Académique Documents
Professionnel Documents
Culture Documents
Edited by
Hans Daiber
Anna Akasoy
Emilie Savage-Smith
VOLUME LXXX
Relational Syllogisms
and the History of Arabic Logic
9001900
By
Khaled El-Rouayheb
LEIDEN BOSTON
2010
Cover illustration: A portrait (of uncertain authenticity) of the Ottoman scholar
Ismail Gelenbev (d. 1791), and a page from Musta f Kutb Srzs commentary on
Gelenbevs manual of logic al-Burhn f ilm al-mzn (printed in Istanbul, ca. 1830?).
The page from Srzs commentary is courtesy of Widener Library, Harvard College
Library.
El-Rouayheb, Khaled.
Relational syllogisms and the history of Arabic logic, 900-1900 / by Khaled
El-Rouayheb.
p. cm. -- (Islamic philosophy, theology, and science. Texts and studies)
Includes bibliographical references and index.
ISBN 978-90-04-18319-3 (hardback : alk. paper) 1. Syllogism--History. 2. Islamic
philosophy--History. 3. Inference. I. Title. II. Series.
BC21.S9E4 2010
166.0953--dc22
2010011774
ISSN 0169-8729
ISBN 978 90 04 18319 3
All rights reserved. No part of this publication may be reproduced, translated, stored
in a retrieval system, or transmitted in any form or by any means, electronic, mechan-
ical, photocopying, recording or otherwise, without prior written permission from the
publisher.
Introduction ............................................................................................ 1
1
See D.D. Merrill, Augustus De Morgan and the Logic of Relations (Dordrecht:
Kluwer Academic Publishers, 1990), 79. De Morgans remarks are often mentioned in
general introductions to logic or the history of logic. See, for example, I.M. Bochenski,
A History of Formal Logic (New York: Chelsea Publishing Company, 1970), 95; I. Copi,
Symbolic Logic, 5th Edition (New York: Macmillan, 1979), 117.
2
On Galen, see J. Barnes, A Third Sort of Syllogism: Galen and the Logic of
Relations, in Modern Thinkers and Ancient Thinkers, ed. R.W. Sharples (Boulder:
Westview Press, 1991). On Ockham and Gersonides, see P. Thom, Termini Obliqui
and the Logic of Relations, Archiv fr Geschichte der Philosophie 59 (1977): 143155;
2 introduction
C.H. Manekin, trans., The Logic of Gersonides (Dordrecht: Kluwer Academic Pub-
lishers, 1992), esp. 2526. On Jungius and Caramuel, see E.J. Ashworth, Joachim
Jungius (15871657) and the Logic of Relations, Archiv fr Geschichte der Philosophie
49 (1967): 7285; P. Dvorak, Relational Logic of Juan Caramuel, in Handbook of the
History of Logic, Volume 2: Medieval and Renaissance Logic, ed. D.M. Gabbay and
J. Woods (Amsterdam: Elsevier, 2008), 645665.
3
N. Rescher, The Development of Arabic Logic (Pittsburgh: University of Pittsburgh
Press, 1964), 8182.
4
The same assessment of the post-thirteenth century tradition of Arabic logic
is given by S. Inati, Logic, in A History of Islamic Philosophy, ed. S.H. Nasr and
O. Leamann (London: Routledge, 1996), II, 802823, esp. 804, by M. Maroth, Ibn
Sn und die peripatetische Aussagenlogik (Leiden: Brill, 1989), 216ff, and more
cautiously by J. Walbridge, Logic in the Islamic Intellectual Tradition: The Recent
Centuries. Islamic Studies 39(2000): 5575
5
T. Street, Arabic Logic, in Handbook of the History of Logic, Volume 1: Greek,
Indian and Arabic Logic, ed. D.M. Gabbay and J. Woods (Amsterdam: Elsevier, 2004),
introduction 3
from the eighth century until the twentieth. In this sense, it may well
appear premature to attempt a survey of a particular topic, especially
given the fact that so few works have been published, and that many
manuscripts are hidden away in uncatalogued private collections or
libraries that are poorly inventoried, difficult to access, and often do
not allow, or have facilities for, the reproduction of manuscripts.
In another sense, however, the term Arabic logic refers to some-
thing more manageable than the entirety of Arabic writings on logic
something that may properly be called a tradition and that primarily
includes works known to have had an influence on subsequent con-
tributors to the field. This influence can be gauged in a number of
ways. Important indicators include the number of extant manuscripts,
references in later works, and the existence of commentaries and
glosses on a work. Another way of gauging influence is to look at the
end-state of the tradition, so to speak. During the nineteenth and
early twentieth century, a number of works on logic were printed or
lithographed in the Islamic world. With very few exceptions,6 these
publications were not motivated by antiquarianism, and they therefore
reveal which works were being studied in various parts of the Islamic
world on the eve of modernity. For example, a plethora of lithograph
editions of the handbook Sullam al-ulm by Bihr (d. 1707) were
printed in India in the nineteenth and early twentieth centuries, and a
number of commentaries on the work were lithographed as well.
These facts indicate that this particular work was a widely studied
handbook in Indo-Islamic scholarly circles. Sullam al-ulm and its
standard commentaries take their point of departure from a number
of earlier works, in particular al-Risla al-Shamsiyya by Ktib (d. 1277),
Matli al-anwr by Urmaw (d. 1283), and Tahdhb al-mantiq by
Taftzn (d. 1390), along with a number of commentaries and glosses.
6
In exceptional cases, works were printed due to antiquarian motivations or the
express desire to reintroduce a forgotten work into the curriculum. For example, the
Egyptian reformer Muhammad Abdu (d. 1905) stated in the introduction to the edi-
tion of Sws al-Basir al-nasriyya that was printed in Cairo in 1898 that he had
come across a single manuscript of the work in Beirut, and that he found it to be bet-
ter than the works used in the Azhar College in his day. Similarly, when the commen-
tary on Tss Tajrd al-mantiq by H ill (d. 1325) was lithographed in Iran in
1311/18931894, the edition was prefaced with the statement that this was a work that
had been almost completely forgotten (kna ani-l-anzri mastran bal kna kaan lam
yakun shayan madhkran). Such exceptions can be detected in a straightforward
fashion by crosschecking with bio-biographical evidence.
6 introduction
account. This applies, for example, to the logical writings of Ibn H azm
(d. 1064), Ghazl (d. 1111), Ibn Kammna (fl. 1268), Taftzn
(d. 1390), Jurjn (d. 1413), Ibn Turka al-Isfahn (d. 1437), Qd Mr
H usayn al-Maybud (d. 1504), Musta f Mstr (d. 1707), Esad Ynyav
(d. 1722), and Muhammad al-Kishnw al-Fln (d. 1740). In general,
I do not systematically mention or discuss those writers who did not
have anything to contribute to the specific problem that I am investi-
gating. I have, however, made an exception in the case of such major
figures as Frb, Averroes, and Suhraward. This is partly because the
stature of such figures today makes it distinctly odd to write a mono-
graph on a theme within Arabic logic without taking their views into
account. I have also found it helpful in the first chapter to present their
general account of syllogism as a background for the discussion that
follows. In chapter four, I deemed it necessary to mention or discuss
some works that had nothing (or nothing new) to say on the issue,
simply by way of supporting my claim that, in certain regions of the
Islamic world in the seventeenth, eighteenth, and nineteenth centu-
ries, the problem of relational inferences appears not to have been dis-
cussed, or not to have been discussed in novel ways.
My focus will remain rather narrow throughout. Many passages
that I cite or discuss will invite broader questions. One might legiti-
mately wonder, for example, if the gradual recognition of relational
syllogisms in some parts of the Islamic world but not others was
related to more general shifts and differences in views on the nature
and aim of logic, or in core logical concepts. In other words, did ear-
lier scholars simply fail to see something that later scholars saw? Or
were there changes in core concepts, or in the general ideas of what
logic is, that may help us understand the development of ideas? One
might also wonder if a particular scholars views on the categorical syl-
logism and its limits were related to his more general positions in epis-
temology and metaphysics. These may well be promising lines of
inquiry, but they will not be systematically pursued here. To para-
phrase what the recently deceased English philosopher Bernard
Williams wrote in a different context: if one addresses such questions
properly then one does nothing else.7
7
B. Williams, Ethics and the Limits of Philosophy (Cambridge: Harvard University
Press, 1985), viii.
8 introduction
A Note on Periodization
The history of Arabic logic after the twelfth century is still to a large
extent unexplored territory. There is therefore no general agreement
on how to subdivide this later history. My research has led me to form
a view on this topic, and this view underlies the chapter divisions of
the present work. It may be helpful to present the reader with this view
in summary form from the outset. Additional details, and fuller evi-
dence for my claims, will be given in the course of the chapters that
follow.
From the perspective of fourteenth-century Islamic scholars, an
important transformation of Arabic logic occurred during the late-
twelfth and early-thirteenth centuries. The famous historian Ibn
Khaldn (d. 1406) associated this transformation with the influence of
especially Fakhr al-Dn al-Rz (d. 1210) and Afdal al-Dn al-Khnaj
(d. 1248), and his remarks are supported by other thirteenth- and
fourteenth-century scholars, who viewed Rz as marking the divide
between the early logicians (al-mutaqaddimn) and the later logi-
cians (al-mutaakhkhirn).8 As part of this transformation, Arabic
logic became decisively dissociated from the exegesis of Aristotles
Organon. The discipline became more narrowly focused on the five
predicables, definition, propositions and their immediate implications
(such as conversion and contraposition), and the formal syllogistic.
A number of extensive summas of logic (in this narrow sense), as well
as a number of condensed handbooks were written in the course of
the thirteenth century by the likes of Khnaj, Ktib (d. 1277),
Urmaw (d. 1283) and Samarqand (d. 1303). These thirteenth century
logicians may be described as post-Avicennian in the sense that they
typically took their point of departure from Avicennas works and
accepted the main innovations of Avicenna in logic, such as the
combinatorial hypothetical syllogism and the distinction between
descriptional (wasf) and substantial (dht) readings of modality
propositions. However, they did not approach Avicennas works with
anything like the veneration accorded Aristotles writings by, for exam-
ple, Averroes or the early Arabic Aristotelians of Baghdad.
8
See the careful discussion of the terms early and late logicians in J. Lameer,
Conception and Belief in Sadr al-Dn Shrz (Tehran: Iranian Institute of Philosophy,
2006), 6468.
introduction 9
9
T. Street, Arabic Logic, esp. 567579.
12 introduction
A Note on Transliteration
I have given the Arabic form of the name of scholars who were active
before the establishment of the Ottoman, Safavid and Mughal
10
See his Tadl al-miyr f naqd Tanzl al-afkr, in Collected Papers on Logic and
Language, ed. M. Mohaghegh and T. Izutsu (Tehran: Institute of Islamic Studies,
1974), 139248.
introduction 13
of h for ( ) , and s instead of th for (). I thus have Tkd, H all and
instead of in case the modern Turkish orthography is o, h instead
1
Unless otherwise indicated, further bio-bibliographic information on the logicians
discussed in this book can be found in the Register of Arabic Logicians in Reschers
The Development of Arabic Logic.
2
For a discussion of this question, see J. Lameer, Al-Frb and Aristotelian
Syllogistics (Leiden: E.J. Brill, 1994), 1318. Lameer concludes that the work is not the
Middle Commentary.
3
Frb. Kitb al-qiys, in al-Mantiq inda l-Frb, ed. R. al-Ajam (Beirut: Dr al-
Mashriq, 1985), II, 19. The Arabic bi-dhtihi would seem to play the same role in the
definition of syllogism as Aristotles (An. Pr. 24b20), translated into
the classical period 15
English as from their being so (Jenkinson) or from the fact that the assumptions are
such (Tredennick) or because these things are so (Smith). In the Arabic translation
of Theodorus, an associate of the celebrated translator H unayn b. Ishq (d. 877),
Aristotles is translated as li-wujdi tilka l-ashyi l-mawdati
bi-dhtih in the definition of the syllogism, but then merely as bi-dhtih in
Aristotles immediately following explication of what he means by .
[See A. al-Badaw, ed., Mantiq Arist (reprint, Kuwait: Wiklat al-Matbt, and
Beirut: Dr al-Qalam, 1980), I, 142143.] Since the Arabic Aristotelians (following
Aristotle) explicitly meant the condition to rule out elliptical arguments with sup-
pressed premises, the translation of the Arabic bi-dhtih as by themselves seems
appropriate. Note that rendering the term into by their essence or essentially would
falsely suggest that the subsequent l bi-l-aradi is a clarificatory reiteration. The
Arabic Aristotelians did not consider it as such, but rather considered it an inde-
pendent condition. It is also far from clear what the essence of two posited premises
would be.
4
On this point, see Lameer, Al-Frb and Aristotelian Syllogistics, 83. Lameer points
out that this view of syllogism is already in evidence in the work of the Syriac scholar
Athanasius of Baldh (d. 686).
5
See A. Speca, Hypothetical Syllogistic and Stoic Logic (Leiden: E.J. Brill, 2001),
3566.
16 chapter one
6
Frb, Kitb al-qiys, 37 (ayyu qawlin lam yakun talfuhu ah ada l-talfti l-lat
dhakarnh thumma zda fhi aw nuqisa minhu wa buddila tartbuhu wa suyyira
talfuhu ah ada l-talfti l-lat dhakarnh wa baqiya l-mafhmu an al-qawli l-awwali
al h latihi qabla l-taghyri kna dhlika l-qawlu qiysan wa ayyu qawlin ubdila
maknahu ah adu l-talfti l-qiysiyyati l-lat dhakarnh wa taghayyara l-mafhmu
al l-qawli l-awwali fa-sra shayan khara fa-inna dhlika l-qawla laysa bi-qiysin).
7
See T. Lee, Die griechische Tradition der aristotelischen Syllogistik in der Sptantike
(Gttingen: Vandenhoeck and Ruprecht, 1984), chapter IV, esp. 108109.
the classical period 17
for the two premises do not have three terms in total, one of which is
the middle term, and the two other of which appear in the conclusion.
Frb might have been willing to call such arguments syllogisms, but
only insofar as they can be regimented into standard syllogistic form
without any change in meaning. He did not, in his extant writings,
indicate how this was to be done.
Some idea of how Frb would have dealt with the argument from
equality may perhaps be gathered from his second-generation student
Ibn Zura (d. 1008), whose Middle Commentary on the Prior Analytics
has survived. Ibn Zura adduced Aristotles definition of a syllogism: a
statement, when things more than one are posited in it, then some-
thing else is implied with necessity by the existence of these posited
things themselves (qawlun mat wudiat fhi ashyu aktharu min
wh idin lazima shayun kharu min al-idti rri li-wujdi tilka l-ashyi
l-mawdati bi-dhtih).8 He added that the stipulation that the con-
clusion should follow from the existence of these posited things by
themselves rules out an argument such as A is equal to J & J is equal
to B, therefore A is equal to B. For this argument, Ibn Zura wrote, has
a suppressed universal premise which is: All things that are equal to
one thing are equal. Ibn Zura apparently conceded that the following
of the conclusion from the original premises is obvious (zhira).
Nevertheless, it is not a syllogism, since:
Aristotle does not consider a syllogism that which has one of its two
premises missing. Rather [he considers a syllogism] that in which all
premises are present.9
Ibn Zura distinguished this case from simple enthymemes what he
called al-maqys al-mudmara. The latter are single-premise argu-
ments that lack the major premise (i.e. the premise which contains
the predicate of the conclusion), such as the rhetorical argument
So-and-so has adorned his appearance, therefore he is an adulterer.
The major premise of the argument has been left out, viz. Everyone
who has adorned his appearance is an adulterer.10 As will be seen
below, later logicians tended to abandon this distinction between an
8
Ibn Zura, Kitb al-qiys, in Mantiq Ibn Zura, ed. J. Jhm and R. al-Ajam (Beirut:
Dr al-Fikr al-Lubnn, 1994), 108.
9
Ibn Zura, Kitb al-qiys, 110 (Arsttls l yataqidu qiysan m kna ih d
muqaddimatayhi nqisatan bal al-ladh muqaddimtuhu kulluh mawjdatun fhi).
10
Ibn Zura, Kitb al-qiys, 109.
18 chapter one
11
Ibn Zura, Kitb al-qiys, 177.
the classical period 19
12
Ibn Sn, al-Shif: al-Qiys, ed. S. Zyid and I. Madkour (Cairo: Wizrat al-thaqfa,
1964), 54.
13
Ibn Sn, al-Shif: al-Qiys, 5960.
20 chapter one
However, the principle in both cases is the same: the two original
premises are merged into one new premise, to which a universal premise
is added.
Avicenna distinguished such incomplete arguments from those
arguments whose premises imply a conclusion but do not do so with
necessity. An example of the latter case is the following argument:
No human is a horse
Every horse neighs
No human neighs
The conclusion is, Avicenna wrote, implied (yalzamu) by the premises.
However, the argument is nevertheless not a syllogism since a
conclusion does not always follow from premises that have this com-
position (talf ), viz. a universal-negative proposition and a universal-
affirmative proposition arranged in such a manner that the predicate
of the first is the subject of the second. Arguments with this form are
not always productive of a conclusion and hence are not syllogisms.14
Avicenna would seem to be invoking a distinction between material
and formal productivity.15 It is not clear, however, why elliptical argu-
ments and arguments that are only materially productive should be
considered two different types. Arguably, the example that Avicenna
gave of a materially productive but formally sterile argument has just
as good a claim to be considered elliptical as the argument A is equal
to B & B is equal to J, so A is equal to J. After all, both examples are
not formally productive, but become so on Avicennas account by
the addition of a further premise. Avicenna may himself have come
14
Ibn Sn, al-Shif: al-Qiys, 64.
15
It is worth noting that Avicenna glossed the stipulation that the conclusion should
follow with necessity (bi-l-darra) as: i.e. always, as opposed to in some matter
but not another matter (ay diman laysa f mddatin dna mdda). In recent years,
there has been some discussion of the question whether the distinction between for-
mal and material consequence was recognized by the Greek commentators or was a
later development within mediaeval Latin logic. For a discussion of this point, see
J. Barnes Logical Form and Logical Matter, in Logica, mente e persona, ed. A. Alberti
(Florence: Olschki, 1990), 39 ff., and J. Barnes, Truth, etc.: Six Lectures on Ancient Logic
(Oxford: Clarendon Press, 2007), 274 ff. It seems to me that one cannot make sense of
the distinction that Avicenna is making here without imputing to him some notion of
formal vs. material productivity, though he may not have drawn the distinction sys-
tematically or using those precise terms. This appears to be in line with Barnes own
conclusion with respect to ancient Greek logic. As will be clear from later chapters, a
number of later Arabic logicians explicitly drew a distinction between productivity by
virtue of form (sra) and by virtue of matter (mdda).
the classical period 21
to see the point, for in his later and enormously influential epitome
al-Ishrt wa l-tanbht he abandoned the distinction between a con-
clusion following from the premises by themselves and a conclusion
following with necessity. He there defined a syllogism as a statement
composed of statements which, if the propositions expressed in these
statements are conceded, implies by itself another statement (qawlun
muallafun min aqwlin idh sullima m rida fhi min al-qady laz-
ima anhu li-dhtihi qawlun kharu).16 In another epitome, entitled
Uyn al-h ikma, Avicenna gave an even more compact definition: a
statement composed of propositions which, if they are conceded,
implies by itself another statement (qawlun muallafun min qady
idh sullimat lazima anhu li-dhtihi qawlun kharu).17 This was, with
some minor variations in formulation, to remain the standard defini-
tion of a syllogism in Arabic logic from the thirteenth to the twentieth
century.
In al-Shif, Avicenna also distinguished incomplete arguments
from arguments that imply a conclusion accidentally.18 In the case of
accidental productivity, a conclusion follows, not from the premises
together with one or more suppressed premises, but from one of the
premises and another proposition that is equipollent to the other
premise. In other words, one of the premises is, in effect, replaced
with another premise that is logically equivalent to it. As an exam-
ple, Avicenna gave the following argument, already mentioned by
Alexander of Aphrodisias:
(i) The part of the substance, its removal necessitates the removal
of the substance ( juzu l-jawhari yjibu raf uhu raf a l-jawhari)
(ii) The removal of what is not a substance does not necessitate
the removal of the substance (irtifu m laysa bi-jawharin l
yarfau l-jawhara)
The part of the substance is a substance (juzu l-jawhari jawharun)
In this case too, Avicenna wrote, the two premises do not imply the
conclusion syllogistically. However, what is required to make a syllo-
gism in this case is not an additional premise, but rather replacing an
16
Ibn Sn, al-Ishrt wa l-tanbht, ed. J. Forget (Leiden: E.J. Brill, 1892), 65; ed.
M. Zri (Tehran: Bustan-e Ketab-e Qom Press, 1423/20022003), 138.
17
Fakhr al-Dn al-Rz, Sharh Uyn al-h ikma, ed. A. al-Saq (Tehran: Muassasat
al-S diq, 1415/19941995), 161.
18
Ibn Sn, al-Shif: al-Qiys, 6061.
22 chapter one
19
Avicenna did not elaborate on why this proposition is equipollent to the one it
replaces. Proposition (ii) has the form No non-A is B and proposition (ii*) has the
form Every B is A. The derivation of (ii) from (ii*) would seem to be a straightforward
instance of what later Arabic logicians called aks al-naqd al-mukhlif and nineteenth-
and early twentieth-century English logicians called obversion. The opposite deriva-
tion of (ii*) from (ii) does not correspond to any category of immediate implication
recognized in the Arabic logical tradition. It rests on the questionable assumption that
the converse of No non-A is B, viz. No B is non-A, is logically equivalent to Every
B is A.
the classical period 23
Could one not equally argue that the premises only produce a con-
clusion accidentally, and that it is only if the second premise is
replaced with its converse, No animal is a stone that the conclusion
follows non-accidentally? Again, Avicenna may have come to recog-
nize some of the difficulties mentioned, for the stipulation that the
syllogism should not produce its conclusion accidentally is left out of
the above-mentioned definitions of syllogism in his al-Ishrt and
Uyn al-h ikma.
In al-Ishrt, Avicenna briefly touched upon what he called the
syllogism of equality (qiys al-muswt). He wrote:
It may be that certain things that are left out [of the argument] are
known from the properties of the premises. The syllogism would then
be constructed in a form that does not accord with that of the [proper]
syllogism. This is as when they say J is equal to B & B is equal to A, so
J is equal to A. What has been left out is that what is equal to what is
equal is equal, and one has abandoned the syllogisms proper appear-
ance, according to which the entire middle-term must be common [to
both premises], for one in which what is common is only a part of it.20
Like much of Avicennas Ishrt, the passage is highly compressed.
Read in light of the Shif and the writings of Avicennas student
Bahmanyr (d. 1065), it would appear that Avicenna is alluding to the
basic position that has already been presented and discussed above.
According to this account, the syllogism of equality is not as it stands
a proper syllogism at all, for it only implies the conclusion together
with an additional premise to the effect that what is equal to what is
equal is equal. Avicenna seems to suggest that the additional premise
needed to formally produce the desired conclusion is somehow known
from the explicit premises. This appears to be the only plausible expla-
nation of his statement that it may be that certain things that are left
out are known from the properties of the premises. Perhaps he meant
that the unstated premise is too obvious to be explicitly mentioned
along with the two premises. Or perhaps he was alluding to the other
additional premise mentioned in al-Shif, namely that A is equal
to what is equal to J, which is somehow known from the mentioned
20
Ibn Sn, al-Ishrt, ed. Forget, 78; ed. Zri, 159 (rubbam urifa min ah kmi
l-muqaddimti ashyun tasqutu wa yubn l-qiysu al sratin mukhlifatin li-l-qiys
mithlu qawlihim J muswin li-B wa B muswin li-A fa-J muswin li-A fa-qad usqita
minhu anna musw l-musw muswin wa udila bi-l-qiysi an wajhihi min wujbi-l-
shirkati f jami l-awsati il wuqi l-shirkati f badihi).
24 chapter one
21
Bahmanyr, al-Tah s l, ed. M. Mutahhar (Tehran: Dnishkdah-i Ilhyt va
Marif-i Islm, 1349H/19701971), 108.
22
Bahmanyr, al-Tah s l, 109 ( fa-innaka idh qultu J muswin li-B wa B muswin
li-D fa-J muswin li-D fa-innaka lam tusarrih bi-ih d l-muqaddimatayni fa-inna srata
hdh l-qiysi yajibu al hdh yaknu [sic] qiysan tmman fa-naqlu J shayun
muswin li-B al-ladh huwa shayun muswin li-D wa muswiyti l-muswiyati [sic]
mutaswiyatun, fa-idhan J muswin li-D kaannahu qla J muswin li-musw D wa
kullu muswin li-musw D fa-huwa muswin li-D fa-J muswin li-D).
the classical period 25
23
Sw, al-Basir al-nasriyya f ilm al-mantiq, ed. M. Abdu (Cairo: al-Matbaa
al-kubr l-amriyya, 1316/1898), 7879.
26 chapter one
from premises by themselves and at the same time follow from the
premises accidentally. However, this is not possible, and hence there
is no need for an additional condition.
Also ruled out by the stipulation that the conclusion should follow
from the premises by themselves are arguments in which the premises
only imply a conclusion if conjoined to another premise that is left
unmentioned entirely. As an example, he mentioned the syllogism of
equality:
This is like our saying A is equal to B & B is equal to J, so A is equal to J.
It is not implied by just this statement [i.e. the premise-pair] that A is
equal to J; rather this [conclusion] is implied by another item that is
omitted, namely that A is equal to what is equal to J, and what is equal
to what is equal is equal. Only then is it implied that A is equal to J.24
In other words, the conclusion A is equal to J does not follow from
the original pair of premises by itself. Rather, it follows from an
unmentioned fact or item (amr) which Sw stated as: A is equal to
what is equal to J, and what is equal to what is equal is equal. Despite
the formulation, it can hardly be the case that he meant this to be
understood as a single premise that, in conjunction with the two men-
tioned premises, produces the desired conclusion. Like Bahmanyr,
Sw must have meant that the additional item needed to produce
the conclusion is actually a new pair of premises: A is equal to what
is equal to J and What is equal to what is equal to J is equal to J.
However, again like Bahmanyr, Sw did not make clear the relation-
ship of these two premises to the original premises. In particular, it is
natural to ask what the relationship is between the original premise-
pair A is equal to B & B is equal to J and the new premise A is equal
to what is equal to J. Has the original premise-pair simply been
replaced with another premise that is equipollent to it? If so, then the
argument from equality would not, after all, be a straightforward
example of an enthymeme, but would rather be partly an enthymeme
and partly an argument in which one or more premise are replaced by
equipollent propositions. Alternatively, the original premise-pair A is
equal to B & B is equal to J and the new premise A is equal to what is
equal to J could be seen as being related like premises and conclusion
24
Sw, al-Basir, 79 (fa-mithlu qawlin A muswin li-B wa B muswin li-J fa-A
muswin li-J fa-l yalzamu min mujarradi hdh l-qawli anna A muswin li-J bal yal-
zamu min amrin kharin h udhifa wa huwa anna A muswin li-musw J wa musw
l-musw muswin fa-yalzamu h naidhin anna A muswin li-J).
the classical period 27
Over the course of the eleventh and twelfth centuries, the influence of
Avicennas philosophical and logical works spread throughout the
Islamic world, and by the thirteenth century they had in most parts
of the Islamic world effectively replaced the works of Aristotle as the
point of departure and reference for philosophers and logicians.25
As will be seen in the following chapters, it was the regimentation of
the syllogism of equality by Avicenna and his followers that was the
starting-point for later discussions, and the slightly different regimen-
tation of Ibn Zura was forgotten.
This spectacular spread of Avicennas influence did not occur
without resistance. In the remaining part of this section I will present
the discussions of syllogisms by some of the more prominent twelfth-
century philosophers whose attitude to Avicennas influence ranged
from skepticism to outright hostility: (1) the Iraqi-Jewish convert to
Islam Ab l-Barakt Ibn Malka al-Baghdd (d. 1165); (2) the famous
25
Street, Arabic Logic, 552554, 560561.
28 chapter one
26
Ab l-Barakt al-Baghdd, al-Mutabar f l-h ikma (Hyderabad, Deccan: Dirat
al-Marif al-Uthmniyya, 1357/19381939), I, 123 (wa dhlika anna l-qarnata
taknu min qawlayni hum muqaddimatni wa f kulli muqaddimatin h addni h addin
mawdin wa h addn mah mlin wa yalzamu anh m yalzamu li-shirkatin bayna
l-muqaddimatayni wa tilka l-shirkatu taknu f juzin l mah lata idh law knat f
l-kulli la-knat ih dhum hiya l-ukhr bi-aynih wa dhlika l-juzu imm an yakna
huwa l-mah mla wa imm an yakna l-mawda f kilayhim wa imm an yakna
mawdan f ih dhum mah mlan f l-ukhr).
the classical period 29
27
Ab l-Barakt al-Baghdd, al-Mutabar, 173.
28
Ab l-Barakt al-Baghdd, al-Mutabar, 170.
29
Ab l-Barakt is here clearly taking the original premises as constituting a second-
figure syllogism of the form:
Every J is B
No non-A is B
No J is non-A
30 chapter one
a familiar syllogistic form: either treat them like enthymemes that call
for the addition of a further premise, or replace one or more of the
original premises with propositions that are equipollent to them.
30
Ibn Rushd, Talkhs kitb al-qiys, ed. M. Qsim, C.E. Butterworth, and A. Hard
(Cairo: al-Haya al-misriyya al-mma li-l-kitb, 1983), 65.
31
Ibn Rushd, Talkhs kitb al-qiys, 65.
32
Ibn Rushd, Talkhs kitb al-qiys, 65.
33
Ibn Rushd, Talkhs kitb al-qiys, 65.
34
Ibn Rushd, Talkhs kitb al-qiys, 190192.
the classical period 31
the subject by means of a single premise, for this single premise must
have either one term or two terms in common with the desired con-
clusion. It cannot have both terms in common, for then one would be
proving a proposition by means of itself. It also cannot have one term
only in common with the conclusion, for there is no syllogistic argu-
ment from the premise A is B or A is not B to the conclusion A is C
or A is not C. The premise obviously cannot have no term in common
with the conclusion, for there is no syllogistic argument from a
premise of the form D is J to the conclusion B is A. In an argu-
ment that meets the aforementioned definition of a syllogism, there
must therefore be two premises. These two premises each have two
extremes. It cannot be the case that the premises have no extreme in
common, nor can they have two extremes in common, so they must
share one and only one term. It also must be the case that both
extremes of the desired conclusion appear in the pair of premises,
though not in the same premise. Thus, each of the two premises must
have one term in common with the desired conclusion. In this way,
it can be shown or so Averroes claimed that a syllogism that seeks
to establish that a predicate belongs or does not belong to a subject
must have the form of a standard Aristotelian syllogism with three
terms.
Averroes argument on this point which is derived from Aristotle
(Pr An. A25) is more developed than that of Ab l-Barakt al-
Baghdd, but is also beset by a number of difficulties. One difficulty
that is particularly relevant in the present context is that an apparent
counter-example is the syllogism of equality. It would seem at least
at first sight that one could show that A is equal to J by adducing the
premises A is equal to B and B is equal to J, and these premises do
not have an entire extreme in common. Averroes must have held such
an argument not to be formally productive, and hence not to be a gen-
uine counter-example. The problem is that as has been seen above
it is far from clear how the syllogism of equality can be supplemented
with additional premises so as to become formally productive from
an Aristotelian perspective. Furthermore, one could present other
relational arguments that cannot plausibly be dismissed as elliptical.
It is for example possible to argue as some later logicians writing
in Arabic did that one may want to prove a proposition such as
Zayd is the brother of a writer by adducing the premises Zayd is the
brother of Amr and Amr is a writer. It is difficult to see why seeking
to establish such a conclusion by means of such premises should be
32 chapter one
35
Ibn Rushd, Talkhs kitb al-qiys, 227.
the classical period 33
36
See N. Shehaby, The Propositional Logic of Avicenna (Dordrecht, Reidel, 1973).
37
Ab l-Barakt al-Baghdd, al-Mutabar, I, 155.
38
See for example his treatise al-Qawl f l-qiys al-h aml wa l-shart wa naqd al-qiys
al-iqtirn l-shart inda Ibn Sn, in Rasil falsafiyya: Maqlt f l-mantiq wa l-ilm
al-tab li-Ibn Rushd, ed. J. Alaw (Casablanca: Dr al-Nashr al-Maghribiyya, 1983),
187207.
39
Ibn Rushd, Talkhs kitb al-qiys, 229230.
40
Ibn Rushd, Talkhs kitb al-qiys, 229 (laysa kullu m yalzamu an shayin bi-
l-idti rri fa-huwa lzimun luzman qiysiyyan bal m lazima bi-idti rrin an muqaddi-
matayni nisbatu ih dhum il l-ukhr nisbatu l-kulli il l-juzi fa-huwa qiys).
34 chapter one
The principle that not all cases of necessary implication are cases
of syllogistic implication had already been recognized by Aristotle
and by the Greek commentators.41 It is not, despite appearances, an
acknowledgement of the existence of formally productive arguments
that are not syllogisms. Such an understanding would make nonsense
of much of what Aristotle, his Greek commentators, and Averroes had
to say elsewhere. Rather, Averroes was simply expressing the view that
there are arguments in which the conclusion follows from the premises
with necessity but are nevertheless not syllogisms because they do not
meet other conditions stipulated in the definition of a syllogism
namely, the condition that the conclusion follow from the premises by
themselves and not accidentally. His previous discussion of the defi-
nition of syllogism clearly shows that he considered both (i) enthyme-
mes with true suppressed premises and (ii) materially productive but
formally sterile arguments to imply conclusions necessarily. Conceding
the existence of such arguments does not in any way amount to a
departure from the orthodox Aristotelian position that any multi-
premise argument that implies another categorical proposition by
itself and not accidentally is a categorical syllogism with three and
only three terms.
Averroes thus recognized the familiar techniques for regimenting
arguments in books and speech into standard Aristotelian syllogisms:
treating an argument as enthymematic, or reformulating one or more
premises. There can be little doubt that he would have used much the
same techniques to solve the anomaly of the syllogism of equality as
those used by Ibn Zura or Avicenna and his followers.
41
J. Barnes, Logical Form and Logical Matter, in Logica, mente e persona, ed.
A. Alberti (Florence: Olschki, 1990), 5557.
42
Suhraward, Mantiq al-Talwh t, ed. A.A. Fayyd (Tehran: Intishrt-i Dnishgh-i
Tehrn, 1955), 46.
the classical period 35
43
Suhraward, Mantiq al-Talwh t, 87.
44
Shahrazr, Sharh H ikmat al-ishrq, ed. H. Ziai (Tehran: Institute for Humanities
and Cultural Studies, 2001), 138; Ibn Kammna, Sharh al-Talwh t, ed. by Najafqoli
Habibi (Tehran: Miras-i Maktub, 2009), I, 328.
45
The point is made by Mull Sa dr (d. 1635) in his gloss on Qutb al-Dn al-Shrzs
commentary on H ikmat al-ishrq, see Mull Sa dr, Talqt ala Sharh Hikmat al-ishrq
(Lithograph: 1315/18971898; reprint, Qom: Intisharat Baydar, n.d.), 140 (margins).
46
Suhraward, H ikmat al-ishrq, ed. and trans. H. Ziai and J. Walbridge (Provo,
Utah: Brigham Young University Press, 1999), 21.
36 chapter one
IV. Conclusion
47
Suhraward, H ikmat al-ishrq, 3435.
the classical period 37
48
Quoted and translated in Barnes, Logical Form and Logical Matter, 83; Barnes,
A Third Sort of Syllogism, 179.
49
Quoted and translated in Barnes, Logical Form and Logical Matter, 84.
38 chapter one
The historian Ibn Khaldn (d. 1406) accorded the Persian scholar
Fakhr al-Dn al-Rz (d. 1210) a prominent place in his brief account
of the history of Arabic logic in the well-known Prolegomena (al-
Muqaddima) to his universal history. Ibn Khaldn noted that, starting
with Rz, logicians ceased to be interested in covering all the books of
Aristotles Organon, and instead reoriented the field toward a more
focussed study of the five predicables, definition, propositions and
their immediate implications, and the formal syllogistic.1 Ibn Khaldn
did not, however, indicate whether Rz should be credited with any
substantial contributions besides this shift in focus.
In The Development of Arabic Logic (1964), Nicholas Rescher, who
was well aware of the passage from Ibn Khaldns Muqaddima, wrote
of Rz that:
It is possible that no genuinely original developments of significant
intrinsic importance can be credited to him, and that the principle origi-
nality of his contribution lies in the organization of materials and in the
anti-Avicennist impetus of his discussion.2
An astonishing aspect of this assessment is that it cannot have been
based upon a study of Rzs works on logic. When Rescher penned
these remarks, Rzs major works in the field (the logic section of his
al-Mulakhkhas and the logic sections of his commentaries on
Avicennas two works al-Ishrt and Uyn al-h ikma) were not availa-
ble in printed editions, and Rescher gives no indication of having con-
sulted any manuscripts or relevant secondary studies. Whence then
did Rescher derive his judgment?
The answer is presumably that he was influenced by certain general
assumptions that have been widespread amongst modern historians of
1
Ibn Khaldn, The Muqaddimah: An Introduction to History. Trans. F. Rosenthal.
New York: Pantheon Books, 1958. I, 142143.
2
Rescher, The Development of Arabic Logic, 185.
40 chapter two
3
This is noticeable in, for example, the influential works by George Makdisi and
Fazlur Rahman.
4
In his Qisas al-ulam, Tonkbon remarked that the influence of Avicennas phi-
losophy waned in the face of criticisms by Ab l-Barkt al-Baghdd and Fakhr
al-Dn al-Rz, before being resuscitated by Ts. This appears to be the principle piece
of evidence for Reschers postulation of a western school of anti-Avicennian logicians
led by Ab l-Barkt and Rz (who in fact differed significantly in their logic).
Rescher seems to have become acquainted with Tonkbons remark through Shlomo
Pines, who cites it in his entry on Ab l-Barkt Hibat Allh b. Malka al-Baghdd
al-Balad in the Encyclopedia of Islam, 2nd ed., ed. P. Bearman et al. (Leiden: E.J. Brill,
1960), I, 111.
5
The Ottoman bibliographer Ktib eleb (d. 1657) recorded an anecdote accord-
ing to which Qutb al-Din al-Rz (d. 1365) presented to his teacher Qutb al-Dn
al-Shrz (d. 1311) some criticisms he had penned of Fakhr al-Dn al-Rz, only to be
told that to pursue with criticisms the author of many works is easy (al-taaqqubu
al sh ibi l-kalmi l-kathri yasr); see Kashf al-zunn an asm l-kutub wa l-funn
(Istanbul: Maarif Matbaasi, 19411943), 95.
the challenge of rz and its aftermath 41
6
Fakhr al-Dn al-Rz, Sharh al-Ishrt wa l-tanbht, ed. A. Najafzada (Tehran:
Amjuman-i sr va Mafkhir-i Farhang, 2005), I, 285 (idh quln A muswin li-B wa
B muswin li-J wa antajn anna A muswin li-musw J fa-f hdh l-qiysi l-awsatu
laysa bi-mutakarririn wa l-natjatu laysat h silatan min ijtimi l-tarafayni amm
anna l-awsata ghayru mutakarririn fa-lianna l-muqaddimata l-l wa hiya qawlun
A muswin li-B fa-l-alifu huwa l-mawdu wa l-muswi li-B huwa l-mah mlu thumma
idh quln B muswin li-J jaaln mawda hdhihi l-qadiyyati l-ba faqat wa laysa
hdh tamma mah mli l-muqaddimati l-l fainna mah mlah kna huwa l-musw
li-B l l-ba faqat fa-zhirun anna l-awsata ghayru mukarrarin thumma idh antajn
anna A muswin li-musw J fa-mawdu hdhihi l-natjati huwa A wa huwa mawdu
l-sughr amm mah mluh fa-huwa muswin li-musw J wa laysa dhlika mah mla
l-kubr fainna mah mlah huwa musw J faqat).
42 chapter two
previous chapter that the first step is somewhat curious. The original
two premises must somehow guarantee the truth of the new amalga-
mated premise, but this should invite the question whether the origi-
nal premises A is equal to B and B is equal to J constitute a syllogism
with respect to the conclusion A is equal to what is equal to J. Rz
saw that the Avicennian regimentation of the syllogism of equality
invites this other, and apparently insurmountable, problem for the
assumption that a categorical syllogism must have three terms, of
which one is the middle term. Why is the inference from A is equal to
B & B is equal to J to A is equal to what is equal to J not a syllogism
in which the conclusion follows from the premise-pair by itself ?
Rz went on to present a possible answer to the objection that has
just been presented.7 It could, he wrote, be objected that though the
syllogism of equality superficially ( f l-zhir) might seem to lack a
standard middle term, upon closer consideration (inda l-tah qq) the
argument should be cast in the following manner:
A is equal to B
Everything that is equal to B is equal to J
A is equal to J
It is curious that Rz suddenly shifted from discussing an argument
whose conclusion is A is equal to what is equal to J to discussing
another argument whose conclusion is A is equal to J. It is obvious
that the arguments are not the same, and that no-one would claim that
the second is a regimentation of the first. Presumably, Rz was dis-
cussing an alternative regimentation of the syllogism of equality that
would bypass the need for an intermediate step in which one derives
A is equal to what is equal to J from the premises. To this alternative
regimentation, Rz made three replies:
(1) If J is not equal to B then the second premise is false. If J is equal to
B, then J is one of the things that being equal to B are equal to
J, but this is false, because equality only obtains between different
entities (lianna l-muswta bada l-mughyara).
(2) The proposed second premise Everything that is equal to B is
equal to J cannot be known to be true unless we already know
that B is equal to J and that what is equal to what is equal is
7
Rz, Sharh al-Ishrt, I, 285286.
the challenge of rz and its aftermath 43
8
Rz, Mantiq al-Mulakhkhas, ed. A.F. Karamaleki and A. Asgharinezhad (Tehran:
ISU Press, 2003), 243.
44 chapter two
the standard position that the condition that the implication should
be by the premise-pair itself rules out arguments such as the syllo-
gism of equality, which are only productive if further premises are
added:
Our stating [in the definition of syllogism] by itself is in order to guard
against two things. First: the propositions should not be in need of
another proposition to be productive. For if you say A is equal to B & B
is equal to J one might superficially think that it produces A is equal to
J. The verified view is that it does not imply this conclusion, rather the
implied proposition is A is equal to what is equal to J. If you then say &
What is equal to what is equal is equal then the desired conclusion is
implied.9
On Rzs account, the syllogism of equality is a complex argument
consisting of two syllogisms, the first of which is:
A is equal to B
B is equal to J
A is equal to what is equal to J
This intermediate conclusion is then conjoined to the initially sup-
pressed premise Everything that is equal to what is equal [to J] is
equal [to J] to produce the conclusion A is equal to J. The analysis
implies accepting that the first argument is a syllogism, even though it
is not a standard syllogism with three terms.
It would thus seem that Rz recognized the difference between the
following two arguments:
(i) A is equal to B (ii) A is equal to B
B is equal to J B is equal to J
A is equal to J A is equal to what is equal to J
Argument (i) is, he wrote, not a syllogism in which the conclusion
follows from the premise-pair by itself. By contrast, it would seem
that Rz considered argument (ii) to be a syllogism in which the
conclusion does follow from the premise-pair by itself, though it is a
9
Rz, Mantiq al-Mulakhkhas, 244 (wa qawlun li-dhtihi ihtirzun an shayayni
fa-A: anna tilka-qady l tahtju f kawnih muntijatan il qadiyyatin ukhr fa-
innaka idh qulta A muswin li-B wa B muswin li-J fa-yuzannu f l-zhiri annahu
yuntiju anna A muswin li-J wa f l-tah qqi l talzamuh hdhihi l-natjatu bal
al-lzimu anna A muswin li-musw J thumma idh qulta wa musw l-musw
muswin fa-h naidhin yalzamu l-matlb).
the challenge of rz and its aftermath 45
10
Rz, Mantiq al-Mulakhkhas, 319 (wa hiya mithlu qiysi l-muswti ka-qawlin
l-jismu fhi sawdun wa kullu sawdin lawnun fa-innahu yalzamu bi-l-darrati min
al-ilmi bi-htayni l-muqaddimatayni l-ilmu bi-anna l-jisma fhi lawnun wa kadhlika
qawlun l-durratu f l-h uqqati wa l-h uqqatu f l-sundqi fa-innahu yalzamu minhum
kawnu l-durrati <fm f> al-sundqi wa l-awsatu ghayru mutakarririn lkinna
qawlan mat kna l-awsatu mutakarriran lazimat al-natjatu l yanakisu naam
inna dhlika yakhtalifu bi-khtilfi l-mawdd).
11
The editors have chosen the manuscript reading al-durra f l-sundq, but this
destroys the formal parallel between the two examples. According to their critical
apparatus, the variant I have chosen is supported by two manuscripts, including the
46 chapter two
earliest used (dated 631/12331234), which is otherwise the base-text of the edition.
The reading I have chosen is also supported by what appears to be the oldest extant
copy of Rzs Mulakhkhas, copied in 603/1206, i.e. during Rzs lifetime, and pre-
served in the British Library, but not used by the editors. See Rz, al-Mulakhkhas f
l-mantiq wa l-h ikma (Manuscript Or. 2360, British Library), fol. 58a.
12
Rz, Sharh Uyn al-h ikma, 167.
the challenge of rz and its aftermath 47
13
Rz, Sharh al-Ishrt, 335.
14
Rz, Sharh al-Ishrt, 335.
48 chapter two
The historian Ibn Khaldn after noting that logicians after Rz nar-
rowed their focus to the study of the five predicables, definition, prop-
ositions and their conversions, and the formal syllogistic observed
that treatment of such topics has become lengthy and wide-ranging.15
It is not difficult to gather why he wrote this. Thirteenth-century
logicians working in the wake of Avicenna and Rz produced a
number of extensive summas on logic. These tended to be formally
oriented: their attention was primarily dedicated to definition and
its preliminaries (the five predicables), as well as the formal syllogistic
and its preliminaries (the immediate implications of propositions,
such as conversion and contraposition). They in particular displayed
a strong interest in exploring the technicalities of modal and condi-
tional logic. By contrast, most of them showed little interest in getting
the interpretation of, say, Aristotle or Avicenna right. The works of
Aristotle in fact ceased to be the point of departure for logicians writ-
ing in Arabic in this period. Instead, Avicennas works came to
be the new paradigm for doing logic almost everywhere in the Islamic
world (Northwest Africa and Islamic Spain held out until the four-
teenth century), but his works tended to be approached with the
same critical and argumentative spirit exhibited by Rz and indeed
by Avicenna himself. The ethos is well-captured in a remark by the
very influential Persian logician Afdal al-Dn al-Khnaj (d. 1248) who
was active in Cairo: it is not our business to preoccupy ourselves with
what people may have meant in these matters, but to verify the issues
and establish what follows in case such and such is the meaning.16
In the thirteenth century, doing logic in Arabic was thoroughly
15
Ibn Khaldn, The Muqaddima, I, 143.
16
Khnaj, Kashf al-asrr an ghawmid al-afkr, ed. K. El-Rouayheb (Berlin &
Tehran: Institute for Islamic Studies & Iranian Institute of Philosophy, 2010), 279
the challenge of rz and its aftermath 49
(l alayn an nashtaghila bi-murdi l-nsi f hdhihi l-umri bal tah qqi l-h aqqi wa
anna l-lzima an kulli murdin mdh).
17
Ibn Wsil al-H amaw, Mufarrij al-kurb f akhbr Ban Ayyb, ed. Rab and
shr (Cairo: Dr al-Kutub, 1977), V, 3940.
18
Besides Daqiq al-h aqiq which will be cited below, he wrote another lengthy
summa entitled al-Nr al-bhir f l-h ikam al-zawhir, published in facsimile by
F. Sezgin (Frankfurt: Publications for the History of Arabic-Islamic Science, 2001).
The brief discussion of the syllogism of equality in that work occurs in volume 1,
p. 243. mid also wrote an extensive commentary on Avicennas Ishrt, which I have
not seen.
19
mid, Daqiq al-h aqiq (Manuscript Garrett 42B, Firestone Library, Princeton
University), fol. 96b (wa rubbam tushukkika al m dhakarnhu min luzmi ttih di
l-h addi l-awsati wa takarrurihi f muqaddimatayi l-iqtirn bi-qiysi l-taswiya).
50 chapter two
20
mid, Daqiq al-h aqiq, fol. 97a.
21
mids phrasing is elliptical and my reconstruction is tentative. All that he wrote
was: al-taqdru l-thlithu an yuqla: al-murdu min qawlin J muswin li-D ay al-jm
al-ladh muswhu B wa inda dhlika fa-l-h addu l-awsatu yaknu mutakarriran. See
fol. 97b.
22
Both the manuscript of al-Nr al-bhir published by Sezgin and the Princeton
manuscript of Daqiq al-h aqiq were copied in the lifetime of the author. There
appears to be no other copies of either work.
the challenge of rz and its aftermath 51
23
Ts, Sharh al-Ishrt (Tehran: Matbaat al-H aydar, 1377/19571958), I, 237.
52 chapter two
24
Ts, Sharh al-Ishrt, I, 237.
25
Ts, Sharh al-Ishrt, I, 237238 (l yakhl imm an yakna bayna mafhmi
l-maqtli bi-l-sayfi wa mafhmi bi-latin h addiyyatin taghyurun yaqtad an yakna
ah aduhum l-mah mula al l-khari aw l yakna baynahum taghyurun aslan bal
hum bi-manzilati lafzayni mutardifayni yuabbirni an shayin wh idin wa al
l-taqdri l-thn l yaknu dhlika qiysan wa l f quwwatihi bal kna qawlun Zayd
maqtlun bi-latin h addiyyatin al-ladh zannanhu natjatan fa-huwa bi-aynihi
qawlun Zayd maqtlun bi-l-sayfi l-ladh zannanhu muqaddimatan wa h naidhin
lam yakun baynahum farqun lianna mah mlahum ismni mutardifni).
the challenge of rz and its aftermath 53
26
Note that Ts did not explain why the two propositions are equipollent. He helped
himself to precisely the inference that would later be presented as a challenge to
Aristotelian syllogistic by De Morgan, but apparently did not feel the need to justify it.
As will be seen in chapter V, the later Safavid scholar q H usayn Khvansr (d. 1687)
addressed this point.
27
Ts, Sharh al-Ishrt, I, 238 (wa al l-taqdri l-awwali kna qawlun Zayd
maqtlun bi-l-sayfi wa l-sayfu latun h addiyyatin fi quwwati qiysin sratuhu Zayd
maqtlun bi-l-sayfi wa l-maqtlu bi-l-sayfi huwa l-maqtlu bi-latin h addiyyatin wa
yuntiju m dhakarnhu).
54 chapter two
28
Ts, Sharh al-Ishrt, I, 280 (wa qawlun wa B muswin li-J laysa bi-juzi l-qiysi
bal huwa baynu h ukmi m li-l-B al-ladh huwa juzun min ah adi h uddi l-qiys).
29
Qutb al-Dn al-Rz, al-Muh kamt. Printed with Ts, Sharh al-Ishrt (Tehran:
Matbaat al-H aydar, 1377/19571958), I, 237 (wa l-h silu anna qiysa l-muswti l
yuntiju bi-l-dhti anna A muswin li-muswin li-J bal bi-wsitati qawlin wa l-musw
li-B muswin li-muswin li-J).
the challenge of rz and its aftermath 55
30
Ts, Tadl al-miyr f naqd tanzl al-afkr, in Collected Papers on Logic and
Language, ed. M. Mohaghegh and T. Izutsu (Tehran: Institute of Islamic Studies,
1974), 201202.
31
Ibn Khaldn, The Muqaddima, I, 143.
56 chapter two
32
Khnaj, Kashf al-asrr, 232 (wa li-qilin an yaqla: alastum sallamtum luzma
anna A muswin li-musw J li-l-muqaddimatayni l-madhkratayni? fa-kna l-sulu
bi-aynihi qiman).
the challenge of rz and its aftermath 57
33
Sirj al-Dn al-Urmaw, Matli al-anwr, printed in the margins of Qutb al-Dn
al-Rzs Lawmi al-asrr fi Sharh Matli al-anwr (Istanbul: Matbaa-i Amire,
1277/18601861), 177 (wa min al-nsi man jaala tilka l-muqaddimata qawlan
musw l-musw muswin wa anta talamu annahu maa hdhihi l-muqaddimati l
yuntiju bi-l-dhti wa l yatakarraru l-wasat).
58 chapter two
34
Ktib, Jmi al-daqiq (Manuscript Or. 11201, British Library), fol. 78a78b,
and (Manuscript Majm Musta f Bsh 162, Dr al-Kutub al-Misriyya), 587588
(l yuqlu idh sullima anna htayni l-muqaddimatayni yuntijni bi-l-dhti anna
A muswin li-musw J kna qiysan li-sidqi m dhakartumhu min al-h addi alayhi
h naidhin li-ann naqlu naam innahu qiysun bi-l-nisbati il hdh l-qawli wa laysa
qiysan bi-l-nisbati il qawlin A muswin li-J wa l tanf f dhlika lianna
l-qiysiyyata amrun idfyyun fa-jza an yakhtalifa bi-l-nisbati il mafhmtin
mukhtalifatin bi-an yakna muqaddimatni qiysan bi-l-nisbati il shayin wa l
yakn qiysan bi-l-nisbati il ghayrihi).
35
The entry on this logician in Rescher, The Development of Arabic Logic, 209210
should be corrected and supplemented on the basis of L.B. Miller, al-Samarqand,
Shams al-Dn, in Encyclopaedia of Islam, 2nd Edition, ed. P. Bearman et al. (Leiden:
E.J. Brill, 1960), VIII, 1038.
the challenge of rz and its aftermath 59
term is repeated in both of the simple syllogisms into which the argu-
ment is analyzed, then it can be objected that the definition of a syllo-
gism does not state that the middle term must be repeated. If it did,
then one could simply rule out syllogisms of equality by definition,
without the need for the tortuous analysis. Furthermore, Samarqand
argued, the problem remains even if one accepts Khnajs analysis:
Does the pair of premises A is equal to B & B is equal to J constitute a
syllogism in relation to the proposition A is equal to what is equal to
J? If it does, then a syllogism need not have a middle term and the far-
fetched analysis is not called for. If it does not, then the definition of
syllogism (a statement composed of propositions that by itself implies
another statement) is defective and must be revised, since evidently A
is equal to B & B is equal to J qualifies as a statement composed of
propositions that by itself implies another statement (A is equal to
what is equal to J) and yet is not a syllogism.36
Samarqand went on to devote a short section of his lengthy work
to syllogisms whose middle-term is not repeated (al-qiysti l-lat l
taknu awstuh mutakarriratan). The section is worth quoting in its
entirety:
These are many, but that which is most frequent is the relational syllo-
gism (qiys al-nisba). This obtains when a relation and one of the two
things related are predicated of the other, in the minor premise or in
both the minor and the major premise. This syllogism produces a state-
ment by itself and another statement not by itself. As for the first, its [i.e.
the conclusions] subject is the minor term and its predicate is the rela-
tion that is in the minor premise together with the major term. For
example, our statement Zayd is a father of Bakr & Bakr is a brother of
Khlid produces that Zayd is a father of a brother of Khlid. This is evi-
dent, since what stands in a relation to what stands in a relation stands
in a relation to that relation. This is also like our statement Zayd is a
father of Bakr & Bakr is a scholar, so Zayd is a father of a scholar, and
like our statement A is equal to B & B is equal to J, so A is equal to what
is equal to J, and A is a part of B & B is a part of J, so A is a part of a part
of J, and A implies B & B implies J, so A implies what implies J, and so
on. This is also like our statement The pearl is in the box & the box is in
the chest, so the pearl is in what is in the chest, and The human is from
a clot & the clot is from the elements, so the human is from what is from
the elements, and The colour is in the body & every body is a substance,
so the colour is in what is in a substance.
36
Samarqand, Sharh al-Qists (Manuscript Landberg 1035, Berlin Staatsbibliothek),
fol. 93a93b.
60 chapter two
As for the second, and this is when the productivity is not by itself :
its [i.e. the conclusions] subject is the minor term and its predicate is the
major term. This is as our statement [in the conclusion] A is equal to B,
or implies B, or The pearl is in the chest, or The human is from the
elements, or The colour is in the substance. These do not result [from
the premise-pair] by itself, for if they did then it would never be the
case that they do not follow, and this is not so. For if we say A is dissimi-
lar to B & B is dissimilar to J, then this is not productive. Or if we say A
is half of B & B is half of J, or if we say The human is from the clot & the
clot is a fluid. Rather, when the mentioned conclusion is implied [by the
syllogism], it is only implied by the mediation of an imported premise.
However, it [i.e. the conclusion in the second case] is more useful in the
sciences than what is implied by itself.37
Both the section-heading and the examples reveal Samarqands
indebtedness to Rz. However, Samarqand obviously had a clearer
sense that the category of syllogisms in which the middle-term does
not recur in its entirety included two distinct types of argument,
depending upon the kind of conclusion one sought to derive. One
type is the classic syllogism of equality:
A is equal to B
B is equal to J
A is equal to J
37
Samarqand, Qists al-afkr (Manuscript 45 Hk 2213, www.yazmalar.gov.tr), sayfa
numeras 58r. The manuscript is unpaginated, and hence I give the digital page
number (sayfa numeras) on which the folio can be located on-line (wa hiya kathratun
lkin aktharuh wuqan huwa qiysu l-nisbati wa huwa an tujala l-nisbatu maa
ah adi l-muntasibayni mah mlan al l-khari f l-sughr aw f l-sughr wa l-kubr wa
hdh l-qiysu yuntiju qawlan bi-l-dhti wa qawlan khara l bi-l-dhti amm l-aw-
walu fa-yaknu mawduhu l-asghara wa mah mluhu l-nisbata l-lat f l-sughr maa
l-akbari ka-qawlin Zayd abun li-Bakr wa Bakr akhun li-Khlid yuntiju Zayd ab
akh Khlid wa dhlika zhirun idh al-muntasibu il muntasibin muntasibun bi-tilka
l-nisbati wa ka-qawlin Zayd abun li-Bakr wa Bakr limun fa-Zayd ab limin wa
ka-qawlin A muswin li-B wa B muswin li-J fa-<A muswin> li-musw J wa
A juzun li-B wa B juzun li-J fa-A juzun li-juzi J wa A malzmun li-B wa B malzmun
li-J fa-A malzmun li-malzmin li-J wa al hdh [sic] wa ka-qawlin l-durratu f
l-h uqqati wa l-h uqqatu f l-sundqi fa-l-durratu f l-sundqi wa l-insnu min nutfatin
wal-nutfatu min al-ansiri fa-l-insnu mimm min al-ansiri wa l-lawnu f l-jismi wa
kullu jismin jawharun fa-l-lawnu f-m huwa l-jawharu.
wa amm l-thn wa huwa m laysa bi-l-dhti fa-yaknu mawduhu l-asghara wa
mah mluhu l-akbara ka-qawlin A muswin li-J wa malzmun li-J wa l-durratu f
l-sundqi wa l-insnu min al-ansiri wa l-lawnu f l-jawhari wa innam lam takun
tilka bi-l-dhti liannah law knat li-dhti hdhihi l-qiysti lam takhallafat anh
wa laysa kadhlika liann idh quln A mukhlifun li-B wa B mukhlifun li-J l yunt-
iju aw naqlu A nisfun li-B wa B nisfun li-J wa kadh law naqlu l-insnu min
al-nutfati wa l-nutfatu mi bal h aythu yalzamuhu l-qawlu l-madhkru fa-innam
the challenge of rz and its aftermath 61
In such arguments, the conclusion has the minor term A as its subject,
and the major term equal to J as its predicate. Arguments of this type
are not, Samarqand wrote, formally productive:
Otherwise such compositions would always be productive, and this is
not the case; for if we substitute equality [in the premises] with dissimi-
larity or halfness then there is no productivity.38
The second type of argument has premises of the same form as the
first. However, the conclusion has as its subject the minor term and as
its predicate the relation mentioned in the minor premise plus the
major term. For example:
A is equal to B
B is equal to J
A is equal to what is equal to J
Here, Samarqand noted, the conclusion does follow from the premise-
pair by itself. Other examples are:
Zayd is a father of Bakr
Bakr is a brother of Khlid
Zayd is a father of a brother of Khlid
Yet another example is one of the inferences mentioned by Rz:
The pearl is in the box
The box is in the chest
The pearl is in what is in the chest
Samarqand thus had a clearer sense than Rz that such arguments
are productive by virtue of their form. Unlike Rz, he made no appeal
to the claim that knowledge of the conclusion follows with necessity
from knowledge of the premises. Instead, he implied that the argu-
ments are such that a conclusion of that form is always true when
premises of that form are true. He apparently considered this to be
obvious (zhir). To be sure, he did appeal to the principle that since
the minor premise tells us that the subject stands in a particular rela-
tion to some other thing, and the major premise tells us that the other
39
Samarqand, Sharh al-Qists, fol. 133r. Note that the precise formulation of this
principle is curious. In the inference Zayd is a father of Bakr & Bakr is a brother of
Khlid, so Zayd is a father of a brother of Khlid it does not make sense to say that
Zayd stands in the relation of being a father to the relation of being a brother.
Samarqand appears to have understood the first relation of the syllogism as the
father of and the second relation as a brother of Khlid . Only in this sense can one
say that Zayd stands in the first relation to the second relation.
40
See on this point J. Lear, Aristotle and Logical Theory (Cambridge: Cambridge
University Press, 1980), 37, and R. Smith, Aristotle: Prior Analytics (Indianapolis/
Cambridge: Hackett, 1989), 111.
41
Samarqand, al-Anwr al-ilhiyya (Manuscript Murad Molla 1394, Sleymaniye
Ktphanesi, Istanbul), fol. 269a269b.
the challenge of rz and its aftermath 63
III. Sadr al-Shara (d. 1346) and Qutb al-Dn al-Rz (d. 1365)
A number of scholars who were active in the final decades of the thir-
teenth century or the first half of the fourteenth also wrote extensive
surveys of logic. A number of them had nothing to add to the posi-
tions outlined in the preceding section. Ibn al-Mutta hhar al-H ill
(d. 1325), in his extensive tripartite survey of philosophy al-Asrr al-
khafiyya, reiterated the regimentation of the syllogism of equality
given by his teacher Ts in Tadl al-miyr.42 Shams al-Dn
al-Shahrazr (fl. 1288), one of the earliest known followers of the
anti-Peripatetic Platonist (or Illuminationist) Suhraward, followed
closely (but without acknowledgment) the discussion in Ktibs Jmi
al-daqiq.43 The theologian and jurist Shams al-Dn Mahmd
al-Isfahn (d. 1348) followed the regimentation of Khnaj and
Urmaw.44 Two prominent scholars from the period who did have
something new to say on the topic of relational inferences were: the
Central Asian jurist, theologian, logician, and astronomer Sadr
al-Shara al-Mahbb (d. 1346)45, whose Tadl al-ulm was recom-
mended by the sixteenth-century Ottoman scholar and judge Ahmed
T kprzde (d. 1561) to anyone desirous of reaching the highest
degree of excellence in logic46; and the Persian scholar Qutb al-Dn
al-Rz (d. 1365), whose writings on logic continued to be studied
intensively throughout the Islamic world until modern times.
Sadr al-Shara began his discussion by noting the controversy
surrounding what additional premise is needed to make the premises
A is equal to B & B is equal to J syllogistically produce the conclusion
42
H ill, al-Asrr al-khafiyya f l-ulm al-aqliyya (Qom: Markaz al-Abh th wa
l-Dirst al-Islmiyya, 1421/20002001), 117.
43
Shahrazr, al-Shajara al-ilhiyya f ulm al-h aqiq al-rabbniyya, ed. N. Grgn
(Beirut and Istanbul: Dr Sdir & Dr al-Irshd, 2007), I, 236237.
44
Mahmd al-Isfahn, Sharh Matli al-anwr (Manuscript Yahuda 187, Firestone
Library, Princeton University), fol. 68b69a. See also his Sharh Tawli al-anwr
(Lithograph, Istanbul: Khuls Efendi Matbaas, 1305/18871888), 41, and his Bayn
al-Mukhtasar, ed. M.M. Baq (Mecca: Dr al-Madan, 1986), 36.
45
He is not mentioned in Reschers Register of Arabic Logicians in The Development
of Arabic Logic. On him, see C. Brockelmann Geschichte der arabischen Literatur
(Leiden: E. J. Brill, 19371949), II, 277279; SII, 300301; and A. Dallal, An Islamic
Response to Greek Astronomy: Kitb Tadl hayat al-aflk of Sadr al-Shara (Leiden:
Brill, 1995), 711.
46
T kprzde, Mifth al-sadah wa mifth al-siydah, Ed. Bakr & Ab l-Nr
(Cairo: Dr al-Kutub al-H adthah, 1968), I, 303.
64 chapter two
47
Sadr al-Shara, Tadl al-ulm (Manuscript Landberg 394. Berlin Staatsbibliothek),
fol. 71b (inna qawlan A musw B wa B musw J laysa qiysan bi-l-nisbati il qawlin
A musw musw J liannahu l yasduqu alayhi h addu l-qiysi wa huwa qawlun
qadiyyatni yalzamuhum ukhr dhtan lianna l-murda bi-qawlin yalzamuhum
ukhr an yalzamah qadiyyatun taqtad nisbatan mughyiratan li-l-nisbati l-madhkrati
f l-qadiyyatayni wa f hdhihi l-srati l-nisbatu l-madhkratu f qawlin A musw
musw J aynu l-nisbati l-madhkrati f qawlin A musw B wa B musw J lianna
madlla htayni l-qadiyyatayni bi-l-mutbaqati nisbatu A il B al-mawsfu bi-kawnihi
muswiyan li-J wa qawlun A musw musw J manhu A musw shayin mawsfun
bi-annahu musw J fa-l yadullu ill al l-nisbati l-madhkrati f l-qadiyyatayni fal
taknu hdhihi l-qadiyyatu qadiyyatan ukhr bi-l-nisbati il qawlin A musw B wa B
musw J).
the challenge of rz and its aftermath 65
48
On the later Greek notion of the connector () as a third element in a
categorical proposition, see J. Barnes, Truth, etc.: Six Lectures on Ancient Logic
(Oxford: Clarendon Press, 2007), chapter 3, esp. 192193. Some later Arabic logicians
wished to add a fourth part: the assertion or negation of the nisba. See A.I. Sabra,
Avicenna on the Subject-Matter of Logic, Journal of Philosophy 77 (1980): 746764,
esp. 760761
49
Sadr al-Shara, Tadl al-ulm, fol. 71b (kam yuqlu kullu J B wa badu B A laysa
qiysan bi-l-nisbati il qawlin A thbitun li-badi shayin huwa thbitun li-kulli
J liannahu tikrru man l-qadiyyatayni l-latayni kam yuqlu Zayd huwa ibnu
Amr wa Amr huwa ammu Bakr fa-Zayd ibnu ammi Bakr fa-l-qadiyyatni l-latni
laysat qiysan bi-l-nisbati il hdhihi).
66 chapter two
50
Qutb al-Dn al-Rz, Lawmi al-asrr bi-Sharh Matli al-anwr (Istanbul:
Matbaa-i Amire, 1277/18601861), 179 (fa-inn l naqilu min al-luzmi bi-l
wsitatin ill anna mujarrada l-muqaddimatayni kfin f taaqquli-l-natjati wa min
al-luzmi bi-wsitatin anna taaqqula l-muqaddimatayni l yakf f taaqquli-l-natjati
wa innam yakf maa taaqquli l-wsita).
the challenge of rz and its aftermath 67
to when it is not true, as in the case of being a half or a third. As for the
mediating propositions that they have invented: there is no need for
them, for we cognize the conclusion of the syllogism of equality even if
any of them do not occur to us. Geometers confine themselves to just
stating the premises and obtain from this the desired conclusion as if it
were self-evident because the mediating proposition that says What is
equal to what is equal is equal occurs quickly to the mind from the pos-
iting of the two premises.51
There is a tension in Qutb al-Dns argument. He insisted that Urmaws
additional premise and numerous intermediate inferences are unnec-
essary because the mind can be certain of the conclusion once it has
cognized the premises without considering these additional premises
and steps. However, it is not clear why the very same argument cannot
be used against the need for any additional premise besides the origi-
nal pair A is equal to B and B is equal to J. It has already been seen
that Fakhr al-Dn al-Rz had advanced this very argument in his
commentary on Avicennas Uyn al-h ikma. The premises A is equal
to B & B is equal to J, Rz wrote, produce the conclusion A is equal
to J, for the mind cannot but know that the conclusion is true when it
knows that the premises are true. It is not at all clear how Qutb al-Dn
would have warded off such an argument. After all, he apparently con-
ceded that geometers derive the conclusion A is equal to J from the
premises without consciously adding a third premise to the effect that
What is equal to what is equal is equal. Yet, he believed that the two
original premises of the syllogism of equality had to be supplemented
with an additional premise before they yielded the conclusion A is
equal to J, and was driven to suppose that this additional premise
occurred to geometers but so quickly that they usually did not stop to
think about it.
Qutb al-Dn went on to accept the position of Fakhr al-Dn al-Rz
and Samarqand, namely that sharing a middle-term is not a necessary
51
Qutb al-Dn al-Rz, Lawmi al-asrr, 179 (wa min al-bayyini anna man taaqqala
anna A muswin li-B wa B muswin li-J wa taaqqala anna kulla muswin li-l-musw
muswin taaqqala jazman anna A muswin li-J wa l ihtiyja il takarruri wasatin
lu l-jazmu bi-dhlika l-qawli h aythu tasduqu tilka l-muqad-
qatan wa li-dhlika yah su
dimatu bi-khilfi m idh lam tasduq kam f l-nisfiyyati wa l-thulthiyyati wa amm
l-wasitu l-lat ibtadah faan tawstih ghinan liann nataaqqalu l-matlba min
qiysi l-muswti wa in lam yakhtiru bi-blin shayun minh bal al-muhandisna
yaqtasirna al rdi l-muqaddimatayni wa yastafdna minhum l-matlba kaanna
istilzmahum iyyhu badhiyyun li-nsibqi l-wsitati l-qilati musw l-musw
muswin ila l-dhihni min wadi l-muqaddimatayn).
68 chapter two
52
Qutb al-Dn al-Rz, Lawmi al-asrr, 179.
53
Qutb al-Dn al-Rz, Lawmi al-asrr, 183 (al-shurtu l-mutabaratu f intji
l-qiysi nawn: m huwa shartun li-tah aqquqi l-intji ka-l-shariti l-mutabarati f
l-ashkli-l-arbaati wa m huwa shartun li-l-ilmi bi-l-intji wa takarruru l-wasati
laysa shartan li-l-intji bal li-l-ilmi bi-l-intji idh al-qiysu innam dubita qawiduhu
wa urifa ah kmuhu idh takarrara fhi l-wasat).
54
Qutb al-Dn al-Rz, Tah rr al-qawid al-mantiqiyya bi-sharh al-Risla al-
Shamsiyya (Cairo: Musta f al-Bb al-H alab, 1948), 139; (Cairo: al-Matbaa al-Amriyya
1323/1905), II, 189.
the challenge of rz and its aftermath 69
55
In the section on logic in the encyclopaedia by Sirj al-Dn al-Sakkk (d. 1229)
entitled Mifth al-ulm, the semantic and rhetorical part of which came to be widely
studied and commented on in later centuries, the grammatical terms mubtada and
khabar are used for subject and predicate, respectively. See Mifth al-ulm (Cairo:
al-Matbaa al-Adabiyya, 1317/18991900), 229273. Of course, from a grammatical
point of view, the semantic dependent is the entire prepositional phrase, not just the
substantive governed by the preposition.
56
The same definition of the syllogism of equality is given in the very influential
commentary by H usm al-Dn al-Kt (d. 1359) on Abhars elementary introduction
to logic sghj. See Kt, Qla aqlu (Cawnpore: Matbaa-i Nizm, 1279/18621863),
43. Kt is a little-known scholar whose reported date of death indicates that he was a
contemporary of Qutb al-Dn al-Rz. It is not clear who influenced whom on this
point, or whether they were both following a common source.
CHAPTER THREE
1
N. Rescher, The Development of Arabic Logic, 73.
epitomes, commentaries and glosses 71
the Avicennian corpus with the same critical and independent spirit
with which Avicenna had himself approached the Aristotelian. They
delighted in raising and discussing doubts, and dismissed exegetical
questions concerning what Aristotle or Avicenna had really meant.
Logicians writing in Arabic after around 1350 seem, on the whole, to
have been more conservative in their approach, though there were
exceptions which will be discussed below and in subsequent chapters.
They also tended again with notable exceptions to equate writing
on logic with the exegesis of logical texts. In both respects, they may
be compared to the Greek commentators of late antiquity, and indeed
to the Arabic commentators on Aristotle from Frb to Averroes.
Dismissals of the Greek and Arabic commentators on Aristotle were
once commonplace. Bertrand Russell, in his classic A History of
Western Philosophy (1945), opined that Arabic philosophy is not
important as original thought. Men like Avicenna and Averroes are
essentially commentators.2 Such opinions are no longer taken seri-
ously by specialists, but it is striking that the very same people who
would reject Russells judgment as deeply uninformed have often
made similar armchair dismissals of Arabic writings on logic (and
philosophy in general) after the twelfth or thirteenth century. As
Tony Street has sensibly urged in his recent survey of Arabic logic for
The Handbook of the History of Logic, edited by D.M. Gabbay and
J. Woods:
I am perfectly willing to entertain the possibility that logical production
went through a radical decline in quality at this time [the fourteenth
century]. But it cannot simply be assumed to be the case because the
preferred genre of logical composition came to be the commentary
We simply have to read these texts.3
2
B. Russell, A History of Western Philosophy (Reprint, New York: Simon and
Schuster, 1964), 427.
3
T. Street, Arabic Logic, 526.
72 chapter three
4
There does not appear to be a single commentary on Ktibs Shamsiyya nor
Urmaws Matli by a North African scholar. In contrast, there is a plethora of North
African commentaries on Khnajs Jumal and Sanss Mukhtasar.
5
For example, the Ottoman bibliographer Ktib eleb (d. 1657) did not know of
Sanss Mukhtasar and falsely noted that Khnajs Jumal was an epitome of a work
entitled Nihyat al-amal by Ibn Marzq, not realizing that the latter is in fact a com-
mentary on the Jumal (see Kashf al-zunn, I, 602). He was obviously not well
acquainted with these works.
6
Ibn Khaldn, al-Tarf bi Ibn Khaldun wa rih latihi gharban wa sharqan, ed. by
M. al-Tanj (Cairo: n.p., 1951), 6264. The entry on this figure in Reschers The
Development of Arabic Logic, 217, is unreliable and should be corrected on the basis
of, for example, the detailed entry on him in Ahmad Bb al-Timbukt, Nayl al-ibtihj
bi-tatrz al-Dbj (Cairo: Abbs b. Abd al-Salm b. Shaqrn, 1351/1932), 255264
(margins).
7
This is the text as it appears in the lemma quoted within the commentary on the
work by Khnajs student Ibn Wsil. The lemma quoted in later commentaries is
slightly different.
epitomes, commentaries and glosses 73
(d. 1298), in his commentary on the Jumal, explained that Khnaj had
excised this condition since:
It is as if in reality there is no need for it, for if propositions imply some-
thing by the mediation of an extrinsic premise, then they do not imply it
at all without this extrinsic premise.8
Tilimsn reiterated the point in his own commentary on the Jumal.
He wrote:
His [i.e. Khnajs] saying implies indicates that its implication of the
conclusion should not depend on an extrinsic premise. For if it depends
on an extrinsic premise then that which implies the conclusion is not
what has been [explicitly] mentioned alone but rather the whole consist-
ing of the mentioned composite statement and the extrinsic premise.9
Tilimsn went on to apply this point to the case of the syllogism of
equality. He stated that the syllogism of equality is excluded from the
definition of syllogism by the condition that the premises imply the
conclusion:
This excludes the syllogism of equality, such as our statement A is equal
to B & B is equal to J, for this implies that A is equal to J by the media-
tion of an extrinsic premise, and this is that things that are equal to one
thing are equal.10
In other words, Ibn Wsil and Tilimsn departed from the hitherto
reigning assumption that the premise-pair of the syllogism of equality
implies the conclusion but does not do so by itself. Their position
seems to have been that there is no implication (luzm) that is not
implication by the premises alone. It is not the case that elliptical
arguments with suppressed premises imply a conclusion but not by
themselves. They simply do not imply the conclusion at all. As will be
8
Ibn Wsil, Sharh al-Jumal (Manuscript Landberg 103, Beinecke Library, Yale
University), fol. 29b (kaannahu f l-h aqqati ghayru muhtjin ilayhi fa-inna l-qady
idh stalzamat shayan bi-wsitati muqaddimatin ajnabiyyatin lam takun bi-dni
l-muqaddimati l-ajnabiyyati mustalzimatan aslan).
9
Tilimsn, Sharh al-Jumal (Manuscript Add. 9617. British Library), fol. 77b
(yushiru qawluhu mustalzimun ann l yaknu mutawaqqifan f istilzmihi li-l-natjati
al muqaddimatin ajnabiyyatin fa-innahu in tawaqqafa al muqaddimatin ajnabiyya-
tin lam yakun al-mustalzimu li-l-natjati l-madhkra faqat bal al-majma min
al-qawli l-madhkri maa l-muqaddimati l-ajnabiyya).
10
Tilimsn, Sharh al-Jumal, fol. 77b (fa-yakhruju anhu qiysu l-muswti
ka-qawlin A muswin li-B wa B muswin li-J fa-inna dhlika yastalzimu A muswin
li-J lkin bi-wsitati muqaddimatin ajnabiyyatin wa hiya anna l-ashya l-muswiyata
li-shayin wh idin mutaswiyatun).
74 chapter three
11
Tilimsn, Sharh al-Jumal, fol. 78a78b.
epitomes, commentaries and glosses 75
The later North African scholar Ibn Marzq al-H afd (d. 1439) wrote
a lengthy commentary on Khnajs Jumal that he intended to be a
synthesis of earlier commentaries, specifically those of Tilimsn and
another North African scholar, Sad al-Uqbn (d. 1408).12 Ibn Marzq
reiterated Tilimsns discussion of the syllogism of equality, but he
also engaged with the views of Uqbn. The latter is quoted by Ibn
Marzq as having written:
Their saying of the syllogism of equality that it produces A is equal
to J because what is equal to what is equal is equal is not a verified
statement. It only becomes a syllogism if we say: If we know that B is
equal to J then it is true that Everything that is equal to B is equal to
what is equal to J and that Everything that is equal to what is equal to J
is equal to J. As for the first proposition, it is manifest once we know
the equality of B to J. As for the second proposition, it is even more
manifest.13
This, according to Uqbn, allowed for the regimentation of the syllo-
gism of equality into two syllogisms, the first of which is:
A is equal to B
Everything that is equal to B is equal to what is equal to J
A is equal to what is equal to J
This intermediate conclusion is then conjoined to the second of the
evident propositions posited by Uqbn:
A is equal to what is equal to J
Everything that is equal to what is equal to J is equal to J
A is equal to J
The analysis is the one proposed by Ts in his commentary on
Avicennas Ishrt. Uqbn went on to write that this regimentation is
an effort to reduce the syllogism of equality to an argument in which
there is a standard middle term. Otherwise, he wrote, there would be
little point to the exercise, for the following of the conclusion from
12
On Ibn Marzq and Uqbn, see Timbukt, Nayl al-ibtihj, 293299 and 125126
respectively.
13
Ibn Marzq, Nihyat al-amal bi-sharh kitb al-Jumal (Manuscript Derenbourg
640, Escorial, Madrid), fol. 3b (qawluhum f qiysi l-muswti l-mutaqaddami: innam
antaja A muswin li-J lianna musw l-musw muswin kalmun ghayru muh aqqaqin
wa innam yatimmu qiysan bi-an naqla idh ulima anna B muswin li-J sadaqa
kullu muswin li-B muswin li-musw J wa kullu muswin li-musw J muswin li-J
amm l-l fa-jaliyyatun bada l-ilmi bi-muswti B li-J wa l-thniyatu ajl).
76 chapter three
14
Ibn Marzq, Nihyat al-amal, fol. 3b (qawluhu inna istilzma hdh azharu min
istilzmi l-shakli l-rbii musallamun lkin laysa l-qiysu ibratan amm kna
istilzmuhu li-ghayrihi zhiran bal huwa ibratun an tarkbin khssi n laysa hdh
minhu).
15
Ibn Marzq, Nihyat al-amal, fol. 3a (wa htaraza bihi an qiysi l-muswti
nahwa A muswin li-B wa B muswin li-J fa-innahu yalzamu an hdh l-qawli A
muswin li-J lkin l li-mujarradi taaqquli l-qawli l-muallafi min qadiyyatayni bal
epitomes, commentaries and glosses 77
19
Sans, Sharh al-Mukhtasar, 89 (an yakna l-luzmu li-dhti talfi l-tasdqayni
ay l yaknu bi-wsitati muqaddimatin ajnabiyyatin ay ghayri lzimatin li-ih d
l-muqaddimatayni luzman darriyyan).
20
Sans, Sharh al-Mukhtasar, 89 (lam yuntij hdh l-talfu fi qiysi l-muswti
bi-dhtihi bal bi-wsitati muqaddimatin ajnabiyyatin wa hiya qawlun kullu muswin
li-B fa-huwa muswin li-kulli m yuswhi l-B).
epitomes, commentaries and glosses 79
21
Jarb, Sharh sghj (Tunis: al-Matbaa al-Rasmiyya, 1309/1891), 22.
22
T skprzde, Mifth al-sada, I, 299; Rescher, The Development of Arabic Logic,
220. Reschers estimated date of birth for this scholar (1340) must be amended to
allow for the fact that Ibn Mubrak was a teacher of Jurjn who was born in 1340.
23
This was printed in Kazan in 1902 with the glosses of Jurjn (d. 1413) and Mrz
Jn Bghnav (d. 1586).
80 chapter three
24
Other commentaries on Ktibs Shamsiyya that I have consulted are those
by: Ibn Mutahhar al-H ill (d. 1325), Qutb al-Dn al-Rz (d. 1365), Sad al-Dn
al-Taftzn (d. 1390), Qd Mr H usayn al-Maybud (d. 1504), Allmek Mehmed
Bsnav (d. 1635), and Musta f Mstr (d. 1707).
25
Ibn Mubrak Shh, Sharh al-Shamsiyya (Manuscript Or. 6318, British Library),
fol. 58b (lakna hdh l-nawu min al-talfi muntijan diman wa laysa kadhlika
fa-inna mubyina l-mubyini l yajibu an yakna mubyinan fa-inna l-insna
mubyinun li-l-farasi wa l-farasu mubyinun li-l-ntiqi maa anna l-insna l yubyinu
l-ntiq). The British Library manuscript is anonymous, but the incipit corresponds to
that of another manuscript copy of the work in Topkapi Library, Istanbul, which is
attributed to Ibn Mubrak Shh. See F.E. Karatay, Topkap Saray Mzesi Ktphanesi
Arapa yazmalar katalou (Istanbul: Topkap Saray Mzesi, 1966), III, nr. 6833.
epitomes, commentaries and glosses 81
(1) A is equal to B
B is equal to J
A is equal to what is equal to J
(2) A is equal to what is equal to J
Everything that is equal to what is equal is equal
A is equal to J
In the second inference, according to the objection, the predicate of
the first premise is equal to what is equal to J, whereas the subject of
the second premise is equal to what is equal. Ibn Mubrak Shh dis-
missed the problem with the following words:
We do not concede that it follows that there is no recurrence of the mid-
dle term for what is meant by our statement what is equal to what is
equal is equal is that everything that is equal to what is equal to J is
equal to J.26
As in the case of Ts, what Ibn Mubrak Shh stated to be the missing
premise is not exactly the same as the premise that he actually added
to the original premises. Rather, the two were held to be such that the
former (What is equal to what is equal is equal) validates the latter
(What is equal to what is equal to J is equal to J).
Much more serious, in Ibn Mubrak Shhs view, was the familiar
objection that the first of the two inferences does not have a middle-
term. This, as we have seen, was the objection first raised by Fakhr
al-Dn al-Rz. Ibn Mubrak Shhs response was surprisingly to
abandon entirely the regimentation of the syllogism of equality that he
had just proposed. He instead gave a variant of the regimentation of
Khnaj and Urmaw:
We say: The mentioned syllogism implies A is equal to J not by itself but
by the mediation of our statement Everything that is equal to B is equal
to everything which is equal to B (kullu muswin li-B muswin li-kulli
m yusw B) for, if this is added to our statement A is equal to B , it
produces A is equal to everything that is equal to B. This [in turn]
implies Everything that is equal to B, A is equal to it. Then, if we
transpose our statement B is equal to J into our statement J is equal to
B and make this the minor premise to our statement Everything that is
equal to B, A is equal to it, this produces J, A is equal to it, and this [in
26
Ibn Mubrak Shh, Sharh al-Shamsiyya, fol. 58b (l nusallimu luzma adami
takarruri l-awsati in lam yahtaj il <muqaddimatin> ukhr li-anna l-murda min
qawlin musw l-musw muswin anna kulla m huw muswin lim yusw J
fa-huwa muswin li-J).
82 chapter three
27
Ibn Mubrak Shh, Sharh al-Shamsiyya, fol. 59a (nah nu naqlu: l-qiysu
l-madhkru yalzamuhu A muswin li-J l li-dhtihi bal bi-wsitati qawlin kullu m
huwa muswin li-B muswin li-kulli m yaknu muswiyan li-B fa-innahu idh
indamma il qawlin A muswin li-B yuntiju A muswin li-kulli m yaknu
muswiyan li-B wa yalzamuhu kullu m yaknu muswiyan li-B fa-A muswin lahu
thumma idh qalabn B muswin li-J il qawlin J muswin li-B wa jaalnhu sughr
li-qawlin wa kullu m yaknu muswiyan li-B fa-A muswin lahu yuntiju J A
muswin lahu wa yalzamuhu A muswin li-J wa huwa l-matlbu wa hdh l-baynu
mukarrarun al-h addu l-awsatu f kulli wh idin min qiysayhi fa-l yutawajjahu m
dhakartum alayhi).
28
Fenr, Fusl al-badi f usl al-shari (Istanbul: Matbaat al-Shaykh Yahy,
1289/1872), 59 (bi-sharti an yuwfiqahu mawdu l-kubr li-yah su la amrun mukar-
rarun jmi). Fenr attributed the insight to Khnaj (d. 1248) and Urmaw (d. 1283).
epitomes, commentaries and glosses 83
29
Fenr, Fusl, 60 (wa min hhun ulima anna takarrura l-awsati shartun
li-l-intji f kulli shaklin li-ruji jamihi il l-shakli l-awwal).
30
Fenr, Fusl, 60 (l kam zannahu badu l-afdili min annahu shartun li-l-ilmi
bi-l-intji amm qiysu l-muswti fa-l-h aqqu anna l-awsata mutakarrirun fhi
bi-l-h aqqati li-anna qawlan A muswin li-B wa B muswin li-J wa musw l-musw
muswin f quwwati qawlin A muswin li-musw J wa kullu muswin li-musw J fa-
huwa musw J fa-A muswin li-J).
31
Fenr, Fusl, 60 (wa kawnu taaqquli l-natjati inda taaqquli l-qawli l-awwali
h silan bidni takarruri l-wasati l yunfhi binan al m marra min anna
mulh azata l-shayi l yastad l-tabra anhu).
32
The scant information on this figure in Reschers The Development of Arabic
Logic, 229230, should be supplemented with the entry in T kprzde, al-Shaqiq
al-numniyya f ulam al-dawla al-uthmniyya (Beirut: Dr al-Fikr, 1975), 62.
84 chapter three
33
Acem, H shiya (Manuscript Yahuda 5609, Firestone Library, Princeton
University), fol. 90a90b (in anaytum bi-qawlikum qiysu l-muswti l yuntiju bi-l-
dhti annahu l yuntiju l-natjata l-madhkrata wa hiya qawlun A muswin li-J
fa-dhlika btilun idh luzmuh darryyun wa in anaytum annahu l yazharu
intjuhu ill bi-wsitati muqaddimatin ukhr fa-l nusallimu li-anna intja qiysi
l-muswti min jumlati l-badhiyyti wa li-hdh addahu l-muhandisna maa diqqati
nazarihim min jumlati l-awwaliyyti wa lain sallamn an yahtja il muqaddimatin
ukhr lkin dhlika l yukhrijuhu an al-qiysiyyati wa ill la-wajaba an yukhraja
l-shaklu l-thn wa l-thlithi wa l-rbii an al-qiysiyyati li-htiyjih f l-intji il l-aksi
wa l-khalf wa ghayrihi).
epitomes, commentaries and glosses 85
34
Acem, H shiya, fol. 90b (inna l-natjata l talzamu min qiysi l-muswti
ill bi-mulh azati muqaddimatin ajnabiyyatin h udduh mughyiratun li-h uddi
muqaddimti l-qiysi wa l yaridu l-naqdu bi-l-ashkli l-lat intjuh ghayru bayyinin
lianna luzma l-natjati anh wa in kna bi-wsitatin lkin tilka l-wsitatu laysat
ajnabiyyatan).
35
Acem, H shiya, fol. 91a (iturida alayhi bi-anna takarrura l-wasati laysa
bi-shartin li-intji l-qiysi liannahum sarrah bi-anna qawlan A muswin li-B wa B
muswin li-J yuntiju bi-l-dhti A muswin li-m yusw J bidna hdh l-shart wa
ujba anhu bi-anna l-qiysa l-iqtirn l-h aml wa in kna aamma min an yakna bi-
takarruri l-wasati lkin li-m kna l-qiysu bi-hdh l-sharti dima l-intji malma
l-burhni mabsta l-qawidi wa l-ah kmi bi-khilfi l-bq min al-aqsmi
l-madhkrati fa-innah ghayru malmi l-burhni wa ghayru madbti l-qawidi wa
l-ah km).
86 chapter three
36
The earliest dated manuscript that I have found of his work is from 898/
14921493 (Manuscript Or. 4316, British Library, Folios 3969). The second-oldest
dated manuscript I have found is from 904/1498 (Manuscript 06 Hk 2890/2. www
.yazmalar.gov.tr). Burhn al-Dn is sometimes referred to in manuscript catalogues as
al-Bulghr. The term Bulghr in Islamic languages was primarily used of the Volga
Tatars until the nineteenth century.
37
Bulghr, H shiya ala l-Fanr (Istanbul: n.p. 1290/18731874), 62 (li-anna
intjahu l-salbiyya an qawlan A laysa muswiyan li-J intjun li-dhtihi l bi-wsitati
epitomes, commentaries and glosses 87
predicated in both premises, such as The honey is inside the pot & The
pot is inside the house or Two is half of four & Four is half of eight.
He then proposed a broader definition of syllogism of equality that
would make all such examples instances of one quiddity (mhiyya):
It is possible to explain the syllogism of equality as that in which one
thing is predicated of two things and belongs to them. This would be
more general than this one thing being equality or difference or some-
thing else. In this case, all the examples would be examples of syllogisms
of equality, and this would be more amenable to apprehension.40
Again, Burhn al-Dns proposal seems to have been original. However,
it is no less problematic than his suggestions regarding syllogisms
of equality with a negative second premise. It immediately got him
involved in a discussion of why his proposed definition of a syllogism
of equality is not true of a standard second-figure syllogism such as:
Every J is B
No A is B
Burhn al-Dns proposed definition thus seems to be true of cases
that it obviously should not be true of, including standard second-
figure syllogisms. It also seems to exclude cases that arguably have the
same form as the syllogism of equality. Qutb al-Dns definition of the
syllogism of equality would include, for example, the following two
premises (mentioned by Samarqand in the thirteenth century):
Zayd is a father of Bakr
Bakr is a brother of Khlid
Burhn al-Dns proposed definition, by contrast, would exclude such
premises from constituting a syllogism of equality since the predi-
cates are not identical.
Thus, while Burhn al-Dns discussion of the syllogism of equality
may have been original, it was also seriously flawed. Nevertheless, it is
of some historical interest. As will be seen in subsequent chapters, his
gloss was known to, and invoked by, at least some of the Ottoman
logicians who explored the logic of relational inferences in the seven-
teenth and eighteenth century.
40
Bulghr, H shiya al l-Fanr, 61 (wa yumkinu an yufassara qiysu l-muswti
bi-annahu qiysun yaknu fhi l-shayu l-wh idu mah mlan al l-shayayni wa
thbitan lahum aamma min an yakna dhlika l-shayu l-wh idu huwa l-muswta
aw al-mubyanata aw ghayrahum fa-h naidhin yasru jamu l-amthilati min qabli
qiysi l-muswti wa huwa aqrabu il l-dabt).
epitomes, commentaries and glosses 89
41
Ridaw, H shiya al l-tasdqt, printed with Dawns H shiya al l-tasawwurt
(Istanbul: n.p., 1307/18891890), 24104. I have not seen this edition.
42
According to Brockelmann, Khall al-Ridaw completed his gloss in 759/1358
see Geschichte der arabischen Literatur (Leiden: E.J. Brill, 19371949), I, 613. This
cannot be true, since Ridaw repeatedly cited from Sharh al-Mawqif by al-Sayyid
al-Sharf al-Jurjn (13401413) and from the gloss on that work by Al al-Dn
al-Ts (d. 1473) who was mentioned as being deceased at the time of writing. The
earliest dated manuscript of Ridaws gloss appears to be from 916/1509 (Manuscript
37 Hk 2035/1 listed on www.yazmalar.gov.tr). In the introduction, Ridaw mentioned
his teacher Kaml al-Dn H usayn. Ktib eleb (d. 1657) identified this teacher as
H usayn Ilh Ardabl (Kashf al-zunn, II, 1064), though it seems to me that assuming
that the teacher is H usayn al-Maybud fits better with the chronology, and with the
fact that Ridaw mentioned his teacher as deceased.
43
Ridaw, H shiya al l-tasdqt (Manuscript Yahuda 4317. Firestone Library,
Princeton University), fol. 47a.
90 chapter three
44
His mammoth commentary entitled al-Atwal on the standard handbook of
semantics-rhetoric Talkhs al-mifth and his commentary on al-Risla al-Samarqandi-
yya f l-istirt were widely studied and repeatedly printed in the nineteenth and
early twentieth centuries in Cairo and Istanbul. His extensive and often critical gloss
on the commentary of Jm (d. 1492) on al-Kfiya (on syntax) was also esteemed and
itself elicited a number of super-glosses by later scholars both the gloss and some
of the super-glosses were printed on a number of occasions in Istanbul in the nine-
teenth century. See on him Brockelmann, Geschichte der arabischen Literatur, II,
410411, SII, 571.
45
Ism al-Dn Isfarn, H shiya al l-tasdqt (Istanbul: Matbaa-yi mire,
1259/18431844), 170 (ikhtalaf f tayni hdhihi l-muqaddimati wa tawwal l-kalma
fh wa l yalqu bi-hdh l-maqm).
epitomes, commentaries and glosses 91
One point that Ism al-Dn did see fit to make is of some interest in
the present context. It occurs in connection with a discussion relating
to Qutb al-Dns definition of a syllogism of equality as a two-premise
argument in which the semantic dependent of the predicate of the
first premise is the subject of the second. The definition implies that
the expression the syllogism of equality is true of a whole range of
arguments with this form, whether or not they involve the relation of
equality. The question discussed by Ism al-Dn is why the expres-
sion the syllogism of equality should be used generically for such
types of argument. This question had been discussed by previous
glossators. The above-mentioned Khall al-Ridaw whose glosses
Ism al-Dn seems to have known had written that the term syllo-
gism of equality is used generically because the argument A is equal
to B & B is equal to J, so A is equal to J is a paradigmatic instance of
such argument-forms, and hence it could be applied to a similar argu-
ment that makes no mention of equality. Isam al-Dn instead sug-
gested that such an argument is called a syllogism of equality because:
Its productivity depends upon the equality of two things and their not
differing in their relation to one thing. The productivity of A implies
B & B implies J depends upon implies B and implies what implies B
being equal with respect to the implication of B.46
Ism al-Dns point is compressed, but his point seems to have been
the following: Of the arguments that satisfy Qutb al-Dns definition of
syllogism of equality, some imply their conclusion (even though they
do not do so formally). This is the case, for example, with the follow-
ing argument:
(1) A implies B
B implies J
A implies J
Ism al-Dn noted that the premises imply the conclusion (albeit not
formally) because the relation of implying and implying what
implies do not differ with respect to their relation of implication to J.
In other words, there is no difference with respect to implying J
between implying what implies J and implying J things that fall
46
Ism al-Dn Isfarn, H shiya al l-tasdqt, 169 (lianna intjahu yatawaqqafu
al muswti l-amrayni wa adami l-tafwuti f l-nisbati il amrin fa-inna intja A
malzmin li-B wa B malzmin li-J yatawaqqafu al muswti malzmi J wa malzmi
malzmi J f l-nisbati il J bi-l-malzmiyya).
92 chapter three
into either category equally imply J. Hence, such arguments are called
syllogisms of equality.
Other arguments that satisfy the definition of the syllogism of
equality do not imply their conclusion, such as:
(2) A is half of B
B is half of J
A is half of J
The conclusion does not follow from the premises since being half of
half of B is not equivalent to being half of B.
The point made by Isam al-Dn was perhaps latent in earlier discus-
sions, since it is not difficult to notice that the missing premises in
arguments like (1) have a conspicuously similar form for example,
Everything that implies what implies something implies that thing or
Everything that is equal to what is equal to something is equal to that
thing. However, Ism al-Dn appears to have been the first to have
tried to capture this similarity in more general terms. It is perhaps
tempting to attribute to him the insight that the implication in such
arguments depends on the relation mentioned in the premises being
transitive. However, his point has a more general application. It is, for
example, applicable to the following case too:
A is equal to B
B is distinct from J
A is distinct from J
The conclusion follows (materially not formally) from the premises
since things that are equal to what is distinct from J and things that
are distinct from J are equally distinct from J.
47
For the lives of these two scholars, and a sample of their debates (though not
dealing with logic), see the informative discussion in R. Pourjavady, A Sh Theologian
and Philosopher of Early Safavid Iran: Najm al-Dn H jj Mah md Nayrz and his
epitomes, commentaries and glosses 93
produces Zayd is the brother of the leader of the town. And similarly:
A is equal to B & B is equal to J produces A is equal to what is equal
to J.49
Dawn thus argued that a middle term can recur with addition and
with subtraction without this impugning the syllogistic implication
of the conclusion. As an example of a middle term recurring with
addition he mentioned the example:
The world is composite (al-lamu muallafun)
To every composite there is a composer (li-kulli muallafin
muallifun)
To the world there is a composer (li-l-lami muallifun)
It is clear that Dawn used middle-term (awsat) of the predicate (or
predicate plus semantic dependent) of the minor premise. In the cited
example, the middle-term (in this sense) is composite and recurs in
the second premise with the addition of the preposition to (li-). As an
example of a middle-term recurring with subtraction, he mentioned
the following:
Zayd is the brother of Amr (Zayd akh Amr)
Amr is the leader of the town (Amr rasu l-balad)
Zayd is the brother of the leader of the town (Zayd akh rasi
l-balad)
This second type of argument is familiar from the earlier discussions
of Fakhr al-Dn al-Rz and Shams al-Dn al-Samarqand. The middle-
term (i.e. the predicate plus semantic dependent of the minor premise)
49
(l nusallimu anna l-akbara f l-h aqqati lahu muallifun bal al-akbaru huwa
l-muallafu fa-l-h addu l-awsatu huwa al-muallafu bi-fath i l-lmi wa huwa mukar-
rarun bi-ziydati h arfi l-lmi wa l-mah mlu alayhi maa hdhihi l-ziydati huwa
l-muallifu bi-kasri l-lmi fa-idh asqatn l-mukarrara taadd l-h ukmu bi-l-akbari il
l-asghari bi-wsitati l-lmi wa l nusallimu anna l-h adda l-awsata yajibu takarru-
ruhu min ghayri ziydatin wa nuqsnin bal takarruruhu bi-l-ziydati wa l-nuqsni l
yukhillu bi-l-intji mithlu l-ziydati m marra wa mithlu l-nuqsni Zayd akh Amr
wa Amr rasu l-baladi yuntiju Zayd akh rasi l-baladi wa kadh A muswin li-B wa B
muswin li-J yuntiju A muswin li-musw J). The passage is in Dawn, al-H shiya al-
ajadd (Manuscript Veliyddin Efendi 2026, Veliyddin, Istanbul), fol. 111a111b, and
quoted in Mr Sadr al-Dns counter-gloss (cited below). I have not found this passage
in the manuscripts that I have consulted of the first set of glosses by Dawn, known
as al-H shiya al-qadma. Hence, it seems the point was first made in the second gloss
(al-H shiya al-jadda), that Mr Sadr al-Dn replied to it, and that Dawn defended
himself in his third and last gloss (al-H shiya al-ajadd). An edition of these glosses is
sorely needed and is crucial to the study of later Islamic philosophy and logic.
epitomes, commentaries and glosses 95
is the brother of Amr and does not recur in the second premise in its
entirety; only a part of it does, viz. Amr.
Dawn did not claim that the point he made was unprecedented,
merely that it was contrary to common opinion (khilf al-mashhr).
There is a difference between advancing a minority position and
advancing a position that no logician had ever suggested, and Dawn
was clearly eager not to appear to be doing the latter. He wrote that he
had seen a treatise on the point by a student of Ts, and he invoked
the fact that Fakhr al-Dn al-Rz and Khnaj had both mentioned
that A is equal to what is equal to J follows syllogistically from the
premise-pair A is equal to B & B is equal to J. Paradoxical as it may
seem to modern readers, he was obviously eager not to appear origi-
nal. His reference to Khnaj is curious, for though Khnaj had men-
tioned the suggestion that the premises of the syllogism of equality
produce the intermediate conclusion A is equal to what is equal to J,
he had gone on to reject it. The reference to the treatise by a student of
Ts is also mysterious. It is for one thing curious that he did not men-
tion the name of this student, suggesting that he had read an anony-
mous treatise that he nevertheless thought had been written by a
student of Ts. Furthermore, the idea that Dawn was advancing
went against the position of Ts, and it also went against the position
of the most well-known of Tss students, Ibn al-Mutahhar al-H ill
(d. 1325) and Qutb al-Dn al-Shrz (d. 1311).50 The treatise in question
may perhaps be recovered, and the precise extent of Dawns debt to
it will then be clear. However, as will be seen from what follows, it was
Dawns discussion that came to be the point of departure for later
scholars who took up the issue of syllogisms in which a middle-term
recurs with subtraction or addition.
Dawns position was rejected by Mr Sadr al-Dn al-Dashtak in
his counter-gloss to Dawns second set of glosses. Mr Sadr al-Dn
wrote:
The people of the discipline have shown the necessity of the recurrence
of the middle term in a syllogism by clear and familiar proofs that
should not be doubted. If, by refusing to concede the necessity of the
50
For H ills discussion of the syllogism of equality, see his al-Asrr al-khafiyya
f l-ulm al-aqliyya, 117; al-Qawid al-jaliyya f sharh al-Risla al-shamsiyya,
ed. F.H . Tabrzin (Qum: Muassasat al-Nashr, 1412/1991), 332; al-Jawhar al-nadd f
sharh mantiq al-Tajrd, ed. M. Bidrfar (Qom: Intashrt Bdr, 1423/20022003),
163164. For Qutb al-Dn al-Shrzs views on the topic, see below.
96 chapter three
51
( fa-lianna l-qawma qad bayyan wujba takarruri l-awsati f l-qiysi bi-dalilin
wdih atin mashhratin l yanbagh an yushakka fh fain arda ithbti qiysin
mughyirin lim athbatahu l-qawmu l yaknu l-wasatu fhi mukarraran fa-l yath-
batu dhlika bi-mani wujbi takarruri l-awsati bal alayhi an yujarridahu wa
yubayyinu intjahu bi-wajhin kulliyyin wa l yakf f dhlika irdatu l-intji f badi
l-amthilati li-htimli an yakna badun kharu l yaknu kadhlika). The passage is
quoted as an objection, but without mentioning the name of the objector, by Dawn
in H shiya, fol. 111b. The passage is in Mr Sadr al-Dn, H shiya al Sharh al-Tajrd
(Manuscript Bijapur 172, British Library), fol. 77a, and (Manuscript Bijapur 298,
British Library), fol. 62a62b.
52
Dawn, H shiya, fol. 111b (amm anna l-qawma bayyan takarrura l-awsati f
l-qiysi fa-l nusallimu annahum iddaaw takarrurahu min ghayri ziydatin wa
nuqsnin wa kayfa wa l yusiduhu l-dallu li-annahu idh h fiza f srati l-ziydati
wa l-nuqsni h adhfu l-mukarrari wa naqlu l-h ukmi il l-asghari bi-l-nahwi l-ladh
thabata li-l-awsati l yatakhallafu l-intju qatan mathalan idh qulta hdh l-kitbu
musannafu Zayd wa Zayd rajulun kadh yuntiju hdh l-kitbu musannafu rajulin
kadh wa qis alayhi jama l-mawddi maa l-muh fazati al l-shariti wa qad bayyana
dhlika l-mutaakhkhirna l-muh aqqiqn).
epitomes, commentaries and glosses 97
53
Qutb al-Dn, Tah rr al-qawid al-mantiqiyya bi-sharh al-Risla al-Shamsiyya
(Cairo: al-Matbaa al-Amriyya. 1323/1905), II, 195196; (Cairo: Musta f l-Bb
al-H alab, 1948), 141142.
98 chapter three
This is also like A implies B & B implies J, for this implies by itself A
implies what implies J and by the other [suppressed premise] A implies
J. Such examples [of syllogisms without a middle-term that imply by
themselves] are few, and by comparison what follows from them by the
other [suppressed premises] is more useful in the sciences.54
Interestingly, Dawn was unsure of the author of al-Anwr al-ilhiyya,
and wrote that he thought it was by the author of al-Sah if, a well-
known handbook on philosophical theology. Both al-Sah if and
al-Anwr are in fact by Samarqand.55 Dawn did not invoke
Samarqands Qists or Qutb al-Dn al-Rzs commentary on Urmaws
Matli which is particularly odd given the eagerness with which he
tried to avert the charge that he was advancing a position that was
unheard of.56 It is also noteworthy that Dawn did not repeat his men-
tion of the treatise by Tss student in his response to Sadr al-Dn.
Could he have been referring to Samarqands al-Anwr al-ilhiyya?
Since (i) Samarqand (d. 1303) may well have studied with Ts
(d. 1274) and indicated this in his writings by for example using locu-
tions such as our teacher when mentioning the latter, (ii) Dawn was
unsure of his name, and (iii) Samarqand defended a view similar to
that of Dawn, it is at least worth considering that this is the work to
which Dawn referred in his first discussion of the issue.
54
Dawn, al-H shiya al-ajadd, fol. 111b (wa l-iqtirniyyu imm ghayru mukarrari
l-wasati ka-qiysi l-muswti idh lazimahu li-dhtihi A muswin li-musw J wa bi-l-
ghayri kam marra yan A muswin li-J wa ka-A malzmun li-B wa B malzmun li-J
idh yalzamu li-dhtihi A malzmu malzmi J wa bi-l-ghayri A malzmun li-J wa
amthlu hdhihi qallatun wa hiya bi-l-nisbati il m yalzamuh bi-l-ghayri anfau f
l-ulm). The passage is found in Shams al-Dn al-Samarqand, al-Anwr al-ilhiyya
(Manuscript Murad Molla 1394. Sleymaniye Ktphanesi, Istanbul), sayfa numeras
269270. The manuscript is unpaginated, and hence I give the digital page number
(sayfa numeras) on which the folio can be located on the digital copy in the
Sleymaniye.
55
Miller refers to al-Anwr in his entry on Samarqand, Shams al-Dn in the
Encyclopedia of Islam, but mistakenly suggests that it is his commentary on his own
Qists al-afkr. Another manuscript in the Sleymaniye Ktphanesi in Istanbul
(Laleli 2432) includes al-Anwr al-ilhiyya (fols. 138153) among a number of
Samarqands works.
56
Dawn wrote at least two sets of glosses on al-Sayyid al-Sharf al-Jurjns gloss
on Qutb al-Dns commentary on Urmaws Matli (see bibliography). Dawn also
cited Samarqands Qists in his commentary on Taftzns Tahdhb al-mantiq. He
merely referred to one of the eminent scholars (badu l-fudal), but the identifica-
tion is made by Mr Ab l-Fath in his gloss on Dawns commentary, see Dawn,
Sharh Tahdhb al-mantiq (Istanbul: al-H jj Muharram al-Bsnav, 1305/1887), 97; and
Mr Ab l-Fath, H shiya al Sharh Tahdhb al-mantiq (printed with Dawns com-
mentary), 22. It is therefore very curious that Dawn did not explicitly invoke either
work in defence of his position in his glosses on Sharh al-Tajrd.
100 chapter three
Dawn did not have the last word in his exchange with Mr Sadr
al-Dn. The latters son Ghiyth al-Dn Mansr Dashtak (d. 1542)
reported that his father responded to Dawns point by drawing a dis-
tinction between what is implied by premises and what follows as a
syllogistic conclusion from these premises. Something may be implied
by a syllogism and yet not be a conclusion of the syllogism. A conclu-
sion of a syllogism must be such that knowledge of the conclusion fol-
lows from knowledge of the premises, and not everything that is
implied by the premises is such. For example, A is half of half of J is
implied by the premise-pair A is half of B & B is half of J, but is not a
conclusion of the premise-pair since knowledge of the conclusion does
not follow from knowledge of the premise-pair. He reportedly wrote:
Not everything that is implied by a syllogism must be a conclusion of
the syllogism. This is only the case if knowledge of the syllogism implies
knowledge of it [i.e. the conclusion]. For example, if you say A is half of
B & B is half of J then the truth of this statement implies that A is half
of half of J. However, this need not be the conclusion of the mentioned
statement, since cognizing the mentioned statement does not imply cog-
nizing it.57
Mr Sadr al-Dns point is surprisingly feeble. It in effect amounts to
abandoning his previous (and fair) objection that Dawn had not
shown the formal productivity of the arguments he invoked, and had
merely given examples and claimed that they are formally productive.
Moreover, the distinction he invoked appears blatantly ad hoc and
would necessitate a revision of the generally accepted definition of a
syllogism for no apparent reason except to avoid conceding Dawns
point. Furthermore, it is not clear how one would go about determin-
ing whether or not knowledge of a proposition follows from knowl-
edge of the premises, as opposed to establishing whether a proposition
follows from the premises.
The point about the middle-term of a syllogism occurring with
addition or subtraction also came up in Dawns annotations to
Qutb al-Dn al-Rzs gloss on Tss commentary on Avicennas
57
Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, in Musannaft Giyth Mansr
Dashtak, ed. A. Nourani (Tehran: University of Tehran, 2007), II, 582. ( fa-inna kulla
m yalzamu min sidqi l-qiysi sidquhu l yalzamu an yakna natjata l-qiysi, wa
innam yalzamu dhlika idh lazima min al-ilmi bi-l-qiysi l-ilmu bihi, mathalan
idh qulta A nisfu B wa B nisfu J lazima min sidqi hdh l-qawli an yakna A nisfu
nisfi J, wa l yalzamu an yakna hdh natjatan li-l-qawli l-madhkri idh l yalzamu
min taaqquli l-qawli l-madhkri taaqquluhu).
epitomes, commentaries and glosses 101
58
This is also shown by the fact that when Avicennas Ishrt was printed in
Istanbul in 1290/1873 and in Cairo in 1325/1907, with the commentaries of Fakhr
al-Dn al-Rz and Nasr al-Dn al-Ts, the editions left out the section on logic.
59
Quoted in Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, 579580 (ilam annahu
l yalzamu f l-shakli l-awwali an yakna l-mukarraru tamma mah mli l-sughr bal
rubbam kna juzan min mah mlih aw mutaalliqa dhlika l-mah mli mathalan
idh quln Zayd f l-dri wa l-dru maknun yuntiju Zayd f maknin wa kadh idh
quln kitbu l-Ishrti tasnfu l-Shayki Ab Al wa l-Shaykhu Ab Al huwa Ibn Sn
102 chapter three
yuntiju kitbu l-Ishrti tasnfu Ibn Sn wa idh quln A muswin li-B wa B muswin
li-J yuntiju A muswin li-musw J wa dhlika mutta ridun f jami l-suwar bi-khilfi
m sammawh qiysa l-muswti wa h akam bi-adami intjihi bi-l-dht).
60
Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, 580 (wa kam yajzu an yakna
l-mukarraru nqisan an tammi mah mli l-sughr yajzu kawnuhu zidan alayhi
bi-sharti l-muh fazati al taadd l-h ukmi kam taqlu Zayd insnun wa izmun f
rasi l-insni sabatu durz).
61
Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, 580.
epitomes, commentaries and glosses 103
62
Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, 584.
63
Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, 584. It is possible that Dashtaks
insulting remarks about Dawns ancestors had sectarian connotations. Dawn
claimed descent from Ab Bakr al-Siddq, the first Caliph of Islam, who was reviled
and cursed by the Sh Safavids. Dashtak, by contrast, embraced Shism after the
Safavid conquest of Shrz. See R. Pourjavady, A Sh Theologian and Philosopher of
Early Safavid Iran, ch. I.
64
Ghiyth al-Dn Dashtak, H shiyat al-Ishrt, 584. I have disregarded the editors
punctuation, which renders the statement incomprehensible.
65
See Manuscript 9698, Ayatollah Marash Najaf Library, Qom, and Manuscript
23594, Astn-i Quds-i Radaw Library, Mashhad. The Astn-i Quds-i Radaw manu-
script only includes the opening sections of the work. The Ayatollah Marash Najaf
manuscript lacks the beginning and end of the work, but contains the books in which
one would expect the discussion to occur, namely the book on syllogism and the book
on demonstration. Another manuscript in the Astn-i Quds-i Radaw library that is
listed as being a copy of Tadl al-mzn (Manuscript 5864) turned out on inspection
to be a copy of Dashtaks epitome of Tadl al-mzn, entitled Miyr al-irfn, which
has been published and which does not discuss Dawns point.
104 chapter three
66
Ghiyth al-Dn Dashtak, Tadl al-mzn (Manuscript 23594, Astn-i Quds-i
Radaw Library, Mashhad), fol. 1b.
67
Ghiyth al-Dn Dashtak, Tadl al-mzn (Manuscript 9698, Ayatollah Marash
Najaf Library, Qom), fol. 57a. It is clear that Dashtak is quoting from Bahmanyrs
Tah s l; compare Bahmanyr, al-Tah s l, 109. The parallel is so close that Bahmanyrs
text can be used to correct the lacunose MS text of Tadl al-mzn.
68
On Mrz Jn, see Mrz Muhammad Bqir Khvansr, Rawdt al-jannt f
ahwl al-ulam wa l-sdt (Qom: n.p., 1391/19711972), III, 12; and Brockelmann,
epitomes, commentaries and glosses 105
the last prominent Sunn Iranian philosopher, and left Sh Iran for
Sunni Uzbek-controlled Central Asia where he died. Perhaps helped
by his sectarian affiliation, he had considerable influence in subse-
quent centuries outside of Iran, both through his works and through
his students, several of whom settled in the Ottoman Empire and
Mughal India. Mrz Jn summarized Dawns point about middle
terms recurring with addition or subtraction in a number of works. In
his own gloss on Qshjs commentary on Tajrd al-kalm, he wrote
that Dawn had shown that the original formulation of Tss argu-
ment as The world is composite & To every composite there is a
composer, so to the world there is a composer did not need regi-
mentation:
He [Dawn] does not concede that the middle-term must recur with-
out addition or subtraction. Rather, it is often the case that something
is added to the middle which is repeated in the conclusion, as in our
statement Zayd is the son of Abd-Allh & Abd-Allh is a writer which
produces Zayd is the son of a writer, without straining to reduce it to
one of the well-known syllogisms.69
Mrz Jn reiterated the point in his own widely-studied gloss on Qutb
al-Dn al-Rzs gloss on Tss commentary on Avicennas Ishrt. Mrz
Jn, following Dawns own annotations on Qutb al-Dns glosses,
demurred from the necessity of the regimentation of the following
argument presented by Ts:
The body contains things that are not divided
Everything that is contained in a body and is not divided is
indivisible
The body contains things that are indivisible [i.e. atoms]
Mrz Jn wrote:
It is possible to respond by saying that such a syllogism is one of those in
which the middle-term does not recur in its entirety; rather a part of it
Geschichte der arabischen Literatur, SII, 594. (Note that Brockelmann followed by
Rescher misread Bghnav as Bghand. Bgh Naw is a town near Shiraz.)
69
Mrz Jn, H shiya al Sharh al-Tajrd (Manuscript Add. 7462, British Library),
fol. 75b, and (Manuscript Yahuda 5637. Firestone Library, Princeton University),
fol. 167a (wa yamnau anna l-awsata yajibu tikrruhu bi-aynihi bil ziydatin wa
nuqsnin bal kathran m zda al l-awsati shayun yudu f l-natjati kam f qawlin
Zayd ibnu Abdillh wa Abdullh ktibun yuntuju qawlan Zayd ibnu ktibin min
ghayri tamah h uli wa irjihi il wh idin min al-aqyisati l-mashhra).
106 chapter three
[recurs]. This is like our statement Zayd is the son of Amr & Amr is a
writer with respect to our statement Zayd is the son of a writer.70
In such an argument, Mrz Jn, wrote:
There is no need to notice an extrinsic premise as [one must do in the
case of] the syllogism of equality. Just as an argument in which the mid-
dle recurs in its entirety is evidently productive, so this [arguments evi-
dent productivity] becomes clear by introspection.71
The point that the validity of the relevant relational arguments is
evident which is implicit in Dawns discussion is here made
explicit. Mrz Jn then wrote:
This is the summary of what one of the verifying scholars has noted, in
accordance with what has been mentioned by al-Allma al-Shrz.72
The reference to al-Allma al-Shrz is curious. The term was usually
used of Tss student Qutb al-Dn al-Shrz (d. 1311).73 As was seen
above, Dawn did not mention the name of Tss student whom
he invoked in his discussion with Sadr al-Dn al-Dashtak. The fact
that a second-generation student of Dawn apparently identified this
student as Qutb al-Dn al-Shrz is worth serious consideration, but it
70
Mrz Jn, al-Muh kamt maa H shiyat Mrz Jn (Istanbul: Matbaa-yi mire,
1290/18731874), 54 (margin) (yumkinu an yujba bi-anna mithla hdh l-qiysi
yaknu l-h addu l-awsatu l yatakarraru fhi bi-tammihi bal bi-badihi mithla qawlin
Zayd ibnu Amr wa Amr ktibun bi-l-qiysi il qawlin Zayd ibnu ktibin).
71
Mrz Jn, Al-Muh kamt, 54 (margins) (wa l yuhtju fhi il mulh azati
muqaddimatin ajnabiyyatin ka-qiysi l-muswti. kam anna m yaknu l-wasatu
bi-tammihi badhiyya l-intji fa-kadh mithlu hdh yazharu bi-l-ruji il l-wijdn).
72
Mrz Jn, Al-Muh kamt, 54 (margins) (hdh talkhsu m afdahu badu
l-muh aqqiqna muwfiqan lim dhakarahu l-allmatu l-Shrz).
73
See J. Walbridge, The Science of Mystic Lights: Qutb al-Dn al-Shrz and the
Illuminationist Tradition in Islamic Philosophy (Cambridge: Harvard University Press,
1992), 26. Dawns opponent Sadr al-Dn al-Dashtak was also referred to in some
sixteenth-century sources as al-Allma al-Shrz, which is not as surprising as it may
seem at first, since he was traditionally known as Sadr al-Dn al-Shrz, and it is only
in more recent times that the attributive Dashtak has come to be used to avoid con-
fusion with the later Sadr al-Dn al-Shrz (d. 1635), also known as Mull Sadr. For
example, in H all al-Tahdhb by the Persian-born Ottoman scholar H asan b. H seyn
Emle (fl. 1548) a point is attributed to al-Allma al-Shrz (see Manuscript Laleli
2644, Sleymaniye, Istanbul, fol. 52a.) and this point can be found verbatim in Sadr
al-Dn al-Dashtaks gloss on Qutb al-Dn al-Rzs commentary on the Shamsiyya
(Manuscript Reid Efendi 1015, Sleymaniye, Istanbul, fol. 47a). However, Mrz Jn
could hardly have been referring to Dawns opponent unless one assumes that the
printed text of his gloss is corrupt and that instead of muwfiqan one should read
something like muridan.
epitomes, commentaries and glosses 107
74
Suhraward, H ikmat al-ishrq, 31. I follow the translation of Walbridge and Ziai,
but the Arabic text of their edition seems corrupt at this place, and I have given the
passage as it appears in the edition of Qutb al-Dn al-Shrzs commentary. See Sharh
H ikmat al-ishrq, ed. A. Nourani and M. Mohaghegh (Tehran: Institute of Islamic
Studies, 2002), 136.
75
Qutb al-Dn al-Shrz, Sharh H ikmat al-ishrq, 136.
76
Qutb al-Dn al-Shrz, Durrat al-tj li-ghurrat al-dbj, ed. M. Mishkt (Tehran:
Majlis 19381945), II, 123124, 146147. For the dating of Shrzs works I have
relied on Walbridge, The Science of Mystic Lights, 175 ff.
108 chapter three
As has been seen above, much logical writing in the eastern Islamic
world during the fourteenth, fifteenth, and sixteenth centuries
assumed the form of commentaries and glosses on the thirteenth cen-
tury handbooks of Ktib (the Shamsiyya), Urmaw (the Matli), and
Abhar (sghj).77 New handbooks were sometimes written in the
period, but the only one that seems to have become widely used is
Tahdhb al-mantiq by the extremely influential Tmrid scholar Sad
al-Dn al-Taftzn (d. 1390). In the fifteenth and sixteenth century, a
number of commentaries on this work were written, and some of these
commentaries came to be studied in Islamic colleges up until the
modern period.78 Particularly influential were the commentaries of the
following scholars: (i) Jall al-Dn al-Dawn, which was particularly
esteemed but also incomplete, stopping before the formal sections
dealing with conversion, contraposition, and syllogism; (ii) Ubayd-
Allh Khabs, whose commentary dedicated to the Uzbek ruler of
Central Asia Abd al-Latf Khn (r. 15401552) became a standard
handbook in the Azhar College in Cairo until well into the twentieth
77
It is sometimes mistakenly assumed that, in Iran, H ills commentary on Tss
Tajrd al-mantiq remained the main handbook on logic in later centuries, in contrast
with the Sunn parts of the Islamic world. In fact, Qutb al-Dn al-Rzs commentaries
on Urmaws Matli and Ktibs Shamsiyya were both lithographed in Iran before
H ills commentary on the Tajrd (the former in 1274/1857, the latter in 1295/1878).
When the latter work was lithographed in Iran in 1311/18934, the edition was
prefaced with the statement that this was a work that had been almost completely
forgotten (kna ani l-anzri mastran bal kna kaan lam yakun shayan madhkran).
In the catalogue of manuscripts on logic in the Astan-i Quds-i Radaw Library in
Mashhad, there are 67 copies of Mull Abd Allh Yazds commentary on Tahdhb
al-mantiq and 50 copies of Qutb al-Dn al-Rzs commentary on the Shamsiyya. By
comparison, there are only 7 copies of H ills commentary on Tajrd al-mantiq. See
Barat Ali Gholam Moghaddam, Fihrist-i Kitbkhna-yi stn-i Quds-i Radaw
(Mashhad: Sazmn Kitbkhne-h, 1383/1963 ff ), XXV. I should also point out that
the logic section of Abhars Hidyat al-h ikma was not normally studied. The most
widely studied commentaries on the handbook, those by Ibn al-Sharf al-Jurjn
(d. 1434), Qd Mr H usayn al-Maybud (d. 1504), and Mull Sadr Shrz (d. 1635),
did not comment on the logic part. As mentioned above, the logic section of
Avicennas Ishrt, though it certainly continued to be read, also seems to have fallen
into disuse as a studied handbook by the late fifteenth century.
78
Besides the commentaries mentioned in the text, I have consulted extant
fifteenth- and sixteenth-century commentaries by Muhy al-Dn al-Kfiyaj (d. 1479);
Sayf al-Dn Ahmad al-Haraw, known as H afd al-Taftzn (d. 1511); Shh Mr Hibat-
Allh H usayn (fl. 1546); Abd al-Rahmn Shrz (fl. 1546); Ghnim al-Baghdd
(d. 1620); H asan b. H seyn Emle (fl. 1548); Jaml al-Dn Shahrastn (fl. 1584); Abd
al-H ayy Astarabd (fl. 1508); Mahmd Nayrz (fl. 14971522).
epitomes, commentaries and glosses 109
79
On Yazd, see Mrz Muhammad Bqir Khvansr, Rawdt al-jannt f ahwl
al-ulam wa lsdt (Qom: n.p., 1391/19711972), IV, 228230. Khvansrs entry is
much fuller than that of Muhammad Amn al-Muhibb (d. 1699), Khulsat al-athar f
ayn al-qarn al-h d ashar (Cairo: al-Matbaa al-Wahbiyya, 1284/18681869), IV, 40.
Muhibb gives Yazds date of death as 1015/16061607, as opposed to Khvansrs
981/15731574, but the latter source though later is clearly much better informed
about the life of Yazd and gives details about his studies alongside Mrz Jn Bghnav
with Dawns student Jaml al-Dn Mahmd Shrz (d. 1554).
80
Little is known about Mr Ab l-Fath. He was apparently a student of Isam
al-Dn Isfarn, and finished his completion of Dawns commentary on Tahdhb
al-mantiq in 972/1564. He also completed a commentary entitled Mifth al-bb on the
Sh creedal work al-Bb al-h d ashar by Ibn Mutahhar al-H ill in 975/1567. The
date of death 976/15681569 or 977/15691570 is given in gh Buzurg Tehrn,
al-Dhara il tasnf al-Sha (Najaf: Matbaat al-Gharr, 1936 ff.), VI, 59. Rescher
gives his date of death as 1543 (The Development of Arabic Logic, 248249), which
cannot be correct. This Mr Ab l-Fath, who gave his own name as Ab l-Fath b.
Makhdm al-H usayn al-Arabshh, is often confused in manuscript catalogues with
Ab l-Fath b. Amn al-Sad al-Ardabl, a fifteenth-century scholar who studied with
Qdzde (fl. 1412), one of the prominent astronomers in the circle of Ulugh Beg
(d. 1449) in Samarqand. On this other Ab l-Fath, see Isml al-Baghdd, Hadiyyat
al-rifn: Asm al-muallifn wa thr al-musannifn (Istanbul: Milli Eetim Basimevi,
1951) II, 207.
110 chapter three
81
Mr Ab l-Fath, Takmilat Sharh al-Tahdhb (Manuscript New Series 272,
Firestone Library, Princeton University), fol. 134b (al-murdu bi-l-luzmi huwa
l-luzmu bi-h asabi nafsi l-amri bi-l-nazari il srati l-qawli l-muallafi l-malzmi maa
qati l-nazari an khussi l-mddati kam yadullu alayhi tadhkru l-damri f qawlihi
yalzamuhu li-yada il l-qawli l-muallafi l-mushiri bi-l-hayati l-talfiyyati l il
l-qady l-khliyati anh wa natjatu l-istiqri wa l-tamthli laysat lzimatan lahum
bi-hdh l-man wa dhlika li-takhallufi natjatihim bi-h asabi nafsi l-amri an
sratihim f badi l-mawddi kam f qawlika aktharu l-h ayawnti tuh arriku fakkah
l-asfala inda l-madghi fa-kullu h ayawnin yuh arriku fakkahu l-asfala indahu liannahu
wa in tah aqqaqa hhun l-luzmu l-ilm aw al-zann lkin qad takhallafa l-luzmu
bi-h asabi nafsi l-amr).
82
Mr Ab l-Fath, Takmilat Sharh al-Tahdhb, 135a (wa lidh tastakhlifu l-natjatu
f badi l-mawddi kam f qawlin A nisfun li-B wa B nisfun li-J wa l yalzamu anna A
nisfun li-J).
83
Mr Ab l-Fath, Takmilat Sharh al-Tahdhb, 135a (inna qiysa l-muswti
yakhruju bi-qaydi l-luzmi maa qati l-nazari an khussi l-mddati ka-l-istiqri wa
l-tamthli fa-l h jata il qaydi li-dhtihi ill li-ikhrji l-qiysi l-mubayyani bi-aksi
l-naqd).
epitomes, commentaries and glosses 111
84
Mr Ab l-Fath, Takmilat Sharh al-Tahdhb, 135b (wa l-mashhru f wajhi
wujbi takarruri l-awsati f l-qiysi l-iqtirn anna l-qiysa l budda fhi min amrin
munsibin imm li-majmi l-matlbi aw li-kulli wh idin min tarafayhi fa-l-awwalu
huwa l-qiysu l-istithniyyu wa l-thn huwa l-iqtirniyyu fa-l budda fhi an yakna
li-dhlika l-amri nisbatun il l-asghari wa nisbatun il l-akbari fa-yakna mushtarakan
bayna l-sughr wa l-kubr mutakarriran fhim).
112 chapter three
There is scope for doubt here, and therefore one of the verifying scholars
[Qutb al-Dn al-Rz] has talked about this point in the commentary on
the Matli.85
It is noteworthy that Mr Ab l-Fath, as well as the other sixteenth cen-
tury commentators on Tahdhb al-mantiq, seem to have been oblivious
of the discussion that had unfolded in the glosses on Qshjs com-
mentary on Tajrd al-kalm two or three generations earlier. Dawns
argument that the middle term in a syllogism may recur with addition
or subtraction was far from being perceived at the time as a dramatic
breakthrough. It was one of countless subtle discussions Dawn had
with Mr Sadr al-Dn that were hidden in various sets of demanding
glosses on an already demanding commentary on a handbook of phil-
osophical theology. As will be clear from subsequent chapters, it was
apparently only in Ottoman Turkey, starting in the seventeenth cen-
tury, that his point came to have a decisive and sustained impact on
logical writings.
85
Mr Ab l-Fath, Takmilat Sharh al-Tahdhb, fol. 135b (wa fhi nazarun wa
li-hdh takallama alayhi badu l-muh aqqiqna f Sharh i l-Matli).
CHAPTER FOUR
The Maronite scholar Butrus al-Tlw (d. 1745) was a graduate of the
Maronite College of Rome.1 His sghj, written in 1688, is one of the
first extant writings on logic by an Arabic-speaking Christian trained
in the Western tradition of logic. His work displays, as do the later
works in this tradition, some terminological similarity with that of the
Arabic-Islamic tradition, suggesting that their authors were familiar
with at least some works in that tradition.2 They nevertheless belong
squarely in the Western tradition of logic. For example, they adopt the
Western custom of mentioning the major premise before the minor.
Syllogisms in which the minor premise is mentioned first and these
appear already in Frbs works are described by these authors as
the syllogisms of the Arabs (qiyst al-arab).3 They also seem oblivi-
ous to the post-Avicennian tradition: they ignore or dismiss the fourth
figure of the syllogism, do not recognize combinatorial hypothetical
syllogisms, and do not seem to be aware of the more than a dozen
modality propositions discussed by Rz, Khnaj and subsequent
logicians. The focus of these works is also much less formal than
those of the post-Avicennian tradition. They sometimes discuss the
Aristotelian categories and devote considerable attention to demon-
stration and fallacies topics that tended to be ignored in the post-
Rz tradition of Arabic-Islamic logic.
Tlws introductory work does not deal with the topic of relational
arguments.4 His account of the categorical syllogism is conserva-
tively Aristotelian. A syllogism is a statement composed of statements
which, when conceded, imply by themselves another statement
(qawlun muallafun min aqwlin mat sullimat lazima anh li-dhtih
1
On him, see G. Graf, Geschichte der christlichen arabischen Literatur (Vatican:
Biblioteca Apostolica Vaticana, 19441952), III, 394400.
2
The terminological parallel is far from perfect, though. For example, Tlw uses
the term malzm of that which is implied (for example, the conclusion of a syllogism).
The Arabic-Islamic tradition uses it of that which implies (the antecedent of a true
conditional or the premises of a productive argument).
3
Butrus al-Tlw, al-sghj, ed. A. Rhn et al. (Beirut: Notre Dame University
Press, 2001), 283. On the order of premises in Farb, see J. Lameer, Al-Frb and
Aristotelian Syllogistics, 9193.
4
Tlw also wrote a longer and more advanced work on logic in which he dealt at
greater length with some of the issues raised in sghj. However, as far as I have been
able to ascertain, he did not expand on the problem of relational inferences in that
work. See Tlw, Kitb al-mantiq (Manuscript Or. 4246, British Library).
distinct traditions of arabic logic after 1600 115
qawlun khar).5 Tlw explained that this definition implies (i) that
the acceptance (taslm) of the conclusion must be due to the ordering
of the syllogism, (ii) that the conclusion must be different in form
from the two premises, and (iii) that the productivity must be from
the terms of the premises, and not be in need of an additional term. In
a later section, he wrote that it is a general condition of syllogisms that
there be no more than three terms. If there are more than three
terms, he wrote, then the syllogism is corrupt (fsid).6 Though he
acknowledged the existence of hypothetical syllogisms i.e. the Stoic
schemata of propositional logic like modus ponens he may well have
thought that the condition that a syllogism must have no more than
three terms applied to these as well, because his examples of such
hypothetical syllogisms involved no more than three terms for
example:7
If Zayd walks, then he moves
He walks
He moves
The Greek Catholic scholar Yuwkm Mutrn (d. 1772) wrote more
extensive and detailed works on logic in which he did discuss the
problem of the syllogism of equality.8 He gave the same definition of
syllogism as Tlw, attributing it to Aristotles Topics.9 He added that
the condition that the conclusion follow from the premises by them-
selves was to rule out (i) cases in which the conclusion follows due
5
Tlw, sghj, 183. Note that sullimat has been rendered as sulibat by the copyist
of the manuscript on which the edition is based. Despite the wording of his definition,
Tlw added that the conclusion is part of the syllogism (wa talfuhu min muqaddi-
matayni wa malzm).
6
Tlw, sghj, 219.
7
Tlw, sghj, 249.
8
On Yuwakm Mutrn, see Graf, Geschichte der christlichen arabischen Literatur, III,
210214. The following is based on his al-dh t al-nutqiyya f l-usl al-mantiqiyya
(Manuscript Islamic Third Series no. 138b, Firestone Library, Princeton University)
and his al-Sah fa al-abqariyya f l-usl al-mantiqiyya (Manuscript Garrett 488H,
Firestone Library, Princeton University), fol. 44b45b. The former manuscript is not
paginated. The relevant discussion is in the second chapter (al-bb al-thn) of the
third part (al-qism al-thlith) entitled On the definition of the syllogism (f h add
al-qiys).
9
In the older Arabic translation of the Topics by Ab Uthmn al-Dimashq (fl. 914),
a syllogism is defined as qawlun idh wudiat fhi ashyun lazima min tilka l-ashyi
l-mawdati shayun kharu ghayruh bi-l-idti rr, see Badaw (ed.), Mantiq Arist, II,
489. The definition given by Yuwkm is by comparison a loose paraphrase of the
Greek of Aristotles Topics I, 100a2527.
116 chapter four
10
Yuwkm Mutrn, al-Sah fa al-abqariyya, fol. 45b (qla Arst rasu l-mashiyyna
anna l-qiysa l yatarakkabu ill min thalthati h uddin faqat fa-l-awwalu wa l-thn
hum l-h addni l-ladhni yubh athu an iqtirnihim wa ittih dihim wa tanfhim
distinct traditions of arabic logic after 1600 117
(ii) Arguments like the syllogism of equality, which only implies the
desired conclusion by means of the suppressed premise What is equal
to what is equal is equal. Despite stating that this is the additional
premise, Ys went on to explicate, without any comment, Sanss
complicated derivation of the conclusion on the basis of the additional
premise What is equal to B is equal to everything which B is equal
to.17 As seen on a number of earlier occasions, what is stated to be the
14
On Ys, see Muhammad b. al-Tayyib al-Qdir, Nashr al-mathn li-ahl al-qarn
al-h d ashar wa l-thn, ed. H ajj and Tawfq (Rabat: Maktabat al-T lib, 19771986),
III, 2549.
15
Ys, Nafis al-durar f h awsh l-Mukhtasar (Manuscript Garrett 485H, Firestone
Library, Princeton University), 131b132a.
16
Taftzn, Sharh al-Shamsiyya (Lithograph, Lucknow: al-Matba al-Ysuf, 1317/
1899), 56.
17
Ys, Nafis al-durar, fol. 132b.
distinct traditions of arabic logic after 1600 119
18
On Ibn Sad, see M. Makhlf, Shajarat al-nr al-zakiyya f tabaqt al-Mlikiyya
(Cairo: al-Matbaa al-Salafiyya, 19301931), I, 350.
19
Ibn Sad, H shiya al Sharh al-Khabs (Cairo: al-Matbaa al-Azhariyya al-
Misriyya, 1318/19001901), 272.
20
Sans, Sharh al-Mukhtasar, 96.
120 chapter four
21
Ys, Nafis al-durar, fol. 139b140a (MS 485) (laysa l-shaklu l-thn mutawaq-
qifa l-intji al l-muqaddimati l-khrijiyyati tawaqqufa <qiysi> l-muswti lianna
qiysa l-muswti l yuntiju ill inda l-ityni bi-tilka l-muqaddimati wa amm
l-shaklu l-thn fa-huwa wa in knat tilka l-muqaddimatu wajha intjihi ghayru muta-
waqqifin f l-intji al l-ityni bi-h li-fahmihim muqatadh min muqaddimtihi wa
farqun bayna tawaqqufi shayin al shayin wa bayna wujdihi fhi wa ill fa-l-awwalu
l-ladh huwa abyanuh shaklan wa aqrabuh taban mabniyyun al muqaddimatin
khrijatin wa hiya anna lzima l-lzimi lzimun wa lkin l yatawaqqafu alayh
li-mafhmiyyatih min al-muqaddimti darratan).
22
On Hill, see Qdir, Nashr al-mathn, IV, 143151.
distinct traditions of arabic logic after 1600 121
thing, is itself subsumed under that third thing (wa hiya anna l-
mundarija tahta l-mundariji mundarijun). Hill wrote in response:
We do not concede that this premise is extrinsic; rather it is what one
understands from the form of the syllogism. It is an explication of the
fact that the conclusion is implied by the form of the syllogism in any
material instance that we consider, not by another premise. If you say,
Every human is an animal & Every animal is a body, the fact that the
minor term, which is subsumed under the middle-term, is subsumed
under the judgment in the major premise is something that is gathered
from the two premises, not from anything extraneous.23
The distinction that Ys and Hill emphasized between a sup-
pressed premise and the rule of inference that a valid argument exhib-
its is an important insight. It does, however, imply the odd position
that first-figure syllogisms embody the evidently true logical principle
If A is subsumed under B & B is subsumed under J, then A is sub-
sumed under J, but that if one were to formulate this principle as a
syllogism, then the result is invalid:
A is subsumed under B
B is subsumed under J
A is subsumed under J
Ys and Hill were obviously constrained by the (not unreasonable)
assumption that it is possible to construct an argument with the same
form which has true premises and false conclusion:
A is half of B
B is half of J
A is half of J
The idea of the formal invalidity of such arguments was clearly
expressed by Yss star-pupil Ibn Yaqb al-Walll (d. 1716) who was
also the prize-scholar of the powerful Moroccan Sultan Mawly
Isml (r. 16721727).24 Ibn Yaqb gained a reputation for his pellucid
23
Hill, al-Zawhir al-ufuqiyya bi-sharh al-Jawhir al-mantiqiyya (Lithograph, Fez:
n.p., 1313/1895). The work is not continuously paginated, but is divided into quires of
eight pages. The passage is on page 3 of the 33rd quire (l nusallimu annah ajnabiyy-
atun bal hiya mafh mu srati l-qiysi fa-huwa baynun li-kawni l-natjati lzimatin
li-srati l-qiysi f ayyi mddatin furida l li-muqaddimatin ukhr fa-idh qulta kullu
insnin h ayawnun wa kullu h ayawnin jismun fa-indirju l-asghari l-mundariji f
l-awsati f h ukmi l-kubr makhdhun min htayni l-muqaddimatayni l min khrij).
24
On Ibn Yaqb, see Qdir, Nashr al-mathn, III, 229233.
122 chapter four
25
His commentary on Taftzns Maqsid (on philosophy and theology) entitled
Ashraf al-maqsid f sharh al-Maqsid, and on Qazwns Talkhs al-Mifth (on
semantics-rhetoric) entitled Mawhib al-fatth f sharh Talkhs al-Mifth were both
printed in Cairo in the late nineteenth and early twentieth centuries.
26
Ibn Yaqb, al-Qawl al-musallam f tah qq man l-Sullam (Manuscript Add.
9622, British Library), 42b (wa kharaja bi-kawni l-istilzmi li-dhti l-qiysi ay li-sratihi
m l yastalzimu li-l-dhti bal bi-wsitati qawlin ajnabiyyin an srati l-qiysi wa huwa
l-qiysu l-ladh kna l-h addu l-wasatu fhi mutaalliqa mah mli l-sughr l nafsa
l-mah ml).
27
Ibn Yaqb, al-Qawl al-musallam, fol. 42b (lianna l-tarkba l-madhkra law antaja
lam yatakhallafu f mddatin sadaqa fh wa laysa kadhlika fa-innaka idh qulta
l-arbaatu nisfu l-thamniyati wa l-thamniyatu nisfu l-sittata ashara lam yuntij
al-arbaatu nisfu l-sittata ashara maa annahu al srati qiysi l-muswti lianna
l-wasata fhi mutaalliqu mah mli l-sughr wa lihdh quln anna l-mutabara
l-sratu l l-mddatu fa-l-mutabaru huwa kawnu l-srati mat sullimat lazima
l-qawlu).
distinct traditions of arabic logic after 1600 123
fact that he did not mention this in any of his three works on logic that
I have consulted suggests that he retained the older view held by
Tilimsn and Sans namely, that such examples are in no way dif-
ferent from elliptical arguments with suppressed premises.28
In his commentary on Sanss Mukhtasar, Ibn Yaqb gave an
account of how to derive the conclusion of the syllogism of equality
from the original premises and the extrinsic premise What is equal to
what is equal is equal. He quietly departed from the regimentation
given by Sans and explicated by Ys, and instead amalgamated the
original premises into A is equal to what is equal to J and conjoined it
to the additional premise Everything that is equal to what is equal to
J is equal to J.29 Ibn Yaqb did not address the problem first raised in
the twelfth century by Fakhr al-Dn al-Rz, namely that the premises
of a syllogism of equality imply by themselves the intermediate con-
clusion A is equal to what is equal to J. I have not managed to find a
discussion of this point by any North African logician from the seven-
teenth, eighteenth, or nineteenth century. One reason for this is pre-
sumably that the relevant works by Fakhr al-Dn al-Rz, Samarqand,
Qutb al-Dn al-Rz, and Dawn appear not to have been studied
in North Africa during the early-modern period. The frame of refer-
ence for scholars like Ys, Ibn Yaqb, and Hill appears to have
consisted of fourteenth-century North African commentaries on
Khnajs Jumal, the works of Sans, and the commentaries on Ktibs
Shamsiyya by Qutb al-Dn al-Rz and Taftzn.30 And none of these
works raise this particular point. The later logical works of Muhammad
b. al-H asan al-Bannn (d. 1780), Ibn Krn (d. 1812), and Al Qass ra
(d. 1843), all lithographed in Morocco during the late nineteenth cen-
tury, are relatively introductory and did not have anything to add to
the discussions of Ys, Ibn Yaqb and Hill.
28
The three works on logic by Ibn Yaqb that I have seen are his commentaries on
Akhdars Sullam and on Sanss Mukhtasar, as well as a commentary on his own
didactic poem on logic entitled al-Lmiyya (extant in the Azhar Library in Cairo,
Mantiq 2576/96175).
29
Ibn Yaqb, Lawmi al-nazar bi-tah qq man l-Mukhtasar (Manuscript no. 2307.
Khuda Bakhsh Oriental Library, Bankipore, India). The manuscript is not paginated.
The passage occurs in the commentary on Sanss definition of syllogism.
30
In the late eighteenth and early nineteenth century, one begins to encounter
references in North African works to Dawns incomplete commentary on Tahdhb
al-mantiq and to Qutb al-Dns commentary on the Matli. However, I have not come
across any such references in the works of Ys, Ibn Yaqb, or Hill.
124 chapter four
31
See my Was there a revival of logical studies in eighteenth-century Egypt? Die
Welt des Islams 45 (2005): 119.
32
On Siylkt, see Abd al-H ayy al-H asan, Nuzhat al-khawtir wa bahjat al-
masmi wa l-nawzir (Hyderabad, Deccan: Majlis Dirat al-Marif al-Uthmniyya,
1955), V, 210211. His gloss on Sharh al-Shamsiyya was printed in Cairo in 1905 and
in Istanbul in 2 volumes in 1902.
33
Siylkt, H shiya al Sharh al-Shamsiyya (Cairo: al-Matbaa al-Amriyya, 1905),
II, 189; H shiya al l-tasdqt (Istanbul: irket-i Sahafiye- Osmniye Matbaas,
1320/19021903), 127.
distinct traditions of arabic logic after 1600 125
34
Siylkt, H shiya al Sharh al-Shamsiyya, II, 194; H shiya al l-tasdqt, 130.
35
On Bihr, see H asan, Nuzhat al-khawtir, VI, 250252.
126 chapter four
36
Qutb al-Dn, Lawmi al-asrr, 179.
37
On the two scholars, see H asan, Nuzhat al-khawtir, VI, 7475 and VI,
247248.
38
Gpamv, Sharh Sullam al-ulm (Kazan: al-Matbaa al-Malakiyya, 1887),
278279.
39
Sandl, Sharh Sullam al-ulm (Dehli: Matbaat al-Mujtab, 1328/19101911),
188 (l burhna lahum f shtirki l-h addi l-awsati bi-tammihi fainnahu l shubhata f
anna muqaddimatay qiysi l-muswti mustalzimatni bil wsitatin li-qawlin A
muswin li-muswin li-J fa-l budda min al-iltizmi wa l-itirfi bi-anna takarrura
l-wasati bi-tammihi laysa darriyyan f l-qiys).
40
Cairo: Matbaat al-Sada, 13271328/19091910 and Lucknow: Nuval-i Kishr,
1299/1882. On Mull Mubn, see H asan, Nuzhat al-khawtir, VII, 403404.
distinct traditions of arabic logic after 1600 127
41
Mull Mubn, Mirt al-shurh , II, 355 (Lucknow ed.); II, 178 (Cairo ed.) (f
l-qiysi l budda min takarruri l-awsati lianna l-mashhra baynahum anna kulla
qiysin iqtirniyyin murakkabun min muqaddimatayni yashtarikni f h addin wa laysa
f qiysi l-muswti hdh l-ishtirku idh mawdu l-kubr huwa mutaalliqu mah mli
l-sughr fa-qawlun A muswin li-B wa B muswin li-J al-mutakkariru fhi huwa B wa
l-mah mlu muswin li-B fa-lam yaqa m kna mah mlan f l-sughr mawdan f
l-kubr fa-lam yatakarrar al-awsat).
42
Mull Mubn, Mirt al-shurh , 2: 355 (Lucknow ed.); 2: 178 (Cairo ed.) (l budda
f l-qiysi min tikrri l-awsati bi-h aythu yatah aqqaqu l-indirju f l-jumlati wa amm
takarruruhu bi-tammihi l yadullu alayhi dallun wa l bayyinun wa l mubayya-
nun wa l shakka anna muqaddimatay qiysi l-muswti wa hum A muswin li-B
wa B muswin li-J yastalzimni bil wsitati amrin kharin li-l-natjati wa hiya A
muswin li-muswin li-J fa-ulima anna l-tikrra laysa darriyyan bal qad yuntiju
bidnihi).
43
At first sight, it may appear that the affirmativeness of the minor premise is not a
condition for subsumption, for it may appear that A is not equal to B & B is equal
to J implies by itself the conclusion A is not equal to what is equal to J. However,
128 chapter four
transmits to the minor premise in the sense that its predicate can take
the place of the recurring term B, thus producing A is equal to what
is equal to B.
There is a tension in Bihrs presentation, a tension that his com-
mentators did their best to smooth out by speaking of the premises
of the syllogism of equality as constituting a syllogism with respect
to one conclusion and yet not a syllogism with respect to another con-
clusion. He wrote that the syllogism of equality is excluded from the
definition of syllogism by the stipulation that the premises must imply
the conclusion by themselves. At the same time, he defined the syllo-
gism of equality as an argument in which the semantic dependent of
the predicate of the first premise is the subject of the second premise.
However, this definition is equally true of the following argument
which he explicitly conceded to be formally productive:
A is equal to B
B is equal to J
A is equal to what is equal to J
Insofar as Bihr accepted the prevalent definition of the syllogism of
equality, he had no grounds for excluding it from being a syllogism. It
is clear that, notwithstanding lip-service to the standard definition of
the syllogism of equality, Bihr in practice understood the term in
the narrower sense of arguments with the following form:
A is equal to B
B is equal to J
A is equal to J
The tension would seem to be rooted in two different points made
by Qutb al-Dn al-Rz in the fourteenth century. On the one hand, he
had defined the syllogism of equality in terms that make no reference
to the conclusion, i.e. as an argument in which the semantic dependent
of the predicate of the first premise is the subject of the second. On the
other hand, he had defended the view that the premises A is equal to
B and B is equal to J evidently produce by themselves the interme-
diate conclusion A is equal to what is equal to J. Strictly speaking, the
two insights should lead to acknowledging the syllogism of equality
as formally productive, and hence there should be no need to exclude
it from the definition of the syllogism.44 As will be seen in the follow-
ing chapters, this was a point with which Ottoman Turkish logicians
struggled during the seventeenth and eighteenth centuries, and they
eventually amended Qutb al-Dns definition of the syllogism of equal-
ity. In the Indo-Islamic tradition of logic there seems to have been no
generally accepted revision of the definition of the syllogism of equal-
ity, and the resulting tension is very much in evidence in eighteenth
and nineteenth century works.
An anonymous commentary on Abhars sghj that was widely
studied in Indo-Islamic circles in the eighteenth and nineteenth cen-
tury was known as Mr sghj.45 The commentator gave the stand-
ard definition of the syllogism of equality: an argument in which
the semantic dependent of the predicate of the first premise is the sub-
ject of the major. He stated that such arguments are excluded from the
definition of syllogism by the condition that the implication should
be by the premise-pair itself. The premise-pair A is equal to B & B is
equal to J implies the conclusion A is equal to J but it does not do
so by itself, he noted, adducing a counter-example with the same
form: A is different from B & B is different from J, so A is different
from J. This is a thoroughly conventional account that is repeated in
44
This appears to be the point made by Mull Muhammad H asan Lakh nav (d. 1755)
in his Marij al-ulm, a longish handbook on logic inspired by Bihrs Sullam but
apparently not widely studied in later times. See Manuscript Delhi Arabic 1519,
British Library, fol. 52b. Unfortunately, the text of the relevant passage in this manu-
script is clearly very corrupt and therefore defies any firm interpretation. On Mull
H asan, see H asan, Nuzhat al-khawtir, VI, 296298. He also wrote an esteemed com-
mentary on Bihrs Sullam. However, all the lithograph editions I have seen of that
commentary only include the first part, covering conceptions (tasawwurt).
45
The commentary was attributed by later Indian scholars to al-Sayyid al-Sharf Al
al-Jurjn (d. 1413), but this is probably a case of misattribution. There are no refer-
ences outside of India to Jurjn having written such a commentary. Had such a prom-
inent scholar written a commentary on sghj then surely other parts of the Islamic
world would have taken notice. It is possible that the author was al-Sayyid al-Sharf s
great-grandson Mr Shams al-Dn Muhammad Jurjn, who is known to have been
active in India in the late fifteenth and early sixteenth century.
130 chapter four
46
Lakh nav, Abd al-H ayy, H shiya al Bad al-Mzn (Lucknow: al-Matba
al-Ysuf, 1311/18931894), 85 (margin); Tahnaw, Kashshf mustalah t al-funn
(Calcutta: W.N. Lees Press, 1862), II, 1192; Ahmadnagr, Dustr al-ulam
(Hyderabad, Deccan: Matbaat Dirat al-Marif al-Uthmniyya, 1329/1911), III, 109.
47
It is possible that he is identical to Ghulm Muhammad b. Wal-Allh b. Ghulm
Muhammad Srat (d. 1824), see H asan, Nuzhat al-khawtir, VII, 360.
48
Ibn Ghulm Muhammad, H shiya-yi jadda-yi Mr sghj (Lithograph, Lahore:
Matbaat Shihb Thqib, n.d.), 133.
distinct traditions of arabic logic after 1600 131
49
On Fazl Imm, see H asan, Nuzhat al-khawtir, VII, 374. The handbook on which
he commented must have been written before the sixteenth century, for the standard
commentary on it, entitled Bad al-mzn, was written by Abd-Allh Tulanb [or
Talnab] (d. 1516). I have considered Siylkt rather than Tulanb/Talnab to have
been the first prominent representative of the Indo-Muslim tradition of Arabic logic,
since the latters work is a short, explanatory commentary on a relatively introductory
handbook.
50
Fazl Imm Khayrabd, Tashh dh al-adhhn bi-sharh al-Mzn (Lithograph,
Tamasganj: n.p., 1869), 72.
51
Yazd, H shiya al Tahdhb al-mantiq (Beirut: Muassasat Ahl al-Bayt, 1988),
141.
132 chapter four
52
Fazl Imm Khayrabd, Mirqt (Lithograph. Delhi: n.p., 1886), 19; (Lithograph.
Lahore: al-Maktaba al-Qdiriyya, 1978), 88.
53
Abd al-H aqq Khayrabd, Sharh -i Mirqt (Lithograph. Cawnpore: Matbaa-i
Intizami, 1333/19141915), 175; (Lithograph. Rampur: n.p., 1866), 306. On Abd
al-H aqq Khayrabd, see H asan, Nuzhat al-khawtir, VIII, 222224.
54
Abd al-H aqq Khayrabd, Sharh -i Mirqt, 176177 (Cawnpore ed.); 308309
(Rampur ed.) (wa l-h addu l-awsatu l yajibu an yatakarrara bi-tammihi f jami
distinct traditions of arabic logic after 1600 133
Abd al-H aqqs explicit reference shows that Dawns point was not
lost on later Indian scholars. The fact that he merely summarized
Dawnis discussion and then moved on to other topics also suggests
that there had been no significant development of this particular point
in the later Indian tradition.
56
Mull Sadr, Talqt, 98 (margins) (hhun bahthun wa huwa anna l-h adda
l-awsata qad l yatakarraru fa-inn idh quln A muswin li-B wa B muswin li-J
yuntiju fa-A muswin li-J wa l-mutakarriru hhun laysa h addan f l-muqaddimatayni
bal juzu h addin min ih dhum wa h addun tmmun min al-ukhr wa l-jawbu anna
kullan min hdhihi l-aqyisati inda l-tah qqi qiysun murakkabun min qiysayni wa
qad intaw fhi natjatu l-awwali wa sughr l-thni wa kubrhu l-lat hiya muqaddima-
tun khrijiyyatun tandammu maah hkadh A muswin li-B wa B muswin li-J
yuntiju fa-A muswin <li-muswin> li-J thumma najaluh sughr qawlin A muswin
li-muswin li-J wa kullu muswin li-muswin li-J muswin li-J fa-A muswin li-J).
57
Mull Sadr, Talqt, 98 (margins) (inna l-h adda l-awsata l yajibu an yatakar-
rara kullan f jami l-mawdii bal f mawdiin yaknu h amlu l-akbari nafsihi al
l-asghari majhlan matlban fa-yahtju il h addin jmiin amm idh kna l-majhlu
sidqa badihi aw ma yazdu alayhi ala l-asghari fa-l budda min takarruri shayin
yarbitu l-akbara bi-tibri dhlika l-shayi bi-l-asghari wa lam yaqum al-burhnu al
anna l-h adda l-awsata yajibu takarruruhu min ghayri ziydatin wa l nuqsnin bal
takarruruhu bi-l-ziydati wa l-nuqsni l yukhillu bi-l-intj).
distinct traditions of arabic logic after 1600 135
58
Mull Sadr, Talqt, 98 (margins) (fa-l-h addu l-awsatu huwa l-muallafu wa
qadd takarrara bi-ziydati h arfi l-lmi fa-idh asqatn l-mukarrara taadd l-h ukmu
bi-l-akbari il l-asghari bi-wsitati h arfi l-lm).
59
Mull Sadr, Talqt, 98 (margins) (fainna al-majhla hhun l-natjatu bi-
tibri juzihi an ktib <fakawnu> Zayd akhan li-shaksin mina l-ashkhsi kaannahu
malmun wa huwa l-mah mlu innam l-matlbu huwa kawnu m udfa ilayhi
l-mah mlu ktiban fayustah sa lu dhlika l-matlbu bi-tah s li amrin yajalu l-ktiba bi-
takarrurihi bi-h aythu yudfu ilayhi dhlika l-mah mlu wa dhlika l-amru huwa Amr
fa-idh takarrara wa asqatn dhlika l-mutakarrara taadd l-h ukmu bi-l-akbari maa
qaydihi ala l-asghari bi-wsitatihi).
136 chapter four
60
Mull Sadr, Talqt, 98 (margins) (wa hdh wa in kna mukhlifan lim huwa
l-mashhru ill annahu h aqqun haqqun bi-l-tasdqi wa m dhakarnhu mutta ridun f
jami l-mawdd).
61
Mull Sadr, al-Tanqh f l-mantiq, in Majmah-i rasil-i falsaf-yi Sadr al-
Mutaallihn, ed. H mid Nj Isfahn (Tehran: Hikmet Publishing, 1999), 217.
distinct traditions of arabic logic after 1600 137
and the emphasis on the implication being formal had been stressed
in Mr Ab l-Faths commentary on it. Mull Sadr explained that the
condition that the conclusion be implied ruled out induction and
analogy. As for the syllogism of equality, Mull Sadr wrote that there
was no need for an explicit condition to rule it out, since it is a com-
plex syllogism (qiys murakkab) rather than a simple syllogism (qiys
bast). This would seem to be in line with the view expressed in Mull
Abd-Allh Yazds influential sixteenth-century commentary on
Taftzns Tahdhb, according to which the premises of the syllogism
of equality without the additional premise are sterile, and with the
additional premise constitute a complex argument that can be resolved
into two syllogisms. Mull Sadr went on to state that in a simple cat-
egorical syllogism there are three terms: the recurring middle-term
and the two extremes. This is not really consistent with the insight
expressed in his own gloss on the commentary on H ikmat al-ishrq,
for on that account the premise-pair A is equal to B & B is equal to
J constitutes a simple categorical syllogism with respect to the con-
clusion A is equal to what is equal to J, and yet does not have three
terms.
Very little is known of Iranian writings on logic between Mull
Sadr (d. 1635) and the nineteenth-century scholar Mull Hd
Sabzavr (d. 1878). It is striking that, in the nineteenth and early
twentieth centuries, a number of Arabic works on logic written by
seventeenth- or eighteenth-century North African, Egyptian, Ottoman
Turkish, and Indo-Muslim scholars were printed or lithographed. By
contrast, no work on logic written by an Iranian scholar active after
Mull Sadr and before Sabzavr appears to have been lithographed
or printed during this period.62 This seems to indicate that no Iranian
work written in the seventeenth and eighteenth centuries gained suffi-
cient influence or recognition to be widely studied in the nineteenth
century when the printing press was introduced. This, in turn, means
that it is much more difficult to get a sense of the Iranian tradition of
logic in this period. This gap in our knowledge means that the survey
that follows is particularly tentative.
62
An exception may be an Iranian lithograph edition from 1323/19051906 of
Yazds commentary on Tahdhb al-mantiq, which is reported to have included a
number of glosses on the work by seventeenth, eighteenth and nineteenth century
Iranian scholars. See gh Buzurg Tehrn, al-Dhara il tasnf al-Sha (Najaf:
Matbaat al-Gharr, 1936 ff.), VI, 59 ff. I have not seen a copy of this edition.
138 chapter four
63
Gln, al-Risla al-muh ta bi-tashkkt f l-qawid al-mantiqiyya maa tah qqtih,
in Collected Papers on Logic and Language, ed. by M. Mohaghegh and T. Izutsu
(Tehran: Institute of Islamic Studies, 1974), 359395.
64
He is mentioned as one of the many eminent students of Dawns student Jaml
al-Dn Shrz (d. 1554) in R. Pourjavady, A Sh Theologian and Philosopher of Early
Safavid Iran, p. 42n33.
65
Gln, al-Risla al-muh ta, 390 (wa iturida alayhim anna qawlan Zayd ab
ktibin wa kullu ktibun yaqrau qiysun muntijun bi-l-badhati l-aqliyyati annahu
Zayd ab qriin).
distinct traditions of arabic logic after 1600 139
66
Gln, al-Risla al-muh ta, 390 (wa amm l-muh aqqiqu l-Dawni fayuwajjihu
al wajhin l yahtju il irtikbi l-musmah ati f hdh l-qiysi wa huwa anna
l-akbara h aqqatan huwa l-muallifu wa l-awsatu huwa l-muallafu wa l-lmu l-jrratu
tudu f l-natjati wa yamnau kawna l-awsati yajibu takarruruhu bi-aynihi bil
ziydatin wa nuqsnin bal kathran m zda al l-awsati shayun yudu f l-natjati
kam f qawlin Zayd ibnu Abdillh wa Abdullh ktibun fa-Zayd ibnu ktibin min
ghayri tamah h ulin wa irjin il wh idin min al-aqyisati l-mashhra).
67
On this scholar, see the editors introduction to q H usayn Khvansr, H shiya
al shurh al-Ishrt, ed. A. bed (Qum: Markaz Intishrt Daftar Tablght Islm,
1378/1999).
68
q H usayn Khvansr, H shiya al H shiyat Sharh al-Tajrd (Manuscript India
Office 1001, British Library), fol. 25b (ilam anna l-h aqqa l-ladh l mah sa anhu anna
kulla qiysin yuntiju natjatan mutta ridatan l budda fhi min an yasih h a irjuhu il
l-aqyisati l-mutabarati l-lat itabarah l-qawmu wa dawwanh wa anna l budda
f l-iqtirniyyti min takarruri l-h addi l-awsati bi-tammihi).
140 chapter four
69
Khvansr, H shiya al H shiyat Sharh al-Tajrd, fol. 25b (naam qad l yatakar-
raru bi-tammihi kam f hdh l-qiysi l-ladh nah nu bi-sadadihi wa yantaqilu
l-dhihnu minhu il l-natjati min dni mulh azati irjihi il l-qiysi l-ladh takar-
rara fhi l-h addu l-awsatu bi-tammihi binan al anna l-irja l-madhkra wa
l-muqaddimti l-matwiyyata al-muhtja ilayh f l-irji malh zatun li-l-aqli ijmlan
kam nushhidu f kathrin min al-mawdii f ghayri mithli hdh l-qiysi mimm l
niza fhi f annahu l budda min irjihi il l-aqyisati l-mutarafati).
70
Khvansr, H shiya al H shiyat Sharh al-Tajrd, fol. 26a (fa-l-qnnu l-ladh
yulimu l-intja f mithli hdh l-qiysi l-ladh nah nu bi-sadadihi huwa hdh
fa-yunzaru annahu hal yarjiu il l-qiysi l-mahdi am l fain rajaa fa-huwa sah h u
l-intji wa in lam yarji fa-l wa hdh l yunf an yakna l-aqlu yantaqilu il
l-natjati min dni irjihi mufassa lan).
71
Khvansr, H shiya al H shiyat Sharh al-Tajrd, fol. 26a.
distinct traditions of arabic logic after 1600 141
(ii) to replace the second premise with another premise that is equi-
pollent to it and that shares a term with the first premise, viz. What is
equal to B is equal to what is equal to J
72
Khvansr, H shiya al H shiyat Sharh al-Tajrd, fol. 26a26b.
142 chapter four
73
Khvansr, H shiya al H shiyat Sharh al-Tajrd, fol. 26b.
distinct traditions of arabic logic after 1600 143
The diagnosis given by Qutb al-Dn al-Rz is that the second premise
is not a universally quantified premise; it does not say that every ani-
mal is a genus. Rather, it is a so-called natural (tab) proposition in
which the subject-term denotes the nature itself rather than the indi-
viduals falling under the universal term.74 Surely, Dawn and Mrz
Jn could have used the very same diagnosis to show what is wrong
with the counter-example, and at the very least they could have chal-
lenged Khvansr to explain why his counter-example could not
equally be aimed at standard first-figure syllogisms with a fully recur-
rent middle-term.
The second counter-example presented by Khvansr is:
The human is distinct from the horse (al-insnu mubyinun
li-l-faras)
The horse is an animal (al-farasu h ayawn)
The human is distinct from the animal (al-insnu mubyinun
li-l-h ayawn)
In the context of Arabic logic, the first premise would most naturally
be understood as stating that the general terms human and horse
have non-overlapping extensions. Mubyana was introduced in hand-
books of logic as one of four ways in which the extensions of two uni-
versals could be related, the others being partial overlap (umm wa
khuss min wajh) like human and white, complete overlap (muswt)
like human and capable of laughter, or one term being more general
and the other more specific (umm wa khuss mutlaq) like human
and animal.75 Thus understood, the argument seems to be:
Human has a non-overlapping extension with horse
The horse is an animal
Human has a non-overlapping extension with animal
Here, the premises are true but the conclusion is false. Yet, the argu-
ment seems, on the face of it, to have a similar form to the arguments
that Dawn claimed were formally productive.
This counter-example was not first formulated by Khvansr, for it
was already referred to by the Ottoman scholar Mehmed Emn irvn
74
Qutb al-Dn al-Rz, Tah rr al-qawid al-mantiqiyya, II, 250 (1905 ed.); 170 (1948
ed.).
75
Qutb al-Dn al-Rz, Tah rr al-qawid al-mantiqiyya, I, 249 ff (1905 ed.); 6364
(1948 ed.).
144 chapter four
(d. 1627) as an objection that had been raised against Dawns point.76
irvn also cited a reply to the objection by another earlier and
unnamed scholar, which suggests that the objection had been made by
the mid-sixteenth century. The identity of the person who first raised
the objection is a matter for further research. It may have been Sadr
al-Dn al-Dashtaks son Ghiyth al-Dn Dashtak (d. 1542).77 The
counter-example will be discussed at some length below and in the
subsequent chapter, in connection with the efforts of later Iranian and
Ottoman scholars to deal with it. However, it may be helpful to note at
this point that it appears to be no less fallacious than the previous
counter-example. It seems for one thing to lack a properly recurring
term at all: in the first premise the universal human is said to be dis-
tinct from the universal horse, but in the second premise it is every
individual horse (and not the universal term horse) that is said to be
an animal.
Another problem with the counter-example is clear from the Arabic
formulation:
al-insnu mubyinun li-l-faras
al-farasu h ayawnun
al-insnu mubyinun li-l-h ayawn
In the second premise, the predicate is in the indefinite form
h ayawnun. In the purported conclusion, this appears with the defi-
nite article al-. In Arabic, the definite article in such a context would
naturally be understood as what grammarians call lm al-jins, i.e. the
definite article employed to indicate the genus, i.e. any individual
bearing the name.78 This is precisely why the conclusion appears to
be false. If the conclusion were formulated as: al-insnu mubyinun
li-h ayawnin (i.e. The human has a non-overlapping extension with
an animal) then it becomes much less obvious that the conclusion is
in fact false.
76
See the following chapter for an account of irvns discussion.
77
As mentioned in the previous chapter, Ghiyth al-Dn Dashtak stated in his gloss
on the gloss of Qutb al-Dn al-Rz on Tss commentary on Avicennas Ishrt that
he had refuted Dawn at length in his Tadl al-mzn. However, I have been unable
to locate such a refutation in the two incomplete manuscripts of Tadl al-mzn that
I have seen.
78
W. Wright, A Grammar of the Arabic Language, 3rd edition (Reprint, Beirut:
Librarie du Liban, 1974), I, 269b.
distinct traditions of arabic logic after 1600 145
79
See his al-Manhij f l-mantiq, ed. I. Dbj (Tehran: Institute of Islamic Studies,
1376/1997).
80
In the introduction to his major work on logic, Tadl al-mzn, Dashtak stated
that he found the views of the older logicians, as distilled in Avicennas Shif, to be
superior to those of the later logicians. See Manuscript 23594, Astn-i Quds-i Radaw
Library, Mashhad, fol. 1b.
81
See the sparse entry in Mhsin al-Amn, Ayn al-Sha (Beirut: Dr al-Insf,
1960 ff.), XXXXVII, 158.
distinct traditions of arabic logic after 1600 147
rejected the fourth figure of the syllogism, accepted by almost all logi-
cians since Rz.82 He rejected the combinatorial hypothetical syllo-
gism first discussed by Avicenna. He rejected the more than a dozen
modality propositions distinguished and discussed by Rz and subse-
quent logicians. He also rejected the formal orientation of post-
Avicennian logicians, and the greater part of his work is dedicated to
discussing topics relating to demonstration that Aristotle had covered
in the Posterior Analytics. His work is idiosyncratic and well-worth
editing and publishing. It is a salutary reminder that further research
into unpublished Iranian manuscripts may force a revision of many of
the general statements made in modern studies of the history of
Arabic logic, the present study not excepted.
Tehrn defined a syllogism as a composite statement which, when
what is presented in it is conceded, implies by itself another statement
necessarily (qawlun murakkabun mat sullima m rida fhi lazima
li-dhtihi qawlun kharu idti rran).83 One may notice the reintroduc-
tion into the definition of a reference to the premises being conceded,
as well as the reintroduction of the condition that the implication be
necessary. The latter condition is, Tehrn wrote, for the purpose of
ruling out cases of implication in particular material instances, such as
the argument No horse is a human & Every human is a speaker
which implies the conclusion No horse is a speaker but not necessar-
ily. The condition that the premise-pair imply the conclusion by itself
is, Tehrn explained, to rule out cases in which a premise had been
left unmentioned. As an example, he gave the syllogism of equality
which produces its conclusion by mediation of an omitted premise
(bi-wsitati muqaddimatin mah dhfa) which is What is equal to what
is equal is equal.84
Another sign of the intentionally archaic organization of Tehrns
work is the later chapter on analysis (tah ll), i.e. on how to reduce
arguments in ordinary language to syllogistic form. This had appeared
in the works of the early Arabic Aristotelians, but rarely in the works
of logicians after Fakhr al-Dn al-Rz. In the section, Tehrn resumed
his discussion of the syllogism of equality, including a lengthy (and
82
The rejection of the fourth figure had been prefigured in Tadl al-mzn by
Ghiyth al-Dn Dashtak and in al-Tanqh by Mull Sadr Shrz.
83
Muhammad Ysuf Tehrn, Naqd al-usl wa talkhs al-fusl (Manuscript New
Series 996, Firestone Library, Princeton University), fol. 13a.
84
Tehrn, Naqd al-usl, fol. 13b14a.
148 chapter four
85
Tehrn, Naqd al-usl, fol. 27a27b.
86
Two prominent examples of this tendency are the anti-Peripatetic philoso-
pher Suhraward (d. 1191) who saw himself as reviving the wisdom of Plato, the pre-
Socratics and the ancient Iranian sages, and the iconoclastic religious purist Ibn
Taymiyya (d. 1328) a hero of modern Sunn fundamentalism who opposed mysti-
cism, philosophy, logic, rational theology and scholastic jurisprudence in the name of
a return to the supposedly pristine Islam of the earliest generations.
87
See the editors introductions to the editions of Averroes, Ibn Zura, and Frb
cited in the bibliography. Note also two Safavid and post-Safavid Iranian manu-
scripts containing Averroes Middle Commentaries on Aristotles Categories, De
Interpretatione, Prior Analytics and Posterior Analytics. One, copied by a certain Abd
al-Azz Abhar in 1084/16731674, is extant in the Quds-i Radawi Library in
Mashhad. See Barat Ali Gholam Moghaddem, Mantiq, vol. 24 of Fihrist-i Kitbkhna-yi
stn-i Quds-i Radaw, 542. The second, copied by Muhammad Mumin Rz in
1177/17631764, is extant in Dr al-Kutub in Cairo. See Fihris al-kutub al-mah fza
bi-l-Kutubkhna al-Khedviyya al-Misriyya (Cairo: n.p., 1305/18871888), VI, 52.
88
Bah al-Dn Isfahn, Hikmat-i Khqniyya, ed. G.I. Dinn (Tehran: Ayene-ye
Miras, 1998); Muhammad Rid Araj, al-Urjza f l-mantiq, in Collected Papers on
Logic and Language, ed. by M. Mohaghegh and T. Izutsu (Tehran: Institute of Islamic
distinct traditions of arabic logic after 1600 149
Quran commentator Abd Al al-H uwayz. Although ethnically Arab, such Sh schol-
ars were very much part of Safavid intellectual life.
94
Ishq al-H uwayz, H shiya al H shiyat Mull Abd-Allh (Manuscript Add.
7461. British Library), fol. 104ab.
95
See Muhammad Mahd Narq, Sharh al-Ilhiyyt min kitb al-Shif, ed.
M. Mohaghegh (McGill & Tehran: Institute of Islamic Studies & Tehran University,
1986). The editor has included biographical information on this scholar in his
introduction.
152 chapter four
united. In that case, the conclusion is that the human is different from
something specific of which animal is true or with which animal is
united.96
Narq thus argued that since the first premise asserts that humans
are distinct from horses, and the second premise that the predi-
cate animal is true of every horse, then the conclusion follows that
humans are distinct from specific entities (namely horses) of which
the predicate animal is true. The implication of this which Narq
did not spell out is that the objection that the premises of the argu-
ment are true and the conclusion false is based on the fallacious
assumption that the only conclusion one can draw from the premises
is that humans are distinct from all entities of which the predicate ani-
mal is true. I will return to this important point in the following
chapter.
Narqs discussion indicates that Dawns claim had not been for-
gotten in Iran, and that there were still scholars there who defended
it against the attacks of Sadr al-Dn al-Dashtak, Ghiyth al-Dn
Dashtak, and q H usayn Khvansr. However, judging from the
late nineteenth- and early twentieth-century Iranian works on logic
that have been published, it nevertheless seems that Dawns point
though it continued to be discussed in certain advanced contexts
left little trace on the standard handbook presentations of logic in
Iran.
The didactic poem entitled al-Lal l-muntazama by Mull
Hd Sabzavr (d. 1878), with its authors commentary, was the first
handbook on logic to achieve wide use in Iranian scholarly circles
since Yazds sixteenth-century commentary on Tahdhb al-mantiq.97
Sabzavrs work had a basically post-Avicennian orientation. To be
sure, it only dealt with three figures of the categorical syllogism, and
the fourth was characterized as being far from nature (yanb an
al-tab). However, the reader was referred to other works should
he wish to know the conditions of productivity of that syllogistic
figure, which suggests that Sabzavrs exclusion was motivated by
96
Mahd Narq, Mushkilt al-ulm (Tehran: Muassasah-i Mutlat va Tahqqt-i
Farhang, 1367/1988), 133 (man qawlin l-farasu h ayawnun anna l-farasa mimm
yasduqu alayhi l-hayawnu aw yattah idu maahu wa h naidhin taknu l-natjatu anna
l-insna mubyinun li-misdqin khssi n li-l-h ayawni aw shayin khssi n yattah idu
maahu).
97
On Sabzavr, see A.J. Newman, Sabzawr, H djdj Mull Hd H djdj Mahd,
Encyclopaedia of Islam, Second Edition, VIII, 695.
distinct traditions of arabic logic after 1600 153
98
Sabzavr, Sharh al-Manzma, edited by M. T lib and annotated by H asanzde
mul (Qom: Nashr Nb 1416/1995), 307.
99
Sabzavr, Sharh al-Manzma, 286287.
100
Haydaj, Talqa al l-Manzma w-sharh ih (Lithograph, Tehran: n.p., 1346/
1928), 54.
101
Fursat Shrz, Ashkl al-mzn (Lithograph, 1307/18891890; reprint, Tehran:
Kitbfursh-yi Islmyeh, 1340/19611962), 109.
102
On him, see Mhsin al-Amn, Ayn al-Sha, XXXIX, 194195.
103
Hamadn, Luluat al-mzn maa sharh ihi Muntaq l-jummn (Lithograph, n.p:
n.p., 1323/19051906), 194. Due to the rarity of this edition, I should indicate that
I have used a copy in the Astan-i Quds-i Radaw Library in Mashhad (Call number
21313).
154 chapter four
104
Hamadn, Luluat al-mzn, 195196.
105
The difference between the two formulations of the missing premise seems to
derive from variants in the manuscripts of Khnajs Kashf al-asrr and Urmaws
Matli. The detailed regimentations offered by Hamadn with these premises are
very similar.
106
Hamadn, Luluat al-mzn, 198.
distinct traditions of arabic logic after 1600 155
did not note that by adopting this position he should, strictly speak-
ing, abandon the view that the syllogism of equality is excluded from
the definition of syllogism.
The discussion of the problem of the syllogism of equality by
Hamadn is much fuller than that of Sabzavr, but it nevertheless
does not add anything to what Qutb al-Dn al-Rz had written in the
first half of the fourteenth century, nor does it incorporate or engage
with the insight of Dawn. Even if more extensive research into
Iranian manuscript collections should bring to light developments of
Dawns point from the seventeenth, eighteenth or nineteenth centu-
ries, it is striking that Sabzavr, Haydaj, Fursat Shrz, and Hamadn
did not register them. This suggests that any such contribution did not
succeed in making an impression on the subsequent Iranian logical
tradition.107
V. Conclusion
107
For a brief discussion of the point by a twentieth-century Sh scholar, see Al
Kshif al-Ghit, Naqd al-r al-mantiqiyya wa h all mushkiltih (Qom: frand,
1427/2006 [reprint of 1954 ed.]), 518. The author briefly accepts that a fully recurrent
middle term is not necessary in a categorical syllogism, and adduces in support of his
view the example Zayd is the son of Bakr & Bakr is a scholar, so Zayd is the son of a
scholar. His discussion supports the impression that there had been little develop-
ment of this particular point since the time of Mull Sadr and Gln.
108
Rescher, The Development of Arabic Logic, 73.
156 chapter four
made before any firm conclusions can be made about the innovative-
ness of Arabic logical works after 1550. Certainly, Muhammad Ysuf
Tehrn cannot be dismissed as a mere commentator or glossator,
since he wrote a lengthy exposition of logic that broke with the prev-
alent post-Avicennian tradition. His work deserves serious study.
Seventeenth and eighteenth century Iranian and Indian scholars who
accepted the post-Avicennian paradigm, and who contributed to the
standard post-classical literary forms of epitome, commentary, and
gloss should also not simply be dismissed off-hand. Even a cursory
perusal of Mull Sadrs gloss on Sharh H ikmat al-Ishrq, or Siylkts
gloss on Sharh al-Shamsiyya, or Bihrs handbook Sullam al-ulm
and its commentaries confirms that these are critical and sophisticated
works. It does seem, however, that most Iranian and Indian scholars
in these later centuries were primarily interested in the non-formal
aspects of logic: the subject-matter of logic, the division of knowledge
into conception and assent, the five predicables, the parts of a proposi-
tion, and a number of paradoxes such as the liars paradox (al-jadhr
al-asamm) and Menos paradox (al-majhl al-mutlaq). North African
logicians from the seventeenth and early eighteenth century such as
Ys, Ibn Yaqb, and Hill appear to have been less interested in such
questions, and still devoted considerable attention to formal topics like
conversion, contraposition, the immediate implications of condition-
als, and the syllogistic. Their treatment of such issues seems to have
been conservative in spirit, but this did not preclude novel insights
into particular topics, including the above-mentioned distinction
between the rule of inference that an argument exhibits and a sup-
pressed premise.
Another reason why the dismissal of later centuries is overhasty
should become clear in the following three chapters. As will be seen,
Ottoman Turkish logicians in the seventeenth and eighteenth centu-
ries did develop Dawns point regarding the middle-term of a syllo-
gism recurring with addition or subtraction. By the late eighteenth
century, they had incorporated what they called unfamiliar syllo-
gisms into their logical handbooks. In the nineteenth century, such
syllogisms came to play an important role in a marked revival of inter-
est in applied logic.
CHAPTER FIVE
Modern historians have often assumed that the seventeenth and eight-
eenth centuries constitute a period of cultural and intellectual decline
in the Ottoman Empire. In classic works such as Halil nalciks The
Ottoman Empire: The Classical Age, 13001600 (1974), the Ottoman
cultural and intellectual florescence of the fifteenth and sixteenth cen-
turies was presented as having come to an end during the seven-
teenth.1 An important element of this supposed decline was what
nalcik dubbed the triumph of fanaticism. The violently puritan
K dzdel movement that emerged on the scene toward the end of
the sixteenth century was, according to this account, responsible for
putting an end to interest in the rational sciences. This dour state of
affairs has been contrasted by other historians with the strong interest
in rational sciences in Safavid Iran and Mughal India during this
period.2
It is beginning to emerge from more recent research that this
account is fundamentally wrong.3 Particularly the second half of the
seventeenth century and the first half of the eighteenth were a period
of considerable intellectual ferment and cultural dynamism in the
central parts of the Ottoman Empire. Numerous libraries were est-
ablished, the number of colleges increased dramatically, and a number
of works on medicine, philosophy, astronomy, and geography were
translated from Latin.4 Ottoman scholars interest in logic, dialectic,
1
Halil nalcik, The Ottoman Empire: The Classical Age, 13001600 (London:
Weidenfeld & Nicholson, 1973), 179185.
2
Francis Robinson, Ottomans-Safavids-Mughals: Shared Knowledge and
Connective Systems, Journal of Islamic Studies 8(1997): 151184, esp. 155 ff., 172.
3
For an insightful and wide-ranging discussion of the supposed decline of
the Ottoman Empire, see C. Kafadar, The Question of Ottoman Decline, Harvard
Middle Eastern and Islamic Review 4 (19971998): 3075. A discussion that is more
narrowly focused on the continued study of philosophy and logic in seventeenth- and
eighteenth-century Ottoman Turkey is contained in my The Myth of the Triumph of
Fanaticism in the Seventeenth-Century Ottoman Empire, Die Welt des Islams 48
(2008): 196221.
4
According to one estimate, the number of Istanbul colleges doubled between 1650
and 1705, see M. Zilfi, The Ottoman Ulema in the Post-Classical Age, 16001800
158 chapter five
I. H usayn Khalkhl (d. 1604) & Mehmed Emn irvn (d. 1627)
(Minneapolis: Bibliotheca Islamica, 1988), 205. Some of the major libraries founded
in this period include Carullah, Amcazade Hseyn Paa, and Laleli (the holdings of
which are all now in the Sleymaniye Library), as well as the Ahmed III collection
(now in Topkapi), Feyzullah, Kprl, and Nuruosmaniye. Translations from Latin in
this period include translations of Paracelsus and Crollius (by Slih b. Sallm, d.1670,
the court physician of Mehmed IV, r. 16481687), the translation of Janszoon Blaeus
Atlas major (by Eb Bekr Dimak, d.1691, commissioned by Mehmed IV), and trans-
lations of Aristotle and of works by the Greek-born Renaissance Aristotelian Ioannis
Kottounios (by Esad Ynyav, d.1722, who enjoyed the patronage of Dmd Ibrhm
P, the Grand Vizier of Ahmed III, r. 17031730).
5
On Khalkhl, see Muhammad Amn al-Muhibb, Khulsat al-athar f ayn al-
qarn al-h d ashar (Cairo: al-Matbaa al-Wahbiyya, 1284/186869), II, 122.
ottoman logic 159
6
Khalkhl, H shiya ala Sharh al-Tahdhb (Manuscript 01 Hk 795, www.yazmalar
.gov.tr), fol. 54b. (wa-hdh al m sarrah a bihi l-ustdhu l-muh ashsh f badi
tasnfihi shaklun awwalun lam yatakarrar al-awsatu fhi bi-tammihi nazruhu Zayd
abun li-Amr wa Amr ktibun yuntiju Zayd abun li-ktibin).
7
On irvn, see Muh ibb, Khulsat al-athar, III, 475476.
8
irvn, H shiya (Manuscript 45 Ak Ze 676. www.yazmalar.gov.tr), sayfa numeras
8 (right margin) (l yuqlu: hdh qiysu muswtin li-sidqi tarfihi alayhi wa qiysu
l-muswti khrijun an tarfi mutlaqi l-qiysi fa-kayfa yaknu hdh shaklan awwa-
lan wa huwa min aqsmihi).
160 chapter five
9
I should stress the seems to be. There is a problem with the inference that will
be noted below.
10
irvn, H shiya, sayfa numeras 8 (right margin). The passage in question has
an indication that it is from a later copy (nuskha jadda) of irvns gloss (l budda
f qiysi l-muswti an yakna l-mah mlu f l-muqaddimatayni amran wh idan
lkinnahum lam yusarrih bi-hdh l-qaydi f tarfihi itimdan al l-amthilati l-lat
dhakarh).
ottoman logic 161
11
irvn, H shiya, sayfa numeras 8 (main text, right) (mithlu hdh l-qiysi wa in
sadaqa alayhi tarfu qiysi l-muswti lkinnahu laysa minhu f shayin li-luzmi
l-qawli l-khari an mithli hdh l-qiysi li-dhtihi fa-huwa shaklun awwalun lkin al
hayatin ghayru mutarafatin).
162 chapter five
As seen in the previous chapter, this was an objection that had been
raised against Dawn by a sixteenth-century scholar. The response
cited and defended by irvn was that the conclusion of the counter-
example is not false but true. irvn wrote:
For it is not impossible that opposing descriptions be true of universal
natures. Rather this is necessary Animal as such, without any qualifi-
cation, is distinct from human in the case of horse, and is not distinct
but identical to it in the case of human.12
In other words, the universal human is distinct from (in the sense of
having a non-overlapping extension with) the universal animal inso-
far as the universal human is distinct from the universal horse, and
yet the universal human is not distinct from the universal animal in
the sense that the universal human is not distinct from, but identical
to, the universal human.
irvns response brings out an ambiguity in the conclusion that
he claimed is true and the objector claimed is false. irvns response
amounts to saying that it is true that The human is distinct from
the animal since the human is distinct from an animal, namely horse.
The objector understood the conclusion The human is distinct from
the animal in the sense that the extensions of the universals human
and animal do not overlap, and hence in the sense that every human is
distinct from every animal. irvn does not appear to have been fully
alive to this point, for if he had then one would assume that he would
have given a more straightforward diagnosis of the fallaciousness of
the counter-instance, for it illegitimately goes from predicating an
animal (h ayawn) in the major premise, to a conclusion featuring the
animal (al-h ayawn), in the sense of every animal, in the conclusion.13
Instead, he was forced to reject the counter-instance by adopting a
non-standard view of the truth-conditions of the proposition al-insnu
mubyinun li-l- h ayawn. In effect, such a proposition becomes true if
the human is distinct from some animal. However, this account of the
truth-conditions of the conclusion was non-standard and hence liable
12
irvn, H shiya, sayfa numeras 8 (main text, left) (lianna ttisfa l-tabii l-
kulliyati bi-l-mutaqbilayni ghayru mumtaniin bal lzimun fa-l-h ayawnu l bi-shartin
mubyinun li-l-insni f dumni l-farasi wa ghayru mubyinin lahu bal muttah idin
maahu f dumni l-insn).
13
The definite article here functions as what grammarians call lm al-jins, i.e. indi-
cating that the term applies to the genus, and by implication any individual belonging
to that genus. See Wright, A Grammar of the Arabic Language, I, 269B.
ottoman logic 163
14
Isml Bagdd, Hadiyyat al-rifn: Asm al-muallifn wa thr al-musannifn
(Istanbul: Milli Eetim Basimevi, 1951), II, 482. The fact that Pehlevn hailed from
the Turcoman tribe Pehlevan is mentioned by his student Mehmed T vskr, see
T vskr, Sharh al-Qiysiyya (Manuscript Yazma Baislar 248, Sleymaniye
Ktphanesi, Istanbul), fol. 109r (marginalia by author).
15
T vskr, Sharh al-Qiysiyya, fol. 119a119b. The passage also occurs on pp.
1213 of the, generally inferior, edition printed in Istanbul by Matbaa-i Amire in
1281/18641865 (wa-yanbagh an yulama hhun annahu kam yaknu li-kulli shak-
lin min al-ashkli mutarafun ka-dhlika li-l-kulli ghayru mutarafun amma ghayru
mutarafi l-awwali fa-huwa an yakna mutaalliqu mah mli l-sughr mawdan f
l-kubr bi-sharti an yakna mah mlu l-sughr mukhlifan li-mah mli l-kubr matha-
lan A muswin li-B wa B J fa-taknu l-natjatu A muswin li-J).
ottoman logic 165
of Qutb al-Dn al-Rz (d. 1365) and Sad al-Dn al-Taftzn (d. 1390);
Matli al-anwr by Sirj al-Dn al-Urmaw (d. 1283) with the com-
mentary of Qutb al-Dn al-Rz (d. 1365); al-Ghurra f l-mantiq by
Muhammad ibn al-Sharf al-Jurjn (d. 1434) and its commentaries by
Khidir al-Jabalrd (fl. 1421) and s al-Safaw (d. 1546); Mukhtasar
al-mantiq by Muhammad ibn Ysuf al-Sans (d. 1490) with the
authors own commentary; and Tahdhb al-mantiq by Sad al-Dn
al-Taftzn with the commentaries of Mull Abd-Allh Yazd
(d. 1573), Ubayd-Allh Khabs (fl. 1540s), and Mr Ab l-Fath
H usayn (f l. 1564). It does not occur in the summa of logic entitled
Qists al-afkr by Shams al-Dn al-Samarqand (d. 1303), who had
used the term qiysu l-nisba or the phrase al-qiystu l-lat l taknu
awstuh mutakarriratan when speaking of relational inferences. The
phrase also does not occur in the extensive summa of logic entitled
Jmi al-daqiq by Najm al-Dn al-Ktib, a work that the Ottoman
bibliographer Katib eleb (d. 1657) praised for not leaving out any of
the principles and issues of logic. Nor does it occur in Tadl al-ulm
by Sadr al-Shara al-Mahbb (d. 1346), a work recommended by the
sixteenth-century Ottoman scholar Ahmed T kprzde (d. 1561) to
anyone who wanted to obtain a complete mastery of logic. The scholar
Ah mad b. Yahy al-Haraw, a great-grandson of the prominent
Timurid scholar Taftzn and therefore better known as H afd
al-Taftzn (d. 1511), used the phrase ghayr mutaraf of certain syl-
logisms, but in a way that confirms that it did not have the technical
sense that it came to have in Ottoman logic after Pehlevn. He used it
of a syllogism in which the conclusion is one of the premises, for
example:16
Every human is an animal
Every animal is an animal
Every human is an animal
The phrase unfamiliar syllogism does not occur in sixteenth- and
seventeenth-century handbooks on logic such as Takml al-mantiq by
the Persian-born Ottoman scholar H asan ibn H seyn Emle (fl. 1548),
nor al-Tanqh by Mull Sadr, nor Sullam al-ulm by Bihr. There is
also no mention of it in the commentaries on Abhars sghj and
16
H afd al-Taftzn, al-Durr al-nadd li-majmat ibn al-H afd (Beirut: Dr
al-Kitb al-Arab, n.d), 336.
166 chapter five
17
K ara all, H shiya ala Sharh al-Tahdhb (Manuscript 06 Hk 2248, www.yazmalar
.gov.tr), fol.4r.
ottoman logic 167
18
T vskr, Sharh al-Qiysiyya, fol. 119b120a (manuscript); 1213 (printed ed.)
(l yuqlu hdh aynu srati qiysi l-muswti liann naqlu hdh l-qawlu ghuflun
an qawlin bi-sharti an yakna mah mlu l-sughr mukhlifan li-mah mli-l-kubr).
19
There are several towns in Anatolia with the name Kalaycik. The largest of them
is in the province of Tokat. The author of the commentary may therefore have been
known in other contexts as mer Tkd. mer Feyz Tkd (d. 1848), author of a
popular commentary on Abhars sghj, referred in that work to a commentary he
had written on Pehlevns treatise. See Tkd, al-Durr al-nj al matn sghj
(Istanbul: irket-i Sahfiye-yi Osmniye-ye Matbaasi, 1302/18841885), 98. Feyz
Tkds commentary on sghj was by his own account completed in 1794, and his
commentary on Pehlevns treatise must therefore have been written earlier.
20
K alaycik, H shiya al l-Risla al-Msawiyya (Manuscript Fatih 3304,
Sleymaniye, Istanbul), fol. 37b; (Manuscript 06MilYzA9363, www.yazmalar.gov.tr),
fol. 13b (lianna akthara l-fudali ghfilna [sic] m f hdhihi l-rislati min al-farqi
bayna qiysi l-ghayri l-mutarafi wa qiysi l-muswti fadlan an tullbi zamnin bal
aktharuhum l yalamna m huwa ghayru l-mutaraf ).
168 chapter five
stipulation was not attested amongst previous logicians, but was rather
a new convention (istilh jadd).21 T vskr tried to answer this
objection by citing a passage from the fifteenth-century gloss of
Burhn al-Dn al-Bulghr on Sharh sghj by Mehmed Fenar
(d. 1431), discussed in Chapter Three. Burhn al-Dn had defined the
syllogism of equality as that in which equality is predicated twice, as
for example: A is equal to B & B is equal to J. Burhn al-Dn then
went on to state that there were cases similar to the syllogism of equal-
ity in which other predicates than equality were predicated in both
premises, such as The honey is inside the pot & The pot is inside the
house or Two is half of four & Four is half of eight. He then proposed
a broader definition of syllogisms of equality that would make all
such examples instances of one quiddity (mhiyya):
It is possible to explain the syllogism of equality as that in which one
thing is predicated of two things and belongs to them. This would be
more general than this one thing being equality or difference or some-
thing else. In this case, all the examples would be examples of syllogisms
of equality, and this would be more amenable to apprehension.22
Burhn al-Dn did not, however, make a distinction between syllo-
gisms of equality thus defined and unfamiliar syllogisms. His pro-
posed definition of syllogisms of equality would exclude cases of
what Pehlevn called unfamiliar syllogisms, but he did not explicitly
do so. He simply did not consider cases in which the two premises
have different predicates.
It is revealing that citing Burhn al-Dns definition of syllogisms of
equality was the best that T vskr could do to answer the objection
that his teacher had introduced a new convention; he apparently
could not find earlier logicians who drew the same distinction as
Pehlevn did between syllogisms of equality and unfamiliar syllo-
gisms. He seems to have been unaware of irvns discussion, and
irvn who appears to have been the first to draw the distinction
in turn missed Burhn al-Dns discussion, for he had written that pre-
vious logicians had not explicitly mentioned the condition that the
21
T vskr, Sharh al-Qiysiyya, fol.122r (manuscript); 14 (printed ed.).
22
Bulghr, H shiya ala l-Fanr, 61 (wa-yumkinu an yufassara qiysu l-muswti
bi-annahu qiysun yaknu fhi l-shayu l-wh idu mah mlan al l-shayayni wa
thbitan lahum aamma min an yakna dhlika l-shayu l-wh idu huwa l-muswta
aw al-mubyanata aw ghayrahum fa-h naidhin yasru jamu l-amthilati min qabli
qiysi l-muswti wa huwa aqrabu il l-dabt).
ottoman logic 169
23
T vskr, Sharh al-Qiysiyya, fol. 120a120b (manuscript); 13 (printed ed.) (law
quln f l-srati l-madhkrati A muswin li-B wa B muswin li-J taknu l-sratu
l-madhkratu surata qiysi l-muswti fa-tutasawwaru l-muqadimmatu l-ajnabiyyatu
l-lat l budda minh f qiysi l-muswt).
24
T vskr, Sharh al-Qiysiyya, fol. 121a121b (manuscript); 1314 (printed ed.)
(bi-khilfi l-ghayri mutarafi kam l yukhf al l-mutatabbi).
25
T vskr, Sharh al-Qiysiyya, fol. 121b122a (wa-amm ghayru mutarafi
l-thn fa-huwa an yakna mutaalliqu mah mli-l-sughr mah mlan f l-kubr bi-l-
sharti l-madhkri mathalan A muswin li-B wa J laysa bi-B fa-taknu l-natjatu laysa
A muswiyan li-J).
170 chapter five
26
T vskr, Sharh al-Qiysiyya, fol. 122a122b (manuscript); 14 (printed ed.)
(hdh mithlu l-awwali kathru l-wuqi f l-kalmi wa in kna kullun minhum ghayra
mutaraf).
27
T vskr, Sharh al-Qiysiyya, fol. 110r (manuscript); 3 (printed ed.).
28
As will be shown in chapter VII, there seems to have been an increasing interest
during the eighteenth and nineteenth century amongst Ottoman scholars in render-
ing into explicit syllogistic form arguments found in the handbooks studied in col-
leges. Despite the reference to speech (kalm), Pehlevanis comment is therefore
almost certainly a reference to the way in which the authors of handbooks argued,
rather than to the way in which ordinary people argued orally.
29
T vskr, Sharh al-Qiysiyya, fol. 122b (manuscript); 1415 (printed ed.)
(wa-amm ghayru mutarafi l-thlithi fa-huwa an yakna mutaalliqu mawdi
l-sughr mawdan f l-kubr bi-l-sharti l-madhkri aydan mathalan kullu muswin
li-B A wa kullu B J fa-taknu l-natjatu badu A muswin li-J).
172 chapter five
30
T vskr, Sharh al-Qiysiyya, fol. 122b123a (manuscript); 1415 (printed ed.)
(wa-amm ghayru mutarafi l-rbi fa-huwa an yakna mutaalliqu mawdi l-sughr
mah mlan f l-kubr bi-l-sharti l-madhkri aydan mathalan kullu muswin li-B A wa
kullu J B fa-taknu l-natjatu badu A muswin li-J).
31
T vskr, Sharh al-Qiysiyya, fol. 123a (manuscript); 15 (printed ed.) (lkin lam
nathur al wuqi l-thlithi f l-kalmi wa in atharn al wuqi l-rbii f l-kalmi
qallan).
ottoman logic 173
32
T vskr, Sharh al-Qiysiyya, fol. 123a123b (manuscript); 15 (printed ed.) (wa
l-shurtu wa l-durbu l-wqiatu f l-mutarafi jriyatun f ghayri l-mutaraf ).
33
T vskr, Sharh al-Qiysiyya, fol. 123b (manuscript); 15 (printed ed.) (taammal
f hdh l-maqmi fainnahu min mazliqi l-aqdm).
174 chapter five
34
In the colophon of the manuscript that I have used (Manuscript Yazma Baislar
248, Sleymaniye Ktphanesi, Istanbul), it is stated that a fair copy of the work was
completed in 1161/1748 in the Nakbiye College in Gaziantep. The same information
is repeated in the colophon of another manuscript, evidently by a different hand, in
Ktahya Vahitpaa I Halk Ktphanes, 43 Va 958 (available on www.yazmalar.gov
.tr). Note that in many extant manuscripts, including the one on which the printed
edition of 1281/18641865 is based, the authors name is given as Mehmed b. Musta f
Erzurm. Tavusker is a small town in the province of Erzurum in eastern Anatolia.
The author seems to have been known in or near his home province as T vskr, but
would probably have been known in far-away Istanbul as Erzurm.
35
T vskr, Sharh al-Qiysiyya, fol. 121a (manuscript); 13 (printed ed.) (l hun
shubhatn wa hiya annahu in urda anna qiysa l-muswti l yuntiju natjatan aslan
fa-butlnuhu zhirun lianna kullan min mawddihi yastalzimu qawlan mughyiran
lahu bi-isqti l-h addi l-awsati wa in knat tilka l-mddati mimm addha min
al-mawddi l-lat l tuntiju aslan ka-qawlun l-wh idu nisfu l-ithnayni wa l-ithnni
nisfu l-arbaati fa-innahu yuntiju li-dhtihi l-wh idu nisfu nisfi l-arbaa).
ottoman logic 175
T vskr was aware of the fact that this point undermined the tradi-
tional idea that the syllogism of equality is excluded from the defini-
tion of syllogism by the stipulation that the premises must imply the
conclusion by themselves. The syllogism of equality does imply a con-
clusion by itself. T vskr conceded that the premise-pair A is equal
to B & B is equal to J does not imply by itself the conclusion A is equal
to J. In order to obtain this conclusion from the premises one would
have to take the intermediate conclusion that the original premises do
imply, and then conjoin it to the additional premise Everything that is
equal to what is equal to J is equal to J. However, this only shows that
the original premises are part of a complex syllogism in which the
conclusion only follows from all premises and not a subset of premises.
It does not, he wrote, show that the original premises are excluded
from the definition of the syllogism.
A second point on which T vskr dissented from Pehlevn was
the form of the conclusion of the third and fourth figures of the unfa-
miliar syllogism. As regards the third figure, Pehlevn had given the
following paradigm:
Every equal to B is A
Every B is J.
Some A is equal to J.
T vskr opined that the proper conclusion is the converse:
Some equal to J is A36
The thought behind this modification was plausibly explained by
Ahmed kir Birgev (fl.1839) in his gloss on T vskrs commentary:
J in the major premise is the predicate of what is the semantic depend-
ent in the minor, i.e. B. Thus, when the semantic dependent is left out of
the conclusion because it is repeated [in the premises], J is put in place
of this semantic dependent, and the original conclusion is as the
commentator [T vskr] said.37
36
T vskr, Sharh al-Qiysiyya, fol.122b (manuscript); 15 (printed ed.).
37
Birgev, H shiya ala Sharh al-Qiys al-Msawi, printed as an appendix to
Mehmed T vskrs Sharh al-Risla al-Qiysiyya (Istanbul: Matbaa-i Amire,
1281/18641865), 84 (J f l-kubr qadd waqaa mah mlan al m kna mutaalliqan f
l-sughr ay B fa-lamm saqata dhlika l-mutaalliqu f l-natjati li-takarrurihi fa-ydau
fh J mawdia dhlika l-mutaalliqi fa-yaknu aslu l-natjati kam qlahu l-shrih ).
176 chapter five
38
T vskr, Sharh al-Qiysiyya, fol. 123a (manuscript), 15 (printed ed.).
39
Levz Hseyn, Sharh al-Risla al-Qiysiyya (Manuscript Yahuda 5333, Firestone
Library, Princeton University), fol. 61. This is the earliest manuscript of the work that
I have found, having been copied by the same hand, and with the same layout, as two
other works that are bound with it, dated 1161/1748 and 1162/1749 respectively and
copied in the town of Amasya by a certain Osmn b. Ibrhm Amsyav. The author
mentioned the Turkish scholar Eb Sad dim (d. 1762) in his treatise, so he cannot
have been writing much earlier than that date. In at least one manuscript, the attribu-
tive Ams is added to the authors name suggesting that he was active in that town.
See: Manuscript Diyarbakr I Halk Ktphanesi 21 Hk 987/15, copied in 1189/1774,
and listed on www.yazmalar.gov.tr.
ottoman logic 177
40
Tkd, al-Durr al-nj, 104105.
41
Osmn Tarss, Risla-yi Istidlliye (Istanbul: Matbaat Dr al-Khilfa,
1258/1842), 1314. Ahmed b. Suleyman Magnisv, Sharh al-Istidlliyya (Lithograph,
Istanbul: Matbaat al-H jj Muharram al-Bsnaw, 1287/18701871), 6263. According
to the commentator Magnisv (fl. 1869), the author of the treatise is Osmn b.
Musta f Tarss known as K abakczde. This means that the authors father was
named Musta f K abakc, and it is tempting to identify him with the famous rebel,
even though the rebel had no known connections to the town of Tarss. In any case,
Osmn Tarss must have been active after T vskr (fl. 1748), on whom he relies,
and before 1842, the date of the first printing of his treatise.
178 chapter five
42
T vskr, Sharh al-Qiysiyya, fol.123b (manuscript); 15 (printed ed.) (ilam
anna l-qiysa l-ghayra l-mutarafi l-wqia f kalmihim l yatarakkabu kathrun
minhu min al-mah s rti bal innam yatarakkabu min al-shakhsiyyti aw min
al-tabiyyti kam kna qiysu l-muswti kadhlika wa qad yatarakkabu minh lkin
yuntiju f badi durbi badi l-ashkli intjan kulliyan wa f badih l yuntiju natjatan
kulliyyan).
ottoman logic 179
T vskr thus suggested that when the premises are taken as qua-
ntified propositions then unfamiliar syllogisms cease to be produc-
tive in all the moods and figures that are productive in familiar
syllogisms. Certainly, this was how Ahmed kir Birgev, who wrote a
gloss on T vskrs commentary, saw the matter.43 Ahmed kir dis-
cussed the very first example of unfamiliar syllogisms mentioned by
Pehlevn:
A is equal to B
B is J
A is equal to J
It seems, Ahmed kir wrote, that a counterexample is the following
argument:
The human is equal to the Homo sapiens (al-insnu muswin
li-l- bashar)
Every Homo sapiens is an animal (kullu basharin h ayawnun)
In this instance, we cannot conclude that The human is equal to the
animal (al-insnu muswin li-l- h ayawn), since it is evident, accord-
ing to Ahmed kir, that The human is not equal to the animal.
Indeed, the classical Arabic logical tradition taught that two universals
A and B are equal if and only if every A is B and every B is A,44 and
this is of course not true if A is human and B is animal. The problem
is the same as in the suggested counter-instance to first-figure unfa-
miliar syllogisms discussed by irvn:
The human is distinct from the horse
Every horse is an animal
The human is distinct from the animal
irvns solution to the counter-instance seems to have been unknown
to Ahmed kir, who opined that the problem here is that the minor
premise is a natural proposition (in which the subject-term stands for
a concept rather than for the particulars that fall under it), but that the
major premise is a quantified proposition. He went on to claim that if
43
Birgev, H shiya, 77.
44
Thus al-Sayyid al-Shard al-Jurjn (d. 1413) wrote: marja al-tasw il
mjibatayni kulliyyatayni, see his H shiya al Sharh al-Shamsiyya (Cairo: Musta f
al-Bb al-H alab, 1948), 65 (lower rubric); (Cairo: al-Matbaa al-Amriyya, 1323/1905)
I, 302.
180 chapter five
the major too had been a natural proposition then the argument
would be productive:
The human is equal to the Homo sapiens (al-insnu muswin
li-l- bashar)
The Homo sapiens is a rational animal (al-basharu h ayawnun
natiqun)45
The human is equal to the rational animal (al-insnu muswin
li-l- h ayawni l-ntiq)
The argument would also be productive, Ahmed kir wrote, if both
premises were singular propositions in which the subject term refers
to a particular entity:
This length is equal to that length (hdh l-miqdru muswin
li-dhlika l-miqdr)
That length is a foot (dhlika l-miqdru dhirun)
This length is a foot (hdh l-miqdru muswin li-l- <dhir>46)
T vskrs and Ahmed kir Birgevs diagnosis would seem to be
ill thought out. Both the claim that (i) the unfamiliar syllogisms of the
first mood of the first figure are unproductive if the premises are taken
as quantified propositions, and the claim that (ii) all valid moods of
familiar syllogisms are also valid in unfamiliar syllogisms as long as
the premises are taken as natural or singular propositions, can easily
be disproved. Concerning the first claim: The counter-argument
adduced by Ahmed akir is:
The human is equal to the Homo sapiens
Every Homo sapiens is an animal
The human is equal to the animal
This apparent counter-example had already been mentioned by
Ibrhm Levz H seyn in his commentary on Pehlevanis treatise.
H seyn had proposed amending the conclusion to be Every human
is an animal.47 Again, the uncertainty about the form of the conclu-
sion confirms the impression that this was an area of logic in which
Ottoman logicians could not appeal to precedent. H seyns proposal,
45
It is not clear why Ahmed kir considered this proposition a natural one.
46
The printed edition has two feet (li-l-dhirayn), which seems to be a misprint.
47
Levz Hseyn, Sharh al-Risla al-Qiysiyya, fol. 58a58b.
ottoman logic 181
48
One could capture the validity of this argument in standard first-order predicate
logic as follows:
$x(x is a human) & "x(x is a human $y(y is a homo sapiens & x = y)
$x(x is a homo sapiens) & "x(x is a homo sapiens x is an animal)
$x(x is a human) & "x(x is a human $y(y is an animal & x = y)
182 chapter five
49
The example has been adopted from Ibn al-Qaradgh (d. 1936) in his gloss on
Gelenbevs Burhn, see Ibn al-Qaradgh, H shiya al l-Burhn, 313314 (lowest
rubric). Ibn al-Qaradgh correctly noted that the conclusion al-insnu muswin
li-l-fasl is false, though both natural premises are true. However, he was clearly
not alive to the fact that al-insnu muswin li-faslin does indeed follow from the
premises.
50
This judgment of invalidity will be qualified below, but T vskr and Ahmed
kir have no ground for thinking that the counter-example involving quantified
premises is invalid while this argument involving singular premises is valid.
ottoman logic 183
51
On lehr, see Murd, Silk al-durar f ayn al-qarn al-thn ashar (Blq: Dr
al-T iba al-Kubr, 1301/18831884), III, 148; and Mehmed T hir Bursal, Osmnl
Mellifleri (Istanbul: Matbaa-yi mire, 1333/19141915), I, 214.
52
lehr, Risla f l-mantiq (Manuscript 60 Zile 170, www.yazmalar.gov.tr), fol.
34a34b (wa-huwa m yatarakkabu min qadiyyatayni mutaalliqu mah mlil-l
mawdu l-khari bi-h aythu yalzamuhu bi-wsitati muqaddimatin ajnabiyyatin
gharbatin sdiqatin thubtu mah mli l-ukhr li-mawdi l-l).
184 chapter five
53
On Drendev, see Bursal, Osmnl Mellifler, II, 28.
54
Drendev, Sharh al-Risla al-H usayniyya (Manuscript Yahuda 5412, Firestone
Library, Princeton University Library), fol. 11b (marginal comment by author him-
self). Drendevs commentary on al-H usayniyya (a handbook on dialectic) should
not be confused with his earlier gloss on the more widely-used and anonymous com-
mentary on the same handbook (mathalan idh qla A muswin li-B wa B muswin
li-J yuntiju A muswin li-J bi-wsitati khussi l-mddati wa amm intjuhu anna A
muswin li-musw J fa-innam huwa li-dhtihi, al tar annahu idh qla mathalan
al-wh idu nisfu l-ithnayni wa l-ithnni nisfu l-arbaati yasduqu anna l-wh ida nisfu
nisfi l-arbaati wa amma anna l-wh ida nisfu l-ithnyni fa-l, wa kadh h lu akhthari
l-aqyisati bal h lu jamih bi-tibri l-jihti yuntiju natjatan li-dhtihi wa ukhr l
li-dhtihi).
ottoman logic 185
55
The manuscript copy in question is MS Haci Mahmud Efendi 5821, Sleymaniye
Ktphanesi, Istanbul, fols. 1117.
56
Mehmed Sreyya, Sicill-i Osmn, III, 322323.
57
[Izmr], Risla f tah qq aqsm al-qiys (Manuscript Haci Mahmud Efendi 5821,
Sleymaniye, Istanbul), fol. 11b.
58
[Izmr], Risla f tah qq aqsm al-qiys, fol. 12ab.
186 chapter five
Figure II
This is ordered by my thinking (hdh mutarattabun al fikr)
Nothing that is sharp is my thinking (kullu qawiyyin laysa
bi-fikr)
This is ordered by what is not sharp (hdh mutarattabun al m
laysa bi-qawiyyin)
Figure III
What is ordered by my thinking is this (al-mutarattabu al fikr
hdh)
My thinking is not sharp (fikr laysa bi-qawiyyin)
This is ordered by what is not sharp (hdh mutarattabun al m
laysa bi-qawiyyin)
Figure IV
What is ordered by my thinking is this (al-mutarattabu al fikr
hdh)
What is not sharp is my thinking (m laysa bi-qawiyyin fikr)
This is ordered by what is not sharp (hdh mutarattabun al m
laysa bi-qawiyyin)
The examples clearly bring out that the second figure can be reduced
to the first by converting the major premise, the third can be reduced
by converting the minor, and the fourth by converting both. It is also
remarkable that Izmr appears to have given some consideration to
the objection raised by T vskr that the first mood of the second
figure is sterile since it is possible to give the following counter-
example:
Every believer is noble in relation to the unbeliever
No donkey is an unbeliever
No believer is noble in relation to a donkey
Izmr seems to have seen that what does follow from two premises of
this form is not a universal-negative proposition but a universal-
affirmative proposition with a negative semantic dependent: in this
case Every believer is noble in relation to what is not a donkey. He
carefully gave as a conclusion to his own example the affirmative prop-
osition This is ordered by what is not sharp, and not the negative
ottoman logic 187
This is not ordered by what is sharp. This insight is, however, rare in
Ottoman writings on unfamiliar syllogisms. As will be clear from sub-
sequent chapters, it resurfaces only much later, in the early-twentieth
century.
Izmr went on to discuss the relationship between the unfamil-
iar syllogism and the syllogism of equality. He stated that in a syllo-
gism of equality the predicates in the two premises are the same,
whereas this is not always the case in an unfamiliar syllogism. This is
in accordance with the amendment to the standard definition of syl-
logism of equality proposed by irvn and Pehlevn. He also stated,
again in line with irvn and Pehlevn, that the syllogism of equality
only implies a conclusion by means of an extrinsic premise, whereas
the unfamiliar syllogism implies a conclusion by itself. Of course, as
pointed out by Drendev and T vskr, this is not a defensible posi-
tion, for the premises A is equal to B & B is equal to J have the same
predicate and yet imply by themselves A is equal to what is equal to
J. Izmr ended his discussion by apparently accepting this view, for
immediately after presenting the view that a syllogism of equality
does not imply a conclusion by itself whereas an unfamiliar syllogism
does, he went on to write that the verified view (tah qqan) is that
unfamiliar syllogism is the more general concept (aamm) in the
sense that any syllogism of equality is an unfamiliar syllogism, but
not vice-versa. For example, he wrote, the premises Zayd is equal
to Amr & Amr is equal to Bakr constitute a syllogism of equality
insofar as they imply by means of an extrinsic premise that Zayd is
equal to Bakr, and they constitute an unfamiliar syllogism insofar
as they imply by themselves that Zayd is equal to what is equal to
Bakr.59
59
[Izmr], Risla f tah qq aqsm al-qiys, fol. 12b.
60
On Velcn, see Brsal, Osmnl Mellifler, II, 34.
188 chapter five
the subject of the major premise. He added briefly that such argu-
ments do not produce a conclusion except when an extrinsic premise
is true (l yuntiju ill inda sidqi l-muqaddimati l-ajnabiyya). By 1793,
when Mustaf Siverek wrote a commentary on the work, such a
view appeared imprecise. According to the insight of scholars like
Drendev, T vskr, lehr, and Izmr, the paradigmatic syllo-
gism of equality A is equal to B & B is equal to J does produce by
itself the conclusion A is equal to what is equal to B. Even the stand-
ard example of a supposedly sterile syllogism of equality, such as
A is half of B & B is half of J produces by itself that A is half of half of
J. Siverek accordingly amended Velcns statement. The syllogism of
equality does produce a conclusion, regardless of the truth of any
extrinsic premise, insofar as it constitutes an unfamiliar syllogism:
For example, if you say A is equal to B & B is equal to J, then if the con-
clusion is A is equal to J the syllogism is a syllogism of equality. If it [the
conclusion] is A is equal to what is equal to J then it is an unfamiliar
syllogism and is productive without the need for an extrinsic premise.61
It is thus clear that Siverek, like T vskr, lehr, and Izmr, held
that any syllogism of equality is also an unfamiliar syllogism. The dif-
ference between a syllogism of equality and an unfamiliar syllogism,
he wrote, is that in the former the predicates in the two premises are
identical, whereas this may or may not be so in the case of unfamiliar
syllogisms. Furthermore, in a syllogism of equality the purported con-
clusion leaves out more than merely the recurrent semantic dependent
of the minor premise. For example, in the paradigmatic A is equal to
B & B is equal to J, so A is equal to J the conclusion leaves out not
only the recurrent B but also one of the two predicates. In an unfa-
miliar syllogism only the recurrent semantic dependent of the minor
premise is omitted in the conclusion, as in A is equal to B & B is equal
to J, so A is equal to what is equal to J.
61
Siverek, Sharh Ismat al-adhhn (Manuscript 06Hk3118, www.yazmalar.gov.tr),
25 (mathalan idha qulta A muswin li-B wa B muswin li-J fa-in jaalta l-natjata A
muswin li-J kna l-qiysu qiysa muswtin wa in jaalta <l-natjata> A muswin
li-muswin li-J kna l-qiysu ghayra mutarafin wa muntijan bil ihtiyjin il
l-qadiyyati l-ajnabiyyati).
62
On dim, see Brsal, Osmnl Mellifler, II, 296298. See also Y. Sarkaya, Ebu
Said Hadimi (Istanbul: Kitap Yaynevi, 2008).
ottoman logic 189
63
dim, Aris anzr al-abkr (Manuscript Yahuda 5310, Firestone Library,
Princeton University Library), fol. 118b.
64
The particular division is not original to dim. It is already found in Aristotle
or at least in early Arabic translations of Aristotles Prior Analytics (see Lameer,
Al-Frb and Aristotelian Syllogistics, 127128). dims source may have been the
introductory section on logic in a widely-studied handbook on theology by Baydw
(d. 1316), see Baydw, Tawli al-anwr min matli al-anzr, ed. Abbs Sulaymn
(Cairo & Beirut: al-Maktaba al-Azhariyya li-l-Turth & Dr al-Jl, 1991), 60.
65
dim, Aris anzr al-abkr, fol. 130ab.
190 chapter five
does not follow from the premises themselves, but only on the added
assumption that these premises are true.66
In a later section on the middle-term, dim wrote that the stand-
ard view is that what recurs in the two premises of a syllogism must be
the term itself. However, he opined that this is too restrictive. A syllo-
gism remains productive if what recurs in the second premise is not
the term itself but another term implied by it. For example, the follow-
ing two arguments are, according to dim, proper syllogisms:67
Zayd is a human
Every speaker is an animal
Zayd is an animal
Zayd is a human
Every animal walks
Zayd walks
By the same principle, dim added, if the recurrent term occurs with
a qualifier (qayd) in one premise and without a qualifier in the other,
then the argument may remain productive. For example, the following
argument is productive:68
This is a rational animal (hdh h ayawnun ntiq)
Every animal breathes (kullu h ayawnin mutanaffis)
This breathes (hdh mutanaffis)
66
See j, Sharh Mukhtasar Ibn al-H jib (Blq: al-Matbaa al-Amriyya, 1316/1898
1899), 42 (margins). The view was supported by Taftzn (d. 1390) and rejected by
Jurjn (d. 1413) in their respective glosses on js commentary. It is interesting that
Frege by all accounts the founder of modern logic also held that one cannot infer
anything from false propositions, see M. Dummett, Frege: Philosophy of Language
(Cambridge, MA: Harvard University Press, 1981), 309.
67
dim, Aris anzr al-abkr, fol. 148b. These examples are mentioned by Ism
al-Dn Isfarn in his gloss on Qutb al-Dn al-Rzs commentary on the Shamsiyya.
Ism al-Dn mentioned them as additional examples of arguments that are excluded
from the definition of syllogism by the condition that the conclusion must follow
from the premises by themselves (li-dhtih). He went on to write that, in both exam-
ples, the conclusion follows by the mediation (bi-wsitat) of an extrinsic premise, viz.
Every human is a speaker and Every human is an animal, respectively. Strangely, he
then wrote that it is plain that there is no ground for excluding such proofs from the
definition of syllogism (wa-l yukhf annahu l wajha li-ikhrji tilka l-adillati min
h addi l-qiys) despite the fact that he had just given a reason. See Ism al-Dn
Isfarn, H shiya al l-tasdqt, 169170.
68
dim, Aris anzr al-abkr, fol. 148b149a. This point appears to be taken from
Saklzdes Taqrr al-qawnn. See Saklzde, Taqrr al-qawnn al-mutadwala f
ottoman logic 191
70
See on this point S. Read, Thinking about Logic: An Introduction to the Philosophy
of Logic (Oxford: Oxford University Press, 1995), 4954. For an extended discussion,
see his Formal and Material Consequence, Journal of Philosophical Logic 23 (1994):
247265.
71
The earliest dated manuscript I have found is dated 1151/17381739 and is not
an autograph (see MS 06 Hk 2319, available on www.yazmalar.gov.tr).
72
On him, see Brsal, Osmnl Mellifler, II, 47.
73
Ayntb, Sharh al-Aris (Manuscript Haci Mahmud Efendi 5733, Sleymaniye
Ktphanesi, Istanbul), sayfa numeras 133 (innahu khrijun an tarfi l-qiysi
li-adami intjihi bi-mujarradi hayatihi li-takhallufi l-intji maa baqi hayati l-talfi
f mddati l-nisfiyya mathalan).
74
Ayntb, Sharh al-Aris, sayfa numeras 133 (fa-qawlun al-rajulu khayrun min
al-marati wa l-maratu maqsratun f l-khiymi qiysun ghayru mutarafin wa
194 chapter five
Ayntb thus maintained the older view of irvn and Pehlevn that
no syllogism of equality is an unfamiliar syllogism, since in the former
case the predicates are identical and in the latter they are different.
Another example of an unfamiliar syllogism given by Ayntb is of
particular interest. dim had, in the section on fallacies, adduced an
example of a fallacy that has been mentioned on a number of occa-
sions in this study. This is the following argument:
Humans have hair (al-insnu lahu sharun)
All hair grows from a place (kullu sharin yanbutu min mah allin)
Humans grow from a place (al-insnu yanbutu min mah allin)
Ayntb noted that even though the mentioned conclusion indeed
does not follow, the premises can be seen to constitute an unfamiliar
syllogism that produces the conclusion: Humans have something that
grows from a place (al-insnu lahu shayun yanbutu min mah allin).75
V. Conclusion
1
On Gelenbev, see Brsal Osmnl Mellifler, III, 293297. He is sometimes
referred to in bio-bibliographic sources with the Arabic transliteration al-Kalanbaw
(for example in the index to Brockelmann, Geschichte der arabischen Literatur, under
Isml b. Musta f al-Kalanbaw). The appearance of unfamiliar syllogisms in
Gelenbevis work has been noted by A. Bingl, who correctly noted that this category
does not appear in the classical handbooks on logic from the thirteenth and four-
teenth centuries. Bingl, however, merely reproduced Gelenbevs examples. He did
not investigate whence the concept came, nor did he discuss the logic of unfamiliar
syllogisms. See Gelenbevinin Mantik Anlay (Istanbul: Milli Eetim Basmevi, 1993),
105106.
2
Abd l-Nfi, Fenn-i mantik: Mzn-i erh -i Mtercem-i Burhn (Istanbul:
Matbaa-yi mire, 1295/1878); arpt, Nms al-qn Sharh al-Burhn (Lithograph,
Istanbul: Litografya Matbaas, 1274/18571858); Msl, Tanwr al-Burhn (Istanbul:
Matbaat al-Sharika al-Murattabiyya li-rtn sdryn, 1307/18891890); Srz,
Sharh al-Burhn (Istanbul: n.p., n.d.). The edition of Gelenbevs al-Burhn that I have
used was printed with the glosses of the Kurdish scholars Panjiyn (d. 1901) and Ibn
al-Qaradgh (d. 1936), see al-Burhn fi ilm al-mzn (Cairo: Matbaat al-Sada,
1347/19281929).
gelenbev and the logic of unfamiliar syllogisms 197
3
Wright, A Grammar of the Arabic Language, I, 269C.
4
It may be added that Gelenbevs Burhn was written after 1763 and before 1776.
In the introduction to the work, Gelenbev wrote that he composed the handbook for
a particularly intelligent student of his, who is identified by commentators as Mehmed
Sad H asanp-zde. Gelenbev received his certificate to teach in the year 1763, and
the dedicatee of the work died in 1776. Gelenbevs commentary on sghj, appar-
ently written when he was himself an advanced student, makes no mention of unfa-
miliar syllogisms. See Isml Gelenbev, Sharh sghj (Istanbul: irket-i Sahafiye-
Osmniye Matbaas, 1307/18891890). A treatise on syllogisms attributed to
Gelenbev and lithographed in Istanbul in 1297/18791880 with the title K iys Risles
also does not mention unfamiliar syllogisms. However, the attribution to Gelenbev is
doubtful, since one manuscript of the work renders the title as al-Risla al-intjiyya
and the authors name as Abbs Efend (Manuscript Laleli 3713, Sleymaniye
Ktphanesi, Istanbul, fols. 7682). This would seem to be K iys Mehmed mid
(fl. 1175/1761), li-Qiys Efend having been mistranscribed as li-Abbs Efend (which
is easily done in Arabic). Another copy of the treatise appears in a series of epistles on
logic attributed to K iys Mehmed mid, see Manuscript Haci Mahmd Efend 1703,
Sleymaniye Ktphanesi, Istanbul, fols. 88 ff.
198 chapter six
5
Gelenbev, al-Burhn, 315317 (top rubric) (ilam anna ghayra l-mutarafi in
ittah ada fhi mah ml l-sughr wa l-kubr fa-lahu natjatni ih dhum bi-ithbti
kil l-mah mlayni fh wa hiya lzimatun lahu li-dhtihi wa l-ukhr bi-isqti ah adi
l-mah mlayni fh wa hiya l-sdiqatu fm sadaqat al-muqaddimatu l-ajnabiyyatu
l fim kadhabat fa-dhlika l-qiysu bi-l-nisbati il l-natjati l-thniyati yusamm
qiysu l-muswti wa amm bi-l-nisbati il l-natjati l-l fa-mundarijun f l-qiysi
l-mustalzimi li-dhtihi ka-l-ladh ikhtalafa fhi l-mah mlni).
6
Gelenbev, al-Burhn, 317318 (top rubric) ( fa-qawlun l-wh idu nisfu l-ithnayni
wa l-ithnni nisfu l-arbaati qiysun ghayru mutarafin mustalzimun li-dhtihi anna
l-wh ida nisfu nisfi l-arbaati wa qiysu muswtin bi-l-nisbati il natjati anna
l-wh ida nisfu l-arbaati lkinnahu ghayru muntijin lahu li-kidhbi l-muqaddimati
l-ajnabiyyati l-qilati bi-anna nisfa l-nisfi nisfun).
gelenbev and the logic of unfamiliar syllogisms 199
7
See his H shiya al H ashiyat al-Lr al Sharh Hidyat al-h ikma (Istanbul:
Matbaa-yi mire, 1270/18531854); H shiya al Mr al-db (Istanbul: Matbaa-yi
mire,1234/18181819); H shiya al Sharh al-Dawn al l-Aqid al-Adudiyya
(Istanbul: irket-i Hayriye-yi Sahfiye-ye 1307/1890).
8
H shiya al Mr al-Tahdhb (Istanbul: n.p., 1288/1871).
200 chapter six
9
Gelenbev, al-Burhn, 314315 (second rubric) (fa-inna l-awsata in kna
mutaalliqa mah mli l-sughr mawdan f l-kubr fa-huwa l-shaklu l-awwalu nahwa
hdh ghulmu rajulin wa kullu rajulin insnun fa-hdh ghulmu insnin wa
yushtaratu bi-jbi l-sughr wa kulliyyati l-kubr li-takhallufi l-intji f qawlin ghulmu
l-marati laysa bi-ghulmi rajulin wa kullu rajulin insnun aw mudhakkarun fa-
l-h aqqu f l-awwali l-salbu wa f l-thn l-jbu wa f qawlin ghulmu l-rm ghulmu
insnin wa badu l-insni abyadu aw aswadu wa l-h aqqu f l-awwali l-jbu wa f l-thn
l-salbu).
10
Gelenbev, al-Burhn, 315 (second rubric) (wa in kna mutaalliqa mah mli
l-sughr mah mlan f l-kubr aydan fa-huwa l-shaklu l-thn nahwa hdh ghulmu
rajulin wa l shaya min al-marati bi-rajulin fa-hdh laysa bi-ghulmi mraatin wa
yushtaratu bi-khtilfi muqaddimatayhi f l-kayfi wa kulliyyati l-kubr li-l-takhallufi f
qawlin ghulmu l-marati ghulmu h ayawnin wa kullu insnin aw farasin h ayawnun
fa-l-h aqqu f l-awwali l-jbu wa f l-thn l-salbu wa f qawlin ghulmu l-marati laysa
bi-ghulmi rajulin wa l shaya min al-muannathi aw al-farasi bi-rajulin fa-l-h aqqu
gelenbev and the logic of unfamiliar syllogisms 201
13
Unless otherwise indicated, the following is based on Gelenbev, al-Burhn,
314317 (second rubric, comprising Gelenbevs own annotations to his work); Srz,
Sharh al-Burhn, II, 118119.
14
Gelenbev, al-Burhn, 159 (top rubric).
15
$x(Bx) & "x(Bx $y(Ay & Ixy) ) & $x(Ax) & "x(Ax Nx) $x(Bx) & "x(Bx
$y(Ny & Ixy)
gelenbev and the logic of unfamiliar syllogisms 203
16
$x(Bx) & "x(Bx $y(My & Sxy) ); "x(Mx ~Wx) $x(Bx) & "x(Bx
$y(~Wy & Sxy) )
17
$x$y(Bx & Ay & Ixy); "x(Ax Nx) $x$y(Bx & Ny & Ixy)
204 chapter six
18
To be read as: For all x, if x is human then there exists a y such that y is a horse
and x is not the offspring of y.
19
To be read as: For all x, if x is a human then there does not exist a y such that y is
a horse and x is the offspring of y.
20
"x (Hx $y(Ey & ~Oxy) ) & "x (Ex Ax) |= "x (Hx $y(Ay & ~Oxy) )
21
"x (Hx ~$y(Ey & Oxy) ) & "x (Ex Ax) | "x (Hx ~$y(Ay & Oxy) )
22
Rm in the Arabic tradition originally denoted the inhabitants of the Byzantine
(Roman) Empire, and continued to be used in post-Byzantine times of the set-
tled Turkish-speaking people of the Ottoman Empire. The Rm was stereotypically
white.
23
Note that this need not be true unless one additionally stipulates that two
people, one a Rm and the other black, cannot co-own a slave. However, the counter-
example can easily be amended and the underlying claim that the major premise must
be universal is correct. Consider:
Every human is the offspring of a man
Some man is childless/Some man is a father
206 chapter six
24
The printed edition has kullu instead of l, which is obviously a misprint.
25
The clearest statement of this is in Ahmed H ayt Mara, H shiya al Sharh al-
Qiysiyya. Printed as an appendix to Mehmed T vskrs Sharh al-Risla al-Qiysiyya
(Istanbul: Matbaa-yi mire, 1281/18641865), 37.
26
Ra & Aa; "x(Dx ~Rx) ~(Da & Aa)
208 chapter six
27
"x(Bx "y(Dy ~Sxy) ) & [$x(Cx) & "x(Cx Dx)] "x(Bx "y(Cy
~Sxy)
gelenbev and the logic of unfamiliar syllogisms 209
28
Ahmed Sidk Brsev, Mzn al-Intizm (Istanbul: Matbaa-yi mire, 1306/1888
1889), 221 (wa sumia min h udhdhqi l-ulami l-almi anna hdh l-sharta innam
huwa li-mutarafi l-shakli l-thni l li-ghayri mutarafihi).
gelenbev and the logic of unfamiliar syllogisms 211
slave of some animal. Had the sense been The slave of a woman is the
slave of all animals then indeed the conclusion would follow that
The slave of a woman is the slave of a (indeed every) horse. Again,
one can see the limitation of not explicitly quantifying the semantic
dependent of the relational premise in an unfamiliar syllogism. This
lack means that the following two propositions were seen as having
the same form:
(i) Affirmation is nobler than privation
(ii) The slave of a woman is the slave of an animal
The fact that (i) clearly says that affirmation is nobler than all and
not just some privation cannot be captured. Hence, the validity of an
argument that depends on understanding (i) in this stronger sense
cannot be shown.
With respect to the second condition for productivity in the second
figure, viz. that the major must be universal, Gelenbev gave the fol-
lowing example:
The slave of a woman is the slave of an animal
Some body is not an animal
Here a true conclusion would have to be affirmative. If in the major
premise we replace body with inanimate object the conclusion
would have to be negative.
29
$x$y(My & Syx) & "x"y(My & Sxy) Hx)] & $x(Mx) & "x(Mx Ax)
$x$y(Ay & Sxy & Hx)
30
Pirizrenl, Sharh al-Burhn (Manuscript Yazma Baislar 257, Sleymaniye
Ktphanesi, Istanbul), sayfa numeras 284. The manuscript is unpaginated, and
hence I give the digital page number (sayfa numeras) on which the folio can be
located on-line (wa l-sawbu fa-badu l-insni ghulmu-h ayawnin lianna l-sughr
idh akastah li-l-raddi il l-shakli l-awwali yaknu l-qiysu hkadh badu l-insni
ghulmu rajulin wa kullu rajulin h ayawnun yuntiju anna bada l-insni ghulmu
h ayawnin kam dhakarn).
31
See Ibn al-Qaradgh, H shiya al l-Burhn. Printed with Gelenbev, al-Burhn
fi ilm al-mzn (Cairo: Matbaat al-Sadah, 1347/19281929), 315 (lowest rubric).
gelenbev and the logic of unfamiliar syllogisms 213
(iii) Converting the major and switching the premises would result in
the following argument:
Some created thing is an accident
Everything implying an accident is a body
gelenbev and the logic of unfamiliar syllogisms 215
sterility of this figure when the minor is negative and the major is
affirmative:32
(1) The offspring of a man is not a horse
Every man is an animal
(2) The offspring of a man is not a horse
Every man is human
In the first case, a conclusion in this figure would have offspring of an
animal as subject and horse as predicate, and would be true as an
affirmative proposition. In the second case, the conclusion would have
offspring of a human as subject and horse as predicate and would
have to be negative.
To show that the figure is sterile with a negative minor and a nega-
tive major, Pirizrenl adduced the following:
The offspring of a man is not a horse
No man neighs/No man is inanimate
A conclusion with the offspring of what neighs as subject, and horse
as predicate, would have to be affirmative. A conclusion with the off-
spring of what is inanimate as subject and horse as predicate would
have to be negative.
To show the sterility of moods in which both premises are particu-
lar, Pirizrenl adduced the following:
The offspring of some animal is human
Some animal is articulate/Some animal is a horse
A conclusion with The offspring of some articulate as subject and
human as predicate would have to be affirmative. A conclusion with
The offspring of some horse as subject and human as predicate
would have to be negative.
32
Pirizrenl, Sharh al-Burhn, sayfa numeras 284.
gelenbev and the logic of unfamiliar syllogisms 217
33
Pirizrenl, Sharh al-Burhn, sayfa numeras 284 (wa l-sawbu fa-badu l-h ayawni
ghulmu rmiyyin liannaka idh akasta tartbah li-yartadda il l-shakli l-awwali
thumma akasta l-natjata wa huwa l-mutabaru f intji l-rbii taknu l-natjatu m
dhakarn hkadh kullu rmiyyin insnun wa ghulmu l-insni h ayawnun fa-
ghulmu kullu rmiyyin h ayawnun wa idh akasta l-natjata taknu badu l-h ayawni
ghulmu rmiyyin).
34
See Qutb al-Dn al-Rz, Tah rr al-qawnn al-mantiqiyya, II, 41 ff (1905 ed.);
9497 (1948 ed.).
gelenbev and the logic of unfamiliar syllogisms 219
However, this does not solve the problem, because the conclusion
should convert to Some animal is the son of a childless man which is
clearly false.
On the second, h aqq reading, an affirmative proposition does not
have straightforward existential import. On this account, a proposi-
tion such as Every phoenix is a bird is true, since it is understood in
the sense that Anything that is a phoenix (if it exists) is a bird (if it
exists). Gelenbev followed the tradition in making this distinction,
and wrote that in the case of a h aqq reading the subject of an affirma-
tive proposition is assumed to have extra-mental existence (al-wujd
al-khrij l-muqaddar).35 This assumption would amount to adding to
the argument a premise stating that the subject of the conclusion
exists. If such a premise is added, then the first mood of the fourth fig-
ure becomes valid.
It is noteworthy that the same problem with existential import
bedevilled Augustus De Morgans attempts at developing the logic of
relations in the nineteenth century. As one of his contemporary critics
pointed out, the inference that was meant to exhibit the limitations of
traditional Aristotelian logic, viz. The human is an animal, so the
head of a human is the head of an animal, falls afoul of the principle
that affirmative propositions have existential import.36 The following
appears to be invalid:
The bachelor is a man
The wife of a bachelor is the wife of a man
In the tradition of Ibrhm H seyn, mer K alack and Osmn
K abakczde Tarss, the conclusion of the first mood of the fourth
figure is: Some slave of an animal is a Rm. This is much more obvi-
ously invalid, as is clear from the following counter-example adduced
by Ibn al-Qaradgh:37
The daughter of a man is an animal
Every male human is a man
Some daughter of an animal is a male human
35
See Gelenbev, al-Burhn, 173 (top rubric).
36
See on this point, D.D. Merrill, Augustus De Morgan and the Logic of Relations
(Kluwer: Dordrecht, 1990), 110112.
37
See Gelenbev, al-Burhn, 315 (lowest rubric).
220 chapter six
38
Ibn al-Qaradgh, H shiya al l-Burhn, 316 (lowest rubric). The example given
by Ibn al-Qaradgh in the printed text is not in the fourth figure at all, but in the
third figure:
ghulmu insnin h ayawn
l shaya min al-h ajari bi-h ayawn
ghulmu badi m laysa bi-h ajarin h ayawn
I take it that this is a slip of the pen, or that the printed text is corrupt.
39
This is missing from the printed edition of Srzs commentary.
222 chapter six
40
Pirizrenl, Sharh al-Burhn, sayfa numeras 285.
gelenbev and the logic of unfamiliar syllogisms 223
The conditions for productivity that Gelenbev gave for the fourth
figure presumes that there are eight productive moods. This was a
common feature of Arabic logical writings after Athr al-Dn al-Abhar
(d. 1265) who had pointed out that three additional moods of the
fourth figure besides the standard five become productive if the
particular-negative proposition converts, which it does in certain
complex modality propositions.41 The three additional moods which
are productive when the particular-negative proposition converts are:
41
See the remarks by Ts in Tadl al-miyr, 193 and in Tajrd al-mantiq, printed
with H ills commentary al-Jawhar al-nadd f sharh mantiq al-Tajrd, ed. Muhsin
Bidarfar (Qum: Intisharat Bidar, 1423/20022003), 156.
224 chapter six
III. Conclusion
42
G. Saliba, Islamic Science and the Making of the European Renaissance (Cambridge:
MIT Press, 2007), 279 n. 21.
43
Gelenbev, Rislat al-imkn (Istanbul: n.p. 1309/1892).
44
Jurjn also wrote a gloss on the commentary by j (d. 1355) on a handbook on
jurisprudence by Ibn al-H jib (d. 1246) which covered the opening section on logic in
that handbook, as well as two short and introductory Persian epitomes of logic. He
did not engage with the more technical issues of modal or conditional logic in any of
these works.
226 chapter six
45
See H asan, Nuzhat al-khawtir, VI, 306308.
46
On K ara H all, see Brsal, Osmnl Mellifleri, I, 403. His extant works on logic
include: (i) a gloss on the gloss of Mr Ab l-Fath on Dawns incomplete commen-
tary on Tahdhb al-mantiq; (ii) a gloss on the gloss by Mull Dad al-Khawf
(fl. 1460) on the first half (dealing with conceptions) of Qutb al-Dn al-Rzs com-
mentary on Ktibs Shamsiyya; (iii) an extensive gloss on a commentary by Mehmed
Emn irvn on remarks made by Mehmed Fenr in the introduction to his com-
mentary on Abhars sghj on what makes the numerous inquiries of logic one
discipline; (iv) a gloss on the gloss of Kl Ahmed (fl. 16th century) on Fenrs com-
mentary on sghj. Only the last of these works deals with formal topics (contradic-
tion, conversion, contraposition, the immediate implications of conditionals, and the
formal syllogistic), and even that work is heavily oriented toward non-formal topics.
In one printed edition of the gloss, for example, only 37 pages (from page 184 to 221)
are dedicated to formal topics, out of a total of 231 pages, see K ara H all, H shiya al
l-Fanr (Istanbul: Yahy Efend 1289/18731874).
47
For example, in the lengthy commentary by Hill (d. 1761) on Qdirs didactic
poem on logic (discussed in chapter 4), around half of the works approximately
355 (unnumbered) pages in the lithograph edition published in 1895 are devoted
to contradiction, conversion, the immediate implications of conditionals, and the
syllogistic.
gelenbev and the logic of unfamiliar syllogisms 227
Appendix
Universal affirmative
Every boy is male .. $x(Bx) & "x(Bx Mx)
Every boy is the son of a man ... $x(Bx) & "x(Bx $y(My &
Sxy) )
Every son of a man is human.. $x$y(My & Sxy) & "x"y( (My &
Sxy) Hx)
Particular affirmative
Some boy is male .. $x(Bx & Mx)
Some boy is the son of a tall man ..... $x(Bx & $y(Ty & Sxy) )
Some son of a man is tall .. $x$y(Mx & Sxy & Tx)
Universal negative
No boy is female . "x(Bx ~Fx)
No boy is the son of a donkey .. "x(Bx ~$y(Dy & Sxy) )
No son of a man is female .. "x($y(My & Sxy) ~Fx)
Particular negative
Some boy is not tall ~$x(Bx) v $x(Bx & ~Tx)
Some boy is not the son of a scribe .. ~$x(Bx) v $x(Bx & "y(Wy
~Sxy) )
Some son of a man is not tall . ~$x$y(My & Sxy) v $x$y
(My & Sxy & ~Tx)
CHAPTER SEVEN
I. Decline?
have been printed. It is, for example, much more advanced than the
perhaps most well-known nineteenth-century Arabic work on logic,
Mull Hd Sabzavrs Sharh al-Lal l-muntazama. However, Srz
devoted less than two pages to unfamiliar syllogisms, out of a total of
324 pages in the printed edition of his work. He gave examples of all
the moods of the four figures, tried to prove the productive moods of
second-figure unfamiliar syllogisms, and showed by counter-example
the sterility of third-figure unfamiliar syllogisms that did not satisfy
the stated conditions of productivity. Srzs commentary was fol-
lowed by two shorter and mainly explanatory commentaries, by Ysuf
kr H arpt (d. 1875) and H asan H sn Msll (fl. 1878), and a
popular and frequently printed two-volume commentary in Ottoman
Turkish by Abd l-Nfi Iffet Efend (d. 1890). These commentaries
did not add anything to the discussion of unfamiliar syllogisms, and
merely incorporated Gelenbevs own glosses as a running commen-
tary to the text.1 Gelenbevs handbook was also printed in Cairo in
1929 with the extensive glosses of the Kurdish scholars Panjiyn
(d. 1901) and Ibn al-Qaradgh (d.1936). Particularly the gloss of Ibn
al-Qaradgh is sometimes incisive, but again there was no significant
development of the logic of unfamiliar syllogisms in these glosses. It is
striking that none of these later commentators bothered to give counter-
examples showing the sterility of third- and fourth-figure syllogisms
that do not satisfy the mentioned conditions of productivity, despite
Gelenbevs specific injunctions.
Gelenbevs presentation of unfamiliar syllogisms was incorporated
almost verbatim in a number of other nineteenth-century Ottoman
works. This is the case, for example, in the commentary on Abhars
sghj by Ahmed Rd K araga (d. 1835), and in the independent
exposition of logic entitled Mzn al-adl by H c Kerm Ams
(d. 1886).2 In the final decades of the nineteenth century, Ottoman
Turkish started to displace Arabic among Ottoman Turkish scholars as
the preferred language for writing on logic. A number of handbooks
in that language were written in this period. Examples are Miyr-i
Sedd by Ahmed Cevdet P (d. 1895); yet l-beyn f ilm il-mzn
1
H arpt, Nms al-qn sharh al-Burhn, 158; Msll, Tanwr al-Burhn, 216;
Abd l-Nfi, Fenn-i Mantik, II, 190192.
2
Karaga, Tuh fat al-Rushd l-Qaraghaj bi-sharh sghj (Istanbul: Matbaa-yi
Amire, 1279/1862), 193; Ams, Mzn al-adl (Istanbul: Ceride-yi Havadis Matbaasi,
1276/18591860), 6970.
230 chapter seven
3
erkezde, yet l-beyn f ilm il-mzn (Istanbul: Mekteb-i Sanayi Matbaas,
1306/18881889), 5153; Ahmed Cevdet, Miyr-i sedd (Istanbul: Karabet ve Kasbar
Matbaas, 1303/1886), 7981; Sleymn Sirr, Mantik (Istanbul: Karabet Matbaas,
1310/1892), 118121; Rid Efendi, Mzn l-makl (Istanbul: Mekteb-i Fnun-
Harbiye-yi ahane Matbaas, 1315/1899), 103104.
the ottoman tradition 231
4
Tkd, al-Durr al-nj, 104105; Tarss, Risla-yi Istidlliye, 1314; Magnisv,
Sharh al-Istidlliyya, 62.
5
Tarss, Risla-yi Istidlliye, 14; Manisv, Sharh al-Istidlliyya, 63.
6
Mustaf Kmil Yemlhzde, Manhij al-ihtijj (Manuscript 03 Gedik 18113,
www.yazmalar.gov.tr), fol. 110a110b; (Manuscript 45 Hk 7831, www.yazmalar.gov
.tr), fol. 150b. The same paradigm is repeated in the commentary on Yemlhazdes
versification of Pehlevns treatise by his nephew H all Esad Yemlhzde, see Jil
al-uyn: Sharh al-Qiysiyya min Nazm al-funn (Istanbul: Cemal Efendi Matbaas,
1308/1890), 85.
232 chapter seven
7
Akehirl, al-Qasda al-Azziyya wa Sharh uh (Istanbul: Matbaat al-H jj
Muharram al-Bsnav, 1292/1875), 4142.
8
H arpt, Nms al-qn, 158.
the ottoman tradition 233
9
Akhisr, Sharh al-Risla al-Qiysiyya (Manuscript 45 Ak Ze 5983. www.yazmalar
.gov.tr), sayfa numeras, 38. The manuscript is unpaginated, and hence I give the
digital page number (sayfa numeras) on which the folio can be located on-line.
234 chapter seven
10
Ahmed Sidk Brsev, Mzn al-Intizm, 221 (wa sumia min h udhdhqi l-ulami
l-almi anna hdh l-sharta innam huwa li-mutarafi l-shakli l-awwali l li-ghayri
mutarafihi).
the ottoman tradition 235
11
R.L. Chambers, The Education of a Nineteenth-Century Ottoman Alim: Ahmed
Cevdet Paa, International Journal of Middle East Studies, 4 (1973): 440464, esp. 440,
445; J. R. Barnes, An Introduction to Religious Foundations in the Ottoman Empire.
Leiden: Brill, 1986, esp. chapter VIII; N. Berkes, The Development of Secularism in
Turkey (Montreal: McGill University Press, 1964), esp. 99 ff. The decline of traditional
Ottoman colleges in the nineteenth century is also amply documented in Y. Sarkaya,
Medreseler ve Modernleme (Istanbul: z Yaynclk, 1997). I would like to thank
Aslhan Grbzel for bringing this source to my attention.
236 chapter seven
12
Tarss, Risle-yi Istidlliye, 2.
13
Mustaf Kmil Yemlhzde, Manhij al-ihtijj, fol. 93b94a (03 Gedik 18113);
fol. 147b (45 Hk 7831). According to the on-line catalogue of Harvard University
Library, there is a printed copy of Yemlhzdes work, printed in 1273/1856 in 32
pages by Matbaa-yi Amire in Istanbul. However, this copy can no longer be located
and appears to be lost.
238 chapter seven
14
Jm, Sharh al-Kfiya (Istanbul: irket-i Sahfiye-yi, 1324/1906), 6 (inna l-kalma
l yah su lu bidni l-isndi wa l-isndu l budda lahu min musnadin wa musnadin ilayhi
wa hum l yatah aqqaqni ill f ismayni aw f ismin wa fil).
15
[Izmr], Risla f tah qq aqsn al-qiys, fol. 16a.
the ottoman tradition 239
Tarss wrote that such a syllogism was called ifte sugral (i.e. double-
minor) in Turkish.16 One can confidently infer from his remark that
he was not the first to recognize such syllogisms. Yet, the term does
not seem to occur in earlier works, and may have originated instead in
oral pedagogical practices common in the eighteenth-century. This is
indicated by the very fact that the term is Turkish; had the syllogism
first been recognized in a theoretical work on logic then one would
have expected it to be known by an Arabic name, for Arabic at the
time of Tarss was still overwhelmingly the language used to write on
logic in the Ottoman Empire. Neither Tarss nor any other nine-
teenth-century Ottoman scholar whose works are known to me dis-
cussed why such an inference is legitimate, or what the conditions of
its productivity are. The double-minor syllogism seems to have been
an innovation of applied logic that was never fully integrated into the
works on theoretical logic in the older, established tradition.
Another innovation that seems to have its origin in the keen inter-
est in applied logic is the unfamiliar unfamiliar syllogism (ghayru
mutarafi l-ghayri mutaraf). Writing in the eighteenth century,
Pehlevn and Gelenbev had only considered relational inferences in
which the first (i.e. minor) premise expressed a relation and the sec-
ond (major) premise is non-relational. In unfamiliar unfamiliar syllo-
gisms the situation is reversed: the first premise is non-relational and
the second expresses a relation. Again, Tarsss treatise seems to be
the first written occurrence of the term unfamiliar unfamiliar
syllogism.17 He gave two examples, the first of which is derived from
glosses on the preamble of Mehmed Fenrs (d. 1431) widely-studied
commentary on Abhars sghj. The glossator, an obscure sixteenth-
century scholar by the name of Kl Ahmed, had discussed the open-
ing sentence of Fenrs preamble, addressed to God: A praising unto
Thou (h amdan laka). He noted that grammatically one should under-
stand there to be a suppressed verb in the sentence, which could be
either I praised or I praise. Kl Ahmed then added that the latter is
preferable since the imperfect indicative indicates continuous activity,
whereas the perfect indicates that the activity is completed.18 Tarss
regimented the argument as follows:
16
Tarss, Risle-yi Istidlliye, 1415.
17
Tarss, Risle-yi Istidlliye, 12.
18
Kl Ahmed, H shiya al l-Fanr. Printed as an appendix to Fenr, Sharh
sghj (Istanbul: al-Matbaa al-H arbiyya, 1278/1861), 28.
240 chapter seven
19
T vskr, Sharh al-Risla al-Qiysiyya, 13 (printed ed.).
20
H ayt Mara, H shiya, 37.
the ottoman tradition 241
annotation by the author.21 However, Tarss did not integrate the con-
cept into his main text, instead just presenting the four figures of the
unfamiliar syllogism. In the work of a later nineteenth-century scholar
like Musta f Kmil Yemlhzde, one can find unfamiliar unfamiliar
syllogisms alongside unfamiliar syllogisms in the main presentation
itself, and each type is divided into the four figures.22 Yemlhzde
defined a first-figure unfamiliar unfamiliar syllogism as a two-prem-
ised argument in which the semantic dependent of the subject of the
major premise is a predicate in the minor, for example:
Magic is knowledge
The possessor of knowledge is precious
The possessor of magic is precious
As mentioned in the previous chapter, inverting the order of premises
of this argument would result in an unfamiliar syllogism of the first
mood of the fourth figure.
In the second-figure unfamiliar unfamiliar syllogism, the semantic
dependent of the predicate of the major is a predicate in the minor
for example:
Every human is an animal
Nothing inanimate is equal to an animal
No human is equal to an inanimate [entity]
Inverting the order of the premises results in a second-figure unfamil-
iar syllogism with the conclusion Nothing inanimate is equal to a
human. For some reason, Yemlhzde gave the converse as the
conclusion.
In the third-figure unfamiliar unfamiliar syllogism, the semantic
dependent of the subject of the major is subject of the minor, for
example:
Every human is an animal
Every instance of human is a speaker
Again, inverting the order of the premises results in a third-figure
unfamiliar syllogism, whose conclusion is: Some instance of an ani-
mal is a speaker.
21
Tarss, Risle-yi Istidlliye, 12 (margins).
22
Yemlhzde, Manhij al-ihtijj, fol. 109b111a (03 Gedik 18113); fol. 150b (45
Hk 7831).
242 chapter seven
23
Brsev, Dharat al-imtih n (Istanbul: Matbaat Ahmed Kemal, 1325/1907), 103.
the ottoman tradition 243
24
Bursal, Osmanl Mellifler, I, 385387.
25
H czde Kilis, al-H shiya al-jadda al l-Tasawwurat (Istanbul: Matbaa-yi
Amire, 1289/1872), 3.
26
C. Neumann, Bad Times and Better Self: Definitions of Identity and Strategies
for Development in Late Ottoman Historiography, 18501900, in The Ottomans and
the Balkans: A Discussion of Historiography, ed. F. Adanr and S. Faroqhi (Leiden: Brill,
2002), 5778, esp. 6667.
244 chapter seven
27
M.O. Alkan, Modernization from Empire to Republic and Education in the
Process of Nationalism, in Ottoman Past and Todays Turkey, ed. K. Karpat (Leiden:
Brill, 2000), 47132, esp. 64, 68.
28
H czde Kilis, Usl-i Cedde Zbdes (Istanbul: Matbaa-yi Amire, 1290/1873).
29
Mehmed T hir Kilisl, sgc Rislesinden Dellet zerne K yslar (Istanbul:
El-Ha Halil Efendi Matbahanesi, 1287/1870); Abd-Allh Edb Ayntb, Usl- akyise
takrri (Istanbul: Saday- Millet Matbaas, 1327/1911).
30
H czde Kilis, al-H shiya al-jadda al l-Tasdqt (Istanbul: Matbaa-yi mire,
1275/1858), 2.
31
H czde Kilis, al-H shiya al-jadda al l-Tasdqt, 4 (wa amm ghayru
mutarafi l-awwali fa-huwa an yakna mutaalliqu mah kmi bihi l-sughr mah kman
alayhi bi-tammihi f l-kubr bi-sharti an yakna mah ml l-muqaddimatayni
mutaghyiratayni li-yakhruja anhu l-muswt).
the ottoman tradition 245
32
H czde Kilis, al-H shiya al-jadda al l-Tasdqt, 304.
33
H czde Kilis, al-H shiya al-jadda al l-Tasdqt, 45.
34
H czde Kilis, al-H shiya al-jadda al l-Tasdqt, 12.
35
This is stated in Ahmed Sidk Brsev, Mzn al-intizm, 222.
246 chapter seven
36
H czde Kilis, al-H shiya al-jadda al l-Fanr (Lithograph, Istanbul: n.p.,
1287/1870), 15.
37
H czde Kilis, al-H shiya al-jadda al l-tasdqt, 237.
the ottoman tradition 247
38
H czde Kilis, al-H shiya al-jadda al l- tasdqt, 9.
248 chapter seven
39
H czde Kilis, al-H shiya al-jadda al l- tasdqt, 45.
40
H czde Kilis, al-H shiya al-jadda al l- tasdqt, 53.
41
H czde Kilis, al-H shiya al-jadda al l- tasdqt, 32. On the same page, there
is another example of an unfamiliar syllogism in the third figure that follows the same
paradigm.
the ottoman tradition 249
42
H czde Kilis, al-H shiya al-jadda al l- tasdqt, 132, 239, 250.
43
H czde Kilis, al-H shiya al-jadda al l- tasdqt, 325 (inna l-maqsda bi-l-
intji ay bi-intji l-qiysi huwa istilzmuhu li-ah adihim ay li-l-natjati l-mjibati
mutta ridan aw li-l-natjati l-slibati mutta ridan wa mat lam yatta rid lam yuntij wa in
sadaqa l-qiysu wa l-natjatu f badi l-mawdd).
250 chapter seven
44
On Maks d, see A. Battal-Taymas, Iki Maksudiler: Sadri-Arsal, Ahmed-Hadi:
kiilikleri, fikir hayatlar ve eserleri (Istanbul: Sralar Matbaas, 1959).
45
Maks d, Mzn al-afkr (Kazan: B.L. Dombrovskago, 1903), 6.
46
Maks d, Mzn al-afkr, 154 (wa inn knat f l-sughr mafhman mutaalliqan
li-mah mlih wa mawdan f l-kubr).
47
A. Bain, Logic: Deductive and Inductive (New York: D. Appleton & Co. 1889);
W. Minto, Logic: Inductive and Deductive (New York: Charles Scribner & Sons, 1894).
the ottoman tradition 251
48
Of course, Maks d may have been following earlier Tatar scholars on this point.
Nineteenth-century Tatar scholarship in the rational sciences is woefully under-
researched, but would seem to have been of high quality and to have maintained con-
tacts with Central Asian, Indian, and Ottoman scholarship. At least this is the
impression one gets from the kinds of works on logic and philosophy published in
Kazan in the late-nineteenth and early twentieth centuries, for example Gpamvs
commentary on Bihrs Sullam al-ulm (in 1887) and Ibn Mubrakshhs commen-
tary on Ktibs H ikmat al-ayn with the glosses of Jurjn and Bghnav (in 1902). The
published works of the Tatar scholar ihb l-Dn Mercn (d. 1889) on theology and
jurisprudence also reinforce this impression. His gloss on Sadr al-Sharas Tawdh
al-Tanqh (on jurisprudence) was published in Kazan in 1889, and his gloss on
Dawns Sharh al-Aqid al-Adudiyya (on theology) was published in Istanbul in
18891890 in 2 volumes along with the gloss of Gelenbev.
49
Maks d, Mzn al-afkr, 165.
252 chapter seven
50
Maks d, Mzn al-afkr, 165166.
51
Maks d, Mzn al-afkr, 168.
the ottoman tradition 253
Zayd is a father & Bakr is his son (Zayd abun wa Bakr ibnuhu)
Every father is older than his son (kullu abin akbaru min ibnihi)
Zayd is older than Bakr (Zayd akbaru min Bakr)
Another example given is:
Zayd is my friend & Amr is my enemy (Zayd sadqun l wa Amr
aduwwun l)
All my friends are more liked by me than my enemies (kullu
asdiq ah abbu ilayya min ad)
Zayd is more liked by me than Amr (Zayd ah abbu ilayya min
Amr)
Maks d did not, in such cases, quantify the semantic dependent. In
both examples, the relational major premise is naturally understood to
mean that Every father is older than all his sons and that All my
friends are more liked by me than all my enemies. On this account,
the inferences are clearly valid. However, as noted above in the discus-
sion of Kilis, in some cases the relational major is not naturally under-
stood in this sense, as for example The slave of a woman is the slave of
a human or The contradictory of the universal-affirmative proposi-
tion is the negative-particular proposition. In such cases the corre-
sponding inferences are not valid. The lack of explicit quantification of
the semantic dependent may indicate that Maks d was not fully aware
of this problem. However, his discussion is too brief to warrant any
confident interpretation of this point. What is clear is that Maks ds
treatment of relational inferences is fully in the tradition of late
Ottoman logic, though with some minor (but significant) modifica-
tions. His work shows that the Arabic tradition of logic was still devel-
oping even as late as the beginning of the twentieth century.
IV. Conclusion
52
Al Sedt, Mzn l-ukl (Istanbul: Karabet ve Kasbar Matbaas, 1303/1887).
CONCLUSION
For much of the twentieth century, it was assumed that the course of
Arabic-Islamic philosophy came to an end with Averroes (d. 1198).
This demise of the philosophical tradition was often presented as a
consequence of the hostility to philosophy exhibited by influential
Sunni Islamic scholars such as Ghazl (d. 1111) and Fakhr al-Din
al-Rz (d. 1210). It was only or so it has often been claimed in
Shi Iranian circles that philosophy continued to be cultivated, reach-
ing a silver age in the seventeenth century with Mull Sadr Shrz
(d. 1635).1 It is beginning to emerge that this grand narrative is woe-
fully inaccurate. It may be that the word falsafa came to have negative
overtones in much of the Islamic world, but philosophy in the mod-
ern sense of the word continued to be studied intensely throughout
the subsequent centuries. One need only glance at the table of con-
tents of widely used handbooks of Sunn theology such as Tawli
al-anwr of al-Baydw (d. 1316), al-Mawqif of al-j (d. 1355), and
al-Maqsid of al-Taftzn (d. 1390) to see that philosophy was still
studied throughout the Islamic world in later periods. These hand-
books and their standard commentaries dealt at length with topics like
epistemology, the soul and its relation to the body, the Aristotelian
categories, modality, the nature of space and time, and the relation
between essence and existence.2 Logic too was studied intensively
throughout the later centuries, not only in Sh Iran but across the
Sunn world, from Morocco and Bosnia to India.
Nicholas Rescher deserves credit for being one of the first to point
out that the century after the death of Averroes was a period of excep-
tional richness in the history of Arabic logic. However, he himself pre-
sented an alternative narrative of catastrophic decline, setting in at
around 1300 and leading to the complete atrophy of the Arabic logical
1
For a recent statement of this view, see Majid Fakhry, A Short Introduction to
Islamic Philosophy, Theology and Mysticism (Oxford: OneWorld, 1997), 114115.
2
See for example the table of contents of volume I of the English translation of
Baydws Tawli al-anwr and its commentary by Shams al-Dn al-Isfahn (d. 1348)
by E.E. Calverley & J.W. Pollock, Nature, Man and God in Medieval Islam (Brill:
Leiden, 2002).
256 conclusion
3
Rescher, The Development of Arabic Logic, 236.
258 conclusion
4
Merrill, Augustus De Morgan, 7980.
conclusion 259
By itself (bi-dhtihi): This phrase would seem to play the same role in the
definition of syllogism as Aristotles (An. Pr. 24b20), trans-
lated into English as from their being so (Jenkinson) or from the fact that
the assumptions are such (Tredennick) or because these things are so
(Smith). In the Arabic translation of Theodorus, an associate of the celebrated
translator H unayn b. Ishq (d. 877), Aristotles is translated as
li-wujdi tilka l-ashyi l-mawdati bi-dhtih in the definition of the syllo-
gism, but then merely as bi-dhtih in Aristotles immediately following expli-
cation of what he means by (see A. al-Badaw (ed.), Mantiq
Arist (Kuwait & Beirut: Wiklat al-matbt & Dr al-qalam, 1980 [reprint])
vol. 1: 142143). Since the Arabic Aristotelians (following Aristotle) explicitly
meant the condition to rule out elliptical arguments with suppressed
premises, the translation of the Arabic bi-dhtih as by themselves seems to
me to be in order. Note that rendering the term into by their essence (as
done in E.E. Calverley & J.W. Pollock (trans.), Nature, Man and God in
Medieval Islam [Brill: Leiden, 2002], I, 8890) would suggest falsely that the
subsequent mention of l bi-l-aradi in definitions of syllogism is a clarifica-
tory reiteration. The Arabic Aristotelians did not consider it as such, but as
an independent condition. It is also far from clear what the essence of two
premises would be.
Conclusion (natja)
Extreme (taraf): The two terms in a syllogism that are not the middle-term
and that are brought together to form the conclusion.
Form (sra): The form of a syllogism is what remains after the uniform sub-
stitution of the terms (h udd) of the syllogism with schematic letters. Thus
the argument Every human is a mammal & Every mammal is an animal, so
Every human is an animal has the form Every J is B & Every B is A, so Every
J is A. Sometimes used interchangeably with structure of the composition
(hayat al-talf). Its opposing concept is that of matter (mdda).
264 glossary
Judgment (h ukm)
Major term (al-h add al-akbar): the predicate of the conclusion of a syllogism.
Middle term (al-h add al-awsat): the term that repeats in both premises of a
classic Aristotelian syllogism and that is left out of the conclusion.
glossary 265
Minor term (al-h add al-asghar): the subject of the conclusion of a syllogism.
Premise (muqaddima)
Proposition (qadiyya)
Proof (dall): In logic, the term dall could be used in two related senses: (i) in
the narrow sense of a step-by-step derivation of a proposition from other
propositions in this sense it was used interchangeably with the term bayn;
(ii) a generic term covering syllogism (qiys), induction (istiqr) and analogy
(tamthl) in this sense it was used interchangeably with argument (h ujja).
Relation (nisba): The term nisba was standardly applied to the relation
between subject and predicate in a categorical proposition. As such the nisba
was considered to be the third element in any categorical proposition, besides
the subject and the predicate. It was held to be indicated linguistically by a
copula (rbita) in some languages (Greek and Persian) and implicitly under-
stood in others (Arabic). In this sense, all categorical propositions express a
nisba. The term nisba was also used of a relational property such as owner of
money or brother of a writer or equal to B. This latter meaning is presup-
posed, for example, in Samarqands using the term qiys al-nisba of rela-
tional syllogisms, and his explication of the validity of such syllogisms by
invoking the principle that al-muntasibu il muntasibin kharin bi-nisbatin
muntasibun il tilka l-nisba. This implies that in a proposition such as A is
equal to B there are two relations: (1) the subject-predicate relation indi-
cated by the copula, and (2) the relation indicated by the relational predicate.
This analysis is also clear when Khnaji, for example, wrote that in a proposi-
tion with predicates such as owner of money (dh ml), a nisba is the predi-
cate. This again implies that in such propositions there is (i) a subject, (ii) a
subject-predicate relation, and (iii) a relational predicate. Both the second
and third element would be referred to as nisba. The term idfa tended to be
used of the Aristotelian category of relation.
Statement (qawl)
Subject (mawd)
Non-Modern Sources
Abd l-Nfi Iffet Efendi (d. 1890). Fenn-i mantik: Mzn-i erh -i Mtercem-i Burhn.
Istanbul: Matbaa-yi mire, 1295/1878.
Ab l-Barakt al-Baghdd, Hibat Allh b. Malka (d. 1165). al-Mutabar f l-h ikma.
Hyderabad, Deccan: Dirat al-Marif al-Uthmniyya, 1357/1938.
Ab l-Salt al-Dn (d. 1134). Taqwm al-dhihn. Ed. C. A. Gonzales-Palencia. Madrid:
Centro de Estudios Historicos, 1915.
Acem, Al (d. 1456). H shiya al Sharh al-Shamsiyya. Manuscript Yahuda 5609.
Firestone Library, Princeton University.
Adaw al-Sad, Al (d. 1775), H shiya al Sharh al-Shamsiyya. Printed with Qutb
al-Dn al-Rz, Tahrr al-qawid al-mantiqiyya bi-sharh al-risla al-Shamsiyya, and
the glosses of Jurjn and Siylkt. Cairo: al-Matbaa al-Amriyya. 1323/1905.
Ahmadnagr, Abd al-Nab (fl. 1759). Dustr al-ulam. Hyderabad, Deccan: Matbaat
Dirat al-Marif al-Uthmniyya, 1329/1911.
. H shiya al Sharh al-Tahdhb li-l-Yazd. Lithograph. Lucknow: Sultn al-Matbi,
1268/1852.
Akhdar, Abd al-Rahmn (d. 1546). Sharh al-Sullam al-murawnaq. Printed as an
appendix to Ahmad al-Damanhr, dh al-mubham min man l-Sullam. Cairo:
Musta f al-Bb al-H alab, 1948.
Akhisr, Mehmed Sad (d. 1902). Sharh al-Risla al-Qiysiyya. Manuscript 45 Ak Ze
5983. www.yazmalar.gov.tr.
Akehirl, H asan Fehm (d. 1880). al-Qasda al-Azziyya wa Sharh uh. Istanbul:
Matbaat al-H jj Muharram al-Bsnav, 1292/1875.
lehr, Osmn (d. 1776). Risla f l-mantiq. Manuscript 60 Zile 170. www.yazmalar
.gov.tr.
Allmek Bsnav, Mehmed (d. 1636). Sharh al-Shamsiyya. Manuscript 19 Hk 2538.
www.yazmalar.gov.tr.
Ams, c Kerm (d. 1885), Mzn al-adl. Istanbul: Ceride-yi Havadis Matbaasi,
1276/18591860.
midi, Abd al-Wahhb (d. 1776). Sharh al-Waladiyya. Cairo: al-Matbaa al-Azhariyya,
1331/19121913.
mid, Sayf al-Dn (d. 1231). Daqiq al-h aqiq. Manuscript Garrett 42B. Firestone
Library, Princeton University.
. al-Nr al-bhir f l-h ikam al-zawhir. Facsimile prepared by F. Sezgin. Frankfurt:
Publications for the History of Arabic-Islamic Science, 2001.
Ansr, Zakariyy (d. 1519). Sharh sghj. Printed in the margins of Ysuf al-Hafns
H shiya ala Sharh Zakariyy l-Ansr ala sghj. Cairo: Matbaat al-Taqaddum
al-Ilmiyya, 1326/19081909.
Araj, Muhammad Rid (fl. 1716). al-Urjza f l-mantiq. In Collected Papers on Logic
and Language, ed. Mehdi Mohaghegh and Toshihiko Izutsu, 321355. Tehran:
Institute of Islamic Studies, 1974.
Astarabd, Abd al-H ayy (fl. 1508). Sharh Tahdhb al-mantiq. Manuscript Delhi Arabic
1474. British Library.
Att r, H asan (d. 1834), H shiya al Sharh al-Khabs.
Printed with Khabs, al-Tadhhb bi-sharh al-Tahdhb, and the gloss of Ibn Sad
al-H ajar. Cairo: al-Matbaa al-Azhariyya al-Misriyya, 1318/19001901.
Printed with Khabs, al-Tadhhb bi-sharh al-Tahdhb, and the gloss of Muhammad
b. Arafa al-Dasq. Cairo: Musta f al-Bb al-H alab, 1936.
270 bibliography
Ayntb, Necb Abd-Allh (d. 1804). Sharh al-Aris. Manuscript Haci Mahmud
Efendi 5733. Sleymaniye Ktphanesi, Istanbul.
Ayntb, Abd-Allh Edb (fl. 1911). Usl- akyise takrri. Istanbul: Saday- Millet
Matbaas, 1327/1911.
Baghdd, Ghnim (d. 1620). al-Wast Sharh al-Tahdhb. Manuscript Laleli 2639.
Sleymaniye Ktphanesi, Istanbul.
Bahmanyr b. Marzubn (d. 1065). al-Tah s l. Ed. Murtada Mutahhar. Tehran:
Dnishkdah-i Ilhyt va Marif-i Islm, 1349/19701971.
Bahr al-Ulm al-Lakh nav (d. 1810). Sharh Sullam al-ulm. Manuscript Yahuda
2678. Firestone Library, Princeton University.
Bjr, Ibrhm (d. 1860). H shiya al l-Sullm al-murawnaq. Cairo: Musta f al-Bb
al-H alab, 1347/1928.
. H shiya al Sharh Mukhtasar al-Sans. Cairo: Matbaat al-Taqqadum al-
Ilmiyya, 1321/1904.
Bannn, Muhammad b. al-H asan (d. 1780), Sharh al-Sullam. Lithograph. Fez: al-
Matbaa al-H ajariyya, n.d.
Baydw, Qd Abd-Allh (d. 1316). Tawli al-anwr min matli al-anzr. Ed. Abbs
Sulaymn. Cairo & Beirut: al-Maktabah al-Azhariyyah li-l-Turth & Dr al-Jl,
1991).
Byezd, Al (fl. 1863). Sharh kawshif qin al-aqyisa. Istanbul: n.p., 1863.
Birgev, Ahmed kir (fl.1839). H shiya ala Sharh al-Qiys al-Msawi. Printed as an
appendix to Mehmed T vskrs Sharh al-Risla al-Qiysiyya. Istanbul: Matbaa-i
Amire, 1281/18641865.
Bukhr, Mahmd b. Nimat-Allh (fl. early 16th century). al-Mughlatt. Manuscript
Delhi Arabic 1068/c. British Library.
Bulghr, Burhn al-Dn b. Kaml al-Dn b. H amd (fl. ca. 1460). H shiya ala l-Fanr.
Istanbul: n.p. 1290/18731874.
Brsal, Mehmed T hir (d. 1926). Osmanl Mellifler. Istanbul: Matbaa-yi Amire,
1333/19141915.
Brsev, Ahmed Sidk (d. 1895). Dharat al-imtih n. Istanbul: Matbaat Ahmed
Kemal, 1325/1907.
. Mzn al-Intizm. Istanbul: Matbaa-yi mire, 1306/18881889.
erkezde, Mehmed Tevfk (d. 1901). yet l-beyn f ilm il-mzn. Istanbul:
Mekteb-i Sanayi Matbaas, 1306/18881889.
Cevdet, Ahmed (d. 1895). Miyr-i Sedd. Istanbul: Karabet ve Kasbar Matbaas,
1303/1886.
Damanhr, Ahmad (d. 1778). dh al-mubham min man l-Sullam. Cairo: Musta f
al-Bb al-H alab, 1948.
Drendev, Mehmed (d. 1739). H shiya al Sharh al-H usayniyya. Manuscript Yahuda
5412. Fols. 4389. Firestone Library, Princeton University.
. Sharh al-Risla al-H usayniyya. Manuscript Yahuda 5412. Fols. 143. Firestone
Library, Princeton University.
Dashtak, Ghiyth al-Dn (d. 1542). H shiyat al-Ishrt. In Musannaft Giyth Mansr
Dashtak, ed. A. Nourani. Tehran: University of Tehran, 2007. Vol. II: 519590.
. Miyr al-irfn. In Musannaft Giyth Mansr Dashtak, ed. A. Nourani. Tehran:
University of Tehran, 2007. Vol. II: 9891068.
. Tadl al-mzn.
Manuscript 9698. Ayatollah Marash Najaf Library, Qom.
Manuscript 23594. Astn-i Quds-i Radaw Library, Mashhad.
Dashtak, Mr Sadr al-Dn (d. 1498). H shiya al Sharh al-Shamsiyya. Manuscript
Reid Efendi 1015. Sleymaniye, Istanbul.
. H shiya al Sharh al-Tajrd.
Manuscript Bijapur 172. British Library.
Manuscript Bijapur 298. British Library.
bibliography 271
Dasq, Muhammad b. Arafa (d. 1815). H shiya al Sharh al-Khabs. Printed with
Khabs, al-Tadhhb bi-sharh al-Tahdhb, and the gloss of H asan al-Att r. Cairo:
Musta f al-Bb al-H alab, 1936.
Dawn, Jall al-Dn (d. 1502). al-H shiya al-ajadd ala Sharh al-Tajrd. Manuscript
Veliyddin Efendi 2026. Veliyddin, Istanbul.
. H shiya al H shiyat Sharh al-Matli. Manuscript New Series 1948. Firestone
Library, Princeton University Library.
. H shiya al H shiyat Sharh al-Shamsiyya. Lithograph. Delhi: al-Matba
al-Mujtab, n.d.
. H shiya jadda al H shiyat Sharh al-Matli. Manuscript 01 Hk 111. www
.yazmalar.gov.tr
. Sharh Tahdhb al-mantiq. Istanbul: al-H jj Muharram al-Bsnav, 1305/1887.
Dihlav, Raf al-Dn b. Wal-Allh (d. 1818), Takml sinat al-adhhn. Manuscript
Delhi Arabic 1513. British Library.
Edirne Mftis, Mehmed Fevz (d. 1890). Khulsat al-mzn [=Sharh Sharh sghj
li-l-Fanr]. Istanbul: al-H jj Muharram, 1284/1867.
Emle, H asan b. H seyn (fl. 1548). H all al-Tahdhb. Manuscript Laleli 2644.
Sleymaniye Ktphanesi, Istanbul.
. Takml al-mantiq. Autograph Manuscript Laleli 2561. Suleymaniye Ktphanesi,
Istanbul.
Eskiczde Edirnev, Al b. H usayn (d. 1827). Sharh sghj. Edirne: n.p., 1287/
1870.
Frb, Ab Nasr Muhammad (d. 950). Kitb al-qiys. In al-Mantiq inda l-Frb, ed.
R. al-Ajam, II, 1164. Beirut: Dr al-Mashriq, 1985.
Fenr, Mehmed (d. 1431). Fusl al-badi f usl al-shari. Istanbul: Matbaat al-
Shaykh Yahy, 1289/1872.
. Sharh sghj. Istanbul: Istanbul: al-Matbaa al-H arbiyya, 1278/1861.
Fihris al-kutub al-mah fza bi-l-Kutubkhna al-Khedviyya al-Misriyya. Cairo: n.p.,
1305/18871888.
Fursat Shrz, Muhammad Nsir (d. 1920). Ashkl al-mzn. Lithograph. 1307/1889
1890. Reprint. Tehran: Kitbfursh-yi Islmyeh, 1340/19611962.
Gelenbev, Isml (d. 1791). al-Burhn fi ilm al-mzn. Cairo: Matbaat al-Sadah,
1347/19281929.
. H shiya al H ashiyat al-Lr al Sharh Hidyat al-h ikma. Istanbul: Matbaa-yi
mire, 1270/18531854.
. H shiya al Mr al-db. Istanbul: Matbaa-yi mire,1234/18181819.
. H shiya al Mr al-Tahdhb. Istanbul: n.p., 1288/1871.
. H shiya al Sharh al-Dawn al l-Aqid al-Adudiyya. Istanbul: irket-i
Hayriye-yi Sahfiye-ye 1307/1890.
. Rislat al-imkn. Istanbul: n.p., 1309/1892.
. Sharh sghj. Istanbul: irket-i Sahafiye- Osmniye Matbaas, 1307/1889.
Ghazl, Ab H mid (d. 1111). Mih akk al-nazar. Cairo: al-Matbaa al-Adabiyya, n.d.
. Miyr al-ilm. Cairo: Matbaat Kurdistan al-ilmiyya, 1329/1911.
. al-Mustasf min ilm al-usl. Cairo: al-Matbaa al-Tijriyya al-Kubr, 1937.
Gln, Abd-Allh (fl. ca. 1600). al-Risla al-muh ta bi-tashkkt f l-qawid al-
mantiqiyya maa tah qqtih. In Collected Papers on Logic and Language, ed. Mehdi
Mohaghegh and Toshihiko Izutsu, 359395. Tehran: Institute of Islamic Studies,
1974.
Gpamv, Qd Mubrak (d. 1747). Sharh Sullam al-ulm. Kazan: al-Matbaa al-
Malakiyya, 1887.
dim, Eb Sad (d. 1762). Aris anzr al-abkr. Manuscript Yahuda 5310. Firestone
Library, Princeton University Library.
. Tajrd fawid Aris al-anzr. Manuscript New Series 361. Firestone Library,
Princeton University Library.
272 bibliography
H afd al-Taftzn, Sayf al-Dn Ahmad b. Yahy al-Haraw (d. 1511). al-Durr al-nadd
li-majmat ibn al-H afd. Beirut: Dr al-Kitb al-Arab, n.d.
. Sharh Tahdhb al-mantiq. Manuscript Delhi Arabic 1553b. British Library.
H afd l-Nsr (d. 1774). al-Fawid f silk al-farid. Manuscript Yahuda 3179.
Firestone Library, Princeton University.
. H shiya al l-H shiya al-Mriyya [al Tahdhb al-mantiq]. Manuscript Yahuda
1084. Firestone Library, Princeton University.
H afn, Ysuf (d. 1764). H shiya ala Sharh Zakariyy l-Ansr ala sghj. Cairo:
Matbaat al-Taqaddum al-Ilmiyya, 1326/19081909.
Hamadn, Abd al-Hd (d. 1913). Luluat al-mzn maa sharh ihi Muntaq l-jummn.
Lithograph. n.p.: n.p., 1323/19051906.
arpt, Ysuf ukr (d. 1875). Nms al-qn Sharh al-Burhn. Istanbul: Litografya
Matbaas, 1274/18571858.
H ayt Mara, Ahmed (d. 1813). H shiya ala Sharh al-Qiys al-Msaw. Printed as
an appendix to Mehmed T vskrs Sharh al-Risla al-Qiysiyya. Istanbul:
Matbaa-yi mire, 1281/18641865.
Haydaj, Muhammad Al (d. 1927). Talqa al al-Manzma w-sharh ih. Lithograph.
Tehran: n.p., 1346/1928.
Herav, Mr Zhid (d. 1689). H shiya al l-Jall al l-Tahdhb. Lithograph. Lucknow:
Matba-i Najm al-Ulm, n.d.
Hill, Ahmad b. Abd al-Azz (d. 1761). al-Zawhir al-ufuqiyya bi-sharh al-Jawhir
al-mantiqiyya. Lithograph. Fez: n.p., 1313/1895.
H ill, Ibn Mutahhar (d. 1325). al-Asrr al-khafiyya f l-ulm al-aqliyya. Qom: Markaz
al-Abhth wa l-Dirst al-Islmyah. 1421/20002001.
. al-Jawhar al-nadd f sharh mantiq al-Tajrd.
Ed. Muhsin Bidarfar. Qum: Intisharat Bidar, 1423/20022003.
Lithograph. Iran: n.p., 1311/18931894.
. al-Qawid al-jaliyya f sharh al-Risla al-shamsiyya. Ed. F. H . Tabrzin. Qum:
Muassasat al-Nashr, 1412/1991.
H usayn Shh (fl. 16th or 17th century). Sharh al-H usayniyya. Lithograph. Istanbul:
n.p., 1267/18501851.
H uwayz, Ishq (fl. 17th century). H shiya al H shiyat Mull Abd-Allh. Manuscript
Add. 7461. British Library.
Ibn Ab al-H add al-Madin (d. 1258). Sharh al-yt al-bayyint. Ed. M. Djebli.
Beirut: Dar Sader, 1996.
Ibn al-Bad al-Bandah (d. 1258). Sharh al-Muqaddima al-Kashshiyya. Manuscript
Arabic 4931. Chester Beatty Library, Dublin.
. Sharh al-Mjaz. Manuscript 45 Hk 2212. www.yazmalar.gov.tr.
Ibn Bjja [Avempace] (d. 1139). Talq Ibn Bja al mantiq al-Frb. Ed. M. Fakhry.
Beirut: Dar al-Mashriq, 1997.
Ibn Ghulm Muhammad (fl. 1802). H shiya-yi jadda-yi Mr sghj. Lithograph.
Lahore: Matbaat Shihb Thqib, n.d.
Ibn H azm (d. 1037). al-Taqrb li-h add al-mantiq. Ed. I. Abbs. Beirut: Dr Maktabat
al-H ayt, 1959.
Ibn Kammna (fl. 1268). Sharh al-Talwh t. Ed. Najafqoli Habibi. Tehran: Miras-i
Maktub, 2009.
. Al-Kshif: al-Jadd f l-h ikma. Ed. H . N. Isfahn. Berlin & Tehran: Institute for
Islamic Studies & Iranian Institute of Islamic Philosophy, 2008.
Ibn Khaldn (d. 1406), The Muqaddimah: An Introduction to History. Trans. F.
Rosenthal. New York: Pantheon Books, 1958.
. al-Tarf bi Ibn Khaldun wa rih latihi gharban wa sharqan. Ed. M. al-Tanj. Cairo:
n.p., 1951.
Ibn Krn, al-Tayyib (d. 1812), Sharh al-Kharda f l-mantiq. Lithograph. Fez: n.p.
1329/1911.
bibliography 273
Ibn Marzq al-H afd (d. 1439). Nihyat al-amal bi-sharh kitb al-Jumal. Manuscript
Derenbourg 640. Escorial, Madrid.
Ibn Mubrak Shh (fl. 1360s). Sharh al-Shamsiyya. Manuscript Or. 6318. British
Library.
Ibn al-Nafs (d. 1288), Sharh al-Wurayqt f l-mantiq. Ed. Talbi, Zidani & Melit. Tunis:
Dr al-Gharb al-Islm, 2009.
Ibn al-Qaradgh (d. 1936), H shiya al l-Burhn. Printed with Gelenbev, al-Burhn
fi ilm al-mzn. Cairo: Matbaat al-Sadah, 1347/19281929.
Ibn Rushd al-H afd [Averroes] (d. 1198). Talkhs kitb al-qiys. Ed. Mahmd Qsim,
Charles E. Butterworth, and Ahmad Abd al-Majid Hard. Cairo: al-Haya
al-Misriyya al-mma li-l-Kitb, 1983.
. Rasil falsafiyya: Maqlt f l-mantiq wa l-ilm al-tab li-Ibn Rushd. Ed.
J. Alaw. Casablanca: n.p., 1983.
Ibn Sad al-H ajar (d. 1785). H shiya al Sharh al-Khabs. Printed with Khabs,
al-Tadhhb bi-sharh al-Tahdhb, and the gloss of H asan al-Att r. Cairo: al-Matbaa
al-Azhariyya al-Misriyya, 1318/19001901.
Ibn Sn [Avicenna] (d. 1037). al-Ishrt wal-tanbht.
Ed. J. Forget. Leiden: E. J. Brill, 1892.
Ed. M. Zri. Tehran: Bustan-e Ketab-e Qom Press, 1423/2002.
. al-Shif: al-Qiys. Ed. S. Zyid and I. Madkour. Cairo: n.p., 1964.
Ibn Turka al-Isfahn (d. 1437). al-Manhij f l-mantiq. Ed. I. Dbj. Tehran: Institute
of Islamic Studies, 1997.
Ibn Wsil al-H amaw (d. 1298). Mufarrij al-kurb f akhbr Ban Ayyb. Ed. Rab and
shr. Cairo: Dr al-Kutub, 1977.
. Nukhbat al-fikar. Manuscript Landberg 103. Beinecke Library, Yale University.
. Sharh al-Jumal. Manuscript Landberg 104. Beinecke Library, Yale University.
Ibn Yaqb al-Walll, Ah mad (d. 1716). Lawmi al-nazar bi-tah qq man l-
Mukhtasar. Manuscript Nr. 2307. Khuda Bakhsh Oriental Library, Bankipore, India.
. al-Qawl al-musallam f tah qq man l-Sullam. Manuscript Add. 9622. British
Library.
. Sharh al-Lmiyya f l-mantiq. Manuscript Mantiq 2576/96175. Azhar Library,
Cairo.
Ibn Zura (d. 1008). Kitb al-qiys. In Mantiq Ibn Zura, ed. J. Jhm and R. al-Ajam,
91200. Beirut: Dr al-Fikr al-Lubnn, 1994.
j, Adud al-Dn (d. 1355). Sharh Mukhtasar Ibn al-H jib. Blq: al-Matbaa
al-Amriyya, 13161317/18981900.
Ilh Ardabl, H usayn (d. 1543). H shiya al Sharh Tajrd al-kalm. Manuscript New
Series 1111. Firestone Library, Princeton University.
Isfahn, Bah al-Dn Fzil-i Hind (d. 1725). H ikmat-i Khqniyya. Ed. G. I. Dinn.
Tehran: Ayene-ye Miras, 1998.
Isfahn, Shams al-Dn Mahmd (d. 1348). Bayn al-Mukhtasar. Ed. M. M. Baq.
Mecca: Dr al-Madan, 1986.
. Sharh Matli al-anwr. Manuscript Yahuda 187. Firestone Library, Princeton
University.
. Sharh Tawli al-anwr. Lithograph. Istanbul: Khuls Efendi Matbaas,
1305/18871888.
Isfarn, Ism al-Dn (d. 1537). H shiya al l-tasdqt. Istanbul: Matbaa-yi mire,
1259/18431844.
[Izmr, Abd l-Rahmn (d. 1779)]. Risla f tah qq aqsm al-qiys. Manuscript Haci
Mahmud Efendi 5821. Sleymaniye, Istanbul.
Jabalrd, Khidir (fl.1421). Sharh al-Ghurra f l-mantiq. Ed. Albert N. Nader. Beirut:
Dr al-Mashriq, 1983.
Jall al-Sann, al-H asan (d. 1673). Sharh Tahdhb al-mantiq. Beirut: Dr al-Masrah,
n.d.
274 bibliography
Jm, Abd al-Rahmn (d. 1492). Sharh al-Kfiya. Istanbul: irket-i Sahfiye-yi,
1324/1906.
Jarb, Sulaymn (fl. 1507). Sharh sghj. Tunis: al-Matbaa al-Rasmiyya, 1309/1891.
Jawnpr, Muhammad Al (fl. 18th century). Mirj al-fuhm bi-sharh Sullam al-ulm.
Manuscript India Office Islamic 1728. British Library.
Jazir, Sad Qaddra (d. 1656). Talqa al Sharh al-Sullam al-murawnaq. Manuscript
Add. 9620. British Library.
Jazir, Muhammad Nr al-Dn (fl. 1875). Kashf al-asrr al-makhfiyya f dumn
al-abyt al-ramziyya. Beirut: Matbaat Jardat Bayrt, 1307/1889.
Jurjn, al-Sayyid al-Sharf (d. 1413). H shiya al Sharh Matli al-anwr. Printed as
an appendix to Qutb al-Dn al-Rz, Lawmi al-asrr bi-Sharh Matli al-anwr.
Istanbul: Matbaa-i Amire, 1277/18601861.
. H shiya al Sharh Mukhtasar Ibn al-H jib. Printed with js Sharh Mukhtasar
Ibn al-H jib. Blq: al-Matbaa al-Amriyya, 13161317/18981900.
. H shiya ala Sharh al-Shamsiyya.
Printed with Qutb al-Dn al-Rzs Tah rr al-qawid al-mantiqiyya bi-sharh
al-Risla al-Shamsiyya. Cairo: Mustaf al-Bb al-H alab, 1948.
Printed with Qutb al-Dn al-Rz, Tah rr al-qawid al-mantiqiyya bi-sharh
al-Risla al-Shamsiyya. Cairo: al-Matbaa al-Amriyya. 1323/1905.
. D mantiqi-i Frs: Kubr, Sughr. Ed. Murtaza Mudarris Chahrdah. Tehran:
Kitbkhneh-i T hr, 1334/1956.
Kfiyaj, Muhy al-Dn (d. 1479). al-Targhb sharh al-Tahdhb. Manuscript India Office
1052. British Library.
K alack, mer [possibly identical to mer Feyz Tkd, below]. H shiya al l-Risla
al-Msawiyya.
Manuscript Fatih 3304. Sleymaniye Ktphanesi, Istanbul.
Manuscript 06MilYzA9363. www.yazmalar.gov.tr.
Knkir, Abd-Allh (d. 1828). Nafis aris al-anzr wa latif fawid al-afkr
[=H shiya al Sharh al-Fanr]. Istanbul: Matbaa-yi Amire, 1313/1896.
K araga, Ahmed Rd (d. 1835). Tuh fat al-Rushd l-Qaraghaj f sharh sghj.
Istanbul: Matbaa-yi Amire, 1279/1862.
K araca Ahmed (d. 1450). Sharh sghj. Manuscript Yahuda 4539. Firestone Library,
Princeton University.
K ara all Trev (d. 1711). H shiya al l-Fanr. Istanbul: Yahy Efend,
1289/18731874.
. H shiya ala Sharh al-Tahdhb. Manuscript 06 Hk 2248. www.yazmalar.gov.tr.
K rs, Dvd (fl. 1740s). Sharh al-sghj l-jadd. Manuscript Garret 472H. Firestone
Library, Princeton University.
. Sharh Takmilat al-Tahdhb. Manuscript Yahuda 5144. Firestone Library,
Princeton University.
Ksh, Imd al-Dn Yahy (fl.1453). Sharh sghj. Manuscript 14 Hk 33/1. www.yaz-
malar.gov.tr.
Kt, H usm al-Dn (d. 1359). Qla Aqlu [=Sharh sghj]. Lithograph. Cawnpore:
Matba-i Nizm, 1279/1862.
Ktib eleb (d. 1657). Kashf al-zunn an asm l-kutub wa l-funn. Istanbul: Maarif
Matbaasi, 19411943.
Ktib, Najm al-Dn (d. 1277). Jmi al-daqiq fi kashf al-haqiq.
Manuscript Or. 11201. British Library.
Manuscript Majm Musta f Bsh 162. Dr al-Kutub al-Misriyya.
Kzim Isfahn (fl. 1924). Zubdat al-kalm f l-mantiq wa l-kalm. Baghdad: Dr
al-Salm, 1343/19241925.
Kefev, Ebul-Bek Eyyp (d. 1684). al-Kulliyyt. Beirut: Muassasat al-Risla, 1992.
Kefev, Mehmed b. H amd (d. 1754). H shiya al Sharh Mukhtasar Ibn al-H jib.
Manuscript Yahuda 5446. Firestone Library, Princeton University.
bibliography 275
Madan, Najb al-Dn (d. 1722). al-Farid al-H asaniyya bi-h all al-fawid al-Abhari-
yya. Manuscript Yahuda 3849. Firestone Library, Princeton University.
Madhr al-H alab, Ibrhm (d. 1776). Sharh al-Manzma al-qiysiyya. Manuscript
Yahuda 4458. Firestone Library, Princeton University.
Magnisv, Ahmed b. Sleymn (fl. 1869) Sharh al-Istidlliyya. Lithograph. Istanbul:
Matbaat al-Hajj Muharram al-Busnawi, 1287/18701871.
Magnisv, Mahmd (d. 1808). Mughn l-Tullb. Istanbul: irket-i Sahfiye-yi
Osmniye, 1319/1901.
Mallaw, Ahmad (d. 1767). Sharh al-Sullam al-murawnaq. Printed on the margins of
Muhammad al-Sabbns H shiya ala Sharh al-Mallaw ala al-Sullam. Cairo:
Musta f al-Bb al-H alab, 1938.
Maks d, Ahmed Hd (d. 1941). Mzn al-afkr. Kazan: B. L. Dombrovskago,
1903.
Mangalr, Al al-Dn (fl. 17th century). Zubdat sharh al-Shmma. Manuscript India
Office Islamic 545. British Library.
Maybud, Qd Mr H usayn (d. 1504). Sharh al-Shamsiyya. Istanbul: n.p. 1289/1872.
Mr Ab l-Fath H usayn (fl. 1564). H shiya al Sharh Tahdhb al-mantiq. [Printed
with Dawns Sharh Tahdhb al-mantiq.] Istanbul: al-H jj Muharram al-Bsnav,
1305/1887.
. Mifth al-Bb. Printed in al-Bb al-h d ashar li-l-Allma al-H ill maa
sharh ayhi. Ed. M. Mohaghegh. McGill & Tehran: Institute of Islamic Studies &
Tehran University, 1986.
. Takmilat Sharh al-Tahdhb. Manuscript New Series 272. Firestone Library,
Princeton University.
Mrz Jn Bghnav, H abb-Allh (d. 1586). H shiya al Sharh Matli al-anwr.
Manuscript Yahuda 3451. Firestone Library, Princeton University.
. H shiya al Sharh al-Tajrd.
Manuscript Add. 7462. British Library.
Manuscript Yahuda 5637. Firestone Library, Princeton University.
. al-Muh kamt maa H shiyat Mrz Jn. Istanbul: Matbaa-yi mire, 1290/
18731874.
Mstr, Musta f (d. 1707). Sharh sghj. Istanbul: Matbaa-yi mire, 1316/1898.
. Sharh al-Shamsiyya. Manuscript Yahuda 1403. Firestone Library, Princeton
University.
Msl, H asan H sn (fl. 1878). Tanwr al-Burhn. Istanbul: Matbaat al-Sharika
al-Murattabiyya li-rtn sdryn, 1307/18891890.
Mftzde Erzincn, Mehmed Sdik (d. 1808). H shiya al l-H usayniyya. Istanbul:
irket-i Sahfiye-yi Osmniye, 1307/18891890.
. H shiya al Sharh al-Shamsiyya: al-Tasdqt. Istanbul: Matbaa-yi mire,
1254/1838.
Muhibb, Muhammad Amn (d. 1699). Khulsat al-athar f ayn al-qarn al-h d
ashar. Cairo: al-Matbaa al-Wahbiyya, 1284/18681869.
Mull Mubn Lakh nav (d. 1810). Mirt al-shurh .
Cairo: Matbaat al-Sadah, 13271328/19091910.
Lithograph. Lucknow: Nuval-i Kishr, 1299/1882.
Mull Sadr Shrz (d. 1635). Rislat al-tasawwur wa l-tasdq [Printed with Qutb
al-Dn al-Rz, Risla f l-tasawwur wa l-tasdq and the gloss of Mr Zhid Herav],
ed. Mahd Sharat. Beirut: Dr al-kutub al-ilmiyya, 2004.
. Talqt ala Sharh Hikmat al-ishrq. Lithograph. 1315/18971898. Reprint.
Qom: Intisharat Baydar, n.d.
. al-Tanqh f l-mantiq. In Majmah-i rasil-i falsaf-i Sadr al-Mutaallihn, ed.
H mid Nj Isfahn, 195234. Tehran: Hikmet Publishing, 1999.
Murd, Muhammad Khall (d. 1791). Silk al-durar f ayn al-qarn al-thn ashar.
Blq: Dr al-T iba al-Kubr, 1301/18831884.
bibliography 277
Nn, Bah al-Dn Muhammad Mukhtr (fl. 1698). Tadl al-mzn f talq ilm
al-mzn. Manuscript New Series 500. Firestone Library, Princeton University.
. Zawhir al-jawhir wa nawdir al-zawjir. Ed. by A. Rawdat. Isfahan: Matbaat
al-H abl al-Matn, 1379/19591960.
Narq, Muhammad Mahd (d. 1795). Mushkilt al-ulm. Tehran: Muassasah-i
Mutlat va Tahqqt-i Farhang, 1367/1988.
. Sharh al-Ilhiyyt min kitb al-Shif. Ed. M. Mohaghegh. McGill & Tehran:
Institute of Islamic Studies & Tehran University, 1986.
Nayrz, Mahmd (fl. 14971522). Sharh Tahdhb al-mantiq. Manuscript 1088. Astn-i
Quds-i Radaw Library, Mashhad.
Pirizrenl, H seyn Necm l-Dn (d. 1924). Sharh al-Burhn. Manuscript Yazma
Baislar 257. Sleymaniye Ktphanesi, Istanbul.
Qdir, Muhammad b. al-Tayyib (d. 1773). Nashr al-mathn li-ahl al-qarn al-h d
ashar wa l-thn. Ed. H ajj and Tawfq. Rabat: Maktabat al-T lib, 19771986.
Qass ra, Al (d. 1843), H shiya al Sharh al-Sullam li-l-Bannn. Lithographed with
Bannn, Sharh al-Sullam. Fez: al-Matbaa al-H ajariyya, n.d.
Qshj, Al al-Dn (d. 1479). Sharh Tajrd al-kalm. Lithograph. Tabriz: n.p.,
1301/1883.
Quwaysin, H asan (d. 1839), Sharh nazm al-Sullam f l-mantiq. Ed. Marhm.
Casablanca: Dr al-Rashd al-H adthah, 2006.
Rahn al-T itwn, Ahmad (fl. 1921). Jarayn al-qalam bi-sharh al-Sullam. T itwn:
al-Matbaa al-Mahdiyya, 1354/1935.
Rmpr, Muhammad Fazl H aqq (d. 1939), al-Kf li-h all sghj. Lithograph. Delhi:
al-Matba al-Mujatb, 1309/18911892.
Rid Efendi (d. 1917). Mzn l-makl. Istanbul: Mekteb-i Fnun- Harbiye-yi
ahane Matbaas, 1315/1899.
Rz, Fakhr al-Dn (d. 1210). Mantiq al-Mulakhkhas. Ed. A. F. Karamaleki and
A. Asgharinezhad. Tehran: ISU Press, 2003.
. al-Mulakhkhas f l-mantiq wa l-h ikma. Manuscript Or. 2360. British Library.
. Sharh al-Ishrt wa l-tanbht. Ed. A. Najafzada, Tehran: Amjuman-i sr va
Mafkhir-i Farhang, 2005.
. Sharh Uyn al-h ikma. Ed. A. al-Saq. Reprint. Tehran: Muassasat al-Sdiq,
1415/1994.
Rz, Qutb al-Dn (d. 1365). Lawmi al-asrr bi-Sharh Matli al-anwr. Istanbul:
Matbaa-yi mire, 1277/18601861.
. al-Muh kamt.[Printed with Ts, Sharh al-Ishrt.]
. Tah rr al-qawid al-mantiqiyya bi-sharh al-Risla al-Shamsiyya.
Cairo: Mustaf al-Bb al-H alab, 1948.
Cairo: al-Matbaa al-Amriyya. 1323/1905.
Ridaw, Khall (fl. ca. 1500). H shiya ala al-tasdqt. Manuscript Yahuda 4317.
Firestone Library, Princeton University.
Ridwn, Musta f (fl. 1871). Hidyat al-jinn f ilm al-mzn. Blq: n.p., 1872.
Sabbn, Muhammad (d. 1791). H shiya ala Sharh al-Mallaw ala al-Sullam. Cairo:
Musta f al-Bb al-H alab, 1938.
Sabzavr, Mull Hd (d. 1878). Sharh al-Manzma. Edited by M. T lib and anno-
tated by H asanzde mul. Qom: Nashr Nb, 1416/1995.
Saklzde, Mehmed (d. 1732). Taqrr al-qawnn al-mutadwala f ilm al-munzara.
Istanbul: n.p., 1289/18721873.
Sadr al-Shara al-Mahbb (d. 1346). Tadl al-ulm. Manuscript Landberg 394. Berlin
Staatsbibliothek.
Safaw, s (d. 1546). Sharh al-Ghurra f l-mantiq. Ed. Albert N. Nader. Beirut: Dr al-
Mashriq, 1983.
Sakkk, Sirj al-Dn (d. 1229). Mifth al-ulm. Cairo: al-Matbaa al-Adabiyya,
1317/18991900.
278 bibliography
Samarqand, Mull Muhammad Sdiq (fl. early 17th century). H shiya al Bad
al-mzn. Manuscript Delhi Arabic 1564/a. British Library.
Samarqand, Shams al-Dn (d. 1303). al-Anwr al-ilhiyya. Manuscript Murad Molla
1394. Sleymaniye Ktphanesi, Istanbul.
. Qists al-afkr. Manuscript 45 Hk 2213. www.yazmalar.gov.tr.
. Sharh al-Qists. Manuscript Landberg 1035. Berlin Staatsbibliothek.
Sammk, Fakhr al-Dn (d. 1576). Risla f ilm al-munzara. Manuscript 60 Hk 362.
www.yazmalar.gov.tr.
Sandl, H amd-Allh (d. 1747). Sharh Sullam al-ulm. Dehli: Matbaat al-Mujtab,
1328/19101911.
Sans, Muhammad b. Ysuf (d. 1490). Sharh al-Mukhtasar f l-mantiq. Cairo: n.p.,
1292/18751876.
. Sharh Mukhtasar Ibn Arafa. Manuscript Mantiq 18. Dr al-Kutub al-Misriyya.
. Sharh Muqaddimat sghj. Manuscript Yahuda 2802. Firestone Library,
Princeton University.
Sw, Umar b. Sahln (f1. 1145). al-Basir al-nasriyya f l-mantiq. Ed. M. Abdu.
Cairo: al-Matbaa al-Kubr al-Amriyya, 1316/1898.
Sedt, Al (d. 1900). Mzn l-ukl. Istanbul: Karabet ve Kasbar Matbaas, 1303/1886.
ehrzde, Mehmed Sad (d.1764). Misbh al-qalbiyya f dh al-mughlaqt
al-Athriyya Manuscript Yahuda 1345. Firestone Library, Princeton University.
evk, Ahmed (d. 1808). Sharh al-Fanr. Istanbul: Matbaa-yi Amire, 1284/1867.
Shh Mr Hibat-Allh H usayn (fl. 1546). Sharh Tahdhb al-mantiq. Manuscript
Garrett 124 L. Firestone Library, Princeton University.
Shahrastn, Jaml al-Dn (fl. 1584). Sharh -i Tahdhb. Lithograph. Cawnpore: n.p.,
1915.
Shahrazr, Shams al-Dn. (fl. 1288) al-Shajara al-ilhiyya f ulm al-h aqiq al-
rabbniyya. Ed. N. Grgn. Beirut and Istanbul: Dr Sdir & Dr al-Irshd, 2007.
. Sharh H ikmat al-ishrq. Ed. H. Ziai. Tehran: Institute for Humanities and
Cultural Studies, 2001.
Shrz, Abd al-Rahmn (fl. 1546). Sharh Tahdhb al-mantiq. Manuscript Yahuda
1410. Firestone Library, Princeton University.
Shrz, Qutb al-Dn (d. 1311). Sharh H ikmat al-ishrq. Ed. A. Nourani and
M. Mohaghegh. Tehran: Institute of Islamic Studies, 2002.
. Durrat al-tj li-ghurrat al-dbj. Ed. M. Mishkt. Tehran: Majlis, 19381945.
Shirwn, Masd (d. 1499). Sharh al-db al-Samarqandiyya. Manuscript Or. 3124.
British Library.
Shuwayh al-Najaf, Muhammad b. Ynus (fl. 1806). Mzn al-uql f kashf asrr
ghawmid h aqiq masil ilm al-maql. Manuscript Or. 7999. British Library.
Srz, Musta f Kutb (fl. early nineteenth century). Sharh al-Burhn. Istanbul: n.p.,
n.d.
Sirr, Sleymn (fl. 1892). Mantk. Istanbul: Karabet Matbaas, 1310/1892.
irvn, Mehmed Emn Sadruddnzade (d. 1627). H shiya ala h shiyat al-Khalkhl
ala Sharh al-Tahdhb. Manuscript 45 Ak Ze 676. www.yazmalar.gov.tr.
. Al-Fawid al-khqniyya. Manuscript 43 Va 2371. www.yazmalar.gov.tr.
Svs, Al Keml (fl. 1902). Mzn al-mantiq. Cairo: al-Matbaa al-H usayniyya,
1327/19091910.
Siverek, Musta f (fl. 1793). Sharh Ismat al-adhhn. Manuscript 06Hk3118. www
.yazmalar.gov.tr.
Siylkt, Abd al-H akm (d. 1657). H shiya al Sharh al-Shamsiyya. Printed with
Qutb al-Dn al-Rz, Tahrr al-qawid al-mantiqiyya bi-sharh al-Risla al-
Shamsiyya, and the glosses of Jurjn and Adaw. Cairo: al-Matbaa al-Amriyya,
1905.
. H shiya al l-tasdqt. Istanbul: irket-i Sahafiye- Osmniye Matbaas,
1320/19021903.
bibliography 279
Modern Sources
. A third sort of syllogism: Galen and the logic of relations. In Modern Thinkers
and Ancient Thinkers, ed. R. W. Sharples, 172194. Boulder: Westview Press,
1991.
. Truth, etc.: Six Lectures on Ancient Logic. Oxford: Clarendon Press, 2007.
Barnes, J.R. An Introduction to Religious Foundations in the Ottoman Empire. Leiden:
Brill, 1986.
Battal-Taymas, A. Iki Maksudiler: Sadri-Arsal, Ahmed-Hadi: kiilikleri, fikir hayatlar
ve eserleri. Istanbul : Sralar Matbaas, 1959.
Berkes, N. The Development of Secularism in Turkey. Montreal: McGill University
Press, 1964.
Bingl, A. Gelenbevinin Mantk Anlay. Istanbul: Milli Eetim Basmevi, 1993.
Bochenski, I. M. A History of Formal Logic. Ed. and trans. I. Thomas. New York:
Chelsea Publishing Company, 1970.
Brockelmann, C. Geschichte der arabischen Literatur. Leiden: E. J. Brill, 19371949.
Calverley, E. E. & Pollock, J. W. (trans.), Nature, Man and God in Medieval Islam. Brill:
Leiden, 2002.
Chambers, R. L. The Education of a Nineteenth-Century Ottoman Alim: Ahmed
Cevdet Paa. International Journal of Middle East Studies 4 (1973): 440464.
Copi, I. Symbolic Logic. 5th Edition. New York: Macmillan, 1979.
Daiber, H. Bibliography of Islamic Philosophy. Leiden: Brill, 1999.
Dallal, A. An Islamic Response to Greek Astronomy: Kitb Tadl hayat al-aflk of Sadr
al-Shara. Leiden: Brill, 1995.
Derenbourg, H. Les manuscrits arabes de lEscurial. Paris: E. Leroux, 1903.
Dummett, M. Frege: Philosophy of Language. Cambridge, MA: Harvard University
Press, 1981.
Dvorak, P. Relational Logic of Juan Caramuel. In Medieval and Renaissance Logic.
Vol. 2 of Handbook of the History of Logic, ed. D. M. Gabbay and J. Woods, 645665.
Amsterdam: Elsevier, 2004.
Fakhry, M. A Short Introduction to Islamic Philosophy, Theology and Mysticism.
Oxford: OneWorld, 1997.
Graf, G. Geschichte der christlichen arabischen Literatur. Vatican: Biblioteca Apostolica
Vaticana, 19441952.
Gutas, D. Aspects of Literary Form and Genre in Arabic logical works, in Glosses and
Commentaries on Aristoelian Logical Texts, ed. C. Burnett, 2952. London: Warburg
Institute, 1993.
. The Heritage of Avicenna: The Golden Age of Arabic Philosophy, 1000
ca.1350, in Avicenna and His Heritage, ed. J. Janssens & D. De Smet, 8197. Leuven:
Leuven University Press, 2002.
Gyekye, K. The Term Istithn in Arabic Logic, Journal of the American Oriental
Society 92(1972): 8892.
H asan, Abd al-H ayy. Nuzhat al-khawtir wa bahjat al-masmi wa l-nawzir.
Hyderabad, Deccan: Majlis Dirat al-Marif al-Uthmniyya, 1955.
Hitti, P., Faris, N.A. & Abd al-Malik, B. Descriptive Catalog of the Garrett Collection of
Arabic Manuscripts in the Princeton University Library. Princeton: Princeton
University Library, 1938.
nalcik, H. The Ottoman Empire: The Classical Age, 13001600. London: Weidenfeld &
Nicholson, 1973.
Inati, S. Logic. In A History of Islamic Philosophy, ed. S. H. Nasr and O. Leamann, II,
802823. London: Routledge, 1996.
Kafadar, C. The Question of Ottoman Decline. Harvard Middle Eastern and Islamic
Review 4 (19971998): 3075.
Kshif al-Ghit, Al, Naqd al-r al-mantiqiyya wa h all mushkiltih. Qomm: frand,
1427/2006 [reprint of 1954 edition].
Karatay, F. E. Topkap Saray Mzesi Ktphanesi Arapa yazmalar katalou. Istanbul:
Topkap Saray Mzesi, 1966.
282 bibliography
Khuda Bakhsh Oriental Public Library. Catalogue of the Arabic and Persian
Masnuscripts in the Khuda Bakhsh Oriental Library. Volume 21: Encyclopaedias,
Philosophy, Logic and Dialectics. Patna: The Library, 1970 ff.
Khvansr, Mrz Muhammad Bqir. Rawdt al-jannt f ahwl al-ulam wa l-sdt.
Qom: n.p., 1391/19711972.
Lameer, J. Conception and Belief in Sadr al-Dn Shrz. Tehran: Iranian Institute of
Philosophy, 2006.
. Al-Frb and Aristotelian Syllogistics. Leiden: E. J. Brill, 1994.
Lear, J. Aristotle and Logical Theory. Cambridge: Cambridge University Press, 1980.
Lee, Tae-Soo. Die griechische Tradition der aristotelischen Syllogistik in der Sptantike.
Gttingen: Vandenhoeck & Ruprecht, 1984.
MacEoin, D. Mull Sadr Shrz Sadr al-Dn Muhammad b. Ibrhm K awm
Shrz. Encyclopaedia of Islam, Second Edition. Edited by P. Bearman et al. Leiden:
Brill, 19602002. VII, 547.
Mach, R. Catalogue of Arabic Manuscripts (Yahuda Section) in the Garrett Collection,
Princeton University Library. Princeton: Princeton University Press, 1977.
Mach, R. & Ormsby, E. Handlist of Arabic Manuscripts (New Series) in the Princeton
University Library, Princeton: Princeton University Press, 1987.
Makhlf, M. Shajarat al-nr al-zakiyya f tabaqt al-Mlikiyya. Cairo: al-Matbaa al-
Salafiyya, 19301931.
Manekin, C. H., trans. The Logic of Gersonides. Dordrecht: Kluwer Academic
Publishers, 1992.
Maroth, M. Ibn Sn und die peripatetische Aussagenlogik. Leiden: Brill, 1989.
Merrill, D. D. Augustus De Morgan and the Logic of Relations. Dordrecht: Kluwer
Academic Publishers, 1990.
Miller, L. B. al-Samarqandi, Shams al-Dn. In Encyclopaedia of Islam, Second Edition.
Edited by P. Bearman, et al. Leiden: E. J. Brill, 19602002. VIII, 1038.
Minto, W. Logic: Inductive and Deductive. New York: Charles Scribner and Sons,
1894.
Moghaddam, Barat Ali Gholam. Fihrist-i Kitbkhna-yi stn-i Quds-i Radaw. XXIV:
Mantiq. Mashhad: Sazmn Kitbkhne-h, 1383/1963 ff.
Neumann, C. Bad Times and Better Self: Definitions of Identity and Strategies for
Development in Late Ottoman Historiography, 18501900. InThe Ottomans and
the Balkans: A Discussion of Historiography, ed. F. Adanr and S. Faroqhi, 5778.
Leiden: Brill, 2002.
Newman, A.J. Sabzawr, H djdj Mull Hd H djdj Mahd. Encyclopaedia of Islam,
Second Edition. Edited by P. Bearman et. al. Leiden: Brill, 19602002. VIII, 695.
ner, N. Tanzimattan sonra Trkiyede ilim ve mantk anlay. Ankara: Ankara
niversitesi Basmevi, 1967.
Pines, S. Ab l-Barkt Hibat Allh b. Malka al-Baghdd al-Balad. In Encyclopaedia
of Islam, 2nd edition, P. Bearman, et al., I, 111. Leiden: E. J. Brill, 1960.
Pourjavady, R. A Sh Theologian and Philosopher of Early Safavid Iran: Najm al-Dn
H jj Mah md Nayrz and his Writings. Unpublished PhD Dissertation. The
Department of History and Cultural Studies. Institute for Islamic Studies. Free
University of Berlin, 2008.
Read, S. Formal and Material Consequence. Journal of Philosophical Logic 23 (1994):
247265.
. Thinking About Logic: An Introduction to the Philosophy of Logic. Oxford: Oxford
University Press, 1995.
Rescher, N. The Development of Arabic Logic. Pittsburgh: University of Pittsburgh
Press, 1964.
Robinson, F. Ottomans-Safavids-Mughals: Shared Knowledge and Connective
Systems. Journal of Islamic Studies 8(1997): 151184.
El-Rouayheb, K. The Myth of the Triumph of Fanaticism in the Seventeenth-
Century Ottoman Empire. Die Welt des Islams 48 (2008): 196221.
bibliography 283
akasta 212, 218; see also aks and h ujja 252, 265; see also dall
yanakis al-h ujja al-nisbiyya 252; see also ifte
aks 84, 240; see also akasta and yanakis sural
aks al-naqd 22(n19), 110
antaj 41, 75, 122; see also yuntij idmr 153; see also mudmar
aql 68, 138, 140; see also taaqqul idti rr 14, 17, 19, 30, 33, 115, 147; see
asqatn 94, 98, 135; see also isqt and also darra
saqata ilm al-istidll 237; see also tartb and
awsat 15, 23, 35, 41, 45, 49, 50, 59, 81, istikhrj
82, 83, 94, 96, 97, 105, 106, 107, 111, indirj 97, 121, 127; see also mundarij
117, 121, 126, 127, 132, 133, 134, 135, intj 68, 83, 84, 85, 86, 87, 94, 96,
139, 140, 159, 164, 165, 174, 200, 263, 97, 102, 106, 120, 133, 134, 140,
264; see also wasat 178, 184, 193, 194, 197, 200, 218,
249, 265
badha 68, 138, 178; see also badh iqtirn 33, 49, 85, 99, 111, 116, 127,
and zhir 139, 196, 262
badh 67, 84, 106; see also badha and irj 105, 139, 140, 266; see also tah ll
zhir and tartb and istikhrj
bayn 54, 82, 121, 265; see also dall isqt 174, 198; see also saqata and
bi-dht 14, 14(n3), 17, 19, 78, 189, asqatn and h udhifa and h adhf
261; see also bi-l-dht, li-l-dht, istidll 119, 189; see ilm al-istidll
and li-dht istilzm 67, 73, 76, 116, 122, 249; see
bi-l-arad 14, 14(n3), 19, 30 also luzm
bi-l-darra see darra istikhrj al-qiys min al-ibra 171;
bi-l-dht 54, 57, 58, 60, 61, 68, 85, 102, see also tartb and irj and tah ll
153; see also bi-dht, li-l-dht, and istiqr 109, 265
li-dht istithn 74, 111, 196, 264, 266
burhn 10, 66, 85, 93, 104, 126, 132133, itti rd 124, 264; see also mutta rid
134, 189, 191, 262
khrijiyya (premise), see muqaddima
ifte sural 239 khrijiyya
khrijiyya (proposition), see qadiyya
dall 96, 125, 127, 189, 196, 265; see also khrijiyya 218, 263
istidll
darra 20(n15), 45, 49, 78, 84, 120, 126, lawzim 11, 119
127; see also idti rr lzim 33, 44, 4849, 78, 110, 120, 121,
162, 198, 264
ghayr mutaraf 161, 164, 165, 166, 167, lazima 14, 17, 18, 21, 30, 33, 43, 45,
169, 171, 172, 173, 178, 183, 188, 193, 4849, 77, 78, 80, 99, 100, 110, 114,
198, 210, 234, 244 115, 120, 121, 122, 147, 198; see also
ghayr mutaraf al-ghayr mutaraf 239 yastalzim and yalzam
li-dht 21, 43, 44, 56, 60, 77, 78, 82, 86,
h adhf 96, 97, 133; see also h udhifa and 99, 109, 110, 114, 122, 125, 132, 147,
isqt and saqata and asqatn 151, 153, 161, 174, 184, 190(n67), 196,
haya 110, 161, 164, 193194, 263; see 198; see also bi-dht, bi-l-dht, and
also sra li-l-dht
h udhifa 26, 51; see also h adhf and isqt li-l-dht 122; see also bi-dht, bi-l-dht,
and saqata and asqatn and li-dht
286 index of arabic and turkish terms
luzm 33, 49, 56, 66, 73, 78, 81, 84, 85, qadiyya h aqqiyya 218, 263
110, 132, 161, 191, 261, 264, 265; see qadiyya khrijiyya 218, 263
also istilzm qadiyya muallafa 117
qadiyya mah s ra 178, 265
mdd 20(n15), 110, 116, 118, 121, 122, qadiyya shakhsiyya 178, 267
174, 184, 263; see also mawdd qadiyya tabiyya 143, 178, 265
malzm 60, 91, 99, 110, 114(n2), 115, qawl muallaf 21, 43, 72, 76, 77, 80, 109,
119, 202, 203, 207, 208, 209, 213, 214, 110, 114, 125, 132, 136
215, 217, 220, 221, 264; see also qayd 110, 135, 136, 151, 160, 190, 267;
mustalzim see also mutaalliq
matlb 16, 44, 67, 82, 111, 117, 134, 135, qiys (definition of) 14, 17, 1819, 21,
185, 262; see also natja 24, 24, 25, 30, 34, 43, 64, 72, 77, 80,
matwiyya 140, 267; see also mudmar 109, 114115, 125, 131132, 136, 147,
and idmr 153, 189, 196
mawdd 45, 47, 96, 110, 136, 174, 249 qiys bast 137, 246
muallaf 93, 94, 132133, 134, 135, 139; qiys ghayr mutaaraf, see ghayr
see also qawl muallaf and qadiyya mutaaraf
muallafa qiys murakkab 134, 137
mudmar 17; see also idmr and qiys al-muswt 23, 45, 54, 67, 73, 75,
matwiyya 76, 78, 83, 84, 85, 87, 88, 91, 99, 102,
mundarij 121, 198; see also indirj 106, 110, 120, 122, 126, 127, 149, 159,
muntij 44, 61, 80, 138, 188, 198; see also 160, 161, 167, 168, 169, 174, 178, 188,
yuntij 198, 244; see also qiys al-taswiya
muqaddima ajnabiyya 73, 78, 85, 87, qiys al-nisba 59, 165, 266
106, 120, 121, 122, 169, 183, 188, 191, qiys al-taswiya 49; see also qiys
194, 198, 263, 264 al-muswt
muqaddima gharba 61, 8687, 183, 263,
264 radd 212; see also tah ll
muqaddima khrijiyya 120, 121, 134,
149 saqata 175; see also isqt and asqatn
murakkab 127, 147; see also qiys and h udhifa and h adhf
murakkab sullima 21, 43, 58, 77, 80, 114, 115, 122,
mustalzim 73, 126, 198; see also malzm 147, 153; see also taslm and nusallim
mutaalliq 68, 69, 87, 101, 122, 127, 164, sra 20(n15), 23, 24, 53, 64, 78, 89, 96,
169, 171, 172, 175, 183, 185, 200, 201, 97, 110, 121, 122, 136, 167, 169, 263;
207, 244, 250, 266; see also qayd see also suwar and haya
mutakarrir 41, 45, 50, 59, 83, 111, suwar 102, 192; see also sra
127(n41), 134, 135, 165; see also
takarrar and takarrur and yatakarrar taadd 82, 97, 135; see also taadd and
mutta rid 47, 102, 136, 139, 192, 249; see yataadd and naql
also itti rd taadd 102; see also taadd and
yataadd and naql
naql (of judgment) 35, 96, 97, 132(n54); taaqqul 66, 76, 83, 100; see also aql
see also yunqal and taadd tah ll 197, 237, 266; see also radd and
natja 41, 44, 45, 52, 66, 73, 83, 84, tartb and istikhrj
85, 100, 105, 110, 117, 121, 127, takarrar 68, 133, 135, 140; see also
134, 135, 139, 140, 151, 152, 164, yatakarrar and takarrur and
169, 171, 172, 174, 175, 178, 184, mutakarrir
188, 196, 198, 218, 231, 249, 262; takarrur 49, 67, 68, 81, 83, 85, 94, 96,
see also matlb 111, 126, 127, 132133, 134, 135, 139,
nisba 33, 58, 59, 60, 62, 64, 65, 91, 99, 140, 175, 191; see also yatakarrar and
111, 165, 177, 198, 262, 266; see also takarrar and mutakarrir
qiys al-nisba and al-h ujja al-nisbiyya takhallaf 60, 110; see also yatakhallaf
nusallim 81, 84, 94, 94, 121 and takhalluf
index of arabic and turkish terms 287
takhalluf 110, 193, 200, 201, 265; yalzam 20, 26, 28, 33, 44, 45, 60, 61, 64,
see also yatakhallaf and takhallaf 7677, 82, 99, 100, 101, 109, 110, 125,
talf 16, 20, 61, 78, 80, 110, 115, 193, 132, 136, 183; see also lazima and
194; see also muallaf and qawl yastalzim
muallaf yanakis 45; see also aks and akasta
tamthl 109, 110, 265, 267 yastalzim 72, 73, 122, 127, 149, 174, 196;
tarkb 76, 87, 122, 262; see also see also lazima and yalzam
murakkab yataadd 82, 97, 135; see also naql and
tartb al-qiys min al-ibra 237; taadd
see also istikhrj yatakarrar 45, 57, 106, 127, 132,
taslm 115, 132; see also sullima and 132133, 134, 140, 159, 164;
nusallim see also takarrar and takkarrur
tastakhlif 110; see also yatakhallaf and and mutakarrir
takhallaf and takhalluf yatakhallaf 96, 97, 122, 133; see also
taswr 185 takhallaf and takhalluf
tawassut 61; see also wsita yunqal 107; see also naql
yuntij 44, 52, 53, 54, 57, 58, 60, 61,
usl-i cedde 236, 243244 68, 78, 82, 84, 85, 94, 96, 101, 102,
120, 122, 127, 130, 133, 134, 139,
wasat 57, 67, 68, 83, 85, 96, 99, 106, 122, 159, 174, 178, 184, 188, 212, 249;
126, 191; see also awsat see also antaj
wsita 54, 66, 67, 73, 78, 82, 85, 86, 94,
98, 120, 122, 126, 127, 135, 147, 183, zhir 17, 41, 42, 44, 60, 61, 78, 174; see
184, 190, 261; see also tawassut also badha and badh
INDEX OF ENGLISH TERMS
Analogy 30, 109, 110, 137, 189, 265, 267 immediate 8, 10, 22(n19), 39, 48, 156,
Arabic Aristotelians 8, 15(n3), 40, 147, 226, 226(n46), 226(n47)
237, 238, 249, 261, 263, 266 material/due to matter 118, 147, 192
Aristotelian 1, 2, 3, 4, 11, 14, 15, 16, 25, necessary 34, 147
28, 29, 31, 32, 33, 34, 35, 36, 37, 38, syllogistic 25, 34, 44, 73, 76, 78, 92,
53(n26), 54, 70, 71, 74, 114, 117, 146, 94, 109110, 125, 129, 132, 153
158(n4), 214, 219, 244, 255, 256, 262, Induction 30, 32, 109, 110, 137, 189, 265
263, 264, 265, 266, 267; see also Arabic
Aristotelians Middle-Term, Recurrence of 35, 41,
Avicennian 11, 18, 25, 33, 34, 38, 40, 4546, 47, 49, 52, 57, 60, 62, 65, 66, 68,
40(n4), 41, 42, 51, 55, 71, 83, 104, 116, 80, 81, 82, 83, 85, 93, 94, 9596, 97, 98,
117, 146, 148, 196, 264, 265, 266; 101102, 103, 105, 111, 112, 125, 126,
see also Post-Avicennian 127, 128, 132, 133, 134, 135, 136, 137,
138, 139, 142, 143, 144, 145, 146, 150,
Contraposition 8, 48, 108, 110111, 156, 151, 154, 155(n107), 156, 159, 164,
225, 226(n46), 264 183, 188, 190, 191, 196, 245
Conversion 8, 48, 108, 111, 156, 172,
209, 212, 220, 225, 226(n46) Post-Avicennian 8, 72, 114, 116, 117,
Copula 65, 248, 262 146, 147, 148, 152, 156, 204, 227, 237,
Counter-example 29, 31, 35, 41, 51, 256, 264, 266
78, 85, 87, 90, 125, 128(n43), 129, Premise
141, 142145, 151, 162163, 177, Extrinsic 73, 78, 84, 85, 86, 106, 120,
179, 180, 181182, 186, 201, 204, 121, 122, 123, 125, 130, 132, 133,
205(n23), 210, 212, 214, 215, 218, 151, 169, 183, 187, 188, 190(n67),
219, 222224, 228, 229, 231, 242, 191(n69), 193, 195, 198, 263
246, 251, 264 Extraneous 120, 121, 149, 263
Implicitly understood 153, 267
Deduction 189, 192, 256 Imported 60, 86, 119, 154, 174, 183,
Demonstration 10, 6, 85, 93, 103(n65), 245, 263, 264
104, 114, 125, 126, 132, 134, 147, 189, Suppressed 15(n3), 17, 18, 19, 21, 24,
191, 262 34, 41, 44, 47, 55, 56, 57, 73, 74, 84,
99, 111, 118, 119, 120, 121, 123,
Enthymeme 17, 19, 24, 26, 30, 34, 37, 130, 132, 133, 139, 141, 153, 156,
163, 164, 171 191, 192, 252, 261, 267
Evident 29, 59, 62, 66, 67, 68, 75, 76, 85, Productivity; see also Implication
97, 106, 119, 120, 121, 127, 129, 138, by accident 14, 18, 21, 2223, 25, 26,
154, 176, 179, 251, 261, 262 30, 34
by (premise-pair) itself 14(n14), 21,
Fallacy 29, 35, 36, 107, 114, 142, 144, 25, 26, 34, 35, 42, 43, 44, 52, 54, 57,
152, 162, 163, 181, 194 59, 60, 61, 62, 68, 72, 73, 74, 76, 78,
80, 81, 84, 85, 86, 98, 99, 100, 101,
Implication 66, 261, 264, 265; see also 109, 110, 122, 127(n43), 129, 131,
Productivity 132, 147, 153, 154, 160, 161, 174,
by premises themselves/by premise- 175, 183, 184, 187, 188, 189, 195,
pair itself 25, 44, 73, 122, 125, 129, 196, 197, 198, 245, 261
132, 153 by mediation 54, 60, 66, 73, 78, 81,
formal/due to form 36, 110, 122, 137, 84, 85, 86, 122, 125, 126, 127, 147,
192 153, 190(n67)
290 index of english terms
by (premises) themselves 14, 14(n3), 185, 186, 195, 198, 239, 248, 252,
17, 19, 21, 24, 26, 30, 32, 34, 56, 58, 255, 262, 264, 266
64, 68, 73, 77, 78, 79, 80, 84, 90, 99,
110, 114, 115, 118, 120, 123, 125, Semantic Dependent 6869, 87, 89,
128, 129, 130, 131, 140, 150, 153, 90, 91, 94, 101, 122, 127, 128, 129,
160, 174, 175, 184, 187, 190(n67), 132, 142, 150, 154, 159, 160, 163,
192, 261 164, 169, 170, 172, 175, 183, 185,
by virtue of structure 193 186, 187, 188, 191, 198, 199, 200,
formal/by virtue of form 20, 20(n15), 201, 203, 204, 205, 206, 207, 208,
23, 25, 30, 31, 32, 34, 35, 36, 37, 209, 210, 211, 213, 218, 221, 224,
38, 47, 61, 78, 85, 91, 92, 100, 107, 240, 241, 242, 244, 246, 247, 248,
110, 111, 121, 128, 129, 136, 137, 250, 251, 252, 253, 266267
138, 139, 140, 141, 143, 149, 150, Stoic 15, 115, 196, 244, 264, 266
159, 160, 161, 169, 176, 181, 183, Subsumption 97, 120121, 127,
192, 193, 194, 202, 204, 209, 210, 127(n43)
214, 217, 220, 228, 230, 231, 232, Syllogism
234, 242, 245, 247, 261, 263, 264, Categorical 2, 3, 7, 15, 16, 25, 28,
267 34, 35, 36, 40, 41, 42, 49, 54, 56,
invariant 47, 101, 124, 136, 139, 192, 57, 62, 85, 98, 114, 127, 132, 133,
249, 264 137, 139, 141, 146, 151, 152,
material/by virtue of matter 20, 155(n107), 173, 196, 228, 244,
20(n15), 25, 30, 34, 35, 4546, 92, 258, 262
118, 122, 184, 192 Combinatorial 8, 33, 49, 111, 114,
regardless of matter/in all matter 147, 196, 262
20(n15), 47, 87, 96, 109, 110, 116, Complex 137, 175, 246
122, 124, 136, 174, 192, 249 Definitions of 14, 17, 1819, 21, 24,
Proof 32, 62, 82, 84, 9596, 97, 126, 127, 24, 25, 30, 34, 43, 64, 72, 77, 80,
138, 154, 185, 189, 190(n67), 196, 208, 109, 114115, 125, 131132, 136,
212, 214, 215, 217, 234, 237, 261, 265 147, 153, 189, 196
Proposition Double-minor 239, 245, 246, 252
Natural 143, 178180, 181, 182(n49), Hypothetical 8, 15, 3233, 74, 114,
199, 202, 265 115, 147, 196, 262, 264, 266
Quantified 117, 143, 178, 179, 180, Reduplicative 74, 111, 196, 244, 262,
181, 182, 199, 202, 211, 265 264, 266
Singular 127, 178, 180, 182, 199, 202, Relational 4, 7, 10, 11, 48, 59, 65, 201,
267 228, 266
Unquantified 211, 267 Simple 25, 28, 35, 59, 131, 137, 149
Unfamiliar 2, 4, 10, 156, 161, 163,
Qualifier (of subject or predicate) 135, 163195, 196227, 228236, 238,
190191, 267; see also Semantic 239, 240, 241, 242
Dependent Unfamiliar unfamiliar 239243, 244
Reduction 90, 139, 140, 176, 215, 266 Transfer (of judgment) 96, 97, 132
Relation 9, 5859, 6162, 62(n39), Transmission (of judgment) 82, 93, 97,
64, 65, 91, 92, 126, 151, 177, 182, 98, 102, 103, 128, 135, 142, 261
INDEX OF PROPER NAMES
Abd al-Latf Khn, Uzbek ruler of Bain, Alexander (d. 1903) 250
Transoxania (r. 15401552) 108 Bjr, Ibrhm (d. 1860) 77(n17), 124
Abd l-Azz, Ottoman Sultan Bannn, Muhammad b. al-H asan
(r. 18611876) 232 (d. 1780) 6, 123
Abd l-Mecd, Ottoman Sultan Baydw, Qd Abd-Allh (d. 1316) 189,
(r. 18391861) 243 189(n64), 255
Abd l-Nfi Iffet Efendi (d. 1890) 229 Bihr, Muhibb-Allh (d. 1707) 5, 12,
Abdu, Muhammad (d. 1905) 5(n6), 124 113, 125126, 128, 129(n44), 130, 131,
Abhar, Athr al-Dn (d. 1265) 11, 12, 132, 156, 165, 192, 226, 235, 251(n48),
69(n56), 79, 86, 108, 108(n77), 129, 256
145, 164, 165, 167(n19), 177, 223, Birgev, Ahmed kir (fl. 1839) 175,
226(n46), 229, 230, 239 179, 180, 251
Ab l-Barakt al-Baghdd, Hibat-Allh Boole, George (d. 1864) 254
b. Malka (d. 1165) 3, 11, 27, 2830, Bulghr, Burhn al-Dn (fl. ca. 1460)
31, 33, 38, 40(n4), 111, 237 8688, 90, 168
Acem, Al (d. 1456) 8385 Brsal, Mehmed T hir (d. 1926) 243
Adaw al-Sad, Al (d. 1775) 124 Brsev, Ahmed Sidk (d. 1895) 210,
Ahmadnagr, Abd al-Nab 234, 242, 245(n35)
(fl. 1759) 166
Akhdar, Abd al-Rahmn (d. 1546) 79, Caramuel, Juan (d. 1682) 1
122, 123(n28) erkezde, Mehmed Tevfk
Akhisr, Mehmed Sad (d. 1901) 230
(d. 1902) 233234 Cevdet, Ahmed (d. 1895) 229
Akehirl, H asan Fehm (d. 1880) 232
lehr, Osmn (d. 1776) 183184, Drendev, Mehmed (d. 1739) 184, 187,
188, 195 188, 195, 198
Alexander of Aphrodisias (fl. 200) 1, 16, Dashtak, Ghiyth al-Dn (d. 1542) 100,
21, 29, 37, 38, 54 101, 102104, 144, 145, 146,
Ams, c Kerm (d. 1885) 229 147(n82), 152
Ammonius (d. ca. 520) 16 Dashtak, Mr Sadr al-Dn (d. 1498) 4,
mid, Sayf al-Dn (d. 1231) 11, 4951, 92, 94(n49), 9596, 97, 100, 102,
55, 70 106(n73), 112, 138, 140, 144, 152, 164
Ansr, Zakariyy (d. 1519) 164 Dasq, Muhammad b. Arafa
Araj, Muhammad Rid (fl. 1716) 148 (d. 1815) 124
Aristotle 8, 9, 11, 14, 14(n3), 15, 16, 17, Dawn, Jall al-Dn (d. 1502) 4,
27, 29, 30, 31, 32, 34, 36, 39, 48, 62, 9295, 9699, 100102, 103, 103(n63),
65, 71, 72, 115, 116, 147, 148(n87), 104, 105, 106, 107, 108, 109, 112, 113,
158(n4), 189, 189(n64), 196, 261 123, 123(n30), 126, 132, 133, 134,
Att r, H asan (d. 1834) 124 136, 138, 138(n64), 139, 140, 141,
Averroes, see Ibn Rushd al-H afd 142, 143, 144, 145, 146, 150, 151,
Avicenna, see Ibn Sn 152, 155, 156, 158, 159, 160, 161,
Awrangzeb, Mughal ruler 162, 163, 164, 16, 173, 191, 194,
(r. 16581707) 148 199, 225, 226, 226(n46), 251(n48),
Ayntb, Necb Abd-Allh 257, 258, 261
(d. 1804) 183, 191(n69), 193194 De Morgan, Augustus (d. 1871) 1, 2, 4,
53(n26), 219, 254, 258
Bghnav, see Mrz Jn H abb-Allh
Bahmanyr b. Marzubn (d. 1065) 3, 23, Emle, H asan b. H seyn (fl. 1548)
2425, 26, 37, 104, 104(n67), 237 106(n73), 108(n78), 165
292 index of proper names
Frb, Ab Nasr (d. 950) 3, 7, 11, Ibn Krn, al-T ayyib (d. 1812) 6, 123
1417, 71, 74, 114, 146, 148, 189, Ibn Malka, see Ab l-Barakt
237, 256 al-Baghdd
Fil-i Hind, see Isfahn, Bah al-Dn Ibn Marzq al-H afd (d. 1439) 72(n5),
Fenr, Mehmed (d. 1431) 8283, 86, 7577, 78
164, 168, 226(n46), 239 Ibn Mubrak Shh (fl. 1360s) 7982,
Fursat Shrz, Muhammad Nsir 251(n48)
(d. 1920) 6, 153, 155 Ibn al-Mutahhar, see H ill
Ibn al-Qaradgh (d. 1936) 182(n49),
Galen (fl. 2nd century CE) 1 196(n2), 212, 219, 220221, 229
Gaspral, Ismail (d. 1914) 249 Ibn Rushd al-H afd [Averroes]
Gelenbev, Isml (d. 1791) 4, 12, (d. 1198) 3, 7, 8, 11, 27, 28, 3034,
113, 158, 182(n49), 183, 196227, 36, 38, 71, 72, 111, 146, 148, 148(n87),
228, 229, 230, 232, 233, 235, 236, 237, 255, 256
239, 244, 246, 247, 248, 251(n48), Ibn Sad al-H ajar (d. 1785) 119
256, 258 Ibn al-Sharf al-Jurjn
Ghazl, Ab H mid (d. 1111) 7, 255 (d. 1434) 108(n77), 165
Gln, Abd-Allh (fl. ca. 1600) 138 Ibn Sn [Avicenna] (d. 1037) 3, 6, 8,
139, 155(n107) 11, 1824, 25, 27, 28, 30, 33, 34, 37,
Gpamv, Qd Mubrak (d. 1747) 126, 38, 39, 40(n4), 41, 43, 46, 47, 48,
251(n48) 49(n18), 51, 52, 54, 57, 67, 69, 70,
71, 72, 75, 93, 100, 101(n58), 103,
H dim, Eb Sad (d. 1762) 158, 104, 105, 108(n77), 136, 140,
176(n39), 183, 184, 188193, 194 144(n77), 145, 146, 146(n80), 147,
H afd al-Taftzn, Sayf al-Dn al-Haraw 148, 151, 189, 237, 257
(d. 1511) 108(n78), 165 Ibn Taymiyya (d. 1328) 40, 148(n86)
H afn, Ysuf (d. 1764) 124 Ibn al-T ayyib al-Qdir, Abd al-Salm
Hamadn, Abd al-Hd (d. 1913) 6, (d. 1698) 120
153155 Ibn Turka al-Isfahn (d. 1437) 7, 145
Hamilton, Sir William (d. 1856) 250, Ibn Wsil al-H amaw (d. 1298) 70,
251, 254 7273, 110
H arpt, Ysuf ukr (d. 1875) 229, 232 Ibn Yaqb al-Walll, Ahmad
H ayt Mara, Ahmed (d. 1716) 6, 121123, 156
(d. 1813) 207(n25), 240 Ibn Zura (d. 1008) 3, 14, 1718, 19,
Haydaj, Muhammad Al (d. 1927) 6, 27, 34, 36, 38, 50, 74, 148
153, 155 j, Adud al-Dn (d. 1355) 16, 189,
Herav, Mr Zhid (d. 1689) 226 190(n6), 225(n44), 255
Hill, Ahmad b. Abd al-Azz Ilh Ardabl, H usayn (d. 1543) 89
(d. 1761) 6, 120121, 122, 123, 156, nalcik, Halil 157
226(n47) Isfahn, Bah al-Dn Fil-i Hind
H ill, Ibn al-Mutahhar (d. 1325) 5(n6), (d. 1725) 148
63, 80(n24), 95, 108(n77), 109(n80), Isfahn, Shams al-Dn Mahmd
164, 223(n41) (d. 1348) 63, 107, 255(n2)
H seyn Shh (fl. 16th or 17th Isfarn, Ism al-Dn (d. 1537)
century) 166 9092, 109(n80), 113, 189,
H uwayz, Ishq (fl. 17th century) 148, 190(n67)
150151 Izmr, Abd l-Rahmn (d. 1779) 183,
184187, 188, 208, 236, 237238
Ibn Ghulm Muhammad
(fl. 1802) 130131 Jabalrd, Khidir (fl. 1421) 165
Ibn H azm (d. 1037) 7 Jm, Abd al-Rahmn (d. 1492)
Ibn Kammna (fl. 1268) 7 90(n44), 238
Ibn Khaldn (d. 1406) 8, 39, 48, 55, Jarb, Sulaymn (fl. 1507) 79
72, 146 Jevons, William Stanley (d. 1882) 254
index of proper names 293
61, 67, 69, 70, 71, 72, 81, 94, 95, 98, Srz, Musta f K utb (fl. early nineteenth
104, 114, 123, 141, 145, 146, 147, 237, century) 4, 202, 202(n13), 203, 206,
255, 257, 263 207, 208, 209, 211, 213, 215, 217, 220,
Rz, Qutb al-Dn (d. 1365) 3, 11, 221(n39), 227, 228229, 236, 243
40(n5), 54, 63, 6669, 77, 78, 79, Sirr, Sleymn (fl. 1892) 230
80(n24), 82, 83, 84, 85, 87, 87(n39), irvn, Mehmed Emn Sadruddnzade
88, 89, 90, 91, 97, 99, 99(n56), 100, (d. 1627) 113, 143, 144, 158, 159163,
101, 102, 103, 105, 106(n73), 107, 164, 16, 167, 168, 169, 173, 179, 181,
108(n77), 112, 123, 123(n30), 124, 187, 194, 226(n46)
125, 126, 128, 129, 131, 142143, Siverek, Musta f (fl. 1793) 183,
144(n77), 150, 151, 154, 155, 160, 187188, 193
165, 190(n67), 192, 225, 226, Siylkt, Abd al-H akm (d. 1657)
226(n46), 243, 244, 245, 257 124125, 131(n49), 156
Rescher, Nicholas 2, 1112, 14(n1), 39, Street, Tony 2, 11, 71
40, 47, 58(n35), 63(n45), 70, 72(n6), Suhraward (d. 1191) 3, 7, 27, 28, 3436,
79(n22), 83(n32), 105(n68), 109(n80), 38, 63, 107, 133, 145, 148(n86)
155, 189, 255, 256, 257
Ridaw, Khall (fl. ca. 1500) 8990, 91, Tabrz, Mull Hanaf (fl. 1516) 166
170, 172 Tahnaw, Muhammad Al (fl. 1745)
Ridwn, Musta f (fl. 1871) 124 130(n46), 166
Russell, Bertrand. 71, 256, 257 Taftzn, Sad al-Dn (d. 1390) 5, 7,
80(n24), 99(n56), 108, 109, 118, 119,
Sabbn, Muhammad (d. 1791) 124 122(n25), 123, 131, 136, 137,
Sabzavr, Mull Hd (d. 1878) 6, 137, 150(n92), 153, 165, 189, 190(n6), 225,
152153, 155, 229 257
Saklzde, Mehmed (d. 1732) 158, T leqn, Mull Muhsin (fl. 1720)
189, 190(n68) 148, 149
Sadr al-Shara al-Mahbb (d. 1346) 3, T arss, Osmn b. Musta f (fl. ca. 1800)
6366, 69, 165, 189, 251(n48) 177, 219, 231, 232, 233, 234, 237, 238,
Safaw, s (d. 1546) 165 239, 240, 241, 248
Samarqand, Shams al-Dn (d. 1303) 3, T kprzde, Ahmed (d. 1561) 63, 165,
8, 5862, 64, 65, 67, 69, 70, 88, 94, 166
98, 99, 102, 123, 146, 165, 166, T vskr, Mehmed b. Musta f (fl. 1748)
257, 266 164(n14), 167, 168, 169, 170, 171,
Sammk, Fakhr al-Dn H usayn 174182, 184, 186, 187, 188, 195, 197,
(d. 1576) 138 198, 199, 202, 206, 212, 213, 220, 221,
Sandl, H amd-Allh (d. 1747) 126 230, 234, 240, 244, 248, 258
Sans, Muhammad b. Ysuf (d. 1490) 6, Tehrn, Muhammad Ysuf (fl. 1694)
72, 72(n4), 72(n5), 7779, 118, 119, 146148, 156, 189, 256
120, 123, 165 Tilimsn, Muhammad al-Sharf
Sw, Umar b. Sahln (fl. 1145) 3, 5(n6), (d. 1370) 55, 7274, 75, 110,
2527, 37 118, 123
Sedt, Al (d. 1900) 254 Tkd, mer Fey (d. 1848) 167(n19),
Selim III, Ottoman Sultan 177, 230, 234
(r. 17891807) 177 Tonkbon, Muhammad b. Sulaymn
Shh Isml, Safavid ruler (d. ca. 1892) 40
(r. 15011524) 146 T lw, Butrus (d. 1745) 114115, 116,
Shahrazr, Shams al-Dn (fl. 1288) 117, 279
35, 63 T s, Nasr al-Dn (d. 1274) 3, 5(n6),
Shrz, Jaml al-Dn Mahmd 6, 11, 12, 40, 5155, 63, 69, 74, 75, 81,
(d. 1554) 104, 109(n79), 138(n64) 93, 95, 99, 100, 101, 102, 103, 105,
Shrz, Qutb al-Dn (d. 1311) 35(n45), 106, 107, 108(n77), 119, 132, 136,
40(n5), 95, 106107, 107(n74), 133 140, 141, 144(n77), 145, 148, 164,
Shirwn, Masd (d. 1499) 166 223(n41), 256
index of proper names 295
Urmaw, Sirj al-Dn (d. 1283) 5, 6, 8, 9, Ynyav, Esad (d. 1722) 7, 158(n4)
57, 58, 63, 6, 67, 70, 72, 78, 81, Yazd, Mull Abd-Allh (d. 1573)
82(n28), 85, 99, 99(n56), 108, 108(n77), 109, 131, 137, 137(n62),
108(n77), 119, 125, 151, 154, 165, 192, 146, 148, 150, 151, 152, 165
225, 257 Yzij, Nasf (d. 1871) 117
Yemlhzde, Mustaf Kmil
Velcn Mara, Mehmed b. Viz (d. 1877) 231, 234, 237, 241, 242, 248
(d. 1746) 183, 187188, 193, 194 Ys, al-H asan (d. 1691) 6, 118120,
121, 122, 123, 124, 156
Williams, Bernard. 7 Yuwkm Mutrn (d. 1772) 115116,
Wisnovsky, Robert 2 117