the
Physicians Therapeutics.16 The specific version of the text he worked on has a
rather rough appearance. This is not only caused by the fact that it is written
in a vernacular dialect; the syntax sounds monotonous, and the vocabulary is
often basic. Here are samples from both versions (the source of the corrected
version is transmitted in three manuscripts, M, A and B):17
14
This phenomenon has been widely discussed in the field of Latin medicine.
For more details and bibliography see Fischers contribution in this volume. The
most comprehensive analysis of the Byzantine material can be found in D. Bennett,
Xenonika. Medical texts associated with Xenones in the late Byzantine period, Diss.
London 2003.
15
See for instance B. Zipser, Die Therapeutica des Alexander Trallianus:
Ein medizinisches Handbuch und seine Uberlieferung,

in R.-M. Piccione-M. Perkams
(edd.), Selecta Colligere II. Beitr
age zur Methodik des Sammelns von Texten in der
Sp
atantike und in Byzanz, Alessandria 2005, 211-234.
16
For a chronology see Zipser, John the Physician cit. 24-26. The text dates
to the last decade of the thirteenth century, and this particular version is transmitted
in two manuscripts, the Vind. med. gr. 33 of the fifteenth, and the Vind. med. gr. 43
of the sixteenth century. Both were acquired in Constantinople.
17
These are the Monac. gr. 288, the Par. gr. 2236 and the Par. gr. 2224.
171
Source: ( , omitted in A)
( ) .18
Corrected version: .19
Source:
( ) ( ) (omitted
in A) ( )
.20
Corrected version:

.21
Certainly, these interventions improved the quality of the work.22 To a textual
critic, such a scenario causes obvious methodological problems, as there is no
assumption in stemmatics that a scribe or editor could improve a text beyond
the quality of the original. He may guess single readings correctly, which would
be termed a conjecture, but overall he would not be able to outperform the
author.
Overall, the three topics discussed above lead us to the following conclusions: stemmatics are based on two main assumptions, first, that a text was
authored by a person at a specific time, and second, that the quality of this
text deteriorated in the process of copying. As the samples discussed above
show, these assumptions do not necessarily apply, as authorship and unity
18
This transcription is based on 3 (177, 6-8 Zipser).

Vind. med. gr. 33 f. 5v.
20
12 (180, 10-12 Zipser).
21
Vind. med. gr. 33 f. 6v.
22
The differences between source and corrected version are difficult to capture
in English translation. Predicates have been turned into participles. It may be best
illustrated with the following simplified example: take A and clean it and chop it and
mix it with B has been replaced with take A, and having cleaned and chopped it,
mix it with B. The corrected version is closer to the classical idiom, which was at this
time still used by the intellectual elite. However, it must be stated that the corrected
version still contains a large proportion of vernacular words. Moreover, the participles
which are used in the corrected version can also be found in the vernacular. Thus,
even though the interventions of the editor can be termed classicising, they mainly
affect the style of the text.
19
172
MARKOWETZ/ZIPSER
of text may be uncertain. Moreover, it could very well be the case that the
quality of a text remained the same in the process of copying, or that it in fact
improved.
The book of life and its history
Unlike physics, biology does not have a strong mathematical theory to
explain and predict observed phenomena. This may be one of the reasons
why biology is very rich in metaphors. The Tree of Life connects all forms
of life on earth. Conrad Waddington famously compared the development of
cell types and tissues to marbles rolling down a grooved slope, the so called
epigenetic landscape.23 And inside every single cell the nucleus contains an
organisms genome, the Book of Life written in the language of DNA. Similar
to a text written in a human language, DNA transfers i n f o r m a t i o n, it
can be t r a n s c r i b e d into a different form (RNA instead of DNA) and
it can be t r a n s l a t e d (into proteins). The idea that the genome can be
r e a d and e d i t e d pervades all molecular biology and forms one of the
most powerful and suggestive metaphors of modern science.
Errors in the book of life. The step from a biological molecule to a written
text seems large, but it is actually quite easy to make. DNA is built of four
nucleotides, the bases Adenine, Cytosine, Guanine and Thymine, which can
form long chains. In higher organisms, two chains together form the strands of
a double-helix. The text of the book of life is a linear abstraction of the threedimensional structure of DNA. The first step is to unwind the double-helix into
two parallel linear strands. Next, we notice that the pairings between them are
not random: Adenine only binds Thymine and Cytosine only binds Guanine.
Thus, the strands are complementary; if I know one, I can reconstruct the
other. This allows us to concentrate on a single strand. If we now, in a final
step, abbreviate each nucleotide of this strand by its first letter, we have a
linear sequence of As, Cs, Gs and Ts the text of the book of life. For
example, the beginning of the gene sequence of FOXP2, which is important
for the development of speech in humans and only found in an altered form in
apes, looks like this:24
23
24
C.H. Waddington, The Strategy of the Genes, London 1957.

Ensembl, Transcript: FOXP2-001 (ENST00000350908).
173
CTGATTTTGTGTACGATTGTCCACGGACGCCAAAACAATCACAGAG
CTGCTTGATTTGTTTTAATTACCAGCACAAAATGCCATCAGTCTGG
GACGTGATCGGGCAGAGGTGTACTCACA...
It is not visible in this very short example, but the genomic text is highly
structured. For example, genes have precise start and stop positions. Around
the genes short sequence patterns indicate the positions where proteins can
bind to the DNA to turn genes on or off, allowing cells to react to different
environments. Thus, even though the text is the same at all times in all cells
of an organism, its meaning is context-specific: a liver cell and a brain cell
both contain exactly the same genome, but in different tissues different parts
of the text are read out. The basic process in genetics is the copying of the
genetic DNA material, which happens every time cells divide. Mistakes in
copying the genomic text in a single cell can have terrible consequences for the
whole organism. For example, a copy mistake that activates an oncogene or
de-activates a tumor suppressor gene can often lead to cancer.
Cancer is a distorted version of our self.25 It took until the middle of the
20th century to realize that cancer does not invade the body from the outside
but is often derived directly from the tissue in which it was first discovered.
Both the normal growth of an organism and the abnormal growth of cancer,
can be traced back to the genome. Observations like these strengthen the book
of life metaphor and make DNA the center of interest in molecular biology.
Compared to a normal cell, the genomes of cancer cells often look completely chaotic: some parts are missing, other parts are repeated many times,
and the order of the text can be completely lost. Think of a vandalized collection of books in a foreign language that has been poorly set together again
by a novice librarian. Currently several international projects26 are under
way to catalogue as many genomic changes as possible in as many cancers as
possible. The hope is that this catalogue will then point us to the drivers
of cancer, the genomic aberrations that cause a cell to become a cancer cell.
This is a difficult task, because not all aberrations are causative; some are
25
This point is powerfully made in Siddhartha Mukherjees Pulitzer-price winning book The Emperor of All Maladies, New York 2011.
26
The two major examples are the International Cancer Genome Consortium
(ICGC) and The Cancer Genome Atlas (TCGA), which gather genomic and other
molecular data on hundreds of patients for many cancer types. In 2011, the first
available data sets described the genomic landscapes of brain and ovarian cancers.
174
MARKOWETZ/ZIPSER
just passengers, which develop because cancer in general has a much lower
genomic stability than the more tightly regulated normal cells. To add to this
complexity, there is not always a single driver for each type of cancer, instead
different (combinations of) aberrations can cause very similar cancers. Untangling this complexity requires large sample sizes: cancer genome projects
involve sequencing hundreds and thousands of genomes. This makes the cancer genome projects a much larger effort than the Human Genome Project27 ,
which took more than a decade to succeed. The hope is that in this mass of
data the drivers of cancer will stand out as very frequently mutated genes,
while the random mutations in the passengers are much less frequent. Success
in these large-scale international projects will depend largely on technological
advances and meticulous book-keeping, less on scientific inspiration and eureka moments.
Evolution and the book of life: textual criticism of genomic sequences. Cancer exploits natural mechanisms that have developed during
evolution to allow a species to better adapt to its environment. Evolution has
several mechanism to act on the genome. Individual letters (nucleotides in
the DNA) can be mutated and changed. These mutations are called single
nucleotide polymorphisms (SNPs; pronounced snips). Counting the number
of SNPs allows us to infer how related two genomic sequences are: the more
SNPs, the further apart they are. Another evolutionary mechanism is recombination, where a region of the DNA is cut out and joined in at a different
position. One of the effects of recombination can be a change in the number of
copies of a DNA region, which can get lost or amplified. In general all humans
have two copies, but every one of us can naturally have more or less copies in
particular regions. This variation is a very mild form of the chaos raging in a
cancer cell. Finally, an evolutionary mechanism to act on the genome is the
incorporation of genes from one population into another, so called gene flow.
When not perverted by cancer, all these genomic changes happen naturally
and contribute to the variety of body sizes, hair and eye colours and the rest
of the phenotypic diversity we see in humans.
The mutations observable in the course of evolution were beneficial and
27
The Human Genome Project launched in 1989, initially funded by the US

National Institutes of Health (NIH) and Department of Energy but quickly joined by
international collaborators. It produced a first draft in 2000 and an almost complete
sequence in 2003.
175
induced an evolutionary advantage. In some sense, they are improvements

of the text. Just a few mutations can have an tremendous impact on the
morphology and behaviour of related species. For example, the genomes of
modern humans, chimpanzees and Neandertals are almost identical. Humans
and chimpanzees differ by only 1.2 percent of all base-pairs in gene sequences,
and Neandertals are even closer.
DNA changes trace the evolutionary history of species. We dont know
all the details of the Tree of Life but collections of genomic sequences allow researchers to estimate how many years ago two species were still one and
when the split between them happened. In 2010 an international research consortium led by Svante P
a
abo published a draft sequence of the Neanderthal
sequence.28 Their goal was to identify genomic features that distinguish modern humans from other hominin forms by comparing the human genome to
the genomes of Neanderthals and apes. In humans and chimpanzees the DNA
could be sampled from current populations. In Neanderthals, however, the
DNA needed to be retrieved from archaeological and paleontological remains,
making it a challenge to prove the authenticity of DNA sequences retrieved
from ancient specimens. Still, obtaining a Neanderthal genome is worth the
effort, because genomic analyses can reach much further back in history than
archeological analyses based on excavated bones and artefacts can. For example, the earliest known remains of anatomically modern humans are 195,000
years old. From genomic data, however, it could be estimated that the split between ancestral human and Neanderthal populations happened 370,000 years
ago, extending the horizon by 175,000 years.29 So far, only around 100 genes
surprisingly few have been identified that have contributed to the evolution
of modern humans since the split. Less surprisingly, several of theses genes
are involved in cognitive function and others in bone structure. Understanding
the functions of these genes better can have a bearing on what it means to be
human or at least not Neanderthal.
28
R. Green et al., A Draft Sequence of the Neandertal Genome, Science CCCXXVIII 5979 (2010) 710-722.
29
J.P. Noonan et al., Sequencing and Analysis of Neanderthal Genomic DNA,
Science CCCXIV 5802 (2006) 1113-1118 on-line. The method compares the Neanderthal genome to the human and Chimpanzee genomes and takes the variability
observed in human populations into account. A computational model then simulates
an evolutionary process and statistics is used to derive the age of the Neanderthalhuman split that is most likely under the model.
176
MARKOWETZ/ZIPSER
Genomic data also allows us to address questions that can be very hard
to solve from archeological data alone. Did Neanderthals interbreed with
anatomically modern humans? Substantial controversy surrounds this question: Morphological features of present-day humans and early anatomically
modern human fossils have been interpreted as evidence both for and against
genetic exchange between Neanderthals and human ancestors. However, Neanderthals mating with humans must have left traces in both the Neanderthal
and human genomes. P
a
abo and his colleagues compared three Neanderthal
genomes with five genomes of present-day humans from different parts of the
world, including Africa, Asia and Europe. If Neanderthals are more closely
related to present-day humans in certain parts of the world than in others, this
would suggest that Neanderthals exchanged parts of their genome with the ancestors of these groups. P
a
abo and his colleagues found that Neanderthals are
equally close to Europeans and East Asians, but they are significantly closer
to non-Africans than to Africans. This can be explained by Neanderthals exchanging genes with the ancestors of non-Africans.
Statistics versus the facts. Svante Paabos study elegant and diligent
as it may be will not be the last word on human prehistory. The dominance
of the genome in phylogenetic studies is not uncontested. DNA evidence can
be contradicted by other, more classical sources of data, like fossil records.
In the question of Neanderthal-human interbreeding, the genomic evidence
points to a period of 100,000-60,000 years ago in the Middle East. However,
the archeological record for an overlap in the populations at this time and
place is very sparse. Archeologists and paleo-anthropologists favour a scenario
in which interbreeding happened in Europe, possibly from 44,000 years ago
(when modern humans first entered Europe) to 30,000 years ago (when the
last Neanderthals fell extinct).
What becomes visible here is the gap between two types of approaches to
answer the same scientific questions: geneticists and computational biologists
prefer to analyze DNA, because it directly shows the traces of evolution. To
them, DNA is an overwhelmingly superior source of information. It can be
analyzed much more precisely than the blur of phenotype in ancient specimens.
Archeologists, on the other hand, prefer to draw conclusions from the fossil
record. While most of them agree that geneticists have been making valuable
contributions to human prehistory, they feel that genetic analysis rely too
177
heavily on computational methods and mathematical statistics. With their

scientific training, archeologists find genetic arguments very hard to follow
and far less solid, informative and convincing than hard archeological fact.
In doubting statistics archeologist are not completely wrong. Reconstructing genomic phylogenies relies on the statistical analysis of genomic data
and like all statistical analyses crucially depends on mathematical assumptions that can sometimes be disputed. No statistical method recognizes the
truth; they only make estimates based on the likelihood of observed events and
a quantification of the uncertainty in the data. As a result, statistical estimates
from genomic data are seldom so clear cut and convincing that they automatically beat conclusions drawn from other sources of information, like for example fossil records. Often, it is a judgement call for the researcher whether or not
to trust the statistical results of a phylogenetic genome analysis. The same is,
of course, true for archeological facts. Facts are theory-bound they need to
be interpreted and put into a bigger context, else they are useless. This requires
no less experience and judgement than the analysis of statistical results. Finding one type of evidence more convincing than the other is a matter of education and training, not of the intrinsic scientific value of different types of data.
The story about Neanderthal-human mating
is not the only example of this gap in scientific backgrounds and approaches. The science writer Carl Zimmer has collected several others in his essay The Genome: An
Outsiders view.30 In some of them, the fossil evidence clearly contradicts the statistical claims; in others, fossil evidence later
validated claims from DNA data that had
Figure 2: Elephant (Illustrator:
seemed far-fetched at first. Carl Zimmer
Caroline Buckley)
compares the clash between different research styles with the fistfight between the
blind men, who had touched different parts of an elephant and could not agree
on an interpretation of what they had felt.
30
ogy
C. Zimmer, The Genome: An Outsiders view, PLoS Computational BiolII 12 (2006) on-line.
178
MARKOWETZ/ZIPSER
A common theme emerging from these examples is the need for integrated approaches that combine different styles of research. In some areas,
like cancer biology, interdisciplinary training and research are already well established. Other areas, like paleo-anthropology, will need to follow if they want
to take advantage of the full range of data and insights. Maybe the abundance
of metaphors in biology actually turns out to be a strength when it comes to
interdisciplinary research, because metaphors if used carefully can bridge
the gaps between disciplines.
Conclusion
Two main findings have come to light. In the section on genetics,
it became clear that statistical estimates can be of value for other related
disciplines, such as paleontology and archeology, even though these are based
on mathematical models rather than an analysis of actual physical remains.
Ideally, these estimates would be followed up in an interdisciplinary setting to
discuss and validate results yielded with different methods. The same may be
said about stemmatics, which often allow us to reconstruct historical data.31
In the section on stemmatics, it emerged that the standard methodology
could not always be applied to Byzantine medical works. This was mainly
caused by the fact that these texts were often updated and developed by
competent editors, whereas stemmatics are built on the assumption that the
quality of a text deteriorates in the course of its transmission. Here, our
comparative approach yielded a number of important findings that help to
overcome these methodological obstacles.
In genetic analysis, changes can be for the better and for the worse,
or neutral a mutation can lead to evolution and therefore progress, or to
a disease such as cancer; an exchange between two groups is usually termed
gene flow (rather than contamination, its counterpart in the philological
field). Any analysis focusses on observed events, regardless of their quality.
This is also expressed in the way logical dependencies between these events
31
For instance, a specific type of mistake can be linked to a handwriting that

was only in use up to the late ninth century AD. Leos Synopsis, which cannot not be
dated with certainty, was shown to contain such a mistake, which proves that the text
was written prior to this date. See Zipser, Uberlegungen

cit. For other approaches to
date the text see Leo the Physician. Epitome on the nature of man, ed. R. Renehan,
CMG X/4, Berlin 1969.
179
are displayed in scholarly publications. The family tree of a species or a single

gene may, for instance, also merge with another family tree.
This more flexible methodology can easily be implemented in stemmatic
research. In addition, we see a strong potential for the implementation of
specialized databases in the philological field. Geneticists have over the past
decade developed solutions to handle complex intricacies in large data sets.
With the help of these applications it would for instance be possible to display
compilations of medical texts, so called iatrosophia, and their sources in various
stages of their history, which is not possible in the conventional output format
in the form of a book.
Cancer Research UK - CRI
Robinson Way, Cambridge CB2 0RE,
UK
Florian Markowetz
[email protected]
Barbara Zipser
[email protected]

02 Zipser Medical Byzantine Books

Uploaded by

Copyright:

Available Formats

02 Zipser Medical Byzantine Books

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

02 Zipser Medical Byzantine Books

Uploaded by

Copyright:

Available Formats

EIKASMOS

Quaderni Bolognesi di Filologia Classica

Medical books in the

BECh Biblioth`eque de lEcole

Prefatory note: the uses of medical manuscripts

The conference from which it derives was indeed conceived as a collective

Prefatory Note: The Uses of Medical Manuscripts

written out, no copy made, automatically without some calculation, however

See V. Nutton, Ancient Medicine, Abingdon-New York 2004, ch. 1.

their unwieldy size, the perceived redundancy of some of their information, or

V. Nutton, Galen in Byzantium, in M. Gr

Prefatory Note: The Uses of Medical Manuscripts

Does it suggest display here too self-validation, image-making? What does

The second version of John the Physician is in effect a commentary on the

Byzantine medicine, genres, and the ravages of time

O. Temkin, Byzantine medicine: tradition and empiricism, DOP XVI

S. Geroulanos, Byzantine surgery, in M. Gr

Byzantine medicine, genres, and the ravages of time

medical literature of earlier Antiquity. Medical historians frequent concern

appears to be the introduction to a commentary on Galens Art of medicine,

Byzantine medicine, genres, and the ravages of time

G. Sabbah-P. Mudry, La medicine de Celse. Aspects historiques, scien

M.H. Marganne, Lophthalmologie dans lEgypte

Byzantine medicine, genres, and the ravages of time

Samama, o.c. 396f.

in den Papyri aus Agypten, Vienna 2007, 23-34.

S. Vogt, Drugs and pharmacology, in R.J. Hankinson (ed.), The Cambridge

Byzantine medicine, genres, and the ravages of time

one of his heirs, a very close friend of mine, voluntarily handed

was no great loss but so also was my treatise on the composition

A.E. Hanson, Text and context in papyrus catechisms on afflictions of the

Byzantine medicine, genres, and the ravages of time

Secondly, students of Latin medical writings from Late Antiquity are

For iatromathematics, derided by Galen, see I. Andorlini, Un anonimo del

Disease and where to treat it: a Byzantine vade mecum

A.K. Orlandos-L. Branouses, , Athens 1973,

V. Nutton, Ancient Medicine, London 2004, 53-102.

Disease and where to treat it

This type of disease aetiology carried very positive connotations for

In the cases of demonic-induced diseases (whether the demons were conjured

Disease and where to treat it

J. Duy, Byzantine medicine in the sixth and seventh centuries: aspects of

G.B. Ferngren, Medicine and Health Care in Early Christianity, Baltimore

Disease and where to treat it

The issue of plagues is touched upon, but a different question (66) is

Disease and where to treat it

occult remedies and amulets in gout.26 In another passage he writes that

Alexander von Tralles. Therapeutika, ed. T. Puschmann, I-II, Vienna 1878.

of physicians.30 There is no need, I think to refer to such examples; they

On physician at the bedside of emperors: see A. Kazhdan, The image of the

Disease and where to treat it

Scholars await with great anticipation the publication of the findings of

This is clearly the case in the xenon of Panteleemon in Constantineople.

Disease and where to treat it

period.44 Patients were healed through incubation or immersion in water, but

See A.-M. Talbot, Pilgrimage to Healing Shrines: The Evidence of Miracle

Disease and where to treat it

Two Latin Pre-Salernitan medical manuals, the Liber passionalis

This article forms part of my research undertaken in the context of the

CorsettiK.-D. Fischer, pref. de M.D. Grmek, Saint-Etienne

O. Temkin, Byzantine medicine: tradition and empiricism, DOP XVI

occult remedies and amulets in gout.26 In another passage he writes that

La science medicale antique. Nouveaux regards. Etudes