Situating Arabic Science: Locality versus Essence

Author(s): A. I. Sabra
Source: Isis, Vol. 87, No. 4 (Dec., 1996), pp. 654-670
Published by: The University of Chicago Press on behalf of The History of Science Society
Stable URL: http://www.jstor.org/stable/235197 .
Accessed: 07/04/2014 05:08

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press and The History of Science Society are collaborating with JSTOR to digitize,
preserve and extend access to Isis.

http://www.jstor.org

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
HISTORYOF SCIENCE SOCIETY
DISTINGUISHED LECTURE

Situating Arabic Science

Locality versus Essence

By A. I. Sabra*

LOCALITY AS A FOCUS OF HISTORIOGRAPHY

I trustthat no one would wish to contest the propositionthat all history is local history-
whetherthe locality is that of a short episode or of a long story. All history is local, and
the history of science is no exception. There can be no history of science that is not
concernedwith a localized episode or a sequenceof such episodes. Philosophersof science,
or some of them, may restrictthemselves to analyzing the formal or logical and timeless
structureof a piece of scientific thought;or they may substitutean emphasison growthas
a basic featureof scientific inquirythat requiresfor its logical elucidationthe additionof
a temporaldimension that retains an abstractcharacter.But historiansof science have a
differenttask. For while they can ignore the cognitive core of scientific practice only at
the cost of forfeiting their claim to a distinctive problematicand a distinctive discipline,
they are especially concernedwith science as a process that takes place in actual time or
science as a series of phenomenathat, owing to their special characterof chronological
and geographicallocality, we call "historical"-this "special character"being due to the
fact thatthe phenomenain questionare not merely in space and time but events associated
with, and indeed producedby, individuals acting in what we broadly call "culturalset-
tings."The thesis outlinedin these few abstractsentences is but a generalizationof a weak
version of the familiarcontextualistthesis in scientific historiography;or, to put it another

* Departmentof the History of Science, HarvardUniversity, Science Center235, Cambridge,Massachusetts

02138.
This lecture was delivered at the annualmeeting of the History of Science Society, Minneapolis,Minnesota,
28 October 1995. I am greatly honoredto have been invited to give this talk. I have since made a few additions
and changes, some of them in response to comments, questions, or criticisms by the anonymous referees, to
whom my thanksare due.

Isis, 1996, 87: 654-670

C) 1996 by The History of Science Society. All rights reserved.
0021-1753/96/8401-0001$01.00

654

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 655

way, contextualismis but an obvious consequence of the simple, undeniablefact of the

local characterof all events, includinghistoricalevents. Since a historicalevent is where
and when and how it is, inseparablytied to all the circumstancesthat combine to define
it for us as historians, then, to be genuinely historical, all history of science must be
contextual,because all historicalevents are local.
I said "weak version" because I do not wish to subscribe to a stronger,reductionist
version that seems to me to misinterpretthe local characterof cognitive expression and
behaviorby appearingto deprivethem of objective import.At the risk of being much too
brief, let me indicatewhat I mean with just a sentence or two before I get on to my subject
proper. I am persuaded,as I hope you are, that locality is not a propertyof a piece of
scientific thought-say, the Pythagoreantheorem. In other words, the Pythagoreantheo-
rem, not being an event in the spatiotemporalworld, has no space-time coordinates.But
recognizing it as a theoremin a given geometricalsystem with certaindefinablefeatures
was the achievementof someone or other who certainlyhad such coordinates.From this
it would seem to follow that the historicalstudy of scientific thoughtis, strictlyspeaking,
not concerned with the thought content of science as such (that would be to engage in
science, logic, or philosophical analysis), but with the historical occurrence of the rec-
ognition of thought.I am aware that all of this has been said before and in various ways
by philosophersand historiansto serve ratherdifferent agendas. So, to avoid misunder-
standing,I will just addthe observationthatrecognitionof thoughtalways involves, among
other things, a cognitive context that itself is location-bound,in the sense of being partof
a problem situationthat depends on the existing state of knowledge at a given time and
place. And this means that engaging in science, logic, and philosophicalanalysis may all
be involved in the historical study of scientific thought.'
My purposehere is to try to illustratethe advantagesof a strict adherenceto the axiom
of locality in situatingthe traditionof Arabic science with referenceboth to the place that
this traditionoccupies in the general history of science and to its place in the civilization
where it emerged and developed. My discussion will thus be concernedwith two contexts
that, though distinct from one another,are, as I shall suggest, intimatelyconnected with
each other. Obviously what I shall present to you can only be the bare sketch of what
might be described as a frameworkfor research, but, I hope, a frameworkthat is not
irrelevantto other scientific traditionsand that may even propose a correctionto other
historiographiesthat seem to pay little or no attentionto the interculturaltransmissionof
scientific knowledge.
Let us begin with an apparentlyneutraland innocent definitionof Arabic, or what may
also be called Islamic, science in terms of location in space and time: the termArabic (or
Islamic) science denotes the scientific activities of individualswho lived in a region that
roughly extended chronologically from the eighth century A.D. to the beginning of the
modernera, and geographicallyfrom the IberianPeninsulaand North Africa to the Indus
valley and from southernArabiato the CaspianSea-that is, the region covered for most
of thatperiodby what we call Islamic civilization, and in which the resultsof the activities
referredto were for the most part expressed in the Arabic language. We need not be
concernedover the refinementsthat obviously need to be introducedeven into this seem-
ingly neutraldefinition.But what about the term scientific in it? What does it mean, and

I For ideas underlyingthe theoreticalstructureof this essay I am indebted to the writings of Gottlob Frege
(on the distinctionbetween thoughtand recognitionof thought),KarlPopper(on methodologicalindividualism,
situationallogic, interactionof Worlds 1, 2, and 3), and Alfred North Whitehead(all about events).

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
656 SITUATINGARABIC SCIENCE

can it be regardedas in any way "innocent"?To me it seems clear that the only correct
answerto this last question must be an unequivocalNo. Science and scientificare our own
terms and they express our own concepts (which, by the way, does not mean thatthey are
sharplydefinedor unproblematic);and,therefore,the studyof any past intellectualactivity
can be relevantto what we call "historyof science" only to the extent that such an activity
can be shown to help us understandthe modes of thought and expression and behavior
thatwe have come to associatewith the word science. This is not anachronism,presentism,
whiggism, or any of the other objectionableisms, but a consequence of the fact that we
who are writing the history also have a location of our own that defines our perspective
and, hence, the questions we pose from our vantage point and the terms in which these
questions are framed.Nor should this admissionto a definitepoint of view discourageor
detractfrom investigatingpast modes of thoughtand expression and behaviorunderother
categoriesdeemed suitablefor elucidatingthese modes "in their own terms,"as the phrase
goes. But, without aiming to replace other approachesthat put the emphasis on certain
concerns of sociology or anthropologyor culturalhistory, our historiographyof science
will always change as a function of our changingposition, being ourselvesforeverlocated
at the end point of the process that is continually shaping and reshaping what we call
"science."And so I am led to combine a self-evident propositionwith anotherthat seems
no more, and no less, than a corollary of it: that all history of science is local, and no
history of science can ever be neutral.
The characterof Arabic science, its strengthsandfailings, the course of its development,
and its ultimatefate have all been variouslyexplained in terms of language as a matrixof
thoughtand expression, of religion as an inexorableshapingforce, of naturalaptitudesor
inclinations of a certain race or inherentmentality, or as one inevitable expression of a
world cultureof which Islamic civilization was a late embodiment.A perceivedemphasis
on algebra in the Arabic traditionhas been attributedto certain features of Semitic lan-
guages that make these languages or their native users prone to "algebraization,"as op-
posed to Greek "geometrization."The persistentattemptsof Islamic astronomersto con-
struct kinematic models primarily designed to save the principles and the logical
consistency of Ptolemaic astronomyhave been seen as a sign of poverty of imagination
or of the tendencyof the "Semiticmind"towardthings it can easily perceiveby the senses.
Islamic religion has been cited both as the origin and source of vigor of medieval Islamic
science and as the majorcause of its final demise. And the "spiritof culture,"in this case
a Magianculturealreadyat work in "so-called"late antiquity,has been invoked to account
for every aspect of Islamic civilization, including its scientific products.2
It is not difficult to expose the weaknesses from which such explanationssuffer. One

2 Roger Arnaldezand Louis Massignon, "ArabicScience," in History of Science: Ancient and Medieval Sci-

ence, from the Beginnings to 1450, ed. Ren6 Taton, trans.A. J. Pomerans(New York: Basic, 1963), Vol. 3, Ch.
2, pp. 385-421, esp. pp. 402-405; Massignon, "R6flexionssur la structureprimitive de l'analyse grammaticale
en arabe,"Arabica, 1954, 1:3-16; Pierre Duhem, Le systeme du monde, Vol. 2 (Paris: Vrin, 1954), pp. 117-
179, esp. pp. 117-119 (on "le r6alismedes arabes");Duhem, To Save the Phenomena:An Essay on the Idea of
Physical Theoryfrom Plato to Galileo, trans.EdmundDolan and ChaninahMaschler (Chicago/London:Univ.
Chicago Press, 1969), pp. 25-35, esp. p. 26; BernardCarrade Vaux, "Les spheres c6lestes selon Nasir-Eddln
Attfisi," in Paul Tannery,Recherches sur 1'histoirede l'astronomie ancienne (Paris: Gauthier-Villars,1983),
App. 6, pp. 337-361; G. E. von Grunebaum,"MuslimWorld View and Muslim Science," in Islam: Essays in
the Nature and Growth of a Cultural Tradition(London: Routledge & Kegan Paul, 1964), pp. 111-126; von
Grunebaum,"Islamand Hellenism,"in Islam and Medieval Hellenism: Social and CulturalPerspectives (Lon-
don: Variorum, 1976), Ch. 1; J. J. Saunders,"The Problem of Islamic Decadence,"Journal of WorldHistory,
1963, 7:701-720; and Oswald Spengler, The Decline of the West, trans. Charles Francis Atkinson, 2 vols.
(London:Allen & Unwin, 1959), esp. Vol. 1, pp. 71-73, 207-216 (and see the index under"ArabianCulture").

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 657

can refer, for example, to the considerableand highly successful efforts of Islamic math-
ematiciansin the fields of geometry and trigonometry.One can relate the theoreticalpro-
gramof Islamic astronomersto the work of Ptolemy himself and to earlierideals of Greek
astronomy. One can point out the great complexity of the relationshipbetween science
and religion throughoutIslamic history and in variouspartsof the Islamic world. And one
can easily show the vacuousness of theories born of the spirit-of-cultureapproach.And,
in fairness to those who have advanced explanationsof these sorts, it must be said that
they tend to be poorly informed(or worse) aboutArabic science and, in many cases, about
Islamic civilization-a fact that, unfortunately,does not seem to have discouragedtheir
influence on minds that seek ready-madeand perhapscomfortingexplanations.
What is wrong with these explanations,and others like them, is not their consideration
of language, religion, and cultureas factors in the formationof a scientific enterprisethat
consciously adoptedearliertraditionswith markedlydifferentlanguagesand religious and
cultural values but, rather,their essentialist character,which has tended to prejudiceor
obstruct historical research. Now locality-that is, the characterof being local-is an
ineradicableor, if you like, an essential propertyof all historical events, but the actual
where and when and how of any such events are happeningscreated by human effort.
With the sure perceptiveness of a true historian, Richard Southern once described the
process of acquisitionand adaptationof Greek learningin Islam as "the most astonishing
event in the history of thought."3The event is astonishingbecause it strikes us as unex-
pected, and the best way I know to explain the unexpectedin history, insofar as it can be
explainedat all, is to try to understandit, not in termsof essences or spiritsor inevitabilities,
but as the outcome of choices by individualsand groups respondingto their situationsas
they perceived and experiencedthem. Let me illustrate.

THE INTERSECTION OF ISLAMISM, ARABISM, AND HELLENISM IN

NINTH-CENTURY BAGHDAD

The powerful drive that eventually led to the transferof the bulk of Greek science and
philosophy (as well as elements of the scientific thoughtof Indiaand Persia)to Islam was
launchedas a massive translationeffort thattook place in the context of empire and under
the patronageof the confident Abbasid court in Baghdad. Translationsinto Arabic had
been made earlier, and these had been preceded in the Middle East by translationsfrom
the Greek into Syriac and Persian,but it was the Abbasids who mounted a concentrated
translationeffort soon after they came to power in the middle of the eighth century and
who furtherorganizedand intensifiedtheir supportduringthe ninth century.Under their
predecessors, the Umayyads who ruled from Damascus (661-750), the Islamic empire
alreadyencompassedlarge areas-including Egypt, Syria, and Persia-that had come un-
der the influence of Hellenism from the time of Alexander;and before the ninth century
was over Islamic rule had reachedKashmirin the east and Khwarazmto the north.In the
early Abbasid period the higher administrationof the court itself was in the hands of
cultivatedPersianswho had gained much favor and influencewith the Abbasidrulersand
whose intellectual interests inclined them to various forms of secular learning and to a
rationalizingapproachfor understandingmattersof religious belief. Some of these Persian
officials acted as translators,especially from Persian, and in general they constitutedan

I R. W. Southern, Western Views of Islam in the Middle Ages (1962; Cambridge,Mass./London:Harvard

Univ. Press, 1978), pp. 8-9. These lectures were delivered at Harvardin April 1961.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
658 SITUATINGARABIC SCIENCE

important,politically influentialpart of Baghdad's intellectual elites. Two other groups

within the empire (and concentratedmainly in Syria, Iraq, and Persia) had maintaineda
long-establishedtraditionof Hellenized Syriac learning.One consisted of Christianphy-
sicians and Christiantheologians, who continuedto pursue their interests in Greek logic
and philosophy in scatteredmonastic schools; the otherwas the pagan Sbians of Harran,
in northernMesopotamia,an ancient Semitic group whose astralreligion connectedthem
to Hellenistic astrology and astronomyand to Hermeticism. It was from these two last
groupsthat the Abbasids were able to recruitthe scholarswho carriedout the translations
of Greek medical, philosophical, and mathematicalworks into Arabic, either from pre-
existing Syriac versions or directly from the Greek.
The survival of these pockets of Hellenic learningduringthe first centuriesof Islamic
rule in the Middle East andAsia, althoughscatteredand limited at firstin scope andappeal,
ensured a certain continuity with the classical tradition-a continuity that was largely
lacking, for example, in the case of the "Renaissanceof the Twelfth Century,"when
Europeanscholars first had to journey to the edges of Western Christendomto acquire
Arabicand Greeklearningfrom acrossthe borderswith Islam andByzantium.In the earlier
Middle Easternepisode, this continuitymeantthe immediateavailabilityof texts, in Greek
or Syriac versions, and of translatorsalready conversantwith these languages and with
Greekthoughtitself in a numberof scientific,medical, and philosophicaldisciplines.And
although much additionalGreek materialwas later to be brought over the borderswith
Byzantium,the continuitywith the Greco-Syriactraditionhelps to explain the high level
of competence, even sophistication,that characterizedscientific writings in Arabic from
an early period that overlappedthe translationmovement.
One might then say, and with muchjustification,thatthe stage was set, at a certainplace
and time, for the translationmovementthatquickly acquiredunprecedentedproportions-
unprecedentednot only in the Middle East but in the world at large. But in orderto explain
the momentum,scope, and multiple dimensions of that movement, it is necessary to go
beyond the availabilityof favorableconditions, and even beyond the importantconsider-
ation of practicalexpectations that must have loomed large at least in the minds of the
Muslim patrons.Islamic religion had introduceda new ideology with sweeping and uni-
versalist claims. Already during the swift expansion of Islamic conquests, that ideology
had come into directcontactwith a large varietyof creeds (Jewish, Christian,Zoroastrian,
Mazdian,Manichaean,etc.) with which it inevitablycollided and againstwhich it had not
merely to defend but-much more importantly-to define itself, often in terms borrowed
from its opponents. The result was a huge intellectual ferment, centered especially in
multiculturalIraq,to which the movements of Islamic theology, philosophy, and science
owed their birth.
Or, should we not rathersay, more accurately,thatthe creationof these fields of thought
representedthe responses of so many groups of individualsto aspects of what was, in the
context of religion and politics and power and the variety of competingways to salvation,
a very complex andlive intellectualatmosphere?Withregardto the creationof the tradition
of science and philosophy in Islam, I am temptedto borrow an obsolete term, aspecting,
in orderto refer to the way in which individualsin a given cultureaspect anotherculture
as they direct their gaze to the other from their own location. Aspecting in this sense is
conditioned both by the interests, aspirations,and aptitudesof the aspecting individuals
and by the accessible aspects of the viewed culture,that is to say, the aspects thathappen
to be disclosed to them by the accidents of history or by their further,determinedeffort.
Thus, for example, through the Sabians of Harran,Muslim thinkers were able to view

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 659

facets of Hellenistic thought that might not have been available to them by way of the
Christiantheologians,who had alreadymadetheirown choices fromtheirown standpoints.
And, as has been plausibly suggested, the absence of Greek literatureand Greek histori-
ographyfrom the translatedcorpusmay be attributableto a lack of acquaintanceor serious
intereston the partof the Christiantranslators.4In a similarway, the twelfth-andthirteenth-
century Arabo-Latintranslationsin northernSpain were understandablylimited to the
types and the levels of the learning that was currentlyavailable in Al-Andalus, with all
the features of that learning that had undoubtedlybeen shaped by a combinationof cir-
cumstancespeculiarto Al-Andalus.5
The scholarsof eighth- and ninth-centuryIraqlooked east to Persia and Indiaand west
to Greece and especially to Alexandria.Both looks deeply affected the characterof Arabic
science, especially in mathematicsand astronomy,in both of which we find combinations
of identifiableelements from the East and the West. But it was the westward gaze that
proved most enticing and, as it turnedout, most consequential.Some years ago I used the
term appropriationto characterizethe attitudeof Muslim scholarsand patronswho made
it their business to get hold of and make their own what they called "the sciences of the
ancients,"an expression that clearly revealed a sense of distance in time between them-
selves, as "the modems" (al-muta'akhkhirun),and the appropriatedlegacy of "the an-
cients" (al-mutaqaddimun),even as the appropriatorsset about gaining possession of the
ancientlegacies with greatenergy.6Withoutaiming here to unfold the full meaningof that
sense of distance(which has frequentlybeen misinterpretedand misused in modem schol-
arship),let me indicatebrieflyhow it was understoodand evaluatedby some of those who
promotedor participatedin the appropriationdrive of the ninth century.
I will not expand,not even briefly, on the role of the Abbasidcaliph al-Ma'miin,whose
contributionsas a patronof astronomicalresearchand as the one who turnedthe library
of Greekphilosophicalsciences collected by his immediatepredecessorsinto an organized
center of translationare well known. But I must emphasize the significance of his deep
involvementin a theological disputethatpromptedhim to initiatethe "inquisition"against
the conservative opponents of the Mu'tazilite school of kalam (or "theology")that he
favored. It is naturalto think of a connection between al-Ma'miin's supportof the Mu'-
tazilite emphasison the role of reasonin elucidatingreligious dogma and his championing

4Paul Kunitzsch, "The Two Movements of Translationinto and from Arabic and Their Importancein the
History of Thought"(in Arabic), Zeitschriftfur Geschichte der Arabisch-IslamischenWissenschaften,1987/
1988, 4:93-105.
5Guy Beaujouan, "The Transformationof the Quadrivium,"in Renaissance and Renewal in the Twelfth
Century,ed. Robert L. Benson and Giles Constable (Cambridge,Mass.: HarvardUniv. Press, 1982), pp. 463-
487, on p. 465: "Having found its principal Arabic sources in Spain, medieval Western science obviously
reflected, at the start,the choices of the Hispano-Moslemculture.The lack of interestin abstractmathematics,
the predominanceof astronomyand astrologyin the early translations,the relativelylate date of the Arabo-Latin
versions of Aristotle's naturalphilosophy, the failure to use importantworks by easternArabic scholars:all are
explained by the evolution of Arabic science in the Iberian peninsula, with its peculiarities of history and
geography,its particularistpride within the Islamic world, its conditioningby the oppressive dominationof the
Malikitefakihs."In additionto the referencescited by Beaujouansee A. I. Sabra,"TheAndalusianRevolt against
Ptolemaic Astronomy: Averroes and al-Bitriujl,"in Transformationand Tradition in the Sciences: Essays in
Honor of L Bernard Cohen, ed. EverettMendelsohn(Cambridge:CambridgeUniv. Press, 1984), pp. 133-153,
rpt. in Sabra,Optics,Astronomy,and Logic (Aldershot:Variorum,1994); Julio Sams6, Islamic Astronomyand
Medieval Spain (Aldershot:Variorum,1994), esp. Chs. 1 and 12; and R. S. Avi-Yonah, "Ptolemyvs. al-Biit4iji:
A Study of Scientific Decision-Makingin the Middle Ages," Archives Internationalesd'Histoire des Sciences,
1985, 35:124-147.
6A. I. Sabra,"TheAppropriationand SubsequentNaturalizationof GreekScience in Medieval Islam,"History
of Science, 1987, 25:223-243, rpt. in Sabra,Optics,Astronomy,and Logic.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
660 SITUATINGARABIC SCIENCE

of Greekscience andphilosophy.We areeven given a hintof a contemporaryinterpretation

of such a connection in a story thatcirculatedin tenth-centuryBaghdad,if not earlier.The
story reporteda dreamthat the caliph was alleged to have had, in which none other than
the pagan Aristotle appearedas an instructorto the "overawed"Commanderof the Be-
lievers. In answerto al-Ma'miin'squestion aboutthe natureof "thegood" (al-hasan), the
Greekphilosopheridentifiesits authoritativesource in the firstplace with reason(al- 'aql),
in the second place with "the (religious) law," and in the thirdplace with what is accepted
by "the majority of people" (i.e., the majority of legal scholars). In one version of the
dreamAristotle ends by urging his Muslim studentto upholdthe unity of God (wa'alayka
bi l-tawhid) !7 Affirmationof the unity of God (tawhrd)was the first article in the Mu'ta-
zilite credo, and it was in the name of divine unity, as interpretedby reason ('aql), that
the Mu'tazilamaintainedthe doctrine,which al-Ma'miinandhis two immediatesuccessors
sought to impose on the conservativelegists, thatthe Qur'an-as the speech of God-had
not existed from all eternity but was created. The suggested connection between Greek
rationalismand Islamic Mu'tazilismis impossible to overlook.
Turningnow from the patronal-Ma'miinto leading Muslim intellectualsof the period,
one quickly detects an attitude of openness and gratefulness to the recently imported
wisdom, mingled with a feeling of high optimism and a certain trust in humanismthat
was quite pronounced.The majorArabic prose writer of the period, and one of the most
influentialfigures in the history of Arabic literature,al-Jahiz(d. 869), was also the leader
of a prominentbranch of Mu'tazilism named after him (al-Jahiziyya) and a reader of
Aristotle. In one of his most importantworks, the monumentalBook of Animals, which
owed much to Aristotle and which he dedicatedto the famous wazjr Ibn al-Zayyat(who
served the Abbasid court in the reigns of al-Ma'miin's successors, al-Mu'tasim and al-
Wathiq [833-842, 842-847]), he thanks "the ancients"(i.e., the Greeks) for their consid-
erableintellectualcontributionto the scantypossessions of the people of his time andplace
and culture.He wrote, in a passage that is characteristicfor its utter lack of inhibitionor
ambiguity: "Our share of wisdom would have been much reduced, and our means of
acquiring knowledge weakened, had the ancients not preserved for us their wonderful
wisdom, and their various ways of life, in writings which have revealed what was hidden
from us and opened what was closed to us, therebyallowing us to add their plenty to the
little we have, and to attainwhat we could not reach without them."8
An exact contemporaryof al-Jahiz, and arguablythe single most importantfigure in
this phase of appropriation,the celebratedMuslim philosopher,scientist, and mathemati-
cian al-Kindli(d. ca. 870), was a member of the Arab nobility (his grandfather,we are
told, was amrrof al-Kulfain southernIraq).As a tutorto a son of Caliph al-Mu'tasim,he
was much closer to the Abbasid court than al-Jdhiz. In his work On First Philosophy,
dedicated to al-Mu'tasim, we find strong acknowledgmentof the accomplishmentsof
ancientGreece that is combinedwith the assertionof truthas the universalgood thatmust
be sought out wherever it may be found, in additionto a clear concept of the growth of
knowledge as a process of accumulationthat requiresthe cooperative effort of different
peoples and successive generations.9These were the deep convictions of a true devotee,
I Ibn al-Nadlim,al-Fihrist, ed. Gustav Fliigel (Leipzig, 1871-1872), Vol. 1, p. 243 (also in an edition by Rid.I
Tajaddud[Tehran,1971], pp. 303-304).
Al-Jahiz, Kitab al-Hayawan, ed. 'Abd al-Salam Hariin, 2nd ed., Vol. 1 (Cairo: MaktabatM. al-Babi al-
Halabl, 1965), p. 85.
9 Alfred Ivry, Al-Kindr'sMetaphysics:A Translationof Yaqiib ibn Ishaq al-KindT'sTreatise "On First Phi-
losophy" (Albany: State Univ. New York Press, 1974); Al-Kindl, FTal-sina'a al-'u;ma, ed. 'Azrir Taha al-
Sayyid Ahmad (Cyprus:Dar al-Shabab, 1987); and Al-Kindli,"FThudiudal-ashya wa rusiimiha,"in Rasa'il al-
KindTal-falsafiyya,ed. M. AbiuRida, Vol. 1 (Cairo:Dar al-Fikral- Arabl, 1950), pp. 163-180.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 661

and they carriedwith them an entireHellenic or Hellenistic world view, a distinctconcept

of wisdom in both the theoreticaland practicalsenses (also borrowedfrom the Greeks),
and a sense of mission on the partof the authorto do his utmostto disseminatethe ancient
and especially Greek heritage in his milieu. Both this world view and this concept of
wisdom, as well as the universalistcharacterof al-Kindli'smission, are recognizableHel-
lenistic themes. But we all know betterthan to push the continuitythesis too far-espe-
cially not in the directionof blatantessentialism:to look on al-Kindlias a latter-dayHellene
or Hellenist would be as helpful as the affirmation,found in Spengler's Decline of the
West,that the Pantheonbuilt by Hadrianin the second centurywas the first mosque ever.
Such a view of al-Kindi and other Muslims who sharedhis outlook in the ninth century
would tear them away from their unique position in history and from the role they con-
sciously andfreely chose to assumein the context of theirown culture.Al-Kind7i'smission,
as he understoodit, clearly reveals his adoptionof the humanistictheme, his avowed intent
being, as he said, "the perfecting of our [human]species."10But as one who lived in an
Arabicculture,his immediaterole, as he also conceived it, was to introduceand,he hoped,
to convert his Arabic-readingcontemporaries(ahl lisanina: the people who speak in our
tongue) to the Greek wisdom that had captivatedhim-a task that he actuallyundertook
to achieve by producinga huge numberof Arabic epitomes and adaptations,with supple-
mentary clarificationsand additions when necessary, of a very large number of Greek
disciplines of science and philosophy. This was a preposterouslyoptimistic project for
anyone to envisage; but not only was al-Kindliable to carryit out, whethersinglehandedly
or with the help of others, it proved to be remarkablysuccessful-so successful, in fact,
as to make him truly worthy of the reputationhe quickly gained as one of the foundersof
the Arabic traditionin philosophy and science.
With these three pivotal figures in mind (and there are others that can be broughtinto
considerationalong with them), I am inclined to portraythat crucial phase in the appro-
priation process as the accomplishmentof individualswho experiencedthe intersection,
at a certainplace andtime, of threemajormovementsat work-namely those of Hellenism,
Arabism, and Islamism. By viewing ninth-centuryBaghdad as a point of intersectionin
the mannerI have tried to outline, I hope to renderuseless such questions as whetherthe
scientific traditionthen being establishedwas essentially Arabic, Islamic, or Greek" and
to open the way for empiricalresearchaimed at identifying the actual workings of these
movements as revealed in the writings and recordsof those individualswho experienced
and respondedto them. No doubt there is reason here to celebratethe creative genius of
a moment in the history of civilization, but my overridingaim is to direct attentionto the
complexity and richness of that extraordinarymoment and away from the misleading
"essences."9

THREE LOCI OF SCIENTIFIC ACTIVITY IN ISLAM: THE COURT, THE COLLEGE,

AND THE MOSQUE

Abundantrichness also awaits the empirical investigatorinto the subsequentcourse of

scientificactivityin Islamiccivilization.WhatI havejust describedis the attitudeof certain

10"idh kunnahirasan 'ala tatmiminaw'ina":al-Falsafa al-dla, in Rasitil, ed. AbiuR7da,Vol. 1, p. 103. For
the translationsee Ivry, Al-KindT'sMetaphysics,p. 58. See also Spengler, Decline of the West (cit. n. 2), Vol.
1, p. 211.
" ErnestRenan,L'islamismeet la science, 2nd ed. (Paris:CalmannLdvy, 1883).

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
662 SITUATINGARABIC SCIENCE

individualswho were favorably disposed to the importedknowledge and who played an

active partin bringingaboutwhat laterprovedto be a long-lastingtradition.But, of course,
a few first steps, even significantones, might not have been followed by othersin the same
direction or with the same determinationand vigor. And, indeed, there were other, con-
temporaryindividualsand groups whose markedlydifferentor contraryattitudesand in-
tellectual commitments,then and in laterperiods, did much to shape the course of Arabic
science. As I turn now to later developments,our story acquiresa degree of complexity
that I cannot hope to convey in a lecture. But I shall try to give you a sense of it. It is a
complexity thatfurtherillustratesthe usefulness of the methodologicalconcept of locality,
and it will lead me at the end of my talk to pose the general question of whether,and in
what sense, Arabic science should be investigatedas a single enterprise.The storyunfolds
in at least threedistinctbut by no means isolated loci whose differentstructuresandmodes
of operationand interactionhave yet to be explored from the standpointof our subject.
These loci are the college or institutionof higher learning,the royal or princelycourt,and
the mosque. I shall arbitrarilyignore the hospital as a result of excluding medicine from
my presentaccount.'2
I will begin with the college, and in orderto bring you closer to an unfamiliarsituation
I will startwith two general observationsby way of comparisonwith more familiarepi-
sodes. The first is this: as far as science and philosophy are concerned, the European
Renaissance of the sixteenth century was in part a reaction, which became more pro-
nounced in the seventeenthcentury, against patternsof thought and argumentassociated
with medieval "scholasticism."In Islamic history, events followed the reverse order:the
"renaissance"(if that is the right word) came first, in the ninth and tenth centuries,and a
form of scholasticismfollowed, though not immediatelyand not uniformlyin all partsof
the Muslim world. My second observationpoints to anothercontrastbetween Islam and
medieval Europethat is cruciallyimportantbut more difficultto describebriefly.In Islam,
whether in ninth- and tenth-centuryBaghdad, eleventh-centuryEgypt and central Asia,
twelfth-centurySpain, thirteenth-centuryMaraghain northwesternIran, or fifteenth-cen-
tury Samarkand,the major scientific work associated with the names of those who were
active at those times and places was carried out under the patronage of rulers whose
primaryinterests lay in the practicalbenefits promised by the practitionersof medicine
and astronomyand astrology and applied mathematics.Many of these practitionerswere
also prolific writers on "philosophy,"a mode of thinkingknown by the Arabicizedterm
falsafa and characterizedto a large extent by a mixture of Aristotelianand Neoplatonic
doctrinesand forms of argument-the kind of mixturewe find, for example, in the works
of al-Kindli,al-Fdrdbl,and Avicenna. In those circumstancesscience and "philosophy,"or
falsafa, were secular activities that were practiced,developed, and propagatedas rational
inquiries completely independentof any religious authority-which, of course, did not
prevent the proponentsof this autonomous,self-legitimizing mode of thinking from of-
fering their own rationalistic(i.e., Hellenic) interpretationsof religious doctrinessuch as
revelation or prophecy or providence and of religious institutionssuch as law. After all,
falsafa was an all-embracingworld view that claimed the right to scrutinizeand account
for everythingwithin the sphere of humanexperience, including religious experience. In

12 The only "excuse"for this exclusion is to avoid further

complicatingan alreadycomplex picture.No story
of Arabic/Islamicscience or philosophyis complete withouttakinginto accounttheirrelationto medicalthought,
the effect of medical patronage,the place of science and naturalphilosophy and logic in the institutionsof
medical educationand practice, and the role of Galenic writings as a source of ideas and doctrinesthat shaped
the minds and attitudesof Islamic philosophersand scientists, as well as physicians.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 663

one case among the prominentdevotees offalsafa, that of al-Kindli,a seriouscompromise

was made by renouncingthe Greekdoctrineof the eternityof the world in favor of creatio
ex nihilo. But unlike most of their Christiancounterpartsin medieval Europe, Islamic
philosophers(the self-styledfalasifa) and philosopher-scientistsin that Greek sense were
not "theologians"or members of religious orders. The one major exception, of sorts, is
the twelfth-centuryAndalusianAverroes, who came from a celebratedtraditionalfamily
of Malikitejurists and practicedthe Malikite version of Muslim law as a judge, but who
neverthelessbelieved himself to have inheritedthe mantle of Aristotle. I shall come back
to him later.
Now, Islamic "theology,"or what has come to be known in Westernscholarshipby this
name, followed a differentcourse-with important,indeed far-reachingconsequencesfor
the developmentof both science andfalsafa. It began to make a conspicuous appearance
in the eighth century (the second Islamic century),well before the patronizedtranslation
movement got under way, as the activity of spontaneouslysproutinggroups of Muslim
intellectuals in the urban centers of Basra and Baghdad who immersed themselves in
probingdiscussions (kalam: speech, discussion, argument),obviously driven by their in-
terestin currentreligious and political controversies.They graduallydeveloped somewhat
varied and sometimes seemingly indecisive but sophisticatedand sophisticatedlyargued
doctrinesconcerninga comprehensivearrayof subjectsthatrangedall the way from God
and his relation to man and the world, to questions of epistemology and morality and
political leadership,to subtle and difficult speculationsabout the ultimateconstitutionof
all createdbeing, which they characteristicallyproposedto understandin atomisticterms.
Thefalasifa laterdubbedthese kalampractitionersas religious apologists,therebyseeking
to downgradetheir rivals or, if possible, to circumscribetheir role by subsuming their
enterpriseunderthe authorityof falsafa. I3
There can be no doubt that the early practitionersof kalam, the mutakalliman,were
influencedby a multiplicityof pre-Islamictraditionsin ways thatstill remainmostly veiled
in obscurity.But whateverthe remote sourcesof theirideas, and despitetheirfundamental
concern with the elucidationand critiqueof religious tenets, it is my conviction (which I
share with a few others) that the discourse of the early "school"of the Mu'tazila,the one
favored by al-Ma'mtin,and of the later and subsequentlydominantAsh'arites,represents
an importantturnin the historyof philosophicalthought-one that gave rise to new styles
of thinkingthat seriously challenged the Aristotelianismand Neoplatonism of falsafa by
proposing a thoroughgoingatomism that viewed the world as a creative process. It was
this new philosophy, the "philosophyof the kalam,"as HarryWolfson called it,'4 that, in
the Ash'arite version, later found its way into the colleges of higher education, the so-
called madrasas that ultimately spreadwide and far over the Islamic world as endowed
or charitableinstitutions, having been first introducedon a large scale in the eleventh
century by the Sunnite Saljiiqs in Iraq and Persia as part of a political agenda and in
response to the Ism'Tll-propagandaemanatingfrom FatimidEgypt and Syria.
The madrasas, it should be noted, were first conceived of as primarilyschools of law,

13 A. I. Sabra,"Science and Philosophy in Medieval Islamic Theology: The Evidence of the FourteenthCen-

tury,"Z. Gesch. Arab. Islam. Wiss., 1994, 9:1-42.

14 HarryAustryn Wolfson, The Philosophy of the Kalam (Cambridge,Mass/London: HarvardUniv. Press,

1976). The view of kalam as religious apologetics has been prevalentin modem literaturebut is currentlybeing
revised. See RichardM. Frank,"The Science of Kalam,"Arabic Sciences and Philosophy, 1992, 2:7-37; and
Sabra,"Science and Philosophy in Medieval Islamic Theology."

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
664 SITUATINGARABIC SCIENCE

an emphasis that they retainedthroughouttheir history.'5But, as creationsof privateen-

dowments,they generallyenjoyed a degree of informalitythatallowed for a variablerange
of intellectual pursuits that depended on local circumstancesand the interests of their
professors and their sponsors. Many, perhapsa large number,of the madrasas included
some teachingin arithmetic,algebra,astronomy,and logic as partof the intellectualequip-
ment of the practicingjurist, along with the indispensable disciplines of language and
rhetoric.Kalam,as a study of the "fundamentaltenets of religion"(usul al-dTn),performed
the dual function of supplying a superstructureof theory for the rest of the "religious
sciences" as well as a substitutefor Greek metaphysicsand naturalphilosophy.
Combiningthese two generalobservationsshould now help us to appreciatethe follow-
ing result. The sciences of the Greeks, which were first welcomed in Islam along with
Greek theories of cosmology and epistemology and metaphysics (be they Hellenic or
Hellenistic), eventually came to be confrontedin the madrasasby a homegrownreligious
philosophy thatclaimed to develop viable alternativesto the Greekparadigms.Thatmuch
we can say in light of what we already know. But there is no end to the questions that
have yet to be examined.When and where and in what circumstancesdid thatprocesstake
place? Was it one process or many? How did kalam manage to subduefalsafa, given the
tentative beginnings of the former and the originally strong and full-grown natureof the
latter?How much, if anything,of falsafa doctrinesand forms of argumentwas absorbed
into kalam? How didfalsafa react to the assault of the mutakallimdn,given thatfalsafa
continuedto pursue its activities long after the advent of the madrasas, at least in some
parts of the Muslim world? And-the question of special importancefor the historianof
science-what was the effect of the kalam point of view on the disseminationand devel-
opment of scientific disciplines such as cosmology and astronomy,about which the mu-
takallimtinhad a lot to say as an integralpartof their own world view?
None of these questions can be answereda priori.They are all empiricalquestionsthat
requireempiricalresearch.Some of my colleagues, I am happyto say, are now beginning
to tackle them in earnest.Othersare reluctantto embracethem, being afraidof the possible
danger of diverting too much attention from the vast quantities of scientific texts that
remain to be edited and analyzed. The skeptics have a point, and I share their concern.
But this is not an either/ormatter.As for the argumentthat "we do not yet know enough
to ask the big questions,"my answer is this: it is only by attemptingto formulateappro-
priate questions that can be fruitfully examined in light of what we now know that we
make it possible for others to come up with deeper and more probing questions in the
future.We do not know much (that is for certain),but the day when we know "enough"
will never come. On the otherhand,by altogetherabandoningall programsof full-fledged
historicalresearch,we only tempt others to fill the vacuum with easy and useless essen-
tialist generalizations.
The madrasas were not, therefore, in general a locus where scientific research was
promoted for its own sake, but one in which science was interpretedand judged and

15 The
literatureon the madrasa is growing rapidly, but see especially the wide-rangingstudies of George
Makdisi: "Muslim Institutionsof Learningin Eleventh-CenturyBaghdad,"Bulletin of the School of Oriental
and African Studies, 1961, 24:1-56 (rev. by A. L. Tibawi, ibid., 1962, 25:225-238), and The Rise of Colleges:
Institutionsof Learning in Islam and the West (Edinburgh:EdinburghUniv. Press, 1981). Most of the recent
publicationsare concerned with the Mamlik period (for which there is abundantmaterial),but they have not
yet directed special attentionto the question of the place of science and philosophy in the madrasas. This must
be due in partto the long-held assumptionthat science and philosophy had no place in the madrasas, which is
not quite true, as is now being realized.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 665

presentedto a large group, indeed the vast majority,of educatedMuslims. When we talk
of scientificadvancein Islam,whetherin mathematics,astronomy,or experimentalscience,
we usually have in mind the contributionsof men who carriedout their work outside of
the madrasas with the supportof kingly patrons-men like al-KhwTrizml,al-Khayyami,
Ibn Yulnus,Ibn al-Haytham,al-Blrtinl,al-Tilsi, and al-Shirazl.As always, there were ex-
ceptions-sometimes importantones, especially in the later period-and in some cases
the systems of patronageand madrasa even seem to merge, eitherpartially,or even com-
pletely, as for example underUlugh Beg's initiativein fifteenth-centurySamarkand.Such
exceptions should direct our attentionto the significant overlap among all three loci of
courtpatronage,college, and mosque. But the generalpictureof scientificadvancein Islam
as a patronizedactivity holds.
What we are still far from understandingis how patronageworked in contexts that
obviously differed from one center of activity to another.Let me give an example with
referenceto two or three situationsaboutwhich we know a little more than aboutthe rest.
Consideragainthe pagan sage, Aristotle,who was said to have inspiredCaliphal-Ma'mun
and with whom the Malikite Muslim jurist Averroes identifiedhimself. In both cases we
encounteracceptanceof the authorityof an alien thinkerand of the intellectualvalues he
represented.In both cases a Muslim theological context was involved: Mu'tazilitekalam
in the earlier episode, and the religious ideology of the Almohad dynasty in the later
Andalusianepisode. But the theological commitmentswere differentin the two cases, with
ratherdifferentimplicationsfor the place and authorityof the law, and so were the patterns
of patronage.The differences were, moreover, enhanced by an emphatic self-conscious
Andalusianidentityvis-a-vis the rest of the Islamicworld.It is not surprisingthatAverroes,
as the intellectual who responded most powerfully to this Andalusian situation, should
develop a new theory of religious authority,a totally negative attitudeto kalam,and a new
valuationof his adoptedAristotelianismthat set him against his Peripateticpredecessors
and earlier mathematiciansin the easternpart of the Islamic world. It is, therefore,with
referenceto this special context in all its geographical,political, and intellectualparticu-
laritiesthat we should try to gain a historicalunderstandingnot only of Averroes'rebuttal
to the attack launched by the Easternal-Ghazali against the falasifa, but also of his im-
portantand explicit divergencesfrom fellow-faylasufslike Avicenna or from a recognized
mathematicalauthoritylike Ibn al-Haythamand of his ultimaterejectionof the hitherto-
dominantPtolemaic astronomy.16
Similar,or greater,contrastswith regardto attitudes,patternsof relationsbetweenpatron
and client, and implicationsfor the practiceof science are what we should expect to find
as we turn our attentionto laterperiods. For example, soon afterthe Mongol Ilkhanshad
capturedBaghdad in 1258, thus bringing the Abbasid caliphate to an end, their leader
Hiilagulwas persuadedto establish an observatoryat Maraghain northwesternIran, an
event that markedthe beginningof one of the longer-lastingand importantepisodes in the
history of Arabic science. Most of the scholarswho were soon to be gatheredat Maragha
were Muslims (thereare reportsof one or more Chinese scholars).The man put in charge
of organizingthe new enterprisewas Nasir al-Din al-Tilsi, a Persianfrom Tus with serious
interestsin Shl'ite theology and Avicennan philosophy. (See Figure 1.) At age fifty-five
when he surrenderedhimself to the Mongols upon their captureof the Isma'Ili stronghold
of Alamut, he was alreadyfamous as a scholarand known to the Mongols as a competent
astronomerand astrologer.Two other scholars, both of them Sunnis, were broughtover

16 Sabra,"AndalusianRevolt against Ptolemaic Astronomy"(cit. n. 5).

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
666 SITUATINGARABIC SCIENCE

....F,' ^" .... '.'........I l. .. ;. . f

*. si. .
........

Figure 1. NasTral Din al-Tujs,andcolleagues at the Maraghaobservatory.(MSF1418, early

sixteenthcentury IstanbulUniversityLibrary.)

from Syria.One of them Mu ayyadat Din al-'Urdl,had a reputationas a buildingengineer

and instrument-maker.The other, the mathematicianMuhyi al Di-nal Maghribiat Anda-
lUS17,was captured by the Mongols during their campaign in Synia in 1259 1260. He
managed to save his life only by presentinghimself to his captors as an astrologerwho
could be of use to "the lord of thieearthi,"thiegreat Mongol Khan.17 The Mongol patron
of the astronomicalenterpniseat Maraghawas not Muslim, and his interestin the work of
the observatorywas undoubtedlyastrological.The immediategoal was to producea new

"7The story of al-Maghribi'scaptureis told by Barhebraeus(in Ta nr-kh mukhtasaral-duwal), who heard it

from al-Maghribiin Maragha;see A. I. Sabra, "Simplicius's Proof of Euclid's ParallelsPostulate,"Joumal of
the Warburgand CourtauldInstitutes, 1969, 32:1-24, esp. pp. 13-14, rpt.-in Sabra, Optics, Astronomy,and
Logic (cit. n. 5). On al-'Urdi see George Saliba, ed., TheAstronomicalWorkof Mu 'ayyadal-Dmnal- 'Urdl.:Kitib
al-Hay a (Beirut:Centerfor Arab Unity Studies, 1990), pp. 27-30.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 667

.}..X o~~~ ... .....1

\i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/t
.............
.XtS,*|
.{{.

Figure 2. Giant armiillaryset up in the open air at the sixteenth-century Istanbul observatory. (MS
1404, late sixteenthcentury,IstanbulUniversityLibrary.)

set of astronomicaltables, based on new observations,of the type that had been used by
Arabic astronomers and astrologers for planetary predictions since the time of al-Ma'miin.
The Zj-i Il-Khani, as the new Persian handbook came to be known, was not completed
until 1272, after Hulagu's death. But in the meantime,and for several decades afterward,
the scholars at Maraghaand nearby Tabriz were able to pursue their individualinterests
in theoretical astronomy and in various branches of mathematics. It was in this singular
situation that the aporetic research in planetary theory, which was initiated by Ibn al-
Haythambefore the middle of the eleventh century and had attractedthe attentionof a
few individualscholarsin Asia and Syria, first found a sustainingatmosphere;and it was
from here that ffiis type of research later spread further east, south, and west, where it was
cafried on in different terms or with different emphases as scholars with different com-
mitmentsrespondedto changing contexts. (See Figure 2.)

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
668 SITUATINGARABIC SCIENCE

The phenomenonof the mosque as a significant locus of scientific activity came into
being at about the same time as the establishmentof the Ilkhanidrule in Iran and Iraq,
being associatedwith the ascension to power of the Mamluiksin Egypt and Syria in 1250.
Most of what is now known aboutthis importantphenomenonis due to David King, whose
work over the last twenty years has been responsiblefor puttingthis phenomenonon the
map of Arabic science.18 It was the longest-lastingepisode within the traditionof Arabic/
Islamic science, having continuouslyenduredin some of the major mosques all the way
up to the nineteenth century, and it possessed interestingfeatures that distinguish it in
several ways from courtpatronageand the madrasas, the two otherloci of consequence.19
Though primarilya place of worship, the mosque, from its inception, and as distin-
guished from the madrasa that was sometimes attachedto it, often served as a forum for
propagationand discussion of subjectsrelatedto Arabiclanguage,grammar,and rhetoric,
as well as the vital issues of law, religion, andpolitics. Throughthe introduction,apparently
for the first time underthe Mamluiks,of the office of muwaqqit,the timekeeperin charge
of regulatingthe times of the five daily prayers,a place was createdfor the utilizationof
one form of scientific knowledge in a permanentreligious institution.
Strictly speaking,it would be wrong to consider the muwaqqita "professional"astron-
omer. His institutionalrole in the mosque was not to pursue the goals of astronomyas
these had been defined and elaboratedby Arabic astronomerssince the ninth centurybut,
as is clearly indicated by his status title, to offer reliable guidance to his local Islamic
communitywith regardto definitereligious observances(mainlyprayertimes) as specified
by religious law. This function the muwaqqitnonetheless performedin his capacity as an
expert in what was called "the science of reckoning time" ('jim al-mTqdt)by means of
exact astronomicalcomputations,and this distinguishedhim from the traditionalmu'ezzin
(the man who called for prayer),who relied on traditionalprescriptions.The main task of
the muwaqqitwas thereforeto use the methodsof sphericalastronomyin orderto construct
tables, usually computedfor a certainlocality or latitude,that would enable anyone who
could operate a simple observation instrument(such as an astrolabe or a quadrant)to
determinethe time of day or night from the altitudeof the sun or a star.A muwaqqitmight
also possess the skill to constructsuch instruments.And some distinguishedmuwaqqitsin
the thirteenthand fourteenthcenturiesaccomplishedthe impressivefeat of providinguni-
versal solutions of timekeeping problems (indeed, all problems of spherical astronomy)
for all latitudes. One muwaqqit,the fourteenth-centuryIbn al-Shatir(d. ca. 1375), who
was attachedto the Umayyad mosque in Damascus, venturedinto the area of theoretical
astronomyto produce the most complete solution to the equant problem, which Ibn al-
Haythamhad forcefully pointed out as a threatto the principles of Ptolemaic astronomy
and which was diligently pursuedby mathematicalastronomersin the thirteenthcentury.
These were all accomplishmentsthat must be regardedas accomplishmentsin astronomy
proper,regardlessof their institutionalsetting. And the same can be said of other equally
impressive investigationsaimed at determiningthe directionof Muslim prayer.As in the
case of timekeeping, these investigations also culminated in universal solutions for all
latitudes.
And yet it is noticeable, as King has pointed out, that the legal scholarsand interpreters

18 David King, "TheAstronomyof the Mamluks,"Isis, 1983, 74:531-555; and King, Astronomyin the Service

of Islam (Aldershot:Variorum,1993).
'9 The distinction has to be maintaineddespite occasional or even frequent overlappings, as, for example,
when a local rulerwas responsiblefor the appointmentof a favored professorin a madrasa.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
 A. I. SABRA 669

of the religious law continuedto apply the simpler considerations,based on observations

of twilight and horizonphenomena(rising and setting)or of shadowlengths, leaving alone
the sophisticatedmathematicaltreatises,which they "generallyconsideredto be too com-
plicated or even completely irrelevant."20 This is not really surprising(mathematicalpre-
cision need not be considered a prerequisiteof religious piety!), but it does renderprob-
lematic the concept of the muwaqqit'smathematicalwork as "serviceto religion."On the
other hand, the theoreticaltriumphof Ibn al-Shatirin planetarytheory does not seem to
have elicited serious attentionfrom othercontemporarymuwaqqits,and this appearsto be
the resultof the fact thattheirinstitutionalposition did not demandor encouragetheoretical
ventures for their own sake. Such paradoxesmay simply reflect our present, inadequate
knowledge of the circumstancesin which a new institutionalstructurebroughttogether
mathematicaland religious interests.But whateverthe correctunderstandingof these par-
adoxes might be, it would be gratuitousto regard the work of the muwaqqitin aiding
religious ritualas constituting"the essence of Islamic science" (as King puts it)21 or even
as the most revealing aspect of scientific activity in Islam. To propose such a view may
have the advantage of highlighting the uniqueness to Islamic civilization of a certain
emphasis on some programsof astronomicalresearch.But the disadvantagesof this pro-
posal are also glaringly conspicuous. It disregardsthe full extent of scientific researchin
Islam, and it ignores the characteristiccomplexity of Islamic civilization itself by neglect-
ing the variety of religious attitudeswith regardto the status, the function, and the value
of scientific knowledge. And it might appearto equate "Islamic science" with narrowly
circumscribedprogramsthat largely developed within the confines of an institutionwith
no commitmentto "science" as such, and this alone would tend to obstructor prejudice
vital questions about scientific practice in Islam by identifying a single locus of activity
with a widespreadand extremely complex phenomenon.And, of course, it would again
open the way into the trapof essentialism.
To come finally and very briefly to the generalquestionformulatedearlier:Was Arabic
science one or many? A similar question has sometimes been asked with reference to
Islamic art, where manifest varieties of styles and functions are displayed in the artifacts
and architecturalmonumentsof the vast Islamic world. As far as science is concerned,it
seems to me that importantconsiderationslead us to say that we have to do with a single,
unitarytradition.These are considerationsof language, which-for science and philoso-
phy-was for the most part one language (Arabic), and of Islamic religion as an ever-
present point of reference though not always a point of departure,in additionto consid-
erationsof the dominanceof dynasticrules over large regions for extendedperiodsof time
and the remarkableease of movetnientand communicationall throughthe Muslimworld-
a featureitself connected to religion and law and language. And, with regardto commu-
nication of learning, we must also keep in mind that crucial Chinese invention, paper,
which took the whole Islamic world by stormfrom the momentof its appropriationin the
middle of the eighth century.
One example will have to suffice as an illustrationof what I mean by these remarks.
Writing in fourteenth-centuryDamascus, Ibn al-Shatir linked his studies in theoretical
astronomyto those of earlier mathematicians,four of whom had worked in thirteenth-
century Maragha,one in eleventh-centuryEgypt, two in twelfth-centurySpain, and two

20
David King, "Science in the Service of Religion: The Case of Islam,"in Astronomyin the Service of Islam
(cit. n. 18), Ch. 1, p. 246.
21
Ibid., p. 245.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions
670 SITUATINGARABIC SCIENCE

who originatedin thirteenth-centurySyria and North Africa. All had written in Arabic,
the languagein which Ibn al-Shatiralso wrote. The example is representativeof situations
that existed before and after Ibn al-Shatir,though there were cases in which Persianand
Turkishwere the languages of composition, especially in later times. But once we direct
our attentionto situations,as distinguishedfrom tradition,ourpictureand ourproblematic
will change with every case, as we turnfrom one set of circumstancesto anotherin which
individualchoices are made with referenceto specific problemsproposedby specific con-
texts. Not, of course, that traditionand individualresponse are separable:on the contrary,
the formerprovides an inseparablepartof the intellectualcontext in which the othermust
take place. When I startedto write this talk I hoped to be able to illustrateand perhaps
also to characterizein some generaltermsthe interplayof traditionandindividualresponse
with referenceto one or two episodes of Arabic science. In the end I am forced to leave
that subjectfor anothertime and place.

This content downloaded from 87.161.248.86 on Mon, 7 Apr 2014 05:08:31 AM

All use subject to JSTOR Terms and Conditions