mindoftheancients.com€¦ · self-awareness in islamic philosophy...

SELF-AWARENESS IN ISLAMIC PHILOSOPHY

This important book investigates the emergence and development of adistinct concept of self-awareness in post-classical, pre-modern Islamicphilosophy. Jari Kaukua presents the first extended analysis ofAvicenna’s arguments on self-awareness – including the flying man,the argument from the unity of experience, the argument againstreflection models of self-awareness and the argument from personalidentity – claiming that all these arguments hinge on a clearlydefinable concept of self-awareness as pure first-personality. Hesubstantiates his interpretation with an analysis of Suhrawardī’s useof Avicenna’s concept and Mullā S· adrā’s revision of the underlyingconcept of selfhood. The study explores evidence for a sustained,pre-modern and non-Western discussion of selfhood andself-awareness, challenging the idea that these concepts are distinctlymodern, European concerns. The book will be of interest to a range ofreaders in history of philosophy, history of ideas, Islamic studies andphilosophy of mind.

jari kaukua is Academy of Finland Research Fellow in theDepartment of Social Sciences and Philosophy at the University ofJyväskylä. He is the author of several articles in journals includingVivarium and History and Theory. This is his first book.

SELF-AWARENESS INISLAMIC PHILOSOPHY

Avicenna and Beyond

JARI KAUKUA

University Printing House, Cambridge cb2 8bs, United Kingdom

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit ofeducation, learning and research at the highest international levels of excellence.

www.cambridge.orgInformation on this title: www.cambridge.org/9781107088795

© Jari Kaukua 2015

This publication is in copyright. Subject to statutory exceptionand to the provisions of relevant collective licensing agreements,no reproduction of any part may take place without the written

permission of Cambridge University Press.

First published 2015

Printed in the United Kingdom by Clays, St Ives plc

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication dataKaukua, Jari.

Self-awareness in Islamic philosophy : Avicenna and beyond / Jari Kaukua.pages cm

isbn 978-1-107-08879-5 (hardback)1. Self (Philosophy) 2. Self-consciousness (Awareness) 3. Islamic philosophy. I. Title.

b745.s35k38 2014

126.0880297–dc232014023812

isbn 978-1-107-08879-5 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy ofurls for external or third-party internet websites referred to in this publication,and does not guarantee that any content on such websites is, or will remain,

accurate or appropriate.

http://www.cambridge.org

http://www.cambridge.org/9781107088795

For Ukko, Touko and Rauha

Contents

Acknowledgements page ix

Introduction 1

1 Preliminary observations: self-cognition and Avicennianpsychology 12

1.1 Self-cognition in the ancient heritage 121.2 Avicennian psychology in outline 22

2 Avicenna and the phenomenon of self-awareness: theexperiential basis of the flying man 30

2.1 The purpose and basis of the flying man 312.2 The validity and plausibility of the flying man 37

3 Self-awareness as existence: Avicenna on the individualityof an incorporeal substance 43

3.1 The problem of incorporeal individuality 433.2 Self-awareness as incorporeal existence 51

4 In the first person: Avicenna’s concept of self-awarenessreconstructed 62

4.1 Three Avicennian arguments from first-personality 644.2 First-personality, the flying man and incorporeal existence 804.3 Self-awareness, reflection and intellection 89

5 Self-awareness without substance: from Abū al-Barakātal-Baghdādī to Suhrawardī 104

5.1 Avicennian material in Suhrawardī 1065.2 Substanceless self-awareness 114

6 Self-awareness, presence, appearance: the ishrāqī context 124

6.1 Self-awareness and knowledge as presence 125

vii

6.2 Self-awareness and being as appearance 1426.3 Degrees of self-awareness 154

7 Mullā S· adrā on self-awareness 161

7.1 Four Avicennian arguments 1647.2 The complicated evidence of self-awareness 181

8 The self reconsidered: S· adrian revisions to the Avicennianconcept 192

8.1 The self and cognitive unity 1928.2 Identity in substantial change 208

Conclusion: Who is the I? 228

Appendix: Arabic terminology related to self-awareness 233Bibliography 238Index 254

viii Contents

Acknowledgements

This book is the distillation of the research conducted during my post-doctoral period, but some of its central insights were already formed duringmy doctoral studies. I therefore owe an immense debt of gratitude to mysupervisors, the late Juha Sihvola, Mikko Yrjönsuuri and Taneli Kukkonen.The extremely conscientious and insightful comments of Jon McGinnisand Simo Knuuttila provided crucial corroboration and realignment at aformative stage. Finally, Peter Adamson not only added his characteristicallypenetrating points but also was pivotal for my full engagement with post-Avicennian philosophy by invitingme to sharemy hesitant first reflections onSuhrawardī at a conference held in London in February 2008.Financially, my research has been enabled by generous support from the

Academy of Finland (through research fellowships under the titles ‘SelfhoodinMedieval Islamic Philosophy’ and ‘Knowledge in Post-Avicennian IslamicPhilosophy,’ and through the two Centers of Excellence led by SimoKnuuttila, ‘History of Mind’ and ‘Philosophical Psychology, Morality andPolitics’), the European Research Council (through Taneli Kukkonen’sresearch project on ‘Subjectivity and Selfhood in the Arabic and LatinTraditions’) and the Riksbankens Jubileumsfond (through the project‘Understanding Agency’ led by Lilli Alanen and Pauliina Remes). I havehad the pleasure of conducting the research at the inspiring environments ofthe Department of Social Sciences and Philosophy at the University ofJyväskylä, my academic home and the Department of Philosophy at theUniversity of Uppsala. During the final revision, I enjoyed the hospitality ofthe SwedishCollegium for Advanced Study inUppsala.My gratitude goes tothe staff and the directors of these organizations, but especially to all the dearcolleagues who have shared their critical insights at the various stages offormation of my ideas, in particular to Vili Lähteenmäki, Juhana Toivanen,Mikko Yrjönsuuri, Taneli Kukkonen, Miira Tuominen, Timothy Riggs andTomas Ekenberg. I have also benefited enormously from the comments ofcolleagues who listened to my talks at various seminars and conferences.

ix

Particularly cherished have been the critical yet encouraging remarksI received during my visit to Iran from ‘Alī ‘Abidi Shahrūdi, Sa‘īd JavadiAmoli and ‘Abd al-Rasūl ‘Ubūdīyat, and I would like to extend my gratitudeto Yasser Pouresmail and Mohammad Javad Esmaeili for organizing thesetalks. Finally, I am heavily indebted to my editor Hilary Gaskin as well as thetwo anonymous referees for Cambridge University Press, whose combinedacumen helped me to improve the book quite considerably.

The librarians at the University of Jyväskylä and the University of Uppsalahave been invaluably efficient in tracing down my frequently obscurerequests. This book would not have been possible without their help. Bythe same token, I would like to recognize Sajjad Rizvi’s and Vasileios Syros’collegial assistance with some particularly unobtainable texts. Deborah Black,Therese Scarpelli Cory, Jules Janssens and Luis Xavier López-Farjeat havekindly shared their own work, even in pre-publication form, for which I ammost grateful. Jessica Slattery and Tim Riggs did a great job in polishing myEnglish; I hold exclusive rights to the inelegancies that remain. Finally, I’mgrateful for Ville Suomalainen’s skilled and reliable preparation of the index.

Most of all, I am grateful for the unfailing love and support of my wifeLotta, who alone has had to bear with the more insecure stages of the book’slabour. I dedicate the finished work to our children.

x Acknowledgements

Introduction

The Western world revolves around the self. A sure sign of this is theproliferation of various neologisms in, for instance, folk psychological,alternative therapeutic or economic parlance. We are all familiar withvarious self-help programmes, self-counselling sessions, prospects of self-development, self-transcendence or self-realization, the conscientious con-sumer’s need of occasional self-compassion, and the rational economicman’s guiding principle of self-interest. This general cultural trend has itsparallels in philosophy and various other disciplines in the humanities andthe social sciences. For the past decades, self-consciousness or self-awarenesshas been a constant concern of philosophers of mind, with the fact of first-personal, self-aware qualitative experience presenting arguably the mostobstinate obstacle for the naturalist explanation of all and everything.Questions of perceived and constructed identity, or identities, have gener-ated a thriving academic industry, with no recession in the foreseeablefuture. Indeed, modernity and post-modernity are often defined preciselyby means of the novel notions of selfhood or individual identity (or thedissolution thereof) to which these epochs are alleged witnesses.As a result of the sustained interest in selfhood, the term ‘self’, as well as

the related psychological terms such as ‘self-awareness’ or ‘self-consciousness’, is a nodal point of both complementary and conflictingintuitions, interests and convictions. It is therefore not a surprise that theterm is extremely ambiguous, and that there are in fact a number of more orless distinct concepts of self; a recent enumeration of variants in thephilosophical scene alone finds no less than thirty-two different epithetsused to characterize the self.1 These concepts range from extremely narrownotions of subjectivity as a structural feature of all experience to consider-ably more complex concepts of the self as a narrative or socially constructedentity; some are motivated by epistemological interests while others emerge

1 Strawson 2009, 18.

1

from research in genetic psychology, sociology or anthropology. On theother hand, extended cases have been made for the thesis that a coherentnaturalistic ontology can do without anything like the self, which at best isan arguably useful psychological or cultural fiction, but more often ahopelessly entangled web of linguistic and conceptual confusions.2

Such heated activity about the self places the historian of ideas, partic-ularly one working with a period and cultural context far removed from ourown, face to face with a set of thorny questions. These arise first of all fromthe ambiguity of the term ‘self’ and the corresponding vagueness of theconcept of self. Which of the many alternative selves are we investigating?What type of self-awareness are we scanning the historical material for? Arewe describing the development of a psychological entity, writing the historyof an epistemic question or an ethical dilemma, or telling the story of aconceptual fiction? Other questions seem even more serious: is it not rathersuspect to set out straightforwardly to study the history of a topic so loadedwith contemporary interest? Even if we were able to dispel the ambiguityabout the self, why should we suppose that thinkers in a period and culturalcontext distant from ours were interested in it in the first place? Indeed, ifinterest in the self is constitutive to modernity, should we rather not assumethat any sustained discussion about it is unlikely to have taken place beforethat particular epoch?

Worries of this sort are by no means exclusive to conscientious historians.Spurred by the ghosts of colonial history, the sociological and anthropo-logical theses of the unimaginable variety of human intellectual and sociallife have penetrated our cultural consciousness and made us particularlysensitive to the diverse values, beliefs, convictions and experiences thatpeople in different cultural contexts can hold and recognize. Indeed, thisconviction of the variability of human being is pivotal to the post-modernidea that human selves or identities are constructed out of elements, manyof which are not determined by our species but are rather open to all sorts ofactive interference by ourselves or by various forces in the cultural and socialcontexts of our lives. As tantalizing as it may initially seem to study anancient Greek thinker’s or a seventeenth-century Iranian philosopher’srespective theories of the self, the first question to ask is why we canlegitimately expect him even to recognize the entity.

In the following, my intention is not to start from any particularcontemporary concept of the self or self-awareness. For this reason, itwould be topsy-turvy to start off by describing the focus of our investigation

2 Cf. Kenny 1988 and 1999; Dennett 1991; Olson 1998; Metzinger 2003 and 2011.

2 Introduction

in specific terms. Rather, I will begin by reconstructing a particular way ofdescribing and conceiving of the self and self-awareness that emergesexplicitly for the first time in Islamic philosophy in the psychologicalwritings of Avicenna (d. 1037). Having laid this basis, I will proceed tostudy the development of this particular description and concept, as well asthat of the arguments applied in its articulation, in the thought ofAvicenna’s most illustrious successors, down to the revisionist philosophicalsystem of Mullā S· adrā (d. 1635/6) in the seventeenth century CE. The pointis to start from the way in which our authors describe, organize and classifytheir experience, asking why they chose to pay attention to this particularaspect of human experience, and what role the concept and the phenom-enon of self-awareness played in their thought.To anticipate the story this approach will yield, it is illuminating to

make a heuristic distinction between the phenomenology and the meta-physics of the self and self-awareness. To borrow Galen Strawson’s suc-cinct demarcation, ‘Metaphysics . . . is the general study of how things areor can be or must be. It’s a matter for scientists and mathematicians as wellas philosophers, and I take it to include physics as an evolving part.Phenomenology is the study of a particular part of how things are or canbe or must be. It’s the general study of the character of experience in all itssensory and cognitive richness.’3 Thus, the metaphysics of the self (or self-awareness) concerns the question of what sort of entity (or event, state orcapacity) it is in reality, whether such things as selves really exist in the firstplace, and if so, whether they are anything like they initially seem to be. Incontemporary terms, the paradigmatic question to ask is whether ournaturalistic framework of explanation needs such entities as selves at all,or whether we can explain them away by reductive recourse to somethingmore foundational. But even if we adopt a reductionist metaphysicalstance towards the self, we need not deny its persistence on the level ofphenomenology. If it is an undeniable fact that people are aware ofthemselves in some sense, and if this is all we mean by their having selves,then the phenomenological level is a matter of discussion of how todescribe the phenomenon. We can make positive assertions about thephenomenon without committing either to realism about a correspondingthing or to the denial thereof; in Strawson’s words, there can be self-experiences on the phenomenological level (perhaps even consensus about

3 Strawson 2009, 1.

Introduction 3

distinctions between their types) even if nothing like selves existed accord-ing to our metaphysics.4

Thus, one way of characterizing the plot of the present story is to say thatit is an unfolding of different metaphysical interpretations on a sharedphenomenological basis. The repetition of familiar Avicennian argumentsrelated to self-awareness, often word for word, sediments the phenomenonof self-awareness into a received foundation of psychology. From the twelfthcentury CE onwards, most philosophical authors will begin their discussionof the human soul, sometimes even their entire psychology, with the famousthought experiment of the flying man, or apply in crucial stages theevidence of the subjective unity of experience or the argument againstreflection-based models of self-awareness. As I will argue in detail, this isbecause they unanimously subscribe to Avicenna’s description of self-awareness and his way of singling out this particular aspect of humanexperience.

The consensus dissolves, however, as soon as the discussion shifts to themetaphysical explanation of the phenomenon and the conclusions that canlegitimately be drawn on its basis. As we will see, Avicenna himself consid-ered self-awareness to be a potent pointer towards, if not a proof of, thetruth of his substance dualist view of human being, but this move wasalready being questioned by the first generation of his students. Thissceptical strand was continued and established as a firm part of the sub-sequent theological tradition by Fakhr al-Dīn al-Rāzī (d. 1209) towards theend of the twelfth century CE. Yet it was not the whole story, for Rāzī’scontemporary Shihāb al-Dīn al-Suhrawardī (d. 1191) not only adoptedAvicenna’s description of self-awareness, but also placed it at the veryfoundation of his new illuminationist concepts of knowledge and being.Thus, from a potent piece of evidence in human psychology, the phenom-enon of self-awareness became the paradigmatic type of knowledge, and thecornerstone of an entire metaphysics.

Not only was the phenomenon of self-awareness open for radically newargumentative applications – there was also room for debate concerning thecorrect metaphysical account of the entity behind self-awareness, thehuman self. This becomes eminently clear in our investigation ofthe thought of Mullā S· adrā, who embeds the received description of self-awareness in a radically revised metaphysical framework. A determinedsubscriber to his predecessors’ means of describing and delimiting the

4 Strawson 2009, 2. Following Strawson, in the present study the terms ‘phenomenology’ and ‘phe-nomenon’ are not used to refer to the thriving tradition of philosophy founded by Edmund Husserl.

4 Introduction

phenomenon, he nevertheless criticized their conception of the self as beinginadequately static. Instead of a stable substance that endures unchangedthrough the constant flux of its attributes and relations to the world, S· adrāpreferred to conceive of the self as a substance in motion that is thoroughlydetermined by the variation of its attributes, and unified only in the sense ofbeing a single continuous stream of existence that is aware of itself.In an approach of this kind, the proof of the pudding can only be in the

eating. I am not studying human selfhood and self-awareness as a ‘perennial’topic of philosophy, but aim instead to describe one historical, and radicallycontingent, trajectory that to me seems best understood by means of ourterminology of self and self-awareness.5 In the end, the texts under studymust not only provide the ingredients for our reconstruction of the Islamicphilosophers’ description of self-awareness, they must also yield sufficientevidence that philosophers writing in Arabic from the eleventh century CEonwards had both the motivation and the conceptual means to pay system-atic descriptive attention to their experience. Although Avicenna’s,Suhrawardī’s and Mullā S· adrā’s concepts of self and self-awareness are notwithout parallels among contemporary classifications of different types ofselfhood and self-awareness, the crucial claim remains that we can and mustreconstruct those concepts, and the underlying preoccupations, interestsand convictions, without taking our primary cue from correspondingmodern concepts.Since the present book is a story of the emergence and development of

one particular concept of self and self-awareness, it is by necessity relativelynarrow in its focus. Consequently, it does not strive to give an exhaustiveoverview of the different possible concepts of self and self-awareness thatone might be able to locate in Islamic intellectual history between theeleventh and the seventeenth centuries CE. If such a general investigationwere conducted rigorously, that is, according to the sort of bottom-upapproach we have just sketched, it would exceed the limits feasible for asingle-volume study, and would most certainly be beyond the capacities ofthe present author. A more liberal charting of the landscape, on the otherhand, could scarcely avoid taking its cue from some contemporary ways ofconceptualizing the self and self-awareness, which would seriously compro-mise its value for the systematically and historically demanding reader.

5 Thus, I wholeheartedly subscribe to the concern over anachronistic rational reconstruction as semi-nally formulated in Skinner 1969, although I believe that it must be qualified by a recognition of thelimits of all historical reconstruction. For an attempt at articulating those limits, with a particular viewto the questions of self and self-awareness, see Kaukua and Lähteenmäki 2010.

Introduction 5

This is by no means the first historical study of the self and self-awarenessin pre-modern philosophy. Charles Taylor’s seminal Sources of the Selfalready set off with a sketch of the opening of the interior space ofexperience in ancient philosophy, even if the book’s main emphasis wason the emergence of the specifically modern notion of selfhood.6 Morerecently, Richard Sorabji published a large volume with the succinct titleSelf, which deals emphatically and at considerable length with ancient andmedieval views on a range of metaphysical, psychological and ethical ques-tions related to selfhood and self-awareness.7 Roughly simultaneous toSorabji’s book, Raymond Martin and John Barresi came out with anintellectual history of personal identity titled The Rise and Fall of Soul andSelf.8 Finally, Alain de Libera’s ongoing Archéologie du sujet is an engagingstory of how the modern notion of subjectivity emerges from the develop-ment of decidedly medieval philosophical concerns.9 These are just some ofthe most prominent recent examples, which can be supplemented by severalexcellent studies focused on a single thinker or a more distant period.10 Yetin spite of the considerable joint merits of these books in covering a vastarray of thinkers, none of them ventures very far into the territory of Arabicor Islamic philosophy, with the stand-alone exception of Avicenna, whosethought experiment of the flying man is often quoted as a perspicacious ifpuzzling attempt at describing and delineating the phenomenon of self-awareness. Similarly, the handful of articles or book chapters that have beenwritten on self-awareness in Islamic philosophy are mostly focused onAvicenna.11

That historians of the self have neglected the post-Avicennian develop-ment in Islamic philosophy is not particularly surprising. Until quiterecently, Islamic philosophy was regarded as a fringe phenomenon in thebroad scope of the history of philosophy, worthy of inclusion only to theextent that it played a role in the transmission and transformation ofthe Greek heritage before its final appropriation by the Latin philosophersand theologians from the thirteenth century onwards. While the absence ofverifiable contacts between the principal proponents of Islamic andChristian philosophy after Averroes’ death in 1198 CEmay have legitimatedthe delegation of the study of the subsequent Islamic tradition to the

6 Taylor 1989. 7 Sorabji 2006. 8 Martin and Barresi 2006.9 De Libera 2007 and 2008, with two more volumes announced.10 Cf., for example, Gill 1996 and 2006 (on Greek literature and Hellenistic philosophy, respectively);

Cary 2003 (on Augustine); Remes 2007 (on Plotinus); Cory 2013 (on Aquinas).11 Cf. Sebti 2000, 100–117; Black 2008 and 2012; and Kaukua 2007. Marcotte 2004 and Kaukua 2011deal with Suhrawardī’s relation to Avicenna.

6 Introduction

orientalists, this was often coupled with the more derogatory thesis thatthere simply was no philosophical activity worthy of the name in the Arabiclanguage after Averroes’ allegedly unsuccessful attempt to defend philoso-phy against Abū H· āmid al-Ghazālī’s (d. 1111 CE) fatal blow dealt in hiscritical Tahāfut al-falāsifa.It has since been conclusively shown that Ghazālī did not put an end to

the development of philosophical thought in the Islamic world, eithersingle-handedly or as the spearhead of a wider opposition from orthodoxtheologians. In fact, the contrary consensus is beginning to emerge accord-ing to which he may not even have intended anything of the sort. Instead,Ghazālī has been argued to have knowingly incorporated a great amount ofphilosophical material, not to mention the philosophical method of rigor-ous argumentation, into his own thought, and to have been followed in thisby Fakhr al-Dīn al-Rāzī, another highly venerated Sunnī theologian.12

Thus, although self-proclaimed philosophers may have grown rare in thesubsequent centuries of Islamic thought, philosophical activity prospered inSunnī theological writing and teaching, quite likely down to our era.13

On the other hand, Iran has fostered a thriving philosophical traditionthrough to the present day. In the light of our increasing knowledge of thedevelopment of this field of intellectual activity, it seems a safe estimate tosay that post-Avicennian Islamic philosophers were not afraid of makingdepartures comparable in extent to their early modern European peers.14

This is especially evident in the thought of Suhrawardī and Mullā S· adrāwhose revisions of received views will be our major concern in the following.Nevertheless, the strictly philosophical value of this tradition is sometimesstill obscured by the fact that some of its most prominent Western scholarshave tended to emphasize other, more mystical aspects of the philosophers’thought. This is especially true of the pioneering work of Henry Corbinwho played a major role in their introduction to theWestern public. Corbinwas an eccentric thinker who developed his own method of phenomeno-logical interpretation, which hinged upon the explicit permission, or indeedrequirement, to give up most of the rigour of the historical method; instead,one was to strive to imaginatively reinvigorate the mystical insights of one’sobjects of study. Instead of an attempt at philosophical understanding, thisoften involved extravagant emphasis on the symbols and myths, which

12 See Wisnovsky 2004a; Shihadeh 2005 and 2006; Griffel 2009.13 See Wisnovsky 2004b; El-Rouayheb 2010.14 For a concise account of the central debates in Iranian philosophy in the fifteenth and sixteenth

centuries, see Pourjavady 2011, 1–105.

Introduction 7

some of the authors frequently employ, to draw daunting connections notonly between different eras of Iranian thought but also between the Islamicphilosophers and historically unconnected European mystics such as JakobBöhme or Emmanuel Swedenborg.15 This approach to reconstructing thehistory of Iranian philosophy has hardly increased the credibility of thetradition in the eyes of less extravagant readers, and the situation has notbeen helped by the fact that a number of influential scholars have main-tained Corbin’s emphasis on mysticism.16 Although many of these scholars,and Corbin in particular, should be lauded for their historical and philolo-gical contributions, their work has been a mixed blessing for the widerrecognition of the philosophical merits of post-Avicennian philosophy.

The recent past notwithstanding, few specialists today will debate theinclusion of post-Avicennian philosophical authors in the class of subjectsmeriting serious philosophical study. But despite several excellent studiessince the 1980s,17 our understanding of later Islamic philosophy is not yet onthe level that we have come to expect in the case of canonical figures such asal-Fārābī, Avicenna or Averroes. In my view, it is crucial for reaching thisgoal that we interpret the ‘post-classical’ authors in close and rigorousconnection to the classical Avicennian framework, in the understandingof which we can rely on several decades of first-rate philosophical scholar-ship on a wide range of topics. As this study of self-awareness suggests, eventhe most original moves of thinkers like Suhrawardī or Mullā S· adrā can befully appreciated only against this background; Avicenna’s insights areneither a model to be slavishly followed nor an antiquated edifice to besimply discarded in favour of supposedly higher mystical ways to reachthe Truth, but rather potent material for revision and reapplication. This ismost obvious in Suhrawardī’s employment of Avicenna’s psychologicalarguments for the irreducibility of self-awareness as the basis for his newmetaphysics of light and appearance in the H· ikma al-ishrāq. It is true thatthe debts are not always acknowledged, and it is not uncommon that wehave to show the Avicennian credentials of an author against his expressdenouncement – again, Suhrawardī’s wholesale rejection of Peripateticmetaphysics and theory of science, preliminary only to the introductionof another piece of Avicennian evidence, is a case in point – but in thisregard the post-Avicennian philosophers are by no means unique.

15 For a prime example of Corbin’s method in practice, see Corbin 1971, vol. ii.16 Cf., for instance, the work of Corbin’s close associate Seyyed Hossein Nasr (such as his 1978 and

1996); and studies like Morris 1981 or Amin Razavi 1997.17 Cf. Ziai 1990; Jambet 2002 and 2008; Bonmariage 2007; Rizvi 2009; Kalin 2010; Rustom 2012.

8 Introduction

Thus, by means of the particular case of self-awareness I hope to sub-stantiate the claim that we should read authors like Suhrawardī and MullāS· adrā as reacting first and foremost to philosophical debates and texts, and tointerpret them with the sort of expectations of conceptual rigorousness andinsight that guide us in the formative case of Avicenna. Conversely, thestudy will also propose that the investigation of the reception of some ofAvicenna’s original ideas may be a considerable asset in our attempts tounderstand those ideas in their inceptor by providing corroboration for ourreconstructions of them. In the case of self-awareness, the novelty ofAvicenna’s concept gives rise to a number of complications in the frame-work of Peripatetic psychology, epistemology, and metaphysics, which inturn has made its reconstruction a matter of considerable difficulty anddebate. While similarity with a particular strand of reception is of courseno evidence for the correctness of any single interpretation of the view thatis being received, the twelfth-century discussion of self-awareness can stillhelp us by showing which interpretation the thinkers temporally andculturally close to Avicenna considered as the most plausible. This isevinced by their devising additional arguments along Avicennian lines,such as the systematic distinction between the subject and the object ofexperience in terms of ‘I’ and ‘it’, respectively, or their introduction ofhighly clarificatory new terms to describe self-awareness, such asSuhrawardī’s ‘I-ness’ (anā’īya).

The first chapter of the book discusses the most prominent pre-Avicennianphilosophical concepts of the self and the various types of self-cognition,and introduces briefly some of the basic doctrines of the Avicennianpsychology that provide the framework for much of the ensuing discussion.Chapter 2 will introduce the phenomenological basis of Avicenna’s new

concept of self-awareness. I will start with one of his most famous argu-ments, the thought experiment featuring the flying or floating man. Byreading the flying man in its immediate context and in close connectionwith the argumentative goals it is intended to reach, I attempt to show thatAvicenna builds his concept of self-awareness upon something he expects usall to be familiar with from our everyday experience. This is an importantpoint to make not only because the nature of the thought experiment hasbeen a matter of scholarly debate, but first and foremost because its familiarphenomenological basis is crucial to my later reconstruction of Avicenna’sconcept of self-awareness.Chapter 3 adopts a parallel line of approach by considering Avicenna’s

possible motives in introducing the new concept of self-awareness. This

Introduction 9

question of theoretical rationale becomes pressing because of the strikingclaim that Avicenna makes in his mature correspondence, namely the claimthat self-awareness amounts to the existence of the immaterial humansubstance. The interpretation I suggest is that Avicenna may have perceivedself-awareness as instrumental to presenting a coherent psychological sub-stance dualism in the Peripatetic framework that founds individuality on astrong connection to matter. In other words, he may have seen in self-awareness a solution to the question of how a human being can be both animmaterial substance and an individual instantiation of the human species.

After these preparatory chapters, I finally present my reconstruction ofAvicenna’s concept of self-awareness in Chapter 4. By considering a numberof Avicennian arguments related to self-awareness, I attempt to show thatthe new concept is intended to capture a very narrow sense of first-personality inherent in all human existence. I argue that this reconstructionof the concept is particularly charitable to Avicenna, because it is capable offulfilling the stringent requirements placed upon the concept by all theargumentative contexts in which the phenomenon is applied. This is furthersupported by a consideration of the scattered remarks Avicenna makes onreflective self-awareness.

Chapter 5 moves on to discuss the treatment of this aspect of theAvicennian heritage in the thought of his twelfth-century critics. Thechapter shows how the critical remarks of Abū al-Barakāt al-Baghdādī(d. 1164/5 CE) and Fakhr al-Dīn al-Rāzī lead eventually to Suhrawardī’sseparation of the phenomenon and the concept of self-awareness fromAvicenna’s metaphysical account of it as the existence of the human sub-stance. Henceforth, self-awareness can be conceived in purely phenomeno-logical terms, that is, without explaining it in more foundationalmetaphysical terms.

As will be shown in Chapter 6, the separation of Avicenna’s phenomen-ology of the self from his metaphysics is decisive for Suhrawardī’s ownattempt at developing a self-styled illuminationist (ishrāqī) alternative toAvicenna’s Peripatetic system. Through a close reading of the passages inwhich he introduces the pivotal concepts of knowledge as presence (hud

˙ūr)

and being as light (nūr) or appearance (z· uhūr), I show that self-awareness ispivotal to the definition of both new terms. Thus, Suhrawardī is witness to aseismic shift in the application of the concept of self-awareness without anychange in the description of the underlying phenomenon. Regardless of itsgreat explanatory power, Avicenna seems to have restricted the importanceof self-awareness to psychological concerns, but in Suhrawardī it becomes acornerstone of both epistemology and metaphysics.

10 Introduction

Chapter 7 proceeds to consider how self-awareness figures in the thoughtof Mullā S· adrā, the great seventeenth-century synthesizer of the variousstrands of earlier Islamic thought. A thorough consideration of S· adrā’smagnum opus al-H· ikma al-muta‘āliya fī al-asfār al-arba‘a al-‘aqlīya showsthat he incorporates most of the traditional arguments revolving on self-awareness as well as a great deal of the critical attention subjected to them inthe intervening centuries of Islamic thought. This is particularly interestingbecause two of the philosophical doctrines that set S· adrā apart frommost ofhis predecessors seem to be severely at odds with the Avicennian concept ofself-awareness. Chapter 8 will therefore turn to consider in detail howS· adrā’s adherence to the ancient theory of cognitive unity, and to hisoriginal theory of thoroughgoing change involving the substantial core ofeach and every created entity, affects his understanding of the self and self-awareness. It will be seen that notwithstanding S· adrā’s initial subscriptionto the traditional material, his concept of the self is significantly differentfrom that of his predecessors. Instead of a narrow and static first-personality,the S· adrian self is thoroughly intertwined with other determinations ofexperience and thereby subject to genuine change and development.To sum up, the story that we are about to tell hinges on a single

philosophically loaded phenomenon, a feature of human experience thatall central actors single out and define by largely the same empirical andargumentative means. Yet it is a genuine story because it incorporatessignificant shifts in the conclusions made on that shared basis. Moreover,as we will see, even the phenomenological basis of the concept of self-awareness is opened for revision. This space for conceptual variationsnotwithstanding, it is important to underline that the storyline itselfemerges from the historical material. It is clear that each subsequentphilosopher latches on to a discussion, the means and motives of whichhe inherits from his predecessors. I do not doubt that other eminentlyinteresting and historically sound stories about the self and self-awareness inthe Islamic cultural milieu remain to be told, but I would like to claim thatthe narrow focus of the present study allows us to reap the benefit of a plotthat marries philosophical suspense to historical truth.

Introduction 11

chapter 1

Preliminary observations: self-cognition andAvicennian psychology

1.1 Self-cognition in the ancient heritage

One of the historical claims of the present study is that Avicenna’s workmarks the point of entry of a new concept of self-awareness into the Arabicphilosophical scene. This does not mean that there was no prior Arabicphilosophical discussion about the self and the various types of self-cognition, nor do I wish to claim that the novel concept was developedon a clean theoretical slate. Like all philosophers, Avicenna builds on theremarks, arguments and doctrinal convictions of his predecessors, but it isan altogether different question whether these were sufficient to determinehis thinking, or whether Avicenna gave the tradition a decided twist of hisown. In order to substantiate the latter alternative, developed in a morepositive vein in the following chapters, let us briefly review some of thoseaspects of the ancient heritage that pertain to questions about the self andself-awareness and that were demonstrably available in pre-AvicennianArabic philosophy.

An obvious point of relevance is Aristotle’s discussion of our perceivingthat we perceive in De anima III.2. Although self-awareness is not the mainfocus of this passage, which is naturally interpreted as addressing the moregeneral problem of phenomenal consciousness,1 it can be argued to implysome sort of self-awareness as well, since Aristotle explicitly states it to beone and the same subject of perception that both perceives something andperceives itself to perceive that something.2 Aristotle’s reasons for introduc-ing the question of the perception of perception are a matter of debate; oneplausible view is that he does it out of a need to distinguish the physical

1 For an influential recent version of this interpretation, see Caston 2002 (and cf. Sisko 2004; Caston2004; Johansen 2005; Polansky 2007, 380–402; and Perälä 2010, 43–101). Earlier studies of the topicare Kahn 1966; Kosman 1975; Hardie 1976; Hamlyn 1978; Modrak 1981, 1987; and Osborne 1983.

2 Cf. NE IX.9, 1170a29–b1, which suggests that Aristotle may have entertained an explicit concept ofsuch a primitive type of self-awareness.

12

changes that result in perception from those that do not.3 This is a problembecause, as Aristotle recognizes in De anima II.12, the very same causes canproduce both perceivable qualities and actual perceptions in another thing –consider, for example, a rose that makes the surrounding air odorous butincites a pleasurable sensation of smelling in a human being.4 The basis ofthe distinction is left strangely underdeveloped in De anima,5 but Aristotledoes state it in the more general context of Physics VII.2, where we can findthe general claim, here translated from Ish. āq ibn H· unayn’s (d. 910/11 CE)Arabic, that ‘that change in other than the senses [that is, the change of whatis not “animate” (al-mutanaffisa)] is not aware of the change’ (takūnaal-istihālata tilka bi ghayri al-h. awāssi lam tash‘ur bi al-istih.āla).

6 Sincepercipient animals are not mere bodies, like rocks whose behaviour can beexhaustively described in general physical terms, the natural philosophermust introduce further differentiating properties as principles for theirstudy in the specific subdiscipline of cognitive psychology. The percipientsubjects’ awareness of the processes of perception can be understood asprecisely such a differentiating property. Thus, in this reconstructionAristotle’s introduction of the question of how awareness of perceptioncomes about is an immediate result of his general concept of perception.In the Greek text, Aristotle speaks of us ‘sensing that we are seeing or

hearing’ (aisthanometha hoti horōmen kai akouomen) and therefore in somesense perceiving ourselves in the act of seeing or hearing, and mentions asense faculty or act which is capable of sensing itself (autē hautēs).7 Heargues that awareness of oneself as perceiving is a kind of perception, andconsequently a corresponding faculty of perception must be ascribed toexplain it. Two options then readily suggest themselves: either there is asecond-order faculty in addition to each sense which perceives the first senseperceiving, or the first-order sense perceives both an external object and itsown perception of the external object. Since the first option will lead to an

3 So Caston 2002, 755–759; for the conflicting view that Aristotle simply takes the phenomenon forgranted, see Perälä 2010, 51.

4 Ar. De an. II.12, 424b3–17.5 Aristotle may also have been thinking about the implicit distinction between mere alteration (inan-imate things) and the actualization of a potency (animate things) in De an. II.5, or the distinctionbetween receiving a quality with matter (inanimate things) and without matter (animate things) inII.12.

6 Arist.ūt.ālīs, al-T. abī‘a VII.2, 244b12–245a2, 751. On Ish. āq’s translation, see Peters 1968, 30–34.7 Ar. De an. III.2, 425b12–17. Whether he meant an act or a faculty is one of the scholarly bones ofcontention; see Caston 2002 and Johansen 2005 for prominent representatives of the activity andcapacity readings, respectively. My use of ‘faculty’ in the following should not be taken as a stand inthis debate, although it does seem the more natural reading from the point of view of Arabicphilosophical psychology, especially that of Avicenna.

Self-cognition in the ancient heritage 13

infinite regress unless a self-perceiving faculty is posited at some point,Aristotle argues that we should posit such a faculty at the first stage. As aconclusion, every faculty of perception is said to perceive its own act.

Unfortunately, the beginning of De anima III.2 is not included amongthe surviving fragments of Ish. āq’s translation, Avicenna’s preferred text. Aninferior anonymous translation, to which he also had access, renders therelevant section as follows:

But since we apprehend when we see and hear (kunnā mudrikīna lammāra’aynā wa sama‘nā), vision’s apprehension when it sees must necessarily beeither through itself (bi nafsihi) or through something else. But it wouldapprehend both itself (yakūna mudrikan nafsahu) and the colour of thesubject. Because of that either two things apprehend a single thing or visionapprehends itself (nafsahu). If vision had a distinct sense, then that woulddescend infinitely in division or turns to apprehend itself (raja‘a fa kānamudrikan nafsahu); as a consequence, this is to be said of the first sense.8

Where the Greek original speaks of sensing that we are seeing or hearing,the anonymous translator renders the point by means of an implicit andlargely ambiguous distinction between seeing or hearing as such, andapprehending when (lammā) we see or hear. Not only is it said that weapprehend when we see and hear things, even the faculty of vision does.9

However, this ambiguity does not obscure the text’s relevance for thequestion of self-awareness, which becomes pronounced in the subsequentreflexive use of nafs in connection with mudrik – there really is somethingapprehending itself here.

Since we lack access to Ish. āq’s allegedly superior translation, it is difficultto determine what exactly readers of the Arabic Aristotle drew from Deanima III.2. I would like to argue, however, that although perception ofperception continued to be recognized as a psychological topic, its ingre-dients seem not to have made their way into explicit discussions on self-awareness proper. Avicenna, for instance, alludes to the problem as aquestion that is pertinent to his theory of the internal sense faculties, withnot so much as a hint that he would have perceived it to have any consid-erable repercussions to his concept of self-awareness.10 Although thethinkers studied in the present volume uniformly agree that all cognitive

8 Arist.ūt.ālīs, Fī al-nafs III.2, 425b12–17, 64. That this version is not Ish. āq’s was established in Frank1958–9 (for Avicenna’s access to the anonymous translation, see 232); the most extensive recentsummary of the scholarly debate is Elamrani-Jamal 2003.

9 An additional difficulty is due to the fact that lammā (‘when’) is orthographically identical to limā(here ‘what [is seen or heard]’), which would also make sense in the present passage.

10 For Avicenna’s brief reference to perception of perception, see Shifā’: Fī al-nafs II.2, 66–67 Rahman.

14 Preliminary observations

activity involves a primitive type of self-awareness, no one seems to connectthis view to De anima III.2. In the end, this is not even particularlysurprising, for the tendency after Avicenna is to emphasize the independ-ence of self-awareness from any kind of perception.Another framework in which pre-Avicennian Arabic philosophy

addresses the self and self-cognition is the theory, also of Aristotelianprovenance, of intellection as an identity or unity of the intellect in actand its intelligible object. If this unity between the two relata is understoodin a strict manner, it has the consequence that, in the final analysis, all actualintellection is self-intellection. Although the theory was the subject of fiercedebate in our period,11 it had an authoritative basis in the translations ofsome of the most highly regarded ancient metaphysical texts.In his discussion of the intellect in the third book of De anima, Aristotle

makes the repeated claim that, in an act of intellection, the intellect is onewith its object. Some of these statements are lost in the anonymous Arabictranslation,12 but the following excursion to the intelligibility of the intellectclearly asserts the unity thesis about pure intellects that are completelyunmixed with matter:

[The intellect] is also understood just like other [things] that are understood.[In the case of] those in which there is no hyle, what understands and what isunderstood is a single thing.13

This is corroborated at length in Metaphysics XII and its account of howGod as ‘thought thinking itself’ moves the world by functioning as its finalcause. Without going into the details of this idea and its transmission intothe Arabic, let us briefly cite two representative formulations in Ust.āth’s(ninth century CE) and Abū Bishr Mattā ibn Yūnus’ (d. 940 CE) trans-lation, respectively:14

That which grasps itself (yafhamu dhātahu) is the intellect through theacquisition of what is understood, and it comes to be understood when itis in contact and grasps.15

11 As will be seen, Avicenna adamantly rejects the theory in his account of human intellection whileMullā S· adrā subscribes to it with equal conviction.

12 Compare Ar. De an. III.5, 430a20 with Arist.ūt.ālīs, Fī al-nafs III.5, 430a20, 75. Ish. āq’s translation ofthis passage has not survived.

13 Arist.ūt.ālīs, Fī al-nafs III.4, 430a3–5, 74. We do not have Ish. āq’s translation of the passage.14 For a concise account of the various Arabic versions of Aristotle’sMetaphysics as well as the attribution

of the passages to Ust.āth and Mattā, see Bertolacci 2006, 5–35; and for the twelfth book in particular,Geoffroy 2003. Avicenna is argued to have read from Ust.āth’s translation (Geoffroy 2003; Janssens2003a), but his version of the second passage does not seem to have survived.

15 Ar. Met. XII.7, 1072b20–21; the Arabic text is in Averroes, Tafsīr mā ba‘d al-t.abī‘a, 1614(cf. Genequand 1986, 157). Cf. Avicenna, Sharh. kitāb h. arf al-lām, 27.


[T]herefore it [that is, the best possible thing] understands itself (ya‘aqiludhātahu) since it is the most powerful, and it understands intellection. But itis always seen that knowledge, sense perception, opinion and intellection areof another, and when it comes to [them being] of themselves (li dhātihi), thisis only by accident. Furthermore, if intellection is one thing and beingunderstood another, to which of them will excellence belong? The thatnessof intellection is not identical with being understood, like knowledge is thething known in some things. In the case of intellectual things, the substancewhich is of no element and the what are [identical] in terms of thatness aswell. As regards theoretical [things] (al-rā’īya), the thing is word and intel-lection, and so what is understood is not different from the intellection.16

What these passages make clear is that the unity between the cognitivesubject and object holds of acts of pure intellection, that is, acts that have norelation whatsoever to anything material. It is also clear that Aristotle holdssuch strong cases of unity to amount to acts of self-intellection. But as thesecond passage indicates, not all acts of intellection are of this type, for thereare cases in which the act of intellection, like acts proper to other modes ofcognition, refers to an external object. Although Alexander of Aphrodisias(second to third centuries CE), perhaps the most venerated of Aristotle’scommentators for the Arabic philosophers, did seem willing to extend thethesis of unity to hold also of human intellects that are related to materialbodies and actualized in time through a process of learning, he too wants toemphasize that the unity only holds of the subject and object in an act ofintellection. In absolute terms the human intellect, which moves discur-sively from one act of intellection to another and is often in the state ofpotency with regard to most acts of intellection, is not identical to theintelligibles it is in principle capable of grasping.17

The idea of all intellection as self-intellection is even more explicit insome of the propositions of the Kitāb al-īd

˙āh. fī al-khayr al-mah. d

˙, a treatise

attributed to Aristotle but in reality translated and adapted from theNeoplatonist Proclus’ (d. 485 CE) Elements of Theology. Consider, forinstance, the twelfth proposition:

Every intellect understands itself (ya‘aqila dhātahu), that is, it is simulta-neously what understands and what is understood. Therefore, since theintellect is both what understands and what is understood, it undoubtedly

16 Ar. Met. XII.9, 1074b33–1075a4; the Arabic text is in Averroes, Tafsīr mā ba‘d al-t.abī‘a, 1692–1693(cf. Genequand 1986, 190–191). Cf. Avicenna, Sharh. kitāb h. arf al-lām, 31–32.

17 Cf. al-Iskandar al-Afrūdīsī, Fī al-‘aql, 181–199 Finnegan; 31–42 Badawī. Alexander also suggests thatthere is a corresponding unity between the subject and object of actual sense perception. As argued byKalin 2010, 17–25, in this regard he is an important predecessor for S· adrā’s broad concept of cognitiveunity. For Alexander’s pivotal role in the transformation of Aristotelian noetics, see Moraux 1978.


sees itself (yarā dhātahu). It knows that it is an intellect understanding itself(‘alima annahu ‘aqlun ya‘aqilu dhātahu). And when it knows itself (‘alimadhātahu), it knows the rest of the things that are under it, because they arefrom it.18

For the present concerns, we can set aside the vast differences betweenProclean and Aristotelian metaphysics; the majority of the original readersof the two Arabic texts perceived them, if not always as originating from asingle pen, at least as essentially compatible with each other. Thus, the pointremains that when we abstract intellection from all concerns related to theperceptible world of concrete material entities, it consists in an undividedact in which the subject that understands is the very object understood, asingle act of intellection with two interdependent structural constituentsdivisible in analysis but not in reality. However, as the end of the crypto-Proclean passage clearly indicates, the paradigmatic case of cognitive unity isagain a type of intellection that is quite distant from the human realm. Weare dealing with an eternally actual intellect that is capable of bringingthe world about through its overabundant act of self-intellection, that is, theNeoplatonic intellectual hypostasis which, when combined with thethought thinking itself that functions as the goal for all existing things inMetaphysics XII, acts as both the source and the point of return of the entirecosmos, including individual human intellects. Consequently, the sort ofself-cognition these texts describe cannot be identified with any ordinarytype of human self-awareness, at least not without a considerable amount ofadditional analysis and interpretation. This is not to deny that the idea of allactual intellection as self-intellection provided an ingredient for the conceptof self-awareness emerging in Avicenna; on the contrary, as I will arguebelow, it seems natural to think that in articulating the novel conceptAvicenna took his cue from precisely this piece of the tradition. However,the theory of all intellection as self-intellection, when considered alone andwithout further qualifications, seems insufficient to account for the emer-gence of the considerably more mundane concept of self-awareness that isour main concern in the present study.One way of describing the relation between human intellection and the

sort of self-intellection we have just characterized is to conceive of the latteras the second perfection proper to human intellects.When a human being isborn, her intellect is at the state of pure potency. However, it is not entirelynon-existent, and since at least the potency exists we can say that the

18 Kitāb al-īd˙āh. fī al-khayr al-mah. d

˙XII, 14–15; corresponding to Proclus, El. Th. 167–169. For very

similar formulations, see XIV, 16, which corresponds to Proclus, El. Th. 83.


intellect proper to that human being has reached a state of first perfectionwhich amounts to saying that it simply exists. As the person acquiresknowledge through an arduous process of learning, her intellect proceedsto a state of greater perfection; this is, however, perfection no longer in thesense of simply existing but in the sense of existing in a more excellent wayaccording to the standard proper to the sort of thing it is. The sense in whichher intellect is more perfect is thus in terms of the second perfection ofexisting more or less well according to intellectual standards.19

The fact that the Aristotelian and Neoplatonic precedents for the Arabicdiscussion of self-cognition were designed to describe the second perfectionproper to us as intellectual entities is particularly evident in the corpusknown as the Arabic Plotinus. These texts are replete with uses of the termdhāt that read naturally as references to various self-relations entailing somekind of self-cognition (or self-recognition). Regular mention is made of, forinstance, entering oneself (dakhala fī dhātī), returning to oneself (raja‘a fīdhātī), being inclined towards oneself (māla ilā dhātihi), and beholdingoneself (naz· ara ilā dhātihi).

20 Once these passages are read in their context,it soon becomes obvious that we are dealing with different aspects of thesecond perfection of human being. For example, the inclination towards theself is always portrayed in contrast to a corresponding mundane orientationthat entails an emphasized attention to and striving for various worldlyobjects and objectives: one enters or returns to oneself precisely by with-drawing from the world and its concerns. The implied notion of the self ispart and parcel of the general emanationist framework of the ArabicPlotinus, and it amounts to the type of upward epistrophe specific tohuman beings, to a turning back towards one’s origin in an imitation ofits cognitive perfection. Extrapolating somewhat, we might say that this isself-cognition insofar as it amounts to the acquisition of a correct concep-tion of one’s proper place in the cosmos of God’s creation, and to actions inaccordance with this conception. Knowing oneself means recognizing one’strue self, what one is or can be according to the highest potencies inherent inone’s essence – not a mundane creature with a variety of ephemeral con-cerns but an intellectual entity capable of gazing at the divine. Thus,

19 For an excellent account of the emergence and maturation of this pair of concepts up to Avicenna, seeWisnovsky 2003, 21–144.

20 See, respectively, ThA I, 22 (corresponding to Plot. Enn. IV.8.1.1–10); VI, 80–81 (corresponding toPlot. Enn. IV.4.43.16–44.6); VIII, 116–117; IX, 132–133 (corresponding to Plot. Enn. V.1.12.12–20);and cf. VIII, 111 (corresponding to Plot. Enn. V.1.4.18–25), 115–117 (corresponding to Plot. Enn.V.8.10.26–11.13); and the pseudo-Fārābian Risāla fī al-‘ilm al-ilāhī, 173–174 (corresponding to Plot.Enn. V.3.8.8–9.20).


although the Arabic Plotinus addresses the human self in considerablybroader terms than the narrow focus at self-intellection in Metaphysics XIIand Kitāb al-īd


˙allows, it remains on the level of what

results from acquired knowledge. The implied concept of self is somethingwe must strive to reach, and hence something that we do not initially have.This is clearly acknowledged by AbūNas.r al-Fārābī (d. 950 CE) who in theMabādi’ ārā’ ahl al-madīna al-fād

˙ila, right after describing the first principle

of all existence as an intellect that understands itself, goes on to characterizeit by distinguishing it from human intellects that need to acquire theintelligibles by means of which they can think themselves.21 In his brieftreatise on the intellect, Fārābī makes the same point by stating thatintellection becomes self-intellection only when the intellect is fully devel-oped and can therefore dispense with any reference to external materialobjects.22

Although this inherited discussion on self-cognition and self-recognitionmostly revolves around the benefits of purely immaterial intellection, it wasbrought to bear on more ephemeral cases of human intellection in a mannerseemingly pertinent to our topic by the Constantinopolitan commentatorThemistius (d. c. 388). In his remarks on De anima III.5, Themistiusattempts to elucidate the distinction between the active and the passiveintellect by means of a related distinction between I and what it is to be me.This he does by asking the question (which in spite of its originality is basedon a perfectly sound Aristotelian distinction between an entity and itsessence) of which of the two intellects, potential or active, we human beingsshould identify with.

I wonder whether we are the intellect in potency or the intellect in act. Wesay that if in the case of all things composed of what is in potency and what isin act the thing is different from the existence of the thing, then I must alsobe different from the existence belonging to me (akūna anā ayd

˙ān ghayra al-

wujūdi lī), and so I am the intellect composed of what is in potency and whatis in act, whereas the existence belonging to me is before that which is in act,that is, that which is [in act] through it. I know and confirm this in writing,for the intellect composed of what is in potency and what is in act writes, andits writing is not by means of what is in potency but rather by means of whatis in act, the act emanating to it therefrom . . . Thus, just as the animal issomething and the existence belonging to the animal something else, the

21 Fārābī, al-Madīna al-fād˙ila I.1.6–7, 70–72 Walzer; 49–51 Cherni. This aspect of the first principle is

then described as trickling downwards in the emanative chain until we reach the active intellect(al-Madīna al-fād

˙ila II.3.1–10, 100–104 Walzer; 81–87 Cherni).

22 Fārābī, Risāla fī al-‘aql, 18–19, 31–32.


existence belonging to the animal being from the soul which belongs to theanimal, similarly I am one thing and the existence belonging to me anotherthing, and the existence belonging to me is from the soul, but not from anysoul, for it is not from the sensitive [soul], that being hyle for the imagi-nation, nor from the imaginative, that being hyle for the intellect in potency,nor from the intellect in potency, that being hyle for the active intellect.Thus, since the existence belonging to me is from the active intellect alone,this alone is genuinely (bi al-s.ih. h. a) a form, or rather the form of forms,whereas the others are both subjects and forms.23

The intriguing example notwithstanding, a close reading of the passagereveals that Themistius is not concerned with the self and self-awareness assuch. Rather, his point is to locate the individual human being in thedynamic relation between the material and the active intellect, which hadbecome the central question of Aristotelian noetics since Alexander’sextended focus on the relevant sections of De anima. Themistius thususes the first person for reasons that are primarily rhetorical or didactic,and the point could just as well be made by means of an individual humanbeing and the essence of humanity. The second perfection proper to humanbeings is intellectual activity, and it is in this sense that the human firstperson, ‘I’ uttered by a human being, properly refers to an actual subject ofintellection. The basis of this particular type of second perfection is to befound in the human essence, or, in Themistius’ terms, in the type ofexistence proper to me as a human being. This existence, the form of mybeing actually me, is provided by the active intellect, because any act of anintellect generated in time owes its actuality to the active intellect. Thus, inspite of the first-personal terms in which Themistius renders his case, thediscussion reverts to the self in the sense of an actualized intellect, an entitythat has already proceeded at least to some extent on the way towards thesecond perfection proper to it. Moreover, in spite of its obvious originality,this passage from Themistius seems never to have been brought up ashaving any relevance to self-awareness by any of our protagonists.Although he was, together with Alexander, absolutely pivotal to Averroes’notorious idea of the shared material intellect,24 Themistius’ influence onthe post-Avicennian discussion on self-awareness was all but negligible.

23 Themistius, Fī al-nafs VI, 182–183; cf. the Greek original in Themistius, in De an., 100,16–100,36,bearing in mind that the manuscript at the basis of Heinze’s edition belongs to a different branch ofthe manuscript tradition (see Lyons 1973, XIII).

24 Cf. Averroes, in De an. III.1, 429a21–24, 387–413; for an English translation and the surviving Arabicfragments, see Taylor 2009, 303–329.


As a final example of pre-Avicennian concepts of self-cognition, let usconsider Aristotle’s statement according to which an actual intellect iscapable of considering itself at will. As a possibility for intellects that arealready actual, this type of intellectual self-reflection presupposes the self-intellection that results from the unity of the subject and object of intellec-tion, and must therefore be distinguished from it. By the same token, itwould seem to be even further removed from what each and every humanbeing receives as her birthright. But since this particular concept of self-cognition resurfaces in one of Avicenna’s arguments for his own concept ofself-awareness, it is worth briefly spelling out. The point is rendered withadmirable clarity in the anonymous Arabic translation:

Thus, when all we have told is in the state we have said before, [namely,] thatthe intellect is like the active knower (this it is when an act can be originatedby it), then after its knowing it knows that it knows (‘alima annahu ‘ālim)within the limits of potency, but not like before it knew and before it foundknowledge; and at that moment it can understand itself (ya‘aqila nafsahu).25

Once an act of intellection is realized, its subject can perform a reflectiveturn to understand its own activity. As Aristotle specifies later on in thesame chapter, this second-order act of intellection is possible because thefirst-order act is immaterial,26 all intellection being based on a process ofabstraction from material constraints. If materiality is the only obstacle ofintellection, then the immaterial first-order act comes readily prepared to beunderstood at will. Importantly though, Aristotle also states that self-reflection must be preceded by actual intellection of some other thing; thehuman intellect is nothing before it understands something,27 and what isnothing cannot reflect on anything, not even itself.This is straightforward enough, and it was generally accepted by most

philosophers in Arabic.28What does, however, become a matter of debate iswhether the potentiality of the individual human being’s material intellect isas pure as Aristotle suggests in the same context. Is there really no actualexistence to it prior to the act of conceiving an abstract intelligible form thatit receives from the active intellect? The problem of course is that any suchexistence would have to be spelled out in cognitive terms, given that theintellect is an immaterial entity whose existence primarily amounts to beinga subject of cognition. If one cannot rely on any actual intellection, very few

25 Arist.ūt.ālīs, Fī al-nafs III.4, 429b6–9, 73. Ish. āq’s translation does not survive, but Avicenna discussesthe passage in al-Ta‘līqāt ‘alā hawāshin kitāb al-nafs li Arist.ū, 103.

26 Arist.ūt.ālīs, Fī al-nafs III.4, 430a3–9, 74.27 Arist.ūt.ālīs, Fī al-nafs III.4, 429a23–25, b31–32; 74.

28 For a rare example of debate, see Rāzī, Sharh. al-Ishārāt, I.169–173.


alternative routes remain open, which seems to confirm Aristotle’s point:self-awareness first arises through reflection.

As we will see shortly, Avicenna does recognize a reflective type of self-awareness that he characterizes as awareness of awareness (shu‘ūr bi al-shu‘ūr), and is followed in this regard by virtually all subsequent Islamicphilosophers. However, this type of self-awareness is frequently discussed inthe context of an argument designed to refute the claim of its foundationalstatus. Thus, instead of subscribing to the Aristotelian thesis that the humanintellect is nothing before it actually understands something, Avicennapresents his alternative concept of self-awareness in striking contrast to it.According to him, self-reflection requires as its condition of possibility thatthe human being capable of reflection is first aware of herself in some moreprimitive and more foundational sense.

To thus sum up this quick foray to pre-Avicennian Arabic concepts of selfand self-cognition, we can say that all the texts we have brought up hinge onthe activity of either human or superhuman intellect. From an Avicennianpoint of view, they deal with something that presupposes, rather thanexplains, what should properly and in the most basic sense be called self-awareness. Although some of the features in the passages we have considereddo arguably play a role in the emergence of Avicenna’s novel concept of self-awareness, they do not suffice to explain that emergence, nor can theyprovide the basis for an exhaustive understanding of that concept. Weshould not draw exaggerated conclusions from the fact that thesepre-Avicennian texts use some of the same linguistic means to describeintellectual self-relations as Avicenna does to characterize self-awareness, forboth the reflexive terms denoting the self (dhāt and nafs in particular) andthe variety of cognitive terms applied to render the cognitive aspect of self-awareness would have been available in the relevant sense from non-technical colloquial speech.29

1.2 Avicennian psychology in outline

It is a widely recognized fact that with some exceptions, most notably theAndalusian tradition spearheaded by Averroes, Islamic philosophy from thetwelfth century CE onwards can be legitimately summed up as ‘post-Avicennian’.30This is due to a seismic shift in terms of the texts that providethe foundation for the emphatically literary practice of Islamic philosophy.

29 For a brief account of these terms, see Appendix.30 Cf., for instance, Endress 1990; Michot 2003a, 2003b; Wisnovsky 2004b; Shihadeh 2005.


If Aristotle, the ‘First Teacher’, was not only the ceremonial master ofphilosophy but also the foremost textual authority for Avicenna and mostof his predecessors, in the later period the role of authority is assumed by the‘Chief Master’ (al-shaykh al-ra’īs). This is certainly true of the science ofpsychology, in the field of which the sixth part of Avicenna’s voluminoussumma of philosophy, the Shifā’, becomes the authoritative textbook.As a result of this shift in authority, the entire discussion that we are

about to chart takes place within the general framework of Avicennianpsychology. Thus, I will briefly review four of its key doctrines in order toprovide a point of quick reference that will assist in the understanding ofsome of the more detailed arguments in the core of our study. Thesedoctrines are the Aristotelian theory of the three different functions ofsouls, Avicenna’s un-Aristotelian substance dualism, the principles of fac-ulty psychology and the theory of internal sense faculties, and the cognitivepsychological theory of the abstraction of forms from matter.

The Aristotelian tripartition of souls and Avicenna’s substance dualism

Avicenna subscribes to Aristotle’s definition of the soul as ‘the first perfec-tion of a natural body possessed of organs that performs the activities oflife’.31Thus, the soul is the formal principle that makes the living body whatit is and maintains it in actual existence. Now, since there are three generaltypes of life – vegetative, which amounts only to self-nutrition, growth andreproduction, animal, which comprises in addition the activities of percep-tion and voluntary motion, and human, which alone manifests rationalcognition and action – Avicenna postulates, again following Aristotle, threecorresponding types of soul, with each more perfect type performing theactivities of the less perfect types in addition to those proper to it.32

In the third book ofDe anima, Aristotle raises the question of whether allthree types of soul are enmattered formal principles, or whether the intel-lectual capacity of the human soul requires it to be an immaterial substancethat performs a formal function but is strictly speaking not a materialform.33 Although he clearly suggests that the intellectual capacity requiresimmateriality, Aristotle leaves the question of the exact consequences thishas for the ontology of human souls sufficiently ambiguous for it to becomea major question of debate in the subsequent tradition. Of the various

31 Avicenna, Shifā’: Fī al-nafs I.1, 12 Rahman; cf. Ar. De an. II.1, 412a26–27.32 Avicenna, Shifā’: Fī al-nafs I.5, 39–40 Rahman; cf. Ar. De an. II.2, 413a20–413b13.33 Ar. De an. III.4, 429a18–21; III.5, 430a17–19.

Avicennian psychology in outline 23

alternatives,34 Avicenna opts for a full-blooded substance dualism: theindividual human essence is an immaterial substance, which is not strictlyspeaking a form of the human body, although it does perform the functionsof a form, animating and using the body for its own ends. This marriage ofAristotelian psychology and substance dualism is not an easy one, and oneproblematic consequence in particular is significant for Avicenna’s discus-sion of self-awareness.

If the human essence is immaterial, why is it connected to a body in thefirst place? Like any other essence considered in itself and in abstractionfrom all relations to other things, such as matter or a body, the humanessence is one, undivided and simple. In order to be instantiated in amultiplicity of numerically distinct individuals, the human essence –again, like any other essence –must be actualized in matter or in a sufficientrelationship to matter. This is because only the unique spatiotemporal co-ordinates afforded by matter can individuate the essentially identical instan-tiations; for a simplified example, if we have two triangles with exactlymatching dimensions, the only way to distinguish between the two is byattributing them with distinct co-ordinates.

Thus, the soul and the body are in mutual need of each other, but fordifferent reasons. The body requires the soul as a formal principle thatanimates it, or makes it a living body. The individual soul, on the otherhand, needs the body as a necessary condition of its initially coming to be asan individual.35 This is because an individual human soul is emanated fromthe active intellect, the principle informing the entire sublunar world, onlywhen a corporeal composite suitable for functioning as its body is formed inthe mother’s womb.36 The material process alone is incapable of developingthe embryo or the foetus to an individual human being capable of living onits own, for this requires a soul that has a separate origin. Thus, theemergence of the body is a necessary but not sufficient condition for theemergence of an individual soul. Nevertheless, the individual soul is inti-mately connected to its proper body through an inherent natural desire togovern, occupy itself with, and take care of the one body that is its own.37

34 For a useful summary of the pre-Avicennian reception of Aristotle’s theory of the intellect, seeDavidson 1992, 7–73.

35 Avicenna, Shifā’: Fī al-nafs V.3, 223–227 Rahman; cf. Najāt II.6, 222–223.36 For a concise account of the Avicennian theory of emanation, see Davidson 1992, 74–83.37 Avicenna, Shifā’: Fī al-nafs V.3, 225 Rahman; cf.Najāt II.6, 222–223. It is precisely on the basis of this

intimacy that Avicenna rejects the transmigration of souls in Shifā’: Fī al-nafs V.4, 233–234 Rahman(cf. Najāt II.6, 227).


Although the human soul is thus genetically dependent on and thereforeincapable of pre-existing the body, Avicenna still wants to maintain itsontological independence from the body after its inception. The relationbetween the soul and its body is, after all, merely accidental to the soul.38

Moreover, as Thérèse-Anne Druart has observed, Avicenna prefers to speakof the co-emergence of the soul and the body as a simple simultaneoushappening; he never asserts that the body is a proper cause of the soul’scoming to be.39 The only cause the soul can have must be an intellectualprinciple that is situated above it in the cosmological hierarchy of emana-tion. But this principle, the active intellect, only contains the pre-individualhuman essence and thus requires the unique individuating conditionsafforded by the suitable body in order to produce an individual soul.Thus, in Avicenna’s view the individual human being is a nodal point oftwo conflicting tendencies; eventually, the thorny question of the soul’sindependence from the body returns to Avicenna in the form of theproblem of how to reconcile individuality with immateriality after thedemise of the body (see Chapter 3).

Faculty psychology and the internal senses

Although the concept of a psychological faculty, power or potency(Gr. dynamis, Ar. quwwa) can be found already in Aristotle, Avicennarepresents a stage of comparative maturity in the development of thepsychological method of faculty analysis. His general definition of thethree types of soul by means of faculties characteristic of each isAristotelian in outline,40 but the field in which he really comes into hisown is the explanation of both human and non-human animal cognition.Here Avicenna’s most important contribution is the classification of five

distinct faculties, the so-called internal senses (al-h. awāss al-bāt.ina), thattake on the different activities of Aristotle’s phantasia.41 The classification isbased on three criteria explicitly stated by Avicenna: (1) each distinct type of

38 Shifā’: Fī al-nafs V.4, 227–228 Rahman; cf. Najāt II.6, 223–227. To be more precise, since Avicennadefines soul in relational terms as something that is in an animating relation to the body, we shouldsay that being a soul is accidental to the individual human essence.

39 Druart 2000, 262–263.40 Cf. Shifā’: Fī al-nafs I.5, 39–41 Rahman, with Ar. De an. II.2, 413a20–413b13.41 An excellent concise account of Avicenna’s system of internal senses is Black 2000, 59–62. For other

studies on some or all of the internal senses, see Wolfson 1973; Gätje 1988; Black 1993; and Hasse2000, 127–141. That Avicenna’s system of internal senses is a differentiation of the distinct activities ofphantasia is corroborated by Avicenna, al-Ta‘līqāt ‘alā hawāshin kitāb al-nafs li Arist.ū, 98 (cf. Averroes,Talkhīs kitāb al-h. iss wa al-mah. sûs, 39; and Tahāfut al-tahāfut [XVII].2, 546–547).


cognitive object requires a distinct faculty, (2) active and passive relations toan object require two respective faculties, (3) a receptive passive relation toan object is distinct from a retentive passive relation to it, and the tworequire two respective faculties.42 Since there are two types of objects,sensible forms (sing. sūra) and meanings inherent in them (sing. ma‘nā),there must be two corresponding classes of objects. Since there is bothpassive reception and active transformation of those objects, as exemplifiedby sense perception and creative imagination, there must be both passiveand active faculties. Finally, since both forms and meanings are not onlyreceived but also retained for subsequent retrieval, there must be bothreceptive and retentive faculties for both types of object. Thus, the appli-cation of the three criteria yields Avicenna a system of five internal senses:1. common sense (al-h. iss al-mushtarak) or fantasy (bant.āsiyā): reception of

forms2. imagery (khayāl) or the formative faculty (al-quwwa al-mutas.awwira):

retention of forms3. estimation (wahm): reception of meanings4. memory (dhikr) or the retentive faculty (al-quwwa al-h.āfiz· a): retention

of meanings5. the imaginative faculty (al-quwwa al-mutakhayyila) or, in human animals,

thought (fikr): separation and combination of forms and meanings.43

Our perceptual experiences are brought about through a co-operation ofthese faculties on the sense data provided by the five external senses (sight,hearing, smell, taste and touch). First, the common sense unites the varioussense data into a synaesthetic whole, which is then retained in the formativefaculty; thus, the formative faculty functions as a sort of memory for thesensible qualities of our experience. When the sense data have beenapprehended as a synaesthetic whole, the estimative faculty is capable ofapprehending the meaning carried in that whole. The exact nature of themeanings and the details of the consequent function of estimation are amatter of debate,44 but a reasonably uncontroversial case of estimativeapprehension is Avicenna’s commonly cited example of a sheep’s appre-hension of enmity in a wolf that it perceives. The enmity is not a sensible

42 Shifā’: Fī al-nafs I.5, 43–44 Rahman. Receptive and retentive faculties are distinguished because ofdifferent material qualities required in their respective organs: a faculty of reception must function bymeans of a relatively malleable organ, and a faculty of retention by means of a relatively rigid organ.

43 This is the standard classification as presented in Shifā’: Fī al-nafs I.5, 44–45 Rahman. For classi-ficatory and terminological variations of the system of internal senses in Avicenna’s works, seeWolfson 1973, 276–282.

44 Cf. Black 1993 with Hasse 2000, 127–141; I have presented my own view in Kaukua 2014a.


form, nor can it be reduced to any of the wolf’s sensible properties, and so itmust be a distinct type of object, a meaning. Nevertheless, the sheep mustsomehow apprehend the enmity in the particular wolf it sees or smells, andso the meaning must be closely connected to, indeed inherent in, thesynaesthetic perception of the wolf. These two features, non-sensibilityand inherence in what is sensible, are the two features definitive of theAvicennian meanings. Once apprehended, the meanings are retained in thefaculty of memory for subsequent retrieval. They are actively analysed andsynthesized by the imaginative faculty, which also makes connectionsbetween meanings and sensible forms.This brief account may give an inadequately atomistic impression of

Avicenna’s theory of the internal senses. To be sure, the distinctions are notmerely conceptual, for Avicenna locates each faculty in a distinct part of thebrain.45 But when he relates the activity of the faculties to their experientialcounterparts that he expects his interlocutors to be familiar with, he usuallydescribes their operation in strikingly holistic terms. For instance, therecognition of a perceived yellow substance as honey involves the retrievalof the corresponding meaning (‘honey’) from memory, an estimativeapprehension of that meaning, the retrieval of complementary sensiblequalities related to that meaning (for instance, sweetness) from the imagery,and the connection between the meaning and the related forms by theimagination.46 Apart from the perception of isolated simple qualities,perceptual cognition involves the system of internal senses as a whole.

Abstraction

According to Avicenna, all cognitive apprehension takes place through theacquisition of the form known, whether it is perceived or intellectuallyunderstood. If the form known is material to begin with, that is, if weperceive an individual human being for instance, her perceived form mustfirst be separated or abstracted from her matter. In fact, Avicenna builds hisentire theory of knowledge on this principle of increasing abstraction(tajrīd) of the form from its material features, classifying different types ofcognition on the basis of how much of the material residue remains in theknown form.47

45 Avicenna, Shifā’: Fī al-nafs I.5, 44–45 Rahman.46 This is my reconstruction of Avicenna, Shifā’: Fī al-nafs IV.1, 166 Rahman. For a more detailed

version, see Kaukua 2014a.47 Avicenna, Shifā’: Fī al-nafs II.2, 58 Rahman; cf. Najāt II.6, 207–208.


The type of cognition most intimately connected with matter is senseperception, because it can only take place at the actual presence of anextramental object that is perceived. This is only reasonable, since senseperception depends for its actuality on the causal influence of the extra-mental object; if the perceivable features of the person in front of me leaveno traces in my sense organs, there is no way I can perceive her. For the samereason, I can only perceive the person’s form as accompanied by a numberof features that are accidental to the form that makes her a human being; bynecessity she will stand in a certain place at a certain time, be of a certainlength, weight, complexion and so forth.48

When the perceived form is transferred from the common sense to theimagery, we arrive at the second stage of abstraction. The increased abstract-ness of the form retained in the imagery is not due to any change in itsqualitative features; the perceived person that remains in this faculty is stillspatiotemporally located and clothed in features due to her embodiment.But the existence of the cognitive form is no longer immediately dependenton the presence of the extramental person, for once acquired in the imagery,it can be retrieved for further consideration at will, without any externalcausal influence.49

The estimative apprehension of meanings is the next stage in the gradualabstraction of the cognitive form. Here, for the first time, we are dealingwith aspects of the person that are not reducible to her perceivable features –the mysterious feature that allows me to recognize a person I am alreadyacquainted with, despite all the interim changes in her appearance. Sincesuch recognition is common to us and non-human animals that lack theconceptual capacities proper to intellectual beings, it must have its basis in ashared psychological faculty, and the only Avicennian alternative is estima-tion. But although estimative apprehension of the form of the person isabove her perceivable features, it is not entirely abstract from them, for itcan only be apprehended as something that is carried by those features, orsomething inherent to them, separable in analysis but not in reality. I canrecognize the person and realize that my recognition is not due to anyparticular thing in her appearance, but I cannot recognize her without thatappearance.50

48 Avicenna, Shifā’: Fī al-nafs II.2, 59 Rahman; cf. Najāt II.6, 208.49 Avicenna, Shifā’: Fī al-nafs II.2, 59–60 Rahman; cf. Najāt II.6, 208–209.50 Avicenna, Shifā’: Fī al-nafs II.2, 60–61 Rahman; cf.Najāt II.6, 39–40. On estimation and this kind of

incidental perception, see Black 1993, 225–227.


The highest grade of abstraction in human cognition is that of intellec-tual apprehension of the form. Here the form is abstracted from all of itsaccidental features, which enables one to understand the essence of theperson, that which really makes the person the human being she is. Sincethis essence is separated from the accidental features that render an indi-vidual distinct from other individuals of the same species, the intellectualform is potentially shared by many perceived individuals, making it auniversal concept.51

To conclude, it is important to point out a crucial difference betweenintellection and the other types of human cognition. All cognition up to thelevel of estimation involves the abstractive activity of the human soul, andfollows a strictly bottom-up chain of causal conditions. Sense perceptiondepends on the activity of the extramental object, the soul’s retention of theform in the imagery depends on the prior existence of the form, and theestimative act of abstraction presupposes some synaesthetic whole thatcarries its proper object. But when we reach intellection, the soul becomesdependent on a higher principle, namely the active intellect which isrequired to emanate either an intellectual ‘light’ that enables the soul toapprehend the intellectual form in the object of the lower faculties,52 or theintelligible forms as such.53

51 Avicenna, Shifā’: Fī al-nafs II.2, 61; cf. Najāt II.6, 209–210.52 The metaphor of intellectual light is of course not particular to Avicenna; in this context, it is

ultimately derived from Ar. De an. III.5, 430a14–17.53 The question of where the soul’s activity ends and that of the active intellect begins is a matter of

debate. According to the traditional view, the soul simply receives intelligible forms when it issufficiently prepared (see, for instance, Goichon 1937, 309; Gardet 1951, 151; Rahman 1958, 15;Weisheipl 1982, 150; Davidson 1992, 93–94; Black 1997, 445), but Hasse 2001 (cf. also Gutas 2012)presents substantial evidence for the view that intellection depends more intimately on the soul’sabstractive activity, and that all that the soul receives from the active intellect is the light that enables itto “see” the intellectual form in the prepared object.


chapter 2

Avicenna and the phenomenon of self-awareness:the experiential basis of the flying man

Abū ‘Alī H· usayn ibn ‘Abd Allāh ibn Sīnā, the Latin Avicenna, is undoubt-edly the most famous of the philosophers writing in Arabic and in thecultural context of Islam.1 What is more, his fame is by and large deserved.By the death of the Persian polymath in 1037 following a series of failedattempts at curing an intestinal complaint, Arabic philosophy had reached alevel of maturity at which the ancient inheritance had been digested into asystem of thought that had proven, and would prove, capable of incorpo-rating indigenous ideas, both critical and constructive, and was strongenough to penetrate most areas of Islamic learning.2 In the subsequentcenturies, authors in search of the philosophical stance would increasinglyrevisit Avicenna before Aristotle, the first teacher.

In this regard, it is relatively uncontroversial to state that Avicenna was aphilosophical renovator. This, however, is not always readily apparent. Likemost of his philosophical contemporaries, he holds on to the Peripateticframework, with the Neoplatonic twist prominent in the late ancientphilosophical schools, and does not aim at a shift of the philosophicalparadigm. But within the received framework, he finds plenty of room toreformulate questions and responses, to develop arguments against novelproblems, and to integrate new phenomena into the system.

The focus of the present study is precisely on such a point of newconcerns. The type of self-awareness Avicenna introduces in his psychologydid not have a ready-made locus of discussion in the Peripatetic framework.And in fact, as witnessed particularly by some of the posthumously collectedcorrespondence investigated below, it did not always sit particularly well,and certainly not without significant conceptual revisions, in the systematic

1 As a result, there are several introductions to his life, thought, and cultural milieu. For two excellentplaces to start, see Gutas 1988 and McGinnis 2010.

2 For this “naturalization” of Avicennian thought, see Endress 1990, 30–37; Michot 2003a, 2003b; andWisnovsky 2004b.

30

niche Avicenna envisioned for it. In light of the reluctance and straightfor-ward opposition of some of his contemporaries, it seems warranted to saythat the concept of self-awareness Avicenna introduces, as well as many ofthe means of describing its foundational phenomenon, was a philosophicalnovelty. This is not to say that he instigated the discussion out of nothing.Meryem Sebti3 has unearthed arguments strikingly similar to Avicenna’s inthe so-called Kitāb mu‘ādhala al-nafs (‘The Book of the Castigation of Soul’,a work of late ancient provenance that was likely a part of the Hermeticcorpus),4 and in the early ninth century Mu‘tazilite Mu‘ammar ibn ‘Abbādal-Sulamī.5 However, the systematic integration of these ideas into thephilosophical tradition is, as far as I am able to tell, a genuinelyAvicennian move. Avicenna’s approach does have similarities with certainancient discussions, for instance in Galen,6 Plotinus and Augustine, but inthe absence of obvious textual predecessors connections will remainspeculative.

2.1 The purpose and basis of the flying man

Avicenna begins the discussion of the soul in the Shifā’ by arguing that thesoul is something the natural philosopher has to postulate due to the factthat she perceives life in the physical world about her. The animateprocesses and actions in which life appears to her – nutrition, growth,reproduction, perception and voluntary movement – are subject to such agreat variation that they cannot be explained merely by recourse to theaccount of the general types of natural motion as described in physics. Thus,there must be a specific agent behind those actions, and that is what thenatural philosopher calls soul (nafs).7

Having shown that there must be souls, Avicenna proceeds to considerthe various traditional attempts at the definition of the term ‘soul’. Theseconsiderations revolve around the necessity to make room in the definitionfor all the different kinds of entities that can function as souls. In thephilosophical tradition, the soul has been defined as the capacity (quwwa)to act in the animate body, the potency (quwwa) to receive cognitiveobjects, and as the form (s.ūra) of the animate body. For Avicenna, all ofthese definitions are too limiting. The two senses of quwwa both pick up

3 Sebti 2000, 118–119. 4 Kitāb mu‘ādhala al-nafs 14, 112.5 An argument based on self-awareness is attributed to Mu‘ammar in Jāh. iz· , al-Masā’il wa al-jawābāt lial-ma‘rifa 4, 51.

6 Cf. Walzer 1954, especially 255–256. 7 Shifā’: Fī al-nafs I.1, 4 Rahman.

The purpose and basis of the flying man 31

one aspect of the soul’s various responsibilities in relation to bodies butexclude others, and so neither can be a general definition applicable to allkinds of souls. Similarly, if we define the soul as a form in the strict technicalsense of a hylomorphic principle which informs matter by inhering in it, weend up excluding from our definition those entities that do not inhere inmatter but do function as souls in relation to it. Thus, if we want to includesuch obviously immaterial entities as the celestial souls in our definition, wecannot define the soul as a form.8

The definition Avicenna arrives at by excluding the alternatives is soul asperfection (kamāl) of the genus ‘living body’. The actually existing soulperfects the genus by first bringing it into existence through the differentiaspecific to the actual soul, that is, as a vegetal, animal or human body.9

Avicenna further specifies this definition by distinguishing between first andsecond perfection, or between ‘that through which the species becomes anactual species, like the shape for the sword’, and ‘the actions and passionswhich follow from the species of the thing, like cutting for the sword’.10

Since a living thing is alive already at the stage of first perfection, the soul inthe most general sense must be a first perfection. By means of theseargumentative moves Avicenna comes to define the soul as the ‘first perfec-tion of a natural organic body which performs the acts of life’.11

So far the moves, including the definition they have yielded as theirconclusion, are of course perfectly familiar. However, it is important to notethat for Avicenna ‘soul’ is a relational term, that is, it refers to the agentbehind animate actions insofar as it is the agent of those actions and there-fore has a relation to the body in which they take place. Thus, although thenatural philosopher has to postulate the existence of something that func-tions as a soul, she will not thereby have made any assumptions about thenature of or the category proper to that thing as it is in itself.12 If that thinghappens to be a substance independent of the body which it animates, beinga soul may be a mere accident to it, yet the natural philosopher will

8 Shifā’: Fī al-nafs I.1, 6–8 Rahman.9 Shifā’: Fī al-nafs I.1, 6–8, 12Rahman. Cf. I.5, 40Rahman, which clarifies that the vegetative soul is thegenus of the animal soul, and the animal soul the genus of the human soul. This sense of perfection isdiscussed in more general terms in Shifā’: al-Ilāhīyāt V.3.8–9, 164–166; and V.6.2, 175–176.

10 Shifā’: Fī al-nafs I.1, 11 Rahman. Robert Wisnovsky (2003, 120–127) has found three different thoughreconcilable distinctions between first and second perfection in Avicenna. The present context is theprimary locus for the distinction between form (first perfection) and function (second perfection).The other distinctions are between the principle of an activity (first perfection) and the activity itself(second perfection) and what can be considered foundational to the other two, between what isnecessary for existence (first perfection) and what is necessary for existing well (second perfection).

11 Shifā’: Fī al-nafs I.1, 12 Rahman. 12 Shifā’: Fī al-nafs I.1, 4–5 Rahman.

32 Avicenna and the phenomenon of self-awareness

nevertheless study it just in that regard, that is, as a soul that causes aparticular kind of motion in the body.13 But in spite of this qualification,Avicenna concludes the first chapter of the psychological section of theShifā’ by considering the question of whether the human soul as the thingwhich performs the animating acts proper to its relation with a human bodyis exhaustively grasped when it is understood as a soul. In other words, is thehuman soul, like the souls of plants and animals, a material form andtherefore reducible to its animating agency, or does it, like the celestialsouls, exist apart from these acts and independent of the body?A demonstrated answer to this question will have to wait until the last

book of the psychological section of the Shifā’. However, Avicenna decidesto lay his cards on the table from the very beginning by presenting a so-called ‘reminder’ (tanbīh) or a ‘pointer’ (ishāra) towards the correct view.

We have now come to know the meaning of the name bestowed on the thingthat is called soul due to a relation it has, and so we should strive toapprehend the quiddity of this thing which has become a soul by theaforementioned consideration; in this place we must point towards(nushīra) establishing the existence of the soul that we have by means of areminder (al-tanbīh) and a call for attention (al-tadhkīr), by a pointer thatwill be found apposite to the situation (ishāratan sadīdata al-mawqi‘) by onewho has the capacity to see the truth by himself without the need to educatehim, prod him onwards, and divert him from fallacies.14

‘Reminder’ and ‘pointer’ are closely related technical terms which denote acall for attention to something one is vaguely or inexplicitly aware of butwhich tends to elude one’s explicit attention. This ‘indicative method’15 canbe used instead of a proper demonstration for educational purposes in orderto have the student arrive at the necessary conclusion by herself and in thecourse of so doing understand the matter at hand more thoroughly thanwould be the case had the teacher given a ready-made demonstration thatcould be learnt by heart. But sometimes a reminder may be the onlyavailable method of argument, as for instance in the case of arguing forthe first principles of all intellection. Since such principles are foundationalto all acts of intellection, they cannot in turn be founded through ademonstration which relies on them. Instead, the only way to convince

13 Shifā’: Fī al-nafs I.1, 9–10 Rahman. 14 Shifā’: Fī al-nafs I.1, 15–16 Rahman.15 See Gutas 1988, 307–311, with a focus on ishāra. Although there may be a nuanced difference in

meaning between ishāra and tanbīh, with tanbīh carrying connotations of correction whereas ishārarather suggests that something is introduced that one has not hitherto considered, from the point ofview of the present study the two terms can be considered synonymous. In any case, both terms figurein our text without an obvious difference in meaning.


one’s interlocutor of their necessity is by indicating their presence in herthinking, that is, by reminding her that she in fact assumes their validityeven when adopting a sceptical stance towards them.16

The reminder Avicenna applies in our context is of course the famousthought experiment of the flying or the floating man.17 It is used to argue forthe initial plausibility of the substance dualist view that there might besomething to human being apart from embodied existence. In other words,it is intended to convince the reader who fails to see that the thing which is asoul for the human body can be something in itself, independent of itsrelation to the body, and it does this by showing that there is a feature in ourexperience that gives a clue of what the incorporeal existence could possiblyconsist in.

As a reminder the flying man is not presented as a demonstration or adefinitive argument for the human soul’s existence as an immaterial sub-stance. Avicenna clearly thinks that substance dualism can be properlydemonstrated, although not without a significant amount of additionalknowledge, part of which will be acquired in the course of the psychologicalresearch we are presently initiating.18 In that respect, the introductorychapter is not the proper place for a demonstrated answer. However, thereason to apply the method of reminding in the present context is verysimilar to that described in the case of the first principles of intellection.Avicenna uses the reminder to make us pay attention to something that isand has always been there for us but that we seldom take heed of. This newfocus of attention is then suggested to be relevant to the question of whetherthe human soul is merely a soul or whether it exists in itself apart from thisfunction.

These methodological remarks are important because once the flyingman is read in this light it will be hard to deny that the argument hinges onevidence that is supposed to be phenomenally or experientially given.Although the argument may be per impossibile, it is still used to point outsomething present to the reader’s experience instead of, say, a mere logicalnecessity or a transcendental condition for the possibility of something else.Indeed, later on in another instance of the flying man, Avicenna explicitly

16 Shifā’: al-Ilāhīyāt I.5.1–5, 22–23. In the context, Marmura translates tanbīh as ‘drawing attention’. Fora similar claim in a psychological context, see Shifā’: Fī al-nafs V.7, 257 Rahman.

17 Avicenna himself does not use the title, and as far as I have been able to ascertain it is only used insecondary literature. The earliest instance seems to be Gilson 1929, 41. The thought experiment comesup in several occasions and in many different works of Avicenna. For a seminal overview of threeinstances, see Marmura 1986. Other passages are introduced in Hasse 2000, 81–82.

18 The demonstration, which is based on the capacity for intellection inherent in the human soul, is firstpresented in Shifā’: Fī al-nafs V.2, 209–216 Rahman; that is, in the last book of the work.


says that the evidence this reminder points towards ‘is exemplified inmyself’(yakūna tamaththuluhu . . . fī nafsī), but I need to be alerted to this examplein myself because of my tendency to neglect it in favour of other, body-related contents of my experience.19

But in order finally to get to what it is that should be experientially givenhere, let us have a closer look at the reminder itself.

So we say: one of us must imagine (yatawahhama) himself so that he is createdall at once and perfect but his sight is veiled from seeing external [things], thathe is created floating in the air or in a void so that the resistance of the air doesnot hit him – a hit he would have to sense – and that his limbs are separatedfrom each other so that they do not meet or touch each other. [He must] thenconsider whether he affirms the existence of his self (wujūda dhātihi). He willnot hesitate in affirming that his self exists (li dhātihi mawjūdatan), but he willnot thereby affirm any of his limbs, any of his intestines, the heart or the brain,or any external thing. Rather, he will affirm his self (dhātahu) without affirm-ing for it length, breadth or depth. If it were possible for him in that state toimagine (yatakhayyala) a hand or some other limb, he would not imagine it aspart of his self (dhātihi) or a condition in his self (shart.an fī dhātihi). You knowthat what is affirmed is different from what is not affirmed and what isconfirmed is different from what is not confirmed.20 Hence the self (dhāt)whose existence he has affirmed is specific to him in that it is he himself (huwabi ‘aynihi), different from his body and limbs which he has not affirmed. Thus,he who takes heed (al-mutanabbih) has the means to take heed of (yatanab-baha) the existence of the soul (wujūdi al-nafs) as something different from thebody – indeed, as different from any body – and to know and be aware of it(annahu ‘ārifun bihi mustash‘irun lahu).21

The general idea of the argument is quite uncontroversial.22 Evidently, thethought experiment aims at an analytical distinction of an aspect of

19 Shifā’: Fī al-nafs V.7, 257 Rahman. Moreover, in his defence of the argument against Abū al-Qāsimal-Kirmānī’s critical remarks, Avicenna explicitly says that it is based on the interlocutor’s contem-plation of his own state, which is something not everyone is capable of (see Mubāh. athāt III.58–59Bīdārfar; Michot 1997, 170). The full importance of this point will be forcefully brought home oncewe get to explicate what Avicenna thinks self-awareness amounts to, for experiential familiarity withthe phenomenon will then be a crucial criterion in our reconstruction of his concept of self-awareness.Whatever other features the concept will have, it must refer to something readily available in our ownexperience.

20 I have chosen to follow Bakoš’ reading of al-muqarru bihi and yuqarru bihi (‘what is confirmed’ and‘what is not confirmed’, respectively) instead of Rahman’s al-maqraba and yaqrabuhu.Orthographically very similar, both readings are supported by manuscript evidence.

21 Shifā’: Fī al-nafs I.1, 16 Rahman; 18–19 Bakoš.22 Indeed, there is a widespread consensus about how the thought experiment should be conducted

(cf. Rahman 1952, 9–11; Galindo-Aguilar 1958; Arnaldez 1972; Marmura 1986; Druart 1988; Davidson1992, 83; Hasse 2000, 80–87), although there are almost as many views about its purpose as there arescholars. For a concise overview of the different interpretations, see Hasse 2000, 86.


experience by means of an imaginative bracketing of other aspects ofexperience which tend to hide the one Avicenna wants us to pay attentionto. Since he is interested in the question of whether we can find anyexperiential support for the idea that the human soul is independent fromthe body, and hence an immaterial substance, he starts by setting aside allthose aspects of experience that involve the body or that require embodi-ment. The most obvious of these aspects is perception, and that is why theflying man is carefully described to be in a position where none of his sensesis capable of transmitting any sensations to him. Even his sense of touch isrendered ineffective by ensuring that there are absolutely no impulses to theorgan of touch from either his own body or the surrounding air. Secondly,since the flying man is supposed to be created immediately to his state, hecannot have any imaginative, estimative or memorative aspects of experi-ence either. This is because all imagination, estimation and recollection arebased on the prior appearance of sensory data, which these so-called internalsenses (al-h. awāssu al-bāt.ina) then subject to their respective actions.23 Bythe same token, this excludes all actual intellection as well, since, inAvicenna’s empiristic epistemology, whatever human beings come toknow, they will have acquired by means of an abstractive operationperformed on sense perceptions, which thus are a necessary condition ofhuman intellection as well.24 Thus, the flying man has no objective contentof experience whatsoever, no acts of perception, imagination or intellection,nothing.

Having imagined this state, I must then ask whether there is anything leftto my experience. Avicenna states as obvious that my answer must beaffirmative: I will still experience my own dhāt to exist, and I will have toaffirm that this dhāt is me. In other words, I myself will still be there assomething I am aware of. This, for Avicenna, points towards substancedualism insofar as it gives us a concrete experiential sense of what theexistence of the thing that functions as our soul could possibly be inseparation from the body of which it is the soul and connected to whichit normally appears. When we consider the way in which Avicenna appliesmy awareness of myself as a pointer towards substance dualism, we have toconclude that he takes self-awareness to be a phenomenal feature of expe-rience, not a mere transcendental or logical condition. Thus, in order tofunction as a pointer, self-awareness has to be accessible to each and everyone of Avicenna’s interlocutors as an uncontroversial constituent of

23 Shifā’: al-nafs I.5, 43–44 Rahman; see Chapter 1.2, 25–27.24 For Avicenna’s cognitive psychology, see Chapter 1.2, 27–29.


experience, quite regardless of whether we ultimately agree that it has thesuggestive power in favour of the substance dualism that Avicenna buildsupon here.25

2.2 The validity and plausibility of the flying man

Although the ultimate goal of the present study is to understand what sort ofphenomenon Avicenna casts in the pivotal role here, we should brieflyconsider whether the argument hinging on it is convincing, or even formallyvalid. This is particularly the case because, as many scholars have pointedout,26 it seems to commit the rather blatant fallacy of proceeding from anepistemic or phenomenological distinction to a metaphysical one. For acontemporary reader versed in the philosophy of mind, this is obvious: thefact that the brain or the extended neural network of my body does notfigure in the phenomenology of my first-personal experience does clearlynot warrant the conclusion that my experience is metaphysically independ-ent of them, for it is perfectly possible, even likely, that my experience isopaque in the sense that mere introspection will never reveal its physicalfoundation.Indeed, if Avicenna intends the flying man to demonstrate the immate-

riality of the entity that functions as a soul in the human body, it is difficultnot to judge the argument as fatally fallacious. Moreover, there are soundreasons for thinking that he may have intended the argument as at least asketch for a complete demonstration. Ibn Kammūna, a thirteenth-centuryCE commentator, attempted to flesh it out into a syllogism,27 and we

25 Thus, my reconstruction of the argument’s procedure is inverse to Marmura’s (1986, 387–388),according to whom the flying man first makes us alert ‘to the existence of the self as immaterialand subsequently to the experiential knowledge of this immaterial existence . . . In other words, wediscern here two stages of knowing. The first is knowing that the self is immaterial, leading to thesecond, the experiential knowledge of one’s self as an immaterial entity.’ I find this order ofprocedure – that is, from a theoretical observation to experiential knowledge – problematic for tworeasons. First of all, it is difficult to see how a piece of personal experiential knowledge, whichMarmura suggests as the ultimate result of the thought experiment, would be a particularly relevantconclusion in the context of scientific psychology dealing with universal natural truths. Rather, itseems clear to me that Avicenna is interested in what the things that function as souls in human bodiesare in general, even if he applies the particular case of the interlocutor’s own soul as a means to reachthe general conclusion. But more importantly, although we may come to learn something about ourparticular selves in the thought process, the experiential level must have been there from the verybeginning as the basis of the entire argument. If there is no uncontroversial shared starting point topay attention to bymeans of the flyingman (a starting point which I argue Avicenna takes to be foundin all human experience), he will be hung in thin air, indeed.

26 For particularly lucid statements of this case, see Sebti 2000, 121–122; and Black 2008, 65.27 See Muehlethaler 2009.

The validity and plausibility of the flying man 37

cannot be certain that he was not thereby merely following a well-established line of interpretation. Furthermore, some of Avicenna’s owndescriptions of the ‘indicative method’ suggest that pointers and remindersshould indeed be so fleshed out by the apt reader.28However, if we allow theflying man to be a less stringent application of the indicative method, thelooming fallacy need not be fatal. In this reading, the argument is onlydesigned to point towards the substance dualism Avicenna will eventuallydemonstrate by other means, or, to put it another way, give a concrete ideaof what the existence of an immaterial human substance could be whenseparated from the body we commonly find it conjoined with. This wouldbe achieved not by painting a narrative picture of an otherworldly type ofexistence, but rather by pointing out a plausible candidate in perfectlycommonplace human experience. Avicenna could then meet the dissatisfiedreader accusing him of the fallacy by simply recognizing that the argumentdoes not work in her case and asking for some patience until the properdemonstration of the human substance’s immateriality in the fifth book ofShifā’: Fī al-nafs.29

But whatever the eventual verdict, the argument loses none of its rele-vance for our present purpose of reconstructing Avicenna’s concept of self-awareness – as long as it is agreed that some kind of self-awareness, theprecise nature of which remains to be determined, is pivotal to it. However,even this is controversial, for in one of the most extended studies ofAvicenna’s psychology, Dag Hasse has argued in extenso that the flyingman is not about self-awareness in the first place, and a fortiori not based ona phenomenal feature of experience. According to him, reading dhāt as ‘self’,although linguistically legitimate, is not warranted by the context. This isfirst of all because the flying man recurs in Shifā’: Fī al-nafs V.7, with anexplicit reference to the passage from chapter I.1 that we have just

28 Cf. Ishārāt, 1; and see Gutas 1988, 307–311.29 Two further fallacies have been detected in the flying man, neither of which is as striking as the one

above. The first of these is perceived in Avicenna’s procedure from the flyingman’s knowledge that heis to the knowledge of what he is, that is, from mere self-awareness to an awareness of the self as animmaterial substance, even though the flying man was supposed not to have any actual intellectualcontent (such as the concepts ‘immaterial’ and ‘substance’). Inmy view, this alleged fallacy is based ona confusion between what the flying man is supposed to be aware of and what the performer of thethought experiment concludes on that basis. The latter is of course not required to lack actualintellection or to be unable to apply concepts to the awareness paid attention to by means of thethought experiment. The second fallacy (noted by Marmura 1986, 388) is claimed to lie in Avicenna’sprocedure from a hypothetical example to categorical ends. This may well be a fallacy, but if it is one,it is common to all thought experiments, and as such far beyond the confines of the present study. Foran excellent extended discussion of the question of the validity of thought experiments in philo-sophical psychology, see Wilkes 1993, 1–48.


investigated but without a single occurrence of dhāt, having us insteadaffirm the existence of our ‘thatness’ (annīya), the bare fact that we arethere.30 Hasse suggests that this is also the sense in which we shouldunderstand dhāt in I.1, and so ‘the common denominator of the twowords is something unspecific like “core being”.’31 Secondly, Hasse thinksthat since the context is concerned with what the soul is in itself, that is, itsessence, in distinction from the soul-relation it has to the body, we shouldread dhāt as denoting ‘essence’.32 Thus, the text is concerned with theaffirmation of an essence, or rather the existence of an essence, independentof the body to which it is related.33 Hasse explicitly addresses the suggestedpossibility that Avicenna’s use of dhāt is ambiguous and that the wordmeans both ‘essence’ and ‘self’, but argues that it univocally stands for‘essence’: ‘To conclude, the Flying Man does not have “immediate access”to himself, nor is he “conscious of his existence” or “fully aware of hispersonal existence”, nor does he “affirm his existence”, but he affirms theexistence of his core entity, his essence, while not affirming the existence ofhis body.’34

Obviously, if Hasse’s categorical denial of the relevance of self-awarenessto the flying man is correct, my claim that in the argument Avicenna relieson an explicit phenomenon of self-awareness will be completely unfounded.But let us pause to consider what argumentative power the thought experi-ment will have once we take the evidence of immediate self-awareness out ofthe equation. Having bracketed all objective content of experience, whatgrounds do I have for claiming that I am aware of my essence? What wouldawareness of essence in this case consist in? It cannot be awareness of theconcept of human being, or any other concept my essence can be subsumedunder, because all intellection was ruled out in the construction of thethought experiment. Since alternatives other than intellection are evenharder to come by, it seems that without a reference to some kind ofself-awareness, to the fact that I will still be there even though I am notaware of my body or indeed anything else at all, the thought experiment isunconvincing at best, downright incoherent at worst. This is not to say thatmy interpretation pushes Avicenna’s attempt to argue for his substance

30 The passage in question is Shifā’: Fī al-nafs V.7, 255 Rahman. I will quote and discuss the passagebelow in Chapter 4, 66–71.

31 Hasse 2000, 83.32 Hasse 2000, 83. This of course accords with the frequent technical use of the term in philosophical

Arabic. See Appendix, as well as Rahman 1991 and Goichon 1938, 134–139.33 Hasse 2000, 83–84.34 Hasse 2000, 86. The quotes of the interpretations Hasse rejects are, respectively, from Druart 1988,

34; Davidson 1992, 83; Pines 1970, 808; Rahman 1952, 10; and Marmura 1986, 387.


dualism beyond debate, but I do claim that the more charitable interpreta-tion will in the very least retain some initial plausibility for the argument.

On the other hand, although the word dhāt does not appear in the flyingman of Shifā’: Fī al-nafs V.7, the context of the argument’s application inthat chapter makes the connection to self-awareness, if possible, even moreprominent. The pressing question here is how to account for the fact thatour experiences tend to be coherent and unified wholes of complex con-stituents in face of Avicenna’s stringent distinctions between the cognitiveand conative faculties responsible for those constituents. In other words,how is it possible that I smell freshly brewed coffee, develop a desire for acup and stand up to fetch one, if the olfactory perception, the desire and themuscular movement are acts of distinct and unconnected faculties, whichproduce distinct and unconnected constituents of experience? Avicenna’sstrategy in the solution is to argue for the necessity of a single subject oragent behind the distinct acts, the inviolable unity of which will be sufficientto connect the acts to the sort of whole we recognize in our experience. Hegoes on to characterize ‘this single thing in which these faculties are con-joined’, the human soul, by appealing to the intuition of his interlocutors,saying that it is ‘the thing that each of us sees as himself (al-shay’u al-ladhīyarāhu kullun minnā dhātahu)’.35 In this general context, the flying manfigures in a more specific argument against the claim that this ‘thing thateach of us sees as himself’ is the body as a whole: since one can be aware ofthe unifying subject that one is oneself without being aware of the body (asthe flying man is supposed to show), the body cannot be the unifyingsubject. Thus, contrary to what Hasse claims, once we have a look at thelarger context of Shifā’: Fī al-nafs V.7, we see that Avicenna explicitly refersto self-awareness, and precisely in a sense which he clearly believes isperfectly familiar to his readers from their own experience, in the immediatecontext of the flying man.

The fact that Avicenna speaks of knowing the existence of one’s annīya inthe flying man in Shifā’: Fī al-nafs V.7 need not be a problem either, forannīya refers to the fact that (anna) a thing exists and can thus be translatedas ‘thatness’.36 As a technical term, it is diametrically opposed to māhīya,which is derived from the question mā huwa (what is it?) and can be

35 Shifā’: Fī al-nafs V.7, 253 Rahman. The point is made more emphatically after the flying man at256–257. I discuss the entire context in detail below in Chapter 4, 66–71. On the contexts of the twoversions, see Marmura 1986, 385–390.

36 See d’Alverny 1959, 73.


translated as ‘quiddity’ or ‘whatness’.37 Now, insofar as an essence that canbe known is an answer to the question what and therefore a quiddity, itwould seem that Hasse’s argument for excluding the interpretation of dhātas ‘self’ in the flying man of Shifā’: Fī al-nafs I.1 falls through. For certainlythe immediate awareness of the fact that one exists is cognitively lessdeveloped than the knowledge of what one is by essence. Moreover, know-ing that one’s essence exists certainly requires some sort of cognitive grasp ofthe essence, which brings us back to the problem we started with, that is, ifwe conceive of this grasp as a normal case of intellection, it will beincoherent with the peculiar conditions posed in the thought experiment.Thus, I argue that Hasse’s arguments are insufficient to rule out the most

obvious interpretation of the flying man, that is, as an argument whichhinges on the phenomenon of self-awareness. The fact that dhāt also meansessence is something that Avicenna may not have seen necessary to rule out,for as I will argue later on, there is a sense in which awareness of oneselfindeed is the existence of an individual essence. In light of texts in which thisidentification is most explicit,38 the question rather becomes whetherAvicenna even recognized a need for a rigorous distinction between thetwo senses of dhāt. If self-awareness is awareness of the very core of one’sbeing, and if the very core of the being of anything is its essence whichprevails unchanged in the flux of the various accidents appended to it, thenthe two indeedmerge into one instead of excluding each other. In this sense,I am in agreement with Hasse’s conclusion: the flying man does hinge onaffirming the existence of one’s essence as separated from the body. Thedifference is that I fail to see how that conclusion can be reached without thepivotal identification of awareness of that essence with self-awareness.But in order to not run too far ahead, let us briefly summarize the main

argument of the present section. I have claimed that a close reading of thethought experiment known as the flying man shows us, first, that Avicennarecognizes the phenomenon of self-awareness as something we all canrecognize in our own experience. But what is more, it shows thatAvicenna thinks this phenomenon has considerable psychological rele-vance. Indeed, considering the fact that self-awareness figures prominentlyin the introductory chapter of the psychological section of Avicenna’s most

37 In this light Avicenna’s formulation that the flying man knows ‘the existence of his thatness’ mayseem strangely redundant, for it could be explicated as knowing the existence of the fact that oneexists, which is one step removed from knowing one’s existence. However, I believe this can beexplained by recourse to the frequent use of annīya in reference to the individual thing instead of itsquiddity. Cf. d’Alverny 1959, 80–81.

38 Cf. Ta‘līqāt, 160–161. This text is discussed in detail in Chapter 3.2.


extensive philosophical summa, and that it provides the starting point forthe investigation of the soul in perhaps his most independent mature work,al-Ishārāt wa al-tanbīhāt,39 he must have perceived it as not having receivedits proper due from his predecessors in philosophical psychology. So far,however, we do not know what exactly Avicenna thinks self-awarenessconsists in, how it should be described, or whether it can be analysed orexplained by recourse to more basic phenomena. All we have found out isthat he believes that such a phenomenon can be determined and pointed to,and perhaps even expects it to be relatively uncontroversial once it has beensufficiently determined and distinguished from other features of commonexperience by such means as the flying man. Indeed, if we recall thesuggested similarity between self-awareness and first intelligibles, it shouldbe as obvious as the laws of the excluded middle or non-contradiction.Avicenna also suggests that as such a basic fact of human mental life, self-awareness provides a pointer towards the truth of psychological substancedualism. In fact, it can be argued that he may have cast it in an even moreprominent role as an indispensable foundation for a coherent dualism.

39 The third namat. of the second part of the Ishārāt (119 ff.), on the soul, begins with the flying man. SeeChapter 4.2, 80.


chapter 3

Self-awareness as existence: Avicenna on theindividuality of an incorporeal substance

3.1 The problem of incorporeal individuality

Avicenna grounds his demonstration of psychological substance dualism ona peculiar feature of the human capacity for intellection.1 All objects ofintellection, insofar as they are actually understood, are indivisible.Although one can analyse the concept of human being, for instance, intoits generic and differential constituents (‘animal’ and ‘rational’, respec-tively), when the concept is actually understood, the constituents mustboth be actual in a single undivided act of intellection. Moreover, theanalysis cannot be pursued infinitely for it will ultimately come acrosssimple intelligibles, such as ‘thing’ or ‘existent’, which can no longer beanalysed or defined.2 On the other hand, all corporeal entities are infinitelydivisible. Thus, if a corporeal organ were the substrate of the intellectualsoul, the soul would be divisible due to the divisibility of its substrate. Thisin turn would entail the infinite divisibility of the object of intellection thatinheres in the soul at the moment of actual intellection, which has alreadybeen argued to be impossible. As a result, insofar as we are capable ofintellection, we cannot be divisible, and hence are not corporeal.3

1 Avicenna also refers to self-awareness after this demonstration (see Shifā’: Fī al-nafs V.2, 216–218Rahman), when he argues that our awareness of perceiving (Aristotle’s perception of perception inDean. III.2, 425b11–26) cannot be due to the corporeal organs of perception but requires an incorporealbasis. Importantly, though, this proof is based not on the evidential force of self-awareness, but on theimpossibility of finding the kind of immediate self-relation, which awareness of perceiving requires, inanything corporeal.

2 See Shifā’: al-Ilāhīyāt I.5.2–21, 23–27.3 Shifā’: Fī al-nafsV.2, 211–216Rahman; cf. Ishārāt, 130. Earlier on (209–211) Avicenna also considers thesuggestion that the body in which the intelligible inheres is an unextended and therefore indivisiblepoint. This is impossible because as a limit of a line or a magnitude the point is accidentally extended,hence divisible, and will transfer this accidental divisibility to the intelligible. The point’s mind-independent existence as such (that is, without anything extended the limit of which it woulddesignate) is refuted by means of familiar arguments against atomism.

43

Avicenna’s dualism is thus ultimately based on the traditional view thatintellectuality entails incorporeality.4 But since he argues for a strong unityof the soul despite the multiplicity of its faculties, Avicenna thinks thatincorporeality is not exclusive to an intellectual ‘part’ of the human being.On the contrary, each of us is incorporeal even when considered as a soul,that is, as the agent of acts that take place in the body. Since we haveexperience of remaining the same entity when we think intellectually andwhen we perceive, desire or move our bodies, Avicenna concludes that thesoul behind all these acts is one, and differentiated only by means of itsfaculties or capacities.5 With the exception of intellection, the soul’s use ofits faculties of course does take place by means of respective corporealorgans, and so the acts are corporeal, but since the agent remains one andthe same from one act to another, and since this one agent is capable ofintellection, it must be incorporeal.

Problems arise when this dualistic concept of the soul is connected to theAristotelian metaphysical framework that Avicenna subscribes to.6 As asubstance dualist concerning human beings, Avicenna departs from astraightforward interpretation of hylomorphic psychology in which thesoul is simply the form of the body; although the soul still retains a formalfunction in relation to its body, it is not a form strictly speaking, since itdoes not inhere in the matter of the body. This departure from orthodoxAristotelian doctrine is decisive, for it leads to a dilemma concerning theindividuation (tashakhkhus.) of human beings. The commonplaceAristotelian account of individuation explains individuality by means ofmatter: matter individuates forms as ostensible things, as unique examplesof primary substances each of which is traditionally referred to as ‘a this’,tode ti. In other words, the primitively determined spatiotemporal co-ordinates afforded by matter are the ultimate guarantee of the individualityof each individual entity, since, unlike all the entity’s other propertiesconsidered in themselves, those co-ordinates are exclusive to it in a primitivesense: no two entities can reside in the same place at a single given momentof time. But in the case of immaterial substances Avicenna cannot rely onmatter to account for their individuality.

Avicenna tackles the problem headlong in Shifā’: Fī al-nafs V.3, sayingthat the emergence of a suitable body is a necessary condition for the

4 See Shifā’: al-Ilāhīyāt VIII.6.6–7, 284–285. This view is explicitly formulated already in Ar. De an.III.4. For a lucid discussion of the view in the Ishārāt and the commentaries by Rāzī and T. ūsī, see nowAdamson 2011a.

5 Shifā’: Fī al-nafs V.7, 253–254 Rahman.6 The following reconstruction of the problem is based on Black 2012.

44 Self-awareness as existence

emergence of an individual soul. Thus, the soul’s individuality requires arelation to a body proper to it and it alone.

It is thus sound that the soul comes to be as corporeal matter suited to beused by it comes to be, and the body that has come to be is its instrument andgoverned by it. In the substance of the soul that has come to be with a body –that body necessitates its coming to be from the first principles – there is aconfiguration of natural tendency to be preoccupied with and use [thatbody], to be concerned about its states and to be attracted to it, which isproper to [the soul] and turns it away from all other bodies. Thus there is nodoubt that when [the soul] exists as individuated (mutashakhkhis.a), theprinciple of its individuation (tashakhkhus.ihā) attaches to it somethingfrom the configurations that designates it as an individual (min al-hay’ātimā tuta‘ayyanu bihi shakhs.an). Those configurations are necessary to make[the soul] proper to that body, and they are in accordance with the mutualsuitability of one to the other, even if that state and that accordance wereunknown to us.7

Thus, human individuals first occur at the occurrence of suitable bodies.When the embryo has developed into a foetus as part of the mother’s bodygoverned by the mother’s soul, and the foetus in turn has reached a stage inwhich it is ‘a natural organic body which performs [or is capable of perform-ing] the acts of life’, a respective soul to govern it and it alone emerges fromthe active intellect. But having said all this, Avicenna immediately goes onto consider the question of the body’s corruption. Given that the individ-uality of the soul is due to its body, does the soul cease to exist as anindividual entity when the body is corrupted? In other words, since thesoul’s individuality requires a relation to a body proper to it, and since anyrelation will cease to prevail when one of its relata ceases to exist, how canAvicenna account for the existence of the human soul post mortem? Adenial of the afterlife would be problematic not only for religious reasons,but also on systematic grounds, for given that matter is the precondition forcorruption, and that the soul is only relationally connected to its body, it ishard to see how the immaterial soul could cease to exist. To avoid thisproblem, Avicenna suggests that the relation to the body has to be under-stood as a property of the soul, as a temporally qualifiable ‘being-related-to-the-body’ that can be the soul’s property whether or not any body actuallyexists: ‘We say that after the separation of souls from bodies, there is nodoubt that every single soul will have existed (takūna qad wujidat) as a dhātmade singular by the difference of the matters which were (kānat), by the

7 Shifā’: Fī al-nafs V.3, 224–225 Rahman.

The problem of incorporeal individuality 45

difference of the times of their coming to be, and by the difference of theirconfigurations in accordance with their different bodies.’8 As a property ofthe soul, the relation to the body can receive different temporal qualifica-tions, and so, although ultimately grounded upon the existence of a bodyfor the soul at a determined moment in time, that property can belong tothe soul at any subsequent moment of time, even when the body nolonger exists. Thus, the soul will be individuated after the corruption ofthe body by its having once been related to that particular body at aparticular time.

What Avicenna thereby ends up with is a theory of individuation bymeans of a special property or a bundle of properties that are uniqueto the immaterial entity that they individuate. In fact, in the imme-diately following section of our chapter he is quick to list a set of suchproperties:

After [the soul] has been individuated as singular, it and another, numeri-cally [different] soul cannot be one dhāt – we have repeated the discussionon the impossibility of this in a number of places. But we are certain that[1] the soul can, when it comes to be with the coming to be of a tempera-ment, afterwards come to have a configuration in rational actions andrational passions that as a whole are distinct from the correspondingconfiguration it would have in another, as two temperaments in two bodiesare distinct; that [2] the acquired configuration, which is called an intellectin act, is also to a certain degree something by which [the soul] is distin-guished from another soul; and that [3] an awareness of its particular self(shu‘ūrun bi dhātihā al-juz’īya) occurs to [the soul], and that awareness isalso a certain configuration in it which is proper to it and to no other. Itmay be the case that [4] a proper configuration also comes to be in it withrespect to bodily faculties. That configuration is connected to moral con-figurations or is the same as them. There [may] also be [5] other propertiesunknown to us that are appended to souls when they come to be andafterwards, just as similar [properties] are appended to individuals ofcorporeal species so that [the individuals] are distinguished by them foras long as they remain; similarly souls are distinguished by their propertiesin them, whether the bodies are there or not, whether we know those statesor not, or know [only] some of them.9

Avicenna mentions five types of individuating properties. First we have a setof ‘rational actions and . . . passions’, which seems to refer to immaterialcharacter traits – or in modern parlance, psychological properties – thatcorrespond with the unique humoral temperament of the soul’s erstwhile

8 Shifā’: Fī al-nafsV.3, 225–226Rahman; emphases added. 9 Shifā’: Fī al-nafsV.3, Rahman 226–227.


body.10 These properties are based on the Galenic theory of temperaments,according to which the relations of domination between the bodilyhumours – blood, phlegm, yellow bile and black bile – cause correspondingemotional dispositions in the soul; for instance, a predominance of blackbile tends to make a person melancholic in character. Given Avicenna’srefutation of atomism, the relative amounts of the humours in a giventemperamental whole allow for an infinity of variations, and so the temper-amental basis alone provides a means of distinguishing an infinite numberof souls from each other.11 Second, each human soul will have developed itsintellect to an extent particular to it, and thereby reached a degree ofcognitive perfection different from that of other souls; each of us hasmore or less knowledge, and knowledge about different things than ourpeers. Our cognitive perfection, although it may seem to be independentfrom the body, also requires the body in its coming to be, for, as we havealready mentioned, Avicenna holds that we acquire knowledge by means oflearning and abstraction through perceptual data. Third, Avicenna men-tions the awareness each person has of her- or himself. We will return to this‘configuration’ in a moment; suffice it to say at this point, however, that, aswe learned from the flying man, Avicenna thinks that self-awareness, unlikethe other properties enumerated here, is at least possibly independent fromthe body. Let us also point out that, unlike the other possible factors ofindividuation, self-awareness is not qualified by any verbal sign of hesitation(such as ‘can’, ‘to a certain degree’ or ‘it may be the case’). Fourth, each of ushas particular character traits that we have acquired through habituation.One person may have been disposed to regular physical hardship from anearly age, having thus developed a character of strong endurance, whereasanother person may have been similarly subjected to a steady diet ofsweetmeats and will subsequently find a life devoid of such substancesdifficult to bear. Avicenna refers to these character traits as ‘moral config-urations’ or configurations related to character (al-hay’ātu al-khalqīya), andit seems quite clear that they derive from the account of dispositions(Gr. hexeis, Ar. akhlāq) in the Nicomachean Ethics.12 Fifth and last,

10 For an extended discussion of the role of the humoral temperament in the individuation andorigination of the immaterial human substance, as well as of the general emanationist context, seeMarmura 2008, especially 123–130.

11 For Avicenna’s refutation of atomism, see Shifā’: al-Samā‘ al-t.abī‘ī III.5, 302–310. Concerning theinfinite number of souls in Avicenna, cf. Marmura 1960, 232–235.

12 For a general account, see Ar.NE II.1. Avicenna discusses habituation as a psychological phenomenonin the context of the faculty of estimation (see Shifā’: Fī al-nafs IV.3, 184–185 Rahman; for discussion,cf. Kaukua 2007, 51–52). Thus, even though the habits themselves are in the immaterial soul, theircoming to be requires corporeal faculties.


Avicenna mentions the possibility of other individuating properties that wemay not be aware of. Since each body has a practically innumerable varietyof properties, there may be some that we will miss; if that is possible, thesame may be true of the soul’s properties. Presumably, however, this fifthgroup will be ontologically similar to the others, in that the factors in it arealso properties of the soul somehow caused by its relation to the body. Theimportant point here is that although they are genetically based on thebody, all of these properties are properties of the soul, and as properties of animmaterial entity they are immaterial as well.

Avicenna therefore seems to suggest a version of what contemporarymetaphysicians call a bundle theory to explain the individuality of theimmaterial human substance. In this account, no single property alone isexpected to provide the foundation for human individuality, but theproperties due to the soul’s relation to its body are collectively thought tobe de facto unique to the soul, and thus capable of singling it out fromamong its kin. Now, this leap from de facto uniqueness to individuation inprinciple entails a well-rehearsed problem: if no single immaterial property iscapable of conveying individuality to the subject whose property it is, whyshould we assume that a collection of such properties will acquire thatcapacity from the mere conjunction of the properties? What is more,Avicenna himself makes the very point explicit in Shifā’: al-Madkhal I.12:13

The individual (al-shakhs.) becomes an individual when accidental properties(khawās.s.), both concomitant and not concomitant, are conjoined with thenature of the species, and designated matter is assigned to them (tata‘ayyanulahā māddatun mushārun ilayhā).14 It is impossible for the intelligible proper-ties, howevermany they are, to be conjoined to the species without an ultimatereference to an individuatedmeaning, so that the individual thereby subsists inthe intellect. For if you said ‘Zayd is the tall one, the writer, the handsomeone’, and so forth for as many descriptions as you wish, Zayd’s individualitywould not be assigned for you in the intellect. Rather, the meaning, which isput together from the collection of all that, can belong to more than one, but itis made concrete (yu‘ayyinuhu) by existence and by reference to an individualmeaning, just as you say that he is the son of so-and-so, exists at such-and-sucha time, is tall, and is a philosopher, and it then happens to be the case that no

13 The incoherence was first raised by Deborah Black (2012), who also suggests in passing that self-awareness may have played a special role in Avicenna’s account of human individuality (see also Black2008, 73–76). The following reconstruction, however, is not based on Black’s article. Anotherassessment of Avicenna’s account of the individuation of human substances as being incompleteand possibly aporetic is Druart 2000, 266–267, 273.

14 One could also read ‘designated matter is assigned to [the nature of the species]’. The nuance isinconsequential for our concerns, since in both cases the point remains that matter is required forindividuation.


one shares these attributes with him and also that you have had a previousacquaintance with the case bymeans of the kind of apprehension which pointstowards [the case] on the basis of sense perception, pointing to the very personand the very moment. Then the individuality of Zayd is ascertained for you,and this statement indicates his individuality.15

Right at the beginning of the passage, Avicenna flatly rejects the bundletheory of individuation that we have just read from the psychologicalsection of the Shifā’. He is adamant that in addition to the conjunction ofaccidental properties to a specific nature – i.e. the conjunction of properties(1), (2), (4) and (5) to the dhāt of a human being – individuation requiresthat ‘designated matter is assigned’ to this bundle of accidental and essentialproperties. However, precisely that assignment is lacking when the bundleis supposed to constitute an immaterial entity. As a result, the allegedindividuation turns out to be insufficiently grounded; at best we mighthave accidental individuality due to the de facto non-existence of anotherentity with exactly the same bundle of properties. But without the irredu-cible material basis, there is nothing to prevent the bundle from belongingto more than one soul in principle.The incongruence hinges on a close connection between intellectuality,

intelligibility and immateriality. Something is an intellect for the samereason as it is intelligible: because it is abstract, either by itself or as a resultof a human intellect’s act of abstraction, from matter and the featuresconcomitant to it, such as having unique spatiotemporal co-ordinates.Furthermore, intelligibility entails potential universality; any intelligible isin principle shareable or participable (mushtarik) by more than one specif-ically identical individual, regardless of whether there actually are more thanone, or indeed any, corresponding individuals.16 Thus, even if the bundle ofthe accidental properties of a given human substance did happen to con-stitute a complete description of a given individual entity, a Leibniziancomplete individual concept if you will, they would not suffice decisively tosingle out that individual, since there will always remain the possibility thatthere is another individual exactly alike but numerically different. What isneeded is a way for this set of properties to refer to an individual, and so tohave an individuated or individual meaning (ma‘nā mutashakhkhas., ma‘nāshakhs.ī),

17 which in the case of Zayd in our passage amounts to an ostensive

15 Shifā’: al-Madkhal I.12, 70. Cf. Shifā’: al-Ilāhīyāt V.8.8–9, 188–189; and Bäck 1994, 48–50.16 Shifā’: al-Madkhal I.5, 26–27; cf. Shifā’: al-Ilāhīyāt V.1.2–3, 148–149.17 The notion of ‘individuated meaning’ or ‘individual meaning’ is not entirely clear, and according to

Black 2012, Avicenna never gives an explicit definition of the term. Black does venture to suggest,however, that individual meanings are involved in Avicenna’s account of self-awareness – a suggestion


reference to the embodied person. This reference enables all the other factsknown about Zayd to be specific to him and him alone.

One might ask, however, whether the incoherence is not dissolved if wetake our cue from the existence proper to a class of immaterial individualswhich Avicenna explicitly asserts, namely the celestial intellects. If they can begenuine individuals, why can we not explain the individuality of the imma-terial human substances along the same lines? The problem is that the types ofindividuality respective to the two classes of being, human and celestialintellects, are different in kind and accountable by drastically divergentreasons. The celestial intellects are individual because each of them is thesole instantiation of its species,18 whereas human substances are multipleinstantiations of the single species of humanity. Thus, although both celestialintellects and human beings are individual for causes external to their quid-dities,19 these reasons are distinct: the lack of other instantiations of thespecies in the case of the celestial intellects, the aforementioned relation tobodies in that of humans. What is more, Avicenna explicitly states that theallegedly individual bundles of the human essence and its unique accidentscannot be understood as subspecies of human: only the human essence canconstitute a species that is capable of subsisting by itself.20 This is clearly anexpression of the Aristotelian belief according to which only natural kinds aresubstances in the strict sense, that is, entities that are capable of subsisting andenduring through the variation of their accidental attributes, and that are thesubjects of definitions in the proper sense.21 Since the variation concernsprecisely the sort of features that form the allegedly individuating bundle, thebundle fails to produce a new species of substance, the sole instantiation ofwhich the individual human would be.

Thus, if we look at the list of the allegedly individuating characteristicsAvicenna mentions in Shifā’: Fī al-nafs V.3, we can see that the majority ofthem suffers from the same problem as the bundle of properties discussed inShifā’: al-Madkhal I.12. If the rational actions and affections that are parallelto the humoral temperament of the soul’s body are immaterial properties,we can immediately ask the question of why they could not in principle beshared by other souls. Even if they were caused by a particular body, it

I will discuss in Chapter 4.3, 91–94. For the present purpose it suffices to point out that, in the case ofmultiple individuals of a single species, individuated meanings seem to presuppose material existence.Indeed, it is possible that Avicenna applies the term simply to mean straightforward reference to amind-independent object. For relevant fluctuations in the meaning of ma‘nā in theories of referencein Arabic philosophy, see Eichner 2010.

18 Shifā’: al-Ilāhīyāt V.2.2, 158; VIII.6.16-17, 288. 19 Shifā’: al-Madkhal I.5, 26–27.20 Shifā’: al-Madkhal I.12, 71–72.21 See, for instance, Ar. Met. VII.12, 1037b27–1038a35; and cf. Shifā’: al-Ilāhīyāt V.7.11–12, 184–185.


would still be possible that there is another soul with precisely the sameproperties yet due to another body that is identical to the first in respect toits humoral temperament. In such a case, we would not have any means ofdistinguishing between the two souls. The soul’s moral configuration issimilar to its humorally induced properties in this regard: certainly we canimagine two souls with exactly matching histories of character formation,only mediated by two spatiotemporally distinct bodies. Again, in such a casewe will have no way of distinguishing between the two souls. Similarproblems are faced when we consider the degree of actualization of thesoul’s capacity of intellection: nothing rules out the possibility of anothersoul having the exact same amount of acquired knowledge. If anything, thismight seem even more plausible than the possibility of two souls withexactly similar humorally induced character traits. The set of unknownattributes, whatever they may be, can be dealt with in the same way: asimmaterial qualities, they are by definition potentially common to manyand therefore have no power to individuate the subject whose qualities theyare. In the end, the problem stands firm: without the relation to its body,which ceases to prevail at the body’s corruption, the human substance lacksa reference to designated matter and its primitively individuating spatio-temporal co-ordinates, and its individuation remains a mystery.

3.2 Self-awareness as incorporeal existence

Even if none of the body-related properties were capable of individuatingthe immaterial human substance, Avicenna’s list leaves us with one possiblefactor to consider: the soul’s ‘awareness of its particular self’ which is aconfiguration ‘proper to it and to no other’. The role of self-awareness in theindividuation of a human being is not developed at any greater length in thepsychological books of the Shifā’, but the Ta‘līqāt, a later compilation ofcorrespondence possibly meant to be appended to the Shifā’,22 does containan extended discussion of self-awareness which perhaps suggests a solutionto the problem of individuation. Many of the points this passage gatherstogether are made separately, and are occasionally expanded upon, in anumber of other contexts, but the way they are here composed into a singlewhole is unique. Moreover, the emphatic repetition, in slight variations, ofthe point that self-awareness is the mode of existence of the humansubstance is to my knowledge without parallel in the Avicennian corpus.It is also the basis on which the passage can be read as suggesting a solution

22 So Gutas 1988, 141–144, whose suggestion is corroborated by Janssens 2012.

Self-awareness as incorporeal existence 51

to the incoherence in the Shifā’. For these reasons I will quote the passage atexceptional length. In order to see how the arguments that figure scatteredin Avicenna’s other works do indeed belong together, it is important not todissect the text into independent parts. (The lettering is added only tofacilitate reference in the discussion below.)

[a] Furthermore, when an instrument is attributed to something, [that thing]acquires by means of [the instrument] what it has potentially, not actually.The self’s self-awareness (shu‘ūru al-dhāti bi al-dhāti) is never23 potential butrather innate (maft.ūra) to it; the human self is an aware self (dhātu al-insānidhātun shā‘iratun), and its awareness of itself (shu‘ūruhā bi dhātihā) is naturalto it. Since that is the case, it is not acquired; and since it is not acquired, it isnot by means of an instrument.

[b] Self-awareness is essential for the soul (al-shu‘ūru bi al-dhāti dhātīyun lial-nafs), it is not externally acquired; it is as if when the self occurs, awarenessoccurs with it (wa ka annahu idhan h. as.ala al-dhātu, h. as.ala ma‘ahā al-shu‘ūr).One24 is not aware of it by means of an instrument, but one is aware of it byitself and from itself (yush‘aru bihā bi dhātihā wa min dhātihā). Its awarenessof it25 is absolute awareness; I mean that there is no condition whatsoever init and that it is constantly aware (dā’imatu al-shu‘ūr), not from time totime.26 Apprehension of the body takes place by way of a sense, and that iseither by means of vision or by means of touch; thus, he who allows thatknowledge of the self (al-ma‘arifatu bi al-dhāt) is from an indication to it bymeans of a sense has the consequence that he does not know himself (lamya‘arif dhātahu) absolutely but knows [himself] (‘arafahā)27 when he per-ceives his body. Furthermore, apprehension by means of a sense requires thatthere is something which is known to apprehend what is sensed by means ofa sense, and which is different from the sense, and it is no doubt the soul. Asregards28 us being aware that we were aware of ourselves (fa ammā annash‘ura bi annā qad sha‘arnā bi dhawātinā), it is from an act of the intellect.

Self-awareness is actual for the soul (al-shu‘ūru bi al-dhāti yakūnu li al-nafsibi al-fi‘l), so that it is constantly aware of itself (fa innahā takūna dā’imata

23 Reading qat.t.u, instead of Badawī’s faqat., in accordance with the forthcoming critical edition bySeyyed Hossein Mousavian, who has kindly bestowed upon me a manuscript version. Both readingsare supported by manuscript evidence.

24 Here and later in the sentence I read in passive (lā yush‘ar) with Mousavian instead of Badawī’s first-person plural (lā nush‘ar), also supported by manuscript evidence. Strictly speaking, it would ofcourse be problematic to ascribe a plural subject (‘we’) for awareness of an individual self. However,such a loose manner of speaking is understandable, and in fact Avicenna does resort to it later on inthe text in places where the manuscripts are unanimous.

25 Reading shu‘ūruhā bihā with Mousavian, instead of Badawī’s shu‘ūrunā bihā, also supported bymanuscript evidence.

26 Cf. Ta‘līqāt 79, which repeats this point almost verbatim.27 Reading the feminine suffix withMousavian, instead of Badawī’s ‘arafahu, which is also supported by

manuscript evidence. The difference between the readings is inconsequential.28 Reading fa ammā with Mousavian instead of Badawī’s fa in mā.


al-shu‘ūri bi dhātihā). As regards awareness of the awareness (al-shu‘ūru bi al-shu‘ūr), it is potential; if awareness of the awareness were actual, it would beconstant and no consideration of the intellect (i‘tibāri al-‘aql) would beneeded for it.[c] My apprehension of myself (idrākī li dhātī) is something constitutive

to me,29 not occurring from a consideration of another thing. Thus, when Isay ‘I did such and such’, I consider my apprehension of myself (idrākī lidhātī) even if I neglect my awareness of [myself] (wa in kuntu fī ghaflatin ‘anshu‘ūrī bihā); how else would I know that I did such and such, if I didn’t firstconsider myself (law lā annī i‘tabartu awwalan dhātī)? Thus, I have firstconsidered my self, not its act (qad i‘tabartu awwalan dhātī, lam fi‘lahā), andI do not consider anything by means of which I would apprehend myself(lam a‘tabar shay’an adraktu bihi dhātī).Our awareness of our self is our very existence (shu‘ūrunā bi dhātinā huwa

nafsu wujūdinā).When we know something, in our knowledge of our apprehension of it

there is awareness of our self (fī ‘ilminā bi idrākinā lahu shu‘ūrun bi dhātinā);for we know30 that our self has apprehended it (na‘alima anna dhātanāadrakathu), and so we have first been aware of our self (fa sha‘arnā awwalanbi dhātinā). How else would we know that we have apprehended it, if weweren’t first aware of our self (sha‘arnā awwalan bi dhātinā)?What is like thatis a reminder,31 not a demonstration, that the soul is aware of itself (annaal-nafsa shā‘iratun bi dhātihā).[d] Those which are primary are not actual for us,32 for otherwise there

would be no need for a consideration about them.[e] Self-awareness is innate to the self, it is [the self’s] very existence

(al-shu‘ūru bi al-dhāti huwa gharīzīyun li al-dhāti, wa huwa nafsuwujūdihā); so nothing external is needed by means of which to apprehendthe self – rather, the self is that which apprehends itself (fa lā yuh. tāja ilā

29 Reading amrun muqawwimun lī with Mousavian, instead of Badawī’s amrun yaqūmu lī, ‘somethingthat arises (or emerges) for me’, which incoherently supposes me to be something to which self-awareness can ‘arise’ and which therefore is separable from it. Avicenna argues for a much strongerconnection between me and my self-awareness, indeed one in terms of constitution.

30 Reading li annā na‘alimawithMousavian instead of Badawī’s lā na‘alim, which does have manuscriptsupport. Mousavian’s reading is coherent with Avicenna’s statement elsewhere in the Ta‘līqāt (79)that, when we know something with certainty, we know that we know.While Badawī’s reading couldperhaps be reconciled with that passage, more interpretative work is required, and thereforeMousavian’s reading appears more probable. For the connection between epistemic certainty andself-awareness in Avicenna, see Black 2008, 76–81.

31 Reading tanbīhwithMousavian instead of Badawī’s bayyanatun, which does have manuscript supportand makes sense. However, tanbīh in the technical sense mentioned above is arguably the lectiodifficilior and provides a clear reference to arguments like the flying man.

32 Badawī omits the lanā retained by Mousavian. The point clearly concerns the explicit presence of thefirst principles of knowledge as objects of our consideration; cf. our discussion on the similaritybetween self-awareness and such first principles, and the corresponding similarities in arguments forthem, in Chapter 2.1, 33–34. The point is made in more explicit terms in Ta‘līqāt, 79–80.


shay’in min khārijin tudraku bihi al-dhātu, bal al-dhātu hiya al-latī tudrikudhātahā).33 Thus, it is not sound for it to exist without there being awarenessof it (takūna mawjūdatan ghayra mash‘ūrin bihā) given that what is aware of itis its very self (yakūna al-shā‘iru bihā huwa nafsu dhātihā), not any otherthing. This is not exclusive (khās.s.an) to human beings, rather all animals areaware of themselves (tash‘iru bi dhawātihā) in this respect.34

For the present purposes, let us reiterate the points on which the allegedsolution to the problem of individuation hinges. Avicenna begins by statingin section [a] that our self-awareness is not acquired but innate. As aconsequence, it is also immediate, and involves no use of cognitive instru-ments of any kind. This could have been inferred from the flying man aswell, based as it was on bracketing all acquired knowledge and shutting outall cognitive faculties by rendering their respective instruments inoperative.In section [b], Avicenna continues by claiming that self-awareness is essen-tial and hence necessary or concomitant to human beings, something thatwill always prevail given the existence of an individual human and thattherefore does not permit intermission. He rephrases the point in section[c], stating that self-awareness is constitutive to all actual human beings andas such never figures as a potency the actualization of which would dependon accidental circumstantial conditions. These considerations are furtherdeveloped by means of an argument against models of self-awareness basedon reflection,35 which is followed by the statement, presented as if inconclusion of what has been said so far, that ‘our awareness of our self isour very existence’. This is repeated in section [e] in the form of a summary:‘awareness of the self . . . is its very existence’. The obvious outcome is that,instead of a human property, self-awareness constitutes human existence. It isthe mode in which individual immaterial human substances exist just asmateriality is the mode in which individual human bodies exist.

If we read this as an attempted solution to the incoherence in the Shifā’,the point is that just as the spatiotemporal co-ordinates make materialexistence the existence of individual things, self-awareness renders immate-rial existence individual. As we recall, Shifā’: Fī al-nafs V.3 listed self-awareness as merely one of the five types of attributes of the immaterialhuman substance, a constituent among others of a bundle of propertieswhich as a whole were argued to individuate the immaterial human

33 I read in passive (lā yuh. tāja, tudraku, tudriku) with Mousavian instead of Badawī’s first-person plural(lā nuh. tāja, nudriku, nudriku), which also has manuscript support and would make sense as a loosemanner of expression.

34 Ta‘līqāt, 160–161.35 The argument resurfaces in the Ishārāt and will be discussed in detail in Chapter 4.1, 72–75.


substance, but which on closer analysis, and in light of Shifā’: al-MadkhalI.12, seemed unfit for the task. In the present passage, self-awareness ispromoted to a unique constitutive status: as the existence of an individualhuman being it is the determining factor in that human being’s individu-ation. To put it another way, instead of one of the many attributes of thehuman substance, Avicenna now presents it as the substrate of all subse-quent attributes. This is entailed by the argument against reflection-basedmodels of self-awareness in section [c].Let us approach this idea by way of an example. Suppose that a colleague

enters my office with a lunch proposal. Caught in the act of contemplating,I decline the offer, uttering ‘I am thinking’ as an excuse. Now, the emphasishere is obviously on ‘thinking’, which expresses my reason for declining,and not on ‘I’, which specifies myself as the agent entertaining a line ofthought. Yet in order for the thinking to be given in the relevant sense in thefirst place, that is as an act I can attribute to myself, I must first be aware ofmyself as something concrete to which I can attribute the act. This priorself-awareness is completely independent of whether or not I ever payspecific attention to it. As a primitive constituent of any individualhuman being, self-awareness is the basis which enables us to attributeanything to ourselves as exclusive to a certain individual. This act ofthinking is mine, even if another individual were thinking an exactly similarthought elsewhere, because it is an attribute ofmy self-awareness. Thus, themetaphysical role of self-awareness is similar to that of the spatiotemporalco-ordinates of the volume of matter which constitutes a material individ-ual. In other words, self-awareness is the existence of an individual soul inprecisely the same sense as corporeality, with the entailed spatiotemporal co-ordinates, is the existence of an individual body.Avicenna does not spell out why he believes that self-awareness can be

cast in this important explanatory role. However, it is perhaps plausible toassume that this is due to its unique phenomenological features. Eachhuman being’s awareness of herself is exclusive to her and cannot be sharedby anyone or anything else; it is something to which only that individual hasaccess.36 But quite apart from whether the reconstructed solution is

36 If this is the basis of the reconstructed solution, then its plausibility hinges on accepting Avicenna’sdescription of self-awareness. Minimally, this includes the claim that self-awareness is primitive andirreducible to more basic psychological or epistemic constituents. Self-awareness does not arise fromthe cognition of a special object, for as Avicenna states in the argument against reflection-basedmodels of self-awareness, recognition of oneself in an object requires prior familiarity with oneself.However, while citing this passage, Black 2012 does reconstruct Avicenna’s concept of self-awarenessas an awareness of the individual intention of Shifā’: al-Madkhal I.12. I will revisit this suggestion inChapter 4.3, 91–94.


plausible in systematic terms, it remains a fact that neither Avicenna nor anyof his immediate interlocutors ever explicitly recognize the described inco-herence, nor do they present self-awareness as its solution. However, thisdoes not necessarily mean that the reconstruction is anachronistic, for asimilar appeal to self-awarenesss as a solution to the problem of individu-ation is recognized as Avicennian a century and a half later by the criticalcommentator Fakhr al-Dīn al-Rāzī.37 Moreover, the connection betweenindividuation and self-awareness is retained in Mullā S· adrā’s discussion ofmental existence.38 In the end, however, it must be admitted that we haveno conclusive evidence for treating the question as Avicenna’s own, let alonereading the passage in the Ta‘līqāt as expressly designed for its solution.Moreover, regardless of the plausibility of this reading, it is problematizedby the context in which the passage is embedded, a complex patchwork ofalternating sections on theology and psychology or epistemology. Ourpresent knowledge of the textual history and the principles of organizationof the Ta‘līqāt allows us no definite assessment of the text, but thisuncertainty notwithstanding, the theological relevance of the discourse onself-awareness is relatively straightforward.39

Avicenna’s description of the characteristics of human self-awarenesstakes place in the wake of an account of God’s knowledge of Himself andHis creation. In a manner familiar from the Shifā’, he first claims that God’sknowledge entails knowledge of the created world because He knowsHimself as its maker. In fact, creation primarily exists in being understoodby God. In a reformulation of the Neoplatonic framework of emanationand return, Avicenna then says that God is both the origin and the end of allthrough His understanding of Himself as the origin and as absolute good,worthy of the pursuit of all things. The usual denial of any kind of passivity,receptivity and impression in God’s knowledge are added in order todistinguish His knowledge from the sort of cognition His creatures maybe capable of. Avicenna thereby asserts that, in the case of God, the subjectand object of knowledge are one, something he has vehemently deniedelsewhere of all other cognitive subjects.40 This is based on the theologicaltenet that all of God’s attributes, including His knowing Himself as theorigin and the good as well as the entailed knowledge of the created world,are identical to His essence or His self (dhāt).41

37 Rāzī, Mabāh. ith II.2.2.5.5, II.402–403. See Chapter 7.2, 185–187.38 See Chapter 7.2, 187–188.

39 Moreover, another similar foray into self-awareness in the Ta‘līqāt (78–80) figures in a similarcontext.

40 Cf. Shifā’: Fī al-nafs V.6, 239–240 Rahman. 41 Ta‘līqāt, 158–160.


This section of the Ta‘līqāt is very close to the account of God’s knowl-edge in Shifā’: al-Ilāhīyāt VIII.6–7, but immediately thereafter Avicennasuddenly moves to consider human cognition insofar as it is similar to God’sknowledge of Himself. If this order of procedure is originally Avicennianand not a result of a later recension, it can be understood as an attempt toelucidate the identity of the different attributes of God with His self on theone hand, and the identity of intellection, its subject and its object on theother. For a devout sceptic could ask how we can even begin to understandGod’s knowledge if it, due to its absolute unity, is fundamentally differentfrom our own, the only phenomenon we can use as a starting point in anydiscussion of knowledge. Thus, it would be in order to make his realistdescription of God’s knowledge more palatable to the sceptical reader thatAvicenna suggests that, notwithstanding the multiplicity inherent inhuman intellection, there is a certain qualified sense in which we canspeak of the unity or identity of what knows and what is known in it aswell. This is stated right before section [a] in a paraphrase of the Aristotelianclaim that they share one and the same actuality. My being an intellect in act(my actual ‘aqlīya) amounts to the actual existence of an intellectual formfor me, and this existence is nothing other than the form’s being actuallyunderstood (its actual ma‘qūlīya) by me.42 Although the intelligible formremains distinct from the intellectual subject that understands it because oftheir metaphysical classification under opposing categories – the form is anaccident which inheres in the intellectual substance – they nevertheless sharein the actuality of intellection.Having thus qualified his denial of the unity of subject and object in

human cognition, Avicenna moves on in our passage to consider an aspectof the human intellect that he seems to think shows a much greatersimilarity to God’s self-intellection. More than anything else we haveimmediate access to, human self-awareness approaches the absolute identitybetween the subject and the object of cognition that is characteristic ofGod’s knowledge. From this perspective, it would be in order to make senseof this attribute of God that our passage contrasts self-awareness so dili-gently with other types of cognition, which take place either by means of agiven intellectual object or, in the case of perception and imagination, byway of corporeal instruments. Similarly, if the context is genuinelyAvicenna’s, it seems natural to assume that Avicenna dwells at such lengthon the identity of our self-awareness and our existence precisely in order to

42 Ta‘līqāt, 160.


elucidate the parallel identity between God’s knowledge and His self.43

Avicenna’s intention would not be unlike that of Augustine in De trinitate,that is, ‘to find in the soul of man [the] image of the Creator’,44 or to cometo understand something that is infinitely above us by means of thesimilarities to it that we can find in ourselves.

In the end, we have to admit that neither of the two discussions to whichwe have embedded Avicenna’s excursion on self-awareness is unequivocallythe correct context for it. Reading the passage in light of the psychologicaldiscussion concerning the individuality of the human substance is madeproblematic by Avicenna’s complete silence about the connection, whereasthe haphazard organization of the Ta‘līqāt throws doubt on the putativeconnection to the theological discussion of God’s knowledge. However, ifthe passage and its blunt identification between human self-awareness andhuman existence belongs to either of the two contexts, this will be due toone and the same idea, that is, that self-awareness is the sort of existenceproper to an immaterial, hence intellectual, substance. This in turn suggeststhat despite the novelty of Avicenna’s concept of self-awareness, his startingpoint is the result of three interconnected ideas, all of which are perfectlytraditional as such.

The first of these ideas is that immateriality is equivalent to intellectual-ity. In other words, if something is an immaterial entity, it is by the sametoken an intellect, and vice versa. The second idea verges on the tautolog-ical: if something is an intellect, its actual existence amounts to intellection.Finally, the third idea expresses a particular concept of intellection: intel-lection always amounts to self-intellection because it consists of an identityor unity of some sort between the subject and the object of intellection.Avicenna subscribes to this set of ideas explicitly, if not without certainqualifications, both in the Shifā’ and in the Ishārāt. For a lucid example,consider the following description of God as the paradigmatic intellect:

The First is understood by [Him]self and subsists [by Himself] (al-awwaluma‘qūlu al-dhāti qā’imuhā).45 Thus, He subsists free from attachments,obligations, matters and other such [things] which would bring the self to

43 In this regard, theTa‘līqāt is not entirely unique in Avicenna’s oeuvre; cf. the discussion of the relationbetween the theological and psychological parts of the Shifā’ and the Ishārāt in Adamson 2011a.Interestingly, Suhrawardī resorts to a similar strategy in the Talwīh. āt and al-Mashāri‘ waal-mut.ārah. āt; see Chapter 6.1.

44 Aug. De trin. XIV.4.6.45 This is a difficult sentence. I read it as consisting of two parallel id

˙āfas, that is, ma‘qūlu al-dhāti (wa)

qā’imu al-dhāt.


be by means of an additional state. And you know that what is judged suchunderstands itself and is understood by itself.46

God’s intelligibility is equivalent with His immateriality, His freedom fromattachments and deficiencies that are due to matter, and as such He isknown by definition to understand Himself and be understood by Himself.In other words, God’s intellectuality amounts to a constant act of self-intellection.Avicenna’s introduction of the discussion of human self-awareness in a

context dealing with God’s intellectuality strongly suggests that heapproaches the phenomenon from the point of view of the traditionaltheory of the intellect. Because the entity that functions as a soul in relationto the human body is an immaterial substance independent of its relation tothe body, its existence in itself must be intellectual. If it is intellectual, itmust be intellectually aware of itself. In this regard, the human substance islike any other intellect.47 However, there is an important differencebetween human and divine intellection which may have motivatedAvicenna to explain the unity of God’s intellection at such great length.God’s knowledge of Himself merges with His knowledge of His creation,whereas we are always emphatically aware of ourselves in the face of another, and this introduces a persistent structural duality to our cognition. InShifā’: Fī al-nafs V.6, Avicenna vehemently denies the theory of cognitiveunity in the case of human intellection, attributing it to Porphyry who ‘wasbent to speak in imaginative, poetic and mystical (s.ūfīya) expressions whichrestricted both himself and others to imagination’. The argument annexedto the denial is admirably lucid. Suppose that I understand catness now attime t1 but horseness a moment later at time t2. According to the theory ofcognitive unity, at t1 I am identical to catness but at t2 identical to horseness,and so, given the obvious difference between catness and horseness, wecannot speak of any me that endures in knowing both at the successivemoments. This, however, violates our intuition that we can, in some

46 Ishārāt IV, 146. Cf. Shifā’: al-Ilāhīyāt VIII.66–8, 284–285. For discussion, see Sebti 2000, 100–103;Adamson 2011a; and cf. Pines 1954, 36–39.

47 The fact that Avicenna occasionally claims that all animals are aware of themselves in the samemanner he has just described human beings to be is of course problematic. I have argued elsewhere(Kaukua and Kukkonen 2007; cf. López-Farjeat 2012) that, on systematic grounds at least, Avicenna’stheory of animal self-awareness (which he is hesitant about; for passages conflicting with the presentone, see Mubāh. athāt 305, 184 Badawī [VI.656–657, 221 Bīdārfar]; 358, 199 Badawī [VI.504–505, 176Bīdārfar]; and most importantly Ta‘līqāt, 79) should be different from his theory of human self-awareness. In the end, I believe that Avicenna failed to develop a fully coherent theory of animal self-awareness, or one capable of reconciling his dualism with intuitions concerning animal psychologythat he seems to have shared with his contemporaries.


sufficiently robust sense, intellectually apprehend different things at differ-ent times, that I who am thinking about horseness now am the very personwho was thinking about catness a moment ago.48 To save this intuition, wehave to postulate our selves to endure unchanged from t1 to t2, and thereforereject the theory of cognitive unity. Instead, we should opt for a theory ofintellection in which the intelligible forms actually understood are con-ceived as accidents inhering in the human intellect, separable from it notmerely in analysis but also in reality.49While this may be compatible with amodest version of the theory of cognitive unity, according to which theintellect and its object share in a single act of intellection, such a unity willbe considerably weaker than that in the divine intellect. As a result, also ourmeans of describing God’s single act of intellection of Himself and Hiscreation are severely inadequate to render the unity as anything other than aconjunction of two analytically distinct constituents, that is, God’s knowl-edge of Himself on the one hand, and His knowledge of His creation on theother. But even though our capacity of intellection will not take us very farin understanding God’s synonymous act, there is one feature in our knowl-edge that is capable of giving us a more concrete idea of the unity in God’sknowledge. It is for this reason that Avicenna emphasizes the immediacyand the primitive nature of our self-awareness, the sole cognitive phenom-enon in us that is capable of resisting the necessity of a real distinctionbetween subject and object.

But even if Avicenna took his cue from the tradition in applying humanself-awareness to account for our individual existence and to make sense ofGod’s knowledge of Himself, his emphasis on the immediacy of self-awareness forced him to depart from the traditional ground. This is becausenone of the available modes of cognition can be used to explain such aprimitive type of awareness. If nothing is allowed to mediate between thehuman subject and her awareness of herself, self-awareness has to beconceived of in a manner different from both perception and intellection.Both types of cognition take place when what is known is impressed orinheres in the knowing substance or its cognitive organs. Depending onthe type of cognition, the known object can be of three different types. Itcan be a proper or a common sensible form, such as colour, sound, motion,shape and so forth, in which case it will be impressed in the corresponding

48 The argument thus implicitly relies on a feature of our self-awareness, which is explicitly broughtforth in an argument from personal identity inMubāh. athāt VI.403, 147 Bīdārfar; cf. 453, 226 Badawī;see Chapter 4.1, 75–79.



sense organ, and by means of that in the faculty of common sense, as amaterial property that corresponds to an extramental object. It can be whatAvicenna calls meaning (ma‘nā), such as the enmity a sheep apprehends in awolf. This is an object proper to the faculty of estimation (wahm), andalthough not sensible as such, it is conveyed in what is sensed. Thus, even ifit no longer has the sensible properties that are immediately due to matter, itdoes reside in what is sensed, and since it also inheres in the material organproper to estimation, it remains material.50 Finally, the object can beintellectual, in which case it is received by the soul from the active intellectwithout any corporeal faculty, although the soul must have been preparedfor this reception by a process of abstraction which it performs by means ofcorporeal faculties.51 But as we have just seen, although the intellectual formis immaterial, it remains distinct from the intellect in which it inheres as anaccidental attribute really separate from it. Since this distinction betweensubject and attribute is thus common to all these types of cognition, theform known is in each case apprehended as an object, which I conceive to bedifferent from myself and which therefore cannot function as the mediumof self-awareness.Thus, self-awareness must be a very peculiar type of cognition. To my

knowledge, Avicenna never explicitly attributes a special cognitive categoryto it, although something along such lines is hinted at in a hesitant remarkhe makes in the Mubāh. athāt: ‘It may be that “intellection” [in the sense ofthat] which grasps intelligibles is not applicable to the purity of completeself-awareness but comes after that. That is worth thinking about.’52 In anycase, although Avicenna’s thesis of self-awareness as the mode of existenceproper to the immaterial human substance is based on the traditionalequivalence between immateriality and intellection on the one hand, andthe equally traditional concept of intellection as self-intellection on theother, he is landed with the task of conceiving human self-awareness in amanner distinct from commonplace intellection. Avicenna never sets out tofulfil this task in a systematic fashion; there are no chapters ‘On self-awareness’ in any of his psychological summae. However, I claim that ifwe gather together the frequent, albeit somewhat scattered, remarks hemakes on the phenomenon, we can reconstruct a coherent concept ofself-awareness that is able to figure in all contexts of its application.

50 For a fuller classification of the cognitive faculties and objects of the sensitive soul, see aboveChapter 1.2, 25–27.

51 For the process of abstraction, see above Chapter 1.2, 27–29.52 Mubāh. athāt 373, 209 Badawī (cf. V.288, 119 Bīdārfar, and VI.886, 316 Bīdārfar); see also 371, 208

Badawī (V.278–281, 117–118 Bīdārfar).


chapter 4

In the first person: Avicenna’s conceptof self-awareness reconstructed

The passages that we’ve considered so far set two rather stringent conditionsfor Avicenna’s concept of self-awareness. The first of these is due to theblunt claim of identity between self-awareness and our existence. It seemsintuitively plausible that our existence is continuous in time, not a series ofdiscrete episodes punctuated by periods of non-existence. If Avicenna wantsto save this intuition, the phenomenon his concept of self-awareness picksout must also be continuous, that is, he must speak of self-awareness in asense that allows no lapses into a state lacking it. One could of courseentertain the possibility that he sets out to deny and argue against theintuition of the continuity of our existence, showing that as a matter of factwe are intermittent entities and our intuition amounts to little more thanprejudice. Such a denial, however, is nowhere to be found, and this for goodreason too, since it is manifestly clear that Avicenna opts for the firststrategy, fully aware of the condition. The above passage from the Ta‘līqātexplicitly states that the self ‘is constantly aware, not from time to time’, andthe reminder that initiates the discussion of the human soul in the Ishārātaddresses the issue with more concrete examples. In connection with theflying man, Avicenna there brings up the cases of a sleeping and anintoxicated person whom he clearly considers to embody states of whichwe would find it natural to deny any kind of awareness, a fortiori explicitself-awareness. The examples are introduced only to argue that self-awareness in the sense he speaks about it is present in them as well.1 Thisprovides explicit support for the view that Avicenna was fully aware of thearguably problematic consequences of the equivocation between humanself-awareness and the existence of the immaterial human substance.2

1 Ishārāt, namat. 3, 119.2 A related passage in the Mubāh. athāt (III.66–72, 61–62 Bīdārfar; 380–381, 210 Badawī) suggests thatthese counterexamples may have been introduced by Abū al-Qāsim al-Kirmānī in a critical series of

62

The second condition is due to the flying man. In the above I suggestedthat the argument hinges on something anyone performing the thoughtexperiment should be able to find in her own experience after bracketing allother constituents of the experience in the required manner. I argued atlength that the remaining constituent is self-awareness, in a sense thatremained to be determined. However, it was sufficiently clear at that stagethat, in order for his argument to work, Avicenna must deal with a perfectlycommonplace type of self-awareness, nothing extraordinary or exclusive tospecial states of mind, although it may well be something we do not oftenpay attention to and so must be aroused to notice by means of a remindersuch as the flying man.When these two conditions are brought together, it becomes clear that

Avicenna’s concept of self-awareness should refer to a phenomenon that is(1) constant and (2) experientially given. This is obviously a rather strongrequirement. In any case it immediately rules out any kind of reflective self-awareness such as your explicit consideration of the fact that you are nowreading this sentence. Although reflective self-awareness is experientiallygiven in the full sense of the word and thereby fulfils condition (2), it is anintermittent and occasional state and therefore fails to fulfil condition (1).Part of the experience of reflection is precisely coming to reflect upononeself, which entails that the reflection was preceded by a state in whichone did not pay reflective attention to oneself. Moreover, reflection requiresthe introspective use of one’s cognitive faculties, which in turn involves anobject (the prior state) that is acted upon, and is often prompted bypeculiarities of the situation one finds oneself in. More complex self-relations, such as those that occur when one reflects on one’s personalidentity and its constituents (am I primarily a present father or an aspiringacademic?), are ruled out a fortiori, since they are evidently founded uponthe simpler type of reflection.Once again, Avicenna is fully aware of the stringency of his situation.

This is evidenced by his distinction of self-awareness from reflective aware-ness that he calls awareness of awareness (al-shu‘ūru bi al-shu‘ūr), addingthat the latter is not constant but instead requires ‘consideration of theintellect’ which is by nature an event with a rather narrow temporal scope.Moreover, in the argument against models of self-awareness based onreflection, which we will consider in greater detail in a moment, Avicenna

questions directed at the flying man. However, the provenance of the examples depends on the datingof the Ishārāt, which is a matter of controversy (see Gutas 1988, 140–141; Michot 1997, 153–163; andReisman 2002, 215–219, 222–224).

In the first person 63

states that reflective awareness of oneself depends for its possibility on themore fundamental type of self-awareness.

How likely is it then that Avicenna’s concept of self-awareness is acoherent one in the first place? Before delving into hasty assessments ofthis question, let us attempt to reconstruct a more fully fledged version ofthe concept by means of three discussions in which the underlying phe-nomenon figures in a prominent role. Despite the occasional use of differentvocabulary, these discussions are explicitly related to the texts on self-awareness we have already considered and can therefore be legitimatelyused to make sense of Avicenna’s concept of self-awareness.3 Once we havefleshed out a reconstruction of the concept, we can assess it against the twoconditions.

4.1 Three Avicennian arguments from first-personality

Argument from the unity of experience

In a pioneering study from 1974, Ben Lazare Mijuskovic charted the out-lines of the modern history of a particular argument which, following Kant,he called the ‘Achilles’ of rationalist psychology (referring both to theinvincibility of and an inherent weakness, the notorious heel, in the argu-ment) and which proceeds from the evident unity of our experience to thesimplicity of the soul that is its subject.4 Very briefly, the argumentgenerally sets off from two premises: (1) that our experience is in fact aunified whole of multiple constituents (such as data due to the five externalsenses, various desires, and acts of intellection, in a perfectly commonplacecase) and (2) that such a unified whole must be due to a single subject that isone in itself, neither divided nor divisible into distinct parts. From these it isconcluded (3) that the human soul responsible for the experience is one.5

As is well known, Avicenna subscribes to the argument and in fact uses abroad version of it to argue onwards, by means of the further premise(4) that all corporeal things are divisible into distinct parts, to the conclusion(5) that the soul is incorporeal.6 This version of the argument can be found

3 Shifā’: Fī al-nafs V.7, the very chapter from which we draw the discussion of the argument from unity,presents a version of the flying man. A brief version of the argument against reflection-based models ofself-awareness is featured in the passage from theTa‘līqāt discussed above. It is also brought to bear in adefence of the flying man in Mubāh. athāt III.64, 60 Bīdārfar; 370, 207 Badawī.

4 See Mijuskovic 1974. For a recent attempt at filling some of the gaps in Mijuskovic’s story, as well as ahistory of the argument prior to the Cambridge Platonists, see Lennon and Stainton 2008a.

5 For a concise exposition of various forms of the argument, see Lennon and Stainton 2008b.6 For a brief discussion of the argument as presented inNajāt II.7, 228–230, see Lagerlund 2008, 78–80.

64 In the first person

in its most extended form in Shifā’: Fī al-nafs V.7, to which we will nowturn. Our primary interest, however, is not in the argument itself, but in thepeculiar role self-awareness plays in Avicenna’s rendering of it.The unity of our experience is a particularly pressing concern for

Avicenna, whose psychology is founded on the idea of distinct faculties,capacities or powers. As he states his method in Shifā’: Fī al-nafs I.5, heproceeds by postulating a corresponding faculty whenever he can determinea discrete psychic act.7 The faculties, their respective acts, and thecorresponding objects are firmly separate from each other, so that no twofaculties can share an object. Thus, the act and object of desire orthe appetitive faculty are different from the act and object of anger or theirascible faculty, and by the same token both are to be separated from therespective acts and objects of the external and internal senses. Avicennainsists on this methodological basis in the very context in which he presentsthe argument from unity.8

Yet the stringent criteria of distinction inherent in Avicenna’s facultypsychology seem to be at odds with common experience. First of all, asAvicenna points out, the faculties that were supposed to be distinct fromeach other are capable of hindering each other in their respective acts. Forinstance, my attempt to read a difficult text can fail because I am distractedby the smell of and desire for freshly brewed coffee. If such a thing ispossible, there must be some connection between the distinct faculties.Furthermore, the act of one faculty often seems to cause an act of another,such as when I smell the freshly brewed coffee and then begin to desire acup. Yet according to Avicenna’s criteria, sense perception and desire are noteven directed to the same thing.What is smelled and what is desired are twoentirely different objects, and even if both are caused by a single externalthing, the two causal chains leading to the two objects are distinct from eachother. Thus, there has to be another connection due to which the twoobjects collapse in the soul and which in this sense is above them.9

Now, the most natural answer is of course that the soul, to which all thedistinct faculties belong and which is the single agent behind them all, iswhat provides the connection. Indeed, this is exactly how the argumentruns in the Najāt: ‘Thus, it remains that that which combines is a soul byitself (nafsan bi dhātihā), or a body in the respect that it really has a soul, andso that which combines is the soul. That soul is the origin of all these

7 Shifā’: Fī al-nafs I.5, 39–51 Rahman.8 Shifā’: Fī al-nafs V.7, 252–253 Rahman. For Avicenna’s faculty psychology, see Chapter 1.2, 25–27.9 Shifā’: Fī al-nafs V.7, 253 Rahman. Cf. Najāt II.7, 228–229.

Three Avicennian arguments from first-personality 65

faculties.’10 This is a paradigmatic case of the argument from unity, basedlargely on received or demonstrated views concerning the soul, and with noreliance at all on any experiential data related to self-awareness. The ambi-guity of the term nafs, owing to which the word can sometimes be read tomean ‘self’, is not relevant here, for the argument can be exhaustivelyexplained by means of the distinction between soul and body.

However, the same argument is dwelt upon at greater length in the Shifā’,where much of the expansion consists of experiential data designed torender the soul’s role as the unifier of the different constituents of experi-ence intuitively more plausible. To begin with, prior to a series of argumentsagainst the claim that the body is the unifying principle, Avicenna intro-duces the soul’s role as follows:

Now this single thing in which these faculties are conjoined is the thing thateach of us sees as himself (al-shay’u al-ladhī yarāhu kullun minnā dhātahu) sothat it is sound that he says ‘since we perceived, we desired’.11

Notice that Avicenna does not straightforwardly say, as he does in theNajāt,that the unifying agent is the soul, that is, the theoretical entity postulatedand studied in psychology. Instead, he starts from the commonplaceexperience of motivated action, which is not obscured by the awkwardformulation in plural; we frequently explain our actions by pointing out aconnection between a perception and a resulting emotional motivation, aconnection that is here expressed by means of the causal ‘since’ (lammā).Suppose, for instance, that I see a car speeding through a crossing that achild is trying to pass. I instantly feel an upsurge of indignation, whichcauses my hand to make a universally understandable gesture and mymouth to emit a related verbal expression. If someone then asks me forthe reason behind the outburst, it is natural for me to answer with a briefdescription of what I just saw, thereby expecting my anger and its expressionto be readily understandable. Avicenna all but agrees: such an explanation isperfectly sound (yas.duqu).

But to what is the connection between the perception and the desireultimately due? Let us consider the constituents of the expression ‘since weperceived, we desired’. The connection expressed by the logical connective‘since’ cannot be the basis here, since that would amount to a circle inexplanation. Moreover, the connection can take many forms, even thecontrary ‘we perceived but did not desire’, which clearly shows that theconnection itself requires a causally determining factor. The words

10 Najāt II.7, 229. 11 Shifā’: Fī al-nafs V.7, 253 Rahman; emphasis added.


‘perceived’ and ‘desired’, in turn, refer to the contents of experience at issue.Now, the contents can be said to play a role in attempts to justify theconnection, insofar as they determine the success of our explanation inpersuading our peers about our reasonability. I make my action under-standable to others and subject it to their approval precisely by referring tosuch constituents of my mental states that I can expect to stand in areasonable connection to each other because of their inherent features.However, no justification of a connection between discrete psychic actscan be a psychological explanation, for clearly there are also connectionsbetween perceptions and desires that are not justifiable in this sense; ourvarying idiosyncratic preferences and tendencies are cases in point. Thepsychological explanation, on the contrary, has to be so general that it canaccount for these as well.This leaves us with but one element in the sentence, the repeated

pronoun ‘we’. The repetition is important because it makes explicit thefact that it is we in one and the same sense who perceive and desire, whichmeans that the subject of perception is in a strict sense identical with thesubject of desire. Thus, what Avicenna points at here by means of common-place experience is an implicit connection, all too obvious for me to notice,let alone pay attention to: it is me that sees, becomes angry, and acts in acorresponding manner. The connection between the distinct constituentsof experience cannot be based on anything else. Without an identicalsubject behind the two acts, perceiving and desiring, there would be noway of bringing forth the logical connective. This single subject is ‘the thingthat each of us sees as himself’.Later on, in his third argument against the claim that what unites the

distinct constituents of experience into a single whole is the body, Avicennaagain resorts to our common experience of ourselves:

[T]his body is either12 the whole body, so that if a part of it was lost, what weare aware of being us (mā nash‘uru annā nah. nu) would not exist. But that isnot the case, for I would be me (akūna anā) even if I did not know that I havea hand, a leg or some other of these organs, as has been told earlier in otherplaces.13 I suppose instead that they are my appendages, and I believe thatthey are instruments for me that I use in needs, and were it not for thoseneeds, I would have no necessity for them. I would be me even if they werenot there (akūnu ayd

˙an anā anā wa laysat hiya). Let us consider what we said

12 The other alternative comes only much later in the text and is the suggestion that the body connectingthe constituents of experience is a specific organ instead of the body as a whole. See the next quotebelow.

13 This is most likely a reference to the flying man in Shifā’: Fī al-nafs I.1, 15–16 Rahman.


earlier and say: if a human being were created in a single instant, and createdwith his limbs separate from each other, without seeing his limbs, and ithappened that he would not feel them, they would not be in contact and hewould not hear a sound, [then] he would be ignorant of the existence of allhis organs yet know the existence of his thatness (annīya) as one thing whilebeing ignorant of all that. What is unknown is not the same as what isknown. These our organs are in reality only like clothes which, because of theduration of their adherence to us, we have come to regard as parts of us.When we imagine ourselves (anfusanā), we do not imagine [ourselves]naked, but rather we imagine [ourselves] to have enveloping bodies(dhawāta ajsāmin kāsiyatin). The reason for this is the duration of theadherence. It is only because we are prepared to strip off and discard clothesin a way we are not prepared to do with the organs that our belief that theorgans are parts of us is firmer than our belief that clothes are parts of us.14

It is important to notice that the present argument hinges on the sameintuition, that is, that our selves are the nexus between the diverse constit-uents of our experience and action. The body cannot be the node because nobody can be a self. This is for two reasons. First, Avicenna argues that what Iperceive to be me will remain unchanged nomatter how drastic any changesthat occur to the body. Quite apart from whether we agree, this is some-thing he considers evident. Secondly, as the brief version of the flying manshows, one can be aware of oneself without being aware of one’s body. Onthis basis Avicenna concludes that since what one is aware of cannot be thesame as what one is not aware of, the body cannot be the self; and that sincethe self has earlier been claimed to connect the diverse constituents ofexperience, no such task is left for the body to perform. Now, as notedabove in our discussion of the flying man, this argument is far fromunproblematic.15 The fallacy of straightforwardly inferring a metaphysicaldistinction from a phenomenological distinction is particularly prominentin this case, because here the flying man hovers in a nodal point in the largerrefutation of the body’s capacity to unite the distinct constituents ofexperience. However, judging the argument to be invalid does not forceus to deem it irrelevant or uninteresting for our present concern, that is, thereconstruction of Avicenna’s concept of self-awareness. On the contrary,the fact that the whole argument, and the flying man as its part, builds onthe intuition expressed in the first passage (namely that the unifying subjectin my experience is ‘the thing that each of us sees as himself’ or ‘what we areaware of being us’) shows that Avicenna perceived a crucial connectionbetween the flying man and our intuitive familiarity with ourselves. In light

14 Shifā’: Fī al-nafs V.7, 255 Rahman. 15 See Chapter 2.2, 37–38.


of this context, it therefore seems natural to assume that here too self-awareness, or the experience of being a self, amounts to the diverseconstituents of my experience all being mine in one and the same sense.The same point is made even more apparent in the immediately follow-

ing refutation of the claim that some specific part of the body is responsiblefor connecting the disparate constituents of experience into a single whole.

If it [i.e. the body in which the powers of the soul converge] is not the wholebody but a special organ, then that organ is the thing which I believe to be mein essence, unless the meaning of that which I believe to be me is not thisorgan even though it cannot do without the organ. If the quiddity of theessence of that organ – i.e. of its being a heart, a brain, some other thing or anumber of organs capable of this – or the quiddity of their collection is thething of which I am aware that it is me, then it is necessary that my awarenessof me is my awareness of that thing. But the thing cannot be, in one and thesame sense, both what [one] is aware of and what [one] is not aware of. Thecase is not like that, anyway. On the contrary, when I know that I have aheart and a brain, this is through sensation, hearing and experience, notthrough my knowing that I am me. Thus, that organ in itself is not the thingof which I am aware that it is me in essence, but it is me accidentally. What ismeant – and by means of which I know of me that I amme – and what I referto in my saying ‘I sensed, understood, acted and combined these character-istics’ is a different thing, and that is what I call ‘I’.16

Avicenna states that the self cannot be corporeal, because there is no bodythat I could not be unaware of while being aware of myself as anI. Moreover, I am always aware of a body as an object of perception, andthus as something other than the subject of that perception, which the I is. Ican of course subsequently identify the perceived body as myself, but insuch a case the body will be me only accidentally, and as brought out in theargument against the reflection theory of self-awareness, such identificationwith one’s body or recognition of one’s self in one’s body presupposes priorawareness of oneself as one of the two relata of identity. This leads to thefurther problem, highlighted in a related passage from the Mubāh. athāt,

17

that if the self were primarily given as an object in an act of a cognitivefaculty, it would never be given as it is in itself. I can recognize myself in anumber of vastly different objects, between which there need not be anyobvious similarity that would account for the recognition in each and everycase. For instance, no perceivable connection can be established betweenthe figure I see in the mirror and the voice I hear on a recording, yet I canlegitimately recognize both as appearances of myself. Even in cases of

16 Shifā’: Fī al-nafs V.7, 255–256 Rahman. 17 Mubāh. athāt 55, 134 Badawī.


self-intellection, the self is understood as an instance of a universal, and notas the singular entity it is in itself. Thus, if we want to hold on to the viewthat the self is given as it really is in itself in self-awareness, the mediation ofany cognitive faculty must be rejected.

However, the most telling passage for our concerns comes at the end ofthe present quote. Again, self-awareness, my knowledge about myself, theutter familiarity of me being me, comes explicitly to the fore in expressionsof motivated action. This time Avicenna’s formulation is in an emphaticfirst-person singular: it is the same I that perceived, understood, acted andconjoined these different constituents of its experience. But if this is thecase, the I must be separate or separable from each of its acts. Even if the selfwere only given in relation to its acts, it would still remain separate from andindependent of any particular act precisely because it can be the self ofdistinct acts in one and the same sense. The sense of ‘I’ in the acts does notchange, only the acts do.

Avicenna makes a further interesting statement about the self in theimmediately following paragraph where he finally attempts to spell out thepsychological relevance of his remarks hitherto. As I have emphasized, hisargument against the body being the unifying factor behind our experiencehas consistently relied on the claim, which he clearly believes every humanagent or subject will find intuitively plausible, that each of us is aware ofherself as the unifying subject of her experience. Yet Avicenna still has toargue for the move from the intuitive data concerning the self to a psycho-logical conclusion concerning the soul:

Now, if someone said that you do not know that [the I] is a soul, I would saythat I always know it according to the sense in which I call it the soul. I mightnot know it as designated by the word ‘soul’, but when I comprehend what[it is that] I refer to as the soul, I comprehend that it is that thing and that it iswhat uses motive and cognitive instruments. I am ignorant of [the I asdesignated by the word ‘soul’] for only as long as I do not comprehend themeaning of ‘soul’. This is not the case with the heart or the brain, for I maycomprehend the meaning of ‘heart’ and ‘brain’ and not know [the I].When Imean by ‘soul’ that it is the thing which is the origin of those motions andcognitions that belong to me and that end in this collection, I know thateither it is really me or it is me as using this body. It is as if I now was not ableto distinguish the awareness of me pure and simple from [its] being mixedwith the awareness that it [i.e. I] uses the body and is associated with thebody.18



Avicenna thinks that the problem is solved by a mere reiteration of the soul’sdefinition. Once the soul is defined as the agent responsible for psychic actsand as the subject of experience, it becomes immediately evident that the I,which each of us is, is in fact the soul. Cashing out the psychological relevanceof self-awareness is merely amatter of being reminded of the definition of souland the fact that the soul’s functions overlap completely with our awareness ofourselves as agents and subjects. Thus, in spite of the fact that our passageonly speaks of soul (nafs) as the entity studied in psychology, its embedded-ness in a context of extended and explicit appeal to our intuitive familiaritywith ourselves warrants reading it as a neat and revealing summary of whatAvicenna means with the sort of self-awareness pivotal to the flying man andthe argument from unity. The self is nothing other than I, and being aware ofoneself is nothing other than being an I, existing in the first person. WhileAvicenna does speak about awareness of the self in two senses, as pure andsimple and as associated with the body, it seems that in either case awarenessof the self is nothing other than being an I; in one case it is being an I pure andsimple and in the other an I associated with the body. This is just as onewould expect in light of the Ta‘līqāt’s equivocation between self-awarenessand the existence of the immaterial human substance. One need not becomeexplicitly or pronouncedly aware of oneself as an I in order to be one, for thebeing of an I only amounts to existence in the first person.I have dwelt on this fairly straightforward chapter at such length only

because I believe it reveals with particular clarity the central idea behindAvicenna’s concept of self-awareness. As first-personality, self-awarenesssimply designates the fact that, regardless of what contents of experience Iam aware of, they will always be given to me in a first-personal perspective asso many aspects of my experience. The commonplace use of the first-personal indexical ‘I’ points towards this primitive fact of first-personality,and as such to the mode in which the immaterial entity that acts as a soul inrelation to the human body exists in itself. This may seem an excessivelyblunt, straightforward or even simplistic claim to make; why should webelieve that self-awareness or first-personality is primitive or unanalysable,something that need not, or indeed cannot, be described or explained bymeans of more basic epistemological or psychological concepts? Be that as itmay, Avicenna does consider it primitive in this sense. But he does not insiston the claim entirely without argument, for as we have seen, he providessystematic reasons for his view that self-awareness cannot be due to the actof any cognitive faculty. A rather elliptic version of one such argumentfigured in section [c] of the extended passage from the Ta‘līqāt, and weshould now turn to investigate it in detail.


Argument against reflection-based models of self-awareness

The target of Avicenna’s argument is a model of self-awareness as a reflexiverelation of the cognitive subject to herself. This model, which resembles thehigher-order theories popular in contemporary debates on self-awareness,19

is tackled in a slightly extended form in the Ishārāt. Earlier in this context,Avicenna has presented a short version of the flying man in order to showthat self-awareness does not depend on any particular object of cognition.He then proceeds to consider as a counterargument the reflection-theoretical thesis that self-awareness requires a prior act of the self ofwhich the self subsequently becomes reflectively aware:

Perhaps you say: I cannot affirm myself except by means of my action. Thenit is necessary that you have an action that you affirm in the said premise, or amovement or something else. In our consideration of the said premise wehave put those out of your reach.

When we regard the more general matter, if you have affirmed your actionas action in the absolute sense, it is necessary that you affirm an agent of it inthe absolute sense, not in the particular sense. [This agent] is your very self. Ifyou have affirmed [your action] as your action, you do not affirm yourselfthrough it. On the contrary, your self is part of the concept of your act insofaras it is your act. The part is affirmed in the conception preceding it and it isnot made any less by being with it but not through it. Thus, your self is notaffirmed through [your action].20

The counterargument is rather elliptic and leaves room for a possibleambiguity that we should first dispel. It seems to me that two senses of theexpression ‘by means of my action’ are possible here, the second of whichyields two further possibilities of interpretation. First of all, (1) the actionmentioned can be understood as an act of reflection. The point then wouldbe simply to say that I first become aware of myself by reflecting on myself.The other interpretation is (2) that the action by means of which onebecomes aware of oneself is an act prior to reflection which then functionsas the medium of self-awareness. This act, for its part can either (2a) be

19 What is more, Avicenna’s argument bears a striking resemblance to the consistent critique ofreflection theories voiced by representatives of the so-called Heidelberg school of philosophers.This critique was seminally formulated by Dieter Henrich, taking his cue from Fichte, in whose‘original insight’ he thought it first emerges (Henrich 1966 and 1970, 280–284). For an extendedsystematic assessment, see Zahavi 1999, 31–37.

20 Ishārāt, namat. 3, 120. Other instances of the argument can be found inMubāh. athāt 64, 60 Bīdārfar(370, 207 Badawī); and in Ta‘līqāt, 79. In the latter text, Avicenna states that I recognize my action(kuntu u‘arifu) through some kind of mark or characteristic (bi ‘alāmatin min al-‘alāmāt), and thatself-awareness is due to the existence of the form of myself to me in a concrete fashion (li wujūdi s.ūratidhātī fī al-a‘ayāni lī).


self-aware in a primitive sense, or (2b) become self-aware through a reflec-tive attention to it, in which case it would be an objective medium for theemergence of self-awareness. As we shall see, (2a) collapses with Avicenna’sown view, and since he clearly sees a genuine difference of opinion here, wecan rule that interpretation out. On the other hand, (1) and (2b) collapseinsofar as both require reflective attention to a state or an act that is not self-aware, that is, a relation the outcome of which is a self-aware state. The wayin which Avicenna addresses the counterargument suggests that either hemade no distinction between these two senses, or, if he did, he found itinconsequential. Thus, the counterargument represents a higher-ordertheory of self-awareness in a broad and unspecific sense.In spite of its imprecision, the counterargument seems quite plausible

from the point of view of Aristotle’s theory of intellection. As we havelearned, Aristotle states that the human intellect is actually nothing before ithas acquired at least some intelligibles, and that once it has acquired these, itis capable of understanding or thinking about itself at will (yumkinuhu fīdhālika al-waqti an ya‘aqila nafsahu).21 There are of course a number ofpossible interpretations of this dense passage, but one plausible reconstruc-tion is to read it as a theory of self-awareness or self-intellection roughlyequivalent with the claim made in Avicenna’s counterargument.22 On theother hand, the counterargument need not be more than a commonsensicalreaction to the striking claim, made just a few paragraphs earlier, that self-awareness is constant. Given that we explicitly consider ourselves only instates of reflection, it seems natural to assume that self-awareness in fact firstcomes to be at the occurrence of such states. If this is the case, Avicennaneed not argue against any particular rival theory here.Regardless of the provenance of the counterargument, it obviously entails

the position of a first-order non-reflective act that precedes and functions asthe object of the reflective attention through which self-awareness is firstclaimed to emerge. Avicenna begins his response by simply stating thatbecause of this position the counterargument is based on a miscomprehen-sion of the flying man. If one understands the conditions posited in thethought experiment, one realizes that there are no acts upon which one

21 Arist.ūt.ālīs, Fī al-nafs III.4, 426b6–9, 73. Unfortunately, Ish. āq’s translation of this passage has notbeen preserved.

22 For a lucid discussion of Aristotle’s theory of self-knowledge, see Lewis 1996. It must be emphasizedthat self-knowledge in this sense should be distinguished from consciousness, or our awareness thatwe perceive, think, act and so forth. Aristotle seems not to have considered in explicit terms thequestion of whether consciousness entails self-awareness in the particular sense meant by Avicenna; inany case the inherent first-personality of all consciousness never becomes a pronounced topic for him.See Chapter 1.1, 12–15.


could reflect. But most important for our concerns, Avicenna decides togive his opponent the benefit of further doubt in order to show that herargument can be refuted on its own terms, independently from the flyingman. Avicenna thus uses the counterargument as a means to corroborate hisown thesis that self-awareness is immediate and epistemically primitive.

Suppose, then, that we first become aware of ourselves through an act ofreflection. In such a case, two possibilities emerge: either the pre-reflectiveact is an act in an absolute sense, that is, not the act of any particular agent,or it is an act of a particular agent. In the first case, the object of reflectionwill be simply an act or a mental state with certain determinations (an act ofthinking, walking, hitting and so forth; or a state of perception, imaginationor intellection) and certain objective content (a thought concerning some-thing, walking somewhere, hitting something; perception, imagination orintellection of something). Such an act or state could belong to anyone, andby the same token, is not actually particular to any agent. This being thecase, Avicenna asks how one can identify such an anonymous act or state asone’s own in reflection. If the state is not particular to any subject, it cannotbemine either. There are no non-arbitrary means of introducing any kind of‘mineness’ in the act of reflection, for if the reflection is supposed to be atruthful act of intellection, it should grasp the first-order non-reflective stateas it is, without any additions or distortions. But if the self is nowhere to berecognized in the object of reflection, why should the act of reflection giverise to any sort of self-awareness any more thanmy current perception of anyrandom object, such as the pine outside my window. In this regard, thenon-reflective state will be an object like any other, not earmarked to anyparticular cognitive subject, and the reflection turns out to not be reflexivein any meaningful sense after all.

If we want to hold on to the possibility of reflective self-awareness, thestate prior to reflection must already be characterized as my state in somemanner. The reflective act will then recognize this characteristic of the statealong with its own determinations, and as a result I will be aware of the priorstate as something I myself have undergone. By all accounts, this would be aperfectly legitimate account of reflection for Avicenna. However, the phe-nomenon to be explained was not reflective self-awareness but our beingself-aware in the first place. Thus, Avicenna can say that his opponent has toconcede a certain characteristic of myness in the non-reflective first-orderstate and therefore all but admit that reflective self-awareness is based on amore primitive type of self-awareness. The order of explanation is invertedand the counterargument turns out to be based on a petitio principii; it failsto explain self-awareness because it ends up presupposing self-awareness in


the very explanation. Avicenna formulates the point very carefully: the factthat self-awareness occurs together with an act or experiential content, thatis, that my first-personality is rarely bare but rather most often connected toacts or objects that are first-personally given, does not make it subsequent tothose acts or dependent on those objects. Rather, in these cases just as in therather less usual case of the flying man, in which self-awareness figuresdivested of all further determinations of the person’s acts or experiences,self-awareness is an irreducible constituent of the act or the experience as anact or an experience that belongs to a particular subject.Again, the point of spelling this admirably clear argument out at such

length is to make explicit the concept of self-awareness it hinges upon. Iclaim that the most natural interpretation here, just as in the case of theargument from unity, is to say that Avicenna’s concept of self-awarenesspurports to grasp nothing more than the fact that all my experience isqualified as mine, that no matter what I apprehend I will apprehend as an I,and that every act of mine is performed by me. In other words, whenspeaking of self-awareness Avicenna attempts to pick out the first-personality inherent in all human beings, and that everything about us isgiven in the first person, to a first-personal subject. Nothing more is meant,but nothing less either. Self-awareness is being an I, which is the mode ofexistence proper to an immaterial entity.

Argument from personal identity

The sixth book of the Mubāh. athāt introduces a third argument revolvingaround the phenomenon of self-awareness. The argument is embedded ina discussion of the endurance of individual substances through the changeof their attributes. Against this background we suddenly come across aclear statement of dissatisfaction with the flying man voiced by one ofAvicenna’s most illustrious students, Bahmanyār ibn Marzubān(d. 1066).23

I want that a method other than self-awareness (al-shu‘ūri bi al-dhāt) beapplied to me in showing that,24 for I have already attempted that myself (fainnī qad jārabtu nafsī fī dhālika),25 and in my opinion it is deceptive – in spite

23 For Bahmanyār’s relation to Avicenna, see Reisman 2002, 185–195; and Janssens 2003b.24 That is, in showing what the endurance of individual substances is due to.25 This refers to the argument on self-awareness, that is, the flying man; Bahmanyār has performed the

thought experiment but is dissatisfied with it.


of its validity26 – and I want to know it in another way, so that my heart willfind peace.27

Avicenna’s answer takes its cue from the question posed by the context:

The persistence of a numerically one thing is that it persists as numericallyone in terms not of its quantity and quality but in terms of its substance.

Then my persistence as a single I in terms of my substantial thatness(thumma thabātī anā wāh. idan bi innīyatī al-jawharīya), [the fact] that whatexisted yesterday has not perished or ceased to exist while numericallyanother has come to be, that I am that observer of what I observed yesterdayand the one remembering what I have forgotten from what I observedyesterday – [all this] is something about which no doubt occurs to me, andsimilarly I have not come to be today, nor is my body something that wascorrupted yesterday, I will not cease to exist tomorrow andmy individual willnot be corrupted, even if my time should come tomorrow and a substanceother than me should come to be.

Thus, if he whose servant I am28 is of the opinion that he has come to betoday from his simile that was corrupted yesterday, and that he is not thatwhich existed yesterday, but that he is renewed in substance just as he isrenewed in states, let him be of that opinion and have that view, and let himask in another place for an additional explanation for this proof.29

Avicenna believes he can point to an indubitable intuition in his inter-locutor’s experience, which provides immediate evidence of the enduranceof the interlocutor’s very substance in the midst of the constant change he issubject to in terms of his attributes. That evidence is provided by theinterlocutor’s remaining the same I, that is, by his awareness of himself asthe subject that is aware of itself now just as it was aware of itself the daybefore.

The argument is not entirely unrelated to the flying man with whichBahmanyār was expressly dissatisfied; it characterizes its nodal point by thevery same term as the version of flyingman in Shifā’: Fī al-nafsV.7, pointingour attention to a single ‘thatness’ (innīya) at the core of our mental life, andsuggesting that the recognition of the correct nature of this thatness isinstrumental to reaching the desired conclusion. But there are also cleardifferences between the two arguments. Indeed, where the flying man is

26 The reference of the suffixes here is ambiguous, but I believe the most sensible reading is: ‘and in myopinion [the flying man] is deceptive – in spite of the validity of [the claim that the individualsubstance endures through the change of its attributes]’.

27 Mubāh. athāt VI.402, 146–147 Bīdārfar.28 That is, Bahmanyār.

29 Mubāh. athāt VI.403, 147 Bīdārfar (453, 227 Badawī); cf. III.94, 68 Bīdārfar (cf. 430, 224 Badawī); andVI.456–460, 163–165 Bīdārfar (39, 128–129, and 454, 226–227 Badawī). For discussion, see Sebti2000, 110–111.


about the independent existence of this thatness, this argument frompersonal identity hinges on its endurance through the variation of itsattributes. Also, the flying man was designed to argue for the independencefrom the body of that which functions as a soul in us, whereas this argumentaims to single out that which is substantial in us and to distinguish it fromits quality, quantity, and other features that are subject to constant varia-tion. I would like to argue however that, these clear differences notwith-standing, the similarities are more important. First of all, both argumentstry to point attention to something readily available from our experience,and to do this by separating this something from the accidental features wecommonly find it associated with, whether particular observations, recol-lections and so forth, or the relation to the body in its entirety. Secondly,and more importantly, it seems that the thing pointed to in the two argu-ments is one and the same, the interlocutor’s awareness of himself as an I. Itis true that they emphasize distinct aspects of this self-awareness: the flyingman its epistemic and metaphysical independence, the argument frompersonal identity its endurance and identity in change. But such differencesin emphasis need not entail a difference in the object of attention; indeed,both arguments are readily understandable if we suppose that they rely onone and the same phenomenon.But if the argument from personal identity builds on the same evidence

as the flying man, why should Avicenna expect it to prove any morepersuasive? We have to notice that in the last paragraph he recognizes thepossibility that Bahmanyār will remain unconvinced even after the consid-eration of this new piece of correspondence, but it is hard to avoid perceiv-ing a scarcely hidden irony in his manner of addressing the student. This isprobably due to the commonsensicality of his new argument: Avicennaseems to hold that common sense and human social life require thisawareness of persistent first-personality of each of us. If we did not rely onthe endurance of ourselves, we would have no basis upon which to plan ourfuture, make agreements or promises, or give reliable accounts of pastobservations. The basis of that reliance, the fact that each of us is first andforemost an I, is all that Avicenna is attempting to point to here. It is for thesake of this purpose that he identifies being an I with being a substance, forhe clearly presumes Bahmanyār to subscribe to the foundations ofAristotelian metaphysics – the sort of thing that endures through changeis a substance. Thus, although the flying man and the argument frompersonal identity are founded on the same evidence, Avicenna does have apoint in supposing the latter to be more palatable; instead of relying on an


imaginative thought experiment, it hinges on a necessary condition of oneof our most commonplace beliefs.

Finally, regardless of the shared ground between the flying man and theargument from personal identity, there remains the question of whether thelatter is really relevant for our present purpose, namely the reconstruction ofAvicenna’s concept of self-awareness. This is a pressing question for tworeasons. First of all, because Avicenna does not mention self-awareness orthe self anywhere in the text, and although he does explicitly speak aboutthe first person, or the I, this is only to provide evidence for the persistenceof the substantial core in each of us. Secondly, in the context ofMubāh. athātVI, the argument from personal identity is provided as an alternative to theflying man, which Bahmanyār in his question describes as an argumentfrom self-awareness. Should we thus not expect this argument to deal withsomething decidedly different?

The second concern can, I believe, be met by the aforementioneddistinction between differences in emphasis on a single phenomenon.Whereas it seems natural to call the flying man an argument from self-awareness – it is, after all, based on distinguishing the mode of our awarenessof ourselves from the mode of our awareness of other things, our bodies inparticular – the present argument points out a distinction in the inter-locutor’s experience between a persisting feature, hinting at his substance,and varying features, hinting at his various accidents, without any mentionof differences in the respective modes of grasping these two types of features.But this can be explained as due to a difference in application, while thefeature applied in the two arguments is one and the same: the fact that eachof us, Bahmanyār included, is first, foremost and always an I.

As regards the first concern, it is clear that the argument from personalidentity is not meant to define self-awareness or simply point our attentionto it. Instead, just as in the case of the flying man and the argument from theunity of experience, the phenomenon of first-personality is cast in anexclusively instrumental explanatory role; it becomes the focus of attentiononly in order to transfer us to the ultimate conclusion, whether that is theindependent existence of the entity that functions as soul in the humanbody or, as here, the endurance of the human substance. Successful per-formance in this explanatory role does not depend on a specific term ofdescription; on the contrary, a technical term such as shu‘ūr bi al-dhāt mayprove counterproductive by revoking the sort of technical questions theflying man had given rise to. For our purpose, it is crucial to recognize thesimilarities in the experiential data on which the arguments rely, and forthese I have already argued at length.


The relevance of self-awareness to the argument from personal identityis also corroborated by the fact that the argument makes explicit ametaphysical entailment which is of considerable importance to all ofAvicenna’s discussions of self-awareness, even when its formulation is leftimplicit. This is the claim that the narrowly defined but enduring Iprovides us with unique information of our respective substances, themetaphysical bases of our very being: any attribute of mine, whether myact or a content of my experience, can change, but the I, the substance towhich and the perspective from which those attributes appear, will remainintact. In this sense, the substantial core of human being, unlike that ofother animals, is not available by abstracting from our third-personalapprehension of the informing functions it performs on the body – afact evidenced by the need for a separate reminder at the end of the reviewof the soul’s functions in Shifā’: Fī al-nafs I.1. This implication is explicitin a passage from the relatively early Risāla al-ad

˙h. awīya fī al-ma‘ād,30 in

which Avicenna explicitly distinguishes between ‘the thing due to which“he” (huwa) is said of [a human being]’, and the way in which ‘he says “I”of himself (yaqūlu li nafsihi anā)’.31 Later on in the same context Avicennadescribes the reference of the first-person indexical in a manner thatexpresses in a concise form the central insight behind all the argumentswe have been considering:

When it comes to truth, the human, or the thing considered of the humanon which the meaning of I is based in him (huwa al-wāqi‘u ‘alayhi ma‘nā anāminhu), is his real self (dhātuhu al-h. aqīqīya); it is the thing of which he knowsthat he is it, and it is certainly the soul.32

In light of the foregoing analysis, the density of the passage is not anobstacle to its clarity: the I is the real self and essence of a human being,and thereby his substance, and it is the thing discussed as soul in psychol-ogy. Now, if the Risāla al-ad

˙h. awīya is indeed an early work, and if the

sixth book of theMubāh. athāt can be dated to the last decade of Avicenna’slife,33 this insight into first-personality and its argumentative potentialinformed the entire period of gestation of his most important philosoph-ical works.

30 For an argument for the early dating (1014–1015) as well as a brief consideration of rival views, seeMichot 2003a, 149–151; cf., however, Marmura 2008, 132.

31 Al-Risāla al-ad˙h. awīya IV, 127.

32 Al-Risāla al-ad˙h. awīya IV, 128.

33 As is suggested by Reisman 2002, 233–239, 252.


4.2 First-personality, the flying man and incorporeal existence

Having thus reconstructed the Avicennian concept of self-awareness, let usconsider it in relation to the two conditions we stated at the beginning ofthis chapter. The questions to ask are (1) whether the reconstructed conceptof self-awareness is plausible given that the phenomenon underlying itshould be constant, and (2) whether it makes sense to suppose that self-awareness in the intended sense is readily available to each and every one ofus in our own experience, so that by means of it we can retain whatevercapacity of persuasion the flying man had to begin with.

(1) Let us approach the question of the constancy of self-awareness byconsidering the cases of the sleeping and the intoxicated person whichAvicenna mentions together with the Ishārāt version of the flying man:

Return to yourself (irja‘ ilā nafsika) and consider whether, if you werehealthy or even in some other state of yours, [but] so that you grasp thematter accurately, you would be ignorant of the existence of yourself(dhātika) and would not affirm yourself (nafsaka). I do not think thiswould happen to the perspicacious. Even in the case of a sleeper in hissleep or a drunk person in his drunkenness, his self will not escape his self (lātaghrubu dhātuhu ‘an dhātihi), even if no representation of him in himself(tamaththuluhu li dhātihi) were left in his memory. If you imagine your self(law tawahhamta dhātaka) to have been initially created perfect in intellectand configuration, and [your self] is supposed to be all in all in such positionand configuration that its parts are not seen and its limbs do not touch eachother but are rather momentarily suspended apart in open air, you will find it(wajadtahā) even though you are ignorant of everything but the persistenceof its thatness (thubūti annīyatihā).34

Considered in isolation, the text allows for two ways of understanding therespective states of the two persons. Either the sleep and the intoxication areso profound that the persons are in states we would now call unconscious,that is, lacking consciousness of any cognitive input whatsoever, or they arein states that are not completely unconscious but still in some relevant sensedifferent from the healthy or sound states mentioned in the text. Avicenna’sdense characterization of the latter (‘so that you grasp the matter accurately’)suggests that the relevant difference between sound and unsound statesconcerns the human capacity (or lack thereof) to reflect discerningly ononeself and to thereby assert the existence of one’s self. This amounts tosaying that the sleeping and the drunk person are incapable of turning theirattention to their self-awareness, but this does not necessarily rule out their

34 Ishārāt, namat. 3, 119.


being non-reflectively aware of a variety of things, from angels and pinkelephants to themselves.But since the sleeper and the drunk are followed by the much clearer and

stricter conditions set in the flying man, one easily wonders why Avicennaconsidered it worth his while to introduce them in the first place. If there issomething going on in their minds, the sleeper and the drunk will be lessproblematic cases than the flying man, for surely it is more plausible toassume that someone actually aware of something is also aware of herselfthan to assume that someone aware of nothing is nevertheless self-aware.On the other hand, if the sleeper and the drunk are lacking awarenessaltogether, their case seems to be identical to that of the flying man.35

Avicenna’s motive for considering the two examples becomes clear in acorresponding section of the Mubāh. athāt where their introduction isattributed to his opponent, Abū al-Qāsim al-Kirmānī.36 Here it becomesobvious that we should understand the sleeper and the drunk to be in theless stringent condition; the sleeper may well be dreaming and thereforeentertaining all sorts of experiential content, and mutatis mutandis this willhold of the intoxicated person as well. Avicenna neglects any mention of thedreaming person’s lack of reflective capacity here, possibly because he findsit sufficient to simply show that something is going on in the person’smind.37 Indeed, this seems a natural move in the light of our reconstructionof his concept of self-awareness: if self-awareness amounts to nothing butthe first-personal perspective to whatever is in one’s mind, then the presenceof any mental content will indeed be enough.However, since the flying man is prominently featured in the context of

the Mubāh. athāt as well,38 we still have to solve the second problem: why

worry about the weaker case if you have at your disposal an argument thatwill clear the table anyway? The need for this extension of the initial

35 Fakhr al-Dīn al-Rāzī, Avicenna’s critical commentator, interprets the different states as exemplifyingdistinct degrees of discernment (fit.na). According to him, in sleep only the external senses arerendered dysfunctional. Intoxication, on the other hand, amounts to the inefficiency of both externaland internal senses, and therefore entails a more serious lapse of consciousness. See Sharh. al-Ishārāt,121–122; and for discussion and translation of the passage, Marmura 1991.

36 This attribution may be historically true, for if Mubāh. athāt III was written after the acrimoniousencounter with al-Kirmānī in Rayy in 1030 (as argued in Reisman 2002, 216–219, with a refutationof Michot’s (1997, 153–163) alternative dating), it may well have incorporated elements of thatdiscussion. Moreover, Ishārāt is also commonly dated to a late period in Avicenna’s career (1030–1034 in Gutas 1988, 140–141), and if this is right, Avicenna may have included a tacit reference to thedebate in it.

37 Mubāh. athāt 380, 210 Badawī; III.68, 61 Bīdārfar.38 Indeed, the entire section consists of Avicenna’s responses to al-Kirmānī’s critical remarks concerning

the flying man.

First-personality, the flying man and incorporeal existence 81

argument is due to a briefly cited remark by al-Kirmānī who, seeminglyunhappy with the highly fanciful situation of the flying man, attempts tobring the discussion of self-awareness down to the level of everyday phe-nomena.39 It is for this reason that he introduces the case of the sleeper:whatever our verdict on the flying man, surely it makes no sense to claimthat a sleeping person is aware of herself. Had I been aware of myselfthroughout the night, even if I was aware of nothing else, surely I shouldbe able to remember at least having been aware of that peculiar nothingness.By the same token, if I was aware of myself throughout my intoxication,should it not have felt like something to pass out, and should I not be ablenow to remember what it was like? Since no such recollections seem to beforthcoming, al-Kirmānī suggests that we should deny Avicenna’s argumentfrom the constancy of self-awareness to the immateriality of the humansubstance, and thereby also reject his concept of self-awareness.

The fact that Avicenna wants to integrate the sleeper and the intoxicatedperson in the Ishārāt’s introductory argument for the immateriality of thehuman substance suggests that he is fully aware of the initial implausibilityof his claim that our self-awareness is constant, and that he believes he candeal with the contrary intuition. True to the general method of the Ishārāt,his answer is condensed, but it can be extrapolated bymeans of the extendedversion in the Mubāh. athāt:

The sleeper operates on his imaginations (khayālātihi) just as he operates on[the things] he senses when awake, and he often operates on intellectual andcogitative things just as when awake. In the state of his operation on that he isaware that he is that operator just as he is at the state of being awake, and so ifhe notices and remembers his operations, he remembers his awareness ofhimself (shu‘ūrahu bi dhātihi), but if he notices but does not remember that,he does not remember his awareness of himself (shu‘ūrahu bi dhātihi). Thatdoes not indicate that he wasn’t aware of himself (shā‘iran bi dhātihi). Rather,awareness of self-awareness is different from self-awareness (al-shu‘ūru bial-shu‘ūri bi al-dhāti ghayru al-shu‘ūri bi al-dhāt), and so remembering self-awareness is different from self-awareness (dhikra al-shu‘ūri bi al-dhāti ghayrual-shu‘ūri bi al-dhāt). Even one who is awake may not remember his aware-ness of himself (shu‘ūruhu bi dhātihi) when the pursuits, which he had andduring which he was not unaware of himself (lam yaghfulu fīhā ‘an dhātihi),are not retained in his memory.40

Avicenna’s argument is admirably clear: it is one thing to be aware of oneselfand another thing to have second-order awareness of this self-awareness,

39 Mubāh. athāt III.65–66, 60–61 Bīdārfar; cf. 380, 210 Badawī.40 Mubāh. athāt III.68, 61 Bīdārfar; cf. 380, 210 Badawī.


such as when one remembers that one was aware of oneself, and no lack ofthe latter warrants an inference to a corresponding lack of the former. In thisregard, self-awareness is not different from awareness of other things, suchas one’s own acts or ‘pursuits’ (muzāwalāt) or the objects of one’s experi-ence; no jury will relieve a defendant simply because she fails to remembercommitting the crime she is charged with. No matter how sincere thisfailure is, it does not exclude the possibility that the defendant in fact didcommit the crime – and if she did, she was certainly aware of herself in theintended sense.This extended version of the argument is fully coherent with Avicenna’s

psychological theory of memory. For him, memory is an internal sense thefunction of which is to retain the peculiar cognitive objects he calls mean-ings (ma‘ānī). The exact nature of meanings is a matter of scholarly debate,41

but in this context it suffices to recognize that they are constituents ofexperience that, among other things, enable us to bring past experiences tomind. Whatever the exact mechanism according to which meanings per-form this task, it is clear that recollection will not take place without them.Moreover, it is equally uncontroversial to say that meanings are determi-nations of the objective content of experience, that is, aspects that areinseparable from the appearance of some object of awareness. Thus, whatdistinguishes the sleeping and the intoxicated person from a human being ina wakeful and sober state is that no meanings belonging to their respectiveexperiences are left in their memories. As a result, they have no access totheir past experiences, and it is only because of this that their awareness ofthemselves as the subjects of those particular experiences is also unavailableto them. While Avicenna does not go into great detail concerning thequestions of why this is the case and what causes the cessation in the actof memory in sleep or intoxication, it is clear that the account would have toinvolve the liberation of the faculty of imagination from the governance ofestimation and its apprehension of meanings.42 But regardless of thedetailed reasons for these kinds of ‘meaningless’ mental states and theconsequent lack of memory in the two cases of the sleeper and the intoxi-cated person, Avicenna’s point remains straightforward: continuity in thesense required of the primitive type of self-awareness does not entail that weshould remember every past moment in the continuity of our self-awaremental life. On the contrary, owing to the nature of the meanings stored inmemory, any recollection of a past moment requires remembering an

41 For memory and meanings, see above Chapter 1.2, 25–27.42 Cf. the scattered remarks in Shifā’: Fī al-nafs IV.2, 170–171, 174–176 and 181–182 Rahman.


experience with some objective content.43 Or to be more precise, remem-bering amounts to bringing that content to mind anew. But when theestimative grasp of meanings is cut out, nothing can be retained in memory,and nothing retrieved from it as content of past experience. There willsimply be nothing to remember.

In his answer Avicenna therefore grants that we have no immediateknowledge of what the states of sleep and intoxication were like, becausewe have no memory of those states and therefore no access whatsoever tothem. I have no idea of what went on in my mind between the moments ofmy falling asleep and waking up. Although I can infer, by taking a look at anexternal indicator of time such as the bedside alarm clock or the sky, thatthere is a yawning temporal gap between my last recollection of the nightbefore and mymorning state of awareness, I will nevertheless lack first-handexperiential knowledge of passing the gap. All I have is knowledge byinference. Avicenna’s point is, however, that the assumption of the gap,while entirely correct and warranted in its own right, does not allow us toconclude a corresponding gap, or indeed any kind of breach, in our self-awareness. Self-awareness and the memory of having been self-aware aretwo entirely different matters. Even if sleep and intoxication can be claimedto prevent us from the latter, they do not warrant our drawing anyconclusions about self-awareness, because the alleged lapses from it can beexplained as mere absence of memory-traces.

It must of course be added that the distinction between being aware ofoneself and remembering having been aware of oneself in no way demon-strates Avicenna’s thesis about the continuity of self-awareness either. Butthis is not a problem for him, because a further demonstration is not whathe is after in this debate with al-Kirmānī. Rather, if his thesis about theconstancy of self-awareness turns out to be a natural consequence of thepsychological tradition and its conception of the intellect, the burden ofproof is returned to the challenging interlocutor. In other words, if humansouls in themselves are immaterial entities – a thesis Avicenna took to bedemonstrable by other means – they must be somehow there and awareeven when the body’s lights are off. Moreover, if the primitive type of self-awareness is in fact the most minimal account of what such existence-cum-awareness might consist of, the interlocutor suddenly appears to stand onmuch less secure ground. It suffices for Avicenna to be able to show thatsleep and intoxication do not force us to infer any straightforward denial ofthe constancy of self-awareness and that the claim of those states lacking

43 Cf. Mubāh. athāt 380, 210 Badawī; III.68, 61 Bīdārfar.


self-awareness is no less an undemonstrated one than his claim – indeed, inthe light of other concerns it turns out to stand on shakier ground. Thus,Avicenna can consistently say that self-awareness, in the narrow sense offirst-personality, is constant and continuous, and therefore something hecan consistently equivocate with human existence.(2) The other condition that the reconstruction of Avicenna’s concept of

self-awareness should fulfil is due to a particular aspect of the flying manargument. As I emphasized in discussing the thought experiment, itsplausibility hinges on the interlocutor’s capacity of locating the phenom-enon of self-awareness in her own experience. Although Avicenna doesdemand a certain amount of wit from the interlocutor, it is important tonotice that this is only required to be able to pay second-order analyticattention to something any normally developed human being will have as anecessary constituent of her experience. This abstractive feat may be beyondthe capacity of some people,44 but its object is not generated in it. On thecontrary, in order that the thought experiment prove plausible in the firstplace, it is crucial that we deal not with any special state of mind but insteadwith something familiar to each of us from a perfectly commonplace humanexperience.45

I believe that the reconstructed concept of self-awareness will survive thetest. Although not uncontroversial, it nevertheless makes reasonable senseto say that if I have successfully bracketed all objective content of myexperience, I will still be left with the peculiar mode of givenness orappearance of all such content, that is, its appearance to me. In otherwords, the fact that I am an I will remain in spite of the absence of allfurther determinations of that I, whether as tasting an apple, thinking aboutthe identity of indiscernibles, fancying a beer or walking about in the park.The flying man’s unusual experience of nothing will be my experience inthis commonplace sense.Furthermore, this reconstruction is capable of retaining the argument’s

motivation as well as the precise technical sense in which Avicenna speaks ofit as a reminder. Consider the brief description of this type of argument inShifā’: Fī al-nafs V.7, that is, in the very context in which Avicenna presentsa shorter version of the flying man.

44 Cf. Avicenna’s scarcely veiled diatribe against al-Kirmānī in Mubāh. athāt III.56–65, 58–60 Bīdārfar(partly in 370, 207 Badawī).

45 Although this point may seem trivial, it has been debated (see Chapter 2.2, 38–41). Moreover, thedebate with al-Kirmānī in Mubāh. athāt III suggests that it was not entirely obvious to Avicenna’scontemporaries either.


[I]t is not when I am investigating whether [the thing which governs thebody] exists and whether it is not a body that I am wholly ignorant of it,rather I do not pay attention to it. It is often the case that knowledge aboutsomething is close at hand, but one does not pay attention to it, so that itverges on the unknown and is investigated at the greatest remove. Sometimesknowledge that is close at hand is like the reminder, which is lost throughinadequate effort, so that one’s wit, due to the weakness of [its] grasp, doesnot find the way to it, and then one needs to approach it from afar.46

As a reminder, the flying man is designed to make us aware of somethingthat is particularly close to us but most often tends to escape our attention.Now, it does not seem far-fetched to say that our first-personality isprecisely such a thing. For the most part, we are immersed in and preoccu-pied with the various contents of our experience, lacking either a reason,time or both to concentrate on the fact that those contents are first-personally given and that each of us is there to experience them in eachcase as our own. In this regard, self-awareness is exactly like the firstintelligibles, to the simultaneous presence and absence of which in ourthought Avicenna explicitly compares it.47 When we apply the law ofnon-contradiction in everyday reasoning, for instance when I attempt topersuade my daughter that she cannot both go out to play and stay sitting infront of the television, we rarely pause to consider the law itself but ratherconcentrate on the application.Were someone to ask us about this aspect ofour reasoning, we may even find ourselves utterly confused and incapable offormulating a coherent general law apart from the concrete instances of itsapplication, let alone demonstrate the law. Yet when a teacher of logicbrings the law to our attention by sufficiently clever pedagogical means,perhaps indeed arguing for it indirectly by means of an Avicennianreminder, we will eventually recognize the intuitive plausibility of the lawas something we have in fact been relying on all along. This is exactly thekind of plausibility or familiarity Avicenna expects his concept of self-awareness to be found to have.

If we have a look at the discussion related to the flying man in theMubāh. athāt, it becomes clear that in this case the teacher’s intention wasnot particularly easy to grasp. Most of the discussion revolves around criticalremarks that arise from the interlocutor’s incapacity or refusal to recognize amode of awareness that is not based on the normal functioning of any of thehuman cognitive capacities.48 This, however, need not be fatal for our

46 Shifā’: Fī al-nafs V.7, 257 Rahman.47 Shifā’: Fī al-nafs V.7, 257 Rahman; cf. Shifā’: al-Ilāhīyāt I.5.1–5, 22–23.48 See Mubāh. athāt III.56–74, 58–61 Bīdārfar; for discussion, cf. Michot 1997, 168–174.


reconstruction of Avicenna’s concept of self-awareness. To begin with, theconcept of self-awareness Avicenna introduces is a novel one, and thephenomenon on which it is based has no ready-made niche in the precedingpsychological tradition. There is certainly nothing extraordinary in a phil-osophical novelty proving to be difficult to palate.But there is also a further difficulty, which is due to a peculiarity of the

flying man, and indeed any attempt to pay explicit attention to primitiveself-awareness, for performing the thought experiment involves at leastthree distinct levels of awareness, a confusion between which is both easyand understandable. On the first-order level we have the constant pre-reflective self-awareness that is prior to the thought experiment and entirelyindependent of whether or not we ever perform anything like it. This levelshould be distinguished from the higher-order consideration involved inactually performing the imaginative manoeuvres which the thought experi-ment requires of us and which are designed to bring us to a state in which wenotice something about the first-order state that we had not paid attentionto before. Furthermore, we can distinguish another higher-order level onwhich we draw conclusions about first-order awareness on the basis of theresults of our second-order state that resulted from a successful performanceof the thought experiment. The two higher-order states of awareness arereflective acts in relation to the first-order awareness, but it is very easy toconfuse at the third level of psychological conclusions the act of reflection,which has the first-order state as its particular object, with self-awarenesspure and simple, by holding that the first-order state is somehow trans-formed from a state lacking self-awareness to a self-aware one by thereflective act. Avicenna’s explicit claim, however, is that the thought experi-ment points towards something that was already there. All that it brings intobeing is our attention to that something.49 Failure to realize this will resultin exactly the kind of criticism the Mubāh. athāt is witness to: that our firstbecoming self-aware requires a particular intellectual process, such as per-forming the thought experiment of the flying man.50 Avicenna, however,attributes the confusion to al-Kirmānī’s lack of required wit, showing quitestriking impatience towards his older peer.Finally, the use of self-awareness in the flying man entails a potential

third problem that we should address. This is because the argument clearly

49 This is also evident from the argument against reflection-based models of self-awareness, whichAvicenna reiterates in the context of the Mubāh. athāt dealing with the flying man.

50 This confusion is addressed in Suhrawardī’s discussion of the flying man in Talwīh. at III.4.3, 156Habībī; see Chapter 5.1, 107–108.


relies on self-awareness as an important piece of evidence for the humanself’s incorporeality. Now, if self-awareness is nothing but being an I andtherefore not based on any constant act of intellection with a specific object,the evidence it provides must be different from that applied in Avicenna’sproof from the indivisibility of intelligible objects and intellectual subjects.51

The question then is, what sort of evidence can self-awareness provide?Whycould it not just as well be a feature of a corporeal subject?

Here again it is important to distinguish clearly between the initial levelof self-awareness and the reflective attention paid to it in the course of thethought experiment. It is not self-awareness as such that provides evidenceof our incorporeality – were that the case, the question would not havebeen debated in the first place. Rather, crucial to the argument is the factthat our awareness of ourselves is readily given to us as an object ofreflection, independent of any mediating cognitive instruments or acts.The capacity to reflect presupposes incorporeality, for corporeal things areby definition incapable of such a wholesome relation to themselves.52 Butit also presupposes that the object reflected upon is available to thereflecting subject, in other words that the subject is already aware of itselfin some manner before it turns its reflective gaze to itself. In the Ta‘līqāt,Avicenna refers to this givenness by means of various formulations basedon the notion of the self’s presence (h. ud

˙ūr) to itself, a term which gains in

stature in the twelfth-century reception of Avicenna that we will soon turnto study.53 The idea is that the self is constantly present to itself, regardlessof whether anything else is present to it, or whether it ever turns to reflecton this presence. The mere capacity of performing this turn, which thethought experiment of the flying man by necessity actualizes for itsperformer, is sufficient to indicate the self’s incorporeality, and asAvicenna states explicitly in the argument against reflection-based modelsof self-awareness, self-awareness in the simple sense of presence is anecessary condition of the reflective capacity. It is in this sense that self-awareness provides evidence of the incorporeality of the self, quite inde-pendently of the inference from the indivisibility of intellectual objects tothe indivisibility, and the consequent incorporeality, of the subject ofintellection.

51 For this proof, see Shifā’: Fī al-nafs V.2, 211–218 Rahman; and Chapter 3.1, 43.52 In the psychological context, this amounts to denying the claim that the corporeal cognitive faculties

are capable of apprehending their own acts. See Najāt II.6, 218.53 Cf., for instance, Ta‘līqāt, 69, 79, 148, 162. For ‘presence’ in Rāzī and Suhrawardī, see Chapter 6.1.


4.3 Self-awareness, reflection and intellection

Our characterization of primitive self-awareness by setting it in contra-distinction to explicit self-reflection naturally gives rise to a question: doesAvicenna have a concept of reflectivity that is developed enough to provide asufficient basis for the distinction? I would therefore like to conclude thischapter with an argument for the view that although his remarks are some-what scattered, we can reconstruct a genuinely Avicennian concept ofreflectivity that is not only systematically and historically plausible butalso intimately connected to, indeed relying on, his concept of primitiveself-awareness. Finally, with the reconstructed concept of reflectivity as ourcue, we can perhaps better dispel some of the doubts caused by certainproblematic passages from theMubāh. athāt that are well known but that wehave so far refrained from addressing.The central insight of Avicenna’s concept of reflection was introduced in

nuce towards the end of section [b] of the lengthy passage from the Ta‘līqāt.

Self-awareness is actual for the soul, so that it is constantly aware of itself. Asregards awareness of the awareness, it is potential; if awareness of the awarenesswere actual, it would be constant and no consideration of the intellect (i‘tibāral-‘aql) would be needed in it.54

Here, Avicenna explicitly distinguishes reflective self-awareness from prim-itive self-awareness by referring to the former as a higher-order ‘awareness ofawareness’, and by describing the higher-order awareness as potential,intermittent and requiring an intellectual effort of consideration. Thebrevity of the passage belies the fact that it expresses a reasonably consideredview, for these statements clearly echo a more elaborate account found inthe Ishārāt:

You know that anything, which understands something, understands by apotentiality close to actuality (bi al-quwwati al-qarībati min al-fi‘l) that itunderstands [that thing], and that is intellection of [the thing] for it (wadhālika ‘aqlun minhu li dhātihi). Thus, whatever understands something isable to understand itself (lahu an ya‘aqila dhātahu).55

Avicenna qualifies the human potential for reflective self-awareness as‘potentiality close to actuality’. This notion stems from Aristotle, who in awell-known passage from De anima II.5 distinguishes between the threesenses in which a human being can be said to know things. In the weakestsense, a person is a potential knower by virtue of the fact that she is a

54 Ta‘līqāt, 161; emphases added. 55 Ishārāt, namat. 3, 132.

Self-awareness, reflection and intellection 89

member of the human species, an essential characteristic of which is thecapacity to know. A person is potentially knowing in a stronger sense, if shehas acquired a piece of knowledge through prior insight, experience orinstruction, and thus has access to that knowledge at will, but is notpresently engaged in the consideration of her knowledge. Finally, a personis actually knowing, if she has acquired knowledge and is presently engagedin the consideration of her knowledge.56 Avicenna rephrases thisAristotelian idea to accord with his theory of knowledge as a combinationof abstractive effort and receptivity to intellectual emanation, but retainsthe central distinction between the two types of cognitive potency. In thediscussion of learning in Shifā’: Fī al-nafs V.6, he defines learning as theacquisition of a capacity to connect to the active intellect at will in order toreceive the content of thought that one has learned, or to put it another way,learning is the capacity to bring to mind the preparedness, for instance acertain mental image or linguistic symbol, that is required for the renewedreception of the learned piece of knowledge. Avicenna then distinguishesthe two senses of potency Aristotle had described by using the very termsapplied in the description of reflective self-awareness:

But if one has turned away from [what one has learned], the faculty [ofunderstanding] recedes and the form becomes potential, albeit potential veryclose to actuality (quwwatan qarībatan jiddan min al-fi‘l). The initial learning[of something] is like treating the eye which, when it has become a healthyeye, can turn according to its wish towards the thing from which it receivessome form. And when it has turned away from that thing, the thing becomespotential close to actuality.57

It is hardly a coincidence that Avicenna uses this concept to describe thecapacity of self-reflection. The reflective capacity in us is naturally conceivedas analogous to a person’s capacity to consider the knowledge she hasacquired but is not presently preoccupied with. As a potentiality close toactuality, it is something one can actualize at will, and in this sense reflectiveself-awareness, or awareness of awareness, is a constantly open possibility forany primitively self-aware subject. At any moment, we can turn our atten-tion to the primitively self-aware act, thought or perception that we arepresently engaged in.

This much is a natural consequence of the received theory of intellection.As Avicenna reiterates time and again, anything that is immaterial in itself isreadily available for an intellect to understand. The only property that can

56 Ar. De an. II.5, 417a22–417b20. 57 Shifā’: Fī al-nafs V.6, 247 Rahman.


prevent a thing from being understood is materiality.58 Yet the human self isonly one possible object of intellection among many others, even to itself,and for this reason its reflective turn towards itself is merely intermittent.Avicenna also specifies that reflexivity in this sense is exclusive to entitiesendowed with intellection; only that which ‘understands something’ iscapable of taking its own understanding as an object of a further act ofunderstanding. Again, this is due to the immateriality of all subjects ofintellection, for only an immaterial entity is capable of a genuine andwholesome relation to herself.However, although Avicenna refrains from mentioning another neces-

sary condition for reflection in this particular pointer, since he mentions itearlier on in the Ishārāt and states it slightly later on in the relevant section ofthe Ta‘līqāt, it seems warranted to incorporate it into his full account ofreflective self-awareness. This condition is of course primitive self-awarenessas the reflective subject’s constant familiarity with herself, as spelled out inthe argument against the reflection-model of self-awareness. Notice that theargument was designed not to deny our capacity of reflection, but to arguethat it has to be founded on a more primitive type of self-awareness thatallows us to recognize the act reflected upon as belonging to ourselves.Thus, self-reflection is possible only if the self is already primitively given inthe object of reflection to which one turns, that is, if it provides anindividuating reference for the concepts by means of which we reflectupon ourselves, as outlined above.But how is the subject given to herself in reflection? Two preconditions

our answer must meet bear mentioning here. First, the subject must begiven as primitively individual in order to be identified with the individualsubject that performs the reflective act. However, at the same time it mustbe possible to subsume it under the universal attributes that are employed inthe reflective act. This is because, in the above passages, Avicenna hasexplicitly described reflection as an act of intellectual consideration, andacts of intellection simply are, by definition, acts constituted by universalcontent. How, then, to meet two such seemingly contradictorypreconditions?Deborah Black has recently made the interesting suggestion that the

individual meanings encountered in Shifā’: al-Madkhal I.12 might play arole in Avicenna’s theory of self-awareness. Although Black’s suggestion is

58 This is stated in the remainder of the very ishāra just quoted (Ishārāt, namat. 3, 132); cf. Shifā’:al-Ilāhīyāt VIII.6.6–7, 284–285.


tentative, I believe it is based on a perspicacious and important insight.59 Aswe saw above,60Avicenna introduces the concept of individual meaning in acontext that deals with the question of our capacity to know individualentities and to refer to them bymeans of universal concepts. Little harm willbe done by revisiting the central lines:

It is impossible for the intelligible properties, however many they are, to beconjoined to the species without an ultimate reference (ishāra) to an individ-uated meaning, so that the individual thereby subsists in the intellect. For ifyou said ‘Zayd is the tall one, the writer, the handsome one’, and so forth foras many descriptions as you wish, Zayd’s individuality would not be assignedfor you in the intellect. Rather, the meaning, which is put together from thecollection of all that, can belong to more than one, but it is made concrete(yu‘ayyinuhu) by existence and by reference (al-ishāra) to an individual mean-ing, just as you say that he is the son of so-and-so, exists at such-and-such atime, is tall and is a philosopher, and it then happens to be the case that noone shares these attributes with him and also that you had a previousacquaintance with the case by means of the kind of apprehension which pointstowards (yushāru) [the case] on the basis of sense perception, pointing (yushāru)to the very person and the very moment.61

The intellect can grasp an individual under a universal description only if areference to an individuated or individual meaning is provided for it. Thisreference is not described by means of any mental content – unsurprisingly,since that would only constitute further additions to the bundle of univer-sals that will fail to individuate itself, no matter how extensive it becomes.Rather, the reference amounts to an acquaintance with a concrete thing thatis made by our perceptual access to the world of enmattered individualentities. Thus, here the term ‘meaning’ is used not in the sense of mentalcontent that is familiar from Avicenna’s psychological texts, but rather todenote the referent of such content in a sense that Avicenna systematicallyemploys in logical semantics.62 In this sense, an individual meaning is a

59 See Black 2012, sections 1–3. Although I confess that the present account is in many ways indebted toBlack’s initial insight, my interpretation of the relevance of the individual meaning to Avicenna’stheory of self-awareness differs from hers in two important respects. First, I do not think Avicennaemployed them in his concept of primitive self-awareness; instead, I claim that they first come intoplay on the level of explicit self-reflection. Secondly, I do not believe they should be understood alongthe lines of Scotian haecceitates, that is, as singular properties or concepts of singulars. Rather, as Iattempt to argue here, the concept of individual meaning denotes a special type of reference, not aspecial type of content.

60 See Chapter 3.1, 48–50. 61 Shifā’: al-Madkhal I.12, 70; emphases added.62 Shifā’: al-‘ibāra I.1, 3: ‘That which is in the soul signifies things, and they are those which are called

meanings (ma‘ānī), that is, intentions (maqās.id) of the soul, just as the traces [that is, content in thesoul] are also meanings in relation to [linguistic] expressions.’ Cf. I.1, 5; and for discussion, see Black2010 and Kaukua 2014b.


concrete individual referent to which we attribute universal content butwhich does not thereby cease to be given to us as a perceived individual.So far so good, but what does our access to enmattered individuals have

to do with self-awareness? If anything, the previous chapters should haveconvinced us that Avicenna is sternly against all attempts to explain our self-awareness by recourse to features of sense perception. My suggestion is,however, that primitive self-awareness provides us with a type of individu-ating reference that is functionally analogous with, but metaphysically andpsychologically different from, the example of the concrete material indi-vidual perceived by way of the senses. The idea is, very simply, that the I ofprimitive self-awareness is the individual meaning, or the point of reference,for the universals employed in an act of reflection. The I as the object of asubject’s reflective intellectual apprehension of herself is individual onlybecause it belongs to the subject’s primitively individual first-personalperspective. As Avicenna’s argument against reflection-based models ofself-awareness put it, a subject can recognize herself in reflection only ifshe already is familiar with herself. Just like the unique spatiotemporal co-ordinates primitively individuate a materially existing substance and enableit, by means of sense perception, to provide an individuating reference foran intellect’s bundle of universals, the subject’s self-awareness primitivelyindividuates her as an immaterially existing instantiation of the humanspecies and can therefore provide an individual reference for her reflectiveact and its universal content.Another way of characterizing this is to say that, instead of a special kind

of intellectual object, primitive self-awareness provides the conceptsemployed in reflection with an indexical referent. Consider, again, thecase described in Shifā’: al-Madkhal I.12. My personal awareness of anindividual, such as my friend Zayd, as an attributee of certain universalproperties that he shares with other individuals, requires that I have anindexical access to the materially existing Zayd. On the most general level, Imust be able to point, by means of perception, to a ‘this’ to which I can thenattribute the bundle of properties I have in mind, whereby my under-standing of the bundle comes to concern an individual human being. Theindividual meaning only indicates the sort of indexical reference mademanifest by utterances such as ‘this here’. The self is available to itself asan individual in an analogous fashion, that is, as something to which onecan point by means of the first-personal indexical ‘I’. And just as one canattribute all sorts of properties to a this, the I can be subject to any variety ofcharacterizations which become individual, or have an individual meaning,through this attribution to a primitively individual subject. I can reflect


upon myself as an individual fan of the local football team just as I canrecognize the individual cheering beside me as another individual fan of thevery same team. What is crucial is that in the case both of myself and of myfellow, the property of supporting the local team would remain universal,and indeed shareable by many, were it not for the reference to a primitivelyindividual this or I.

It must be emphasized that, according to Avicenna, I do not needexplicitly to utter or think the indexical ‘I’ in order to be primitivelyaware of myself, that is, in order to exist as I – as he has repeatedly asserted,primitive self-awareness is both prior to and independent of reflection. Yetat the same time, Avicenna holds that I can utter or think the first-personindexical, and thereby reflectively focus my attention on myself, at will.Moreover, our applications of the first-personal indexical in reference toourselves are all but self-evident, neither open to doubt nor in need oflearning. This is forcefully suggested by the way in which Avicenna speaksof our intuitive recognition of ourselves as subjects and agents that providethe unifying point of reference for a variety of states and acts. Moreover, theargument from personal identity asserts unequivocally Avicenna’s beliefthat each of us is intuitively and indubitably aware of a stable I as thesubstantial core enduring throughout our lives. This indubitability is anatural concomitant of the epistemically primitive nature of self-awareness:just as I cannot not be aware in a first-personal perspective, I cannot fail torecognize that perspective as my own in reflection.

This reconstruction of Avicenna’s concept of reflection is coherent withhis way of dealing with two related questions, namely of whether ourprimitive self-awareness is reduplicated in the act of reflection, and ofwhether the inherent potentiality of reflection in every primitively self-aware state is liable to give rise to an actually infinite series of reflective acts.The first question is addressed in the context of a discussion of thedistinction between external existence, or existence in concrete (fīal-a‘yān), and mental existence, or existence in the mind (fī al-dhīhn).Having briefly presented the distinction in the paradigmatic case of exter-nally existing individual substances, which when known come to exist in themind as accidental attributes of the knowing subject, Avicenna turns to themore peculiar case of our reflective awareness of ourselves. Here the dis-tinction between external and mental existence is complicated by the factthat, in both modes of existence, the human subject is immaterial and cantherefore be reasonably described as mental. Nevertheless, Avicenna holdsthat it is still valid to distinguish the subject’s existence as a substance fromher existing as an attribute of a knowing substance. This is intimately related


to our mode of existence as immaterial substances, that is, our self-awareness:

A human being may be inattentive to his self-awareness and then remindedof it, but he is not aware of himself twice. As regards awareness of awareness, itmay be by acquisition, not by nature.63

In other words, when my primitively self-aware state becomes an object ofreflection, it is no longer an act of existence proper to me as a substance anddue to me by nature, but a piece of acquired cognitive content and hence anaccidental attribute of me. Because of this change, the acquired awareness ofmy awareness does not result in redoubled self-awareness; rather, I remainaware of a single self in exactly the same sense as before, only now I realizethat I was aware of it all along.I do not think Avicenna means that the prior state, in which I was

primitively self-aware without paying attention to my awareness, is pre-served intact in the posterior act of reflective attention towards it. Rather,his point is that for each human being there is always one and only oneprimitively self-aware state. Now, an act of reflection involves minimallytwo states: a prior state which functions as the object of reflection and apresent state which is the act of reflecting upon the prior state. If all of theself’s acts are self-aware, then surely the present state of reflection should besuch. But if one is ‘not aware of oneself twice’, then the state one is reflectingupon can no longer be self-aware when it has been made an object ofreflection. In becoming an object of reflection, the prior state is transformedfrom a primitively self-aware state to mental content representing such astate. What is then immediately experienced is the act of reflection; if Ireflect on my writing, I am no longer simply writing, but reflecting on mywriting. Provided I am sufficiently capable of cognitive multitasking, thewriting may go on during my reflection, but this does not change the factthat my initial immersion in the act of writing is lost.In this respect it is not correct to say that reflective self-awareness is

fundamentally different from primitive self-awareness, because reflection isprimitively self-aware as well. It of course has a particular type of object, anact or state indexically attributed to oneself, that sets it apart from mostother primitively self-aware states, and this difference entails an epistemicascent to a higher-order level of consideration in comparison to the statereflected upon, but it does not entail either ascent or descent with regard to

63 Ta‘līqāt 147; emphasis added. Cf.Mubāh. athāt 422, 221 Badawī (VI.435–437, 158 Bīdārfar); 425–426,222–223 Badawī (VI.444–446, 160–161; and VI.892, 318 Bīdārfar). For a brief discussion, see Black2008, 79.


primitive self-awareness as such. Any reflecting subject is primitively, butnot reflectively, aware of herself in her reflective state, and just like any otherprimitively self-aware state, her reflecting state can in turn become theobject of a reflective act of a yet higher order. True, a subject of reflectionrecognizes herself in the object, when she attributes the state reflected uponto herself as her own past state. This recognition, however, is not immediatebut due to a relation of identity (huwīya) which the subject of reflectiongrasps intuitively as prevailing between herself and the object of her reflec-tion.64 This is neatly in line with Avicenna’s explicit denial of immediateidentity between the subject and the object of human intellection in Shifā’:Fī al-nafsV.6.65 If the subject of reflection were primitively self-aware of thesubject of the state reflected upon, this denial of identity would be violated.This need not rule out the self-evidentiality of the reflective identificationthrough recognition, something Avicenna explicitly states, but it does entailthat the recognition is not immediate. Rather, the relation of identity is anintelligible constituent of the act of reflection, albeit one that is not open tosincere doubt.

The second concern, that is, whether the inherent reflexibility of allprimitively self-aware states will result in the possibility of an infinitelyascending series of higher-order reflective states, is addressed in Shifā’: al-Ilāhīyāt V.2:

Because it is in the soul’s power to understand, and to understand that it hasunderstood, and to understand that it has understood that it has understood,and to construct relations upon relations, making different relational statesfor a single thing potentially ad infinitum, it is necessary that there be no endto these intellectual forms that are arranged upon each other, and as a resultthey will go on infinitely, but in potency, not in actuality.66

The text is unambiguous: there is no preordained limit to the amount ofreflective steps that we can possibly take in relation to our mental states andacts. This is a natural consequence of the claim that each self-aware state,that is, each moment in human existence, contains the potency close toactuality for the subject to turn reflectively towards herself. But this infinityof ascending steps is merely potential; it is not possible ever actually toascend an infinite number of times. Most often this is probably due to

64 For this particular use of the term huwīya, see Ta‘līqāt, 147–148. I realize that this is an anomalousreading of the term, but in this particular context Avicenna contrasts it with ‘otherness’ (ghayrīya),which strongly suggests that the relation of identity is what he has in mind. For the more commonuses of huwīya in Avicenna, see Goichon 1938, 411–413.

65 Shifā’: Fī al-nafs V.6, 239–240 Rahman. 66 Shifā’: al-Ilāhīyāt V.2.8, 160.


accidental distractions that seize one’s attention, but there is also a moreprincipled reason for the limitation of the number of actual reflective states.Because each reflective ascent is a temporal occurrence, because each humanbeing capable of reflection has a temporal beginning to her existence, andbecause each mental state is only primitively self-aware in itself, the actualexistence of every human being at any given moment will by necessity haltat a finite degree in the scale of higher-order states. This idea becomes clearfrom Avicenna’s comparison of the potential series of reflective steps withour knowledge of the entailments of the piece of knowledge we are actuallyconsidering.67 Suppose I know the quadratic formula, for instance. Byknowing the formula I know the solution to every single quadratic equationeven though I do not – indeed cannot – thereby consider them all inactuality. In the same way, the possibility of reflective self-awareness isconcomitant to every single self-aware act of mine only in the sense thatas an immaterial entity I can always turn to consider that I am the oneacting. The actualizations of these possibilities must take place successively.This exposition of Avicenna’s concept of reflection gives us a solid basis

for investigating the questions and problems concerning self-awareness thatare raised in theMubāh. athāt, a collation of Avicenna’s correspondence withhis critically gifted students. This material easily reads like a series ofhesitant, often ad hoc reactions to seemingly fundamental problems;indeed, with an exclusive focus on the Mubāh. athāt, one would quite likelyend up with a rather dismal assessment of Avicenna’s theory of self-awareness.68

A comprehensive analysis of the Mubāh. athāt material shows that thecritical points largely revolve around the difficulties, already familiar to us,that Avicenna faced when attempting to locate the phenomenon of prim-itive self-awareness in the scheme of the capacities described in his cognitive

67 See Shifā’: al-Ilāhīyāt V.2.8, 160–161. Cf. Shifā’: Fī al-nafs V.6, 241–243 Rahman.68 This is exemplified by Pines 1954, the first serious scholarly attempt at a comprehensive reconstruc-

tion of Avicenna’s concept of self-awareness. This pioneering study was largely motivated by ‘Abd al-Rahmān Badawī’s 1947 publication of his edition of the Mubāh. athāt. Relying extensively on thecritical remarks contained therein, Pines arrives at the suggestion that, despite the wealth of hisinsights, Avicenna’s concept of self-awareness was aporetic, incoherent, and inferior to those of suchancient predecessors as Aristotle and Plotinus. Allegedly, this is because Avicenna modelled hisconcept of self-awareness on the received account of intellection, which landed him with a load ofproblems when dealing with self-awareness as it appears on the lower levels of cognition. (Pines 1954,36, 39, 43.) Although my interpretation is diametrically opposed to Pines’, it must be admitted in hisdefence that he lacked access to the Ta‘līqāt, another collection of Avicennian Nachlass, which wasfirst edited by Badawī in 1973. Although plenty of material on self-awareness can be derived from theShifā’ (as recognized in, for instance, Pines 1954, 25, 29–30, 34) and the Ishārāt, the relevant passages intheTa‘līqāt are quite unique in their explicit emphasis on self-awareness and in the way in which theybring the scattered arguments together to a thematically defined discussion.


psychology. Perhaps the most pressing, and lucid, case in point is thefollowing piece of correspondence between Bahmanyār and Avicenna:

[Avicenna] was asked: by means of which faculty are we aware of ourparticular selves (bi ayyi quwwatin nash‘uru bi dhawātinā al-juz’īya)? Forthe soul’s apprehension concerns meanings, either by means of the intellec-tual faculty, but particular self-awareness is not understood (al-shu‘ūru bi al-dhāti al-juz’īyu laysa huwa yu‘aqalu), or by means of the estimative faculty,but the estimative faculty apprehends meanings connected to what is imag-ined, and it has been shown that I am aware of myself (annī ash‘ura bi dhātī)even if I am not aware of my organs and do not imagine my body.

So he answered: it is clear that the universal meaning is not apprehendedby means of a body, and it is clear that the individual meaning,69 which isindividuated by hylic accidents – a kind of defined determination (al-qadr)and defined position – is not apprehended by anything other than a body;but it is not clear that the particular is not at all apprehended by anythingother than a body or that the particular does not figure in a judgment of theuniversal; rather when the particular is not individuated by a determinationand a position and what embroiders them, nothing prevents that one is awareof that particular,70 and it is not clear that this is impossible in a position.There is nothing to object to [the fact] that the cause of this individual is hyleand something hylic in some manner, when the consequent individuatingconfiguration itself is not hylic but rather one of the configurations thatindividuate what is not through a body. On the contrary, the intellect or theunderstanding soul does not apprehend particulars that are individuated bydetermined hylic configurations. As regards to what is free of that, it may beapprehended, and this is apprehended both when it is peeled from individ-uating things and [when] the individuating things, taken as universals, arerelated to it. Separate things are either individuals of a species which aredistinguished by properties and the essences of which (dhawātuhā)71

are apprehended as such, or singulars (afrād) the species of which is notdivided into individuals (bi mukhas.s.as.āt), but the species is in one essence (fīdhātin wāh. idatin) which does not need to be distinguished except byspecificity, and so also their essences (dhawātuhā) are apprehended by theirspecificity. Then here it is to be contemplated whether the first sort areapprehended by their individuality.72

69 Because of the way in which ‘individual meaning’ (al-ma‘nā al-shakhs.ī) is contrasted with ‘universalmeaning’, I believe that in this case the term denotes mental content, albeit one that owes itsindividuality to the sort of reference we have just discussed.

70 Some manuscripts, including the one Badawī based his edition on, add the specification ‘I presume[that this is] the intellect.’

71 The word dhāt is here used in the sense ‘the thing itself’ or ‘the very thing’, with a reference to anindividual entity. The same holds of the instance of dhāt in the next sentence.

72 Mubāh. athāt 371, 208 Badawī; V.278–281, 117–118 Bīdārfar; for discussion, see Pines 1954, 47–48.


Bahmanyār’s question is unambiguous: how can we explain self-awarenessin terms of Avicenna’s cognitive psychology, given that both of thealternatives, intellectual and estimative apprehension, are fundamentallyproblematic, albeit for different reasons. Estimation necessarily involvesperceptual mental content and thereby the body, whereas self-awarenesshas been shown to be independent of both by means of arguments we,like Bahmanyār, have already discussed at length. Intellection in turnconcerns universals, but the self of which each of us is aware is a uniqueindividual.At this stage of investigation, we should expect Avicenna’s answer to be

ready and clear: self-awareness is due to neither an estimative nor anintellectual act, but rather constitutes a category of its own which, as thevery existence of the cognitive subject, is foundational to and independentof both types of cognition. In this light, the response section of the passageis especially problematic, for although the question is posed in the veryterms by means of which Avicenna usually describes primitive self-awareness, his answer unequivocally sets out to attempt a solution interms of intellection. Avicenna seems to think that if only we can makeconceptual room for the intellection of particulars, the question will solveitself. His attempt is thus very close to the account of the individuation ofimmaterial human substances found in Shifā’: Fī al-nafs V.3: although theself I am aware of is individuated by a bundle of properties that owe theirgenesis to matter, the properties as such and the self they serve toindividuate are immaterial and therefore readily intelligible. It is painfullyevident that none of the worries expressed elsewhere are so much as hintedat in Avicenna’s answer. The passage can thus be read as strong counter-evidence to our interpretation of self-awareness as primitive or irreducible:as far as texts like this are concerned, when all is said and done, Avicennaseems to have conceived of self-awareness as a quite commonplace act ofintellection.However, if we read his response closely, we see that Avicenna does not

explicitly claim that self-awareness should be understood as an act ofintellection with a certain individual as an object. All he does is entertainthe possibility. Since the resulting account is so obviously incoherent withAvicenna’s explicit arguments elsewhere, for instance for the constancy ofself-awareness (is not the explicit understanding of oneself as a bundle ofproperties a rather rare occasion?) and against the reflection-model of self-awareness (how will one recognize the apprehended individual as oneself?),we should be wary of adopting the passage as expressive of Avicenna’s


considered view.73 Indeed, the very last sentence of his answer clearly signalsthat Avicenna hesitates to assert the sketched solution and terminatesinstead in a problem that is postponed for further consideration. If theattempted solution is close to Shifā’: Fī al-nafs V.3, the problem is an ellipticexpression of the contrary insight we found from Shifā’: al-Madkhal I.12: inthe absence of a primitive factor of individuation analogous to the spatio-temporal co-ordinates afforded by matter, do we have any reason to assumethat the immaterial bundle of properties I understand as myself is anythingbut a bundle of universals? Thus, instead of a straightforward account ofself-awareness as due to an act of intellection, we seem to be left with theaporia characterized in the previous chapter, albeit with the difference thatself-awareness is now ruled out as a way out and forwards.

This hesitation is a recurrent feature in theMubāh. athāt questions relatedto self-awareness.74 It may be perceived as a sign of the unease Avicenna feltin the face of the problems an overtly intellectualist account of self-awareness is bound to run into when dealing with corporeal acts andperceptual cognition, that is, with acts and mental states the embodiedagents and subjects of which seem quite indubitably to be aware of them-selves.75 But although the symptoms are undeniable, different diagnoses ofthe cause remain possible. First of all, it is not at all clear that the unease wasdue to Avicenna’s insistent subscription to the view that self-awareness is theresult of an intellectual act, and not rather because his new conception ofprimitive self-awareness lacked a ready-made niche in the received cognitivepsychological framework. Secondly, quite regardless of the fact thatAvicenna subscribes to a strongly dualist conception of human being, and

73 The same vagueness shrouds the immediately following question in Mubāh. athāt 372, 208–209Badawī; V.282–285, 118–119 Bīdārfar (cf. Pines 1954, 50–51). As for Mubāh. athāt 426, 222 Badawī(VI.892, 318 Bīdārfar), another similar passage (cf. Pines 1954, 52–53), it deals not with primitive self-awareness but with our capacity of understanding the human essence (dhāt) of which each of us is aninstantiation. The point is that my understanding my own humanity (my own dhāt) differs not incontent, but only in reference, from my understanding of the humanity of another human being.Humanity is the same in both cases, and in reflection I attribute it to myself just as I attribute it toanother individual in an act constituted by perception and intellection.

74 Its most pronounced expression is Mubāh. athāt 373, 209 Badawī (cf. V.288, 119 Bīdārfar), whereAvicenna entertains the possibility that self-awareness is not an act of intellection at all.

75 Pines 1954, 54–55, takes up the question of animal self-awareness (raised in Mubāh. athāt 375, 209Badawī (V.291–293, 120 Bīdārfar); and 421, 222 Badawī (VI.891, 317–318 Bīdārfar)) as a particularlytelling case in point. The question of animal self-awareness, brought up again by Rāzī in his criticalcommentary to the Ishārāt (Sharh. al-Ishārāt, 122), is complicated andmust be set aside for the present.Suffice it to say that I believe the question shows that the sort of self-awareness we have beendiscussing here is exclusive to immaterial, and hence intellectual, entities. For a more detailedtreatment of the question of animal self-awareness, see Kaukua and Kukkonen 2007 and López-Farjeat 2012.


thus one that places a decided emphasis on our intellectual capacities, it isnot clear that self-awareness as the mode of existence proper to the imma-terial aspect of us is incompatible with activity and cognition in or by meansof the body.It has been suggested, indeed quite plausibly in the light of such incon-

sistency, that Avicenna’s view of the cognitive class proper to self-awarenesswas subject to development throughout his career, and that he may neverhave managed to arrive at a definite solution.76 If this is the case, the naturalquestion to ask is how we should date texts like the present piece ofcorrespondence with its aporetic conclusion. Do they represent a passingphase in the course of Avicenna’s labour with self-awareness, or are theyrather expressive of a mature view? The details of the genesis and dating oftheMubāh. athāt texts are extremely complex, but, in the light of our currentknowledge, it seems warranted to hold that the set of questions to which thepresent one belongs was drafted by Bahmanyār and answered by Avicennabefore 1030 and so quite possibly before the excursions on self-awareness inthe Ishārāt and the Ta‘līqāt.77 The text may therefore represent a stage inwhich Avicenna had gained the insights that were to provide the basis of hisconcept of self-awareness (note the veiled reference to arguments like theflying man at the end of Bahmanyār’s question) but had not yet drawn theconclusions about its immediacy and consequent uniqueness as a mode ofcognition that are exemplified in the later works. In any case, regardless ofthe ultimate order of production of Avicenna’s works, our present knowl-edge does not warrant us to consider the present text as an expression ofAvicenna’s final stand.If we approach the Mubāh. athāt from this perspective, they bear witness

to a constant labour in finding a niche for the newly discovered phenom-enon of first-personality in the Peripatetic psychological framework, andportray a thinker in a sustained attempt to twist the received concept ofintellection as self-intellection to tally with such a narrow concept of self-awareness.78 But although Avicenna may not have been entirely successful,and certainly failed to persuade all his interlocutors, his tentative consid-eration of the possibility that self-awareness is not an act of intellection at all

76 Sebti 2000, 116–117. 77 I infer this on the basis of Reisman 2002, 212, 221–224, 227, 231.78 Cf. the largely concurring assessment in Sebti 2000, 113–117. It should also be noted that certain post-

Avicennian developments, such as Suhrawardī’s concept of knowledge as presence and Mullā S· adrā’sidentification of knowledge with existence (see Chapters 6.1 and 8.1 below, respectively), can be readas further steps in the pursuit of a concept of awareness that is no longer reducible to the ancientdichotomy between perceptual knowledge of particular objects and intellectual knowledge ofuniversals.


may have been positively loaded with sustained consideration.79 While it istrue that Avicenna’s concept of self-awareness is based on the view that anindividual human dhāt is an immaterial substance and therefore an intel-lectual entity, the texts we have discussed in extenso strongly suggest thatthat substance’s awareness of itself is not an ordinary act of intellection. Ifour reconstruction of Avicennian self-awareness as first-personality is any-where close to the mark, it is rather an inherent feature of all humanexperience from the highest echelons of intellection down to the loweststrata of sense-perception. All of these drastically different experiences, as somany accidental determinations of an individual entity, are always given asmine to the respective individual, or, to put it another way, the individualwill always undergo or live through them as an I. In this limited respect,there is no difference whatsoever between her various states: as far as myprimitive self-awareness is concerned, it is the same regardless of whether Iam engaged in physical labour, a rapturous erotic act or the contemplationof the law of non-contradiction. Explicit reflective intellection of oneself isan altogether different phenomenon, but also one that Avicenna explicitlydistinguishes from the more primitive type of self-awareness.80 As we’velearned, reflective self-awareness is an act of intellection, which presupposesprior familiarity with oneself as its condition of possibility. Hence, primitiveself-awareness as this familiarity must be distinguished from reflection, justas Avicenna repeatedly does.

In the end, it has to be admitted that, despite its coherence, our recon-struction of Avicenna’s concept of self-awareness remains a rational recon-struction, pieced together from somewhat (though not entirely) fragmentedpieces of discussion. It remains a fact that Avicenna never explicitly defineshis concept of self-awareness, nor does he devote a single chapter of hispsychological works to the phenomenon. At the same time, the reconstruc-tion is hardly marred by a shortage of material, and as I hope to have shown,the material is not only ample, but highly coherent as well. Finally, many ofthe arguments that are presented separatim in Avicenna’s other works werein fact brought together in the Ta‘līqāt.

79 Here I refer especially to passages likeMubāh. athāt 373, 209 Badawī (cf. V.288, 119 Bīdārfar): ‘It maybe that “intellection” [in the sense of that] which grasps intelligibles is not applicable to the purity ofcomplete self-awareness (mujarrada al-shu‘ūri al-mujmali bi al-dhāt) but comes after that. That isworth thinking about.’

80 Pace Pines 1954, 46. The distinction seems to have been there in some of the texts circulated under therubric of Mubāh. athāt (cf. VI.549, 185 Bīdārfar), but apparently not in the recension represented bythe manuscript Badawī used for his edition.


As we have seen, the Avicennian shu‘ūr bi al-dhāt alludes to the first-personal perspective inherent in all our experience. As such, it is the mode inwhich we exist as immaterial substances. The way in which we have seenAvicenna develop this insight has two consequences, which may be rela-tively unproblematic, even trivial, in their Avicennian setting, but whichbecome problems and points of critique for his followers whomwe will nextturn to discuss. For this reason, these consequences merit being brieflyspelled out.First of all, Avicennian self-awareness is static and allows no room for

development. Since self-awareness is our birthright, it cannot be relied on tomake sense of the process of acquisition of our second perfection, thecognitive goal proper to the human species. Our self-awareness notwith-standing, we will still be imbued with the task of developing our charactersby consistently doing the right things andmaking the right choices as well aslabouring in the pursuit of knowledge. But regardless of the eventual degreeof our success, we will taste its fruits as subjects aware of ourselves in exactlythe same sense as when we first came to be. I will be no more and no less an Iwhen I have reached the degree of a virtuous sage than I was as a peckish lad.I may be able to pay attention to my being an I, to understand why that isthe case, and to fully grasp and attempt to realize the sort of duties towardsmyself it entails, but the fact itself of being an I is not changed at any stage ofthis development. Intimately related to this point is a second one, namelythat the self indicated by the phenomenon of primitive self-awareness is asimilarly static entity, that is, an Aristotelian substance. As the argumentfrom personal identity in the sixth book of the Mubāh. athāt makes partic-ularly explicit, the human self remains immune to change throughout thecourse of its existence, despite the fluctuation of its attributes. As substan-ces, Avicennian selves do come to be, but they are not subject to any kind ofalteration, change or development in the proper senses of these terms.In the subsequent centuries, both of these features come to be fiercely

contested. Intriguingly enough, however, these critical moves are made in aframework that is entirely based on Avicenna’s ways of describing self-awareness and of arguing for its primitivity. I believe that a close study ofthe reception of this part of the Avicennian heritage will further corroboratethe reconstruction proposed here.


chapter 5

Self-awareness without substance: fromAbū al-Barakāt al-Baghdādī to Suhrawardī

In many ways, the twelfth century CE is decisive for the solidification ofthe Islamic philosophical scene, duly characterized as post-Avicennian.Instigated by Abū H· āmid al-Ghazālī’s (d. 1111 CE) critical appropriationand Abū al-Barakāt al-Baghdādī’s (d. 1164/5) radical scrutiny of Avicenna’sphilosophy, two closely related, if fundamentally different ways of applyingthis heritage emerge towards the turn of the century, both of which wouldprove formative for the subsequent centuries. These are Fakhr al-Dīn al-Rāzī’s systematic theology and Shihāb al-Dīn Yah. yā al-Suhrawardī’s ishrāqī,or illuminationist, school of philosophy.1

Although many authors of the period address Avicenna’s remarks on self-awareness, it is in Suhrawardī’s thought that their developmental potentialis most fully accomplished. Information about Suhrawardī’s youth is scarce,but we do know that he was born in 1154, possibly in the village of Suhrawardin northwestern Iran, and acquired an education in philosophy and theologyfrom a number of notable teachers of his time. After a period of itinerantexistence in Syria and Anatolia, we find him in Aleppo from 1183 onwards,gathering a circle of notable students about him. One of these was theyoung prince al-Malik al-Z· āhir, recently appointed as the ruler of the cityby his father, the great Saladin. This seems to have aroused the ire andjealousy of Aleppo jurists who complained to Saladin. The conflict withcrusaders growing increasingly tense, the leader ordered his son to executeSuhrawardī, which the latter eventually submitted to in 1191, though notwithout reluctance.2

Despite his early demise, Suhrawardī left a considerable corpus, the mostimportant texts of which are often divided into three main classes: (1)mystical allegories, (2) Peripatetic philosophical treatises, most importantly

1 For a general account of the early reception of Avicenna, see Eichner 2009, 3–95; and for more specificstudies, cf. Wisnovsky 2004a; Shihadeh 2005; Eichner 2011; and Treiger 2012.

2 Walbridge and Ziai 1999, xv–xvii; Walbridge 2005, 201.

104

the Talwīh.āt and the Mashāri‘ wa al-mut.ārah.āt, and (3) ishrāqī treatises,in particular the H· ikma al-ishrāq.3 The question of the exact relations ofimportance between these different types of texts is a matter of debate.Some scholars, most notably Henry Corbin, perceive the mystical narrativesas the culmination of Suhrawardī’s thought and read the ishrāqī treatises astheir systematic counterpart, while assigning the so-called Peripatetic worksa status of secondary importance.4 Other scholars have argued for a muchstronger connection between the different types of philosophical treatises.According to Hossein Ziai, ‘the impetus behind the composition ofeach of these works was nothing other than the systematic presentation ofthe philosophy of illumination. This means that when Suhrawardī statesthat the Intimations [al-Talwīh.āt], for example, is written according tothe “Peripatetic method”, it is not an independent work written solely asan exercise in Peripatetic philosophy, nor does it represent a Peripatetic“period” in Suhrawardī’s life and writings. Rather, it points to the fact thatcertain parts or dimensions of the philosophy of illumination are acceptedPeripatetic teachings.’5 By the same token, it has been argued that instead ofa culmination of the Suhrawardian corpus, the mystical narratives amountto little more than introductory treatises designed for students who are notyet able to savour the technical and conceptually rigorous argumentationproper to the philosophical works.6

I see no reason to hide my sympathy with the latter line of interpretation.Had the Peripatetic works been merely propaedeutic in character, whyshould Suhrawardī have spent so much precious paper and ink on them –particularly in comparison to the decidedly less imposing volume of theH· ikma al-ishrāq? Indeed, why did he choose to write two such introductoryworks when the Talwīh.āt, for instance, would alone seem to have beenperfectly apposite to the task? Furthermore, recent work which situatesSuhrawardī in the context of the twelfth-century reception of Avicenna hasshed some direly needed light on the question of how the discussion in thePeripatetic treatises paves the way for Suhrawardī’s transition to the overtlyishrāqī philosophy of the H· ikma al-ishrāq.7 Since the Talwīh.āt and theMashāri‘ wa al-mut.ārah.āt are comparably closer to the terminologicalframework of Avicennian philosophy, they provide us with invaluableevidence for assessing why and to what extent the H· ikma al-ishrāq departs

3 Walbridge and Ziai 1999, xviii. For a voice of concern, cf. Marcotte 2012.4 See, for instance, Corbin 1971, vol. II, 187–200. This view is manifested in Corbin’s decision to omitboth the logical and the physical sections of the Talwīh.āt and theMashāri‘ wa al-mut.ārah.āt from hisedition of these texts.

5 Ziai 1990, 10–11. 6 Walbridge 2000, 109–112. 7 A prime case in point is Eichner 2011.

Self-awareness without substance 105

from that tradition. This interpretative stance will, I hope, be amplysubstantiated also by the following two chapters.

5.1 Avicennian material in Suhrawardī

From both the historical and the systematic point of view, the mostinteresting aspect of Suhrawardī’s treatment of self-awareness is his novelapplication of the phenomenon in introducing his alternatives to thePeripatetic concepts of knowledge and existence. However, to bring thisaspect of his thought into clear focus, we must delineate and chart theground he shares with Avicenna, the epitome of that tradition. Fortunatelyfor this concern, Suhrawardī’s arguments for the constancy and immediacyof self-awareness are for the most part familiar from the previous chapter,which strongly suggests that Suhrawardī is indeed referring to exactly thesame phenomenon as Avicenna. Furthermore, on the occasions when heventures beyond his predecessor, he articulates even more explicitly theAvicennian thesis that self-awareness amounts to bare first-personality as anirreducible and necessary constituent of human experience.

When one sets out to read the section on the rational soul in thepsychology of the Talwīh. āt, one cannot fail to notice a close resemblanceto Avicenna’s method of procedure in al-Ishārāt wa al-tanbīhāt. Just like hispredecessor, Suhrawardī initiates the discussion of the properly humanaspect of the soul with the thought experiment of the flying man:

Is it not so that you are not absent from yourself (dhātika) in the two states ofyour sleep and wakefulness, your sobriety and intoxication? If you weresupposed [to be] created all at once to a perfection of your intellect, and yoursense were not preoccupied with anything from you or from another, and themembers [were] extended in the air so that they would not touch, you wouldignore everything but your thatness (annīyatika). Thus, you will not haveacquired either bodies or accidents when they were not included in your self(dhātika) which you understand without them [and] without the need of amedium, or a corporeal indication or notification, and so you know forcertain about your self (ma‘arifatuka li dhātika) that it is incorporeal.8

The two examples of sleep and intoxication give away Suhrawardī’simmediate source, and he follows Avicenna in all the essential aspects ofthe thought experiment. The conclusion is just as familiar: because self-awareness can be there in the absence of the awareness of one’s body or any

8 Talwīh.āt II.4.3, 155–156 Habibi; 337–338 Ziai and Alwishah. Cf. Avicenna, Ishārāt, namat. 3, 119, inChapter 4.2, 80.

106 Self-awareness without substance

other cognitive object, it cannot be based on or brought about by anythingelse. Two signs of departure from Avicenna are Suhrawardī’s tacit intro-duction of terminology related to his own concept of knowledge aspresence (‘you are not absent (lā taghību) from yourself’) and the increasein the argument’s demonstrative power (‘you know for certain (d

˙arūrīya)

about your self that it is incorporeal’),9 but since these aspects of thepassage are intimately related to Suhrawardī’s original application of self-awareness, which we will discuss below, we will have to set them aside forthe time being. In the remainder there is little that we have not alreadyencountered in Avicenna. If anything, Suhrawardī’s formulation of theflying man is even more elliptic than the already condensed version of theIshārāt, and he seems to direct no extended attention at the phenomenonthe argument builds upon; in fact, one could argue that Suhrawardī’sterminology is less indicative of first-personality than that of Avicenna.However, Suhrawardī dovetails the passage with a sceptical question thatwe cannot find in Avicenna and that gives away the clarity of his grasp ofthe argument’s basis.

Question: I have come to know myself (‘araftu dhātī) by means of this act,and so it is a medium.Answer: In your supposition you were divested of your act, and so there isno medium.

Question: This supposition is an act, and so it is a medium.Answer: It is an omission of an act, or rather it is an idea, and when youcome across it, you come across the one having the idea (dhā al-khat.ra)behind it and known before it, not by means of it, and that is your self(dhātuka).10

Suhrawardī has his interlocutor raise the question of whether the flyingman itself, as a thought experiment, first brings self-awareness about, and ifit does, whether this does not constitute a problem for the claim that it isimmediate and not acquired. Although Suhrawardī’s answer is rather con-densed, it belies a solid grasp of the Avicennian method of reminding.Although the flying man stages our awareness of ourselves in a mannerthat forces it into the focus of our attention, it thereby only enables therecognition of what must have already been there. Supposing oneself in thesituation of the flying man is a very peculiar operation of thought, a most

9 The latter aspect is evenmore pronounced in Ibn Kammūna’s thirteenth-century commentary on theTalwīh.āt (Muehlethaler 2009).

10 Talwīh.āt II.4.3, 156 Habībī; 340 Ziai and Alwishah.

Avicennian material in Suhrawardī 107

unlikely ‘idea’ (khat.ra), but its peculiarity lies precisely in the fact that it isdesigned to foreground the subject that has the idea (dhū al-khat.ra), whichcannot be supposed to be generated by the idea it has. If this interpretationof Suhrawardī’s dense little dialogue is correct, he had an accurate grasp ofthe basis of the flying man: the thought experiment is a means of payingattention to something that tends to elude our consideration but is con-stantly given to each of us in perfectly commonplace experience.

If the main purpose of the Talwīh.āt was to present the outlines of thePeripatetic system of thought, it can be asked whether Suhrawardī is merelyusing the Ishārāt as an authoritative source of the traditional doctrinewithout himself subscribing to its contents. Such concerns are underminedby the same section of the Ishārāt being mined for arguments in Suhrawardī’spositive exposition of his alternative system. Consider, for example, how theargument against reflection-based models of self-awareness is articulated inthe H· ikma al-ishrāq:

The thing that subsists by itself and apprehends itself (al-qā’ima bi dhātihial-mudrika li dhātihi) does not know itself by means of an image of itselfin itself (lā ya‘alamu dhātahu bi mithālin li dhātihi fī dhātihi) . . . if [that]were by means of an image that it does not know to be an image of itself(li nafsihi), it would not know itself (nafsahu); if it does know that it is animage of itself (nafsihi), it must have already known itself (qad ‘alimanafsahu) without the image. How could it not, when it is not conceivablethat the thing knows itself (ya‘alama nafsahu) by means of somethingadded to itself (‘alā nafsihi), for [that] would be an attribute of it? Thus,when it judges that every attribute added to itself (‘alā dhātihi), whetherknowledge or something else, belongs to itself (li dhātihi), it has alreadyknown itself (qad ‘alima dhātahu) prior to and apart from all the attributes,and so it has not known itself (lā yakūna qad ‘alima dhātahu) by means ofthe added attributes.11

The alien context notwithstanding,12 the argument is clearly familiar. Ifwe suppose that self-awareness is due to an awareness of a special type ofobject, here an image of oneself, we will come across the problem of how toexplain the recognition of the image as an image of oneself. Since there is nonon-arbitrary way of introducing the recognition in the case of some imagesbut not others, we have to suppose that the recognition is based on a morebasic familiarity with oneself. Suhrawardī sums this up in a general claim:

11 H· I II.1.5.115, 80 Walbridge and Ziai; 111 Corbin.12 In this section of the H· ikma al-ishrāq Suhrawardī is engaged in the characterization of the concept of

pure or incorporeal light by means of Avicennian arguments for the immediacy of self-awareness. Fora discussion of this material in its context, see Chapter 6.2, 146–154.


any image or attribute that I recognize as my own, as attributable orbelonging to me, has to be qualified as mine to begin with.We have seen Avicenna claim that this mineness or first-personality

inherent to our attributes is a constant feature of our experience. As aconsequence, it cannot be based on anything that we can be intermittentlyaware of, such as our bodies. This basic difference between the intermit-tency and constancy respective to the awareness proper to bodies on the onehand and to selves on the other is repeated by Suhrawardī immediately afterthe above argument against the reflection-model of self-awareness:

You are never absent from yourself and your apprehension of it (anta lātaghību ‘an dhātika wa ‘an idrākika lahā), and since the apprehensioncannot be by means of a form or an addition, you do not need in yourapprehension of yourself anything other than your self, which appears toitself, or is not absent from itself (fa lā tah. tāja fī idrākika li dhātika ilāghayri dhātika al-z· āhirati li nafsihā aw al-ghayri al-ghā’ibati ‘an nafsihā).Thus, your apprehension of it must be due to itself as such (fa yajibu anyakūna idrākuka lahā li nafsihā kamā hiya),13 and you are never absent fromyourself or any part of yourself (lā taghību qat.t.u ‘an dhātika wa juz’idhātika). What your self is absent from (mā taghību dhātuka ‘anhu), suchas organs like the heart, the liver and the brain . . . do not belong to that ofyou which apprehends, and so that of you which apprehends is not bymeans of an organ . . . since otherwise you would not be absent from itinsofar as you have enduring self-awareness that does not cease (kāna lakashu‘ūrun bi dhātika mustamarrun lā yazūlu).14

Again, the metaphysical context does not betray the familiarity of thisseries of statements, which reads readily as a paraphrase of the centralinsights behind the lengthy excursion on self-awareness from the Ta‘līqāt.Suhrawardī’s emphasis on the constancy and continuity of self-awarenesssuggests that he is aware of Avicenna’s striking and potentially counter-intuitive claim that we are always aware of ourselves, and that he knowinglyadopts the Avicennian concept of self-awareness. This is further corrobo-rated by a related argument, characterized as a reminder (dhikrun tanbīhī),in the psychological section of the Talwīh. āt.

Your skin is such that you have the power to change it while aware of thepersistence of your thatness (annīyatika), and the same holds of your fleshand your bone, and so they have no share in it. You have understood it while

13 I read idrākuka (instead of idrākuhā as preferred by both Corbin and Walbridge–Ziai) which issupported by some manuscripts and is not as awkward as the alternative reading. Insofar as ‘you’ issynonymous with ‘your self’, as is the case here, there is no difference in meaning.

14 H· I II.1.5.116, 80 Walbridge and Ziai; 112 Corbin.


ignoring the heart, the brain and the liver, which are known to you throughdissection; something has occurred to your mind about them once or twiceevery year, and you are not through what is distant from your thatness, andso what is not included in it in the intellect is not constitutive of it. Thus,[the] you behind the whole is [such] that it is a self (anta warā’a al-jamī‘iannahā dhātun) that exceeds the confusion of the ignorant about it; it is asubstance which we rightfully call ‘the rational soul’.15

Just as in the passage from the H· ikma al-ishrāq, Suhrawardī here makesthe distinction between the self and any of its cognitive objects by meansof the constancy and intermittency respective to them. If anything, thistime the connection to Avicenna is even more pronounced, for both theargument and the context in which it is embedded read naturally as aparaphrase of a passage from Shifā’: Fī al-nafs V.7.16 What is more, theargument is followed by one based on another aspect of difference betweenour awareness of our bodies and our selves, and with an equally Avicennianprovenance.

If nothing has dissolved from your body and the nourishing [faculty] hasproduced what it has produced so that the volume of your body has grownvery large up to what you are familiar with, or up to the position of theresting [faculty], and if dissolution is then asserted and you are aware of thepersisting heat in decomposition, corruption and other causes, then will younot be another self (dhāt) every year, or a self whose thatness is diminished(aw dhātun muntaqis.atu annīyatihā)? Or is it a self that does not dissolve?Thus, it is not embodied; I have reminded you! Be reminded that it is adivine fire, undescended, exalted above both impression [in] matter andbeing the same as the temperament.17

This argument relies on the interlocutor’s intuitive certainty of theendurance of her self through time despite the indubitable changes thattake place in her body. As we recall, Avicenna had brought forth the verysame intuition as a response to Bahmanyār’s dissatisfaction with the flyingman. And even though nothing in Suhrawardī’s formulation gives away theMubāh. athāt as his immediate source, all the ingredients of the presentargument were already laid on the table by other Avicennian texts. It isalso worth pointing out that Suhrawardī briefly resorts to illuminationistterminology in spelling out the dualist conclusion of the argument. Thissuggests, if somewhat obliquely, that he perceived a connection between

15 Talwīh.āt II.4.3, 156 Habībī; 342 Ziai and Alwishah.16 Avicenna, Shifā’: Fī al-nafs V.7, 255–256 Rahman.17 Talwīh. āt II.4.3, 157 Habībī; 342–343 Ziai and Alwishah; cf. H· I II.1.5.116, 80 Walbridge and Ziai;

112 Corbin.


this aspect of Peripatetic psychology and his own alternative system ofphilosophy.Let us finally consider another argument based on our awareness of our

enduring selves. This time the context is epistemological and the argumentoriginal to Suhrawardī, notwithstanding that it figures in an argument foran entirely Avicennian thesis and has an equally Avicennian basis in its useof self-awareness. Suhrawardī now relies on the phenomenon in a piece ofevidence brought forth to refute the identity or unity theory of cognition.He begins by arguing that if the identity theory is correct and the subject ofknowledge indeed becomes the known form, or inversely, if the knownform is assimilated into the subject, then there remains no sense in which wecan speak of cognition anymore; rather, all we have is the subject or the form(depending on which of them becomes the other) as it was before theidentity. On the other hand, if both remain, there is no identity.18 ThisAvicennian case19 is then supported by a corroborating piece of evidence:

Then the self-aware substance of you (al-jawharu al-shā‘iru bi dhātihi minka)is not such that it is renewed every moment, but it is a single thing whichendures before, with and after the form, and the form is something thatoccurs while it endures. Thus, you are you with the apprehension andwithout the apprehension, and so there is no meaning for unification.20

Self-awareness is here used to substantiate a commonplace Peripateticclaim: a necessary condition for understanding any kind of process is thatwe are aware of an unchanging substance in relation to which the processtakes place or to which the process somehow belongs. To take learning as anexample, I as the subject of learning must not be replaced when I come toknow what I did not know before, for were that not the case, whom couldwe say to have learned anything thereby? I must be I in some unchangingsense with and without any attribute of mine, such as the knowledge I havemanaged to acquire.This last argument therefore highlights the factor that is common to all

four types of arguments we have cursorily analysed, namely that they allhinge on the same phenomenon. It shows that the self and its awareness ofitself can be distinguished from its awareness of objects or attributes bywhich it is qualified, and, as became clear from Suhrawardī’s considerationof the intricacies of the flying man as well as from the contrast he drewbetween the endurance of the self and the fluctuation of the body, self-

18 Mashāri‘ III.7.1, 474. 19 See Avicenna, Shifā’: Fī al-nafs V.6, 239–240 Rahman.20 Mashāri‘ III.7.1, 474–475.


awareness, in an immutable sense, is attached as a necessary constituent toany awareness of objects. Thus, even without resorting to as laborious anargument as we did in the previous chapter, I dare propose that the require-ments these arguments amass on the concept of self-awareness that figuresin them forcefully point towards the view that Suhrawardī adopts theAvicennian concept of self-awareness in the sense reconstructed above. Inother words, for Suhrawardī just as for Avicenna, self-awareness amounts tobare first-personality, or the simple fact that all we are aware of is given to usin a first-personal perspective.

This is underlined by a conceptual distinction between the self-awaresubject and any object it is aware of to which Suhrawardī consistently sticksthroughout his philosophical works. Let us consider two examples, the firstfrom the already familiar section on the rational soul from the Talwīh.āt:

Every corpus ( jirm) and accident in it of your body (badanika) and others isreferred to from your perspective by its being an it (mushārun ilayhi minjihatika bi annahu huwa), and whatever is referred to from your perspectiveby its being an it is distinct from you by being different from both you as awhole21 and [any] part of you; thus, every corpus and accident in it is distinctlike that. Since you are distinct, you are not them as a whole, for not a part ofthem exists for you (li ‘adami juz’īyatihā laka); thus, your self (dhātuka) isincorporeal altogether and not merely in some respect.22

Unlike ourselves, each object we are aware of appears to us, or ‘from ourperspective’, as an ‘it’, something in front of our regard to which we canrefer ostensively by means of the third-person indexical pronoun. Since allbodies are given to us in this manner, Suhrawardī believes that the differentmode of givenness of ourselves allows us to conclude that our selves are notbodies. The argument is not entirely without parallel in Avicenna’s psycho-logical works; one is reminded especially of the sustained claim in Shifā’: Fīal-nafs V.7 that our intuitive use of the first-personal indexical is anindication of our incorporeality.23Moreover, we found a similar distinctionin the Risāla al-ad

˙h. awīya fī al-ma‘ād.24 However, Suhrawardī’s consistent

reliance on the distinction is without parallel in his predecessor. The noveltyof this move is particularly clear in the following passage from the H· ikmaal-ishrāq:

21 I read kullika with Habībī instead of the dhātika in Ziai and Alwishah, because it fits better with theimmediately following juz’ika, ‘part of you’. I take it as obvious that the reading in Ziai and Alwishahis perfectly compatible with my interpretation of the passage as a whole.

22 Talwīh. āt II.4.3, 156 Habībī; 341 Ziai and Alwishah.23 Avicenna, Shifā’: Fī al-nafs V.7, 256–257 Rahman; see Chapter 4.1, 66–71.24 Avicenna, al-Risāla al-ad

˙h. awīya IV, 127–128; see Chapter 4.1, 79.


The thing that subsists by itself and apprehends itself does not know itself bymeans of an image of itself in itself, for if its knowledge were by means of animage and if the image of its I-ness (anā’īya) is not [the I-ness], [the image]would be an it in relation to [the I-ness], but then what is apprehendedwould be the image, and as a consequence, apprehension of I-ness would beidentical to apprehension of what is an it (yakūna idrāku al-anā’īyati huwa bi‘aynihi idrāka mā huwa huwa), and apprehension of self would be identicalwith apprehension of another (yakūna idrāku dhātihā bi ‘aynihi idrākaghayrihā), which is absurd, unlike in the case of external [things], wherethe image and what is that for it are both an it (al-mithāla wa mā lahu dhālikakullāhumā huwa).25

Suhrawardī’s argument is designed to refute the claim that self-awarenessis due to the cognition of a specific object, here an image (mithāl) represent-ing a subject to itself. As such, the argument is clear enough. But the termsin which Suhrawardī stages it are of prime importance for our topic, for herewe find the object of cognition explicitly distinguished from its counterpart,the subject. Any object will be given as an it facing our attentive regard,whereas the subject will be given to itself as an I (anā). From this first-personindexical Suhrawardī derives the abstract noun ‘I-ness’ to single out themode of being proper to a cognitive subject. Having made this categoricaldistinction, he can rephrase Avicenna’s argument against reflection-basedmodels of self-awareness in more general terms: it is not possible to proceedfrom the awareness of any object or it, however specific we suppose it to beto the subject in question, to an awareness of the subject, an I or me. In thisrespect self-awareness is fundamentally different from awareness of objects,for as Suhrawardī’s last clause states, nothing prevents a representationalrelation between an image appearing to the mind and the thing it is animage of (‘what is that for it’, that is, what the image is an image of), becauseboth the image and its referent exist in the third person, each being an it.I would like to claim that Suhrawardī’s introduction of the abstract

noun derived from the first-person indexical anā in an argument that is inclear and considerable debt to Avicenna supports my reconstruction ofAvicenna’s concept of self-awareness in the previous chapter. The newterm signals that by the end of the twelfth century the discussion of self-awareness had become a firm enough fixture in the philosophical traditionto warrant the coinage of a novel highly technical term.26 Let us now shiftour focus to material which shows that this part of the Avicennian heritagewas still very much in formation.

25 H· I II.1.5.115, 80 Walbridge and Ziai; 111 Corbin; cf. Mashāri‘ III.7.1.208, 484.26 Cf. also the parallel passages in the Persian Partaw-nāma: III.28, 24–25, and V.43, 38–39.


5.2 Substanceless self-awareness

As we have seen, Avicenna considered self-awareness to be a potent pointertowards the human being’s existence and subsistence independent of herbody. In Avicenna’s metaphysics, this amounts to the view that the entitywhich functions as a soul in relation to the human body is an immaterialsubstance, for an entity that subsists by itself is by definition a substance.Avicenna does not dwell on the basis of his inference at any great length,presumably because he saw no need to reassess the foundational role ofsubstance in Peripatetic metaphysics and therefore considered the conclu-sion of the soul’s substantiality to be a natural outcome of its incorporeality.This is corroborated by his summary account of the different kinds ofsubstances in the metaphysical section of the Shifā’:

[E]very substance either is or is not a body, and if it is not a body, either it isa part of a body or it is not a part of a body but rather altogether separatefrom bodies . . . If it is separate, it is not a part of a body, and then it eitherhas a connection of some governance to bodies in terms of motion, andwe call it soul, or it is free from matters in every respect, and we call itintellect.27

Once that which functions as a soul in the human body is established toexist as a subject that is aware of itself independently of its body, itsclassification within this remarkably unequivocal metaphysical scheme is amatter of course.28

But regardless of how innocent the move from incorporeal self-awarenessto the self’s substantiality may have seemed to Avicenna, it was questionedby Abū al-Barakāt al-Baghdādī in his highly original, and equally critical,Kitāb al-mu‘tabar:

[P]eople use expressions in their conversations each according to what hemeans. No one means by his expression what he does not conceive of andgrasp in his mind. No one says ‘my soul’ (nafsī) or ‘your soul’ in a con-versation to refer to anything but his self (dhātihi) or his reality (h. aqīqatihi).If he says ‘my soul rejoices’ ( farih. at nafsī) or ‘your soul is suffering’ (ta’alla-mat nafsuka), there is no difference for him between that and his saying‘I rejoice’ ( farih. tu) or ‘you are suffering’ (ta’allamta). Similarly, he says ‘mysoul knows’ and ‘[my soul] is ignorant’ as if he were saying ‘I know’ or ‘I amignorant’ – there is no difference for him between his saying ‘my soul’ or ‘myself’ (dhātī) and his saying ‘I’ (anā) . . .This is the correct understandingof people’s uttering the expression ‘soul’ in their conversations. If this

27 Avicenna, Shifā’: al-Ilāhīyāt II.1.10, 48. 28 Cf. also Avicenna, Ishārāt, namat. 3, 125, 132–133.


expression were said against the first two [ways of] understanding,29 it wouldbe said in truth which convinces them both of this understanding by argu-ment and evidence (bi h. ujjatin wa bayān). When a human being grasps thismeaning from the expression ‘soul’, he does not thereby know whether thesoul is the entire body, or one of its internal or external parts which differsfrom it in nature, or an accident in the body, or whether it is an incorporealsubstance. Rather, most of them refer by means of this expression, and use itin their conversation to refer to a grasp of themselves (mafhūmin bi ‘aynihi),and they do not thereby consider or think about any of that. Thus, it isproper that this is the first grasp of this expression, I mean the expression‘soul’, and according to this grasp it is evidence of existence for everyoneuttering this expression. Thus, not a single human being needs an argumentto assert the existence of his soul. Who would doubt that he exists so that thiswould be shown to him bymeans of an argument? How could this not be thecase when according to every human being nothing is more evident thanthat, I mean more evident than the existence of his self (dhātihi)? Similarly,he does not need to be shown that another human being has a soul ora self (nafsan ay dhātan), which is his itness (huwīyatuhu) and thatness(annīyatuhu), even though he does need to be shown what his existing selfor soul is (mā dhātuhu wa nafsuhu mawjūda) and what another person’s self(dhāt) is.30

In this perspicacious analysis of common language utterances expressiveof the Avicennian phenomenon of self-awareness, Abū al-Barakāt questionsthe great Peripatetic’s claim that the connection between the incorporealsubstance of dualist psychology and the I one is constantly aware of is self-evident. Rather, he claims, no man in the street will feel compelled tocommit either to the hylomorphic theory of the soul as the enmattered formof the body or to the dualist notion of the self as an independent entity thatacts by means of the body but does not exist in it. Although the analysispresented in support of this claim seems to be motivated by the ambiguityof the Arabic nafs, which is both a common language reflexive noun and atechnical term in psychology, we should pay careful attention to its clearand precise characterization of what the common language term expresses.Abū al-Barakāt states that in common use nafs signals the speaker’s aware-ness of his individual existence, that is, of the fact that he exists (his annīya orthatness) as an individual (his huwīya, itness or heness) – all terms that had

29 This is a reference to the preceding context where Abū al-Barakāt has first introduced the hylomor-phic notion of soul as a form that functions as the first perfection of a potentially living body and thenthe Avicennian idea that the soul is an immaterial substance which animates the body. See Baghdādī,Mu‘tabar: al-‘ilm al-t.abī‘ī VI.1, II.298–300.

30 Baghdādī, Mu‘tabar: al-‘ilm al-t.abī‘ī VI.1, II.300–301.

Substanceless self-awareness 115

figured prominently in Avicenna’s discussion of self-awareness.31 The targetof the critique is therefore clear: the phenomenon of self-awareness thatAvicenna appeals to, although quite uncontroversial as such, simply doesnot have sufficient persuasive power in regard to the question about theproper category and correct metaphysical classification of the self. To put itanother way, the self-evidentiality and indubitability of one’s existencenotwithstanding, it fails to give away what it is that exists thereby.

A similar point is raised by another critical commentator, the famoustheologian Fakhr al-Dīn al-Rāzī, in his al-Mabāh. ith al-mashriqīya. Thefollowing passages are from a chapter that addresses arguments for andagainst the view that the human soul is an incorporeal substance.

Fourth, substance is a genus for what is under it, and so if the soul is asubstance, knowledge of its substantiality will be intuitive (badīhīyan) andoccur without acquisition, but the consequence is false, and the premiselikewise.32

This dense argument against the substantiality of the soul is based on theAvicennian claim of the soul’s transparency to itself: if the soul is constantlyaware of itself, and if it is a substance when considered in itself independentof its relation to the body, then surely it should be aware of its ownsubstantiality just as incontestably as it is aware of itself. However, sincesubstantiality is not intuitively evident in the soul’s awareness of itself, thesoul is not a substance.

Although the premise that the soul’s self-awareness does not give away itssubstantiality is familiar from Abū al-Barakāt, he never drew the conclusionpresented here but rather intended to make the point that the substancedualist view requires arguments that go beyond the phenomenon of self-awareness. Rāzī’s defence of the soul’s substantiality elaborates on thedistinction between self-awareness and substantiality along similar lines.

[T]he master [that is, Avicenna] has said that we only know of the soul that itis something governing the body, but as regards the quiddity of that thing,it is unknown, and the substance essential to that quiddity (al-jawharual-dhātīyu li tilka al-māhīya) is not grasped to be something which governsthe body, and so what constitutes it as a substance is not known to us, andwhat is not known to us constitutes the substance, and therefore theobscurity remains.

Someone may say that according to the above, my knowledge of my soul(‘ilmī bi nafsī) does not occur by acquisition, and so either I do not know my

31 See, for instance, Avicenna, Ta‘līqāt 160–161; Shifā’: Fī al-nafs I.1, 16 Rahman; V.7, 253 Rahman.32 Rāzī, Mabāh. ith II.2.2.1.3, II.246.


soul except insofar as it has a relation to my body, or I know its reality, andthe first is false, for it has been established above that my knowledge of mysoul is prior to my knowledge of its relation to my body. Moreover, howcould this statement be sound from someone who says that my knowledge ofmy soul is identical to my soul (nafsu nafsī) and that it is always present(h. ād

˙ir) in act? It is amazing that someone says something like these two

statements and then neglects their incoherence, albeit not affirmatively.The sound answer is to say that substantiality is not one of the essential

things (al-umūri al-dhātīya), and because of that it can remain unknown ashas been shown.33

The passage begins with a paraphrase of the situation in which Avicennaintroduces the flying man in Shifā’: Fī al-nafs I.1.34 We know that the soulexists from its action in the body it animates, but we lack information aboutthe entity itself that functions as a soul, that is, whether or not that entity isan incorporeal substance independent of its relation to the body. Thecounterargument that accuses Avicenna of incoherence is somewhat lessclear. The idea seems to be that the two statements that my self-awareness isconstant and that I am my self-awareness, when taken together, amount tothe claim that the self is fully transparent to itself. This, however, is inblatant contradiction with the confession that we lack knowledge ofwhether the self is a substance or not. The particularly puzzling feature inthe accusation is that it seems to all but ignore the argument of the flyingman which is precisely designed to indicate the human soul’s substantialityon the basis of its awareness of itself.35 But regardless of this puzzle,Rāzī’s defence is unambiguous: substantiality is not self-evidently givenin self-awareness because it is accidental to the phenomenon. This is astriking claim and an obvious departure from the Avicennian theory itallegedly defends. Although Rāzī’s statement seems to be instigated byAbū al-Barakāt’s critical insight, as well as his own materialist theory ofthe human soul,36 he takes the idea a step further, for instead of merelyquestioning the demonstrative force of self-awareness in the question of

33 Rāzī, Mabāh. ith II.2.2.1.3, II.246–247.34 Cf. Chapter 2.1.

35 Perhaps the accusation is based on the epistemological status of the reminder, which is perceived to beinferior to that of a demonstration proper. If self-awareness is as pervasive and immediate as Avicennaclaims, surely our substantiality too should be indubitable. This would be incompatible with Rāzī’scritique of the corresponding section of the Ishārāt in his commentary (see Sharh. al-Ishārāt, 121–122;for discussion, see Marmura 1991). At least some later commentators attempted to develop a logicallyvalid demonstration on the basis of the flying man; for an account of this process in the thirteenth-century commentator Ibn Kammūna, see Muehlethaler 2009. Later on, Mullā S· adrā presents hisversion of the flying man as a demonstration of the soul’s immateriality (see Asfār IV.2.2, 47; and cf.Chapter 7).

36 See Marmura 1991, 630–633.


substantiality, Rāzī expressly denies any necessary connection between self-awareness and substantiality. As a result, the phenomenon of self-awarenessis abstracted from the particular metaphysical basis of the concept ofsubstance.

I am not claiming that this abstraction was Rāzī’s aim. But be that as itmay, the theoretical possibilities opened by such a conceptual move did notpass unnoticed by Suhrawardī.37 Bearing in mind Suhrawardī’s generalcriticism of the concept of substance as a foundation of metaphysics, it isnot surprising to come across a critique of Avicenna along similar lines. Butparticularly interesting is the manner in which Suhrawardī develops theingredients laid on the table above. Moreover, the critical argument againstthe substantiality of the self-aware human being is presented in both thecomparably more Peripatetic setting of the Talwīh.āt and the overtly ishrāqīframework of the H· ikma al-ishrāq, which suggests that Suhrawardī consid-ered it to be of prime importance to his departure from the tradition.

The beginning of the last, eschatological section of the Talwīh. āt is alengthy eulogy, rife with symbols, for knowledge and its soteriologicalpower. Much of its terminology is familiar from Suhrawardī’s less technicalworks, and it therefore signals a move away from a strictly Peripateticconceptual framework.38 Just before the passage below, Suhrawardī haspraised a group of thinkers he characterizes as ‘the people of wisdom’,seemingly a reference to representatives of the Peripatetic philosophicaltradition that has been the object of study for the preceding sections ofthe book, for having revealed the most important doctrines concerning theworld’s origination from and return to its Creator, that is, the principles ofthe knowledge required for human perfection. However, he then adds apoint he has qualms about.

You have made the greatest destination evident by the clearest argument . . .were it not for one word here, namely that I was separated by myself(tajarradtu bi dhātī), gazed into it and found it to be thatness and existence,and it was added to ‘not be in a subject’ – which is like a description ofsubstantiality – and [to have] relations to the body – which is a description ofsoulness. Regarding the relations, I found them external to it, and regarding

37 I do not claim that the following discussion in Suhrawardī derives from this particular work by hiscontemporary Rāzī, whom he certainly knew, for both had studied theology and philosophy withMajd al-Dīn al-Jīlī (Walbridge 2005, 201). It is perfectly possible that Rāzī and Suhrawardī had acommon source, or that both were addressing a topic pertinent to the twelfth-century reception ofAvicenna. For a brief but lucid characterization of this crucial period, see Eichner 2011, 117–119.

38 Talwīh.āt III.5.1–11, 276–279 Habībī; 105–110 Corbin. Chapter 10 is especially interesting because itspeaks of isolation or retreat into oneself from externally caused appearances in order to reach a highertype of knowledge, and thereby paves the way for our passage in chapter 15.


its ‘not being in a subject’, it is something negative. If ‘substantiality’ hasanother meaning, I did not come across it, whereas I did come across my selfand was not absent from it (uh. as.s.ilu dhātī wa anā ghayru ghā’ibin ‘anhā), andit did not have a difference ( fas.l), and so I know it by the very non-existenceof my absence from it (u‘arifahā bi nafsi ‘adami ghaybatī ‘anhā). If it had adifference or a property (khus.ūs.īya) beyond existence, I would have appre-hended them when I apprehended it, for nothing is closer to me than me,and upon analysis I only see in myself ( fī dhātī) existence and apprehension,nothing else. [That] is distinct from another through accidents, and appre-hension is as has been [explained] above, and so only ‘existence’ remains.Furthermore, if apprehension is supposed to have an acquired grasp – againstwhat has been said – it will be apprehension of something, and [my self] isnot constituted by an apprehension of itself (bi idrāki nafsihā) since it [that is,the apprehension] would be after [the self] (huwa ba‘ada nafsihā), nor by anapprehension of another since that is not concomitant to it. Preparednessfor apprehension is accidental, and anyone who apprehends himselfunder the concept ‘I’ (man adraka dhātahu ‘alā mafhūmi anā) will not findupon analysis and inspection anything but existence that apprehends itself(nafsahu), and he is it. The concept of I as the concept of I (mafhūmu anāminh. aythu mafhūmu anā) – as far as it is common to both the Necessary andothers – is that it is something which apprehends itself (dhātahu). Thus, if Ihad a reality other than this, the concept ‘I’would be accidental to it, and so Iwould apprehend the accidental due to my not being absent from it whilebeing absent from myself (akūnu anā udriku al-‘arad

˙īya li ‘adami ghaybatī

‘anhu wa ghibtu ‘an dhātī), and that is absurd. Thus, I judge that my quiddityis identical with existence, and my quiddity in the intellect is not divisibleinto two things, with the exception of negative things – which have beengiven existential names – and relations.39

Suhrawardī has no objections to the Avicennian phenomenon of self-awareness; his qualms are exclusive to the added qualifications that the self isnot in a subject (that is, that it is an incorporeal substance) and that it hasrelations to the body (that is, that it is a soul). The latter qualification isquickly done away with as something unessential to the self, which of courseis not adverse to Avicenna who applied self-awareness precisely to pointtowards the existence in itself of the entity that functions as a soul in relationto the body. The rejection of the self’s substantiality, on the other hand, ismotivated by a metaphysical departure from Avicenna. Since substantialitycan only be defined negatively as not being in a subject, it depends on amind that is capable of negating a property of its object. As a negativeproperty, substantiality is not a real feature of the object considered

39 Talwīh.āt III.5.15, 282–283 Habībī; 115–116 Corbin.


anymore than a dog that I presently distinguish from being human isconstituted by this perceived non-humanity.40 The dog’s constitutive fea-tures are independent of what is specific to human beings, and althoughthey entail the negative property of non-humanity, this entailment can bemade actual only by a mind that is distinct from the real particular dogness.

Notice that Suhrawardī applies the argument found in Abū al-Barakātand Rāzī – substantiality is not evident in self-awareness – in a new system-atic framework. Substantiality is no longer merely accidental to self-awareness. As a negative property it is not even an accident but somethingimported to the primitive phenomenon of self-awareness in a considerationthat is external to it. Suhrawardī’s point in accentuating this critical depar-ture from Avicenna is brought out by his additional consideration of self-awareness as mere apprehension and existence, that is, separated from anysubstance whose apprehension and existence would be at stake. Even assuch a bare actuality, apprehension is found problematic. This is first andforemost because apprehension entails a relation to what is apprehended.There is no apprehension absolutely speaking, because apprehension mustalways be realized (made ‘distinct from another’) by its content, whereasself-awareness is characterized in the arguments discussed above as lackingany specific content. Interestingly, however, Suhrawardī grants the targetof his argument the benefit of the doubt by considering the possibilitythat self-awareness is due to an acquired content (mafhūmun muh. as.s.al, or‘acquired grasp’) of something specific to the self.41 This is refuted by meansof a very condensed version of the Avicennian argument against reflection-based models of self-awareness: the apprehension of the content specific tothe apprehending subject must be subsequent to the subject, for the subjectcan recognize the content as itself only by means of a prior familiarity withherself. Finally, Suhrawardī considers, albeit in a rather condensed argu-ment, the special case of the first-person indexical ‘I’ as a concept specific tothe self-aware subject. While self-awareness in the primitive sense (as‘existence that apprehends itself’) does provide the reference to this concept,it cannot be brought about by the concept, because our explicit use of thefirst-personal indexical, both in speech and in thought, is intermittent andhence accidental to our existence. Consideration of a concept in the absenceof reference is consideration of a mere concept, and if the concept ‘I’ has no

40 For a more detailed account of Suhrawardī’s critique of the Peripatetic concept of substance, seeChapter 6.2, 142–144.

41 This is exactly parallel to Avicenna’s rhetorical manner of procedure in Ishārāt, namat. 3, 120; seeChapter 4.1, 72–75.


primitive self-awareness as its basis,42 then in my use of it I will apprehendsomething that is accidental to myself without apprehending myself.Suhrawardī clearly thinks he has the weight of the epistemological traditionand common sense behind him when he declares the view as absurd.Thus, this series of remarks connects a familiar Avicennian argument

for the immediacy of self-awareness to a decidedly un-Avicennian view ofthe metaphysics underlying the phenomenon. Self-awareness is no longerthe existence of a substantial self but merely a certain type of existence, anexistence that apprehends itself, which does not belong to any entitydistinct from it. The first-personality that Avicenna carefully delimited asthe existence of a certain type of existent is now divested of that existent.That these remarks amount to Suhrawardī’s final take in the discussion ofself-awareness is corroborated by the emergence of a very similar argumentin theH· ikma al-ishrāq, right at the heart of a section in which self-awarenessis cast in an entirely new theoretical role. Having emphasized the immedi-acy of self-awareness by the familiar method of separating it from all possibleobjects of cognition, including one’s own body, he goes on to consider thecase of substantiality.

If substantiality is the perfection of [the self’s] quiddity or is taken to meanthe denial of a subject or a substrate, it is nothing independent that your selfcould be identical to (laysat bi amrin mustaqillin takūna dhātuhā nafsuhāhiya). If substantiality is taken to be an unknownmeaning (ma‘nā) and if youapprehend yourself continually and not by means of something additional,then this substantiality, which is absent from you, is neither your selfas a whole (kulla dhātika) nor a part of your self (juz’a dhātika). On closeexamination you will not find anything by means of which you are you43

except something that apprehends itself (shay’an mudrikan li dhātihi):this is your ‘I-ness’ (anā’īyatuka). Everything that apprehends itself and itsI-ness (kullu man adraka dhātahu wa anā’īyatahu) shares this with you.Apprehensiveness (al-mudrikīya), therefore, is not by means of an attributeor anything additional, whatever it is like. It is not a part of your I-ness, sincethe other part would then remain unknown as long as it remains beyondapprehensiveness and being aware (al-shā‘irīya), and so it would be unknownand would not belong to your self whose awareness is not additional to it(lā yakūna min dhātika allatī shu‘ūruhā lam yazid ‘alayhā). Thus, it is evident

42 Cf. Talwīh.āt III.3.1, 239 Habībī, 70–71 Corbin, where Suhrawardī describes the concept ‘I’ as acommonplace universal.

43 I readmā anta bihi anta with Corbin instead of the rather awkwardmā an bihi anta in Walbridge andZiai. The latter may be a typographical error, for Walbridge and Ziai’s translation accords withCorbin’s reading.


in this way that thingness is not added to being aware either, and so it44

appears to itself by itself (huwa al-z· āhiru li nafsihi bi nafsihi), and there is noproperty with it so that appearing (al-z· uhūr) would be its state. Rather, it isthat which appears itself (huwa nafsu al-z· āhir), nothing else.

45

The argument is familiar from the Talwīh.āt. If substantiality merelyindicates that the attributee is not in a subject, it is a negative property andtherefore nothing that can be found in reality independent of the act ofattribution. If, on the other hand, we suppose substantiality to be a realproperty of ourselves, we come across Abū al-Barakāt’s observation that thesoul’s substantiality is simply not given in the experience of a self-awaresubject. When people refer to themselves by means of first-person indexicalsand related expressions, they do not intend to commit to any ontologicaltheory of their selves. All they do is point out that they are aware of themselves,and that is what Suhrawardī urges the philosopher to settle with as well.

Earlier on in the metaphysical section of the H· ikma al-ishrāq, Suhrawardīhas allowed for a certain qualified use of the concept of substance. This is inthe case of corporeal entities, which owing to the strictures of their materialconstitution can never appear completely or in their entirety. For a simpleexample, I cannot see the underside of the table on which I write unless I benddown to look at the table from below, but this will in turn rob me of the viewof its upper surface. Yet in spite of all surfaces of the table never appearingsimultaneously, it seems intuitively plausible to attribute them all to asingle thing as its simultaneous aspects. This opacity or ‘darkness’ (z· ulma)of corporeal things can be overcome by means of the concept of substance towhich the different aspects can be attributed. It is important to notice that theconcept of substance is fundamentally linked to this opacity in the H· ikmaal-ishrāq, where Suhrawardī constantly uses the conjunctive formula ‘duskysubstance’ (jawhar ghāsiq), and that substantiality is added in intellectualconsideration and never appears as such.46

But when it comes to my presence to myself, we are dealing with fulland constant transparency, which bereaves us of all reasons to attributesubstantiality to ourselves – nothing hidden is required for understandingself-awareness. As a parallel conceptual move Suhrawardī also denies thatself-awareness is a property, or even existence, that belongs to a thing. Bydenying thingness (shay’īya) of self-awareness – ‘thingness’ being one of the

44 It is not obvious what the masculine third-personal pronoun refers to here. I take the reference to beto ‘everything that apprehends itself and its I-ness’ two sentences earlier.

45 H· I II.1.5.116, 80–81Walbridge and Ziai; 112–113 Corbin; cf. II.1.5.118–119, 81–82Walbridge and Ziai;114–115 Corbin.

46 H· I II.1.3–4, 77–79 Walbridge and Ziai; 107–110 Corbin.


most general concepts in Avicenna’s metaphysics47 – Suhrawardī denies theviability of all metaphysical accounts of the phenomenon, beyond mereostension and naming, for if the self-aware subject cannot even be said to bea thing, we can hardly expect to reach very far in attempting to define whatkind of thing it is. Moreover, slightly later in the same section Suhrawardīargues against the view that self-awareness, like any act of intellection, couldbe explained by means of the incorporeality of its subject,48 which amountsto a rejection of another explanatory feature in Avicenna’s approach to self-awareness.As the outcome of these critical remarks Suhrawardī concludes that self-

awareness is self-awareness, period. It is not a property of any thing, not evena mode of existence, which could yield it to alternative metaphysicalaccounts. All we can do, and should even aspire to, is point towards it inan analytical description of our experience. But such a description willmerely pick out something that is immediately obvious to us. Once it hasbeen so picked out, it can be named but it can never be defined or explainedby means of anything more basic or better known.Suhrawardī had a reason for rejecting both the inference from self-

awareness to the substantiality of the human soul and the idea thatself-awareness is definable or explainable in any way. As we will see, heresorts to the phenomenon of primitive self-awareness in giving content tothe foundational notion of his metaphysics, that of pure incorporeal light,and for this purpose substantiality would be quite superfluous, indeedpotentially problematic. Moreover, the simultaneous obviousness andundefinability of self-awareness provides Suhrawardī with a foundationaltype of knowledge, on which he builds his revisionist concept of knowledgeas presence.

47 For this status of thingness, see Avicenna, Shifā’: al-Ilāhīyāt I.5, 22–29.48 H· I II.1.5.119, 82–83 Walbridge and Ziai; 114–116 Corbin. Suhrawardī’s somewhat sophistical argu-

ment is based on the idea that prime matter is also incorporeal in the sense that it is not determinedinto discrete bodies. Thus, if self-awareness is due to incorporeality, primematter will also be aware ofitself. Since the conclusion is absurd, we must reject the explanatory connection between self-awareness and incorporeality.


chapter 6

Self-awareness, presence, appearance:the ishrāqī context

In spite of his major debt to Avicenna’s philosophy, Suhrawardī departsfrom his predecessor in two interrelated questions of fundamental impor-tance. The first of these concerns the foundation of epistemology: what inthe final analysis is primarily known, epistemologically primitive and foun-dational, and therefore the basis of our knowledge? The second questionpenetrates the corresponding foundation of metaphysics: what in the finalanalysis is primarily there, metaphysically primitive and foundational, or, inAristotelian terms, the primary sense of the verb ‘to be’?

Interestingly for our topic, self-awareness functions in a most prominentrole in the articulation of the Suhrawardian alternative in both of the twoquestions. The epistemological point is stated most explicitly in the Talwīh.ātand the Mashāri‘ wa al-mut.ārah.āt, that is, in those of Suhrawardī’s philo-sophical works that are considered ‘Peripatetic’ in comparison to the full-blown illuminationism of theH· ikma al-ishrāq. In fact, the positive treatmentof the concept of knowledge in the latter is relatively scant, the focus beinginstead on the metaphysical question. The presence of its overarching termi-nology shows, however, that the concept of knowledge articulated in the twosystematically ‘prior’ works is preliminary to illuminationist metaphysics.One can therefore speculate upon the argumentative order between the twodepartures and say that the concept of knowledge as presence is a condition ofthe concept of being as appearance in much the same way as the generalaccount of being comes after – indeed much later than – the theory of sciencein the structure of the Peripatetic system. This would tally neatly withSuhrawardī’s own advice to study his philosophical treatises in an orderculminating in the H· ikma al-ishrāq.1 Systematically speaking, however, itseems more natural to conceive of the two as fundamentally interdependent.Indeed, it does not seem to be a coincidence that Suhrawardī’s metaphysicshinges on the concept of appearance (z· uhūr), derived from a term

1 See Mashāri‘ III, muqaddama, 194 Corbin; cf. Ziai 1990, 9–11, 14–19.

124

with a primarily epistemic meaning, whereas the corresponding epistemologyis based on the concept of presence (h.ud

˙ūr), a term with existential

connotations.2

6.1 Self-awareness and knowledge as presence

Although the Talwīh.āt follows a rather traditional Peripatetic order ofprocedure, Suhrawardī introduces the notion of knowledge as presenceneither in the sections devoted to the theory of science nor in thosediscussing cognitive psychology. Rather, it is only towards the end of thethird, metaphysical part, when tackling the problem of God’s knowledge,that he has recourse to the term and the idea behind it. The term figuressystematically only in the enigmatic passage in which Suhrawardī recountshow Aristotle, as an emphatically mythical figure, appears to him in adreamlike vision after he had exhausted himself labouring with ‘the ques-tion of knowledge’.3

Much has been written about this passage, but few writers have paid closeattention to its context. Two observations in particular are quite crucial for afull comprehension of why and in what sense Suhrawardī introduces theconcept of presence here. First of all, the passage is embedded in themetaphysical discussion of God, the particular explanandum being God’sknowledge of particular things.4 Secondly, this section of Talwīh. āt showsconsiderable similarities with the way in which self-awareness figures in thediscussion of God’s knowledge in Avicenna’s Ta‘līqāt.Before introducing the appearance of Aristotle, Suhrawardī briefly

reviews the problems in some of the available alternatives for makingsense of knowledge in general and of God’s knowledge in particular. Hefirst argues against the identity theory of knowledge in a manner that isentirely derivative of Avicenna.5 But he also has qualms with Avicenna’sown theory, according to which cognitive forms inhere in the knowingsubject, universal forms in an intellectual subject, and particular forms ina corporeal organ of perception.6 According to the standard Avicennianphrase, God knows particular things ‘in a universal manner’, which is often

2 The rest of this chapter is a refined adaptation of Kaukua 2013 and 2011, respectively.3 Talwīh. āt III.3.1, 238–239 Habībī; 70 Corbin.4 This observation is somewhat controversial; I will discuss Eichner’s (2011) alternative interpretationbelow.

5 Talwīh. āt III.3.1, 237–238 Habībī; 68–69 Corbin. Cf. Avicenna, Shifā’: Fī al-nafs V.6, 239–240Rahman.

6 Eichner 2011 (119–127) shows that the problems Suhrawardī raises were commonplace in the twelfth-century reception of Avicennian epistemology.

Self-awareness and knowledge as presence 125

qualified by saying that He knows His creation by knowing Himself as itscause.7 The universality of God’s knowledge is due to His immateriality,which entails that He is a subject of intellection and that therefore theproper objects of His knowledge are universal.8 The equivalence of imma-teriality and intellectuality is explicitly borne out in Avicenna’s discussionof human access to particulars in Shifā’: al-Madkhal I.12.9 Given that thehuman soul is intellectual, the manner of cognition proper to it is theapprehension of universals, and so as an intellect, like God, it only hasaccess to particular entities in a universal manner. For instance, I can onlyknow my friend Zayd as a human being with a particular complexion,build, gait, humoral character and so forth – with as many other universalattributes as I like to add. But the problem is that the Zayd that I therebygrasp is not a particular person but a bundle of universal properties whichcan in principle be shared by individuals other than Zayd. Yet I find itintuitively plausible that my friend is a unique person whose individualitycannot be reduced to the accidental fact that there happen to be no otherhuman beings with the exact same bundle of properties. The problemSuhrawardī seems to perceive here is that I am somehow certain thatI apprehend an individual in this strong sense and that our theory ofknowledge should be able to save this intuition.10

But Avicenna did propose a solution to the dilemma: human beings arenot merely intellects but also souls that function in and engage with thematerial world by means of corporeal instruments proper to them. Thus,their faculties of sense perception allow them an ostensive reference to theunique spatiotemporal co-ordinates which are the foundation of the indi-viduality of material entities. The person I am conversing with can benone other than my friend Zayd because I perceive him as this individualhere right now.11 Now, Suhrawardī is perfectly aware of this attempt at asolution.12 Why does he find it unsatisfactory?

It has recently been suggested that this is because of problems related toAvicenna’s substance dualism, more precisely his inability to make lucidsense of the relation between the immaterial human substance and itsbody.13 Since the case is exclusively epistemological here, the relevant

7 For the relevant texts and discussion, see Marmura 1962 and Adamson 2005.8 For this traditional tenet in Avicenna, see Adamson 2011a.9 Avicenna, Shifā’: al-Madkhal I.12, 70; cf. Chapter 3.1, 48–50.10 Talwīh.āt III.3.1, 237–238 Habībī, 69 Corbin; for discussion, see Eichner 2011, 129.11 Avicenna, Shifā’: al-Madkhal I.12, 70; cf. Ishārāt, nahj 1, 5–6. For discussion, see Black 2012 andEichner 2011, 130.

12 Talwīh.āt I.1.4, 8 Habībī; cf. Eichner 2011, 129. 13 Eichner 2011, 135–136.

126 Self-awareness, presence, appearance

aspect of the mind–body relation is of course how a material process in theorgans of perception can cause an immaterial appearance of a particular objectin a cognitive subject that is designed to apprehend universal objects. Merelystating that the organs are causally related to particular things does not explainhow those things can be given as appearances to the soul.Although this problem is a real one for any substance dualist, it is difficult

to see how Suhrawardī’s proposed solution could meet with any greatersuccess. I will revisit this point once we have a clearer idea of whatthe concept of knowledge as presence is about, but let it now be said thatI believe a rather different motive for Suhrawardī’s dissatisfaction emergesfrom the context of discussion. Two conditions are relevant here. First of all,since God is absolutely one, His knowledge of Himself and the world of Hiscreation cannot be two pieces of knowledge inHim. Secondly, since God isthe supreme knower, He has to somehow know the world lest there be anydeficiency to His knowledge.14 As a result, Suhrawardī needs a concept ofknowledge that is capable of making sense of a subject’s simultaneousknowledge both of itself and other objects, and allows for both particularand universal objects to be given to the same subject. Avicenna’s theory ofknowledge, because it is based on the inherence of cognitive forms in theknowing subject and makes the apprehension of particular objects condi-tional to a relation to matter, fails on both accounts as an explanation ofGod’s knowledge. This is why Suhrawardī attempts to carve the conceptualmap anew by means of the notion of knowledge as presence.In fact, it is precisely in the solution of the problem of how to account for

the possibility of an immaterial subject apprehending particular objects thatAristotle comes to help Suhrawardī:

[a] So he said to me: Return to yourself (irja‘ ilā nafsika), and it will besolved for you.

I said: How?And he said: You apprehend yourself, and your apprehension of yourself is

either by yourself or by means of another (innaka mudrikun li nafsika, faidrākuka li dhātika bi dhātika aw ghayrihā), but then you would haveanother faculty or self that apprehends yourself (dhātun tadrukudhātaka), and the discussion would revert, and so its absurdity is evident.[b] Since you apprehend yourself by yourself (adrakta dhātaka bidhātika), is that by considering a trace of yourself in your self (bii‘tibāri atharin li dhātika fī dhātika)?

I said: Of course.

14 For these conditions, see the discussion immediately following the account of Aristotle’s appearance,especially Talwīh.āt III.3.1, 244–245 Habībī; 75–76 Corbin.


He said: Then if the trace does not correspond to your self (dhātaka),it will not be its form, and you will not apprehend [your self].I said: Thus, the trace is the form of my self (dhātī)?He said: Is your form of an absolute soul or one individuated by other

attributes?And I opted for the second.

So he said: Every form in the soul is universal – even if it werecomposed of many universals – and it does not prevent participation initself; if it is supposed to be prevented, that is due to another preventing[factor]. You apprehend yourself (anta mudriku dhātika), and it preventsparticipation in itself, and so this apprehension is not of form.So I said: I apprehend the concept ‘I’ (udriku mafhūma anā).And he said: The concept ‘I’ as the concept of I does not prevent

participation from occurring in it, and you know that the particular,insofar as it is nothing but a particular, is universal; ‘this’, ‘I’, ‘we’ and‘he’ have universal intelligible meanings with respect to their separateconcepts without particular reference.So I said: How then?He said: Since your knowledge of yourself (‘ilmuka bi dhātika) is not

by means of any faculty other than your self (dhātika) and you know thatyou are nothing but the one apprehending your self (anta al-mudriku lidhātika lā ghayr), not by a trace that does not correspond and not by onethat does, your self (dhātuka) is an intellect, that which understands andthat which is understood.15

Aristotle relies on the indubitable phenomenon of self-awareness in anextended argument against the validity of the impression theory of knowledgeas an account of our apprehension of particular things such as ourselves. Hebegins in section [a] by countering the view that self-apprehension is by meansof something apart from the self that apprehends and is apprehended. Theargument is a condensed version of the familiar refutation of reflection-basedmodels of self-awareness, construed here as a regress argument: if awareness of aspecial object is supposed to render me aware of myself, then I must somehowrecognize the other as myself, which forces us again to face the question ofwhether this recognition is due to the same self being both that whichapprehends and that which is apprehended or due to a further special object.Since the regressmust be ceased at the earliest possible stage, the thesis that self-awareness is bymeans of a special object distinct from the self can be ruled out.

Section [b] addresses the possibility that although self-awareness isdue to the self instead of any distinct object, it should be explained by aspecial feature in the self, such as a trace or a form, that is caused by andcorresponds to the self. Here we come across the concern extrapolated

15 Talwīh.āt III.3.1, 239–240 Habībī; 70–71 Corbin.


above. Since the soul is immaterial, all forms in it will be universal. Thus, ifI were aware of myself by means of an impression of the form of myself,I would be aware of something universal, and this has to be rejected in theface of my intuitive certainty of my uniqueness as an individual. Along linesfamiliar from his denial of the self’s substantiality, Suhrawardī also suggeststhat self-awareness is induced by an apprehension of the first-person index-ical, but Aristotle cogently argues that there’s nothing individual in theconcept ‘I’when taken alone as a concept. All subjects of experience share inbeing an I, and although the concept expresses in each case an inherentlyindividual self’s awareness of itself, as a point of reference this awareness isprior to the concept.Thus, self-awareness is presented here as a paradigmatic example of

knowledge that cannot be explained as a case of the inherence of what isknown in the knower. As we know, this is something Avicenna would haveagreed with. However, as we have also learned, Avicenna was ill at ease in hisattempts to articulate the cognitive category proper to self-awareness. Thistension in Avicennian cognitive psychology is increased by further phenom-ena that the mythical Aristotle introduces to Suhrawardī. The first of theseis our constant awareness of our bodies as being unique to ourselves.Thus, our awareness of them cannot be due to a universal form of thebody inherent in us.16 Further corroboration for the claim that we must beaware of individual objects is provided by a brief excursion into Avicennianfaculty psychology and its account of discursive thought by means of thesystem of internal senses. The human faculty of thought (mufakkira) canonly operate by means of particular objects of cognition, which Suhrawardīconsiders especially problematic because thought plays an indispensablerole in the acquisition of the universal objects of intellection proper to theimmaterial subject of cognition. Were it not for this faculty, we could neverabstract from the particular features of our percepts, nor could we arrangeacquired propositions into syllogisms that render us capable of understand-ing new propositions as their conclusions. Thus, Avicennian epistemologyitself requires that we are aware of both universal and particular objects.Finally, Suhrawardī’s Aristotle suggests that we are indubitably aware of theactivity of our internal senses in such processes of thought, and this aware-ness is something that none of these faculties is capable of, for since theyact through a corporeal organ, they cannot establish a transparent relationto themselves. On the other hand, if the incorporeal self only knows by

16 Talwīh. āt III.3.1, 240Habībī; 71Corbin. Suhrawardī uses the expression lā taghību ‘anhu (‘you are notabsent from it’) of the body, indicating thereby his concept of knowledge as presence.


means of universal forms inhering in it, it cannot grasp this particularactivity either.17 The unstated conclusion is that a new concept of knowl-edge is direly needed.

Having dealt with these phenomena as so many problems for theimpression theory of knowledge, Aristotle moves on to present a newpositive definition:

[c] He said: Since you know that [the soul] apprehends neither by means of acorresponding trace nor by means of a form, know that intellection is thepresence of the thing to a self (al-dhāt) separate from matter, or if you like,you can say: [the thing’s] not being absent from [the self]. This is morecomplete because it includes the apprehension of something of both itself(li dhātihi) and another, for the thing is not present to itself, but it is notabsent from [itself] either (al-shay’u lā yah. d

˙uru li nafsihi wa lākinna lā

yaghību ‘anhā). As regards the soul, it is separate and not absent from itself(ghayru ghā’ibatin ‘an dhātihā), and in accordance with its separation itapprehends itself (dhātahā) and what is absent from it, which when it isnot made present to [the soul] in concrete (‘aynihi), such as heaven, earth andtheir kind, [the soul] makes present its form. As regards particulars, they arein faculties that are present to [the soul], and as regards universals, they are in[the soul] itself ( fī dhātihā), for among those that are apprehended theuniversal is not impressed in bodies, and what is apprehended is the veryform that is present, nothing external to conception. If it is said of theexternal that it is apprehended, that is in a secondary sense. [The soul’s] self isnot absent from itself (dhātuhā ghayru ghā’ibin ‘an dhātihā), nor is its body[absent from the soul] in any regard whatsoever, nor are any facultiesapprehending its body [absent from the soul] in any regard whatsoever.18

At the outset, knowledge is defined as the presence of what is known tothe knower, and so it is contrasted with the Avicennian theory based onthe inherence of a representation or form of what is known in the knower.The notion of presence is then rephrased by means of a semantic doublenegation as the known object’s not being absent from the knower. Thelatter definition is stated to be more appropriate because it is inclusive ofboth self-awareness and the awareness of other objects, which suggeststhat presence is a special case of the non-existence of absence. All objectsof cognition, whether universal or particular, as well as the cognitivecapacities of the subject that is aware of them, are known by the subjectthrough their presence to her. Since they are present to the subject, bydefinition they cannot be absent from her. On the contrary, the subject’s

17 Talwīh. āt III.3.1, 240 Habībī; 71 Corbin.18 Talwīh.āt III.3.1, 240–241 Habībī; 71–72 Corbin; cf. Mashāri‘ III.7.1, 487.


awareness of herself is not due to presence, for presence being characteristicof objects, nothing can be present to itself, but the subject is not absent fromherself either, and in this latter sense her self-awareness can also be spoken ofas knowledge. Thus, it seems that knowledge in the more specific senseof presence means being an object of knowledge for a subject, or, to use aSuhrawardian expression, an it that appears for an I. The non-existence ofabsence on the other hand seems to be a more vague phrase for simply beinggiven as a matter of experience, appearing either as an it or as an I.Suhrawardī’s epistemological use of the concept of presence is not unique

in the twelfth-century reception of Avicenna. Fakhr al-Dīn al-Rāzī appliesthe term profusely in al-Mabāh. ith al-mashriqīya19 and especially in thesummary of Avicenna’s Ishārāt known as the Lubāb al-Ishārāt.20 In thelatter work, Rāzī begins the section on cognitive psychology by a definitionof apprehension (idrāk) in terms of presence: ‘Apprehension amounts to thepresence of the form of what one is aware of in the one that is aware (h. ud

˙ūru

s.ūrati al-mash‘ūri bihi fī al-shā‘ir).’21 The notion of presence is then used to

argue for a mental mode of existence for the immediate objects of knowl-edge that is independent of whether anything corresponds to them inextramental reality. Later on in the same section, in presenting an argumentfor the existence of ‘the holy faculty’ (al-quwwatu al-qudsīya),22 Rāzī equiv-ocates between presence and awareness:

When we have apprehended an intellectual form and then forgotten it, it issaid that after forgetting that form either is or is not present in our souls. Thefirst is impossible, because if it were present in our souls, one would be awareof it (law kānat h. ād

˙iratan fī nufūsinā la kānat mash‘ūran bihi), for there is no

other meaning of awareness than that very presence (lāma‘nā li al-shu‘ūri illānafsu dhālika al-h. ud

˙ūr).23

Rāzī here refutes the existence of intellectual memory and then explains theempirical fact that we do recall matters of understanding by means of arenewed connection to the active intellect. The holy faculty, around whichthe argument as a whole revolves, is then defined as the spontaneity of sucha connection. Our interest, however, lies in the connection of awareness topresence. If the argument against intellectual memory is to hold, my being

19 Cf. Rāzī, Mabāh. ith II.1.2.3.1.3, I.444; and see Eichner 2011, 126–127.20 See, for instance, Rāzī, Lubāb, namat. 3, 2, 235–242.

21 Rāzī, Lubāb, namat. 3, 2, 235.22 This is a paraphrase of Avicenna, Ishārāt, namat. 3, 127. The holy faculty is the rare human capacity of

connecting to the active intellect spontaneously, without recourse to learning or discursive thought.Avicenna’s explanation of prophecy hinges on this capacity; cf. Avicenna, Shifā’: Fī al-nafs V.6,248–250 Rahman.

23 Rāzī, Lubāb, namat. 3, 2, 239.


aware of an object must mean that the object appears to me as an object ofmy actual consideration, for were that not the case, it is difficult to see whatthe intuitive evidence could be which Rāzī here clearly relies on. Thus, both‘awareness’ and ‘presence’ entail that something is experientially given.

Rāzī does not define knowledge as presence – that is a uniquelySuhrawardian move – but his way of characterizing the term corroboratesthe interpretation according to which Suhrawardī’s concept of presence isdesigned to focus on knowledge as a matter of experience.We are intuitivelycertain that we have experience of individual entities, or to put it anotherway, that such entities are experientially present to us, and in this sense itmakes sense to say that we know them. As is suggested by the strong relianceon intuitive certainty in the discussion leading to section [c], Suhrawardīthinks that it should not, indeed cannot, be explained away in the face ofproblems that are due to the Avicennian combination of substance dualismand the inherence theory of knowledge. Diametrically opposed to Avicennain this regard, Suhrawardī takes his cue from this intuitive certainty and isdetermined to make room for it in his concept of knowledge. His method ofprocedure thereby all but neglects the entire question of truth, absolutelycrucial to all normative concepts of knowledge. The question of the corre-spondence between the experiential presence of what is known and itsextramental existence is set aside by the simple statement that it is subse-quent to presence. Only once something is present to me in experience can Ibegin to consider the question of whether anything like what is presentexists without my experience.24

But why would one want to design a concept of knowledge shorn of thisnormative dimension? As I stated in the beginning of this section, the entireappearance of Aristotle is embedded in a theological context, and it isthereby motivated by the task of explaining how, or in what sense, Godcan be said to know His creation, constituted as it is by an infinite numberof individual entities, in spite of the fact that He is absolutely one in amanner that our conception of His knowledge must not violate. In view ofthis aim, it is not relevant to consider the difference between truth andfalsity in human knowledge. In the Peripatetic framework, this difference isdealt with naturalistically in terms of a psychological account of howknowledge comes to be, and Suhrawardī has related a standard, if somewhatsimplified, Avicennian version of such an account earlier on in theTalwīh.āt.

25 But if one is to explain God’s knowledge by means of its

24 This is pointed out by Ha’iri Yazdi 1992, 44–47.25 Talwīh.āt II.4.2, 151–154 Habībī; II.4.3.1–2, 157–166 Habībī.


similarities to the more readily available phenomenon of human knowledge,one will want to focus on aspects common to the two. From this point ofview, questions of truth and falsity and the peculiarities of human cognitivepsychology are not merely irrelevant but downright counterproductive.Moreover, Suhrawardī’s criticism of Avicenna is difficult to grasp without

assuming a shift of approach of the sort I have described. As HeidrunEichner has suggested, Avicenna’s account of intellection and universalshouses a conflict of theoretical interests between logic and psychology.26 Ifintellects are confined to the exclusive apprehension of universals, as thecase is made in logic, then Avicenna will run into seemingly insurmountableproblems in his psychological account of the apprehension of particularthings by human subjects, who because of their immateriality are intellectsby definition. However, pace Eichner, I do not believe that this incoherenceis Suhrawardī’s target here, for were that the case, it would be hard to avoida rather grim assessment of his philosophical acumen. Why shouldSuhrawardī’s solution – that particulars are present to the immaterial selfby means of the presence of its corporeal faculties of cognition – be any lessobscure or problematic than that of Avicenna? In fact, isn’t his allegedsolution a mere paraphrase of the Avicennian constituents of the problem?Corporeal faculties will still be required for human awareness of particulars.But if the point is to clear conceptual room for the knowledge of particularsin absolute terms, that is, including the special case of God and departingfrom the connections between immateriality, intellection and universalityon the one hand and materiality, perception and particularity on the other,then Suhrawardī’s move appears much sounder. The concept of knowledgeas presence performs this task by focusing on the givenness or appearance ofwhat is known, entirely setting aside questions about the manner in whichthe appearance is brought about. This shared basis of givenness, appearanceor presence allows Suhrawardī to use ordinary human knowledge, with itssimultaneous apprehension of the knowing self and the objects known, as ameans of elucidating God’s knowledge.In the above, I have claimed that Suhrawardī builds his concept of

knowledge as presence on the phenomenon of self-awareness he inheritsfrom Avicenna. But how precisely is self-awareness foundational in thecrucial section [c]? Is it not simply one type of knowledge among othersthat the new concept is inclusive of? To answer this question, it is importantto notice that the enumerated types of non-existence of absence are hier-archically ordered. The self’s not being absent to itself is the basis for the

26 Eichner 2011, 130–131.


presence of its faculties to itself, and it is only by means of the faculties thatthe objects of the self’s operations can be present to it.27 A natural way ofinterpreting this is to say that my body, the corporeal faculties of cognitionand the objects of their activity are present to me by entering my experience,that is, by entering ‘the field of presence’ opened bymy self-awareness, or bymy not being absent from myself. In other words, presence only meansbeing given for a first-personal subject or appearing to a subject that exists inthe first person. I admit that this interpretation requires reading quite a bitbetween the lines of the dense section [c] above, but it helps us to see howSuhrawardī could have expected the concept of knowledge as presence toavoid the contradiction between God’s absolute unity and the multiplicityintroduced by the world of particular creatures as the object of His knowl-edge. Moreover, this aspect of the concept of knowledge as presence islaid out in a more extended fashion in a parallel chapter of theMashāri‘ waal-mut.ārah.āt.

The chapter in question is the first of the seventh mashra‘ in themetaphysical section of the book, and it is devoted to the task of explaining‘the apprehension and knowledge of the Necessary Existent and those thatare separate’. Just as in the Talwīh.āt, Suhrawardī begins by arguing againstthe identity theory of knowledge, this time applying the permanence of self-awareness in an interesting argument which we have already analysed.28 Hethen proceeds to refute in passing, referring to a discussion earlier on in thebook, the theory according to which knowledge amounts to a union withthe active intellect. Following this section, he addresses an argument for theself-awareness of any immaterial entity, which he attributes somewhatenigmatically to ‘people . . . who are stronger in investigation than’ thoseproposing the union with the active intellect. These people

have set a premise within the question of knowledge. Thus, they haveestablished a principle – that what is separate must apprehend itself (annaal-mufāraqa yajibu an yakūna mudrikan li dhātihi) –which is that whatever isunderstood has a subsisting essence (dhāt), the existence of which outside themind is like its existence in the mind, that is, separate from matter. Thus,when it is understood, its form can be connected to another intelligible in thesoul, and as a result it is understood along with another thing. Since its

27 Cf.Mashāri‘ III.7.1, 487. A parallel hierarchy is put forth in the account of God’s knowledge later onin the Talwīh. āt: God is primarily aware of (or not absent from) Himself, but this entails thesimultaneous presence of His concomitants in a descending order from the celestial intellects tothe celestial souls, and subsequently to whatever takes place under their guidance below them(Talwīh. āt III.3.2, 244–245 Habībī; 75–76 Corbin).

28 Cf. Mashāri‘ III.7.1, 474–475; and see Chapter 5.1, 111.


essence, as its form, is not surrounded by material accidents, its quiddity canbe connected to an intellectual form, and so it can be rendered understood. Ifintellection were impossible for substance – and what is impossible in thegenus of its nature is impossible in the species – self-apprehension (idrākudhātihi) would not be possible for any substance, and that is not the case.Since the intellection of an intellectual form is not impossible for it, theintellection of that form entails that it understands itself (ya‘aqila dhātahu).Thus, that which understands something is to understand that it is thatwhich understands (lahu an ya‘aqila annahu huwa alladhī ya‘aqila). Sincethis thing is such that it is actual in every respect, what is not impossible for itis not possible for it as a non-occurring possibility, rather it must havenecessity through itself (bi dhātihi) or through another in some things,such as the intellects. Thus, whatever is understood has an essence (dhāt)that is separate from matter and subsists by itself (bi nafsihā), and it under-stands itself (dhātahu) and another.29

This particular version of the argument derives from Avicenna,30 but assuch it expresses a view with a much longer history according to which theexistence proper to an intellect entails self-intellection. Suhrawardī, how-ever, does not accept the argument, based as it is on the idea of the inherenceof the known form in the knowing subject. His main problem is that Godtoo understands Himself, but in His absolute unity He cannot be a subjectin which any known form could inhere and therefore represents an excep-tion to the principle upon which the argument hinges.31 But as a follow-upto this critical assessment, Suhrawardī adds that he has no objections to theequivalence between self-cognition and immateriality. It is perfectly soundto say that since God is immaterial, He is not absent from Himself andtherefore understands Himself. However, this piece of truth must bedetached from the Avicennian concept of knowledge as the inherence ofwhat is known in the knower. Rather, God’s self-understanding is a simplefact, and although we can distinguish in analysis between the act of Hisunderstanding, His being a subject of understanding and His being theobject of understanding, no such distinction exists in reality.32

This emphasis on God’s unity is followed by a discussion of God’sknowledge of His creation. In a rather traditional manner, Suhrawardīstates that He knows particular creatures through His knowledge ofHimself, that is, as His concomitants. This is analogous to the way in

29 Mashāri‘ III.7.1, 475–476. 30 Avicenna, Ishārāt, namat. 3, 132.31 This argument derives from Rāzī, Sharh. al-Ishārāt, I.169. Suhrawardī also presents two further

counterarguments based on a distinction between mental and extramental existence. (Mashāri‘III.7.1, 476–477.)

32 Mashāri‘ III.7.1, 477.


which we, knowing the human essence, will also know its concomitants,with the difference that where our knowledge of the relevant concomitantsis potential in the sense that we can give an account of them if presentedwith the task, God’s knowledge of His concomitants is constant andactual.33 Even though Suhrawardī goes through an array of counterargu-ments to this traditional claim,34 he eventually seems to accept it, thoughnot without a qualification. Again, he emphasizes that God’s knowledge ofHis concomitants is not to be explained as the inherence of the concom-itants in Him in the manner of the inherence of accidents in a subject.35

After this analysis of the sound and unsound aspects of the alternativeconcepts of knowledge, Suhrawardī states that his own view can be read in theH· ikma al-ishrāq and that his present motive has been to show to what extentthe Peripatetics have been in the right. He adds that the next best account isgiven in the appearance of Aristotle in theTalwīh.āt, the method of which canbe summarized in the maxim ‘that the human being first investigate hisknowledge of himself (fī ‘ilmihi bi dhātihi) and then rise to what is higher’,36

and he goes on to give an extended paraphrase of it. Thus, the followingdiscussion of presence takes place in a context exactly similar to that in theTalwīh.āt; God’s knowledge is elucidated by considering the entire range ofhuman knowledge from a point of view that sets aside differences in causesrespective to each type of knowledge and focuses instead on their sharedfeatures, that is, presence and lack of absence, which are then relied on in anew account of God’s knowledge of Himself and His creation.

Suhrawardī starts as he did in the Talwīh.āt, that is, by presenting self-awareness as a type of knowledge that the Avicennian inherence theory ofknowledge cannot explain:

So we say that when our souls apprehend their self (nufūsanā idhan adrakatdhātahā), their apprehension of it is not by means of form due to [several]respects. The first of them is that the form which is in the soul is notidentical with [the soul] (laysat bi ‘aynihā hiya hiya), whereas that whichapprehends itself (al-mudriku li dhātihi) apprehends what is identical to hisI-ness (mudrikun li ‘ayni mā bihi anā’īyatuhu), not something correspondingto it, and every form, which in the apprehender is additional to its self (‘alādhātihi), is in relation to it an ‘it’, not being an ‘I’ for it, and so theapprehension is not by means of form.37

33 Mashāri‘ III.7.1, 478. Interestingly, an Avicennian version of this argument figures in the Ta‘līqāt inthe very context in which the most striking remarks about human self-awareness are made (Ta‘līqāt158–160).

34 See Mashāri‘ III.7.1, 478–480. 35 Mashāri‘ III.7.1, 480–483. 36 Mashāri‘ III.7.1, 483–484.37 Mashāri‘ III.7.1, 484.


This argument rules out the possibility that self-awareness can beexplained as an apprehension of a special object by means of the distinction,familiar from theH· ikma al-ishrāq, between a subject’s appearance to itself asan I and an object’s appearance to the subject as an it. The case is made interms that take greater liberties with its Avicennian heritage than the parallelargument in the Talwīh. āt, but the immediately following second argumentis closer to that text.

Secondly, if the soul’s apprehension of itself (idrāka al-nafsi li dhātihā) wereby means of a form, then since every form occurring in the soul is universal,its correspondence to many is not impossible, and even if one takes acombination of universals, the whole of which is exclusive to a singleindividual among souls, one will not do away with its being universal.Every human apprehends himself (yudriku dhātahu) in a manner whichprevents sharing, and so his intellection of his particular self (ta‘aqquluhu lidhātihi al-juz’īya) cannot be by any form whatsoever.38

The crux of the question is familiar from section [b] of the appearance ofAristotle. Since the human subject is immaterial, any form inhering in it,including the form that is alleged to make the subject aware of itself, mustbe universal. This extrapolation of the Avicennian equivalence betweenimmateriality and intellectuality, however, violates our intuition accordingto which the self each of us is aware of is a unique individual. Moreover, itmakes impossible our exclusive operative access to our bodies and thecognitive faculties that rely on corporeal organs:

Furthermore, the soul apprehends its body, and it apprehends its estimationand imagery, and if it apprehended these things by means of a form in itself(fī dhātihā) – that form being universal – the soul would move a universalbody and operate universal faculties, and it would have neither apprehensionof its body nor apprehension of the faculties of its body. This is not correct –how could it be when estimation ignores itself (nafsahu) and also ignores theinternal faculties! In that case it would not reject their traces. Thus, sinceestimation does not apprehend these faculties, since none of the corporealfaculties apprehend themselves (nafsahu), and since the soul apprehendsnothing but universals, the human must not apprehend his body, hisestimation and his imagery which are exclusive to him as particulars. Thatis not the case, for he is not human who does not apprehend his particularpresent body and his particular present faculties and operate his particularfaculties. Thus the human does not apprehend himself (mudrikun li nafsihi)by means of a form, nor does he apprehend his faculties as a whole by meansof a form, nor his body as a whole by means of a form.39

38 Mashāri‘ III.7.1, 484. 39 Mashāri‘ III.7.1, 484–485; emphases added.


In comparison with the corresponding passage in the Talwīh.āt, this textemphasizes that the basis of the body’s presence is in the prior self-awarenessof the soul that acts in and through the body. Nothing in the body is capableof apprehending or making present the body or anything in it, not evenestimation, the highest cognitive faculty operative in the body. Were it notfor the soul that is aware of itself, the body could never be present toanything, precisely because presence requires a subject that is already awareof herself and to which other things can therefore be present. In otherwords, the body enters the soul’s experience because that experience belongsto a subject that is aware of herself to begin with.

But importantly, it is the body itself that is present to the soul, and onlytherefore can it be present as my particular body, the locus of operation formy particular faculties. If the body was present only through the mediationof a representative form inhering in me, the inhering form would of coursebe mine in the same sense, but because of my immateriality and theconsequent universality of whatever is supposed to inhere in me, the formwould not be uniquely representative of a single body. Thus, only theknowledge of the body and the acts taking place in it would be mine inthis strong sense, but this would exclude neither the possibility that what Iknow is applicable to other bodies nor the possibility that a single body isknown in the same respect by two different subjects.

This aspect of the concept of knowledge as presence, namely that theknown thing itself can be present to the knower without the mediation ofinhering representations, is particularly prominent in Suhrawardī’s discus-sion of the phenomenon of pain.

What confirms that we have apprehensions, in which there is no need for anyform other than the presence of [the thing] itself that is apprehended (h. ud

˙ūri

dhāti al-mudrak), is that the human being suffers pain because the cohesionis impaired in an organ of his, and he is aware of it, not because theimpairment of the cohesion occurs for him as another form in that organor in another; rather, what is apprehended is identical with that impairment(nafsu dhālika al-tafarruq), it is sensed and pain by itself (bi dhātihi), not bymeans of a form occurring from it. It has thus been indicated that there aresuch apprehended things that the occurrence of [the thing] itself (h. us.ūludhātihā) to the soul, or to something with a proper presential connection tothe soul, is sufficient for apprehension.40

When we feel pain, we do not feel an effect of a cause that does not appearbut ‘the thing itself’. Pain is not a representation of a wound, a lesion or

40 Mashāri‘ III.7.1, 485.


anything like that, but rather the very wound itself as it is present throughthe faculty of touch. Of course, I can later on see the wound and infer thatwhat I see is causing the pain, but I should not draw the conclusiontherefrom that what is seen is in any sense prior to or more real thanthe pain. The two appearances, the wound seen and the pain felt, can beidentified because both are present to the same subject, not because thewound that is seen causes the pain.Importantly, Suhrawardī does not consider pain to be an anomaly in

perception. Rather, owing to its pronounced immediacy pain provides abrilliant example of a fact true of all perception: perception is not consti-tuted by the occurrence of forms that represent extramental objects ofperception but is rather to be conceived as the entrance of the very thingsperceived through the soul’s faculties of perception into the field of presencethat the soul is in itself. This is summarized in Suhrawardī’s concludingremark on the various types of human knowledge:

What the party of Peripatetics has to admit is this: they allow that the formmay occur in the instrument of sight without the human being aware of it –when he is immersed in his thought or in what another sense is bringingforth – and so there is no doubt about the soul’s attention (iltihāt) to thatform, and apprehension is only by means of the soul’s attention to what itsees as being looked at (al-idrāku laysa illā bi iltifāti al-nafsi ‘inda mā tarāmushāhadatan), and what is being looked at is not by means of a universalform, rather what is being looked at is by means of a particular form, and sothere is no doubt that if the soul has presential illuminational knowledge(‘ilmun ishrāqīyun h. ud

˙ūrī), it is not by means of a form.41

It is not sufficient for perception to arise that a physical process takes placein the organ of perception. The process has to be attended by the soul, or ithas to be present to the soul. In other words, what goes on in the organ ofperception has to enter the field of presence constituted by the soul, that is,its experience, in the manner proper to the particular faculty in whose organthe process takes place. In itself, the soul is nothing but this field ofpresence, this first-personal experience, which is entered by the variouscontents that are thereby made present.When Suhrawardī turns to elucidate God’s knowledge by means of these

observations about human experience, it becomes obvious that his point isnot to present a natural philosopher’s account of how knowledge comes tobe. Rather, the concept of knowledge as presence is an attempt at spellingout what knowledge is when it is understood in exclusively experiential

41 Mashāri‘ III.7.1, 485; emphasis added.


terms, as an appearance to a subject of experience. It is in this specific sensethat Suhrawardī uses human knowledge as a means to elucidate his theo-logical subject. Self-awareness is foundational for this purpose preciselybecause it provides the basis on which other cognitive content can becomepresent without the presence of that content in any way compromising theconstancy, stability and unity of the self and its awareness of itself. God’sknowledge is defined as

illuminational knowledge (al-‘ilmu al-ishrāqī) without a form or a trace butmerely by means of an individual relation which is the illuminationalpresence of the thing (h. ud

˙ūru al-shay’i h. ud

˙ūran ishrāqīyan), like for the soul,

but in the Necessary Existent it is more primary and more complete.42

And because of this greater completeness, God

apprehends Himself (dhātahu) not by means of anything additional toHimself (dhātihi) – as came up in the soul’s case – and He knows thingsby presential illuminational knowledge (bi al-‘ilmi al-ishrāqīyi al-h. ud

˙ūrī).43

Thus, the only difference between knowledge as presence as found in Godand in human beings is due to how the presence of other things is broughtabout; human subjects will passively receive their content from what isgenuinely other through their faculties, but God will have His as His ownconcomitants, just as is proper for a supreme agent. But this difference is allbut irrelevant here, for the reason why the kind of presence we find inourselves can assist in elucidating the presence of God is that it is similar to itin some fundamental respect.

Suhrawardī is in considerable debt to Avicenna in this theological appli-cation of the phenomenon of self-awareness. Not only does he derive thenarrow concept, particularly fitting for the purpose, of self-awareness fromhis predecessor, but quite possibly he was also inspired to apply it in thisparticular question by the context in which Avicenna’s excursion into itwas embedded in the Ta‘līqāt. As we have seen, Avicenna may have beenmotivated to make some of his most striking claims about self-awareness bythe task of accounting for the absolute unity and immediacy of God’sknowledge of Himself, for which he may have found a simile in humanself-awareness.44 But in Suhrawardī, human experience, and its basis in self-awareness, is used to make sense not only of God’s knowledge of Himselfbut also of His creation. The human case can be cast in this explanatory roleonly if we disregard the psychological approach and focus on knowledge as a

42 Mashāri‘ III.7.1, 487; emphasis added. 43 Mashāri‘ III.7.1, 487. 44 See Chapter 3.2, 56–61.


given matter of experience, that is, as presence or the non-existence ofabsence. As the paradigmatic case of the latter, human self-awareness givesus an idea, as imperfect as it may be, of what immediate knowledge is like atits most intense. But it also enables Suhrawardī to explain in what sense itcan be both one with and distinct from the presence of other cognitivecontent. By not being absent from itself as the first-personal field in whichother things can become present, the self is there in immediate relation tobut also independent of the content of its experience. At the same time, thatcontent owes its presence to the self-aware subject for which it appearsbecause, were it not for the subject as the field that enables presence, nocontent would be present in the first place. Furthermore, considered merelyas such a field, the subject persists unaltered by any of the content thatenters and leaves the field.45 It will be an I in an immutable sense regardlessof what attributions the I will be connected to.In this sense, Suhrawardī can claim that God knows particular things

as particulars, without the Avicennian qualification ‘in a universal way’,because as the same sort of first-personal field for the presence of objects ashuman subjects are, He is not changed by the entrance of particular thingsinto His knowledge. Moreover,

He has absolute illumination and dominion, and nothing escapes from Him,and past and future things – the forms of which are established by the celestialdirectors – are present to Him because He encompasses and illuminates thesubstrate of those forms, and the same holds of intellectual origins.46

God is aware of the universe in a hierarchical continuity, mediated by thecelestial intellects and souls that are the causes of the sublunary realm. But thisneed not confine Him to cognition in universal terms, for the mediation ofthe celestial intellects and souls can be conceived in the same manner as themediation of the human soul’s faculties in its knowledge of objects externalto it. In spite of the mediation, the very individual things are brought into thefield of presence that is God’s knowledge. By the same token, God retains Hisabsolute unity as the pure non-existence of absence, because the qualificationsdue to the attribute of knowledge do not affect that lack of absence in any way.God remains Himself, because ‘His apprehension of Himself is His life’ as thelight in which created things come to be.47

If the aforesaid is a correct interpretation of Suhrawardī’s concept ofknowledge as presence and the decisive role of self-awareness in cancelling itout, his so called ‘Peripatetic’ works seem remarkably consistent with the

45 Mashāri‘ III.7.1, 488. 46 Mashāri‘ III.7.1, 488. 47 Mashāri‘ III.7.1, 489.


full-blown illuminationism of the H· ikma al-ishrāq.48 In the passages wehave considered, Suhrawardī has described how the self’s not being absentfrom itself, its self-awareness, provides the foundation for its cognitiverelation to its body and by means of the body to the world. The body andthe world, in both particular and universal terms, can be present to the selfbecause the self is not absent from itself. If this relation suffers comparableneglect in the H· ikma al-ishrāq, this is merely due to the fact thatSuhrawardī’s main concern in that work is to make sense of the metaphys-ical counterpart of knowledge as presence, that is, of appearance (z· uhūr).

6.2 Self-awareness and being as appearance

The H· ikma al-ishrāq begins with a dense but substantial criticism of thePeripatetic theory of science and metaphysics. When Suhrawardī’s subse-quent account of the philosophical system he presents as its alternative isread in the light of this critical introduction, it is hard not to receive theimpression that he intends to replace the intimately connected Peripateticconcepts of substance and definition with the new concepts of light (nūr)and appearance (z· uhūr). As we will see, Suhrawardī’s strategy of defininglight and appearance in a manner that is independent of these Peripateticfoundational concepts is heavily reliant on the phenomenon of self-awareness. But in order to understand the introduction of the new concept,we have to describe briefly how Suhrawardī paves the way for it by criticiz-ing the Peripatetic foundations of being and knowledge.

In his discussion of the concept of substance, or substantiality, Suhrawardīsets out from the Peripatetic definition as transmitted by Avicenna. Accordingto this received view, ‘“substance” is what does not exist in a subject, regardlessof whether it is altogether independent of a substrate or inheres [in asubstrate], but the substrate is not independent of it like “forms”’.49 Thisdefinition from the Talwīh.āt contrasts substance with the accident (al-‘arad

˙)

that exists in a subject. It also hints at the traditional division between differenttypes of substances, that is, matter, form and their composite. The corre-sponding definition in the H· ikma al-ishrāq adds a reference to extramentalreality; according to it, substance is ‘something that has an existence outsidethe mind’ and that ‘does not inhere in another so that it is entirely diffused’.50

This definition distinguishes substance from a state (hay’a) that existsin another, adding the qualification ‘so that it is entirely diffused’ (‘alā sabīli

48 Pace Eichner 2011, 139–140. 49 Talwīh.āt III.1, 177 Habībī; 5–6 Corbin.50 H· I I.3.3.52, 42 Walbridge and Ziai; 61 Corbin.


al-shuyū‘i bi al-kullīya) in order to distinguish states from independentlysubsisting parts of a whole, which can nevertheless be said to be inthe whole and thereby to exist in another. In light of this definition ofsubstance, Suhrawardī detects four different kinds of substances in theworld: three-dimensional substances with unique spatiotemporal locations,that is, bodies,51 the two constituents of bodies, that is, matter and form, andsubstances that are separate from matter, that is, intellects.52

For Suhrawardī, the problem with both of these definitions of substanceis that they hinge upon negation: substance is that which does not exist in asubject. Now, negations do not refer to anything real immediately; there isno non-being as such. Instead, their truth requires a mind or an intellectwhich is capable of conceiving something that exists and considering it inrelation to something else that is also conceived by the mind in somemanner but that does not exist in the reality without the mind. When themind considers these two things properly related, it then proceeds to denyor negate the non-existing of the existing thing. Because negations, anddefinitions based on the use of negation, are dependent on a mind or anintellect, Suhrawardī calls them mental or intellectual considerations(i‘tibārāt dhihnīya, i‘tibārāt ‘aqlīya).53 If the concept of substance cannotbe defined positively, it will be a mere intellectual consideration, and as aresult it will not refer immediately to anything real. And if substancesdepend for their very existence on a mind that postulates them, they cannotpossibly provide a foundation for metaphysics as that which is said to be orto exist in a primary sense.Suhrawardī presents a sustained argument for his claim against the reality

of mental considerations. If we postulate something like substantiality thatis independent of the mind and as such the basis of all being, we end up withan infinite regress of substantialities. Suppose that substantiality1 is basic tothe existence of the tree in front of me. If substantiality is a real basis of allthat exists, then substantiality1 must have an ontologically prior substan-tiality2 at its basis, and so forth ad infinitum. Barring the possibility ofan ordered infinite series in act, Suhrawardī suggests that we give upthe supposition of substantiality as metaphysically foundational. The infin-ite regress argument is Suhrawardī’s frequent strategy to refute the

51 For this characterization of bodies, see H· I I.3.3.53, 43 Walbridge and Ziai; 62 Corbin.52 Cf. Talwīh. āt III.1, 177–178Habībī; 6 Corbin. This division is also the basis for Avicenna’s discussion

of substance in Shifā’: al-Ilāhīyāt II, see especially II.1.10, 48.53 Cf. H· I I.3.3.68, 50–51Walbridge and Ziai; 71–72 Corbin; and Talwīh.āt III.2, 193, 196Habībī; 22, 26

Corbin.

Self-awareness and being as appearance 143

mind-independent reality of mental considerations.54 In the Talwīh. āt heconcludes his discussion of them by stating it as a general principle: any suchthing, the supposition of which as a real entity leads into infinite regress,does not exist really but is a mere consideration of the mind or theintellect.55

Corresponding to his critique of the notion of substance, Suhrawardīfinds fundamental problems with its epistemological counterpart, the con-cept of definition as the foundation of knowledge. He defines definition inrather traditional terms, though not without certain important qualifica-tions. At the beginning of the section on expository propositions in theTalwīh.āt, the complete definition is said to be ‘a statement (al-qawl) whichindicates the quiddity of something and combines all of its constituents, andin the case of principal realities (al-h. aqā’iqi al-as.alīya) [that is, substances], itis composed of their genera and differentiae’.56 Comparison with corre-sponding definitions in Avicenna shows that Suhrawardī has subtly addedthe word ‘all’ (kulluhā) where Avicenna merely speaks of the proximategenus and the differentia of the thing to be defined.57 Suhrawardī’s strongerrequirement gives natural rise to the question of whether such stringentconditions can ever be met and whether we are likely to acquire completedefinitions of anything. In the Talwīh.āt, these consequences remainimplicit, but Suhrawardī’s explicit criticism of the concept of completedefinition suggests that that is the direction he is heading towards.58

The crux of the critique becomes explicit in H· ikma al-ishrāq I.1.7, whereSuhrawardī bluntly states that we can never be absolutely certain of havingarrived at a complete definition which collects together all essential charac-teristics of the definiendum. To put the same in extensional terms, we cannever be sure that our definition picks those and only those individuals thatinstantiate the essence we are interested in and that this collection is due tothose and only those of their properties that really are essential and notmerely accidental or concomitant to the essential properties.59 Related tothis point, Suhrawardī also criticizes the Peripatetic method of arriving at adefinition by means of an inductive inference. In order to be certain,

54 For an extended use of this argument in relation to various mental considerations, includingsubstantiality, see H· I I.3.3.56–68; 45–51 Walbridge and Ziai; 65–72 Corbin.

55 Talwīh.āt III.2, 196 Habībī; 26 Corbin.56 Talwīh.āt I.2.1, 18 Habībī. For discussion, see Ziai 1990, 100.57 See Ziai 1990, 100–102, who contrasts the Talwīh.āt formulation with Avicenna’s Kitāb al-h. udūd and

Shifā’: al-Burhān.58 Ziai 1990, 113–114.59 H· I I.1.7.13–15, 8–11 Walbridge and Ziai; 18–21 Corbin. For Suhrawardī’s debt to Abū al-Barakāt al-

Baghdādī in this regard, see Ziai 1990, 115–116.


induction must encompass all relevant cases, but since it is not possible tocover every individual instantiation of a species occurring in an infinitecourse of time, we can never be certain of not having overlooked an excep-tional case that would force us to qualify our inductive definition.60

The definition of definition cited from the Talwīh.āt makes anothersubtle point by means of the phrase ‘is composed of’ (yutarakkaba), whichis explicated in the Mashāri‘ wa al-mut.ārah. āt. In order that a completedefinition be true and thus capable of producing knowledge of the thingdefined, it must correspond to that thing. However, the external thing isone, an absolute unity with no independently subsisting constituents. Thisis not the case with the definition: it consists of the really distinct conceptualunits of genera and differentiae, which are merely conjoined to each other inan act of composition, and this is incapable of bringing about a degree ofunity corresponding to that of the external thing we strive to define. In viewof this incoherence between absolute unity and unity through the compo-sition of discrete parts, the best our definitions are capable of is describingthe definienda.61

What is more, definitions will always rely on constituents that must beknown before they can be applied in a definition. Since we cannot proceedinfinitely with steps that define the terms used in subsequent definitions,we must admit that at least some things, such as the first intelligibles, mustbe known without definitions.62 Thus, our knowledge is not ultimatelyfounded upon definition,63 and in the final analysis nothing will be knownif we subscribe to a concept of knowledge based on definition.Suhrawardī’s intention is not to give up the notion of definition as

something utterly useless in the pursuit of knowledge, but his criticismdoes amount to a straightforward rejection of definition as the foundation ofknowledge. From here on, definitions come to serve in an instrumentalrole, perhaps a pedagogical one, in the acquisition of knowledge. Once thisis understood, they should be conceived in conceptualist terms: theydefine linguistic concepts, but, in order to be capable of this, they require

60 H· I I.1.7.15, 10–11Walbridge and Ziai; 20–21 Corbin. This problem of induction was seized upon byAvicenna, who developed a systematic notion of ‘experience’ (tajriba) as its alternative (see McGinnis2003), a suggestion with which Suhrawardī unfortunately does not engage. For empirical premises inSuhrawardī, see Talwīh.āt I.3.3, 28–30 Habībī; H· I II.2.7.30, 27 Walbridge and Ziai; 41–42 Corbin;and Ziai 1990, 54.

61 Mashāri‘ I.2.2, as edited in Ziai 1993, 115–116. For discussion, see Ziai 1990, 104, 108, 110–111.62 See Ziai 1990, 137–138. Related to this point, Suhrawardīmakes a fourfold distinction between innate

and acquired conception (tas.awwur) and assent (tas.dīq) in Talwīh.āt I.1.1, 4–5 Habībī.63 Cf. H· I I.1.7.14–15, 9–11 Walbridge and Ziai; 19–21 Corbin. See Ziai 1990, 116–117, 120–121.


the full presence to the defining mind of that to which the relevant conceptis applicable.64 In this function even ostensive definitions, which are notconcerned with grasping the essential properties or causes of the thingdefined, may serve a perfectly legitimate purpose, although they are ofcourse not sufficient in their own right to provide knowledge of the thingdefined.65

But how do we grasp the thing our definitions can merely describe?Whatis primarily known in Suhrawardī’s epistemological alternative? By the sametoken, if substances are not real, what is the primary sense in which being isspoken of?

Suhrawardī builds his alternative philosophical system quasi-axiomaticallyby bluntly stating, first, his conception of what is primarily known, andsecond, his conception of what should be said to primarily exist. Let usconsider the two axioms in some detail.

If there is in existence that which does not need to be defined or elucidated,it is that which appears (al-z· āhir). Nothing is more apparent than light(al-nūr). Thus, nothing is more independent of definition.66

Notice that this introduction of the illuminationist technical term ‘light’describes it solely in terms of appearing or being evident. All that is said isthat light appears and is evident before and in spite of anything else, andas a result knowledge about it cannot be acquired through a definition orelucidation by means of other things that are known prior to it. Becauseof the opacity of Suhrawardī’s dense formulation, it is worth spelling outthe rather obvious point that what he means by the most apparent thingcan hardly be the quiddity of the optical phenomenon we usually speak ofas light. Were that his intention, the very fact that the quiddity of lightwas a point of serious contention would have been a sufficient cause tojudge his statement philosophically unsatisfactory.67 If we want to inter-pret him in philosophically charitable terms instead, we should stick tohis words: in this context, the primary meaning of ‘light’ is the fact ofappearing, that which appears insofar as it appears. Although the termwill be described at greater length in the immediately following chapters,it is important to notice that the term is used in a very precise and rathernarrow technical sense.

64 Mashāri‘ I.2.2, as edited in Ziai 1993, 115–116. See Ziai 1990, 108, 110–111, 125.65 Mashāri‘ I.2.5–6, as edited in Ziai 1993, 118–119.66 H· I II.1.1.107, 76 Walbridge and Ziai; 106 Corbin.67 Cf. Avicenna’s discussion of light in Shifā’: Fī al-nafs III.1–4, 91–115 Rahman. For an excellent

overview of the theory, see Hasse 2000, 108–113.


The second axiom concerns independence in existence:

The independent (al-ghanī) is such that neither its essence (dhātuhu) nor itsperfection is due to another. The dependent (al-faqīr) is such that somethingof it is due to other than its essence and perfection.68

If the first axiom addressed the question of what is primarily known, itsfollow-up makes an intimately connected point about the foundation ofmetaphysics: what should be said to exist in the primary sense of the word isthat which is not dependent in any manner on anything other than itself.The subsequent exposition of the relation of the concepts of light (thatwhich is most apparent) and of darkness (that which does not appear assuch) makes it clear that the two points are inseparable: that which is mostapparent and primarily known is also the only thing that can be said to existindependently, and therefore it is that which exists in the primary sense ofthe word. Prime examples of dark things are the material substances that wesuppose to subsist out there in the world. Suhrawardī denies them of anypositive reality; corresponding to the negativity at the core of the definitionof substantiality, the darkness of material substances merely amounts to alack or a privation of light. The only positive appearance they can have iswhen an adventitious light (al-nūr al-‘ārid

˙) falls upon them andmakes them

appear for another.69 Thus deprived of any positive reality, it is clear thatmaterial substances cannot be said to exist in the primary sense of the word.But the adventitious light that renders them apparent is also dependent onanother, in fact doubly so, for it not only requires the other in which itadvenes, it also must have a cause that is separate light (al-nūr al-mujarrad)or pure light (al-nūr al-mah. d

˙), pure appearance in and for itself.70

This discussion was pertinent to the seventeenth-century debate over the‘primacy of quiddity’ (as.āla al-māhīya) as opposed to the ‘primacy ofexistence’ (as.āla al-wujūd).

71 It should be pointed out that although appear-ances are always appearances of essences, Suhrawardī’s adoption of theconcept of appearance as the foundation of metaphysics does not amountto a flat denial of the reality of existence. What is under attack is thePeripatetic view according to which substantiality is the paradigmaticmode of existence. The point comes out in the way in which the continuity

68 H· I II.1.2.108, 76 Walbridge and Ziai; 107 Corbin.69 Suhrawardī refers to these encounters between dark substances and adventitious lights as ‘barriers’

(barāzikh), presumably because they stand at the border of complete non-appearance andappearance.

70 H· I II.1.3.109, 77–78 Walbridge and Ziai; 107–109 Corbin.71 For a concise outline of the debate, see Bonmariage 2007, 30–52.


and identity of individual things is explained, for, according to Suhrawardī,the continuous existence of things such as the pine outside my windowshould be conceived not in terms of an underlying material substance thatpersists intact through the constant change of its accidental attributesbut rather in terms of an essential unity due to the persistent cause of thestream of the pine’s constantly changing appearance. It was clear to both histhirteenth-century commentator Qutb al-Dīn al-Shīrāzī (d. 1311) and MullāS· adrā, perhaps the most prominent voice in the debate, that Suhrawardīsimply speaks of existence as appearance; the stream of appearance isnothing but existence.72

In the fourth chapter, Suhrawardī emphasizes the dependence of allappearance on pure appearance by means of an anti-Avicennian pointconcerning the existence of individual entities. He states that their individ-uality is due to their apparent properties, and since these are real properties,they can only be due to light, which alone is real. As something inherentlyunreal, matter cannot provide the basis for individuality, and by the sametoken there is no individual substantiality beyond the appearances, any-thing like that is a mere intellectual consideration.73 The two principles inwhich Suhrawardī concludes the axiomatic introduction to his new systemreiterate the dependence of adventitious lights, or appearances for another,on pure lights, or appearances in and for themselves. They also state that,unlike adventitious lights, pure lights are not accessible to a spatio-temporalostensive reference.74

At this point, pure light or pure appearance has been axiomatically laidout as the basis of Suhrawardī’s new illuminationist metaphysics. Light ismetaphysically foundational, and since pure light is appearance in and foritself, it is also immediately and primarily known. The Platonic overtones ofall this are clear, but it can still be legitimately asked whether the concept ofpure light or appearance is anything more than a play of words, a mentalfiction every bit as fanciful and unreal as the concept of substance it isdesigned to replace. Moreover, in both of the axioms cited above, the crucialconcept is characterized in strikingly negative terms: what is most apparentis not in need of definition or elucidation, and what exists independentlyis not due to another. If there is no way of making positive sense of these newconcepts, Suhrawardī’s new theory only amounts to so much heated air.

72 See Qutb al-Dīn al-Shīrāzī, Sharh. H· ikma al-ishrāq, ad II.1.1.107, 283–284; and S· adrā, al-Ta‘līqāt ‘alāSharh. H· ikma al-ishrāq, ad II.1.1.107, 283. Ziai 1990, 166–171, seems to agree, but for a voice of concernsee Walbridge 1992, 49–55.

73 H· I II.1.4.111, 78–79 Walbridge and Ziai; 109–110 Corbin.74 H· I II.1.4.112–113, 79 Walbridge and Ziai; 110 Corbin.


It seems that Suhrawardī is fully aware of his situation, for the decisivechapter II.1.5 considers the newly introduced concept of light from acompletely different perspective:

Whatever has a self which it is not unaware of (dhātun lā yaghfulu ‘anhā) is notdusky (ghāsiq) because of the appearance of itself to it (li z· uhūri dhātihi ‘indahu).Nor is it a dark state in another, for even a luminous state is not light for itself(li dhātihā), let alone a dark [state]. Thus, it is a separate pure light whichcannot be pointed at (nūrun mah.d

˙un mujarradun lā yushāru ilayhi).75

The basis of the new approach is signalled in the very first sentence. Inspite of the doubly negative expression we are now suddenly dealingwith a positive phenomenon, that of appearance to self or self-awareness.Through a rhetorical process of excluding the alternatives, Suhrawardī statesthat whatever is self-aware is an instance of what he calls separate or purelight. Thus, after the axioms related to the concept of light discussed above,this new beginning can be characterized as an attempt at an ostensiveelucidation of the new concept. That this is Suhrawardī’s strategy is corro-borated by this brief elucidation being immediately followed by a series offamiliar arguments for the immediacy and irreducibility of self-awareness.The first of these is the argument for immediacy by the rejection of anyobject corresponding to the self as well as the distinction between the twotypes of awareness of I and it.76 Added to this is an emphatic insistence onthe constancy and continuity of self-awareness, a related denial of theidentification of the self with the human body we find it to govern, and adenial of the substantiality of the self.77 This suggests that Suhrawardī’sconcern is to find a commonplace phenomenon to which his newly intro-duced concept of light can refer and thus to find plausible support for hisrevisionist metaphysics, which henceforth has been introduced in an exclu-sively axiomatic manner.Suhrawardī sums up these elucidations in a new principle concerning

light:

[L]ight is that which appears in the reality of itself (al-z· āhiru fī h. aqīqatinafsihi) and which makes another appear by itself (al-muz· hiru li ghayrihi bidhātihi),78 and it is more apparent in itself (az· haru fī nafsihi) than all [that] to

75 H· I II.1.5.114, 79Walbridge and Ziai; 110–111 Corbin. Cf. the summary reiteration of this principle inII.1.5.120, 83 Walbridge and Ziai; 116 Corbin.

76 H· I II.1.5.115, 80 Walbridge and Ziai; 111 Corbin. Cf. Chapter 5.1, 111–113.77 H· I II.1.5.116, 80–81 Walbridge and Ziai; 112–113 Corbin. Cf. Chapter 5.2, 121–123.78 An alternative would be al-maz· haru li ghayrihi bi dhātihi, ‘appearance for another by itself’. This

would make sense in relation to the reference (omitted in the present quote) to adventitious lights,which appear to another but are light or appearance in themselves. However, I have chosen to follow


whose reality appearance is additional . . . It is not so that light occurs and isthen accompanied (yalzamuhu) by appearance, so that it would not be lightin the definition of itself (nafsihi) and would be made apparent by anotherthing. Rather, it is that which appears, and its appearance is its lightness(nūrīyatuhu).79

According to this principle, light is nothing but appearing. Appearing is notan additional feature of light, not even a concomitant characteristic, but thevery thing that the light’s being light amounts to.What is interesting, just asin the above, is that this claim is elucidated by the example of human self-awareness and the reiterated thesis that there is no substance behindself-awareness to which the phenomenon could be attributed as either anaccidental or a concomitant feature.80 As a conclusion, Suhrawardī statesthat ‘whatever apprehends itself (kullu man adraka dhātahu) is a pure light,and every pure light appears to itself and apprehends itself (z· āhirunli dhātihi wa mudrikun li dhātihi)’.81

This reveals Suhrawardī’s motive for rejecting the Avicennian inferencefrom self-awareness to a self-aware substance. If self-awareness belonged to asubstance, it could not be applied in an ostensive elucidation of a novelmetaphysical concept designed precisely to replace the very concept ofsubstance. In order to use the phenomenon as the basis of his metaphysics,Suhrawardī has to render it primitive and denuded of all prior metaphysicalcommitments. A related abstractive operation is performed in relation tothe claim, also familiar from Avicenna, according to which immediateself-awareness is concomitant to the incorporeality of an incorporeal sub-stance.82 Suhrawardī’s point, supported by a somewhat suspicious argu-ment according to which this claim entails the self-awareness of primematter,83 is clearly to argue that as self-awareness, pure light or pureappearance cannot be reduced to or explained by means of any otherallegedly more foundational concepts. And if that is the case, he shouldbe warranted in adopting the phenomenon of self-awareness and the con-cept of light or appearance as the foundation of his metaphysics.

the reading preferred both by Corbin and by Walbridge and Ziai, which emphasizes the activity ofpure lights in all appearances to another. Since the two readings are systematically compatible witheach other, it is of course possible that Suhrawardī is purposefully ambiguous here.

79 H· I II.1.5.117, 81 Walbridge and Ziai; 113–114 Corbin.80 H· I II.1.5.118, 81–82 Walbridge and Ziai; 114 Corbin.81 H· I II.1.5.118, 82 Walbridge and Ziai; 114 Corbin.82 H· I II.1.5.119, 82–83 Walbridge and Ziai; 114–116 Corbin.83 This is because prime matter, divested from the most elementary form of body in general, is

incorporeal. If self-awareness is a concomitant of incorporeality, prime matter too will be aware ofitself – about as implausible a claim as one can make in a Peripatetic context.


But if the concept of light or appearance is supposed to be metaphysicallyfoundational, a mere argument that there is such a thing as pure light willnot suffice. In addition to bringing forth human self-awareness as aninstance of pure light, one must also be able to explain how the concretethings in the world, whichmost certainly are not instances of self-awareness,can nevertheless be founded upon it. Suhrawardī tackles this task in aninvestigation into the different types of lights, which he embarks uponimmediately after having laid out the foundation described above. Thisdense section contains in a nutshell Suhrawardī’s account of the world ofobjects that we perceive around us.Suhrawardī distinguishes between two types of light or appearance, light

that is for itself (nūr li nafsihi) and light that is for another (nūr li ghayrihi),84

stating that light for another is based on themetaphysically prior light for itself.The two lights are hierarchically related for two reasons. First of all, the otherto which the light for another appears has to appear to itself or be aware of itselfin order to be able to apprehend the appearance of a light distinct from itself.Suhrawardī is unambiguous about this requirement: ‘What has no awarenessof itself is not aware of another (lā yash‘uru bi ghayrihi man lā shu‘ūra lahu bidhātihi).’85 But more importantly, nothing can appear to another unless it firstappears to itself. In other words, the light appearing to another is caused by alight that appears to itself, or, formulated in a more accurate way, it is theexternal appearance of something that is pure appearance in and for itself.86

While this may sound unsettling, the historically informed reader will findit arguably less awkward once she perceives the Platonic underpinnings ofSuhrawardī’s metaphysics. These are laid out in the second maqāla of themetaphysical section of theH· ikma al-ishrāq. The multiplicity manifest in theuniverse in spite of its absolutely one source is explained in chapter II.2.9 asdue to a series of complex refractions and reflections of the single pure light ofall lights.87 Towards the end of this account Suhrawardī mentions the lightsthat cause the appearance to another of concrete composite and single things.These ‘dominating luminous species’ (al-anwā‘ al-nūrīya al-qāhira), whichsubsist by themselves in the world of pure light,88 are the formal causes ofconcrete things, but unlike the Peripatetic material forms endorsed by

84 H· I II.1.6.121, 83–84 Walbridge and Ziai; 117–118 Corbin.85 H· I II.1.6.121, 84 Walbridge and Ziai; 117 Corbin. Cf. II.1.6.122, 84 Walbridge and Ziai; 118 Corbin:

‘When a thing makes something appear to another, that other must appear to itself before anythingelse can appear to it.’

86 H· I II.1.8.128, 86 Walbridge and Ziai; 120 Corbin.87 H· I II.2.9.150–152, 99–101 Walbridge and Ziai; 138–143 Corbin.88 H· I II.2.9.153, 101–102 Walbridge and Ziai; 143–144 Corbin.


Avicenna, these forms subsist independently of their appearances in con-crete.89 Each of them ‘is universal, not in the sense that it is predicated, butin the sense that it is equal in terms of the relation of emanation to thesenumerically many individuals (al-a‘dād), as though it were both the whole(al-kull) and the principle (al-as.al). This universal is not such that theconception of its meaning does not prevent the occurrence of sharing, forthey recognize that it has an individual self and knows itself (lahu dhātanmutakhas.s.as.atan wa huwa ‘ālimun bi dhātihi).’90

In a word, then, the pure lights or appearances behind the lights or appear-ances for another are Platonic forms,which account for the identity and stabilityof concrete appearances by being their immediate causes in a downwardemanative process of illumination.91 Given the constancy of the emanativeagency of the pure lights or specific forms, Suhrawardīno longer needs to rely onthe supposition of a substance behind the external appearance, a substance thatcan never appear as such and that cannot even be positively defined, since theappearance itself, being the appearance of a self-subsistent form, is that whichguarantees its identity. On the other hand, the self-subsistent form that isdistinct from us cannot appear to us as it appears to itself; our sole access to itis by way of its individual instantiations, its appearances to another. Repeatedencounters with several such instantiations will of course allow us to formulate auniversal concept of the form, expressed in a definition, but this concept is notthe same as the form’s appearance in itself to itself. We can grasp the commonfactor between the distinct individual instantiations only in an intellectualconsideration that subsumes them under a single concept which they all areasserted to have a share in.This aspect of sharedness betweenmanyor common-ness to many, indispensable for universality, must therefore be added as anintellectual consideration. But what is based on an act of the mind can bereductively explained by recourse to the mind and is therefore neither meta-physically nor epistemically primitive. The appearance itself that was subsumedunder the universal, on the other hand, cannot be further analysed or explained,and in this sense it is foundational in both senses; in the metaphysical sensebecause it is due to nothing but appearance, and in the epistemic sense becauseit is the very thing known, the form itself in its appearance to another.

89 H· I II.2.12.168–169, 109 Walbridge and Ziai; 159–161 Corbin.90 H· I II.2.12.169, 109 Walbridge and Ziai; 160 Corbin.91 H· I II.3.4.193, 123Walbridge and Ziai; 186 Corbin. The different properties of concrete things are due

to the simultaneous agency of several pure lights (H· I II.2.11.172, 111Walbridge and Ziai; 165 Corbin).On the hierarchical structure of the emanated pure lights, seeH· I II.3.3.182–183, 118–119Walbridge andZiai; 177–179 Corbin. For an extended discussion of Suhrawardī’s subscription to the Platonic theoryof forms, see Arnzen 2011, 135–143.


Thus, here we also encounter Suhrawardī’s alternative to the epistemo-logical concept of definition. He does not intend to deny our knowledge ofthe quiddities common to individual entities; according to Suhrawardī, likethe majority of his philosophical contemporaries, we can grasp the equinitycommon to all horses that we perceive or imagine. But instead of a thoughtprocess leading to the definition of the concept of horse, this grasp should beunderstood in more straightforward terms: one simply grasps the appear-ance of the quiddity in the field of one’s awareness. There are no media thatcan be resorted to in an epistemological or psychological explanation ofwhat takes place. Appearance cannot be demonstrated, it is unshareable, noone can make another aware of an appearance, and yet it is self-evident forthe one aware of it.92 All sorts of preparatory processes, including syllogisticreasoning and attempts at definition, may assist in the occurrence of theappearance; they may help me see the horse in front of me in a manner I wasnot capable of before, but once the conditions are there, equinity simplyappears to me in the individual horse. The preparatory processes do notcause the intuition in the proper sense of the term, because no awareness ofappearance will necessarily follow from them. An exemplary instantiation ofequinity has to appear to me as a single concrete object that I am presentlyaware of, for the definition requires something that it helps me to becomebetter aware of. Moreover, the validity of the definition can only be assessedagainst the intuitive certainty I have on the basis of my indubitable aware-ness of the appearance of the definiendum.93 And finally, awareness of theappearance of equinity is not dependent on definition or syllogistic reason-ing, for it is perfectly possible that someone simply sees the equinity in anexcellent representative of the species without any prior preparation orinstruction. Thus, the theories of definition and syllogistics concern whatare at best educational tools.94 The intelligible form is a simple thing, and itis grasped in a simple act of intuition. We can express what is simple bymeans of the complex, but this does not render the reference of theexpression complex in itself. In the final analysis, knowledge is not basedon definition but amounts to the simple appearance of what is known, andthe one who is aware of the appearance can dispense with the definition.95

In light of the above, the originality of Suhrawardī’s application of thephenomenon of self-awareness is difficult to exaggerate. Instead of anexplanatory factor in the psychological investigation of human being, asin Avicenna, it furnishes him with the basis of an original reformulation of

92 H· I I.3.4.70, 51–52 Walbridge and Ziai; 73 Corbin. 93 Cf. Ziai 1990, 121–122.94 Ziai 1990, 124–125. 95 H· I I.3.70, 52 Walbridge and Ziai; 73 Corbin. Cf. Ziai 1990, 130, 134–135.


Platonic metaphysics. Admittedly, Suhrawardī is not entirely withoutprecedents in this regard, for Ghazālī had already spoken of existence interms of appearance and light, and of self-awareness as a particular typeof light, in the first chapter of the Mishkāt al-anwār.96 However, as Ihope to have shown, Suhrawardī constructs his novel metaphysics uponself-awareness in full awareness of the originality of his move and with anargumentative rigour that is all but missing in Ghazālī’s work, which isultimately little more than a paraphrastic account, in light of the famousverse from the s.ūra of light (Q 24:35) and the h. adīth of veils, of Neoplatoniccosmology and the situation of human being in it, designed to edify andinvigorate its reader rather than establish a secure foundation for a philo-sophical system. Suhrawardī, on the contrary, attempts to elucidate hischosen foundational concept of metaphysics in an ostensive fashion bystating that pure light is the self-awareness which each of his readers will,upon introspective analysis, find herself to consist in.

For Suhrawardī, self-awareness is something that no theoretical accountcan explain. The best he can do is try to make it evident for his interlocutorsby means of the argumentative pointers he inherits from Avicenna.However, what is thereby paid attention to can of course be constructivelyapplied in subsequent theorizing, and this is exactly what Suhrawardī doesin the metaphysical section of the H· ikma al-ishrāq. Despite the nobleIslamic and Zoroastrian history of light metaphors to which the bookoccasionally refers,97 ‘pure light’ is first and foremost introduced as aname for a constant feature of experience, or, to put it another way, theterm is characterized by purely experiential means, ostensively as it were.

6.3 Degrees of self-awareness

As we have seen, Suhrawardī cashes out the concept of pure light orappearance to self, the foundation of the illuminationist metaphysics, bymeans of the phenomenon of self-awareness as delineated by Avicenna.

96 Ghazālī, Mishkāt al-anwār I.32, 12–13; I.40–41, 16; and I.49, 19.97 Cf.H· I II.2.2.138, 92Walbridge and Ziai; 128–129 Corbin; II.2.10.159, 104Walbridge and Ziai; 149–150

Corbin; II.2.12.166, 108 Walbridge and Ziai; 106–108 Corbin; II.4.1.201, 128 Walbridge and Ziai; 193Corbin; II.4.3.208–210, 131–132 Walbridge and Ziai; 198–201 Corbin. The systematic use of thismythology is scant in comparison to the philosophical ingredients, and to my mind the Corbinian‘hierohistorical’ account of Suhrawardī as a reviver of an ancient Persian Platonism (see Corbin 2009a,XLIII–LXII; Corbin 2009b, 33–55; an extended treatment of this conviction is Corbin 1971, vol. II)drastically overstates its importance. I am neither alone nor the first in claiming this; for a consistentargument that the philosophical tradition is much more determinative of Suhrawardī, see Walbridge2000. Corbin’s interpretation is set in a historical context in Walbridge 2001, 90–91, 105–110.


Later on he revisits this connection, stating explicitly that every pure lightshares this particular type of self-awareness with us human beings andmoreover that this is because self-awareness is constitutive to pure lights,that in which being a pure light consists:

It has been shown that your I-ness is a separate light and apprehends itself(mudrikun li nafsihi), and that separate lights do not differ in their realities(h. aqā’iq). Thus, the whole must apprehend itself (mudrikan li dhātihi), sincewhat is necessary for something is also necessary for that which shares thereality with it.98

Thus, insofar as there are other pure lights, I can assume that they areessentially similar to what I have found myself to be. This is because as purelights we share or participate in one and the same reality, that of being light.Furthermore, since I have found myself to be nothing but self-awareness,I am warranted to conclude that the being of other pure lights also consistssolely of their awareness of themselves, in exactly the same sense as I amaware of myself.99

It is important to note that the shared reality does not entail any kind ofdissolution or merging into the radiance of a single light. Instead,Suhrawardī holds on steadfastly to the idea that the multiplicity apparentin the world is genuine and real, as a consequence of which pure lights,although participating in a single reality, are distinct from each other. Thisgives rise to a version of the problem of individuation which Avicennatackled in Shifā’: Fī al-nafs V.3 and which we have discussed in some detailabove:100 if pure lights are such only because of their self-awareness, howcan they be distinguished from each other?Interestingly enough, Suhrawardī tacitly refers to Avicenna’s solution in

the psychological section of the Shifā’ when he addresses the question of theindividuality of human beings post-mortem.101 His version of the list offactors that allow an intellectual sort of distinction (imtiyāzan ‘aqlīyan)between individual human beings is made up of ‘their self-awareness(shu‘ūrihā bi dhātihā), their awareness of their lights and their illumination(shu‘ūrihā bi anwārihā wa ishrāqātihā), and individuation based on thegovernance of fortresses’, the second and third factor presumably summingup the properties due to the body (or ‘fortress’) elaborated at greater length

98 H· I II.1.8.127, 86 Walbridge and Ziai; 120 Corbin; cf. II.1.8.126, 86 Walbridge and Ziai; 120Corbin.

99 This point is also noted by Ziai 1990, 150–152. 100 Chapter 3.1.101 H· I II.5.2.243, 148 Walbridge and Ziai; 228–229 Corbin.

Degrees of self-awareness 155

by Avicenna. None of the factors other than self-awareness can, however, beused to account for the individuation of pure lights in general terms, sincenot every pure light is a ‘managing light’, or, in more familiar terms, a soulthat governs a body. Thus, pure lights have to be distinguished from eachother by features inherent to their being light.

Suhrawardī locates this feature in the degree of luminosity unique toeach of the pure lights. According to Suhrawardī, lights form a hierarchi-cally ordered series of emanation, reflection and refraction in which aproper place is designated to each and every light by virtue of its degreeof luminosity. If their mutual differences were due to anything externalto luminosity, each pure light would have a constituent in itself, which,not being light, would not appear to itself – a consequence in blatantcontradiction to the manner in which the concept of pure light wasaxiomatically sketched to begin with.102This idea of degrees of luminosityultimately allows Suhrawardī to conceive of the entire realm of existence,including God as the Light of all lights, by means of a single term: inthe final analysis, all is light in one and the same sense but in differentdegrees.

But if light and luminosity amount to self-awareness, and if luminosityallows degrees, then Suhrawardī finds himself committed to the view thatself-awareness allows degrees, that one can be more or less aware ofoneself. Such a view is diametrically opposed to Avicenna’s central insightaccording to which self-awareness amounts to the first perfection ofhuman beings. As we have seen, according to Avicenna each of us isaware of herself in the very precise sense that she will find all content ofexperience given to her in the first person in a manner which enduresintact and unchanged from the very beginning of her existence until theproverbial end of time. This first-personality is not subject to change; onewill be an I regardless of whether as a flying man, a newborn, a vile sinneror a virtuous sage fully developed in terms of knowledge, the secondperfection of human being. On the contrary, Suhrawardī not only allowsfor degrees in human self-awareness, but in full knowledge embraces thefarther reaching point that all pure lights, including God, are each an I inthe same sense as we are, only somehow more so.

Although it is part and parcel of the metaphysical tradition to conceiveof God as a cognitive agent similar but superior to ourselves, the shift ofthis idea to the basis of self-awareness allows Suhrawardī to make sense of

102 Cf. H· I II.1.8.126, 86 Walbridge and Ziai; 120 Corbin.


God’s awareness of Himself in a manner arguably more profound than hisPeripatetic predecessors.103 By means of a variation of the traditionalargument from contingency, Suhrawardī first argues that there must bea necessary being: given that each of the manifold lights is caused byanother light, they are all necessary through another; since an orderedinfinite series is impossible, and since nothing can exist without beingnecessary in itself or having been made necessary through another, therehas to be an end to the postponement of necessity, an end which isnecessary in itself.104 Furthermore, since pure lights are highest in thehierarchy of being, and since existence can only be bestowed in a down-ward order of causation, only another pure light is capable of bringingabout a pure light.105 The necessary being must therefore be a pure light,an appearance to itself in itself, and this to the highest possible degree – aLight of all lights. Finally, there can be only one such light, for were theretwo or more, these would have to differ due to something additional totheir merely being lights. Since such a differentiating factor could beneither a light (for this would render the Light of lights necessary throughanother) nor a non-apparent state (for in that case the Light of lightswould not be a pure light), nothing is left to differentiate between twolights necessary in themselves, and thus it makes no sense to conceive ofmore than one Light of lights. As a result,

[i]t is thus established that the Light of lights is separate from all else andnothing is associated with it. It cannot be conceived that there is [anything]more splendorous than it. Since the point [of speaking about] something’sknowing itself goes back to its self’s appearance to itself (raja‘a h. ās.ilu ‘ilmial-shay’i bi nafsihi ilā kauni dhātihi z· āhiratan li dhātihi), which is purelightness the appearance of which is not through another, the life and self-knowledge (‘ilmuhu bi dhātihi) of the Light of lights are not added to its self(dhātihi). This has already been shown to you in the case of every separatelight.106

The passage speaks of self-awareness as the appearance of the self to itselfalong the lines familiar from the section in which the concept of pure light

103 For an example of Avicenna’s use of human cognitive categories to make sense of God, cf. Ta‘līqāt,158–161, and the discussion in Chapter 3.2, 56–61. Suhrawardī’s explication of God’s awareness ofHimself by means of our self-awareness also has an obvious parallel in his introduction of the conceptof knowledge as presence.

104 H· I II.1.9.129, 87Walbridge and Ziai; 121–122Corbin. For an Avicennian formulation, cf. for instanceAvicenna, Shifā’: al-Ilāhīyāt VIII.1.4–6, 258–259.

105 H· I II.1.9.132, 88 Walbridge and Ziai; 123 Corbin.106 H· I II.1.9.134, 89 Walbridge and Ziai; 124 Corbin.


was introduced by means of the arguments delineating the Avicennianphenomenon of primitive self-awareness. All that is new is the connectionof this phenomenon to God’s awareness of Himself. Although it is possible,indeed natural, to interpret the description of God as a self-aware subject tobe but a variation of the familiar Peripatetic theme of thought thinkingitself, the explicit connection to the most primitive level of self-awarenessresults in an important difference. As a subject of perfect intellection inwhich the subject and the object completely coincide, God will at bestremain a somewhat remote example of the cognitive second perfectionproper to though seldom reached by human beings. Moreover, God isnot only a self-knower, but also a creator: his self-intellection is the cause forall other things, which makes it highly doubtful whether our self-understanding can really be compared to His. For Suhrawardī, however,God is self-awareness in exactly the same sense as I, regardless of mycognitive progress or lack thereof, always find myself to be – only to agreater degree. Although Avicenna too can be seen to ground his account ofGod’s self-intellection on his psychological theory of intellection in generaland human intellection in particular,107 the difference in emphasis is clear.Even if self-awareness does play a role in intellection by making each act ofunderstanding exclusive to the first-personal subject to whom it belongs,Avicenna’s account revolves around the content of intellection – and in thisregard there is literally a world of difference between God and the humanbeing in its cognitively undeveloped state.

In addition to this theological concern, Suhrawardī speaks of humandevelopment towards the perfection proper to us in terms of an increaseof light and therefore an increase in self-awareness. The illuminationistterminology fails to hide the fact that he here presents a ratherstraightforward paraphrase of the traditional doctrine of emanation andreturn. According to Suhrawardī, all lights are driven by an innate love(mah. abba), desire (shawq) or passion (‘ishq) for the higher lights that aretheir source;108 this unique upward drive, as exclusive to each light as itsself-awareness is, accounts for the particular features of the type ofepistrophe proper to it. In the case of human beings, the return to theperfection of our origin amounts to a detachment from the body and a

107 Cf. for instance the reference to psychology in the theological context of Ishārāt, namat. 4, 146. Thisconnection between the two parts of the Ishārāt is discussed in Adamson 2011a.

108 H· I II.2.6.147, 97–98 Walbridge and Ziai; 135–137 Corbin.


corresponding increase in terms of knowledge.109 This is further qualifiedby a claim, inherited from Avicenna,110 that the desire for the particulartype of pleasure that is proper to human beings is conditional upon thehuman subject’s apprehending (idrāk) the proper object of desire aspleasurable. Thus, a human being may fail to apprehend her presentstate in the light of her true self, in relation to what she really is andwhat she should therefore strive to be; in other words, one may fail torecognize oneself as the sort of thing one really is, that is, as a pure light,and consequently fail to pursue the goal, proper to a pure light, ofbecoming more luminous and better aware of oneself.111

But merely stating that self-awareness allows degrees is not a partic-ularly convincing move unless one appends to it at least an elementaryelucidation of what those degrees amount to. How am I to conceive ofthe difference in degree between God’s self-awareness and my own?How is it possible to be more or less an I than I presently am?Moreover, isn’t it possible that such a difference, although seeminglyconceived in purely quantitative terms, will ultimately make God’sawareness of Himself just as inconceivable and inaccessible to me asHis self-intellection, which brings the world into existence, is differentfrom my self-intellection as a subject in and passive to that world?Finally, does all this not compromise the validity of Suhrawardī’sapplication of self-awareness, along the lines described above, in layingthe foundation for his new metaphysics?Unfortunately, the elucidation of the sense in which self-awareness is

gradual is awkwardly underdeveloped in Suhrawardī. While he clearlyendorses the idea, it is not obvious whether he recognizes that it amountsto a departure from the Avicennian concept of self-awareness. He does makeenough observations to enable a rational reconstruction of at least theelements of a theory behind the idea,112 but in the end such reconstructionsare bound to remain more or less speculative. A historical study of thedevelopment of the concept of self-awareness in Islamic philosophy mustremain closer to Suhrawardī’s explicit words, carefully charting the looseends but refraining from tying them up too neatly, as awkward as the resultmay seem. This is particularly the case with the idea of the gradation of self-awareness, since it seems to be picked up and explicated at far greater length

109 H· I II.5.2.237, 145 Walbridge and Ziai; 223–224 Corbin. Cf. II.2.12.171, 110–111 Walbridge and Ziai;162–165 (with a reference to the Arabic Plotinus); and II.4.8.226, 139Walbridge and Ziai; 213–214Corbin.

110 Cf. Avicenna, Ishārāt, namat. 8, 191–194.111 H· I II.5.2.238, 145–146 Walbridge and Ziai; 224–225 Corbin.112 I have suggested one possible reconstruction in Kaukua 2011, see especially 151–156.


by Mullā S· adrā, though not without a fundamental revision to the conceptof the self involved in self-awareness. In light of that departure, Suhrawardīappears to represent a point of conflict between the Avicennian descriptionof self-awareness as a static feature of I-ness and the idea that one candevelop in terms of one’s I-ness and thereby to be more or less of an I. As wecan see, the conflict can be resolved only by surrendering one or the otherthesis.


chapter 7

Mullā S· adrā on self-awareness

Since the vigorous promotion of his philosophy by his Qajar commentatorsin the nineteenth century, the name of Muh. ammad ibn Ibrāhīm ibn Yahyāal-Qawāmī al-Shīrāzī, better known by the honoraries S· adr al-Dīn, S· adr al-Muta’allihīn and Mullā S· adrā, has been virtually synonymous with Islamicphilosophy in the Shi’ite seminaries of Iran. He is venerated for his synthesisof the rival philosophical, theological and mystical currents of his time intoan original system that represents the summit of the philosophical inter-pretations of the Muh. ammadan revelation. In an œuvre of encyclopedicproportions, S· adrā does indeed develop a philosophy that, as we will soonsee, is genuinely novel yet at the same time brewed from recognizableingredients, mainly from three traditions of learning. The first of thesegoes back to Avicenna and lives on in the theological tradition of criticalcommentary on his thought, while the second is represented by Suhrawardīand his ishrāqī commentators and the third by the akbarī tradition oftheoretically oriented Sufism founded upon the heritage of Muhyī al-DīnIbn ‘Arabī (d. 1240). But in spite of his considerable learning, S· adrā’s ownassessment is that the most significant influence came not from his prede-cessors but straight from the Source, for time and again he emphasizes thecrucial impact of personal inspired intuition upon the development of hisphilosophical system.1

S· adrā’s philosophical career is closely interwoven with the cultural pol-itics of the early S· afavid state. Born in 1571 or 1572, S· adrā left his nativeShīrāz in his early twenties to pursue the studies in philosophy and theIslamic sciences that he seems to have begun on his own, eventually arrivingat Isfahān, the S· afavid capital.

2Under the tutelage of Bahā’ al-Dīn al-‘Āmilī(d. 1620/1), he is reported to have acquired a level of knowledge in theIslamic sciences that was unprecedented among his philosophical

1 For an example of this topos, see Asfār, muqaddima, I.14–15, 17–18; cf. Rizvi 2005, 231.2 Rizvi 2007, 5–8, 10.

161

predecessors,3 and this was complemented by a profound schooling inphilosophy by one of the most original Islamic thinkers, Mīr Muh. ammadBāqir Astarabādī (d. 1631), better known asMīr Dāmād and often referred toby the honorary ‘Third Teacher’ (after Aristotle and al-Fārābī).4

S· adrā’s rise to prominence in the intellectual milieu of Isfahān seems tohave been met with considerable opposition from the more conservativescholars of the city.5 Prompted at least in part by the adverse reception, hereturned to Shīrāz in 1601–2, but sustained opposition in his hometown,once renowned for its thriving philosophical scene,6 eventually inspired himto retreat to the small village of Kahak near Qom, the present centre ofShi’ite learning.7 S· adrā himself reports that in Kahak he largely retired frompublic and collegial life, focusing instead on ascetic andmeditative practices.As he recalls in the preface to his magnum opus, his constant efforts inprayer and ascesis led to a series of intuitions which may have beeninstrumental in convincing him that the ishrāqī doctrine, which he hadadopted from his teacher, was prone to fundamental problems. Thesehinged especially on Mīr Dāmād’s adoption of the notion of quiddity asthe basis of his metaphysics, the famous theory of as.āla al-māhīya, or‘primacy of quiddity’, against which S· adrā eventually developed an alter-native based on existence.8 He held firmly to this in both writing andteaching for the rest of his career through the itinerant years that finallysettled him at the turn of the 1620s in a prestigious teaching position inShīrāz, where he remained until his death in Basrā on a seventh pilgrimage,most likely in 1635/6.9

The S· adrian corpus consists of more than fifty works, ranging frommulti-volume summae to brief treatises on strictly defined topics. In roughterms, these can be divided into two main classes.10On the one hand, S· adrāwrote a number of works in the so-called Islamic sciences, the mostprominent among which are his three treatises on the principles ofQur’ānic exegesis (Asrār al-āyāt wa anwār al-bayyināt, Mafātīh. al-ghayband Mutashābihāt al-Qur’ān), his large commentary on selected s.ūras of

3 Ziai 1996, 636; Nasr 1996, 643. 4 Rizvi 2007, 9–13; Dabashi 1996, 621–632; Ziai 1996, 636.5 Corbin 1964, 3; Morris 1981, 16; Dabashi 1996, 623, 627. For a considered critique of this receivedview, however, see Rizvi 2007, 31–36.

6 For a concise account of philosophy in early sixteenth-century Shīrāz, see Pourjavady 2011, 1–44.7 Rizvi 2007, 14.8 Asfār, muqaddima, I.7–14;Mashā‘ir VI.85, 35; cf. Kalin 2003, 27–28. For an excellent overview of thedebate between as.āla al-māhīya and as.āla al-wujūd, see Bonmariage 2007, 28–53.

9 Rizvi 2007, 14, 22–30.10 Cf. Kalin 2003, 35–60. Rizvi 2007, 52–111, to my knowledge the most complete bibliography of S· adrā,

opts for a more nuanced classification.

162 Mullā S· adrā on self-awareness

the Qur’ān, and his commentary to Abū Ja‘far ibn Muh. ammad ibn Ya‘qūbal-Kulaynī’s (d. 941) Us.ūl al-Kāfī, the first section of one of the mostauthoritative Shi’ite collections of ahādīth. We can also include a numberof practically oriented works of ethical guidance in this class.11 The otherclass consists of S· adrā’s philosophical works, among which the uncontestedpride of place belongs to the immense mature work, al-H· ikma al-muta‘āliyafī al-asfār al-arba‘a (‘The Transcendent Wisdom in Four Journeys’,1606–1628), our main source in the present study. Although modelled onearlier endeavours at the summary presentation of the philosophical explan-ation of the world, S· adrā’s magnum opus deviates from the classicalPeripatetic model by its decided emphasis on metaphysics and eschatologyat the cost of the natural sciences.12 Other important philosophical treatisesare al-Mabda’ wa al-ma‘ād (completed in 1606), a relatively early work onthe origination of the cosmos from God and its return to Him, the Kitābal-mashā‘ir, a concise presentation of S· adrā’s theory of existence as thefoundation of metaphysics written after 1628, al-Shawāhid al-rubūbīya fīal-manāhij al-sulūkīya, a more condensed summa than the Asfār completedbefore 1631, and al-H· ikma al-‘arshīya (completed between 1631 and 1634), alate work with an eschatological emphasis.13 S· adrā also wrote several minortreatises in philosophy as well as commentaries on works by Avicenna,Suhrawardī and other philosophers.14

Much like in Suhrawardī’s case, earlier Western scholarship, with Corbinand Seyyed Hossein Nasr at the spearhead,15 has tended to emphasize themystical aspects of S· adrā’s thought and thereby to downplay the importanceof his consistent systematic and analytic striving, immediately evident inworks like the Asfār.16 For a long time, the only exception was FazlurRahman’s monograph The Philosophy of Mullā S· adrā,

17 now obsolete inmany regards, but a number of studies published since 2000 are beginningto consolidate the image of an original and profound philosopherwho shouldbe considered a peer of his most luminous European contemporaries.18

Reading S· adrā, it is difficult not to recognize the analytic acuity that he

11 Cf. Kalin 2003, 35–41; Rizvi 2007, 69–91.12 The structure of the Asfār is analysed in Arnzen 2007. For changes in philosophical summae between

Avicenna and S· adrā, see Eichner 2007.13 Rizvi 2007, 52–68. 14 Rizvi 2007, 69–77, 91–111.15 Corbin 1964, 1971; Nasr 1963, 1978. A later but influential representative of this strand is Morris

1981.16 For a critique of this line of interpretation, see Ziai 1996, 638–639; and for a concise defence, Nasr

1996, 645 and 659. The different approaches in contemporary scholarship on S· adrā are lucidlydescribed in Rizvi 2009, 4–14.

17 Rahman 1975. 18 Jambet 2002, 2008; Bonmariage 2007; Rizvi 2009; Kalin 2010; Rustom 2012.

Mullā S· adrā on self-awareness 163

applies to his vast reading, most evident in the manner, ubiquitous in theAsfār, in which he painstakingly develops his own thought in a criticalrelation to one or several of his predecessors. The authors brought into playare not exclusively philosophers, but even when dealing with theological orSufi interlocutors S· adrā’s aims and foci remain thoroughly philosophicaland systematic.19 Methodologically, S· adrā ceaselessly asserts the indispens-ability of intellect and reason.20

But instead of dwelling on this general level of debate, I would ratherintroduce the following study of S· adrā’s treatment of self-awareness as a casein point to support the claim that he should be read first and foremost as aphilosopher, albeit one that departs from the tradition in many importantrespects. In order to highlight the departures in the topic of self-awarenessand selfhood, we must begin by establishing that he stands on a sharedground with his predecessors. A brief glance into S· adrā’s discussion ofknowledge and intellection, as well as into the arguments for the immate-riality of the soul in the psychological section of the Asfār, suffices to showthat he imports most of the arguments Avicenna and Suhrawardī designedto describe and delimit the phenomenon of self-awareness. In the following,I will briefly revisit the S· adrian versions of the flying man, the relatedargument from the constancy of self-awareness, the argument againstreflection-based models of self-awareness and the argument from theunity of experience.

7.1 Four Avicennian arguments

The flying animal

One of the most amusing features of S· adrā’s attempt at establishing theimmateriality of the animal soul in the second chapter of the psychologicalsection of the Asfār is his sustained reliance on Avicennian arguments. Thisis amusing because these arguments were originally designed precisely todistinguish the entity that functions as a soul in the human body from itsfunctional counterparts in non-human animals. This difference in applica-tion is most striking in S· adrā’s reappropriation of the flying man, now recastas an animal.

19 Moreover, as Rustom 2012 shows, there seem to be no doctrinal breaches between S· adrā’s philo-sophical and theological works.

20 This is substantiated by his hierarchical classification of the cognitive methods available to humanbeings. See, for instance, Asfār IV.10.4, IX.315; and IV.11.26, IX.464–465.


Another demonstration that the animal is not the sensible structure is that wesay: if an animal is supposed such that it is created all at once, and is createdperfect, but is veiled in its senses from beholding what is external, and that itis floating in a void or in open air so that the air’s volume does not collidewith it and it does not sense any qualities, and its limbs are separated so thatthey do not touch each other, then in this state it will apprehend itself(dhātahu) and ignore all of its external and internal organs, or rather, it willaffirm itself (dhātahu) without affirming a dimension for it, neither lengthnor breadth nor any direction; even if it imagined a position, a direction orsome organ in that state, it would not imagine it to be a part of itself(dhātihi). It is evident that what one is aware of is different from what oneignores; and so its itness (huwīyatuhu) is different from all the organs.21

Notwithstanding the argument’s reappropriation for the animal case, anumber of clues betray that S· adrā is reading from Shifā’: Fī al-nafs I.1.22

However, it is worth pointing out that by replacing the human being withan animal S· adrā loses whatever plausibility the original argument may havehad. Avicenna’s thought experiment hinged on the givenness of self-awareness in the interlocutor’s own experience; the flying man was ameans of bracketing aspects of experience that prevent us from directingour attention at the type of narrow self-awareness each of us has due to ourshared nature. Other animals differ from us in many respects, not least interms of cognitive capacities, and as a consequence we cannot claimintuitive access to animal experience. Thus, animal experience is not avail-able as the sort of immediate evidence one can plausibly build an argumentupon. On the contrary, if self-awareness is intimately related to our beingintellectual subjects, as Avicenna seems to have thought, its application inan argument for the immateriality of the sub-intellectual animal soulappears even more suspect.23

Another sign of S· adrā’s relapse from Avicenna’s argumentative rigour ishis characterization of the argument as a demonstration (burhān) instead ofa reminder (tanbīh) or a pointer (ishāra), as Avicenna had presented it. S· adrāalso sidesteps Avicenna’s emphatic requirement that the interlocutor must

21 Asfār IV.2.2, VIII.47.22 There are both direct quotes (‘created all at once and . . . perfect’, ‘floating in a void or in . . . air’,

‘limbs are separated so that they do not touch’, ‘affirm itself without affirming . . . neither length norbreadth’) and considerable similarities on the level of thought, for instance the added qualification ofa per impossibile act of imagining a body, and the explication of the argument’s logical basis (what isaffirmed as an explicit object of awareness is different from what is not so affirmed). For theAvicennian text, see Chapter 2.1, 35.

23 Cf.Asfār IV.6.1, VIII.327–328, where S· adrā suggests that animals are incapable of self-awareness in theabsence of other contents of experience. This difference, however, is not due to their allegedcorporeality but due to the fact that they are not capable of intellection.

Four Avicennian arguments 165

personally perform the act of imagining herself into the flying man’s situationand chooses to speak in passive about supposing (law furid

˙a) the animal in

the void or the open air.Two possible reasons for S· adrā’s less rigorous take on the thought experi-

ment readily suggest themselves. First of all, by the seventeenth century theargument had become a firm part of the psychological tradition. Earliercommentators, such as Ibn Kammūna, had attempted to increase thestrength of the argument by developing it into a demonstrative syllogism.24

Informed by this line of development, S· adrā may simply have consideredthe argument worthy of the status of demonstration. On the other hand,S· adrā’s claim that non-human animals are aware of themselves in the samesense as human beings may have been based on an argument earlier on inthe same section which infers self-awareness from the perception of painand pleasure. An animal’s perception of pain and pleasure, which S· adrāseems to consider obvious from elementary observation of animal behav-iour, concerns pain and pleasure not in absolute terms but rather as specificto the very perceiving animal itself. Thus, the animal must be aware of itselfin addition to its perception of the object causing the pain or the pleasure.25

In either case, S· adrā’s move remains a significant departure fromAvicenna.26 However, as I think will become obvious from his consistentuse of familiar arguments, this does not mean that the phenomenon hisconcept of self-awareness was based upon was different from Avicenna’s.Rather, the departure signals S· adrā’s different application of the phenom-enon; instead of an Avicennian attempt at making sense of the individualexistence of an intellectual, hence immaterial entity, S· adrā believes it shouldbe an elementary feature of a much broader scope of mental, and therebyimmaterial existence.

This expansion of the scope of self-awareness becomes more explicit in alater chapter designed to argue for the independence of the soul’s cognitivefaculties from their corporeal instruments. For this purpose, S· adrā bringsforth the following piece of evidence:

24 See Muehlethaler 2009. However, Ibn Kammūna’s attempt, although a masterful combination of anumber of Avicennian insights, did not do away with the fact that the premises of the demonstrationhad to be acquired by means of a first-personal performance of the thought experiment.

25 Asfār IV.2.2, VIII.45–46. This argument is suggested to provide the basis of the flying animal byJambet 2002, 229.

26 If Avicenna had a theory of animal self-awareness, it must have been based on acts of estimativeapprehension. Thus, despite certain similarities, the phenomenon would have been different fromthe self-awareness instantiated in human beings. For a reconstruction of Avicenna’s theory of animalself-awareness, see Kaukua and Kukkonen 2007. S· adrā was quite probably aware of this, for heparaphrases an Avicennian discussion on animal awareness in Asfār IV.6.1, VIII.326–328.


When the human being’s external faculties and corporeal senses are still dueto sleep, lack of consciousness (al-ighmā’) or [something] else, he will oftenfind of himself (yajidu min nafsihi) that he hears, sees, smells, touches, strikesand walks. Thus, he has in himself (fī dhātihi) these sensations (al-mashā‘ir),faculties and instruments without deficiency or the need of anything otherthan them.27

Curiously enough, this argument seems to be based on an intuition that isdiametrically opposed to that behind the flying man. At the same time,one cannot help thinking of the sleeping or the intoxicated personfeatured in the Ishārāt version of the flying man; yet here we have S· adrāclaim that even in such seemingly unconscious states one will remainaware of one’s capacity to perceive. For S· adrā, this is evidence of theabsolute immateriality not just of the intellectual, but also of the imagi-native and even the perceptual mode of mental existence. Immersed inacts of perception that demand our complete attention, we may fail tonotice that the perceptions themselves are modes of our own existence, notthat of an external object supposedly independent of and causally activeupon us. Nevertheless, both the capacity to perceive and the act ofperception are in us. Later on in the same context S· adrā formulates thecase even more straightforwardly with important corroboration from thepseudo-Aristotelian Theology:

Thus, his self is by itself (dhātuhu bi dhātihi) sight for the apprehension ofwhat is seen and hearing for the apprehension of what is heard, and similarlyfor every species of sensibles. Thus, in itself (fī dhātihi) it is hearing, sight,smell, taste and touch for itself (li dhātihi). You already know from thepreceding the unity of sense with what is sensed, and so he is the sense of allsenses.28

Thus, although S· adrā recognizes the validity of the flying man, andthereby the concept of self-awareness it hinges upon, he subscribes to thisaspect of Avicennian heritage only with two important qualifications. Thescope of self-awareness must be broader and include both the intellectualand the sub-intellectual mode of mental existence. On the other hand, thelast passage in particular seems to call for a full reassessment of the questionof whether self-awareness is really separable from other constituents ofhuman experience. The motive and the immense consequences of thesequalifications will be our main concern in the next chapter.

27 Asfār IV.8.7, IX.90. 28 Asfār IV.8.7, IX.92.


Argument from the constancy of self-awareness

Closely related to the intuition behind the flying man is the Avicenniandistinction between the constancy of self-awareness and the intermittencyof all acts of awareness concerning the body. This distinction yields thefamiliar argument: were the soul or the self corporeal, it would have to beaware of the organ in which it resides and which is therefore indispensablefor its existence,whenever it is aware of itself; but since we are not constantlyaware of any organ, let alone the entire body, although we are constantlyaware of ourselves, we must reject the supposition and assert that the soul isincorporeal.29

This argument appears repeatedly in the psychological section of theAsfār,30 mostly with rather insignificant additions to or deviations from theAvicennian original. An exception to this rule, however, is a passage inwhich S· adrā engages in a debate that the argument had aroused between thetwo commentators of the Ishārāt, Fakhr al-Dīn al-Rāzī and Nas.īr al-Dīn al-T. ūsī (d. 1274).

31 In one of his counterarguments, Rāzī had challenged thevalidity of Avicenna’s inference by saying that it presupposes the humansoul to be always aware of all of its attributes and concomitants, which doesnot seem plausible on the basis of our experience.32 The simple fact that Iam sometimes unaware of the body is sufficient to establish that my body isnot constitutive of or concomitant to me only if we suppose that I amconstantly aware of whatever is constitutive of or concomitant to me. Inspite of S· adrā’s extremely elliptical rendering of the argument, it seems torely on the salient point that the soul is not in every respect transparent toitself. As we have seen, Rāzī himself had argued that the soul’s substantialityis not obvious to the soul merely from its awareness of itself.33 Earlier on,Abū al-Barakāt al-Baghdādī had pointed out that, the soul’s constantawareness of itself notwithstanding, some of its acts in the body, such asdigestion and growth, are very rarely, if ever, given to it as its own.34 In thelight of such counterexamples, it is only natural to ask why the self’sinherence in the body could not be both constitutive or concomitant and

29 See Avicenna, Shifā’: Fī al-nafs V.7, 255–256 Rahman; Ishārāt, namat. 3, 120.30 Cf. Asfār IV.2.2, VIII.46–47 (in an argument for the incorporeality of the animal soul); IV.6.1,

VIII.347–348 (with an interesting remark against the Aristotelian thesis that the faculty of senseperceives its own act of perceiving); and IV.6.1, VIII.353–355 (in a variation of the argument from theunity of experience).

31 For the argument, which S· adrā seems uncharacteristically critical of, see Asfār IV.6.1,VIII.338–340.

32 Asfār IV.6.1, VIII.343.33 See Rāzī, Mabāh. ith II.2.2.1, II.246–247, as discussed in Chapter 5.2, 116–118.34 Baghdādī, Mu‘tabar: al-‘ilm al-t.abī‘ī VI.5, II.319–320.


opaque to the self. The choice, decisive for the argument, between thoseopaque aspects of the self that are constitutive to it and those that are notseems all but arbitrary.T. ūsī’s answer in Avicenna’s defence is based on a distinction between

two types of constituents of experience:

The attributes and concomitants are divided into what is necessary for thesoul due to itself (li dhātihā), such as its apprehending itself (kawnihāmudrikatan li dhātihā), and what is necessary for it after comparing it withthings that are different from it, such as its being separate from matter andnot existing in a subject. The soul apprehends the first sort constantly, just asit apprehends itself constantly (kānat mudrikatan li dhātihā dā’iman), but itdoes not apprehend the second sort except in the state of comparing, becausethe condition is lacking in states other than that.35

T. ūsi’s point is clear: in order to be aware of its relation to and separationfrom the body, the soul has to compare and relate itself explicitly to thebody by means of considerations such as the present argument. Thiscomparison requires information acquired through the operation of thesenses and is therefore not essential, innate or constitutive to the soul, unlikethe soul’s awareness of itself.S· adrā’s assessment of the debate is admirably clear in its density. If we

hold, as he does, that the self’s perceptions (apprehension in T. ūsī’s secondclass) are acts of the self that it is aware of in and due to itself, we must rejectT. ūsī’s distinction as based on inadequate evidence.36 Again, S· adrā’s basicinsight will become clearer when we chart the consequences of his theory ofcognitive unity for his conception of human selfhood. Suffice it now to saythat his analysis of this debate generated by the argument from the con-stancy of self-awareness shows that S· adrā latches on to a well-establisheddiscussion on self-awareness without pointing out any pressing need toreject the underlying phenomenon. That he understood this phenomenonin exactly the same sense as his aforediscussed predecessors becomes clearfrom his treatment of the two remaining types of argument.

35 Asfār IV.6.1, VIII.343–344. Cf. T. ūsī, Sharh. al-Ishārāt II.346–348.36 Asfār IV.6.1, VIII.344. Let it be added that S· adrā’s refusal of T. ūsī’s line of defence does not entail that

he takes Rāzī’s criticism to be conclusive. If one differentiates categorically between the material andmental existence of one and the same entity, as S· adrā does, the opaque aspects of a human being canbe exhaustively relegated to material existence so that they no longer pose a problem for the fulltransparency of every act of mental existence. As a material form, the human soul’s existence doesconsist in unaware acts in the body, but that is simply because material existence is below the level ofthe mental. Cf. Asfār IV.2.5, VIII.77–79.


Argument against reflection-based models of self-awareness

This Avicennian argument figures in two different contexts in the Asfār.S· adrā reproduces the intuition underlying the original in a chapter dedi-cated to showing that an intellect’s act of self-intellection is identical withthe intellect itself, and therefore temporally coterminous with it. Theargument is preceded by a closely related statement according to which allhuman apprehension and action take place in the first person:

When a human is engaged in an apprehensive or a motive act, his purpose isnot absolute apprehension or movement but an individual apprehensionwhich originates in him and occurs to him, and the same is to be said ofmovement. When someone escapes from an enemy or heat or cold, hisescape is not from an absolute enemy but from his individual enemy, nor is itfrom absolute heat but from an individual heat which has hurt him andhappened to his self (dhātihi), and knowledge of the occurrence of the heat orthe cold to him entails knowledge of him. Similarly, someone who intends toact in some manner or to acquire something desired does not intend that theact occurs in an absolute sense but that it occurs with respect to him, nor tosatisfy an absolute desire but a desire particular to him, and all this is derivedfrom his knowledge of himself (dhātihi). It is thus evidently shown that ahuman’s knowledge of his soul and his self (bi nafsihi wa dhātihi) is the firstand oldest knowledge, it is always present and he is never without it.37

The argument relies on commonplace examples of motivated action: nodisinterested perception of a potentially painful quality will cause anyreaction in us, but only a quality that we actually perceive as painful andtherefore concerning ourselves will. By the same token, we will be driven toneither fight nor flee at the sight of an armed person unless we perceive himto be hostile towards ourselves. Thus, the first-order perceptions thatprompt reactions in us must entail some sort of self-acquaintance, theyhave to be uniquely our own. Now, if such self-acquaintance is a necessaryfeature of such commonplace acts and related apprehensions, S· adrā argues,we have no reason to presume that they are not there even in apprehensionsthat do not demand immediate reactions of us. In this precise sense, there-fore, each human being, as an intellectual entity, is constantly aware ofherself.

Thus, S· adrā’s argument hinges on the fact that all our experience andaction take place in the first person, in much the same way as Avicenna’sargument against reflection-based models of self-awareness. It is therefore

37 Asfār I.10.2.4, III.505.


not surprising that he supports his case with what amounts to an uncreditedquote:

No one can say: my knowledge of myself (bi nafsī) is due to a medium whichis my act, I am informed of myself by my act (ustudilla bi fi‘lī ‘alā dhātī). Thatis because I can neither be informed of myself (dhātī) by an absolute act norbe informed by an act which originates from myself to myself (s.adara minnafsī ‘alā nafsī). If I am informed by an absolute act, an absolute act onlyrequires an absolute agent, and only an absolute agent can be established bymeans of it, not an agent that would be me. If I am informed of myself (‘alay)by my act, I can only knowmy act after knowingmyself (nafsī). Thus, if I canonly knowmyself (nafsī) after knowing myself (nafsī), a circle results, and it isfalse. This therefore indicates that a human being’s knowledge of himself (binafsihi) is not by means of his act.38

This version of the familiar argument is clearly derived from the Ishārāt andis used to make exactly the same Avicennian point.39 If my self-awareness issupposed to be due and subsequent to my reflective consideration of myown act, we will not be left with anything by means of which the first-orderact can be rendered an act that belongs uniquely to me. Thus, the state hasto be somehow earmarked as mine to begin with, and as the earlier argu-ment spelled out, this can be achieved because of the inherent first-personality of the act, its being my act.Another occurrence of the same argument can be identified in the

chapter that has already yielded us the flying animal. The purpose is,again, to argue for the immateriality of the animal – and, a fortiori, thehuman – soul. One of S· adrā’s demonstrations is a syllogism relying on apremise that the animal is constantly aware of itself, not by means of anacquired object of knowledge. The argument against reflection-based mod-els of self-awareness is relied on as support for that premise.

Were knowledge of the soul’s existence acquired, it would be either by meansof sense-perception, which is false . . . or by means of thought, and thus nodoubt from evidence (dalīl), the evidence being either a cause of the soul orits effect. The first is false, because the cause of souls is something tooelevated to be encompassed by the animal’s knowledge; furthermore, mostpeople know themselves (anfusahum) even though the cause of their souls(anfusihim) does not occur to their mind.The second is also false, because eitherthe medium in the evidence is an absolute act or [the animal’s] act related to it;thus, if it considers an absolute act, consequently it asserts an absolute agent, notan agent that is it; and if it considers a related act, knowledge of an act related toan individual depends on knowledge of [the individual], so that if knowledge of it

38 Asfār I.10.2.4, III.505. 39 Cf. Avicenna, Ishārāt, namat. 3, 120; and see Chapter 4.1, 72–75.


were acquired from knowledge of an act related to it, a circle would result. Thus,it is established that the animal’s knowledge of itself (bi nafsihi) is not derivedfrom sense-perception or evidence.40

In all its density, the argument is a remarkably faithful rendering of theAvicennian original. However, the fact that it is embedded in a contextdealing with the animal soul has consequences that may seem problematicat first sight. In Avicenna the argument is designed to refute a thesis thatarises from a commonplace phenomenon of human psychology: we arecapable of reflecting upon ourselves, and since it is in an explicit reflectivestance that we first come to pay attention to ourselves, it may seem naturalto adopt reflection as the primary sense in which we choose to speak of self-awareness. The capacity of reflection, however, is an exclusive prerequisiteof intellectual subjects and thereby something that non-human animals arebereft of. Since nothing indicates S· adrā’s willingness to give up this tradi-tional tenet,41 it seems natural to read this version of the argument againstthe reflection model of self-awareness as an argument per impossibile. Butunlike the flying animal, this re-appropriation of a discussion on humanbeings in the animal case is not fatal to the argument’s plausibility. Acounterfactual argument can be perfectly valid – provided that the thesisit is intended to support can be corroborated by other means. And S· adrāclearly does not base his claim of the immateriality of the animal soul on theargument against reflection, which instead serves the significantly moremodest purpose of rejecting the suggestion that animal self-awareness isdue to acquired knowledge.

There is also a less obviously Avicennian variation of the argument, in thecourse of which S· adrā makes an interesting qualification. This version isembedded in an extended case for the immediacy of self-awareness.Suppose, for the sake of the argument, that self-awareness is due to a specialobject of awareness, a cognitive form which refers to one’s self and theawareness of which will therebymediately amount to self-awareness. In sucha case, this form will either have to be identical to one’s self or in somerespect different from it. If it is identical to one’s self, then a form (of the selfas object of awareness) will occur to itself (the same self as the subject ofawareness). The problem is, given that the self-aware human being isimmaterial, we have no means of distinguishing between the two identicalforms, and so the duality required in the presupposition can no longer bemaintained. If, on the other hand, the form is somehow different from the

40 Asfār IV.2.2, VIII.46–47; emphasis added.41 That animal self-awareness is not intellectual is explicitly stated in Asfār IV.6.1, VIII.326–328.


self it represents, awareness of the form is not a case of self-awareness sensustricto. S· adrā is elliptic about why exactly this is the case, but it seems naturalto read his statement as a tacit reference to the argument against reflection-based models of self-awareness: if the form is simply different from me,there is no non-arbitrary way to recognize it as myself. In any case, when allpossibilities of an alleged medium of self-awareness have been ruled out, sothe argument runs, the self must simply occur to itself. Since this cannot bea question of acquisition, self-awareness has to be constant.42

In light of what we have learned from S· adrā’s predecessors, one mightexpect the discussion to be settled at this point. However, S· adrā finds theargument problematic because it threatens to rule out the possibility ofreflective self-awareness, which he says ‘many of us’ are perfectly familiarwith. This is because when I take a prior mental state or act of mine intoexplicit consideration in a higher-order act of reflection, a form that isidentical to me will have to occur to me. Were it not for this identity, Iwould not be able to recognize myself in the object of reflection; were it notfor the distinction between the first- and second-order states, no reflectiverelation could be established. Thus, S· adrā thinks that a rephrasing of theargument is in order.43

According to S· adrā, the real problem behind the claim that a specificform acts as a medium for self-awareness is that any form occurring to asubject that is aware of it is an accident of that subject. But if the self is asubstance and the object of knowledge an accident, I should be aware ofmyself as an accident of myself. This is evidently not the case. On thecontrary, if I choose to apply the Aristotelian system of categories to myself,I will inevitably classify myself under ‘substance’. Thus, self-awareness hasto be due to an immediate presence of the self to itself, just as Avicennacogently, if in somewhat inaccurate terms, had argued.44 However, it is

42 This argument, with slight variations, figures in two contexts in the Asfār: first in a discussionconcerning the self-intellection of all intellects (Asfār I.10.2.4, III.501), and later on in the psycho-logical section of the book in a chapter designed to show that the rational soul is incorporeal (AsfārIV.6.1, VIII.320–326). In the latter, S· adrā informs us that he is drawing from Avicenna’sMubāh. athāt.While certain specific points in the discussion can be located in that compilation of Avicenniancorrespondence (cf. Avicenna,Mubāh. athāt 55–58, 134–135 Badawī (VI.446, 121; and VI.493, 172–173Bīdārfar) with Asfār IV.6.1, VIII.320–324; Avicenna, Mubāh. athāt 426, 222–223 Badawī (VI.892, 318Bīdārfar) with Asfār IV.6.1, VIII.324–325; the discussion on animal self-awareness in Avicenna,Mubāh. athāt 357–358, 199 Badawī (VI.502–505, 175–176 Bīdārfar) and 374–375, 209 Badawī(V.289–293, 120 Bīdārfar) seems to be vaguely related to the considerably more developed views inAsfār IV.6.1, VIII.326–328), I have not been able to find precedents for all steps in the series ofquestions S· adrā discusses.

43 Asfār I.10.2.4, III.501. This follow-up is not included in the psychological context.44 Asfār I.10.2.4, III.501. This is reminiscent of Avicenna’s response to his interlocutors as reported in

Asfār IV.6.1, VIII, 320–321; cf. Avicenna, Mubāh. athāt 56, 135 Badawī; VI.446, 121 Bīdārfar.


important to notice that S· adrā clearly thinks that his manner of phrasing thepoint is better equipped to make sense of commonplace acts of reflection.When I reflect upon myself, I realize that my act of reflection is accidentaland that its object is just as fortuitous as any other object of my awareness.Thus, the form that is me as an object of reflection does occur to meaccidentally. Yet it is equally important to point out that this qualificationis inconsequential for S· adrā’s account of the primitive self-awareness at itsbasis, an account which remains thoroughly Avicennian. His only qualm isthat we have to allow for the possibility that a cognitive subject is presentedwith an object she apprehends as distinct from yet identical to herself. Theidentity between the subject and the object is grounded in first-orderprimitive self-awareness, the subject of which recognizes herself in reflectionprecisely because she was already there in the object of reflection, whereasthe distinction is due to the hierarchical relation between the subject-substance and the object-accident.

The Avicennian argument against the reflection model of self-awarenesshinges on an implicit distinction between the types of awareness respectiveto a subject and an object. In the previous chapter we saw Suhrawardīexplicate this distinction by means of the pronouns ‘I’ and ‘it’. S· adrā revisitsthis method in a loaded chapter that gathers together the premises requiredin an investigation into God’s knowledge. Having established a distinctionbetween the epistemological concept of knowledge as a mental form thatrefers to an external object and the psychological concept of knowledge assimply a feature of the world, he makes the following observation on self-awareness:

We apprehend ourselves (dhawātanā) through our very form through whichwe are we, not through a form additional to it. Thus, every human beingapprehends himself (dhātahu) in a manner which prevents sharing. If thisapprehension were through a form which occurs in our soul, [the form]would be universal, and even if it were a collection of universals the whole ofwhich is individuated by a single self (dhāt), then nevertheless its veryconception would not rule out the possibility of being true of many.Besides, we refer to every universal concept and mental form – even if itwere something which subsists through our self (dhātinā) – by ‘it’ whereasour self (dhātinā) we refer to by ‘I’, and our knowledge of our self is identicalto the existence of our self and our individual itness (‘ilmunā bi dhātinā ‘aynuwujūdi dhātinā wa huwīyatinā al-shakhs.īya).

45

45 Asfār III.1.3.1, VI.149.


Self-awareness is a unique cognitive phenomenon because in it theepistemological and the psychological concept of knowledge coincide: weknow ourselves simply through our existence, or, in the Avicennian termsS· adrā borrows here, our knowledge of ourselves is our very existence. Butthe way in which S· adrā argues towards this statement is a mixture of twoSuhrawardian ideas. As we recall,46 Suhrawardī argued against the impres-sion theory of knowledge by saying that whatever is impressed in animmaterial subject, such as the human soul, is universal, because immate-riality entails intellectuality, and intellects have universals as objects. S· adrāconnects this to the statement that objects always appear in the mode of anit, whereas appearing as an I is exclusive to our respective selves, unshareableby and inaccessible to any other subject.47 The interesting feature in thiscombination of the two Suhrawardian insights is that S· adrā seems to denyhis predecessor’s claim according to which the first-personal indexical is auniversal in its own right.48 This impression is corroborated later on in thesame context: ‘whatever is composed of universal concepts can only bereferred to by means of “it”, not by means of “I”’.49

Suhrawardī spoke of the ‘I’ as a universal concept presumably because itrefers in one and the same sense to primitive first-personality in the case ofeach of its multiple utterers. When Zayd, ‘Umar and Khālid say ‘I’, eachindicates his being a first-personal subject of experience; although theindividual referent of the pronoun is given exclusively to the subject utter-ing it, everyone refers to individual instantiations of the same I-ness. S· adrā,however, quite probably owing to the needs of his context, emphasizes theindexicality of the pronoun instead of the alleged meaning that is preservedfrom one context to another. Whenever someone utters the expression ‘I’,she must refer to her unshareable and exclusive awareness of herself, and inthis sense the indexical cannot be used of any object. Moreover, strictlyspeaking S· adrā does not even say that the first-personal indexical does notfunction as a universal in the sense expounded by Suhrawardī; all he states isthat whatever is explicitly given in universal terms is by necessity given as anobject and thereby distinct from the I or the self.50

46 Cf. Chapter 6.1, 128–130.47 This familiar distinction is not obscured by the fact that S· adrā uses the technical term ‘itness’

(huwīya) at the end of the passage. In the sense intended here, ‘itness’ does not denote the oppositeof I-ness but rather refers to the individuality of any existing thing, whether the thing is a subject (‘I’)or an object (‘it’). Cf. Goichon 1991.

48 Cf. Suhrawardī, Talwīh.āt III.5.15, 282–283Habībī; 115–116 Corbin; and see Chapter 5.2, 119–121; andChapter 6.1, 128–129.

49 Asfār III.1.3.1, VI.149; cf. III.1.3.1, VI.150.50 Cf. Asfār IV.2.3, VIII.50–51, where S· adrā speaks of the concept ‘I’.


Our brief consideration of S· adrā’s use of the argument against thereflection model of self-awareness shows, again, that he speaks of self-awareness in the same sense as his predecessors. This is not obscured eitherby his somewhat less rigorous terminology or by the qualifications he makesfor the argument. If anything, the qualifications help to underline the factthat by the seventeenth century this particular way of delimiting thephenomenon had undergone considerable refinement.

Argument from unity of experience

The traditional argument from the unity of experience surfaces in a con-densed form quite frequently in the psychological part of the Asfār. I willconsider only two somewhat more extended treatments of it; the first comesfrom a chapter designed to demonstrate the soul’s unity despite its seem-ingly distinct faculties, while the second is embedded in an eschatologicalrefutation of transmigration. I will conclude by considering a qualificationS· adrā is driven to make in a consideration of Abū al-Barakāt’s and Rāzī’scounterexample of unconscious vegetative acts such as digestion.

S· adrā’s discussion of the Avicennian theory of the internal senses in thefifth section of the psychology of the Asfār culminates in three demonstra-tions for the soul’s unity, notwithstanding the faculty psychological analysisof its functions. Of these demonstrations – from the object known, theknowing subject and knowledge itself, respectively – the second is a varia-tion of the argument from unity.

You do not doubt that you see things, hear sounds and apprehend intelli-gibles, nor do you doubt that you are numerically one. If that whichapprehends intelligibles were different from that which apprehends sensibles,then the substance of your self (jawharu dhātika), which is strictly speaking(‘inda al-tah. qīq) you, would not perceive the two together. If it doesapprehend both, then that which apprehends them is one self (dhātan),and that is what was sought for, otherwise you would be two selves (dhātayn)instead of one self (dhātan). The same is to be said about desire and anger, foryou do not doubt that you desire intercourse or something else, and that youare angry at your adversary.51

There is little extraordinary about this version of the argument. S· adrā relieson the familiar phenomenon of distinct objects being given to a subject thatone will constantly identify with regardless of the fluctuation in the objec-tive content the subject is presented with. That subject, the persistent first-

51 Asfār IV.5.4, VIII.265.


personal perspective to that content, is what the interlocutor will intuitivelyrecognize as herself.However, slightly later on, when dealing with a counterargument accord-

ing to which the real subject of each apprehension is the respective faculty ororgan, S· adrā is led to make an interesting specification regarding the properrealm of the experiential unity that the argument relies on.

If you say: ‘I do not apprehend after the service’,52 then you do not see, youdo not hear, nor do you find in yourself (min nafsika) your pain, yourpleasure, your hunger or your thirst, but you know that the eye, which isyour instrument and the seeing faculty, has apprehended and seen some-thing, and this knowledge is different from the reality of seeing and vision.Thus, knowledge that the eye sees, the ear hears, the foot walks and the handstrikes is not seeing, hearing, walking or striking, just as knowledge thatanother is hungry, in pain or joyful is not intuition (wijdānan) of hunger,pain or joy. But even those in the very beginning of their understandingknow that they hear, see, suffer pain, rejoice, strike and walk, and if thisknowledge can be denied, then all that is perceived and beheld can be denied.It is therefore known that the faculty of our hearing, seeing, striking andwalking is through the soul, and through it we hear, through it we see,through it we strike and through it we walk. By means of this it is establishedthat the substance of your soul (jawhara nafsika) – through which you areyou – hears, sees, suffers pain, rejoices, understands, comprehends, strikesand walks, even if in each species of these acts it needs a proper naturalinstrument. There is no dispute about that as long as we remain in the worldof nature, but if the soul casts away the body and becomes independent inexistence, these acts emerge from it without instruments.53

Knowledge of the relevant corporeal processes which the Peripateticphilosophers took to cause perception or amount to action, that is, knowl-edge ‘that the eye sees, the ear hears, the foot walks and the hand strikes’, isnot awareness of actually perceiving or acting. Yet our awareness of ourseeing, hearing, walking and striking is one of the first facts we recognize asintuitively evident, and therefore the foundation upon which all our furtherknowledge must lie. This is of course an oblique subscription to theAvicennian claim of the primitivity of self-awareness, but here in thecontext of the argument from unity it serves to distinguish the corporeal

52 ‘Service’ (al-ta’diya) here refers to the operation of the instrument of the soul’s act, in this case ofapprehension. Earlier on S· adrā has stated that even if we allow, for argument’s sake, that theinstrument apprehends its objects, the apprehension of the soul or the self still remains distinctfrom it, and it is the latter that the argument from unity hinges upon. Here he proceeds to tackle thestronger claim that there is no subsequent apprehension of the single self at all.

53 Asfār IV.5.4, VIII.266–267.


processes allegedly underpinning our experiences from those experiencesconsidered as such. The argument from unity derives its force from therealm of the latter, and the self-awareness whose recognition it hinges uponis precisely the sort of commonplace phenomenal fact that Avicenna hadintroduced. S· adrā’s specification intensifies Avicenna’s argument in a fash-ion that rings peculiarly modern in its explicit distinction between third-personally describable objective things and events, such as corporealprocesses, and first-personally experienced qualitative experiences. By thesame token, S· adrā takes his predecessor’s substance dualism a step further:for him, mental existence, or the realm of self-aware experience, is ulti-mately self-subsistent and entirely independent of its erstwhile corporealbasis. The body is an instrument which the soul, having reached the level ofmental existence, can leave behind, not merely in intellection but equally inexperiences of the perceptual mode.54

In the last, eschatological part of the Asfār, S· adrā devotes an entire sectionof considerable length to a refutation of transmigration, with one of thechapters building upon the fact that ‘every human individual has a singleself (dhātan) that is his soul’,55 that is, on the argument from unity. This factis established at the very beginning of the section as follows:

Each one of us knows intuitively (bi al-wijdān), before resorting to demon-stration, that his self (dhātahu) and reality is one thing, not many things.Thereby he knows that he understands, apprehends, senses, desires, is angry,prefers, moves, is at rest and is characterized by a combination of attributesand names, some of which are of the class of the intellect and its states, someof the class of sense-perception and imagination and their states, and some ofthe class of the body and its accidents and passions. Although this is some-thing intuitive, most people cannot know it with respect to the art ofknowledge but deny this unity when they embark on inquiry and scrutiny,except the one whom God assists by a light from Him. How will one who isincapable of the unity of his self (nafsihi) have strength for the unity of hisLord? What has reached us from the ancients regarding this question is thatwhen they distributed types of acts to types of faculties and related each oneof them to a different faculty, they needed to show that in all of them there issomething like root and origin and that the rest of the faculties are like itsconsequences and branches.56

The argument itself is perfectly familiar, but the methodological pointS· adrā raises is worth spelling out. As the end of the passage clearly indicates,he diagnoses the problem of unity as due to the faculty psychological

54 For corroborative passages, see Asfār IV.3.8, VIII.150–151; IV.11.1, IX.270; and IV.11.13, IX.372.55 Asfār IV.8.5, IX.72. 56 Asfār IV.8.5, IX.72–73.


postulation of distinct faculties as theoretical counterparts of distinguishableacts. Thus, despite the vague reference to ‘the ancients’, S· adrā seems to haveShifā’: Fī al-nafs V.7 on his desk here.57 Particularly interesting from thispoint of view is S· adrā’s reservation about Avicenna’s claim that the identitybetween the soul, that is, the entity the functions of which are studied inpsychology, and the self one is aware of is self-evident. Notwithstanding thefact that every human being knows intuitively and indubitably from herown experience that as a subject she persists from onemode of apprehensionor action to another, psychological inferences of the soul’s substantiality onthis basis may not be quite as straightforward as Avicenna claimed. As wehave seen, this suspicion had already been raised by Abū al-Barakāt who wasthen followed by Rāzī and Suhrawardī.58 But S· adrā’s stance is somewhatdifferent. He does not deny the validity of the move as such, but points outthat not everyone is capable of applying intuitive evidence in theoreticalargumentation in the proper manner. For this reason alone, the move ismore involved than Avicenna thought.Moreover, the theory of really distinct psychological faculties can be set

in two alternative relations to the soul’s unity, and the one that Avicennaopted for will make the soul’s unity ultimately unattainable. This is becausehe set out by analysing experience into atomary content units, postulatingon that basis a corresponding distinction in reality between faculties respon-sible for each of them, and only subsequently attempted to bind the discreteunits back into a unified whole that corresponds to our experience.According to S· adrā, we should instead start from the fundamental unityof the soul and proceed to explain the distinction between faculties and therespective organs as a step subsequent to that unity. This method of proce-dure has a parallel in theology: only by taking our cue from God’s absoluteunity can we expect to make sense of His manifold attributes, whereas theinverse order of explanation that begins with attributes that are supposedreally to exist as such will never attain an adequate conception of His unity.The soul’s unity signalled by the self’s unity has to be the foundation andthe starting point of psychology. We must anchor our psychology to thisfundamental unity of the soul and conceive of the distinction betweenfaculties as subsequent to it, a fact that is exclusively due to the soul’s bodythe deficiency of which enables it to reproduce the mental unity of

57 Cf. Avicenna, Shifā’: Fī al-nafs V.7, 252–257 Rahman; and see Chapter 4.1, 66–71. The connection tothis chapter of the Shifā’ is also belied by formulations such as ‘knows that he understands . . . and ischaracterized by a combination of attributes and names’, as well as the explicit discussion of the validityof psychological inferences on the basis of this aspect of self-awareness.

58 See Chapter 5.2.


experience only through a structural unity of spatially distinct organs.59 Yetdespite this difference from Avicenna, it is clear that S· adrā has no objectionsto his description and application of self-awareness in the argument fromunity. He merely believes that the phenomenon can function in theexplanatory role devised for it only if the faculty psychological approach isassigned to its proper realm of relevance.

To conclude this discussion of S· adrā’s application of the argument fromunity, let us have a brief look at his treatment of a related counterargumentwhich, though originally presented by Abū al-Barakāt al-Baghdādī, he cullsfrom Rāzī’s al-Mabāh. ith al-mashriqīya.

60 The argument can be summarizedas follows. If the soul’s status as the single agent of all acts in the animatedbody is evidenced by our awareness of ourselves as single subjects and agentsbehind our experience and action, then given that Avicenna holds the samesoul to be equally responsible for vegetative acts such as growth anddigestion, should we not be equally aware of ourselves as agents of suchacts as well? But that is evidently not the case; on the contrary, for the mostpart those acts remain below the threshold of any sort of awareness. Thus,such subconscious animate processes as digestion or growth threaten toinvalidate the crucial connection between the experience of self and thetheory of soul.61

S· adrā tackles this challenge by means of three metaphysical principles ofhis own. The first concerns what he calls active knowledge (‘ilm fi‘lī):knowledge is active if it causes an effect, whether or not it entails knowledgeof that effect.62 This rather dense principle expresses the manner in whichimmaterial things, such as the human self, exist and act as causes: theirexistence is self-cognition, and in cognizing themselves they bring abouteffects in the material world. The important point is that knowledge of theeffect is not necessary either for the self-cognition or for the causal power ofthe immaterial things. The second principle states that knowledge (‘ilm) inthe most general sense is a perfection proper to the mental level of existenceand thereby something that material existence is by definition devoid of.63

Because the human soul exists both materially and mentally, it can havecharacteristics that are proper for one mode of existence but not for theother. In this sense acts such as vegetation and growth on the one hand, andself-awareness on the other, fall on opposite sides of the fence. Finally,

59 Asfār IV.8.5, IX.76–77, 80.60 For the original, cf. Baghdādī, Mu‘tabar: al-‘ilm al-t.abī‘ī VI.5, II.319–320. S· adrā’s source is Rāzī,

Mabāh. ith II.2.2.1.5, II.257.61 For S· adrā’s paraphrase of the argument, see Asfār IV.2.5, VIII.77. 62 Asfār IV.2.5, VIII.77–78.63 Asfār IV.2.5, VIII.78.


according to S· adrā’s third principle, knowledge and its object can be asym-metrical in terms of their respective levels of existence. The subconsciousanimate processes exist on the lowest level of material existence, and sincetheir existence is therefore knowledge neither in nor of itself, they can onlybe known when a form represents them for a subject that is capable ofknowledge.64

These principles allow S· adrā to state that the selves we are aware of are thetrue agents of even subconscious animate processes, even though they arenot aware of their agency in these cases in the same manner as they are awareof themselves or such acts of theirs as intellection, perception or deliberateaction. The self’s awareness of itself and its innate desire for the perfectionproper to it causes the subconscious acts. In this sense the self, by beingaware of itself, is aware of the cause of these acts just as it is aware of thecause of its conscious acts. The fact that it is not aware of itself as the agent ofthose acts is due not to any opacity in the self, but rather to the weak degreeof existence those thoroughly corporeal acts are inherently confined to –indeed, as far as the mode of existence proper to cognitive phenomena isconcerned, they border on non-existence. They are, as it were, inconse-quential material concomitants of the self’s existence, not unlike reflectionsof the Sun’s light, which although caused and fully dependent on theirorigin, are incapable of penetrating to the level of its intensity.S· adrā’s answer to the evidence that was designed to counter the argument

from unity shows that he approaches the phenomenon of self-awareness in anew conceptual framework. This was also signalled in some of the qual-ifications we have seen him make to otherwise familiar arguments hingingupon the phenomenon. But they also show that he latches on to a traditionof sustained discussion that stays within the framework established byAvicenna’s original definition of self-awareness. As I believe the wealth ofshared material shows, all the additions and qualifications notwithstanding,the concept of self-awareness with which S· adrā operates amounts to thesame very narrow concept of first-personality that we have encountered inAvicenna and Suhrawardī.

7.2 The complicated evidence of self-awareness

The familiarity of S· adrā’s discussion of self-awareness goes beyond a meredescription of the phenomenon, for he casts self-awareness in a theoretical

64 Asfār IV.2.5, VIII.78–79.

The complicated evidence of self-awareness 181

role that resembles those we have considered in both Avicenna andSuhrawardī. This debt is signalled by the following passage:

We know through our intuition (bi wijdāninā) that when we apprehend ourself (dhātinā) we may be unaware of all universal concepts and signs, not tomention the concept of substance or rational, or other [concepts]. Whateverwe know of these things, we do not refer to it as ‘I’, and from this it is knownthat all is absent from us except our simple itness, and no doubt this simpleitness is nothing but existence. Whatever is other than it no doubt falls underone of the categories and is composed of universal things. Existence is notlike that, for as has repeatedly come up, it does not enter under a universalmeaning, even if many of those meanings are true of it.

Herefrom emerges the remark by one of them to the people – when theyassert the soul’s separation by [the fact] that we are unaware of the body andother corpora as well as their accidents while not unaware of our self(dhātinā), and so our self (dhātunā) is a separate substance, no body or anyof their accidents – where he says in opposition that we often apprehend ourself (dhātanā) without the meaning of separate substance having occurred toour mind at all, how then can our self (dhātunā) be identical to a separatesubstance?65

The central claim that self-awareness is nothing but existence alreadycomes close to Suhrawardī’s denial of substantiality behind the phenom-enon,66 and S· adrā pays his dues in the anonymous reference to the shaykhal-ishrāq in the second paragraph. Even if we were forced to classifyourselves under the category ‘substance’ in a self-reflective act, we mustnot infer that substantiality is a real, observer-independent feature of ourself-awareness. However, S· adrā applies the phenomenon with anAvicennian twist, for he does seem willing to allow for a mode of existencethat is not equivalent to self-awareness. Rather, self-awareness is first andforemost a characteristic of mental existence in much the same way asAvicenna conceived it as exclusive to intellectual, and therefore immaterial,entities. But the S· adrian concept of mental existence is articulated in asystem that renders it fundamentally different from Avicenna’s concept ofimmaterial human existence, which cannot but have important consequen-ces for his understanding of self-awareness. In order to get a grasp of thisaspect of S· adrā’s thought, we have to start from one of his most famous – ornotorious – ideas, the theory of change in the category of substance.

Islamic philosophers seem to have almost univocally subscribed toAristotle’s thesis according to which motion or change takes place only in

65 Asfār III.1.3.1, VI.150; emphases added. 66 Cf. Chapter 5.2, 118–123.


categories other than substance.67 Avicenna, for instance, accepts motionwith respect to quality (alteration), quantity (augmentation and diminu-tion, densification and rarification), place (locomotion) and position (forinstance, the motion of the celestial spheres).68 All these types of motion,like their respective categories, take place in or by means of something else,namely a substance that is their subject. Substances themselves, on the otherhand, of course come and cease to be, but this is not a change properlyspeaking. It is not a temporal process with an intermediate state of perfec-tion which could be correctly called motion, but an instantaneous, dura-tionless replacement of a form by another in a material substrate. Thus, thegeneration of a substance is an instantaneous move from pure potentiality(with regard to the emerging form) to full actuality (of the same form),69

and since there is nothing actual in relation to which we could mark thesetwo moments as phases in a continuum, the move does not count as amotion properly speaking.S· adrā departs from the mainstream view by constantly resorting to his

postulation of substantial change (h. araka jawharīya). This idea diametri-cally contradicts the traditional claim: change is possible, and even primarilytakes place, in the category of substance. Yet despite this departure from thetradition, S· adrā’s claim is an expression of the fundamentally Aristoteliantenet that all natural processes are teleological, or ultimately based onteleological processes, innately directed towards a specific goal. Every nat-ural thing, having come to exist, by its very nature strives to exist well,pursuing a manner of perfection proper to it. But whereas this was tradi-tionally conceived as change within the set of the concomitant accidents, asdistinguished from completely fortuitous accidents, of the substance withthe substance itself remaining static, S· adrā understands it as a process withinthe very category of substance, a process that concerns the substance as awhole down to its very core.The most impressive example of substantial change is provided by the

human development from an exclusively material embryo to a perceivinginfant, and ultimately to an adult with a more or less perfectly developedintellect. In making sense of this development, S· adrā subscribes to theAvicennian idea according to which individual human souls first come to

67 Cf. Ar. Phys. V.2, 225b10–12; and Avicenna, Shifā’: al-Samā‘ al-t.abī‘ī II.3.1–6, 136–141. Avicenna doesmention, as a conceptual possibility, the ‘extreme’ view according to which there is motion insubstance, but in the end this merely amounts to calling generation and corruption a kind of motion,that is, to an improper use of the term (Shifā’: al-Samā‘ al-t.abī‘ī II.2.1, 128).

68 Avicenna, Shifā’: al-Samā‘ al-t.abī‘ī II.3.7–20, 141–151.69 Avicenna, Shifā’: al-Samā‘ al-t.abī‘ī II.3.2, 136–137.


be because of their connection to a material body. Of these two constituentsof early human existence, the soul is of course the ontologically higher one –that which gives the human being a stable identity in the face of thecontinual transformation of its material basis70 – but its superiority not-withstanding, the soul still owes its individual genesis to the body it governs.All of the foregoing is familiar to any student of Avicenna’s psychology, but,unlike his predecessor, S· adrā thinks that in this early stage of its existencethe soul is a material form pure and simple, not at all different from theforms of minerals or plants in this respect.71 The difference is that the soulsof neither human nor non-human animals are not restricted to exist asmaterial forms. Rather, it is part and parcel of the substantial change properto them that they develop in perfection until they reach the level of mentalexistence (wujūd dhihnī), really distinct and independent from materialexistence. This happens when the soul becomes capable of perceptual, andsubsequently imaginative and intellectual, apprehension, and it is only atthe acquisition of this level of existence that self-awareness is introduced.

Thus, at the beginning of its existence as a material form, the human soulis not aware of itself. This is because self-awareness amounts to the presenceof the self-aware entity to itself. Now, the only presence a material entity canhave to anything, whether to itself or another, is the spatial relation ofproximity or contiguity. But neither proximity nor contiguity allows amaterial entity to have a holistic relation to itself; we can of course imaginea material thing folding upon itself like a piece of cloth, but it is obvious thatsuch a folding will be neither complete (for no part of the cloth will be incontact with every other part) nor properly described as a relation of theentity to itself (only a contact between its parts). These limitations, inherentto spatiality, preclude any self-relation, let alone presence to self, on thematerial level of existence. Moreover, S· adrā states that strictly speakingthere is no presence of any material thing to any other material thing, forspatial juxtaposition does not entail presence unless one or both of thejuxtaposed things apprehends the juxtaposition.72

The situation is completely different in the case of immaterial, or mental,existence:

Whatever exists incorporeally occurs to itself (h. ās.ilun li dhātihi), for its self(dhātahu) is not veiled from itself (dhātihi). Thus, it understands itself(dhātihi), for knowledge is the same as existence provided that there is noveiling. But the only veil there really is is non-existence, and likewise, non-

70 Asfār IV.7.1, VIII.380–384. 71 Asfār IV.7.2, VIII.385; cf. IV.7.4, VIII.440.72 Asfār I.10.2.1, III.483.


existence of the veil comes back to a confirmation and intensification ofexistence until there is no weakness borne by privation, which is a kind ofnon-existence.73

This passage reads quite naturally as an expression of the traditional doc-trine according to which immateriality entails intellectuality, and intellec-tuality in turn amounts to self-intellection. But I would argue that S· adrā’sexclusive regard to intellection here is primarily due to the context, which isdevoted to a discussion of intellection. As we have already seen in such casesas the flying animal, S· adrā transformed the traditional idea to hold true of allapprehension: whatever apprehends exists mentally, and since mental exis-tence is immaterial, the subject apprehending any object whatsoever willthereby apprehend herself.So far so good: S· adrā departs from Avicenna by taking the material

constituent of human existence to be essential to it in the beginning, buthe subscribes to the latter’s dualismwhen it comes to the mental existence ofthe same human being. This similarity naturally gives rise to two questions.First, does S· adrā identify mental existence with self-awareness in the man-ner of Avicenna? And if he does, is this in order to deal with the problem ofthe individuation of immaterial existence that we found looming inAvicenna?The question of individuation does indeed come up in the psychological

section of the Asfār in an argument for the thesis that the individual humansoul is generated in time. The discussion is particularly interesting because itincorporates, by way of refutation, Fakhr al-Dīn al-Rāzī’s critical remarks onAvicenna. In al-Mabāh. ith al-mashriqīya fī ‘ilm al-ilāhīyāt wa al-t.abī‘īyāt,Rāzī presents six arguments against Avicenna’s demonstration of the indi-vidual human soul’s generation in time.74 As we have seen, Avicenna’sdemonstration was based on the Peripatetic idea that distinctions betweengenerated individual entities are always due to matter, and so, in order tobecome instantiated in multiple distinct persons, the human essence com-mon to them all must be related to a corresponding number of bodies.75 In

73 Asfār I.10.2.1, III.483. Cf. I.10.2.1, III.484–485, where the second argument for the self-awareness of allimmaterial entities is based on essentially the same idea. This argument dwells at some length uponthe rejection of the claim that presence to self entails a real relation. Instead, S· adrā states that therelation is merely apparent and due to our manner of speech; in reality the self as that which is presentand the self as that to which it is present are one and the same thing.

74 Rāzī, Mabāh. ith II.2.2.5.5, II.401–403. S· adrā quotes the arguments at length in Asfār IV.7.2,VIII.389–391, but he seems to attribute them to Rāzī’s Mulakhkhas. fī al-h. ikma. Unfortunately, thiswork has not been edited, and I have not been able to find the arguments in the sole manuscriptavailable to me (Berlin Staatsbibliothek Or. Oct. 623).

75 Shifā’: Fī al-nafs V.3, 223–227 Rahman; see Chapter 3.1.


his sixth and final argument Rāzī brings forth the case of souls that havebeen actualized as individuals by being generated through relations to theirrespective bodies but that, having passed away in infancy, have failed toacquire any further distinguishing characteristics. According to Avicenna,even such souls will have some kind of individual existence, albeit a very‘thin’ one, after the demise of their bodies.76 But, Rāzī counters, thisamounts to saying that they subsist as mere material intellects, with noaccidents to account for their individuality. In other words, they are statedto be individuated by themselves (lā yakūna fīhā shay’un min al-‘awārid

˙i illā

mujarrada dhātihi). Yet if this suffices for individual existence in the after-life, why cannot souls be individuated by themselves before their connec-tion to bodies?77

At first glance, Rāzī’s paraphrase seems rather uncharitable to Avicenna’saccount of body-induced accidents in the immaterial soul. However, hemay have been guided by the insight behind his second argument againstthe generation of souls, which can be read as a rephrasing of Avicenna’sdiscussion of individuation in Shifā’: al-Madkhal I.12.78 Since all attributesof an immaterial intellect are universal by definition, Rāzī can reconstructthe situation of those deceased in infancy in terms of the material intellect.This would tally well with the generous return he grants to Avicenna.Perhaps the thesis of the temporal individuation of human beings can besalvaged by recourse to their self-awareness: even undeveloped souls, whoseexistence borders on nothingness, will be aware of themselves in a uniqueand unshareable manner (li kullin minhā shu‘ūran bi huwīyatihā al-khās.s.a).Thus, their state would resemble that of the flying man; they would indeedexist in an extremely narrow sense, but this would nonetheless be a genu-inely individual existence. Rāzī’s clever answer turns on the ambiguity of thepivotal term dhāt. Earlier on, Avicenna has identified a thing’s awareness ofitself with the very dhāt of that thing (shu‘ūru al-shay’i bi dhātihi huwa nafsudhātihi), that is, with its self or essence; the human soul cannot exist withoutbeing aware of itself because self-awareness is a necessary constituent of theexistence proper to human beings. But if the soul’s self-awareness is iden-tical with its dhāt, then souls must differ by their very essence! And if that isthe case, individual souls will be essentially distinct before their respectiveconnections to bodies.79

76 In Najāt III.2, 333, Avicenna describes them as ‘coming to a wealth of God’s mercy and a kind ofrest’.

77 Rāzī, Mabāh. ith II.2.2.5.5, II.402–403; Asfār IV.7.2, VIII.391.78 See Chapter 3.1, 48–50; cf. Rāzī, Mabāh. ith II.2.2.5.5, II.401; and Asfār IV.7.2, VIII.389.79 Rāzī, Mabāh. ith II.2.2.5.5, II.402–403; Asfār IV.7.2, VIII.391.


In his conclusion S· adrā attempts to save Avicenna’s insight according towhich self-awareness plays a crucial role in the individuation of immaterialentities, but not without giving Rāzī’s perspicacious point its rightful due.As we recall, Avicenna held that connection to a proper body immediatelybrings forth an immaterial substance whose subsequent existence is inde-pendent of that body and amounts to its awareness of itself. As a result, theself-awareness in question is of an extremely narrow, indeed contentlesskind, and it is precisely this feature that Rāzī seizes upon. S· adrā, on thecontrary, believes that Rāzī’s challenge can be met if Avicenna’s insight isincorporated into the theory of substantial change.Let us rehearse the essential phases of human development. Over the

course of her substantial motion, a single human being comes to exist invarious distinct modes. In first coming to be, she is a material form thatdiffers from the forms of minerals or plants only with respect to hercognitive and motive potencies. When these potencies are actualized, andshe begins to perceive and to move voluntarily, informed by what she hasperceived, she comes to exist in the mental mode.80 From this moment on,the human being is an immaterial entity whose existence is always charac-terized by self-awareness, or more accurately, amounts to self-awareness.But having first been individuated as material forms, human souls arealways already determined as individuals by a multitude of attributes thatthey inherit from the period they spent inhering in matter. Only after thatperiod,

it then follows that each of them is determined by their individual existence(thumma yalzamu ta‘ayyunu kullin minhā bi wujūdihā al-khās.s.), which isidentical to their self-awareness (‘aynu shu‘ūrihā bi dhātihā), and that is whatendures permanently though with a kind of existential renewal.81

Thus, S· adrā’s strategy in meeting Rāzī’s challenge is to incorporate theaccidental features due to the body into mental existence as necessaryconstituents of the self-awareness uniquely exclusive to each humanbeing. Although self-awareness, in the narrow sense of first-personality,does not vary from one person to another, the experiential content andpersonal attributes that we inherit from our bodies render us individualsubjects. On the other hand, S· adrā does recognize the need, as expressed inShifā’: al-Madkhal I.12, to account for the individuality of these features that

80 Asfār IV.7.3, VIII.430–432; IV.7.6, VIII.449–450; IV.8.2, IX.31–32; IV.9.4, IX.127; IV.11.9,IX.317–318. In IV.9.5, IX.150, S· adrā specifies that mental existence begins at some point during thefourth month of pregnancy.



now belong to an immaterial subject. His solution is that the body-inducedattributes, even when divested of their spatiotemporal co-ordinates, remainindividual because they are given in first-personal perspectives as in eachcase uniquely mine. Mental existence thus necessarily consists of both self-awareness and its various determinations.

As will become forcefully clear below, when we turn to discussing theconsequences of the theory of cognitive unity for his concept of self-awareness, S· adrā opts for the phenomenologically plausible view that weare always aware of ourselves as engaged in an act or undergoing anexperience, not as naked first persons divested of all predicates. Thisplausibility, however, does not come without a price, which is cashed outas a serious weakness in S· adrā’s final stance on human individuality. If self-awareness is individuated by its body-induced content, and that content inturn by belonging to an individual self, S· adrā’s account of the individuationof mental existence turns out to be viciously circular. In this regard it seemsthat Avicenna was not entirely misguided in emphasizing the primitive andirreducible nature of individual human self-awareness. But notwithstandingthe circularity in his reasoning, it is clear that S· adrā’s discussion of thisparticular application of self-awareness is witness to his thorough familiaritywith the subtleties of the traditional concept. What is more, this familiarityis shown compellingly in the treatment of another inherited question.

As we recall, Avicenna relied on self-awareness in an inference to thesubstantiality of the human self: because I can be aware of myself whileunaware of all other things, my body in particular, I am independent ofthose things, my body included, and therefore I exist as an immaterialsubstance. As we have also seen, the inference to the self’s substantiality wassternly criticized by Abū al-Barakāt and Rāzī, and its conclusion wasultimately rejected by Suhrawardī.82 S· adrā latches on to the debate, again,by means of Rāzī. The argument in question is the one we have alreadyexamined in the course of our discussion of Suhrawardī’s critique of thehuman self’s substantiality. It seems that Rāzī sets out to contest the flyingman’s argumentative power in the context of Shifā’: Fī al-nafs I.1. In thatchapter, the thought experiment was designed to make us pay attention toour self-awareness, which in turn would readily indicate the substantiality ofthe self we are aware of while unaware of the body.83 But according to Rāzī,nothing in my self-awareness forces me to lean on one view rather thananother; yet if the self really is a substance, I should be aware of my

82 See Chapter 5.1. 83 See Chapter 2.1.


substantiality whenever I am aware of myself.84 That this is not the case isclear from the bare fact that philosophers debate the issue.S· adrā’s manner of saving the Avicennian inference in the face of Rāzī’s

critique follows a pattern pervasive in the Asfār: we have to distinguishbetween the first-order act of existence, here constituted by human self-awareness, and the second-order act of an intellectual apprehension of it. Asan act of mental existence, a mode of presential knowledge, self-awareness isa state of mental existence which is able to provide the basis for manypossible kinds of conceptual analysis without being actually constituted bydiscrete constituents corresponding to the results of the analysis. Rather, theanalytic act of intellection introduces the distinctions into what is one tobegin with; although the fact that the act of existence in question has a givendegree of perfection does provide the basis for those distinctions, forinstance for conceiving the corresponding existent as a substance of somesort, the existence in itself is not a composite of substantiality and a given setof specific differentiae.85 To take Avicennian primitive self-awareness as anexample, it is different from and prior to any act of reflective self-intellection(such as the act of performing a thought experiment), which involvesapplication of universal concepts and therefore does not entail awarenessof oneself as a substance. But in spite of this, it is still sound to say thatsubstantiality is necessarily and self-evidently introduced in the act of self-intellection, because the existent that one finds when reflecting on oneselfturns out to be an independently subsisting essence for the kind of which‘substance’ is the summum genus. Regardless of which aspect of myself Icome to focus upon, whether my humanity, my intellectuality or my agencyin and through my body, I can only perform this consideration by conceiv-ing of a genus and a differentia, and upon further analysis these will beshown to entail the most general concept of substance as their foundation.In such an analysis, the fact that substantiality is not given in the first-

order act of self-aware existence need no longer be fatal to Avicenna’sargument. In order to function in the role Avicenna had devised for it, itis enough if our first-order self-awareness is such that our second-orderintellectual attention to it will necessarily entail the concept of substance.86

This entailment itself, although necessary, need not be obvious, and I donot believe Avicenna himself thought it is, for it is precisely in order to

84 Rāzī,Mabāh. ith II.2.2.1.3, II.246–247; Asfār IV.2.3, VIII.50. S· adrā also mentions the critical point inAsfār III.1.3.1, VI.150.

85 See Asfār I.1.1.7, I.78–79. A most lucid account of this aspect of S· adrā’s metaphysics is Bonmariage2007, 43–47.



render it such that he brings forth arguments like the flying man, designedas it is to rule out the possibility that the self inheres in a body. Thus, thenecessity is not to be understood in psychological terms, for the very fact ofour disagreement about the question suffices to refute it. On the contrary,we are dealing with a logical necessity which can and will become obvious insound reflection. If one fails to grasp it, one has simply not reflected uponone’s self-awareness in a correct manner but has rather focused on one oranother of its accidental features, such as the seemingly constant presence ofthe body and one’s agency in and through it.

Nevertheless, it is important to realize that S· adrā subscribes to Avicenna’sinference only under these qualifications. When it comes to the founda-tional level of primitive self-awareness, he seems rather to side withSuhrawardī’s deflationary position, as the following conclusion shows:

when a human turns to himself (raja‘a ilā dhātihi) and makes his itnesspresent (ah. d

˙ara huwīyatahu), he may neglect all universal meanings, even the

meaning of his being a substance, an individual, or a governor for the body.Thus, in examining myself (‘inda mut.āla‘ati dhātī) I only see an existencewhich apprehends itself as a particular (wujūdan yudriku nafsahu ‘alā wajhial-juz’īya), and whatever is other than the individual itness to which I refer by‘I’ is external to myself (dhātī), even the concept ‘I’, the concept of existence,the concept of what apprehends itself (al-mudriki nafsahu), and the conceptof governor of the body, or soul, and so forth; for they all are universalcognitions, to which all I refer by ‘it’, whereas I refer to myself (dhātī) by ‘I’.Thus, neglect or ignorance of substantiality does not refute that it is one ofthe predicates that are essential for the quiddity of a human being, and also ofan animal; and it is possible to apply the Master’s [that is, Avicenna’s]discussion here so that the aforementioned contradiction is expelled.87

Even if it can only be understood as the existence of a substance, first-orderself-awareness as such consists in nothing but being an I, and it involves noconsideration of what kind of thing this I is or what its quiddity is. Thesecond-order reflective consideration of this I as an instantiation of somequiddity differs from the first-order self-awareness because it regards the I asan object, as an it, not as the I it was to begin with. It does this byintroducing one of the many possible universal concepts under which theI is subsumed, whether that of a human, individual, soul, existence, orindeed subject or self (the concept ‘I’). That one thereby necessarily alsointroduces the concept of substance, which is the highest genus of anythingsubsumed under a specific concept, suffices to save Avicenna’s argument,

87 Asfār IV.2.3, VIII.50–51.


but this does not hide the fact that S· adrā’s analysis is tacitly derivative ofSuhrawardī.88 Thus, in the end S· adrā conceives of self-awareness as aparticular mode of existence (that is, mental existence) rather than anindication of substantiality. But on the other hand, he seems reluctant tofollow Suhrawardī to the extreme point of identifying self-awareness withexistence (or appearance and light) pure and simple.89 In this regard, herepresents a middle stance between his two predecessors, an Aristotelianattempt to do justice to both by incorporating them, under the aforemen-tioned qualifications, into his own account.

88 This is betrayed by the distinction between ‘I’ and ‘it’ as well as the mention of the concept ‘I’. S· adrāhere comes very close to Suhrawardī, Talwīh. āt III.5.15, 282–283 Habībī; 115–116 Corbin. For adiscussion of this chapter, cf. Chapter 5.2, 118–121. It also provides the basis for Asfār I.1.10.2.1,III.503–504, where S· adrā endorses the denial of the self’s substantiality in more unqualified terms.However, in this context he does not explicitly address the relevant question, which compromises thedemonstrative value of the endorsement.

89 This assessment must, however, be qualified by noting that there are passages which suggest thatS· adrā holds all existence to entail a more or less developed type of self-awareness (cf. Asfār III.1.3.1,VI.147–148; IV.2.5, VIII.79; IV.4.1, VIII.184; IV.4.2, VIII.192; IV.11.13, IX.358–362, 364–365). Thisquestion, alas, is both so wide and so debatable that I must postpone its treatment to a later occasion.


chapter 8

The self reconsidered: S· adrian revisionsto the Avicennian concept

As our discussion of S· adrā’s sustained strategy shows, he is familiar with thesubtleties in his predecessors’ applications of self-awareness in their respec-tive theoretical concerns. This familiarity shows that when he incorporatesthe inherited concept of self-awareness into his own system of philosophy,he must be conscious of the precise meaning of that concept. Thus, as weturn to discuss his departures from the narrow concept of self-awareness atwork in both Avicenna and Suhrawardī, we can rest assured that thedepartures are made in full awareness.

8.1 The self and cognitive unity

S· adrā’s epistemology can be succinctly characterized as a sustained attempt torehabilitate the ancient theory of knowledge as a unity between the subjectand the object of the actual act of cognition.1 An early version of it wasAristotle’s theory of intellection, which famously holds that the intellect is allthe intelligibles and that actual intellection collapses with the actual existenceof what is understood.2 Later on, the doctrine of cognitive unity became partof the Neoplatonic theory of the intellect as a single indivisible entity, albeitone with an internal structure.3 Yet the venerable provenance notwithstand-ing, the theory may seem strangely counterintuitive at close inspection. Oneof its problems, perspicaciously formulated by Avicenna, is that it contradictsthe commonsense intuition according to which I am perfectly capable ofperceiving or understanding different things at different moments of time. Ifat time tx I understand and am therefore identical to an object x, whereas attime ty I understand and am therefore identical to object y, then either x and y

1 For S· adrā’s epistemology, see now Kalin 2010. 2 Ar. De an. III.5, 430a20; III.7, 431a1.3 This version of the theory is most prominently present in the crypto-Proclean Kitāb al-īd

˙āh. fī al-khayr

al-mah. d˙(see, for instance, XII, 14–15, which corresponds to Proclus, El. Th. 167–169) and the Plotinian

Theology of Aristotle (see, for instance, II.21, 32, which corresponds to Plot. Enn. IV.4.2.4–8; cf.Adamson 2002, 120–121, 152).

192

are transitively identical or I do not endure as a substance from tx to ty. Thefirst option is obviously false (an apple is not an orange), and the second entailsa rejection of the eminently plausible view according to which the subjects at txand ty are bothme in one and the same sense. According to Avicenna, our onlyalternative is to deny the theory of cognitive unity, a suspicious innovationwhich he disparagingly attributes to Porphyry’s intent ‘to speak in imagina-tive, poetic and mystical (s.ūfīya) expressions’,

4 although he does allow thatGod’s intellection, owing to His absolute unity and immutability, may be aspecial case for which the theory of cognitive unity is capable of giving its due.Yet in spite of such problems, S· adrā holds the theory of cognitive unity to

be true not only of intellection, but of all modes of cognition, including themost elementary types of sense-perception.5 Addressing the case of intellec-tion, he states that the existence of each immaterial form that is actuallyunderstood is identical with its existence for the corresponding subject ofintellection. In other words, the intelligible ‘human’ that I am thinking ofright now is nothing but my act of understanding; the act of understandingexhausts the intelligible existence of the human being. An intelligible canonly exist actually if it is actually understood, and this requires a subject ofunderstanding actually understanding that very intelligible. Thus, the twoare not really separable from each other.6

This insight is not exclusive to intellection but pertains to all types ofcognition:

What is sensed . . . is divided into what is sensed potentially and what issensed actually, and what is sensed actually is united in existence with theactually sensing substance.7

S· adrā corroborates his thesis by a brief argument against alternative theoriesof perception, that is, the theory of the impression of forms from thematerial objects in the perceiving soul, the extramission theory of vision,8

and the theory of primitive relationality between the subject and the object

4 Avicenna, Shifā’: Fī al-nafs V.6, 239–240 Rahman5 S· adrā seems to think that once his theory of substantial change has been established, the solution to theproblem raised by Avicenna will be all but self-evident. If any entity subject to change has a staticidentity only when the process of change is considered from an extratemporal perspective as a four-dimensional whole (time being one of its defining dimensions), it will no longer be a problem if anentity, even the human intellect, does not endure unchanged from one act of intellection to another.The intuition of the subject’s enduring identity can be saved by recourse to S· adrā’s view that laterphases in the process of substantial change include the earlier ones by transcending them. I will discussthis topic in greater detail in Chapter 8.2.

6 Asfār I.1.10.1.7, III.340–341. 7 Asfār I.1.10.1.7, III.342; cf. IV.10.7, IX.200.8 According to the extramission theory, endorsed by the optician Ibn al-Haytham as well as somephilosophers, such as Abū al-Barakāt al-Baghdādī, the act of vision consists in the emission of extremely

The self and cognitive unity 193

of vision held by both Rāzī and Suhrawardī.9This is followed by his positiveaccount of perception:

On the contrary, sense perception occurs so that an apprehensional lumi-nous form, of which there occurs apprehension and awareness, is emanatedfrom the giver (al-wāhib),10 so that it is both the percipient in act and theperceived in act; prior to that there is neither a percipient nor a perceivedexcept in potency. As regards the existence of a form in proper matter, [thematerial form] is a preparation for the emanation of that form which is theperceived and the percipient in act. The discussion of this form’s beingperception, percipient and perceived by itself (bi ‘aynihi) is like the discussionof the intellectual form’s being intellection, that which understands and thatwhich is understood.11

The passage is remarkably unambiguous: the existence of the per-ceived form, just as that of the form understood, is nothing but itsbeing perceived which in turn is unqualifiedly identical to the corre-sponding act of perceiving. We should, however, pay careful attentionto S· adrā’s somewhat oblique statement towards the end of the passage,namely, that cognitive unity only concerns the mental mode of exis-tence.12 While not an invention of S· adrā’s, mental existence is a conceptthat he articulates at a considerably greater length than any of his

subtle yet nonetheless material rays from the eye. When these rays encounter extramental objects, theybring back information to the eye, which then results in vision. For an extended discussion of theimpressionist and extramission theories of perception in S· adrā, see Kalin 2010, 118–135.

9 Asfār I.1.10.1.7, III.342. These critical comments are further developed in the psychological section ofthe Asfār (see IV.4.6, VIII.210–212), where S· adrā explicitly refers to the present context.

10 This is a reference either to the active intellect, which was standardly referred to as the giver of forms(wāhib al-s.uwar), or more vaguely to the immaterial origin of both the human soul and all materialthings, and thereby ultimately to God as the origin of all (Asfār IV.4.6, VIII.212 speaks of ‘the powerof God’). This ambiguity notwithstanding, the crucial point is clear: the perceived, imagined andunderstood forms, that is, the content of experience as experienced, are not caused by external materialthings but brought about by the immaterial origin of the soul. In this sense, all experience is generatedfrom within the soul.

11 Asfār I.1.10.1.7, III.342–343; cf. IV.1.1, VIII.18; IV.2.5, VIII.74; IV.3.3, VIII.98; IV.4.5, VIII.208–209;IV.4.6, VIII.212; and IV.8.7, IX.92.

12 Cf., however, Asfār IV.3.9, VIII.155, where S· adrā says that in its descent to govern the body, the soul‘becomes, for instance, in touching identical with the touching organ, and in smelling and tastingidentical with that which smells and tastes’ (my emphasis). I believe that this passage can beinterpreted in a manner that salvages S· adrā’s consistent distinction between the material and themental. As a material form, the soul is of course responsible for the proper functioning of the cognitiveorgans, and in a sense those organs in their action can be identified with their form. Yet while thisaspect of the soul’s action may be parallel to the mental act of perception, it is not an instance ofcognitive unity in the sense expounded here, but an instance of the soul’s action as form that is inprinciple not different from its corresponding action in the organs of digestion, for example. S· adrā’sfocus in this context is precisely on the soul’s descent to matter, and for this purpose it may be relevantto emphasize the material circumstances parallel to but not causally active upon the mental event ofperception.

194 The self reconsidered

predecessors,13 and that is pivotal to his explicit distinction between thetwo possible descriptions of perception. We can of course describeperception by means of the material process that takes place in theperceptual organs of the body, but such a description will not grasp theexperiential reality, or the mental phenomenon, of perception. In thislatter sense, perception is an indivisible act of mental existence, thesubject of which cannot be really distinguished from its object, becausethe subject only exists in unity with the object as the subject perceivingthis object, and inversely the object only exists as an object for thissubject. This is corroborated by S· adrā’s comparison, towards the end ofthe chapter devoted to the unity theory of knowledge, of the relationbetween the subject and the object of cognition to the hylomorphicrelation between matter and form. The two are relations only in aqualified sense and must be distinguished from relations proper,which prevail between two independently subsisting things such as ahouse and its inhabitant, property and its owner, or a parent and heroffspring, precisely because in them the relata are interdependent inexistence. Matter can only exist as actualized by form just as formrequires matter for its subsistence; similarly, just as there can be noobject of cognition without a subject to cognize it, the subject ofcognition needs something to cognize in order actually to be such asubject.14

Finally, S· adrā’s account of mental existence denies any causal power toextramental objects, even in the case of perception.15 The content ofperceptual experience is produced by them only in the sense that theyprovide the circumstantial conditions for the emanation of experientialcontent from the higher principle that is the origin of both the soul itselfand its content.16 As a consequence, S· adrian mental existence is thoroughlydetached from matter. Even though its objects may have a spatial structureand location and in this sense be analogous to material things, their

13 For an overview of S· adrā’s arguments for mental existence, see now Marcotte 2011. Knowledge isdiscussed as a mode of existence in Asfār III.1.3.1, VI.143–144.

14 Asfār I.1.10.1.7, III.345; cf. I.1.10.2.3, III.498–499.15 Asfār IV.4.6, VIII.212–213; cf. IV.4.7, VIII.217; IV.11.13, IX.376.16 Asfār IV.10.5, IX.189; IV.10.10, IX.245–246. In true Neoplatonic fashion, this source is also conceived

to be the soul’s goal in its development towards the perfection proper to it, which S· adrā seems toidentify with the active intellect. In Asfār I.1.10.2.3, III.498–499, S· adrā says that there ‘should be’(yanbaghī an yakūna) a principle for sense-perception analogous to the active intellect in the case ofintellection. Finally, in Asfār IV.4.12, VIII.239–240, he states that objects of perception depend onimagination for their subsistence, whereas the objects of imagination depend similarly on theintellect. The idea is that the higher principles are thereby potentially present in the actuality of thelower – what is perceived can also be imagined and ultimately understood.


spatiality is not material but experiential, a matter of either imagination orperception.17

The consequences of the theory of cognitive unity to S· adrā’s discussionof self-awareness are clear. As Avicenna must have perceived when he setout to argue against the theory, the claim of a strong unity between the selfas the first-personal subject of experience on the one hand, and theobjective content of its experience on the other, amounts to a denial ofthe reality of the narrow type of self-awareness endorsed by both Avicennaand Suhrawardī. If the I is one with its determinations, we can neverencounter pure I-ness divested of all its possible acts and passions.I would now like to argue that S· adrā is clearly aware of this consequenceby means of two separate but mutually corroborative discussions that bearon the topic. The first consists in an extended discussion on self-awarenessthat S· adrā engages in in order to lay out the principles needed for under-standing God’s knowledge, while the second is a discussion of the soul’sfaculties from the psychological section of the Asfār.

The fifth principle S· adrā presents as requisite for making sense of God’sknowledge of Himself and the world of His creation focuses on theimmediacy of our awareness not just of ourselves in the narrow sense, butalso of our various faculties:

Just as the soul apprehends itself (dhātahā) by means of the very form of itself(bi nafsi s.ūrati dhātihā), not by means of another form, similarly it appre-hends many of its apprehensive and motive faculties, not by means ofanother mental form.18

The idea that our faculties are immediately given, or present, to us is ofcourse familiar from Suhrawardī’s introduction of the concept of knowledgeas presence. That S· adrā is relying on a Suhrawardian source here, mostlikely al-Mashāri‘ wa al-mut.ārah. āt III.1.7, is suggested by substantial sim-ilarities between his arguments for the principle and those Suhrawardībrought forth in that chapter.19 But instead of the similarities, let us focuson S· adrā’s development of the inherited material.

17 Asfār I.1.10.2.6, III.512–515. In his eschatology, S· adrā puts forth an argument for our embodiedexistence in the hereafter, only the body there is immaterial, an exclusively experienced body (cf. AsfārIV.11.4, IX.303–304; IV.11.17, IX.389–390).

18 Asfār III.1.3.1, VI.150–151.19 Suhrawardī, Mashāri‘ III.7.1 is discussed in extenso in Chapter 6.1, 134–141. For the similarities

between the two texts, compare Asfār III.1.3.1, VI.151–152 (S· adrā’s second argument) withSuhrawardī, Mashāri‘ III.7.1, 484; Asfār III.1.3.1, VI.152 (S· adrā’s third argument from the evidenceof pain) with Suhrawardī, Mashāri‘ III.7.1, 485; and finally, Asfār III.1.3.1, VI.153–154 (attention as acondition of actual apprehension) with Suhrawardī, Mashāri‘ III.7.1, 485.


The first thing to notice is that S· adrā’s formulation of the principle is notentirely unambiguous. Are the other faculties apprehended simply by beingaware of oneself, that is, as necessary constituents of one’s self, or are theyapprehended by themselves, as an immediate consequence of their operation,so that awareness of them is merely structurally analogous to self-awareness,not caused by it? Dissolving the ambiguity is pivotal to our topic, for the firstalternative would clearly be incompatible with the venerable argument of theflying man and the narrow concept of self-awareness it was designed tocorroborate. The second alternative, on the contrary, would merely amountto saying that by operating one’s cognitive faculties one is necessarily aware ofthem, and thus it would be perfectly coherent with the flying man, for it isprecisely the non-operation of the faculties that the thought experimenthinges upon. Fortunately for us, S· adrā substantiates the principle with fivearguments and an explication of considerable length, which provide the keyfor unravelling the ambiguity. The first four arguments deal with the imme-diacy of sense perception in general terms and therefore do not immediatelybear upon our topic. But a fifth, thronal (‘arshī) argument, that is, one basedon S· adrā’s own intuition,20 is more consequential:

[I]n the beginning of its creation, the soul is devoid of both conceptual(tas.awwurīya) and assentual (tas.dīqīya) knowledge. There is no doubt thatthe employment of instruments – such as the senses – is a voluntary act, not anatural act, and so it undoubtedly depends on knowledge of those instru-ments. If all knowledge were through the impression of a form from what isknown, then as a result its dependence on the employment of instrumentswould depend on knowledge of those instruments, and so the discussionwould be reiterated; and both circle and regress are impossible.21

S· adrā starts from a denial of innate knowledge, stating that the acquisitionof knowledge requires a knowing use of cognitive faculties and organs. Theproblem is, if one is not immediately aware of the faculties to begin with, ifone is completely lacking in innate knowledge about how to operate them,one will never get underway with the learning process. Were the familiaritywith the faculties not innate, then in order to learn that one has the faculties,let alone how to operate them, one would have to acquire knowledge aboutthem. Since those very faculties are the means by which one acquiresknowledge of anything, one would have to acquire knowledge that onehas cognitive faculties, and learn how to use them, by having, knowing andusing them, which of course results in either a vicious circle or an infiniteregress. Thus,

20 For the term ‘arshī in S· adrā, see Nasr 1978, 56, and Corbin 2009a, liii–liv. 21 Asfār III.1.3.1, VI.152.


by necessity the soul’s first knowledge is its knowledge of itself (bi dhātihā),and then its knowledge of its faculties and instruments, which are theexternal and the internal senses; these two are presential knowledge.22

The conclusion is reminiscent of Aristotle’s account of phenomenal con-sciousness as a concomitant of the basest acts of perception.23 According toAristotle, our awareness that we perceive is indubitable but needs nonethe-less to be given a psychological explanation. Since adding a higher-orderfaculty to perceive that we perceive would result in infinite regress, hesuggests that we close the inevitable reflexive circle at the first possibleinstant and build phenomenal consciousness into our account of first-order sense-perception as one of its necessary constituents.

But in spite of the similarity in argument, S· adrā differs from Aristotle byexplicitly qualifying that it is not the individual faculties as such that weshould conceive to be self-aware. Following Avicenna, he holds instead thatonly the primitively self-aware soul, which operates them all, can have thempresent to itself and thereby be aware of them.24 This unity of subject in allexperience had already been anticipated in the first argument for theprinciple under discussion, where the soul was described as the singlesubject of awareness in the operations of its various faculties, such asthinking, imagination and sense-perception. S· adrā here states that theseoperations are immediately present to the soul, and what is more, theirrespective objects, constitutive as they are to the operations, are present to itthrough or in the operations. This becomes particularly evident in thestraightforward identification of the power of thought with the concreteindividual forms that it has as its objects: ‘what receives these operations andtransformations [that is, the operations characteristic of thinking] is nothingbut concrete individual forms, I mean the power of thought’.25 This is ofcourse a consequence of S· adrā’s broad subscription to the theory of cogni-tive unity, and, as if to underline this, he concludes the argument by sayingthat in the final analysis the soul sees and is aware of the objects of itscognitive operations ultimately ‘by seeing itself (bi bas.ari dhātihā)’.

26 This isbecause the soul’s operation and the objects constitutive to that operationare in turn constitutive to the soul’s awareness of itself as a self engaged inthis particular operation. Thus, in the end S· adrā’s emphasis on the unity ofexperience leads him to depart from Avicenna: engaged in its operations, thesoul is not a detachable I or pure first-personality that can attach to various

22 Asfār III.1.3.1, VI.153. 23 Ar. De an. III.2, 425b11–17. See Chapter 1.1, 12–15.24 Cf., for instance, Avicenna, Shifā’: Fī al-nafs V.7, 253–254 Rahman; and see Chapter 4.2, 66–71.25 Asfār III.1.3.1, VI.151; my emphasis. 26 Asfār III.1.3.1, VI.151; my emphasis.


contents without being dependent on or determined by them. In S· adrā, thesoul is literally one with and inseparable from its acts.This point is made more forcefully in the explication of the fifth argu-

ment. Immediately after the passage quoted above, S· adrā says:

Then after these two knowledges [that is, the soul’s knowledge of itself and ofits faculties], there emerges from the soul’s self to itself (min dhāti al-nafsi lidhātihā) the employment of instruments without any conception of this act,that is, of the use of instruments, or assent of its usefulness, like in the case ofother voluntary acts which originate from us outside the body. Thus, this[employment of instruments] is another type of volition, not by means ofintent (al-qas.d) and deliberation (al-ruwīya), even though it is not separatefrom the knowledge of it; for here volition is identical with knowledge,whereas in other voluntary acts originating from the soul it is preceded by aknowledge of them and by an assent of their usefulness.As regards this act which is like the soul’s employment of faculties, senses

and the like, it emerges from itself (‘an dhātihā), not from its deliberation;thus, its self by itself (dhātuhā bi dhātihā) necessitates the employment ofinstruments, not by means of an additional volition or additional knowledge.Rather, the soul in the very beginning of its creation knows itself (bi dhātihā)and loves [itself] and its action by a love generated from the self (al-dhāt), andit is forced to employ the instruments which is all it is capable of.27

This rich passage not only emphatically corroborates the claim that humanself-awareness is not separable from but built in to our various acts, but alsogives us a clue to how S· adrā thinks the awareness of acts follows from self-awareness. In order to explicate this point, S· adrā makes an interestingphenomenological distinction between two types of voluntary acts. Themore commonplace type involves deliberation and conscious intentionality,and it is exemplified by the all too familiar situation of having to choosebetween two compelling but mutually exclusive goals of action, for instance,whether to head home and spend time with one’s offspring or to stay sitting atthe office in the labour of a delayed paper. Regardless of what would be thecomplete account of the potentially very complex set of reasons and motiveswhich contribute to making the eventual choice, it seems safe to assume, asS· adrā does, that this type of conscious volition involves minimally twocognitive acts: a conception (tas.awwur) of the alternative courses of action,and a belief in or assent to (tas.dīq) their respective values.

28 It is because ofthese explicit acts of conception and belief that the choice can be called aconscious one, or as S· adrā says, performed ‘by means of intent and deliber-ation’. But having cleared that, we must note that S· adrā describes the more

27 Asfār III.1.3.1, VI.153. 28 For tas.awwur and tas.dīq in S· adrā, see Lameer 2006.


familiar case only in order to characterize the other, less commonplace type ofvoluntary action and that it is precisely this other type that is involved inthe explanation of how the human soul’s self-awareness entails its awarenessof its operations. This less familiar type of voluntary action is not character-ized by explicit deliberation and belief but is rather identical to theknowledge of actually performing the voluntary act. Since in the absence ofexplicit conception of the pursued goal there can be no volition in the sense offree choice, this sort of act is voluntary only in the sense that it is not forced,that in the moment of acting I am aware of causing the act myself, eventhough I have never decided to do so. To consider an example, whenI suddenly come to think of my deceased grandfather, I am fully aware ofboth that I did not decide beforehand to think of him now and that I myselfnevertheless am the agent and the subject thinking that very thought. It is inthis sense that the soul employs its faculties: nothing forces it to do this, it actsby itself, but at the same time it never chooses whether or not to employits faculties, for it always finds itself in the process of already actually employ-ing them.

It is true that in the second paragraph S· adrā does speak of the soul beingin some sense forced (ud

˙t.urrat) to put its faculties to use because its

awareness of itself entails a loving, interest-laden relation to itself and itsacts. However, I believe that in order to make any sense of the passage, wehave to distinguish this sense of being forced from being forced by anexternal cause, for S· adrā consistently describes the soul’s operation of itsfaculties as free precisely because of the absence of external causes. Thesoul’s free voluntary operation of its faculties is forced only in the sense thatit follows naturally, and therefore necessarily, from the soul’s essence, and isthus not a matter of choice. The soul by itself necessarily proceeds to operateby means of its faculties.

But if we find ourselves ‘from the beginning of our creation’ acting freelyby ourselves, no one will ever be aware of herself as a mere I detached fromall possible acts. Even the flying man will be aware of its interest-ladenrelation to the body – if not otherwise, then at least by being frustrated in hisattempts to act. However, I believe S· adrā’s point goes deeper still, for he isnot content to say that our self-awareness is always accompanied with theawareness of performing some of the acts characteristic to us. Rather, asbecomes clear from the explication shortly after the present passage, heinsists on the stronger claim that our self-awareness is one with the actualperformance of those acts. In an argument against the implied suggestionthat his theory of knowledge fails to account for the differences in people’sacquired knowledge, S· adrā says:


[When] being a knower and being known is realized between two things,there is no doubt about an essential connection between them in accordancewith existence, and so a unificational connection or existential bond of oneknowing the other is realized between both of the two things, provided thatthere is no obstacle due to one of them being deficient in existence or mixedwith non-existence and veiled by dark appendices, and that connectionrequires the occurrence of one of them to the other and its being revealedto it. It may take place between the very self of what is known (nafsi dhātial-ma‘lūm), in accordance with its concrete existence, and the self of theknower (dhāti al-‘ālim), like in the soul’s knowledge of itself (fī ‘ilmi al-nafsibi dhātihā), its attributes, its faculties and the forms established on the tabletsof its awareness, or it may be between a form which occurs from what isknown and is additional to its self (dhātihi) and the self of the knower (dhātial-‘ālim), like in the soul’s knowledge of what is external to itself (dhātihā)and the self of its faculties and its awareness (dhāti quwanhā wa mashā‘irihā),and it is called ‘occurrent knowledge’ or ‘emergent knowledge’. What isreally apprehended is also here the very form that is present, not what isexternal to it, and when it is said of the external that it is known, this is in asecondary sense. Similarly ‘existent’ is said of both existence itself and of theexisting quiddity, but what really exists is the first division. It is reallyconcrete and distinct without the quiddity, for [the quiddity] in itself issomething obscure, unconcrete in essence. Thus, when the word ‘existent’ isapplied to it, this is in a secondary sense in respect to a bond to existence.29

S· adrā begins by asserting in general terms his thesis of the unity in existencebetween any knowing subject and the object she knows. The two can onlyexist as an actual knower and something actually known in a single act ofmental existence. This amounts to saying that the connection betweenthem must be understood not as a relation prevailing between two inde-pendently existing things, but rather as a mode of existence proper toknowledge: the knower and what is known literally first come to be in anact of this type of existence. S· adrā conceives of this shared actuality inmental existence abstracted from any reference to or causal dependence onwhat is external to it. On the level of mental existence, that is, in the absenceof any darkness or deficiency due to matter, actual existence simply requiresthat what is known occurs or is revealed to the knower. The paradigmaticexample of such a union is the subject’s knowledge of herself, her facultiesand their operations, along the lines we have just described. But as S· adrāstrikingly asserts here, it also holds of all forms known, that is, of theobjective constituents of acts of mental existence. No matter how naturalit may seem to believe that the objects of our experience, the figures,

29 Asfār III.1.3.1, VI.154–155.


colours, sounds, smells and textures that we are aware of as interdependentproperties of substance-like objects, are mere representations of extramentalthings that exist independently of our consideration, and that the latter arethe true objects of our knowledge, S· adrā explicitly states that this convictionis due to an implicit secondary consideration, a tacit addition to the basiclevel of mental existence that we have to start from. Ultimately it is nothingbut an unfounded hypothesis that we have formed on the basis of experi-ence acquired over a long period of time, neither a necessary constituent ofthe act of mental existence nor something necessarily entailed by theawareness inherent to that act of existence. On the contrary, the phenom-enal object, abstracted from any possible reference to a world outsideexperience, is just as constitutive to the act of mental existence as its self-aware subject.

Thus, for S· adrā self-awareness is nothing less than a constitutive part ofmental existence (all mental existence is aware of itself), but it is also notmore than a constituent of it. I can of course distinguish myself as thesubject of my experience from any object I am aware of, and I can coinvarious terms that reify this result of my analysis into a substance conceivedindependent of the cognitive attribute it was distinguished from. S· adrā doesnot deny that we can think and speak of the two constituents of mentalexistence as two independent existents which can survive their mutualconnection in the particular act of mental existence that we started from.Indeed, this is precisely what Avicenna and Suhrawardī had done in theirarguments for a narrow concept of self-awareness as pure first-personality orI-ness. Rather, S· adrā’s point is that such a postulation of a mutuallyindependent subject and object is always secondary to a primordial experi-ence in which the two figure as one and requires that a second-order analyticoperation is performed on the first-order act of mental existence. Butdistinguishability in analysis does not entail distinction in reality. Neitherthe subject nor the object of the act of mental existence exists reallyindependent of the other, but only as twin constituents of the act ofexistence that alone is real, the experience of this subject knowing this object.

The inseparability of self-awareness from the unified act of mentalexistence, of which it is one constituent, resurfaces in the context of anargument for the metaphysical priority of the mental cognitive faculties inrelation to their counterparts among the organs of the body.

[W]hen a human being’s external faculties and his corporeal senses are stilledby sleep, loss of consciousness (al-ighmā’) or [something] else, he will oftenfind of himself (nafsihi) that he hears, sees, smells, touches, strikes or walks,


and so he has in himself (fī dhātihi) these sensations, faculties and instru-ments with no deficiency or lack of any of them, although they are notestablished in this world – that is, the world of sense and observation – forotherwise anyone with a sound sense would observe them, which is not thecase. It is thus known that their dwelling is another world, which is the worldof the hidden and the internal.30

The argument relies on a piece of evidence which S· adrā clearly believes to beobvious on the basis of his interlocutor’s personal experience: any adulthuman being must have imagined or dreamt herself as perceiving, acting orreacting in all manner of ways without a corresponding physical processtaking place in the body that could be observed by another person. Thisleads S· adrā to the conclusion that the faculties, the sufficient means toperform acts proper to them and the acts themselves are in the soul or theself that is aware of itself as their subject.Of course, any modern reader will consider the case inconclusive, at least

in the form S· adrā has given here, for he has failed to address the possibilitythat the sleeper’s experiences are dependent on physical processes takingplace in the brain. In spite of his aversion to the pursuit of such practicalsciences as medicine, S· adrā certainly knew of the cognitive functions post-Galenic authors like Avicenna had accorded to the brain, and he was fullyaware that inferences concerning the brain are hardly a matter of straight-forward observation. Moreover, the examples of sleep and loss of conscious-ness readily bring to mind the Avicennian discussion of the self-awareness ofa sleeping or intoxicated person in the Ishārāt.31 If it is not entirely unlikelythat S· adrā had the Ishārāt on his table here, it seems strange that he shouldhave so thoroughly failed to grasp Avicenna’s central intuition according towhich the possible lack of any experiential content in the sleeper’s mind isdue precisely to the broken connection between her self-aware soul and thecognitive organs in the body, the brain in particular.In the final analysis, I do not believe that S· adrā’s argument, if taken

alone, survives this criticism. However, he may not have intended it to beconclusive, but merely as one piece of evidence which, when supported byfurther evidence and proper arguments, provides valid support for dualism,although on its own it can only provide a pointer of sorts towards that view.The qualification towards the end of the passage suggests that S· adrā’simmediate point is to highlight the strict distinction between mental andcorporeal existence. For him, the fact that a complete breakdown of theconnection between the soul and the cognitive organs of the body entails no

30 Asfār IV.8.7, IX.90. 31 Avicenna, Ishārāt, namat. 3, 119; see Chapter 4.2, 80–84.


cessation of mental activity corroborates his strategy of discussing the varietyof experiential phenomena – the awareness we have of ourselves as hearing,seeing, smelling, touching, striking, walking and so forth – in exclusivelymental terms.

This move, on the other hand, is crucial for understanding anotherdifference from Avicenna’s discussion of the sleeper. As we recall,Avicenna employed the special cases of the sleeping and the intoxicatedperson to argue for a narrow type of self-awareness that can plausibly be saidto remain even in states lacking all experiential content. S· adrā’s strategy isthe polar opposite: because the sleeper’s experience is not empty, hercognitive faculties and their acts must take place in the sleeper herself, notin the body. Later on in the same chapter, he ventures to formulate the samepoint in even stronger terms:

the psychic human (al-insānu al-nafsī) has sense-perceptions of things byhimself (bi dhātihi) and judges them by himself (bi dhātihi), not by means ofa natural instrument that he would need in his apprehension and act; thus,although his apprehension of external percepts is by means of an additionalform that is present to him or occurs in him, his apprehension of it is by meansof those very forms, not by means of another form, for otherwise a regress to amultiplication of apprehensional forms would result. Thus, his self by itself(dhātuhu bi dhātihi) is sight for apprehending what is visible and hearing forapprehending what is audible; and in this manner for every species of what isperceived, and so he in himself (fī dhātihi) is hearing, sight, smell, taste andtouch for himself (li dhātihi). You know from the preceding the unity of thesense with what is sensed, and so he is the sense of all senses. Furthermore, hejudges by himself (bi dhātihi) about estimative and other premises, not bymeans of anything additional to the forms of the premises, and he is desire forhimself (li dhātihi) of what is desirable and anger for himself (li dhātihi) of whatis repulsive, with no additional desire or anger.32

The somewhat unusual term ‘psychic human’ is derived from theTheology of Aristotle, which provides the framework for much of the chapterour passage is embedded in.33 It refers to a human being that has left themode of existence proper to a material form and ascended to the mentalmode of existence, being therefore a self-aware existence, but that has notyet become a pure intellect.34 In other words, S· adrā is describing a rathercommonplace human experience when he states that the psychic human’s

32 Asfār IV.8.7, IX.92.33 Cf., in particular, the extended discussion on different levels of human being in ThA X.52–136,

142–154, based on Plot. Enn. VI.7.2.53–VI.7.11.36.34 Asfār IV.8.7, IX.89.


self is by itself, for itself and in itself the five ‘external’ senses as well as all ofthe Avicennian internal senses, represented here by the governing faculty ofestimation, the motive faculties of desire and anger, and ultimately theintellect as well, if that is what the ‘other premises’ refers to. He doesacknowledge the fact that we commonly take sense perception to be causedby extramental entities and to have those entities as its proper objects, forlater on in the chapter he qualifies his thesis by saying that sense perception,unlike imagination and intellection, requires the presence of extramentalobjects as a necessary circumstantial condition for the creation of mentalpercepts in the soul.35 However, he does not seem to think that this in anyway alters the fact that when considered exclusively in terms of mentalexistence, or as phenomenal experience, perception has a mental origin andshould be understood as a unified act of mental existence consisting of bothself-awareness and objective content.36 After all qualifications, the centralpoint remains that the self-aware subject of mental existence is in a veryrobust sense one with and inseparable from her various experiential con-tents. The way in which S· adrā repeatedly insists on the point suggests thathe is fully aware of departing from the traditional, narrow concept of self-awareness, and the explicit reference to his general theory of cognitive unitygives out his motive for the departure.The same point is revisited in a later chapter designed to elucidate the

subsistence of the human soul in the hereafter. Having dealt with alternativeinterpretations of the strikingly sensualistic descriptions of the afterlife inthe Qur’ān, S· adrā offers his own account according to which the afterlife isjust as real and concrete as this life, only it is lived purely mentally, free fromany connection to matter. As a consequence, the afterlife is stronger inexperiential terms than our mundane existence here and now. Since thehuman soul is of an immaterial origin, it can create immaterial mental formsby itself. Its connection to the body sets certain conditions to this creation,and since these conditions are constantly changing due to the unceasingmaterial process in the body, the soul’s existence in the mental sphere is alsofleeting, impermanent and weak. But when the soul is separated from thebody, intermittently in states like sleep or ecstasy and permanently in death,it can create its percepts free from those material conditions, as a result ofwhich they will be stronger and more stable.37

35 Asfār IV.8.7, IX.94–95.36 Cf. Asfār IV.10.10, IX.237–241, where S· adrā, following Ghazālī, makes an explicit distinction between

the content and the cause of a mental state. The experience of pain or sexual pleasure is the sameregardless of whether it is caused by an external wound or intercourse, or one that is merely imagined.

37 Asfār IV.10.10, IX.244–245.


To corroborate this argument S· adrā describes the ascent to the afterlife bymeans of the traditional trope, of Plotinian origin and familiar from bothAvicenna and Suhrawardī,38 of returning to oneself.

Do you not see that whenever the soul is relieved from the concern and thenecessary movements in preserving this body made up of incongruent thingson the verge of falling apart, and the external senses are suspended of theiract . . . it seizes the opportunity and returns to itself (dhātihā) to someextent – though not entirely, because the natural, vegetative and otherfaculties [remain] in operation . . .

Thus, through this return the soul becomes an originator for forms andobserves them by means of senses which are in itself (dhātihā), with noparticipation of the body, and so it has in itself (dhātihā) hearing, sight, smell,taste and touch; for if it didn’t have these five in itself (dhātihā), how could ahuman in the state of sleep or lack of consciousness see, hear, smell, taste ortouch while his external senses are suspended of their apprehensions? . . .

Just as these five corporeal ones go back to a single sense, which is thecommon sense, so all the senses and the apprehensive and motive faculties ofthe soul go back to a single faculty which is its luminous self (dhāt) that isemanative of God’s permission, might and power, and during its return fromthis world to itself (dhātihā) its apprehension of things becomes identicalwith its power. If its return to itself – even though it [remains] operative inthe body to some extent – gives rise to the creation of forms in this manner,what do you think when it leaves the connection and all the obstacles behindand returns to itself (dhātihā) and to the self of its origin (dhāti mabda’ihā)entirely?39

Since the argument is remarkably clear as it stands, let us concentrate onthe important formulations: the soul has its perceptual content in itself, and,what is more, its self is the very faculty of perception and motion in a singleact of mental existence. If that is a description of the soul as it is in itself,temporally or permanently cut off from all conscious action related to thebody,40 then we have here a clear parallel to the state of the flying man. ButS· adrā’s description of the state could scarcely be more drastically at variancewith that of Avicenna and Suhrawardī; for him, the soul’s awareness ofitself, that is, its very existence in the mental mode, necessarily entailsawareness of its cognitive and motive capacities as well as the correspondingacts. When this entailment is understood in the framework of the

38 Cf. ThA I.21, 22 (based on Plot. Enn. IV.8.1.1–4); VIII.150, 116 (based on Plot. Enn. V.8.10.39–40);Avicenna, Ishārāt, namat. 3, 119 (see Chapter 4.2, 80); and Suhrawardī, Talwīh. āt III.3.1, 239Habībī, 70Corbin (see Chapter 6.1, 127–128).

39 Asfār IV.10.10, IX.245–246.40 Cf. Asfār IV.8.3, IX.53. For the unconscious nature of vegetative acts, see S· adrā’s answer to Abū al-

Barakāt’s related comments in Asfār IV.2.5, VIII.77–79.


epistemological theory of cognitive unity, as we have seen S· adrā explicitlydo, it becomes clear that the narrow concept of self-awareness that Avicennaintroduced and Suhrawardī subscribed to appears fundamentally problem-atic, for consistent systematic reasons.According to S· adrā, self-awareness is always part and parcel of a complex

experiential whole, a necessary constituent of a unified act of mentalexistence. While Avicenna and Suhrawardī would probably have beenready to admit that in the majority of normal human cases self-awarenessis conjoined to manifold experiential content that one is thereby aware of asone’s own, there remains a principal divide that separates S· adrā from themand that can be spelled out bymeans of the thought experiment of the flyingman. The question we should ask is whether the flying man is an argumentper impossibile.41 For Avicenna and Suhrawardī, the case of the flying manmay well be impossible but only in the sense that no one ever actualizes thatparticular potentiality by coming to exist in such a state, that his is anexclusively imaginary case. As a consequence, the flying man’s situation canstill be called logically possible, that is, the inherent features of the essencesdenoted by the relevant concepts allow that someone can have such a thinexperience that it consists of nothing but being an I. What is more, thereremains a sense in which the case is nomically possible as well, for, given theomnipotence of God, He can presumably create anew a fully developedhuman being to float in the air without any perceptions at all, if only Hehappened so to choose.For S· adrā, however, the impossibility runs deeper. While the flying man

may be a useful means of analysis, and can even provide the basis for a validargument for the soul’s immateriality,42 it entails no real consequences for theself-awareness involved. It is not a real possibility for a mentally existing humanbeing to have first-personal experience without content, because the existenceof the first person necessarily gives rise to content it experiences as its own. TheS· adrian flying man, or animal, would be minimally aware of being capable ofcognition and voluntary motion and would be driven by its nature to put itscapacities to use. Even if we suppose that he were prevented from actually usingthem, he would, unlike his Avicennian and Suhrawardian peers, be aware ofwhat he is lacking, subject to all the anguish of a soul divested of its body.

41 Let it be emphasized that the following is entirely a product of rational reconstruction and should beunderstood as a heuristic device designed to articulate the difference between S· adrā and his prede-cessors. I realize that our protagonists’ interpretations of the modal terms (‘possible’ and ‘impossible’)may not yield to this manner of exposition.

42 As we saw in Chapter 7.1: 164–166, S· adrā presents it as one of the arguments for the immateriality ofthe animal soul in Asfār IV.2.2, VIII.47.


Thus again, S· adrā would maintain, in order to arrive at the narrowconcept of self-awareness, we must perform a second-order analysis of theimmediately given first-order act of mental existence that each of us finds inher own experience. The flying man is a potent means of performing thatanalysis, but we must be careful not to postulate a really existing pure I tocorrespond with the result of the analysis. Each of us is an I, that much istrue, but always an I engaged in a concrete act, immersed in a concreteperception or thought, driven by a concrete desire, and so forth. This criticaldeparture is, I presume, not entirely implausible in light of availablephenomenological evidence, but S· adrā’s primary motivation for making itseems nevertheless to have been his firm adherence to a particularly broadversion of the theory of cognitive unity. The picture that emerges from thechapters in which he makes that departure is one of a thinker who is willingto save as much of the inherited discussion of self-awareness as is possible inlight of a commitment he has made in another context. The same is true ofthe revision his theory of substantial change drives him to make to hispredecessors’ understanding of human selfhood.

8.2 Identity in substantial change

The narrow self that Avicenna’s concept of self-awareness entails is inti-mately connected to a particular view of the identity of an individual humanbeing. This view provided the basis for Avicenna’s primary argumentagainst the unity theory of knowledge, but as we have seen, it was alsoexplicitly connected to the topic of personal identity over time in an argu-ment for psychological substance dualism found in the Mubāh. athāt.

43

What endures unchanged through the constant variation of my corporealacts and experiential content is the I in me, the narrow first person that is theagent and subject of them all in one and the same sense. This concept of selfis also familiar from another argument for dualism, that is, the one from theunity of experience, which we have seen S· adrā apply in much the samemanner as his predecessors. If I first smell coffee, then begin to desire a cup,and in the end proceed to stand up in order to walk to the coffee room, arenot these three constituents of the explanation of my action quite obviouslytemporally consecutive states? And if that is the case, does not the plausi-bility of the explanation hinge on the fact that the I remains one and thesame despite the changes that result in its three distinct states? Finally, sincemuch of the power of these arguments is derived from the interlocutors’

43 Avicenna, Mubāh. athāt VI.402–403, 146–147 Bīdārfar (cf. 453, 227 Badawī); see Chapter 4.1, 75–79.


shared intuitions, it must be presumed that the idea of a stable I in each of usis not entirely lacking in commonsense plausibility either.It therefore seems that human self-awareness and selfhood pose a major

problem for S· adrā’s theory of substantial change, according to which allaspects of created entities such as human beings, their respective substantialcores included, are subject to change. Even if the I or the self is in some sensethe core and essence of my being, its stability turns out to be but an illusion.If everything in me, down to the very core of my first-personality, is subjectto change, can I in any reasonable sense identify with the former phases ofmy substantial development?What makes the present I the same as the I tenyears, ten days or even ten minutes ago? If there is no stable I in me, butmerely a series of consequent and distinct I-phases, can any of theselegitimately claim to be the proper subject of the change – in what sensecan I now say that I am living my life? To conclude our investigation ofS· adrā’s concept of self-awareness, I would now like to argue that he not onlyis aware of these problems but also attempts to provide a solution to them.Faced with the choice of whether to qualify the doctrine of substantialchange or to revise the concept of human selfhood, he clearly opts for thelatter, despite the fact that he has found use for a great deal of the relevantAvicennian and Suhrawardian insights.S· adrā’s solution is based on a new concept of identity introduced in his

theory of substantial change. To begin with, he defines the concretelyexisting individual nature, which corresponds to the substantial essencethat provides the basis for the identity of an individual entity in thePeripatetic framework, as a continuum of substantial change.44 Now, wecan of course pause to consider any given phase of the continuum, and byanalysing it into its constituents we can distinguish an essence for it that isdistinct from other features that provide its accidental determinations. Wecan also compare the essences derived from the analyses of two or more suchphases and judge that the phases are instantiations of a single essence andtherefore identical in terms of what they are. However, we must not therebyneglect the fact that the identity we then perceive between the essences ispossible only because they share in a quiddity that we understand. In otherwords, the identity is intelligible and therefore a case of the instantiation of auniversal quiddity in two or more particulars. But since it is the identity of atemporally existing individual entity that we are concerned with, we cannotsettle with this, for, considered as individuals, the essences in the phases we

44 Asfār I.1.7.19, III.74–76.

Identity in substantial change 209

have analysed remain numerically and specifically distinct.45 The I thatwoke up this morning is the same as the I reading these words only in thesense that both are considered in abstraction from all other features of thetwo individual acts of existence, that is, as instantiations of a quiddity suchas ‘I-ness’; but considered in themselves, the I-that-woke-up-this-morningis distinct from the I-that-reads-these-lines.

However, the temporal continuum of existence to which the phasesbelong does have a genuine individual identity. This identity is abovechange, but not in the sense of enduring through time, for it is based onan extratemporal consideration of the continuum as a temporally extendedwhole.46 In his articulation of this new concept of identity, S· adrā relies onthe notion of ‘fixed essence’ (‘ayn thābit) introduced by the great Sufitheoretician Ibn ‘Arabī. As the ultimate foundations of the identity ofindividual existents, the fixed essences are atemporally present in God’smind, but as such they are mere potentialities that lack the kind of existenceproper to them. Their mode of existence in reality is to be unfolded in timeas processes of substantial development towards increasing perfection.47 Buta fixed essence that can only exist as a process of change will not be availablefor intellectual abstraction from any particular phase in the process. If ametaphor is allowed for clarification, our concrete existence amounts toperforming in time all the acts of a script that is fixed once and for all, butsince the play is about our own development, we are confined to theprotagonist’s closed perspective and the plot will remain a secret to usuntil the very end.

S· adrā does seem to think that on the primary level of merely existing, thatis, when one is not attempting to understand explicitly what it is that exists,the identity of the process of existence is somehow given in the very act ofexistence.48 In the same context, he describes the extratemporal essence incausal terms, stating that the essence enters existence by directing it to thesort of goal proper to that essence and thereby determining the correspond-ing act of existence to be of the type it is.49 It thus seems that the existence ofeach fixed essence amounts to a restricted instantiation of the Neoplatoniccircle of origination (mabda’) from and return (ma‘ād) to the Origin. As thesingle point of origin and return, the essence is present at each stage of the

45 Asfār I.1.7.19, III.76; I.1.7.22, III.83; I.1.7.24, III.95.46 This bears an intriguing resemblance to four-dimensionalist theories of identity in contemporary

analytic metaphysics; cf. Gallois 2012. According to Rahman 1975, 109, ‘Allāma Sayyid Muh. ammadT. abāt.abā’ī (d. 1981) has presented an interpretation of S· adrā along these lines.

47 Asfār I.1.7.25, III.105–107; IV.7.3, VIII.398–404; IV.7.6, VIII.449–452.48 Asfār I.1.7.22, III.86; I.1.7.25, III.107. 49 Asfār I.1.7.23, III.89–92; cf. IV.9.5, IX.143–144.


process, but since each instant is a unique phase of the process that consistsof all the determinations it has, an act of understanding cannot straightfor-wardly identify two phases of one process with each other. Rather, theidentification must take place by means of some quiddity, which in an idealcase would be the very fixed essence actually governing the process one isconsidering.50

In the remainder of this subsection I will start by investigating sometextual evidence for the claim that S· adrā was aware of the consequences hisnew theory of identity bears for the discussion of self-awareness. My aimhere is to show that he deliberately attempted to import change into the veryI each of us is aware of ourselves as being. I will thenmove on to consider therelated question of whether we are transparent to ourselves in the mannermost of S· adrā’s predecessors, Avicenna and Suhrawardī in particular, seemto have thought, or whether the new concept of selfhood rather introducesan amount of opacity to our self-awareness.Much of the positive account in the chapter of the psychological

section of the Asfār in which S· adrā promises to explain the endurance ofthe human faculties, the related acts, and the human body in theirconstant transformation, consists of a series of quotes from the Theologyof Aristotle. The thrust of the material is that the identity in change is dueto an atemporal, unchanging and unified principle governing the change,quite in line with our brief account of the theory of substantial change.However, S· adrā concludes the chapter with an extended critique ofAvicenna who, failing to grasp this crucial crypto-Plotinian point, wasincapable of dealing with a number of psychological problems related toidentity in change.51

One of these problems goes straight to the heart of our concern. S· adrātakes Avicenna to task by citing a question from theMubāh. athāt:

52 how canAvicenna say that a human soul becomes an intellect when an immaterialintellectual form actually occurs to it, for does this not entail that a soulwhich is not yet immaterial can nevertheless apprehend something that isimmaterial, and would this not be a rather obvious violation of Avicenna’s

50 S· adrā clearly allows the possibility of subsuming two or more phases of a process under differentquiddities, as shown by his remarks on accidental motion, that is, motion considered as notimmediately concomitant to the essence, such as a particular change of place, a particular qualitativechange or a particular amount of growth (see Asfār I.1.7.19, III.75). For instance, I can conceive of asparrow’s flight from one tree to another as the movement of a bird or as the movement of somebrownish thing. Both quiddities will enable me to attribute the two stages in the locomotion to asingle entity in this restricted case.

51 Asfār IV.9.5, IX.139–146. 52 Avicenna, Mubāh. athāt 377, 209 Badawī; V.302–303, 122 Bīdārfar.


own dearly held principles? Avicenna attempts to explain the problem awayby appealing to a metaphorical sense of becoming, but S· adrā is, unsurpris-ingly, not convinced. According to him, the question points to a realproblem in Avicenna’s theory of the human soul, which arises fromAvicenna’s claim that the human being is immaterial and hence intellectual,in an immutable sense from the very beginning of its coming to be – amoment which S· adrā, presumably relying on a later source, locates at thefourth month of pregnancy when the foetus has reached a sufficient stage ofdevelopment to enable the emanation of an immaterial entity capable offunctioning as a soul for it.53 Avicenna did of course allow for two momen-tous events in the course of human existence, that is, the eventual cessationof the soul’s relationship to the body and the actualization of its intellectualpotency through the acquisition of knowledge. However, in both cases thesense in which the human being exists as an immaterial and thereforeintellectual thing, her self-awareness, remains immutable. Insofar as I aman I, I with the body am exactly the same as I without the body. Similarly,whether or not I have acquired knowledge, I am and remain I. For S· adrā, onthe contrary, the soul

is at first an imagination in act, an intellect in potency, and then, by means ofthe repetition of apprehensions and the extraction of intelligibles fromsensibles and universals from particulars, it comes from the degrees ofpotential intellect to be at the degree of actual intellect; its self (dhātuhā)evolves and is transferred in this substantial transformation from an imagi-native faculty to an intellectual faculty.54

S· adrā is emphatic: the soul literally becomes an intellect by changing initself from an imaginative capacity to an intellectual one. This develop-ment is not a matter of intellectual objects merely replacing what wasimagined, although that is involved as well. Because of the theory ofcognitive unity, to which S· adrā refers elsewhere in the chapter,55 thesubject and object of all cognition are fundamentally interdependent,and the ascent to a higher level of knowledge and existence concerns thesubject and object alike.

One may duly ask whether we are not reading too much into S· adrā’schoice of the ambiguous dhātuhā here. More precisely, is there anythingthat forces us to bring self-awareness into play, instead of interpretingthe formulation as a vague reference to change in the essence or thesubstantial core of a human being? The present context alone provides no

53 Asfār IV.9.5, IX.150. 54 Asfār IV.9.5, IX.151. 55 Asfār IV.9.5, IX.148.


definitive giveaway,56 but corroborative material can be found from relatedpassages elsewhere in the Asfār.In the part of the Asfār devoted to the theory of substantial change S· adrā

addresses the problem of identity in change by taking human existence as acase in point. This is instigated by an argument for dualism that relies onour self-awareness as evidence for an enduring immaterial self that, byfunctioning as a soul, underpins the identity in flux of its body.

Some of them have said: ‘By means of this it is known that the soul is nottemperament; for temperament is something flowing and renewed, and inwhat lies between any two extremes it has potentially infinite species. Thesense of “potentially” is that no species is actually distinct from its neighbour,just as points and parts in a line segment are not actually distinct. Everyhuman being is aware of himself (dhātihi) as one individual and unchangingthing, even if he is one in the sense of continuity (al-ittis.āl) to the end of life.’I say: it is as if something of the scent of the self’s (al-dhāt) renewal in the

human came upon this speaker when he said ‘even if he is one in the sense ofcontinuity to the end of life’; for temporal continuity is not incompatiblewith transformation in that very continuum (al-muttas.il), you will know thiswhen we resume the discussion.57

According to the argument S· adrā cites, the body is a temperamentalconstitution of elements, subject to ceaseless fluctuation of its constituents,and therefore requires an external principle of stability that keeps it fromfalling apart. This external principle is the soul, which can function as sucha principle precisely because it is stable and unchanging in itself. Whatmakes the argument interesting for us is the reference it makes to self-awareness. The argument explicitly assumes that human beings universallyshare an intuition of their respective dhawāt as enduring unchangedthrough time, and, in order to save the plausibility of this assumption,I believe we have to read dhāt here to mean ‘self’.58 If that is the case, thennotwithstanding the argument’s anonymity, the insight it builds upon is

56 In Asfār IV.9.4, IX.155. S· adrā does criticize Avicenna for having incoherently held that the immaterialhuman substance is capable of really separating from the body and for having nevertheless deniedsubstantial change. According to S· adrā, separation from the body is an instance of substantial change:what first existed as a soul comes to exist as an independent immaterial substance. He then character-izes the motion as istih.ālatun dhātīya, which could be translated as ‘transformation in terms of theself’, but equally well, indeed perhaps less awkwardly, as ‘essential transformation’. Thus, the passagedoes not help to resolve our dilemma. Cf. also Asfār IV.11.1, IX.265–266.

57 Asfār I.1.7.24, III.95.58 My argument would be the same, mutatis mutandis, as in the discussion of Hasse’s interpretation of

the flying man (see Chapter 2.2, 38–41). If we read dhāt as ‘essence’ here, we face the question of whatsort of awareness of essence could plausibly be regarded as intuitively obvious by anyone.


of course familiar fromAvicenna.59However, the addition at the end, whichqualifies that the self remains one throughout the change ‘in the senseof continuity’ (bi ma‘nā al-ittis.āl) departs from the common ground.Whatever the origin or the motive for the qualification, S· adrā finds itparticularly apposite because he rejects any straightforward notion of stabil-ity in the human self. It is crucial to notice, though, that he does not denythe plausibility of our intuition of our selves remaining the same in spite ofthe change about and related to them – only this identity is a property that isdue to the continuity of the process of change, not a relation between anytwo instants isolated from that continuity in intellectual analysis.

This is a point S· adrā has laboured at some length earlier on in thechapter. Another example by means of which he attempts to clarify histhesis is qualitative change, such as the gradual darkening of a colour.Suppose you are sitting under a tree at dusk, observing the shadows growincreasingly thick by the minute. Although the shadow was visibly lighter amoment ago than it is at present, which enables you to distinguish theparticular shade or species of its earlier black from that of the present one,you can still say that the two are instances of the same shadow or the samecolour. This is legitimate because primarily, that is, prior to any intellectualdistinction between the two shades, they were embedded in one process ofchange, which is potentially divisible by an intellect but actually one initself.60 Our selves are similar to the thickening shadow: I can distinguishbetween two phases in the temporal development of my being, and evenidentify one with the other in subsequent reflection, but that is only becausethe development was one continuous whole to begin with. If I consider thetwo phases in isolation, I can never identify them with each other, becausethey are distinct individuals. I may of course subsume both under a singleuniversal, but then I will lose grasp of their individuality – and it is preciselyas an individual that I am aware of myself. The way out of the impasse is bylocating the identity in the first-order self-awareness prior to the intellectualconsideration. On the level of primitive self-awareness, I am primitivelyaware of myself as a continuous existence that remains a single processdespite the thoroughgoing changes it is subject to. But ascending from this

59 A parallel case, though in a different formulation, can be found in Avicenna, Mubāh. athātVI.402–403, 146–147 Bīdārfar (cf. 453, 227 Badawī), and Avicenna, al-Risāla al-ad

˙h. awīya IV,

127–128; cf. Chapter 4.1, 75–79. I have not been able to determine an immediate source for S· adrā’sargument here; it may be a summary of the interchange between Rāzī and T. ūsī over a directive fromthe third namat. of the Ishārāt. For the passages in question, see Avicenna, Ishārāt, namat. 3, 120–121;Rāzī, Sharh. al-Ishārāt 125–127; and T. ūsī, Sharh. al-Ishārāt II.350–356.

60 Asfār I.1.7.24, III.94–95.


primitive awareness to explicit intellectual consideration of myself, I lose theawareness of identity, and the only substitute I can come up with is auniversal quiddity that divests its instantiations of their individuality.61

That the substantial change of human beings concerns our very first-personality is also supported by means of proverbial evidence in a psycho-logical chapter on the generation of individual human souls. Earlier in thechapter S· adrā has stated a preference for the view that, prior to theirconnections to bodies, human souls are fixed essences in God’s knowl-edge.62 In a corollary to this statement he tackles Suhrawardī’s refutation ofpre-existence, particularly the critical claim that attributing the two statesproper to a fixed essence and to a soul, respectively, to one and the samething leads to asserting that the very reality of the thing is subject totransformation (inqilāb).63 Unsurprisingly, S· adrā says that as long as thistransformation is understood in the framework of his doctrine of substantialchange, there will be nothing to object to in it. Having brought forth anumber of physical and cognitive processes as examples of perfectly legit-imate transformations from one mode of existence to another,64 he con-cludes with the following chapter.

What is reported from Pythagoras supports this claim. He said: ‘A spiritualself (dhātan) radiated knowledge (al-ma‘ārif) upon me, and I asked: Who areyou? It said: I am your complete nature.’Ohmy beloved, if you were enabledto ascend in the layers of your existence, you would see numerous itnesses,different in existence. Each of them is a completion of your itness, notlacking anything of you, and each one of them is referred to by means of‘I’. This is like in the famous proverb: ‘You are I, so who am I?’65

S· adrā’s explication is almost as enigmatic as Pythagoras’ dictum, but thecontext in which it is introduced warrants us to interpret it as a condensedexpression of his novel theory of identity. If I were, per impossibile, able toascend to later and therefore more perfect phases of my continuous exis-tence, all the while remaining what I presently am, I would encounter themas distinct individual things or itnesses (huwīyāt) separate from me. Yet atthe same time each of them would be me in as full and complete a way asI now am me, for once I reach these phases in the continuous developmentof my existence, I will be aware of myself and refer to myself by the

61 For other passages on the individual human identity as a continuity, see Asfār IV.10.8, IX.227–228;and IV.11.1, IX.267–268.

62 Asfār IV.7.3, VIII.398–401. 63 Asfār IV.7.3, VIII.417–420. 64 Asfār IV.7.3, VIII.420–422.65 Asfār IV.7.3, VIII.422–423. The saying reported from Pythagoras is ascribed to Hermes by

Suhrawardī in Mashāri‘ III.6.9.193, 464. The context is an argument for Platonic forms, andSuhrawardī spends no time at all on the implications of the quote for the concept of self-awareness.


first-personal indexical in exactly the same manner as now. The seemingcontradiction can be dissolved by means of the distinction we have justintroduced; the phases are distinct from each other in a second-orderconsideration, such as the depicted ascent, but in the continuity of primitivefirst-order self-awareness one remains aware of oneself as a single selfthroughout the development.66 The playful proverb at the end of thepassage yields to a similar interpretation. If each of the numerous phasesof my existence has a legitimate claim to provide the reference for the use ofthe first-personal indexical exclusive to me, we can ask which of the phases,if any, is really me in the most fundamental sense.67 The true answer, ofcourse, is two-tiered. In the first-order level of continuous primitive self-awareness where there is no actual distinction between them, they are oneand therefore all me in one and the same sense, whereas on the level ofsecond-order consideration, none of them has a more valid claim to be methan any other.

Occasionally, S· adrā applies the phenomenon of self-awareness as evi-dence for the way in which our identity is due to the single fixed essence thatis both the origin and the point of return pursued in our existence. Aparticularly explicit case is the following version of the argument from theunity of experience that concludes an Avicennian classification of the soul’sfaculties:68

This notion [that is, the soul’s unity] will be better revealed to you if youregard yourself (dhātika) separate from all else. You will then see it tounderstand and to apprehend universals and particulars, to estimate, toperceive, to hear, to see, to have desire, anger, love, joy, will, and otherattributes which are innumerable and uncountable. You know that a form iseither simple or composite and that matter too is either simple or composite;yet the composite existent is united in somemanner of unity, for what has nounity has no existence. Thus, multiplicity is caused by unity, because it is theorigin of multiplicity, its principle, model and goal.69

At first glance, the passage reads like a straightforward rephrasing of theAvicennian argument from unity: the self is the immutable subject of all theacts attributed to distinct faculties, which provides an important insightconcerning the corresponding unity of the psychological entity known as

66 Cf. Asfār IV.8.2, IX.17; and IV.8.5, IX.72–73.67 Notice that the pivotal man anā is ambiguous and can be translated both as ‘who am I’ and as

‘who is I’.68 The bulk of Asfār IV.3.8, VIII.146–150, is a paraphrase of Avicenna, Shifā’: Fī al-nafs I.5, 39–51

Rahman.69 Asfār IV.3.8, VIII.151.


the soul. This impression dissolves, however, if we pay careful attention tothe qualifications S· adrā appends to the argument. There is a unity under-lying the multiplicity that appears in the self’s involvement in divergent acts,but instead of the stability of a Peripatetic substance it is a unity proper to acontinuous process, which is made one by the origin and the goal whoseunfolding in temporally extended existence it is. The acts, passions andother determinations of the self are one manifold because they all areconstituents of one and the same existence, phases in a single processfrom the origin to the goal. In a closely related context, S· adrā characterizesthis process of human development governed by a fixed identity in theorigin and goal as our way of imitating God’s unity.70 God is mysteriouslyone by being identical to all of His semantically distinct names; the best Ican do to approach this absolute unity is to be one by living through theentire course of the unbroken continuum of my existence.When the passage is set into its context, it becomes clear that S· adrā is

using the Avicennian insight into the force of evidence provided by the self’sunity in experience to make a point that is in polar opposition to Avicenna.Instead of pointing towards a substantial core that endures through thefluctuation of attributes its existence consists in, the narrow selfhood uponwhich the argument from unity hinges is now used to provide us with a clueto the principle that governs our existence as a continuous whole. But theself we are experientially aware of within the continuity can only beidentified with that principle if we realize that the fluctuation is an essentialand inseparable part of it. The narrow concept of selfhood may help us toarrive at this realization, but only if we understand it in heuristical terms andresist the temptation straightforwardly to identify its abstract stability withthe much more contentful principle. The I is there all along, yes, but only asa dimension of continuous development.This point is corroborated by another characterization of the self as a

continuity, this time in an eschatological context. S· adrā inaugurates thesection of the Asfār devoted to an argument for corporeal afterlife byrecounting the principles required in the argument. The sixth of theseprinciples states that the individual unity (al-wah. da al-shakhs.īya), or iden-tity, varies according to the different modes of existence. In the case oftemporally extended existence, the unity is ‘identical with its renewal andcompletion (‘aynu tajaddudihā wa taqad

˙d˙ihā)’, that is, with the continuity

of the respective individual existence as a whole.71 According to this prin-ciple, the human soul or self is one by proceeding through the entire course

70 Asfār IV.3.7, VIII.136; and IV.3.9, VIII.154–155. 71 Asfār IV.11.1, IX.265.


of her existence at all of the levels due to it, from sense-perception throughimagination to intellection.72 The closely related seventh principle, accord-ing to which the principle of identity in human existence is the immaterialsoul, formulates the point with particular explicitness:

the human itness in all these transformations and variations is one andidentical to itself (wāh. idatun hiya hiya bi ‘aynihā), because it takes place inthe manner of unitive gradual continuity (‘alā sabīli al-ittis.āli al-wah. dānīyial-tadrījī). What is decisive is not the substantial individualities and existen-tial definitions which take place by way of this substantial motion; on thecontrary, what is decisive is that which persists and remains, and it is the soul,for it is the perfectional form in human being, which is the principle of hisitness and self (huwīyatihi wa dhātihi).73

An individual human being is realized as a continuity of constant gradualtransformation. As a result, we can grasp ourselves as one individual only byfocusing on the principle of our development towards perfection, not byattempting to locate, by means of abstractive analysis, some sort of sub-stantial core to our individuality or definitional limits to our existence. S· adrāclearly does not deny the possibility of such analysis, but he does reject itsability to get to the core of our individual identity. It is experientiallyavailable to us in our very first person, but only by living through all thegradual phases of our first-personal existence.

Let us pause to recapitulate. The evidence we have considered suggeststhat S· adrā’s solution to the problems raised by his simultaneous subscrip-tion to both the Avicennian arguments that rely on the stability of self-awareness and the strong concept of thoroughgoing substantial change in alltemporal entities is based on a novel concept of individual identity based onthe continuous whole of temporally extended existence instead of anenduring substantial feature. As a result of this concept, the change throughtime that each of us is bound to undergo is now built in to our very selves.On the first-order level of primitive self-awareness, that is, on the level inwhich each of us simply is an I without pausing to consider what being an Iminimally amounts to, each of us is aware of herself as one because she isliving through the development by which her identity can only be realized.But on a second-order level of reflective attempts to grasp that first-orderunity, that is, in maneouvres of thought (such as the flyingman) designed toisolate the phenomenon of first-personality, we are bound to lose eitherour unity, by finding the first-order unity fragmented into distinct slices offirst-personal existence determined by different sorts of content, or our

72 Asfār IV.11.1, IX.266. 73 Asfār IV.11.1, IX.267.


individuality, by abstracting from those determinations and subsuming thedistinct slices under a universal quiddity such as ‘I-ness’. The tacit criticismtowards Avicenna and Suhrawardī would therefore be that they mistake aresult of abstraction (the narrow concept of selfhood as pure I-ness) for a realindividual (first-personality embedded in a process of constantly fluctuatingdeterminations).That, however, is not all, for S· adrā’s insistence on the change built

into the very self of every human being results in another importantdifference from his predecessors. Avicenna’s and Suhrawardī’s narrowconcept of selfhood as pure I-ness divested of all accidental determina-tions has a feature common to the vast majority of dualist theories ofself-awareness, namely, that the self is thoroughly transparent to itself.This is of course only proper for an immaterial, and thereby intellectual,entity: if there are no material obstacles preventing me from regardingmyself, then I should be fully manifest to myself in all my immaterialglory. What could there possibly be about being me that I can beunaware of? But if I-ness is conceived as a dimension of change, Isuddenly become decisively less apparent to myself. Indeed, if I pauseto anticipate the future phases of my career of development, I soonrealize that to a large extent this aspect of me is not only unclear butentirely unavailable to me. Thus, S· adrā’s concept of selfhood as subjectto substantial change introduces a fundamental sort of opacity to ourselves. If change is indeed crucial to our being, then there will be more instore of us in the future, something about us that we are not presentlyaware of, our immediate awareness of ourselves notwithstanding.74

As we have learned by now, the direction of substantial change is uni-linear: like all creatures, we strive to reach ever higher levels of perfectionproper to our type of existence. In general schematic terms, this amounts tofirst leaving behind the level of mere sense perception and ‘becoming actualimagination and what is actually imagined’, and at a still more developedstage, ‘becoming an intellect and what is understood in act’.75 If at theimaginative level I am in a strong sense one with the imagined content I amaware of, as S· adrā states, then the higher level of intellection must be hiddento me. Even if my existence as an imagination in act contains the

74 The rest of this chapter is adapted from the more extensive discussion in Kaukua 2014c.75 Asfār I.1.10.2.2, III.495. This is S· adrā’s standard account of cognitive substantial change, for other

formulations, cf., e.g. Asfār IV.4.5, VIII.208–209; IV.4.12, VIII.239–240; IV.10.1, IX.167–168. S· adrāalso describes human development in moral terms (e.g. Asfār IV.8.3, IX.41), by means of a distinctionbetween relative activity and passivity (Asfār IV.11.24, IX.445), and by comparing it to variousmethods in the sciences (e.g. Asfār IV.10.4, IX.184–186; IV.11.9, IX.315–316).


potentiality to develop into an intellect in act, being potentially somethingis not the same as being that something in full act and awareness.

But if a lower stage of development contains the higher in potency, can Inot conceive the higher stage by conceiving the potency in myself? Those ofus who have the experience of catching themselves in the act of daydream-ing will certainly recognize the possibility of this sort of anticipation;perhaps with some added rigour we could be more accurate in our antici-pations, and this could be a means of penetrating the opacity in ourselves,albeit not without a significant amount of mediation. S· adrā does not discussthis possibility as such, but he does make an interesting remark in a chapterdealing with the self-intellection due to all intellectual subjects. The insti-gation for the remark is, again, provided by Rāzī who argues against theAvicennian identification of God’s essence with His existence. Suppose wegrant that God’s essence is His existence and then describe His existence aspure existence, divested of all determinations that would diminish it in oneway or another. Since we can understand what existence is, and since we canalso understand what those determinations are to which we apply theequally understandable logical operator of negation, the outcome is thatwe can understand God’s essence. But this amounts to saying that we canunderstand God as He understands Himself, which is about as outrageous aclaim as one can make.76

In his answer, S· adrā is unwilling to reject any of the premises as such.Instead, he opts to qualify our understanding of existence by distinguishingbetween existence and the concept of existence. Existence in reality allowsan infinite variation of degrees of intensity and weakness, and the concept ofexistence can be common to all instantiations of existence in spite of theirdifferences in intensity only by abstracting from those differences. Thus,understanding a particular existence, in this sense God’s, as an instantiationof the concept of existence is not the same as being that particular exis-tence.77 In other words, although a less intense existence can conceive of amore intense existence, it does not thereby know what it is like to exist moreintensely. Moreover, given that the sort of existence proper both to intellec-tually capable human beings and the infinitely capable Creator is cognitive,we can also rephrase S· adrā’s point by stating that knowing God as Heknows Himself, and not as we now do by means of the concept of existence,would amount to existing at His level, or existing as Him.

76 Asfār I.1.10.2.1, III.488. S· adrā seems to be paraphrasing from Rāzī, Mabāh. ith I.1.5, I.124.77 Asfār I.1.10.2.1, III.488.


Although the discussion concerns the infinite difference in degreebetween us and the Creator, there seems to be no reason not to believethat the statement holds also of the finite difference between a less and amore developed phase of our own existence.We cannot know what we havethe potency to become before we actually are that; full awareness of theinnate goal of development can only be had upon arrival at the goal. In thecontext of interpreting the topic of the ‘straight path’ (al-s.irāt. al-mustaqīm)as addressing the ethical implications of the human return to God, S· adrāframes the point in a rather charming symbol:

the journeyer towards God – I mean the soul – travels in itself (fī dhātihā),and passes through sojourns and stations that occur in itself through itself (fīdhātihā bi dhātihā). Thus, at every step it lays its foot upon its head, or ratherits head upon its foot, and this is something astonishing; yet it is notastonishing upon verification and knowledge (al-‘irfān).78

Earlier in the chapter, S· adrā has stated that the straight path is the soul itself,suggesting that the path is only paved when it is first trodden upon.79 Theidea is that the soul’s choices and acts determine its lot in the afterlife;treading on the path of its self, as straight or crooked as it turns out to be, itgenerates itself as an entity with the corresponding character traits. Ourpassage then connects this eschatological theme to the idea of developmentin the most intimate dimension of self-awareness. The earlier phases of thesoul’s self-aware existence provide the basis for the later ones; one can onlydevelop by surpassing one’s present self, or, in terms of the simile, by usingone’s head as a stepping stone upon which one can step to reach higher. Yetat the same time, the later phase is later precisely because it ascends theearlier one. When we consider the fact that proper human developmenttakes place as an increase in knowledge – as an ascent from the lower modesof perception and imagination towards the summit of intellection – werealize that one really only progresses by standing as erect as a human beingshould, that is, by placing one’s head, which after Avicenna is the uncon-tested seat of the highest body-related cognitive faculties, in its proper placeabove one’s feet.If we connect this metaphorical description of the properly human

substantial change to S· adrā’s general claim that the higher stages in thesubstantial development of any act of existence somehow include the loweror preceding ones,80 it begins to seem that in the human case this inclusionamounts to a superior cognitive perspective to oneself. In a process whose

78 Asfār IV.11.19, IX.403. 79 Asfār IV.11.19, IX.394–396, 402. 80 Cf., e.g., Asfār I.1.7.25, III.106.


logic is not altogether unlike that of the Hegelian Aufhebung, abstractedfrom the grand historical story, the later self (a head above feet) is aware ofitself as the present state of a developmental process, and thereby grasps theearlier phases as so many steps that have led it to the present. As patronizingas it may be, one will always know better when one comes of age intellec-tually, as one progresses from a predominantly perceptual subject to onecapable of imagination, and ultimately to a fully mature intellect. In thissense, I can legitimately say that I am the same person as I was ten days, tenmonths or even ten years ago, in spite of all the changes I have been subjectto, for I would not be my present self were it not for those earlier stages inthe development, which in this sense belong to myself. But this inclusion inawareness of course implies a corresponding exclusion from the perspectiveof the earlier phases of self-aware existence. The foot that is about to stephigher is not aware of the superior perspective it will thereby gain for thehead, and similarly, the I a decade ago can only be aware of the present I bydeveloping to the level of existence corresponding to it – that is, bybecoming it.

The inaccessibility of future development in self-awareness is not theonly source of opacity in the S· adrian self, for the inclusion of the past in thepresent also comes in degrees of intensity of awareness. As most of us knowall too well, we do not remember our entire personal histories. On the otherhand, although I can have a recollection of myself as I was a decade or twoago, I may already have lost touch with myself ten minutes into the past.Moreover, there are many things I now know that I know I must havelearned in the past, although I have no recollection whatsoever of everhaving learned them. Were it not for the blatant implausibility that aconcept like ‘my wife’s favourite licorice’ could be hard-wired into mypsychological make-up, it might just as well be a priori for my presentself. S· adrā discusses phenomena like these in terms of habituation which,given his theory of substantial development, does not concern merely theaccidental character traits of one’s soul but one’s very self.81

Habituation is addressed especially in the eschatological part of the Asfārand is consistently approached by means of two Qur’ānic topoi, that of theopening of the book of the soul on the final day,82 and that of the sinners’

81 Cf. Asfār IV.7.1, VIII.382–383; IV.10.4, IX.184; IV.11.4, IX.303–304; IV.11.9, IX.315; IV.11.220,IX.404–405; IV.11.33, IX.537–538. Different types of habituation are described in IV.9.2, IX.115–122.

82 Cf., e.g., Q 7:187; 17:13–14, 71; 18:49; 50:22; 69:19–20, 25–26; 81:10; 84:10–12. For a concise account,see Asfār IV.11.20, IX.404–412. The topos figures constantly also in the brief and derivative Iksīral-‘ārifīn (see, e.g., I.5, 9, 12–13; II.5, 22; II.7–8, 25–27; IV.3, 67–68). For an extended discussion, seeJambet 2008, 150–203.


appearance in the afterlife in a variety of animal forms.83 On the final day,each soul is read its proper judgment from a book, which the soul itselfconstitutes and which contains a complete account of all its acts that areworthy of praise or blame. Prior to the opening of the book, the soul wasoblivious to these acts despite the fact that it had been their agent and thatthe acts had become part of the soul itself as its habitudes. This suggests thatthe previous phases of an individual human existence need not be presentlygiven as explicit cognitive objects. Instead, as habituations to perceive,estimate and react to the world, they often determine the present in allkinds of ways one is not aware of, and it is precisely such hidden aspects ofone’s self that will be painfully clear to read on the final day. By the sametoken, the habituations can be made apparent in the afterlife when theirbearers are transformed to exist in ‘angelic, satanic, animal or bestial forms’that are depictive of their respective habituated states, henceforth tortuouslyexplicit for all to see.84

The relevance of this eschatological discussion from the point of view ofour topic can be assessed by considering two questions. First, does S· adrāexplicitly state that our character traits, or at least some of them, are belowthe threshold of awareness? Second, does he identify the unaware traits withthe self? The answer to the first question seems to be affirmative, not only onsystematic grounds (what opening or revelation would there be unless therevealed traits were somehow hidden to begin with?) but also in light oftextual evidence. Consider as an example the following variation on theopening of the soul’s book:

We also say that the intoxication of nature and the soul’s stupor in thisabode – due to its preoccupation with the deeds of the body – prevent itfrom apprehending the harms and pains of the soul that occur to it and thatare acquired from among the results of its deeds and the concomitants of itsdestructive character traits (akhlāqihā) and habits (malakātihā), by a trueapprehension which is not spoiled by what the senses convey to it and whatthey are engaged in, forgetting and ignoring. Thus, when the veil is liftedfrom the human by death and the cover is removed, on that day his sightfalls upon the consequences of his acts and the results of his deeds, so thatthey then end up – if he is mean in character traits, evil in deeds anddestructive in beliefs – in strong pain and great disaster, as in His saying,

83 Cf., e.g., Q 5:60 (apes and pigs) and 7:166 (apes). S· adrā also cites a number of ahādīth from theProphet and the imams. Cf. the extended discussion in Asfār IV.8.2, IX.12–32.

84 Asfār IV.8.2, IX.29–30; cf. IV.8.3, IX.43–44; IV.10.4, IX.184 (where S· adrā also discusses habituationin terms of traces and blemishes on the mirror of the soul); IV.11.20, IX.404–405, 407–408; andIV.11.26, IX.464.


praised be He: therefore We have now removed your covering, so your sighttoday is piercing.85

Immersed in perception and its various investments in the perceived world,the human soul is incapable of directing its attention to the habituating andaccustoming consequences of its acts in and relations to the world. As aconsequence, it lacks a true grasp of this opaque aspect of itself and isunaware of its habituations and character traits as such, apprehendinginstead their traces in its present objects of experience.86 The meaning ofthe opening of the soul’s book, ‘when the veil is lifted . . . and the cover . . .removed’, is precisely that the human being is forced to face the charactertraits by which she has been determined all along but which have thithertoescaped her explicit awareness. The pleasure or pain she is due in the afterlifeconsists precisely in her immediate awareness of herself down to themurkiest recesses of her being.87

This distinction between the opacity and transparency of the human selfis particularly prominent in Asfār IV.11.20, the eschatological chapter dedi-cated to explaining the symbol of the opening of the soul’s book. It alsocontains formulations that clearly indicate S· adrā’s sustained insistence onthe unity of the self and its characteristics. Having stated that all morallyrelevant acts, even ‘an atom’s weight of good or evil’ that a person hasperformed in this world are inscribed in her soul, S· adrā states:

When the resurrection takes place and the time comes for his sight to fallupon the face of his self (dhātihi), due to his emptiness of the preoccupationsof this mundane life and what the senses have brought him, and to turn tothe page of his interior and the tablet of his conscience (ilā s.afh. ati bāt.inihi walawh. i d

˙amīrihi), which is the meaning of His saying, praised be He: when the

pages are spread out, then he who is heedless of the states of his soul and theaccount of his vices and virtues will say at the unveiling of his cover, presenceof his self (h.ud

˙ūri dhātihi), and reading of the page of his book: what is in this

book leaves neither the minute nor the great unaccounted, and they will findpresent what they have done, your Lord does no injustice to anyone, on the day onwhich every soul finds present what good it has done and what vicious it has done,wishing that there were a great distance between it and itself.88

In the very sentence that emphasizes the soul’s lack of awareness of its owncharacteristics, S· adrā states that they are recognized by turning the gaze ofawareness upon oneself. There are thus aspects of the self that can become

85 Asfār IV.8.3, IX.48–49; S· adrā quotes Q 50:22.86 See also Asfār IV.10.2, IX.169–170; and IV.10.3, IX.177–182. 87 Asfār IV.11.33, IX.539.88 Asfār IV.11.20, IX.408; S· adrā quotes Q 81:10, 18:49 and 3:30, respectively.


present to the self, but this is possible only because they were not such tobegin with. The wish in Q 3:30 at the end is thus counterfactual in a veryradical sense, for there can hardly be a distance, great or small, between theself and its characteristics if they are one with each other.That S· adrā intends his eschatology to be coherent with his theory of

cognitive unity is corroborated by a later passage from the same chapter:

What gardens, trees, rivers and so forth there are in the other [world] arespirits, which are nothing other than (hiya bi ‘aynihā) suspended forms thatsubsist by themselves (bi dhātihā) and whose life is identical with theiressence (nafsu dhātihā), and every human soul, together with what houris,castles, trees and rivers are attached to it, the whole of them exists throughone existence and lives through one life, the whole with its individual unitybeing manifold in form.When the human has departed from the world and has divested of the

garb of this nethermost, and this cover is removed from his sight, hisapprehensive faculty is a power, his knowledge hidden, and what is hiddenof him is manifest, so that he comes to see the consequences of his deeds andthoughts, to behold the traces of his movements and acts, reading the scrollof his deeds and the tablet of his book, his virtues and vices, as He said,praised be He: and upon every human being we have forced his bird upon hisneck, and we shall bring forth for him on the day of resurrection a book he shallfind wide open. Read your book, your soul suffices you today as an accountant.89

A close reading of the passage will reveal virtually all the peculiarities ofS· adrā’s conception of human selfhood in operation. The unity of thesubject and object of knowledge provides the basis for identifying the soulwith its particular lot in the hereafter; the soul does not merely have ‘whathouris, castles, trees and rivers are attached to it’, rather the whole of the souland its characteristics ‘exists through one existence’ as an ‘individual unity . . .manifold in form’. On the other hand, the idea of substantial change in thehuman self, the very core of the human being, though perhaps less empha-sized, seems nevertheless to be required in order to account for the accu-mulation of the determinations that are the individual’s afterlife. S· adrā’ssystematic interpretation of the Qur’ānic theme of a superior cognitiveperspective reached at the apex of one’s development in death, when onefully realizes not only what one was on the way to becoming but also whatthe road to that apex was really all about, relies on the idea that later andhigher developmental stages contain the earlier and lower, an idea which isexplicable only in the framework of the theory of substantial change.

89 Asfār IV.11.20, IX.411; S· adrā quotes Q 17:13–14. Cf. IV.11.21, IX.413–414; IV.11.26, IX.469; IV.11.27,IX.480–481.


A later formulation makes this point even more explicit by undoing thestrict difference between our present life and that post-mortem: ‘Know . . .that the garden, to which he who is of its people will arrive, is visible to youtoday with respect to its substrate (min h. aythu mah. allihā), not with respectto its form, and you dwell in it in the state you are in, yet you do not knowthat you are in it (wa anta tutaqallibu fīhā ‘alā al-h.āli al-latī anta ‘alayhā walā ta‘alimu annaka fīhā).’90 In our present state we already are living theafterlife in the sense that we bear in us the potency out of which the superiorcognitive perspective evolves, or to put it another way, it is our present selfthat we come to regard in the afterlife. It will be different from the way it isnow in the sense that it will then be informed by the superior perspectiverevealed in death, but it will also be the same in the sense that participationin the same act of existence will enable the later self to recognize itself in andidentify with the earlier. The counterpart of this idea is, however, that thereis something in the very core of me that I presently lack awareness of despitemy constant awareness of myself.

Although he treads faithfully along the path opened by Avicenna’s andSuhrawardī’s arguments related to self-awareness, its primitiveness, and itsirreducibility to any other type of cognition, S· adrā demotes the phenom-enon from the pivotal role it once had, particularly in Suhrawardī whoplaced the phenomenon at the foundation of his revisionist epistemologyand metaphysics. As a result, his discussion of self-awareness is somewhatless systematic, and we have been forced to gather together the passagesunder investigation from a number of different contexts of discussion. Thisnotwithstanding, I hope to have shown that S· adrā’s treatment of self-awareness is not only systematic but also highly original. This is due tohis sustained attempt to fit the phenomenon into the general frameworkestablished by the two theories that, in addition to the foundationality ofexistence (as.āla al-wujūd), are his true labours of love.

The theory of cognitive unity, as we have seen, is problematic for thenarrow concept of selfhood as pure first-personality or I-ness. If the subjectand the object of knowledge are inseparably one in an act of cognition, itseems unwarranted to postulate that the I we can separate from the it inanalysis could exist as such, isolated and alone, in reality. As we have seen,this is precisely the conclusion S· adrā draws: selfhood as pure first-personality is a mere concept, whereas in reality the I is always determinedby its acts, perceptions, desires and so forth. This does not mean that the

90 Asfār IV.11.26, IX.468.


concept is purely fictional, for it does refer to a structural feature in all ouracts and experiences: all that we do we do as first-personal agents, and allthat we are aware of we are aware of in the first person. What he does reject,however, is the postulation of a real entity that corresponds exclusively toour first-personal reference.On the other hand, S· adrā’s theory of substantial change further under-

mines the Avicennian identification between the self and the stablesubstantial core of the human being. If everything in us is subject todevelopment, then there seems to be no reason to assume that I remainthe same throughout my existence. Earlier on, we saw Suhrawardī tenta-tively introduce the idea that self-awareness comes in degrees, but it is in theframework of S· adrā’s theory of substantial change that the gradation of self-awareness blooms in full. As we develop in the manner proper to humanbeings from perceiving entities to ones capable of imagination, and ulti-mately towards increasingly complete subjects of intellection, we therebyacquire more perfect modes of being in the first person. This, however,requires S· adrā to forge a new explanation for the phenomenologicallyplausible view that we have a first-personal identity that is capable ofsome stability in the constant fluctuation of its properties. And as we haveseen, S· adrā no longer ascribes our identity to a stable substantial core to ourbeing but conceives of it as a feature that belongs and is due to our existenceas a continuous whole. A corollary of this novel concept of identity is thatwe are no longer as transparent to ourselves as Avicenna and Suhrawardīheld; there are determining features in us that we are not fully aware of, andinstead of merely accidental appendices, these are constitutive to our veryselves. I do not want to claim that S· adrā was intent on developing anythinglike a theory of the unconscious, for it seems clear that the discussion of theself’s opacity is indeed a corollary of other, more important concerns. Yetthe scattered but systematic remarks he makes seem almost epochal fromthe point of view of our present focus.


conclu s ion

Who is the I?

Much of the preceding hinges on the reconstruction of shu‘ūr bi al-dhāt aspure first-personality, or the sort of existence proper to an I abstractedfrom all other considerations. Such a result may appear disappointinglymeagre and short of explanatory force, for surely the natural thing for anyphilosophically oriented mind is to ask, first, how first-personality isbrought about, and second, what being an I consists in. I believe, however,that when the plausibility of the reconstruction is tested in the widercontext of post-Avicennian Islamic philosophy, we find that the contin-uous discussion not only corroborates the legitimacy of the reconstructionbut also alleviates the initial disappointment by providing some flesh overits bare bones. It is true that I-ness remains a primitive fact throughout ourperiod; it is not explained by recourse to any more foundational type ofexistence, or anything more evident or better known, either in itself or tous. All our experience is simply taken to be given to us in the first person,as is proper to a certain class of beings, namely immaterial intellects. Fromthe point of view of a contemporary physicalist, this may seem a strikinglyunreflected, dogmatic and unphilosophical account. But if we adopt abroader and more charitable stance, we can read Avicenna as indeedpresenting a kind of answer, for when he states that self-awareness is theexistence of an intellectual substance, he assigns the phenomenon itsproper place in the available categories of being. His answer is arguablyunnuanced, but in essence it is of the same form as that of the contem-porary physicalist whose reductionism he would sternly reject; the mini-mal account just is the most thorough explanation of self-awarenessavailable in the framework of his metaphysics. This is highlighted bySuhrawardī’s antipodal strategy in which self-awareness is taken to befoundational to all explanation, something that cannot be explained butrather provides the Archimedean point for the knowledge and explanationof all other things.

228

But from the historian’s point of view, the second question concerningthe constitution of I-ness is perhaps the more interesting one, and corre-spondingly, the failure to perceive its relevance a clear sign of the philo-sophical barrenness of a context of discussion. Again, it is true that Avicennaheld the human substance at its barest, in the undeveloped state of mere firstperfection, to be nothing but a first-personal perspective to a variety of itspotential determinations, that is, to the various acts human beings arecapable of, or to the perceptions, volitions and cognitions they have themeans to acquire. Considered in isolation, this perspective is not constitutedby anything, and thus cannot be described or defined by means of anythingmore elementary. This, however, does not mean that Avicenna consideredsuch pure first-personality to be the whole story about our possibilities toexist in the first person. Although he did not present a sustained analysis, theremarks he makes on self-reflection and the individuation of human beings,for instance, suggest that his account of the concrete first-personalityinstantiated in persons like you and me would have been considerablymore complex and inclusive of the various accidental determinations wein fact have. The I-ness he focuses on is an abstraction, a minimal conditionwe must fulfil in order to exist in the first place, and it is only this aspect ofus that he holds to be unanalysable. From this point of view, S· adrā’srejection of the self’s separability from its determinations, as well as hisalternative conception of first-personality as a structural constituent of thecomplex whole of experience with its variety of objective content andepistemic or conative attitudes, is not so much a contrary theory of theself as a refusal to admit the self-sufficient existence of the transcendentalconditions of our empirical existence – if a characterization in foreign termsis pardoned. For S· adrā, although we are always minimally aware of ourselvesin the sense of being in the first person, there are no minimal selves. All thathas led me to the present state, as well as all that is to become of me, isessential to me and should therefore be featured in a complete account,indeed the only absolutely valid (even if humanly unattainable) account, ofmy particular self.Notwithstanding the variety of views with regard to the constitution of

the self, it remains the case that the self of all Islamic philosophers issomething each of us simply has to accept as given. In a sense, I am whatI am irrespective of my own choice and effort. Nothing like the malleableself which is a product of social construction, various contingent economic,historical and libidinal factors, or even the individual’s reflective efforts toguide her own existence, and which becomes increasingly to the fore in earlymodern European thought, seems to emerge in the Islamic context. As I


hope the foregoing clearly indicates, this is not simply because thinkers inclassical Arabic lacked the conceptual means to arrive at such a radicallyopen concept of the self. S· adrā’s new concept of individual identity isconsiderably detached from the strong species-specific determination thatarguably still governs the intuitions of Avicenna and Suhrawardī, andalthough it was based on the idea of a fixed essence known to God, theexplication of which the existence of an individual human self is and whichgoverns the course of the self’s development in a strongly deterministicfashion, there seems to be no conceptual necessity why further analysiscould not have separated the idea of a constructible self from that particulardoctrinal fixture. Yet it remains a fact that this final move was never made,and a natural question to ask is why that is the case.

Questions of this sort bring us to the limits of the explanatory power of ahistory of philosophy narrowly conceived. To even attempt an answer, oneshould adopt Charles Taylor’s seminal method of procedure and take intoaccount as much as possible of the complex set of extraphilosophicalhistorical factors, stemming from and related to the spheres of religiousand ethical intuitions, natural science, economy, urban and rural planning,politics, and so on and so forth. I will be the first to confess that such aninvestigation is beyondmy capacity, knowledge and inclination. However, Iwould also like to add that it may be hasty, and slightly too easy for comfort,to jump to the conclusion that the discussion on self-awareness provides uswith another example of the stifling influence of religion, and in particularof the stagnation of Islamic intellectual life after the classical and golden eraof science and philosophy. Regardless of our ethical convictions, it is notclear whether the liberal individualistic intuition of the construability of theself stands on a solid theoretical ground. Perhaps the openness of the self toour own efforts of construction is merely an illusion, albeit an arguablybeneficial one, and perhaps we really are determined down to the core of ourbeing. Perhaps the self, which I have chosen to identify with and whichtherefore provides me with a narrative bridge from my present person tothose in my past, is a mere conceptual fiction that fails to reproduce the trueconnections that are exclusively due to the continuous stream of existence,not to any enduring characteristics, regardless of how central to my person Itake them to be. Perhaps the me that I now reflectively grasp is connected tothe me ten years, ten days or even ten minutes ago in ways my grasp fails toencompass; and perhaps, regardless of the decisions I now believe I ammaking, that stream of existence will lead to a me ten minutes, days or yearslater that is connected to the present in all sorts of ways that will remainobscure to me – until the Final Day, in any case. Perhaps you, aware of

230 Conclusion: Who is the I?

yourself reading this line, really are but a phase in a continuity the logic ofwhose inevitable development is simply inaccessible to you.From this point of view, S· adrā’s conception of the self and its true

identity suddenly appears decidedly less dated. On the contrary, the trajec-tory of the Islamic discussion on the self seems to have led us to a set ofquestions that remain a matter of acute debate in the philosophy of mindboth westwards and eastwards from the Islamicate world. Prominent phi-losophers of mind with a strong physicalistic bent have recently found aninteresting piece of shared ground with Buddhist contesters of the self ’sreality, and the question of whether there is such a thing as the self isarguably more in vogue than ever before, at least in the history of thephilosophical tradition that goes back to Plato and Aristotle. Interestingly, itseems that in this debate between the challengers and the defenders of theself,1 S· adrā would represent a middle position that is opened by means of afurther question. Supposing that there is some reality to the self, do I haveany reason to believe that my concept of my self is capable of matching thatreality? S· adrā’s radical doubt of the endurance of any of the momentaryselves that we can delineate in reflective investigation should not bestraightforwardly identified as a denial of the self, for he clearly recognizesthat there is a principle of unity operative in our conscious experience,which we can legitimately characterize as a self. Our stream of consciousnessis not a haphazard series of abrupt cuts between unrelated moments, but acontinuity with an internal logic of development that corresponds to a unityof origin and purpose. However, that unity is not due to any principle thatendures immune to change within the stream of conscious experience;rather, the principle is embodied in the stream as a whole. As a consequence,our self as this principle of unity is not available to our consideration fromany single moment of the stream. Trapped within the constant change ofthe stream, we are limited to a participant’s point of view. Thus, for S· adrāthe self is unreal if we mean by it some stable thing corresponding to theconcept that I form through a process of conceptual abstraction from thecontinuous stream of self-aware experience, but this is of no consequencefor the self as the principle of unity embodied in the continuity as a whole.Neither a no-self doctrine like that of the Buddhist philosophers and thecontemporary deniers of the self, nor its opposite according to which the selfis a transcendental principle that endures experiential change and can beaccurately grasped from within the empirical stream of consciousness that itstructures, we should perhaps characterize S· adrā’s view as revolving on the

1 Summarized with exemplary conciseness in Zahavi 2011.


idea of a ‘super-self’ underlying all the efforts, thoughts and characteristicsthat I consider as my own. I am a product, or a creature, brought about inthe unfolding of that greater self, and although I have my momentary shareof its light, I should eventually realize my own limits and withdraw to theshadows from obscuring the respective share of all the other I’s that remainto be generated.

Finally, perhaps this distinction between an unreal and an eminently realself provides us with a vantage point from which to reassess the problematicrelation between the rational and philosophical on the one hand, and themystical and religious on the other, in S· adrā and later Islamic philosophy atlarge. Perhaps S· adrā’s emphatic exhortation to follow the sharī‘a strictly andto engage in supererogatory acts of abstinence and devotion, an undeniablefeature of his writing that is sometimes supposed to represent a mystical orirrational aspect of his thought and to somehow transcend his efforts intheoretical reasoning, can be understood as referring to a means of identify-ing with the greater self. According to this view, these practices, whichcontinue to be venerated by a great many Muslims, are reappropriated as amethod of divesting oneself of erroneous limited conceptions and beliefsregarding an enduring self, and of connecting instead more immediately tothe basis and origin of one’s experience. If that is the case, it is important tobe clear about the question of in what sense exactly these aspects of S· adrā’swriting suggest a step beyond philosophy and rational thought. This is notbecause they represent a superior type of thinking or knowledge, for thesupreme theoretical account of the world of God’s creation remains to beattained through a rigorous process of learning and thought governedby the best principles of philosophy. Rather, they transcend philosophybecause they pertain to being a self, whereas philosophical theorizingremains at the level of what it is to be one. This, however, should notblind us to the fact that it is in works of philosophy that we find thequestions, and answers, of what the individual self is and why she findsherself in the situation that calls for the prescribed remedy.

232 Conclusion: Who is the I?

app end i x

Arabic terminology related to self-awareness

The present study as a whole should stand as an argument for the claim thatthe Arabic philosophers had at their disposal sufficient terminologicalmeans for an analytic discussion of self-awareness. This appendix is there-fore not intended as a further substantiation of this claim, but is rathermeant simply to gather together the central terms for quick reference.

Dhāt. The Arabic term our authors most often use to refer to the self is theambiguous dhāt (pl. dhawāt) which, in addition to its colloquial reflexivemeaning (for instance in phrases like al-rajul bi dhātihi, ‘the man himself’),functions in philosophy as a technical term denoting ‘essence’.1 This it does,moreover, in a most general sense, referring simply to the principle thatmakes a thing the sort of thing it is, and not specifying whether we aredealing with essence as an object of knowledge (māhīya, or ‘quiddity’), asexisting in an active formal principle in matter (t.abī‘a, ‘nature’, or s.ūra,‘form’), or from the point of view of its actualization in reality (h. aqīqa,‘reality’, or annīya, ‘thatness’).2 Thus, in this sense too, dhāt could legit-imately, if somewhat imprecisely be rendered as ‘(the thing) itself’ whencontrasted with the accidental attributes it can lose without turning into adifferent kind of thing. However, in the present study I have adopted amuch more strict translation of dhāt as ‘self’ in roughly the sense in whichthe term is used in contemporary discussions of self-awareness or self-consciousness. I defend this translation in some detail in Chapter 2, butmy final argument for its legitimacy in most appearances of the term in thetexts discussed relies on the massed evidence of those very appearances – itsimply makes eminently good sense to interpret these passages as dealingwith what we would now call a self. Be that as it may, I have made a point ofrigorously providing the Arabic equivalents of all relevant phrases for thecritical reader’s consideration.

1 Cf. Afnan 1964, 101; Rahman 1991. 2 Goichon 1938, 134–137.

233

Nafs. Another ambiguous term often used in the reflexive sense of ‘self’ isnafs (pl. nufūs, or anfus), for example in phrases like al-rajul bi nafsihi, ‘theman himself’. This term is potentially even more problematic than dhātbecause it was adopted as the standard translation of the Greek word psychē,meaning ‘soul’, which is prone to generate confusion in psychological textsdealing with self-awareness. As a frequently occurring Qur’ānic term, it wasalso common in popular ethical and religious texts; for instance, there are anumber of different Sufi classifications of the various types of nafs, a matterwhich in no way alleviates the translator’s arbitration between ‘soul’ and‘self’. In the present study, in order to try and beg as few questions aspossible, I have striven to translate nafs consistently as ‘soul’ whenever thecontext allows, and opt for some variation of ‘self’ only in those cases inwhich ‘soul’ seems obviously ill-fitting. Again, in the latter cases I alwaysprovide the Arabic for the critical reader’s evaluation.

Anā. The first person indexical pronoun anā is of obvious relevance toour topic. The pronoun itself presents no particular difficulties for trans-lation, but difficult questions of interpretation do arise concerning the levelof abstraction at which it is used. Most importantly, is the pronoun used, ina quasi-technical manner, in a generalized description of self-awareness(‘what am I?’ in the sense of ‘what is the I?’) or does it refer in a more casualway to the subject of writing?3 Although Arabic does not standardly demandthe explicit utterance of indexical pronouns, as a result of which their mereappearance often carries a certain air of emphasis, the exact sense in whichanā is used in each of the passages under consideration will have to bedetermined by a close analysis of the context. This, however, is not the caseabout the abstract term anā’īya (‘I-ness’), which was introduced bySuhrawardī and which, as far as I have been able to determine, the philos-ophers use exclusively in discussions on self-awareness.4

‘Ayn. A less frequently occurring term is ‘ayn, which is usually embeddedin phrases like bi ‘aynihi (‘by himself’ or ‘he himself’) or in idioms designedto identify one thing with another such as huwa ‘aynu al-nafs (‘he is[identical with] the soul’), or employed in distinctions between what existsconcretely in extramental reality (‘aynī or fī al-‘ayn) and what exists in themind (dhihnī or fī al-dhihn).5 This term seems not to have been usedtechnically in the philosophical discussion of self-awareness, but it doesoccur in looser turns of phrase, which merits its brief consideration here.

3 A case in point is the extended context of Avicenna, Shifā’: Fī al-nafs V.7, discussed in Chapter 4.1,66–71.

4 For Sufi uses of the term, however, see Afnan 1964, 93–94. 5 Afnan 1964, 101–102.

234 Appendix: Arabic terminology

Annīya, huwīya. Two other expressions, which are tangentially related toself-awareness, are the technical terms annīya (‘thatness’) and huwīya(‘itness’), both of which refer to the existence of an individual thingconsidered as an individual, that is, to the fact that (an or anna) the thingis there or to its being an individual object that can be ostensively referred toas an it (huwa), in distinction from the question of what the thing is, ananswer to which would require subsuming it under some universalconcept.6 However, when the terms were forged in the first wave of thetranslations of Greek philosophical works, they were used rather loosely ascorrelates of clearly distinct concepts in the original, and, as a consequenceof this, their precise sense in subsequent philosophical texts is a constantproblem of interpretation. For example, in Ust.āth’s translation of theMetaphysics, in Ibn Nā‘ima al-H· imsī’s (d. early ninth century CE) renderingof the pseudo-Aristotelian Theology, and in the crypto-Proclean Kitābal-īd


˙, both terms were relied on to translate to on

and to einai, as a result of which they can be used in the sense of ‘being’ (inthe senses of both verbal noun and agent). Sometimes annīya is even used torender to ti ēn einai, Aristotle’s term for ‘essence’, which in theMetaphysics isused synonymously with ousia.7 Finally, the modern Arabic use of huwīya inthe sense of ‘identity’ already figures in al-H· imsī’s Theology, where the termoccasionally translates the Greek tautotēs, but this seems to be a somewhatanomalous case; for instance, Ust.āth’s Metaphysics never uses huwīya fortautotēs, opting instead for the closely related huwa huwa. This staggeringvariety of meanings is to some extent held together by the fact that annīyaand huwīya focus on the individual existence of their reference. Thus, theterms naturally figure in discussions on self-awareness – the entire phenom-enon does, after all, hinge on a cognitive relation of an individual thing toitself – but the Arabic philosophers’ conceptual grasp of the phenomenon isnot primarily dependent on these terms. In most cases we can simplytranslate ‘thatness’ or ‘itness’ without needing to specify in which precisemeaning the terms are used.8

Sha‘ara, shu‘ūr. The cognitive aspect of self-awareness (the ‘awareness’)is most often denoted by the verb sha‘ara and the corresponding verbalnoun shu‘ūr, which I have standardly translated as ‘to be aware’ and‘awareness’, respectively. They figure especially frequently in combination

6 Goichon 1938, 1991; d’Alverny 1959; Afnan 1964, 94–97; and van den Bergh 1991. Cf. Aristotle’sdistinction between to hoti and to ti estin in An. post. II.1, 89b23–25.

7 D’Alverny 1959, 72–74; Goichon 1991; van den Bergh 1991.8 Exceptions to this rule are Avicenna’s controversial use of annīya in the flyingman (see Chapter 2.2, 38–41),and Suhrawardī’s use of huwa and huwīya as contraries of anā and anā’īya (see Chapter 5.1, 111–113).

Appendix: Arabic terminology 235

with dhāt; indeed, although not explicitly defined as such, shu‘ūr bi al-dhātemerges from the texts as a quasi-technical term that very naturally trans-lates as a compound such as ‘self-awareness’. The choice of this particularterm seems to be motivated by its apparent connotation of the phenomenal,experienced or felt aspects of different types of cognition.9

Adraka, idrāk. Another cognitive term often used in discussions of self-awareness is the fourth form verb adraka and its cognate noun idrāk. Thispair of terms commonly functions as a generic expression for all types ofcognition, and I have attempted to capture this sense of generality andvagueness by translating them consistently as ‘to apprehend’ and‘apprehension’.

Other cognitive terms. Many of the most common cognitive terms figureoccasionally in the philosophical discussion of self-awareness. In order ofincreasing specificity, the three most important of these are the verbs ‘arafa,‘alima and ‘aqala, and their cognate verbal forms and nouns.

The somewhat ambiguous verb ‘arafa and the corresponding nounma‘arifa are sometimes taken to have mystical connotations,10 but in theclassical philosophical literature the most prominent use seems to be rathermore ephemeral, simply denoting knowledge, usually in the sense of thepreliminary starting point provided by the senses for more developed typesof knowledge,11 but occasionally also in the sense of recognizing somethingone knew from prior acquaintance.12

The verb ‘alima and the cognate noun ‘ilm, translated here as ‘to know’and ‘knowledge’, respectively, may also be used in an ambiguous and loosemanner, but more commonly they have the technical meaning of scientificknowledge, as is proper for the standard Arabic translation of Aristotle’sepistēmē. A closely related psychological verb is ‘aqala (‘to understand’,sometimes also in the fifth form ta‘aqqala) and the corresponding noun‘aql (‘intellect’). Since the human subject, around which most ofthe discussion on self-awareness revolves, is an intellectual entity, it isunsurprising to come across these terms in our texts. Worth mentioning,however, is the fact that they are rarely used of self-awareness, and thus

9 Cf., for instance, Avicenna’s discussion of pleasure in Ishārāt, namat. 8, 192. For other examples, seeGoichon 1938, 161.

10 This is particularly manifest in the use of the cognate noun ‘irfān for a mystical type of philosophy,particularly in Iran; see Mutahhari 2002, 89–141.

11 So, for instance, in Avicenna, Shifā’: al-Burhān I.3, 12; and I.6, 26–27. For a brief discussion of this useof ma‘rifa, see Adamson 2005, 266–268.

12 Cf., for instance, Avicenna, Ta‘līqāt 147, 161.

236 Appendix: Arabic terminology

when they do figure in the relevant contexts they merit special attention.13

But although both ‘alima and ‘aqala suggest a considerably developed typeof knowledge, and therefore something above ‘mere awareness’, they areoccasionally used in a loose manner that renders them virtually synonymouswith sha‘ara or adraka.14

Finally, many other cognitive terms, such as ra’ā (‘to see’) or fahima (‘tograsp’ or ‘to conceive’) and its passive participlemafhūm (‘concept’), featureinfrequently in the discussions on self-awareness, but in light of the presentstudy it seems warranted to assess that the terminological variation does notsignal any systematic classification of different types of self-cognition. Themost consistent distinction along such lines is that between reflective self-awareness and the more primitive type of self-awareness it presupposes, butthat distinction is usually made by means of the expressions shu‘ūr bi al-dhāt(‘self-awareness’) and shu‘ūr bi al-shu‘ūr (‘awareness of awareness’), and eventhen there is significant terminological vacillation.H· ad˙

ara, h. ud˙ūr, z· ahara, z· uhūr. Finally, the ishrāqī tradition instigated

by Suhrawardī is witness to the emergence of an entirely novel pair ofconcepts, that is, the verb h. ad

˙ara (‘to be present’) and the cognate noun

h. udūr (‘presence’), and the verb z· ahara (‘to appear’) and its cognate nounz· uhūr (‘appearance’), in the introduction of which the phenomenon of self-awareness plays a major role. Since the investigation of these novel conceptsand their relation to the phenomenon of self-awareness is one of the centraltasks of Chapter 6, I refrain from characterizing the concepts here.

13 Such is the case, for instance, in our discussion of the epistemic category proper to Avicennian self-awareness in Chapter 4.3, 97–103.

14 Cf., for instance, S· adrā, Asfār I.10.2.4, III.505, where a familiar Avicennian argument about self-awareness is rendered in terms of knowledge (‘ilm) of the self.

Appendix: Arabic terminology 237

Bibliography

PRIMARY SOURCES

Abū al-Barakāt al-Baghdādī

Mu‘tabar

Yaltkaya, Ş. (ed.). Al-Kitāb al-Mu‘tabar fī al-h. ikma Sayyid al-H· ukamā’ Awh. ad al-Zamān Abī al-Barakāt Hibat Allāh ibn ‘Alī ibn Malkā al-Baghdādī. 3 vols.Hyderabad: Dā’ira al-Ma‘ārif al-‘Othmānīya. 1357–1358 H.

Arist.ūt.ālīs (Aristotle)

Al-T. abī‘a

Badawī, ‘A. (ed.). Arist.ūt.ālīs al-T. abī‘a, tarjama: Ish. āq ibn H· unayn, ma‘a shurūh. IbnShamh. wa Ibn ‘Adī wa Mattā ibn Yūnus wa Abī al-Faraj ibn al-T. ayyib. 2 vols.Cairo: Al-Dār al-Qawmīya li al-T. ibā‘a wa al-Nashr. 1964–1965.

Fī al-nafs

Badawī, ‘A. (ed.). Arist.ūt.ālīs Fī al-nafs, ‘Al-ārā’ al-t.abī‘a’ al-mansūb ilā Flūt.arkhus,‘Al-H· iss wa al-mah. sūs’ li Ibn Rushd, ‘Al-Nabāt’ al-mansūb ilā Arist.ūt.ālīs. Cairo:Maktaba al-Nahd

˙a al-Mis.rīya. 1954.

Averroes

in De an.

Crawford, F. S. (ed.). Averrois Cordubensis Commentarium magnum in Aristotelis deanima libros. Cambridge, MA: The Mediæval Academy of America. 1953.

Tafsīr mā ba‘da al-t.abī‘a

Bouyges, M. (ed.). Averroès Tafsir ma ba’d at-tabi’at. 4 vols. Beirut: Dar el-MachreqÉditeurs. 1973.

238

Tahāfut al-tahāfut

Bouyges, M. (ed.). Averroès. Tahafot at-tahafot. Beirut: Imprimerie Catholique. 1930.

Talkhīs kitāb al-h. iss wa al-mah. sūs

Blumberg, H. (ed.). Abū al-Walīd ibn Rushd. Talkhīs kitāb al-h. iss wa al-mah. sūs.Cambridge, MA: The Mediæval Academy of America. 1972.

Avicenna

Ishārāt

Forget, J. (ed.). Ibn, Sīnā. Le Livre des théorèmes et des avertissements publié d’après lesmss de Berlin de Leyde, et d’Oxford. Leiden: Brill. 1892.

Mubāh. athāt

Badawī, ‘A (ed.). Arist.ū ‘inda al-‘arab. Cairo: Maktaba al-Nahd˙a al-Mis.rīya. 1947.

118–239.Bīdārfar, M. (ed.). Al-Mubāh. athāt. Abū ‘AlīH· usayn ibn ‘Abd Allāh ibn Sīnā. Qom:

Intishārāt Bīdārfar. 1371 (solar) H.

Najāt

Fakhry, M. (ed.). Al-H· usayn Abū ‘Alī ibn Sīnā. Kitāb al-Najāt. Beirut: ManshūrātDār al-Āfāq al-Jadīda. 1985.

al-Risāla al-ad˙h. awīya

‘Āsī, H. (ed.). Al-Ad˙h. awīya fī al-ma‘ād li Ibn Sīnā. Beirut: Al-Mu’assasa al-Jāmi‘īya

li al-Dirāsāt wa al-Nashr wa al-Tawzī‘. 1987.

Sharh. kitāb h. arf al-lām

Badawī, ‘A. (ed.). Arist.ū ‘inda al-‘arab. Cairo: Maktaba al-Nahd˙a al-Mis.rīya. 1947.

22–33.

Shifā’: al-Burhān

Badawī, ‘A. (ed.). Avicennae De Demonstratione ex libro ‘Alchifā’’. Cairo:Association of Authorship, Translation and Publication Press. 1954.

Shifā’: Fī al-nafs

Bakoš, J. (ed.). Psychologie d’Ibn Sīnā (Avicenne) d’après son œuvre Aš-Šifā’ I: Textearabe. Prague: Éditions de l’Academie Tchécoslovaque des Sciences. 1956.

Bibliography 239

Rahman, F. (ed.). Avicenna’s De anima (Arabic Text). Being the Psychological Part ofKitāb al-shifā’. London: Oxford University Press. 1959.

Shifā’: al-‘ibāra

El-Khodeiri, M. (ed.). Ibn Sina. Al-Shifa. La Logique III De l’interpretation (al-ibara). Cairo: Dar el-Katibi al-‘Arabi. 1970.

Shifā’: al-Ilāhīyāt

Marmura, M.E. (ed.). Avicenna. The Metaphysics of The Healing. Al-Shifā’: al-Ilāhīyāt.A Parallel English–Arabic Text. Provo: Brigham Young University Press. 2005.

Shifā’: al-Madkhal

Madkour, I., El-Khodeiri, M., Anawati, G. and El-Ahwani, F. (eds.). Ibn Sīnā. Al-Shifā’. La logique I – L’Isagoge (al-madkhal). Cairo: Imprimerie Nationale. 1952.

Shifā’: al-Samā‘ al-t.abī‘ī

McGinnis, J. (ed.). Avicenna. The Physics of the Healing. Al-Shifā’: Al-Samā‘ al-t.abī‘ī. 2 vols. Provo: Brigham Young University Press. 2009.

Ta‘līqāt

Badawī, ‘A. (ed.). Ibn Sīnā. Al-Ta‘līqāt. Cairo: Al-Hay’a al-Mis.rīya al-‘Āmma li al-Kitāb. 1973.

al-Ta‘līqāt ‘alā hawāshin kitāb al-nafs li Arist.ū

Badawī, ‘A. (ed.). Arist.ū ‘inda al-‘arab. Cairo: Maktaba al-Nahd˙a al-Mis.rīya. 1947.

57–116.

Descartes

Discours

Adam, C. and Tannery, P. (eds.). Œuvres de Descartes. Vol. VI. Discours de laméthode & essais. Paris: Vrin. 1983: 1–78.

Al-Fārābī

al-Madīna al-fād˙ila

Walzer, R. (ed.). On the Perfect State (Mabādi’ ārā’ ahl al-madīnat al-fād˙ilah). Abū

Nas.r al-Fārābī. Revised Text with Introduction, Translation, and Commentary.Oxford: Oxford University Press. 1985.

240 Bibliography

Cherni, A. (ed.). Abū Nasr al-Fārābī. Opinions des habitants de la cité vertueuse.Texte, traduction critique et commentaire. Paris: Éditions Albouraq. 2011.

Risāla fī al-‘aql

Bouyges, M. (ed.). Alfarabi. Risalat fi’l-‘aql. Texte arabe integral. Beirut: ImprimerieCatholique. 1988.

Al-Ghazālī

Mishkāt al-anwār

Buchman, D. (ed.). Al-Ghazālī. The Niche of Lights. Mishkāt al-anwār. A ParallelEnglish–Arabic Text: Provo: Brigham Young University Press. 1998.

Al-Iskandar al-Afrūdīsī (Alexander of Aphrodisias)

Fī al-‘aql

Badawī, ‘A. (ed.). Commentaires sur Aristote perdus en grec et autres épîtres. Beirut:Dar el-Machreq Éditeurs. 1971: 31–42.

Finnegan, J. ‘Texte arabe du Peri nou d’Alexandre d’Aphrodise’. Mélanges del’Université Saint Joseph 33/2 (1956): 159–202.

Al-Jāh. iz·al-Masā’il wa al-jawābāt li al-ma‘rifa

Hārūn, ‘A. (ed.). Rasā’il al-Jāh. iz. Vol. IV. Cairo: Maktaba al-Khānjī. 1979. 45–65.

Mullā S· adrā

Al-Ta‘līqāt ‘alā Sharh. H· ikma al-ishrāq

Hirātī, A. A. (ed.).Mullā S· adrā: Ta‘līqāt li Sharh. H· ikma al-ishrāq li Qutb al-Dīn al-Shīrāzī. Tehran: n.p. 1313 H.

Asfār

Khamene’i, S.M. et al. (eds.). Sadr ad-Din Muhammad Shirazi Mulla Sadra(979–1050 A.H.). Al-Hikmat al-muta’aliyah fi’l-Asfar al-arba’a al-aqliyyah(Transcendent Philosophy in Four Intellectual Journeys). 9 vols. Tehran: SadraIslamic Philosophy Research Institute. 1380–1383 (solar) H.

Iksīr al-‘ārifīn

Chittick, W. C. (ed.). Mullā S· adrā. The Elixir of the Gnostics. Iksīr al-‘ārifīn. AParallel English–Arabic Text. Provo: Brigham Young University Press. 2003.

Bibliography 241

Mashā‘ir

Corbin, H. (ed.).Mollâ Sadra Shirazi (980/1572–1050/1640). Le Livre des Pénétrationsmétaphysiques (Kitâb al-Mashâ’ir). Texte arabe publié avec la version persane deBadî’ol-Molk Mirza ’Emadoddawleh. Traduction française et Annotations.Tehran and Paris: Departement d’Iranologie de l’Institut Franco-Iranienand Librairie d’Amerique et d’Orient Adrien-Maisonneuve. 1964.

Ps.-Aristotle

Kitāb al-īd˙āh. fī al-khayr al-mah. d

˙Badawī, ‘A. (ed.). Neoplatonici apud Arabes. Cairo: Maktaba al-Nahd

˙a al-Mis.rīya.

1955. 1–33.

ThA (Uthūlūjiyā Arist.āt.ālīs)

Badawī, ‘A. (ed.). Plotinus apud Arabes. Cairo: Maktaba al-Nahd˙a al-Mis.rīya. 1955.

1–164.

Ps.-Fārābī

Risāla fī al-‘ilm al-ilāhī

Badawī, ‘A. (ed.). Plotinus apud Arabes. Cairo: Maktaba al-Nahd˙a al-Mis.rīya. 1955.

167–183.

Ps.-Hermes (or ps.-Plato)

Kitāb mu‘ādhala al-nafs

Badawī, ‘A. (ed.). Neoplatonici apud Arabes. Cairo: Maktaba al-Nahd˙a al-Mis.rīya.

1955. 51–116.

Fakhr al-Dīn al-Rāzī

Lubāb

Shahāby, M. (ed.). At-Tanbihât val-Eshârât by Shaikh Rais Abu-Ali Sinā andLobab-ol-Eshârât by Fakhrod-Din Rāzi. Tehran: Tehran University Press.1339 H.

Mabāh. ith

Al-Baghdādī, M. (ed.). Al-Mabāh. ith al-mashriqīya fī ‘ilm al-ilāhīyāt wa al-t.abī‘īyātli al-imām Fakhr al-Dīn ibn ‘Umar al-Rāzī. 2 vols. Beirut: Dār al-Kitāb al-‘Arabī. 1990.

242 Bibliography

Sharh. al-ishārāt

Anon. (ed.). Sharh. ay al-Ishārāt li al-khawāja Nas.īr al-Dīn al-T. ūsī wa al-imāmFakhr al-Dīn al-Rāzī. N.p. 1325 H.

Qutb al-Dīn al-Shīrāzī

Sharh. H· ikma al-ishrāq

Hirātī, A. A. (ed.).Mullā S· adrā: Ta‘līqāt li Sharh. H· ikma al-ishrāq li Qutb al-Dīn al-Shīrāzī. Tehran: n.p. 1313 H.

Suhrawardī

H· I (H· ikma al-ishrāq)

Corbin, H. (ed.). Œuvres philosophiques et mystiques (Shihābuddīn YahyāSuhrawardī). Tome II: 1. Le Livre de la Théosophie Orientale, 2. Le Symbol defoi des Philosophes, 3. Le Récit de l’Exil Occidental. Textes édités avecProlégomènes en Français. Tehran: Institute d’Études et des RecherchesCulturelles. 2009. 1–260.

Walbridge, J. and Ziai, H. (eds.). Suhrawardī. The Philosophy of Illumination.H· ikma al-ishrāq. A New Critical Edition of the Text of H· ikmat al-ishrāq withEnglish Translation, Notes, Commentary, and Introduction. Provo: BrighamYoung University Press. 1999.

Mashāri‘

Corbin, H. (ed.). Œuvres philosophiques et mystiques (Shihābuddīn YahyāSuhrawardī) Tome I: La Métaphysique: 1. Kitāb al-Talvīh. āt, 2. Kitāb al-Muqāvamāt, 3. Kitāb al-Mashāri’ va’l-Mut.ārah. āt. Textes édités avecProlégomènes en Français. Tehran: Institute d’Études et des RecherchesCulturelles. 2009. 193–506 (contains only the section on metaphysics).

Partaw-nāma

Ziai, H. (ed.). Sohravardī. The Book of Radiance. Partaw nāma. A Parallel English–Persian Text. Costa Mesa: Mazda Publishers. 1998.

Talwīh. āt

Habibi, N. (ed.). Sohrawardī. Al-Talwīh. āt al-Lawh. iyyah va al-Aršiyyah. Tehran:Iranian Institute of Philosophy. 2009.

Ziai, H. and Alwishah, A. (eds.). Ibn Kammūna. Al-Tanqīh. āt fī Sharh. al-Talwīh. āt.Refinement and Commentary on Suhrawardī’s Intimations. A ThirteenthCentury Text on Natural Philosophy and Psychology. Costa Mesa: MazdaPublishers. 2003. Contains only the section on natural philosophy.

Bibliography 243

Corbin, H. (ed.). Œuvres philosophiques et mystiques (Shihābuddīn YahyāSuhrawardī) Tome I: La Métaphysique: 1. Kitāb al-Talvīh. āt, 2. Kitāb al-Muqāvamāt, 3. Kitāb al-Mashāri’ va’l-Mut.ārah. āt. Textes édités avecProlégomènes en Français. Tehran: Institute d’Études et des RecherchesCulturelles. 2009. 1–121 (contains only the section on metaphysics).

Themistius

Fī al-nafs

Lyons, M. C. (ed.). An Arabic Translation of Themistius’ Commentary on Aristotle’s‘De anima’. London: Cassirer. 1973.

in De an.

Heinze, R. (ed.).Themistius in libros Aristotelem de anima paraphrasis. Commentariain Aristotelem Graeca V/3. Berlin: Reimer. 1899.

Nas.īr al-Dīn al-T. ūsī

Sharh. al-ishārāt

Dunya, S. (ed.). Al-Ishārāt wa al-tanbīhāt li Abī ‘Alī ibn Sīnā ma‘a Sharh. Nas.īr al-Dīn al-T. ūsī. 4 vols. Cairo: Dār al-Ma‘ārif bi Mis.r. 1960–1968.

SECONDARY LITERATURE

Adamson, P. 2002. The Arabic Plotinus: A Philosophical Study of the Theology ofAristotle. London: Duckworth.

2005. ‘On Knowledge of Particulars’. Proceedings of the Aristotelian Society 105/3:257–278.

2011a. ‘Avicenna andHis Commentators onHuman andDivine Self-Intellection’.97–122 in D.N. Hasse and A. Bertolacci (eds.). The Arabic, Hebrew, and LatinReception of Avicenna’s Metaphysics. Berlin and Boston: de Gruyter.

(ed.) 2011b. In the Age of Averroes: Arabic Philosophy in the Sixth/Twelfth Century.London and Turin: The Warburg Institute and Nino Aragno Editore.

Adamson, P. and Taylor, R. C. (eds.) 2005. The Cambridge Companion to ArabicPhilosophy. Cambridge: Cambridge University Press.

Afnan, S.M. 1964. Philosophical Terminology in Arabic and Persian. Leiden: Brill.d’Alverny, M. T. 1959. ‘Anniyya – anitas’. 59–91 inMélanges offerts à Étienne Gilson

de l’Académie Française. Toronto and Paris: Pontifical Institute of MedievalStudies and Vrin.

Amin Razavi, M. 1997. Suhrawardi and the School of Illumination. Richmond:Curzon Press.

Arnaldez, R. 1972. ‘Un précédent avicennien du “Cogito” cartésien?’ Annalesislamologiques 11: 341–349.

244 Bibliography

Arnzen, R. 2007. ‘The Structure of Mullā S· adrā’s al-H· ikma al-muta‘āliya fī l-asfāral-‘aqliyya al-arba‘a andHis Concepts of First Philosophy andDivine Science:An Essay’. Medioevo 32: 199–204.

2011. Platonische Ideen in der arabischen Philosophie: Texte und Materialien zurBegriffsgeschichte von s.uwar aflātūniyya und muthul aflātūniyya. Berlin andBoston: de Gruyter.

Bäck, A. 1994. ‘The Islamic Background: Avicenna (b. 980; d. 1037) and Averroes(b. 1126; d. 1198)’. 39–67 in J. J. E. Gracia (ed.). Individuation in Scholasticism:The Later Middle Ages and the Counter-Reformation 1150–1650. Albany: StateUniversity of New York Press.

Bergh, S. van den 1991. ‘Anniyya’. 513–514 in Lewis et al. 1991.Bertolacci, A. 2006. The Reception of Aristotle’s Metaphysics in Avicenna’s Kitāb al-

šifā’: A Milestone of Western Metaphysical Thought. Leiden and Boston: Brill.Black, D. L. 1993. ‘Estimation (Wahm) in Avicenna: The Logical and Psychological

Dimensions’. Dialogue 32: 219–258.1997. ‘Avicenna on the Ontological and Epistemological Status of FictionalBeings’. Documenti e studi sulla tradizione filosofica medievale 8: 425–453.

2000. ‘Imagination and Estimation: Arabic Paradigms and WesternTransformations’. Topoi 19: 59–75.

2008. ‘Avicenna on Self-Awareness and Knowing that One Knows’. 63–87 inS. Rahman, T. Street, and H. Tahiri (eds.). The Unity of Science in the ArabicTradition: Science, Logic, Epistemology and their Interactions. N.p.: Springer.

2010. ‘Intentionality in Medieval Arabic Philosophy’. Quaestio 10: 65–81.2012. ‘Avicenna on Individuation, Self-Awareness, and God’s Knowledge ofParticulars’. 255–281 in R. Taylor and I. Omar (eds.). The Judeo-Christian-Islamic Heritage: Philosophical and Theological Perspectives. Milwaukee:Marquette University Press.

Bonmariage, C. 2007. Le Réel et les réalités: Mullā S· adrā Shīrāzī et la structure de laréalité. Paris: Vrin.

Cary, P. 2003. Augustine’s Invention of the Inner Self: The Legacy of a ChristianPlatonist. Oxford: Oxford University Press.

Caston, V. 2002. ‘Aristotle on Consciousness’. Mind 111: 751–815.2004. ‘More on Aristotle on Consciousness: Reply to Sisko’.Mind 113: 523–533.

Corbin, H. 1964. ‘Introduction’. 1–86 in H. Corbin (ed.).Mollâ Sadra Shirazi (980/1572–1050/1640): Le Livre des pénétrations métaphysiques (Kitâb al-Mashâ’ir).Texte arabe publié avec la version persane de Badî’ol-Molk Mirza’Emadoddawleh. Traduction française et annotations. Tehran and Paris:Departement d’Iranologie de l’Institut Franco-Iranien and Librairied’Amerique et d’Orient Adrien-Maisonneuve.

1971. En Islam iranien. 4 vols. Paris: Gallimard.2009a. ‘Prolégomènes’. I–LXXXI in H. Corbin (ed.). Œuvres philosophiques etmystiques (Shihābuddīn Yahyā Suhrawardī) Tome I: La Métaphysique: 1. Kitābal-Talvīh. āt, 2. Kitāb al-Muqāvamāt, 3. Kitāb al-Mashāri’ va’l-Mut.ārah. āt.Textes édités avec prolégomènes en Français. Tehran: Institute d’Études et desRecherches Culturelles.

Bibliography 245

2009b. ‘Prolégomènes II’. 1–102 in Corbin, H. (ed.). Œuvres philosophiques etmystiques (Shihābuddīn Yahyā Suhrawardī). Tome II: 1. Le Livre de lathéosophie orientale, 2. Le Symbol de foi des philosophes, 3. Le Récit de l’exiloccidental. Textes édités avec prolégomènes en Français. Tehran: Instituted’Études et des Recherches Culturelles.

Cory, T. S. 2013. Aquinas on Human Self-Knowledge. Cambridge: CambridgeUniversity Press.

Coseru, C. 2012. ‘Mind in Indian Buddhist Philosophy’. In E.N. Zalta (ed.). TheStanford Encyclopedia of Philosophy (Winter 2012). http://plato.stanford.edu/archives/win2012/entries/mind-indian-buddhism/.

Dabashi, H. 1996. ‘Mīr Dāmād and the Founding of the “School of Is.fahān”’.597–634 in Nasr and Leaman 1996.

Davidson, H. A. 1992. Alfarabi, Avicenna, and Averroes, on Intellect: TheirCosmologies, Theories of the Active Intellect, and Theories of the HumanIntellect. New York: Oxford University Press.

de Libera, A. 2007. Archéologie du sujet: naissance du sujet. Paris: Vrin.2008. Archéologie du sujet: la quête de l’identité. Paris: Vrin.

Dennett, D. 1991. Consciousness Explained. Boston: Little, Brown.Druart, T.-A. 1988. ‘The Soul and Body Problem: Avicenna and Descartes’. 27–48

in T.-A. Druart (ed.). Arabic Philosophy and the West: Continuity andInteraction. Washington, DC: Center for Contemporary Arab Studies.

2000. ‘The Human Soul’s Individuation and its Survival after the Body’s Death:Avicenna on the Causal Relation between Body and Soul’. Arabic Sciences andPhilosophy 10: 259–273.

Eichner, H. 2007. ‘Dissolving the Unity of Metaphysics: From Fakhr al-Dīn al-Rāzī to Mullā S· adrā al-Shīrāzī’. Medioevo 32: 139–197.

2009. Towards the Construction of Islamic Orthodoxy: Philosophy in the Post-Avicennian Period and Islamic Theology as Literary Traditions. Habilitationthesis. Martin Luther Universität.

2010. ‘Das Avicennische Corpus Aristotelicum: Zur Virtualisierung desAristotelestextes in der postavicennischen Tradition’. 197–235 in P. Derron,R. Goulet and U. Rudolph (eds.). Entre orient et occident: la philosophie et lascience gréco-romaines dans le monde arabe. Geneva: Fondation Hardt.

2011. ‘“Knowledge by Presence”, Apperception and the Mind–BodyRelationship: Fakhr al-Dīn al-Rāzī and al-Suhrawardī as Representatives ofa Thirteenth Century Discussion’. 117–140 in Adamson 2011b.

Elamrani-Jamal, A. 2003. ‘De anima. Tradition arabe’. 346–358 in R. Goulet et al.(eds.). Dictionnaire des philosophes antiques: supplément. Paris: CNRSÉditions.

El-Rouayheb, K. 2010. Relational Syllogisms and the History of Arabic Logic,900–1900. Leiden and Boston: Brill.

Endress, G. 1990. ‘The Defense of Reason: The Plea for Philosophy in theReligious Community’. Zeitschrift zur Geschichte der arabisch-islamischenWissenschaften 6: 1–49.

246 Bibliography

http://plato.stanford.edu/archives/win2012/entries/mind-indian-buddhism/

http://plato.stanford.edu/archives/win2012/entries/mind-indian-buddhism/

Frank, R.M. 1958–9. ‘Some Fragments of Ish. āq’s Translation of the De anima’.Cahiers de Byrsa 8: 231–251.

Galindo-Aguilar, E. 1958. ‘L’Homme volant d’Avicenne et le ‘Cogito’ deDescartes’. Institut des belles lettres arabes 21: 279–295.

Gallagher, S. (ed.) 2011. The Oxford Handbook of the Self. Oxford: OxfordUniversity Press.

Gallois, A. 2012. ‘Identity over Time’. In E. N. Zalta (ed.). The StanfordEncyclopedia of Philosophy (Winter 2012). http://plato.stanford.edu/archives/sum2012/entries/identity-time/.

Gardet, L. 1951. La Pensée réligieuse d’Avicenne. Paris: Vrin.Gätje, H. 1988. ‘Gedächtnis und Erinnerung bei Avicenna und Averroes’. Acta

orientalia 49: 7–36.Genequand, C. 1986. Ibn Rushd’s Metaphysics: A Translation with Introduction of

Ibn Rushd’s Commentary on Aristotle’s Metaphysics, Book Lām. Leiden: Brill.Geoffroy, M. 2003. ‘Remarques sur la traduction Ust.āt_ du livre Lambda de la

Métaphysique, chapitre 6’. Recherches de théologie et philosophie médiévales 70:417–436.

Gill, C. 1996. Personality in Greek Epic, Tragedy, and Philosophy: The Self inDialogue. Oxford: Clarendon Press.

2006. The Structured Self in Hellenistic and Roman Thought. Oxford: OxfordUniversity Press.

Gilson, É. 1929. ‘Les Sources gréco-arabes de l’augustinisme avicennisant’. Archivesd’histoire doctrinale et littéraire du moyen âge 1: 5–149.

Goichon, A.-M. 1937. La Distinction de l’essence et de l’existence d’Ibn Sīnā(Avicenne). Paris: Desclée de Brouwer.

1938. Lexique de la langue philosophique d’Ibn Sīnā (Avicenne). Paris: Desclée deBrouwer.

1991. ‘Huwiyya’. 644–645 in Lewis et al. (eds.) 1991.Griffel, F. 2009. Al-Ghazālī’s Philosophical Theology. Oxford: Oxford University

Press.Gutas, D. 1988. Avicenna and the Aristotelian Tradition: Introduction to Reading

Avicenna’s Philosophical Works. Leiden: Brill.2012. ‘The Empiricism of Avicenna’. Oriens 40: 391–436.

Ha’iri Yazdi, M. 1992. The Principles of Epistemology in Islamic Philosophy:Knowledge by Presence. Albany: State University of New York Press.

Hamlyn, D.W. 1978. ‘Aristotle’s Cartesianism’. Paideia 7: 8–15.Hardie, W. F. R. 1976. ‘Concepts of Consciousness in Aristotle’.Mind 85: 388–411.Hasse, D.N. 2000. Avicenna’s De anima in the Latin West: The Formation of a

Peripatetic Philosophy of the Soul. London and Turin: The Warburg Instituteand Nino Aragno Editore.

2001. ‘Avicenna on Abstraction’. 39–72 in R. Wisnovsky (ed.). Aspects ofAvicenna. Princeton: Markus Wiener Publishers.

Henrich, D. 1966. ‘Fichtes ursprüngliche Einsicht’. 188–232 in D. Henrich andH. Wagner (eds.). Subjektivität und Metaphysik: Festschrift fur WolfgangCramer. Frankfurt am Main: Vittorio Klostermann.

Bibliography 247

http://plato.stanford.edu/archives/sum2012/entries/identity-time/

http://plato.stanford.edu/archives/sum2012/entries/identity-time/

1970. ‘Selbstbewußtsein, kritische Einleitung in eine Theorie’. 257–284 inR. Bubner, K. Cramer, and R. Wiehl (eds.). Hermeneutik und Dialektik.Tübingen: Mohr.

Jambet, C. 2002. L’Acte d’être: la philosophie de la revelation chez Mollâ Sadrâ. Paris:Fayard.

2008. Mort et resurrection en Islam: l’au-delà selon Mullâ Sadrâ. Paris: AlbinMichel.

Janssens, J. 2003a. ‘Avicenne et sa “paraphrase-commentaire” du livre Lambda(Kitāb al-Ins.āf)’. Recherches de théologie et philosophie médiévales 70: 401–416.

2003b. ‘Bahmanyār ibn Marzubān: A Faithful Disciple of Ibn Sīnā?’ 177–197 inD. C. Reisman and A.H. Al-Rahim (eds.). Before and After Avicenna:Proceedings of the First Conference of the Avicenna Study Group. Leiden: Brill.

2012. ‘Ibn Sīnā’s Ta‘līqāt: The Presence of Paraphrases of and Super-Commentaries on the Ilāhiyāt of the Šifā’’. 201–222 in D. Reisman andF. Opwis (eds.). Islamic Philosophy, Science, Culture, and Religion: Studies inHonor of Dimitri Gutas. Leiden: Brill.

Johansen, T. K. 2005. ‘In Defense of Inner Sense: Aristotle on Perceiving that OneSees’. Proceedings of the Boston Area Colloquium in Ancient Philosophy 21:235–276.

Kahn, C. 1966. ‘Sensation and Consciousness in Aristotle’s Psychology’. Archiv furGeschichte der Philosophie 48: 43–81.

Kalin, I. 2003. ‘An Annotated Bibliography of the Works of Mullā S· adrā with aBrief Account of His Life’. Islamic Studies 42/1: 21–62.

2010. Knowledge in Later Islamic Philosophy: Mullā S· adrā on Existence, Intellect,and Intuition. Oxford: Oxford University Press.

Kaukua, J. 2007. Avicenna on Subjectivity: A Philosophical Study. Jyväskylä:University of Jyväskylä.

2011. ‘I in the Light of God: Selfhood and Self-Awareness in Suhrawardī’sH· ikmat al-ishrāq’. 141–157 in Adamson 2011.

2013. ‘Suhrawardī’s Knowledge as Presence in Context’. Studia orientalia 114:309–324.

2014a. ‘Avicenna on Active Perception’. In J. F Silva and M. Yrjönsuuri (eds.).Active Perception. Dordrecht: Springer.

2014b. ‘The Problem of Intentionality in Avicenna’. Documenti e studi sullatradizione filosofica medievale 25: 215–242.

2014c. ‘A Closed Book: Opacity of the Human Self in Mullā S· adrā’. Vivarium52: 241–260.

Kaukua, J. and Kukkonen, T. 2007. ‘Sense-Perception and Self-Awareness: Beforeand After Avicenna’. 95–119 in S. Heinämaa, V. Lähteenmäki, and P. Remes(eds.). Consciousness: From Perception to Reflection in the History of Philosophy.Dordrecht: Springer.

Kaukua, J. and Lähteenmäki, V. 2010. ‘Subjectivity as a Non-textual Standard ofInterpretation in the History of Philosophy’. History and Theory 49/1: 21–37.

Kenny, A. 1988. The Self. Milwaukee: Marquette University Press.

248 Bibliography

1999. ‘Body, Soul, and Intellect in Aquinas’. 33–48 in J. Crabbe (ed.). From Soulto Self. London: Routledge.

Kosman, L. A. 1975. ‘Perceiving that We Perceive: On the Soul III, 2’. PhilosophicalReview 84: 499–519.

Lagerlund, H. 2008. ‘The Unity of the Soul and Contrary Appetites in MedievalPhilosophy’. 75–91 in Lennon and Stainton 2008a.

Lameer, J. 2006. Conception and Belief in S· adr al-Dīn Shīrāzī (ca 1571–1635): Al-Risāla fī l-tas.awwur wa-l-tas.dīq. Introduction, Translation and Commentary.Tehran: Iranian Institute of Philosophy.

Lennon, T.M. and Stainton, R. J. (eds.) 2008a. The Achilles of RationalistPsychology. N.p.: Springer.

2008b. ‘Introduction’. 1–18 in Lennon and Stainton 2008a.Lewis, B., Ménage, V. L., Pellat, C., and Schacht, J. (eds.) 1991. The Encyclopedia of

Islam: New Edition. Leiden and London: Brill and Luzac Co.Lewis, F. A. 1996. ‘Self-Knowledge in Aristotle’. Topoi 15: 39–58.López-Farjeat, L. X. 2012. ‘Self-Awareness (al-shu‘ūr bi-al-dhāt) in Human and

Non-Human Animals in Avicenna’s Psychological Writings’. 121–140 inA. G. Vigo (ed.). Oikeiosis and the Natural Bases of Morality: From ClassicalStoicism to Modern Philosophy. Hildesheim, Zurich, and New York: GeorgOlms Verlag.

Lyons, M.C. 1973. ‘Introduction’. VII–XVI in M.C. Lyons (ed.). An ArabicTranslation of Themistius’Commentary on Aristotle’s ‘De anima’. London: Cassirer.

Marcotte, R.D. 2004. ‘Irja‘ ilā nafsika: Suhrawardī’s Apperception of the Self inLight of Avicenna’. Transcendent Philosophy 5/1: 1–21.

2011. ‘Al-Masā’il al-qudsiyya and Mullā S· adrā’s Proofs for Mental Existence’.Journal of Islamic Studies 22/2: 153–182.

2012. ‘Suhrawardī’. In E. N. Zalta (ed.). The Stanford Encyclopedia ofPhilosophy (Summer 2012). http://plato.stanford.edu/archives/sum2012/entries/suhrawardi/.

Marenbon, J. 2006. Medieval Philosophy: An Historical and PhilosophicalIntroduction. London: Routledge.

Marmura, M. E. 1960. ‘Avicenna and the Problem of the Infinite Number ofSouls’. Medieval Studies 22: 232–239.

1962. ‘Some Aspects of Avicenna’s Theory of God’s Knowledge of Particulars’.Journal of the American Oriental Society 82: 299–312.

1986. ‘Avicenna’s “Flying Man” in Context’. The Monist 69: 383–395.1991. ‘Fakhr al-Dīn al-Rāzī’s Critique of an Avicennan Tanbīh’. 627–641 inB. Mojsisch and O. Pluta (eds.). Historia philosophiae medii aevi: Studien zurGeschichte der Philosophie des Mittelalters. Amsterdam and Philadelphia:B. R. Grüner.

2008. ‘Some Questions Regarding Avicenna’s Theory of the TemporalOrigination of the Human Rational Soul’. Arabic Sciences and Philosophy 18:121–138.

Martin, R. and Barresi, J. 2006. The Rise and Fall of Soul and Self: An IntellectualHistory of Personal Identity. New York: Columbia University Press.

Bibliography 249

http://plato.stanford.edu/archives/sum2012/entries/suhrawardi/

http://plato.stanford.edu/archives/sum2012/entries/suhrawardi/

McGinnis, J. 2003. ‘Scientific Methodologies in Medieval Islam’. Journal of theHistory of Philosophy 41/3: 307–327.

2010. Avicenna. Oxford: Oxford University Press.Metzinger, T. 2003. Being No-one: The Self-Model Theory of Subjectivity.

Cambridge, MA: MIT Press.2011. ‘The No-Self Alternative’. 279–296 in Gallagher 2011.

Michot, J. R. 1997. ‘La Réponse d’Avicenne à Bahmanyâr et al-Kirmânî:présentation, traduction critique et lexique arabe-français de la Mubâh. athaIII’. Le Muséon 110: 143–221.

Michot, Y. J. 2003a. ‘A Mamlūk Theologian’s Commentary on Avicenna’s Risālaad˙h. awiyya: Being a Translation of a Part of the Dar’ al-ta‘ārud

˙of Ibn

Taymiyya, with Introduction, Annotation and Appendices, Part I’. Journalof Islamic Studies 14/2: 149–203.

2003b. ‘A Mamlūk Theologian’s Commentary on Avicenna’s Risāla ad˙h. awiyya:

Being a Translation of a Part of the Dar’ al-ta‘ārud˙of Ibn Taymiyya, with

Introduction, Annotation and Appendices, Part II’. Journal of Islamic Studies14/3: 309–363.

Mijuskovic, B. L. 1974. The Achilles of Rationalist Arguments: The Simplicity, Unity,and Identity of Thought and Soul from the Cambridge Platonists to Kant: A Studyin the History of an Argument. The Hague: Martinus Nijhoff.

Modrak, D. 1981. ‘An Aristotelian Theory of Consciousness?’ Ancient Philosophy 1:160–170.

1987. Aristotle: The Power of Perception. Chicago: University of Chicago Press.Moraux, P. 1978. ‘Le De anima dans la tradition grecque. Quelques aspects de

l’interprétation du traité, de Théophraste à Thémistius’. 281–325 inG. E. R. Lloyd and G. E. L. Owen (eds.). Aristotle on Mind and the Senses:Proceedings of the Seventh Symposium Aristotelicum. Cambridge: CambridgeUniversity Press.

Morris, J.W. 1981. The Wisdom of the Throne: An Introduction to the Philosophy ofMulla Sadra. Princeton: Princeton University Press.

Muehlethaler, L. 2009. ‘Ibn Kammūna (d. 683/1284) on the Argument of theFlying Man in Avicenna’s Ishārāt and al-Suhrawardī’s Talwīh. āt’. 179–203 inY. T. Langermann (ed.). Avicenna and His Legacy: A Golden Age of Science andPhilosophy. Turnhout: Brepols.

Mutahhari, Murtada 2002. Understanding Islamic Sciences: Philosophy, Theology,Mysticism, Morality, Jurisprudence. London: ICAS Press.

Nasr, S. H. 1963. ‘S· adr al-Dīn Shīrāzī (Mulla S· adra)’. 932–961 in M.M. Sharif(ed.). A History of Muslim Philosophy with Short Accounts of Other Disciplinesand the Modern Renaissance in Muslim Lands. 2 vols. Wiesbaden: OttoHarrassowitz.

1978. Sadr al-Dīn Shīrāzī and His Transcendent Theosophy: Background, Life andWorks. Tehran: Imperial Iranian Academy of Philosophy.

1996. ‘Mullā S· adrā: His Teachings’. 643–662 in Nasr and Leaman 1996.Nasr, S. H. and Leaman, O. (eds.) 1996. History of Islamic Philosophy. London:

Routledge.

250 Bibliography

Olson, E. 1998. ‘There is No Problem of the Self’. Journal of Consciousness Studies 5:645–657.

Osborne, C. 1983. ‘Aristotle,De Anima 3.2: How DoWe Perceive that We See andHear?’ Classical Quarterly 33: 401–411.

Perälä, M. 2010. ‘Aristotle on the Perception of Perception: Seeing, Remembering,and Sleeping’. Ph.D. dissertation. University of Helsinki.

Peters, F. E. 1968. Aristoteles Arabus: The Oriental Translations and Commentarieson the Aristotelian Corpus. Leiden: Brill.

Pines, S. 1954. ‘La Conception de la conscience de soi chez Avicenne et chez Abu’l-Barakat al-Baghdadi’. Archives d’histoire doctrinale et littéraire de moyen âge 29:21–98.

1970. ‘Philosophy’. 780–823 in P.M. Holt, A. K. S. Lambton, and B. Lewis(eds.). The Cambridge History of Islam. 2 vols. Cambridge: CambridgeUniversity Press.

Polansky, R. 2007. Aristotle’s De anima. Cambridge: Cambridge University Press.Pourjavady, R. 2011. Philosophy in Early Safavid Iran: Najm al-Dīn Mah.mūd al-

Nayrīzī and His Writings. Leiden: Brill.Rahman, F. 1952. Avicenna’s Psychology: An English Translation of Kitāb al-najāt,

Book II, Chapter VI with Historico-Philosophical Notes and TextualImprovements on the Cairo Edition. Oxford: Oxford University Press.

1958. Prophecy in Islam: Philosophy and Orthodoxy. London: George Allen andUnwin.

1975. The Philosophy of Mullā S· adrā (S· adr al-Dīn al-Shirāzī). Albany: StateUniversity of New York Press.

1991. ‘Dhāt’. 220 in Lewis et al. 1991.Reisman, D. C. 2002. The Making of the Avicennan Tradition: The Transmission,

Contents, and Structure of Ibn Sīnā’s al-Mubāh. athāt (The Discussions). Leiden:Brill.

Remes, P. 2007. Plotinus on Self: The Philosophy of the ‘We’. Cambridge: CambridgeUniversity Press.

Rizvi, S. 2005. ‘Mysticism and Philosophy: Ibn ‘Arabī and Mullā S· adrā’. 224–246in Adamson and Taylor 2005.

2007. Mullā S· adrā Shīrāzī: His Life and Works and the Sources for SafavidPhilosophy. Oxford: Oxford University Press.

2009. Mullā S· adrā and Metaphysics: Modulation of Being. London: Routledge.Rustom, M. 2012. The Triumph of Mercy: Philosophy and Scripture in Mullā S· adrā.

Albany: State University of New York Press.Sebti, M. 2000. Avicenne: L’âme humaine. Paris: Presses Universitaires de France.Shihadeh, A. 2005. ‘From al-Ghazālī to al-Rāzī: 6th/12th Century Developments in

Muslim Philosophical Theology’. Arabic Sciences and Philosophy 15/1: 141–179.2006. The Teleological Ethics of Fakhr al-Dīn al-Rāzī. Leiden: Brill.

Siderits, M. 2011. ‘Buddhist Non-Self: The No-Owner’s Manual’. 297–315 inGallagher 2011.

Sisko, J. E. 2004. ‘Reflexive Awareness Does Belong to the Main Function ofPerception: Reply to Victor Caston’. Mind 113: 513–521.

Bibliography 251

Skinner, Q. 1969. ‘Meaning and Understanding in the History of Ideas’. Historyand Theory 8/1: 3–53.

Sorabji, R. 2006. Self: Ancient and Modern Insights about Individuality, Life, andDeath. Chicago: University of Chicago Press.

Strawson, G. 2009. Selves: An Essay in Revisionary Metaphysics. Oxford: ClarendonPress.

Taylor, C. 1989. Sources of the Self: The Making of the Modern Identity. Cambridge,MA: Harvard University Press.

Taylor, R. C. 2009. Averroes (Ibn Rushd) of Cordoba: Long Commentary on the Deanima of Aristotle. Translated and with Introduction and Notes. New Havenand London: Yale University Press.

Thiel, U. 2011. The Early Modern Subject: Self-Consciousness and Personal Identityfrom Descartes to Hume. Oxford: Oxford University Press.

Treiger, A. 2012. Inspired Knowledge in Islamic Thought: Al-Ghazālī’s Theory ofMystical Cognition and its Avicennian Foundation. London and New York:Routledge.

Walbridge, J. 1992. The Science of Mystic Lights: Qutb al-Dīn Shīrāzī and theIlluminationist Tradition in Islamic Philosophy. Cambridge, MA: HarvardUniversity Press.

2000. The Leaven of the Ancients: Suhrawardī and the Heritage of the Greeks.Albany: State University of New York Press.

2001. The Wisdom of the Mystic East: Suhrawardī and Platonic Orientalism.Albany: State University of New York Press.

2005. ‘Suhrawardī and Illuminationism’. 201–223 in Adamson and Taylor 2005.Walbridge, J. and Ziai, H. 1999. ‘Translators’ Introduction’. xv–xxxvii in

J. Walbridge and H. Ziai (eds.). Suhrawardī: The Philosophy of IlluminationH· ikma al-ishrāq. A New Critical Edition of the Text of H· ikmat al-ishrāq withEnglish Translation, Notes, Commentary, and Introduction. Provo: BrighamYoung University Press.

Walzer, R. 1954. ‘A Diatribe of Galen’. Harvard Theological Review 47: 243–254.Weisheipl, J. A. 1982. ‘Aristotle’s Concept of Nature: Avicenna and Aquinas’.

137–160 in L. D. Rogers (ed.). Approaches to Nature in the Middle Ages.Binghamton: Center for Medieval and Early Renaissance Studies.

Wilkes, K. V. 1993. Real People: Personal Identity without Thought Experiments.Oxford: Oxford University Press.

Wisnovsky, R. 2003. Avicenna’s Metaphysics in Context. Ithaca, NY: CornellUniversity Press.

2004a. ‘One Aspect of the Avicennian Turn in Sunnī Theology’. Arabic Sciencesand Philosophy 14/1: 65–100.

2004b. ‘The Nature and Scope of Arabic Philosophical Commentary in Post-Classical (ca. 1100–1900 AD) Islamic Intellectual History: Some PreliminaryObservations’. 149–191 in P. Adamson, H. Balthussen, and M.W. F. Stone(eds.), Philosophy, Science and Exegesis in Greek, Arabic and LatinCommentaries. London: Institute of Classical Studies.

252 Bibliography

Wolfson, H. A. 1973. ‘The Internal Senses in Latin, Arabic, and HebrewPhilosophical Texts’. 250–314 in H. A. Wolfson. Studies in the History ofPhilosophy and Religion. Cambridge, MA: Harvard University Press.

Zahavi, D. 1999. Self-Awareness and Alterity: A Phenomenological Investigation.Evanston: Northwestern University Press.

2011. ‘Unity of Consciousness and the Problem of Self’. 316–335 in Gallagher2011.

Ziai, H. 1990. Knowledge and Illumination: A Study of Suhrawardī’s H· ikmat al-ishrāq. Atlanta: Scholars Press.

1993. ‘Definition and the Foundations of Knowledge in IlluminationistPhilosophy: Section on Expository Propositions (al-Aqwāl al-Shārih. a), ofthe Text, al-Mashāi‘ wa al-Mut.ārah. āt’. 108–130 in B. I. Yousef (ed.). Jashnnāma-i Ostād Javād Mosleh. Los Angeles: Research and Education Center.

1996. ‘Mullā S· adrā: His Life and Works’. 635–642 in Nasr and Leaman 1996.

Bibliography 253

Index

‘Āmilī, Bahā’ al-Dīn, 161Abū Bishr Mattā, 15Alexander of Aphrodisias, 16, 20argument against reflection models of

self-awareness, 4Avicenna, 54–55, 63, 64, 69, 72–75, 87, 88, 91,

93, 99Mullā S· adrā, 170–176Suhrawardī, 108–109, 113, 120, 128

argument against the soul’s substantiality,116, 120

argument from personal identityAvicenna, 75–79, 94, 103Suhrawardī, 110–111

argument from the unity of experience, 4Avicenna, 64–71, 75, 78Mullā S· adrā, 176–181, 198, 208, 216–217

Aristotle, 12–16, 21, 23–25, 30, 73, 89, 97, 162, 182,192, 198, 231, 235–236

perception of perception, 12–14, 43soul, 23

Augustine, 6, 31, 58Averroes, 6, 8, 15–16, 20, 22, 25Avicenna

abstraction, 27, 29awareness of awareness, 22, 63, 74, 82, 89–91,

95, 97, 102dualism, 10, 23–24, 32, 34, 36–39, 42, 44, 51, 59,

61, 64, 82, 114, 115, 126and self-awareness, 51, 54–56, 58, 62, 69, 71,

82, 88, 93, 95, 101–102and individuation, 10, 25, 44–51, 47–48,

54–56, 58, 93, 99–100, 186Ishārāt, 38, 42, 42–44, 54, 58–59,

62, 63, 72, 80, 81, 82, 89, 91, 97, 101,106–108, 114, 117, 120, 126, 131, 135, 158,167–168, 168–169, 171, 203, 206, 214, 236

Mubāh.athāt, 35, 59, 60–62, 61, 64, 69, 72,75–76, 78–79, 81–82, 84–87, 86–87, 89,95, 97–98, 100–101, 102, 103, 110, 173, 208,211, 214

Najāt, 24–25, 27–29, 64–66, 65–66, 88, 186pointer and reminder, 4, 33–36, 42, 53, 62–63,

79, 85–86, 91, 114, 165Risāla al-ad

˙h. awīya fī al-ma‘ād, 79, 112

self, 35, 38, 41, 46, 52–54, 62, 68–72, 78–80, 93,95, 99, 103, 114

and presence, 88self-awareness, 24, 41–42, 47, 51–55, 57–64,

71–75, 78–80, 82–87, 89, 93, 95, 97,98–101, 102, 104, 106, 228

and individuation, 229and intellectuality, 58–61, 100, 228and presence, 88as existence, 10, 51–56as first-personality, 10, 69–71, 74–75, 80–88concept of, 31, 38, 61–64, 68, 71, 75, 78,

80–82, 85–88, 91, 92, 97, 99–103, 113,202, 208

in animals, 100phenomenon of, 30, 36, 39–41, 62–63,

65–66, 69–71, 77, 80, 85, 87–88, 165primitive, 55, 60–61, 63–64, 70–71, 73–74,

83–84, 87–91, 92, 93–97, 99–100, 102–103,158, 188

Shifā’, 23, 51–52, 54, 56, 58, 97al-‘ibāra, 92al-Burhān, 144, 236al-Ilāhīyāt, 32, 34, 43–44, 49–50, 57, 59, 86,

91, 96, 96–97, 114, 114, 123, 143, 157al-Madkhal, 48–50, 55, 91–93, 100, 126,

186–187al-Samā‘ al-t.abī‘ī, 47, 183Fī al-nafs, 14, 23–29, 31, 31–36, 33, 35, 38,

39–40, 40–41, 43–47, 44–46, 49–51, 54, 56,59, 60, 64–70, 76, 79, 83, 85–86, 88, 90,96, 97, 99–100, 110, 110–112, 112, 116, 117,125, 131, 146, 155, 165, 168, 179, 185, 188, 193,198, 216, 234

soul, 23–25, 31–33, 35–37, 39, 42–43, 45–46, 48,51, 59, 61, 64–66, 70–71, 77–79, 89, 92,96, 98, 114

254

and body, 25, 32, 34, 36, 44–48, 50–51, 61, 66,69, 114

and form, 32, 44and self-awareness, 52–53animal, 23, 32human, 23–25, 29, 32, 33–34, 34, 36, 40, 45,

47, 62, 64, 186, 211, 212incorporeality of, 44unity of, 44vegetative, 23, 32

Ta‘līqāt, 41, 51–54, 56–58, 62, 64, 71, 72, 88–89,91, 95, 95–97, 101–102, 109, 116, 125, 136,140, 157, 236

Baghdādī, Abū al-Barakāt, 10, 104, 114–117, 120,122, 144, 168, 176, 179–180, 188, 193, 206

Bahmanyār Ibn Marzubān, 75–78, 98–99, 101,110

bundle theory, 46, 49, 92, 99, 126

dhāt, 18, 22, 35–36, 38–41, 45–46, 49, 52–53, 56, 58,68, 75, 78, 80, 82, 98, 100, 102–103, 106,108–110, 112, 115, 119, 121, 127, 130, 134,136–138, 140, 149, 155, 157, 167, 169,170–171, 174, 176, 178, 182, 184, 186, 190,196, 198–199, 201, 203–204, 206, 212,213, 215–216, 221, 224–225, 228,233, 236

emanation, 24, 25–26, 90, 152, 156, 158–159,194–195, 212

epistemology, 9–10, 36, 56, 124–125, 129, 192, 226cognitive unity, 17, 59–60, 111, 125, 134, 192–193impression theory, 128, 130, 175, 193

Fārābī, Abū Nas.r, 8, 19, 162first-personality, 20, 71, 73, 75, 77–80, 85–86,

101–102, 106–107, 109, 112, 121, 156, 175,181, 202, 218–219, 226, 228–229

flying man, 4, 168, 186, 188, 213Avicenna, 6, 9, 30, 34, 36–42, 47, 53, 54, 62–63,

63–64, 67, 68, 71–78, 80–82, 85–88, 101,110, 117, 165–167, 190, 207, 235

Ibn Kammūna, 117, 166Mullā S· adrā, 117, 164–167, 171, 197, 200,

206–207, 218Suhrawardī, 87, 106–108, 111

Ghazālī, Abū H· āmid, 7, 104, 154, 205God’s knowledge

Avicenna, 56–60Mullā S· adrā, 174, 196, 215Suhrawardī, 125–127, 132–136, 139–141, 156,

158–159

H· imsī, Ibn Nā’ima, 235

Ibn ‘Arabī, 161, 210Ibn Kammūna, 37, 107, 117, 166illuminationism, 4, 10, 104–105, 110, 118, 124, 146,

148, 154, 161–162internal senses, 25, 65, 81Avicenna, 25–28, 83–84common sense/fantasy, 26, 28, 61, 206estimation, 26–29, 36, 47, 61, 83, 137–138, 205imagery/formative faculty, 26–29, 137imaginative faculty, 26–27, 36memory/retentive faculty, 26–27, 36, 80,

82–84Mullā S· adrā, 176, 198, 205Suhrawardī, 129, 137–138thought, 26, 129

Ish. āq Ibn H· unayn, 13–14, 15, 21, 73

Kirmānī, Abū al-Qāsim, 35, 62, 81–82, 84,85, 87

Kitāb al-īd˙āh. fī al-khayr al-mah. d

˙(Liber de causis),

16, 17, 235Kulaynī, Abū Ja‘far ibn Muh. ammad ibn

Ya’qūb, 163

metaphysics, 3–4, 8–10, 17, 77, 114, 118, 121,123–124, 142–143, 147–151, 154, 159, 162,163, 189, 210, 226, 228

Mīr Dāmād, 162Mullā S· adrā, 3–5, 7–9, 11, 15–16, 56, 148al-H· ikma al-‘arshīya, 163al-H· ikma al-muta‘āliya fī al-asfār al-arba‘a, 117,

161, 163–165, 167–178, 182, 185, 187,189–190, 193–194, 196–199, 201,203–204, 206, 211–213, 215–217, 221–222,223–225, 237

al-Mabda’ wa al-ma‘ād, 163al-Shawāhid al-rubūbīya fī al-manāhij

al-sulūkīya, 163al-Ta‘līqāt ‘alā Sharh. H· ikma al-ishrāq, 148cognitive unity, 16, 169, 174, 188, 193–196, 198,

201, 226dualism, 178

individuation, 185existence, 168, 175

gradation of, 181modes of, 169, 180, 182, 184–185, 187, 191, 201,

203–204, 210, 215, 217primacy of, 162–163, 226

God, 163Kitāb al-mashā‘ir, 163life, 161–162Mafātīh. al-ghayb, 162

Index 255

Mullā S· adrā (cont.)mental existence, 94, 101, 166–167, 169,

177–178, 180, 182, 184–185, 187, 188, 194,195, 201–202, 205–206

and self-awareness, 5, 56self, 160, 167, 169–170, 174, 178, 181–182, 190,

197, 201, 203–206, 215, 221, 224and agency, 180and cognitive unity, 11and identity, 216–217and mental existence, 181and opacity, 219and presence, 224and substantial change, 212–213, 216–217,

219, 221, 225–227and substantiality, 173, 176, 182, 188–190,

202, 227concept of, 213, 219, 222, 229–231gradation of, 181opacity of, 211, 222–224, 226–227

self-awareness, 166–167, 169, 171–174, 181, 197,204, 213, 226

and agency, 199–200and cognitive unity, 188, 196, 198–199, 202,

205, 207–208and existence, 182and first-personality, 175, 187and identity, 214, 216and individuation, 187–188and mental existence, 167, 182, 184–185,

187–189, 191, 202, 205, 208and presence, 173, 184and substantial change, 187, 209, 211,

213–215, 218, 221–222, 230and substantiality, 182, 189–190as existence, 182, 191concept of, 166–167, 176, 181–182, 197, 205,

207–208gradation of, 227in animals, 166, 172phenomenon of, 181primitive, 174, 177, 189–190, 198, 214,

216, 218reflective, 173

soul, 164, 168–171, 174, 177, 195–198, 200–201,206, 217, 221, 223

and agency, 200and body, 166, 178–179, 182, 184, 205, 213and cognitive unity, 225and form, 184and substantial change, 212, 217–218and substantiality, 168, 177, 179and unity, 179animal, 164–165, 171–172, 207human, 170, 205

substantial change, 182–183, 187, 193, 208, 211,213, 219, 227

and identity, 209–211, 217works, 162–163

nafs, 22, 31, 35, 52, 66, 68, 71, 75, 80, 108, 114–116,122, 127, 130, 136–137, 149, 155, 157,170–172, 177–178, 199, 201–202, 225, 234

Neoplatonism, 17–18, 30, 56, 154, 192, 195, 210

Peripatetic, 8–10, 30, 101, 104–106, 108, 111,114–115, 118, 120, 124–125, 132, 141–142,144, 147, 150, 151, 157–158, 163, 177, 185,209, 217

Plato, 231Plotinus, 6, 18–19, 31, 97, 158Porphyry, 59, 193presence (h. ud

˙ūr), 131–132, 185, 237

primacy of existence, 147primacy of quiddity, 147, 162Proclus, 16, 17psychology, 2, 4, 9, 12–13, 23–25, 30, 36–37, 38, 42,

44, 56, 59, 64–66, 71, 79, 98–99, 106, 111,115, 125, 129, 131, 133, 158, 172, 176, 179, 184

Rāzī, Fakhr al-Dīn, 4, 7, 10, 56, 88, 104, 116–118,120, 131–132, 168, 169, 176, 179, 185–189,194, 214, 220

Lubāb al-Ishārāt, 131Mabāh. ith, 56, 116–117, 131, 168, 180, 185,

185–186, 189, 220Sharh. al-Ishārāt, 21, 44, 81, 100, 117,

135, 214

self-intellection, 15–17, 19, 21, 57–58, 61, 70,91, 135

Shīrāzī, Qutb al-Dīn, 148Suhrawardī, Shihāb al-Dīn, 4–5, 7–10, 58,

104–105, 161, 163, 179, 194appearance (z· uhūr), 8, 10, 122, 124, 137, 140,

142, 147–153, 157definitioncritique of, 144–145, 153

dualism,and individuation, 126

existence, 118, 121–123, 134–135, 142–143,147–148, 156–157

H· ikma al-ishrāq, 8, 105, 108–110, 112–113, 118,121–122, 124, 136–137, 142, 144, 146–147,149–151, 154–155, 157

individuation, 126, 148, 155–156I-ness, 9, 113, 121, 122, 136, 155, 160, 175, 202knowledge, 106–107, 123–125, 127–130, 133–134,

136, 138–142, 157, 196life, 104

256 Index

light (nūr), 8, 10, 108, 142, 146–151, 154–159Mashāri’ wa al-mut.ārah.āt, 58, 105, 111, 124, 130,

134–136, 141–145, 196pointer and reminder, 109presence (h.ud

˙ūr), 10, 88, 101, 107, 123–125, 127,

130–134, 136, 138–142, 157, 196quiddity, 121, 135, 144, 146, 153self, 106, 107, 110–113, 119–121, 127–130, 133,

136–137, 140, 149, 152, 157, 159and presence, 109, 119, 130, 133, 141–142and substantiality, 119–123, 129, 149–150,

188, 191self-awareness, 106, 109, 138, 149–150, 154, 159and appearance, 149, 154, 157and incorporeality, 123and light, 149, 151, 154, 156and presence, 122and substantiality, 182, 190, 228as a challenge to Avicenna’s theory of

knowledge, 136, 138as a foundational concept, 154concept of, 109, 112, 120–123, 136

phenomenon of, 106, 119–120, 153primitive, 121, 123

soul, 106, 110, 119, 128–130, 134, 136–140and body, 127, 130, 137–138and presence, 138–140human, 141rational, 112

substance, 111, 118–119, 120, 122, 135, 142critique of, 142–144, 148, 152

Talwīh.āt, 58, 105–110, 112, 118–119, 121, 122,124–125, 127–128, 130, 132, 134, 136–138,142, 144–145, 175, 191, 206

vision of Aristotle, 125, 127–130, 132,136–137

thatness (annīya), 16, 39–40, 41, 68, 76–77,80, 106, 109–110, 115, 118, 233, 235

Themistius, 19, 20T. ūsī, Nas.īr al-Dīn, 44, 168–169, 214Sharh al-Ishārāt, 169, 214

Ust.āth, 15, 235

Index 257

mindoftheancients.com€¦ · self-awareness in islamic philosophy...

Documents