THE TELEOLOGICAL ETHICS OF FAKHR AL-D^N AL-R$Z^

ISLAMIC PHILOSOPHY
THEOLOGY AND SCIENCE
Texts and Studies
EDITED BY

H. DAIBER and D. PINGREE

VOLUME LXIV

THE TELEOLOGICAL ETHICS OF

FAKHR AL-D^N AL-R$Z^
BY

AYMAN SHIHADEH

BRILL
LEIDEN BOSTON
2006

This book is printed on acid-free paper

Library of Congress Cataloging-in-Publication Data

A C.I.P. record for this book is available from the Library of Congress.

ISSN 0169-8729
ISBN 90 04 14991 0
Copyright 2006 by Koninklijke Brill NV, Leiden, The Netherlands
Koninklijke Brill NV incorporates the imprints Brill Academic Publishers,
Martinus Nijhoff Publishers and VSP.
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,
mechanical, photocopying, recording or otherwise, without prior written
permission from the publisher.
Authorization to photocopy items for internal or personal
use is granted by Brill provided that
the appropriate fees are paid directly to The Copyright
Clearance Center, 222 Rosewood Drive, Suite 910
Danvers MA 01923, USA.
Fees are subject to change.

printed in the netherlands

contents

CONTENTS
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Al-Rzs Biography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Development of al-Rzs Thought and the
Chronology of his Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1
1
4

Chapter One. Al-Rzs Theory of Action . . . . . . . . . . . . . . . . . .

The Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Theory of Human Action . . . . . . . . . . . . . . . . . . . . . . . . .
Destiny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Theme of the Creation of Human Acts . . . . . . . . . . . . . .

13
13
17
29
39

Chapter Two. Al-Rz on the Ethics of Action . . . . . . . . . . . . . .

The Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ethical Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Revealed Law and Ethical Value . . . . . . . . . . . . . . . . . . . . . . .
Moral Obligation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Objections to Mu#tazil Ethics . . . . . . . . . . . . . . . . . . . . . . . . .
God and Ethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

45
47
56
63
73
83
96

Chapter Three. Al-Rzs Perfectionist Theory of Virtue . . . . . .

Perfectionism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Human Perfection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Prophecy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

109
109
116
129

Chapter Four. Al-Rzs Later Pessimism:

Commentary on Rislat Dhamm al-ladhdht . . . . . . . . . . . . . . . . . .
The Nature and Extent of Sensory Pleasure and Pain . . . . . .
Socio-Political Pessimism . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Epistemological Pessimism . . . . . . . . . . . . . . . . . . . . . . . . . . . .

155
156
169
181

Appendix: Al-Rzs Rislat Dhamm ladhdht al-duny . . . . . . . . . . . 205

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Index of Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Index of Subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

vi

contents

contents

vii

PREFACE
The present study aspires, first and foremost, to make a contribution
to two main areas of interest in Islamic intellectual history, namely
ethical philosophy and the thought of Fakhr al-Dn al-Rz. I hope
to have demonstrated that al-Rz is one of the most important
ethicists in Muslim history, and to have produced some stimuli for
further research on his thought, as he remains one of the most
influential, yet sorely understudied, medieval thinkers.
The present monograph is based, often remotely, on a doctoral
thesis that I submitted in 2002 to the faculty of Oriental Studies,
Oxford University, under the supervision of Professor Yahya Michot
and Dr Fritz Zimmermann. I would like to express my gratitude
to both my supervisors, from whom my research has benefited in
numerous ways. I am indebted to Professor Michot for more recent
exchanges of ideas, many of which are relevant to questions tackled
in this thoroughly revised version.
I would like also to extend my sincere thanks to Professor Wilferd
Madelung, Professor Hans Daiber and Dr Tony Street, who provided
me with valuable comments on this study, to Professor Yahya Ibn
Junaid and Dr Nizam Yaquby, who both assisted me in acquiring
copies of some manuscripts, and to Sobia Syyed for reading the
monograph and suggesting some corrections in style. Needless to
say, I alone am responsible for its present form.
Finally, I am truly grateful to the following institutions: to the
Muslim Academic Trust, Cambridge, for generously funding my
graduate study; to Oxford University for funding provided during
my doctoral study; to the British, Berlin State, Mar#ash-Najaf and
Princeton University Libraries, for facilitating access to manuscripts
of Rislat Dhamm ladhdht al-duny, allowing me to produce a critical
edition thereof; and to Brill for undertaking this publication.

viii

contents

introduction

INTRODUCTION
General Introduction
The first centuries of Islam witnessed the emergence of different
traditions of ethical thinking, within which several distinct ethical
theories were propounded. The most sophisticated philosophical
theories were developed within the two largely independent traditions of kalm and falsafa, which clashed at the level of ethics and
in other related areas, including metaphysics, cosmology, psychology
and epistemology.
The main ethical concern of the classical mutakallimn was to
investigate the nature of Gods justice and the goodness of His acts
generally, which they approached through analytical discussions of
ethical language, metaphysics and epistemology. Similar discussions,
likewise with an emphasis on action, can also be found in ul al-fiqh,
and concern the establishment of general normative principles for
human conduct. The falsifa, on the other hand, were chiefly, but by
no means solely, interested in the development of human character,
primarily by engendering virtues, which are essentially dispositions
internal to the individual.
The gap separating the two traditions was initially so wide that
many notions central to one tradition of ethical theory were completely
alien to the other, in which they would normally be dismissed in toto,
without engagement in any proper dialogue. Yet there then emerged
signs of increasing, and more positive, interaction between kalm
and falsafa, culminating in the efforts of al-Ghazl (d. 505/1111),
who was both a severe critic of the falsifa and deeply influenced by
them in many respects. A century later, Fakhr al-Dn al-Rz was
to open the gates widely, allowing a more liberal exchange of ideas,
a synthesis even, between kalm and falsafa. This feature manifests
no less in his ethics than in other areas of his thought.
The present study is thus, at once, both a comprehensive analysis
of one major facet of al-Rzs thought, viz. his ethical theory, and an
exploration of the main trends and debates in its wider intellectual
background. It shows that he sets forth a sophisticated and original
ethical theory, which is both eclectic and highly consistent internally.

introduction

In this theory, he departs with classical Ash#ar voluntarism.

Al-Rz is significant in this regard not as a moralist (al-Ghazl
is evidently a more elaborate moralist than he is), but mainly as
an outstandingly analytical and thoroughgoing ethical philosopher.
In this respect, his discussions of certain ethical themes are among
the most penetrating in Islamic history and will easily match corresponding discussions in any extant Mu#tazil texts. This owes partly
to his firsthand familiarity and engagement with the writings of
the Mu#tazila, especially the school of Ab l-\usayn al-Bar (d.
436/1044), to an extent unprecedented among their earlier critics
(and which was apparently unsurpassed in later Sunni theology, to
which Mu#tazilism became less relevant). Al-Rz also had a great
deal of influence on later ethical thought in Islam, especially in kalm;
and his main work on the science of character, Kitb f #ilm al-akhlq,
is listed by Ibn al-Akfn (d. 749/1348) as a major representative of
its genre (however, its circulation seems to have become limited in
later centuries).1
Al-Rz develops a metaethical theory that underlies both his falsaf
ethics of character and his kalm and juristic ethics of action.2 This
underlying theory manifests differently in these two different contexts:
as a consequentialism in relation to action, and as a perfectionism
in relation to character. Although at the level of normative ethics
his elaboration of the relation between his ethics of action and his
ethics of character remains in certain respects underdeveloped, the
relation between the two at the metaethical level is made clear:
consequentialism and perfectionism are two aspects of the same
teleological ethics, rather than separate ethical theories.3
These two areas of ethical enquiry are discussed separately in al-

1 Ibn al-Akfn, Irshd, 401. The two other representative works that he lists are
Ibn Sns Rila f l-akhlq and Miskawayhs Al-Fawz al-aghar.
2 Contemporary ethical philosophy is normally divided into the sub-fields of
metaethics and normative ethics. The distinction is not universally accepted, primarily because the two fields are arguably not mutually exclusive. Metaethics, or
so-called second order ethics, seeks to understand the nature and justification
of ethical judgement. Normative, first order ethics denotes attempts to defend
or establish ethical judgements on specific types of human action, or norms and
principles to guide human action (cf. Sh. Kagan, Normative Ethics, esp. 16; articles
Analytic ethics, REP; Metaethics, EE).
3 These different types of ethical theory are defined p. 478 infra.

introduction

Rzs writings for several reasons. First of all, the classical mutakallimn focused on action exclusively, as they were concerned with
investigating how Gods acts relate to His creatures. Al-Rz, too,
discusses divine action, which he approaches on the basis of a thorough analysis of human action. Moreover, in developing his ethical
theory, he works within separate established traditions, each having
its own scope and ethos. He does not attempt to produce a synthesis
between the science of character and jurisprudence. Yet he does
provide some general guidelines on how the ethics of action and the
ethics of character should be viewed in relation to each other.
The starting point for this study will be al-Rzs ethics of action
(Chapter II), for which first we will need to examine his theory of
action, which is central to his metaethics (Chapter I). Since he starts
as a classical Ash#ar theologian with little interest in the examination
of character, his interest in action will have chronological precedence
in his intellectual career. Even in his later thought, his analysis of
action does not presuppose a theory of character (which normally
would give moral primacy to character over action), although it will
be complemented by such a theory. Chapter III will then examine his
theory of virtue, including his ethics of character and the influence
of his theory of virtue on his later theory of prophecy.
Chapter IV will focus on the epistle entitled Dhamm ladhdht alduny (Censure of the Pleasures of This World), which is published here
and studied for the first time (a critical edition can be found as an
Appendix). In this immensely interesting short text, which al-Rz
wrote towards the end of his life, he expresses pronounced moral
pessimism and intellectual scepticism. The background of this stance
in his writings more widely will also be explored.
The narrower theoretical themes covered in this study will be
introduced in their appropriate places in the following chapters.
First, however, we should provide a short biography of al-Rz and
a brief descriptive bibliography of his main writings that are cited
in this study.

introduction
Al-Rzs Biography4

Ab #Abdullh Muammad Ibn #Umar Fakhr al-Dn al-Rz was

born in 544/1149 in the city of Rayy to its main preacher; whence
his title, Ibn Khab al-Rayy, or more commonly Ibn al-Khab.
His father 4iy" al-Dn (d. 559/1164), who was a prominent Shfi#
and Ash#ar,5 was Fakhr al-Dns first teacher in both kalm and fiqh.6
Al-Rz, thus, began as a very traditional Ash#ar, as is clear from
his earliest books. Among those he studied with after his father died
were Amad Ibn Zarnkum al-Kaml al-Simnn (d. ?) of Simnn,
not far from Rayy, and the then famous philosopher-theologian Majd
al-Dn al-Jl (d. ?) of Margha in Azerbaijan, who was a student
of Muammad Ibn Yay al-Naysbr (d. 548/1153), al-Ghazls
student (d. 505/1111). Reportedly, he then continued to study the
philosophical sciences independently.7
Al-Rz travelled widely throughout his life, mainly in Persia, parts
of central Asia and northern India.8 At some stage in his travels, he
established a close relationship with the Ghrid sultan Ghiyth al-Dn
(d. 599/1203) of Ghazna, and reportedly worked for him, and then
for his brother and successor, Shihb al-Dn (d. 602/1206). Later,
while maintaining a good relationship with the Ghrids, he became
close to their opponent, the Khwrizm-Shh #Al" al-Dn Tekesh
(d. 596/1200), worked for him, and taught his son, Muammad (d.
617/1219). When the latter inherited the sultanate, al-Rzs status

4 Some of the main classical biographies of al-Rz are: Ibn Ab Uaybi#a, #Uyn,
3, 3445; al-Qif, Ikhbr, 1902; al-#Asqaln, Lisn, 4, 4269; Ab Shma, Dhayl,
68; Ibn Khallikn, Wafayt, 4, 24852; Ibn al-#Ibr, Trkh, 240; 254; Ibn al-S#,
Jmi#, 9, 3068; Ibn Kathr, Bidya, 13, 556; al-afad, Wf, 4, 24858; al-Dhahab,
Trkh, 43, 21123; al-Subk, abaqt, 8, 8196; al-Yfi#, Mir"t, 4, 711; al-Shahrazr, Nuzha, 2, 14450. See also: Muammad al-Zarkn, Fakhr al-Dn al-Rz, 855;
Georges Anawati, Tamhd, 193201; Elements; Fakhr al-Dn al-Rz in EI2;
John Cooper, al-Razi, Fakhr al-Din, REP; Yasin Ceylan, Theology, 113; Tony
Street, Life and Works; Jacques Jomier, Quranic Commentary.
5 Cf. al-Subk, abaqt, 3, 159 ff.; 7, 242; Ibn Khallikn, Wafayt, 4, 252.
6 Al-Rzs Ash#ar chain extends from his father, to Ab l-Qsim al-Anr,
to al-Juwayn, to Ab Isq al-Isfar"n, to Ab l-\asan al-Bhil, to al-Ash#ar
(Tal al-aqq, 52; cf. al-Yfi#, Mir"t, 4, 11). He also presents his Shfi# juristic
chain, starting with his father.
7 Ibn Ab Uaybi#a, #Uyn, 3, 34.
8 E.g. Munart, 7; Malib, 7, 38891. On his travels, see J. Jomier, Quranic
Commentary, passim.

introduction

increased greatly. It appears that although he was initially poor,9

al-Rz acquired great wealth later on, perhaps as a result of his
high-profile links. In Herat, the Ghrid sultan built him a school,
where he taught a great number of students.
Accounts of his travels also abound with debates he had with proponents of various other theological schools, especially the Karrmiyya,10
Mu#tazils and \anbals. His debating abilities brought him notoriety.
He himself recorded some of his debates in a separate collection,
the Munart, while others are referred to by later sources.11 On
one occasion, a debate he had with the head of the Karrmiyya in
Frzkoh, the capital of a branch of the Ghrids, resulted in a riot,
forcing him to depart.12 The extent of the enmity that this sect had
towards him was such that, in his deathbed will, he asked for his
body to be buried secretly in a remote place, apparently for fear that
supporters of this sect may exhume and mutilate it. Al-Rz died in
606/1210 in Herat.

The Development of al-Rzs Thought and the Chronology of His Works

Most studies on al-Rzs thought rely on a more or less narrow
selection of his writings, often with little attention paid to their
chronology. The sheer volume of his writingshe is a main contender to being the most prolific of all Muslim theologians and
philosophers13and the fact that some of the most important among
them remain unpublished, are normally sufficient to place the student of his thought under a formidable burden, more so than is
the case with most other major Muslim thinkers. Yet, positively, in
contrast to some other thinkers, most of al-Rzs works, including
the most important ones, have survived. In the absolute majority
of these texts, one encounters no problems in attributing them to

E.g. al-Qif, Ikhbr, 190.

On the Karrmiyya, an anthropomorphist sect, see I#tiqdt, 101; al-Shahrastn,
Milal, 1, 10813; article Karrmiyya in EI2.
11 E.g. al-Qazwn, $thr, 2523; cf. Tafsr, 7, 88; 8, 6971.
12 Ibn al-Athr, Kmil, 12, 1512.
13 Biographers report that he authored more than 200 works (Ibn al-S#, Jmi#,
9, 307; Ibn Kathr, Bidya, 13, 55). Around a third of these are known to be extant.
The most useful (though incomplete and out of date) published bibliography of alRzs known works remains: M. al-Zarkn, Fakhr al-Dn al-Rz, 56164.
10

introduction

their author. Most, including all the important ones, are in Arabic;
some are in Persian.
The present study uses the most comprehensive selection of alRzs works to date; several are used for the first time. This has
allowed a more accurate understanding of the complex developments
that took place in his thought, which are often subtle, but sometimes
striking. Some of his earliest and latest works give the impression
of being written by very different authors (yet they are his, without
doubt). To give one curious example, while in the early Ishra, Ab
l-#Al" al-Ma#arr is cursed for his heresy, he is referred to appreciatively in the late Tafsr and Malib as the philosopher-poet (akm
al-shu#ar"). It is an all too familiar mistake for studies to cite a view
presented in one or two of al-Rzs books as being simply his view,
when in fact contrasting views may be readily found in his other,
or later, published works.
Moreover, as was the practice of many other authors, al-Rz
will often treat a problem thoroughly in one book, and will then
refer to it in passing in later books, sometimes without mention of
the earlier discussion. In one book, he writes that he will avoid discussing topics that have been discussed elsewhere.14 It is, therefore,
not uncommon for readers of a given work by al-Rz, pre-modern
and modern alike, to conclude that he fails to address a particular
problem, when in fact the author himself is satisfied in having done
so elsewhere.
At this stage, we are able to determine the exact dating of some
of al-Rzs works and to estimate the general, relative chronology
of others. Some texts remain very difficult to date, except sometimes
very tentatively. Internal evidence of dates, such as references made
to other works, may often mislead, since some works underwent
revision by their author years after they were first written. Thus,
e.g. we find references in the Mabith to Shar Kulliyyt al-Qnn and
vice versa, which leads us to conclude either that both were written
in the same period, or that references in at least one of them were
inserted in a later revision.
Al-Rzs earliest works are strictly classical Ash#ar in style and
content.15 This is most evident in Ul al-n and, to a slightly lesser

14

Kamliyya, (Ar.), 88; (Per.), 114.

See M. al-Zarkn, Fakhr al-Dn al-Rz, 61926; A. Shihadeh, From alGhazl to al-Rz, 163.
15

introduction

degree, the Ishra. In the later Nihyat al-#uql, he introduces Aristotelian logic into kalm, but still proclaims, in line with classical
Ash#arism, that the purpose of his theological enquiry is to defend
the orthodox creed. Later, he wholeheartedly embraces the growing
eclecticism of his milieu, and starts to write works on falsafa, logic,
medicine and the occult. At this stage, he writes the Mabith and
the Mulakhkha, which he describes as falsaf, or ikm, works, in
contrast to his kalm works. He then returns to writing in the latter
genre, gradually synthesising kalm and falsafa.16
The following is a chronological list, with brief descriptions, of
some of the most important works used in the present study. Most of
the internal evidence for the dating has been omitted; a systematic
evaluation thereof will go well beyond the scope of this bibliographic
introduction.
[Kitb f Ul al-dn, #aq"id ahl al-sunna]. The original title of this
unpublished volume is uncertain; hence, it will be referred to as Ul
al-dn.17 Given its classical Ash#ar content and approach, it appears
to be the earliest known theological book authored by al-Rz. He
demonstrates great familiarity with the works of al-Ash#ar (d. 324/
936) and the main proponents of his school, to whom he refers as
our masters (a"immatun), including al-Bqilln (d. 403/1013), his
student Ab Ja#far al-Simnn (d. 444/1052), Ab Isq al-Isfar"n
(d. 418/1027) and al-Juwayn (d. 478/1085).
Al-Ishra f #ilm al-kalm.18 This book represents al-Rzs earliest,
highly Ash#ar stage, but is slightly later than Ul al-dn.19
Al-Mal f #ilm al-ul. Completed in 576/1180,20 this is al-Rzs
most important, and highly influential, work on ul al-fiqh.
Nihyat al-#uql f diryat al-ul. Still unpublished, this large kalm

16

Shihadeh, From al-Ghazl to al-Rz, 164 ff.

It could be Tahdhb al-dal"il f #uyn al-mas"il, mentioned in I#tiqdt, 146.
18 The original title is most probably Ishrat al-nur il la"if al-asrr, a kalm
book that al-Rz mentions in I#tiqdt, 146.
19 E.g. al-Ash#ar is referred to as Shaykhun Ab l-\asan raiya Allhu #anhu (e.g.
Ishra, fol. 3b; 36b; 62a).
20 Mal, ha al-#Alwns editorial introduction, 48. The Mal is mentioned
in Nihyat al-#uql, fol. 200b.
17

introduction

work is one of the most influential texts in the history of Islamic

theology. It represents a crucial transitory stage between al-Rzs
earlier Ash#ar thought and his later philosophical theology.21
Al-Sirr al-maktm (f mukhabat al-nujm, or f asrr al-nujm, or f
#ilm al-sir wa-l-alsim wa-l-nujm, etc.). Still an intriguing aspect of
his career, and apparently written at a relatively early stage thereof,22
this book discusses the theory and practice of magic. It may give
credence to the report that he spent some time in his youth experimenting with alchemy.23
Al-Mabith al-mashriqiyya. The main influences on what appears
to be the earliest of al-Rzs surviving falsaf books are Ibn Sn (d.
429/1037), Ab l-Barakt al-Baghdd (d. 560/11645), and Ab
Bakr al-Rz (d. 313/925). This work does not represent al-Rzs
philosophy in its maturity, but contains many views that differ considerably from his later and earlier views.24
Al-Mulakhkha f l-ikma wa-l-maniq. Al-Rz seems to have revised
this falsaf work in 579/1183.25 It has many similarities to the Mabith,
but is generally more concise (hence its title) and also includes a
section on logic. It shows greater consistency and independence
from Ibn Sn than the Mabith.
Shar al-Kulliyyt min kitb al-Qnn. A commentary on the theoretical introduction of Ibn Sns Qnn, on medicine, probably written
around 580/1184.26
Kitb al-Nafs wa-l-r. Alternatively entitled Kitb f #ilm al-akhlq,
this book has two parts: the first on the theory of the science of
character, the second on practical ethics in detail. Its dating is uncertain, but appears relatively early.

21

Shihadeh, From al-Ghazl to al-Rz, 163 ff.

Mentioned in: Mulakhkha, fol. 323a; Shar al-Ishrt, 2, 143; Shar #Uyn alikma, 2, 1934.
23 Ibn al-#Ibr, Trkh, 240.
24 Mentioned in Mulakhkha, fol. 337a.
25 Dated in the authors colophon of MS. 1510, Leiden University Library.
26 Ibn al-#Ibr, Trkh, 240. The Mabith is mentioned in Shar al-Qnn, fol.
22b; 43b. The latter is mentioned in Mabith, 2, 258; 2, 409.
22

introduction

Shar al-Ishrt wa-l-tanbht. Apparently written around 580/1184,27

this work presents a critical commentary on Ibn Sns philosophy.
The logic section is still unpublished.
Munart f bild m war" al-nahr. A collection of autobiographical accounts of several debates on theological, philosophical and
juristic themes, in which al-Rz participated during his travels in
Transoxiana around 582/1186.28
I#tiqdt firaq al-Muslimn wa-l-mushrikn. A short heresiography.
Immediately after the section on falsafa, al-Rz concludes the book
by affirming his orthodoxy.
Muaal afkr al-mutaqaddimn wa-l-muta"akhkhirn... A compendium
of philosophical theology, and one of al-Rzs most influential and
widely studied works.29
Al-Arba#n f ul al-dn. Another of al-Rzs influential kalm works.
Internal evidence suggests that it was written after the Muaal.30
Lubb al-Ishrt. A critical abridgement of Ibn Sns al-Ishrt
wa-l-tanbht, probably written in 597/1201.31
Manqib al-Imm al-Shfi#. Authored in 597/1201,32 this work
is a defence of al-Shfi# and the Shfi# school, apparently against
\anaf critics.
Lawmi# al-bayyint f (tafsr) al-asm" wa-l-ift. A work on divine
names and attributes.33

27 Mentioned in Muaal, 202. Also, both the Mabith and the Mulakhkha are
mentioned in Shar al-Ishrt, 1, 153. It was taught in Bukhara in 582/1186 (see
following note).
28 The date is mentioned in Munart, 32. The Mabith, the Mulakhkha and
Shar al-Ishrt, are mentioned in Munart, 60.
29 Cf. Shihadeh, From al-Ghazl to al-Rz, 1712.
30 The Tafsr is mentioned in Arba#n, 423. This suggests that al-Rz wrote the
Arba#n around 595600/11991204, after starting the Tafsr.
31 Ktib elebi, Kashf al-unn, 1, 94.
32 Manqib, 538.
33 Mentioned in Tafsr, 15, 55; 22, 13.

10

introduction

Al-Jabr wa-l-qadar. This work is dedicated to the problem of destiny

and human choice, and discusses rational, scriptural and traditional
evidence.34 Its contents suggest that it was written at a late stage.
Al-Tafsr al-kabr, or Mafat al-ghayb. This huge commentary on
the Qur"n was started around 595/1199. Chapters 1730 were
authored in 6013/12057.35 Reports that al-Rz did not complete
this work, and that it was completed later by one of his disciples,
appear to be unfounded.36
Ma#lim ul al-dn. This concise work on kalm is perhaps al-Rzs
last work in the genre. It is clearly of late authorship, but whether
it was written before or after the completion of the Tafsr is unclear.
Al-afad writes that it is the last of al-Rzs smaller books.37
Ma#lim ul al-fiqh. A work on ul al-fiqh, which is shorter than
the Mal. The two Ma#lim works appear to be parts of a larger
series.
Asrr al-tanzl wa-anwr al-ta"wl. An unfinished work which examines theological themes in the light of Qur"nic statements. It appears
to have been written after Al-Tafsr al-kabr, and is sometimes referred
to in later sources as the Smaller Commentary on the Qur"n (Al-Tafsr
al-aghr).
Al-Malib al-#liya min al-#ilm al-ilh. One of the lengthiest of alRzs philosophical and theological works, and in many ways the
most interesting. Books 12 were finished in 603/1207, soon after
the completion of the Tafsr. More than a year later, in 605/12089,
books 37 were apparently finished over a period of 5 months,
34 Although published as book 9 of the Malib, there is much evidence to show
that this is a separate book. Its style differs from that of the Malib. It is not listed
in its initial plan (Malib, 1, 634). There are references to it elsewhere (e.g. Shar
#Uyn al-ikma, 3, 96; biographers list a work entitled al-Qa" wa-l-qadar: al-Qif,
Akhbr, 192; Ibn Ab Uaybi#a, #Uyn, 3, 44; al-afad, Wf, 4, 255). Also, many
manuscripts of the Malib do not include it.
35 Tafsr, 9, 127. For detailed dating of various parts of this work, see Michel
Lagarde, Index, 517.
36 Cf. Jomier, Maft al-Ghayb, 25390; Qur"nic Commentary, 467;
Lagarde, Index, 57 ff.
37 Al-afad, Wf, 4, 258.

introduction

11

whereas book 8 seems to be unfinished and is undated.38 Yet the

latter date is puzzling, since Dhamm ladhdht al-duny (below) is dated
earlier, but refers to discussions in books 3 and 4 of the Malib. One
explanation is that books 37 were written in 603/1207, but were
later revised in 605/12089. However, if this is the case, then why
was the Malib not finished before it was revised? The question is
significant, since an earlier dating of the Malib will give greater
weight to Dhamm ladhdht al-duny.
Shar #Uyn al-ikma. This commentary on Ibn Sns #Uyn alikma was apparently authored after the Malib.39
Rislat Dhamm ladhdht al-duny. A short ethical work finished in
604/1208. For its exceptional interest, a critical edition of this text
is included in the appendix, and the work is examined in detail in
Chapter IV below.

38 Dates are given at the end of each book. Book 3 is not dated in the printed
edition, yet the authors colophon in MS Chester Beatty Ar 3114 (fol. 147a) dates
it to 605/1208.
39 The Malib is mentioned in Shar #Uyn al-ikma, 3, 100. The discussion
referred to appears in book 7 of the Malib.

12

introduction

al-rzs theory of action

13

CHAPTER ONE

AL-R$Z^S THEORY OF ACTION

The Historical Background
Al-Rzs earliest, strictly Ash#ar kalm works will form the ideal
starting point for exploring both his classical Ash#ar background
and the direction in which his thought develops therefrom. Conveniently, a complete sketch of the various other positions and trends
in his intellectual background is presented in his later works, where
he will normally begin his enquiry into the nature of human action
by outlining and scrutinising current and preceding opinions.
In Ul al-dn, al-Rz accepts the common kalm division of created
existents into atoms and accidents, maintaining that human power
(qudra), or capability (isti#a), is an accident that subsists in the atoms
of the human body. He argues, as do previous Ash#ars, that power,
being an accident, cannot endure for more than a single moment,
the minimum discrete unit of time.1 God directly creates the power
to perform an act within man, as well as the act itself, which occurs
at his bodily organs. He institutes and preserves the habitual order
(#da) of created things, such that certain types of human action follow uniformly from certain types of human power.2
At this early stage, al-Rz adheres to the classical Ash#ar doctrine of acquisition (kasb), which attempts to establish a link (ta#alluq)
between the power of the human agent and his acts, for the purpose
of affirming responsibility and obligation. In Ash#ar occasionalism,
the causal link between the human agent and his acts is severed by
the doctrine that only Gods pre-eternal power (qudra qadma) brings
substances and accidents into existence, whereas human power, being
temporally originated (ditha), cannot produce any effects. Yet, in the
doctrine of acquisition, man is said to acquire the act that occurs

1
2

Ul al-dn, fol. 220. Cf. Harry Wolfson, Philosophy of the Kalam, 522 ff.
Cf. passage from the Mal, p. 100 infra.

14

chapter one

at his limbs by virtue of possessing the power that relates to it.3

The mutakallimn debated whether power exists before the act, or
simultaneously with it. In the atomist physics of classical kalm, if
an immobile object is set into motion, it will be immobile at time t1,
and mobile at time t2= t1+t, where t is the smallest unit of time.
Now, if the object is moved by a human being, will the human power
involved in this change exist at t1 or t2? The Mu#tazila were almost
unanimous on the former: the effect (motion) occurs immediately
after the cause (human power). Man, thereby, produces his acts.
By contrast, Ash#ars (including the early al-Rz) maintained that
power and action are simultaneous (al-isti#a ma#a l-fi#l).4 Otherwise,
if power exists at t1, but not at t2, when the act takes place, there
will no longer be a cause to produce the effectwhich, they argued,
is inconceivable.
The same analysis of human action can be found in the Muaal,
which al-Rz wrote at a fairly later stage:
Power is an accident; so it does not last [for more than one moment].
If it precedes the act, it will be impossible [for the agent] to be capable
of acting. For, at the moment that power exists, the act will be nonexistent; and continual non-existence cannot possibly be an object of
power (maqdr). Moreover, at the moment action occurs, there will
be no power.5

The classical mutakallimn also debated whether a particular instance

of human power may relate to opposite acts, or a single act only.
Al-Rz writes in Ul al-dn:
Temporally originated power relates to one object of power (maqdr)
[i.e. one act] only. According to the Mu#tazila, it relates to opposites.
Moreover, according to most of them, it relates to different [objects
of power], i.e. those that are not opposed. They say, Temporally
originated power relates to an infinite number of objects of power at
successive moments.6

For Ash#ars, only one, specific act could follow from a specific
instance of power. The accident of the power to move my arm will
exist in my arm at the time of motion and will relate to that specific

3 Ul al-dn, fol. 231. On the doctrine of acquisition, see Daniel Gimaret, Thories
de lacte humain, 69 ff.
4 Ul al-dn, fol. 222; 225.
5 Muaal, 253.
6 Ul al-dn, fol. 227.

al-rzs theory of action

15

act only; it cannot relate to the omission of this movement, or to a

different act. Contrarily, the Mu#tazila argue that this view leaves
no room for free choice in the agent.
The early al-Rz also rejects the Mu#tazil argument that the
occurrence of the agents act in accordance with his intention (qad),
or motive (d#), proves that it is caused by it. This claim, he objects,
is undermined by several lines of argument. For example, many an
act will occur contrary to, or without, the agents intention, such as
the acts of the sleeper and the unaware. Moreover, if one accepts
that God is the creator of human acts, the above Mu#tazil contention will imply that God will be necessitated to create the human
act in accordance with the human motive.7 A precedent to al-Rzs
rejection of the link between human action and motivation can
already be found in al-Shahrastns (d. 547/1153) Nihyat al-aqdm,
which belongs to the classical Ash#ar tradition.8 We will return to
this problem below.
The classical Ash#ar theory reproduced in al-Rzs earliest works
continues to be a major influence on him, although his later theory of
human action will become highly eclectic. In later works, he provides
a fuller account of other contemporary theories of action, elements
of which he will incorporate. He lists the following four positions.9
(1) Al-Ash#ar and most Ash#ars, including al-Bqilln and Ibn
Frak (d. 404/1015), maintain that God is the only effecter (mu"aththir)
of human action. Al-Rz notes that whereas al-Ash#ar denies that
human power has any effect in human action, al-Bqilln holds
that the acts attribute of being obedience (#a) or disobedience
(ma#iya) depends on the human agents power, though the act itself
is created by God.10
(2) According to the Ash#ar theologian, Ab Isq al-Isfar"n,
human action is produced by the combination of divine power and
human power. He reportedly argued that though human power,
in contrast to divine power, cannot produce acts independently,
it may effect with assistance (mu#n). Therefore, when divine

Ul al-dn, fol. 2034.

Al-Shahrastn, Nihyat al-aqdm, 803 (published as Nihyat al-iqdm).
9 Jabr, 913; Arba#n, 2278; Muaal, 455; Barhn, 1, 2167; Tafsr, 4, 878.
10 Cf. al-Bqilln, Tamhd, 286. Cf. D. Gimaret, Thories de lacte humain,
92 ff.
8

16

chapter one

power combines to human power, the latter becomes effective.11

(3) Another theory is that action occurs necessarily when power
and motive combine in the agent. The majority of the falsifa,12 Ab
l-\usayn al-Bar the Mu#tazil, and al-Juwayn, are variably given
as the main proponents of this view.13 Al-Rz discusses whether the
combination of power and motivation effects action, or only constitutes a condition for it. The latter view is supported by the falsifa,
who hold that the Originator of the world is a necessitating cause
and that this combination produces preparedness (isti#dd), which
allows the Active Intellect to bring the act into being. It is also a
view held by those who maintain that God is a voluntary agent, who
directly creates power, motivation and action in humans, without any
intermediation.14 Al-Rz adds that he accepts the last view, though
he will also attempt to synthesise both positions.
(4) Most Mu#tazils maintain that humans act with autonomy
(istiqll) and choice (ikhtiyr), arguing that this is known through
reflection (naar), or inference (istidll). However, according to Ab
l-\usayn al-Bar and his school-member Mamd Ibn al-Malim
al-Khuwrizm (d. 536/1141), we have immediate (arr) knowledge
that we produce our acts.15 Al-Rz objects that this contention
contradicts Ab l-\usayns view that human action is determined
by the combination of power and motive. He writes:
I am extremely puzzled by the way he upholds, simultaneously, both
this view and the view that the act depends on (mawqf #al) the motive.
[The latter] amounts to an extreme determinism (ghuluww f l-jabr); so
how would he reconcile with it such an extreme qadarism!16

According to this division, human action is brought into being by

divine power alone, human power alone, the combination of both, or
by divine power, with the combination of human power and motivation as a precondition. In what follows, we will see how each theory,
to the exception of al-Isfar"ns, contributes to shaping al-Rzs

11

Cf. al-Shahrastn, Nihyat al-aqdm, 87; D. Gimaret, Thories de lacte humain,

1189. According to al-^j (Mawqif, 8, 147), this accords with al-Isfar"ns view
that one effect can be the product of two separate causes.
12 Cf. Salmn al-Budr, Naariyyat Ibn Sn f l-fi#l, 65 ff.
13 In Tafsr, 4, 88, al-Juwayns work al-#Aqda al-Nimiyya is specified.
14 Jabr, 112.
15 Jabr, 12; Arba#n, 228.
16 Arba#n, 228. cf. Jabr, 228; Barhn, 1, 2167.

al-rzs theory of action

17

later theory of action, which he develops as he emerges gradually

from his classical Ash#arism towards a highly eclectic stance.

The Theory of Human Action

Al-Rzs later theory of human action has two main salient themes.
On the one hand, it has a concern in cosmogony, specifically in
the question of whether human acts come into being by human
or divine power, this being the cardinal problem in classical kalm
discussions of human action. The so-called problem of the creation
of acts (khalq al-af #l) will be examined briefly in the last section of
the present chapter. On the other hand, al-Rzs theory of action
has a psychological concern (more pertinent to his ethics), primarily
with respect to the nature and mechanism of human choice and
motivation. Related to this is the question of whether the human
agent is determined (mujbar), or has free choice (mukhtr).
The most important of al-Rzs later discussions of human action
can be found in the Arba#n, the Ma#lim, the Tafsr and, most importantly, the Malib.17 He also dedicates Kitb al-Jabr wa-l-qadar to
discussing destiny and human choice.
All these discussions revolve around a simple, key contention that
is central to al-Rzs later theory of human action: that action is
produced when human power and motivation combine. Once this
combination occurs, it will act as a natural cause, which necessitates its effect. In what follows, the two elements will be examined
separately.
Power
Debate over the nature of human power (qudra) has a long history
before al-Rz. The Mu#tazila differed on whether human capability is mere health (ia) and bodily fitness (salma), or an extra
accident that subsists in the atoms of the human body.18 Classical
Ash#ars took the latter view. Al-Ash#ar maintained that health is
the soundness of constitution, fitness is the absence of defects and

17
18

Arba#n, 12254; 22746; Ma#lim, 7885; Malib, 3, 973.

Al-Ash#ar, Maqlt, 229; Ab Rashd, Mas"il, 241 ff.

18

chapter one

obstructions, and power is an aspect that is additional to that,19

and that capability is the same as power, and is an accident that
cannot subsist by itself, [but] subsists in the live atom (jawhar ayy).20
Accordingly, the difference between the healthy person and the disabled one is the existence of the attribute of power in the former
and its absence in the latter.
In his later works, most notably the Ma#lim, al-Rz rejects this
common Ash#ar notion, that human power is an accident, directing
his criticism at al-Ash#ar specifically. Human power, he will argue, is
not an accident that necessarily precedes a particular act, but is the
fitness of the physical organs and the balance of the humours.21
As to whether a particular instant of power may produce only
one act, or different types of action, he writes:
Ab l-\asan al-Ash#ar says that power does not relate to opposites.
In my view, if [power] refers to the balance of humours and the fitness of organs, it will relate to both performance and omissionthis
is known immediately. However, if it means that unless a decisive
and preponderating motive combines to power, it will not produce
the effect, and that when this combination occurs it will not relate to
opposites, then this indeed is true.22

These two interpretations of the statementneither of which is

intended by al-Ash#arfollow from al-Rzs abandoning the definition of power as an accident. Though, by power, he normally
refers to the balance of the humours and physical fitness, he also
uses the term in the second sense above, i.e. as referring to active
power, which is the combination of potential power and motivation. As such, al-Rz will preserve the traditional Ash#ar doctrinal
formula, Capability relates to one act only, often presenting it as
the accepted Sunni position, despite his untraditional re-interpretation thereof.
Also in the Ma#lim, he cites al-Ash#ars view, related to his definition of power, that capability is simultaneous with action.23 Power,
defined as physical fitness and balance in humours, is not an accident
that appears in the body at one moment and disappears at its end.

19
20
21
22
23

Ibn Frak, Mujarrad, 117.

Ibn Frak, Mujarrad, 107.
Ma#lim, 812; cf. Jabr, 40.
Ma#lim, 834.
Ma#lim, 83.

al-rzs theory of action

19

Rather, it may exist continuously, before, during and after the act,
though it acts as a cause only when it combines with motivation.
The act, therefore, is produced by the combination of power and
motivation, rather than the mere presence of power. When this combination occurs a complete cause (mu"aththir tmm) emerges, which
produces the effect at exactly the same instant.24 Al-Rz concludes
that if capability denotes power, it will indeed exist before action;
however, if it denotes the combination of power and motive, it will
be simultaneous with (#inda) action. As such, he affirms the traditional
Ash#ar formula, Capability and action occur simultaneously.
Al-Rzs analysis of human action is in agreement with the classical
Ash#ar stance with respect to its atomism and occasionalism, though,
essentially, it constitutes a very different theory. On the other hand, it
is mainly on account of its atomism and occasionalism that al-Rzs
analysis of action contrasts with that of Ibn Sn, with which it is
otherwise in general agreement. In the Mabith, Ibn Sn, for whom
time is a continuum, is cited criticising those who claim that power
is simultaneous with action: they will have to accept that someone
who is seated is incapable of standing up (i.e. it is not possible, in his
constitution, for him to stand up), unless he does stand up; so how
could he stand up!25 In response, al-Rz argues that potentiality
(quwwa) is the principle of change and exists simultaneously with the
change it effects, whereas power is only a constituent of potentiality
and exists before the act. He adds, If it is possible to interpret the
claims of those people in the way we explain, what need is there to
attack them and to deride their claims!26 Again, however, al-Rz
must be aware that he does not merely introduce an interpretation
of the Ash#ar claim, but a fundamentally different theory.

24

Ma#lim, 83.
Ibn Sn (Shif", Ilhiyyt, 1, 1767) attributes this view to some ancients,
including Ghrqw (sic.) and some people who came very long after him, probably in reference to Ash#ars. The reference to the 4th3rd century BC Megarian
school (Ghrqn) is taken from Aristotles Metaphysics (IX.3). Ibn Rushd (Tafsr
m ba#d al-ab#a, 2, 11246) comments: Nowadays, among [the followers of] our
religion, this view is upheld by the Ash#ars, and it is a view that contradicts human
nature, with respect to both beliefs and acts. (See Megarian School, REP).
26 Mabith, 1, 3823; cf. Mulakhkha, fol. 159b160a; Ibn Sn, Shif", Ilhiyyt,
1, 1767.
25

20

chapter one

Motivation
According to al-Rz, the production of human action requires the
combination of power and a preponderator (murajji); power, on its
own, is passive potentiality that will produce an effect only when
prompted by an external factor.27 Since human acts are contingent, and since a nonexistent contingent will require a cause for
its existence to preponderate (tarajjaa) over its non-existence, the
occurrence of the act will be inconceivable without such a cause,
and necessitated by it. Al-Rz rejects the view that the cause will
merely make acting more probable (awl) than omission; for the two
probabilities will be either equivalent, in which case no change will
occur, or not equivalent, in which case the preponderant possibility
will prevail.28 As soon as acting preponderates, the act will occur
instantly.29 Al-Rz summarises the argument in the Mulakhkha:
On that the Acts of Servants are [Subject to] Divine Determination. Human
capability either has the potentiality to produce opposite [acts], or not.
The latter [view] amounts to the affirmation of compulsion (jabr).
In the former [view], the production of one of the two [possible]
objects of power (maqdr) from [human capacity], to the exclusion of
the other, will either depend on a preponderator, or not. [If not,] it
would follow that one of the two possible alternatives preponderates
over the other without a preponderator.
[If acting depends on a preponderator,] that preponderator will be
produced by either the servant (and the above disjunction will re-apply),
or God. Then, the production of that effect, when that preponderator occurs, is either necessary, or not. The latter is inconceivable; for
otherwise [the effect] would leave the state of equivalence, without
reaching the state of necessitationwhich we have refuted elsewhere.
The former amounts to compulsion.30

Al-Rz repeatedly argues that to deny the principle of preponderation will be tantamount to denying the existence of the Creator,
which is proven mainly on the basis of this principle.31
The preponderator in question is identified as will (irda, mash"a),
which acts as a decisive will (irda jzima), or inclination (mayl jzim),

27
28
29
30
31

Mabith, 1, 383; Mulakhkha, fol. 160a.

Mabith, 2, 5167; cf. Jabr, 334.
Jabr, 24.
Mulakhkha, fol. 349b.
Jabr, 16; 22; cf. Tafsr, 13, 187.

al-rzs theory of action

21

since it will necessitate its effect.32 This will originates from a final
cause within the agent, which is his intention, or objective. Al-Rz
writes, commenting on Ibn Sn:
How excellent is the Shaykhs saying that the final cause is an active
cause for the causality of the active cause (al-#illa al-gh"iyya #illa f#iliyya
li-#illiyyat al-#illa al-f#iliyya)! For the living being may be able to move
right or left. Therefore, before the preponderance of one over the
other, it will be an agent in potentiality. If it imagines benefit in either
of the two movements, that imagination will become an efficient cause
for that potentiality becoming an actual cause for one of the two
movements.33

In al-Rzs later works, the notion of motive (d#, d#iya) becomes

central to his discussions of human action; and he devotes a lengthy
section to this topic in the Malib.34 This marks the extent of his
departure under Mu#tazil and falsaf influence from his early Ash#ar
stance in Ul al-dn, where, as we saw, he categorically rejects the
Mu#tazil notion that motivation plays a role in action.
Al-Rz defines motive as the presence, in the agent, of the
knowledge, conviction (or belief, i#tiqd) or presumption (ann) that
he has a preponderant advantage (malaa rjia) in performing a
particular act, which thereby brings about, in the agents heart, a
decisive inclination to perform that particular act.35 Conversely, a
motive can be the knowledge, conviction or presumption that the
agent has a preponderant disadvantage (mafsada) in an act, resulting
in a decisive aversion (nafra) from performing it. The latter type is
also termed deterrent (rif).36
Knowledge, al-Rz maintains, is a stronger motive than belief,
since the latter, no matter how firm, will always remain liable to
doubt.37 Although belief on the basis of uncritical imitation (taqld) will
often constitute a stronger motivation in some agents than knowledge
in others, this is due to the interference of other motives, such as

32

Mabith, 2, 5178; 3, 13; Muaal, 257; Jabr, 25; 41.

Lubb, 856; cf. Shar al-Ishrt, 1, 1924; Ibn Sn, Shif", Ishrt, 3, 301.
34 Malib, 3, 773. #Abd al-Jabbr previously dedicated a book to the subject of
motivation, which is not known to have survived (see Ibn al-Malim, Mu#tamad,
510).
35 Malib, 3, 9.
36 Malib, 3, 13.
37 Malib, 3, 146.
33

22

chapter one

habit, social custom, and an agents expectation of praise or benefit

in consequence to his acts. Also, each of these three types of cognitive states (knowledge, belief and presumption) includes motives
that vary widely in strength; hence, the presumption that relies on a
single informants account of an event generally constitutes a weaker
motive than that which relies on multiple accounts, which, in turn,
constitutes a weaker motive than witnessing the event oneself.
Al-Rz distinguishes between objective (ghara), which is the final
cause of an act, i.e. the desired benefit itself, and motive, which is
the knowledge or belief that relates to it.38 This is expanded upon
in a section in the Malib entitled, That nothing but conviction
and presumption (sic.) has an effect in motivating action, where he
addresses some widespread notions.39 For instance, it is said that
ones motive for performing an act is its being beneficial (manfa#a) and
advantageous (malaa) in itself, while the motive for omitting an act
is its being harmful (maarra) and disadvantageous (mafsada) in itself.
Such statements, he argues, should be treated as metaphorical.
The acts being beneficial and good is never a cause for the agents
performance thereof. Rather, what effects agency (f#iliyya) is the
agents knowledge that the act in question involves such goodness
and benefit.40

Similarly, it is not Gods promise and threat as such that motivate a

believers acts, but the knowledge of what afterlife advantages and
disadvantages may follow in consequence.
So does knowledge, al-Rz asks, motivate action directly, or
through the intermediation of will (irda), or inclination (mayl) (aversion
(karhiya, nafra) in the case of a deterrent)?41 According to the latter
view, which he accepts, knowledge will still be the sole determinant
of motivation, and may be referred to as the motive in this sense.
This view will explain the process of preponderation between multiple
convictions existing simultaneously in the mind, which produce only
one will, or inclination, for action. Al-Rz gives the example of one
who leaves his home to visit a friend, but then remembers that he
has to attend to something important at home, in which case he will

38
39
40
41

Malib,
Malib,
Malib,
Malib,

3,
3,
3,
3,

10.
178.
18.
1920.

al-rzs theory of action

23

have two conflicting motives. If he finds himself convinced that the

benefits that would follow from visiting the friend and returning home
are equal, he will incline to neither choice of action and will remain
in his place, undecided and lacking will to act. Then, if it occurs in
his mind (waqa#a f khirih) that one of the two benefits preponderates
over the other, he will move immediately in its direction.
According to #Abd al-Jabbr, a motive is the knowledge, conviction
or presumption of what is either morally good in itself, or beneficial
to the agent, though he accepts that most human action is directed
towards the latter.42 By contrast, al-Rz maintains that the consideration of personal benefit is the only ultimate basis for all human
motivation. He argues that it is known immediately (bi-l-arra) that
human nature (ab#, fira) is undoubtedly inclined towards something
that it finds desirable in itself, and repelled from something else
that it finds undesirable in itself.43 All other things are sought or
avoided ultimately for the sake of one or both of these two things.
He adds,
If we then contemplate and return to ourselves, we will know that
the thing that is sought in itself is one of two things, either pleasure
(ladhdha) or joy (surr), and that the thing that is avoided in itself is
either pain or grief (ghamm). All that the occurrence of which leads to
pleasure and joy is sought for the sake of another, whereas all that
the occurrence of which leads to pain and grief is avoided because of
another. The words good (khayr) and advantage (malaa) refer to
all that is sought, whether in itself or for the sake of another; while
the words evil (sharr) and disadvantage (mafsada) refer to all that is
avoided, whether in itself or because of another.44

Seeking pleasure and avoiding pain are the two simple, primal instincts
which underlie all human motivation.
These two primal instincts constitute the rudimentary elements
for the calculations involved in more complex motives, as follows.
Harm in general, which is avoided in itself, is of three main types:
(a) real harm, which is the immediate experience of pain, (b) the
negation of benefit, and (c) the negation of what prevents harm (daf #
dfi# al-arar). Benefit in general, which is sought for its own sake, is

42
43
44

E.g. #Abd al-Jabbr, Mughn, 6/1, 196.

Malib, 3, 21; Jabr, 41.
Malib, 3, 212; cf. Nafs, 1920; Ma#lim, 86.

24

chapter one

also of three types: (a) real benefit, which is the immediate experience of pleasure, (b) the prevention of harm, and (c) the prevention
of what impedes benefit. Some of these constitute generally stronger
motives than others; e.g. expecting harm is a stronger motive than
expecting the negation of an equal benefit. Moreover, overall motives
vary in strength according to whether they are based on knowledge,
conviction or presumption.45
The agents mind weighs the various advantages and disadvantages
it conceives within an act. If the agent then believes that the act will
be purely advantageous, he will be compellingly motivated (mulja")
to perform it, and will perform it by necessity. If he finds it purely
disadvantageous, he will omit it.46 However, if he believes that the
act involves both advantageous and disadvantageous aspects, he
will be compellingly motivated to perform or omit it, according to
which of the two choices he conceives to be preponderant. If both
are equal, he will abstain from acting (tawaqquf ).47
All the motives and deterrents that influence the agent at a given
situation will add up and subtract in an often complex calculation
within the agents mind. Equal conflicting motivational quantities
will, as it were, cancel each other out, leaving, as their end product,
only a single quantity that effects action. Al-Rz himself uses mathematical language to describe this process; he writes that if equivalent
opposed motives cancel each other out, the remaining quantity
[lit. number (#adad z"id)] will remain as a pure (khli) motive for
action.48 Such calculative processes often involve multiple stages.
Suppose that an agent, in a given situation, has to choose between
more than one course of action, when he expects that each would

45

Malib, 3, 23 ff.
Malib, 3, 25. Al-Rzs use of the term compelling motivation (ilj") comes
directly from its use by Mu#tazils. #Abd al-Jabbr seems to borrow the term from
Ab Hshim al-Jubb", to denote the particular case of the intentional act, which
the agent finds himself virtually forced to choose (Mughn, 11, 395; 12, 128; cf.
Richard Frank, The Autonomy of the Human Agent in the Teaching of #Abd
al-@abbr, 340; D. Gimaret, Thories de lacte humain, 56 ff.). According to #Abd
al-Jabbr, when the agent is subject to only one sufficiently strong motive, it will
be impossible for him to act in any other way; as such, he will be compellingly
motivated. Mu#tazils, therefore, viewed compelling motivation as an extreme case,
unrepresentative of human action in general. Al-Rz generalises it to include all
human action.
47 Malib, 3, 26; Jabr, 2233.
48 Malib, 3, 26.
46

al-rzs theory of action

25

lead to various advantages and disadvantages. His mind would first

assess each possible course of action individually by calculating the
sums of its expected advantages and disadvantages.49 The second
stage will be to find the final sum of the sums of all conceived courses
of action, which then produces the will that motivates action.
It will be instructive to compare al-Rzs theory of motivation with
its main influence, viz. the views of some previous Baran Mu#tazils.
The most pertinent are #Abd al-Jabbr, Ab l-\usayn al-Bar, and
Ibn al-Malim.50
According to #Abd al-Jabbr, the automatic process of adding
up, described by al-Rz, does not take place. Rather, the agent
may choose between separate motives that influence him at a given
situation and that could only make some choices of action more
preferable to him, or more likely to be performed (awl). Yet, though
motives, according to #Abd al-Jabbr, provide the agent with a sense
of the various possible courses of action at a given situation, they
are neither necessary for, nor effective in, his action.51 Only when
the agent has one sufficiently strong motive will he be compellingly
motivated to act according to it.
Ibn al-Malim criticises #Abd al-Jabbr for the view that the
agent may act without motivation, and asserts that the function
of the motive is akin to that of physical organs, viz. it constitutes
a necessary condition for the production of action by the efficient
causality of power alone.52 While motives make one act more likely
(awl) than another,53 the agent may choose freely among his various
motives. According to Ibn al-Malim, motives do not automatically
produce a single will, or inclination, that then motivates the agent to
act, as al-Rz maintains.54 Rather, when confronted with a number
of motives (which consist of knowledge, convictions or suppositions),
the agent may choose a course of action according to one or more
of them. Ibn al-Malim can thus plausibly contend that a motive

49

E.g. Malib, 3, 345.

The influence of Ab l-\usayn and Ibn al-Malim on al-Rz is noted by
Ibn al-Murta (abaqt, 119).
51 Cf. R. Frank, Autonomy of the Human Agent, 3413.
52 Ibn al-Malim, Mu#tamad, 5104.
53 Ibn al-Malim, Mu#tamad, 514.
54 Ibn al-Malim, Mu#tamad, 240 ff.
50

26

chapter one

makes the occurrence of an act more likely than others, but does not
necessarily determine it.
Al-Rz points out a crucial difference between the theories of
action of Ibn al-Malim and his master Ab l-\usayn: the latter
holds that the act depends (mawqf ) on the motive, whereas the
former holds that the motive does not necessitate the act, but makes
it more likely to occur.55 However, it seems that Ab l-\usayn does
not state explicitly that the motive necessitates the act, as al-Rz
criticises him for shying away from admitting this obvious implication to his position. One may wonder here whether there was indeed
such a disagreement between Ab l-\usayn and Ibn al-Malim,
as al-Rz claims.56 In his work al-Mu#tamad f ul al-dn, Ibn alMalim does not seem to refer to this point of divergence with
Ab l-\usayn; yet he does point out another relevant disagreement
with him of which al-Rz too is aware, viz. that Ab l-\usayn
affirms the role of will in human action. Will, according to him, is
produced by motives; for when man knows that a thing contains a
preponderant benefit, he will find that his self seeks it, and he will
find this seeking (alab) to be as though produced by this knowledge
and following from it.57 The immediate motive for action becomes
will, or inclination in the heart (mayl al-qalb).58 Therefore, since
Ab l-\usayn maintains that action presupposes the presence of
motivation and that the presence of a single motive in the agent
at a given moment will compel him to act in accordance with it, it
should follow (though Ab l-\usayn appears not to admit this) that
the emergence of a single will from multiple partial motives will
constitute a decisive will. Al-Rz argues that Ab l-\usayn will
have to admit that motivation necessitates action. As such, he may
appear justified in accusing him of inconsistency.59
In the Malib, al-Rz includes a section on this view, which he
attributes to Ab l-\usayn, entitled On whether the production of

55

Jabr, 123; Arba#n, 228.

According to Wilferd Madelung (Late Mu#tazila, 254), it is unlikely that
Ibn al-Malim modifies his masters theory fundamentally.
57 Ibn al-Malim, Mu#tamad, 2401; al-Rz mentions the views of both authors
in Nihya, fol. 125a.
58 Ibn al-Malim, Mu#tamad, 251.
59 On Ab l-\usayns theory of action, cf. Madelung, Late Mu#tazila, 2506;
and, recently, Martin McDermott, Abul-\usayn al-Bar on Gods Volition.
56

al-rzs theory of action

27

action depends (tawaqqafa) on motivation or not, and another on that

the production of the act requires a motive.60 The conclusion of the
two sections is that the act occurs if, and only if, its motive occurs.
He then includes another section entitled On demonstrating that,
at the occurrence of motivation, the production of action becomes
necessary, and that no probability remains whatsoever, which is
directed against those who claim that when motivation occurs, the
production of action becomes more likely, but does not reach the
level of necessitation,61 viz. Ibn al-Malim.62
Al-Rz, therefore, maintains that motivation is not only necessary for action, it necessitates action. He addresses some objections
based on observations of apparently unmotivated acts. With respect
to unconscious action, he argues that the act will not occur unless the
agent intends (qaada) it.63 The seemingly purposeless act of moving
ones finger while engrossed in deep thought is in fact purposeful,
since it will involve a slight, though often unconscious, benefit or
negation of harm. For instance, the constant immobility of the finger
may cause tedium, or the movement may be habitual, in which case
immobility will cause unease. Even in the total unconsciousness of
sleep, the sleeper may have imaginations which motivate him to
do some acts, such as turning from side to side and talking. These
imaginations may be dreams, or may be produced by physical pains,
resulting, e.g. by prolonged sleep on one side. Similar explanations
of the action of the sleeper are advanced by Ibn al-Malim in his
objections to #Abd al-Jabbrs contention that man may act without
motivation, and by Ibn Sn in his refutation of Mu#tazil views on
autonomous human choice.64 Al-Rz accepts Ibn Sns distinction
between imagination and the agents awareness of that imagination
and his remembering it.65 It is possible, therefore, that the agent acts
according to an unconscious motive, or a conscious motive that he
later forgets.

60

Malib, 3, 3743; 3, 4553.

Malib, 3, 5560.
62 Arba#n, 228.
63 Malib, 3, 42.
64 Cf. Ibn al-Malim, Mu#tamad, 510-4; Ibn Sn, Al-Qa" wa-l-Qadar, 55;
Madelung, Late Mu#tazila, 251.
65 Malib, 3, 423; Shar al-Ishrt, 1, 1889; Shar #Uyn al-ikma, 3, 43; Mulakhkha, fol. 186a; cf. Ibn Sn, Ishrt, 2, 460.
61

28

chapter one

Also, many Mu#tazils, including #Abd al-Jabbr, will define vain

action as unmotivated action.66 Yet, following a common falsaf view,
al-Rz defines vain action (#abath), not as unmotivated action, but
as non-rational action, which is motivated by the imagination, and
is often unconscious and related to trivial ends.67 Hence, the section
in the Mabith entitled, On demonstrating that there must be an
end (ghya) for vain and fortuitous (juzf) [action].68
Al-Rz also comments on the phenomenon of hesitation (taraddud),
which the Mu#tazila cite as a sign that the agent has free choice.
They claim, he writes, that obligation (taklf) is inconceivable when
the agent is compellingly motivated, and conceivable only when he is
hesitant among a number of conflicting motives (mutaraddid al-daw#
il l-fi#l wa-l-tark).69 He responds by arguing that hesitation is due
to the complexity of mental processes. Motives are in constant and
subtle motion in the agents heart, since convictions are momentary
and quick to disappear and alter.70
Natural causes are constant, continuous and unchanging. The cause
of heating remains forever characterised by that property by which
it produces heating, and never changes ... in contradistinction to voluntary (ikhtiyr) acts. For that which produces movement leftwards71
is a combination of power and the motive for leftward movement.
These motives are quick to alter and change. Thus, if the motive to
turn right occurs in the capable [agent]s heart, that combination will
produce rightward movement. Then that motive disappears quickly
and is replaced by the motive for leftward movement. This combination now produces leftward movement.72

Hesitation, thus, occurs when the agent has a preponderant motive

to perform a certain act at one instant, and a preponderant motive
to abstain, or to perform a different act at the next instant, each

66 In this sense, vain action will be inconceivable in Ibn al-Malims theory

of action (although he does use the term), given his insistence on the impossibility
of unmotivated action (cf. Madelung, Late Mu#tazila, 253), unless he defines vain
action as action that accords, not with the most preponderant (awl) motive the
agent has, but with a much weaker one.
67 Shar al-Ishrt, 1, 1889; Shar #Uyn al-ikma, 3, 413; cf. Ibn Sn, Ta#lqt,
83; 141; Ishrt, 2, 459.
68 Mabith, 1, 5357; cf. Shar #Uyn al-ikma, 3, 423.
69 Malib, 3, 268.
70 Malib, 3, 27; cf. 3, 312.
71 Reading yusr instead of yumn.
72 Jabr, 26; cf. Malib, 3, 278.

al-rzs theory of action

29

constituting a compelling motive. People call one who is generally hesitant and undecided dh badawt, since, at one moment, a
certain act will seem (bad) to him to be of preponderant benefit,
while, at another moment, he will conceive another act to be so; this
indecision may be due to his nature. Also, the agent may perceive
something that may remind him of, or invoke in him, a certain
knowledge or conviction that motivates him, whereas perceiving
something else, one moment later, may invoke a different conviction
in his mind.73 Considerations of benefit and harm are many and
virtually countless; the greater the minds reckoning of them, the
greater the perplexity and confusion.74

Destiny
Al-Rz presents various arguments against Mu#tazil claims of free
human choice, one of the most important of which goes as follows.
The combination of power and motivation either necessitates, or
does not necessitate, the act. The former possibility will lead to a
type of determinism (jabr), since the agents motives cannot rely
on other motives ad infinitum, but should ultimately originate from
motives that God creates. Al-Rz then continues with an argument
ad hominem, viz. that the latter possibility will entail that when power
and motivation combine, action may or may not follow. In other
words, if we imagine that a given situation involving an agent, who
has to make a choice, is repeated a number of absolutely identical
times, the choice that the agent makes will follow sometimes, but
not at all times; i.e. it will occur randomly, by pure chance and
for no reason. This would result in another type of determinism,
whereby human action is determined by pure chance.75
Having demonstrated that the combination of motivation and
power determines the act, al-Rz discusses the only preceding stage
in the production of human action, viz. the emergence of motivation
from cognitive states. By arguing that both stages are deterministic,
he proves that human action as a whole is determined. He writes:

73
74
75

Jabr, 43.
Malib, 3, 28.
Jabr, 134.

30

chapter one
Man finds in himself that unless he wants to act he will not be able to
act, and unless he wills to abstain he will not be able to abstain. That
will is not due to another will; for, otherwise, this would entail infinite
regress. It is thus certain that those wills go back to a necessary will
that occurs in the heart without will. And if that decisive will occurs
in the heart, the occurrence of the act will undoubtedly be necessary.
Therefore, neither the occurrence of will in the heart is of the persons
[own doing], nor is the occurrence of the act after the occurrence of
will in his heart of the persons [own doing]. All is thus from God.
And man is compelled under the guise of a voluntary agent.76

In one place in the Malib, al-Rz is confronted with the question

of what determines the motive, i.e. what makes a particular motive
at one moment preponderant over other motives.77 He responds that
God brings motives into being; and it is thus His will that makes
one motive preponderant over others. Elsewhere, he qualifies this
by dividing motives that appear in the agent into motives that the
agent himself causes to occur (bi-q# al-#abd), and motives that God
initiates (ibtid"an) in his heart.78 The former type is possible when
the agent wants to remove a particular motive, or inclination, from
his heart, and then strives to do so. If anyone examines himself, he
will find in himself this ability to change or eliminate motivesalRz treats this as an indisputable, self-evident truth. Motives that are
brought into being in the heart immediately by God include motives
that motivate the agent to change other motives; otherwise, if the
appearance or change of each motive required another motive, the
chain of motives would continue ad infinitum.
Al-Rz also argues that motives stem from the agents cognitive
states, which are determined by both internal and external factors.79

76

Manqib al-Imm al-Shfi#, 122.

Jabr, 289.
78 Malib, 3, 612. Cf. Arba#n, 127: Inclinations and desires go back to an
inclination and a motivation that occurs in the heart, either by Gods creation or
by a heavenly cause (sabab samw). A thirsty persons choice between two identical glasses of water placed in front of him relies on these heavenly causes (mustanid
il l-asbb al-falakiyya).
The devil may also play a role in human motivation, by reminding the agent
of the pleasures associated with a particular disobedience that he is otherwise forgetful of. An angel may remind him of the happiness associated with a particular
obedience. Both, however, depend on the psychological preparedness of the agent
for such influences (Malib, 7, 32931).
79 Jabr, 3546.
77

al-rzs theory of action

31

The main internal ones are: (a) variations in the essences of human
souls;80 (b) variations in temperance; and (c) physiognomic variations. The three main external factors are: (d) habituation; (e) the
expectation of certain benefits or harms from adhering to certain
beliefs; and (f) theoretical training. These six factors will determine
the agents character, while accidental (ittifq) external circumstances,
perceived through the senses, will act as more direct factors in the
production of motives.81 All these are factors that the agent will
have no control over.
In Kitb al-Jabr, we find a lengthy section containing proofs for
the view that man has no power whatsoever over the contents of his
knowledge.82 Even religious belief and disbelief are products, not of
the agents choice, but purely of Gods creative power. A number
of arguments are advanced.
It is first argued that all acquired (muktasab) knowledge will depend
on self-evident (badh) knowledge that the mind knows immediately
and spontaneously, not by choice. So, at a given moment, all that
is necessary for self-evident knowledge to entail acquired knowledge either exists within the agent, or does not exist. If it exists,
that acquired knowledge will come into being immediately, not by
choice. If not all that is necessary exists, something will still be
needed for the appearance of acquired knowledge. That something
cannot be self-evident knowledge (which is already assumed). Nor
can it be acquired knowledge, in the case of the first bit of acquired
knowledge. Al-Rz seems to imply that it can only be divine, rather
than human, will (since an agents will depends on his knowledge). It
follows, he concludes, that self-evident knowledge is not produced by
human will, nor are the first, second and following bits of acquired
knowledge.
Al-Rzs second proof is as follows. Knowledge is either conception (taawwur) or assertion (tadq), the former being apprehension
(idrk) without judgement (ukm), and the latter being, according to
al-Rz, apprehension with a judgement of fact. He first argues that
conceptions are not acquired. For if I try to acquire a conception, I
will either have awareness (shu#r) of the essence to be conceptualised,

80
81
82

This is discussed p. 11820 infra.

Jabr, 434.
Jabr, 10110; cf. Arba#n, 2357; Tafsr, 24, 179.

32

chapter one

or not. If I have awareness, I will already have conceptualised it.

If not, my mind will be heedless of it, and thus cannot seek it. If
it is objected that the mind may be aware of only some aspects of
the essence, but not of it completely, the same argument will apply,
though awareness of essence will be substituted with awareness of
the unknown aspect of the essence.
Al-Rz also argues that if we examine ourselves, we will find that
conceptions have to be: (a) derived from sense perception, or (b)
based on self-evident conceptions, such as our knowledge of pleasure
and pain, or (c) constructed by the mind or the imagination from
the first two types, which are not acquired. Conceiving something
beyond these will be impossible. Therefore, conceptions cannot be
acquired.83
Having demonstrated that conceptions are not acquired, al-Rz
proceeds to prove that no knowledge of assertion can be acquired.84
Each statement of assertion, he argues, will require two conceptions,
viz. of subject and predicate. Now, the presence of these elements in
the mind is either sufficient to produce an assertion, which will thus
be self-evident (such as knowing the concepts of one and half of
two); or not, in which case the resultant assertion will be discursive
(naar) knowledge (such as knowing the concepts of world and
temporal; since the presence of both in the mind is insufficient
for concluding that the world is temporal). In the latter case, if all
that is necessary, for self-evident knowledge to entail some discursive
knowledge, is present in the mind, that discursive knowledge will follow necessarily, not by choice. The argument then continues along
the same lines as the first proof.
Al-Rz provides three more proofs in support of his contention
that all the knowledge and ignorance that occur in peoples hearts
and minds are from God, and are brought into being by God.85
In showing that the agent has no control over, or choice of, his
knowledge, al-Rz covers all parts of the theory of human action in
affirmation of absolute determinism: the acquisition of knowledge,
the emergence of motivation from knowledge, and the emergence of

83
84
85

Jabr, 105.
Jabr, 1057.
Jabr, 108.

al-rzs theory of action

33

action from the combination of motivation and power.86 For #Abd

al-Jabbr, by contrast, knowledge that derives from reflection (naar)
is generally not determined.87
Having argued for the determination of human acts on largely
logical and metaphysical grounds, al-Rz is faced with the accusation that he treats human motives as natural causes. He replies
that motives are distinct in two main ways. First, natural causes are
continuous and uniform, whereas voluntary acts are in constant
motion and change, and their effects are thus not uniform or always
predictable.88 Second, natural causes do not have awareness, knowledge or comprehension of their effects, unlike voluntary agents who
are aware of their acts and whose knowledge of a possible, prospective act may itself be part of the cause that produces it. We, therefore, describe the act as being voluntary (ikhtiyr), and the agent as
being a voluntary agent (mukhtr), because the act will occur only
when the agent conceives it as being good (khayr). This etymological
analysis removes the emphasis in the word ikhtiyr from choosing,
or choice, to the other related sense of the root kh-y-r, viz. finding
something to be good or better than others.89
Hence, for instance, to the objection, If man were destined, sending prophets, obligating men by commands and prohibitions, and
afterlife punishment and reward would all be pointless, he responds
that both peoples obedience and disobedience to divine command
and their afterlife consequences are included in destiny. There is no
end to, or culmination of, destiny, at which point the injustices of
this world are resolved, and justice (as conceived by the Mu#tazila)
is served. Rather, as one may be destined to have a bad day today
and another bad day tomorrow, which will only add to the badness
of today, let alone compensate for it, one may be destined to have
an infinite number of bad days in the afterlife.
In their criticism of such a deterministic view the Mu#tazila had

86
87

Cf. Muaal, 1367; Arba#n, 3302.

See D. Gimaret, Thories de lacte humain, 48 ff.; Madelung, Late Mu#tazila,

246.
88 Jabr, 26; cf. p. 28 supra. This same point is made by Ibn Sn; cf. S. al-Budr,
Naariyyat Ibn Sn f l-fi#l, 67.
89 Cf. Lawmi#, 360; Malib, 7, 329. Ibn Sn (Ta#lqt, 523) gives a similar
distinction between natural causes and voluntary ones, and adds that human acts
will not be voluntary in reality.

34

chapter one

yet another strong objection, which they based on analogy (qiys), or

judging the unobservable in accordance with the observable (radd
al-gh"ib il l-shhid). If human action occurs only with the combination of power and motivation, the same should be true in the case of
divine action. Consequently, if all that is required for Gods agency
(f#iliyya) existed from pre-eternity, the world would have been preeternal. Otherwise, if not all that is required for Gods agency existed
from pre-eternity, some of that would have come into being at some
point. That would have required yet another cause for it to come into
being; that cause would have required another cause; and the chain
would continue ad infinitum. The argument concludes that al-Rzs
determinism will have to apply universally, to both the observable
(shhid) and the unobservable (gh"ib),90 and will inevitably lead to
the view that the world is eternal and that God is a necessitating
cause (#illa mjiba), rather than a voluntary agent. This, al-Rz states,
is the strongest Mu#tazil argument in this context.91
His immediate response is that rejecting determinism and affirming free human choice will lead to a more heinous consequence,
viz. denying the very existence of the Creator. For that position will
undermine the main proof for His existence, which is based on the
contingency of this world and its need for a preponderator to bring
it into being. However, al-Rz recognises that this counterargument
is unsatisfactory and he promises to return to this topic later in Kitb
al-Jabr to elucidate the difference between the observable and the
unobservable with respect to this issue.
Elsewhere in this work, he replies to the same objection by proposing a distinction between human will, which is temporally originated
(dith) and thus in need for an originator (mudith), and divine will,
which is pre-eternal and uncaused.92 But this does not solve the
problem; for the Mu#tazila may argue that a pre-eternal divine will
that is totally uncaused will necessitate a pre-eternal creation. I have
not found other places in Kitb al-Jabr where the main Mu#tazil
objection is addressed.
Indeed, this implication of the falsaf influence that al-Rz intro-

90 The observable world (#lam al-shahda)in contrast to the unobservable

world (#lam al-ghayb)al-Rz writes, is all that relates to bodies and bodily things
...; for one observes these things with ones sight (Tafsr, 1, 275).
91 Jabr, 156; Arba#n, 230. Cf. D. Gimaret, Thories de lacte humain, 14951.
92 Jabr, 27; cf. Arba#n, 230.

al-rzs theory of action

35

duces into his theory of action appears to be one of the places where
his synthesis between falsafa and kalm proves most difficult,93 especially that he had to face well-established criticisms of falsafa. However, in the Malib, his last major work, he solves this problem,
alongside other problems relating to divine action, by adopting a
highly unusual stance, as follows.
In all works earlier than the Malib, al-Rz affirms the common
Sunni doctrine that will is a real attribute of divine essence that is
distinct from the attributes of knowledge and power; and he locates
divine choice (ikhtiyr) in this attribute.94 Thus, the attribute of will
consists of both the ability to choose one of two possible options
freely and the actual choices themselves. Al-Rz rejects the views
of some Mu#tazils on divine will, such as the Baghdds95 and Ab
l-\usayn l-Bar, who argues that, unlike human will, which is real
and produced by motives, divine will refers to Gods motives only,
which are included in His knowledge.96
However, al-Rz defines will, in animate beings (i.e. in the
observable), in terms of the inclination (mayl) that follows naturally
from motives, and he thereby excludes the notion of choice from
it. As such, irda, as the mere natural product of motives, becomes
decisive (jzima), just as inclination is decisive (mayl jzim), and totally
contrary to free choice. Therefore, if our knowledge of the divine
attribute of will depends on qiys, whereby the unobservable is judged
in accordance with the observable, a clear implication will follow.
Al-Rz arrives at this in a section on the nature of will (aqqat alirda), in his discussion of divine attributes:
All we know in the meaning of will and aversion (karhiya) is the
inclination of the natural disposition (ab#) to attaining benefits and
its inclination to avoiding harms. Since this is inconceivable in relation to God, it will be inconceivable to affirm the notions of will and
aversion to Him.97

93

See Madelung, Late Mu#tazila, 2567.

Ishra, fol. 23b24a; Nihya, fol. 125a; Muaal, 391; Khamsn, 478; Arba#n,
147; Ma#lim, 54. On that the classical Ash#ar notion of irda includes choice, see
e.g. al-Baghdd, Ul, 102.
95 Ul al-dn, fol. 243; cf. al-Shahrastn, Nihyat al-aqdm, 238 ff.
96 On Ab l-\usayns position, see also Ibn al-Malim, Mu#tamad, 2401;
252.
97 Malib, 3, 178. Hence, he does not list will among Gods positive attributes
(Malib, 3, 5).
94

36

chapter one

God is not compellingly motivated (and, thereby, not a necessitating cause) because He does not have will, i.e. natural inclination
(mayl), which follows from motivation. As we will see in the next
chapter, al-Rz also maintains that Gods acts are not motivated.
Therefore, while Ab l-\usayn considers divine will to refer to
nothing but Gods motives, al-Rz denies that God has either will
or motivation.
This highly unorthodox later position brings al-Rz into conflict
with mainstream Sunni theology, which affirms a real and distinct
divine attribute of will.98 Yet it is an implication of his theory of action
that he accepts sincerely, not simply for dialectical consumption to
address what, he admits, is a strong Mu#tazil argument. Strangely,
by his denial of the divine attribute of will, al-Rz does not seem
to locate divine choice in any other attribute, though he still affirms
that God is a voluntary agent.
Having reached a consistent position as regards human and
divine action, the charge of inconsistency he directs at Ab l-\usayn
becomes more tenable:
Ab l-\usayn was confused on [whether the production of action by
power depends on motivation]. Whenever he discourses with the falsifa
on their saying, Why did God single out a specific time, rather than
a preceding or succeeding one, to create the world? he says, Action
does not depend on motivation. Whenever he discourses with his
companions on all other matters, he says, Action depends on motivation, and preponderation without a preponderator is self-evidently
inconceivable.99

98

Ibn Taymiyya (Majm#, 13, 128; cf. 6, 245) writes, Al-Rz discusses the
problem of will in al-Malib al-#$liya, and chooses to deny [Gods] will. For he
was unable to answer the argument of the falsifa using the principles of his companions of Jahms and Mu#tazils, so he ran off to their side! Ibn Taymiyya also
dedicates a work to refuting this position put forward in the Malib (Ibn #Abd
al-Hd, #Uqd, 51).
99 Jabr, 12. In a later episode of this debate, Ibn Taymiyya (Minhj, 1, 111; cf.
Dar", 1, 326) adopts this argument, but directs it at al-Rz:
Whenever ... al-Rz and his followers debate with Mu#tazils on matters of
destiny, they ... hold that one of the objects of the capacity of the voluntary
agent can preponderate only by a complete preponderator. Whenever they
debate with the falsifa on the questions of the creation of the world, the
affirmation of Gods choice and the refutation of [their notion of God being]
a necessitating cause, they follow the route of Mu#tazils and Jahms in saying
that one of the objects of the capacity of the voluntary agent may preponderate over the other without a preponderator.

al-rzs theory of action

37

In addition to the main argument from preponderance, al-Rz uses

the traditional argument for predestination from Gods pre-eternal
knowledge of all events.100 He treats this argument and that based on
the theory of motivation as his two strongest in this regard.101 One
clever Mu#tazil even reportedly admitted that these two arguments
are the enemies of Mu#tazilism.102 In Kitb al-Jabr, both receive
extensive scriptural backing, with evidence from the Qur"n, the
adth and the statements of companions and religious scholars.103
With the combination of these two main arguments, one primarily falsaf, the other commonly Sunn, al-Rz concludes that man,
though a voluntary agent in the sense of acting with awareness, is
ultimately compelled. He writes, Man is compelled under the guise
of a voluntary agent (muarr f rat mukhtr);104 Man is compelled
in his choosing (muarr f ikhtiyrih);105 and The acts of men occur
by compulsion (wqi#a #al sabl al-iirr).106 He does not shy from
affirming his determinism in the most explicit terms; e.g. he writes,
Affirming determinism is inescapable (al-qawl bi-l-jabr lzim); There
is nothing in existence but determination (m f l-wujd ill l-jabr);
and Man is compelled in his willing (majbr #al l-irda).107
It was almost unprecedented in Sunni theology that such an uncompromisingly bold and systematic determinism (jabr) be affirmed, given
the problems it creates with respect to human responsibility and

Yet Ibn Taymiyya himself then relies on arguments similar to al-Rzs in criticising
the position of Mu#tazils and Qadars on human free choice (Dar", 1, 329).
100 Arba#n, 3434; Muaal, 471; Ma#lim, 8990; Jabr, 4665.
101 See, e.g., Jabr, 341, where he speaks of mas"alat al-d# wa-mas"alat al-#ilm.
102 Muaal, 471.
103 Jabr, 113389. Al-Rz clarifies his general methodology in dealing with
scriptural evidence, and he is clear about their being generally less conclusive than
rational evidence (Jabr, 113118). To the objection that the Qur"n contains verses
that support both the view of the determination of acts and that of free choice, he
states that Gods speech includes both explicit truth (aqqa) and metaphor (majz),
and one should try to see the truth that lies behind the latter by interpretation
(ta"wl) (Jabr, 1323).
104 Lawmi#, 242; Tafsr, 24, 179; Malib, 3, 60; Jabr, 25; 258; Manqib, 122; cf.
Ibn Sn, Ta#lqt, 51; 53. Al-Rz (Tafsr, 15, 64) also quotes this from a defence
of predeterminism by al-Ghazl (Iy", 4, 235).
105 Jabr, 43; Mabith, 2, 517; Mal, 1/2, 389. Cf. al-Ghzal (Iy", 4, 6): All is
from God. Indeed even choice is from God. Man is compelled in the choosing that
he has (muarr f l-ikhtiyr alladh lahu). Al-Rzs statements left a strong impact on
some later authors (e.g. al-Ifahn, Kshif, 4, 356; al-Taftazn, Shar, 4, 263).
106 Jabr, 41.
107 Respectively: Jabr, 43; Mabith, 2, 217; Khalq, fol. 46a.

38

chapter one

divine justice. According to al-Rz, such ethical problems should

not detract from the truths of metaphysics. Nonetheless, he argues
that Revelation does not call average believers to believe in an absolute determinism because of the detrimental effect it could have on
their faith.108
Nevertheless, al-Rz often assumes a negative view of the designation determinists (mujbira, jabriyya) and uses it polemically against
Jahm Ibn afwn, \usayn Ibn Muammad al-Najjr, 4irr Ibn
#Amr and others.109 In some places, he distinguishes between different determinisms, of which he accepts one and rejects others.110
Elsewhere, he argues that he is not a determinist, since he maintains
that man acts in accordance with will (albeit it is ultimately determined), whereas determinism-proper is to hold that man acts not in
accordance with his will.111
Despite his explicit determinism, al-Rz admits, in one place in the
Tafsr, one of his later works, that the question of human destiny and
choice does not have an unproblematic and conclusive solution:
There is a mystery (sirr) in [this issue]; viz. that proving the existence
of God compels one to uphold determinism (jabr), ... while proving
prophecy compels one to uphold [human] autonomy (qudra).112 For if
man does not act autonomously, what use is there in sending prophets
and in revealing scriptures?
Indeed there is even yet another mystery here, which surpasses all;
viz. that if we return to sound primordial nature and primary intellect
(al-#aql al-awwal), we will find that when existence and non-existence
are on a par in relation to something, neither will preponderate over
the other without a preponderatorwhich leads to determinism. Yet
we also find a self-evident distinction between voluntary movements
and movements by compulsion, and a self-evident certitude in the
goodness of praise and the badness of blame, and in command and
prohibition; and this leads to the doctrine of the Mu#tazila.
It is as though this question falls in the sphere of contradiction (f
ayyiz al-ta#ru), in relation to both immediate and discursive knowledge, in relation to proclaiming the greatness (ta#m) of God, with

108

Cf. p. 145 infra.

I#tiqdt, 1036.
110 E.g. he responds to Mu#tazils: If you mean, by determinism, the necessary
entailment of the act from the occurrence of motivation and the non-occurrence
of deterrents, then ... this is our position (Nihya, fol. 227a).
111 Manqib, 127.
112 Al-Rz sometimes uses qudra in the Mu#tazil sense to denote both the
agents choice and his power.
109

al-rzs theory of action

39

reference to His power and wisdom, in relation to upholding divine

unity and deanthropomorphism (tanzh), and in relation to scriptural
proofs. ... The question is thus difficult and obscure. We pray God
guide us to truth!113

Elsewhere, al-Rz writes that when no solution for a problem seems

to preponderate, we will leave [that problem] captive in the sphere
of contradiction (f ayyiz al-ta#ru), i.e. by suspending judgement.114
It seems that in the above passage from the Tafsr, al-Rz contemplated taking a sceptical stance on this problem, which (as we will
see in the last chapter of the present study) could have stemmed
from a wider scepticism that appears in his later works.

The Theme of the Creation of Human Acts

We shall now turn briefly to the question of the causality that underlies the human act, i.e. whether the act is produced by human or
divine powerthis being the essence of the early kalm problem
of the creation of human acts (khalq al-af #l). To maintain that
human power effects the act will imply an affirmation of some form
of natural causality, which classical Ash#ars judged to be contrary
to divine omnipotence. In contrast, they maintained that human
acts are created directly by God, and that natural uniformity relies
on a habitual order (#da), maintained ultimately by Him, rather
than on properties that are inherent in physical objects, including
human beings. In order to avoid unequivocal determinism and to
affirm religious obligation, they then attempted to bridge the gap
that consequently appeared between man and his acts, by developing the doctrine of acquisition (kasb). According to al-Ash#ar, God
creates the agents act as well as the choice by which he acquires
both it and its attribute of being an obedience or a disobedience.
For al-Bqilln, the agents power itself effects these attributes of
action and thereby the agents acquisition thereof.
However, the later Ash#ar theologian al-Juwayn departs in his
late work al-#Aqda al-Nimiyya with the common Ash#ar position,
and contends that the human act is produced, not by divine power

113
114

Tafsr, 2, 523.
Mulakhkha, fol. 1a.

40

chapter one

directly, but by the agents power alone, provided to him by God.115

He thus discards the notion of acquisition, which he considers to be
a mere word, and pure jargon, without reference to any meaning.116
Al-Juwayn is then criticised by al-Shahrastn, a classical Ash#ar,
for implying a form of natural causality, which detracts from Gods
omnipotence.117
Turning to al-Rzs early works, we find that he adheres to the
classical Ash#ar denial of natural causality: he contends that the
agents power does not existentiate the act in any respect, though
both are ordinarily concomitant. He thus rejects, not only al-Juwayns
later position, but also al-Bqillns doctrine of acquisition on the
ground that it acknowledges the effectiveness of created power, which,
al-Rz objects, accords with the Mu#tazil position.118
Yet, from an early stage, he will also abandon the doctrine of
acquisition, which he accepted previously. He finds that both human
choice and action are created by God directly,119 and that acquisition does not refer to anything real, but is a word without a referent
(ism bi-l musamm).120 This echoes an objection that was frequently
levelled against this doctrine by many critics of the Ash#ars, including
the Mu#tazila.121 By contrast, al-Rz argues, the Qur"nic expression
kasb refers to nothing but the consequences of acts for the agent:
Acquisition denotes what man attains by his own doingso it is his
kasb and his muktasabon the condition that that involves realising
a benefit or preventing a harm. As such, one says that profits are the
acquisition of so-and-so, and that he acquires much or little, since it
refers to profit only. As for the claims of our fellow associates (abun)
that acquisition is intermediate between determinism (jabr) and human
autonomy (khalq), this appears in the old books of kalm.122

He does not find the need to resort to this (apparently outdated)

doctrine, since he considers that his model of human action does
115 Muaal, 455; Tafsr, 4, 88; cf. al-Juwayn, Nimiyya, 305; D. Gimaret,
Thories de lacte humain, 1489.
116 Al-Juwayn, Nimiyya, 32.
117 Al-Shahrastn, Nihyat al-aqdm, 7889.
118 Muaal, 471.
119 Nihya, fol. 91b.
120 Muaal, 470; Jabr, 2601.
121 Tafsr, 4, 878. Cf. #Abd al-Jabbr, Mughn, 8, 83 ff.
122 Tafsr, 5, 208 (commenting on Qur. 2:202); cf. Tafsr, 1, 69; 12, 157; 17,
110.

al-rzs theory of action

41

not lead to the same problems that arise in the Ash#ar model of
action.
Hence, in the late Kitb al-Jabr, he cites a Mu#tazil argument
from Qur"nic evidence for the view that man produces his acts
and against the doctrine of acquisition. The Mu#tazila conclude, he
notes, that bringing the word acquisition into this subject is pure
falsification (ma al-tazwr). He replies, referring to the contrary
view that God creates and determines all events:
It is necessary to harmonise between the two positions. We say: The
combination of power and motivation effects (mu"aththir) the act; and
the creator of this combination is God, exalted. Since this combination
implies (mustalzim) the occurrence of the act, the attribution [of acts to
human agents] will be real. Since this combination necessitates (mjib)
the occurrence of these acts, it becomes true that all is by Gods decree
and determination. In this way, the contradiction between rational
evidence and this Qur"nic evidence disappears. Know that this is true
according to my position in particular, as I hold that the combination
of power and motivation implies the act. As for one who does not hold
this, it will be hard for him to accept these verses.123

Far from defending the Ash#ar position, al-Rz presents his position
as a superior alternative.
Yet, crucially, he seems to contend, in this last passage, that the
act is caused by the combination of power and motivation, rather
than by the agency of God directly. This would come into serious conflict with classical Ash#arism, which affirms, not only Gods
absolute determination of all beings, but also that He creates them
all directly: they sought to affirm both qa" and khalq. This begs
the question: does al-Rz, in the relatively late Kitb al-Jabr, only
affirm qa"?
Though he rarely discusses the problem of natural causality directly,
there are indications in these later works that al-Rz indeed departs
with the standard classical Ash#ar position on this question. For
instance, in the course of his commentary on Qur. 2:22, [God]
sent down rain from the heavens and brought forth therewith fruits
for your sustenance, he posits the question of whether God creates
fruits following rain by sustaining the habitual order of created things,

123

Jabr, 3001.

42

chapter one

or whether He creates the inherent natures of water and earth such

that they produce growth in vegetation once they combine. Al-Rz
writes that although God is indeed capable of creating these fruits
directly (ibtid"an), without any intermediation (wsia), this fact does
not contradict His ability to create them through the intermediation of natural processes. Later mutakallimn, he adds, appear to
consider these two views contradictory, though they lack evidence
for this contention.124
Nonetheless, al-Rz appears to deny that the effect is existentiated
by these natural processes, including those that underlie human
action. He does so under the influence of the majority of falsifa,
who hold that,
... at the occurrence of power with motivation, complete preparedness (isti#dd tmm) for the production of the act will occur. Yet these
physical capacities do not have the ability to existentiate and effect
[anything]. When complete preparedness occurs, existence emanates
from the Bestower of Forms (whib al-uwar) upon these essences, and
they become existent. Thus, the occurrence of power and motivation
produces complete preparedness, while existence and occurrence (ul)
are from the Bestower of Forms.125

While rejecting the falsaf doctrines of emanation and the cosmogonic

intermediation of the Active Intellect,126 al-Rz accepts Ibn Sns
notion that physical objects may produce the preparedness for the
occurrence of an event in the physical world. The event, however,
will be brought into being by God directly. He writes:
The combination of power and motivation necessarily implies (istalzama) the occurrence of the act. Yet both the antecedent (malzm) and
the consequent (lzim) occur by the power of God. Similarly, although
substance and accident are concomitant (mutalzimn), they come to
exist by the power of God alone.127

As such, all created causes, including human power, are in reality

mere conditions for the efficiency of divine power. They act, not as
efficient causes, but as material causes, in producing human acts. In

124

Tafsr, 2, 110.
Jabr, 11. Cf. Ibn Sn, Najt, 136 ff.
126 E.g. Kamliyya (Ar.), 63; (Per.), 74. On Ibn Sns notion of the cosmogonic
role of the Active Intellect, cf. Herbert Davidson, Alfarabi, Avicenna, and Averroes, on
Intellect, 74 ff.
127 Jabr, 112.
125

al-rzs theory of action

43

the Mabith, al-Rz maintains that all created beings are brought into
being directly by God; yet, whereas some things can be existentiated
only by virtue of being possible, others will require the presence of
some preparing (mu#idd) factors.128 He stresses, however, that though
this is the manner in which effects are produced ordinarily, it does
not contradict Gods voluntary agency, since He is in principle able
to suspend natural order and to create effects without any preparing
causes, as, for instance, in the case of prophetic miracles.
Later than Kitb al-Jabr, in the Malib, al-Rz summarises his
position as follows:
On Explaining How the Servant is an Agent. We hold that the combination
of power and a specific motive implies (mustalzim) the occurrence of the
act. Our saying implies refers to a denominator that is common (qadr
mushtarak) to that combination being a preparing cause (sabab mu#idd)
for the occurrence of that act and its being an efficient cause (sabab
mu"aththir) for it. Since the existentiator129 of power and motivation
is God, exalted, and since it has been proven that their combination
implies the occurrence of the act, the servant will become a real agent
(f#il f l-aqqa); for the effecter of that act is his power and motivation.
All mans acts will thus occur according to Gods decree. Not as much
as an atom in the heavens or the earth will escape the chain of Gods
decree and predestination (silsilat qa" Allh wa-qadarih). The demonstration for this is that it has been proven that the preponderance of
one of two alternatives depends on the preponderator; and it has been
proven that that preponderator is [produced] by Gods agency.130

This passage, however, is not entirely clear on whether the act is

brought into being by the combination of the agents power and
motivation, or by divine power. In any case, although al-Rzs views
in this respect deserve a closer examination than is possible in the
present monograph, we can safely conclude that, in his later works,
he accepts some form of natural causality that is at odds, fundamentally, with the classical Ash#ar position. He establishes a real,
direct, uniform and (ordinarily) necessary connection between the
event and its natural cause. As such, an act will be produced, in a
real sense, by its human agent; and there will be a real and direct
link between the internal state of the agent and his action.

128
129
130

Mabith, 2, 5078.
Reading mjid instead of muwajjih.
Malib, 3, 73.

44

chapter one

Al-Rz develops his theory of action under the influence of various

sourcesfalsaf, Ash#ar and Mu#tazilbut he puts forth a unique
solution. Although he maintains a number of central Ash#ar doctrines,
he often preserves their formulaic, almost creedal, expressions, without
much of their theoretical content and details. Whether his position
on the problem of human acts is essentially Ash#ar or falsaf was a
matter of debate among later Ash#ars. One debate is concluded by
Muammad #Abduh (d. 1323/1905), who remarks, With [al-Rzs]
explication, the position of the Shaykh [al-Ash#ar] unites with that of
the falsifa, ... and the position of the Imm [al-Rz] is the position
of critical investigators (muaqqiq).131
131

#Abduh, \shiya, 76.

al-rz on the ethics of action

45

CHAPTER TWO

AL-R$Z^ ON THE ETHICS OF ACTION

Much of classical kalm relates to the so-called problem of judgements
of goodness and badness (al-tasn wa-l-taqb), which, according to
the early al-Rz, is the source from which most heretical doctrines
(bida#) spring.1 This enquiry has an essentially metaethical concern,
as it investigates issues such as the nature of morality, moral reasoning and moral language. In this context, most mutakallimn will
focus primarily on the acts of human agents, which pertain to the
more accessible and fathomable observable level (al-shhid), on the
basis of which they will then attempt to discuss the unobservable,
divine level (al-gh"ib).2 In classical Ash#ar kalm, this metaethical
enquiry usually introduces the larger, theological discussion of divine
justiceoften under the rubric, judging [acts] as just or unjust (alta#dl wa-l-tajwr)which also includes the problem of whether God
is obligated to perform certain acts, the problem of God causing
pain to the innocent (representative of the wider problem of evil),
and the problem of advantage and the most advantageous (to be
discussed below).3
Classical Ash#ar kalm ethics is then complemented by a closelylinked enquiry into normative ethics within the science of the principles of jurisprudence (ul al-fiqh), which seeks to establish general
principles for deriving specific rules and guidelines for human conduct. As the task will require an appreciation of various features
of revealed scripture, this discipline will often include a theological
component aimed at an ethical analysis of divine command (to the
exclusion of divine action). Among the other main ethical themes
commonly discussed in ul al-fiqh for their normative pertinence
1

Ul al-Dn, fol. 345.

Acts of other agents may be considered, e.g. angels and Satan, who, by arguing
for his superiority to Adam and acting upon it in his refusal to prostrate, is said
to be the forerunner for Mu#tazil ethics (al-Shahrastn, Milal, 1, 168; al-f,
Dar", 678; 945; 195).
3 E.g. Ul al-Dn, fol. 261 ff.; al-Kiy al-Harrs, Ul al-Dn, fol. 198b ff.; cf.
al-Ghazl, Iqtid, 160.
2

46

chapter two

are the problems of thanking the benefactor and obligating the

impossible (both to be discussed below).4
The mainstay of the classical mutakallimn continued to be what we
will classify as theories of the ethics of action. Their focus on action,
as opposed to character, is due partly to historical factors, viz. the
early theological debates from which classical kalm emerged, and
the legal influence on theology. Yet, more immediately, it relates to
features that are more internal to the discipline of kalm itself (in its
classical form), especially its focus on the nature of divine action,
and the common notion that human nature is essentially physical
(since, according to classical kalm physics, created beings consist
exclusively of atoms and accidents). As such, humans are morally
differentiated, not in any essential attributes (viz. of soul or character), but in the values and merits of their acts, which will eventually
be reckoned by God.
The mutakallimn advanced theories of moral ontology, epistemology
and language, tackling questions such as the nature of moral value, the
meaning and reference of moral terms, and the grounds and criteria
of moral reasoning and judgement. Most importantly, it was debated
whether moral value terms refer to real and objective attributes that
are intrinsic to acts, and whether or not moral judgements may be
discovered or established by the means of unaided reason (al-#aql
al-mustaqill). These two questions define the main outlines of the
classical positions of the Mu#tazil and Ash#ar schools, which form
the background to al-Rzs ethics. It is noteworthy that in ethical
theory, as in most other subjects, al-Rz pays almost no attention
to the views of the Mturd school, despite his familiarity with its
theology (as is clear from his debates in the Munart).5

4 Cf. al-f, Dar" al-qawl al-qab bi-l-tasn wa-l-taqb, a compendium of kalm

ethics.
5 In contemporary ethical philosophy a distinction is normally made between
teleological ethical theories and deontological theories. These have further subdivisions, only some of which are relevant in the present study. The following are
brief definitions:
(1) Teleological ethical theories define ethical value by reference to some final
purposes of acts (cf. Teleological ethics, REP). They are further subdivided into
two major types:
(a) Consequentialist theories define ethical value in terms of the favourability,
or instrumentality, of the consequences that an act promotes (cf. Consequentialism, REP and EE).
(b) Perfectionist theories, or ethics of virtue, define the good in relation to an

al-rz on the ethics of action

47

The Historical Background

The Mu#tazila
Mu#tazils hold that ethical values are real attributes of acts in the
world of objects, independently of the subjective judgements, decisions, emotions, or conventions of agents or observers, whether human
or divine.6 This stance is a type of ethical realism, or objectivism:
value judgement, according to #Abd al-Jabbr, is related to something about the act itself (amr ya#du #al l-fi#l; yakhtau bi-hi). The
Mu#tazila also maintain that some ethical truths are knowable by
unaided reason, others by the aid of Revelation.7 The mind recognises some ethical truths about acts in the same way it recognises
non-ethical facts about the external world.8 This knowledge can be
either immediate (arr) or discursive (naar).9
The Mu#tazila presented two distinct stances as regards the nature
of moral value. The Baghdd schoolrepresented by Ab l-Qsim
al-Ka#b (d. 319/931)upheld a form of ethical absolutism, according to which the moral value of an act is a real attribute in the
essence of the act, which is unaffected by the agents circumstances,

objective notion of the perfection of human nature (cf. Th. Hurka, Perfectionism;
Perfectionism, REP and EE).
(2) Deontological theories treat certain acts as good or bad, or as duties, regardless to some extent of their consequences (cf. Deontological Ethics, REP). They
have further subdivisions, including the following two:
(a) Ethical realism, or objectivism, treats ethical value as intrinsic to acts, thus
neither willed by agents, nor reducible to non-rational inclinations (cf. D. Brink,
Moral Realism; Moral Realism, REP and EE).
(b) Divine command ethics rest on the single, basic non-teleological principle
that an act is good or bad if and only if, and because, it is commanded or prohibited by God. Often, good and bad are said to mean, respectively, commanded
or prohibited (W. Frankena, Ethics, 28; cf. Voluntarism, REP; P. Rooney, Divine
Command Morality). However, if a divine command theory bases obligation on afterlife
consequences of acts, it will become fundamentally a teleological theory.
6 On the Mu#tazil refutation of Ash#ar voluntarism, see George Hourani, Islamic
Rationalism, 97 ff.; #Abd al-Jabbr, Mughn, 6/1, 102.
7 #Abd al-Jabbr, Mughn, 6/1, 58. Al-Rz sums the core Mu#tazil ethical position,
as do many of their other critics, by reference to both their realist and epistemological
stances: Mu#tazils claim [a] that the bad is bad by virtue of something that relates
to it (amr #"id ilayh), and [b] that reason either knows that aspect autonomously
(istaqalla bi-ma#rifatih) or not (Nihya, fol. 193b).
8 #Abd al-Jabbr, Mughn, 6/1, 18.
9 #Abd al-Jabbr, Mughn, 6/1, 63.

48

chapter two

or the acts consequences.10 Hence, lying and killing are bad and
prohibited absolutely, whatever their circumstances.
The Baran Mu#tazila (the later and more widespread branch
of Mu#tazilism) reject this ethical absolutism.11 They maintain that
ethical judgement depends on the aspects (wajh) upon which acts are
performed. Inflicting pain, for instance, is bad only within certain
sets of circumstances, e.g. that it neither leads to a greater benefit,
nor averts a greater harm.12
In Mu#tazil ethics, the fundamental ethical principles are universal,
in that they are immutable and apply equally to all agents, including
humans and God. Hence, Mu#tazils consider God to have certain
obligations towards humans, which stem from His essential justice,
wisdom, beneficence (isn), truthfulness, etc. According to #Abd alJabbr, although God is a voluntary agent and has the power to perform both good and bad acts, it is inconceivable for Him to perform
the latter, since He will have knowledge of their intrinsic badness.13
God, thus, acts for the wellbeing, or advantage, (al, malaa) of
humans, never in contradiction to it. The Baghdd Mu#tazila took
the more radical view that God is obligated to do what is most
advantageous (ala) to humans in some respects.
As in all deontological theories, account has to be made of the
consequences of some forms of action. In the ethics of the Barans,
this appears, for instance, in their theory of aspects, which often
takes account of considerations of benefit and harm that follow from
acts, though they maintain that an acts ethical value remains ultimately intrinsic. Wrongdoing (ulm) involves, among other aspects, the
delivery of harm to another, while beneficence involves the delivery
of benefit. Yet their respective badness and goodness are somehow
intrinsic, and not due to their actual consequences, which are often
subjective.
Benefit (naf #) and advantage (al) are defined in terms of sensual
pleasure (ladhdha) and joy (surr), whereas harm (arar) and disadvantage ( fasd) are defined in terms of pain (alam) and grief (ghamm).14
These constitute the primary elements for more complex calcula-

10
11
12
13
14

Ab Rashd, Mas"il, 357; cf. Reinhart, Before Revelation, 1413.

#Abd al-Jabbr, Mughn, 6/1, 7780; cf. Hourani, Islamic Rationalism, 634.
#Abd al-Jabbr, Mughn, 13, 298; cf. Hourani, Islamic Rationalism, 6281.
#Abd al-Jabbr, Mughn, 6/1, 1278.
#Abd al-Jabbr, Mughn, 14, 34 ff.

al-rz on the ethics of action

49

tions of benefit and harm, as Ibn al-Malim writes in his book of

definitions, Kitb al-\udd (quoted by al-Rz):
Harm is pain and what is akin to it, such as grief, the loss of benefit,
or what leads to either. ... Benefit is pleasure, joy, what is a means to
them, and what is a prerequisite for them. Gold and silver are examples of the means (mu"add). Life is an example of the prerequisite
(muai). Loss refers to preventing (man#) something from entering
into existence at the occurrence of the determinant, or to negating
(izla) it after its occurrence. ...
It follows from this division that harms are of ten types: [a] pain,
as with beating; [b] grief, as in the case of swearing; [c] prevention
of pleasure, as with preventing a man from eating food or sleeping
with his wife; [d] prevention of joy, as with preventing someone from
meeting his loved ones; [e] negation of pleasure; [f] negation of joy
([the last two] are clear); [g] prevention of what leads to pleasure, as
with preventing one from earning; [h] negation of what makes pleasure
possible, as with usurping someones money; [i] negation of what leads
to joy ...; and [j] negation of what makes joy possible.15

Unlike moral judgement, which is objective, considerations of benefit

and harm are agent-relative (if).16 #Abd al-Jabbr clearly distinguishes between the two stances, maintaining that an act can be
both advantageous and bad, or disadvantageous and good. Hence,
the infliction of harm can be good, as in the case of a deserved
punishment inflicted in an act of justice. And if a famished person
gives food that he has to someone who is not in immediate need for
it, that act will be advantageous to the latter, but not good.17
Despite the fundamentally deontological ethics of #Abd al-Jabbr
and other Baran Mu#tazila, their analysis of the teleological dimensions of human action were a major influence on al-Rzs ethics,
which is otherwise diametrically opposed to their central ethical
doctrines.
The Ash#ars
Classical Ash#ars have two main difficulties with Mu#tazil normative ethics and their analysis of the nature of morality and moral

15 Nihya, fol. 194b195a. Ibn al-Malims Kitb al-\udd, apparently nonextant, is also mentioned in another work of his, the F"iq (cf. editorial intro. to
his Mu#tamad, iv).
16 #Abd al-Jabbr, Mughn, 14, 35.
17 #Abd al-Jabbr, Mughn, 14, 36.

50

chapter two

knowledge. On the human, observable level, they maintain that

objectivist ethics will undermine the status of Revelation as the sole
basis for legislation. On the supra-sensory, unobservable level, it
will apply what Ash#ars hold to be purely human ethical norms
and duties to divine action.
Ash#ars, therefore, reject the moral objectivism and rationalism
(tasn al-#aql wa-taqbuh) of the Mu#tazila. Moral language, they
maintain, does not refer to any real properties of acts in the external
world; for goodness and badness do not have any objective reality
at all. Unaided reason, therefore, may provide knowledge of metaphysical truths (e.g. that the world is created, that God exists, and
that prophecy is conceivable), but it will never perceive moral truths
in acts and things.18
However, if this is the case, then what does moral language refer to?
In answering this metaethical question, Ash#ars distinguish between
two classes of moral expressions in ordinary language.
The first class includes expressions that may be said to have moral
meanings, or are sometimes used with moral senses, when in fact
they have non-moral lexical meanings (which may indeed refer to
some objective properties of acts or things). These expressions are
particularly relevant to analysing the meanings of expressions relating
to divine attributes, which are normally interpreted on the basis of
their lexical meanings. Someincluding justice, injustice, wrongdoing and wisdom (ikma)are said to describe the perfection or
imperfection of acts. Ibn Frak writes:
[Al-Ash#ar] maintained that, in ordinary language (ilq al-lugha),
describing an act as injustice (jawr) or wrongdoing is not congruent
to describing it as bad (qab). For the lexical meaning of jawr is
departure from the normal standard and standard measure (al-zawl #an
l-rasm al-masnn wa-l-add al-marsm), be the departer obligated (mukallaf)
or not. The saying, The arrow deviates from (jra #an) the target, if
it misses it, is always considered literal. Its missing the target is said
to be jawr in the literal sense, although it is not an act by one who
is obligated, or one who is prohibited from it.19

Also, the lexical meaning of justice is balance (i#tidl) and equilibrium (istiw") in everything, i.e. that there be neither excess nor

18
19

Tafsr, 7, 146; al-Baghdd, Ul al-Dn, 2025.

Ibn Frak, Mujarrad, 96; cf. al-Baghdd, Ul, 132.

al-rz on the ethics of action

51

remissness in it. Ab Isq al-Isfar"n goes even further by interpreting good and bad in the same manner:
Justice is to put things in their appropriate places, and this is the
fundamental sense of goodness (usn); injustice is to put things in
other than their appropriate places, and this is the fundamental sense
of badness (qub).20

Ash#ars accordingly interpret divine names, such as Just and Wise,

in non-moral senses, mainly as related to perfection, or masterly
production, (ikm) in Gods creation, which originates from His
knowledge, power and wisdom (whence ikma).21
The second class includes expressions that undeniably have moral
senses in ordinary language (whereas their lexical meanings do not
seem to be at issue). However, Ash#ars argue that these expressions
do not refer to intrinsic attributes of acts, but to the subjective judgements of individuals. Ibn Frak again writes:
[Al-Ash#ar] maintained that there is only one sense for bad and good
in the observable: that what is bad is avoided for the imperfection and
harm that it results in for one who does it, and that the good and wise
act is chosen because of the benefit and perfection that it results in
for one who does it. There is no ground for the acts performance or
omission, in the observable, but this or its like.22

Ordinary moral language, thus, is completely agent-relative and

reducible to the benefit and harm (defined in terms of pleasure
and pain), and the perfection and imperfection, of the individual.
Something is good for me if it provides me with some sort of benefit or perfection, bad for me if it results in some sort of harm or
imperfection for me. It follows from this position that if what one
wills is realised, one will consider that to be good; and vice versa. If
this will is dependent on the performance of some action by another
person, then one may ask or command that person to do so. As such,
correct (awb) may mean agreement (muwfaqa) with command,
or attainment of what is willed (ibat al-murd) .23

20

From: R. Frank, Creation and the Cosmic System, 64, with some modification.
Al-Ghazl writes, The Wise (\akm) means the knower of the realities
of things and the one capable of creating them perfectly according to His will
(Qudsiyya, 90; cf. Iqtid, 1656; Ibn Frak, Mujarrad, 967).
22 Ibn Frak, Mujarrad, 1412; cf. 967.
23 Ibn Frak, Mujarrad, 96.
21

52

chapter two

Ash#ars are then challenged to explain why people will often

make moral judgements and perform acts that are not self-centred,
but appear altruistic. If a lone traveller in a desert comes across a
man or an animal who is dying of thirst, he will help him, even if
the traveller is an atheist and does not expect any worldly benefit
from this help. Classical Ash#ars normally explain such acts as being
motivated by generic sympathy (riqqa jinsiyya), a purely emotional,
subjective motive; when one sees another man in pain, one will feel
pain oneself and may be motivated to help.24
Al-Ash#ar reportedly contrasts the second, subjective class of
expression with the former, pseudo-moral class:
Describing something as advantageous, advantage or most advantageous is not akin to describing it as wise, correct or true. ... Do
you not see that the corruption of bodies by maladies, diseases, death,
torture in fire, and the occurrence of pains and various harms in them,
is related to what is true, wise and correct, but is not advantageous, most
advantageous or advantage for the agent or the object of action! Had
describing the thing as advantageous and advantage been equivalent
to describing it as wise, it would have followed it in relations (ift
wa-nisab), such that [something] will be an advantage for whoever it is
wise, or wise for whoever it is advantageouswhich is false.25

However, this Ash#ar analysis of moral expressions only describes

the psychological motives that they commonly express in the mundane realm of everyday language (in the observable, as they say),
but does not ascribe any normative, prescriptive authority to them.
This is why classical Ash#ars attach little importance to this subjective sense of value terms, and invoke it solely in their refutation of
Mu#tazil ethical realism, as does al-Kiy al-Harrs:
We say: Blame and praise rely on ends (ghara), i.e. benefits and harms.
Something may be good in relation to the ends of one person, so he
praises it, harmful in relation to the ends of another, so he censures it.
For instance, ... if one informs a man about the infidelity of his wife,
[that man] will praise him ..., whereas the woman will censure him. ...
What is at issue is not this, but [value judgements] without consideration
of ends, in which case goodness and badness will not be essential [to
acts]; for only Revealed Law makes [acts] good or bad.26

24 Cf. Ibn Frak (quoting al-Ash#ar), Mujarrad, 142; al-Juwayn, Nimiyya, 27;
al-Ghazl, Iqtid, 171; Mustaf, 1, 59.
25 Ibn Frak, Mujarrad, 127.
26 Al-Kiy al-Harrs, Ul al-Dn, fol. 200b; cf. 201b.

al-rz on the ethics of action

53

Anti-Mu#tazil dialectic aside, classical Ash#ars do not show interest

in developing their analysis of ordinary moral language. Their real
interest (to which al-Harrs refers) lies in the supra-mundane status
of Revelation, which alone they regard as the source for duties and
rules of conduct.
Relying, often implicitly, on their aforementioned metaethical view
that value expressions could be understood in relation to will and
command, Ash#ars develop a divine command (or a voluntarist)
theory of normative ethics. As such, they argue that the technical
definitions they give to value expressions are not entirely independent
of their uses in ordinary language, though they will also stress that,
unlike human commands, divine commands are not motivated by
subjective inclinations.
According to Ash#ars, morality is Legal (shar#): it consists of the
adherence to rules of conduct, including duties and recommendations,
that derive from divine prescriptions. The bad is bad by virtue of
the relation of Gods prohibition to it (ta#alluq nahy Allh #anh bih).27
Al-Juwayn writes, Good is what the Law presents praise for one
who does it; Obligatory (wjib) is that act that the Law presents
a categorical command for [performing it]; Good is not an attribute additional to Law, which becomes known through it; rather,
it is the same as the Laws presentation of praise for one who does
it.28 And, according to Ibn Frak, To say that our acquisition
(kasb) is good done by us (asan minn) is equivalent to saying that
it is commanded by God.29
Revelation, therefore, is the sole, or primary, source for moral
knowledge. Most instances of lying are bad, not intrinsically, but
because God prohibits them. Some types of lying are not bad, since
God permits them. If He decides that lying is good and obligatory,
then so it will be.30 All that humans can, and should, do is to interpret revealed texts, analyse their prescriptions and investigate how
to apply them (these being the concerns of the jurist mainly).
In later classical Ash#ar theology, we find signs of a growing
teleological trend, which appears to be a development of the earlier

27
28
29
30

Ibn Frak, Mujarrad, 94.

Al-Juwayn, Irshd, 228; cf. Ibn Frak, \udd, 11; al-Mutawall, Mughn, 43.
Ibn Frak, Mujarrad, 95.
Al-Ash#ar, Luma#, 170.

54

chapter two

subjectivist analysis of moral language, under slight falsaf influence.

This trend is hinted at most rudimentarily in al-Juwayns late work
al-#Aqda al-Nimiyya:
[The correct] approach [to the question of ethical value] is to consider
only one excellent, decisive and concise premise that will erase confusion from within one who understands it. What the followers of whims
(ahl al-ahw") consider to be good in itself, such as belief [in God] and
thanking the benefactor, or bad in itself, such as lying and wrongdoing, can only apply to ... one who is liable to harm and benefit. The
reality of benefit, pleasure and sorrow (hamm) is the sensing of fear
from pains and evils, and of comfort from pleasures.31

Al-Juwayn here does not mention any Legal definitions of good

and bad alongside this principle. Also, al-Rghib al-Ifahn (d.
probably ca. 503/1110), normally close to Ash#ar theology, writes,
under falsaf and Sufi influence:
Many mutakallimn [viz. Mu#tazils] hold that truthfulness is good in
essence, and lying bad in essence. Many falsifa and Sufis hold that
lying is bad because of its connection to the harms that result from it,
and that truthfulness is good for the benefits that result from it. This
is so, since speech is an act; and no act can be good in itself. Rather,
the good is good because of the benefits related to it, and the bad is
bad for the harm that is related to it, which preponderates over the
benefit that follows from it.32

Though this statement appears in a work on the science of character

(akhlq), not kalm, it is nonetheless instructive.
The most significant development in pre-Rz kalm towards teleological ethics is introduced by al-Ghazl (d. 505/1111), who writes
in the Iqtid:
Concerning the good (asan), its range of meanings is such that acts,
in relation to the agent, are divided into three divisions. The first is
for [the act] to be agreeable (wfaqa) to him, i.e. that it fits (l"ama) his
end (ghara). The second is for it to contradict (nfara) his end. The
third is for him not to have a purpose for either performing or omitting it. This division is evident to the mind. Thus, what is agreeable
to the agent is called good as far as he is concerned (asan f aqqih);
and there is no meaning for its goodness other than its agreement to
his purpose. What contradicts his purpose is called bad; and there is

31 Al-Juwayn, Nimiyya, 26; cf. 456. G. Houranis article, Juwayns criticism

of Mu#tazil ethics, examines the Irshd only, but not the Nimiyya.
32 Al-Rghib al-Ifahn, Dhar#a, 272.

al-rz on the ethics of action

55

no meaning for its badness other than its contradicting his purpose.
What neither contradicts nor agrees [with his purpose] is called vain
(#abath), i.e. that there is no benefit ( f"ida) in it at all.33

This marks a crucial turn towards emotivism, which al-Ghazl

advances as an alternative to Mu#tazil ethical objectivism. He argues
that all moral judgements are reducible to the subjective considerations of self-interest, viz. agreement and contradiction to the agents
ends. These arise ultimately from emotions: inclinations (mayl) that
consist of estimation (wahm) and imagination (khayl), and stem from
the natural disposition (ab#), rather than reason.34 Goodness and
badness, for all human beings, are expressions of relative things
(amr if) that differ according to relations.35 But why do humans
sometimes perform apparently altruistic acts? Al-Ghazl explains
that such acts are motivated, first, by generic sympathy (the standard
Ash#ar explanation), and second, by the agent being accustomed to
some acts deserving praise or blame.36
With al-Ghazl, this subjectivism is not regarded solely as a
dialectical notion that serves to refute Mu#tazil ethical realism, but
it constitutes the grounds for a consequentialist normative ethics
that contrasts with the classical Ash#ar deontological stance. The
consequences of acts become the grounds for morality, on which the
Legal authority of Revelation will rest. Al-Ghazl, therefore, writes
of three senses for the expression good:
[a] Some use it to designate all that agrees with an end (ghara ),
whether immediately or in the distant future. [b] Some specify that
to what agrees with an end in the hereafter, which is what Revealed
Law made good (assana), i.e. it exhorted the performance thereof and
promised reward for it. This is the technical sense (iil) of our fellow
associates (abun). ...
[c] There is a third technical sense for it. It may be said, The action
of God, exalted, is good, whatever it may be, and though He has no
ends. This will mean that [His action] can result in neither consequence
for, nor blame on, Him, and that He acts in His dominion in which
He has no co-sharer.37
33 Al-Ghazl, Iqtid, 163; cf. Mustaf, 1, 568; Qudsiyya, 90; Hourani, Ghazl
on the Ethics of Action, esp. 714. On relevant sections in Mi#yr al-#ilm see Michael
Marmura, Al-Ghazl on Ethical Premises.
34 Al-Ghazl, Iqtid, 1724.
35 Al-Ghazl, Iqtid, 164.
36 Al-Ghazl, Iqtid, 170 ff.
37 Al-Ghazl, Iqtid, 165.

56

chapter two

The third sense is clearly related to the first two. Al-Ghazl then
goes on to give the Ash#ar definition of wisdom, in terms of the
perfection of knowledge and action.
The influence of al-Ghazls rudimentary consequentialism on
later mainstream kalm was apparently limited. We find classical
Ash#ar voluntarism being defended by al-Shahrastn, who also
rejects the theory of motivation entirely.38 This same stance appears
also in al-Rzs earliest kalm work, Ul al-dn, where he writes:
Nothing is good in itself or its species. Indeed, judgements (viz. goodness, badness, obligation, prohibition, recommendation (nadb), reprehensibility (karhiya) and permissibility) rely on the statements of the
Lawgiver. Good is what Revealed Law presents praise for one who
does it, and bad is what Revealed Law presents blame for one who
does it. ... Goodness is the same as the Laws presentation of praise
for one who does [an act].39

Despite al-Ghazl, deontological ethics remained, until al-Rzs time,

the dominant trend in kalm among both Mu#tazils and Ash#ars.
Yet, al-Ghazls approach signals greater eclecticism, which will
lead to al-Rzs definitive developments in kalm ethics.

Ethical Value
Whereas al-Rzs earliest known kalm work, Ul al-dn, presents a
classical Ash#ar theory of value, we encounter a more sophisticated
ethical theory in his later works. In the Muaal, he defines the
central value terms as follows:
Goodness (usn) and badness (qub) may be intended to refer to [a]
agreeability (mul"ama) and disagreeability (munfara) to disposition (ab#),
and to [b] something being an attribute of perfection or imperfection.
In these two senses, they are rational (#aql) [conceptions].
They may also be intended to refer to [c] something entailing reward
or punishment, and praise (mad) or blame (dhamm). For us, this sense
is Legal (shar#), contra the Mu#tazila.40

These three definitions represent distinct ethical stances that alRz, as we will see, eventually develops into a coherent teleological
38
39
40

Al-Shahrastn, Nihyat al-aqdm, 37091.

Ul al-Dn, fol. 262.
Muaal, 4789.

al-rz on the ethics of action

57

ethics. We need to analyse these stances separately, starting with the

two that relate to the ethics of action: the subjectivist stance in the
present section, and the Legal stance further below in this chapter.
The perfectionist stance, which relates to attributes of essence and
character, will be the focus of the next chapter.
The earliest account of al-Rzs subjectivist definition of value can
already be found (alongside the two other definitions) at a slightly
later stage than Ul al-dn, in his early kalm work the Ishra:
Man designates the word good (asan) for what agrees with his ends
(wfaqa gharaah), and the word bad (qab) for what contradicts (khlafa)
his ends. Accordingly, a thing may be good in relation to one person,
bad in relation to another.41

This same view is advanced in later works. Al-Rz writes in the

Malib, What we understand by goodness and badness is nothing
but benefit (manfa#a) and harm (maarra);42 and in the Ma#lim, There
is no meaning for [judging acts as] good or bad (tasn wa-taqb)
but the acquiring of benefits and the avoidance of harms.43 The
same applies to other value terms, such as good (khayr), advantage
(malaa), evil (sharr) and disadvantage (mafsada).44 Moral judgement, therefore, is subjective and reducible to self-interest. As the
values of acts are connected to their consequences for the individual,
whether past or expected, they are completely agent-relative. The
same act will relate differently to the ends of different agents; it
may result in favourable consequences for one person, for whom it
will be good, but unfavourable consequences for another, for whom
it will be bad.
This analysis of moral judgement, which appears from the Ishra
onwards, seems initially to be a borrowing from al-Ghazl, who too
defines value in terms of agreeability and disagreeability to the agents
ends. This Ghazlian influence highlights al-Rzs increasingly
psychological approach to ethics, action and the theory of human
nature in general. At the early stage of the Ishra, however, al-Rz
still adheres to the classical Ash#ar theory of action, which lacks any
significant psychological component.
By contrast, in none of his later works do we find moral value
41
42
43
44

Ishra, fol. 32b.

Malib, 3, 291.
Ma#lim, 87.
Malib, 3, 212 (see p. 23 supra); cf. Ma#lim, 86.

58

chapter two

being defined with reference to end, or objective (ghara). With his

increasingly psychological theory of action, his approach to moral
value will require clearer psychological reference points than the
notions of agreement and disagreement with ends. In the Malib,
al-Rz points out that while end refers to the external, objective
state of affairs that the agent attains or expects, motive has a subjective, psychological referent.45
Thus, in Nihyat al-#uql, as in the above passage from the Muaal,
al-Rz writes that goodness and badness may be intended to
designate agreeability (mul"ama) and disagreeability (munfara) to
disposition.46 Later, in the Arba#n, he writes,
We know that, within ourselves, some things are agreeable (mul"im)
to our dispositions, and that some are disagreeable (munfir) to our
dispositions. Pleasure (ladhdha) and what leads to it are agreeable. Pain
and what leads to it are disagreeable.47

Benefit and harm, therefore, are ultimately defined in terms of the

primal sensations of pleasure and pain that the agent experiences,
or expects to experience.
Al-Rz considers pleasure and pain to be real (aqq; thubt)
phenomena that are among the most immediate and self-evident
items of human knowledge. They have simple essences and cannot
be defined in terms of anything else, not even (as Ibn Sn defines
them) in terms of our perception (idrk) of what is agreeable or
disagreeable to ourselves.48
Therefore, al-Rz maintains that the notions of good and bad
are rational, in the sense that they refer to mental perceptions of
some simple, internal sensations (in contrast to these notions being
essentially Legal, defined with reference to Revelation). Yet, though
perceived internally by the mind, these sensations do not stem from
the mind, but from the agents natural disposition (ab#), i.e. not from
reason, but desire. Al-Rz frequently challenges the Mu#tazila to
show that the aversion that one finds within himself [towards bad acts]
is rational, rather than originating in natural disposition (ab#).49
As such, al-Rzs ethical rationalism (tasn al-#aql wa-taqbuh)
45
46
47
48
49

Malib, 3, 10.
Nihya, fol. 195a.
Arba#n, 246; cf. Nafs, 1920; Ma#lim, 86.
This is discussed in more detail, p. 156 ff. infra.
Nihya, fol. 199b.

al-rz on the ethics of action

59

is at odds with Mu#tazil ethical rationalism. Rather than a realist,

cognitivist ethics, he appears to present what in contemporary philosophy is described as an emotivist theory of ethics: moral statements
merely express non-rational inclinations and impulsessensations
of attraction and repulsion towards certain acts and thingsthat
emerge from ones natural disposition. If I say, Lying is bad, I will
mean, I hate lying, or Lying is repulsive to me. In his criticism of
Mu#tazil ethical realism, al-Rz writes:
If the statement, Beneficence (isn) is good, referred to [beneficence]
being liked by the disposition and desired by the self (nafs), since it
constitutes a cause for the occurrence of benefits, this would be true
and correct. We do not dispute with you that knowledge of its goodness in this sense is immediate. Also, if the statement, Wrongdoing
is bad, referred to its being hated by the disposition and detested in
the heart, since it constitutes a cause for the occurrence of pains, grief
and sorrows, then there would be no dispute in that knowledge of its
badness in this sense is immediate. ... Goodness and badness become
interpreted (mufassar) through benefit and harm, and advantage and
disadvantage.50

Someone with a sensitive disposition and subtle humours will

tend to be charitable in his conduct and attitudes. By contrast, alRz adds,
I saw one of the greatest of kings, who was utterly ruthless, and whose
sole pleasure was to watch killing and pillage. The more brutal the
form of torture he watched, the more complete would be his joy and
the jovial expression on his face.51

For such a person, killing and pillage are good, the more brutal
the better.
Now, we saw in the previous chapter how al-Rz concludes that
all human motives and value judgements are rooted in, and reducible
to, the two basic sensations of pleasure and pain. He contends that we
know immediately (bi-l-arra) that we seek some things and incline
towards them and are repulsed by others. From these inclinations,
our value judgements are formed. He posits the question of whether
these inclinations and judgements have, at their core, some primal
inclinations from which they ultimately emerge, or not. If not, then
either infinite regress or circular interdependence of inclinations on
50
51

Malib, 3, 347.
Malib, 3, 3501.

60

chapter two

each other will follow, both of which are inconceivable. Therefore,

there must be something that is sought in itself and another that is
hated in itself ultimately. He continues,
Having meditated and investigated (ta"ammaln wa-baathn), we have
found nothing that is sought in itself but pleasure and joy, or the prevention of pain and grief, and nothing that is hated in itself but pain
and grief, or the prevention of pleasure and joy.52

According to al-Rz, that we do make value judgements independently is a self-evident and inarguable fact. However, as the true
nature of value can only be discovered through rational reflection, our
knowledge of it is not immediate, but discursive. For the Mu#tazila,
by contrast, both the truth of moral judgements and the essences
of moral values can constitute immediate knowledge. A Mu#tazil
may argue that all rational men accept that lying is bad, and that
when anyone is asked to explain this statement, he will affirm that
the badness of lying is intrinsic. Al-Rz rejects the latter claim and
argues that in generally-accepted convention (al-#urf al-#mm al-mashhr), what people mean by the expression evil (sharr) is pain and
what leads to it .53 Yet, in itself, this conventional usage does not
constitute evidence for the immediacy of this knowledge. Indeed,
he accepts that some people will make value judgements, believing
(wrongly) that value is intrinsic to acts.
A metaethical theory of moral judgement and motivation of the
sort that al-Rz advances is often described as an egoism, or, given
its emphasis on the primacy of the prudential and pleasure-seeking
motives, a hedonism. Yet, al-Rzs theory does not amount to a
crude hedonism; for he outlines a hierarchy of divisions of pleasure
and pain, which (as we will see below) differ according to which
part of the person experiences them. Joy is a non-physical type of
pleasure, and grief a non-physical type of pain, whereas the expression pleasure is often used narrowly for purely physical pleasure.54
He continues:
The benefit that is sought for its own sake is either pleasure or joy,
and the harm that is avoided in itself is either pain or grief. So it is
necessary to investigate which is stronger: pleasure or joy, pain or grief.

52
53
54

Malib, 3, 348; cf. Nafs, 20.

Shar al-Ishrt, 2, 80.
E.g., Nafs, 1920 (translated p. 111 infra).

al-rz on the ethics of action

61

We say that this varies from soul to soul. There are those who prefer
bodily pleasure, while others prefer joy.55

Wherever al-Rz speaks of pleasure and pain as the primal bases

for moral attitudes, these should be understood as referring, not to
physical pleasures and pains only, but to pleasure and pain generally,
regardless of their type.
The same processes and factors that govern motivation for action
(including the agents internal disposition and external circumstances)
also govern the making of moral judgements. For instance, harm is
divided into (a) harm proper, or positive harm, (b) the negation of
benefit, and (c) the negation of what prevents harm; while benefit
is divided into (a) benefit proper, (b) the prevention of harm, and
(c) the negation of what prevents benefit. There are also degrees of
harm and benefit, and consequently of goodness and badness. Lying
is generally considered less bad than wrongdoing, since it normally
results in milder harms than those normally resulting from the latter.56
A value judgement on a certain act will depend on a prudential and
pleasure-seeking calculation that occurs in the agents mind.57 This
notion of a mental calculation behind motivation and value judgement appears to be borrowed primarily from Mu#tazil discussions
of the theme of advantage and disadvantage.58
Al-Rz informs us that both Ash#ars and the falsifa reject Mu#tazil
moral realism,59 which indicates that the main influences on his later
ethical thought are not simply classical Ash#ars and al-Ghazl.
As we will see below, he also adopts Ibn Sns view that moral
statements, such as Lying is bad, are widely-accepted (mashhra)
statements, constituting little more than conventional beliefs, not
items of knowledge.60 It appears that, in his falsaf works, al-Rz
makes another, related borrowing from Ibn Sn. He contends that
statements that have value terms, such as good, noble and base,
as predicates are essentially rhetorical (khab).61 Value statements
tell us nothing about the reality of the things or acts they describe,

55
56
57
58
59
60
61

Malib, 3, 24.
Malib, 3, 349.
Malib, 3, 34850.
Cf. p. 489 supra.
Malib, 3, 289.
See p. 89 infra.
Mulakhkha, fol. 326a.

62

chapter two

but, when used in certain context, they only aim to persuade (iqn#)
common people.
Moral statements, thus, either express emotions, or evoke them
in an audience, in which case they will be prescriptive. As such, the
seemingly factual, descriptive statement, Lying is bad, will be tantamount to the imperative, Do not lie! This stance, which al-Rz
adopts form Ibn Sn, is reminiscent of contemporary non-cognitivist
moral theories (both emotivist and prescriptivist), including the socalled Boo! Hoorah! approach to moral expressions. It is interesting
that al-Rz takes it to heart more than Ibn Sn does, whom he
criticises in his discussion of the theory of emanation for using such
a value statement as a premise:
He states in the Book of Demonstration of the Shif" that if you find the
knowledge-seeker62 saying, This is noble; that is base, then know that
he is confused.63 Somy goodness! (layta shi#r)how does he allow
himself here to use this rhetorical premise in the present scientific
discussion!64

Rhetoric has its appropriate places. But discussions of objective

science are not among them.
Al-Rzs views outlined so-far constitute a descriptive metaethical
theory. But what kind of normative ethics does he advance? Most
immediately, his moral subjectivism forms an ideal framework for
a self-centred consequentialist ethical theory, the view that an act is
good if it serves the agents own interests, bad if it contradicts them.
Yet al-Rz considers it superfluous to provide reasons for such a
prudential and pleasure-seeking normative stance. He takes it for
granted that all rational human beings (indeed even animals) will
seek pleasure and avoid pain by their very natures, and will know
immediately that they ought to do so. Avoiding harm as much as
possible is rationally imperative (malb f l-#aql).65 As such, a basic
self-centred consequentialism (a so-called ethical egoism) will follow
immediately from psychological egoism.
Yet al-Rzs normative ethics is more complex. Firstly, while
this basic self-centred consequentialism seems to concern mainly the
62

Reading #ilm instead of #mm.

Reading yakhliu instead of ghalia.
64 Shar al-Ishrt, 2, 50; cf. 2, 58; 2, 39; Mabith, 2, 488; Ibn Sn, Shif",
Maniq, 5, 131.
65 Malib, 3, 25.
63

al-rz on the ethics of action

63

mundane, private sphere, the hierarchy of pleasures and pains that

he elaborates allows for a perfectionist ethics that goes beyond this
stance and relates to the supra-mundane, private sphere (the subject
of the following chapter). Two more normative stances, both based
on al-Rzs core self-centred consequentialism, are presented in
his writings: a contractualist stance, relating to the mundane public
sphere, and a Legal stance, relating to both the private and public
spheres. These last two stances will be examined in the two following sections.
Revealed Law and Ethical Value
The second sense of value terms that al-Rz lists in works later
than Ul al-dn relates to Revealed Law. He initially describes it
as being a Legal (shar#) sense, in contrast to the two other senses,
which are rational (#aql). In the Ishra, he presents the subjectivist
and perfectionist definitions of value terms, and writes:
Good also designates what the agent is not prohibited from doing.
What is permissible (mub) is good in this sense. The same applies
to the acts of God.
It also designates what Revealed Law attaches praise (than"), laudation
(ta#m) and desert (istiqq) of reward for one who does it. Conversely,
bad is what Revealed Law attaches blame (dhamm), derision (ihna)
and desert of punishment for one who does it. What is permissible,
in this sense, is not good.
Know that, according to the doctrine of the upholders of truth, these
last two senses do not constitute attributes for what is good or bad.
Rather, they are purely the relations of the discourse of Revealed Law
to it (mujarrad ta#alluq khib al-shar# bi-hi). Therefore, if permission (idhn)
for [performing] an act is presented, it will be described as good; if
prohibition from it is presented, it will be described as bad.66

This distinction between the two definitions of good concerns the

sense in which Gods acts are said to be good, though not as being
commanded by Him. In the Mal, al-Rz prefers the former
definition, giving primacy to the notion of bad, which is defined
more straightforwardly than good:
We mean by bad what is Legally prohibited (manh #anh shar#an); and by
good what is not Legally prohibited. Included in [this definition] will
66

Ishra, fol. 32b33a.

64

chapter two
be Gods acts, and the acts of obligated agentsbe they obligatory acts
(wjibt), recommended acts (mandbt) or permissible acts (mubt)as
well as the acts of the unaware, the sleeper and animals. It is more
appropriate than saying, The good is what is Legally permitted,
since it follows from it that Gods acts are not good.67

Notwithstanding, these definitions are essentially nuances of the

same Legal approach to value, the differences being only in detail.
Thus, in his early works (before the Ma#lim), al-Rz continues to
adhere to the voluntarism of classical Ash#arism, reproduced most
traditionally in Ul al-dn: value is defined in terms of divine command: bad is what God prohibits; obligatory is what He commands categorically.
The central ethical problem thus remains the same as that of classical kalm: whether the foundation of morality is divine command
or unaided reason. Al-Rz writes in the Arba#n:
[There is] dispute over whether the connection of some acts to blame
in this world and to punishment in the hereafter, and the connection of others to praise in this world and to reward in the hereafter,
are due to an attribute of the act itself or, contrarily, purely to the
judgement of Revealed Law to that effect, or the judgement of those
knowledgeable in it.
Mu#tazils say that the causes of these judgements are attributes in
acts themselves. Our position is that they are purely the judgements
(ukm) of Revealed Law.68

The commands and prohibitions of Revealed Law are prescriptions

that relate to acts, but do not unveil any real moral attributes of
them. There will be no meaning to badness but a specific relation to the discourse of Revealed Law (khib al-shar#),69 viz. being
forbidden by it as unlawful (nahy al-tarm). Al-Rz then addresses
a main Mu#tazil objection:
[Objection:] Had goodness and badness not been knowable by reason,
it would have been impossible for us to know them after Revealed
Law presented them; for assertion (tadq) must be preceded by conception (taawwur). Therefore, the basis of goodness and badness must
be knowable by reason.

67 Mal, 1/1, 136; cf. Nihya, fol. 199b200b, where the point is made in
reply to a Mu#tazil objection relating to the same problem (cf. #Abd al-Jabbr,
Mughn, 6/1, 107).
68 Arba#n, 246; cf. Mal, 1/1, 15960; Muaal, 479.
69 Nihya, fol. 200b.

al-rz on the ethics of action

65

[Reply:] We do not claim that we acquire conceptions of the essences

of goodness and badness from Revealed Lawwhich may entail the
error you refer to. For we may say that obligatory is what the Lawgiver commands the performance thereof and forbids us from omitting. This much is conceived by us prior to our reception of Revealed
Law. Therefore, in saying that these judgements become affirmed only
by Revealed Law, we do not imply that they become intelligible by
Revealed Law alone.70

Following classical Ash#ars, al-Rz analyses the notion of obligation

as used in ordinary language by reference to will and command
command (amr) being defined as requesting [the performance of]
an act verbally by way of superiority (#al sabl al-isti#l").71 This
much is understood by humans independently of revelation. The
fact that a rule is commanded by God in itself constitutes its being
a duty, rather than a reason for its obligatoriness. Obligatory simply
means commanded by God.
There is nothing strikingly original in this voluntaristic notion of
ethical value, except that the Legal definition is juxtaposed with
two other definitions. In al-Rzs early works, this juxtaposition
appears to serve the negative purpose of showing that the Legal sense
is independent of the two other senses, and hence non-rational. It is
non-consequentialist; for the sole fact that an act is commanded by
God will make it good, regardless of its consequences in this world
or the hereafter. Al-Rz writes in the Mal:
The essence of obligatoriness (taklf) becomes established at the prohibition of omitting the act. To establish this, it suffices that blame
follows from omission; and there is no need for punishment to follow
from omission.
It is strange that al-Ghazl discusses this problem after scrutinising
the definition given for obligatory as punishable if omitted, and
stating that it is better to say that obligatory is leading to blameworthiness if omitted. This is an admission by him that establishing
the essence of the obligatory does not depend on punishment, but that the desert of
blame suffices to establish it. Then, ... he says that the essence of obligation is established only by the preponderance of the performance [of
the act] over omission; and preponderance is realised only through
punishment. Doubtless, this is an obvious contradiction.72

70

Objection: Nihya, fol. 199b; Reply: Nihya, fol. 200b.

Mal, 1/2, 22. Command, al-Rz maintains, cannot be defined as conveying (ikhbr) that the punishment of one who omits [the act] is willed.
72 Mal, 1/2, 3402; cf. al-Ghazl, Mustaf, 1, 66; 1, 83.
71

66

chapter two

The accusation of incoherence aside, al-Rz here clearly objects

to al-Ghazls consequentialist approach to Legal obligation, and
favours classical Ash#ar deontology, specifically the view taken by
al-Bqilln, that obligation should not be based on the desert of
punishment.
As al-Rz becomes increasingly eclectic and independent of traditional Ash#ar thought, he becomes more aware of the problematic nature of this so-called Legal notion of value. If there is no
cognizable link between the standard, non-technical meanings of a
certain value expression and its technical definition, the latter will
seem arbitrary and unjustifiable. Why does good mean what God
commands? Why, for instance, can we not define bad as what
God commands? And how do we arrive at obligation from will
and command, human or divine, without recourse to consequent
reward and punishment? Arguably, the fact that scriptural prescriptions are made by an absolutely superior being neither means, nor
immediately entails, that they are obligatory. But if this obligation
has to be established by another divine commandObey My commands!, Thou ought to obey Me!then what makes obedience to
this command obligatory?
In his later works, the Ma#lim and the Malib, al-Rz will abandon deontology in favour of teleology. As his later motivational
psychology leaves no room for non-teleological action, he adopts a
thoroughly consequentialist ethics of action. One crucial premise will
be introduced into the prudential and pleasure-seeking calculation:
that afterlife pains are by far the severest of all possible pains, and
afterlife pleasures are the most excellent pleasures that humans may
experience. Since this-worldly pains and pleasures are comparatively
mundane, the most rational course of action for the believing agent
will be to seek his own personal salvation by aiming to avoid afterlife
punishment and attain afterlife reward. The former aim is prudentially the least the agent ought to do, since, according to al-Rz,
minimising pain is more crucial than maximising pleasure.
He, therefore, abandons the distinction between rational and
Legal conceptions of value, and writes in the Ma#lim:
Though debauchery provides a type of pleasure, reason prohibits
(mana#a #an) it. And it prohibits it only because of its conviction that it
will consequently result in greater pain and grief. This shows that the

al-rz on the ethics of action

67

modes of goodness, badness, and inculcating fear and desire (targhb

wa-tarhb) are precisely as we have described. ...
Those who view goodness and badness as being based on Revealed
Law (bi-asb al-shar#) define bad as leading to punishment if done.
It may then be said to them: Do you accept that reason demands the
obligatoriness (wujb) of avoiding punishment, or do you claim that
this obligation can be affirmed only by Revealed Law? If you hold the
former, then you accept that goodness and badness in the observable
are affirmed by the judgement of reason. If you hold the latter, then
it would not become obligatory upon one to avoid that punishment
except by another obligation. This [second] obligation will mean the
successiveness of punishments, which will result in an infinite regress
of punishmentswhich is absurd. Therefore, reason does make judgements of goodness and badness in the observable.73

As such, to speak of categories of value and of obligation as being

non-rational, but Legal, does not make sense. Revealed Law (as
representing divine command) cannot define obligation as such, but
provides some grounds (viz. prescriptions, alongside promises and
threats of afterlife consequences), which the mind deliberates upon.
Al-Rz writes in the Malib:
There is no meaning to badness in the Legal sense (qub shar#) except
[as follows]. Revealed Law tells [the agent], If you perform such-andsuch an act, you will become punishable for it. So his mind tells him,
Ought I to judge the avoidance of punishment obligatory, or ought
I not? If [his mind] judges it as such, rational goodness and badness
will be affirmed. If his mind does not judge it as such, then he will
need Revealed Law to obligate him to avoid punishment. The latter
case will be the same as the former, leading to infinite regresswhich
is absurd.74

Al-Rz, therefore, argues for a rational notion of religious moral

value, which relies on a process of rational deliberation, informed
by Revealed Law. That the human psyche ultimately seeks pleasure
and avoids pain is evident through pure reason (thbit f ma al#uql), whether Law (al-Shar#a) exists or notreason judges some
things good, others bad.75 Yet moral value is rational only in the
sense of being based on internal perceptions, grasped and reckoned

73
74
75

Ma#lim, 867.
Malib, 3, 28990.
Malib, 3, 290.

68

chapter two

by the mind, not in the sense of being rationally intuited, as the

Mu#tazila maintain.
Al-Rz thus abandons classical Ash#ar voluntarism categorically.
This follows from his more fundamental rejection of classical kalm
deontology altogether in a bold and definitive way, in which alGhazl before him shows little interest. Essentially, morality does
not consist of non-subjective duties and rules of conduct, whether
based on intrinsic moral qualities of acts, or on the command of a
certain lawgiver. Rather, the measure of moral value becomes the
subjective interests of the individual agent, though Revealed Law is
reinstated as an objective source for normative judgements.
One may wonder, however, whether this is not merely a nontraditional route to what is fundamentally a very traditional stance:
a classical Ash#ar divine command ethics. Is it not voluntarism in
consequentialist garb? And does this stance not lead to the traditional
view that normative ethics is practiced solely within ul al-fiqh, primarily through scriptural exegesis?
This is very much the conclusion that al-Ghazl reaches. The
primary sources he accepts for establishing rules of conduct are the
Qur"n, the adth, consensus and the method of analogical reasoning
(qiys).76 Though qiys will normally rely on rules prescribed explicitly
in one or more specific statements in the first two sources, al-Ghazl
maintains that it could be based also on more general guidelines that
are derived inductively from the wider body of scripture.77 These
concern the objectives (maqad) of the Law, which serve the interest
of humans in certain general ways, namely, the preservation of religion, soul, intellect, offspring and property.78 Yet al-Ghazl contends
that this form of qiys may be applied only in cases that fulfil certain
strict conditions, namely where a certain course of action is (a) necessary (arr) for preserving one of these five central objectives of the
Law,79 (b) universally (kull),80 and (c) beyond doubt. Conventionally,
76

The method of qiys involves the examination of a particular case to which a

scriptural ruling applies, with the aim of determining a Legal ground (#illa shar#iyya)
for the applicability of the ruling to that case. This ground may then be used to
establish further rulings on other, secondary cases, on which no scriptural rulings
can be found, and which share that aspect with the original case.
77 Al-Ghazl, Mustaf, 1, 3113.
78 Al-Ghazl, Mustaf, 1, 287.
79 This is contrasted to acts found advantageous merely for serving a need
(ja), or for improvement (tasn) and embellishment (tazyn) (al-Ghazl, Mustaf,
1, 286 ff.).
80 Al-Ghazl (Mustaf, 1, 296) gives the example of saving the entire Muslim

al-rz on the ethics of action

69

establishing a ruling by this method is not said to be an instance of

qiys, but an application of the principle of unsupported interest
(malaa mursala).81
According to al-Ghazl, where a course of action may serve an
interest that scripture neither endorses nor rejects, and that falls under
none of the five objectives of the Law, or fails to fulfil the conditions of unsupported interest, no Legal ruling in relation to it may
be established, as this would constitute unfounded legislation.82 For
al-Ghazl, therefore, guiding human action in accordance with the
consequences of acts (i.e. the normative principle of istil) must be
completely subservient to the primary sources of Revealed Lawviz.
the Qur"n and the adthmainly in the statements they make, and
secondarily in the central objectives of the Law that are derived
from them inductively. Even then, this approach may be reverted
to only in extraordinarily severe circumstances, and, as a form of
independent Legal reasoning (ijtihd), it will yield conclusions that
many jurists will have good reasons to question and reject. It appears,
therefore, that though al-Ghazl starts with a teleological metaethics,
his normative ethics of action is fundamentally a divine command
ethics, not a consequentialism.
Al-Rz takes a very different normative stance in his work on ul
al-fiqh, the Mal. He maintains that the aspect (waf) of convenience
(munsaba) in an act will constitute a valid ground (#illa) that may be
used in qiys (al-munsaba dall al-#illiyya).83 Convenience is defined in
terms of benefit and harm, defined in turn in terms of pleasure and
pain.84 Benefits, according to al-Rz, can be:
1. Worldly benefits, which divide into:
(a) Necessary (arr) benefits (viz. preserving the five objectives
of the Law: religion, soul, intellect, offspring and property).

community when attacked by an invading army, in contrast to the case of saving

the lives of a limited number of men, which will not be of universal concern.
81 Cf. Hourani, Ghazl, 847. On the notion of unsupported interest, see
Mohammad Kamali, Principles of Islamic Jurisprudence, 267 ff. It is common to translate
malaa mursala as public interest. Yet mursal does not mean public, but lacking the support of scriptural evidence; and indeed, as this notion is elaborated by
jurists, it normally concerns both public and private types of interest.
82 Al-Ghazl, Mustaf, 1, 3101.
83 Mal, 2/2, 247. On this, see p. 97100 infra.
84 Mal, 2/2, 218; Kshif, 513.

70

chapter two

(b) Benefits that are needed (ja), but do not reach the level of
necessity (e.g. allowing property rental).85
(c) Complementary benefits (tasn), which are neither necessary
nor needed (e.g. public morality).
2. Religious and other-worldly (ukhraw) benefits, served by spiritual
and moral discipline.86
The Lawgiver endorses (i#tabara) some forms of convenient action
and proscribes (algh, abala) others.87 In qiys, if a scriptural ruling
on a given act appears to serve a certain benefit, commensurate
rulings may be established on other acts that serve similar benefits.
What, then, of convenient acts that the Lawgiver neither endorses
nor proscribes? We are referred to a subsequent discussion in the
Mal, on unsupported interest.
So we turn to a chapter, presented after discussing the four primary Legal sources, on other types of Legal evidence that jurists
have disputed. The first type regards the original state (al) of acts.
Al-Rz writes:
Early in this book, we showed that no rulings (l ukm) [apply to acts]
before the Law [is established], and we responded to the objections
of opponents.88 Now we want to show, using Legal evidence, that the
original state for benefits (manfa#a) is permission (idhn), and for harms
it is proscription (man#).89

The view that benefits are originally permitted is implied in the

Qur"nic verses, He created for you all that is in the earth, and
Good things are made permissible to you,90 and by analogy with
the cases of benefiting from the light of someone elses lamp, and
resting in the shade of his wallpermitted acts that benefit the agent
without harming the other person.91 This evidence only confirms
the pre-scriptural condition of acts, where no Legal rulings apply.92

85

Cf. al-Juwayn, Burhn, 2, 923 ff.

Mal, 2/2, 2203; cf. Kshif, 523.
87 Mal, 2/2, 22630.
88 Cf. Mal, 1/1, 209 ff.
89 Mal, 2/3, 131.
90 Qur. 2:29; Qur. 5:4. Note that the former verse refers to the act of creation,
rather than divine command.
91 Mal, 2/3, 13143.
92 Mal, 1/1, 20921.
86

al-rz on the ethics of action

71

If no Legal ruling is presented in, or can be inferred from, scripture

with respect to a certain act, it will retain its default, pre-scriptural
permissibility (iba), in which case one will have a choice in performing it.93 As regards permissible (mub) acts, the role of Revelation is to confirm (taqrr), rather than alter (taghyr), their default,
pre-scriptural state.94
On the other hand, the view that harms are originally proscribed
appears to rest on scriptural evidence, rather than on some pre-scriptural, default condition.95 It is indicated by the adth, No harm shall
be inflicted or reciprocated in Islam.96 As such, benefits are prima
facie permissible; harms are prima facie proscribed. The second Legal
method that al-Rz discusses is the presumption of the continuity
of the initial state (istib al-l), which he accepts as a valid Legal
principle, confirming that unless a specific Legal ruling abrogates
the original state of an act, this state will persist.97
Yet he goes further when, after discussing other Legal methods,
he examines the principle of unsupported interest (malaa mursala),
citing the views of al-Ghazl and Mlik Ibn Anas (d. 179/796), one
of the earliest proponents of this principle. Again, this concerns acts
on which no ruling can be established using specific evidence from
the Qur"n, the adth and standard qiys. He provides the following division of acts with respect to the benefit or harm they involve:
(1) purely beneficial acts; (2) predominantly beneficial acts; (3) acts
that are equally beneficial and harmful; (4) acts that lead to neither
benefit nor harm; (5) purely harmful acts; and (6) predominantly
harmful acts.98 Necessarily, the Law will sanction the first two types,
will not sanction the third and fourth for being vain acts, and will
not sanction the last two for their harmfulness. Al-Rz goes on to
endorse this principle fully:
These rules, outlined in these six divisions, are known almost immediately (bi-l-arra) to be the religion that the prophets called to (dn

93

Mal, 1/2, 35960.

Mal, 1/2, 360.
95 Al-Rzs above statement on the original state of harmful acts contains
some ambiguity. Yet the essential purpose of this discussion in the Mal remains
normative, not theoretical.
96 Mal, 2/3, 1437. Cf. Kamali, Principles of Islamic Jurisprudence, 269.
97 Mal, 2/3, 148 ff. On istib al-l, see Kamali, Principles of Islamic Jurisprudence, 297 ff.
98 Mal, 2/3, 2223.
94

72

chapter two
al-anbiy") and the objective of revealed laws (al-maqd min al-shar"i#).
The Qur"n and the sunna indicate that, sometimes explicitly, at other
times on account of rules that are prescribed in accordance with what
we outlined.99

This principle is supported by rational and scriptural evidence, as

well as by the consensus of the Companions, who did not adhere
to the method of contemporary jurists in analysing the conditions
of qiys. Rather, they only considered benefits, because they knew
that the objective (maqad) of revealed laws is to serve benefits.100
It is thus acceptable to rely on this principle in jurisprudence.
In contrast to al-Ghazl, al-Rz endorses it unreservedly, without
laying down conditions that limit its application. By this principle,
it will become possible to go beyond the default condition of permissibility, by judging acts to be obligatory, recommended (mandb),
prohibited, or reprehensible (makrh) purely on the basis of their
consequences.
So, if scripture does not present an explicit rule in relation to a
given act, other normative methods and procedures may apply to it.101
Yet the most important are consequentialist, and may be summarised
as follows. (1) If the act is beneficial, it will be prima facie permitted;
if harmful, it will be prima facie proscribed. (2) A ruling on the act
may then be established by analogy with an existing scriptural rule
on another act, if both lead to similar consequences (provided that
the scriptural ruling is found to be connected to that consequence).
(3) If the act leads to a type of benefit or harm that scripture does
not address directly, then a ruling may still be established purely
on the basis of the acts consequences. The last normative principle
is essentially non-scriptural (despite the contention that scripture
sanctions it in principle).
Al-Ghazl lays down extremely strict conditions to the principle of
unsupported interest to the extent that he lists it among false normative methods: whoever uses it (uncritically) will be acting as lawgiver
(man istalaa fa-qad sharra#a).102 This stance practically results in a

99
100
101
102

Mal, 2/3, 223.

Mal, 2/3, 225.
See Mal, 2/3, 129 ff. Al-Rz rejects some of the methods he lists here.
Al-Ghazl, Mustaf, 2, 315.

al-rz on the ethics of action

73

standard divine command ethics.103 By contrast, al-Rz endorses a

thoroughly consequentialist stance in his ul al-fiqh, which he clearly
bases on his philosophical and theological metaethics.

Moral Obligation
The distinction between the moral is and ought was recognised by
classical mutakallimn. The Mu#tazila contended, not only that acts can
be intrinsically good (asan) or bad (qab), but also that unaided reason
knows (sometimes immediately) that human agents are obligated to
perform some good acts, which thus constitute duties (wjib), and to
omit some bad acts.104 On the other hand, though Ash#ars reject
objectivist conceptions of moral value entirely, they often declare
that their most serious point of contention with Mu#tazil ethics
concerns how moral obligations could be established by unaided
reason. By contrast, they maintain that both value judgements and
obligations are established by divine command.
Acts that, according to the Mu#tazila, constitute rational obligations include some that seem to affect none but the agent. One is
obligated, for instance, to omit vain acts and self-injury. However,
most rational obligations that the Mu#tazila affirm concern acts
that essentially do not serve the interests of the agent himself, but
relate purely to other creatures (especially human beings), or to God.
These include the obligation to enjoin good and forbid evil (al-amr
bi-l-ma#rf wa-l-nahy #an l-munkar), to thank the benefactor (shukr almun#im), and to omit lying, wrongdoing and wrongful killing.
The deontological view that certain acts constitute duties on
account of their intrinsic moral value is rejected completely in a
self-regarding consequentialism of the type that al-Rz upholds. In
line with classical Ash#arism, he maintains that theological reflection
(naar) and thanking the benefactor are not obligatory on the agent
prior to the reception of revelation.105 Notwithstanding, we find that
103 Unsupported interest was of course discussed by other pre-Rzian jurists.
Al-Ghazl is particularly relevant since he appears to be the first Ash#ar jurist to
present a consequentialist metaethics. One would expect him to develop a consequentialist normative ethics. Yet he does not.
104 #Abd al-Jabbr, Mughn, 14, 150.
105 The obligatoriness of rational speculation is based on Revelation (sam#),
contra Mu#tazils and some Shfi# and \anaf jurists (Muaal, 134; cf. Mal,

74

chapter two

al-Rz indeed uses the term wjib in a non-scriptural sense, viz. that
it is incumbent on the agent to avoid harms that may affect him.106
Al-Ghazl too uses wjib in this sense; George Hourani translates
it as prudentially necessary, and explains that it is predicable of
an act, when from the standpoint of self-interest its performance is
preferable to its omission in a decisive way, i.e. when severe and
certain harm to the agent is to be expected from omission of the
act.107 He then notes:
Any theory in which obligation is related entirely to the interest of the
agent is regarded as analysing obligation in a prudential, rather than
ethical, way; or rather as totally substituting the concept of obligation
with a totally different one.108

For both al-Ghazl and al-Rz, duties are imposed on human

agents firstly by scripture. Yet their subjectivist approach to Legal
obligation makes it an essentially private affair, based on the agents
desert of consequent reward or punishment in the hereafter for
his acts and omissions. An agents contravention of the precepts of
Revealed Law will be contrary only to his own interests, as it may
incur afterlife punishment on him alone in consequence. But why
should one then blame (dhamma) someone elses contravention of
Revealed Law when it does not affect him personally?
We find that, in contrast to a common trend in classical kalm,
al-Rz does not relate the notions of good, bad and obligatory to
the agents desert of praise (mad) or blame (dhamm) in this world for
his acts or omissions (instead of, or alongside, his desert of afterlife
consequences). Praise and blame are themselves acts in their own right,
performed by external observers, purely by subjective motivations
on their part.109 As such, they do not constitute essential aspects of
the moral value of the act praised or blamed.
Yet, having recognised Revelation as a source for norms, the
agent will accept its prescriptions, including duties that relate to
other humans and entities, while being in themselves of no worldly
consequence to the agent himself. Some prescriptions will obligate,
or encourage, the agent to direct praise or blame at certain acts
1/1, 2068; cf. al-f, Dar", 98). On thanking the benefactor, see e.g. Mal,
1/1, 193208.
106 Ma#lim, 86.
107 Hourani, Ghazl, 71.
108 Hournai, Ghazl, 73.
109 Malib, 3,22; 3, 342.

al-rz on the ethics of action

75

performed by other agents. In a late work on ul al-fiqh, al-Rz

explains the statement that an agent deserves (istaaqqa) praise or
blame as his being in a state for which it will be good (yasunu lah) that he be praised or blamed;110 and for an act to be good in
this sense, it will be commanded by God.
Similarly, the obligation to enjoin good and forbid evil becomes a
strictly Legal obligation (in contrast to most Mu#tazils).111 Countering
their view, al-Rz writes in the very early Ul al-dn:
Our view is that enjoining good and forbidding evil are made obligatory by consensus (ijm#). ... Al-Jubb" said that they are known to
be obligatory by reason. Ab Hshim said that if there is any harm
that relates to the servant in abandoning them, their obligation will
be rational; if there is no harm in that, they will become obligatory
by textual evidence alone.
We rely on the consensus of Muslims, prior to these opponents, in
commending the enjoining of good and censuring one who abandons
it. ... And we have already responded to the Mu#tazila as regards
judgements of goodness and badness.112

He then cites scriptural statements that affirm this obligation. As

expected, at the early, classical Ash#ar stage of Ul al-dn, al-Rz
completely rejects even the prudential notion of wjib, suggested by
Ab Hshim al-Jubb".
In one later objection to Mu#tazil moral realism, al-Rz appeals
to both the Legal and prudential notions of wjib. Following classical Ash#ars, he asks: If one hides a prophet from a mob that is
bent on killing him, will his lying to the mob about the whereabouts
of the prophet in order to save him be good, or (according to the
Mu#tazila) bad?113 Al-Rz argues that the man ought to lie for two
reasons. First, Muslims have a consensus on the obligatoriness of
saving the prophets life; one who can do this only through lying will
be obligated to lie. Second (a point not made by classical Ash#ars),
since that mob is likely to kill that person because of his hiding the

110

Ma#lim ul al-fiqh, 26.

E.g. al-Zamakhshar, Minhj, 778. Cf. Cook, Commanding Right, 195 ff.
112 Ul al-dn, fol. 332. Cf. Ishra (fol. 59b): Enjoining the good and forbidding
the evil, repentance, judgments of unbelief and dissoluteness, and the imamate,
are all Legal matters.
113 Nihya, fol. 197ab. Cf. al-Shahrastn, Nihyat al-aqdm, 372. #Abd al-Jabbr
(Mughn, 6/1, 342) argues that even in such situations lying will still be bad, and that
one ought to use insinuation (ta#r) instead (cf. Hourani, Islamic Rationalism, 78).
111

76

chapter two

prophet, it will be prudentially obligatory on him to lie.

The Mu#tazila, however, argue that not all observable human
behaviour can be reduced to self-interest. Rather, the observation of
human behaviour shows that humans often find themselves obligated
to act altruistically, purely with the purpose of benefiting others, not
for their own benefit. Does this behaviour not prove that humans
know through unaided reason that the performance or omission of
certain acts is obligatory? Rejecting this Mu#tazil explanation, alGhazl explains this behaviour as being motivated by simple psychological factors, mainly generic sympathy,114 desire for praise,115
habit and various dispositions.116 As we saw already, al-Rz too uses
these explanations of human motivation and moral judgment. But
does he go beyond this basic emotivism? Two relevant discussions,
in Nihyat al-#uql and the Malib, deserve especial attention.
In the latter work, al-Rz devotes one section to a crucial Mu#tazil
distinction, namely the division of motives into what is a motive of
need (d#iyat al-ja) and what is a motive of beneficence (d#iyat alisn),117 which stems from the agents wisdom (ikma), i.e. knowledge
of moral truths. He writes that the meaning of motive of need is
the pursuit to deliver good and benefit to oneself, and the meaning
of motive of wisdom is the pursuit to deliver good to others.118 AlRz of course recognises only the former, subjective and self-centred
type of motivation, to the exclusion of the latter, altruistic motive of
beneficence, which presupposes that the agent may act principally
and purely to deliver benefit to others, without any element of selfinterest. This would amount to moral objectivism.119
He declares that there are two main stations (maqm) at which
the Mu#tazil views on the motive of beneficence may be contested.
The first is metaphysical, and concerns the view that moral value is
intrinsic to actswhich will be the subject of the following section.
Interestingly, the second station concerns the question of whether
having knowledge of value judgements could motivate the agent
to act accordingly. What concerns us at this point is not the criti-

114
115
116
117
118
119

Al-Ghazl, Iqtid, 1701.

Al-Ghazl, Iqtid, 171; 174.
Cf. Hourani, Ghazl, 83.
Malib, 3, 6572. Cf. #Abd al-Jabbr, Mughn, 14, 24; 14, 38.
Malib, 3, 6970.
Malib, 3, 65.

al-rz on the ethics of action

77

cism, but the alternative explanation that al-Rz provides for the
phenomenon of beneficent action. This will provide insight into his
approach to non-Legal obligation.
He first affirms his commitment to his own theory of value:
We have shown that benefit is sought in itself, and that harm is avoided
in itself. All that leads to preponderant benefit is good, and there is no
other meaning to its goodness. All that leads to preponderant harm is
bad, and there is no other meaning to its badness.120

The Mu#tazila, however, point out that we know by our minds that
acts can be both bad and beneficial, or both good and harmful. AlRz lists four such cases that they cite. (a) Though the wrongdoer
benefits from his wrongdoing, he will still know that wrongdoing is
bad. Elsewhere, he cites the following Mu#tazil argument:
Though wrongdoing is harmful to the one wronged, it benefits the
wrongdoer. The establishment of justice will benefit one individual,
but harm another. Paying a deposit back will benefit the recipient, but
harm the giver. If these cases are judged by the inclination or aversion
in the disposition, the wrongdoer will judge wrongdoing good, while
the one wronged will judge wrongdoing bad.121

However, the Mu#tazila maintain that this is not the case. They also
cite (b) the case of a highly eloquent poem, which is composed well,
written in fine calligraphy and recited beautifully, but which contains
blasphemy. The mind (#aql) will recognise that such a poem is bad,
despite the pleasure that will be found at the level of the natural
disposition (ab#) in reading, and listening to, it.122 Also, (c) though the
liar benefits from his lying, he will know, by unaided reason, that it
is bad.123 (d) One who finds a sick blind man on the verge of death
in a desolate desert will know by reason alone that he is obligated
to help, even where he expects that his help will only burden him
and not benefit him in any way (e.g. he may expect never to meet
him again for his favour to be returned, nor to be praised by other
people, as no other person is present, nor to receive afterlife reward,
as he may be an atheist).124
120

Malib, 3, 66.
Nihya, fol. 199a.
122 Cf. Mal, 1/1, 169.
123 Al-Rz cites a similar argument from #Abd al-Jabbrs Qur"nic commentary
(Tafsr, 16, 222).
124 Malib, 3, 67.
121

78

chapter two

Al-Rz, however, contends that these observations are in fact

explainable from a consequentialist standpoint, and he advances
two lines of reasoning.125 The less important of the two is the traditional Ash#ar explanation, by reference to the motive of generic
sympathy, which he here appeals to only in relation to the example
of the blind man in the desert. Man is created (jubila) such that the
adverse conditions and pains he observes in other humans normally
produce a comparable pain within him, motivating him to help.126
Contrasting between divine mercy and human compassion, al-Rz
writes in his book on divine attributes:
Some sort of pity (ra"fa) has to appear in the servants heart for him
to act compassionately. If one contemplates [this, he will find] that
the only end (maqd) the servant can have is to avert [the pain caused
by] generic sympathy in his heart. In reality, he will be compassionate
towards others only in order to rid himself of the pain that sympathy
produces. God is exalted above this sympathy, and His compassion
(rama) does not rely on it, but will rely on pure favour and beneficence.
We will support this with some examples.
When a father does good (asana) to his child, he will actually be
doing good for himself. For if the wellbeing (mali) of that child is
undermined, the fathers heart will be pained. If he does good for the
child, his wellbeing will be preserved, and the pain will be negated
from the fathers heart. Therefore, the father does good for the child
only in order to realise that end for himself.
When a master does good to his slave, he does so only in order that
[the slave] may benefit him, so that he gains him profit and serves
him. Therefore, the masters end of that beneficence is the procurement of good for himself.
When man acts generously and pays his alms, he does so only to
gain fame among people as a generous and noble person, or to attain
reward and escape punishment in the hereafter. In reality, he does
good for himself alone.127

Such emotive, psychological explanations attribute acts directly to

simple motives in the agents psyche.
Yet, in the Malib, the main explanation that al-Rz advances
for beneficent action is more sophisticated. He addresses the first of

125
126
127

Malib, 3, 68.
Malib, 3, 69; Nihya, fol. 198b199b.
Lawmi#, 15960.

al-rz on the ethics of action

79

the above Mu#tazil examples as follows:

Were the wrongdoer to judge wrongdoing good, he would be unable to
avert wrongdoing being done against him. His life would be imperilled,
and his wealth liable to plunder. It will thus become obligatory on
him (yajibu #alayh), in considering the benefits of his soul and wealth, to
judge wrongdoing bad, in order that his soul and property be preserved
from annihilation and detriment.128

Al-Rz accepts that even wrongdoers will generally consider wrongdoing as such to be bad. But how would the agents interests be
undermined by his denial of such norms?
Al-Rzs response to the second Mu#tazil example then introduces
the most crucial notion in this discussion. Judging the beautiful,
blasphemous poem to be good, he argues, will be contrary to the
advantages of the world, or public wellbeing, (mali, or malaat
al-#lam) in various ways.129 First, if such blasphemy were tolerated,
Gods commands and prohibitions would lose their influence on
peoples hearts, leading to bloodshed and anarchy (al-harj wa-l-marj).130
Second, God is the highest of all beings and the most beneficent to
the needy. If blasphemy against Him were allowed, it would not be
possible, as al-Rz puts it, to use the aid of virtues (fa"il) in averting
harmswhich contradicts public wellbeing. Third, he argues in the
Mal that tolerating such poems would lead to the degradation of
the virtuous, which would be contrary to human wellbeing.131 All
three (non-Legal) grounds for judging the poem bad are non-deontological: the badness of blasphemy is in no way intrinsic.
This is in clear contrast to the classical Ash#ar approach, whereby
the blasphemous poem would be bad purely by Gods prohibiting
blasphemy. It is also noteworthy that al-Rz does not approach

128

Malib, 3, 68.
He also speaks of the order of the world (nim al-#lam) (Mal, 1/1,
180; Nihya, fol. 200a; Tafsr, passim.). Though the falsifa used this expression in
a cosmological sense, it was widely used in reference to social order, which is the
sense that al-Rz intends (e.g. al-Mward, Adab, 2, 242 ff.; Miskawayh, Tahdhb,
127; al-Tawd, Hawmil, 233; 310).
130 Elsewhere (Malib, 3, 291), he refers to this discussion: [We stated that]
wrongdoing, though beneficial to the wrongdoer, is extremely harmful to the condition of the world (wa# al-#lam). For if we were to judge wrongdoing good, bloodshed
and anarchy would prevail, and no one would trust ones spouse and partner. This
is one of the greatest harms.
131 Malib, 3, 68; Mal, 1/1, 170.
129

80

chapter two

this problem from his own Legal standpoint, despite its applicability; viz. that the agent finds the poem and its recitation deplorable
solely because tolerating and enjoying it would be contrary to his
own afterlife benefits.
His response to the third Mu#tazil example follows a similar line:
the liar benefits from his lying, but knows that lying is bad, since to
allow it would contradict human wellbeing. Lying would undermine
the bases for much of human action and pursuit, rendering them
futile. The person lied upon will waste his time, or in some cases
much of his life, suffering much anguish consequently. For this reason,
lying contradicts human wellbeing, whereas truth-telling preserves
and advances it.132
On the fourth Mu#tazil example (the blind man in the desert),
al-Rz writes:
Among the conditions that contribute to preserving human wellbeing
is inculcating the desire (targhb) of beneficence among people, with the
hope that were the like of that state to occur to [the agent] himself,
they would act compassionately toward him. Since this aspect was
found to serve public wellbeing, people evidently consented (ialaa) on
[considering] it good and [considering] omitting it bad. Since people
have become accustomed to this convention (iil), agreed upon it,
and adhered to it throughout their lives, these states have unmistakably
become well-established in their hearts and minds.133

We need now to look into the nature of this link that al-Rz makes
between moral norms and public wellbeing (a theme that does not
appear in classical kalm). First of all, it is clear that the preservation of social order and wellbeing requires a general acceptance
of these norms. However, it is not immediately clear, within a selfcentred consequentialism, how any individual would be motivated
to contribute to the preservation of social order by accepting such
altruistic norms. For these will be of no immediate benefit to him,
but may delimit his freedom to further his own interests, or may
even disadvantage him.
Al-Rz appears to hold that virtually every individual (including wrongdoers) will be aware of the fact that it is not in their own

132 Malib, 3, 68; Mal, 1/1, 170; cf. Tafsr, 16, 222: Lying undermines the
wellbeing of both the world and the self.
133 Malib, 3, 69. Contrast this to the Ash#ar explanation through generic
sympathy.

al-rz on the ethics of action

81

interest (nor in anyone elses) that anarchy prevails. From this, the
individual will realise that the preservation of social order will serve
his own interest (along with the interests of other members of society),
and will thus be necessary from a prudential standpoint. Al-Rz
writes in Nihya al-#uql:
If the wrongdoer declares (aft) that wrongdoing is good, someone
else will soon wrong him. Since, in his natural disposition, he hates
that anticipated wrongdoing, and since all that leads to what is hated
is itself hated in the natural disposition, wrongdoing will indeed be
hated by the wrongdoer. The same applies to all other [commonlyaccepted moral norms].134

And in the Malib:

The wrongdoer knows that if he were to declare that wrongdoing
is good, he would have conceded that others may wrong him. He
would then no longer have security for himself, his property and his
family. Therefore, since he knows that judging wrongdoing to be good
will open the gates to pains and grief upon him, he will undoubtedly
judge wrongdoing to be bad. The same is true in the case of lying.
As for judging beneficence and saving the poor from hardship to be
good, it [i.e. the judgement] will be good since it opens the gates to
benefits and goods.135

The individual will realise that the preservation of social order

requires that everyone, including himself, accepts such moral norms;
for to deny their validity would immediately undermine their general
acceptance. As the last passage indicates, the very act of pronouncing moral judgements becomes itself a good act by virtue of its
favourable consequences.
The agent thus finds himself motivated to consent to moral norms,
which will hinder his liberty to further some of his immediate interests, but will also preserve his more fundamental interests from being
undermined by others. Ultimately, the motivation to act beneficently,
in accordance with these norms, is neither altruistic nor based on a
sense of duty stemming from the agents knowledge of the intrinsic
goodness of beneficence, but is a self-centred, prudential motivation.
In Rislat Dhamm al-ladhdht, al-Rz writes that a human agent will
attempt to procure benefit for others, only on the condition that such

134
135

Nihya, fol. 199a.

Malib, 3, 349.

82

chapter two

help will bring benefit to him.136 He also writes:

What is sought by the animate being (ayawn) is pleasure only; what
is avoided is pain only. All that leads to the attainment of pleasure,
directly or indirectly, will be judged good by [the individual]. All that
leads to pain, directly or indirectly, will be judged bad by him.
Examples of the former are filial piety, fairness, justice and respect for
agreements; for if someone else does these things to him, he will find
them pleasurable and will benefit from them. Thus, for that expected
benefit, [people] consent (tawa#) to holding them good.
Examples of the latter are the badness of wrongdoing; for everyone
knows that if one were to judge it good, he might be wronged and
might experience pain himself. Therefore, [people] consent to holding
it bad in order to avoid that expected harm.137

After people agree upon (iala), or consent to (tawa#; ajma#),

these moral norms, the conventionality of these norms will be forgotten:
[The reason for the agents attempt] to save [the life of the blind man
on the verge of death in a desolate place] is that since people consent
(ajma#) to considering such rescue to be goodas it contributes to the
benefit of a stable world (malaat baq" al-#lam)[the agent] grows up
accepting that consensus (ijm#), becomes habituated to it, and never
hears anything that contradicts it. That [convention] will become like
an a priori proposition (qaiyya awwaliyya) [for him].138

An implicit social agreement lies at the background of widely-accepted

moral conventions, which is then, as it were, forgotten.
Both classical Ash#ars and al-Ghazl maintain that many moral
norms are mere habitual social conventions. Al-Rz takes a step
further by providing a rudimentary account of the socio-psychological
process by which norms appear and are sustained. In short, (a) the
individual realises, out of fear and prudence, the necessity of having
a widely-accepted set of norms and duties; (b) people, thus, consent
to a convention of moral norms, (c) which then become perceived
as objective truths, mainly through habituation. Despite the many
136

Dhamm al-ladhdht, 17.

Nihya, fol. 199b; cf. fol. 198b.
138 Nihya, fol. 200a. He continues: If we assume that [this act] is performed
by someone who has never heard any [moral] opinions, the cause [for his act]
will be the pain caused in his heart by generic sympathy. ... If we assume that it is
performed by one who is not [motivated] thus, we do not accept that, in this case,
he will consider rescue preferable to the omission thereof. Rather, how can he be
expected to act when saving is arduouswhich constitutes a deterrent!
137

al-rz on the ethics of action

83

seemingly altruistic acts observed in everyday human behaviour,

the underlying motives at work within human nature are necessarily
self-regarding, never altruistic:
It becomes evident ... that there is no evidence for any type of motive,
in the observable, except the one you [the Mu#tazila] term the motive
of need (d#iyat al-ja). As for what you term the motive of beneficence, its existence in the observable has never been proven.139

Al-Rz advances this rudimentary utilitarianism as an alternative

to the Mu#tazil theory of beneficence and moral obligation. In the
final chapter, we will see how he bases his political thought on this
same utilitarian stance.

Objections to Mu#tazil Ethics

Classical Ash#ar discussions of ethics have two primary concerns:
negatively, to refute the Mu#tazil view that value terms refer to real
attributes of acts; and positively, to prove that Revealed Law alone
has moral authority at the human level and that ethical value terms
refer strictly to its command and prohibition. Al-Rz has a different
positive concern, as he abandons the voluntarism of classical Ash#ars.
And though he shares their negativist concern, this theoretical leap
in his thought from voluntarism to consequentialism demanded a
commensurate leap in the dialectic with Mu#tazil ethical realism,
especially that al-Rz commits himself to a more thorough and
objective critical method than that of classical kalm.140
This development manifests most immediately in his criticism of
the Ash#ar schools failure to recognise and address the dominant
position in the Mu#tazil school. He writes in the Malib:
A Great many of the followers of Ab l-\asan al-Ash#ar, may God
have mercy on his soul, refuted ethical rationalism by saying: Had
killing been bad in essence (li-dhtih), every killing would, necessarily,
have been bad, and killing by way of Legal punishment would have
been bad. Had benefit been good in essence, every benefit and pleasure
would have been good; so fornication and sodomy would, necessarily, have been good. This not being the case, we know that claiming

139
140

Malib, 3, 70.
See Shihadeh, From al-Ghazl to al-Rz, 1649.

84

chapter two
the good to be good in essence and the bad to be bad in essence is
a false claim.141
Mu#tazils then say: This argument does not apply to what we say.
For we do not say that the good is good in essence and that the bad
is bad in essence. Rather, we say that the good is good on account
of aspects (wajh) that relate to it, and the bad is bad on account of
aspects that relate to it.
At this point, Ash#ars say that explanation with reference to aspects
is neither comprehensible (mafhm) nor conceivable (ma#ql).
We will here explain these aspects, such that what is intended in
them may become comprehended. We say: ... Man may want to take
a particular medicine. What determines that want is not the essence of
that medicine, but a combination of things, including the occurrence
of certain humours in the body, the unavailability of other medicines
and the appearance of symptoms of illness in the body. The combination of these states (l) will produce the want to take that medicine. If
you know this, know that there is nothing below the lunar sphere that
will necessitate anything by its essence (li-dhtih wa-li-#aynih). Rather,
necessitation and determination will occur at the combination of states,
aspects and conditions.
Mu#tazils, therefore, say: We do not say that what makes killing bad
is solely its being killing. Rather, what makes it so is the combination
of a number of things, namely its being an occurring pain, not for a
previous offence, nor for a subsequent compensation (#iwa). And so
forth with all good and bad things.
It thus appears that those who have relied on this argument to refute
the claims of the Mu#tazila did not know their position and did not
address all their views.142

Al-Rz himself appears to be aware of the central difference between

the two Mu#tazil positions from his earliest theological work, where
he distinguishes between the ethical absolutism of earlier Mu#tazilism
and the more sophisticated theory of aspects developed by later
Mu#tazils.143 The range and quality of objections he advances against
the theory of aspects will undergo immense progress, from the often
topical Ash#ar-style arguments in his earliest works, to the sophisticated dialectic developed in his later works, most importantly in
Nihyat al-#uql and the Malib, on which the present section will
focus.144
141 Cf. al-Juwayn, Irshd, 2334; al-Mutawall, Mughn, 43; al-Ghazl, Mustaf,
1, 57; al-Rghib al-Ifahn, al-Dhar#a, 272; Hourani, Reason and Tradition, 1313.
142 Malib, 3, 3389.
143 Ul al-dn, fol. 265.
144 Mainly Nihya, fol. 193b203a; Malib, 3, 279358; cf. Ma#lim, 859; Arba#n,
24653; Muaal, 479 ff.; Mal, 1/1, 1327; 1/1, 159219; Khamsn, 603; Tafsr,
passim. Cf. Kevin Reinhart, Before Revelation, 1634; and Sherman Jackson, The

al-rz on the ethics of action

85

Al-Rzs observation is correct: classical Ash#ar criticisms refuted

mostly the ethical ontology of what is actually the primitive ethical
absolutism of the Baghdd branch of Mu#tazilism. This is true also of
al-Ghazls criticism of Mu#tazil ethics, which is not fundamentally
different from the classical Ash#ar criticism.145 Yet the Baran branch
(which, unlike the Baghdd branch, still enjoyed a considerable following in al-Rzs time146) did not present a clear ethical ontology,
but an ethical epistemology: the Barans had much more to say
on the nature of our knowledge of morality than on the nature of
morality itself. Despite their denial that ethical value can be pinned
down metaphysically amid the various aspects upon which the
act takes place, which determine its moral status, the Barans will
maintain that it has an objective reality outside the minds perception
thereof. They will, therefore, dismiss classical Ash#ar objections as
irrelevant, not only since they constitute an outdated criticism of the
earlier Baghdd school position, but more fundamentally because
they relate to ethical ontology. On his part, al-Rz will address the
ethics of the Barans by developing the metaphysical line of criticism and, more importantly, by advancing what seems to be the first
epistemological criticism.
Ethical Metaphysics
One interesting argument concerning the metaphysical status of
ethical value can be found in the early Ishra:
The attribute that [the Mu#tazila] relate goodness and badness to is not
the acts being an accident, or created (mudath), or temporal (dith),
or existent, or will, or knowledge, or power.147 So it is an unidentified
attribute (ifa majhla). Yet it is unacceptable to resort to affirming this;
for affirming one unidentified attribute cannot be more acceptable
(awl) than affirming a third and a fourth, and so on to an infinity of
attributes.148 However, if such an attribute is not affirmed, it will be
inconceivable to ascribe goodness and badness to it.

Alchemy of Domination, 1914, both examining some objections to Mu#tazil

ethics in the Mal.
145 Cf. Hourani, Ghazl, 81 ff.
146 He writes that the Mu#tazila existing at his time are followers of either Ab
Hshim al-Jubb" or Ab l-\usayn (I#tiqdt, 42).
147 Cf. #Abd al-Jabbr, Mughn, 6/1, 77 ff.
148 On this argument ad ignorantiam, which al-Rz later considers fallacious, see:
Shihadeh, From al-Ghazl to al-Rz, 165.

86

chapter two
This being the case, the goodness or badness of things is not due to
the essence of the act, or to an attribute concomitant to it. ... Therefore, it must be an attribute of the agent. Yet this attribute cannot
be [a] the agents being a knower, for knowledge follows from what
is known, but does not affect it; nor [b] his having power, for power
only effects existentiating the act, which does not vary among good
and bad [acts]; nor [c] his having will, for that would make wrongful
killing ... good if the agent intended it to be good.149

In other words, if we assume that the perceived ethical value of

the act is real and objective, then what exactly is the nature of this
reality? Al-Rz argues that it is in fact unreal and inconceivable:
beyond the agents subjective judgement, no sign (athar) of ethical
value can be identified whatsoever.150 We are unable to perceive or
find goodness and badness outside ourselves. Along this general
line of criticism, he advances a number of narrower arguments
relating to specific acts or situations.
One such argument relates to the act of lying, defined as a statement (khabar) that is contrary to a true state of affairs.151 Al-Rz
argues that since a statement, as speech, consists of a string of discrete consecutive sounds, what exists of it will be the single sound
only, never whole words or phrases. Strictly speaking, the spoken
word does not exist, since when the speaker utters a sound, the
sound uttered immediately before it will cease to exist (though its
effect in the mind of the listener persists). The verbal lie could be
each individual sound (which is absurd), or the total (majm#) of all
the sounds of the statement (but this total does not exist). Therefore,
untruthfulness (kadhibiyya) and truthfulness (idqiyya) are not real
and objective (thubt) attributes of speech. Yet something that does
not exist cannot have an existent attribute, or be the ground (#illa)
for that attribute.
In the Nihya, al-Rz considers the possible objection that speech
expresses the mental judgement of the speaker, which, strictly speaking, is the lie. He replies that if this mental judgement refers to a
belief (in the mind of the liar) that is contrary to truth, it will be
ignorance, not lying. On the other hand, it could refer to the statement within the self, which could be willed; yet the Mu#tazila reject

149
150
151

Ishra, fol. 34ab.

Malib, 3, 291; 3, 334; cf. 3, 348.
Nihya, fol. 195ab; Malib, 3, 3356.

al-rz on the ethics of action

87

the notion of speech within the self (kalm al-nafs), a doctrine upheld
by Ash#ars, relevant mainly to the question of the uncreatedness of
the Qur"n.152
In the Malib, he cites another possible Mu#tazil counter-argument: that, within his own consequentialist framework, he himself
admits that speech may be judged as good or baddoes this not
contradict his argument? Al-Rz responds that he does not evaluate the act of speaking in itself. Rather, the meanings that lying
produces within the listener, which do not communicate to him a
correct understanding of the true state of affairs, may motivate him
to act in a way that is disadvantageous (e.g. if futile or harmful) to
him. For this reason, people will judge lying as such to be bad.153 As
such, consequentialist judgement does not face this same difficulty
as deontological judgement, since it does not concern the act itself,
but its subjective consequences.
Therefore, while the ethical objectivism of the Baghdd Mu#tazila
is refutable on account of the ethical ontological claims that they
make, the ethical objectivism of the Barans is refutable for their failure to justify their claims ontologically. As regards the metaphysical
status of morality, if the ethical ontology of the Baghdds is false,
the ethics of the Barans appears utterly nonsensical.
Ethical Epistemology
Nonetheless, al-Rz gives his opponents the benefit of the doubt. He
realises that Mu#tazil ethical theory is most seriously challenged at the
epistemological level, which becomes his primary focus. Thus the last
two sections of Book 3 of the Malib (introduced by al-Rzs above
criticism of classical Ash#ar objections to Mu#tazil ethics) provide
an account and a refutation of Mu#tazil ethical epistemology.154
He cites the common Mu#tazil view that the knowledge of some
ethical truths is immediate (arr), that of others acquired (kasb),
or inferential (istidll).155 For the latter type, the Mu#tazila rely on
arguments ad hominem (ilzm) and purported proofs (m jar majr al-

152 Al-Bqilln (Inf, 158) writes: Real speech is the meaning (ma#n) that exists
in the self, and has signs (amra) that indicate it, e.g. verbal and written speech.
153 Malib, 3, 3356; cf. 3, 349.
154 Malib, 3, 34158.
155 Malib, 3, 341; Nihya, fol. 198a; cf. #Abd al-Jabbr, Mughn, 6/1, 63.

88

chapter two

istidllt).156 After citing several such arguments in the Nihya, he

writes, They have other widespread arguments that are too weak,
and there is no use in recounting them.157 Indeed, we need only
to examine the discussion surrounding the view that some ethical
knowledge is immediate.158
Mu#tazils argue that knowledge of some ethical truths is immediate
and self-evident (badh) by citing numerous examples.159 They point
out that all rational beings have immediate knowledge, e.g. of: (a) the
goodness of thanking the benefactor, (b) the desert of blame for one
who throws a brick that injures another person (and that the thrower
of the brick, rather than the brick itself, is to blame), (c) the goodness
of helping the thirsty blind man in the desert, and (d) knowledge
of the goodness of truthfulness. Hence, any agent confronted with
a choice between truthfulness and lying, where the consequences
(this-worldly and otherworldly) for him personally of both choices
are absolutely equal, will necessarily choose truthfulness.160
In al-Rzs earlier works, including the Ishra and Nihyat al#uql, he follows a traditional Ash#ar critical strategy, which is to
deny that ethical judgements could constitute immediate knowledge.
For instance, he rejects the above Mu#tazil argument that every
agent will prefer to tell the truth over lying since his mind will know
immediately that lying is bad and truth-telling good. He argues that
this claim cannot be established, and that it is in fact conceivable
that the agent would choose lyingthere is no evidence to believe
otherwise.161
In this early stage, he also advances an argument under the indirect influence of Ibn Sn (apparently, at this early stage, through
al-Ghazl): that one can distinguish between immediate knowledge and conventional beliefs through a mental leap. According

156
157
158

Nihya, fol. 199b.

Nihya, fol. 200a.
Arguments for the other view are discussed in: Malib, 3, 3445; 3, 355

8.
159
160

Cf. Hourani, Islamic Rationalism, 20 ff.

This argument is made by #Abd al-Jabbr (Mughn, 6/1, 181 ff., esp. 214

5).
161

Ul al-dn, fol. 263; fol. 267; Ishra, fol. 34b35a; Mal, 1/1, 1801; Nihya,
fol. 201b. Cf. M. Marmura, A Medieval Argument for the Intrinsic Value of the
Moral Act, esp. 1237, an examination of this Mu#tazil argument and some classical Ash#ar refutations thereof.

al-rz on the ethics of action

89

to Ibn Sn, it is possible to conceive mentally, through a process

of abstraction, of a pristine state in which one is detached from all
acquired dispositions, notions and inclinations. It will then be found
that while the mind has immediate knowledge of the truth of the a
priori (awwal) proposition, 2 is greater than 1, and will not be able
to doubt its truth no matter how hard it tries to, it will easily doubt
the truth of the proposition, Lying is bad, which, by contrast, is a
mere widely-accepted statement (qaiyya mashhra).162 In other words,
such moral propositions are beliefs that stem from social convention,
and do not constitute immediate knowledge.
In anticipation of such objections, the Mu#tazila advance a similar
argument: that if we imagine a situation in which the agent is completely free from social conventions and self-centred inclinations, he
will still reach these ethical truths and will choose truth-telling over
lying. Interestingly, in Nihyat al-#uql, al-Rz will respond to this
argument by going further than all previous critics of the Mu#tazila.
He does not only object that it is in fact possible for the agent to
choose lying in the imagined situation and that one who examines
himself will find that the statement Lying is bad is a mere conventional belief, he also explains why the argument is fundamentally
flawed, as follows:
[The Mu#tazil] says, If we assume (faraa) the above state, we will
find ourselves certain about preferring truth-telling over lying. We
say: What we find now within ourselves cannot be used as evidence
for your contention. For from the time we were born to the present
we heard nothing but affirmations that truth-telling is good and lying
bad. Thus the badness of lying and the goodness of truth-telling became
rooted in our minds. This disposition (hay"a) became extremely firm
and entrenched, such that although we may assume the disappearance
of these dispositions, they do not disappear by that mere assumption. Rather,
they remain, despite our assumption, as they were. Therefore, with
this assumption, we judge truth-telling to be preferable over lying only
because our normal dispositions, which remain within us despite our
assumption, will determine this preference. However, if we posit a man
who is free from all these dispositions, we will not know whether he
will prefer good over bad. For we are unable to grasp such a state;
what we are capable of doing is only to assume our selves to be free from those

162 Ishra, fol. 35a; Nihya, fol. 198b; cf. al-Ghazl, Mi#yr, 14950; Mustaf,
1, 49; Ibn Sn, Ishrt, 1, 1878. On al-Ghazls borrowing from Ibn Sn in
this regard, see M. Marmura, Al-Ghazl on Ethical Premises, 395; Hourani,
Ghazl, 81.

90

chapter two
dispositions, rather than to make them actually free from them. The experience
(tajriba) occurring at the assumption of being free [from dispositions] is
other than the experience occurring at being truly free [from them]. We
experience the former case, though what is conducive to our purpose
is the latter. Therefore, this [Mu#tazil] argument is false.163

Mu#tazils may posit situations in which an agent acts morally without

the interference of any conventions or self-centred motives. Nonetheless, the actual mental leap required to conceive of the situation posited is, according to al-Rz, impossible. Therefore, even
if one scrutinises these ethical notions in this way, and still finds
them indubitable, this will not prove that they constitute knowledge,
immediate or otherwise.
Clearly, this stance will equally undermine Ibn Sns method to
distinguish widely-accepted statements from statements of a priori
knowledge. Al-Rz cites this method in his very late work Shar #Uyn
al-ikma: If I assume (faraa) myself as if I have just been created,
have never interacted with anyone, and have never seen anything,
and if I assume (qaddara) that the influence of custom and habit (ilf
wa-#da) is omitted from myself, then when I present to my mind
the statements 1 is half of 2 and Lying is bad, I will find my mind
certain (jzim) in relation to the former, incapable of judgement
(mutawaqqif) in relation to the latter.164 This, al-Rz here replies,
is a very weak argument; for the individuals habit, custom and
character traits may make some common beliefs as indubitable within
the self as a priori knowledge. One may scrutinise examples of both
types of statements and find no difference in the selfs certainty in
relation to them. Indeed (as he objects to the Mu#tazil view), many
character traits and acquired habits and notions cannot be omitted
merely by assuming the omission.165
In his later works, therefore, al-Rz puts less emphasis on denying that normative judgements constitute immediate, or self-evident
(badh), knowledge. (By contrast, in his very early Ul al-dn, he states

163

Nihya, fol. 201b.

Shar #Uyn al-ikma, 1, 199. The argument originally is in the plural: If
we assume ourselves....
165 Shar #Uyn al-ikma, 1, 199200. This marks a crucial development in alRzs epistemology generally, not only in his ethical epistemology. As we will see
in the final chapter, his contention that it is often impossible to distinguish a priori
knowledge from conventional belief will contribute to his arrival at a moderate
scepticism towards the end of his career.
164

al-rz on the ethics of action

91

that the primary concern in the context of ethics should be to refute

the Mu#tazil claim that one knows some moral truths immediately.166)
This later critical approach will indeed suit his later consequentialism,
in which he rejects only the Mu#tazil theory of the meanings of moral
judgements, not the view that these judgements may constitute selfevident knowledge. Thus, that the statement, Wrongdoing is bad,
is immediately true does not entail that wrongdoing is intrinsically
bad because it is wrongdoing. One should ask: Why is this statement true? Is it because wrongdoing is bad in itself? Or because of
another reason, e.g. that it somehow leads to harm?
This transition in al-Rzs ethics is evident in an argument he
presents in the Malib. He first cites the classical Ash#ar objection
that self-evident knowledge cannot be disputed by rational people;
yet both Ash#ars and falsifa deny that acts are good or bad in
themselves.167 He then cites Ab l-\usayn al-Bar, who responds
that though denying immediate knowledge is inconceivable for the
multitude of people, it is conceivable for small numbers of people
to deny it.168 All common people (common Ash#ars included) hold
that beneficence is good and wrongdoing bad, whereas specialist
Ash#ars, who deny this, are few in number. Al-Rz retorts that what
is at issue is not whether the majority of people make these judgements, but what meanings (tafsr) these judgements have. He turns
Ab l-\usayns empirical argument against him: that in fact what
common people (common Mu#tazils included) refer to by ethical
value terms, in ordinary language, are benefit and harmcontra a
handful of specialist Mu#tazils.169
The main point of contention is no longer whether one knows the
truth of ethical judgements immediately, but what conceptions ethical value terms refer to. Al-Rz, therefore, goes beyond the classical
Ash#ar preoccupation with denying the truth of ethical statements of
assent towards analysing the value conceptions that constitute these
statements. He presents a more sophisticated criticism of Mu#tazil
ethical realism without undermining his own consequentialism.

166

Ul al-dn, fol. 264.

Malib, 3, 351. The argument can be found in al-Rzs early Ul al-dn,
fol. 262; al-Juwayn, Irshd, 22930; al-Mutawall, Mughn, 43; al-Ghazl, Mustaf,
1, 57. Cf. Hourani, Ghazl, 812.
168 Cf. #Abd al-Jabbr, Mughn, 6/1, 23.
169 Malib, 3, 3512.
167

92

chapter two

Al-Rzs most thorough critical analysis of value conceptions

can be found in the beginning of his discussion of ethics in Nihyat
al-#uql. He examines the definitions that the Mu#tazila present for
the central ethical value terms:
Ab l-\usayn says: Bad is what one who is capable of performing
it and who knows its badness may not perform (laysa la-hu an yaf #alah).
Our saying, He may not perform it, is known by reason and does
not require explanation. This entails that he will deserve blame for
performing it. It is also defined as what has an attribute that is effective with respect to the desert (istiqq) of blame....
In defining good, he says that it is what one who is capable of
performing it and who knows its [moral] status may perform.170

Al-Rz asks: What does saying, one may (la-hu)171 or may not
perform a particular act, mean? It could have a number of senses,
including: (1) permission for, or prohibition against, performing it;
(2) capability or inability to perform it; (3) a certain attribute of the
agent, as we say of a body that it has (la-hu) the attributes blackness and motion; (4) the acts being good or bad, which entails
that the agent may or may not perform it accordingly; and (5) some
sort of permission or prohibition that issues from unaided reason.
Yet the first sense will result in a divine command notion of moral
value, which Ab l-\usayn does not intend. The second and third
are obviously not intended. The fourth is a tautology. In the fifth,
if rational prohibition is other than the agents desert of blame or
punishment (as in the second definition), then what is it exactly?
What else, al-Rz adds, could la-hu here possibly mean? This crucial element in Ab l-\usayns definition appears meaningless and
renders the whole definition meaningless.
The thrust of al-Rzs argument is that rights and duties depend
on the moral status of acts. One would normally answer the questions, What may I do?, What ought I to do?, What ought I
not to do?, etc., by reference to what is good or bad, but not vice
versa. To say that the bad is what one may not do presupposes that
one knows what one may not do and why one may not do it. Yet
knowing why one may not do it is to know why it is bad; al-Rz

170

Nihya, fol. 193b194a; cf. Mal, 1/1, 1326; al-Bar, Mu#tamad, 1, 363

69.
171

La-hu is difficult to translate into English while preserving its various senses
and connotations. I find may preferable to has the right to, or is entitled to.

al-rz on the ethics of action

93

points out that Ab l-\usayn commits circularity by including the

agents knowing the goodness and badness of acts in his definitions
of ethical value terms.172
Al-Rz then turns to the second (and more common) Mu#tazil
definition of ethical value, in terms of desert (istiqq) of praise or
blame. The term istiqq, he argues, is derived from aqq, which,
in Ibn al-Malims book of definitions, is given a number of senses
depending on context: (a) the lexical meaning of aqq is what is
true and real (thbit); (b) in the conventional idiom of ordinary usage
(#urf), it refers to correct belief and truthful statement; (c) in religious
Legal convention (#urf al-shar#), it refers to each circumstance of the
agent that will make it good for him to produce a certain act. The
first and last senses cannot be intended by desert in the Mu#tazil
definitions of good and bad. By interpreting desert according to
the second sense, goodness will be defined in terms of goodness, and
Ab l-\usayn will fall again into circularity.173
Al-Rz concludes that despite their claims that the conceptions
of moral value terms are self-evident and known immediately to the
mind, the Mu#tazila fail to define them. As their notions of ethical
value are incomprehensible, their ethical realism as a whole will
appear profoundly irrational.
Whence, his main criticism of Mu#tazil ethical epistemology in the
Malib and Shar #Uyn al-ikma. Though he accepts that knowledge of
the goodness or badness of some acts is indeed immediate, he argues
that a statement of assertion (tadq) will require prior conception
(taawwur) of both its subject and predicate.174 If the words good
and bad in the statements, Beneficence is good and Wrongdoing
is bad, refer to favourableness and unfavourableness to the self, then

172 #Abd al-Jabbr defines bad as what the agent deserves blame for (Mughn,
6/1, 7; 6/1, 267) and rejects defining it as what the agent may not perform since:
(a) there are bad acts that one cannot avoid performing, e.g. the acts of the child
and the sleeper (these further highlight the ambiguity of the phrase, laysa lahu an
yaf #alah); (b) knowledge that the agent may not perform an act is more properly an
entailment (ka-l-tbi#) of its badness (cf. al-Rzs objection); and (c) this definition
does not unveil what [the act] is bad on account of, and does not draw attention
to the judgement (ukm) connected to it, viz. the desert of blame.
173 Nihya, fol. 194ab.
174 Cf. Muaal, 81; Mulakhkha, fol. 1b2a.

94

chapter two

these conceptions are indeed self-evident and known by the agent

immediately. However:
If [Mu#tazils] intend, by beneficence being good and wrongdoing being
bad, something other than the consideration of benefit and harm, then
that is not conceived (mutaawwar), let alone being asserted (muaddaq
bih), or the assertion of it being self-evident.175

It is the conception of value terms, such as good and bad, that

is at issue. If one examines these conceptions, one will be able to
conceive of them in terms of pleasure and pain, but, as we saw,
not in any objective sense. This key criticism of Mu#tazil ethical
realism captures the essential distinction between the ethical theories of al-Rz and the Mu#tazila: it is the station of truth that
the intelligent critical investigator ought to contemplate, so that the
truth in this topic becomes manifest to him.176 Instructively, this
criticism immediately follows his dismissal of classical Ash#ar refutations of absolutist ethics as being misinformed, and it introduces an
account of his own alternative, viz. his consequentialist conceptions
of ethical value.
In another argument, al-Rz examines the conception of the
subject of such statements as Wrongdoing is bad. He quotes Ab
l-\usayns Ghurar al-adilla177 on its definition:
Our masters have said: Pains become good if they have certain conditions. These include [1] that they be deserved; [2] that they [lead to] a
benefit that preponderates over the pain; [3] that they involve averting
a greater harm; [4] that they be supposed to [lead to] benefit and the
averting of harm; [5] that they be performed for the purpose of averting
[harms];178 and [6] that they be performed by way of habit, or they
be considered as performed by other [than the immediate agent].179
So, if pain has any of these conditions, it will become good, rather
than wrongdoing. If it has none, it will be wrongdoing. Therefore, we

175

Malib, 3, 3478; cf. Shar #Uyn al-ikma, 1, 201; 1, 237.

Malib, 3, 348.
177 The edition of the Malib has al-Qudar, instead of al-Ghurar.
178 This seems to concern harm directed at others, whereas the third condition
concerns harm directed at oneself.
179 An example of number 6 is one who places an infant into snow. Although
the pain suffered by the infant from the coldness of the snow will be either created
by God immediately or generated from the coldness of the snow, the responsibility
for the pain will be that of the human agent. Gods preservation of natural habit
will still be good.
176

al-rz on the ethics of action

95

have restricted wrongdoing to being [1] undeserved harm, [2] [not

leading to] a preponderant benefit, [etc.]...180

Only after knowing all these complex details, after a process of subtle
and specialised speculation, will one grasp the complex conception
of wrongdoing. However, al-Rz objects:
The mutakallimn are unanimous on that whenever knowledge of the
principle (al) is discursive (naar), it will be inconceivable for derivative knowledge (far#) to be immediate. Knowledge of the essence of
wrongdoing is undoubtedly a principle for knowledge that it is bad.
Knowledge of the essence of wrongdoing can only be attained through
this definition that only critical investigators can conceptualise by the
means of intricate speculation. Since knowledge of this principle is
discursive, it will be inconceivable for knowledge that wrongdoing is
bad to be self-evident.181

In other words, it follows from the Mu#tazil claim (that such ethical statements are self-evidently and immediately true) that one
who states, Wrongdoing is bad, without having a conception of
wrongdoing, will effectively be saying: Wrongdoing, which I do
not know what it is, is self-evidently bad. If the conception itself
is unknown, then no attributes of it can be self-evident. For selfevident assertion requires only the presence of the two conceptions
of subject and predicate in the mind, and will follow from their
presence immediately, without syllogistic intermediation. If at least
one of these conceptions is not present in the mind, the truth of
any statement of assertion that involves both cannot possibly be
known self-evidently.
The Mu#tazil may then modify his position by contending that
the badness of wrongdoing will be known immediately by one who
already conceives the essence of wrongdoing, not by everyone. This,
according to al-Rz, will be to admit that the badness of wrongdoing is not self-evident universally for all rational humans, as the
Mu#tazila originally claim. And indeed he rejects even this modified
position.182
Lastly, al-Rz also cites the Mu#tazil argument that when we

180 Malib, 3, 353; cf. #Abd al-Jabbr, Mughn, 13, 298. On Mu#tazil views on
how the infliction of pain becomes bad, cf. M. Heemskirk, Suffering in the Mu#tazilite
Theology, 122 ff.
181 Malib, 3, 354; cf. Nihya, fol. 198b.
182 Malib, 3, 3545.

96

chapter two

know [the act] to be wrongdoing, we will know that it is bad, even

if we are heedless of all other considerations; therefore, the ground
(#illa) for this badness has to be wrongdoing.183 Wrongdoing is the
ground (muqta) for the badness of wrongdoing, since knowledge
of badness is coextensive (d"ir), both affirmatively and negatively,
with knowledge of [the act] being wrongdoing.184 He objects that
the concomitance of two notions is not evidence for a direct causal
link (#illiyya) between them. Indeed, the ground of the concomitance
of the conceptions wrongdoing and bad may be something other
than wrongdoing itself, e.g. sensations of pain and pleasure within
the agents experience of actual instances of wrongdoing.

God and Ethics

In the beginning of his discussion of ethics in the Malib, al-Rz
writes:
The Mu#tazila and the Karrmiyya are unanimous on the affirmation
of ethical rationalism (tasn al-#aql wa-taqbih), while the falsifa and
the Determinists are unanimous on denying it. Our position is that it
applies to human beings, but does not apply to God.185

The ultimate concern of al-Rzs kalm discussions of ethics is

the ethical status of Gods acts, more so than human acts. Yet,
as mentioned, the analysis focuses more on the more fathomable
observable human level, before finally moving to the unobservable, supra-human level. In Chapter 1, we saw that al-Rz develops
a theory of human action and motivation, concluding that divine
action cannot follow from similar processes. As such, his view of
the ethical status of divine action is in overall agreement with the
central Ash#ar doctrine that no cognitive ethical judgements can be
made in relation to God, and that He is not under any obligation
to perform, omit, command or prohibit any acts. Previous Ash#ars
based this view on their anti-realism and voluntarism: good and
bad refer not to real attributes of acts, but to divine command

183
184
185

Malib, 3, 355.
Mal, 1/1, 171.
Malib, 3, 289.

al-rz on the ethics of action

97

and prohibition. Al-Rz, by contrast, bases it on his anti-realism

and consequentialism.186
The Problem of Divine Motivation
The motive to perform or omit a certain act will ultimately involve
an expectation of a preponderant benefit or harm, and hence of the
experience of a preponderant pleasure or pain. According to al-Rz
and classical Ash#ars, since God does not experience pleasure or
pain, it will be inconceivable for Him to be subject to prudential or
pleasure-seeking considerations. His acts and commands, therefore,
cannot follow from motives.187
Moreover, whoever acts for an objective (ghara), i.e. with a motive,
will be seeking to be perfected (mustakmal) by it, and will be imperfect
in himself; however, God is absolutely perfect. It may be objected
that though performing or omitting an act will be on a par for God
Himself, He may perform the act for the benefit of humans. Yet,
since al-Rz denies that beneficence can in itself constitute a motive,
he simply reiterates that if the performance and omission of such an
act are absolutely on a par for God, He will not have a motive to
perform it. Otherwise, it would follow that, by benefiting humans,
God would aim to perfect Himself.188
Also, al-Rz rejects the notion that the objective of divine action
and command is to realise the advantage and wellbeing of humans,
because God is able to create pleasure in humans immediately. It
follows that the intermediation of His acts and commands will be
superfluous and vain (#abath).189
According to al-Rz, God may choose to benefit some, all, or
186 This general stance does not appear to be entirely novel (cf. al-Juwayn,
Nimiyya, 267; al-Ghazl, Iqtid, 1747; Ibn Sn, Risla f l-qa" wa-l-qadar,
dedicated to refuting Mu#tazil ethics, theodicy and doctrine of free human choice).
But al-Rzs treatment is much more developed.
187 Cf. al-Bqilln, Tamhd, 301.
188 Cf. Mal, 2/2, 1857; Mulakhkha, fol. 156ab; Muaal, 483; Arba#n, 24953;
Khamsn, 613; Ma#lim, 878; Malib, 3, 2901. Al-s (Talkh, 205) comments
that al-Rz takes the view that whoever acts for an objective is perfected by it
from the falsifa, but uses it in a different context. Also, part of the background
to this doctrine, though of secondary importance, is Ibn Sns view that higher
causes never act for the sake of lower entities (cf. Najt, 3048; 320; Shif", Ilhiyyt,
2, 3935; 2, 4145; Ishrt, 3, 150; 1545).
189 E.g. Mal, 2/2, 188; Muaal, 483.

98

chapter two

no humans. He may communicate to them the courses of action

that He chooses for Himself, including rewards and punishments
promised for human acts and omissions. But His choice will be
absolutely free and neither directed at ends, nor subject to moral
guidelines. His acts and commands are not teleological, though not
vain either.190 Al-Rz seems to reach a point where the nature of
divine action becomes very much an unfathomable mystery. We do
not understand why God acts and commands. Nor do we understand
how God chooses to act and to command, especially that al-Rz
will deny the divine attribute of will.
In one kalm work, he writes, It is inconceivable for God to do
something for an objective, contra the Mu#tazila and most jurists.191
This reference to jurists draws attention to one of the most serious problems that transpire from this doctrine, which concerns the
relation of divine command to human wellbeing. Is there an ethical rationale in divine command? And how could this doctrine be
reconciled with al-Rzs foregoing contention that Revelation serves
human advantage?
In Nihyat al-#uql, he notes that despite this theological position,
most divine commands do in fact accord with human wellbeing.192
The reader is then referred to a closer treatment of the problem in
his work on the principles of jurisprudence the Mal, which we
indeed find in the context of establishing the principles for the Legal
method of analogy (qiys).
Al-Rz considers three definitions for the Legal ground (#illa
shar#iyya), central to the method of qiys: (a) it could somehow be a
cause (mu"aththir, mjib) for establishing a ruling, or (b) a motive (d#)
for it, or (c) it could merely indicate (mu#arrif ) it.193 He rejects the first
for a number of reasons; e.g. that revealed rulings are Gods preeternal, uncaused speech, which is not affected by any factors (such
as the intoxicating property of wine). He rejects the second definition, since God cannot be motivated in His action and command.
He then accepts the third definition for the ground, as the aspect
(waf ) of the act, which functions as a ratio (sabab) for indicating that

190

Muaal, 483.
Muaal, 483; cf. Arba#n, 249.
192 Nihya, fol. 200b.
193 Mal, 2/2, 17990; cf. the good commentary by al-Ifahn, Kshif, 6,
289301.
191

al-rz on the ethics of action

99

a particular Legal judgement applies to the act.194 This last notion

of the Legal ground is non-theological; a ground tells us something
about the contents and implications of scriptural texts, but nothing
about their source. Scriptural rulings provide knowledge of human
conduct, but not of divine nature. The first two definitions, by contrast, rest on theories of why God commands and prohibits certain
acts, viz. that God is either motivated or necessitated to establish
scriptural rulings in the form in which we have them.
Grounds are of several types. Some, for instance, are specified
more or less explicitly in scripture, where it is stated that a given
act is commanded or prohibited for the sake of (min ajli), or for (li-),
a certain reason.195 These are straightforward scriptural indications
of rulings, which fit the above definition of the Legal ground easily.
However, another, more problematic type includes grounds based on
the previously-discussed principle of convenience (munsaba), which
concerns the consequences of acts. Al-Rz dedicates a lengthy section in the Mal to establishing the validity of convenience as a
Legal ground without implying that it is based on divine motivation. One may argue as follows: God establishes Legal rulings for
human benefit; we observe that ruling x serves benefit y; therefore,
the presumption (ann) that ruling x is established for the sake of
benefit y will follow. Obviously, the most problematic part of the
argument is the premise, God establishes Legal rulings for human
benefit. Al-Rz notes that jurists absolutely reject explaining for
human benefit as with the objective (ghara) of realising human benefit,
although both phrases have the same meaning. There is no real difference, therefore, between their position and that of the Mu#tazila,
who proclaim explicitly both that God is obligated not to perform
bad acts nor to make bad commands, and that He is motivated to
deliver advantages to men. Jurists also assert that although God is
not obligated to consider [human] benefits, He does what is beneficial
for His servants only out of favour (tafaul) and beneficence (isn),

194

Al-Rz gives the example of fornication, which is the ratio for the desert of
punishment. The link between act and punishment is based on divine command:
Whenever you see a man fornicating, then know that I have made it obligatory
(awjabtu) that he be punished. So fornication merely indicates the applicability of
punishment, but does not cause it (Mal, 1/1, 13940).
195 Mal, 2/2, 191 ff.

100

chapter two

rather than obligation (wujb).196 In refuting this argument for the

validity of the principle of convenience, al-Rz argues at length
against this problematic premise in particular; for Gods commands
and acts cannot be aimed at human benefits.197
In the Mal, he presents the following alternative theological
basis for the validity of the principle of convenience:
We accept that Gods acts and rulings are not explainable by motives
and objectives. Nevertheless, we assert that convenience provides a
basis for the presumption of grounds. This is so, since the position
of Muslims is that the rotation of the spheres, the rising and setting
of constellations, and their persistence upon their forms and luminescence, are not necessary (wjib), but that since God, exalted, preserves
His custom (#da) by sustaining them upon a uniform state, there will
undoubtedly follow the presumption that they will remain tomorrow
and the day after tomorrow upon these attributes. The same is true in
the cases of the fall of rain at (#inda) [the formation of] humid clouds,
satiation after eating, quenching of thirst after drinking, and burning
from contact with fire. Yet since custom persists continuously in this
manner, there will undoubtedly follow a presumption, verging on certainty, that they will persist in their normal courses (minhj). In short,
the repetition of one thing many times will entail the presumption that
whenever it occurs, it will only occur according to specific aspects.
Therefore, when we examine revealed laws (shar"i#), we will find
rulings and benefits concomitant and inseparable. This is known after
[examining] the various aspects of revealed laws inductively. Therefore,
knowing the occurrence of either will entail presuming the occurrence
of the other, and vice versa, without either being a cause (mu"aththir)
for the other, or a motive for it. Therefore, convenience is [valid]
evidence for Legal grounds, though it is denied absolutely that Gods
rulings may be explainable by recourse to objectives.198

Al-Rz compares divine command to divine action: the Law to the

created world. As one can detect a divinely-sustained uniformity
in the behaviour of created things, which is empirically qualifiable
and quantifiable, one will find a comparable uniformity in scriptural rulings. Neither observation, according to al-Rz, will entail
that God is motivated or necessitated (logically, metaphysically, or
ethically). That scriptural rulings generally serve human wellbeing
does not entail that God commands for the sake of human benefit,
whether out of duty or favour. Nonetheless, this observation will

196
197
198

Mal, 2/2, 23746.

Mal, 2/2, 24871.
Mal, 2/2, 2467.

al-rz on the ethics of action

101

allow humans to derive general teleological principles inductively

from scriptural rulings; hence, the validity of Legal qiys using the
principle of convenience.
Objections to Mu#tazil Theological Ethics
The Mu#tazila developed highly elaborate doctrines on the ethical goodness of Gods acts (including creation, obligating humans
(taklf ) and afterlife reward and punishment), which they maintain
to be cognizable. Their critics confront them with a wide range of
arguments, mostly unsystematic sceptical and topical ones that take
the form of analogical, parable-like and situational arguments ad
hominem (ilzm). These were commonly used by classical Ash#ars (the
best-known being al-Ash#ars own three brothers problem, which
he advanced against Ab #Al al-Jubb"199), as well as the falsifa,
most notably Ab Bakr al-Rz and Ibn Sn.200 Let it suffice here
to provide some representative examples of the arguments that alRz uses.
One argument in the Malib addresses the Mu#tazil notion of
divine beneficence (in#m, isn).201 Al-Rz argues that an act is said to
be an act of beneficence only when the recipient of benefit has prior
need and desire for it. A dog will find pleasure in a bone, but has no
use for a precious necklace. Yet benefit is connected ultimately to
the recipients conscious experiences, and is subjectively measurable
in terms of the amount and quality of the pleasure experienced, or
the pain alleviated, rather than in terms of external things, which
one can accumulate unlimitedly. Therefore, a persons potential
for experiencing pleasure will be delimited by psychological factors,
which are in turn determined by the nature and extent of his needs
and desires. One cannot experience pleasure without having prior
need and desire for it.
Yet need and desire, al-Rz argues, are imperfections in an entity
and in themselves harms and sources of pain and anguish to it.202 If

199

See my forthcoming article on the debate between al-Ash#ar and al-Jubb"

on this problem, and on al-Rzs use thereof.
200 Esp. Ibn Sn, Risla f l-qa" wa-l-qadar, from which al-Rz seems to borrow
one argument (Malib, 3, 317).
201 Malib, 3, 2916.
202 Malib, 3, 2934; cf. p. 156 ff. infra.

102

chapter two

God creates man with numerous imperfections, then bestows upon

him commensurate pleasures, then (according to Mu#tazil ethics)
the whole affair will appear, at best, futile (#abath) and thus bad and
unwise, rather than beneficent; for the Mu#tazila hold that an act
of beneficence will involve bestowing a preponderant benefit upon
the recipient. They cannot argue that God may bestow upon the
individual benefits that are greater than the harms caused by his
needs and desires; for the maximum benefit one may receive will be
commensurate with (never more than) his needs and desires.203
Another argument put forward concerns the Mu#tazil doctrine
of the goodness of Gods obligating (taklf) men to perform certain
acts.204 How can the Mu#tazila justify Gods obligating men to know,
praise and thank Him, when to achieve such knowledge is extremely
difficult and practically impossible for most men? Moreover, God
Himself does not benefit when humans know, thank and praise Him.
Nor are these arduous acts necessary means to human benefit; for
God can deliver all benefits to all humans without them. Therefore,
by Mu#tazil ethical standards, obligating humans will be futile, and
hence bad.205
Al-Rz challenges the Mu#tazil notion of obligation also by referring to Gods promise of afterlife reward for obedience and threat
of punishment for disobedience. When a man obligates another
to do and omit certain acts in return for benefits he bestows upon
him, he will be considered a vile and lowly person. However, when
these acts and omissions are beneficial to neither person, but may
result in punishment or rewards for the obligated person, the obligator, according to Mu#tazil ethics, will clearly be a wrongdoer. The
Mu#tazila cannot escape making the same judgement in relation to
Gods acts.
Among the more analogical problems is one that al-Rz cites
from a debate between Ab Bakr al-Rz and Ab l-Qsim al-Ka#b,
the Baghdd Mu#tazil, who was reportedly unable to reply to arguments put to him on theodicy and the nature of divine justice (ta#dl

203

Malib, 3, 2956.
Malib, 3, 298304; cf. Mal, 1/2, 36398; 2/2, 2578; 2/2, 2602; Nihya,
fol. 202b.
205 On the badness of futile action according to Mu#tazil ethics, see Oliver
Leaman, #Abd al-Jabbar and the Concept of Uselessness.
204

al-rz on the ethics of action

103

wa-tajwr).206 Ab Bakr al-Rz argues that if a man teaches his son

swimming, then obligates him to swim across a river, knowing that
the son will choose to quit swimming in the middle of the river and
drown, he will be inconsiderate of the wellbeing of his son. Al-Ka#b
seems to reply that the correct analogy (mithl) for the God-man
relationship is that the father has many sons, whom he commands to
cross to the other bank in order for each to attain a benefit, knowing that only some will choose to drown. Yet, Ab Bakr al-Rz
retorts, this would be the correct analogy of the God-man relationship only if it fulfils a number of conditions, e.g. (1) that the father
is not responsible for their initial need for that benefit; (2) that, if he
knows with certainty who will choose to drown, he will be unable to
exclude them from the command; and (3) that he is ignorant of the
fact that there will ultimately be greater overall harm than benefit,
since only a minority will cross safely.207
Mu#tazils may respond by affirming that God creates men with
free will and gives them the ability to believe and do good before
He obligates them. However, al-Rz develops Ab Bakr al-Rzs
problem and contends that believers and disbelievers are not given
equal opportunities; for people vary psychologically and in their
circumstances, which, undeniably, will at least make some outcomes
more probable than others. Moreover, what would the Mu#tazila say
of a disbeliever who suffers poverty, blindness, calamities and pains
in this life, only to find himself in the lowest pit in hell afterwards?
Did God somehow intend to benefit this man, as they claim?
Finally, al-Rz defends the doctrine of obligating what is above
capacity (taklf m l yuq), which utterly contradicts the view that
God is motivated to benefit humans. He argues that it is conceivable for God to obligate humans to do what is above their capacity.
For instance, He obligates all humans to become believers, yet He
states in the Qur"n that some (most notably Ab Lahab, an enemy
of the Prophet) will never believe. As such, they are commanded to
206 Malib, 3, 31820; 4, 419. On the debate between al-Ka#b and Ab Bakr
al-Rz, cf. the latters Ras"il, 1678; Mahdi Mohaghegh, Flsf-i-Rayy, 315. On
al-Ka#b, see Abu l-Qsem al-Balk al-Ka#b, Encyclopdia Iranica.
207 Though Ab Bakr al-Rzs cited problem is primarily topical, it also supports
his denial of Gods responsibility for the great evils abundant in this world. Since
He too accepts a type of ethical objectivism, he does not find any justification for
the act of creating such a bad world that could make it a good act fit for a good
deity (cf. p. 1689 infra).

104

chapter two

believe that they will actually never believe, which is inconceivable

and above their capacity.208
Gods Truthfulness
Having rejected ethical objectivism, the Ash#ars still had to address
one crucial and unavoidable problem. They needed to prove that
God is truthful, and that revelation does not contain untruth, without
appealing to the authority of revelation itself, which would lead to
circularity and a fideism that no serious Muslim theologian (premodern, at least) would admit. It would seem, as the Mu#tazila
argue, that one has to accept that lying is intrinsically bad.209 Classical Ash#ars, however, attempted a non-ethical, theological proof
of divine truthfulness.
Al-Bqilln argued that Gods speech is truthful, not because
lying is morally bad, but because truthfulness is an attribute of His
essence.210 Similarly, al-Rz writes in his earliest kalm work:
[The Mu#tazila] say: If reason is incapable of making [judgements of]
goodness and badness, it will be possible for God to lie in his speech,
without it being bad. We say: Good and bad are attributes of acts,
whereas Gods speech is an attribute of His essence, which is pre-eternal
and attributed with neither goodness nor badness.211

Therefore, if Gods speech were untruthful, it would entail an imperfection (naq) in His essence, which is inconceivable.
Al-Rz also quotes Ab Isq al-Isfar"ns argument that the
speech within the self (kalm nafs) relates to a thing in accordance with
the knowledge of [that thing]; therefore, it cannot contain lying.212

208 Mal, 1/2, 37880; Arba#n, 237; Munart, 512; Tafsr, 2, 42 ff.; Ma#lim,
856; Malib, 3, 30515. Cf. al-Ash#ar, Ibna, 195; al-Juwayn, Irshd, 2034. One
commentator (al-Ifahn, Kshif, 4, 3) writes: Al-Juwayn writes, in the Shmil,
that what most of the answers of our shaykh ... al-Ash#ar ... are inclined to, and
what those knowledgeable among his companions have adopted, is that obligating
the impossible (mul) is rationally conceivable, and so is obligating something and
determining the constant obstruction from it.
209 Cf. #Abd al-Jabbr, Mughn, 6/1, 67.
210 Lying is inconceivable in relation to God, not because of [its] badness
(qub), but because of its impossibility in relation to Him by rational evidence
(al-Bqilln, Tamhd, 343; 105).
211 Ul al-Dn, fol. 268.
212 Ul al-Dn, fol. 109; cf. al-Juwayn, Irshd, 27882; al-Kiy al-Harrs, Ul
al-Dn, fol. 140b.

al-rz on the ethics of action

105

After al-Isfar"n, his student al-Juwayn advanced a rigorous defence

of the Ash#ar position on this problem, which al-Rz develops further in his later works.
In Nihyat al-#uql, he cites the traditional argument:
[Gods] speech subsists (q"im) in His self. Lying is inconceivable in the
speech of the self (kalm al-nafs) in the case of one in whom ignorance
is inconceivable, since a statement (khabar) will subsist in the self in
accordance with knowledge. Yet ignorance is inconceivable in relation
to God.213

But could one who is not ignorant have an untruthful statement in

his self (khabar nafsn)? Following al-Isfar"n, al-Rz responds that
when one knows something, it will be inconceivable for him to have
an untruthful statement in his self about it. One can easily imagine
an untruthful verbal statement (e.g. the English statement, The world
is pre-eternal, or its Arabic equivalent). But one cannot imagine
having an untruthful statement within the self (i.e. the non-verbal
statement that the verbal statement expresses) about something that
one knows (hence, if I know that the world is temporally originated,
I can state, The world is pre-eternal, without being convinced in
it).
The more serious problem is: could one who has a truthful statement in his self not produce verbal lies? How does one know that the
statements of revealed scripture do not contain untruth? There does not
appear to be a necessary link between knowledge, or speech within
the self, and the truthfulness of verbal speech. Whatever I have in
my mind, I can still choose to utter a lie; and so can God if He is a
voluntary agent. Al-Rz here responds that the problem of the truthfulness of Gods speech is similar to the problem of the truthfulness
of the messenger.214 He refers us to his discussion of human action,
where he cites the Mu#tazil argument that if God created human
acts, including bad ones, He might create miracles at the hands of
impostors as well as messengers. Al-Rz cites al-Juwayns response
that if one knows that the producer of a supernatural event on the
hands of the claimant of messengership is God, one will have an
immediate knowledge that the claimant is truthful. For doubt enters
only into whether the event is a divine act, rather than whether God

213
214

Nihya, fol. 134b; cf. 134b136a; Muaal, 434.

Nihya, fol. 135b.

106

chapter two

creates it in support of the truth-teller.215 Al-Rz adds to al-Juwayns

argument, that though the latter contention is inductive, one will
know its truth immediately and with certainty; for the contrary possibility (that God creates miracles at the hands of an impostor) will
be so improbable to the extent of being inconceivable.216 However,
his account of the argument is brief, and seems to refer the reader
to al-Juwayns books.
Al-Juwayn argues that a necessary condition for a genuine prophetic miracle is that it be immediately preceded by the prophets
proclamation that he will perform the following miraculous act to
prove his messengership. Mu#tazils object that Ash#ars fail to prove
that God would not create such miracles at the hands of impostors.
In reply, al-Juwayn proposes the following analogy. A person claims,
in front of a royal assembly, to be the kings spokesman. He then asks
the king to break his habit of remaining seated, and to stand up in
confirmation of his claim. If the king does that, his unusual act will
confirm the claim; and it will not occur to anyone in the audience
that the king stood up in order to mislead them.217
However, from al-Rzs point of view, this conclusion will only
provide conviction beyond reasonable doubt, yet not certainty, in
the truthfulness of the claimant of messengership, who supports his
claims with supernatural acts; for the simultaneity of the claim and
the kings act could be coincidental. The king could have stood up
for an unrelated reason. Al-Juwayn neither addresses this possibility,
nor attempts to stress the conclusiveness of his conclusion. Al-Rz
does. But he considers that though this possibility is conceivable in
principle, it is actually impossible: Thus, conclusiveness (al-qa# wa-ljazm) may be realised, though there may be this possibility (tajwz).218

215

Mu#tazil objection: Nihya, fol. 86a; al-Rzs reply: Nihya, fol. 92ab. Cf.
al-Juwayn, Irshd, 25782.
216 Al-Rz explains: If I close my eyes for a moment, then open them, I will
know for certain that God is capable of turning the walls into gold during that
moment, and then, when opening my eyes, returning them as they were. This
possibility will not undermine the immediate knowledge that that did not take
place. The same is true of all customary events. ... Therefore, the possibility of
things and events departing from their normal courses (majr-h al-#diyya) does not
undermine the immediate knowledge that they will continue upon their courses
(Nihya, fol. 92b).
217 Al-Juwayn, Irshd, 275; cf. Arba#n, 317; Malib, 8, 614 (where al-Rz does
not consider it to be a strong argument).
218 Malib, 8, 979. Cf. Arba#n, 324; Nihya, fol. 92b.

al-rz on the ethics of action

107

Perhaps this conclusiveness can thereby reach the level of certainty

(yaqn).
The obvious link between this argument for how a miracle proves
messengership and proving the truthfulness of Gods speech is that
though it is theoretically conceivable for God to mislead men (in
speech or action, by creating miracles at the hands of impostors), this
is a trivial possibility: it is actually inconceivable that He would do
so. The thrust of this defence seems to be that lying is an extremely
trivial, lowly and ignoble act that an agent would resort to only
because of his weakness, and for which he may be viewed with
contempt. Lying contradicts Gods absolute power and majesty,
affirmed by Ash#ars, without being considered intrinsically bad. A
sensible person, therefore, may doubt whether a certain claimant
of messengership is a true messenger of God; however, once he has
verified that he is, he will not have any doubts as to the veracity of
his message, i.e. of both Gods speech that he communicates and
his communication thereof.
Al-Juwayn develops a rather complex defence for the truthfulness
of messengership, partly because he rejects the argument that lying
is inconceivable in divine speech since it constitutes an imperfection. He considers this to admit a type of ethical objectivism.219 By
contrast, al-Rz fully endorses this stance, which he relies on in his
later works. He writes in the Muaal: The statement (khabar) of
God is truthful because lying is an imperfection; and [imperfection]
is inconceivable in relation to God.220 And in the Malib: That
lying is inconceivable in relation to God, exalted, is known immediately, since it is an attribute of imperfection; and [human] primordial
nature testifies (shahdat al-fira) that attributes of imperfection are
inconceivable in relation to God, exalted.221
Thus, returning to the early Ash#ar argument, expressed by alBqilln, al-Rz maintains that lying implies an imperfection of
essence in the agent, though, as an act, it is not intrinsically bad.
This solution, from a perfectionist angle, to the problem of divine
truthfulness brings us to the subject of the next chapter.

219
220
221

Al-Juwayn, Irshd, 279.

Muaal, 434; cf. Tafsr, 3, 143.
Malib, 8, 99100.

108

chapter three

al-rzs perfectionist theory of virtue

109

CHAPTER THREE

AL-R$Z^S PERFECTIONIST THEORY OF VIRTUE

Perfectionism
Referring to the consequentialist and Legal senses of value terms,
Ibn Taymiyya, one of al-Rzs most outspoken critics, writes:
There are some who affirm a third sense for good and bad, and
claim that there is unanimity over it, viz. the acts being an attribute
of perfection or an attribute of imperfection. This sense was not mentioned by the majority of early mutakallimn in this context, but was
mentioned by some later ones, such as al-Rz, who took it from the
falsifa.1

Al-Rz is indeed the first theologian to juxtapose these three definitions of value terms in such a succinct manner. This appears in a
work as early as the Ishra, where he writes that good and bad,
... also designate attributes of perfection (kaml) and imperfection (nuqn).
So it is said, Knowledge is good; ignorance is bad. By perfection,
we mean that a thing has something it is supposed to have (wujd shay"
li-shay" min sha"nih an yakna lahu), with respect to its species, type, or
essence.2

The same listing of the three definitions is also found in his later
works.3 Though it is unprecedented in kalm to list these contrasting
definitions of value terms in this way, this third definition seems
related partly to a sense that classical Ash#ars recognised in a main
class of value terms as used in ordinary language and some technical contexts. Nonetheless, as we have seen, they explained this sense
as being descriptive of non-moral facts about beings, and often
appealed to it in explaining divine attributes, such as justice and
wisdom.4 Thus, Ibn Taymiyya appears to be right in his remark
that this third definition represents a falsaf influence.
1
2
3
4

Ibn Taymiyya, Itijj, 2, 104.

Ishra, fol. 32b.
E.g. Mal, 1/1, 159; Nihya, fol. 195a; Arba#n, 246; Muaal, 479.
See p. 501 supra.

110

chapter three

Earlier than the Ishra, in Ul al-Dn, al-Rz accepts the above

classical Ash#ar stance concerning this class of value terms, as well as
classical Ash#ar voluntarism (the doctrine that value terms could be
defined only in terms of divine command). Though soon afterwards
he abandons the latter doctrine in favour of consequentialist ethics,
the emphasis in both the Ishra and Nihyat al-#uql is on acts, whereas
the third definition of value terms, in terms of perfection, which he
lists, remains unconnected to a moral stance. This is due to al-Rzs
acceptance in these early works of the common classical kalm doctrine that the essence of man is purely physical, and consequently his
rejection of both the possibility of human perfection and the notion
of intellectual pleasure (ladhdha #aqliyya).5 According to Ibn Sn, man
experiences this pleasure at a supra-corporeal level upon attainment
of theoretical perfection. Instead, he presents a soteriology consisting
of a creedal approach to knowledge, an ethics of action oriented at
duties and rules of conduct, and a purely physical notion of human
resurrection in the hereafter. Man ought to have the correct set of
beliefs, ought to act in accordance with the dictates of Revealed
Law, and should expect afterlife reward or punishment accordingly
as promised.6 Even to love God is considered an attribute of action
(ifat fi#l), viz. being obedient to Him.7
At a later stage, al-Rz abandons the physicalism of classical
kalm and adopts a completely new theory of human nature. As we
shall see in the following section, he will maintain, under falsaf and
Sufi influence, that man has an unphysical soul and may experience
intellectual pleasure as well as a spiritual afterlife alongside the physical one. Moral and theoretical perfection become real possibilities
and viable human ends. The connection between the doctrine of
the separate, rational soul and perfectionist ethics is underscored in
al-Rzs minor work, Risla f l-nafs, where, he writes, commenting
on the adth, He who knows his self will know his Lord: Had
self in this adth referred to the physical body, everyone would have
known his Lord completely.8 This view, which al-Rz attributes
to most #ulam" and mutakallimn, is then contrasted with the view of
the falsifa and Sufis, viz. that the soul,
5
6
7
8

Ul al-Dn, fol. 351 ff.; Ishra, fol. 63a; Nihya, fol. 248a; 263b265b.
Cf. Shihadeh, From al-Ghazl to al-Rz, 1723.
Ul al-Dn, fol. 243; cf. Tafsr, 4, 2314; 8, 189.
Risla f l-nafs, fol. 2a; cf. Nafs, 48.

al-rzs perfectionist theory of virtue

111

... is neither the physical body nor physical, but is a spiritual substance
that emanates on this frame, animates it, and uses it as an instrument
to acquire sciences and knowledge. Once it perfects its substance by
them and knows its Lord and the rights of His creatures, it will become
prepared to return to His presence and to become one of His angels
eternally happy.9

This highlights the link between the nature of the soul and the
purpose of its coming into being.
Al-Rz frequently argues that, as with pleasure and pain, likewise
perfection and imperfection are simple primary human ends that
are sought or avoided in themselves. The two types of motivation
are not mutually exclusive: That pleasure is liked in itself does not
conflict with perfection being liked in itself.10 Indeed, in Kitb al-Nafs
wa-l-r, he argues that both are mutually explanatory:
We know spontaneously (bi-l-badha) that we like (aabba) one thing and
dislike (kariha) another. Therefore, we say: Either there is one thing
that is liked in itself, and another that is disliked in itself, or ... each
thing is liked because it involves something else, or disliked because
it involves something else. The latter division is absurd, since it will
lead to either infinite regress or circularity. ...
Having reflected and meditated (baathn wa-ta"ammaln), we have
found nothing that can be said to be liked in itself except pleasure
and perfection. In reality, there is no difference between them; for
what is pleasurable will bring about the perfection of state for the
experient of pleasure (kaml l al-multadhdh), and what constitutes an
[aspect of] perfection will be pleasurable. However, we refer to what
is physically pleasurable pleasure, and to what is spiritually pleasurable perfection.
Also, what is disliked in itself is pain and imperfection. In reality,
there is no difference between them, as explained already; for perfection is liked for its own sake, in itself, qua perfection (mabb li-dhtih
bi-dhtih min aythu anna-hu kaml), and imperfection is disliked for its
own sake, and in itself, qua imperfection.11

Thus, from the psychological and metaethical standpoints, al-Rz

considers pleasure and perfection to be two concomitant sides of
the same coin. The perfection of a given aspect of the individual
is an objective feature, yet it results in the subjective experience of
pleasure.12 The latter, in turn, serves to perfect the subjective state
9
10
11
12

Risla f l-nafs, fol. 2b.

Tafsr, 4, 232.
Nafs, 1920. Cf. Malib, 3, 212; 3, 3489.
On the nature of pleasure and pain, see p. 156 ff. infra.

112

chapter three

of the individual, which appears to refer to a sense of gratification,

fulfilment or happiness.
However, he also contends that the experience of pleasure is not
restricted to the perception of subjective aspects of perfection. He
writes:
Induction indicates that perfection is loved in itself. Therefore, it should
be said that all that is more perfect will be more deserving of being
loved (awl bi-l-mabbiyya). The most perfect of things is God, exalted;
so He is the most deserving of being loved. The perception of what is
loved, qua being loved, effects pleasure. And since the perception of the
rational soul for the True, exalted, is more perfect than the perception
of the physical faculties for their objects of perception, and since the
True, exalted, is the most perfect being, the pleasure resulting from
perceiving Him will be more perfect than all other pleasures.13

Thus, popular stories of brave heroes will invoke a great sense of

admiration and love in peoples hearts, without them receiving any
benefit from those heroes.14 Al-Rz also criticises Ibn Sn for implying that man may seek to know God in order to perfect himself
(istikml), and responds that man may love and seek to know God
in Himself, rather than for the sake of attaining greater subjective
perfection, as he may become heedless of all other than God, including himself.15 Already, in the Iy", al-Ghazl distinguishes between
subjective perfections, which are normally desired and sought by the
agent out of self-regarding motives, and external perfections that
the agent appreciates for their own sake, not for the benefit that
he gains from them.16
As regards subjective perfections, al-Rz argues that an entity
can have aspects of perfection that pertain to its being, attributes, or
acts. The perfection of an entity in its being implies, first, that the
entity exists necessarily by virtue of its essence and is self-sufficient
in all respects, and second, that it is completely unique in its essence.
Therefore, God alone has this type of perfection.17 Human beings may
only perfect attributes of their souls, viz. knowledge and power.18 As

13
14
15
16
17
18

Shar #Uyn al-ikma, 3, 1678; cf. Tafsr, 4, 232.

Nafs, 20; Tafsr, 4, 232; 8, 19; cf. al-Ghazl, Iy", 4, 300.
Shar al-Ishrt, 2, 1089.
Al-Ghazl, Iy", 4, 297306.
Nafs, 212; 131; Tafsr, 4, 232.
Nafs, 22 ff.; 1313.

al-rzs perfectionist theory of virtue

113

for human acts, they are said to be perfect or imperfect on account

of their consequences, i.e. the resultant pleasure or pain.
Al-Rzs exposition of the relation between perfection and pleasure
allows him to treat the consequentialist and perfectionist stances as
two aspects of the same teleological ethics, though obviously they
have to be reconciled at the normative level. In a late work, he
contextualises these two moral stances, as follows:
The things being good (khayr) is due to either that things essence or
its acts. Likewise, its being evil (sharr) is due to either its essence or its
acts. As for the goodness (khayriyya) that pertains to the things essence,
it only means that it actually has every perfection and majesty (jall)
that it may have. Evil is what is contrary to this. ... As for the goodness
that pertains to acts and deeds (athar), it means pleasure and joy, and
what is a means (wasla) to both or either of them.19

He, thereby, makes a sharp distinction between his ethics of action,

which is a subjectivist consequentialism, and his perfectionist ethics
of character. In the latter outlook, value judgements assess aspects of
an entitys essence with respect to an objective notion of the perfect
essence thereof. Moral perfectionism posits an objective notion of
human nature, such that something is said to be good if it serves
human perfection, or constitutes an aspect thereof. Before turning,
in the following section, to al-Rzs notion of human perfection,
his theory of virtue, we shall consider two theological themes where
his character-oriented perfectionism contrasts markedly and fundamentally with the classical kalm emphasis on action.
The first concerns his Sufi-inspired interpretation of Qur"nic
allusions to the believers love of God, which he contrasts with the
dogmatism and duty-oriented Legalism of the majority of the mutakallimn. The latter argue that love, being a type of will (irda), can only
relate to contingent things. Therefore, if we say, We love God,
we will mean that we love to obey God and serve Him, or that we
love His reward and beneficence.20 This will involve praising God,
worshiping Him alone, having correct belief (i#tiqd), seeking His
reward, and fearing His punishment.21 By contrast, knowers (#rif )
maintain that both pleasure and perfection are sought and loved for

19
20
21

Shar #Uyn al-ikma, 3, 131.

Tafsr, 4, 232. Cf. Joseph Bell, Love Theory, 47 ff.; 230, n. 13.
Tafsr, 4, 232; 8, 19.

114

chapter three

their own sake. Therefore, one may love God in Himself and for
His own sake (f dhtih wa-li-dhtih), whereas loving His service and
reward constitutes an inferior level of love.
Pleasure being loved in itself does not conflict with perfection being
loved in itself. This being the case, we say: Those who interpret love
(maabba) of God, exalted, as love of obedience (#a) to Him, or love
of His reward, know that pleasure is loved for its own sake, but not
that perfection is loved for its own sake. As for knowers, who maintain
that God is loved in Himself and for His own sake, they have realised
that perfection is loved for its own sake.22

God is the most perfect being, in His necessary existence and self-sufficiency, and in His attributes of knowledge and power. The greater
the knowledge of God that one attains, the more will his yearning
(shawq) and love of Him be.
Similarly, the classical mutakallimn interpret Gods love for human
beings in terms of His will (irda) that reward (thawb) be delivered
to them. In the Tafsr, al-Rz writes that they are able to support
this interpretation only be arguing that since there is no evidence
to affirm love as a distinct divine attribute, it ought to be denied.
He replies, In Nihyat al-#uql, we have shown that this method is
weak and vacuous (a#fa sqia).23 By this method, he refers to the
argument ad ignorantiamthat if something has no proof, it should
be negatedwhich indeed he rejects in the Nihya.24 By contrast, he
asserts that it is likely that the love of God, exalted, for the servant is
an attribute other than His will that reward be delivered to him.
The second theme is the question whether angels or prophets,
who are the best of mankind, are superior (afal). Most previous
Sunnis (with the notable exception of al-Bqilln and Ab #Abdullh
al-\alm) and Sh#s maintain that prophets are superior, while the
falsifa and Mu#tazils hold that angels are superior.25 In most of his
writings, al-Rz maintains that prophets are superior, though, in the
Arba#n, he seems undecided and more inclined to the opposite view.26

22

Tafsr, 4, 232.
Tafsr, 14, 132.
24 Nihya, fol. 6a; cf. A. Shihadeh, From al-Ghazl to al-Rz, 165.
25 Arba#n, 368. In Tafsr, 2, 2156, he writes that the majority of Sh#s contend
that angels are superior.
26 Arba#n, 36884; cf. Ul al-dn, fol. 325; Muaal, 531 ff.; Khamsn, 667;
#Ima, 33; also Tafsr, 2, 215 ff., which is directly inspired by al-Shahrastn, Milal,
2, 944; see p. 1378 infra.
23

al-rzs perfectionist theory of virtue

115

While much of the evidence given for both views is Qur"nic, the
most important rational argument for the superiority of prophets is
as follows. Humans have more encumbrances than angels: they are
afflicted with appetite, irascibility, forgetfulness and other internal
motives for disobedience, as well as the corruptive influence of the
devil and external causes of doubt. Therefore, it will be harder
(ashaqq) for humans to be obedient [to God]; and what is more difficult
is superior.27 Angels do not suffer from the same shortcomings as
humans, which affect both their will and knowledge, so they perform
good acts effortlessly out of their perfectly good and pure nature, and
on the whole have a rather easy life. Yet, typically in action-oriented
ethics, effort and hardship are the main criteria for moral superiority.
The generosity of a poor man donating a small sack of wheat may
be much more praiseworthy than that of a wealthy man donating a
thousand pieces of gold. Al-Rz explains superiority ( fal) here in
terms of the size of the reward (thawb) earned.28
In his latest works, he turns away from this action-oriented stance
to a character-oriented perfectionism, and contends that angels are
superior to prophets. He writes: Know that the mind of one who
knows what an angel is, and what its attributes are, will not allow
him to delve into this question. Yet most people think that angels
are birds that fly in the heavens! So they delve into this question.29
The essence and attributes, rather than the acts, of an agent become
the primary criteria for superiority, which no longer refers to the
consequences of acts, but the excellence of the agent, his intrinsic
worth of character.30 Responding to an argument for the superiority
of prophets to angels, al-Rz writes that angels possess perfection in
both power and knowledge: Yet excellence ( fala) has no meaning
except this. Thus, since they are superior to men [in these respects],
they will be more excellent than men.31 In the Arba#n, he records
the following argument for the superiority of angels:
Angels are free from appetite, irascibility, imagination and estimation
attributes which act as thick veils from the manifestation of the lights
of God, exalted. There is no perfection except by that manifestation,

27
28
29
30
31

Arba#n, 36970.
Arba#n, 370; 371; Tafsr, 2, 216; 2, 232.
Malib, 7, 405 ff.; cf. Ma#lim, 1012.
E.g. Malib, 7, 414; 7, 418.
Malib, 7, 4101; cf. 7, 421 ff.

116

chapter three
and no imperfection except by the intervention of that veil. Since
[angels] always contemplate this manifestation, while human souls
are usually veiled from [it], we conclude that there is no comparison
between their perfections and human perfections.
To say that service (khidma) when many obstacles are present is more
indicative of sincerity than service without obstacles is fanciful (kalm
khayl). For the bulk of acts of worship and obedience are intended for
the realisation of that manifestation. The greater and the further from
obstacles the realisation of that manifestation is, the more complete
will the perfection and happiness be.32

The contemplative ideal replaces the earlier voluntarist trend in alRzs earlier thought, which gives primacy to the notions of volition,
choice, effort, practice, responsibility and desert. Thus, adherence to
theological creeds and performing acts of worship become means,
rather than ends in themselves.

Human Perfection
The question of the reality of man (aqqat al-insn) is one that alRz found to be highly problematic. He acknowledges in his late
work the Malib that it is a dilemma, about which the strangest fact
is that something so close to us could be so obscure.33
In his earliest works, he accepts the common classical kalm view
that the essence of man is purely physical. In the human body, he
writes in the Ishra, there are core parts (ajz" aliyya) that will neither
cease to exist nor be replaced; this is the reality of man.34 Slightly
later, in Nihyat al-#uql, the reality of man is said to be nothing but
the physical body (hdhih al-bunya, hdh al-badan).35
Later, under philosophical and Sufi influence, he rejects this physicalism and advances a thorough critique thereof, in favour of a
dualism of a physical body and a separate, rational and unphysical
soul. He often refers to previous Sunni scholars, and sometimes to
others as well, apparently to illustrate that this later doctrine is neither unorthodox nor irrational.36 The present section will examine

32
33
34
35
36

Arba#n, 3823; cf. Malib, 3, 302.

Malib, 1, 412; Asrr al-tanzl, 1401; cf. p. 190 infra.
Ishra, fol. 62b63a; cf. Ul al-dn, fol. 351 ff.; Mal, 1/1, 338.
Nihya, fol. 252b.
In the Mabith (2, 22432; 2, 34582), he maintains that the human soul is

al-rzs perfectionist theory of virtue

117

this later psychology briefly, only in so far as it bears on his theory

of virtue; a closer study of his psychology and epistemology will be
left for future studies.
Al-Rz goes further in rejecting kalm physicalism by maintaining
that, essentially, man is the rational soul alone (which is my true referent when I use the expression I), whereas the body is accidental to
it.37 In itself, the soul is an intellect (#aql), and is referred to as soul
(nafs) with respect to its relation to the physical body.38 The body,
on the other hand, is the locus of the non-rational aspects in man.
The ultimate purpose in the rational souls connection to the
human body is that it uses it as an instrument (la) for acquiring
knowledge and hence for attaining perfection.39 It retains this perfection after death, when it departs from the physical world, where it is
in a state of estrangement (ghurba), and returns to its primordial realm
(waan al).40 A soul that has perfected itself to a certain extent in this
world may continue to increase in perfection in the afterlife.41
separate (mujarrada) from the body, but finds Ibn Sns proofs for this weak (cf. Shar
al-Ishrt, 1, 124; M. Marmura, Fakhr al-Dn al-Rzs Critique of an Avicennan
Tanbh). He attributes this theory of the soul to: (a) the falsifa, (b) most Sufis,
(c) some Sunnis, viz. al-Ghazl (on his views on the soul, cf. Timothy Gianotti,
Al-Ghazls Unspeakable Doctrine of the Soul, esp. 117 ff.), al-Rghib al-Ifahn, and
Ab #Abdullh al-\alm (d. 403/1012), (d) some Mu#tazila, viz. Mu#ammar Ibn
#Abbd (d. 215/830) and Ab Zayd al-Dabs (d. 432/1041), (e) some Sh#s, viz.
al-Nawbakht (d. ca. 300/912), Muammad Ibn al-Nu#mn (d. 413/1022) and the
Akhbrs, (f) Ab l-Hayam (d. 407?/10167) from the Karrmiyya, (g) Christians
and (h) reincarnationists (Nihya, fol. 262b; Muaal, 539; Arba#n, 267; Ass al-Taqds,
6; Tafsr, 21, 45; I#tiqdt, 63; Malib, 7, 38; cf. Duncan Macdonald, The Development of the Idea of Spirit in Islam; M. Marmura, Soul: Islamic Concepts,
Encyclopaedia of Religion).
Later, he considers the human soul to be neither the physical body nor separate,
but a subtle, luminescent physical substance that exists in this body (Ma#lim,
109; cf. Arba#n, 266). This view was held by al-Nam (d. 231/846) (cf. al-Ash#ar,
Maqlt, 2, 331) and some Ash#ars, including al-Juwayn (Irshd, 318). Al-Rz
attributes this view to Thbit Ibn Qurra (d. 289/901) (Tafsr, 21, 45).
Finally, from the Tafsr (21, 51) onwards, he reverts to his earlier view that the
soul is separate. In the Malib, he presents his most thorough discussion of the
subject, refuting proofs for this doctrine by Ibn Sn, Miskawayh (d. 421/1030)
and Ab l-Barakt (7, 6999), and advancing a modified version of one of Ibn
Sns arguments (7, 5768) and a refutation of physicalism (7, 10138; cf. Risla
f l-nafs, fol. 2b3b).
37 Nafs, 27 ff.
38 Mabith, 2, 222.
39 E.g. \ikmat al-mawt, fol. 81a; Risla f l-nafs, fol. 2b.
40 \ikmat al-mawt, fol. 84a86a.
41 Nafs, 133.

118

chapter three

Following Ab l-Barakt al-Baghdd and talisman specialitsts,

al-Rz maintains that human souls vary in their essences.42 He contrasts this view to: (a) Ab Bakr al-Rzs view, following Plato and
other ancient philosophers, that humans and animals have essentially
identical souls, but differ only in their various physical natures; and
(b) Ibn Sns view, following Aristotle, that animal souls are physical,
whereas humans have essentially identical and non-physical souls,
but vary in their ethical properties due to variations in their bodily
humours.43 Al-Rz advances several inductive arguments, which
mainly suggest the independence of the capacity for perception and
character traits from bodily conditions.44
But do particular human souls fall under a limited number of
types, or does each soul have its own distinct essence? In his later
works, al-Rz adopts the former view, apparently under the influence of Ab l-Barakt and talisman specialists. He writes that the
souls of the moving planets (and perhaps also fixed constellations)
are the causes, or sources, (#illa; mabda"; al; ma#din; yanb#) for human
souls (contra the falsifa, for whom the producer of human souls is
the Active Intellect).45 Each planetary soul (r kawkab) has a distinct
essence, which determines the essences of human souls originating
from it. Talisman specialists, al-Rz writes, refer to these heavenly
souls as archetypes (or complete natures, ib# tmm), since each
possesses a set of qualities (ifa; kha) in a perfect way, which then
manifest imperfectly in the human souls it produces.46 It will be to
them like a father is to his children, providing them with assistance
during their lives, and affecting their characters and acts.47
Since a certain planetary soul may be magnanimous (urr), noble,
virtuous, and distinguished in its apprehension and good acts, while

42

Cf. Ab l-Barakt, Mu#tabar, 2, 38894.

Nafs, 85; Mabith, 2, 383; Mulakhkha, fol. 313b314a; Muaal, 543; Malib,
7, 1412. Cf. Ibn Sn, Najt, 222; Shif", ab#iyyt, 6, 198.
44 Malib, 7, 1458.
45 Malib, 7, 2637. He rejects the whole notion of the Active Intellect (Shar
#Uyn al-ikma, 2, 2814).
46 Al-Sirr al-maktm, 111.
47 Pseudo-Majr, a representative of talisman specialists, discusses complete
natures in Ghyat al-akm, 187 ff. Cf. references to discussions of this notion in other
primary and secondary sources in the German translation of this work (Picatrix Das
Ziel Des Weisen Von Pseudo-Mar, 198, n. 1). By contrast to al-Rz, al-Suhraward
(d. 586/1191) (Haykil, 65) considers mankind to have a single archetype (ib# tmm)
(cf. al-Zarkn, Fakhr al-Dn al-Rz, 4823).
43

al-rzs perfectionist theory of virtue

119

another type may be vile, evil and stupid, these same characteristics
will appear in their respective human souls, both in their essences
and in the effects they receive from them in their lives. This is why
two strangers may meet and immediately feel either affinity or enmity
toward each other, depending on the types of their souls. Al-Rz
adds that while this model is merely a plausible hypothesis, it is
verified empirically by talisman specialists.48
On the basis of his theory of the variation of human souls, al-Rz
sometimes expresses a psychological determinism with respect to
human character, to the extent that change in character traits appears
impossible (this is to be distinguished from his standard metaphysical
determinism with respect to human action). The differences among
human souls are akin to those that exist among different animal
species: donkeys, horses, wolves, lambs, etc.49 Many people with
malevolent traits have inherently malevolent psychological essences;
though they may modify their behaviour, their essential natures will
remain unchangeable:
We see a man, who may be naturally evil and abject. If he were to
undergo the maximum possible discipline, he would not depart from
his malevolent nature. Rather, he may become, through discipline
and admonition, such that he will avoid those acts and not perform
them. Nevertheless, if he then abandons his self to its original nature,
it will incline to that evil. Also, his temperament may change from
warmth to coldness, from wetness to dryness, and vice versa, yet what
is concomitant to his inborn nature will not change.50

In one place, al-Rz argues that since the essential traits of human
souls can be neither eliminated nor modified, one who is happy
(sa#d) will not become miserable (shaq); and vice versa.51 However, he seems to maintain such radical ethical determinism only
in relation to the extremities in the scale of human psychological
excellence: some individuals are doomed by the sheer baseness of
their souls, whereas others are blessed by the excellence of theirs.
Despite his insistence that altering the essence of the human soul is

48

Tafsr, 16, 183; 19, 20; 19, 112; Malib, 7, 1423; 7, 2667; 7, 272; 7, 400; 8,
136; 8, 144; Ma#lim, 115; Shar #Uyn al-ikma, 2, 284; Jabr, 356; Al-Sirr al-maktm,
1104. Cf. Ab l-Barakt, Mu#tabar, 2, 38894.
49 Malib, 7, 1478.
50 Nafs, 86; cf. Jabr, 36; Kamliyya (Ar.), 901; (Per.), 118.
51 Tafsr, 14, 144; cf. Malib, 1, 557.

120

chapter three

impossible, al-Rz still considers it possible to further the perfection

of those in the middle levels of the scale to the extent permitted by
the particular essence of each soul.52
Al-Rz adopts Ibn Sns division of human psychological faculties,
which is as follows: (a) vegetative faculties, which serve the functions
of nourishment, growth and reproduction; (b) animal faculties, which
are either motive (including the appetitive (shahwa) and the irascible
(ghaab)), or perceptive (the five external senses, and the inner faculties
of perception); and (c) human, rational faculties, viz. the theoretical
intellect and the practical intellect, which allows the soul to govern
the body.53 However, in contrast to Ibn Sn, al-Rz maintains that
both the perceptive and active animal faculties are not bodily and
physical, but are purely functions, or attributes, of the rational soul
itself. On the other hand, the vegetative faculties are physical.54
These attributes serve both the soul directly, in its pursuit of theoretical perfection, and the wellbeing of its physical body. The soul
acquires knowledge by perceiving particulars through its various
attributes of perception and extracting abstract conceptions from
them. Self-evident (badh) knowledge is acquired when the mind
recognises connections immediately, in either affirmation or negation,
among some of these conceptions as soon as they become present
in it. On the basis of self-evident knowledge, the mind may then
arrive at discursive (naar) knowledge of other connections between
these acquired mental conceptions. As for these attributes serving
mans physical wellbeing, this ought to be aimed at preserving the
body in a state that allows it to fulfil its essential function as the
instrument that the soul uses to perfect itself. The attributes of the
soul serve mans physical wellbeing by allowing man to recognise
what is beneficial and harmful to him, and to seek the former and
avoid the latter.55
According to al-Rz, man experiences pleasure mainly when he
attains greater perfection in some subjective aspect and perceives this

52

Asrr al-tanzl, 5489.

Nafs, 747; Mabith, 2, 235 ff. Cf. Ibn Sn, Shif", ab#iyyt, 6, 32 ff.; Fazlur
Rahman, Avicennas Psychology.
54 Mabith, 2, 245; Nafs, 778; 29; cf. Ab l-Barakt, Mu#tabar, 2, 302. This
stance has a profound impact on al-Rzs epistemology, to be examined in a
future study.
55 Nafs, 7980.
53

al-rzs perfectionist theory of virtue

121

attainment. Pleasure may occur at the perception of a change in the

state of the perceiver towards a more perfect state, and pain may
occur at the change towards a less perfect state. Yet neither will be
experienced after the state becomes stable. Al-Rz speaks of three
main subjective types of perfection that produce pleasure in man: (a)
the perfection of the theoretical aspect of the soul; (b) perfections in
the physical body, especially by the gratification of its appetites for
nourishment and coition; and (c) perfections that man achieves in
his association with other human beings, especially in the attribute
of power (qudra), which is related to leadership and social status.
When these three types of subjective perfection are perceived by
three forms of human perception, three types of pleasure may be
experienced. (a) Intellectual pleasure is experienced when the rational
soul perceives its own theoretical perfection. (b) Sensory pleasure is
experienced when the external senses perceive the gratification of
bodily needs. (c) The perfections that man attains in his association
with other human beings are perceived by the inner, animal faculties of perception, and produce a type of pleasure that al-Rz refers
to sometimes as imaginative (khayl) pleasure.56
When man has awareness (shu#r) of what is agreeable (mul"im) and
what is disagreeable (munf) to him of the last two types of perfection,
his motive attributes will determine his conduct: his appetite (shahwa)
will motivate him to seek the former, whereas irascibility (ghaab)
will motivate him to avoid and repel the latter.57 With respect to
the first, intellectual attribute, man may acquire a sense of yearning
(shawq) to progress in this respect, as we will see. Normally, engagement (ishtighl) in any of these pleasures and the forms of activity to
which they are connected will lead to the appearance of dispositions
that prevail over (istawl #al) the soul and divert man from pursuing other engagements and from developing different dispositions.
This is the case, al-Rz argues, since these faculties of perception
are different attributes of the one, unitary substance of the soul.58
The central human predicament is thus that the greater the hearts
preoccupation with other than God, the greater its deprivation from

56
57
58

Cf. p. 155 infra.

Nafs, 76; Mabith, 2, 236; Asrr al-tanzl, 533.
Malib, 7, 160; cf. Nafs, 289; Tafsr, 12, 70; Asrr al-tanzl, 5334; 5501.

122

chapter three

witnessing the lights of Gods majesty.59

In supporting his theory of virtue, al-Rz refutes commonplace
(#mm) hedonism.60 With regard to sensory pleasures, he maintains:
(a) that they are contemptible both in themselves and in various
other concomitant respects,61 and (b) that they are inferior to intellectual pleasures.62 This general line of reasoning is apparently borrowed from Ibn Sn, although al-Rz proceeds to support it with
a very different set of arguments, apparently drawing on a variety
of sources, including Ab Bakr al-Rz.63 For instance, he argues
that sensory pleasure is not real, but is merely the alleviation of
pain and thus cannot constitute a ground for human happiness.64
He also argues that sensory pleasures conflict with humanness as
such, which is purely intellectual: when man becomes preoccupied
with them, his rational faculty will be obscured and he will descend
into bestiality.65
Nonetheless, most humans are naturally and originally inclined
to sensory pleasures.66 Unlike the two other types of pleasure, these
pleasures are produced by the perception of perfections strictly in
the physical body, by the gratification of certain bodily needs. Yet
the perfection of these physical needs is extremely transient, as they
will require constant gratification. And the pleasure experienced
in this perpetual cycle of physical needs and gratification thereof
strengthens the souls attachment to sensory pleasures.67 Therefore,
it is more appropriate to speak of them, in the moral context, as
sensory pleasures, rather than aspects of perfection. By contrast, alRz normally discusses the two other types of pleasure primarily
qua perfections.
He argues that the main subjective attributes that constitute

59

Asrr al-tanzl, 550.

Nafs, 88; Mabith, 2, 4267; Malib, 7, 297; cf. Ibn Sn, Ishrt, 4, 7.
61 Nafs, 96107.
62 Nafs, 8895; Malib, 7, 297302; Tanbh, 459.
63 Compare Mabith, 2, 4412, and Ibn Sn, Ishrt, 4, 710. On Ab Bakr alRzs refutation of hedonism, see his Al-ibb al-rn, 39 ff.; Al-Sra al-falsafiyya (in
the Ras"il), 1012; Shukk #al Jlns, 17; cf. Majid Fakhry, Ethical Theories, 723.
64 This is discussed at length, p. 156 ff. infra.
65 Nafs, 989.
66 Nafs, 105; Tanbh, 445; Tafsr, 18, 220.
67 Nafs (MS), fol. 279b. The passage is missing from the poor printed edition
(p. 106).
60

al-rzs perfectionist theory of virtue

123

perfections in the rational soul and that humans seek to perfect in

themselves are two, viz. knowledge and power.68 The latter attribute
concerns the souls involvement in the external physical world and
manifests in various ways, such as the individuals power over inanimate objects and other human beingsthe possession of wealth
and of hearts, respectively.69 One furthers his power over other
human beings primarily by attaining status ( jh) and leadership,
which leads one into competition with others. Al-Rz argues that,
initially, the individual may have little desire for these pleasures, but
may easily become disposed to seeking them through habituation, as
he begins to experience them. Once the desire and disposition are
acquired, the degrees to which the individuals power over external
beings may be furthered will be endless.70
Though al-Rz accepts that power is a real perfection in the soul
and leads to the experience of pleasure, he nonetheless censures it for
various reasons; e.g. it is worldly and ceases with the souls departure
from the body, leading to severe pain in the soul, and it fails to bring
about true happiness. By contrast, the attribute of knowledge survives
the death of the body and leads to the souls experience of happiness
posthumously. Moreover, as with the pursuit of sensory pleasures,
which revolve around the individuals physical body, involvement
in the external physical world also constitutes an engagement of the
soul that distracts it from furthering its attribute of knowledge.71
The human good (khayr), therefore, is identified with perfecting the
spiritual attribute of knowledge, to the exclusion of the souls engagement in the physical body and the external world. The soul can be
directed either to the higher, divine realm, as in the case of those
who are spiritually advanced and truly happy, or to the lower, physical world, which is the case of those who are completely engrossed
in worldly pleasures. In the middle, some souls constantly alternate
attention between the upper world, which they approach through
piety and worship, and the lower world, in which they manage various worldly affairs.72 Al-Rz adds that the science that guides to the

68 Nafs, 22; 131; Malib, 1, 260; Tafsr, 12, 173; 17, 216. Cf. al-Ghazl, Iy",
4, 3007.
69 Nafs, 223; 12830.
70 Nafs, 245.
71 E.g. Nafs, 144.
72 Nafs, 256.

124

chapter three

higher level is the science of spiritual discipline (al-riya al-rniyya),

by which he means advanced Sufism.73 That which guides to the
middle level is the science of character (#ilm al-akhlq), which is a more
elementary discipline that describes the way in which the average
person may acquire or evade various character traits (khuluq).
The science of character concerns the pursuit of moral virtue, i.e.
virtue with respect to the non-rational aspects of the human soul. It
allows one to purge his soul from what should not be (m l yanbagh),
by detaching it from its inherent physical engagements.74 Al-Rz
presents a practical ethics of virtue in the second part of Kitb al-Nafs,
which he also refers to as a book on akhlq. Three main vices are
discussed, viz. the love of wealth, the love of status, and hypocrisy
(riy"), for each of which he prescribes both contemplative (#ilm) and
practical (#amal) therapy (#ilj).75 Contemplative therapies consist of
mental meditations and exercises that produce a conviction within
one that one ought to avoid these vices, which produces a sense of
repulsion and detachment from them. Practical therapies consist of
recommendations to adhere to certain forms of conduct, which loosen
the individuals attachment to the activities connected to these vices
and promote contrary dispositions in the self. While these three vices
are related to engagements in the external world, al-Rz seems to
consider his preceding discussion on censuring sensory pleasures to
provide sufficient reasons for turning away from them: perhaps these
may be viewed as contemplative therapies for the souls engagement
in the physical body.
In his theory of virtue, al-Rz places greater emphasis on ridding
the soul of vices, which it acquires by its engagement in the physical
world, than on engendering positive moral virtues, other than those
that should replace these vices. This feature stems from his definition of happiness (sa#da) purely in terms of intellectual pleasure,
which transpires from theoretical perfection. In this, he follows Ibn
Sn, who maintains that the purpose of the souls attachment to
matter, and of its having a practical intellect, is the perfection of its
theoretical aspect, which alone leads to happiness.76 For Ibn Sn,

73
74
75
76

On spiritual discipline, see Shar al-Ishrt, 2, 110 ff.

Malib, 8, 111.
See Nafs, 114; 121; 141; 147; 164.
Ibn Sn, Shif", ab#iyyt, 6, 3741; cf. 6, 1846.

al-rzs perfectionist theory of virtue

125

practical (including moral) development may only reduce the misery

(shaqwa) that the soul experiences in its attachment to the physical
world, and constitutes only a precondition, rather than a cause, for
happiness.77 Therefore, in contrast to Aristotle and al-Frb, he
considers dispositions of intermediacy (tawassu) to be good only on
account of their efficacy in purifying the rational soul from its physical attachments.78 For neither acts nor character traits, according to
Ibn Sn, can be good or virtuous intrinsically.
This is a stance that al-Rz adopts. He presents his definitive
discussion of happiness and intellectual pleasure in the Mulakhkha,79
where he writes with respect to practical perfection, in the section
on eschatology:
Virtuous character traits do not produce happiness; for their objective
is that the soul does not become firmly attached to the body. Their
effect is only that the soul does not become tormented (mu#adhdhab); as
for [the attainment of] happiness, [they do not have any effect].80

While bad character leads to torment, virtuous character is only a

condition for happiness, which is attained through theoretical perfection. The happiness of someone who is advanced theoretically
will be impaired by bad (rad") character traits.81
Al-Rz presents a hierarchy of human souls according to their
theoretical perfection, the happiest being souls that have reached
demonstrative knowledge, followed by uncritical imitators (muqallid), who possess correct beliefs and will be in a state of wellbeing
(salma) from torment.82 Only demonstrative knowledge attainable
through metaphysical enquiry, which provides immutable knowledge
of eternal truths, contributes towards the souls theoretical perfection,
to the exclusion of other fields of enquiry, which provide mutable
knowledge, such as the disciplines of language, Qur"nic exegesis,

77

Ibn Sn, Shif", ab#iyyt, 6, 186; Ilhiyyt, 2, 4279.

Ibn Sn, Shif", Ilhiyyt, 2, 430.
79 This definitiveness is also evident from his references in later works: Shar
al-Ishrt, 2, 92; Shar #Uyn al-ikma, 3, 167.
80 Mulakhkha, fol. 327b328a.
81 Cf. Mabith, 2, 426 ff.; Ibn Sn, Ishrt, 4, 7 ff.; Najt, 326 ff.
82 Mulakhkha, fol. 327b328a. After the souls separation from the body, demonstrative knowledge will become immediate (arr) and will be perfected even further
(Malib, 7, 275).
78

126

chapter three

jurisprudence and traditions (akhbr).83 For this, in order to acquire

discursive (naar) knowledge, one needs learning (ta#allum), assistance
from others, and reliance on the craft of [discursive] procedures (alqnn al-in#), i.e. logic.84
Yet, in most of his works, al-Rz describes a level higher than the
demonstrative level, viz. that of those rare intellects that have a special
ability to attain knowledge with little effort and discursive reflection.
These intellects are unique both quantitatively and qualitatively: they
are able to combine conceptions to apprehend more primary premises
than normal, and they are then able to combine these premises to
arrive at obscure conclusions effortlessly.85 Thus, those individuals,
who include prophets and awliy", will require neither learning nor
the instrument of logic to attain theoretical perfection. Al-Rz here
appears influenced partly by Ibn Sns notion of intuition (ads). He
writes that though all human souls are naturally capable of attaining
discursive knowledge quickly and intuitively without any pursuit,
they are hindered by their engagement in governing their bodies
and by imagination. Because of these hindrances, they will depend
on thought (fikr) to arrive at discursive knowledge.86
Returning to al-Rzs identification of happiness with the experience of intellectual pleasure, which he adopts from Ibn Sn, we find
that he nonetheless rejects his proofs for the possibility of this form
of pleasure and its superiority to other pleasures.87 In the Mulakhkha,
he goes on to explain his own approach to this question:
83 Nafs, 133. In the following chapter, we will see that at a later stage in his
career, al-Rz will change this assessment with respect to Qur"nic exegesis. He
is, after all, the author of one of the largest and most outstanding Qur"nic commentaries of all time.
84 Asrr al-tanzl, 530.
85 Asrr al-tanzl, 52930; cf. Malib, 7, 280; Ma#lim, 1134.
86 Shar al-Ishrt, 1, 157. Cf. Ibn Sn, Ishrt, 1, 3905; Dimitri Gutas, Avicenna
and the Aristotelian Tradition, 159 ff.; Intuition and Thinking.
87 E.g. Ibn Sn argues that since pleasure is the perception of the agreeable, the
more intense the perception and the nobler the object of perception, the more
perfect will be the pleasure. Intellectual perception delves deeper into the essences
of things than do other forms of perception. God, His attributes, the angels and
the cosmos as a whole are nobler than other objects of perception. Therefore, the
intellectual pleasure that results from perceiving them is superior to other pleasures.
However, al-Rz rejects defining pleasure as the perception of the agreeable (see p.
15960 infra), and points out that since the nature and objects of intellectual perception are different from those of sensory perception, there is no evidence that it should
produce pleasure in the perceiver (Mulakhkha, fol. 323b326a; cf. Shar #Uyn al-ikma,
3, 135 ff.; Ibn Sn, Ishrt, 4, 1125). Ibn Sn also argues that though angels do

al-rzs perfectionist theory of virtue

127

We do not deny intellectual pleasure, nor that it is more intense than

other [pleasures]. But this is not provable by logical proofs. Nonetheless,
not all that cannot be proved in this way should be rejected. ... The
only way to accept it with certainty is to experience it. The more one
is detached from physical attachments, and the more attracted he is
to theological knowledge, the greater will his share of it be. God has
bestowed it upon me numerous times, in both sleep and wakefulness,
after my belief in it strengthened and my soul became more attuned
to it.
However, it seems that the falsifa provided the aforementioned proofs
only as directives and attractives. I add to this and say:
Perfection is sought for its own sake[this is known] by induction.
For in every engagement, be it noble or base, one will love (ubb) what
is perfect in it more than what is imperfect. As the levels of perfection
are many, the levels of love too are many. And as the highest perfection
is that of God, exalted, intense love will be for Him alone. Intense
love will then result in two consecutive states: heedlessness (ghafla) of
all but the beloved, and finding pleasure in perceiving the beloved.
Induction indicates this. Loving God, exalted, intensely will undoubtedly produce these two states. The people of [spiritual] experience
(ab al-dhawq) call heedlessness of all but Him annihilation (fan").
... Intense love will thus be reserved for God alone. The heart does
not find tranquillity except in His remembrance (dhikr). ...
What the dilettanti think, that knowledge of any intellectual matter
can be a cause for intellectual pleasure, is false. Rather, pleasure only
results from knowing God, exalted, and from being immersed in loving
Him. Therefore, since human intellects attain knowledge of God only
by knowing His acts, the greater the knowledge of them, and the more
complete the perception of His design, the more complete will loving
Him and finding pleasure in loving Him be.88

This Sufi-inspired notion of theoretical perfection contrasts with Ibn

Sns stance, which al-Rz outlines in the earlier Mabith:
Concerning the extent of knowledge at which this happiness is attained,
the Shaykh says: I cannot explicate this matter.89 However, in the
Mubatht, he is content with the intellection of separate substances
(mufriqt).90 In the Shif" and the Najt, he considers it to be that the

not eat or copulate, their state is superior to that of beasts. Al-Rz retorts that this
is a highly rhetorical argument and rests on the assumption that angels experience
pleasure (Mulakhkha, fol. 325b6a; Shar al-Ishrt, 2, 87; cf. Ibn Sn, Ishrt, 4, 10.
Al-Rz accepts Ibn Sns arguments earlier in Mabith, 2, 4267).
88 Mulakhkha, fol. 326ab (also quoted by al-Bursaw, Tahfut, 1245); cf. Ma#lim,
1134.
89 Cf. Ibn Sn, Shif", Ilhiyyt, 2, 429.
90 Cf. Ibn Sn, Mubatht, 1978; 210.

128

chapter three
human soul acquires true conceptions of all separate primary principles,
asserts them apodictically and demonstratively, knows the final causes
for universal motions to the exclusion of particular ones, fathoms the
form of the entirety [of being], the relations among its various parts,
and the order that initiates from the First Principle to the lowermost
existents in its hierarchy, understands providence and its manner, and
recognises how the entity that is superior to all is distinct in its existence
and oneness, how it knows, without any multiplicity or change whatsoever being attributed to it, and how existents are related to it.91

Attaining these cognitions will require the mastery of both metaphysics and physics, which implies that only a philosopher can attain
theoretical perfection. By contrast, al-Rz maintains that happiness
may be attained only by knowing God, whereas other intellectual
pursuits may distract from this end. One ought to,
... view His creatures qua their requiring a producer who possesses
the attributes of perfection, since knowing them in any other respect
will oppose [the attainment of] this happiness. I have learnt this from
experience (tajriba).92

Human souls may acquire a sense of yearning (shawq) towards knowing God by becoming aware of His being and by contemplating
signs of design in His creatures. When the love (#ishq; maabba) of
God becomes established in the heart, the heart will be repulsed
by all else. Yearning, al-Rz explains, transpires when man gains
some awareness of God, but perceives Him neither constantly nor
completely; he will then desire to attain purer and more complete
and constant knowledge of Him. In this world, mans knowledge
of God will inevitably be obscured to some extent by the distorting
veil of the imagination, which will be lifted in the afterlife. Yet man
will not attain complete knowledge of God in either world, since
the degrees of knowing God are endless. The journey is thus endless, and the degrees of human perfection are infiniteno human
being has ever reached complete knowledge of God. If one attains
complete knowledge of an object of knowledge, his yearning for it

91 Mabith, 2, 429; cf. Ibn Sn, Shif", Ilhiyyt, 2, 429; 2, 4256; Najt, 331.
On Ibn Sns views on this question, see Yahya Michot, La destine de lhomme,
467, n. 78.
92 Mulakhkha, fol. 327a; cf. Tafsr, 4, 233. Cf. al-Ghazl, Iy", 4, 30810.

al-rzs perfectionist theory of virtue

129

will cease; yet, al-Rz points out, attaining greater knowledge of

God will only increase the yearning.93
Prophecy
One may wonder, however, how al-Rz conceives the relation between
this perfectionist stance and revealed religion. Defining goodness in
terms of spiritual perfection does not seem to square easily with
defining it in terms of adherence to Revealed Law, with reference
to the afterlife reward it may lead to. This problem is addressed
most directly in discussions of the nature of prophecy.94
Al-Rzs Traditional Approach to Prophecy
From his earliest writings, al-Rz follows the traditional Ash#ar
approach to proving prophecy, which relies on miracles: whoever
claims to be a prophet and performs miracles, under certain conditions, is a true prophet. God ordains prophets and supports them
with miracles that prove their ordinance.95 There are then lengthy
discussions of what qualifies as a true miracle and of the conditions
that should be met in the character of the claimant of prophethood, and in the contents and circumstances of the claim that he
makes.96
The divine origin of revelation and the truthfulness of a given
claimant of prophethood are recognised from his performance of
miracles, evidence that is not essential to prophecy as such, but
external to it. A particular claim of prophethood may then be
scrutinised internally, not in order to prove its truthfulness, but
93

Tafsr, 4, 2334; Nafs, 59; Shar al-Ishrt, 2, 100.

Only relevant aspects of his prophetology will be examined. On the infallibility of prophets, see his #Imat al-anbiy" (cf. Ceylan, Theology and Tafsir, 1725). On
prophetic miracles, especially the Qur"n, see Nihyat al-jz f diryat al-i#jz.
95 In Ash#ar theology, prophecy is established by divine ordinance, rather than
any characteristics of the prophet himself. People recognise a persons prophethood
only through miracles. In one place in the Tafsr (18, 110), al-Rz writes that
prophecy is not acquired (muktasaba). Elsewhere in the Tafsr (19, 96), he states
that the matter is more complex than prophecy being either a gift (hiba; minna)
from God, as the exoterics (hiriyyn) of the Sunnis maintain, or its following from
properties of the prophet himself, as some Muslim falsifa maintain.
96 E.g. Ishra, fol. 46b51b; Nihya, fol. 207b233a; Muaal, 492 ff. (where he
refers to the discussion in the Nihya); Arba#n, 30229; Khamsn, 64.
94

130

chapter three

to ascertain whether some of the claims it makes are plainly false,

should they contradict the certainties of unaided reason (e.g. that
the world is temporal, and that God is one).
In many works, al-Rz will then devote long sections to proving prophecy from miracles and to refuting objections advanced by
falsifa, atheists, followers of other religions, and sceptics of various
sorts. Even in the Malib, a very late work, he defends this proof
against a long list of objections, basing it on both the premise that
God is a free agent who creates all existents directly and his rejection
of moral realism.97 The former addresses the objection that miracles
are unusual, yet perfectly natural, events, which are caused, e.g., by
the persons psychological qualities, higher causes, or lower spirits.
The latter principle relates to various objections, e.g. that obligating humans (taklf) is morally reprehensible, and that revelation is
superfluous, since morality is knowable by autonomous reason.
This traditional Ash#ar proof competed with an influential falsaf
theory of prophecy that Ibn Sn developed and articulated. According to this theory, a prophet will have certain psychological faculties
that are actualised in a more superior way than those of the average person. He will have a perfect and intense imaginative faculty,
which will receive intelligible forms from the higher heavenly souls
and present them in the form of imagery, as well as a perfect intellectual faculty, which, by its contact with the Active Intellect, will be
able to arrive intuitively at conclusions from premises without the
mediation of syllogisms.98 Revelation (way) is the emanation from
these higher causes to the prophets psychological faculties, and the
message (risla) is the result of this contact.99 The angelic image
seen by the prophet also results from this contact and is formed by
his imaginative faculty.100
However, this psychological model only shows that prophecy is
possible. Ibn Sn then argues that the appearance of prophets is
necessary for the spiritual advancement of common people, and (following Ikhwn al-af and al-Frb) that prophecy is also a sociopolitical necessity. Only prophets are capable of introducing laws

97

Malib, 8, 7100.
Cf. Ibn Sn, Awl, 11426; Ithbt al-nubuwwt, 417.
99 Ibn Sn, Ithbt al-nubuwwt, 47.
100 On Ibn Sns theory of prophecy, see: Y. Michot, La destine de lhomme, 118
ff.; H. Davidson, Alfarabi, Avicenna, and Averroes on Intellect, 116 ff.
98

al-rzs perfectionist theory of virtue

131

that organise human life and association, and are thus necessary for
human wellbeing (al). Therefore, he concludes, divine providence
(#inya) will necessarily provide prophets for mankind.101
Despite his early Ash#arism, al-Rz seems to accept this theory
of prophecy in the Mabith, one of his earliest falsaf works, where
he often follows Ibn Sn closely. His discussion of prophecy here is
generally paraphrased from Ibn Sns works, including the notions
of the appearance of the prophet being a socio-political necessity,
the special psychological nature of the prophet, and the process of
revelation.102
However, in the later Mulakhkha and Shar al-Ishrt, al-Rz totally
rejects Ibn Sns theory of prophecy. Against his psychological explanation of supernatural events (including miracles, the visual perception of angels and the reception of revelations), al-Rz argues, for
instance, that such events could have other natural explanations.103
Also, against Ibn Sns teleological explanation to prophecy, he
presents the following argument in Shar al-Ishrt:
What is the meaning of your saying, Since people in this world need
a lawgiver, his existence will be necessary (wjib)? ... If you mean that
it is obligatory on God to create and existentiate him (as the Mu#tazila
say, Compensation is obligatory upon God, i.e. that if He does not
deliver it, He will deserve blame), then that will go against what the
falsifa claim in the first place.
However, if you mean that since the existence of the prophet brings
about the order of this world, and since it has been proved that God
is the source of every perfection and good, it will be necessary that
God causes that person to come into being, then this too is false. For
we say: Not all that is most advantageous (ala) to this world will
happen necessarily in this world. For had the people of this world
been naturally disposed to goodness and virtues, that would have been
more advantageous than their present condition; yet that is not the
case. Therefore, it is conceivable for the existence of the prophet to

101

Ibn Sn, Shif", Ilhiyyt, 2, 4416; Najt, 33840; Ishrt, 3, 2267.

Mabith, 2, 5234; cf. Ibn Sn, Shif", Ilhiyyt, 2, 4412. On prophetic
psychology, see Mabith, 2, 41724; cf. Ibn Sn, Karmt, 22540.
103 Mulakhkha, fol. 320b323a; Malib, 8, 1367. The producer of miracles
could be angels, jinn, or heavenly bodies. Or, according to al-Rzs view that
the essences of human souls differ, the prophet could have a unique psychological
property that would enable him to perform miracles. (On classical Ash#ar objections to falsaf conceptions of prophecy, see M. Marmura, Avicennas Theory of
Prophecy in the Light of Ash#arite Theology, esp. 1604).
102

132

chapter three
be more advantageous than his non-existence, yet for [no prophet]
to ever exist!
If he meant something else, then that ought to be explained, so that
we may investigate its plausibility or falsehood.104

The argument underlines an epistemological weakness in Ibn Sns

teleological proof of prophecy. Assuming that this world is indeed the
best of all possible worlds, there is no way of ascertaining whether a
certain human advantage, no matter how urgently needed, is necessary or not. Indeed, we can conceive of other possible worlds that
would be more or less advantageous to humans than this world (e.g.
a world with more or fewer human illnesses); yet we will be unable
to judge any as necessary. A world in which perfect lawgivers, each
with a perfect psyche that would enable him to receive knowledge
from above and communicate it to humans in a way that would
serve their interests, is, according to al-Rz, perfectly conceivable,
as indeed are other possible worlds in which human wellbeing is
catered for in a more perfect manner than in this world.
In the Mulakhkha, al-Rz presents occasionalism and voluntarism
as the correct alternatives to psychological and teleological approaches
to explaining and proving prophecy. He argues that the only way to
proving prophecy is by affirming Gods unconstrained will and His
absolute omnipotence, and using the proof from miracles alone.105
In this falsaf work, we are thus referred to kalm.
On the Proof of Prophecy in the Muaal
At a strictly theological level, al-Rz and all classical Ash#ars maintain that not all of Gods commands and acts that affect humans
are aimed at human advantage. However, is it possible that at least
some of His acts are aimed at human advantage? In some of his
earlier works, al-Rz answers in the affirmative: it is conceivable
that some divine acts are performed for the sake of human advantage. However, unaided reason has no means to arriving at more
detailed knowledge beyond affirming this possibility. He writes that
it is conceivable that [God] sends prophets for the sake of a benefit
that we do not know.106
Yet, as we saw in the previous chapter, al-Rz maintains that
104
105
106

Shar al-Ishrt, 2, 106.

Mulakhkha, fol. 323a.
Ishra, fol. 46b; cf. Arba#n, 3289.

al-rzs perfectionist theory of virtue

133

although it is inconceivable to affirm motives (hence purposes) for

Gods acts and commands, it is often possible to study the contents
of revealed texts and to recognise, inductively, aspects thereof that
serve certain human advantages. Such induction is required in jurisprudence, especially in the method of qiys.
In the Muaal, al-Rz takes this same approach in responding to
the Barhima, who argue that prophecy is pointless; for our minds are
able to know the good, the bad, the beneficial and the harmful, without the aid of revelation.107 If prophecy agrees with this knowledge,
it will be superfluous; if not, it will be wrong.108 Though he would
normally reply to such arguments by rejecting ethical rationalism
and asserting that Gods acts and commands cannot be judged ethically, here he chooses to cite examples of how humans in fact receive
two main types of great benefit from revelation. First, it confirms
knowledge that unaided reason may attain. For instance, while the
mind may know that the world requires a wise maker, prophets will
bear further witness to this and make belief in it obligatory: this is a
great benefit for humans.109 Second, revelation provides knowledge
that unaided reason cannot attain. Al-Rz cites several examples
that people have given for this type of knowledge: e.g. (a) knowledge
of some divine attributes, such as hearing, sight and speech, (b) the
organisation of society and the reduction of disorder through legal
systems, and (c) knowledge of poisonous plants (which would otherwise require dangerous experimentation).110 Some examples (such
as the last) are ones that al-Rz himself does not accept. The thrust
of this general argument is that indeed there do appear to be many
benefits from revelation. Nonetheless, he does not suggest that any
of these are necessary aspects of prophecy, or can be relied upon
in proving prophecy.
In this same work, he also cites the following proof of prophecy,
alongside the proof from miracles:
The second way to proving the prophecy of [Muammad], peace be
upon him, is the inference (istidll) from his character traits, acts, judgements and conduct. Although each of these alone does not indicate

107 On them, cf. article Barhima, EI2; Norman Calder, The Barhima:
Literary Construct and Historical Reality.
108 Muaal, 5034; cf. Arba#n, 3279.
109 Muaal, 5124.
110 Muaal, 5149.

134

chapter three
prophecy, their combination is known, with certainty, to be found only
in prophets. This was the preferred way of al-Ji and was adopted
by al-Ghazl in his book the Munqidh.111

Al-Rz, in the Muaal, does not endorse this proof from the characteristics of the claimant of prophethood, and the listed benefits of
prophecy are not given in support of it. Instead, he writes:
What should be relied upon (al-mu#tamad) in [proving] the messengership of Muammad, peace be upon him, is his presentation of [the
miracle of] the Qur"n. All other approaches may be cited as complimentary evidence.112

Al-Rz here prefers the proof from miracles, but shows some interest in teleological approaches to prophecy.113
The Teleological Approach to Prophecy
In works later than the Muaal, probably from the Tafsr onwards,
al-Rz adopts a proof of prophecy akin to that attributed to al-Ji
and al-Ghazl, which he will now consider to be superior to the
proof from miracles. In the Ma#lim, he introduces it as follows:
In the first method [to proving prophecy], we prove [the claimants]
prophethood through miracles. Once his prophethood is affirmed, we
infer the truth of his sayings and deeds from this affirmation. As for
this [second] method, we show that all that he presents in speech and
action are the deeds of prophets; therefore, he will necessarily be a
truthful and true prophet, who is sent by God.114

And in the Malib:

The second [method] is to say: We know initially which doctrines are
true and which acts are correct (awb). Knowing these, if we see a
person calling people to true religion, and if we see his speech having
a powerful effect in diverting people from falsehood to truth, we will
realise that he is a truthful prophet, whom it is obligatory to follow.
This method is more rational, and fewer doubts surround it.115

111

Muaal, 491; cf. Tafsr, 32, 41.

Muaal, 508.
113 Frank Griffel (Al-[azls Concept of Prophecy, 106) notes that this proof
is rejected in the Muaal. However, the objection that al-Rz cites is not his own
(Muaal, 5001).
114 Ma#lim, 93. Cf. Tafsr, 17, 1147; 19, 98; Ma#lim, 935; Malib, 8, 10325;
Tanbh, 34 ff.
115 Malib, 8, 103.
112

al-rzs perfectionist theory of virtue

135

Al-Rz refers to this method as the proof of prophecy from the

prophets ability to perfect those who are imperfect (takml al-nqin).
Human perfection, at which happiness is attained, is realised in
the perfecting of the individuals theoretical and practical faculties.
The perfection of the practical faculty involves the acquisition of
a disposition (malaka) to perform good acts, which draw one away
from sensory pleasures towards the hereafter and a spiritual mode
of being.
Al-Rz posits a hierarchy of human souls: (a) the majority of
people are imperfect (nqi); (b) the awliy" have both theoretical and
practical perfection, but are incapable of perfecting others; and (c)
prophets are perfect in both respects, and are capable of perfecting
those who are imperfect. There are also infinite degrees of perfection
and imperfection, varying in weakness and intensity;116 the awliy"
vary in their degrees of perfection, and prophets vary in their capacities to perfect others.
The distinction between prophets and awliy" in both qualities is
a matter of degree: though a wal may perfect some people, only a
prophet will be able to perfect multitudes of people.117 Hence, prophets are more perfect than awliy"; the wal is perfect and complete,
yet the prophet will have absolute perfection (kaml mulaq), and will
be complete and above completeness (tmm wa-fawq al-tamm).118
However, it appears that there are cases where prophets and awliy"
from different times are equal in their ability to perfect others.119
Al-Rzs proof at once attempts to prove both the necessity of
prophecy as such and the truth of a particular claim of prophethood.
He argues that this second proof of prophecy is superior to the proof
from miracles, since it is a demonstration turning on the question
why? (burhn al-li-ma). It relies on understanding the essence of what
is being proved, and showing that it is true by virtue of that essence
itself. He writes that when we know the essence of prophethood to
116

Malib, 8, 104.
Ma#lim, 101.
118 Tanbh, 38. Ibn Sn (Shif", Ilhiyyt, 1, 186 ff.; cf. 2, 355) defines tmm wafawq al-tamm as the thing being perfect in a certain respect and providing benefit
of the same respect to other beings.
119 This, al-Rz writes, is the meaning of the adth, The scholars of my community are the likes of the prophets of the Israelites (Tafsr, 17, 115; cf. 19, 98).
However, he makes it clear that a non-prophet never reaches the level of any
prophet, and he criticises those who place #Al above some previous prophets (Tafsr,
8, 86; cf. Lawmi#, 3201).
117

136

chapter three

be the perfection of imperfect people by a perfect person, we will

be able to judge whether a particular person is a prophet in essence
or not. By contrast, the proof of prophethood through miracles is a
demonstration of the fact (burhn al-inna), since it involves our knowing the cause from one of its effects, without necessarily knowing its
essence. We may know that the prophet is given a high status by
God; yet we will not comprehend the nature of that status.120
In its most general and basic framework, this theory of prophecy
draws on a number of influences. In the Muaal, al-Rz specifies
al-Ji (d. 255/869) and al-Ghazl as two previous exponents of
this proof of prophecy. Al-Ji indeed appears to be one of the first
to advance such an approach to prophethood, albeit in rudimentary
form. In his \ujaj al-nubuwwa, he proclaims to defend prophecy against
objections and to fill gaps that are due, not to lack of evidence, but
to the absence of a systematic analysis of evidence.121 As well as
using the argument from miracles, al-Ji argues that since men are
created imperfect (nqi) and heedless of their own interests, it will
be necessary that prophets be sent to guide them, so that people do
not have any excuse (ujja) against God on the day of judgement.122
The highly Mu#tazil character of this approach is apparent, since it
effectively places God under a moral imperative to act in the interest of humans. Al-Ji then writes that a prophet fit to guide and
perfect others will have to be someone with both high moral traits
and excellence in conduct.123
In contrast to al-Ji, whose main concern lies in moral and practical aspects of human life, al-Ghazl, in the Munqidh, emphasises the
role of the prophet with respect to human spiritual and theoretical
perfection. The proof of prophethood, for him, lies in evidence of
the prophets success in perfecting multitudes of people who are in
a state of natural ignorance.124 This proof, he writes, is superior to

120

Tafsr, 17, 115; Ma#lim, 945; Malib, 8, 123. On these two types of demonstration, see al-Frb, Burhn, 26; Ibn Sn, Najt, 1034.
121 Al-Ji, \ujaj, 12834.
122 Al-Ji, \ujaj, 1367. A Qur"nic reference is implied (Qur. 4:165).
123 Al-Ji, \ujaj, 1567.
124 Al-Ghazl writes: If you understand what it is to be a prophet, and have
devoted much time to the study of the Qur"n and the traditions, you will arrive
at an immediate knowledge of the fact that [Muammad], peace be upon him,
is in the highest grades of prophethood. Convince yourself of that by trying out

al-rzs perfectionist theory of virtue

137

the proof from miracles.125 Interestingly, al-Rz also indicates that

he adopts the distinction between the prophet and the wal from an
unspecified work by al-Ghazl. According to this model, al-Rz
writes, the wal is the perfect man (al-insn al-kmil) who is incapable
of perfecting others, while the prophet is the perfect man who at
the same time perfects others (al-insn al-kmil al-mukammil). At the
other end of the spectrum, there are those who are astray (ll), or
astray and lead others astray (ll muill).126
Notwithstanding the influence of these two earlier theologians, later
commentators note that al-Rz adopts his later theory of prophecy
from the falsifa.127 As we saw, Ibn Sn indeed conceives the prophet
as a person of high spiritual nature, whose existence is necessary
for the realisation of both the perfection of society and the spiritual
perfection of the individual. Yet the spiritual role of prophecy is
relatively minor for Ibn Sn, who (as we will see below) maintains
that the prophets immediate purpose is the practical, rather than
the theoretical, perfection of humanity.
Al-Rzs later theory of prophecy also has much in common with
al-Shahrastns discussion of prophecy in al-Milal wa-l-nial, in which
he showed particular interest.128 Al-Shahrastn writes that prophecy

what he said about the influence of devotional practices on the purification of the
heart. ... When you have made trial of these in a thousand or several thousand
instances, you will arrive at an immediate knowledge beyond all doubt (Munqidh,
149; cf. 14450; W. Montgomery Watts translation, with modification, Faith and
Practice, 67).
125 Al-Ghazl, Munqidh, 14950.
126 Tafsr, 19, 98. I have not found this distinction in any of al-Ghazls works.
The closest seems to be al-Risla al-Laduniyya (cf. 69 ff.), which al-Rz refers to
expressly in a different context (Tafsr, 21, 149).
127 E.g. al-^j, Mawqif, 8, 261 (referring to the Malib); Ibn Taymiyya, Majm#,
4, 99; Khwansr, Rawt, 701 (quoting an earlier author, in reference to the
Ma#lim).
128 See p. 114 supra. Al-Shahrastn (Milal, 2, 944; cf. Mura#a, 1302) designates
this notion of prophecy as that of the \unaf", prophetic monotheists, who call to
the return to human primordial nature (fira), and are represented in his time by
(at least some) Muslims. (Elsewhere, he refers to Islamic sects that have departed
from al-Milla al-\anfiyya, 1, 36). They defend prophetology as an alternative to the
Sabian deification of spiritual entities (rniyyt). These are the two main creeds
that existed at the time of Abraham (1, 230). The discussion is represented as a
summary of debates that occurred between the two groups, yet it is clear that the
author is on the side of the \unaf" (cf. e.g. Hermes Trismegistos is not Sabian,
God forbid!, 2, 44; Abraham refutes Sabianism, 2, 53; cf. Nihyat al-aqdm, 4269;
4645), and that, regardless of whether he considers the debates real or not, the

138

chapter three

involves two perfections: that the prophet is perfect in himself, and

that he is a perfecter (mukammil) of others. The prophet has both theoretical and practical perfection, as well as the ability to perfect others
in both respects.129 The prophetology presented by al-Shahrastn
here manifests clear, albeit mild, Ism#l undertones.130
Al-Ji, Ibn Sn, al-Ghazl and al-Shahrastn all conceive
of prophecy in terms of attributes that are essential to the prophet
himself, and argue that the existence of the prophet is necessary by
virtue of what he is and does. But how does al-Rz conceive of the
necessity of prophethood? He writes in the Malib:
We have shown that perfection and imperfection appear in various
degrees and disparate levels among people. Therefore, as we are able
to see individuals, who have reached great proximity to cattle and
beasts in imperfection, stupidity and heedlessness, similarly, on the side
of perfection, there will have to exist perfect and virtuous individuals.
Necessarily, there will exist among them an individual who is the most
perfect and virtuous of them. He will be at the last stages of humanity
and the first stages of angelhood.131

Earlier in this work, we find a brief, milder form of the argument:

since representatives of all the lower human levels exist, it is highly
probable (l yab#udu) that representatives of the most perfect levels
exist.132
Curiously, al-Rz then goes much further in explicating this
hierarchy of human perfection. There will have to exist one perfect
person in every temporal cycle (dawr), whom the Sufis call the pole

arguments directed at the Sabians are his own; yet he does not admit this, since
the Milal is meant to contain unbiased accounts of creeds (1, 16).
129 Al-Shahrastn, Milal, 2, 11 ff. Cf. Nihyat al-aqdm, 4634, where he refers
only to the prophets perfection, not his ability to perfect others.
130 Cf. al-Shahrastns brief account of Ism#l prophetology (Milal, 1, 193 ff.). On
his Ism#lism, see W. Madelungs introduction to his Mura#at al-falsifa (27).
In one place, al-Rz explains prophecy in terms akin to al-Shahrastns. He
writes that two worlds exist: the lower world and the world of spiritual beings
(rniyyt). There necessarily exists in the latter a being, the angelic messenger
(al-rasl al-malak), who is the highest of all beings in it, and the obeyed (mu#)
among them. Similarly, there will have to exist, in this lower world, one who is
the highest and the one obeyed in it, viz. the human messenger. The former will
be the source (madar), the latter the manifestation (mahar) (Tafsr, 1, 264; cf. Asrr
al-tanzl, 310. Cf. al-Shahrastn, Milal, 2, 15).
131 Malib, 8, 105; cf. 1, 261.
132 Malib, 7, 280.

al-rzs perfectionist theory of virtue

139

(qub).133 He will be what is intended of this elemental world (i.e. the

final cause of creation), all other creatures being intended accidentally (bi-l-#ara). In each group of cycles (perhaps in each thousand
years, perhaps more, or less), there will be one person who is the
most perfect of all those high individuals, who will be able to perfect
others and present a law and a message, viz. a prophet. Al-Rz does
not explain how he arrives at these conclusions, which betray Sufi
and Ism#l, gnostic influences, to be added to the expanding list of
influences on his theory of prophecy. It appears that in formulating
his teleological notion of prophecy, which by his time seems to have
become a rather commonplace notion, he draws upon a wide and
eclectic range of sources.
Poles and cycles aside, how does the view that the prophet
represents the highest level of perfection in the human species imply
that his existence is necessary? Al-Rzs argument is akin to an
argument advanced by al-Shahrastn, viz. that each cosmic hierarchy will, necessarily, have actual representatives of each possibility
it contains, from the absolute highest to the absolute lowest, from
the most perfect to the most imperfect.134 Both arguments are then
reminiscent of a proof of prophecy that Ibn Sn advances, viz.

133

On this notion, cf. |ub, EI2. The Sh#a, he adds, call him the infallible
(ma#m) imm and the lord of the age (ib al-zamn), and say he is hidden (gh"ib).
Al-Rzs interpretation is nevertheless heretical from the Imm point of view,
since he refers, not to the twelfth imm, Muammad al-Mahd, but to members
of a line of Sufi poles. He writes:
[The Imms] are correct in [their] descriptions [of this perfect person]. For
since he has none of the imperfections that are in others, he is exempt (ma#m)
from imperfections. He is the lord of the age because that individual is what
is intended in itself in that age. ... He is hidden (gh"ib) from people; for
people do not know that that individual is the best and most perfect of the
people of that cycle. I would also add that that individual probably does not
know that he is the best of the people of that cycle; for though he knows his
own state, he will not be able to know the state of others. Therefore, others
do not know who that individual is, and he does not know who he really is
(Malib, 8, 106; cf. 1, 260).
Al-Rz here argues that a main factor affecting this human hierarchy is climate,
the most perfect people being the inhabitants of Irnshahr, who will produce the
most perfect person of his time. Could this be a claim to be himself the renewer
(mujaddid) of faith at the end of the 6th century A.H.? Cf. Tony Street, Life and
Works, 415.
134 Al-Shahrastn, Milal, 2, 11 ff.

140

chapter three

that since there is a hierarchy of possible human forms of perception (idrk), there will have to appear actual representatives of each.
The prophet will be the human being in whom the highest type of
perception is actualised most perfectly.135
Yet, al-Rz introduces a subtle, but crucial, modification to Ibn
Sns proof of prophecy. For Ibn Sn, the necessity of prophecy is
ontological, and rests on his view that for a property to be potential
in some beings, it must be actual in at least one other being. In the
case of prophecy, the human psychological property in question is
intuition (ads).136 This is not an argument that al-Rz makes.137 In
contrast to Ibn Sns argument from metaphysical necessity, al-Rz
points out expressly that he bases his view that the appearance of
prophets is necessary on induction (istiqr"), which is of a type that
provides complete certainty (al-qa# wa-l-yaqn).138 He argues that
from observing natural beings, we recognise a hierarchy of beings
that constitutes a cosmological order. When we consider the hierarchy of elemental objects, we find that its main divisions (viz. the
mineral, the vegetative and the animal) are connected (muttail), such
that the highest species in the lower division is very similar to the
lowest species in the higher division. The same is true of their subdivisions and further subdivisions. Therefore, having found, through
observation, actual representatives for each level of possible beings,
except one level, we conclude that that level too must have some
actual representation. In the case of the human species, which is
the most superior subdivision of the animal type, since we find men
who are very close to beasts in their characteristics, we will expect
that some men who are very close to angels (the next higher form
of beings) in their characteristics should exist. In them, both the
theoretical and practical faculties will be so perfect and intense that
they will have access to higher knowledge and be able to perfect
others. They will possess angelic properties and be detached from

135

Ibn Sn, Ithbt al-nubuwwt, 427; cf. M. Marmura, Avicennas Psychological Proof of Prophecy.
136 Cf. M. Marmura, Avicennas Psychological Proof of Prophecy, 53;
D. Gutas, Intuition and thinking, 29 ff.; F. Griffel, Al-[azls Concept of
Prophecy, 1123.
137 Griffel (Al-[azls Concept of Prophecy, 1123) seems to suggest that
al-Rz uses the same line of argument as Ibn Sn.
138 Malib, 8, 105; 8, 107.

al-rzs perfectionist theory of virtue

141

matter and have control over it, which will enable them to perform
miracles. Al-Rz writes:
[Such a person,] who is the most perfect of those perfect, and the best
of those most virtuous and knowledgeable, will be at the frontier of the
supernal horizon (ufuq) of humanity. You have come to know that the
end of each species is connected to the beginning of the more superior
species. Since the angels are superior to the human species, the end of
humanity will be connected to the beginning of angelhood. Since we
have shown that that man exists in the highest stages of humanity, it
becomes necessary that he be connected to the realm of angels and
that he be in their midst.139

Since it rests on induction, al-Rzs cosmological explanation of

prophecy is of a lower metaphysical order than Ibn Sns argument.
Rather, it appears more reminiscent of one theory of prophecy
presented by al-Rghib al-Ifahn, who nonetheless does not use
it as a proof of prophecy.140
Likewise, al-Rzs notion of the necessity of prophecy contrasts
with the Baghdd Mu#tazil view that prophecy is necessary because
God has to do what is most advantageous (ala) to human beings.
A comparable moral consideration is also hinted at in al-Jis
argument. Although al-Rzs proof for the necessity of prophecy
starts with an objective notion of the human good, and reaches the
conclusion that there necessarily has to appear a person or an event
whose purpose is to serve this good, it does not imply moral assumptions in relation to God (e.g. that He has to do and command what
is good to humans). It rather rests both on the above induction from
the observation of natural beings and on a conception of the nature
of the perfect human being who has angelic properties.
Yet the latter conceptionfrom which follows that the prophet
and his message must have certain essential qualities, which are
known by autonomous reasonis also problematic when set against
the background of al-Rzs earlier theology. For it will imply that a
prophet will necessarily possess theoretical and practical perfections
and the ability to perfect others in both respects. Such a view is
rejected expressly in the classical Ash#ar notion of prophecy, which
only allows for some conditions that should be fulfilled in a true

139
140

Malib, 8, 107.
Al-Rghib al-Ifahn, I #tiqdt, 1201.

142

chapter three

prophet, but in themselves do not prove his prophethood.141 Thus,

a true message will not contradict certain theological truths that
are known to reason (none of which are moral), e.g. monotheism.
Prophecy may then be proved through miracles only. By contrast,
in al-Rzs later proof, the style and contents of a message and the
character of the claimant of prophecy may themselves constitute its
proof. Moreover, these qualities of both the prophet and his message
include moral ones. In this instance, al-Rz clearly departs from the
classical Ash#ar position.

Prophecy and Human Perfection

So how does the prophet perfect others? This question will lead us
into one of the most important aspects of al-Rzs later thought,
which lies at the background of his exegetical approach in his great
commentary on the Qur"n, the main manifestation of the prophethood of Muammad.
In some places, al-Rz seems to present the view that the prophets
ability to perfect others lies in the prophets soul itself, which emanates lights upon the souls of those who are imperfect and thereby
perfects them.142 He writes:
The soul of Muammad, peace be upon him, was a powerful, luminous, pure and radiant soul. So if Muammad supplicates [God] on
[peoples] behalf ... effects of his spiritual power will emanate (fa)
upon their souls. Their souls will become illuminated by this, their
spirits will become purified, and they will be transformed from darkness into light, and from corporeality into spirituality.143

This notion echoes an earlier, commonplace gnostic view, which finds

expression, for instance, in the works of Ikhwn al-af.144
Yet to say that the prophet perfects others does not refer mainly
to this notion, but to his effect on other human beings through his

141

Cf. al-Mward, A#lm al-nubuwwa, 56 ff.

Tafsr, 17, 1167.
143 Tafsr, 16, 183; cf. Asrr al-tanzl, 546.
144 Ikhwn al-af (Ras"il, 3, 347) write, If complete and perfect souls leave
their bodies, they become involved in supporting (ta"yd) imperfect souls attached to
matter, so that the latter become complete and the former become perfect.
142

al-rzs perfectionist theory of virtue

143

teachings, guidance and exhortation.145 Al-Rz explains this in a

crucial section in his discussion of prophecy in the Malib entitled
On the manner of this invitation (da#wa) to God, which follows his
explanation of how the prophet is perfect in himself. At the beginning of this section, he notes that although the essential purpose of
prophecy is to perfect people by directing them towards matters of
religious concern (dn), viz. spirituality and the hereafter, the prophet
will also be required to address some matters of this-worldly concern (duny), since these too are aspects of the human condition. His
subsequent discussion, however, concentrates on the former, otherworldly concern, which has three main components, or functions: (a)
a metaphysical component (the past), corresponding to theoretical
perfection, (b) a soteriological component (the present), corresponding to practical perfection, and (c) an eschatological component (the
future), corresponding to the final end of man.146
With respect to the first component, the prophet has to teach
certain doctrines, which relate, e.g., to the world being created and
the Creators existence, pre-eternity, attributes, transcendence, etc.
However, al-Rz argues that a prophet should present neither kalm
nor falsafa in affirming and explicating these truths. The style and
approach of this first component in any effective prophecy should
necessarily have two key features.
Firstly, abstract and complex theological analysis and dialectic
will provoke the audience to advance objections and counterarguments, leading to a descent into disputation and squabble. This
would completely undermine the aim of prophecy, which, instead,
relies on a balance between demonstrations (burhn) and rhetorical
(khab) arguments, which inculcate fear and desire (targhb wa-tarhb). For this reason, in all of his later works, al-Rz maintains that
an effective soteriology will utilise a mixture of demonstrative and

145

Malib, 8, 1112; Tafsr, 17, 117.

Malib, 8, 11520; cf. 1, 604. This highlights a main difference between
the theories of prophecy of al-Rz and Ibn Sn, who is influenced by al-Frbs
emphasis on the socio-political aspect of prophecy. One of the main aims of the
prophet, for Ibn Sn (Shif", Ilhiyyt, 2, 4413; Ishrt, 3, 2267), is to provide
laws that will maintain social order and wellbeing. For al-Rz, who, as we will
see, is a socio-political pessimist, the prophet addresses purely social matters, of no
immediate relevance to religion, only because they are unavoidable, not because
their perfection is an end in itself.
146

144

chapter three

rhetorical arguments, to the exclusion of dialectic (jadal), which may

be left as a last resort.
In the logic section of Shar #Uyn al-ikma, he compares rhetoric,
dialectic and demonstration. Rhetoric, he argues, is more effective
than demonstration on common people, who are persuaded by arguments that use widely-accepted (mashhrt) and admitted (musallamt)
premises,147 more so than by those based on a priori premises:
The ignorance and misguidance of the denier of truth may reach such
a level where he will not shy from denying certainties. Yet he will not
be able to deny widely-accepted beliefs. For one who denies them will
be, as it were, in conflict with most people in this world, which will
make him vulnerable to sorts of detriment and calamity.148

Rhetorical discourse presupposes the listeners constant and reverential acceptance of the orators words. However, if one uses the
same sorts of premises to discusses opinions (al-ql wa-l-ql), he will
become a dialectician and will compromise the halo of reverence
associated with rhetoric.149
Al-Rz thus argues that the art of rhetoric is superior to dialectic,
since dialectic persuades neither specialists nor common people,
whereas rhetoric persuades common people. Dialectic fails to persuade common people for three reasons. (a) Dialectical deductions
are subtle and beyond the minds of common people. (b) When a
common person becomes cornered by a dialectical deduction and
unable to reply, he will explain that by his ignorance of the reply,
not by the strength of the argument. (c) The dialectician assumes
a competitive and aggressive stance; therefore, if one believes that
another tries to subdue him and expose his weakness, he will be
moved to refute him and not to be persuaded by him. Al-Rz
continues:
Indeed, the arts that provide assent (iqrr) are demonstration and rhetoric. Demonstration provides certainty for specialists. Rhetoric provides
persuasion for common people. God, therefore, says: Call unto the
path of your Lord with wisdom (ikma) (i.e. demonstration) and kindly
exhortation (maw#ia asana) (i.e. rhetoric) and argue (jdil) with them in
the best manner.150 He mentions dialectic after these two arts, since

147
148
149
150

On these premises, see: Ibn Sn, Ishrt, 1, 289 ff.

Shar #Uyn al-ikma, 1, 229.
Shar #Uyn al-ikma, 1, 2256.
Qur. 16:125.

al-rzs perfectionist theory of virtue

145

they aim at positive persuasion, whereas dialectic aims at refutation.

The first two provide what should be (m yanbagh), and the third is for
the negation for what should not be [i.e. of opinions].151

Al-Rz maintains that this method, which combines demonstration

and rhetoric is exemplified in the content and literary style of the
Qur"n.
In the Malib, he then explains the second feature of prophecy
with respect to its cognitive component in a crucial passage, which
deserves to be cited in full. This passage, in my view, is key towards
fully appreciating the authors exegetical approach in the Tafsr.
It is inconceivable (l yajz) for the [prophet] to proclaim pure deanthropomorphism (tanzh ma);152 for the hearts of most people repulse
from accepting such exposition. If it were proclaimed, it would cause
the aversion (nafra) of most people from it. Rather, it will be necessary
for him to explain that [God], most transcendent and high, is exalted
above similitude to temporal things and resemblance to contingentsso
He, exalted, says, Naught is as His likeness;153 then, afterwards, He
says, He is the Subduer (al-Qhir) over His servants;154 Unto Him
good words ascend;155 and The Beneficent is established on the
Throne.156 He should prohibit them from investigating these defiles
and from wading into these subtleties (daq"iq), except for one who is
highly intelligent and accomplished, since, by his profound intelligence,
he will comprehend the realities of things.
[The prophet] also shows to them mans being a producer [of acts]
and an agent, capable of both acting and omitting, and of both good
and evil; and he does that to the maximum extent (blagha). For were
he to present to them pure determinism,157 they would abandon it and
not pay any attention to it. He also explains to them that although
things are such, all is by Gods decree and determination (qa" waqadar); so not even as much as an atom in the heavens and the earth
will escape His knowledge and judgement. He then prohibits them, in
the strongest terms, from wading into these subtleties; for the natures
of most men are alien to these matters.
In short, the best method to call people to servanthood to God is the

151

Shar #Uyn al-ikma, 1, 2523. Cf. Ibn Rushd, Fal al-maql, 45 ff.
Notwithstanding, al-Rz maintains that exalting God above all aspects of
similarity to created beings is, theologically, the correct position. Hence, Book 2
of the Malib is devoted to tanzh (cf. Malib, 8, 115).
153 Qur. 42:11.
154 Qur. 6:61.
155 Qur. 35:10.
156 Qur. 20:5.
157 Reading al-jabr, instead of al-khayr.
152

146

chapter three
method that was introduced by the master of all prophets, Muammad, peace be upon him. For he proclaims Gods greatness (ta#m)
in all respects to the maximum extent, in a general, concise way (#al
sabl al-ijml), and prohibits [men] from wading into details. So he
presents, in affirming deanthropomorphism, His saying, God is the
Self-sufficient and you are the needy.158 And if He is absolutely selfsufficient, it will be inconceivable that He be composite of parts; it
follows that He is not spatial; therefore, it will be inconceivable that
He exists in place and space. He also states, Naught is as His likeness. If He were a body, His essence would have been similar to all
bodies, since we maintain that all bodies are alike [in physicality]. He
then provides many statements in affirmation (ithbt), and he does that
to the maximum extent. This is necessary; for were He not to present
such statements, He would not have established [Gods] existence for
most people.
Also, he goes to the maximum extent in establishing His knowing
all objects of knowledge. So he says, With Him are the keys of the
unseen; none but He knows them;159 and He says, God knows that
which every female bears and that of which the wombs fall short of
completion.160 He then does not explain whether He knows by His
essence or by [an attribute of] knowledge.161
In many verses, He also explains that man is an agent, a knower, a
producer, a creator and an effecter. He then shows in all other verses
that all good and evil is from God, exalted. Yet He does not explain
how these two contentions may be reconciled; rather, He made general (#al sabl al-ijml) belief in both obligatory. He also explains that
nothing whatsoever escapes the decree of God, exalted, and His will,
judgement and determination. He then explains that [God] neither
wills nor performs wrongdoing, vainness or futility.
Therefore, the method (arqa) of our Prophet in his invitation (da#wa)
is to proclaim Gods greatness in all conceived respects, and to prohibit
wading into explaining whether these respects conflict or not. For if we
say, Bad acts are of the agency of men, and are created by God, we
proclaim His greatness with respect to power, but we do not proclaim
His greatness with respect to wisdom. And if we say, They are not
from God, we proclaim His greatness with respect to wisdom, but
we do not proclaim His greatness with respect to power.
As for the Qur"n, it proclaims Gods greatness with respect to both
power and wisdom together. Concerning the former He says, Say,

158

Qur. 47:38.
Qur. 6:59.
160 Qur. 13:8.
161 Which is a major problem in classical kalm. Cf. H. Wolfson, The Philosophy
of the Kalam, 112 ff.
159

al-rzs perfectionist theory of virtue

147

all is from God,162 and concerning the latter He says, Whatever of

good befalls you it is from God, and whatever of ill befalls you it is
from thyself.163 He then prohibits people from wading into the investigation of this conflict and eliminating it. Rather, what is obligatory
upon common people (#awmm) is absolute belief in proclaiming Gods
greatness in both power and wisdom. So, in reality, what He states is
most correct. For a universal invitation (da#wa #mma) to God becomes
effective only by this method.164

Therefore, for prophecy to be universal and all-encompassing, and

to effectively transform as many people as possible, its approach
should have the following main characteristics. (a) It should avoid
explicating some subjects in unequivocal and formal theological
terms, which will go beyond the mental capabilities of most men,
who will find it highly problematic and repulsive. Presenting pure
deanthropomorphism to the public (jumhr) is ineffective; therefore,
it is necessary to appeal to a middle way between presenting anthropomorphism (tashbh) and presenting pure deanthropomorphism,
so that [the prophets] message be accepted by the public.165 (b)
Revelation should have a holistic approach to the cognitive aspects
of religious experience, by providing a comprehensive and simple
expression of the main divine attributes, including power and wisdom.
The believer should recognise all these attributes summarily, which
will lead him to glorify God in all these respects. (c) The prophetic
message should then prohibit common people from delving deeply
into these theological defiles and subtleties, which would only cause
confusion in their hearts and undermine their unquestioning acceptance of the message.
Having discussed the first main component of prophecy, which concerns belief, al-Rz then briefly explains the second main component
of prophecy, viz. soteriology, which concerns the actual furthering
of the theoretical and practical perfections of human beings.166 As
the majority of men need guidance (irshd) to their good, prophecy
will provide a detailed, tangible and practical path to the realisation
of human perfection, mainly through acts of worship, both external

162
163
164
165
166

Qur. 4:78.
Qur. 4:79.
Malib, 8, 1168.
Malib, 8, 1245.
Malib, 8, 119.

148

chapter three

and internal. It should also present these within the framework of

a law that makes the pursuit of perfection obligatory upon men.
According to al-Rz, part of the mission of the prophet is to make
belief in prophets and prophethood obligatory.
As regards the third component of prophecy, al-Rz does not
describe it in detail, but only mentions that it relates to eschatology.
In the Ma#lim, he contends that autonomous reason may only realise
that the afterlife is conceivable, and that the only way to know of it
and of what it involves with certainty is through revelation.167
In the case of the message of Prophet Muammad, al-Rz refers
to its approach as a whole, especially its cognitive concern, as the
way (arqa) of our Prophet, or the way of the Qur"n. He argues
that if one examines the efficacy of various religions in perfecting
humans, one will find the prophecy of Muammad superior to other
prophecies.168
In his commentary on Ibn Sns #Uyn al-ikma, written in the
same late period as the Malib, al-Rz underlines one of his main
inspirations for his theory of the perfectionist efficacy of prophecy. He
comments on Ibn Sns views on the definition of human perfection,
the division of philosophy into practical and theoretical parts, and
the division of practical philosophy into politics, economics (tadbr
al-manzil) and ethics. He writes:
The Shaykh says: The principles (mabda") for these three are received
from the divine Law (al-shar#a al-ilhiyya), and their ultimate developments are described by the divine Law. ...
Interpretation: ... The principles for these three sciences are received
from the divine Law; for the objective (maqd) of sending prophets
to men is to guide men to the correct mode and most advantageous
path of action. Since all possible forms of action are encompassed in
these three sciences, it should be said that the prophets, on whom be
peace, were only sent to indicate (ta#rf) the principles for these three
sciences and to delineate their perfections.169

Then, on theoretical philosophy and perfection, he writes:

The Shaykh says: The principles for these divisions of theoretical
philosophy are received from the establishers of the divine religion
(arbb al-milla al-ilhiyya) through intimation (tanbh), and they may be

167
168
169

Ma#lim, 112.
Malib, 8, 1212; cf. Ma#lim, 1034.
Shar #Uyn al-ikma, 2, 134.

al-rzs perfectionist theory of virtue

149

acquired in their perfection by the rational faculty through argumentation (ujja).

Interpretation: The Shaykh states that both the principles and the
ultimate developments of the practical sciences are received from
the establishers of revealed laws (arbb al-shar"i#). With regard to the
principles of these theoretical sciences, he mentions that they too are
received from the establishers of revealed laws, whereas their perfections and ultimate developments are established by the rational faculty
through argumentation.
He describes the difference between the two in the work that he entitles
the Aawiyya. He says: The lawgiver is obligated to invite [people]
to affirming the existence of God, exalted, His being transcendent
(munazzah) over imperfections and faults, and His being attributed by
the attributes of perfection and the aspects of majesty. As for God
not being spatial, and not existing in ___location or position, these are
not among what he has to explicate (tar). For such topics are among
those at which the wits of most men cannot arrive. Were he to invite
people to that, it would repulse them from accepting his invitation
(da#wa). Undoubtedly, therefore, it is necessary for him to be content
with that summary (mujmal) invitation. As for minute details, he should
not explicate them, but should delegate (tafw) them to the minds of
intelligent men. ...170

According to al-Rz, while revelation is the source for both the

general principles and the details of the practical sciences, and consequently for the means to realising practical perfection, it is the
source for only the initial stages of human theoretical perfection.
Revelation should present a minimal and concise (mujmal) creed for
men, which is necessary for their theoretical perfection, as well as
pointers and hints that turn their attention towards furthering
their theological knowledge, without providing a comprehensive
theology. Beyond these basic stages, theological knowledge may be
acquired primarily by extra-revelatory means.171
This stance contrasts sharply with the classical kalm approach to
theology. Previous mutakallimn viewed theological reflection (naar)
as a form of action, to which a ruling may apply, viz. obligatoriness
(wujb). Al-Rz describes this voluntarist view of theological knowledge and enquiry in the Muaal:
170 Shar #Uyn al-ikma, 2, 1921 (I made much use of Yahya Michots translation of this passage, A Mamlk Theologians Commentary on Avicennas Risla
Aawiyya, 154); cf. Ibn Sn, Aawiyya, 97103; 1103. On the theory of prophecy
presented in the Aawiyya, see Y. Michot, La Destine de lhomme, 3043.
171 Cf. Ibn Sn, Shif", Ilhiyyt, 2, 4423; M. Marmura, The Islamic Philosophers Conception of Islam, 989.

150

chapter three
The commonly-accepted explanation of the obligatoriness of rational
reflection is [as follows]. Knowing God, exalted, is obligatory, and can
only be attained through reflection. What is a basis for a categorical
obligation, and falls within the capacity of the obligated agent (mukallaf), is obligatory, as will be explained in the [science of] the principles
of jurisprudence.172

Having the correct creed is Legally obligatory, in the same way that
performing certain acts is obligatory. According to classical Ash#ars,
this obligation stems purely from divine command, whereas the
early al-Rz departs with this view in favour of a consequentialist
notion of Legal obligation. The latter line of argument is used in
the Muaal:
Problem: The obligatoriness of rational reflection is based on scripture, contra the
Mu#tazila and some Shfi# and \anaf fuqah". Our evidence is: [1] [Gods]
saying, We would not punish until we had sent a messenger.173 [2]
Since the basis of obligation is [afterlife] reward and punishment, and
since none of Gods acts can be [morally] bad, the mind alone will
be unable to make conclusive judgements with respect to [afterlife]
reward and punishment. Thus, it will not be possible to arrive [by
autonomous reason] at the obligatoriness [of reflection].174

In classical kalm and classical theology generally, theological knowledge is necessary in as much as it is incumbent upon men to know
certain things about God and His relation to His creation. He is
worshiped fully only when this knowledge (or belief) is combined to
correct action. This stance was the main motive for the compilation
of creeds (#aqda). As such, the main function of Revelation, as far as
believers are concerned, is to communicate knowledge of the correct
creed and practice to which humans ought to adhere. Revelation acts
as a primary source for all types of theological knowledge, except
what is required for proving the veracity of Revelation itself. The
problem that all theologians had to deal with was to harmonise what
seem to be conflicting theological assertions in revealed texts. Most
mutakallimn choose to interpret some statements metaphorically
(ta"wl). However, in his later works, al-Rz maintains that theological
statements in revealed texts provide only a minimal theology, since
their primary purpose is soteriological, rather than epistemic.

172
173
174

Muaal, 130.
Qur. 17:15; cf. Qur. 10:101, which al-Rz cites elsewhere.
Muaal, 134. Cf. al-f, Dar", 98.

al-rzs perfectionist theory of virtue

151

According to al-Rz, the perfectionist concern of religion, which is

its central and ultimate objective, is constant throughout the history
of revealed religion. Therefore, this dimension in each particular
prophecy is never abrogated by a subsequent one. He describes
this concern as being related to both the proclamation of the greatness (ta#m) of the nature and stature (amr) of God and compassion
(shafaqa) towards His creatures, which, elsewhere, he explains as
referring to the perfecting of the theoretical and practical faculties
of the soul, respectively.175 Apart from this essential concern, other
features of a revealed religion are contextual (wa#), and may be
abrogated by a subsequent revelation according to changing historical circumstances.176
Significantly, after presenting his later theory of prophecy, al-Rz
totally rejects a crucial part of Ibn Sns theory, viz. his psychological explanation of revelation .177

175 Malib, 7, 230; Tafsr, 12, 534; 20, 104; Lawmi#, 170. He often mentions
that ta#m and shafaqa are specified in a adth, which I have been unable to find.
Instead, #Abd al-Qdir al-Jayln (d. 561/1166) (Fat, 106) cites it as a non-Prophetic
aphorism: Good lies entirely in two words: al-ta#m li-amr Allh #azza wa-jalla,
wa-l-shafaqati #al khalqih. Al-Rz writes that all things should be viewed as Gods
creatures and as signs for His oneness and greatness, and should consequently be
given due respect. Living creatures, which experience benefit and harm, should
be treated with great compassion, e.g. through filial piety and enjoining good and
forbidding evil (Tafsr, 15, 24).
176 Malib, 8, 124. Cf. al-Mward, A#lm al-nubuwwa, 52. Al-Rz replies by
this to an argument against the possibility of prophecy, made from the same perfectionist notion of the human good that he uses to justify religion: assuming that
revealed religion aims to serve human perfection, one religions abrogation of a
previous one is unjustifiable. For the later religion will abrogate either the perfectionist aspects of the previous one, thereby contradicting its purpose, or inessential
details, in which case great harms will follow from frivolous differences (Malib,
8, 879). Al-Rz also accepts the traditional justification for abrogation from the
corruption of previous religions (Malib, 8, 1212; cf. Ma#lim, 1034).
177 Malib, 8, 12737. Thus, revelation is not a consequence of the psychological faculties of the prophet, but depends on his hearing Gods speech directly, or
receiving it via Gabriel, who acts as an intermediary (Malib, 7, 3899; 7, 421).
In contrast to Ibn Sns theory, whereby the imagery that appears in scriptural
texts stems partly from the prophets faculty of imagination in its interaction with
the Active Intellect and the heavenly souls, and partly from the prophets own
judgement and manipulation of the form and content of scripture, al-Rz considers both the form and content of scripture as completely divine. Gabriel receives
Qu"nic revelation in exactly the same form he communicates to the Prophet
(Tafsr, 27, 188). Cf. Tafsr, 2, 32: Gabriel heard Gods speech in the heavens, and
brought it down unto the Prophet. ... If it is said, How did Gabriel hear Gods
speech, when His speech, according to you, is of neither letters nor sounds? we

152

chapter three

Before concluding the present chapter, we should consider how

this perfectionism in al-Rzs later thought relates to his consequentialist ethics at the normative level. Already, in the beginning
of this chapter, we saw how he explains the relation between these
two teleological stances at the metaethical level.
Al-Rz does not develop a unified normative discipline that combines the different normative stances that he considers to be valid,
nor does he present comprehensive guidelines as to how these may
be applied consistently and systematically. The relation between ul
al-fiqh, which represents his normative ethics of action (including the
general consequentialist stance), and both akhlq and riya, which
represent his ethics of character, remains on the whole unclear. The
following is a brief interpretation of how the different normative
stances in al-Rzs later thought relate to each other.
It is first of all clear that (in sharp contrast with al-Frb) he
considers the perfectionist stance to pertain exclusively to the private
realm. Thus, the furthering of human perfection is the objective of
the disciplines of riya and akhlq, but cannot be an objective for the
art of politics. As we will see, this owes partly to the great pessimism
with which al-Rz regards mans socio-political condition.
It is also clear from the foregoing that he analyses prophecy from
a teleological standpoint: he considers Revelation to serve more than
one main function, including a perfectionist purpose. He outlines
a Sufi-inspired soteriology, in which the Law (shar#a) is the initial
stage, aimed at purifying the exteriors of people from what should
not be (m l yanbagh); then the spiritual path (arqa) is a more
advanced stage aimed at purifying souls from corrupt beliefs and
vile traits; the realisation of truth (aqqa) is the stage of theoretical
perfection in which truth manifests in the hearts of spiritual men;
and prophecy (nubuwwa) is a stage reserved for the elect few, in
which the soul becomes absolutely perfect such that it may perfect
those imperfect.178
say, it could be that God created, for him, a hearing for His speech and provided
him with an ability to express that eternal speech. Or it could be that God created, in the Preserved Tablet, a writing in the exact composition [of the Qur"n],
which Gabriel then read and memorised. Or it could be that God created distinct
sounds, in this exact composition, which Gabriel then received, and that He then
created for him necessary knowledge that this is the expression of the meaning of
that eternal speech.
178 Tafsr, 17, 117.

al-rzs perfectionist theory of virtue

153

Yet, alongside serving the other-worldly and spiritual end of man

(dn), Revealed Law also serves matters of this-worldly concern (duny),
which relate to the advantage and wellbeing of both the individual
and the external world, especially society. Alongside Revealed Law,
the more general normative consequentialist ethics (especially as
embodied in the principle of unrestricted interest) will also apply to
both these worldly, mundane spheres: the private and the public.
Table. How the main normative approaches (columns) apply to the private and
public spheres.
Consequentialism (mundane,
physical wellbeing)

Private sphere

Public sphere

179
180

Revelation: serves the

wellbeing of the individual
in this world and the
hereafter.
Extra-revelatory
consequentialist ethics:
serves this-worldly private
wellbeing.

Revelation: serves worldly

public wellbeing.179
Utilitarian politics: serves
public stability and
wellbeing.180

See p. 177 infra.

See p. 1767 ff. infra.

Perfectionism (supra-mundane,
spiritual wellbeing)
Revelation (shar#a, Qur"n).
Practical ethics (akhlq).
Rational reflection (naar).
Spiritual discipline (riya).
All serve to perfect the
individual both theoretically and
practically.

154

chapter three

a commentary on rislat dhamm al-ladhdht

155

CHAPTER FOUR

AL-R$Z^S LATER PESSIMISM:

COMMENTARY ON RIS$LAT DHAMM AL-LADHDH$T
The science of character (#ilm al-akhlq) concerns both the human
souls attribute of power and the sensory appetites of eating and
coition: aspects of the souls association with the physical world,
including the physical body and the external world. On the other
hand, the souls attribute of knowledge is perfectible by the theological sciences and spiritual discipline.
In his main book on practical ethics, Kitb al-Nafs wa-l-r, al-Rz
discusses two types of moral therapy: (a) contemplative therapy,
which involves descriptions of human vices and reasons for why
they ought to be replaced with virtues, as well as prescriptions of
a contemplative, intellectual nature; and (b) practical therapy, consisting of practical prescriptions. This stance represents a form of
moral optimism: morality is based on a rational science, and moral
perfection is humanly achievable. Al-Rz, of course, was likewise
an epistemological optimist, as he maintained that metaphysical
knowledge is humanly attainable through the exercise of rational
reflection.
However, in a later ethical work, Rislat Dhamm ladhdht al-duny
(Censure of the Pleasures of This World), written in 604/1208 towards
the end of al-Rzs life, a very different stance emerges. This work
(which is brought to attention and published for the first time in
the present volume) consists of three sections, relating to pleasures
experienced in the main faculties of perception in man: (a) sensory
(iss) pleasure, corresponding mainly to the physical appetites of
eating and coition; (b) imaginative (khayl) pleasure, corresponding
to the pursuit of rule (ri"sa) and social status ( jh); and (c) intellectual
(#aql) pleasure, corresponding to intellectual pursuit. Normally, the
first two are the chief ___domain of the science of character; the last is
served ultimately by theological and philosophical learning. However, in Dhamm al-ladhdht, al-Rzs stance in this regard is radically
different. For not only does he offer no prescriptive moral therapies
with respect to the first two forms of human activity, he argues that

156

chapter four

they present utterly irresolvable moral dilemmas and that intellectual

pursuit inevitably culminates in stalemate. This work underscores a
pronounced moral and epistemological pessimism in the later stage
of his career. As we will see towards the end of the present chapter,
al-Rz appears to propose an alternative soteriology, which emphasises spiritual discipline and guidance through the Qur"n.
Dhamm al-ladhdht is, first and foremost, a moral, rather than a
theoretical, text written in response to a question put to the author
by someone who is unknown to us. This character is reflected in the
style and the forms of argument used, which often appear rhetorical.1
Nonetheless, as we will see, the integrity of the main contentions in
the work is not compromised. In what follows, we will examine the
views presented in this text, as well as their fuller and wider theoretical background in al-Rzs thought generally.

The Nature and Extent of Sensory Pleasure and Pain

Al-Rz starts the section on sensory pleasures in Dhamm al-ladhdht
by recalling the premise that human action is produced by either of
two primary motives: the averting of pain and the seeking of pleasure
(no mention of perfection is made here). However, he adds, pleasure
is in fact marginal in the overall human sensory experience, since
most of the time man is preoccupied with averting pains. Though
garments and houses protect humans from a variety of harms, they
form little more than lesser evils, since in themselves they are sources
for pain and discomfort (e.g. by the garments weight), rather than
pleasure or benefit. They are akin to foul-tasting medicines taken
to remedy illnesses.
Of all human sensory activity, only eating and coition seem to
produce real pleasure. Al-Rz thus makes these two bodily functions
the focus of this section, informing the reader from the start that
he aims to draw attention to the vileness, baseness, degradation,
and similitude to ignoble beasts that they involve. He writes that
the justification that poets and orators advance for this assessment
may be summarised in three basic points: (a) these pleasures are

1 By contrast, al-Nafs wa-l-r (3) is a book on the science of character, written

in a demonstrative, apodictic method, rather than a rhetorical, persuasive style.

a commentary on rislat dhamm al-ladhdht

157

transient and (b) inextricably mixed with pains and detriments, and
(c) vile people have a greater share of them than virtuous people.
However, he notes, this reasoning presumes that though they ought
to be avoided for these accidental features, these pleasures are essentially good.2
By contrast, the falsifa argue that sensory pleasures are inherently
contrary to the human good and human happiness. In this regard,
two distinct propositions are advanced: (a) that these pleasures are
real, but essentially base; and (b) that they are not real, but are mere
relief from pain.
As mentioned, al-Rz considers the former proposition and arguments for it to be rhetorical.3 And, indeed, many of the arguments
he lists are akin to the three he attributes to poets and orators.4 For
example, he cites the inductive argument that observation shows
that all people hallow ascetics and despise indulgent pleasure-seekers,
which will indicate that human primordial nature (fira) attests to the
baseness of these pleasures.5 Other arguments emphasise that the
physical pains and ailments that result from eating and coition are
severe, numerous, long-lasting and often spread in more than one
organ in the human body. By contrast, the pleasure experienced in
eating decreases sharply after the first morsel or two, whereas coital
pleasure occurs only during orgasm, which is as though an instant
(al-n alladh l yanqasimu).6 And so forth; many of these arguments
are commonplace.7
In what follows, we shall focus on the latter, possibly demonstrative
proposition, that sensory pleasures, thought to be [real] pleasures,
are actually not pleasures, but little more than relief from pain.8
Al-Rz argues that the experience of sensory pleasure presupposes
the existence of prior needs (ja) and appetites (shahwa). Thus, if
one were to offer jewellery to a dog and a bone to a man, neither
would find pleasure in the object offered, since neither would have

Dhamm al-ladhdht, 6.
See p. 61 supra.
4 Cf. Nafs, 88107; Malib, 7, 297302.
5 Dhamm al-ladhdht, 11. This is an argument from widely-accepted beliefs, which
al-Rz will consider to be non-demonstrative, but rhetorical (cf. p. 88 ff. supra).
6 Dhamm al-ladhdht, 13.
7 Cf. p. 122 supra.
8 Dhamm al-ladhdht, 6.
3

158

chapter four

desire for it. Also, the greater the hunger, or the longer the period
of abstinence from coition, the greater the pleasure experienced at
gratificationwhich indicates that the extent of pleasure depends
on the extent of the preceding urge.9
Yet, other parts of the text seem non-committal with respect to
this view; e.g. the statement, These states are either not pleasures,
or, if they are pleasures, they are extremely base and most vile.10
It is not entirely clear in Dhamm al-ladhdht to what extent al-Rz is
committed to the view that pleasure is unreal. This ambiguity appears
to stem from the rhetorical style of this moral text, which was written
with a view to persuade the reader into shunning sensory pleasures,
rather than to analyse the nature of pleasure as such.
In any case, al-Rz concludes this section by stating the global,
quantitative assessment that pain is predominant in this world,
which is closely related to his qualitative view on the nature of
pleasure and pain. He writes: What prevail over the people of this
world are anguish, affliction and grief, whereas pleasure and good
are very rare.11
But does al-Rz truly adhere to such a pessimistic stance, or is it
merely a ploy that largely serves the rhetorical purpose of a moral
text? It will be instructive to explore his treatment of these themes,
viz. the nature and extent of pleasure and pain, in his philosophical
works, which lie at the background of Dhamm al-ladhdht.
The Nature of Pleasure and Pain
In the Mabith and the Mulakhkhatwo relatively early falsaf
worksal-Rz opens the section on pleasure and pain by quoting Ab Bakr al-Rzs definitions of pleasure as departure (khurj)
from an unnatural state (i.e. towards the natural state), and pain
as departure from a natural state.12 The only two possibilities,
therefore, are the natural state, which is neutral, and unnatural
states, which are comparatively bad. But there cannot be a state
superior to the natural state. From this, Ab Bakr al-Rz arrives at
the view that pleasure is nothing but relief from pain; and there is

9
10
11
12

Dhamm al-ladhdht, 67; cf. Tafsr, 18, 219.

Dhamm al-ladhdht, 11.
Dhamm al-ladhdht, 15.
Mabith, 1, 387. Cf. Lenn Goodman, Rzs Psychology, 31 ff.

a commentary on rislat dhamm al-ladhdht

159

not a pleasure that occurs but after a pain,13 a view that he also
attributes to Plato.14 Al-Rz reports:
Some deny the existence of pleasure, viz. Muammad Ibn Zakariyy
al-Rz. He claims that the reality of pleasure is nothing but the negation of pain. Thus, the reality of the pleasure found in eating is nothing
but the negation of the pain of hunger; and the reality of the pleasure
found in drinking is nothing but the negation of the pain of thirst. ...
Most falsifa and mutakallimn reject this claim.15

In the Mabith and the Mulakhkha, he rejects this view and cites
examples of pleasures that cannot be explained in terms of a return
to a natural state from a previous unnatural state, thus appearing to
be real. One such instance is the enjoyment of a pleasant sight that
one does not conceive of beforehand, which, he argues, contradicts
the claim that pleasure from sight is due to the negation of the pain
of longing.16 The change in the state, al-Rz argues, is in fact only
accidental to pleasure, rather than essential.
He then turns to Ibn Sns definitions of pleasure as perceiving what is agreeable (idrk al-mul"im), which is the realisation of
perfection in an aspect of the perceiver, and pain as perceiving
what is disagreeable (munf).17 However, is pleasure this type of
perception, or a product thereof (which, he notes, is suggested by
Ibn Sn in at least one place)?18 Al-Rz argues for the latter view:
We perceive, in the acts of eating, drinking and coition, a specific
state that is distinct from all other psychological states, the same
being true of the perception of pain.19 Therefore, as sensory perceptions (issiyyt),
... the essences of pleasure and pain are perceived as a priori, self-evident
conceptions (taawwur awwal badh). Indeed, conceiving them is among
the clearest, most evident and most incontrovertible cognitions. But

13

Ab Bakr al-Rz, extracts from Kitb al-ladhdha in Ras"il, 148.

Ab Bakr al-Rz, Shukk #al Jlns, 17. On the Platonic background to this
view, see M. Fakhry, Ethical Theories, 723.
15 Shar al-Ishrt, 2, 80; cf. Malib, 4, 417.
16 However, Ab Bakr al-Rz (Ras"il, 155) argues that looking at unattractive
faces produces boredom and thus a departure from the natural state. Seeing a
pleasant face afterwards will only relieve that pain.
17 Cf. Ibn Sn, Shif", Ilhiyyt, 2, 369; 2, 424.
18 Cf. Ibn Sn, Al-Adwiya al-qalbiyya, 2289.
19 Shar al-Ishrt, 2, 889.
14

160

chapter four
the purpose of definition and description is to define what is obscure
with what is clear.20

Just as statements of assertion (tadq) relating to ones sensory experiences do not require demonstration, the conceptions (taawwur) they
involve do not require definition. Hence, the essences of pleasure
and pain are known without definition.21
Every rational being, indeed every sense-perceptive being (asss), perceives [pleasure and pain] in oneself and, necessarily, differentiates
between both and between these and other things. What is so is indefinable. This shows that the view of those who maintain pleasure to be
the perception of what is agreeable, and pain the perception of what is
disagreeable, is false. For the conceptions pleasure and pain are more
evident than the conceptions agreeability and disagreeability.22

Al-Rz also cites medical evidence, which shows that though some
physical ailments are sensed by the individual, they do not produce
pain. This indicates that pain is not simply the perception of an
unnatural state, and that the latter is not even a sufficient condition for pain.
Objections to Neoplatonist Theodicy
Discussions of the nature of pleasure and pain are often pertinent
to more macro-level questions regarding cosmogony, the problem
of evil and the assessment of human existence as a whole. Perhaps
the most striking example of such a link can be seen in Ab Bakr
al-Rzs conclusion that evil is prevalent in this world, which follows from his notion that pleasure is purely relief from pain. This
conclusion is not expressed clearly in any of his surviving works,
but is reported in the following hostile account by Maimonides (d.
601/1204):
[He] has written a famous book, which he has entitled [the Ilhiyyt].
He filled it with the enormity of his ravings and his ignorant notions.
Among them there is a notion that he has thought up, namely, that
there is more evil than good in what exists; if you compare mans
wellbeing and his pleasures in the time span of his wellbeing with the
pains, the heavy sufferings, the infirmities, the paralytic afflictions, the

20
21
22

Shar al-Ishrt, 2, 89.

Mabith, 1, 3889.
Mulakhkha, fol. 161a.

a commentary on rislat dhamm al-ladhdht

161

wretchedness, the sorrows, and the calamities that befall him, you find
that his existencehe means the existence of manis a punishment
and a great evil inflicted upon him. He began to support this opinion
by inductively examining these misfortunes.23

Ab Bakr al-Rz advances a theodicy that justifies this state of

affairs, to which we return further below in this section. We shall
first examine al-Rzs chief concern, which is Neoplatonist theodicy,
particularly as advanced by Ibn Sn.24
The existence of evil in this world, which emanates from the First
Cause, which is absolutely good and whose effect should also be
good, constitutes a serious problem to Neoplatonist cosmogony. Ibn
Sn tries to diminish the reality and amount of evil in the physical
world (which he considers to be the best of all possible worlds) and
the cosmos as a whole: he contends that it only appears in the sublunar world and only affects individuals for limited periods rather
than whole species perpetually. He also distinguishes between what
is evil essentially and what is evil accidentally. The former type
consists of imperfections in entities (e.g. the absence of sight, life or
knowledge in a human being) and is hence fundamentally non-existent. Accidental evils include acts and events produced by qualities
that constitute perfections in their agents or causes (e.g. the quality
of burning in fire), and that accidentally cause harm to some other
beings, though on the whole their good outweighs their harm. If the
entity harmed has perception, it may perceive the harm it receives;
and this perception will be its experience of pain.
According to Ibn Sn, the qualities that may cause harm to some
entities accidentally are brought into being for the sake of their serving
the greater overall good: the cosmic order. Not to bring them into
being would constitute a greater evil than the evil they may cause

23 Maimonides, Guide, 3, 443 (Piness translation); cf. Ab Bakr al-Rz, Ras"il,

17980; al-^j, Mawqif, 5, 137; Ibn Taymiyya, Majm#, 6, 308. In a section on Ab
Bakr al-Rzs cosmogony, al-Rz quotes him stating that the world is full of evils
(Malib, 4, 409; 4, 413).
24 Mabith, 2, 51923; Shar al-Ishrt, 2, 78 ff. Cf. Ibn Sn, Shif", Ilhiyyt, 2,
3556; 2, 41422; Najt, 3206; Ishrt, 3, 299 ff.; Ikhwn al-af, Ras"il, 4, 108;
cf. Shams Inati, Ibn Sns Theodicy, 65 ff. Anyone who reads al-Rzs account of
Ibn Sns theodicy in the Mabith will conclude that he completely accepts it,
except that the last few lines hint at a different stance, and refer the reader to
earlier discussions of whether God is a necessitating cause or a voluntary agent (cf.
citation from the Mulakhkha, p. 1678 infra).

162

chapter four

accidentally. And while an absolutely good being exists (viz. God),

purely, or predominantly, evil beings are inconceivable. This appeals
to the Neoplatonic doctrine that existence is good, and non-existence
evil; as such, a thing cannot, at once, exist and be purely evil.
For Ibn Sn, therefore, evil may be abundant, yet it is unprevailing (aqall), rather than predominant (akthar), in the entirety of
being.25 Or, as al-Rz reports, evil is unprevailing (maghlb) and
good is predominant (ghlib).
In refuting this theodicy, al-Rz starts by objecting to the Neoplatonic association between good and existence, and between evil
and non-existence. He writes in the Mulakhkha:
On that the Existent is Good, and the Non-Existent Evil. This is a generallyaccepted (mashhra maqbla) premise, which I have found no one support
with a demonstration. Instead, they are only content with relying on
an example; viz. killing is not evil in so far as it is produced by the
agency of a killer, or by virtue of the weapon being injurious, or the
killed persons limb being susceptible to injury, but by virtue of the
departure of life from the body; therefore, only this negation is evil,
whereas all other, real aspects are good. You know that examples are
insufficient in establishing scientific premises.26

Al-Rz completely dismisses Ibn Sns ontological notion of good

and evil, which many others accept uncritically. What do good
and evil, he argues, mean in the propositions, Good is existence
and Evil is non-existence? If existence and non-existence are
predicates in these propositions (rather than simply explanatory of
what good and evil mean), then the conceptions of their subjects,
viz. good and evil, will be required. Yet, al-Rz argues, Ibn Sn
fails to define them.27 Also, in his commentary on #Uyn al-ikma, we
find the following argument that highlights a similar failing:
Minds and hearts become perplexed by the abundance of pain they
witness in this world. The justification that the falsifa advance for
this is that pleasure and benefit are more [widespread] than pain and
harm, and that this minor amount of pain occurs because it is impossible to realise those preponderant benefits except with these minor
harms. They say: Omitting a great good for the sake of a little evil
is a great evil.
[But this is] a proposition, in which the subject is omitting a great

25
26
27

Ibn Sn, Najt, 326. Cf. Y. Michot, La destine de lhomme, 634.

Mulakhkha, fol. 89a.
Shar al-Ishrt, 2, 7980.

a commentary on rislat dhamm al-ladhdht

163

good for the sake of a little evil, and the predicate is a great evil.
If, by a great evil, which is the predicate, [they] mean omitting a
great good, the predicate of the proposition will be identical to its
subjectwhich makes the statement meaningless. Or, if [they] mean,
by great evil, great pain, then this statement will mean Omitting a
great good for the sake of a little evil is a great pain. However, this
is obviously absurd; for had these things not existed, pain, pleasure,
evil and happiness would not have occurred.28

In the Mulakhkha, Ibn Sns views and arguments in this regard

are described as nonsense (mmt, lit. calamities).29
One may wonder, however, whether this refutation does not come
into conflict with al-Rzs own definition of good and bad in terms
of essential perfection and imperfection. This does not seem to be the
case; for, as we saw, he does accept a perfectionist definition of value,
but only one that takes the human perceiver as its primary reference
point, with respect to the perception of either a subjective aspect of
perfection, or an objective aspect of perfection in an external entity.
To say that the courage of a given epic hero is good only refers to
the perceivers love of, and attraction to, this quality as an aspect of
perfection, but not to this quality being somehow intrinsically good,
e.g. in an ontological sense, irrespective of any perceiver. Yet, for
al-Rz, even subjective perfection is not at issue in the problem of
evil. Instead, he appeals to his other notion of good and bad in terms
of the experiences of pleasure and pain. He writes:
In generally-accepted convention (al-#urf al-#mm al-mashhr), what people
mean by the expression evil (sharr) is pain and what leads to it. And it
is known immediately that pain is real (wujd)this is not disputed by
sensible men. Indeed, some even claim that pleasure is the mere relief
from pain; so they consider it unreal. This being the case, it becomes
evident that what people mean by the word evil is real, whereas good
may refer to the non-existence of pain, ... or to pleasure.30

By affirming what good and evil mean in ordinary language, alRz counters Ibn Sns ontological and cosmological theodicy with
the very human experience of suffering. This is what minds and
hearts become perplexed by, and consequently what any attempted
theodicy ought to address. Ibn Sns theodicy merely circumvents
the real problem of evil.
28
29
30

Shar #Uyn al-ikma, 2, 7981; cf. 3, 1312.

Mulakhkha, fol. 350b.
Shar al-Ishrt, 2, 80; cf. Mulakhkha, fol. 350a.

164

chapter four

Al-Rz, therefore, wonders why Ibn Sn attempts a theodicy in

the first place, given that he is not a moral realist and that he does
not consider the Creator to be a voluntary agent, in which case He
cannot be morally responsible for His acts.31 His introduction of the
terms good and evil into the context of ontology is superfluous
(ful) and inapt. Instead, he should have left the attempt to justify
evil in this world to those for whom it is a real problem, since they
adhere to these two doctrines, particularly the Mu#tazila.
Now, on pain, Ibn Sn writes, rather ambiguously, that though
pains and grief are real properties, rather than unreal (#adam), they
seem to be in the same situation as (tatba#u) non-existent things and
imperfections.32 He recognises that pain, as a psychological perception, is real. Yet it is a perception of something unreal, viz. imperfection, and should thus be treated effectively as another unreal evil.
Alternatively, he could have intended to say that, though real, pain
should in the final analysis be treated as a mere imperfection in
man. In either case, Ibn Sn makes a very cursory dismissal of pain,
and disregards the individuals experience of suffering. Instead, he
approaches theodicy with a purely ontological conception of evil.33
By contrast, al-Rz affirms the reality of pain, as a real, simple
and primary perception, which is recognised immediately and cannot be explained away. Having made a clear distinction between
the perception of pain and the perception of its objective cause (e.g.
injury in the body of the experient of pain), al-Rz highlights the
subjective experience of suffering, which is brushed aside in Ibn
Sns ontological analysis. Moreover, for al-Rz, pleasure and pain
define the most fundamental conceptions of good and evil. And it is
this subjective experience of pain, rather than the objective imperfection perceived, that is evil and that any theodicy should attempt
to justify.34
Al-Rz goes further than defining evil in terms of pain, to contend-

31

Shar al-Ishrt, 2, 789.

Ibn Sn, Najt, 323; Shif", Ilhiyyt, 2, 419.
33 Cf. Ibn Sn, Al-Ishra il fasd #ilm akm al-nujm, 10 ff.
34 This criticism became influential after al-Rz. Al-Shrz (d. 1050/1640)
(\ikma, 7,627) cites al-Dawwn (d. 908/1502) arguing that the existence of pain
contradicts the view that evil is non-existent. Al-Shrz defends Ibn Sns definition of pain as perceiving the disagreeable and argues that pain is nevertheless
unreal, since it is an instance of knowledge by presence (#ilm ur) with its object
being unreal.
32

a commentary on rislat dhamm al-ladhdht

165

ing that evil is indeed prevalent in this world. The view is presented
most definitively in Shar al-Ishrt, which al-Rz considers to contain
his most comprehensive refutation of Neoplatonic theodicy.35 In
comparison to his earlier works, the Mabith and the Mulakhkha,
he here seems more favourable to Ab Bakr al-Rzs denial of the
reality of pleasure, though he does not accept it. He writes that
even if it is conceded that pleasure can be real, observation shows
that most pleasures are unreal and that real pleasures are extremely
rare, whereas all pains are real. As does Ab Bakr al-Rz before
him, al-Rz argues from particular empirical examples from daily
human experience.36 Health is merely the absence of physical ailments and pain, and is, therefore, not pleasurable. Eating, drinking,
coition, wearing garments, dwelling in houses, and experiencing
a cooling breeze on a warm day, are not pleasurable, but merely
alleviate the pains of hunger, thirst, coital desire, cold and heat,
respectively. He adds:
What is prevalent over the conditions of men is either pain or the
relief from pain, whereas the occurring pleasure, which is an existent
phenomenon (kayfiyya wujdiyya) other than the negation of pain, is
extremely rare, as if a drop in an ocean, such as finding pleasure in a
sight, sound, etc. that is unknown before being perceived. As for pains,
they are untold, though some are extremely severe, such as illnesses,
while others are mild, which are things that one is preoccupied with
most of his time, such as afflictions, anxieties, fear, dread, shyness, anger,
pains that occur at hunger, thirst, working in trades and professions,
smelling pungent smells, meeting imbeciles, seeing odious sights, the
harms caused by flies, flees, lice and ants, as well as all other things
that are too numerous to recount.
If it is proven that pains are preponderant (ghlib) and that real
pleasures are unprevailing (maghlb), and if those few pleasures were
the purpose of creation and existentiation, then, given the Creators
knowledge of the occurrence of numerous pains, ... the prevalent evil
would be for the sake of an unprevailing good. It becomes evident
that what [the falsifa] claim, viz. that good is preponderant in this
world, is false. Let it not be said that though disease is widespread,
health is prevalent. For we will say that we have shown that health
is a non-existent state (la #adamiyya), which is mans being free from

35

Therefore, he omits the discussion in the Muaal (478) and simply refers
the reader to Shar al-Ishrt.
36 Cf. Maimonides citation, p. 1601 supra. On this kind of argument against
optimism, cf. Eric Ormsby, Theodicy, 9 ff.

166

chapter four
both pleasure and pain. ... Therefore, remaining in non-existence is
preferable (al-baq" #al l-#adam awl).37

In the Ishrt, Ibn Sn also considers evil in human nature, having

explained evil generally. Why are humans predominantly ignorant
and under the sway of desire and irascibility? He argues that humans
fall into three main groups: (a) those who are advanced theoretically
and morally, and will experience great happiness in the hereafter; (b) those who are not advanced in these respects, but are in a
mediocre state, and who will have a neutral outcome (salma) where
they may experience some happiness in the hereafter; and (c) those
who possess evil traits and will experience misery in the hereafter.
While the two extremes are rare, most humans fall into the second
group.38 If we consider the first two groups, who will be saved (najt)
from torment, we will find that they constitute the overwhelming
majority of people. Therefore, for Ibn Sn, evil in human nature,
too, is unprevailing.
However, al-Rz argues that, as with physical health, the neutral
state denotes the absence of both pain and pleasure, and, as such,
is not good in itself and cannot justify creation. Yet, he continues,
even if this justification is conceded, Ibn Sns quantitative premise cannot be. For while it is the case that most human souls are
devoid of both true and false opinions, most are not morally neutral,
but characterised with a wide range of moral vices, such as desire,
irascibility, and attachment to wealth, material things and status.

37 Shar al-Ishrt, 2 801; cf. Arba#n, 2945. This view is echoed in various
places in al-Rzs writings; e.g. Tafsr, 19, 71 (meter: awl):

I behold the scenes of this perishing world,

Intertwined with dreads and sorrows;
Its goods, as if frightening dreams,
Its evils, on men, perpetual and thorough.
Cf. Tafsr, 17, 118:
Physical pleasures do not last. So the greater the pleasure found in them, the
greater and more severe the sorrows that result from the fear of their negation.
Al-Ma#arr [Saq al-zand, 51 (meter: khaff)] thus said: Indeed, grief at the hour
of death is times as much as happiness at the hour of birth. It is well known
that the happiness that occurs at the time of the birth of a child is not equivalent
to the grief that occurs at its death.
38

Ibn Sn, Ishrt, 3, 3068.

a commentary on rislat dhamm al-ladhdht

167

Indeed, both the virtuous and the morally neutral, combined, are
as a drop in the ocean.39
In his wider assessment of the human condition and his criticism
of Neoplatonic theodicy, al-Rz only asserts that most pleasures are
unreal, specifically in relation to worldly pleasures. Yet he avoids Ab
Bakr al-Rzs claim that pleasure as such is unreal, which will exclude
even the possibility of spiritual pleasure.40 For this reason, he does
not arrive at global pessimism. At the end of the section on theodicy
in Shar al-Ishrt, he concludes on a highly personal note:
As for me, though God has bestowed wellbeing (salma) upon me in most
conditions, so that I am as though distinct in this from most [people],
if I go back to myself and compare real (aqq) pleasures (i.e. excluding
unreal (#adam) ones) to pains, both external and internal, as we have
listed, we will find pleasures so wretched in comparison to pains. This
being the case, how could such pleasures, in relation to these pains,
be desirable! Were it not for the great afterlife pleasures that we long
for, it would have been best to have remained non-existent.41

The possibility of spiritual happiness is the only exit from despair.

Yet this can only be a solution to a very human problem, rather
than a premise for a theodicy. Al-Rz would argue that the possibility of spiritual pleasure still does not make creation rationally
justifiable. For though God is capable of creating rational souls in
a state of complete perfection and happiness, without attachment
to matter, He chose to create human beings in this physical world
with all its misery.
Al-Rz summaries his stance in the passage that concludes the
Mulakhkha, as follows:
With respect to the elements [i.e. the sublunar world], we do not
accept that the good is prevalent in them. For the good, according
to [our] present convention, is pleasure; and we do not accept that
what prevails in the world of generation and corruption is pleasure.
Indeed, there are some who even deny [the reality of] pleasure per
se, and claim that it has no reality (ma#n) other than the negation of
pain. According to this view, no state would exist but either pain or
relief from it. Pain is not good; and relief from it is unreal, so it too
is not good.

39
40
41

Shar al-Ishrt, 2, 823.

This is implied in Nafs, 97.
Shar al-Ishrt, 2, 812.

168

chapter four
Some then contrived (itla) to find cases where pleasure was affirmed
without involving relief from pain. However, assuming such cases are
valid (though their validity is in fact questionable), they are rare.
As such, it does not become true that good prevails. Rather, pain
and the negation thereof prevail, whereas pleasure is rare. Thus, their
argument42 will backfire, since evil is prevalent (ghlib); or, if not prevalent, it is on a par [with good]. This being the case, to initiate creation
would be either foolish (safah) or futile (#abath).
[As for Ibn Sn]s claim that separating good from evil is not possible, we say that this presupposes the denial of divine choice. Otherwise, [God] is capable of creating an object such that it is hot when it
is needed for a benefit, and not hot when it becomes harmful.43 Know
that deliverance from these confusions is possible only by upholding
one of two claims: [a] that God is a necessitating cause, in which case
this whole dispute will be nullified; or [b] that God is a voluntary agent,
alongside the denial of ethical realism (al-usn wa-l-qub), and that He
cannot be questioned for His acts,44 which is the clear truth.45

If one were to judge God on account of the magnitude of creaturely

suffering in this world, one could only conclude that bringing this
world into being was a sorely cynical undertaking. Al-Rz maintains
that the only stance that will not face this difficulty will be one in
which Gods free choice is affirmed and moral realism is rejected.
God creates what He wishes, and His acts cannot be judged morally. As such, any attempt to advance a working theodicy will be
futile.
For this reason, al-Rz also rejects Ab Bakr al-Rzs theodicy,
in which he attempts to relieve God, who is absolutely good, of
responsibility for this miserable world where creatures only suffer.
Ab Bakr al-Rz argues that if this physical world were created by
God, why would He censure it in revealed religions and warn people
from becoming attached to it? Rather, the coming of this world
into being is the responsibility of the Soul, which, out of its sheer

42 I.e. the Neoplatonic argument that since the world is predominantly good,
not to create it would have been bad. God thus creates the world necessarily, since
He is absolutely good.
43 He explains this more clearly in the Mabith (2, 523): Burning that occurs
after fire is not necessitated by fire. Rather, God chooses to create it immediately
after (#aqba) contact with fire. If the occurrence of burning after contact with fire
is by the choice and will of God, exalted, then He can choose to create burning
when it is good, and not to create it when it is bad.
44 Qur. 21:23.
45 Mulakhkha, fol. 350b51a.

a commentary on rislat dhamm al-ladhdht

169

ignorance, desired to attach itself to matter.46 Ever since then, the

living manifestations of the Soul in this world suffered, and they will
continue to suffer until the Souls union with matter finally ceases.
Once the Soul is completely liberated from matter, the world will
disappear and evil will be no more. Yet, having witnessed the result of
this union, God ameliorates the physical world and inserts order and
various comforts into it, in as much as possible. He also introduces
reason, which serves to emancipate the Soul from matter, gradually
as it advances through metempsychosis.47
For al-Rz, the answer to this view is straightforward:
The argument for this doctrine revolves around [the view that as
God is] merciful and beneficent, He ought not to commit an act
that would lead to pain and harm. However, if we contend that
moral rationalism is false, and that He does whatever He pleases,
and decrees in [whichever manner] He wills, this position will be
completely invalidated.48

There is neither a need nor a basis for theodicy, even if one were
to accept such an extreme pessimism as advanced by Ab Bakr alRz. Gods acts need not, and cannot, be justified morally.
Socio-Political Pessimism
It transpires from the foregoing that the view presented in the first
section of Dhamm al-ladhdht concerning the nature of pleasure and
the prevalence of suffering in this world represents a genuine stance
to which al-Rz is committed, especially in his later philosophical
writings. Yet, as a moral text, Dhamm al-ladhdht employs a rhetorical
form of discourse and pays less attention to rigour, exactitude and
the overall perspective. The same is true of the second section of this
work, which deals with what al-Rz terms imaginative pleasures,
i.e. pleasures experienced within the faculty of imagination in connection to the individuals involvement in the external world.
Much of the interest of this second section of Dhamm al-ladhdht
46

For Ab Bakr al-Rz, Soul is one of the five eternals, the others being
God, matter, space and time. Cf. his Al-Qawl f l-qudam" al-khamsa in his Ras"il,
191216.
47 See Malib, 4, 40119. Cf. Marwan Rashed, Ab Bakr Al-Rz et le Kalm,
416.
48 Malib, 4, 419.

170

chapter four

lies in its being one of the few places in which we find glimpses of
al-Rzs socio-political thought. Again, what we find in this section
is not a complete, or even rudimentary, socio-political theory, but
a moral treatment of aspects of the human socio-political condition,
focusing especially on the subjective experience, which is nonetheless
a crucial part of al-Rzs socio-political thought.
The Original State of Human Social Nature
Introducing this second section of Dhamm al-ladhdht, al-Rz states
that his purpose is to show (a) that the pleasures of rule and high
status (ri"sa wa-jh) can only be accompanied with great hardship
and severe pains, and (b) that these pleasures are in themselves
vile and unworthy.49 For these contentions, twenty arguments are
advanced.
The first argument is most instructive. Al-Rz writes:
Every man desires that he becomes the ruler over others, and that all
else falls under his domination, control and power. For mans being
dominant over the other and in effective control over him is an attribute of perfection; and attributes of perfection are desired for their
own sake. Ones being dominated by another and under his control is
an attribute of imperfection; and attributes of imperfection are hated
in themselves. Therefore, the natural disposition (ab#) of each person
drives him to be the ruler over the other and in control over him,
and to disallow the other from becoming a ruler over, and controller
of, him.
Therefore, the seeker of the attainment of rule for a particular man
is that particular man alone, whereas all others seek to undermine that
rule and to obliterate it. That single man is thus the only seeker of
the attainment of that rule for himself. As for all other people, from
east to west, they all seek to undermine it, negate it, and obliterate it.
Hence, on the one hand, those who seek to realise this aim cannot be
fewer in number, for there cannot be fewer than one, while, on the
other hand, those who seek to undermine and negate it are very great
in number, for it is evident that everyone other than that one will seek
to negate that rule and to undermine that persons supremacy.50

The same argument is summarised in the Tafsr.51 As mentioned,

al-Rz holds that human nature seeks to acquire attributes of per-

49
50
51

Dhamm al-ladhdht, 16.

Dhamm al-ladhdht, 167; cf. Nafs, 143.
Tafsr, 18, 220.

a commentary on rislat dhamm al-ladhdht

171

fection for their own sake and to avoid attributes of imperfection in

themselves. The perfection of the individuals attribute of power is
realised by furthering his domination over external entities, especially
other human beings, and by minimising the control and influence
of others over him.52
Inevitably, the self-regarding motives of different human beings
will come into conflict, as each attempts to advance his own interests
at the expense of others. Moreover, each person will have his own
character traits (khuluq), and his own dispositions (ab#); what is good
for this person is bad for that person, and vice versa. This could only
lead to enmity and discord.53 Al-Rz argues that, fundamentally,
every man will be involved in a ceaseless struggle of one against all,
and all against one.
Yet this view of human nature will face the same empirical objection that Mu#tazil moral realists advance against his psychological
egoism: viz. observation shows that people do in fact assist others
without expecting any benefit for themselves in consequence.54 Thus,
multitudes of people may support a single individual in attaining
leadership for himself, sacrificing their own lives and wealth in the
course.55 Al-Rz replies that this happens only when two conditions
are satisfied: (a) that one be hopeless of attaining this leadership for
himself; and (b) that one uses such assistance as means to attaining
some benefit for himself, which otherwise would not have been possible. Whenever any of those allies finds it opportune to turn against
the leader to seize power for himself, he will be motivated to do so
without delay.
By the same principle, the greater ones efforts to dominate over
others, or the greater his domination over others, the greater will
be the risk of detriment that he will face.56
The more pleasurable a thing, the greater the desire to acquire it and
to eliminate obstacles from it. When others attain leadership, that will
be one of the greatest obstacles from my attainment thereof. So the
desire to eliminate that obstacle will be the greatest of desires. Therefore,

52
53
54
55
56

Cf. Nafs, 223; 12933.

Dhamm al-ladhdht, 26.
Cf. p. 76 ff. supra.
Dhamm al-ladhdht, 17.
Dhamm al-ladhdht, 20.

172

chapter four
whoever aspires to attain leadership will invite people to kill him and
will increase their interest in destroying and eliminating him.57

With respect to any given attribute, every individual will be superior, equal, or inferior to others. Superiority to others in a given
attribute will motivate them to undermine it as much as possible,
either by attempting to eliminate it if it is a destructible attribute,
such as power, or by attempting to obscure it if it is an indestructible
attribute, such as knowledge. Equivalence to others in a particular
attribute will lead to a similar outcome, since uniqueness (wadniyya;
tafarrud) is an attribute of perfection, while sharing an attribute with
others is an attribute of imperfection. Finally, though by inferiority,
one escapes the active and determined enmity of others, one will
become treated despicably, as any other vile and lowly thing. This
type of relation is not only symptomatic of human interaction, but
is a universal principle:
Physicians even say that whenever a particular organ becomes weak,
stronger organs will send all [their] refuse to it. On the whole, the
domination of the strong over the weak is of the concomitants of being
(min lawzim al-wujd). So it becomes evident that the state of man always
falls in one of these three categories. ... Therefore, this corporeal life
is never dissociated from sorrow, grief and heartbreak.58

Rarely does the psychological egoism at the centre of al-Rzs ethics

receive such pronounced expression. He describes human nature as
being motivated almost uncontrollably by the endless drive for power,
domination, uniqueness and the elimination of competitors. Those
who do experience the pleasures associated with power and rule will
become desirous of experiencing more; the more they experience,
the greater and more urgent will the desire become.59 Moreover,
since man finds pleasure only during a change in his state, but not
afterwards as the new state settles, he will continue to seek more
and will never feel contented in any state, even if he were to possess the treasures of the heavens and the earth.60
Given this view of human social nature, al-Rz concludes that
the great harms that result from the human social condition and
the brutality of the ceaseless struggle and competition will guarantee
57
58
59
60

Dhamm
Dhamm
Dhamm
Dhamm

al-ladhdht,
al-ladhdht,
al-ladhdht,
al-ladhdht,

20.
21.
189.
23.

a commentary on rislat dhamm al-ladhdht

173

all, both ruler and ruled, oppressor and oppressed, great pains and
detriments. Human social existence, as one facet of mans wretched
worldly existence, is inescapably a life of ceaseless strife, tension
and merciless struggle to prevail, evade detriment and survive. The
greater the struggle, the greater the risks of detriment and death,
and the more violent the likely outcome.
The following passage, which appears in the Tafsr as a commentary
on the greeting Peace be upon you (al-salmu #alaykum), presents a
vivid depiction of the original state of human nature.
There is no doubt that this world is a hotbed for evils, detriments,
calamities and terrors. Scholars who investigate the obscure aspects of
human character (akhlq) differ on whether the origin of the nature of
the animal (al-al f jibillat al-ayawn) is good or evil. Some say that its
origin is evil, and that all human individuals are as though in perfect
agreement on this.
We may complement this by adding: Indeed, even all animals are
as though in perfect agreement on this! The evidence to this is that
whenever a man sees another man whom he does not know running
towards him, he will be driven by his nature to take guard against him
and to prepare to repulse him. Had his natural disposition (ab#) not
testified that the original nature of man is evil, primordial mind (firat
al-#aql) would not have necessitated the preparation to repulse the evil
of that comer. They even say that this feature appears in all animals;
for every animal, which finds another animal running towards it, will
escape and take guard against it. So, had it been evident in its natural
disposition that the original nature of this comer is good, it would
have necessitated it to stay where it was; for the original disposition
of [animal] nature motivates it towards desiring the attainment of the
good. ... So we know that the original state in the animal is evil.
Therefore, we say: Averting evil is more important (ahamm) than
attaining good. This is indicated by a number of aspects. First, averting evil involves preserving the original state, which is more important
than attaining more. Second, delivering the good to everybody is not
within capacity, whereas abstaining from harming anybody is within
capacity, since the former is action, whereas the latter is abstention. ...
Third, if evil is not averted, evil will occur; and this will result in the
occurrence of pain and grief ...; but if goodness is not attained, man
will remain in neither good nor bad, but in original wellbeing. ...
So it becomes evident [1] that the averting of evil is more important
than the attainment of good; [2] that this world is the hotbed of evils,
detriments, calamities and afflictions; and [3] that the animal, in the
origin of its created nature and the necessity of primordial disposition,
is a source for evil. Therefore, when a man reaches another man, the
most important consideration will be to inform him that, in relation to
himself, he is in a state of peace, security and safety. So, it has become

174

chapter four
conventional (iil) that [the comer] should initiate the greeting, which
is for him to say, Peace be upon you!61

From the foregoing, al-Rz seems to conclude that good and just
government is virtually impossible, and that political activity involves
great unavoidable detriment to both subjects and political agents.
The ruler will often find himself in great dilemmas, forced to choose
between two or more equally evil options.62 For instance, he may
act either humbly and benevolently, or despotically and brutally,
towards his close associates: either way, he will live in constant fear.
Humility and benevolence are bound to weaken his power, whereas
brutality will motivate others to kill him, or to eliminate his rule.
If it is said that balance (tawassu) is required, then, al-Rz argues,
the nature of this balance is unknown.
Also, just rule should involve appointing qualified individuals in
positions of power. Yet, al-Rz argues, such rule is an ideal that
either never occurred, or is extremely rare. Unjust rule, on the other
hand, will bring upon the ruler the enmity and condemnation of all
sensible and pious men.63 A just and widely satisfactory distribution
of wealth also appears to be an unsolvable dilemma.64
Some of the arguments put forth in Dhamm al-ladhdht seem to
conclude that the individual has many good reasons to minimise his
social association (mukhlaa) and political involvement, or even to
abandon it altogether (farra; i#tazala). However, man is political by
nature (madan bi-l-ab#). For if one were to live in reclusion (tafarrada)
in a desert, away from the material, moral and intellectual benefits
of urban life, he would lose his human attributes and descend into
bestiality. Thus, whether or not one becomes socially involved, one
is bound to suffer great harms.65 Indeed, even the most rudimentary
forms of social association will involve unsolvable dilemmas: Man
may live in this world with or without spouse and offspring: each
option will be a cause for suffering detriments and calamities!66
Family life will cause suffering to the individual in various ways,
including the constant obligation upon him to manage the familys

61
62
63
64
65
66

Tafsr, 16, 1823.

Dhamm al-ladhdht,
Dhamm al-ladhdht,
Dhamm al-ladhdht,
Dhamm al-ladhdht,
Dhamm al-ladhdht,

245.
29.
2930.
256.
27.

a commentary on rislat dhamm al-ladhdht

175

affairs and to secure its livelihood, and having malevolent members

in the family. However, by choosing to live without a family one will
either maintain all his needs by himself, which will involve severe
hardship, or will resort to relying on strangers, an option that will
lead to other sorts of harm.
Man is bound to suffer, whether as ruler or ruled. Man suffers if
he acts, and suffers if he attempts to refrain from acting altogether,
choosing to remain idle (mu#aal). A most baffling fact that invites
contemplation, al-Rz notes, is that, despite this inescapable suffering,
which is concomitant not only to the human condition, but to animals
in general, every animal will be extremely fearful of death, and will
always struggle to avoid or postpone it as much as possible.67
Some of the themes of the second section of Dhamm al-ladhdht are
akin to those presented in a section on contemplative therapies for
desiring status in Kitb al-Nafs wa-l-r.68 These therapies are aimed at
minimising ones desire for, and seeking of, higher social status, and
attempt to solve some of the dilemmas of the social aspect of human
existence. In a way, this section of Dhamm al-ladhdht recommends
the minimisation, or even the total abandonment, of socio-political
pursuit and ambition. At the end of one argument, al-Rz concludes
that the sensible person ought not to pursue leadership.69 In some
places, he concludes that both the pursuit of leadership and actually
having it are highly detrimental and unrecommended.
Yet the text goes further. It contends that it will be impossible
to reach an agreeable state for the individuals social being, or any
adequate solutions to its many moral dilemmas; even attempting
to abandon it, if at all possible, will be highly detrimental. There
is no way to avert such detriment. Man will never become free
from grief, anxieties and sorrows,70 and will never manage to
avert harm.71 One ought to contemplate the reality of mans social
condition to discover its sheer hopelessness and to find oneself utterly
alienated and estranged.

67
68
69
70
71

Dhamm al-ladhdht,
Nafs, 1416; cf. p.
Dhamm al-ladhdht,
Dhamm al-ladhdht,
Dhamm al-ladhdht,

28.
124 supra.
19.
22.
25.

176

chapter four

Social Contract
Had the purpose of the second section of Dhamm al-ladhdht been to
present a comprehensive account of socio-political motivation, we
would have judged the conclusion (that the agents socio-political
involvement is based completely on a calculation that takes into
account only his immediate interests) as fairly crude. What it presents,
rather, is a view of human social nature at its most basic and indications as to how this view may be developed into the foundation for a
full-fledged socio-political theory. Thus, the argument that individuals
may cooperate for a non-personal cause each out of sheer self-interest may become a ground for a more sophisticated utilitarianism. It
will do justice to al-Rz to append the above examination of his
notion of human nature with an outline of what elements could be
found in his kalm works of a fuller socio-political theory.72
Calculations of self-interest may yield more complex conclusions,
whereby the interests of the social agent are often not gained in any
immediate way. We saw above that al-Rz presents a theory of
consent in order to explain both moral norms and the sense of moral
obligation in human beings.73 The sheer brutality and insecurity
of the original, unchecked state of human association, as described
in Dhamm al-ladhdht, will drive individuals to the conclusion that
adhering to a set of norms will define the framework to realising a
state of security to each individual. Their primary motive will be
prudential, since each individuals life, freedom, wellbeing, property
and interests will be at risk from being undermined by others, which
is a risk that even the opportunity to gain at the expense of others
will not make worthwhile. According to al-Rz, reason is generally
more inclined to prudence than to the pursuit of pleasure.
People thus consent to a normative convention, the conventionality of which is then forgotten by time, as it transforms into objectified moral truth and as people start treating these norms as a priori
truths. In such a convention, each individual will accept, e.g., that
wrongdoing as such is universally bad, whether done against him
by others or against others by himself, and that he ought, therefore,

72 See also: Ann Lambton, State and Government, 1307, chapter on Fakhr al-Dn
Rz: The Dissociation of Religious and Temporal Power, which examines his
views in Jmi# al-#ulm and relates them to his historical context.
73 See p. 80 ff. supra.

a commentary on rislat dhamm al-ladhdht

177

to restrain his natural motive to limitlessly seek his own self-interest. Al-Rz indicates that every individual will realise that for such
social order to endure effectively, all or most individuals, including
himself, should adhere to such a convention.
However, on their own, such conventions cannot bring social
order about. The institution and maintenance of order will require
the combination of a legal code, based on a moral convention, with
a recognised sovereign political power that will enforce it on people.
Al-Rz writes in the Ma#lim:
Installing an imm guarantees the prevention of harm that can only
be prevented through it: therefore, it is obligatory (wjib).
The first [premise] stems from our immediate (arr) knowledge
that if a mighty and potent (bi) leader appears in a country, the
condition of that country will become closer to good order (al) than
if this leader did not exist.
As regards the second [premise], since the prevention of harm from
the self (nafs) is obligatory (wjib), whatever is necessary for the prevention of this harm will itself be obligatory.74

Elsewhere, he argues that both law and political authority are necessary for avoiding strife and disorder. The best form of law for
serving this end will be a revealed law (shar#a) that warns wrongdoers of afterlife punishment, which will form a deterrent for both
the mob (awbsh), who will not be completely deterred by judging
bloodshed and anarchy (harj wa-marj) to be morally bad (on prudential
grounds), and the ruler, who will be inclined naturally to oppression. This explains the adth, Islam and political power (suln) are
twin brothers.75
The appointment of a powerful despot will constitute a lesser
evil for each individual than the state of lawlessness, and is thus
prudentially necessary. Therefore, individuals will consent to making a compromise for the sake of the advantage of overall security,
by accepting the authority of a powerful sovereign. In the Arba#n,
al-Rz writes:
If it is said: As there are such advantages in the appointment of
this leader (ra"s), there are also a variety of disadvantages: e.g. [1] if

74

Ma#lim, 134. Cf. Muaal, 574; Arba#n, 4278; Khamsn, 701.

Tafsr, 6, 205; 17, 201. The adth is recorded by al-Hind, Kanz al-#umml,
6, 45.
75

178

chapter four
[subjects] refuse to obey him, corruption will increase;76 [2] he may
become despotic and oppress them; and [3] he may increase taxation77
to strengthen his leadership; so he resorts to extracting money from the
weak and poor.we will say: There is no doubt that such wrongs will
occur. Yet every rational person will know that if the disadvantages
produced from the non-existence of the obeyed leader are compared
to the disadvantages produced from his existence, they will be much
greater. When we find such a dichotomy, consideration should be given
to [the question of which option is] preponderant; for the omission of
a greater good for the sake of a lesser evil is a great evil.78

The question of whether the establishment of the immate is necessary or not was debated from an early stage in Islamic history.
The Khawrij maintained that it is never obligatory, whereas the
Mu#tazil Ab Bakr al-Aamm (d. 225/840) reportedly argued that
only at times of strife does it become obligatory to appoint an imm
for the sake of imposing order. The common Twelver Sh# view
is that God is morally obligated to establish the immate for the
sake of human advantage.79 According to the common Sunni view,
also held by most Mu#tazila and the Zayds, the establishment of
the immate is made obligatory on people by Legal evidence only,
normally consensus. Al-Rz accepts this view in his early kalm
works.80
Yet others were of the view that the establishment of the immate
is known to be obligatory on humans by unaided reason: al-Rz
names some Mu#tazils, viz. al-Ji, Ab l-\usayn al-Khayy (d.
ca. 300/913), Ab l-Qsim al-Ka#b, Ab l-\usayn al-Bar and his
followers.81 These theologians base the obligatoriness of establish-

76

This may imply that when such disobedience occurs, oppression and civil
war may follow.
77 Reading kharj instead of araj. Cf. paraphrase from the Arba#n in Ibn
#Arafa, Mukhtaar, 192.
78 Arba#n, 428; cf. Jmi#, 218. Since good and evil in the statement, The
omission of a greater good for the sake of a lesser evil is a great evil, should, in
my opinion, be understood in terms of benefit and harm, his use of this statement
in this context will not contradict al-Rzs rejection of the original, Neoplatonic
sense thereof (cf. p. 1623 supra).
79 See al-Rzs refutation of al-Sharf al-Murta in this regard (Nihya, fol.
291b295b; Arba#n, 42833; Ma#lim, 135; cf. al-Murta, Dhakhra, 409 ff.).
80 Ul al-dn, fol. 41820; Nihya, fol. 290a291b. Cf. #Abd al-Jabbr, Mughn,
20, 16 ff.
81 Muaal, 574; Arba#n, 4267; Barhn, 2, 199. Cf. al-Ji, Istiqq, 1947;
Lambton, State, 58 ff. I have not found any reference to this view by al-Khayy or

a commentary on rislat dhamm al-ladhdht

179

ing the immate and their justification of political authority on the

prudential necessity of having a powerful ruler who is capable of
enforcing social order.
Al-Ghazl, too, adheres to a comparable position in the Iqtid,
though with a crucial difference.82 He argues that one of the objectives
(maqd) of the Lawgiver is the preservation of the order of religion
(nim al-dn). However, this is possible only by the establishment of
the order of the world (nim al-duny), which is achievable by the
instalment of a recognised ruler. Therefore, the instalment of such a
ruler is a religious necessity (min arriyyt al-shar#). As an application of the principle of unsupported interest (malaa mursala), this
line of argument relies ultimately on evidence from scripture.
By contrast, al-Rz bases his doctrine of the necessity of the
immate, not on an obligation to serve the objectives of Revealed
Law, but on purely prudential necessitya stance that is yet another
expression of his core normative consequentialism. He seems to
draw on these earlier Mu#tazil sources (especially, one presumes,
Ab l-\usayn and his school) in arguing for the necessity of political
authority. Yet, in contrast to al-Ji, al-Rzs argument emphasises
the individuals concern for his own safety and interests, rather than
public wellbeing, as the immediate ground for obligation.
Al-Rzs Hobbesian political theory contrasts sharply with alFrbs utopian optimism, according to which humans associate
primarily to gain various perfections, and only secondarily to protect
themselves from the evils of the asocial state. Al-Frb states that
human innate nature (fira) motivates man to interact and co-operate socially, so as to perfect himself and the conditions of his being,
with the aim of attaining happiness. It is unnatural for humans to
be inclined to strife (taghlub).83
By contrast, Ibn Sn sees the purpose of the polity to be the
establishment of social order and the maintenance of a basic degree
of human wellbeing.84 Hence, the goal (of the elite few) of attaining

al-Ka#b in a source earlier than al-Rz (Josef van Ess cites this view by al-Khayy
from a source that paraphrases al-Rzs Arba#n; see Khayy in EI2, 1163; Ibn
#Arafa, Mukhtaar, 190).
82 Al-Ghazl, Iqtid, 2347; cf. Lambton, State, 110 ff.
83 Al-Frb, $r", 117 ff.
84 Ibn Sn, Shif", Ilhiyyt, 2, 44155; Miriam Galston, Realism and Idealism, esp. 568.

180

chapter four

happiness is confined to the private realm, and is not to be pursued

by the art of politics. Government is necessary to realise justice and
security, since people are naturally inclined to consider just what
others owe them, and unjust what they owe others. Therefore,
divine providence will necessarily provide circumstances conducive
to establishing government, most importantly by sending prophets.
Galston writes that, in contrast to al-Frbs political philosophy,
for Ibn Sn, the virtuous individual replaces the virtuous city as
the highest concern of practical philosophy and, concomitantly, the
just city (al-madnah al-#dilah) replaces the virtuous city as the ultimate
goal of political science.85
Al-Rz moves further away from al-Frbs political optimism. As
he states in Dhamm al-ladhdht, the establishment of just government
is so rare that it is almost an impossible and irrational pursuit; even
if established, it will be short-lived. Political pursuit should aim, not
at the realisation of human happiness or wellbeing, but at securing
the most fundamental necessities for human existence, viz. a degree
of security, law and order.
Ibn Sn maintains (generally following al-Frb) that the polity is
established by divine providence, through the vehicle of prophecy. A
contract of the polity (#aqd al-madna), to which people should consent, is then introduced, as a final stage in the establishment of the
polity, to define its general principles and structure, and the rights
and duties of each of its political strata.86 By contrast, according to
al-Rz, the political contract itself establishes the polity and is the
most essential aspect thereof. It does not merely affirm a previous
political process (whether initiated by prophecy or innate human
nature), but is intended to radically alter the previous state. It represents reason curbing human nature.
The centrality given to the social contract appears in the view that
the only ground, or cause, for the establishment of the immate
(sabab ul al-imma) is the contract of pledging allegiance (#aqd albay#a), which, al-Rz writes, is the view of Sunnis and Mu#tazils,
rather than designation (na), or characteristics pertaining to the
person of the imm, such as lineage, according to Imms.87 However,

85
86
87

M. Galston, Realism and idealism, 564.

Ibn Sn, Shif", Ilhiyyt, 2, 447.
Arba#n, 4378. On #aqd al-bay#a, see e.g. #Abd al-Jabbr, Mughn, 20, 251 ff.

a commentary on rislat dhamm al-ladhdht

181

following the common Sunni position,88 al-Rz maintains that the

pledge of allegiance requires the satisfaction of certain conditions
(shar), including character traits, in the prospective imm, which
will qualify him to fulfil the duties of leadership. Nine attributes are
listed in Nihyat al-#uql: religious knowledge, practical wisdom and
competence in administration and war, bravery, justice, maleness,
free status, puberty, reason, and membership of the Quraysh tribe.89
One who meets the conditions and is given the pledge of allegiance
becomes the legitimate imm, even though he might not be the best
living person for the position. Responding to the argument that a
lesser person (mafl) should never become a leader when someone
with superior qualities (fil) can be found, al-Rz writes:
If the lesser person has the qualities of leadership, but is inferior in
these virtues to another, and if we know that were leadership to be
bestowed on another who is superior, unrest and disorder would result,
whereas if it is bestowed on the inferior person, the social condition will become orderly and the public good will be guaranteed,
then reason will judge that leadership ought to be bestowed on that
lesser person. For the purpose of installing the imm is to preserve
the public good. Therefore, if the preservation of the public good is
realised only by bestowing leadership to the lesser person, then that
will be necessary.90

The conditions of leadership are thus not absolute, but are means
to an end. Al-Rz also argues that while the ascetic (nsik) is ultimately the most virtuous person, neither he nor the jurist (faqh) will
be able to rule better than the political expert (syis), who is inferior
in virtue to both of them.91

Epistemological Pessimism
The third section of Dhamm al-ladhdht concerns intellectual pleasure,
which, in contrast to sensory and imaginative pleasures, al-Rz associates with the human good. However, if his expression of pessimism

88

Al-^j (Mawqif, 8, 349) writes that this is the view of the majority (jumhr).
Nihyat, fol. 295b269a; fol. 301a302b (quoted in Ibn #Arafa, Mukhtaar,
1934); cf. Ul al-dn, fol. 4378.
90 Arba#n, 460; cf. Jmi#, 205.
91 Nihya, fol. 302b303b.
89

182

chapter four

in this work with respect to human sensory activity is not out of

the ordinary, and if his pessimism with respect to social association
is highly unusual, the pessimism he expresses here in relation to
intellectual activity is downright surprising. Al-Rz, of course, has
a reputation for being an exceedingly confident rationalist, which
indeed he lives up to in the absolute majority of his works.
Scepticism in Dhamm al-ladhdht
He begins the section by maintaining that intellectual pleasure is
associated with the pursuit of the rational (#aql) sciences, to the
exclusion of positive (wa#) sciences, which are practiced for their
utility in serving the physical dimension of human existence, and are
thus inferior. Rational sciences either serve other sciences, such as
logic, or have ends of their own. The latter are of four types, which
concern: (a) knowing God, (b) knowing spiritual entities (rniyyt),
(c) knowing the higher world, and (d) knowing the lower world.
With respect to the highest of these sciences, viz. theology, alRz writes:
Yet, who has reached the threshold of that lofty presence! And who
has smelled the aroma of that sacred eminence! Indeed, the yield of
all minds is but presumptions (ann) and conjectures (isbn), and the
culmination of this pursuit is but estimations (wahm) and imaginations
(khayl)!92

With this sceptical note, which is then supported by several arguments,

the tone is set for the rest of this third section of Dhamm al-ladhdht.
Particularly noteworthy is al-Rzs use of the expression isbn,
which is often associated with sceptics, the isbniyya.93
The first argument is as follows. A demonstration is apodictic only
if it has apodictic premises and valid syllogistic form. If the premises
are apodictic propositions, they will be either immediately apodictic
(yaqn ibtid"an), i.e. self-evident, or deduced ultimately from premises
that are themselves immediately apodictic, through one or more stages

92

Dhamm al-ladhdht, 32.

On the isbniyya, see van Ess, Skepticism in Islamic Religious Thought, 1;
15, n. 3. Ibn #Abd Rabbih (Al-#Iqd al-fard, 2, 407) relates that one of the isbniyya
summarised his persuasion by stating that all things are [conceived] through
mere phantasm (tawahhum) and conjecture (isbn). Al-Rz, too, makes use of the
expression wahm here.
93

a commentary on rislat dhamm al-ladhdht

183

of reasoning. In either case, the proposition should be subject to the

agreement of all rational people. The same should be true of the
form of the proof. Therefore, if a proof has both apodictic premises
and a valid syllogistic form, there should be no disagreement on its
truth. However, al-Rz adds:
Had so-called demonstrations been in themselves real demonstrations,
anyone who hears and comprehends them should accept them and,
first of all, should not reject them. However, since we observe that
what one disputant calls demonstration, another disputant will hear it,
understand it, but will then reach not even a weak presumption by it,
we will realise that these things are not in themselves demonstrations.
Rather, they are weak premises to which partisanship and sentiment
are combined. Therefore, someone will claim that [his argument] is
a demonstration, although in itself it is not so.94

The second and third arguments reach a similar conclusion. Al-Rz

argues that when two arguments lead to two opposing conclusions,
at least one will, necessarily, be false. This will be due to at least
one false premise in the argument, which someone will judge to be
self-evident.
This indicates that the mind may judge the truth of a fallacy to be
immediately apodictic. This being the case, the minds pronouncements
as regards self-evident statements will become inadmissible. Therefore,
all proofs will be vitiated (tafsudu jam# al-dal"il).95

Al-Rz gives the specific examples of two debates, in which each

side claims to produce conclusive evidence for their position: the
theological debate on anthropomorphism (tashbh) and the reductionism of divine attributes (ta#l); and the debate on the nature
of matter between atomists and hylomorphists. Al-Rz states that
the arguments for each of the two opposing positions in the latter
debate are very convincing and based on supposedly self-evident
premises. The conflict seems irresolvable; for though each position,
taken separately, appears rational, to accept two mutually-exclusive
positions will be completely irrational.96

94

Dhamm al-ladhdht, 33.

Dhamm al-ladhdht, 34.
96 Al-Rz genuinely found the debate on atomism and hylomorphism highly
problematic. In the Mabith (2, 1138), he supports hylomorphism and refutes
atomist physics, including the fallacies (shubha) and doubts (shakk) that its supporters
direct against hylomorphism. At a later stage, he seems undecided and suspends
95

184

chapter four

These three arguments highlight the view that the mind is able
to judge some false opinions as constituting apodictic knowledge,
which casts doubt on the very notion and possibility of certainty as
such. If we can trust neither immediate nor discursive knowledge,
we will be unable to affirm any knowledge at all: Reason will be
undermined.97 The sceptical mode that underlies these first three
arguments invokes disagreement (ta#ru) among views taken by various parties, especially philosophers and theologians, pointing to the
poor track record of reason in the history of ideas.98
Al-Rzs fourth argument is taken from the sophists, and is one
that he cites and rejects in earlier works.99 It goes as follows:
If we reflect, contemplate and investigate, and if following this reflection a conviction arises, then our knowledge that this conviction constitutes knowledge cannot be immediate (arr); for truth frequently
turns out to be contrary to it. If it is [said to be] discursive (naar),
it will need another proof; and infinite regress (tasalsul) will follow,
which is absurd.100

Following from the conclusions of the previous argument, this sceptical argument asks for a particular criterion that will enable us to
overcome this general doubt, and to distinguish some of our opinions
as constituting certain knowledge, rather than mere beliefs. However,
any criterion will in turn require further criteria to justify it, which
will lead to infinite regress.
Both sceptical modes, that of interminable disagreement and that
of the infinite regress of proofs, have a long history before and after
al-Rz, and are both referred to by Sextus Empiricus as being the
two most basic sceptical strategies.101 Al-Rz appears here to arrive

judgement (Mulakhkha, fol. 226a; also fol. 216b226a). He later adopts atomist
physics and refutes hylomorphism (e.g. Muaal, 268 ff.; Arba#n, 317), apparently
dedicating a work to this subject (Rislat ithbt al-jawhar al-fard, mentioned in Arba#n,
264; cf. al-Zarkn, Fakhr al-Dn al-Rz, 70).
97 Dhamm al-ladhdht, 35.
98 This is reminiscent of the notion of the equipollence of opposing proofs (takfu"
al-adilla) (on this, see van Ess, Skepticism, 7). In this context, al-Rz does not seem
to use the latter notion, which implies the affirmation of equivalencesomething
that, as we will see, he may prefer to avoid.
99 Nihya, fol. 13a; fol. 16a; Mulakhkha, fol. 83b; Muaal, 119; Ma#lim, 212;
Khalq, fol. 4a.
100 Dhamm al-ladhdht, 36.
101 Sextus Empiricus, Outlines of Scepticism, 1.178-79. Cf. Jonathan Barnes, Some
Ways of Scepticism, 20610.

a commentary on rislat dhamm al-ladhdht

185

at global scepticism, whereby people only have beliefs, never knowledge, which is defined in apodictic terms.
Having presented these four general arguments, al-Rz supports
his sceptical stance with a specific argument concerning actual controversies among the main metaphysical and theological worldviews.
He outlines the following:
1.
Atheistic views deny the existence of a producer for the physical
world, but explain its existence in one of three ways:
(a)
Physical objects in this world are necessary in themselves.
(b) Their existence is preponderant to their non-existence.
(c)
They are temporally originated. But temporally originated
things do not require a cause.
2.
Some maintain that there is an essentially necessitating First
Cause. They adhere to one of two possible positions:
(a)
The First Cause produces one effect. (Al-Rz specifies
most of the falsifa).
(b) The First Cause may produce more than one effect.
3.
Some maintain that the Creator has choice and considers human
advantage in His acts. They then hold one of the following views
with respect to the problem of evil:
(a)
Two gods exist: a good one, who produces all good, and
an evil one, who produces all evil. (Dualists).
(b) The world came into being when the Soul, out of its ignorance, attached itself to matter. All good in the world is
from God, and all evil is from the Souls ignorance. (Ab
Bakr al-Rz).
(c)
The suffering endured in this world is not bad, since God
will compensate humans for it. (Mu#tazils).
(d) The world is predominantly good, and is the best possible
world. Not to have created it would have been a great evil.
(Al-Ghazl?)
4.
Some maintain that the Creator is a voluntary agent, who
does not consider human advantage in His acts. He sometimes
delivers benefits to humans, sometimes harms. They fall under
two groups:
(a)
Some deny prophecy, the afterlife and religious obligation.
(b) Some affirm prophecy and religious obligation.102
102

Dhamm al-ladhdht, 379.

186

chapter four

If al-Rz were to remain consistent with the global scepticism implied

in his previous arguments, he would argue for the inability to demonstrate the truth of any of these positions. This would inevitably
lead to an extreme agnosticism, whereby even the existence of a
creator for this world is not affirmed. Metaphysical enquiry would
be futile.
Yet al-Rz immediately begins to assess the praiseworthy and
blameworthy aspects of each position. He states that the denial of
the existence of the Producer is the most heinous of all positions,
and the furthest among them from reason and good sense; for the
need of existent things for an effecter may be proved through the
argument from the contingence of physical objects. After briefly
discussing these premises, al-Rz notes that they involve complex
and obscure discussions, referring the reader to his philosophical
works, particularly the Malib.103
However, for the second, third and fourth main stances, we are
not given final conclusions, but are simply referred to the Malib.104
He even states that the last position (4b), which he normally accepts,
has its obscurities and problems. Hence, al-Rz rejects atheistic
positions, but points out that there are many confusions and complications to be found in the investigations of other positions, though
his references to his discussions in the Malib seem to suggest that
some sort of assessment can be made in relation to some or all of
them. He concludes:
When you have grasped these lofty stations and fine, elevated preambles,
and discerned what intricate problems and obscure objections that each
involves, you will realise that [attaining] certain knowledge is difficult,
and that conclusiveness in each topic, such that it becomes free from
[propensity for further] contention and confusion, is very rare. As
such, the yearning is severe, the privation prevalent, the instrument
feeble, and the goal insuperable!105

Certainty may thus be attainable, yet rarely and with great difficulty.
In relation to the almost absolute and global scepticism of the first
four arguments in this section of Dhamm al-ladhdht, the scepticism
expressed here appears qualified and restricted. Al-Rz then writes,
in conclusion to Dhamm al-ladhdht:
103
104
105

Dhamm al-ladhdht, 40; cf. Malib, 1, 7490.

Cf. Malib, 3, 77100; 4, 37397.
Dhamm al-ladhdht, 42.

a commentary on rislat dhamm al-ladhdht

187

If you recognise these conditions, sensory pleasures will become vile,

and imaginative pleasures will become abject. As for intellectual pleasures, there will be no way to attain them, approach them or rely
on them. For these reasons we say: Would that we had remained in
primordial non-existence! Would that we had never seen this world!
And would that the soul had never become attached to this body! On
this theme I say,
Entanglement, the acme of minds pursuit,106
Most human endeavour is but straying;
Our souls are estranged from our bodies,
The yield of our world, but harms and bane;
All weve gained from a lifelong research,
Is but collecting quotations and sayings;
Many a man and dynasty have we seen,
That all quickly perished and expired;
Many a mountaintop was surmounted,
By men, who perished, yet the mountains remain.107
Know that after deep penetration into these defiles, and delving deeply
(ta#ammuq) in exploring the secrets of these matters, I have found the
most correct and advantageous [method] (al-awab al-ala) in this
regard to be the method of the holy Qur"n (arqat al-Qur"n), the
noble Furqn, which is the abandonment of delving deeply, and of
inferring the existence of the Lord of the Worlds from the divisions
of bodies in the heavens and the earth, and then proclaiming the
greatness [of God] to the maximum extent (al-mublagha f l-ta#m),
without wading into details.
Thus, I read, on deanthropomorphism (tanzh), [Gods] saying, God
is the Self-sufficient and you are the needy,108 His saying, Naught
is as His likeness,109 and His saying, Say, He is God, the One.110
And I read, on the affirmation [of divine attributes] (ithbt), The

106 Nihyat aqdm al-#uql #iql. Interestingly, this line echoes two kalm titlesalRzs own Nihyat al-#uql, and al-Shahrastns Nihyat al-aqdman apparently
symbolic pun.
107 Meter: awl. Cf. the following Persian quatrain attributed to him (quoted in
translation by Seyyed H. Nasr, Fakhr al-Dn al-Rz, 653, without reference):

My heart was never deprived of science,

There is little of the mysteries that I did not understand.
For seventy-two years I thought night and day,
Yet I came to know that nothing is to be known.
One problem in these lines, however, is that al-Rz died at the age of sixty-two.
108 Qur. 47:38.
109 Qur. 42:11.
110 Qur. 112:1.

188

chapter four
Beneficent is established on the Throne,111 His saying, They fear
their Lord above them,112 and His saying, Unto Him good words
ascend.113 And I read, on that all is from God, His saying, Say,
all is from God,114 and, on exalting Him above what is inappropriate, His saying, Whatever of good befalls you, it is from God; and
whatever of ill befalls you, it is from thyself.115 And so forth, by this
same rule (qnn).
I say, from the depth of my heart, and the inmost of my soul: I
confirm that all that is most perfect, most virtuous, greatest and most
glorious pertains to You, and that You are exalted above all that involves
defect or imperfection. I confirm that my mind and comprehension
fall short of attaining the true reality of a single atom of the atoms of
Your creatures. And I confirm that I have failed to praise You with
what befits You. For eulogies are of either of two types; they either
expound attributes of majesty (jall), which is exalting God above what
is inappropriate, or expound attributes of beneficence (ikrm), which
is attributing to God His being the creator of this world. Yet the first
involves ungraciousness in some respects, for if a man tells a king,
You are not blind, deaf, or leprous, he will deserve reprimand and
confinement. As for the second, it involves ungraciousness; for all the
perfections of creatures are imperfections in relation to the Creators
perfection; hence, expounding the Creators perfection using relative
attributes that relate [Him] to creatures involves ungraciousness.
O, Lord of might! I admit that I am incapable of praising You
except in either of these two ways! And I admit that neither befit Your
majesty and might. Yet I am as though excusable; for I know naught
but this, and I am unable to find anything superior to it!116

The beginning of this statement is by far the most explicitly pessimistic statement to be found anywhere in al-Rzs writings, and
is among the most extreme expressions of pessimism that can be
found in medieval Arabic prose. It remains surprising even as a
conclusion to the pessimistic trend throughout Dhamm al-ladhdht.
And indeed, the interest of this statement, which clearly expresses
a genuine sentiment and sincere conviction of its author in one of
his latest texts, made parts of it among the most frequently quoted
pieces of his writings.

111
112
113
114
115
116

Qur. 20:5.
Qur. 16:50.
Qur. 35:10.
Qur. 4:78.
Qur. 4:79.
Dhamm al-ladhdht, 423.

a commentary on rislat dhamm al-ladhdht

189

Now, the scepticism presented in Dhamm al-ladhdht will initially

appear inconsistent. The first four general arguments imply a sweeping, global scepticism, yet they are followed by an expression of a
significantly milder and more restricted scepticism. The latter stance
does not exclude the possibility of arriving at knowledge in rational
theology, but considers it to be rarer and much more difficult than
normally believed to be the case. However, these two trends may
not be as contradictory as they seem. The first four arguments can
be seen to represent (rather sketchy) attacks on the notion of certain
knowledge, but do not necessarily lead to a rejection of the assertion
of more or less justified belief. Despite al-Rzs disillusionment with
philosophical and theological pursuit, he does not conclude that it is
entirely futile and that it calls for the total suspension of judgement
(tawaqquf). Indeed, his contention that the way of the Qur"n is
superior indicates that the difference is one of degree.
Scepticism in the Malib
We may test the above interpretation of the scepticism expressed
in Dhamm al-ladhdht by considering relevant discussions in the
Malib, most importantly its epistemological introduction, written
shortly before Dhamm al-ladhdht.117 The book opens with a section
entitled On explaining how this science [i.e. theology (al-#ilm alilh, uthljy)] is the best of all sciences absolutely, which clearly
expresses a favourable view of philosophical theologyhardly a
sign of scepticism.
The following section, however, is entitled On whether there
is a way for human minds to attain certainty (al-jazm wa-l-yaqn) in
this science, or whether it suffices with respect to some of its topics and questions to accept what is most probable and apt (al-awl
wa-l-akhlaq). Al-Rz here argues that certain knowledge in many
(rather than some) important theological questions is unattainable,
and that one should be content with the most probable conclusions
in them. He advances several arguments for this stance.
The first line of argument is that our knowledge of some items
that should be among the most evident and immediate objects of
knowledge to our minds is highly dubious and problematic. We may

117

Malib, 1, 3764.

190

chapter four

know of them (#ilm), but we cannot properly understand and conceive

them (ma#rifa).118 It follows that our knowledge of less evident items,
including divine nature, must be much more obscure and problematic, and that our minds may only attain probable belief in such
cases, never certain knowledge. Al-Rz considers this argument to
be based on an induction from numerous particular cases, of which
he provides four illustrative examples: knowledge of the nature of
the self, time, space and the principles of geometry.119
He first argues that the knowledge that is supposed to be the clearest and most immediate knowledge to the individual is that of the
nature of his own self. Presumably, this follows from his view that one
knows immediately that his self exists. Yet this knowledge is of the
utmost obscurity and uncertainty; for many conflicting theories on
the nature of man and the human soul have been advanced, involving complicated discussions, which leave the mind bewildered.120 So,
how could it be claimed that one is able to attain certain knowledge
of things that are epistemically remote? Moreover, a theory of knowledge should rely on a notion of the nature of the self, which is the
knower. If the nature of the knower is unknown with certainty, then
the notion and integrity of knowledge itself will come into question.
One is reminded here of the common view, which al-Rz expresses
frequently, that knowing God presupposes knowing oneself.121 The
same argument is advanced in Asrr al-tanzl, followed by two lines
of Persian poetry from San" (d. ca. 548/1152):
O thou, who art incompetent to know thine own nature, how wilt thou
ever know God? Since thou art incapable of knowing thyself, how wilt
thou become a knower of the Omnipotent?122

Al-Rz then underscores the interminable controversies on the nature

of time, space and matter. He concludes, with reference to the last
controversy, which is already cited in Dhamm al-ladhdht:
Whoever examines that problem and realises the strength of each
sides evidence will know that the mind will necessarily culminate in

118 On this distinction between #ilm and ma#rifa, cf. al-Tahnaw, Kashshf, Ma#rifa;
Ma#rifa, EI2.
119 Malib, 1, 416.
120 Cf. p. 1167 supra.
121 E.g. Risla f l-nafs, fol. 2a.
122 Asrr al-tanzl, 1412; cf. San", \adqat al-aqqa, 63; M. Stephensons translation, 4.

a commentary on rislat dhamm al-ladhdht

191

perplexity and bewilderment and in [having to] accept what is probable and apt.
This induction shows that, in its attempt to understand the most evident
objects of knowledge, the mind will culminate in utter perplexity and
bewilderment (ma al-ayra wa-l-dahsha) and will have to accept what
is probable and apt. So how do you reckon the mind will fare when it
[attempts] to ascend to the gates of Gods greatness, and when it tries
to delve into investigations of the nature of His [attributes]!123

Another example that al-Rz then provides is from geometry, which

is seen as one of the most exact and rigorous rational disciplines.
Although Euclid, according to al-Rz, contends that an infinite
number of regular polygons may be conceived, he was only able
to actually demonstrate the constructability (i.e. by the means of
compass and ruler only) of five: the equilateral triangle, the square,
the pentagon, the hexagon, and the pentadecagon (the polygon of
fifteen sides).124 Demonstrations of the constructability of two more,
the heptagon and the nonagon, were later attempted by specialists in
conics, but the constructability of other regular polygons remained
unproven.125 This shows, al-Rz argues, that reason has a severe
handicap in understanding some of the most tangible, basic and
presumably rational aspects of the nature of this physical being. It
is likely to have an even greater handicap in understanding divine
nature.126
Al-Rz also advances the following main argument. The conceptions (taawwur) that may be perceived by reason, estimation (wahm)
and imagination are of four types: (a) essences perceived through the
five physical senses; (b) essences perceived internally and immediately
in the self, such as pain, pleasure, hunger, joy and anger; (c) essences
that are perceived by the mind innately, such as conceptions of existence, non-existence, oneness, multiplicity, necessity, possibility and
impossibility; and (d) essences that reason and imagination combine
(rakkaba) from the simple essences of the three previous types. Since
it will be impossible for reason and imagination to perceive anything
other than these types, al-Rz concludes elsewhere that human

123

Malib, 1, 44; cf. Asrr al-tanzl, 142.

I am grateful to Dr Sonja Brentjes for informing me that in fact Euclid does not
make the claim that an infinite number of regular polygons may be conceived.
125 Cf. Jan Hogendijk, Greek and Arabic Constructions of the Regular Heptagon.
126 Malib, 1, 446.
124

192

chapter four

conceptions are all self-evident (badh), none are acquired (kasb,

muktasab).127 Therefore, if one is to affirm the statement, There
does not exist a partner for God, one needs to have a conception
of partner for God. He will have a conception of partner in some
contexts and a general conception of divinity (al-ilh). So he will
be able to combine both to conceive the meaning of partner for
God, and to conclude that it is inconceivable for a thing to exist
that the relation of which to God is similar to the relation of my
partner to me.128
Al-Rz argues that the range of possible conceptions that the mind
may perceive restricts the range of possible assertions (tadq) that it
will be able to make. This recalls his usual notion of assertion as
consisting of a combination of conceptions with a judgement (ukm)
in either affirmation or negation.129 Without having their constituent
conceptions in mind, true and meaningful statements of assertion
cannot be made.130 In other words, they would be nonsensical.
In the process of intellectual reflection (naar), or thought (fikr), the
mind will only acquire assertions on the basis of other assertions.131
When the mind considers problems of a high epistemic order, such
as those relating to divine nature, it will still use the same above types
of conception in its deductions. Yet, since divine nature is different
in all respects from these conceptions, both simple and combined,
it will be impossible to make statements of assertion that refer to it
essentially.132 This seems to render most statements on divine nature
as effectively nonsensical. Al-Rz concludes:

127

Muaal, 814; Arba#n, 4789.

Malib, 1, 4950.
129 Muaal, 81; Mulakhkha, fol. 1b2a.
130 Knowledge, it should be recalled, is defined as an apodictic statement of
assertion that corresponds to reality, whereas non-apodictic statements may constitute beliefs, or presumptions, depending on the degree of conviction one has in
them (Mulakhkha, fol. 155a).
131 Muaal, 121; Ma#lim, 21.
132 Al-Rz (Lawmi#, 35) uses this same argument to refute a Mu#tazil position
on divine attributes:
128

If we say: His being knowing and powerful are two positive items, distinct from
[His] essence, Ab Hshim [al-Jubb"] will say: Being knowing (#limiyya) and
being powerful (qdiriyya) cannot be said to be existent or non-existent, knowable
or unknowable. Most sensible men agree on that his saying is false.
For, necessarily, every assertion has to be preceded by conception. So, if these
two attributes are not conceived, it will not be possible to maintain that both are

a commentary on rislat dhamm al-ladhdht

193

Minds fall short of knowing Him, and perceptions do not reach Him.
Rather, the highest aim [for reason] is that if we perceive the meaning
of perfection and imperfection in relation to ourselves, ... we will
understand the meaning of perfection and imperfection as such.
For the unconditional [conception] (mulaq) is part of the essence of
the conditional (muqayyad) [conception]. In this way, the meaning of
perfection and imperfection may be perceived. If we perceive this
meaning, we may accept to affirm the designation perfect with respect
to [God], provided that we purge this designation of all concomitants
associated to it when it applies to us.
Most people will have only this much knowledge of [Gods] majesty. With this explanation, it becomes clear that human minds come
to achieve nothing but these general items of knowledge, which are
affirmed only according to what is most probable and apt, but not
in detail.133

The scepticism that al-Rz reaches in the Malib finds expression in his numerous references to bewilderment and perplexity
(al-ayra wa-l-dahsha).134 Though his sceptical stance, viewed within
the history of ideas generally, will appear rather moderate, it is
indeed a radical scepticism in the context of kalm and falsafa. In
the wider Islamic context, it was generally seen that no less than
knowledge, defined in terms of certainty (qa#, yaqn), should be
contented with in theological questions, to the exclusion of mere
belief and presumption (ann), which contain an element of doubt
(shakk, shubha).135 Al-Tahnaw, for example, records the common
view that doubt is a subspecies of ignorance (jahl).136 Those who
favoured a discursive approach did so with the conviction that it
does provide certainty. Thus, in his earlier works, al-Rz writes,
with reference to metaphysics, that affirming what is most prob-

attributes of essence. Also, if an attribute is not conceived, it will not be possible

to assert that it is not conceived. For our saying, This is not conceived, is a
proposition; and every proposition should be preceded by the understanding of its
subject and predicate. Also, what is judged to be unknowable is not the essence, but
an attribute; so this attribute is singled out and said to be not conceivedwhich
is contradictory.
133 Malib, 1, 501.
134 E.g. Malib, 1, 42; 1, 44; 2, 98; 4, 368; 4, 426, where he almost classifies
himself as one of ahl al-ayra wa-l-dahsha. Cf. A. Shihadeh, Review of `skenderolu,
Fakhr al-Dn al-Rz, 215.
135 Cf. Franz Rosenthal, Knowledge Triumphant, 300 ff.; R. Frank, Knowledge
and Taqld, 43.
136 Al-Tahnaw, Kashshf, Shakk.

194

chapter four

able (awl) is inapt in apodictic disciplines (l yalqu bi-l-qa#iyyt).137

Therefore, in the Islamic context, sceptical outlooks that consider
divine nature to be beyond the grasp of discursive reasoning normally
lead to the total rejection of rational theology and metaphysics as
futile. This will be followed by either a sceptical suspension of judgement (tawaqquf), or, more commonly, by the acceptance of an alternative epistemology, be it scriptural, esoteric, or traditional.138
However, these two outcomes are not logically necessary. For
scepticism, it has been argued, may be characterised by two distinct
features: a thesis, asserting the impossibility of knowledge, which
can be either global or restricted to specific classes of knowledge;
and a recommendation that one ought to suspend judgement.
These two statements, the is and the ought, are logically independent of each other, as the thesis is not sufficient to justify the
recommendation.139
Now, in the Maalib, al-Rz goes against the current trend by
putting forth a restricted sceptical thesis, yet without recommending
an automatic suspension of judgement. He contends that although
knowledge cannot be attained discursively with respect to many
theological questions, theological reflection may still show that some
non-apodictic theological statements are more justified and more
plausible than others. When the mind is unable to attain apodictic
knowledge in relation to a given theological problem, it may have
reason to affirm, tentatively, the most probable, plausible, or apt
(awl, akhlaq, ashbah) belief (or conviction) that it can attain, while
recognising its fallibility. This position, which al-Rz adopts at this
late stage, explains the seeming contradiction in the third section of
Dhamm al-ladhdht, noted previously.
In the Malib, one finds that this scepticism leads to a surprising
expression of theological tolerance by a thinker who is notorious for
his debating tenacity. Having found all rational and revealed evidence for the question of the pre-eternity of the world inconclusive,
al-Rz writes:

137

Mabith, 2, 482.
In earlier works, al-Rz refutes these alternative approaches in favour of theological and metaphysical reflection (Nihya, fol. 19b22b; Muaal, 122; 1267).
139 Gisela Striker, Sceptical Strategies, 54; cf. Charlotte Stough, Greek Skepticism,
4.
138

a commentary on rislat dhamm al-ladhdht

195

Now that these various positions have been summarised in this manner,
and that the praiseworthy and blameworthy aspects of each have become
evident, at this point those who are perplexed and bewildered (ab
al-ayra wa-l-dahsha) will say: These proofs are not clear and strong
enough to dispel doubt (shakk), preclude excuses and enlighten the
mind with their strength and vividness. Rather, each betrays a degree
of obscurity. One who is merciful and magnanimous should excuse
one who errs in such defiles.140

By contrast, al-Ghazl, who emerges out of his brief phase of doubt

with very firm certainties, declares the falsifa to be unbelievers for
maintaining that the world is pre-eternal. The attitude of tolerance on the basis of doubt, which al-Rz expresses here, was not
uncommon in the discipline of fiqh; yet it was generally alien to
earlier Islamic theology.
It goes well beyond the scope of the present study to convey how
this sceptical stance affects al-Rzs treatment of specific theological and philosophical questions.141 So does the examination of his
extensive discussions in his earlier works of a wide variety of sceptical
outlooks and arguments. We should also leave aside, in the present
study which focuses on the ideas, the (probably worthwhile) attempt
to speculate on al-Rzs circumstances or motives that may have contributed to his reaching this scepticism. Suffice it in the remainder of
the present subsection to examine briefly the intellectual background
to this particular sceptical stance (that in most cases in metaphysics,
one may only reach probable belief, rather than certainty), which
al-Rz adopts, to the extent permitted by the extant textual sources.
While his exact influences cannot at this stage be pinpointed, relevant
general trends in this background are identifiable.
The sceptical arguments he refutes in his earlier works, especially
Nihyat al-#uql, the Mulakhkha and the Muaal, mostly advocate the
suspension of judgement, either in relation to knowledge as such,
discursive reasoning generally, or discursive reasoning in metaphysics
specifically. Yet one also finds cursory references to the contention
that while certain knowledge in metaphysics is unattainable through
discursive reasoning, some views may be accepted on account of being
more plausible, probable, or apt than others. Therefore, instead of
140

Malib, 4, 426. Cf. Tafsr, 2, 523.

On al-Rzs treatment of the question of the eternity of the world in the
Malib, as an example, see Muammer `skenderolu, Fakhr al-Dn al-Rz; also my
review of this monograph.
141

196

chapter four

suspending judgement in these occasions, one may affirm the most

plausible view, while acknowledging that it is not certain, but probable
(mutamal). For example, the following argument in the Mulakhkha,
advanced by an imaginary sceptic against al-Rz, proposes this as
an alternative to the trend of suspending judgement:
In your present book, you suspend judgement (tawaqquf ) on many
questions, because of the opposition of proofs (ta#ru al-adilla). This
undermines [the notion of] immediate knowledge. For strong, opposed
proofs will be constructed of premises. If the mind is certain about
each premise and unable to undermine it in any way, though we know
immediately that some are false (since it is inconceivable that contradictory premises are all true), then we will find that the mind makes
certain and decisive judgement, without hesitation or probability (itiml),
in relation to [a premise], despite it being false. Therefore, confidence
in the decisive judgement of the mind will cease. Which will lead to
undermining a priori knowledge.
However, if the mind is able to undermine any of those premises,
suspension of judgement will become untenable. One then ought to
show that some of the premises of one side are probable (mutamal)and
the probable does not provide certaintyin which case there will be
no need to suspend judgement with respect to it.142

We may discern a twofold inspiration for this alternative scepticism,

which al-Rz adopts, viz. a metaphysical scepticism among some
mathematicians, and a current which originates in ancient philosophical scepticism.
The metaphysical scepticism to be found among mathematicians
finds expression in sceptical arguments that al-Rz uses in his later
works, most notably his foregoing argument from proofs of the constructability of polygons. It also appears in arguments he cites in
earlier works, such as the following sceptical argument from the
certainties of mathematics, recorded in the Mulakhkha (which, at this
early stage, al-Rz rejects). Statements like 1 is half of 2 constitute
immediate knowledge. But philosophers use premises like It is only
with a preponderator that either the [existence or non-existence] of
the contingent thing will preponderate, and What applies to one
thing applies equally to its like, which, they claim, constitute immediate and certain knowledge. It is clear, however, that these are not
as evident as the proposition 1 is half of 2, which the mind will

142

Mulakhkha, fol. 83b84a.

a commentary on rislat dhamm al-ladhdht

197

be more certain about. Yet there cannot be degrees of certainty,

since the very notion of certainty excludes the slightest probability.
Therefore, the premises that philosophers use are not certain.143
In the Muaal and the Mal, al-Rz writes of a group of geometricians (muhandisn) who accept the use of discursive reasoning
only in arithmetic and geometry, but not in metaphysics, where
the highest aim, they maintain, should be to attain probable belief,
rather than certainty.144 This reference has already been linked both
to an account of a similar view expressed by some mathematicians,
which is recorded in John Philoponos Commentary on the Isagoge,
and to a report of an early debate between mathematicians and
mutakallimn, recorded by al-Ji.145 In the latter, arithmeticians
and geometricians are said to argue that the way of kalm rests on
opinion, conjecture, approximations and delusions, whereas true
science is natural, immediate, and free from inexact interpretations
and equivocations. Similarly, Ptolemy writes, in the then widely read
Almagest, that of the three divisions of theoretical philosophy (viz.
theology, physics and mathematics),
... the first two divisions of theoretical philosophy should rather be
called guesswork than knowledge, theology because of its completely
invisible and ungraspable nature, physics because of the unstable and
unclear nature of matter; hence there is no hope that philosophers
will ever be agreed about them; and that only mathematics can provide sure and unshakeable knowledge of its devotees, provided one
approaches it rigorously.146

There can be little doubt that this sort of metaphysical scepticism,

which rests on a contrast with the certainties that the rigour of
mathematics provides, had a direct influence on al-Rz.147

143

Mulakhkha, fol. 83ab.

Muaal, 122; cf. Mal, 1/1, 207; 1/2, 444.
145 Anton Heinen, Mutakallimn and Mathematicians, 6572, esp. 72; Josef
van Ess, Die Erkenntnislehre des #Audaddin al-^c, 2746; al-Ji, in#a, 57.
146 Ptolemy, Almagest, 36. Cf. F. Jamil Ragep, Freeing Astronomy from Philosophy, 58, which cites an unpublished manuscript of a work by Qub al-Dn
al-Shrz (d. 710/1311), in which Ptolemys above statement is echoed: Astronomy is the noblest of the sciences. ... Its proofs are securebeing of number and
geometryabout which there can be no doubt, unlike the proofs in physics and
theology. I am grateful to Dr Sonja Brentjes for a very helpful discussion we had
on this subject.
147 This metaphysical scepticism had a parallel among physicians, who found
certainty in empiricism to the exclusion of metaphysical speculation. Ab \ayyn
144

198

chapter four

Indicating his other main inspiration for this type of scepticism,

he notes in the Mulakhkha that he read somewhere that Aristotle
maintained that one cannot attain certainty in theological questions,
and that the highest aim in them is to attain belief according to what
is most apt and plausible.148 In the introduction of the Malib, he
indicates that he also found this view attributed to certain ancient
eminent philosophers.149 Aristotle does not seem to have expressed
this view, but this epistemological stance was rather espoused by some
ancient sceptics, including members of the New Academy. According
to Carneades, a head of the Academy, one cannot attain certainty,
but ought not to suspend judgement in all matters when plausible,
or probable positions may be reached. It remains unclear, however,
how this trend in ancient scepticism reached al-Rz.150
In the context of Islamic philosophy and theology, his adoption of
this sceptical stance appears to be unprecedented. Some discussion
of this position continued after him; yet to what extent it had been

al-Tawd (Imt#, 1, 38; cf. Roshdi Rashed, Qh vs. Aristotle, 9) refers to the
aversion that some contemporaries from both groupsphysicians and mathematicianshad towards metaphysics: This one studies illness and health, disease and
medicine, and that one examines the sun and the moon. Yet none of them will
have anything to say on soul, intellect or deity, as though these [subjects] were
forbidden to them, or blameworthy among them.
It is possible that al-Rz, who wrote at least two major works on medicine, was
inspired also by this metaphysical scepticism propounded by physicians, which he
must have come across (although we have found no evidence in his works to confirm
this influence). For instance, in On the Doctrines of Hippocrates and Plato (5779), which
al-Rz read (cf. Nafs, 74), Galen notes that whereas one finds many interminable
disagreements among philosophers, disagreements among physicians are often
resolved: In philosophy it is not surprising that most disagreements have not been
resolved, as the matters it deals with cannot be clearly judged by an empirical test,
and therefore some say that the universe did not have a beginning, others that it
had, and again some say that there is nothing outside surrounding it, others that
there is something, and of the latter, some say that what surrounds it is a void that
has no substance in it, others that it is surrounded by other universes numerous
beyond calculation, so that their number reaches to infinity. Such disagreement
cannot be settled by clear sense-perception. But the case is not the same when a
disagreement arises among physicians about the benefit or harm of remedies applied
to bodies; physicians, at least, can judge by empirical test which of them is helpful
and which is harmful. (Cf. J. Barnes, Some Ways of Scepticism, 206).
148 Mulakhkha, fol. 83b.
149 Malib, 1, 41.
150 For sources on ancient sceptical influences on Islamic thought generally,
see: D. Gutas, Pre-Plotinian Philosophy, 4963.

a commentary on rislat dhamm al-ladhdht

199

influential on later Islamic thought remains to be seen.151 It appears

to have had a profound impact on Ibn Taymiyyas polemics, who
often argues that the falsifa and the mutakallimn can only reach
presumption, not certainty; yet, for him, one should never settle for
less than the latter in theology.152
Epistemological Pessimism and Human Perfection
In the third section of the introduction to the Malib, entitled On
whether this sacred knowledge may be attained by one or more methods (arq), al-Rz contends that there are in fact two approaches
to this knowledge. The first is the discursive, philosophical and theological approach, which was subject to his sceptical scrutiny in the
preceding section. In this approach, one may prove the existence
of God only by deduction from facts about created things.153 This
may yield indirect knowledge of Gods existence, as we will conclude
that created things require an external cause, without knowing the
essence thereof. This limitation in the discursive mode of knowing
God is underscored in another work, where al-Rz quotes Ab lBarakt al-Baghdd as follows:
The knower may know the thing essentially ... or ... non-essentially
(ma#rifa #araiyya). ... When we infer the existence of the Necessary
Existent from the existence of contingent things, this will be a nonessential type of knowledge. For what is known of It will be that It is a
particular essence whose nature is unknown. But we know two features
of It, viz. the dependence of all else on It, and Its independence from
all else. As for essential knowledge, we will not have it, ... neither of
Its essence (since we do not know how Its essence is distinct), nor of
aspects of Its essence (since It is one and non-composite; so Its essence
does not consist of parts).154

The second path to this knowledge, al-Rz writes, is the method of

spiritual discipline, which allows direct and supra-mundane knowledge of God. If one purifies his heart and perseveres in recollect-

151

E.g. al-^j, Mawqif, 1, 13840.

Ibn Taymiyya, Bayn, 1, 372; 2, 252; 2, 478; Majm#, 9, 36; Dar", 1, 159
(French trans. Yahya Michot, Vanits intellectuelles, 607).
153 Malib, 1, 534. Al-Rz also rejects Ibn Sns claim that his ontological
argument for the existence of God does not rely on any empirical premises (cf.
Ibn Sn, Ishrt, 3, 3652).
154 Lawmi#, 99; cf. al-Baghdd, Mu#tabar, 3, 122 ff.
152

200

chapter four

ing (dhikr) God, both inwardly and outwardly, his soul will witness,
directly, divine illumination and knowledge. These are stations that
man will not fathom in detail (#al sabl al-tafl) unless he attains
them.155 Certainty, in other words, may only be reached through the
method of spiritual discipline (arqat al-tafiya wa-l-riya), leading to
spiritual perfection, rather than the discursive method, the method
of reflection and inference (arqat al-naar wa-l-istidll). Nonetheless, proficiency in both paths will allow one to critically assess the
knowledge and stations that he arrives at, which will enable him to
recognise pitfalls along the spiritual path, such as believing that he
has arrived at the end of the path when experiencing very powerful
and unfamiliar states and revelations.156
Al-Rzs metaphysical scepticism, as we saw, rests partly on the
following line of reasoning, in relation to the epistemic limitations
of man: (1) conceptions cannot be acquired; (2) the conceptions
that humans ordinarily perceive fall into certain mundane types of
essences; (3) the range of meaningful statements of assertion is determined by the range of their constituent perceived conceptions; (4)
therefore, statements of assertion that humans can ordinarily make
will be mundane and can refer to supra-mundane items only in a
very general sense. Therefore, the inability to attain supra-mundane
knowledge is not essential to the soul itself, but is due to an accidental limitation in its range of perception, which is determined by the
nature of its association with the physical body. While the soul cannot attain supra-mundane knowledge discursively, it is, in principle,
capable of direct supra-mundane perception, when its attachment
to the body loosens. This is the only way for the soul to attain true
theoretical perfection.
Al-Rz thus concludes that the method of spiritual discipline
is the superior alternative to kalm and falsafa, and the solution to
epistemological pessimism and utter despair. How then should we
understand his numerous statements, in his later writings, that what
he refers to as the method (arqa) of the Qur"n should be regarded
as the superior alternative to kalm and falsafa? He constantly describes
this method as one that involves both the abandonment of delving
deeply (ta#ammuq) into theoretical complications and subtleties, and

155
156

Malib, 1, 545.
Malib, 1, 589; cf. Tafsr, 21, 14950.

a commentary on rislat dhamm al-ladhdht

201

proclaiming Gods greatness to the maximum extent (al-mublagha

f l-ta#m). This finds expression in the above-quoted conclusion of
Dhamm al-ladhdht,157 as well as his last testimony (Waiyya), dictated
at his sickbed shortly before his death:
I tried the methods of kalm and falsafa, and I did not find in them
the profit which I found in the great Qur"n; for it calls to ascribing
all greatness and majesty to God, and prevents from delving deeply
into the preoccupation with objections and contradictions. This is so
only because of our knowledge that human minds come to nothing
and fade away in these treacherous defiles and hidden ways.158

The same view also appears in Book One of the Malib, where
he states, Whoever abandons obstinacy and experiences the like
of my experience will realise that truth is as I have described.159
But what is the method of the Qur"n? And in what way is it a
superior alternative?
For this, we need to revisit al-Rzs later theory of prophecy.
Revelation, he argues, will have spiritually transformative features
in its style and content that are ideal for perfecting human souls.
For instance, the method of revealed religions generally, and Islam
in particular, is to attribute all perfections to God and to exhort
people to proclaim His greatness to the maximum extentmainly by
affirming attributes of majesty (jall) and beneficence (ikrm)while
avoiding intricate theological problems, which may cause confusion
in peoples minds. It will then prohibit believers from delving deeply
into theological details of the various aspects of Gods greatness and
divinity, some of which may appear irreconcilable. These and other
transformative features of prophecy correspond to al-Rzs description of the method of the Qur"n.160 This is described in his Asrr
al-tanzl, where he comments on Qur. 7:545, as follows:

157

Cf. p. 1878 supra.

Waiyya, 640. I made much use of Tony Streets translation (Life and works,
1367).
159 Malib, 1, 236. Ibn Taymiyya (Nubuwwt, 523; cf. Dar", 1, 15960) sometimes
highlights al-Rzs scepticism by blending these two statements and the conclusion
of Dhamm al-ladhdht loosely: I have contemplated the methods of kalm and the
systems of falsafa, and have not found them capable of curing the ill or quenching the thirsty. I have found the most immediate method to be the method of the
Qur"n. I read in affirmation, [etc., see p. 1878 supra]. Whoever experiences the
like of my experience will realise the like of my realisation.
160 Cf. Malib, 1, 216.
158

202

chapter four
His saying, Verily, His are the creation and the command, contains
a remarkable secret and a profound wisdom; namely, that if you witness signs of [divine] governance and design in the various parts of
the heavens, the planets and the earth, your heart will desire to know
the aspects of wisdom in each one of them in detail (#al sabl al-tafl).
It will then be said to you: Halt at your level, do not venture beyond
your limits, nor plunge your mind into an endless ocean, nor set off
to surmount a summit-less mountain, nor desire to fathom what is
above your understanding, imagination, mind and soul! You are not
one of those who could attain these illuminations. Rather, admit your
incapacity, humility, and shortcoming, and proclaim the perfect majesty
and ultimate greatness of the Creator of these beings summarily (#al
sabl al-ijml); and say: Verily, His are creation, command, wisdom,
might, exaltedness, dominion and greatness, blessed be God, the Lord of
the worlds, He who governs all bodies, souls, higher beings and lower
beings! Having abandoned wading into those details and recognised
this greatness summarily, at this stage, you should return to yourself
and consider your incapacity and shortcoming. Then busy yourself
with prayer and supplication.
Herein lies the ultimate level attainable by the spiritually advanced
ones (iddqn) and the acme of the thoughts of knowers (#rif), beyond
which minds cannot venture. ... To this God, exalted, alludes towards
the end of this verse in His saying: Supplicate your Lord humbly
and secretly.
How beautiful these pointers (talw), contained in the great Qur"n,
are! ... No method (arq) occurs to the mind or imagination of this
humble person (miskn), who is the author of the present book, which
is better, more advantageous or more attractive (ajdhab) for human
souls and intellects towards the presence of Him who is Holy, One
and Everlasting, than these divine expositions and lofty secrets!161

A detailed account of this notion of the method of the Qur"n goes

beyond the scope of the present study and will require a dedicated
and comprehensive study of both the Tafsr and Asrr al-tanzl, in
light of al-Rzs later theory of prophecy.
His statement at the end of Dhamm al-ladhdht, that the method
of the Qur"n is superior to the discursive method, and his citations
of Qur"nic verses that affirm seemingly irreconcilable aspects of
divinity, should be understood as stemming from his view that the
Qur"n is spiritually transformative. Rational theology, it seems, will
in many ways often conflict with this process. Though, in places,

161 Asrr al-tanzl, 3723; cf. 376. The same points, he goes on to add, are
highlighted in Qur. 3:1901.

a commentary on rislat dhamm al-ladhdht

203

al-Rz argues that this applies specifically to the level of average

believers, who constitute the public (jumhr), he often indicates that
the method of the Qur"n is effective at all stages of human development, including the most advanced. He writes in the Tafsr:
Sciences are either theoretical or practical. The most sublime and
complete among the theoretical sciences is the knowledge of Gods
essence, attributes, acts, judgements and names. You cannot find these
sciences more complete and sublime than in [the Qur"n]. The practical
sciences concern either bodily acts or acts of the heart, also known as
moral purity and the purification (tazkiya) of the self . You cannot
find these two sciences as you would in this Book.
God decreed that one who studies it and holds fast to it will gain
worldly honour and after-worldly happiness. I have transmitted various rational and transmitted sciences, but I have not gained as much
religious and worldly happiness from any of these sciences as I have
from serving this discipline [i.e. Qur"nic exegesis].162

Having found refuge in spiritual discipline and in the guidance and

inspiration of the Qur"n, al-Rz does not arrive at despair. What
Dhamm al-ladhdht expresses, rather, is a sense of utter alienation
(washa) in this world.
162

Tafsr, 13, 80.

204

chapter four

appendix

205

APPENDIX

RIS$LAT DHAMM LADHDH$T AL-DUNY$

The Manuscripts and the Critical Edition
I have been able to locate seven manuscripts of al-Rzs Rislat
Dhamm ladhdht al-duny, of which I have been able to have access
to five. Two others are currently inaccessible (assuming they have
survived contemporary devastation in Baghdad and Kabul). The
following are brief descriptions of the manuscripts used, along with
the abbreviations for those used in the edition.
1. Berlin State Library, Petermann II, 10 (pp. 24472):1
Title: Rislat Dhamm ladhdht al-duny.
Dimensions: 18.5 14.2; 16.6 11.8 cm. 1920 lines.
Dated: Friday, 12 Dh al-\ijja 610 A.H. (1214).
Copyist: ufayl Ibn Muahhar Ibn Ab Sa#d al-Fal al-afl.
Handwriting: naskh.
Abbreviation: .
Beginning:

...   .   
End:

 
  % ( ' 
-   
 @ AB
.C D  
A'
.D 

1 W. Ahlwardt, Die HandschriftenVerzeichnisse der Kniglichen Bibliothek

zu Berlin, XVI, p. 25, no. 5426. This is the only MS of this work listed by
Brockelmann (GAL I, 669).

206

appendix

2. Princeton University Library, Garret (Yahuda section)

308 (fol. 130b142a):2
Title: Kitb Taqr al-ladhdht.
Dimensions: 24 18; 17.8 11.8 cm. 23 lines.
Dated: 677 A.H. (12789).
Handwriting: naskh.
Abbreviation: L.
Beginning:

M P QD .@( 
 .   
T
U @ D %D
CD 
... Z . P .D \ D
End:

.  CD 
 D .@ P ...
.(
. @ AB   % a(

3. Mar#ash-Najaf Library 4416(3) (fol. 110b129b):3

Title: Al-Ladhdht.
Dimensions: 15 20.5; approx. 9.5 16.1 cm. 1922 lines.
Dated: 731 A.H. (1331).
Copyist: Najm al-Dn Ab Bakr Ibn #Abd al-Wahhb Dahistn.
Handwriting: naskh.
Abbreviation: .
Beginning:

 P c@ ' P  .   

 
g\  D


D@
L AB  
P D D
...   :
2
3

R. McChesney, Catalogue of Arabic Manuscripts, 213, no. 2486.

M. Mar#ash, Fihrist nuskheh-ye kha, 12, 17.

appendix

207

End:

C D 
 D .@ P ...
.
4. Mar#ash-Najaf Library 286(26) (fol. 247b255b):4
Title: Taqr al-ladhdht.
Dimensions: 18 27.5; 11 13.5 cm. 27 lines.
Dated: 1072 A.H. (1662).
Copyist: Shh Murd Farhn.5
Handwriting: nasta#lq.
Beginning:

M P QD .@( 
 .   
T
U @ D %D
CD 
... Z . P .D \ D
End:

.  CD 
 D .@ P ...
. @ AB   % a(
5. The British Library, I.O. (India Office Library) Islamic
3832 (fol. 85b95b):6
Title: Kitb Taqr al-ladhdht.
Dimensions: 27.7 14.8; 19.5 8.5 cm. 21 lines.
Dated: 1063 A.H. (1653).
Handwriting: exquisite nasta#lq.

M. Mar#ash, Fihrist nuskheh-ye kha, 1, 3223.

Dating and copyist: Mar#ash, Fihrist nuskheh-ye kha, 1, 333.
6 P. Stocks, Subject Guide, 217. The Library, I was informed, has no other record
of the MS (despite the reference in the Subject Guide to a handwritten entry). The
MS bears an India Office stamp dating to 1913.
5

208

appendix

Beginning:

D M % P P ( M AB( 
.%D %D %D %a mD P
M P QD .@( 
 .   
T
U @ D %D
CD 
... Z . P .D \ D
End:

oD 'p D D  D .@ P ...

 % a ( .  CD s
.QD AB
The two other manuscripts located are:
6. Al-Qdiriyya Library (Baghdad) 654(3) (fol. 1927), entitled Rislat Dhamm ladhdht al-duny.7
7. Ri"sat al-Mab#t Library (Kabul) 87, entitled Risla f
l-ladhdht al-malba f l-duny.8
For the edition, MSS 1, 2 and 3 are used. The variants among them
are numerous, but generally minor. MSS 4 and 5 appear to be copied,
or derived from, MS 2, and do not improve the text in any way. On
the other hand, MSS 1 and 3 have much in common, and probably
have a common source; yet neither could be derived from the other.
In this edition, the text is corrected in accordance with modern
conventions for spelling, without note of the original spelling in the
manuscripts. For example, is changed to ;C to w @ ;to
;@to ; and Q to Q. Nunation and some diacritical
marks have been added where needed.

7
8

I. Rauf, Al-$thr al-Khaiyya, 3, 134.

P. De Beaurecueil, Al-Makht al-#Arabiyya f Afghnistn, 20.

appendix

209

The Title
The epistle is given the following titles in manuscripts and later external
references:
Dhamm ladhdht al-duny (Censure of the Pleasures of This World).9
Risla f Dhamm al-duny (Censure of This World).10
Taqr al-ladhdht (Degrading Pleasures).11
Risla f l-ladhdht al-malba f l-duny (On the Pleasures Sought in
This World).12
5. Aqsm al-ladhdht (Divisions of Pleasure).13

1.
2.
3.
4.

The fourth and fifth titles appear to be taken directly from the epistles
introduction and describe its contents superficially. In my opinion, the
first title appears most authentic. It describes the works main parts
and conclusions, without being simply drawn from its introduction.
It also agrees with the authors description of the second section as
the section on the censure (dhamm) of imaginary pleasures, in his
final note in the work, preserved only in the Berlin MS. The second
title seems to be a shortened version of the first.
Finally, al-Rz refers to the work loosely both as a book (kitb)
and as an epistle (risla).14

9
10

The Berlin MS; and the Baghdad MS catalogue entry.

Al-Qif, Akhbr, 192; al-afad, Wf, 4, 255; Ibn Ab Uaybi#a, #Uyn, 2,

29.
11 The Princeton MS; the British Library MS; and the Mar#ash-Najaf 286 MS.
All, however, share a common source.
12 The Kabul MS catalogue entry.
13 Ibn Taymiyya (e.g. Dar", 1, 159) and Ibn Qayyim al-Jawziyya (Ijtim#, 120).
However, both almost certainly used the same manuscript.
14 Dhamm al-ladhdht, 3; 44.

210

appendix

211

appendix

@ '   %
 AB

appendix

212

  
@    o D o
  

D
 D

D D D P D ~ D .o

P~
 D o  D
D \ .D
D   .


 \ . DP \ g

@oP
 
D
P

@
@ . Q P  M%

\D @D
 T \D 
D @  U 
D.

D
P Z P  %  M

PD m D %aD %D %%

  @D .   M
D.%

p .
PD m \ : % M~  Q @P M Q  .

D  wDPD .  %
Do  %
D  % D wPD [P :] . %

: .
 : .

@( D . Q P M
" :L    .
 T\D 
 %DD @  U
 CD
 D".
. L
B  M :Qo @  .

213

appendix

D   % D w .

 M  (.

 .
 o  DP M m \ ~  %
PD U
\ g D
D~
\' D

 o %D  D '
D .'
D


 D % % \g
B@D

 D  %D o ~  w@
 .

( P :
D    @ 
D

D M :  %D [ :] m

D %a.%

D M :  %D  Q D'
D.
D m  % a
D M :  %D  Q D
D w m  % D.Q
D   mP  w@  .

"  % D" .
"  % D w PD  % D "%
.L
:  .
: L @.
 : .
.
@ :L .
.L

appendix

214



 :D  M  %
 m P ] [( :   @ M Q.
Ps o D - .Q

P o   \

P . (
 Do

'D m  m


@
o M    g\ w (D . m P (

\( \ .m m P
a
  ( D  
@ 
@  @ '
( ] [P :L o

 @  A
( o U @ P PD .@

o  @  U D aD Q MD .

 \ ' m P \o g @ 

@ .
D m P o L \ @


  D m
 M D ~C~
 wD

 :L  .
: .
 :L (.
: .
:L PD.
:L .U
: .
' :L ~.
.
. L
:L ~ M~  wD  .

215

appendix

oP %
"P @o M
C%
D . .:

D m
D
D ."%  A

M o
  
P .

 wD @DD . o
D  - o
 %

D  [P :] .%'
 m P m M  %
w U
U
 M w U D DPD .~o'
(D \ M
o D M @D

C U ] [ : ~ D . m o
 w D @D o M  % M  . D
D  .
  \ g  % @ M Q w Q
D .
D w Q D
D Qw D D .
PD  M m  m \ :D wD ww
 Do@P '\ MP a  D m oDw@

:L ~.
.
: .
.
: .m
D :L .
" w "Q .L
:L .
:L  : ..QwD
: .
: .

appendix

216

 % g\
%.
P \" :m a .

D ". m P @P g\ Q M
 mP
Q \ P  DD w 
  B.
D P  m \" : D  @D
D ."% P U
D%o
%

 P mD P
\ P ] [( :D % D %
  B.

D P  m \" :m@'  L ] [( :L@m

 M D %%   m P@' P

 D % B Q o
P  M~  %a
  B"D . P U
 P m P mD~ %
D%o \ m P  m@' @ BD
%a

@o M
. w

@

:L .
:L @.Q
 : D w.
:  .
.L
 :L  .
:L .
: .
:L .

"  "w . L

217

appendix

~  D w M w P D .

D\ %a P
D

 P P M
P ~  B D PD .  w 

%a
D [ :] %o
m P~  P M
P   D o D' M.
'\D m D mP   M (
P .s M m P m~  D'D.
D M - m P m P \ P mPD B
D .

 D M :  ( P .ms m P :~
P 
 %  M
\D %
  P
\o g  . :- P m m\D
%
\ % P ] [P :
Q a .
~ D~.
 D m
P 

 \ : @P  m P m P m PD U g\ U
  m'~ P D m ~ P m

.
 :L D .
M" "m .
D : .m
:L .
 :L  D.-
.
 : '.

appendix

218

'~
PD .
 P PD . U 
m P
\ g .P @P P ~  P PD o m

o o
P PD
 P ~ D %\ %
 .

 w @ Q
@D


M
P m P c

@ o ' m  m
'
(oD ' (

  w@ o' M ! P '\D UD U

' (  w o Q ! D ' \

@  M A D P '
 w
 w@ B o ] [P :L @ Q
D \
   .
  .@ @ 
 w@ B o '

D . (

D
( ' w
@ Q  Q
 '  w@ o
@.o
 \ o ~  DD'D .
P ~  %

 : '.
 P : .
: m  ...o m.
:  .
.M :
: .m
' :  .
.L
 : L .
 : @.M Q
@.
:

219

appendix

 % P M
%
 m P~   P
] [( : \ o  DD .  :'\ m
( 
o@P w  D @ '

@ D P ]\ [ : .%

  '\ Q@ o D Q @D
D . m \ P  m'~ P
m P a  L" :~ P Q c  Q D( D."
'D
 P ~ \ % o .
  M m( ' A~ m P  P
 Q BP wPD c   Q
D PPD c . m  Q  Q
D .
P mP c @ Qo
@D g\ U
@  g\

.L
' : .
:L ( .
:L .
.P:
' :L .
.
 :L D . :.
:  .m
 :L .
D :L  P.
" m  Q  Q D " .L
.
.L

appendix

220

@D m
\ m  Q  m
@ g\ m


 P .
@D  Qo

 \ g
D  D . @

 . P  \ gP w D Q

D  P PD  m B P '\D .  @  Q
 Q  \  M '\D .m
 M D @
   Q  M   M'\D .

][P :
D @  . m P
\P m DD  ' m D Q
Qo
  @ U c . .
  M mP m~  % 'D P

@ % @ . Q
\  Q

.  m m 
P  Q [( :L]
\ \' D MD .
' m P  M  
D D
@.
P M
w

 : .
" @" . L
: L .
: .
.m :L
:L .w
: mP@.
: .
: .
: .Q@D
 :L  %   @.Q
: .

221

appendix

D m P     mm
@D mP o ] g\ [ :'D .
  m PQ
  \ ' m  DP 
 % [email protected]

D w
PD m \ U '\ m D 

D(
 w  M D Q   Q

D  \ P w Q a D wD wo

 D %DD %

 M Q
 D woD w D .
M @
P
D .
D  m '\ m w ' m  w M m @.
D   \  w Q   .  m PQ
  \ P  w ][( :
 D m PD  Q  \  wQ
'  @ .

(  a ~ .

 : .
:L .
: .
" D  \" .
 M : .
: \.m
.
D"  m ... m@" .
B  M  :Qo@.
 :L  w.

appendix

222

 \  Q%
~  m P %~  %

 @'D .Q
 m P~  M Q   m PD \
  P ~ D Q o P U
P A 
P@D m P 
\D .  \ P .

 :
P
w c


mP

D(D

PD  Q P mP \ gD .m A

P P  w m ~  wD . m p
m P

 D ~ \D w m
 p m P D o  %o m
( D m  M Q mD~ 'D .w


P ]P [P :L w.m
 Qo m~  w


~ PD .w U~
\
\ m Q
\ Q
  %o M Q .w
D \ [ :] Q   D P@ U

: .
:L .
: .
 .
 : .
:L .
.L
B  M .mP :Qo
 :L  Q.
.

223

appendix

D '
 m P~  % \
@ Q .
~

 % @ Q 'D .
 P ] [P :
D
D %aD % .%o  m a \

D
\ g .' o  DD

D.%a
%

P
B\ '%
D
  m P \'  M   \ gD w@
\ PD   M Q

D w .' ( \ gD w@ o 
D Q o @\DD .  P \ Q \B
DP %
 .% . m P~  \ mP m
\ m
 % M  D @.Q
  M m ~  : w
 \
p  w ~  P . \ a Q 'D
\' \ D U a   M D ~DD

\ D w@BD  D . " :

 "D .  \ m P U P P DPD @ Q \o

 : @.Q
"  " .L
' : .
' : .
: D %a.%
.D :L
.L M
" ~  P \" .
:L .~D

appendix

224

D %D % D D D m BD U.
P \' \ M D P m  D(PD \ D
BD U
D mDo D mD P B QD .

 m Pp Q  Q m~  .

PD \ a Q D ~[( :] .
\ :D
  m P  P 

P P  D D'D .w
 \ ][ :
a m~ .%
 D m \ :-
  @

 L \ M D~ ] [( :L P D .L
\ \@  D \@  .
\
 P   .
\  w m P % D P 
'
Q a \ .

  M m ~  mP P :  Q
 Q D . \  M D gD

:L D ~D~.
D : .
.L
.L
.
" " .L
.
B  M .mP :Qo
: .Q
P : .

225

appendix

 ' .
 
   .'
B m P m ~ B Q m .
D  m L m \" :m @ g  Q P w
oDPD P M D ."w  m P w D

'  PD .  w

D c ' D  '\D .
  : Q Q a D~:
 m P D~  Q \ DP M    

D   .Q
D  .'

 [P :] m P -  ~  Q \  .' @
\ g wA  .Q
 D m P Q M Q Q .
m P~  Q Q a ~ .%

 m P D~  Q
PD  Q Q a D~ .
\ D M  D  PD .  ~

D : : ..
.L
.m :L
: .
.'D : L
: .
. L
: .%
.
DP : .
: .
 :L .

appendix

226


% PD .   ~ 

D  mD o w
D .m 

 D  Q . ] [ : 

 Dw
 D
\  

  m A
. ' m P  m .
PD    'D . m 
\  mm P 
D  w

D A D Q DP 
 . 
DD  %   Mw
D'[P :L] m

 D DP D D a P  D
DP
 . ' p m P  w Q ~
D p  w  .%
Q
PD U~  Q D . ] M [( :
\ g
PD .% U D - D' PD U PD .

 % D( D w.
m P  D M %~   %

: .
 : .
:L  .m
.
D :L  D D : .a P  D .a
D :L .
: .Q
: .Q
D :L .%

227

appendix

 .   ~ \ !% \ m
 D a  QD .a m P  P c
D D P .  BP
 M Q

.

' m P~  % Q \ g.%
  M mP m~  % Q a'D : mP

 % ( Q a D \
  P . m P m
  'D
 mP a

 m \ : m
p M % '\D
\ ' a U@ o M.
P U

\cP@ '\D .
\' @ cP U o @ ~C
 % D . 
\D .  p M
U U
Q
 m P
] @ D o [ : P %o  PD .

:L  .
.
D :L .
D : .
.L
: .m
" " . L
.
.
 M : .%
: .o
 " .  M @ "~C .L

appendix

228

BD D  '\  m P 
\ g .' @  B 'D c '\D . m

'~ D w .
 
D ] [P : A c Q C @D m
' @ %o  . w

\ m\D .@
P . m m U 
'  w
D @ o .
m\D .
  m P mPD mP U M  D D.
( o D ] [( :L.~ D

PD   \ \'
'D M a.a
 m P P    DD  PD .m
D Q a D .   m P M @o  D

.
.mP :
' : .
 : .%o
D : .
:L .
.
: L .m
.'D :L
 : L  D.
.L
" " B. M Q

229

appendix

\ g .o  m P P   \ @D
DD .% P.

  :-  M  %
D  Q D~

 mP D P
D  m P  M D .
D  aD . mP -
 

\
P M     
 .
~  D %
  M :D m P m (
D D~.

mP m  mPD a m
  m P :
P
~ @ D
D'D . m m m @o

P : .
D" P" .L
. L
: .
P : s.
 : L  a"D . "a .
P : .
. L
: .
: .
.

appendix

230
D

 D   .

 a

@D a
D
D
 a ] [( :

P mP m
  . m P
D M [ :] mPD ~a ~a mPm
 ~a

@ UD U '\D . m M 
 m \ '  PD .m
~ M m
\  \ MD '\D . m   m
 M  PD p  P  D(
\ M m oD \D '\D . m  m
M  ( M 
P  D \ M
' c 
oD M  Q m P
 D
'  .
D(  A
 MD  \ M m

 .
D m  m M 

: .
 :L .
.
.L
.L
D : L .
.
.L
 M  D a "".
.
 :L .

231

appendix

DD  m
D U(  B.
m" : @ m \" :  m M \
 m
 
'   m"D 

@D D M  "

 D m

(DP m \ : D m@ \ ] [P :L m
D QP P .s mP C ' 
.  P   DQ


D@ D P  L Q D  . mP [P :] -
 D g\ Q \
 .    D\ .Q \m

 P  \ M  P m PD
 M  \ ~a mP m D
] g\ [ :  D .  @~.

  M m D :  m P @ Q '

D m  U

 :L  D : .  D.
:L .
 : .
: .
 :L  D@.D
' : .
 :L (D : .(.
D : .-
.L
.L
.

appendix

232

@ a D@ Q ~ 
 D  

D w D
 ~    \ . \ m

D@ D .

  :  B  D@ Qm
\'
D Q DQo m

 D .' 'DD  .
 '\D ooB \D @ DP a PD U
U m ~ .
 m P M  D B
 P DB
D . m ~B@%
' A  D '

D
 D@P Q  m'~ P '\D .c m m
oB o P PDc  m  ' 
P A.c

] m P [( : M  \ P

 w
 D 'D . \ m m
 m   P m\D . B  m'~

" ~   " .
 :L .
.L
: .
 '\D" " .L
oB :L .o
D : .m
D : .
.' :
:L .

233

appendix

Pc m B Pc  m w '
P .c
D .
m P   P D
D m  @ Q    Q
 '\D .
  D mP  
 DP :
 \ g
C
@ cP

DP

C ~

'D . m
] [( :L  M m
@D .
 m mo U ~a ~D%

D(
m o mD U PD .U m   '
] [ :P . P m ~D % m m mo U~a

D ~ D % PD . Q @D P

(

:L
"  m w ' P "c .
D : .
: .
: .a

@
@ )o) (/ % m
 @(D . o
"(.
"@ cP
:L .
: L .a
: L .a
.L
: .

appendix

234

 g\ a D .~ 
\ ~  \'
  @  D
PD . m P~ 

P
m m m m U  o
 .D m P m moU
@

' B
@.
a ~D%

D(
PD m P m m o mD U P'D U m
D m
 ' QD @
@ Q U ] [P :  . m~  o

.  m P
D D' c
P DP . DB
D .
D D U ~a 
  M m  m \ : w P m
 MP D  B\ M  P D .
 a P  w U  \ M
P  .% m P
M 
' 

P w
D c \\ g \D D .
@
 M L
D  m P  m' A .   mPQ

.L
 . M
 :L .
.D : L
p  :.
p  :.
.
@ : .
: .
' :L ~.

235

appendix

 m M ~
P D B .Q

 \ M m m
  Q

P \\ g  m
.
  M m  ( m P \ m mPP m
DP ~a mD U DP ~a m
P .~a mm
P ~a P ~a D  .a

@ \ M  :] .
[ m' m  m D U

D  w\ M m B\D @ mD.
m\D ' m  m \B
 w
\ w ] [( :
P .ms ]P [P :L \P m
 ~'D .  w M %D o.

DP - \ m D ~ \P g   a

.L
D :L .
:  D .M "" \ \ g  M
.
" m
P \\ g  m" .L

:L ~ ( MD . L "(" "."%
.
: L D .m .
" mP m "~a .L
: L  D.
: .

appendix

236

 D U   D .

  mPDm
~ .

PD \' mD U ~a
 D  
D  D  '\D . 
D .
w m ~ \ M \D @ P P PD
'   A U'D .  m@ Q
 w M % U   D D @ Q
o w U  % Dm ' a U
' m  a U  M.

PD \' m DoP m ~a  \ D 

 m BD A c w D \ D a.
 w U  m w
 w\" : @

P
\ p D ."% w
 A 
 B.o

m P  m ~   D
mP

.L
mPD" "m .L
.m\ :
D : ~D.
.L
: .
: .a
.
.
: .
 :L .w

237

appendix

B DP DP UDD . m Pp  %

 % D .m m P~  Q

D  PD.
  m
  ] M [P : m ~  Q m P
 \ m mP M m M DP DP Q m U . mM m
D Q P QD m\D . M m MD aQ


P P m ~  a ] [ :
 D  D .w   m
 mP PD m P D \ U PD a
M   .   P PD 
  P DPD c PD . \ m m mU

: D :L . DD.
.L
.
: .
"  % D m mP~  Q " .L
.
.
.
"M" .
:L D.
.L
 : L .Q
.
.

appendix

238

D .~  Q D
 D Q m

'\D . m m P   . m P
 m
  D D.m
 ] [( :L m P@  m %
@
 \ m D P 

P PD c

D .   m \" :@ Q
 a \ \  m@ Q a @
@ Q


 a
D % PD . @ Qo

P
@ ".
Q o 

\'   : % ~  % \

 @
D  @ \ DP MD.

 m P ' ~  % \ DP M :] . D

: Q ~D :L .D QD .
: .
D : L .
:L : .
.
. L
.L
: @ .
@  :  . m
@  :  . m
 : .
 :L .
D"   @ " .L
: .
B P :L M Qo M .D

239

appendix

@ 
P D [( M D D w

'
 ' '\ D . ' D
p 
a U C w .   m P m
D %D@ '~ \ m M D . 
C UD D D .Qo '  D
D  Q . m P w \g
M  P
o .
 D   m P '\ m ( 
' \ mP g a U  D .  m P@ Q
 D
   D .% w  a  ] [ :m
 .%
\'   m \ : \'  \ gD QDo
D QwD~  QD P

 m P a
D D ~o P] .[ P
\ g

. L
.L
 :  '.
: L  p.
: '.
: .
: ~.
. L M
: .%
D :L ..Qo
. L
: .

appendix

240

 D ~D  D   M U@Q

 D  M m@ Q ~ D@ % QD ~
 D pD .Q  m~  . 
\ g D D w  m D oD .QD  %@Q
 '  . 
 M
~ '   D .
@ D .
  [P :] -D .  M
~

 LD oQD
D \ g  
DD QwD UP M @ B D w
\ @  ] [P :L . P M P 
P om
P M P 

D '\D .
 w ~  \  M a M DP .

(
   m P \ m mP (  D a
\  g\ aa \D mP  (  

D" D ~o P  D ~"
.L
D : .
.
.L
: L .
: .
 : .L .
\D :L .
.
.L
\ :  :L ..
.

241

appendix

 MD .
D D
D 
C % P .Qa 
CDP .% m P

mP  (
D @ U
 L  D a P~

 \ P @   mD .D m P
 '\ L  ~ PD~ D \
D
@D.
P L \ \ mD  D w@B D
D P \' a@ ] [ :D oD .~
Q .
( \  g\ a a
PD  D - mP

CP .% \ m P  g\ a
  ] [( : \

\ g mDo 'D . a U QD .Q  :

"DP  g\ aD m Z" D  mPDP A \ a

P ' M U .
Q a D'

\D \ a M

@ ' a U QD.Q
D

 : .
D :L .
: .
:L D.
:L .
:L .
D : L .
:L .
: .
.
 :L  aD : . \ a.

appendix

242

P \"D .  m P@ D A \

D " :
Qo
 a a oD U a

 . M
@D . @ % Qo ' a U M U .QD m P
 P - Z ( D %a  m
 @.

  M ~  Q D :  m P \ m mP
p  D \D mP DD . 
D D @  D @
@ P . B 
'D . B  m m U D   m
 o
. m @

.'
D  PD . [( :L] U  m m
   M  . \'
 M

P D [P :] L  P
 M m D Qac

  %D Q  m
m

  % o  \ g M  %Q
D   % .  \ a U
P . \'

:L .
. L
.
: ( .
: L .
.
 :L .
 :L .
 : .
: L .

243

appendix

 D M Q B U   . m P
 D .
 D Q  m ] [ :

P
PD  D -     P U

\ .% mP  B '\D .% m m

\ @   U '\D .% m  m 
\@  'D ( . a 
(.

 a P D~
\ \  m( \ 
( .( D D D DP w  \
 a P \'  m m o a
~a
m U .
\w
D  m PD m m  .m

.'  mD
p P PD

D m .' D (  ~ 

: L.
.L
 :L .%

D . 
" "m  .
 .
:L .
@ : .
: .a
:L .
 :L .
: '  m.
.'\ :

D  AP U .~D
D :~ .

appendix
@D

P PD .
(.
D m P

D  L U ]D [( : U

  D '\D . m   @ Q

 P( a
D D QDPD .  
D '\D . m p  @ Q 

A  .
DD .w Q  

 
  @( D P
 m
 D.

 D   m P \ m mP  M U D

D DP .

D D 
D D  w mDD
D % P .%  ] [P :L
D  P U  .w m   \ mP
\D mP P .~a  [ :] D
 m

 PD . w \ m ~a
 DD ww 'D

D : .
.
:L .
 : L .
" A " .
: D a.
. L M
: .
.
.mDo :
 :L @.

\ D~w

244

245

appendix

m
  P m D
PD .  D -  w D D  U
 w C 'D . \ g  D DD

D % m \ .% \ m m m

 D % m\D .%o@ m U m 
 DD .% @ cP @ 

@
D D~ " : ! \ m m\D [P :] %
".
  M m ~  mP Q :~  Q
M Q DP D  D .  mw
(   P  m' Pc
PD m mP m m@
 D
  @DD % %  U
'~

D : .m
D : .
: .
p : D .  : M(.
. L
B : M Qo' .
:L  : . P   D.
.L
 :L '~.
:L .
D : D.%
: .

appendix

246

m     D .  ' m P

m

'D . m
 D
 P m DD oD  %
' @
 %

D .

PD  D - mP :~  M Q a D
Q m\ : m  m P M
 m~
% '\ D
D
D

 %o P c~
P ~   P U  .
  M ~  mP Q :  \ m mPm
@ U  DP a MD .m
P  P .%
D 
D D mP@ m U : \ ] [ :m


'\ QP m  D D \@ Qo D . mo

 P \@ %o PD . @P %o

" : "[email protected]
 :L .
:L .m
D : L D o.
"  "%  :L . .%
. L
.L
:L .m
.L
.L
: .Qo@\ m
 : @.Qo

247

appendix

  ] [( :L % P %o .
 m P  [( :] %o P m
P    D  D  D w
 . P  P
 \ g
 

PD D %D % D PPD c PD . 

D - mP @ m U \ mP m

D \D U mP
mo UU Ds .m
  M P    Mm\ :

  \ mP m  DP  Ds  P .
@ m \ g  \ g
m
 
gDP PD  UD . ~  m\D .
\ \ M @ PD .Q   P U

 : @.%o
: L .
: .
.m :L
 : L .

.%
:L
: .m
: .m
: L .
.
: .
:L .
.

appendix

248

' m
 DD a P  D.


@ ~  P U .
D

  M D : P \
 \ ( DD .  m  

P
PD mDo D mm \ @ 
@ D D . mPD
P\" :

P PD L PD Ao -\D  @P mP

@P"a P m P
D @DB .m m P

( D
\ ~  .%


 D  M m ] [P : Q

D@'D :Q  m P \  Q D '\D .

 ] g\ [ : 'D .a \
D  m

m
 D .a
P .

  a

D \  D Qa


.
 :L .
: .@D
: .
.L
. L
.L
.
:L .m
 :L .
:L .

249

appendix

  D ~ ' D U . PD \ g

U DD A
 D a ! a

o '\ . \ g
o \ @ 
 '  @  M  P M
 .  ]@ [P :L DD \
 MD D
\D  (D 
 D M
 D.

  m P    \ mP m
 Q 
\D mP m U
MD .U
D
a
 D P \'  D ~

C P .% 
M D mDo D \ mD . \'
 @ ' U D D( .
PD  D - mP m U  Q 

D :L .
.L
D : .
D : L .
:L @.
 M :L D D D D(.
B DD . M Q
 : .D Q
.
:L ~ .D
D : .
: (.

appendix

250

P \' ~ D D PD . \ m\" :
 a [( :]

 "

. P  m M  m 
D@~ a

DQ @D m .   m P U
D
P P ( P P .
A p P  M
   m P : \ mP
 D P
D \D mP

  P . 
(   D .  m\" :  

 \P g . ] [ :@ ".

 D
PD   - P U
D  M  'D

 D ~
w @'D .Q P U P
.%

   m P  m P   -\" :

: .
:L .~D
.
.
: m  D.gDP Q
@. M Q
.L
D"
P P( P P" D .
.
.L
:L .
D : .L.

251

appendix

P p
 %"D .  m

 %! P 

.
D

 " : m P~  M D Q\ . \ m
 %
.  P QPD c
 
PPD c.

D @
.
D(  w 


 :
 M  %  

  m P \ \D D P .  
\   QD . [( :L] m
P
 Q   \ o
 .
~  [P :] Q gDPD .  P m P

.
.
: .
.L
:L .
.
:L  (  : .  (.
: .
D :L .

252

appendix

Q \

 
  c
@  .
PD   \ mP m  DP a P . 
 ( ~a D . m U ~a m

@
'  P D .a     % @Q
P : D DD %  D
@P M
 .
D :D D P D .  D A


@ ' (  !L
\ g  Q! D  A

@ '
D %  DP 
D

  m
DD .% A
D ~ \pD P [ :] . p D~.

]~ p[

 D

 DP w
\ m
 m \ \' o \
D
 \ w  D DP Q m
D

: L .
\ : .
B  : M Qo.A
.L
B : M Qo.
: L .D

253

appendix

 \D w D . 
' D @ P U \
D   '\D
 w
 m P '\ %
D P  U '\D . @Q
\ w
Z
  D
 PD U  .

'\ D   :
m U U m DD
PD [( :] mP B .m
 D mP m m P 

DP m P
 '.

" Q " M
'\ D   :~  w
P  m
mPD
DD D mP

P  " "U m
PD .  m P c A
  - DD U U m P~ w
%   D \
P M 
U  m P M .

\ w D P  U"

Z
" '\D  D
.L

 :L .
. L
: .L.
.D :L
:L  .
.
: L .
: .
 :L .A
: .

appendix

254

 

 Po  
D m P

\  m

 D m P ]:L
D .

D P U
 Po   D . mP m \" :
D Q
[P

D . \D .
 " 

mP\" :s D cQ \ \  m

 M
D Q %
 ."w
 


 m P [ :] A
  mP   U

 '\D .w m

 m a
  M '\D .% m
p .
'  " :
m" : \ 
M   .Q m m

.
]P [P : c
 w

 m\D . m

.L
: .
: .
\ : \ : .s.
 : .
:L .
.L
: '.
.L
: .

255

appendix

 

\  c  M   @

m a@~
 .o  :
D m

m P

.
 a \D .
 : m

D m .~o
C
'\D m a  DP m m C M
D .%  m
U  w .P
 m m@D .m  U UP sD . m P

P  .P m 
D

 .w@
 U.


 

m
\ \' D DD
'   o
'  o m\ U m@D U a mP

:L .
: .
P :  w .P
: .
P : s D P m P  .
 : L .
: .
. L

appendix

256

 D
C D
 m\D . m U \ g
m P
o
 .

]~ [

PD ~   : \ mP  MM DP %
 M DP % P .o  M %\ : ~ 

 D .  :~  \ w mP \
DP m\D . m
 ' 
@ Q 'D @ Q M ~a  P . ] [ :@
M
 D ' m P \ mP m
D \D mP

] [(: PD .m ] [( :L@  - ' m P  \

mP m
U \D % mP m
@ P D

@ P 
 \D mP m
 .

.
C :L D.
" " B.L M Q
.L
.
.
: .m
.
.L
" D  \D mP m
@ P" .L

257

appendix

D D   m 
 P : @

 mP D :
mP \ g
P  . %

"~  D  oD  " .: -

"\ m\D D  oD \ \  m o .gDP ~
 D  ".\" :  \  m P
\ g D."
P D @ D - m P
U %
P  \D mP
. \ mP mP B@ 
 ~\ : g  D Mo
.B
 D 
 %   D . -  ' \ P

D .
PD @  D m P
@D P m


D D  D Q o   p
P \ . P D  D : P  c w 
D % '   
P 
\ g.

.D :L
 : .
:L .
D :L .-
D : .
.L
B .D :L M Qo
:L .
.
D :  .

appendix

258

@D -

P D " : \P ms

' ".

D P - " [P :] : \
 \ .g
 D .g D  m mP
P  DP ' ] [ :  
( .
M D a .

 D ".
D

    D mP
D  M 
 .
D %  m


D
  m P :   a
 m U .
D  a a
D
 .a
 a U

  @D
.a 
 

. Qa
PD @  D m P  \ U
@ 

 D o [P :L] @ Q  \D @ Q

: '.A
: .
D : .
.
 : D %a.%
.
:L .
: .
.L
.

259

appendix

@ \ D w a
 MD  @

 D Q D D :
P U D .
 : D P D

"\  o D~ DD~D D " D

 DD . :-
D %.
P D

P D . M
~  ' \
D
 D.
D D  P P ~ PD
P @
 DD .D \ M   mP" :~  ]:
[(
" .  Z
D
 P . P m .m
D 
P  m :m~  
D
P .  mP m  BP

 M\ B mP m\D D .  MB

B .mP : M Qo
.
. L
 :L .A
.L
 : .
: L .
D" " B. M Q
: .
: .m
mP" "m .L
.DP :
 :L .D

appendix

260

@D 
D BPD . o

.
PD   \ m D~ .P mP D 

Qa ]DD [ :  o P m DD .m P -
P   Do a
D m .
D P m P  \ gw
D m

% D @ M
D .~ ~

.
PD m P
@ Q  D @ Q

@  M  MD .
D . 

PD  D -  m P  M
 % 
@  M .
D P @ DP a

@
P @ \  D  M 

 M .

PD   D \ % @  A D

 MD B P.U
 :
 :L .
: L .
D" P mP  \ g w
D m "
.L
" " : .L .
:L .
 .
 .
D P.
 :  DP a
 .

261

appendix

w  m P

 D

D m P ] [P :
P M m  PD D M  ~
D

 @ 

 M  .
D D %
' MD
.

PD   D ]\ [( :L % @  A
o@PD w P U D
 

D.%

 D ~  %@ mP m

D @D
  g DP mP B :
D   D % mP
 w
 \ g  D  PU

.

P ~  % D  % 
D P

DD M
D  % D%

 m P   m PD  M

( m


 :L
D.
 .
.
.L
:L .
.L
.m :L
.DP : 
.
.L
.M :L

appendix

262

U D ( '\D . m
.
D m
D D (

][

D ~ ] [ : @ % %

'\D

 \
D % PD .Q   % \g
D
.~ (  : 
( D
 D !D  ! D  
] [(:
:
 MD !m 
P  

D@PD D M
D


@ DoD
D @P
D


 D : D.
.
D :L .
.L
D :L (.
D" " .L
: .
:L .
 .



PD

DD
c'P
D o
p
D
mP c
p Do U BD


D
@  

263

appendix

D -P  M~ D
  M P@

 M (  mC Dm
(
@P
~ 

D

D  P  D %@


 M .P
@(   M a
D o
 M "  PD "wD " "w

 P"PD . M P" %  "c

D"
".

D" @ m " D\"


  "
PD m P M [P :L] P
"MD .

 D P
" P

.
"D .   m

PD D o ] [P : -\ : D
D DP

m   

" "D .L
.D :L
B : M Qo.g
 .
 @. c
.
 .
.
 . w
 :L .
 . w
o :L D D .
. L
.D :

appendix

264

~ D . m D
 \ g

'@
D -  m
'@ Q

 % .

@ M Q \ . M
 %D 

D  U .
 % D
\D M
 D w
(oP ~  m \'  ][ :
P

D "

P" :m
D
D PD .  - w
(oP p m  % %


\  g\


\ g 
 

.(oP
w

- P@ \ P
@(  -\ !Q

D m
D D D . Z@D
 .
P U~D P g\ A P .
( DD o D o PD 
D   P \ PD  %
P \ P mP M Z
P mPD % Z

:L D.
: .
 :L .
:L .
:  L \.
 : .%
@ :m a. a
 :L .
.L
D :L .
.

265

appendix
D

P ] [( : o
@ Z  
D @P !
 DP DP g\ ( '   % M Q.Q

D \C g~ M Q@ @o M BC M m  m

D D D m @P D.

.
@D : .

 CD  .
 :  @ D :L . 
 D C  .
D :  
" @P ... D"  M.

266

bibliography

bibliography

267

BIBLIOGRAPHY
1. Al-Rzs Works
(a) Published Works
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.

Akm al-basmala, M. Ibrhm, Cairo, 1988.

Al-Arba#n f ul al-dn, Hyderabad, 1353 A.H.
Ass al-taqds, Cairo, 1328 A.H.
Asrr al-tanzl wa-anwr al-ta"wl, ed. M. Muammad et al., Baghdad, 1990.
Al-Barhn dar #ilm-i kalm, 2 vols., ed. M. Sabzawr, Tehran, 1962 (Persian).
Dhamm ladhdht al-duny (See Appendix).
Al-Firsa, ed. Youssef Mourad, La physiognomonie arabe et le Kitb al-firsa de Fakhr
al-Dn al-Rz, Paris, 1939.
#Imat al-anbiy", Beirut, 1988.
I#tiqdt firaq al-Muslimn wa-l-mushrikn, ed. . Sa#d et al., Cairo, 1978.
Al-Jabr wa-l-qadar, published as vol. 9 of al-Malib al-#liya (page numbers in
citations refer to the listed Malib edition).
Jmi# al-#ulm, Bombay, 1908 (Persian).
Al-Risla al-Kamliyya f l-aq"iq al-ilhiyya, ed. S. Sabzawr, Tehran, 1956
(Persian).
Al-Risla al-Kamliyya f l-aq"iq al-ilhiyya, ed. A. Muy al-Dn, Beirut, 2002
(Arabic).
Dar Khud shins, in Chahrdah risleh ..., ed. S. Sabzawr, Tehran, 1961, 1750
(Persian).
Al-Kshif #an ul al-dal"il wa-ful al-#ilal, ed A. al-Saqq, Beirut, 1992.
Lawmi# al-bayyint f l-asm" wa-l-ift, ed. . Sa#d, Cairo, 1396 A.H.
Lubb al-Ishrt, ed. #A. #Aiyya, Cairo, 1355 A.H.
Ma#lim ul al-dn, ed. . Sa#d, Cairo, n.d.
Al-Ma#lim f ul al-fiqh, ed. #A. #Awa and #A. #Abd al-Mawjd, Cairo, 1994.
Al-Mabith al-mashriqiyya, 2 vols., Tehran, 1966.
Al-Mal f ul al-fiqh, 6 vols., ed. . al-#Alwn, Riyadh, 1981.
Manqib al-Imm al-Shfi#, ed. A. al-Saqq, Cairo, 1986.
Al-Mas"il al-khamsn f ul al-dn, ed. A. al-Saqq, Beirut, 1990.
Al-Malib al-#liya min al-#ilm al-ilh, 9 vols., ed. A. al-Saqq, Beirut, 1987,
(except vol. 9: see al-Jabr wa-l-qadar, above).
Muaal Afkr al-mutaqaddimn wa-l-muta"akhkhirn min al-#ulam" wa-l-ukam" wal-mutakallimn, ed. H. Atay, Cairo, 1991.
Munart f Bild m War" al-Nahr, see Kholeif, Controversies.
Al-Nafs wa-l-r wa-shar quwhum, ed. M. Ma#m, Islamabad, 1968.
Dar Nafy-e makn va jihat, in Chahrdah risleh ..., ed. S. Sabzawr, Tehran, 1961,
216 (Persian).
Nihyat al-jz f diryat al-i#jz, ed. B. Amn, Beirut, 1985.
Sayr-e nafs, in Chahrdah risleh ..., ed. S. Sabzawr, Tehran, 1961, 98105
(Persian).
Shar al-Ishrt, 2 vols., Cairo, 1325 A.H.
Shar #Uyn al-ikma, 3 vols., ed. A. al-Saqq, Cairo, n.d.
Al-Sirr al-maktm f asrr al-nujm, Cairo, n.d.
Al-Tafsr al-kabr, 32 vols., Cairo, 1938.

268
35.
36.
37.
38.

bibliography
Tal al-aqq, in Chahrdah risleh ..., ed. S. Sabzawr, Tehran, 1961, 5178
(Persian).
Risla f l-Tanbh #al ba# al-asrr al-mda#a f ba# suwar al-Qur"n al-#am, ed.
B. Drtm, Amman, 2004.
Al-Waiyya, see al-Zarkn, Fakhr al-Dn al-Rz, 63843.
Ziyrat al-qubr [published as Falsafa-i Ziyrat], ed. A. #$bid, Qom, 1998.

(b) Manuscripts
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.

Al-$yt al-bayyint f l-maniq, Ms. Or. 2917, Leiden University Library.

\ikmat al-mawt, Ms. 4821, Aya Sofia Library, Istanbul (Persian).
Al-Ishra f #ilm al-kalm, Ms. 519(2), Kprl Library, Istanbul.
Al-Khalq wa-l-ba#th, Ms. 816, Kprl Library, Istanbul.
Al-Mulakhkha f l-ikma wa-l-maniq, Ms. Hunt 329, Bodleian Library, Oxford.
Risla f l-Nafs, Ms. 9508(1) (fol. 1b8b), Mar#ash-Najaf Library, Qom.
Al-Nafs wa-l-r, Ms. Hunt 534 (fol. 257a295b), Bodleian Library, Oxford.
[Unless otherwise stated, citations are from the printed edition].
Nihyat al-#uql f diryat al-ul, Ms. 1874, Ahmet III Library, Istanbul.
Shar al-Ishrt [the logic section], Ms. Or. 746(10), Cambridge University
Library.
Shar al-kulliyt min kitb al-Qnn, Ms. Seldon A. 64, Bodleian Library,
Oxford.
Risla f l-Tawd, Ms. 5426, Fti Library, Istanbul (Persian).
[Kitb f Ul al-dn #aq"id Ahl al-Sunna], Ms. 297:R27kA, American University
of Beirut.

2. Other Primary Sources

#Abduh, Muammad. \shiya... #al Shar... al-Jall al-Dawwn #al l-#Aq"id al-#Audiyya,
Cairo, 1322 A.H.
#Abd al-Jabbr Ibn Amad al-Asadabd. Al-Mughn f abwb al-tawd wa-l-#adl, 10
vols., ed. I. Madkour et al., Cairo, 196068.
#Abd al-Jabbr, (pseudo-). see Ibn Mnakdm.
Ab l-Barakt al-Baghdd, Hibat Allh. Kitb al-Mu#tabar, Hyderabad, 1358 A.H.
Ab Rashd, Sa#d Ibn Muammad al-Nsbr. Al-Mas"il f l-khilf bayn al-Bariyyn
wa-l-Baghddiyyn, ed. M. Ziyda, Tripoli, 1979.
Ab Shma, Shihb al-Dn #Abd al-Ramn Ibn Ism#l. Al-Dhayl #al l-rawatayn, ed.
M. al-Kawthar, Cairo, 1947.
$mid, Sayf al-Dn #Al Ibn Ab #Al al-. Al-Ikm f ul al-akm, ed. I. al-#Ajz, 4 vols.,
Beirut, n.d.
Aristotle. Al-Akhlq, trans. Isq Ibn \unayn, ed. #A. Badaw, Kuwait, 1979.
Metaphysics, in The Complete Works of Aristotle, trans. and ed. J. Barnes, Princeton,
1984.
Ash#ar, Ab l-\asan #Al Ibn Ism#l al-. Al-Ibna #an ul al-diyna, ed. F. Mamd,
Cairo, 1977.
Kitb al-Luma# f l-radd #al ahl al-zaygh wa-l-bida#, ed. R. McCarthy, Beirut, 1952.
Maqlt al-Islmiyyn wa-ikhtilf al-mualln, ed. H. Ritter, Wiesbaden, 1980.
#Asqaln, Shihb al-Dn Amad Ibn \ajar al-. Lisn al-mzn, Hyderabad, 1330
A.H.
Baghdd, #Abd al-Qhir Ibn hir al-. Ul al-dn, Beirut, 1980.
Al-Farq bayna al-firaq, Cairo, 1910.

bibliography

269

Bqilln, Muammad Ibn al-ayyib al-. Al-Inf fm yajibu i#tiqduh wa-l yajzu al-jahl
bih, ed. I. \aydar, Beirut, 1986.
Tamhd al-aw"il wa-talkh al-dal"il, ed. R. McCarthy, Beirut, 1957.
Bar, Ab l-\usayn Muammad Ibn #Al al-. Al-Mu#tamad f ul al-fiqh, 2 vols., ed.
M. \amdullh et al., Damascus, 1964.
Busraw, Muaf Ibn Khall Khwja Zde al-. Tahfut al-falsifa, Cairo, 1302 A.H.
Dhahab, Muammad Ibn Amad al-. Trkh al-Islm wa-wafayt al-mashhr wa-l-a#lm,
52 vols., ed. #U. Tadmur, Beirut, 1997.
Frb, Ab Nar Muammad Ibn Muammad al-. $r" ahl al-madna al-fila, ed.
A. Ndir, Beirut, 1968.
Kitb al-Burhn, ed. M. Fakhry, Beirut, 1987.
Galen. On the Doctrines of Hippocrates and Plato, 3 vols., ed. and trans. P. de Lacy, Berlin,
1980.
Ghazl, Muammad Ibn Muammad al-. Iy" #ulm al-dn, 4 vols., Beirut, n.d.
Al-Iqtid f l-i#tiqd, ed. \. Atay et al., Ankara, 1962.
Al-Risla al-Laduniyya, in Majm#at ras"il al-Imm al-Ghazl, Beirut, 1994, 3, 5774.
Mi#yr al-#ilm, ed. S. Duny, Cairo, 1961.
Al-Munqidh min al-all, ed. F. Jabr, Beirut, 1959.
Al-Mustaf min #ilm al-ul, 2 vols., Cairo, 1322 A.H.
Al-Risla al-Qudsiyya, ed. A. Tibawi, in The Islamic Quarterly, 9, 1965, 7994.
Tahfut al-falsifa, ed. S. Duny, Cairo, 1957.
Hind, #Al al-Mattaq #Al" al-Dn al-. Kanz al-#umml, Hyderabad, 1958.
Ibn #Abd al-Hd, Muammad Ibn Amad. Al-#Uqd al-durriyya min manqib Shaykh alIslm Ibn Taymiyya, Beirut, 1974.
Ibn #Abd Rabbih, Amad Ibn Muammad. Al-#Iqd al-fard, ed. A. Amn, 7 vols., Cairo,
1940.
Ibn Ab Uaybi#a, Muwaffaq al-Dn Amad Ibn Qsim. #Uyn al-anb" f abaqt alaibb", 3 vols., Cairo, 1981.
Ibn al-Akfn, Mamd Ibn Ibrhm, Irshd al-qid il asn al-maqid, ed. J. Witkam,
Leiden, 1989.
Ibn #Arafa al-Wargham, Muammad. (Bb al-Imma min kitb) al-Mukhtaar al-shmil,
ed. S. Ghurb, in \awliyyt al-Jmi#a al-Tnisiyya, 9, 1972, 177234.
Ibn al-Athr, #Al Ibn Muammad. Al-Kmil f l-trkh, 13 vols., Beirut, 1966.
Ibn Frak, Muammad Ibn al-\asan. Early Islamic Theological and Juristic Terminology:
Kitb al-udd fi l-ul, ed. M. A. S. Abdel Haleem, London, 1991.
Mujarrad maqlt al-Shaykh Ab l-\asan al-Ash#ar, ed. D. Gimaret, Beirut, 1987.
Ibn al-#Ibr, Gregorios Ab l-Faraj. Trkh mukhtaar al-duwal, ed. A. ln, Beirut,
1958.
Ibn al-#Imd, Shihb al-Dn #Abd al-\ayy Ibn Amad. Shadhart al-dhahab f akhbr man
dhahab, ed. M. al-Arn", Damascus, 1991.
Ibn Kathr, Ism#l Ibn #Umar. Al-Bidya wa-l-nihya, 14 vols., Beirut, 1977.
Ibn Khallikn, Amad Ibn Muammad. Wafayt al-a#yn wa-anb" abn" al-zamn,
6 vols., ed. M. #Abd al-\amd, Cairo, 1948.
Ibn al-Malim, Mamd Ibn Muammad. Al-Mu#tamad f ul al-dn, ed.
M. McDermott and W. Madelung, London, 1991.
Ibn al-Murta, Amad Ibn Yay. Al-Munya wa-l-amal, ed. I. Muammad #Al,
Alexandria, 1985.
Ibn Qayyim al-Jawziyya, Muammad Ibn Ab Bakr. Ijtim# al-juysh al-islmiyya,
Amritsar, 1314 A.H.
Ibn Rushd, Muammad Ibn Amad. Fal al-maql fm bayn al-ikma wa-l-shar#a min
al-ittil, ed. M. #Amra, Beirut, 1981.
Tafsr m ba#d al-ab#a, 3 vols., ed. M. Bouyges, Beirut, 1942.

270

bibliography

Ibn al-S#, #Al Ibn Anjab. Al-Jmi# al-mukhtaar, Baghdad, 1934.

Ibn Sn, Ab #Al \usayn. Al-Aawiyya f l-ma#d, ed. \. #$, Beirut, 1987.
Al-Adwiya al-qalbiyya, in Min Mu"allaft Ibn Sn al-ibbiyya, ed. M. al-Bb, Aleppo,
1984, 20794.
Awl al-Nafs, ed. A. al-Ahwn, Cairo, 1952.
Al-Ishra il fasd #ilm akm al-nujm, ed. Y. Michot in Avicenne, Rfutation de lastrologie,
Paris, 2006.
Al-Ishrt wa-l-tanbht, ed. S. Duny, 3 vols., Cairo, 19478.
Ithbt al-nubuwwt, ed. M. Marmura, Beirut, 1968.
Al-Karmt wa-l-mu#jizt wa-l-a#jb, in \. #$, Al-Tafsr al-Qur"n (see below), 224
40.
Al-Mubatht, ed. #A. Badaw, in Aris #inda al-#Arab, Cairo, 1947, 122239.
Al-Najt, ed. M. Fakhry, Beirut, 1985.
Al-Qnn f l-ibb, 3 vols., Beirut, n.d.
Risla f #Ilm al-akhlq, in Tis# Ras"il f l-\ikma w-l-ab#iyyt, Bombay, 1318 A.H.
Rislat al-Qa", in Y. Michot, Ibn Sn, (see below), 1037.
Risla f l-Qa" wa-l-qadar, in Jmi# al-Bad"i#, Cairo, 1917, 4368.
Risla f l-sa#da wa-l-ujaj al-#ashara #al anna al-nafs al-insniyya jawhar, ed. A. al#Alaw, Hyderabad, 1353 A.H.
Al-Shif", Al-Ilhiyyt, 2 vols., ed. G. Anawati et al., Cairo, 1960.
Al-Shif", Al-Maniq, 9 vols., ed. G. Anawati et al., Cairo, 195270.
Al-Shif", Al-ab#iyyt, 8 vols., ed. S. Zyid et al., Cairo, 195283.
Al-Ta#lqt, ed. #A. Badaw, Cairo, 1973.
Ibn Taymiyya, Amad Ibn #Abd al-\alm. Bayn talbs al-Jahmiyya f ta"ss bida#ihim
al-kalmiyya, 2 vols., ed. M. al-Qsim, Makka, 1392 A.H.
Dar" ta#ru al-#aql wa-l-naql, ed. M. Slim, 11 vols., Riyadh, 1979.
Rislat al-Itijj bi-l-qadar, in Majm#at al-ras"il al-kubr, Cairo, 1966, 99155.
Majm# fatw Shaykh al-Islm Amad Ibn Taymiyya, 37 vols., ed. #A. Qsim, Riyadh,
1991.
Minhj al-sunna al-nabawiyya f naq kalm al-sh#a wa-l-qadariyya, Cairo, 1321 A.H.
Al-Nubuwwt, ed. A. uwayyn, Riyadh, 2000.
^j, #Au al-Dn #Abd al-Ramn Ibn Amad al-. Al-Mawqif, 8 vols., Cairo, 1325
A.H.
Ikhwn al-af. Ras"il, 4 vols., ed. Kh. al-Zirikl, Cairo, 1928.
Ifahn, Ab #Abd-Allh al-. Al-Kshif #an al-mal f #ilm al-ul, 6 vols., ed. #A. #Abd
al-Mawjd, Beirut, 1998.
Ji, Ab #Uthmn Ibn Bar al-. Kitb \ujaj al-nubuwwa, in Ras"il al-Ji; al-ras"il
al-kalmiyya, ed. #A. Ab Mulim, Beirut, 1987, 12555.
Kitb Istiqq al-imma, in Ras"il al-Ji; al-ras"il al-kalmiyya, ed. #A. Ab Mulim,
Beirut, 1987, 17798.
Risla f in#at al-kalm, in Ras"il al-Ji; al-ras"il al-kalmiyya, ed. #A. Ab Mulim,
Beirut, 1987, 538.
Jayln, #Abd al-Qdir Ibn Ms al-. Al-Fat al-rabbn wa-l-fay al-ramn, Beirut,
1980.
Jurjn, #Al Ibn Muammad al-. Al-Ta#rft, Beirut, n.d.
Juwayn, #Abd al-Malik al-. Al-#Aqda al-Nimiyya, ed. M. al-Kawthar, Cairo, 1948.
Al-Burhn f ul al-fiqh, 2 vols., ed. #A. al-Db, Cairo, 1400 A.H.
Al-Irshd, ed. A. Tamm, Beirut, 1996.
Al-Shmil f ul al-dn, ed. R. Frank, Tehran, 1981.
Khwansr, Muammad Bqir al-. Rawt al-jannt f awl al-#ulam" wa-l-sdt, 8 vols.,
Tehran, 1928.

bibliography

271

Kiy al-Harrs, #Al Ibn Muammad al-. Ul al-Dn, Ms. 290 Kalm, Dr al-Kutub
al-Miriyya, Cairo.
Ma#arr, Ab l-#Al" al-. Dwn Saq al-Zand, ed. U. al-abb#, Beirut, 1998.
Majr, (Pseudo-) al-. Ghyat al-akm wa-aaqq al-natjatayn bi-l-taqdm, ed. H. Ritter,
Leipzig, 1933.
Picatrix Das Ziel Des Weisen Von Pseudo-Mar, trans. H. Ritter and M. Plessner,
London, 1962.
Maimonides, Moses. The Guide of the Perplexed, 2 vols., trans. S. Pines, Chicago, 1974.
Mward, #Al Ibn Muammad al-. Adab al-duny wa-l-dn, ed. A. Ab Zayna, Cairo,
1979.
A#lm al-nubuwwa, ed. M. al-Baghdd, Beirut, 1987.
Miskawayh, Amad Ibn Muammad. (Min risla ...) f l-ladhdht wa-l-lm, in #A.
Badaw, Dirst wa-nus f l-falsafa wa-l-#ulm #inda l-#Arab, Beirut, 1981, 98104.
Tahdhb al-akhlq, ed. C. Zurayq, Beirut, 1967.
Murta, #Al Ibn al-\usayn al-Sharf al-. Al-Dhakhra f #ilm al-kalm, ed. A. al\usayn, Qom, 1411 A.H.
Mutawall, #Abd al-Ramn Ibn Ma"mn al-. Al-Mughn, ed. M. Bernand, Cairo,
1986.
Plato. The Republic, trans. D. Lee, London, 1987.
Ptolemy. Almagest, trans., G. J. Toomer, New York, 1984.
Qif, Jaml al-Dn Ab l-\asan al-. Ikhbr al-#ulam" bi akhbr al-ukam", Cairo, 1326
A.H.
Ktib elebi, Muaf Ibn #Abd Allh. Kashf al-unn #an asm al-kutub wa-l-funn, ed.
S. Yaltkaya et al., Istanbul, 1941.
Rghib al-Ifahn, al-\usayn Ibn Muammad al-. Al-Dhar#a il makrim al-shar#a, ed.
A. al-#Ajam, Cairo, 1985.
Al-I#tiqdt, ed. Sh. Al-#Ijl, Beirut, 1988.
Rz, Ab Bakr Muammad Ibn Zakariyy al-. [Ras"il falsafiyya], ed. P Kraus, Cairo,
1939.
Al-Shukk #al Jlns, ed. M. Mohaghegh, Tehran, 1413 A.H.
Al-ibb al-rn, ed. #A. Muammad, Cairo, 1978.
afad, Khall Ibn Aybak al-. Al-Wf bi-l-wafayt, 29 vols., ed. S. Dedering, Damascus,
1959.
San", Majdd Ibn $dam al-. \adqat al-aqqa wa-shar#at al-tarqa, ed. M. Razav,
Tehran, 1368 A.H.
The First Book of the \adqat al-\aqqat or the Enclosed Garden of the Truth, trans. M.
Stephenson, Calcutta, 1910.
Sextus Empiricus, Outlines of Scepticism, trans. J. Annas and J. Barnes, Cambridge,
2000.
Shahrastn, #Abd al-Karm al-. Al-Milal wa-l-nial, ed. M. al-Kln, 2 vols., Cairo,
1961.
Kitb al-Mura#a, ed. and trans. T. Mayer and W. Madelung, London, 2001.
Nihyat al-iqdm f #ilm al-kalm, ed. A. Guillaume, London, 1934.
Shahrazr, Muammad Ibn Mamd al-. Nuzhat al-arw wa-rawat al-afr, ed. Kh.
Amad, Hyderabad, 1976.
Shrz, Muammad Ibn Ibrhm adr al-Dn al-. Al-\ikma al-muta#liya, 9 vols.,
Beirut, 1990.
Subk, Tj al-Dn #Abd al-Wahhb Ibn #Al al-. abaqt al-Shfi#iyya al-kubr, 10 vols.,
ed. M. al-an et al., Cairo, 1965.
Suhraward, Yay Ibn \abash al-. Haykil al-nr, ed. M. Ab Rayyn, Cairo, 1957.
Taftazn, Mas#d Ibn #Umar Sa#d al-Dn al-. Shar al-Maqid, 5 vols., ed. A. #Umayra,
Beirut and Cairo, 1989.

272

bibliography

Tahnaw, Muammad #Al Ibn #Al al-. Kashshf iilt al-funn, 4 vols., Beirut,
1998.
kprzde, Amad Ibn Muaf. Mift al-sa#da wa-mib al-siyda, 4 vols., ed. K.
Bakr et al., Cairo, 1968.
Tawd, Ab \ayyn #Al Ibn Muammad al-. Al-Hawmil wa-l-shawmil, ed. A.
Amn et al., Cairo, 1951.
Al-Imt# wa-l-mu"nasa, 3 vols., ed. A. Amn et al., Cairo, 1953.
f, Sulaymn Ibn #Abd al-Qaw al-. Dar" al-qawl al-qab bi-l-tasn wa-l-taqb, ed. A.
Shihadeh, Riyadh, 2005.
Al-Ta#yn f shar al-Arba#n, ed. A. #Uthmn, Beirut, 1998.
s, Muammad Ibn Muammad al-. Shar al-Ishrt, printed with Fakhr al-Dn alRzs Shar al-Ishrt.
Talkh al-muaal, ed. A. Nrn, Beirut, 1985.
Warrq, Mamd Ibn al-\asan al-. Dwn al-Warrq, ed. W. Qab, Ajman, 1991.
Yfi#, #Abd Allh Ibn As#ad al-. Mir"t al-jinn wa-#ibrat al-yaqn, 4 vols., Hyderabad,
1338 A.H.
Zamakhshar, Mamd Ibn #Umar al-. Al-Minhj f ul al-dn, ed. S. Schmidtke,
Stuttgart, 1997.
3. Modern Works
Ahlwardt, Wilhelm. Die HandschriftenVerzeichnisse der Kniglichen Bibliothek
zu Berlin, XVI, in Verzeichniss der Arabischen Handschriften, V, Berlin, 1893.
Anawati, Georges C. Fakhr al-Dn al-Rz; Elments de biographie, in Melanges H.
Masse, Tehran, 1963, 110.
Fakhr al-Dn al-Rz; tamhd li-dirsat aytih wa-mu"allaftih, in Melanges h \usain,
ed. #A. Badaw, Cairo, 1962, 193234.
Barnes, Jonathan. Some Ways of Scepticism, in Epistemology, ed. S. Everson,
Cambridge, 1990, 20424.
Bell, Joseph N. Love Theory in Later \anbalite Islam, Albany, 1979.
Black, Deborah L. Practical Wisdom, Moral Virtue, and Theoretical Knowledge:
The Problem of the Autonomy of the Practical Realm in Arabic Philosophy, in
Les philosophies morales et politiques au Moyen ge, ed. B. Bazn et al., Ottawa, 1995,
45164.
Brink, David O. Moral Realism and the Foundations of Ethics, Cambridge, 1989.
Brockelmann, Carl. Geschichte der arabischen Litteratur, 5 vols., Leiden, 19371949.
Budr, Salmn al-. Naariyyat Ibn Sn f l-fi#l, in Dirst, 15, 1988, 6273.
Calder, Norman. The Barhima: Literary Construct and Historical Reality, in
Bulletin of the School of Oriental and African Studies, 57(1), 1994, 4051.
Ceylan, Yasin. Theology and Tafsir in the Major Works of Fakhr al-Dn al-Rz, Kuala
Lampur, 1996.
Cook, Michael. Commanding Right and Forbidding Wrong in Islamic Thought, Cambridge,
2001.
Davidson, Herbert A. Alfarabi, Avicenna, and Averroes, on Intellect, Oxford, 1992.
De Beaurecueil, P. S. L. Al-Makht al-#Arabiyya f Afghnistn, in Majallat Ma#had alMakht al-#Arabiyya, 2(1), 1956, 333.
Encyclopedia of Ethics, 2 vols., Chicago and London, 1992. [EE].
Encyclopaedia of Islam (The), New Edition, 12 vols., Leiden, 19602004. [EI2].
Encyclopdia Iranica, 13 vols. to date, London, 1982.
Ess, Josef van. Die Erkenntnislehre des #Audaddin al-^c, Wiesbaden, 1966.
The Logical Structure of Islamic Theology, in Logic in Classical Islamic Culture;
Giorgio Levi Della Vida Conferences, ed. G. Grunebaum, Los Angeles, 1967, 2150.

bibliography

273

Skepticism in Islamic Religious Thought, in Al-Abth, 21(1), 1968, 118.

Fakhry, Majid. Ethical Theories in Islam, Leiden, 1991.
Frank, Richard M. The Autonomy of the Human Agent in the Teaching of #Abd
al-@abbr, in Le Museon, 95(4), 1982, 32355.
Creation and the Cosmic System: al-Ghazl and Avicenna, Heidelberg, 1992.
Moral Obligation in Classical Muslim Theology, The Journal of Religious Ethics,
2(2), 1983, 204223.
Frankena, William K. Ethics, New Jersey, 1973.
Galston, Miriam. Politics and Excellence; The Political Philosophy of Alfarabi, Princeton,
1990.
Realism and Idealism in Avicennas Political Philosophy, in The Review of Politics,
41, 1979, 56177.
Gardet, Louis, and Georges C. Anawati. Introduction la thologie musulmane: essai de
thologie compare, Paris, 1948.
Gianotti, Timothy J. Al-Ghazls Unspeakable Doctrine of the Soul, Leiden, 2001.
Gimaret, Daniel. Thories de lacte humain en thologie musulmane, Paris, 1982.
Goodman, Lenn E. Rzs Psychology, in The Philosophical Forum, 4.i (New Series),
1972, 2648.
Griffel, Frank. Al-[azls Concept of Prophecy: the Introduction of Avicennan
Psychology into Aarite Theology, in Arabic Sciences and Philosophy, 14, 2004,
101144.
Gutas, Dimitri. Intuition and Thinking: The Evolving Structure of Avicennas
Epistemology, in Aspects of Avicenna, ed. R. Wisnovsky, Princeton, 2001, 138.
Pre-Plotinian Philosophy in Arabic (Other than Platonism and Aristotelianism):
A Review of the Sources, in Aufstieg und Niedergang der Rmischen Welt, II.36.7, ed.
W. Haase, Berlin, 1994, 493973.
Heemskerk, Margaretha. Suffering in the Mu#tazilite Theology; #Abd al-Jabbrs Teaching on
Pain and Divine Justice, Leiden, 2000.
Heinen, Anton. Mutakallimn and Mathematicians; Traces of a controversy with
Lasting Consequences, in Der Islam, 55, 1978, 5773.
Hogendijk, Jan P. Greek and Arabic Constructions of the Regular Heptagon, Archive
for History of Exact Sciences, 30, 1984, 197330.
Horten, Max. Moral Philosophers in Islam, trans. J. Hager, in Islamic Studies, 13(1),
1974, 123.
Hourani, George F. Ghazl on the Ethics of Action, in Journal of the American Oriental
Society, 96, 1976, 6988.
Islamic Rationalism; the Ethics of #Abd al-Jabbr, Oxford, 1971.
Juwaynis Criticism of Mu#tazilite Ethics, in Muslim World, 65, 1975, 161173.
Reason and Tradition in Islamic Ethics, Cambridge, 1985.
Hurka, Thomas. Perfectionism, Oxford, 1993.
Inati, Shams C. The Problem of Evil: Ibn Sns Theodicy, New York, 2000.
`skenderolu, Muammer. Fakhr al-Dn al-Rz and Thomas Aquinas on the Question of the
Eternity of the World, Leiden, 2002.
Jackson, Sherman A. The Alchemy of Domination? Some Ash#arite Responses
to Mu#tazilite Ethics, in The International Journal of Middle Eastern Studies, 31,
1999, 185201.
Jomier, Jacques. Les Maft al-Ghayb de lImm Fakhr al-Dn al-Rz, in Mlanges
de lInstitut Dominicain dEtudes Orientales, 13, 1977, 253290.
Qui a Comment lEnsemble des Sourates al-#Ankabt Ysn (2936) dans Le
Tafsr al-Kabr de lImm Fakhr al-Dn al-Rz?, in International Journal of Middle
East Studies, 11, 1980, 467485.
The Qur"nic Commentary of Imm Fakhr al-Dn al-Rz: Its Sources and Its Originality,

274

bibliography

in International Congress for the Study of the Quran, Canberra, 1980, 93111.
Kagan, Shelly. Normative Ethics, Boulder, 1998.
Kamali, Mohammad Hashim. Principles of Islamic Jurisprudence, Cambridge, 1997.
Kholeif, Fathallah. A Study on Fakhr al-Dn al-Rz and His Controversies in Transoxiana,
Beirut, 1966.
Lagarde, Michel. Index du Grand Commentaire de Fakhr al-Dn al-Rz, Leiden, 1996.
Lambton, Ann K. S. State and Government in Medieval Islam, Oxford, 1991.
Leaman, Oliver. Abd [sic.] al-Jabbar and the Concept of Uselessness, Journal of
the History of Ideas, 41(1), 12931.
Macdonald, Duncan B. The Development of the Idea of Spirit in Islam, Muslim
World, 22, 1932, 2542; 15368.
Madelung, Wilferd. The Late Mu#tazila and Determinism: The Philosophers
Trap, in Yd-Nma: In Memoria di Alessandro Bausani, ed. S. Amoretti, Rome,
1991, vol.1, 24557.
Mar#ash, Mamd. Fihrist nuskheh-ye kha Kitbkhane #Umm \azrat $yat Allh al#Um Mar#ash Najaf, Qom, 1975.
Marmura, Michael E. Avicennas Psychological Proof of Prophecy, in Journal of Near
Eastern Studies, 1963, 22(1), 4656.
Avicennas Theory of Prophecy in the Light of Ash#arite Theology, in The Seed of
Wisdom, ed. W. McCullough, Toronto, 1964, 159178.
Al-Ghazl on Ethical Premises, in The Philosophical Forum, 1.iii (New Series),
1969, 393403.
Fakhr al-Dn al-Rzs Critique of an Avicennan Tanbh, in Historia Philosophiae
Medii Aevi II, ed. B. Mojsisch et al., Amsterdam, 1991, 627637.
The Islamic Philosophers Conception of Islam, in Islams Understanding of Itself;
Eighth Giorgio Levi Della Vida Biennial Conference, ed. R. Hovinnisian, California,
1981, 87102.
A Medieval Islamic Argument for the Intrinsic Value of the Moral Act, in Corolla
Torontonensis; Studies in Honour of Ronald Morton Smith, ed. Emmet Robbins et al.,
Toronto, 1994, 11331.
McChesney, Robert. Catalogue of Arabic Manuscripts (Yahuda Section) in the Garret Collection,
Princeton, 1977.
McDermott, Martin J. Abul-\usayn al-Bar on Gods Volition, in Culture and
Memory in Medieval Islam: Essays in Honour of Wilferd Madelung, ed. F. Daftary et al.,
London, 2003.
Michot, Jean R. Ibn Sn. Lettre au Vizir Ab Sa#d, Beirut, 2000.
La Destine de lhomme selon Avicenne: le retour Dieu (ma#d) et limagination, Louvain,
1986.
A Mamlk Theologians Commentary on Avicennas Risla Aawiyya, being a
translation of part of Dar" al-Ta#ru of Ibn Taymiyya, with introduction, annotation
and appendices, Part I, Journal of Islamic Studies, 14(2), 2003, 149203.
Vanits intellectuelles... Limpasse des rationalismes selon le rejet de la contradiction dIbn Taymiyya, in Oriente Moderno, 19, 2000, 597617.
Mohaghegh, Mahdi. Flsf-i-Rayy Mohammad Ibn-i-Zakary-i Rz, Tehran, 1970.
Nasr, Seyyed H. Fakhr al-Dn al-Rz, in A History of Muslim Philosophy, ed. M. Sharif,
2 vols., Wiesbaden, 1963, 1, 64256.
Ormsby, Eric L. Theodicy in Islamic Thought; The Dispute over al-Ghazalis Best of All
Possible Worlds, Princeton, 1984.
Ragep, F. Jamil. Freeing Astronomy from Philosophy: an Aspect of Islamic Influence
on Science, in Osiris, 2001, 16, 4971.
Rahman, Fazlur. Avicennas Psychology. An English Translation of Kitb Al-Najt, Book II,
Chapter VI, Oxford, 1952.

bibliography

275

Rashed, Marwan. Ab Bakr Al-Rz et le Kalm, in Mlanges de lInstitut Dominicain

dtudes Orientales du Caire, 24, 2000, 3954.
Rashed, Roshdi. Al-Qh vs. Aristotle: On Motion, in Arabic Sciences and Philosophy,
9, 1999, 724.
Ra"f, #Imd. Al-$thr al-Khaiyya f l-Maktaba al-Qdiriyya f Jmi# al-Shaykh #Abd al-Qdir
al-Jln bi-Baghdd, Baghdad, 1974.
Reinhart, A. Kevin. Before Revelation; The Boundaries of Muslim Moral Thought, New York,
1995.
Rooney, Paul. Divine Command Morality, Aldershot, 1996.
Rosenthal, Franz. Knowledge Triumphant; The Concept of Knowledge in Medieval Islam,
Leiden, 1970.
Routledge Encyclopedia of Philosophy, 10 vols., London, 1998. [REP].
Shihadeh, Ayman. From al-Ghazl to al-Rz; 6th/12th Century Developments
in Muslim Philosophical Theology, in Arabic Sciences and Philosophy, 15(1), 2005,
14179.
Review of: Muammer `skenderolu, Fakhr al-Dn al-Rz (see above), in Journal of
Islamic Studies, 15(2), 2004, 2135.
Suffering, Medieval Islamic Civilization: An Encyclopedia, 2 vols., ed. J. Meri, New
York, 2005.
Stocks, Peter. Subject Guide to the Arabic Manuscripts in the British Library, ed. C. Baker,
London, 2001.
Stough, Charlotte. Greek Skepticism, Berkeley, 1969.
Street, Tony. Concerning the Life and Works of Fakhr al-Dn al-Rz, in Islam:
Essays on Scripture, Thought and Society; A Festschrift in Honour of Anthony H. Johns, ed.
Peter Riddell et al., Leiden, 1997, 135146.
Medieval Islamic Doctrine on the Angels: the Writings of Fakhr al-Dn al-Rz,
in Parergon, 9(2), 1991, 11127.
Striker, Gisela. Sceptical Strategies, in Doubt and Dogmatism; Studies in Hellenistic
Espistemology, ed. M. Schofield et al., Oxford, 1980.
Watt, W. Montgomery. The Faith and Practice of al-Ghazl, London, 1953.
Wolfson, Harry A. The Philosophy of the Kalam, Cambridge, Massachusetts, 1976.
Al-Zarkn, Muammad. Fakhr al-Dn al-Rz wa-r"uh al-kalmiyya wa-l-falsafiyya,
Beirut, 1963.

276

bibliography

index of names

277

INDEX OF NAMES
#Abd al-Jabbr 21 n. 34, 23, 24 n. 46,
25, 27, 28, 33, 4749, 75 n. 113, 77
n. 123, 93 n. 172
#Abduh, Muammad 44
Abraham 137 n. 128
Ab l-Barakt al-Baghdd 8, 117 n. 36,
118, 199
Ab l-Hayam 117 n. 36
Ab Lahab 103
Adam 45 n. 2
Akhbrs 117 n. 36
#Al Ibn Ab lib 135
Aristotle 19 n. 25, 118, 125, 198
Al-Aamm, Ab Bakr 178
Al-Ash#ar, Ab l-\asan 18, 19, 20, 83
Ash#ars; Ash#arism 24, 6, 7, 8,
1315, 1719, 21, 35 n. 94, 3941,
4347, 4957, 61, 64, 65, 66, 68,
73, 75, 7879, 8285, 87, 88, 91,
94, 96, 97, 101, 104107, 109, 110,
117 n. 36, 129132, 141, 142, 150
Al-Bhil, Ab l-\asan 4 n. 6
Al-Bqilln 7, 15, 39, 40, 66, 104, 107,
114
Al-Bar, Ab l-\usayn 2, 16, 17, 25,
26, 35, 36, 9194, 178, 179
Barhima 133
Carneades 198
Christians 117 n. 36
Al-Dabs, Ab Zayd 117 n. 36
Al-Dawwn 164 n. 34
Determinists 38, 96
4irr Ibn #Amr 38
4iy" al-Dn al-Rz 4
Dualists 185
Euclid

191

Falsifa 1, 16, 36, 42, 44, 54, 61, 79

n. 129, 91, 96, 97 n. 188, 101, 109,
110, 114, 117 n. 36, 118, 127, 129
n. 95, 131, 137, 157, 159, 162, 165,
185, 195, 199

Al-Frb 125, 130, 143 n. 146, 152,

179, 180
Gabriel 152 n. 177
Galen 198 n. 147
Al-Ghazl 1, 2, 4, 37 n. 104, 45 n. 3,
51 n. 21, 54, 57, 61, 65, 66, 68, 69,
71, 72, 73 n. 103, 74, 76, 82, 85, 88,
89 n. 162, 112, 134, 136, 137, 138,
179, 185, 195
Ghiyth al-Dn al-Ghr 4
Ghrids 45
Al-\alm, Ab #Abdullh
n. 36
\anafs 9, 74 n. 105, 150
\anbals 5
Al-Harrs, al-Kiy 52, 53
isbniyya 182

114, 117

Ibn al-Akfn 2
Ibn Frak 15, 34, 50, 51, 53
Ibn al-Malim 16, 2528, 49, 93
Ibn al-Murta 25 n. 50
Ibn al-Nu#mn, Muammad 117 n. 36
Ibn Qayyim al-Jawziyya 209 n. 13
Ibn Rushd 19 n. 25
Ibn Sn 2 n. 1, 8, 9, 11, 19, 21, 27,
42, 58, 61, 8890, 97 n. 188, 101,
110, 112, 117 n. 36, 118, 120, 122,
124127, 130132, 137141, 143 n.
146, 148, 151, 159, 161164, 166, 168,
179, 180
Ibn Taymiyya 3637 n. 99, 109, 199,
201, 209 n. 13
Ikhwn al-af" 130, 142
Al-Isfar"n, Ab Isq 4 n. 6, 7, 15,
16, 51, 104, 105
Ism#ls 138, 139
Israelites 135 n. 119
Al-Ji 134, 136, 138, 141, 178179,
197
Jahm Ibn afwn 38
Jahms 36 n. 99

278

index of names

Al-Jayln, #Abd al-Qdir 151 n. 175

Al-Jl, Majd al-Dn 4
Al-Jubb", Ab #Al 101
Al-Jubb", Ab Hshim 24 n. 46, 75,
192
Al-Anr, Ab Qsim 4 n. 6
Al-Juwayn 7, 16, 39, 40, 53, 54, 104
n. 208, 105107, 117 n. 36
Al-Ka#b, Ab l-Qsim 47, 102, 103,
178
Karrmiyya 5, 96, 117 n. 36
Khawrij 178
Al-Khayy, Ab l-\usayn 178, 179
n. 81
Al-Ma#arr, Ab l-#Al" 6
Maimonides 160
Mlik Ibn Anas 71
Mturds 46
Megarians 19 n. 25
Miskawayh 2 n. 1, 117 n. 36
Mu#ammar Ibn #Abbd 117 n. 36
Mutakallimn 1, 3, 14, 42, 45, 46, 54,
73, 95, 109, 110, 113, 114, 149, 150,
159, 197, 199
Mu#tazila; Mu#tazilism 2, 5, 1416, 17
18, 21, 24 n. 46, 2528, 29, 3338,
40, 41, 44, 45 n. 2, 46, 4750, 5256,
5861, 64, 68, 7377, 79, 80, 8399,
101106, 114, 117 n. 36, 131, 136, 141,
150, 164, 171, 178180, 185, 192 n.
132
Mu#tazila, Baghdd 35, 47, 48, 85,
87, 102
Mu#tazila, Baran 25, 48, 49, 85, 87
Al-Najjr, \usayn 38
Al-Nam 117 n. 36
Al-Nawbakht 117 n. 36
Al-Naysbr, Muammad Ibn Yay 4
Neoplatonism 160168, 178 n. 78
New Academy 198

Philoponos, John
Plato 118, 159
Ptolemy 197

197

Qadars 37 n. 99
Quraysh 181
Qub al-Dn al-Shrz

197 n. 146

Al-Rghib al-Ifahn 54, 117 n. 36

Al-Rz, Ab Bakr 8, 101, 102, 103,
118, 122, 158, 160, 161, 165, 167169,
185
Sabianism 137 n. 128
adr al-Dn al-Shrz 164 n. 34
Al-afad 10
San" 190
Satan 45 n. 2
Sextus Empiricus 184
Al-Shfi# 9
Shfi#s 9, 74 n. 105, 150
Al-Shahrastn 15, 40, 56, 114 n. 26,
137139
Al-Sharf al-Murta 178 n. 79
Shihb al-Dn al-Ghr 4
Sh#s; Imms; Twelvers 114, 117 n. 36,
139, 178, 180
Al-Simnn, Ab Ja#far 7
Al-Simnn, al-Kaml 4
Sufis, Sufism 54, 110, 113, 116, 117
n. 36, 124, 127, 138, 139, 152
Al-Suhraward 118 n. 47
Sunnis 2, 18, 35, 36, 37, 114, 116, 117
n. 36, 129 n. 95, 178, 180, 181
Al-Tahnaw 193
Al-Tawd, Ab \ayyn 197198
n. 147
Tekesh, #Al" al-Dn 4
Thbit Ibn Qurra 117 n. 36
Zayds

178

index of subjects

279

INDEX OF SUBJECTS

absolutism, ethical 4748, 8485

advantage; benefit (malaa, manfa#a) 21
7, 35, 489, 5152, 54, 55, 5761, 72,
7683, 91, 9495, 97103, 112, 156,
162, 168, 171, 177
action, divine 3, 3436, 45, 46, 50, 96
98, 100
afterlife 33, 6667, 74, 110, 148, 150,
167
al-amr bi-l-ma#rf wa-l-nahy #an al-munkar
73, 75
analogical argument (qiys) 34, 35, 68
72, 98101, 103, 106, 133
angels 114116
atomism 14, 19, 183
benefit (see advantage)
causality 21, 3943
character 1, 23, 46, 113, 115, 116153,
1556
consequentialism 2, 46 n. 5, 56, 62, 63,
80, 91, 113, 153, 179
d# (see motivation)
deontological ethics 47 n. 5, 4849, 55
56, 68, 73
destiny 10, 16, 17, 2940, 43, 119, 145
determinism (see destiny)
disadvantage; harm (mafsada, maarra) 21
5, 489, 5152, 5762, 6975, 94, 120,
156, 161162, 165, 172178
divine command ethics (voluntarism) 2,
47 n. 5, 4953, 56, 6373, 83, 96, 98,
110, 132, 150
emotivism 55, 59, 78
epistemology, ethical 8796
happiness 125128
harm (see disadvantage)
hedonism 60, 122
iba (see permissibility, Legal)
#illa (Legal ground; see analogical argument)

irda (see will)

istiqq (desert) 63, 66, 7475, 9293
istib al-l 71
isti#a (capacity; see power)
kasb (acquisition)

13, 14 n. 3, 3941

language, ethical 501, 612

love of God 113114, 127128
lying 53, 6061, 75, 8587, 90, 104
105, 107
maqid al-shar#a (objectives of the Law)
6872, 179
malaa (see advantage)
malaa mursala (unsupported interest)
6972, 179
malaat al-#lam 7982
metaethics 2, 3, 45, 50, 53, 60, 62, 69,
73, 111, 152
miracles 105107, 129130
motivation 1530, 3234, 3638, 41
43, 56, 76, 78, 97101, 111, 121, 156
munsaba (convenience; see analogical argument)
normative ethics 2, 2 n. 2, 45, 49, 53,
55, 62, 68, 69, 73 n. 103, 152
objectives of the Law (see maqid al-shar#a)
objectivism, ethical (see realism, ethical)
obligating what is above capacity (taklf
m l yuq) 103104
obligation 13, 28, 39, 48, 53, 56, 6567,
7383
occasionalism 13, 19, 132
original state of human nature 170175
pain 234, 489, 5152, 54, 5863, 66
67, 69, 78, 8182, 9497, 121123,
156170
perfectionism 2, 467 n. 5, 57, 63, 109
163
permissibility, Legal (iba) 56, 702

280

index of subjects

pleasure 234, 489, 5863, 97, 110

114, 120127, 155172
power (qudra) 136, 179, 43
principles of jurisprudence (see ul al-fiqh)
prophecy 105107, 114115, 129153,
180, 201
qudra (see power)
Qur"n 145, 146, 148, 187, 189, 200
203
rationalism, ethical (tasn al-#aql wa-taqbuh) 47, 50, 589, 83, 96, 133
realism, ethical; objectivism, ethical
4750, 52, 5556, 59, 61, 7576, 83
96, 168
Revealed Law (shar#, shar#a) 5256, 58,
6373, 74, 75, 83, 98101, 113, 148
149, 152153, 177,
rhetoric 6162, 143145, 158
riqqa jinsiyya (generic sympathy) 52, 55,
76, 78, 80 n. 133, 82 n. 138

soul, human 31, 109112, 116128,

135, 142, 190
arqat al-Qur"n (method of the Qur"n;
see Qur"n)
teleological ethics 467, 536
theodicy 102103, 160169
ul al-fiqh

1, 3, 45, 6373, 98101, 152

vain action (#abath) 28, 55, 71, 73, 9798,

146
value, ethical 4648, 54, 5664, 8586,
9194
voluntarism (see divine command ethics)
will (irda) 2026, 3031, 346, 38, 113
115
wujb (see obligation)