Islamic Philosophy of Religion: Analytic Perspectives 1032356227, 9781032356228

"This volume focuses on Islamic philosophy of religion with a range of contributions from analytic perspectives. It

164 69

English Pages [317] Year 2023

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Islamic Philosophy of Religion: Analytic Perspectives
 1032356227, 9781032356228

Citation preview

Islamic Philosophy of Religion

This volume focuses on Islamic philosophy of religion with a range of contributions from analytic perspectives. It opens with methodological discussions on the relationship between the history of Islamic philosophy and contemporary analytic philosophy. The book then offers a philosophical examination of some specific Islamic beliefs as well as some approaches to general beliefs that Islam shares with other religions. The chapters address a variety of topics from the existence and attributes of God through to debates on science and religion. The authors are predominantly scholars from Muslim backgrounds who tackle philosophical issues concerning Islam as their own living religion, representing internal perspectives that have never been vocal in analytic philosophy of religion so far. This is valuable reading for scholars and students of philosophy, theology, and religious studies. Mohammad Saleh Zarepour is a Lecturer in the Department of Philosophy at the University of Manchester, UK. His research interests include medieval Islamic philosophy, philosophy of religion, philosophy mathematics, and philosophy of mind. He is the author of Necessary Existence and Monotheism (CUP 2022) and a co-editor of Global Dialogues in the Philosophy of Religion (OUP 2024) and Mathematics, Logic, and their Philosophies (Springer 2021). He won a Philip Leverhulme Prize in Philosophy and Theology (2023).

Routledge Studies in Religion

Interreligious Encounters in Europe Sites, Materialities and Practices Edited by Jan Winkler, Laura Haddad, Julia Martínez-Ariño and Giulia Mezzetti Modern Debates on Prophecy and Prophethood in Islam Muhammad Iqbal & Said Nursi Mahsheed Ansari Religious Responses to the Pandemic and Crises Isolation, Survival, and #Covidchaos Edited by Sravana Borkataky-Varma, Christian A. Eberhart, and Marianne Bjelland Kartzow Gender Inequality in the Ordained Ministry of the Church of England Examining Conservative Male Clergy Responses to Women Priests and Bishops Alex Fry Tolerance and Intolerance in Religion and Beyond Challenges from the Past and in the Present Edited by Anne Sarah Matviyets, Giuseppe Veltri, and Jörg Rüpke Interreligious Dialogue Models From the Life of the Prophet Muhammad Alwani Ghazali and Muhammad Kamal Islamic Philosophy of Religion Essays from Analytic Perspectives Edited by Mohammad Saleh Zarepour For more information about this series, please visit: https://www.routledge.com/religion/series/SE0669

Islamic Philosophy of Religion Essays from Analytic Perspectives

Edited by Mohammad Saleh Zarepour

First published 2024 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2024 selection and editorial matter, Mohammad Saleh Zarepour; individual chapters, the contributors The right of Mohammad Saleh Zarepour to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 9781032321127 (hbk) ISBN: 9781032356228 (pbk) ISBN: 9781003327714 (ebk) DOI: 10.4324/9781003327714 Typeset in Sabon by codeMantra

For my sisters, Somayeh and Sajede – M. S. Z.

Contents

List of figures List of tables List of contributors Introduction

xi xiii xv 1

MOHAMMAD SALEH ZAREPOUR

PART I

Methodology5   1 What Can Contemporary Analytic Philosophy Do for the Historians of Islamic Philosophy?

7

SAJJAD RIZVI

  2 What Can the History of Islamic Philosophy Do for Contemporary Analytic Philosophers?

24

ANTHONY ROBERT BOOTH

PART II

Existence and Uniqueness of God39  3 Kalām and Cognition

41

MAHRAD ALMOTAHARI

  4 God’s Existence and the Uniqueness Claim NAZIF MUHTAROGLU

64

viii Contents PART III

Attributes of God81   5 Divine Power and Divine Knowledge

83

EDWARD OMAR MOAD

  6 The Doctrine of Divine Simplicity Considered in Light of Averroes’s Conception of Divine Knowledge

98

BAKINAZ ABDALLA

PART IV

Miracles123   7 The Metaphysics of Miracles: Avicenna on Natures and Prophetic Powers

125

EMANN ALLEBBAN

  8 Of Miracle of Miracles: An Introduction to the Qur’anic View of Miracle

147

ROUZBEH TOUSERKANI

PART V

Faith and Its Obstacles163   9 Faith, Knowledge, and Certainty: An Islamic Philosophy Perspective

165

HAMID VAHID

10 Islam and the Problem of Divine Hiddenness

181

IMRAN AIJAZ

11 An Analysis of Some Problems of Evil from an  Islamic Perspective

203

HAJJ MUHAMMAD LEGENHAUSEN

PART VI

Ethics and Morality223 12 Islamic Philosophers’ Idea of Heaven: A Moral Analysis AMIR SAEMI

225

Contents  ix 13 Towards Epistemic Justice in Islam

241

FATEMA AMIJEE

PART VII

Science and Religion259 14 Adam, Eve, and Human Evolution: Is There a Conflict?

261

SHOAIB AHMED MALIK

15 Islamic Occasionalism within the Context of Modern Religion and Science Debate

282

OZGUR KOCA

Index297

Figures

7.1 14.1 14.2 14.3

The main moves of Ghazālī’s argument in Discussion 17 Differences between HE and AE An illustration of genetic dissipation A phylogenetic tree showing the possibility of multiple universal genealogical ancestors, of which one could be Adam while still being created miraculously (same goes for Eve)

130 263 274 276

Tables

14.1 Classification of Muslim perspectives on evolution 262 14.2 Possibilities in Ashʿarism265 14.3 Differences between experimental and historical sciences (adapted but significantly modified from de Santis 2021) 270

Contributors

Bakinaz Abdalla is a Postdoctoral Research Fellow in Jewish Philosophy of Religion at the University of Birmingham. Imran Aijaz is an Associate Professor of Philosophy at the University of Michigan-Dearborn. Emann Allebban is an Assistant Professor of Philosophy at Providence College. Mahrad Almotahari is a Senior Lecturer in the School of Philosophy, Psychology, and Linguistics at the University of Edinburgh. Fatema Amijee is an Assistant Professor of Philosophy at the University of British Columbia. Anthony Robert Booth is a Professor of Ethics and Epistemology at the University of Sussex, UK. Ozgur Koca is an Assistant Professor of Islamic Studies and Philosophy at Bayan Islamic Graduate School. Hajj Muhammad Legenhausen is a Professor of Philosophy at the Imam Khomeini Education and Research Institute, Qom, Iran. Shoaib Ahmed Malik is an Assistant Professor of the Natural Sciences at Zayed University, Dubai. Edward Omar Moad is a Professor of Philosophy in the Department of Humanities at Qatar University. Nazif Muhtaroglu is an Instructor in the Philosophy Department at Mimar Sinan University, Istanbul, Turkey.

xvi Contributors Sajjad Rizvi is a Professor of Islamic Intellectual History and Director of the Centre for the Study of Islam at the University of Exeter. Rouzbeh Touserkani is a Researcher in the School of Computer Science at the Institute for Research in Fundamental Sciences (IPM) in Tehran. Amir Saemi is an Associate Professor of Philosophy at the Institute for Research in Fundamental Sciences (IPM) and an adjunct faculty at Rutgers University. Hamid Vahid is a Professor of Philosophy at the Institute for Research in Fundamental Sciences (IPM), Tehran. Mohammad Saleh Zarepour is a Lecturer in the Department of Philosophy of the University of Manchester.

Introduction Mohammad Saleh Zarepour

The philosophical issues related to Islam can be divided into two main ­categories: (1) Some issues are related to exclusively Islamic beliefs (e.g., the belief that the Quran is the miracle of the Prophet Muhammad). (2) The others are related to beliefs that Muslims share with the followers of some other traditions (e.g., the existence and unity of God and the existence of the afterlife). However, over the last 50 years, the analytic philosophy of ­religion has been led almost exclusively by scholars in the Christian ­tradition, and there has been little attention to other traditions like Islam, which is the focus of this book. As a result, neither the issues of group (1) nor the Islamic perspectives towards the issues of group (2) receive much attention from ­analytic philosophers of religion. The issues of both groups have of course been studied by Muslim thinkers since the early centuries of Islam. And the historical engagement of prominent Muslim thinkers (e.g., al-Fārābī, Ibn Sīnā, al-Ghazalī and Ibn Rushd) with these issues is still of interest to the intellectual historians in general and to the historians of Islamic ­philosophy in particular. But these engagements have rarely been addressed in the ­context of the analytic ­philosophy of religion. An important exception to this general phenomenon is the extensive and fruitful engagements of analytic ­philosophers of religion with the Kalām cosmological argument(s). But many other significant aspects of the philosophical encounters of Muslim thinkers with the issues of the two aforementioned groups have remained untouched by analytic philosophers of religion. Moreover, many exclusively Islamic beliefs seem to be worthy of detailed philosophical investigations that can in principle be conducted independently of the approaches taken by historically ­significant Muslim thinkers regarding those beliefs. They can be addressed by philosophical tools and theories that are developed in contemporary analytic philosophy, regardless of whether or not we can find any trace of such tools and theories in the history of Islamic philosophy. This is again something that has rarely been touched upon previously.1 The aim of the present book is to take a step towards filling some of these important gaps. The present collection offers analytic perspectives on various topics in the Islamic philosophy of religion. It must be emphasised that the title ‘Islamic Philosophy of Religion’ without the subtitle ‘Essays from Analytic Perspectives’ DOI: 10.4324/9781003327714-1

2  Mohammad Saleh Zarepour could be misleading. This is because ‘Islamic Philosophy’ is u ­ sually taken to refer to the history of the philosophical discussions developed by Muslim philosophers—or, even more generally, by the philosophers who lived in the Islamic world, regardless of whether or not they are themselves Muslims. On this interpretation, ‘Islamic Philosophy of Religion’ would refer to the history of the philosophical discussions about religious issues developed by those philosophers. Understood as such, the title does not encompass, for instance, a philosophical treatment of a religious subject that does not engage with the historical developments of relevant discussions in the Islamic world.2 So, ‘Analytic Perspectives’ is put in the subtitle to prevent this confusion and give the impression that this book includes analytical discussions of philosophical issues concerning Islam and Islamic beliefs regardless of whether or not they directly engage with the history of Islamic philosophy. Moreover, since this book is a work in the analytic philosophy of Islam, its domain is narrower than what is called ‘analytic Islamic philosophy’. This is because Islamic philosophy is usually understood as including all the ­philosophical thoughts and works developed in the Islamic world regardless of whether or not they are about religious issues. Thus, for example, Muslim thinkers’ discussions of the nature of mathematical objects are included in Islamic philosophy in the aforementioned sense. Nevertheless, it does not seem plausible to take them as included in the philosophy of Islam. Apart from the first two chapters—which address general methodological issues about the interaction of Islamic philosophy and analytic philosophy—all the chapters of this collection engage with issues from the two groups mentioned in the first paragraph and, therefore, lie in the more restricted domain of the philosophy of Islam or the Islamic philosophy of religion.3 Many authors of the history of Islamic philosophy in the West address Islam as ‘external observers’ and consider its historical and p ­ hilosophical significance from general, non-Muslim perspectives. An important ­ ­characteristic of the present collection is that it collects papers by analytic philosophers most of whom come from Muslim backgrounds and tackles philosophical issues concerning Islam as their own living religion. There is no doubt that studying Islam from external perspectives has its own significant and nonnegligible values. But this book represents internal perspectives that have never been vocal in the analytic philosophy of religion so far. This book is divided into seven parts. Part I includes general m ­ ethodological discussions on the relationship between the history of Islamic ­philosophy and contemporary analytic philosophy. Sajjad Rizvi explores how the h ­ istorians of Islamic philosophy can benefit from contemporary analytic p ­ hilosophy. On the other hand, Anthony Booth investigates how c­ ontemporary ­analytic philosophers can benefit from (the studies on) the history of Islamic ­ philosophy. Part II is dedicated to arguments for the existence and uniqueness of God. Mahrad Almotahari argues that the Kalām Cosmological Argument fails

Introduction  3 because its second premise—according to which everything that begins to exist has a cause for its existence—is not true. On the other hand, Nazif Muhtaroglu reconstructs and defends Saʿd al-Dīn al-Taftāzānī’s ­argument for the uniqueness of God, which can be considered a version of the a­ rgument from ‘the mutual hindrance by two omnipotent beings’. Part III focuses on the attributes of God. Edward Omar Moad discusses al-Ghazali’s and Ibn Rushd’s accounts of divine power and knowledge. Bakinaz Abdalla ­investigates Ibn Rushd’s theory of divine simplicity in light of his theory of divine knowledge. Part IV consists of two chapters on miracles. Emann Allebban investigates the metaphysics of miracles in the context of Ibn Sīnā's discussion of natures and divine power. Following a less historical approach, Ruzbeh Tusserkani develops and defends a new analysis of miracles in which the observer(s) for whom a miracle has been done (by a personified deity) plays a significant role—so that an event can be miraculous for someone and non-miraculous for another. The issue of faith and its obstacles are addressed in Part V. The link(s) between faith, knowledge and certainty are explored by Hamid Vahid. His treatment relies on an analysis of al-Fārābī’s conception of certainty. In the next two chapters of this part of the book, the Islamic approaches to the problems of divine hiddenness and evil are studied, respectively, by Imran Aijaz and Hajj Muhammad Legenhausen. Part VI concerns ethics and morality. Appealing to Isaiah Berlin’s idea that any notion of the ideal is incoherent and immoral, Amir Saemi ­criticizes Islamic philosophers’ conception of heaven. In the other chapter of this part, Fatema Amijee discusses the problem of epistemic justice in Islam by focusing on verse 2:282 of the Quran, which ‘implies that the worth of a man’s ­testimony is twice that of a woman’s testimony’. The final part of this book concentrates on the relationship between science and religion. Shoaib Ahamed Malik provides a detailed analysis of different approaches one might take to make the Islamic version of the story of Adam and Eve compatible with the scientific theory of human evolution. Engaging with a more general issue, Ozgur Koca examines the (in)compatibility of Islamic occasionalism with the modern scientific picture of the world. Admittedly, many important and fascinating topics and approaches have remained uncovered. Nevertheless, I hope that this book demonstrates that engagements with philosophical issues about Islam can be very fruitful. Notes 1 On the negligence of Islam in the contemporary philosophy of religion, see ­Zarepour (2021). 2 For example, that is how ‘Islamic Philosophy of Religion’ is understood by McGinnis and Acar (2023). 3 Islamic philosophy includes topics from but is not restricted to the p ­ hilosophy of Islam. For a book on analytic Islamic philosophy, in a sense more ­general

4  Mohammad Saleh Zarepour than merely the philosophy of Islam, see Booth (2017). For books which lie in the ­narrower domain of the philosophy of Islam, see Aijaz (2018) and Zarepour (2022).

References Aijaz, Imran, 2018, Islam: A Contemporary Philosophical Investigation, New York: Routledge. Booth, Anthony Robert, 2017, Analytic Philosophy of Islam, London: Palgrave MacMillan. McGinnis, Jon, and Rahim Acar, 2023, “Arabic and Islamic Philosophy of ­Religion”, Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/ arabic-islamic-religion/ Zarepour, Mohammad Saleh, 2021, “Islamic Problems and Perspectives in ­Philosophy of Religion”, Religious Studies Archives 5, 1–9. Zarepour, Mohammad Saleh, 2022, Necessary Existence and Monotheism, ­Cambridge: Cambridge University Press.

PART I

Methodology

1 What Can Contemporary Analytic Philosophy Do for the Historians of Islamic Philosophy? Sajjad Rizvi Is philosophy primarily characterised by critique and analysis? Or perhaps more ‘traditionally’ a systematic study of metaphysical reality? Or even in our decolonial times, a process of dialogue and conversation through ­competing, contradicting, and overlapping conceptual schemes producing newer and more universal forms of understanding? The exciting nature of this part of this book is an act of translation, and the desire for openness between two traditions of philosophy that attempt to bring conceptions of ­philosophy into conversation without imposing the one over the other. To ask what ­contemporary analytic philosophy can do for the historian of Islamic ­philosophy requires an understanding of the terms – while r­ ecognising that both analytic and Islamic capture multivocal traditions. The c­ ommonality of philosophy is a given, even if we can dispute whether philosophical p ­ roblems, terms, conceptual frameworks, and the very sense of ‘philosophy’ are truly perennial and universal. How can we hold onto the meaning of these labels while acknowledging their elasticity? If we reduce the analytic to anglophone and Islamic to arabophone, we have potentially the problem of linguistic and conceptual incommensurability (not least through the famous S­ apir-Whorff thesis) as well as clashing ‘ontological and linguistic nationalisms’ (Diagne 2022: 122–133). If the decolonial turn in the study of the humanities is of any value, it will set aside two types of extremes: on the one hand, the ­insistence that the Anglo-Austrian tradition of analytic philosophy can only be expressed in English and the universal language of logic and anything that cannot be so captured and transformed may be a cute cultural a­ rtefact but is not philosophy (it may constitute ‘wisdom’ or even ‘­paraphilosophy’ ­neither of which possess power as departments in the hierarchy of the humanities in metropolitan ­universities) and, on the other hand, a philological purity and authenticity that would insist upon ­reading Islamic philosophy as unique, almost a revealed language of metaphysics mediated by Greek and Arabic. From the incommensurability of ­linguistic and ontological ­categories, we can move to the opening of language as translation, of conversation based on equals devoid of the dominance of the ­ coloniser over the ­ colonised (Diagne 2022: 17–19). This process of rejecting the monolingualism of DOI: 10.4324/9781003327714-3

8  Sajjad Rizvi coloniality might lead to a mutually respectful, decentred, and rational ­linguistic ­multiplicity of ­philosophy. In that humanistic spirit of mutuality, both the analytic and Islamic traditions may be willing to open that process, the former through respect and the principle of ­charity and hospitality, and the latter from a ­position of powerlessness seeking a voice (Ganeri 2016; Baggini 2018; Gordon 2019). Historians of Islamic ­philosophy are already habituated to analyse, evaluate, critique, and reformulate ­argument in order that their peers interested in the history of philosophies might be able to comprehend them. But of what value is that qualifier ‘analytic’? Linked to that is a double-sided concern: on the one hand, analytic philosophers are not ­ known for their ­attention to historical issues or a historical understanding of ­questions, and it is difficult to define contemporary analytic philosophy in a ­reasonable ­manner, and on the other hand, most historians of philosophy in an ­anglophone ­context do draw upon the analytic as ‘method’. While the ­earlier history of the analytic tradition was clearer in that it defined the ­tradition as a method predicated on the precise use of logic and the linguistic turn (one thinks of the famous proposition that understanding of thought can and must be produced through a careful account of language) and one might even argue being clear in its rejection of the ‘abstract’, imprecise, and ‘poetic’ approach of continental philosophy, it does indeed seem that analytic philosophy is devoid of a fixed doctrine, method, and definition. Perhaps it is more an attitude characterised by a focus on the precise articulation of language, a disavowal of historicism, a marginalisation of poesis and florid language, a focus on analysis and explanation, on coherence and clarity, and the careful attention to definitions and concepts from the most primitive to the complex. This chapter will begin with an important preliminary discussion on what we mean by the history of philosophy and how that is practiced by those doing Islamic philosophy as well as analytic philosophers, while b ­ uilding to some normative considerations. I will then develop this by ­distinguishing the ­analytic as a historical and culturally specific tradition from an ‘­analytic’ ­attitude or method that is marked by a desire to cut through poesis and obscure language to find the coherence and syllogistic form of ­reasoning embedded, setting aside historicism (and even conventionalism) by e­ xamining texts as part of wider conversations on philosophical problems and fi ­ nding precisely what the philosophical significance of the argument is, and not being ­committed to a particular doctrinal position but rather being open to ­revision and reconsideration of a position in terms of the merit of the ­argument. Analysis and inquiry in analytic philosophy are entirely consistent with the notions of baḥth (discourse) and taḥqīq (verification, ­reconsideration) in the Islamic philosophical traditions. To illustrate this attitude, I will examine the old problem of the semantics of existence in Islamic philosophies and proffer a language for explaining it. But finally, I will also consider whether an analytic

Contemporary Analytic Philosophy for Historians  9 method is enough either to capture the whole scope of what we might mean by philosophy in the Islamic contexts or whether it is ­commensurable as a tradition with other forms of philosophising in a decolonial turn in which I want to locate the history of Islamic philosophical traditions within the quest for a properly global history of philosophy. 1.1 History of Philosophy A common misnomer is that the analytic tradition does not engage with ­history or with philosophical texts in the past; rather, it is commonly alleged that a crucial feature of the continental tradition is that the major thinkers in that tendency engage in exegesis and historical analysis to make sense of ­philosophical problems and discourse in the present. Analytic philosophers do engage with history and are interested in it for differing purposes and ends. There have been thinkers who have rejected history – one thinks of quips ­projected upon canonical figures like Quine, Harman, and even Searle – because they have insisted that philosophy is about argument posited in the present; in fact, one could say that the self-image of analytic ­philosophy is entrenched in bias against engaging in history (Rieck 2013: 1–2). Or, as others have put it, ‘extreme forms of the analytical history of philosophy do now belong to the demonology of the subject’ (Williams 2006: 259). The rejection of ­history is about ‘liberation from the dead weight of past errors and ­illusions’ (Taylor 1984: 17). In fact, using history suggests to some a reduction to the argument from authority – and the primary posture of the analytic tradition is the notion that philosophy is being done ‘anew’ (which is rather different to the ‘humble’ eschewal of ‘originality’ by many Islamic ­philosophers in the past) (Sorrell 2005: 5–6). Allied to this is the notion that modern logic has changed the field and hence made prior forms of ­argumentation redundant; whether this assumes the broad line of progress is a different matter – and certainly, decoloniality is rather distrustful of the claims to progress (Gracia 1992: 315). A common mode of distinguishing the continental from the analytic ­traditions of philosophy is to claim that the former is exegetical and ­historical in essence as a mode of doing philosophy while the latter is problem and ­argument based and does not necessarily consider history and context s­ eriously (Sorrell 2005: 1–2; Montefiore 2014: 35). Concomitant with that classic ­stereotype is the idea the while both traditions do read ­philosophical texts in the past, they differ on how they read the texts and how much of them they read – there is always a sense that analytic philosophers do not read h ­ istorical texts carefully, extensively, or deeply (Montefiore 2014: 31–32). There are many exceptions. Brandom – if consider pragmatism as analytic, certainly he himself defines what he does as ‘philosophical r­ationalism’, ­reasoning, and conceptual use (Brandom 2009: 1, 8) – extensively and s­ erious engages with Kant and Hegel among others in his work, and Dummett – a quintessential

10  Sajjad Rizvi English analytic philosopher – does the same with Frege (Dummett 1993; Brandom 2009, 2019). There are important pedagogical reasons for e­ ngaging in history – and pretty much every modern philosopher takes Kant as their starting point (Rieck 2013: 4–5). So how does an analytic philosopher think one ought to do the history of philosophy? In the first instance, we have the idea that history is a way of doing philosophy (Rieck 2013: 22). But do most philosophers reflect on their method of doing history (Frede 2022: 3)? There is a difference of opinion on whether history for ‘its own sake’ is a valid enterprise. While ­acknowledging that the term history itself might be ambiguous, what we mean here is the process by which historians produce an account and as such denotes an activity and indeed a discipline with an emphasis on explanation (Gracia 1992: 42–44). A common schema posits broadly three ways of doing the history of philosophy (Frede 2022: x–xi). The first is intellectual history (Geistesgeschichte), or what one might call philosophical doxography. In this approach, one engages in conceptual translation, rearranges and reorganises texts, and traces both the sources and influences of ideas – and as such it focuses on meaning expressed in the idea (Gracia 1992: 235, 241–242). One possible drawback could be that the concern is antiquarian predicated on the dissimilarity and even the ineffability of the past since no one can truly be disinterested and knowing (Gracia 1992: 66–72). The emphasis on the past might also bring with it a polemic against the present and an ideological stance (Gracia 1992: 253–274). The second is the historical history of philosophy or what one might call rational reconstruction with an emphasis on description and interpretation (Rieck 2013: 6–7, 12). For Frede and Williams, this is the best way of doing things; it makes the strange familiar, seeks to remove misunderstandings, and avoids prolepsis and anachronism (Frede 2022: 124–129; Gracia 1992: 81; Williams 2006: 257, 260–261). It is about approaching what is rational in its context; to use Skinner’s terms, rational beliefs are what are suitable to the agents whom we study in the circumstances in which they find ­themselves, not on our terms, and which are obtained through some established and accredited process of reasoning (Skinner 2002: 27–29, 31). False beliefs therefore constitute the failure of their reasoning not what we consider to be false; we need to take the texts seriously on their terms and avoid prolepsis (Skinner 2002: 30, 72). Still, a historical study must be willing in the quest for intelligibility to ascribe truth values to arguments (Gracia 1992: 85). The third is the philosophical history of philosophy, where the e­ mphasis is on the evaluation and relevance for us now in the present and future (Gracia 1992: 39). Philosophy involves being willing to say that the c­ urrent mind knows better and can consider some past arguments and positions are ­irrelevant and even wrong (Gracia 1992: 108). It involves understanding justifications and explanations not just on their own terms and beyond the confines of historicism (Gracia 1992: 158). One possible way of doing this is what Gracia calls the framework approach – an evaluative and philosophical

Contemporary Analytic Philosophy for Historians  11 approach that takes asks much: one ought to analyse definitions of the main concepts, present precise formulations of the issues at stake, explain the ­solutions to those issues, present arguments and objections to those ­solutions and articulate criteria for the evaluation of arguments and solutions (Gracia 1992: 279–282). This is very much a manifesto for doing history of ­philosophy for those trained in analytic philosophy. Insofar as the goal of philosophy is to increase awareness of the ­complexity of problems, such analysis may also consider how askesis entails therapeutic investigation of concepts and their classification (Gracia 1992: 319). How have historians of Islamic philosophy approached philosophical texts and ideas in terms of the analytic scheme set out? Let us take the case of ­studies on Avicenna as an example. First, we have those primarily interested in a ­contextual, doxographical, and philological approach. They consider the study of philosophy in Islam to fall within the rubric of the history of science and ideas and seek an honest but textually careful description, p ­ hilologically authentic, and interpretative reconstruction because we are not yet at the stage of doing the evaluation. This often involves locating Avicenna within the Aristotelian tradition or more broadly within the reception of that ­tradition in medieval thought exemplified in the work of Dimitri Gutas and his ­students (Bertolacci 2006; Gutas 2014; Hasse 2000). This often comes with a strong emphasis on empirical and analytic aspects of Avicenna, ­rejecting the mysticising, poetical, and rhetorical aspects as ‘esoteric’ and ‘oriental’ and proposing a ‘sober’ approach (Gutas 2014: xix−xxvi). A ­parallel approach to this analyticising one is to emphasise the location of Avicenna within the ‘commentators on Aristotle’ tradition (Wisnovsky 2003). That critique is usually directed to the approach of Henry Corbin whose work, stemming from the phenomenological tradition in which he was steeped and emphasised the poetical, the visionary, and the non-discursive aspects of Avicenna’s thought (Corbin 1960). In that approach, he was ­followed by Seyyed Hossein Nasr whose own spiritual commitments meant that he wished to read Avicenna as a spiritual master and magus and as a reflection of a ‘prophetically revealed philosophy’ (Nasr 1964). A critique of this tendency but one influenced by a continental approach is Nader el-Bizri’s approach to the question of being that juxtaposes Avicenna with Heidegger. His concern is with a global history of philosophy that grapples with the question of being and its oblivion and takes a holistic approach to Avicenna that comprises the discursive as well as the mystical, linguistic, and rhetorical, thus bringing Avicenna into conversation with phenomenology (el-Bizri 2000: xiii−xvii). The third approach involves analyticising Avicenna and bringing him into a direct conversation with contemporary analytic and post-­analytic ­philosophy in the anglophone context. Lenn Goodman is somewhat d ­ ifferent; his work is a contextual one but also willing to defend aspects of Avicenna’s thought such as reading the theory of the active intellect divorced from its ­cosmology as an account that posits a transcendental foundation for knowledge that accords

12  Sajjad Rizvi with some elements of modern European a­ nalyticising thought (Goodman 1992). Perhaps the best examples here are very distinct works with differing readings of the analytic tradition. Nusseibeh is interested in Avicenna as a philosopher of language and logic who uses these tools for understanding the world and describing it albeit with limits, and setting aside Aristotelian essentialism (Nusseibeh 2018: 309). This in fact is the reason, according to Nusseibeh, for the two main positions of Avicenna for the l­ atter tradition: the priority of existence over essence in objects and the concept of mental existence (Nusseibeh 2018: 312). While sympathetic to Gutas’ ­holistic ­reading of Avicenna, he reads ‘oriental philosophy’ as ­ something distinct from Aristotelianism but already present in al-Shifāʾ – not for him Gutas’ understanding of Avicenna’s ‘philosophical project’ (Nusseibeh 2018: ix). The work is aimed at the analytic philosopher, hence the comparisons of Avicennian essentialism with Quine, truth validity with Williamson, and ­formalising Avicennian logic in terms of quantifiers, operators, and ­modalities. He is not afraid to evaluate Avicenna critically either (Nusseibeh 2018: 275–281). An element of his analytic approach lies in his emphasis on possibilities and that true ‘oriental philosophy’ lies in being liberated from the need to intuit and ‘guess’ the middle term of the syllogism – epistemic perfection is not possible in this world (Nusseibeh 2018: 363). Philosophy thus became for Avicenna analytic, naturalised, and limited (Nusseibeh 2018: 366). Azadpur pins his own commitments within a more pragmatic form of the analytic ­tradition with Sellars and MacDowell whom he wishes to put in a dialogue with Avicenna, because the common language of ­philosophy is analytic (Azadpur 2020: 1). He attempts to overcome the traditional ­historical approaches to Avicenna that pit emanationism against (Lockean) empiricism in favour of a non-mystical, non-inferential knowledge of matters of fact drawing on Sellars (Azadpur 2020: 2). His concern is with problems in the use of language and the understanding of reality, and while he does not engage in ahistorical or decontextualised readings of Avicenna, he is not restricted by the context or the philology. In many ways, his work is along the lines of ‘what can analytic philosophers learn from philosophers in the past’. 1.2

Analytic Tradition

One important distinction to draw is between analytic philosophy as a ­tradition that emerges at a historically contingent time and context and between what I would call an analytic approach or method that has been universalised beyond that Eurocentric cultural context. It might be worth considering how the two are often conflated. In the first instance, the tradition is somewhat illusory. What really brings the disparate and distinct traditions together is defined negatively; they are not ‘continental’. The latter is also defined negatively as what is not analytic or Anglo-American with the assumption of a growing gulf that is perhaps unbridgeable despite both emerging from European philosophies at the turn

Contemporary Analytic Philosophy for Historians  13 of the 20th century – not least because the likes of Frege and Husserl could arguably be claimed by both. Both analytic and continental are problematic labels for traditions as they are each very disparate and imprecise (Montefiore 2014: 27). To consider analytic and continental to be divided, schismatic, and incommensurable traditions in need of bridging is quite absurd (Dummett 1993: xi). For others the roots of the analytic tradition are in fact the same as those for continental phenomenology, namely German philosophy with the likes of Frege, Meinong, Bolzano, and Brentano (Beaney 2013: 7; Dummett 1993: ix–x). The historical origins of the analytic seem to lie in Austria and, in ­particular, through the responses to different thinkers to a key question. Is there a distinction between minds and mental processes and extra-mental objects and events – the classic problem of dualism raised by the Aristotelian and Cartesian traditions? But even here there was no agreement on positions among Austrian thinkers and those that later influenced Britain and beyond on major issues such as the nature of language, the role of logic, the ­existence of souls and selves, the possibility of metaphysics, and so forth (Textor 2021). Traditional metaphysics as the study of the ‘truly real’ in Platonic terms was set aside – but as we note later metaphysics did make a comeback. The ­analytic tradition as philosophy done in English still has a complicated and non-linear history and therefore what one can draw upon will depend very much on the thinker being engaged. 1.3

Analytic Method

If not the tradition, then what about the method? A common account would suggest that the analytic embraces precision and clarity of l­anguage and ­expression, the use of logic as rigorous implementation, and the notion of ­ argumentation predicated on these, all of which constitute analysis (Beaney 2013: 25). Consequently, the areas of concern seem to be language, ­epistemology (especially the problem of perception), and modality (which somewhat opens the door to metaphysics) (Beaney 2013: 18–20). Earlier in the tradition, analysis sought to be scientific with its promise of certainty and actual knowledge but that seems to have been discarded (Rieck 2013: 123– 134). Other notions included the idea that philosophy engaged in the ­freedom of thought and the eschewal of prior authority (Brandom 2009: 130). This was related to one central feature of post-Kantian philosophy namely critique, which is sceptical of the claims to metaphysics and indeed claims to certain knowledge because it rejects the notion that the mind is a proper ­mirror to reality (Gracia 1992: 9–10). Philosophical analysis cannot ­therefore focus on reality but on language (Gracia 1992: 21–25). This famously is the linguistic turn that analyses language, rejects metaphysics, postulates that philosophy is not identical to science, analysis, science oriented, argument, and clarity, and is grounded in formal logic. Dummett famously reduced this to two propositions ‘first, that a philosophical account of thought can

14  Sajjad Rizvi be attained through a philosophical account of language, and secondly, that a comprehensive account can only be so attained’ (Dummett 1993: 4). Thus, the focus on language is about method that contrary to the earlier ­tradition insists that philosophy is not science – in fact, it concerns the c­ areful considerations of propositional truth and the limits of human ­ ­ language (Hacker 2013: 932, 941). In that sense, we can see a development from Frege to the Vienna circle through to Wittgenstein, logical positivists, and down to Oxford ordinary language philosophers (Beaney 2013: 23; Hacker 2013: 932). But again, not all agree on these or even on what is understood by the linguistic turn (Preston 2007: 99–123). Ultimately, we are left with the idea that the analytic is a style of doing ­philosophy, of doing ‘good’ and sound philosophy based on the ­principles discussed above (Preston 2007: 9). Most Islamic philosophers of the ­ past would have no objection to these concepts and principles. But the ­fundamental question is whether their assumptions about the purpose and end of p ­ hilosophy and their ethical and theological commitments ought to be marked or bracketed when applying the analytic method to their thought. 1.4

One Example: Existence

How might we apply an analytic method to the study of the history of Islamic philosophy? Let us consider a few examples of the issue of existence. Perhaps the main issue is a distinction between the conception of philosophy and its relation to ethical and theological commitments as well as the scope of what philosophy might discuss (with, for example, a heavy emphasis on ­metaphysics) and the actual method of analysis used, in particular, problems and issues. On the former, the broad divergence is clear, while for the latter, there is a clear compatibility. The foundational figure in many ways for Islamic philosophy is Avicenna (d. 1037) and his modification and development of Aristotle from whom he took the general conception of metaphysics as the first philosophy and the classification of the sciences, both theoretical and practical, and the idea of logic as an organon for understanding sound reasoning. While he wrote on Qurʾanic exegesis, he integrated theological issues within philosophy such as the nature of the soul, the existence of God and her attributes and agency, and the nature of prophecy, miracles, and the afterlife. This had a profound impact on the later tradition which considered the role of p ­ hilosophy as ­critical to explaining theological issues. Metaphysics remains the first philosophy for Avicenna, and it studies the ‘existent as such’ (Avicenna ­ 2005: 9–10). On existence, his position was taken from Aristotelian pros hen homonymy of the term ‘exist’ but taken in a direction of radical contingency. The primary division of existence – which as a term was incapable of being defined or demonstrated – was between the necessary, namely God, and contingent, everything apart from God. God alone is pure existence, simplex,

Contemporary Analytic Philosophy for Historians  15 immutable, timeless, unknowable as such and infinitely benevolent and wise; he bestowed existence on contingents, who are conceptual dyads of ­existence and essence, knowable, mutable, and capable of striving for ­ perfection (Avicenna 1973: 47−75, 2005: 29−44, 257−298). This was his ‘necessary existent theology’ (Zarepour 2022). The problem arises because Avicenna, like Aristotle before him, uses existence (which, in itself, is a ­linguistically awkward, abstract notion in Arabic) ambiguously: it refers to the copula, it refers to substances and the categories, it refers to being qua being, but also to concrete ­existence, and to existence in the mind (Avicenna 1973: 157−201). Even though ­existence is self-evident, the necessary existent qua God needs to be demonstrated in metaphysics. Similarly, the fundamental question is whether God exists and Zayd exists are the same sort of proposition and what they exactly mean. In the early modern period, there were two dominant ‘readings’ of Avicenna. This first is Mīr Dāmād (d. 1631), the leading Avicennian ­philosopher and jurist of the Safavid period. His conception of ­philosophy is based on the ­ harmonisation of religious experience with an analytic approach to the nature of reality that he described as the Yemeni ­philosophy. By this he meant ­something distinct from the Greek philosophical tradition that dominated most philosophers in the Islamic period, and a philosophical method ­predicated on wisdom that was revealed; philosophical and r­ eligious ­metaphysics and ways of life are congruent (Mīr Dāmād 2002: 2–3). The revealed aspect of his philosophical approach indicates the embrace of the non-propositional and even mystical experience. In doing so, he d ­ istanced himself from the ‘Greek’ approach of Avicenna (Mīr Dāmād 2012: 170– 171). His conception of the scope of metaphysics and the nature of ­existence extends Avicenna’s necessary existent theology and divine simplicity to embrace two distinct albeit related positions. The first is that existence as a reality refers to God and only God (Mīr Dāmād 1977: 17, 24, 49–50). Such a position is consistent with much Christian philosophy that is contemporary to him and with perfect being theology (de Ray 2022). At the same time, for contingents, existence is a quantifier and a conceptual ‘accident’ of essences that obtain, and therefore as such, the ‘value of a variable’ to use Boolos’ language (Boolos 1984; Mīr Dāmād 1977: 39–43). Such an approach to the meaning of the term ‘exist’ is probably the dominant position in Islamic ­philosophy before the 17th ­century and even after. The second reading of Avicenna was provided by Mīr Dāmād’s student, Mullā Ṣadrā (d. 1636), who reversed the position on existence. He extended his teacher approach to philosophy as a way of life that embraces revelation and mystical experience. His definition of philosophy and metaphysics in particular proposes a dual method: of analysis and demonstration on one hand, and on mystical insight and intuition on the other (Shīrāzī 2005, I: 21). The goal of philosophy is to understand reality as it is guided by ­reading the c­ osmos as well as other divine books including scriptures. The

16  Sajjad Rizvi end of p ­ hilosophy is godlikeness which for Mullā Ṣadrā lies in p ­ hilosophical askesis as well as spiritual practices peculiar to the Islamic traditions. On existence, he tries to reconcile Avicennian metaphysical pluralism with his teacher’s ­ monism. Existence is univocal but modulated in that singular meaning between the necessary and the contingent. Applying existence to the ­contingent is neither empty nor merely indexical, and he rejects ­existence as quantification. His approach continues the Avicennian decentring of Aristotelian category theory and hylomorphism. Two further examples. The first is the modern seminarian Iranian ­philosopher Mihdī Ḥāʾirī Yazdī (1923–1999), one of the first to study ­analytic philosophy and obtain a doctorate from a major North American ­philosophy department, namely the University of Toronto, and famous for his Kantian t­ rilogy on metaphysics, pure, and practical reason. In the preface to his textbook on metaphysics, or universal science, written for his students at Tehran University in the 1950s, he criticises two types of people: first, those post-Kantian philosophers who think that metaphysics is an impossible and futile pursuit and, second, ‘orientalist’ philologists who study Avicenna and others but do not engage with the thought or the way in which he is ­studied in the seminary (Ḥāʾirī Yazdī 2017: 49–52). For Ḥāʾirī, any denial of ­metaphysics is a denial of two realities: the self and the extra-mental reality that it seeks to understand. Furthermore, he takes a traditional approach to analysis and reflection at whose heart lies the pursuit of concepts; the practice of ­metaphysics is the means of ‘becoming human’ and perfecting the self, a way of life to use the paradigm of another modern philosopher (Ḥāʾirī Yazdī 2017: 55). On existence, he puts forward the position of the school of Mullā Ṣadrā: existence is univocal but modulated in intensity; it is ­conceptually distinct from essence; it can neither be defined nor pointed out since there is no term more common or more extensive than it; it applies both to mental existence which is the realm of thought and extra-mental existence which is affirmed against any sceptical doubt; existence is a causal link in reality and also the critical link between subject and predicate that allows for the s­ oundness of propositions (Ḥāʾirī Yazdī 2017: 63–85). Importantly, he includes arguments for his positions most of which focus on the analysis of existence. The second modern example is Sayyid Muḥammad Ḥusayn Ṭabāṭabāʾī (1903–1981), the leading seminarian philosopher and exegete in Iran. In the seminary textbook that he wrote in the 1970s, he merely begins with an ­assertion about the definition and necessity of doing philosophy because if a person reflects, they discern that they are a reality and that there is also a ­reality beyond that them that they seek to inquire (Ṭabāṭabāʾī 2018: 1). He does not polemicise here against detractors of metaphysics or the possibility of inquiry since that is central to his earlier project on realism and the ­refutation of materialism and naturalism that arose from his circle in the 1950s and 1960s. In the main text, he discusses the same positions on existence that Ḥāʾirī did; however, he extends arguments and considers a number

Contemporary Analytic Philosophy for Historians  17 of counter-arguments and objections to his analysis of existence (Ṭabāṭabāʾī 2018: 7–43). He also considers problems of predication including that of conceptuals and non-existent things, as well as an attempt to define the realm of the fact (lit. the thing in itself) that is critical to the practice of metaphysics, a proper evaluation of perception, and the truth validity of propositions. If we approach the question of existence in these thinkers purely from the style and concerns of analytic philosophy, what might we miss? For one, we might miss the broader context of their philosophical concerns: the theological, the mystical, the juristic, and the exegetical. In most cases, these figures are all more than just philosophers. Mullā Ṣadrā’s rather intuitive sense of existence – which is not absent even in Avicenna – cannot simply be captured by linguistic, propositional, or even perceptional analysis. But at the same time, there is enough material in their work that can be presented in analytic language to engage philosophers of that tradition. 1.5

Objections: Decoloniality, Metaphysics, and Poesis

Is analytic philosophy capable of embracing the global and the d ­ ecolonial? The new challenge is how do we effect a new pluralistic philosophical ­practice oriented to metaphysics, ethics, epistemology, and so forth that is also ­recognisably analytic without being Eurocentrically so. One of the best cases for a properly global and cosmopolitan philosophy is made by the ­ eminent philosopher and specialist of Indian philosophy, Jonardon Ganeri who argues that we are living in a new age of the re-emergence of ­competing reasons based on four observations (Ganeri 2016). First, c­ ultural forms of reason ­ predicated on lived experience and self-reflection are ­finding ­authentic voices. It is very much about (re)discovery of philosophical ideas and a­ llowing the space to engage on their own terms and present their own ‘ways or forms of life’. Second, academic philosophy – much like ­metropolitan ­academia ­accepting the post-colonial and multicultural critique – is ­becoming more polycentric and embracing plural and diverse forms of intellectual ­production. Third, colonial powers and epistemologies are being provincialised; Westerners are no longer accepted as privileged interlocutors in a global dialogue. Fourth, anglophone philosophy is becoming more self-reflective about its obligations to consider an international context and cognisant that its role is less a result of intellectual superiority and more a legacy of colonialism and empire; hence, it needs to facilitate space and open participation for all. What is thus needed is a new map of philosophy and a form of pluralist realisms in the academic and even popular philosophical circles. This does not entail that global philosophy should be ­motivated by white guilt and a compensatory inclusion, nor should it be a vehicle for the dissemination of essentialised ethnic chauvinism. Nor is the call for global philosophy a manifesto for (cultural and epistemological) relativism. Rather, non-European traditions of philosophy, including Islamic ones, have something to offer by way of argument, thought experiment, ontology,

18  Sajjad Rizvi hermeneutics, narrative, mimesis, and even poetics. Recognising the extent of the coloniality of philosophy is critical to affecting an emancipatory and reparative decolonial turn against the ethnocentrism, commodification, and solipsistic complacence of ­metropolitan academic (Anglophone) philosophy (Gordon 2019). One can no longer accept the idea of Europe as a ‘biotype’ of philosophy predicated on the exclusivity of its liberty, secularity, and analytic and reflective rationality (König-Pralong 2019: 85–87). By decolonial, I mean much more than the narrowly political and ­historical process of decolonisation; rather it concerns ontology, epistemology, and praxis – how we are and how we think and act authentically, with a clear critical notion of history itself, without falling back on essentialist notions of cultural traditions and ways of being, or even reverting to nativism. It can p ­ resent ‘local’ d ­ efinitions of the human (Rivera 2021; Wynter 2003). Coloniality ‘used in recent scholarship indicates the epistemic dominance of western and Eurocentric modernity, including its suppression of ­methodological ­critiques and alternate modes of thought’ (Yountae and Craig 2021: 3). It is a regime of power that survives formal historical and colonial governance and ­continues to reproduce Eurocentrism, an idea of Europe, and privileged notions of race (Yountae and Craig 2021: 4). Decoloniality, by contrast, is a non-linear movement towards possibilities of modes of being, thinking, knowing, sensing, and living in the plural that are other than those complicit (without necessarily entirely overcoming) coloniality (Mignolo and Walsh 2018: 81). As such, it proposes forms of relationality and interdependence that contest the totalising, epistemic claims and power of modernity and its abstract universals that set up hierarchical binaries: mind and matter, science and magic, reason and superstition, orality and literacy, male and female, and modernity and tradition among other major concepts. Significantly, it is a process and an endeavour – much like philosophy in this conception. Such a conceptual approach is not without its critics. Decolonisation remains a Eurocentric and hierarchical pursuit of the Western academy. As a form of identity politics, it suffers from ‘elite capture’, furthering the interests of metropolitan elites, and fails to live up to Marx’s call for philosophy to change the world and not merely interpret it (in the eleventh of his ‘Theses on Feuerbach’) (Táíwò 2022a). From another perspective within Africana thought, decolonialisation denies history and agency and fails to deal with incommensurabilities of language (Táíwò 2022b). These are reasonable and one cannot fail to notice that there is a ‘decolonial bandwagon’ (Moosavi 2020). However, there are different modes of decoloniality that do not need to valorise native knowledge systems in an uncritical manner, nor do they need to ignore the question of agency and choice among intellectuals in the global south. A historical approach to the study of those systems will reveal internal critiques and contestations as well as the emergence of a long ­tradition alongside minor ones. Furthermore, decoloniality as mentioned before entails a praxis as well as an ontology and epistemology. How we do philosophy should change us and the world around us.

Contemporary Analytic Philosophy for Historians  19 Global philosophy dislodges us from our parochial and highly s­ pecialised, perhaps trivial, concerns and reorients us towards the notion that the true task of philosophy is to consider, debate, and keep alive the major, ­fundamental, urgent problems that we face as embodied and self-reflective beings, to ­examine attempts to make sense and even solve them, and then to evaluate the success of those possible solutions. The study of the humanities in the world of Islam is precisely concerned with what it means to be human and that cuts across different modes of expression – poetry, prose, plastic arts, music, and ­ hilosophical ­theology calligraphy – as well as different genres from Sufism to p to law and legal theory. Sometimes the most philosophically p ­rofound work that changes our assumptions about knowing, being, and doing arise from compositions that do not self-identity as philosophy. Furthermore, an approach to philosophy as a way of life, f­ollowing Pierre Hadot, may well be a better comparative for some traditions (Ogunnaike 2017). Philosophy in pursuit of wisdom needs to embrace the affective and the broad sense of our cognitive, embodied lives that cannot be ratiocinatively reductionist. It needs to be more than what can be inscribed in Greek, or German, or Arabic. Ganeri suggests that the analytic tradition and method can embrace these new realities and continue to reinvent itself – and this is a traditional area of weakness since it is often assumed that the ­continental tradition is far better at dealing with eurocentrism and other forms of ethnocentrism in philosophy (Derrida 2016: 3–5, 345–368). Beyond the larger question of decoloniality, two further questions arise. Does analytic philosophy do metaphysics? Analytic metaphysics is no longer an oxymoron (Beaney 2013: 5), but a vibrant field, whether in terms of Strawson’s descriptive versus speculative or revisionary m ­ etaphysics, Wittgenstein’s descriptive metaphysics, Lewis’ famous possible worlds ­ontology, or Yablo’s ontology (Lewis 1986; Rieck 2013: 14–15; Simons 2013: 717–720; Strawson 1959; Yablo 2010). In fact, these philosophers suggest that the famous bias against systematisation is similarly misplaced (Preston 2007: ix; Simons 2013: 226–227). Finally, can analytic philosophy deal with poetics and non-discursive argument and knowledge? Arguably, continental philosophy deals with ­ ineffability, unsayable, poetical, and rhetorical much better (Montefiore ­ 2014: 36). Can a speech acts approach, following Searle and Grice, or a ­language games and forms of life approach, following Wittgenstein, deal with such language (Skinner 2002: 2–3)? Once again if we set aside our biases for what analytic philosophy essentially is, the potential for all manner of ­consideration of linguistic analysis and perception beyond the narrowly logical opens up. 1.6 Conclusion Does the study of the history of Islamic philosophy need analyticising? For me, the answer is somewhat equivocal. Given the professional near

20  Sajjad Rizvi monopoly that analytic philosophy has, the historian of Islamic philosophy cannot afford to forgo it (Preston 2007: 16). There are Muslim thinkers who engage in ­different ways as we have seen. Interesting are the ‘Islamic ­analytic ­theologians’ who drawing on that movement within British theology of those trained in the analytic tradition who study historical figures and ­arguments such as Fakhr al-Dīn al-Rāzī and Ibn Taymiyya reading them through Wittgenstein and reformed epistemology (Abdelnour 2023; Ahsan 2021; Turner 2021, 2023). But for some of them, it is a procedural method and not a substantive use of the analytic. Others use analytic methods in order to affect an engaged conversation between the two traditions albeit with their primary interest in speaking to the analytic (Azadpur 2020; Nusseibeh 2018). Others still – ­perhaps the most successful – are those who consider the analytic to be the most effect language into which one can translate the a­ rguments and conceptions of Islamic philosophers in the past in order to open up new ­conversations (Morvarid 2022; Mousavian 2021, 2022; Zarepour 2022). But does that mean analytic philosophy is now the universal language of the discipline? If I am honest about my own decolonial commitments, there is much that we can engage with the analytic not least as an open t­radition capable of transforming and acting as a broad tradition. I would like to see the emergence of a properly global, and pluriversal approach to ­philosophy in which we can bring together different traditions in conversation and open ­different kinds of conversations, not just Islamic analytic but also ones in which we place Avicenna alongside Nāgarjuna, Suhrawardī alongside Dogen, and Mullā Ṣadrā juxtaposed with Madhusudan Saraswati. If analytic ­philosophy can provide a language for those conversations and can continue to develop and embrace globality, then it will become i­ndispensable for all historians of philosophy regardless of the cultural and intellectual c­ ontext in which they find themselves. Let us return to Diagne who does point out that the word of God and indeed her existence is a problem posed to ­translation. The use of language and schemes are not in themselves obstacles to ­conversation; what is required is a clear acknowledgment of positionalities and the ­particularities of cultural, social conditions, and human capabilities that determine the production of philosophy (Diagne 2022: 134). Philosophy as a pluriversal human practice is thus also a local one. References Abdelnour, Mohammed Gamal (2023). ‘The Qurʾān and the future of Islamic a­ nalytic theology’. Religions 14: 556ff. Ahsan, Abbas (2021). ‘The possibility of analytic philosophy in United Kingdom ­madrasas’. Journal of Islamic and Muslim Societies 6: 56–83. Avicenna [Abū ʿAlī Ibn Sīnā] (1973). The Metaphysica of Avicenna (Ibn Sīnā). Transl. by Parviz Morewedge. Persian Heritage Series no. 13. London: Routledge, Kegan & Paul.

Contemporary Analytic Philosophy for Historians  21 ——— (2005). The Metaphysics of the Healing. Ed. and transl. by Michael E. ­Marmura. Provo, UH: Brigham Young University Press. Azadpur, Mohammad (2020). Analytic Philosophy and Avicenna: Knowing the ­Unknown. Abingdon: Routledge. Baggini, Julian (2018). How the World Thinks: A Global History of Philosophy. London: Granta. Beaney, Michael (2013). ‘What is analytic philosophy?’. In Michael Beaney (ed.) The Oxford Handbook of the History of Analytic Philosophy, 3–29. Oxford: Oxford University Press. Bertolacci, Amos (2006). The Reception of Aristotle’s Metaphysics in Avicenna’s Kitāb al-Shifāʾ. Leiden: Brill. Boolos, George (1984). ‘To be is to be the value of a variable’. The Journal of ­Philosophy 81: 430–449. Brandom, Robert (2009). Reason in Philosophy: Animating Ideas. Cambridge, MA: Belknap Press at Harvard University Press. ——— (2019). A Spirit of Trust: A Reading of Hegel’s Phenomenology. Cambridge, MA: Harvard University Press. Corbin, Henry (1960). Avicenne et le récit visionnaire. Paris: Verdier. de Ray, Christopher (2022). ‘Existence exists, and it is God’. Religious Studies ­advance article 1–16. Derrida, Jacques (2016). Of Grammatology. Transl. Gayatri Chakravorty Spivak. 402h anniversary edition. Baltimore, MD: Johns Hopkins University Press. Diagne, Souleymane Bachir (2022). De langue à langue. L’hospitalité de la traduction. Paris: Albin Michel. Dummett, Michael (1993). Origins of Analytical Philosophy. London: Duckworth. El-Bizri, Nader (2000). The Phenomenological Quest between Avicenna and ­Heidegger. Binghamton: Global Academic Publishing. Frede, Michael (2022). The Historiography of Philosophy. Edited by Katerina ­Ierodiakonou with a Postface by Jonathan Barnes. Oxford: Oxford University Press. Ganeri, Jonardon (2016). ‘Why philosophy must be global: A manifesto’. Confluence 4: 134–186. Goodman, Lenn E. (1992). Avicenna. London: Routledge. Gordon, Lewis (2019). ‘Decolonizing philosophy’. The Southern Journal of ­Philosophy 57 Spindel Supplement: 16–36. Gracia, Jorge J.E. (1992). Philosophy and Its History: Issues in Philosophical ­Historiography. Albany: State University of New York Press. Gutas, Dimitri (2014). Avicenna and the Aristotelian Tradition. 2nd edition. Leiden: Brill. Hacker, P.M.S. (2013). ‘The linguistic turn’. In Michael Beaney (ed.) The Oxford Handbook of the History of Analytic Philosophy, 926–947. Oxford: Oxford ­University Press. Ḥāʾirī Yazdī, Mihdī (2017). Universal Science [ʿElm-e kollī]. Transl. John Cooper. Ed. Saiyad Nizamuddin Ahmad. Leiden: Brill. Hasse, Dag (2000). Avicenna’s De Anima in the Latin West. London: Warburg Institute. König-Pralong, Catherine (2019). La colonie philosophique. Paris: Éditions EHESS. Lewis, David (1986). On the Plurality of Worlds. Oxford: Blackwells.

22  Sajjad Rizvi Mignolo, Walter and Catherine Walsh (2018). On Decoloniality: Concepts, A ­ nalytics, Praxis. Durham, NC: Duke University Press. Mīr Dāmād, Sayyid Muḥammad Bāqir (1977). Al-Qabasāt. Eds. Mahdī Muḥaqqiq et al. Tehran: McGill Institute of Islamic Studies Tehran Branch. ——— (2002). Muṣannafāt I. Ed. ʿAbdullāh Nūrānī. Tehran: Anjuman-e āsār va mafākher-e farhangī. ——— (2012). Al-Ufuq al-mubīn. Ed. Ḥāmid Nājī Iṣfahānī. Tehran: Mīrās-e maktūb. Montefiore, Alan (2014). ‘The ‘continental’ tradition?’. Royal Institute of Philosophy Supplement 24: 27−43. Moosavi, Leon (2020). ‘The decolonial bandwagon and the dangers of intellectual decolonisation’. International Review of Sociology 30.2: 332–54. Morvarid, Mahmoud (2022). ‘Truthmaking, resemblance, and divine simplicity’. ­Religious Studies 58: 547−63. Mousavian, Seyed N. (2021). ‘Is Avicenna an Empiricist?’. In Shahid Rahman, et al. (eds.) Mathematics, Logic, and Their Philosophies, 443–474. Logic, Epistemology, and the Unity of Science Series. Dordrecht: Springer. ——— (2022). ‘Avicenna on the Semantics of Maʿnāʾ’. In Christina Thomsen Thörnqvist and Juhana Toivanen (eds.) Forms of Representation in the Aristotelian Tradition. Volume 3: Concept Formation, 95–140. Philosophia Antiqua: A Series of Studies on Ancient Philosophy. Boston, MA/Leiden: Brill. Nasr, Seyyed Hossein (1964). Three Muslim Sages: Avicenna, Suhrawardi, Ibn ʿArabi. Cambridge: Harvard University Press. Nusseibeh, Sari (2018). Avicenna’s al-Shifāʾ: Oriental Philosophy. Abingdon: Routledge. Ogunnaike, Oludamini (2017). ‘African philosophy reconsidered: Africa, religion, race, and philosophy’. Journal of Africana Religions 5.2: 181–216. Preston, Aaron (2007). Analytic Philosophy: The History of an Illusion. London: Continuum.Rieck, Eric (2013). ‘Analytic philosophy and philosophical history’. In E. Rieck (ed). The Historical Turn in Analytic Philosophy, 1–38. Basingtoke: Palgrave Macmillan. Rivera, Mayra (2021). ‘Embodied Counterpoetics: Sylvia Wynter on Religion and Race’. In An Yountae and Eleanor Craig (eds). Beyond Man: Race, Coloniality, and Philosophy of Religion, 57–85. Durham, NC: Duke University Press. Shīrāzī, Mullā Ṣadrā (2005). Al-Ḥikma al-mutaʿāliya fīʾl-asfār al-ʿaqliyya al-arbaʿa. Eds. G. Aʿwānī et al. Tehran: Sadra Islamic Philosophy Research Institute. Simons, Peter (2013). ‘Metaphysics in analytic philosophy’, in Michael Beaney (ed.) The Oxford Handbook of the History of Analytic Philosophy, 709–728. Oxford: Oxford University Press. Skinner, Quentin (2002). Visions of Politics 1: Regarding Method. Cambridge: ­Cambridge University Press. Sorrell, Tom (2005). ‘Introduction’. In Tom Sorrell and G.A.J. Rogers (ed). Analytic Philosophy and History of Philosophy, 1–11. Oxford: Clarendon Press. Strawson, Peter F (1959). Individuals: An Essay in Descriptive Metaphysics. London: Methuen & Co. Ṭabāṭabāʾī, Sayyid Muḥammad Ḥusayn (2018). The Elements of Islamic Metaphysics [Bidāyat al-ḥikma]. Transl. Ali Quli Qaraʾi. Ed. Janis Esots. 2nd edn. London: ICAS Press. Táíwò, Olúfémi O. (2022a). Elite Capture: How the Powerful Took Over Identity Politics (and Everything Else). London: Pluto Press.

Contemporary Analytic Philosophy for Historians  23 Táíwò, Olúfémi (2022b). Against Decolonisation: Taking African Agency Seriously. London: Hurst & Co. Taylor, Charles (1984). ‘Philosophy and its history’. In Charles Taylor, et al. (eds). Philosophy in History, 17–30. Cambridge: Cambridge University Press. Textor, Mark (2021). The Disappearance of the Soul and the Turn against M ­ etaphysics: Austrian Philosophy 1874–1918. Oxford: Oxford University Press. Turner, Jamie (2021). ‘An Islamic account of reformed epistemology’. Philosophy East & West 71.3: 767–92. ——— (2023). ‘Ibn Taymiyya’s “common-sense” philosophy’. In Amber Griffioen and M. Backmann (eds) Pluralizing Philosophy’s Past: New Reflections on the ­History of Philosophy, 197–212. Cham: Springer/Palgrave Macmillan. Williams, Bernard (2006). The Sense of the Past. Essays in the History of Philosophy. Ed. Myles Burnyeat. Princeton, NJ: Princeton University Press. Wisnovsky, Robert (2003). Avicenna’s Metaphysics in Context. London: Duckworth. Wynter, Sylvia (2003). ‘Unsettling the Coloniality of Being/Power/Truth/Freedom’. The New Centennial Review 3: 257–337. Yablo, Stephen (2010). Thoughts and Things. New York: Oxford University Press. Yountae, An and Eleanor Craig (2021). ‘Introduction’. In An Yountae and Eleanor Craig (eds) Beyond Man: Race, Coloniality, and Philosophy of Religion, 1–31. Durham, NC: Duke University Press. Zarepour, Mohammad Saleh (2022). Necessary Existence and Monotheism. ­Cambridge: Cambridge University Press.

2 What Can the History of Islamic Philosophy Do for Contemporary Analytic Philosophers? Anthony Robert Booth 2.1 The Corrective View about the Value of History of Philosophy In this chapter, I seek to propose that the history of philosophy is valuable for analytic philosophy because it is at least sometimes essential in presenting a coherent narrative, a world picture, or way of understanding the world. Analytic philosophy’s value, it will be suggested, is tied up to how we think through how true we are to our own ways of understanding the world. I do this with the help of Ibn Sīnā’s concept of al-wahm and al-Fārābi’s ideas regarding the epistemic value of Prophecy. Studying the history of Islamic philosophy, then, helps the analytic philosopher have a better sense of its own aims and purpose. In Section 2.1, I argue against a competing ­conception of the value of studying history of philosophy (that I label ‘the corrective view’). In Section 2.2, I present an interpretation of Ibn Sīnā’s concept of al-wahm, along with how it ties up with al-Fārābī’s ideas regarding the epistemic value of Prophecy, wherein al-wahm in presented as being a cognitive, but nonproposition and pre-reflective judgement that I term ‘understanding-as’. In Section 2.3, I then show how these ideas can usefully inform a conception of what world views are as well as give us a rich idea of what the value of ­history of philosophy is in a way that makes the latter essential to the former. Section 2.4 concludes. Here is a way of thinking about how doing the history of philosophy is useful for contemporary analytic philosophy. Contemporary analytic ­philosophers routinely make historical claims about philosophical ideas and figures. Often, these claims are plain false, or underspecified, or not properly understood given the context in which they were written. The historian of philosophy then provides a corrective to these errors. That is what of value for analytic philosophers: truth about matters of history of philosophy. For ­ ublication example, analytic epistemology is often thought to begin with the p of Gettier’s classic 1963 paper. And it is often thought to begin then because up to the publication of Gettier’s paper, the consensus view had been that knowledge is justified, true belief. That consensus, according to many (if not all) epistemology introductory undergraduate courses in the analytic tradition until recently, was an historically sustained one. That historical consensus, DOI: 10.4324/9781003327714-4

Islamic Philosophy for Contemporary Analytic Philosophers  25 so it was claimed, even stretched as far back so as to include Plato (with the Theatetus cited as the corroborating source). Part of why Gettier was held to be so important is that he upended centuries of ­philosophical consensus with one short paper. However, historians of philosophy have now shown that claim to be at the very least overstated. Plato did not hold the so-called ‘­tripartite view’ about what knowledge is (rather he likely thought knowledge was unanalysable - his and the concomitant ancient view was then more akin to Tim Williamson’s contemporary ‘Knowledge First’ fad – see Antognazza (2015)).1 Further, indeed a more careful history reveals that the idea that there was any consensus at all around the tripartite view is really nothing more than a myth. Analytic philosophy was working on the ­assumption of a falsehood which doing some proper history of philosophy exposed. …if one returns to examining main figures and traditions in the history of philosophy, the ‘standard’ analysis begins to look surprisingly nonstandard. The ‘traditional’ view, it appears, is not a tradition stretching back to the infancy of philosophy, the epistemic naivety of which was finally exposed in 1963 by Gettier’s short paper. Instead, it begins to look like a twentieth century view, prominent especially in the AngloAmerican world, which after the middle of the century, has repeatedly tried and, according to many, failed to correct itself. (Antognazza 2015, p. 174) In fact, according to Dutant, the corrective does not require even a very c­ areful history, just attendance to history: Even though the Legend [that Gettier refuted the ‘traditional’ account of knowledge’] figures in almost every epistemology handbook, I do not expect a strong resistance to the claim that it is false. The Legend is not widespread because it has been powerfully defended—it has hardly been defended at all—but because no better picture is available. (Dutant 2015, p. 95) The historian of philosophy should then provide us with this better historical picture, and in doing so proves their worth. We value truth, and the historian of philosophy brings that to the table about historical claims in philosophy. I think this picture of what is the value of the history of philosophy for the analytic philosopher (let’s call this ‘the correction picture’2), while extremely tempting, is wrong-headed. I will now say why and then give an ­alternative account, using medieval Islamic philosophy. My argument against the ­correction picture begins by asking the following question: are ­historical claims essential to analytic philosophy? They are either essential or they are not. Let’s consider the claim that they are essential to analytic ­philosophy. The claim here, to be clear, is not merely that one cannot do analytic philosophy without making historical claims, but rather that part of analytic

26  Anthony Robert Booth philosophy’s epistemic value is tied up to what it says about its history. If the correction picture was committed merely to the claim that analytic philosophy ­cannot but make historical claims, but that these are irrelevant to its epistemic value, then the correction picture would make the historian of philosophy a mere pedant. I think there is more to the ­history of ­philosophy than mere pedantry. But if that is true, then the value of history of p ­ hilosophy for the analytic philosopher cannot merely be to provide truth about historical claims. Their contribution would be more inexorably tied to whatever the value of analytic philosophy turns out to be. Correcting errant historical claims may be part of the ­historian’s modus operandi but the value brought to analytic philosophy cannot end there if history of philosophy is inescapable for analytic philosophy, and the historian of philosophy is not just a boring pedant. Now let’s consider the claim that historical claims are not essential to ­analytic philosophy. That claim is incompatible with the correction p ­ icture too. Fairly obviously, when the analytic philosopher makes no historical claims, there is no historical claim available for the historian to correct. Since the analytic philosopher does not make any historical claim, the work of the historian of philosophy would look pretty superfluous. But what about, more interestingly, the historical claims that the analytical philosopher ­contingently indulges in? Would it not be valuable for the analytical p ­ hilosopher to be ­corrected in these cases, when they inevitably get the ­history wrong? The trouble here is that we have retreated once again to the view that the ­ ­historian of philosophy is no more than a pedant, since a­ nalytical ­philosophy could well have proceeded without making the historical claims. From this ­perspective, the historical claims made by the analytic philosopher will add no more than rhetorical emphasis to their work. Correct history (that is, a history that ­contains no falsehoods, according to the correction view) will not contribute to its epistemic value. History of philosophy then, to the ­analytic philosopher, is just a series of quibbles about the manner in which their work is presented. Work whose value is unrelated to how it is presented. Now, we could move to the claim that the manner in which a work is ­presented and the epistemic value of a work cannot be disaggregated in the way that was implicit in the point above.3 Rhetoric is important. It tells us, very finite creatures with limited cognitive recourses and time, what evidence we ought to consider. Gettier’s paper is considered important in large part because it overturned a large consensus. That there was a large consensus about an issue assumes that the issue is an important one, since it had ­occupied the minds of the most famous philosophers all the way back to Plato. I am very inclined to agree with this line of thought (I think one can appeal here to the idea that what evidence we ought to consider is something that sits tightly within the epistemic purview – see Friedman (2020) for a recent defence, for instance). But it is incompatible with the correction picture. This is because it tells us that where the analytic p ­ hilosopher ­(contingently) makes ­historical claims, the epistemic value of their enterprise is in part dependent on the way they present history. The epistemic value of

Islamic Philosophy for Contemporary Analytic Philosophers  27 history of ­philosophy then a ­forteriori cannot only be to bring truth about matters of ­history of ­philosophy. History of philosophy must also have value in so far as ­analytic philosophy has value. Consider that here analytic philosophy’s value would not be simply reduced to truth or knowledge, but rather truth and ­knowledge about things that are important to us (at least). Since – very plausibly – we are not able to read off what is important simply from an accurate h ­ istory, the value of history of philosophy would consist in more than giving us that accurate history. For example, as I will propose for the rest of this c­ hapter, it is at least sometimes essential in presenting a coherent narrative, a world p ­ icture, or way of understanding the world. It is this that analytic ­philosophy, I will ­suggest, aims to give us (and thus why history is essential to it). Moreover, I want to try to show that this picture of analytic ­philosophy’s aims can be drawn by doing a rational reconstructive kind of history of Islamic ­philosophy. Doing history of Islamic philosophy then helps us ­understand what contemporary analytic philosophy is, or ought to be. In the next ­section, I attempt this bit of history, drawing together an interpretation of Ibn Sīnā’s al-wahm and al-Fārābī’s thinking on the unique epistemic value of Prophecy. In Section 2.3, I apply the concepts learnt here to the question regarding the aims of analytic philosophy and its relationship to its history. 2.2

Ibn Sīnā on Estimation and ‘Understanding-as’

One of the many (many) innovative ideas that the famous medieval Islamic Philosopher Ibn Sīnā developed was the concept of al-wahm – often translated as ‘estimation’ (I prefer ‘understanding-as’, as we will see). The most famous example of Ibn Sīnā’s is of a sheep who apprehends a wolf as a dangerous being, and of someone’s being aware of a child’s love. This may bring to mind Tamar Gendler’s concept of ‘alief’ as compared to belief (where the former and not the latter is characterised essentially by being affective and resistant to counter-evidence). One of Gendler’s examples here is of the ‘alief’ – when one is standing atop the glass walk-way over the Grand Canyon – that one is in an unsafe place, along with the rational belief that one is in a safe space, having heard convincing evidence in its favour (see Gendler 2008). For Ibn Sīnā, however, al-wahm need not be affective, and its deliverances can be overturned by a rational mind. It is neither perception nor imagination, and we share it with animals: Further, there is in animals a power (estimation) that passes judgement, upon such or such a thing that it is so or not so. It does this decisively. As a result, the animal flees away from shunned evil and seeks chosen good. If is also evident that this sense is other than the imaginative, since this last imagines images of the sun in accordance with what it has obtained from the opposite sense, to be of the size of its disc and so forth… The matter stands in this sense quite otherwise. So, too, the lion

28  Anthony Robert Booth finds its prey from far away, even the size of a small bird, yet its form and size in no way perplex it, but it makes for it. It is also evident that this sense is other than the imaginative. This is because the imaginative power performs its manifold deeds without belief and conviction on its part that matter are in accordance with its imagining. This power is what is named the estimative or the surmising faculty or judgement, the ability to sense an intention. (Ibn Sīnā 2013, p. 36) Just as the lion is able to ‘shun’ its visual perception of its prey as being the size of a small bird such that in no way perplexes it and inhibits his hunting (so too are those who have the rational faculties able to shun and overturn the workings of estimation). The estimation is not to be identified with the imagination, but it uses the imagination in order to do the overriding work against perception. The rational mind can also use the imagination in turn to override and ‘shun’ the ‘sway’ of the workings of estimation. Again, the imaginative power is called by this name – imagination – if the ability to sense an intention or surmising sense alone uses it. If the rational soul uses it, it is called the thinking or cogitative sense. (Ibn Sīnā 2013, p. 37) Using Gendler’s example, we can overturn the impression of the glass walkway as an unsafe place by – actively, and voluntarily – using the ­imagination to our advantage, in line with our better judgement that it is a safe place.4 I think, then, that al-wahm is best understood as the cognitive corollate of ­seeing-as (or ‘incidental perception’): hence, why I think it is useful to ­translate this as ‘understanding-as’. When we see our friend John we see him as John rather than as a human with a particular physical shape (a point made by Aristotle in the De Anima). And this is somewhat under our control – we can choose for example to see the Jastrow’s Duck-Rabbit as a Duck. Thus, we might be able to choose whether we understand a glass walk-way as a safe place or not. Similarly, we can have brute rational intuitions (say that p and not p cannot both be true) or we can understand that intuition as the logical principle of non-contradiction, and this may be in part up to us. Now, for Ibn Sīnā al-wahm can work with both the deliverances of perception and ­intellectual intuitions,5 and as such can work with propositions. Ibn Sīnā heavily had set a precedent for this kind of combinatorial role for estimation in his remarks on how what he calls the ‘the common sense’6 functions: Not one of the outward senses unites within itself the perception of color, odor, and softness, and yet, we often come upon a body that is yellow, and perceive at once so much about it, namely that it is honey sweet, nice of smell, and fluid, although we have neither tasted nor smelt, nor even touched it. It is clear that we posses a power wherein

Islamic Philosophy for Contemporary Analytic Philosophers  29 are assembled the perceptions of the four senses, and have thus become summed up into a single form. (Ibn Sīnā 2013, p. 35) So, we can think of understanding-as as putting together the objects of thought and perception to create a world view which the intellect can then in reflection (wilfully, using the imagination) assent to or deny.7 This is because, Ibn Sīnā holds, the workings of the understanding-as (al-wahm) are ­themselves not reflective: It is known that if the sensibles have principles and foundations, then these are prior to the sensibles, and are not themselves sensed, nor is their existence according to the mode of the existence of sensibles. So it is not possible for us to represent this existence in the estimation. So for this reason, neither the estimation itself, nor its activities, are ­represented in the estimation.8 We should be careful, however, to note that all this is not to say that for Avicenna estimation grasps universals (intelligibles) directly. Rather, that once universals have been grasped and are retained in memory, estimation can put them together in various ways. How they fit together is an intentional (though non-propositional) object that is apprehended by the estimation. So it isn’t the universals that are being grasped in estimation, rather the relations between them which I think can be thought to be particulars.9 Interestingly, Ibn Sīnā often talks about revelation while talking about al- wahm. For instance, take this passage: Again, the rational soul is sometimes so fitted out in a few persons through vigils and conjunction with the universal mind to be quite independent of taking refuge in syllogistic argument and reflection, but rather is sufficiently stored with inspiration and revelation to r­ ender it wholly absolved from such ordinary means as mental ­ratiocination. This peculiar property of the reasoning mind is called the sacred soul and it is accordingly called the Spirit. Unto such a favoured rank and degree none shall attain save prophets upon whom be peace and blessing. (Ibn Sīnā 2013, p. 43)10 Leaving to one side the complex issue of ‘the universal mind’, the idea here is that for some extremely unique individuals – Prophets – they can get a direct understanding-as that is correct without the need to be corrected by the ­intellect. To understand what Ibn Sīnā means here, I think it is useful to turn to the work of his predecessor, the also esteemed medieval Islamic Philosopher Abu Nasr al-Fārābī. I think that a unifying problem in a lot of Falsafa (medieval Islamic Philosophy) has to do with the issue of verifying Prophecy. An early ­theological

30  Anthony Robert Booth school (Mutazilite) in Islam has held what I have labelled in ­previous work a form of ‘Evidentialism’ – the view that only epistemic ­reasons are relevant in settling the question regarding what one ought to believe. This gave them an easy answer to how one is to verify the Prophet: ­verisimilitude. The true Prophet’s message will accord with the truth. However, this brings up the ­further issue – very familiar to modern epistemologists working on ­testimony – of explaining the epistemic uniqueness of Prophecy. Why did God send down a Koran, if we can independently reason our way to a­ nything that is contained within it? Correlatively, the purpose of experts seems o ­ bviated if the only way to ascertain a genuine expert is to become an expert. As has been argued in other work (e.g. Booth 2016), al-Fārābī’s solution to this problem is ingenious: any single proposition uttered by the Prophet can be c­ orroborated by reason and evidence. However, Prophecy remains s­omewhat esoteric in the way that it presents how the various truths of the world hang together – how it understands the world, in other words. The Prophet, through a s­uperior faculty of the imagination is able to apprehend all the intelligibles (all a priori truths, roughly) at once, and as such u ­ nderstands their interrelation and so the world as a whole better than any other human. Further, it affords the Prophet a level of warranted certainty that goes way beyond that of any ­ordinary human. This is because in ­grasping the totality of the world, the Prophet uniquely meets stringent (broadly ‘K-K’) ­conditions on ­certainty11: they not only have rational intuitions (such as that p and not p cannot both be true) but (unlike the rest of us) understands why they have those i­ ntuitions. This was of a piece with Aristotelian e­ pistemology that demanded that the middle term of a demonstrative syllogism be e­ xplanatory – indeed it aims to take Aristotelian epistemology to its logical conclusion. Returning to Ibn Sīnā’s idea of Understanding-as (al-wahm), it seems then that what al-Fārābī is claiming is that Prophecy, qua Prophecy, is really a kind of narrative testimony. That is not to say that it is fiction – some ­narratives are meant to be factual. Rachel Fraser recently makes the point with the following examples: Gibbon’s The History of the Decline and Fall of the Roman Empire is a narrative, but is supposed to be read as history. Mary Prince’s The History of Mary Prince is a narrative, but it is supposed to tell us the truth about Prince’s life. (Fraser 2021, p. 4027) Al-Wahm for Ibn Sīnā was precisely that faculty that brings together our diverse impressions and beliefs (such that a dog, in another example of his, is able to learn to avoid being beaten by a stick by putting together its ­experiences of pain and sticks). Presumably, the more comprehensive the bringing together, the richer the world view (richer than sticks and masters and pain). Further, presumably the Prophet’s is the most comprehensive of that bringing together that is possible such that it alone achieves ­reflectiveness

Islamic Philosophy for Contemporary Analytic Philosophers  31 (so even satisfies Fārābi’s strict K-K demands). I think that explains why Ibn Sīnā thought that there was no need then for the Prophet to engage in ­philosophical or reflective thought. There is a way that the different parts of the world hang together (an ‘intention’) and the Prophet directly and ­effortlessly grasps this without having to think about it. For the rest of us, our estimative faculty is such that it regularly leads us astray and we need to do a lot of reflection to get it to even vaguely resemble the state of the Prophet’s. This sketch I have just given should generate several questions. For ­example, if there is a way that the world hangs together, such that the Prophet correctly apprehends it in estimation, then is there not an evidential proof of Prophecy after all? Gibbon’s History of the Decline and Fall of the Roman Empire is meant to get right the history of the decline and fall of the Roman Empire, after all. Another way to put the question: is the right way to understand the world itself a proposition, or set of propositions? If not, in virtue of what is it the right way to understand the various components of the world? Further, Ibn Sīnā – following Aristotelian precedent – feels the ­ istinct faculty (or ‘inner sense’) when it corresponds to a need to identify a d distinct kind of object apprehended. For there to be a need for an ‘estimative’ faculty, then there needs to be an intention, an object of thought for that faculty. If the object is not a proposition, then what is it? After all, if there is no intentional object to the thought, can it really be classified having content at all? In response, we could here appeal to modern philosophy of mind and metaphysics. For instance, Jubien (2001) denies that there are such things as propositions, so that none of our mental states can have propositions as their content. Instead (to avoid the charge of ‘psychologism’) he argues that the objects of thought are ‘Platonic properties and relations’ (p. 47 my ­emphasis). Unlike Jubien, Ibn Sīnā does not seem to deny that some of our mental states have propositions are their object, but nevertheless, ­similarly to Jubien, could be interpreted as holding that mental states can have r­ elations as their object (and these relations are not themselves propositions) – these are the objects of understanding-as (al-wahm). So-called ‘Predictive Processing’ accounts of the mind (see e.g. Clark 2016) also reject that all ­cognitive states must have ­propositions as their object. As Daniel Williams puts it in a useful e­ xplanation of the development of the view A long, alternative tradition in philosophy and psychology rejects this propositionalist account of mental representation in favour of an  iconic  or  analogue  understanding of the mind’s representational capacities…. Advocates of this view contend that much of sophisticated internal representation is founded on similarity or physical analogy with the mind’s objects. Instead of language, the relevant paradigms from everyday life here are representational tools such as pictures, diagrams, graphs, maps, and models. (Williams 2018, p. 513)

32  Anthony Robert Booth Further, the idea that the relations between all the propositions of the world are themselves propositions threatens an infinite regress (much like the one in Aristotle’s famous ‘Third Man Argument’). If the relations are ­propositions, then they themselves will stand in a relationship with the set of all ­propositions, and that relationship will also be a proposition… and so on… Rather, I think we are best thinking of the way these propositions fit together as akin to how Wittgenstein saw the logical form of a singular proposition: as something that can only be shown, and not said. Be that as it may, we are still left with the problem – for al-Fārābī and Ibn Sīnā – as to whether Prophecy is really verifiable (and the concomitant more general problem with testimony and expertise). What makes Prophecy really distinctive is that it provides us with the right understanding-as of how the various propositions of the world fit together. But verifying whether it is the right understanding-as is beyond any of us, ordinary humans. Of course, we can verify the truth of the individual propositions constituting that ­understanding-as, but it must be an open question whether there could be a false Prophecy that contains only true propositions but put together in the wrong way (an example may be that there is nothing to choose, e­ pistemically speaking, between science understood as sectarian vs. ­science understood as secular). We will not be able to tell apart the true from the false Prophecy. Perhaps we can say that the two world views are on an epistemic par, but are not on a par on non-epistemic criteria – that we need to e­ valuate them on, say, practical grounds (after all, it looks like we can choose to make gestalt-switches as when we choose to see a Duck in Jastrow’s ­Duck-Rabbit). I do not have the space here to explore whether that move is available to al-Fārābī or Ibn Sīnā, nor deal with the many other ­questions that may arise here. Would such a move really solve the issue of what is ­epistemically unique about Prophecy? Must the understanding-as of true Prophecy involve all ­(presumably infinite) propositions or can it deal with an important sub-set only (e.g. only the necessary, or a priori truths)? Can a sharp distinction between seeing-as and understanding-as really be drawn? If the content of al-wahm is at least partly non-propositional, then what does non-Prophetic thought and reflection look like? My aim here has not been to give such a ­comprehensive account of Ibn Sīnā and al-Fārābī so as to address these ­questions. Rather, to weave out of their views here an account of what Philosophy and Philosophical reflection is. I move to flesh that out further, and return to the question of the value of the history of philosophy, in the next section. 2.3

World Views and the History of Philosophy

As I mentioned in Section 2.2, Antognazza claims that the value of the ­history of philosophy outruns what is claimed by what I called the correction view. In particular, she claims that the value of history of philosophy for the

Islamic Philosophy for Contemporary Analytic Philosophers  33 analytic philosopher has to do with keeping the latter open-minded. As she puts it: Doing history of philosophy is a way to think outside the box of the latest philosophical orthodoxies or commonly held beliefs. Somewhat paradoxically, far from imprisoning its students in outdated and ­ ­crystallized views, the history of philosophy trains the mind to think differently and alternatively about the fundamental problems of ­philosophy. It keeps us alert to the fact that latest is not always best, and that a genuinely new perspective often means embracing and ­developing an old insight. (Antognazza 2015, p. 168) I think this view is broadly right, but requires supplementation in order for it not to be vulnerable to the following issue. Even if the historian of ­philosophy could cover everything in the history of philosophy, surely the analytic philosopher will not be able to pay attention to all of the history while simultaneously doing their thing. Note: we cannot retreat to the view that the historian need only cover what is historically important, since pace Antognazza, the point is that we may currently be wrong about what is important. We are clearly finite creatures, and so prone to fashions and fads not only about what are the important ‘perennial problems’ but also about what historical periods and figures are worthy of our attention. And the problem is that the former is likely to overdetermine, or at least strongly influence, the latter. Further, in order to bring home the point that history of philosophy is worth doing, Antognazza illustrates how contemporary views in the history of philosophy (e.g. current ideas about ‘knowledge-first’) have important historical precedents, that perhaps have been overlooked. But this has the corollary of making history of philosophy redundant, where its aim is to keep us open-minded as to views different from the mainstream. We are able to keep our minds open just by doing first-order (non-historical) philosophy, else there would hardly be historical precedents to some of our current ways of thinking. Timothy Williamson, for example, did not seem to construct the knowledge-first view via doing the history of philosophy. So what is really the unique epistemic value of history of philosophy? I think here is where the picture of what is a world view that is suggested by Ibn Sīnā and al-Fārābī can be of help. Recall that this picture was meant to solve a structurally similar problem regarding what is the unique value of Prophecy when we can just engage first-order reason and evidence to come up, for ourselves, anything that is in genuine Prophecy. The idea is that a world view is a composite of both sets of propositions and the manner in which those propositions are understood-as related to one another. The ­manner in which those propositions are understood to be related to one another is not something that can be expressed in propositional form, and

34  Anthony Robert Booth as such will not be evaluable in epistemic terms. This is not to say that any world view is as good – epistemically speaking – as another: they are made up of ­propositions after all, and these can be true or false. Thus, we can engage in first-order philosophical thinking to grasp the putative truth of, or at least have a ­conception of, the proposition that knowledge is an unanalysable primitive (for example). This idea may or may not have historical precedents. But ­discovering whether it does is not why we study philosophical history. We study history because while our ancestors may have had a similar idea to ours, the way they understood a given idea as related to others in their world view is likely to be radically different to ours. We do not look to ­history in search of novel ideas so that – in Skinner’s famous phrase – it ‘does our ­thinking for us’. We can do our thinking for ourselves! But we cannot employ first-order reason to know a priori our ancestors’ world views, and where they are on an epistemic par with ours, there is nothing epistemically to choose between them. Yet what is our world view will likely overdetermine what evidence we seek and what we find important, and so it is necessary to engage world views other than our own to keep us open-minded.12 Now, to be clear, I am not saying that history of philosophy is like Prophecy! Rather that Ibn Sīnā’s and al-Fārābī’s reflections on what is ­distinctive about Prophecy bring to the fore a rich account of what a world view is. That it contains something that cannot be grasped directly by rational reflection, but nonetheless will affect the way that we do ­philosophy (for instance, what ­philosophical problems we think are worthwhile). But is not painting world pictures part of the philosophical exercise – even if it is something that employs the imagination perhaps more than ­reason? And why cannot we paint these pictures without doing history? The answer is that we can, but that doing history keeps us open-minded in the ­manner Antognazza ­suggested. What propositions we deem worthy of our ­consideration, and so form part of the alternative world views we might construct to consider in contrast to our own, will in part be overdetermined by our current world view. Such that we have our best chance of coming up with radically ­alternative ones by ­engaging in history. To the first ­question, I answer it in the ­affirmative. I do not see any reason why ‘­philosophy’ needs to be ­straightjacketed inside ­theoretical ­reason. Nevertheless, the ­conception of philosophy that is ­emerging here is looking like anathema to the c­ontemporary analytic p ­ hilosopher. But this need not be a problem, since this last move is optional and we can stick to ­thinking about philosophy in the way that the analytic philosopher does – as the employment of reason to uncover truth – while heeding the notion of a world view and the value of history of ­philosophy that is on the table. The ­analytic ­philosopher’s job is to attend to the truth of the individual ­propositions that make up any world view such that we can ever determine, for example, that two d ­ ifferent world views are on an epistemic par. The historian’s is to show us how we have (wilfully or not) put the different ­propositions together throughout ­history. Since what are our world views will affect what we find important, and what evidence we ought to seek, there is value for the ­analytic

Islamic Philosophy for Contemporary Analytic Philosophers  35 philosopher to be cognisant of this. Not least because it should help to keep us open-minded. Furthermore, ­perhaps most ­importantly, taking a cue from Ibn Sīnā’s ideas on reflection and the o ­ verturning of understandings-as we think we ought not have, ­analytic ­philosophy helps us be able to judge in that our tacit world views are in tension with what we reflectively think our world views should look like. This is perhaps one of the most valuable things ­analytic philosophy can do: help us be true to ourselves. To come to a ­position, metaphorically speaking, where we do not experience a tension in how we understand our situation staring down a glass walk-way – to judge in the manner we think we ought. To do this, we need to know what in fact are our tacit world views. In so far as the historian identifies these, by distinguishing them from our historical ones, we cannot realise analytic philosophy’s value without doing history of philosophy. 2.4

The History of Islamic Philosophy

The titular question this chapter asks is specifically about the value the ­history of Islamic philosophy has for the analytic philosopher. A tempting way, and perhaps the most natural way, of answering this question is to find some technical philosophical moves and innovations in some part of non-­contemporary Islamic philosophy that can then be usefully mined to solve some problem(s) occupying the analytic philosopher. But the question would then be: why could the analytic philosopher not have discovered that ­technical device for themselves? It would seem a matter of sheer contingency that the historical Islamic philosophers got there first. Perhaps the idea here would be that the value of philosophy comprises the uncovering of truths about abstract matters such that it becomes irrelevant how or who came up with them. Since what is of value is the truth, then we should get it by any means at our disposal and we are likely to find many such truths in the labours of our historical peers. According to this view, in that sense, there is nothing particularly special about Islamic philosophy per se. I think this view is a close cousin of the correction view we discussed earlier and its ­concomitant value monism about the point of philosophical and historical inquiry. In this chapter, I have tried to show that looking to the history of Islamic ­philosophy gives us an alternative picture of what the value of ­philosophy is: tied up with self-knowledge, understanding the world, and being true to ourselves. And that this picture of the value of philosophy makes the study of its h ­ istory indispensable. What about the history of Islamic philosophy in particular? Well, as I have said, I think we can find from Islamic philosophy ­alternative and rich ideas about what the aims and values of philosophy are. Yet, this alternative picture allows room for analytic ­philosophy as it is ­currently dominantly conceived and practised and indeed has been explicated here using some concepts from analytic philosophy. Part of the value of h ­ istory of philosophy that this picture espouses has to do with keeping ourselves open-minded – not so much to ideas that can be worked out a priori, but

36  Anthony Robert Booth rather with alternative conceptions or world views of how those ideas fit together. From this perspective then, philosophical ideas that are conceived in an historical, geographical, socio-political, religious, and general intellectual, context that at least looks as different as that in which the medieval Islamic philosophers were working, and that within which analytic philosophy was forged, will be especially valuable. Notes 1 She cites Fine (2004) as a corroborating source. 2 I think this view is most apparent in Quentin Skinner’s famous 1969 paper (later a manifesto for the ‘Cambridge School’ of intellectual history) and its ­concomitant excoriations of the ‘rational reconstructive’ way of doing history of ­philosophy. While I have cited Dutant and Antognazza as providing work of a similar ­register, it should be noted that neither would likely consider themselves as ­members ­Cambridge School intellectual history. For example, Antognazza ­explicitly ­disagrees with Skinner’s claim that ‘there are no perennial problems in ­philosophy’ (see Antognazza 2015, p. 166) and holds that while that part of the value of history of philosophy is to provide such correctives, part of its value has to do with keeping our minds open to views outside an established orthodoxy. 3 I won’t consider Skinner’s view that analytic philosophy can at most provide us with a kind of Wittgensteinian therapy and expose philosophical problems as ‘pseudo-problems’. Since there are no ‘timeless truths’ for analytic philosophy to discover, according to Skinner, analytic philosophy as it is practised today has no real object of study. For Skinner then, all we can really aspire to is pedantry. I won’t discuss this view here, since the title of the question assumed that there is something meaningful in the way that analytic philosophy currently operates in such a way that the history of Islamic philosophy can benefit it. 4 According to Ibn Sīnā, we have direct voluntary control over our imagination. As Kaukua puts it: ‘As the only genuinely active faculty among the human ­cognitive capacities, it can function both under the governance of other faculties ­(estimation and intellection) and on its own’ Kaukua (2014, p. 237). 5 Corroborating this is the fact that Ibn Sīnā speaks of intellectual i­ntuitions as ‘sensata’ and that estimation works with ‘sensata’: ‘As ­regards those that are ­intuitively evident, they are like sensata (al-mahsusat), that is, propositions the assent to which we acquire from the sense, such as our judgment that the Sun exists and is shining or our judgment that ire burns, or like propositions relative to the intuitive evidence of faculties other than sense, for example our familiarity with [the fact] that we have a thought, that we are in fearor anger, or that we are aware of ourselves and the acts of ourselves’ Ibn Sīnā Isharat, VI, p. 56, cited in Kaukua (2014, p. 228). 6 The ‘common sense’, like al-wahm, and the imagination are what Ibn Sīnā calls the ‘internal senses’ – compared with the external senses such as visual ­perception. It’s a bit of an open question where incidental perception fits in since, as Kaukua puts it: ‘Avicenna takes the experience of incidental perception to be a case of perception only in a strongly qualified sense, for it relies heavily on faculties other than the senses’ (Kaukua 2014, p. 233). This resembles W ­ ittgenstein’s ­characterisation in his Philosophical Investigations of seeing-as as having the character of ‘an echo of a thought in sight’ (212e). 7 As Deborah Black puts it, it is that ‘coordinating faculty in the animal soul that can unite all the intentional and formal properties of an object together into the perception of a concrete whole’ Black (1993, p. 9).

Islamic Philosophy for Contemporary Analytic Philosophers  37 8 This is from his Remarks and Admonitions, cited in Black (1993, p. 65). About this point, Black says: ‘Avicenna’s comment on the absence of self-awareness in the estimative faculty here helps to explain further why the estimative sense is unable to rein itself in, and so leads to the sorts of errors with which Avicenna charges it. Reflectiveness, and therefore self-correction, are marks of the intellect alone, because only the intellect can perceive the non-sensible as such, and all faculties of the soul are, in themselves, non-material’ (Black 1993, p. 65). This does not mean, of course, that the workings of the estimation cannot be corrected – just that they are not corrected by itself and require the assistance of the intellect working with the imagination. 9 This accords well with Zarepour’s (2021) recent account of Avicenna’s ­empiricism about mathematical concepts where mathematical objects are specific ­‘connotational’ properties of physical objects that are grasped by estimation – in the language adopted in this chapter estimation is the sense that allows us to understand physical objects as mathematical. See also McGinnis (2017) for an account of the role of estimation in our grasp of mathematical concepts and in thought experiments. 10 That Ibn Sīnā mentions Prophetic vision in a discussion of al-wahm also evidences the idea that a priori truths and universals (intelligibles) can in some sense be the contents of al-wahm, since the ‘universal mind’ deals only with intelligibles. Though again with the caveat that it’s the relation between them that is the unique object of estimation. 11 See especially his Conditions on Certainty. 12 One way that understanding-as, or ‘aspectual thought’ as Camp calls it, is through what Camp calls ‘characterisation’. Fraser summarises the view helpfully as ­follows: ‘Aspectual thought involves using one’s characterization of one object to structure one’s thinking about another. A characterization is a structured set of properties. For example, Camp’s characterization of a quarterback includes the following properties: being natural leaders, being affable, dumb, shallow, and having a ready smile Camp (2017). My characterization of a computer science student includes properties like being shy, being good at maths. ­Characterizations structure properties along at least two axes: prominence and centrality. A ­feature is prominent (in a given context) to the extent that it is (i) diagnostic, that is, h ­ elpful in classifying an object and (ii) intense, that is, s­ alient, or prominent (Camp 2017). A feature of a characterization is central to the ­extent that I take it (implicitly) to explain other features of the characterization. When I use my c­ haracterization of a quarterback to structure my thinking about ­Daniel, the ­features of Daniel which ‘match’ with the characterization will matter more – and guide my thinking – about Daniel than do those which do not match the ­characterization (Camp 2017). Suppose that I use my characterization of a ­quarterback to structure my thinking about Daniel, who is both shy and affable. His affability will ‘stick out’ more than his shyness. By contrast, if I use my computer scientist characterization of Daniel, his shyness will stick out more than his affability’ (Fraser 2021, p. 4039).

References Antognazza, M. R. 2015: “The Benefit to Philosophy of the Study of its History”, British Journal for the History of Philosophy 23(1), pp. 161–184. Black, D. 1993: “Estimation (Wahm) in Avicenna: The Logical and Psychological Dimensions”, Dialogue 32, pp. 219–258. Booth, A. R. 2016: Islamic Philosophy and the Ethics of Belief (London: Palgrave Macmillan).

38  Anthony Robert Booth Camp. E. 2017: “Perspectives in Imaginative Engagement with Fiction”, P ­ hilosophical Perspectives 31(1), pp. 73–102. Clark, A. 2016: Surfing Uncertainty: Prediction, Action, and the Embodied Mind (Oxford: Oxford University Press). Dutant, J. 2015: “The Legend of the Justified True Belief Analysis”, Philosophical Perspectives 29(1), pp. 95–145. Fine, G. 2004: “Knowledge and True Belief in the Meno”, Oxford Studies in Ancient Philosophy 27, pp. 41–81. Fraser, R. 2021: “Narrative Testimony”, Philosophical Studies 178, pp. 4025–4052. Friedman, J. 2020: “The Epistemic and the Zetetic”, Philosophical Review 129(4), pp. 501–536. Gendler, T. 2008: “Alief and Belief”, Journal of Philosophy 105(10), pp. 634–663. Gettier, E. 1963: “Is Justified True Belief Knowledge?”, Analysis 23, pp. 121–123. Ibn Sīnā, A. A. 2013: Fil Ilm Al-Nafs translated by E. A. van Dyck, adapted by L. Bakhtiar as Avicenna on the Science of the Soul: A Synopsis (Chicago, IL: KAZI Publications). Jubien, M. 2001: “Propositions and the Objects of Thought”, Philosophical Studies 104(1), pp. 47–62. Kaukua, J. 2014: “The Problem of Intentionality in Avicenna”, Documenti e Studi sulla Traditzione Filosofica Medievale 25, pp. 215–242. McGinnis, J. 2017: “Experimental Thoughts on Thought Experiments in Medieval Islam”, in Stuart, M., Fehige, Y. & Brown, J. R. (eds.) The Routledge Companion to Thought Experiments (London: Routledge), pp. 77–91. Skinner, Q. 1969: “Meaning and Understanding in the History of Ideas”, History and Theory 8(1), pp. 3–53. Williams, D. 2018: “Predictive Processing and the Representation Wars”, Minds and Machines 28, pp. 141–172. Zarepour, M. S. 2021: “Avicenna on Grasping Mathematical Concepts”, Arabic ­Sciences and Philosophy 31, pp. 95–126.

PART II

Existence and Uniqueness of God

3 Kalām and Cognition1 Mahrad Almotahari

One initially compelling feature of the kalām cosmological argument is its apparent simplicity: 1 Everything that begins to exist has a cause for its existence. 2 The universe began to exist. 3 Therefore, the universe has a cause for its existence (namely, “The First Cause”). Many hundreds of pages have been written about this little argument in the last few decades alone, and there’s good reason to think the argument has been around since at least the 6th century CE (Shihadeh 2008, p. 206). The earliest formulation that I’ve studied appears in al-Ghazālī’s book, al-Iqtiṣād fī al-I’tiqād (Moderation in Belief). But its most influential contemporary defender is probably William Lane Craig.2 Not too long after Craig’s initial presentation of the argument, in 1979, it became the subject of a great deal of controversy in the Anglo-­American ­philosophical community. Despite its apparent simplicity, the argument raises ­ odality, all sorts of questions about time, infinity, explanation, causation, m and the relationship between The First Cause and the universe (that is, the totality of space, time, and matter). Endorsing both premises requires taking a stand on some of the hardest questions in metaphysics, cosmology, and the foundations of mathematics.3 So the argument’s apparent simplicity belies a great deal of complexity. This might lead one to think that, even if sound, informed rational skepticism is possible. After all, knowledge in the relevant areas of metaphysics, cosmology, and mathematics is sufficiently equivocal to sustain decades of expert disagreement; how then can I (not really an ­expert, not quite a layman) justifiably take a stand where only the wise tread with caution? This chapter is a plea for skepticism, but one that approaches the issue in, perhaps, a slightly unorthodox way: not by the familiar observation that expert disagreement is incompatible with a high degree of confidence, but by the application of some recent work in cognitive science and philosophy DOI: 10.4324/9781003327714-6

42  Mahrad Almotahari of language. Briefly put, my central claim is that the credibility of the kalām argument’s first premise rests on the generic overgeneralization effect— a ­fallacious mode of thought to which we all seem susceptible.4 Insofar as the rejection of (1) is judged to be self-evidently untenable, friends of the kalām argument are exploiting a stubborn cognitive vulnerability. Somewhat ­ironically, the claim’s enduring appeal undermines its epistemic standing. To be clear, my strategy isn’t to falsify premise (1), but to debunk it. This means telling an empirically plausible story according to which its ­apparent truth derives from an erroneous or unreliable form of thought. To that e­ xtent, my project is continuous with a growing body of work exploring the ways in which cognitive science might constrain metaphysical inquiry (Goldman 2015; Goldman and McGlaughlin 2019). A clear and concise statement of the animating thought behind this work appears in Paul (2010): “…one role for cognitive science in ontology is to identify places where our ordinary judgments might not be appropriately generated, as with illusions” (p. 470). In the discussion that follows, I’ll argue that the disposition to accept premise (1) probably isn’t, in the relevant sense, “appropriately generated”. Along the way, I’ll sketch a picture that’s manifestly coherent, empirically informed, and yet incompatible with the truth of (1). At no point will I argue that (1) is false. In a domain as contested as this one, it seems to me that ambitious deductive arguments are bound to be unconvincing. Better to sketch a picture and leave it to others to judge how appealing the picture is. I’ve said my strategy is to debunk premise (1). There is, of course, a m ­ assive literature in contemporary epistemology about the mechanics of ­debunking arguments. Important questions have been raised about how, ­ exactly, ­arguments of this kind are supposed to work.5 What are the ­underlying ­principles on which they rely? I’m unable to engage with this literature in any detail here. As a result, the case for my central claim will be somewhat ­provisional, but not at all unique in this respect. Philosophical inquiry is ­ typically provisional. Often, one is forced to bracket certain issues in order to address others. The hope is that, in the long run, things will work out. A striking feature of the contemporary debate about the kalām ­argument is that, overwhelmingly, participants focus on the second premise.6 For example, in their contribution to The Blackwell Companion to Natural ­ ­Theology, Craig and Sinclair (2012) provide a comprehensive 101-page ­defense of the kalām argument in which only about six or seven pages are devoted to its first premise. And in their even more recent two-volume ­anthology on the subject—a collection of essays that’s just under 700 pages in length—Copan and Craig (2018a,b) reserve only about 60 pages for ­critical examination of the premise (and even many of these pages are concerned with peripheral issues). The presumption seems to be that the first premise is just ­self-evidently true. I say this is a “striking” feature of the contemporary debate because some of the most influential Muslim thinkers of the classical period (both before and after al-Ghazālī) rejected the presumption that premise (1) is self-­evident. “Al-Juwaynī’s al-Shāmil, for

Kalām and Cognition  43 instance, ­testifies to the heated debates among the mutakallimūn about this ­ rinciple” (Erlwein 2017, p. 43). In the Metaphysics of his Shifā, Avicenna p claims that the ­principle requires proof.7,8 And Fakhr al-Dīn al-Rāzī seems to have ­suspended ­judgment regarding its truth altogether.9 In fact, according to Majid Fakhry, the kalām argument “was never viewed with favor in the Muslim world […] since the very validity of the causal principle is challenged by the Mutakallims” (1957, pp. 135–136).10,11 My aim in writing this chapter is to provoke a critical reexamination of premise (1). But that won’t be possible unless the claim’s hold over us is ­considerably weaker than it presently seems. So I want to encourage the ­somewhat skeptical frame of mind that was normal during the c­lassical period of Islamic philosophy. To reiterate, the central claim for which I’m arguing is that the credibility of the kalām argument’s initial premise issues from the generic overgeneralization effect. A natural place to begin the ­discussion, then, is with the science of generic thought. Once the idea of generic ­overgeneralization has been adequately unpacked (Section 3.1) and defended (Section 3.2), I’ll explain how it bears on the argument that ­interests me (Section 3.3). 3.1

The Idea of Generic Overgeneralization

Generic sentences are constructions like ravens are black, lions have manes, and sea-turtles are long-lived.12 These constructions are about the members of a certain genus or kind, but unlike sentences beginning with all, every, most, many, and some, generics lack an overt quantifier. They remain silent regarding how many members of the relevant kind have the property at issue; thus they express a distinctive form of generality that philosophers and ­linguists often call genericity. The meaning of some generics is roughly preserved by the insertion of an adverb like generally, typically, or normally (as in: generally, ravens are black). But some generics express significantly weaker generalizations. For example, sea-turtles are long-lived is intuitively true, though generally, sea-turtles are long-lived is false. A female sea-turtle will typically lay between 1,900 and 2,300 eggs in her lifetime, and only a handful will survive to ­maturity. Some generic sentences express generalizations that are so strong that even if every member of the relevant kind were a conformer, the g­ eneralization would still be false. Consider, for example, Supreme Court Justices have a prime Social Security number (Cohen 2012). It’s not part of the job description that ­members of the Court have prime Social Security numbers. As the examples in the previous paragraph suggest, one salient feature of a generic sentence is that its truth tolerates exceptions.13 But the exceptions can vary in prevalence. Only a minority of ravens are non-black (albinos, for example, are white); fewer than half of all lions lack manes (specifically, immature or unfortunate males, and just about every female); and almost all turtle hatchlings die shortly after birth. But these exceptions are, so to speak,

44  Mahrad Almotahari harmless. The truth of the corresponding generic sentence remains intact. There’s no pressure on us to give them up or to soften our commitment to them once we acknowledge the exceptions. Generics raise many fascinating questions for philosophers, linguists, and cognitive scientists. One particularly important question is, what does their truth demand of the world? When is an exception harmless and when is it a genuine counterexample? More abstractly, what does genericity consist in? (Compare: what does universality consist in?) One would like a systematic answer, an answer that’s not terribly disjunctive or piecemeal. But it’s not clear how to simultaneously account for the strength of the strong generics and the weakness of the weak ones. They seem to push and pull in o ­ pposite directions. Expert opinion is largely divided, even about the a­ppropriate method with which to pursue the inquiry.14 It’s no wonder, then, that a ­uniform theory remains elusive. Interestingly, young children understand generic sentences before they develop an understanding of every F is G and some F is G (Leslie 2007, 2008). One reason why this pattern of acquisition is interesting is that, from a theoretical perspective, universal and existential quantifiers are far more tractable. One can easily specify their meanings in simple set-theoretic terms. (Every F is G is true if and only if the set of Fs is a subset of the set of Gs; some F is G is true if and only if the set of Fs intersects the set of Gs.) Here is how Leslie puts the point: “While generics have proved nigh i­ntractable for theorists, they turn out to be a cakewalk for language learners. Young ­children engaged in learning their native language grasp and produce ­generics far more quickly and readily than they do explicitly quantified sentences” (Leslie 2008, p. 2). Another reason why this pattern of language acquisition is interesting is that generics lack an overt marker. There’s no explicit generic quantifier akin to all or some.15 The absence of a dedicated generic operator ought to make the acquisition of generics more difficult. Children are not even provided with an explicit object of study; there is no part of the sentence with which they could learn to associate the meaning of generics. Quantifiers like ‘all’ and ‘some’ are not only truth-theoretically simpler; they are actually articulated! (p. 2) In general, children find it massively difficult to associate meanings with absences (p. 19, fn. 13). Even still, children start using generics by the time they’re two years old, which is much earlier than their use of all and some (Gelman 2003).

Kalām and Cognition  45 How might we explain these remarkable facts? Here’s one influential hypothesis: generic sentences are the language faculty’s way of articulating an innately given and default system of generalization. That is, the cognitive system may have an automatic, early-­developing way of generalizing information from individuals to kinds. These ­primitive kind-based generalizations are, according to this ­hypothesis, later articulated in language as generics. If correct, this hypothesis would explain why generics are understood and produced by young children, despite the semantic complexity that linguists have claimed generics exhibit. The generalizations expressed by [‘all’ and ‘some’], in contrast, ­ rimitive default represent more sophisticated generalizations—not the p one expressed by generics. (Leslie et al. 2011, p. 16)16 This hypothesis derives considerable support from its predictive success. “For example, if understanding [‘all’ and ‘some’] statements requires d ­ eviating from the default mode of generalization, then both children and adults should sometimes fail to execute this deviation, and so should incorrectly treat [‘all’ and ‘some’] statements as generics” (p. 16). The idea behind this prediction is that default patterns of thought are automatic and, therefore, easier to engage in than nondefault patterns of thought. If just the right sort of pressure were introduced in context, a subject might lighten the cognitive load for himself by regressing to the default. One relevant sort of pressure that Leslie and her collaborators discuss is the difficulty in assessing universal generalizations. It may be harder, though, for people to confirm that a universal s­ tatement is true, since they are only true if every single member of the kind has the property in question. Adults may thus save cognitive effort by relying on the generic to evaluate the universal. (p. 17) This sort of difficulty is only compounded when the universal ­generalization is as abstract and sweeping as the general premise in a cosmological ­argument. Be that as it may, the prediction to which I’m drawing attention is largely confirmed. Hollander et al. (2002) and Tardif et al. (2012) found that young children are highly disposed to treat sentences beginning with all and some as generics. More relevant for my purpose, researchers have observed statistically significant patterns of the error among informed adults, as well (Jönsson and Hampton 2006; Connolly et al. 2007; Khemlani et al. 2007; Leslie et al. 2011; Leslie and Gelman 2012).

46  Mahrad Almotahari The take-away point is this: competent language users are inclined to agree with some false universal generalizations when the corresponding generic sentences are true, despite knowing about the falsifying counterexamples (Leslie et al. 2011, pp. 17–18). Manifestations of this widespread tendency exemplify the generic overgeneralization effect. Sorensen describes the effect as follows: “Once you pick up a generic generalization you tend to wield it nearly as strongly as a universal generalization” (2012, p. 447). Recent examination of the effect raises doubts about whether its experimental confirmation is conclusive, suggesting possible contextual ­ interference (Lazaridou-Chatzigoga et al. 2017, 2019). However, even more recent experimentation vindicates the initial findings of Leslie and her ­collaborators, showing the effect to be robust in a wider range of cases, even cross-linguistically (Karczewski et al. 2020; Wajda and Karczewski 2020). Further investigation is required, but the results to date are, at the very least, highly suggestive. At any rate, it’s not clear that friends of the kalām ­argument should be relieved by the findings of Lazaridou-Chatzigoga and her colleagues, since even they indicate that subjects have a tendency to ­misinterpret universal generalizations, softening them so that all and every are roughly equivalent to almost all and just about every. My central claim can be reformulated accordingly. Leslie (2007, pp. 394–397) persuasively embeds her theory of generic ­cognition in the dual-systems framework of Tversky and Kahneman (1982) and Kahneman (2012). This influential framework says that our ­judgments issue from two different systems of thought. System 1 is the default, a­ utomatic, effortless, unreflective, lower-level system that’s largely screened off from one’s total body of evidence. It delivers firmly held and difficult-to-abandon intuitions. System 2 is the slower, effortful, conscious, evidence-based, higher-level system. It can yield strongly counterintuitive judgments. Naturally, the two systems often compete. Kahneman illustrates this with riddles. “A bat and ball cost $1.10 in total. The bat costs $1 more than the ball. How much does the ball cost?” Most of the undergrads surveyed at Harvard and MIT say $.10. The correct answer is $.05, since (i) Bat + Ball = 1.10, (ii) Bat = Ball + 1, and therefore (iii) (Ball + 1) + Ball = 1.10. But many people continue to feel the strong pull of the intuitive yet incorrect answer even after the calculation demonstrates its falsity. Erroneous System 1 thinking is highly seductive; correcting it requires training. And since there’s reason to believe that generic cognition is under the control of System 1, we should expect some manifestations of the overgeneralization effect even after the corresponding universal generalization is judged to be false or unacceptable. As if all this weren’t bad enough, the epistemic impact is made worse by a troubling psychological regularity: “Generic statements are often judged true on weak evidence” (Cimpian et al. 2010, p. 1477). So, by shifting to a default generic interpretation, we put ourselves in a position to accept a

Kalām and Cognition  47 universal generalization on weak grounds and at the same time shield the resulting belief from falsification by counterexample. To adapt a phrase from al-Ghazālī, the generic overgeneralization effect reveals immoderation in belief. Interestingly, the immoderate character of generic thought was anticipated in the mid-sixties. Abelson and Kanouse observed that generic sentences are often accepted on the basis of very little evidence, but “once accepted psychologically they appear to be commonly taken in a rather strong sense, as though the quantifier always had implicitly crept into their interpretation” (1966, p. 172; citation in Sterken 2015a, p. 80; Leslie 2017, p. 398). 3.2

In Defense of the Idea

Support for the generic overgeneralization effect is both experimental and ­theoretical. The experimental results confirm its reality, but the c­ orresponding theory makes its reality intelligible. Unintelligible experimental results are vulnerable to doubt. But, with Leslie’s theory in view, overgeneralization is precisely what we should expect, since the comprehension of a n ­ ondefault generalization requires the interpreter to override or inhibit her innately given default pattern of thought. For a conceptual system to perform a nondefault generalization, such as a universal one, it must be explicitly commanded to do so. Absent such instruction, the conceptual system simply  employs its default. …Inhibitory processes, by their very nature, are more taxing for the  ­conceptual system to implement than noninhibitory ones. As the demands on the conceptual system are increased, it becomes more likely that it will be overtaxed, and so fail to implement the inhibitory process, and revert to the default. (Leslie 2008, pp. 23–27) The case for Leslie’s theory of generics relies on several different strands of argumentation. The one that I emphasized in the previous section began with the observation that generic sentences vary in strength: some appear to be about as strong as a universal; others, almost as weak as an existential. Cases of the latter sort are particularly difficult to handle. They seem to falsify even the most sophisticated and promising semantic theories. Leslie calls them “troublesome” generics, and she distinguishes between two basic types, “A” and “B”. In this section, I’ll focus entirely on type B generics—those that attribute a potentially dangerous, appalling, or striking property, e.g., sharks attack bathers, mosquitoes carry the West Nile virus, and rottweilers maul children. To streamline the discussion even further, I’ll restrict my attention to the example about mosquitoes. Other type B generics raise fundamentally the same issue, and the example about mosquitoes has achieved a kind of celebrity status.

48  Mahrad Almotahari Less than 1 percent of mosquitoes carry the West Nile virus, but ­mosquitoes carry the West Nile virus is intuitively true. Why is that? According to Leslie (2008), It should be evident upon reflection that the criteria that govern t­roublesome generics reflect our psychology. This is especially evident in the B group: the more striking, appalling, or otherwise gripping we find the property predicated in the generic, the more tolerant the generic is to exceptions. …If the above classification of troublesome generics is appropriate, then this suggests that our goal of understanding the quirky behavior of generics might be furthered by looking carefully into some of the relevant psychological literature. (pp. 15–16) Leslie’s proposal, as I understand it, is that our default mode of ­generalization facilitates our intuitive assessment of a generic. If the generic predicates a “striking” property, then that property’s infrequent manifestation among members of the relevant kind may not matter. I say it “may” not matter because there’s more to the story. Although mosquitoes carry the West Nile virus is true, animals carry the West Nile virus is false. Leslie is aware of this …we do not accept ‘animals carry the West Nile virus’ or even ‘insects carry the West Nile virus’, even though both kinds have some members that carry it (namely those few mosquitoes). …The mechanism [that facilitates our semantic evaluation], I suggest, looks for a good ­predictor of the property in question; it avoids generalizing to overly broad kinds or to irrelevant kinds. In particular, for a kind to be the locus of a striking-property generalization, it seems that the members of the kind that lack the property must at least be disposed to have it. (p. 41, Leslie’s emphasis) In light of this complication, Leslie proposes that a type B generic, Ks are F, is true if: (i) the exceptions are negative17 and (ii) if being F is striking, then some Ks are F and the others are disposed to be F (p. 43). This is both a ­metaphysical proposal about the truthmaker for such generalizations and a psychological theory about the way in which our default mode of ­generalization functions. One might wonder, what accounts for this happy convergence between mind and world? My hunch is that Leslie would deny that genericity is, as it were, an altogether “worldly” phenomenon. That is, according to her theory, the principles that normally control our most primitive system of generalization constitute genericity; they don’t merely track the phenomenon.18 Unlike ­universality, genericity is fundamentally psychological. The central point for my purpose is that troublesome type B generics appear to create insurmountable problems for every theory of generics except Leslie’s. To that extent, they provide support for the theory that generics

Kalām and Cognition  49 express a psychologically primitive mode of generalization, because they make the acquisition problem harder to solve. The harder it is for alternative theories to solve the problem, the more plausible Leslie’s theory will be. In short, type B generics motivate an understanding of genericity in psychological terms. Remember: “It should be evident upon reflection that the criteria that govern troublesome generics reflect our psychology. This is especially evident in the B group…” (op. cit.).19 But this line of thought has been forcefully criticized in recent years (Sterken 2015a,b, 2017). If the criticism sticks, then one might begin to doubt the generic overgeneralization effect, since its intelligibility rests on Leslie’s psychological account of the subject.20 And, as Sterken is aware, type B generics are “plausibly the best evidence for a psychologically based ­theory” (2015b, p. 2503). So I’d like to quickly present the criticism and explain why it fails. Actually, the upshot will be slightly more interesting: the criticism to which I’ll respond inspires a line of thought that ultimately ­supports the theory it’s meant to undermine.21 Sterken’s criticism relies on conjunctions of the following sort: 4 #Mosquitoes carry the West Nile virus, but typically they don’t.22 Sentences of this sort sound contradictory. But if the initial conjunct were nearly as weak as an existential, one would expect them to be coherent. Consider Leslie’s analysis: 5 Some mosquitoes carry the West Nile virus, and the rest are disposed to, but typically mosquitoes don’t carry the West Nile virus. It sounds fine. Sterken suggests that (4) sounds contradictory because it is contradictory (2015a, p. 83,b, pp. 2508–2509, 2017, p. 9). The initial conjunct expresses a generalization that’s basically equivalent to the claim that mosquitoes ­typically carry the West Nile virus. And this generalization is straightforwardly denied in the second conjunct. Furthermore, since mosquitoes carry the West Nile virus is basically equivalent to mosquitoes typically carry the West Nile virus, our intuitive assessment of its truth is mistaken. Far too few mosquitoes are carriers of the virus for the typicality generalization to be true. Similar reasoning applies to all type B generics. If the argument succeeds in one case, then plausibly it succeeds in all.23 However, if (4) were a genuine contradiction, as Sterken suggests, then we should be unable to eliminate its inconsistency by simply adding more information. In general, one can’t achieve coherence by supplementing p ∧ ~p with q. But the apparent inconsistency of (4) is eliminable in precisely this way: 6 Mosquitoes carry the West Nile virus, but because the overwhelming majority inhabit circumstances that fail to trigger their disposition to do so, they typically don’t.

50  Mahrad Almotahari This sentence is perfectly fine. There are other contexts in which (4) is coherent. Think of a setting in which someone tries to soothe her aerophobic friend’s anxiety before an upcoming flight. She does so in a tactful way, acknowledging the validity of her friend’s fear but communicating that the risk is negligible. She says, 7 Air travel is a lot safer than it used to be, but you’re right: airplanes still crash. Though, typically, they don’t. This sentence seems completely unproblematic. And it’s quite easy to adapt the general form of this example to fit (4). Suppose you’re trying to soothe your anopheliphobic friend’s anxiety before he heads to a swamp teeming with mosquitoes. You don’t want to be a jerk about it, so you make sure to acknowledge the validity of your friend’s fear. But you also want to unequivocally assure him that the risk is low. So you say, 8

Mosquitoes aren’t as dangerous as we thought, but you’re right: they still carry the West Nile virus. Though, typically, they don’t.

Again, this seems completely unproblematic. The felicity of (6)–(8) can’t be squared with the hypothesis that all type B generics are “close in meaning” (Sterken’s phrase) to a corresponding ­typically sentence. But there’s more. Consider: 9 a. Bob might be in his office; in fact, he must be. b. #Bob must be in his office; in fact, he might be. The contrast between the felicity of (9a) and the infelicity of (9b) is evidence that must is logically stronger than might. One can easily qualify an utterance by saying something stronger; but it’s very difficult to qualify an utterance by weakening what one says (Horn 2021; cf., Lewis 1979). Now, for the sake of argument, assume that 90 percent of mosquitoes carry the West Nile virus and consider the following pair of sentences: 10 a. Mosquitoes carry the West Nile virus; in fact, they typically do. b. #Mosquitoes typically carry the West Nile virus; in fact, they do. The relationship between genericity and typicality in (10) patterns with might and must in (9), as Leslie’s theory predicts.24 Sterken’s hypothesis (that mosquitoes carry the West Nile virus is basically equivalent in meaning to mosquitoes typically carry the West Nile virus) implies that (10a) and (10b) shouldn’t significantly differ with respect to acceptability. But they clearly do.

Kalām and Cognition  51 3.3

An Application of the Idea

Recall the first premise of the kalām argument: everything that begins to exist has a cause for its existence. The falsity of this premise is compatible with the truth of the corresponding generic: 11 Things that begin to exist have a cause for their existence. Furthermore, the falsity of (1) is compatible with the necessitation of (11). 12 Necessarily, things that begin to exist have a cause for their existence. I emphasize these points because Craig defends the first premise of the kalām argument by highlighting implausibilities that follow only from a rejection of (11) and (12). For example, Mackie (1982, p. 94) famously questions the first premise of the kalām argument. Craig’s response to Mackie is telling: Does Mackie sincerely believe that things can pop into existence uncaused, or out of nothing? Does anyone in his right mind really believe that, say, a raging tiger could suddenly come into existence uncaused, out of nothing, in this room right now? (2008, p. 113) Similar rhetoric is used in Craig and Sinclair (2012, p. 182) and Craig (2018, pp. 15–16).25 But Mackie’s doubts about (1) are compatible with the generic truth that things can’t pop into existence uncaused; for the belief that things can’t spontaneously pop into existence is merely (12). Nobody who denies (1) is committed to the possibility of uncaused tigers, nor to widespread cases of something coming from nothing. All one would be committed to is a ­single (and singular) exception to (11), namely, the origin of the universe.26 And, as we’ll see momentarily, there’s a compelling reason (internal to the ­materialist’s worldview) to think of it as an exception. Why might Craig exaggerate the consequences of rejecting premise (1)? For rhetorical effect? –Perhaps. But overblown rhetoric is just as likely to be counterproductive, and a polemicist as skilled as Craig knows that. His ­rhetoric isn’t meant to be hyperbolic; it’s meant to be proportional. So then why does he exaggerate? The generic overgeneralization effect provides a plausible explanation: on Craig’s implicit interpretation of premise (1), the sentence is assigned a generic reading such that its rejection would be ­tantamount to a rejection of (11). And a rejection of (11) would license the kind of rhetoric that Craig deploys, since it would involve repudiating what we all know to be the case: that there’s a large-scale pattern/regularity ­impeding spontaneous, uncaused generation. This point deserves emphasis. Craig has been thinking and writing about the kalām argument for nearly

52  Mahrad Almotahari half a century. Probably no one has spent as much time as he has evaluating its merits. If his understanding of premise (1) is susceptible to the generic overgeneralization effect, how much more susceptible to it must the neophyte be who reads his work as an introduction to the subject? Craig isn’t unique. Prominent Muslim philosophers have indulged in similar forms of exaggeration while discussing the kalām argument’s ­initial premise. For example, al-Kindī acknowledges just two possibilities: either the general premise of the kalām argument is true or things pop into e­ xistence by chance (Erlwein 2017, p. 43 and fn. 76). If you take this to be a reasonable assessment of the logical landscape, that’s only because you’ve erroneously confused the kalām argument’s initial premise with (11). And c­ onfusions of this sort are to be expected. They’re manifestations of the generic overgeneralization effect. In a moment, I’ll present another contemporary instance of this effect at work. Still, one might think that my reasoning is objectionably ad ­hominem. But that would be a mistake. My reasoning provides a charitable i­nterpretation of a demonstrable error in Craig’s criticism of Mackie. The rejection of (1) doesn’t require the rejection of (11); so Craig is just wrong to impute c­ ertain exaggerated consequences to his interlocutor. This raises an i­nterpretive ­ ­question: why does Craig make this mistake? It’s only the attribution of ­inexplicable or brute error that’s uncharitable (Lewis 1974). By explaining Craig’s mistake in terms of the generic overgeneralization effect, I’m ­interpreting him in a way that coheres with the demands of charity. Charitable interpretation is not objectionably ad hominem. Consider this analogy: intuitive assessments of reasoning with ­universal quantifiers often fail to conform to the principles of logic. For example, Johnston and Leslie report studies in which adult participants judge that argument (A) is stronger than argument (B). A All animals use norepinephrine as a n ­eurotransmitter; ­ therefore, all ­mammals use norepinephrine as a neurotransmitter. B All animals use norepinephrine as a ­neurotransmitter; ­therefore, all ­reptiles use norepinephrine as a neurotransmitter. These naive evaluations are just plain wrong, according to classical logic. But they make sense on the assumption that participants are interpreting the universal premises as generic claims: …‘animals use norepinephrine as a neurotransmitter’ can be true even if some animals are exceptions to the claim. If one judges then that reptiles are more likely than mammals to be exceptions to the generic, then argument (A) is indeed stronger than argument (B). Hence these results are as one would expect if adults have a tendency to evaluate universals as generics. (2012, pp. 126–127)

Kalām and Cognition  53 My interpretation of Craig is no more problematic than Johnston and Leslie’s interpretation of the subjects who participated in these experiments. But I think it would be unfair to leave the matter here and simply move on. In the context of addressing doubts about premise (1), Craig raises an important challenge that deserves a response from anyone who, like me, flirts with the idea that the origin of the universe is a harmless exception to (11). He asks, Why is it only universes that can come into being from nothing? What makes nothingness so discriminating? There cannot be anything about nothingness that favors universes, for nothingness does not have any properties. Nor can anything constrain nothingness, for there is not anything to be constrained. (Craig and Sinclair 2012, p. 182) Craig reiterates this challenge in his more recent defense of the kalām a­ rgument (Craig 2018). Something ought to be said here, and the answer had better not rely on the nature of nothingness. But it might plausibly rely on a feature of universes: perhaps it’s ­essential to causation that a universe—a maximally specific distribution of matter over space and time—is a condition for its very possibility. More concisely: no ­ universe, no spatiotemporal manifold; no spatiotemporal manifold, no ­causation. That seems like a thought perfectly suited to a materialist ­worldview, the sort of worldview that friends of the kalām argument ­typically want to undermine. In fact, Kim (2003) presents an ingenious argument for roughly this conception of causation. As one might expect, the argument has met with some resistance (see, for example, Bailey et al. 2011), but my claim here isn’t that one is rationally obliged to accept Kim’s conception of ­causation; my claim is far more modest: if causation requires a system of space, time, and matter for its very possibility—a thesis for which Kim’s ­argument provides independent (if inconclusive) motivation—then there’s hardly anything arbitrary or ad hoc about that very requirement being an exception to the demand for causal explanation. A somewhat pleasant ­feature of this response, from a dialectical point of view, is that it relies on nothing more than the familiar strategy that the cosmological arguer relies on to avoid an infinite regress of causes: the nature of The First Cause precludes It from requiring a causal explanation because It cannot not exist. Both the cosmological arguer and I suggest that, because something has a nature of a certain sort, the demand for a specific sort of causal account doesn’t arise. But I’m not saying that no explanation can be given for the existence of the universe. Perhaps the explanation is probabilistic,27 or perhaps it takes some form still awaiting discovery. I don’t know. Does anyone? I suspect that Craig and Sinclair’s challenge—“Why is it only universes that can come into being from nothing?”—is really an assertion disguised as

54  Mahrad Almotahari a question. It seems that what they’re really saying is that a rejection of the kalām argument’s first premise would involve a commitment to ­something far too arbitrary to be credible: that everything except for the biggest thing there is requires a cause for its existence.28 In the previous paragraph, I ­suggested that there’s a better way to think about the position. But if Craig and Sinclair’s point is about the arbitrariness of rejecting premise (1), then their case is fundamentally no different from the sort of rationale one finds for the induction step of a sorites argument. This is noteworthy because the induction step appears to exploit the generic overgeneralization effect. People deduce that there must be an exception to the induction step of the sorites: Base step. One is a small number. Induction step. Small numbers have small successors. Conclusion. A billion is a small number. Yet they are reluctant to concede that the notional exception constitutes refutation of the induction step. They confidently apply the ­generalization to any number you pick. Recent work on the ‘Generic Overgeneralization Effect’ suggests a psychological explanation of this loyalty to the induction step. Although the propounder of the sorites paradox intends the induction step to be a universal generalization, hearers assimilate universal generalizations to generic generalizations (for instance, ‘All birds fly’ tends to be remembered as ‘Birds fly’). Most generic generalizations permit ­exceptions—especially when those exceptions are rare, abnormal or difficult to imagine. Any counter-example to the induction step will have all of these features. After all, sorites arguments are crafted to ensure a smooth ride down the slippery slope. (Sorensen 2012, p. 444) The ride is made smooth by the thought that any stopping point would be implausibly arbitrary or ad hoc. But this form of argument trades on our ­vulnerability to overgeneralize. Any exception to (11) will also be rare, abnormal, and difficult to imagine. Our expectations about how things ­originate have been conditioned by both evolution and a lifetime of personal experience with “medium-sized” objects. So it appears no less likely, given our psychological propensities, that we’re apt to assimilate premise (1) to principle (11). If the appeal of premise (1) is, at least partially, due to the ­overgeneralization effect, then we should expect people who deny it to continue to feel its pull even after they acknowledge its probable falsity. And this expectation is ­easily confirmed. Despite his doubts concerning the first premise of the kalām argument, Mackie confesses that he can’t help but find it attractive: “Still, this [causal] principle has some plausibility, in that it is constantly confirmed in our experience (and also used, reasonably, in interpreting our experience)”

Kalām and Cognition  55 (1982, p. 89). Mackie’s ambivalence is intelligible: unearned c­redibility is assigned to a false or objectionable universal generalization because it’s ­automatically interpreted as a true or plausible generic. The cosmological arguer might respond to all of this with puzzlement: Okay, but so what? I mean, can’t I easily skirt your objection by ­simply reformulating premise (1)? What if I say instead that nothing that begins to exist does not have a cause for its existence? This claim is l­ogically equivalent to (1) but not universal in form. So there’s no danger of ­erroneously treating a generic as if it were a universal.29 But this response is based on a misunderstanding of the triggering ­conditions for overgeneralization. The presence of a universal quantifier isn’t necessary. If interpreting or evaluating a generalization (whether universal or not) is a sufficiently demanding task, then one’s conceptual system is likely to fail at implementing the inhibitory process that suppresses the default generic interpretation. After all, inhibitory processes are more ­taxing than default ones (compare how easy it is to conform to a habit with how hard it is to eliminate one). “As the demands on the conceptual system are increased, it becomes more likely that it will be overtaxed, and so fail to implement the inhibitory process, and revert to the default” (Leslie 2008, p. 27).  Sufficient ­cognitive strain results in a failure to curb the default r­ eading and thus ­triggers ­overgeneralization. Now, the claim that nothing that begins to exist does not have a cause for its existence is manifestly more difficult to parse than the generalization that (1) expresses. (The interpretation of every F is G is less demanding than it’s not the case that some F is not G. Presumably, that’s why the kalām argument is canonically formulated in terms of a ­universal ­quantifier.) So the claim with which the cosmological arguer ­proposes to replace (1) isn’t less likely to exploit our susceptibility to generic ­overgeneralization; it’s actually more likely to do so. At this stage, one might reasonably wonder whether the conclusion of the kalām argument receives enough support from (11) and (2) to be rationally compelling.30 After all, isn’t the inference from ravens are black, and Rupert is a raven, to Rupert is black a reasonable one to draw, even though it’s less than fully conclusive? This line of thought rests on a faulty analogy. An apt analogy with (11) would rely on a generic sentence that specifies a temporally sensitive relation between ravens, just as (11) specifies a t­ emporally sensitive relation between things that begin to exist and their causes. So the ­analogue generalization would be something like this: ravens are the i­mmediate ­offspring of ravens. And the analogue of (2) would, presumably, be the claim that membership in the kind raven began with an original specimen—call her Rowan. Given these claims, how rationally compelling is the conclusion that Rowan is the immediate offspring of ravens? I submit that it’s not at all compelling, because it’s incompatible with our assumption that Rowan is the original raven. With this assumption held fixed, the proper conclusion to draw is that Rowan is a harmless exception to the generic truth that ravens

56  Mahrad Almotahari are the immediate offspring of ravens. For many of us, the kalām argument’s initial premise suffers from a parallel issue: the erroneous demand to give a causal account of the condition for the possibility of causation. The cosmological arguer is unlikely to concede. She might defend the ­initial premise of the kalām argument (or something in the vicinity) on a priori grounds. Here’s my best attempt on her behalf. Lions are birthed; sea-turtles hatch; and stars are formed when ­massive clouds of gas coalesce. But what about the members of a much more ­ general kind? How do contingent beings characteristically originate? One answer immediately suggests itself: contingent beings originate by means of ­causation. But if we think of a cause for the existence of something as that on which its being is contingent, then our answer seems to be a ­conceptual truth: ­contingent beings originate by means of that on which their being is ­contingent. How could something be a contingent being without being contingent on something? Isn’t that like being a mother without being the mother of anyone? And isn’t that inconceivable? Surely every mother is someone’s mother. Likewise, because our reasoning depends on nothing more than the nature of contingency per se, one might think it obvious that every contingent being is contingent on something. On what, exactly? –On that by means of which it originates. Equivalently, every contingent being originates by means of causation. We then obtain premise (1) by restricting the domain of this conclusion to a subset of temporal contingents, specifically, those that have a beginning. This line of thought appears to vindicate al-Ghazālī’s view that premise (1) is an a priori truth. But, as compelling as it may be, one can plausibly challenge the assumption that the relationship between being contingent and being contingent on something is aptly modeled by the relationship between being a mother and being the mother of someone. A genuinely ­random state of affairs (such as the result of radioactive decay) is, presumably, contingent but uncaused; therefore, it’s not in the relevant sense contingent on anything. Perhaps there are no genuinely random states of affairs. It doesn’t really ­matter. What matters is that there might have been. Random occurrences aren’t ruled out by the nature of contingency. If contingency is a kind of modal instability, then random occurrences are among the most c­ ontingent of all contingencies. So a better model of the relationship between being ­contingent and being contingent on something is given by the relationship between being precious and being precious to someone. Being precious doesn’t require being precious to anyone. I don’t know of a promising a priori argument for (1). Insofar as it c­ ompels assent, it does so because it seems to be incompatible with our experience of causal order. That, at any rate, is what the cosmological arguer often ­emphasizes—the absence of things “popping into existence” (Pruss 2009, p. 125). But this observation is compatible with the denial of (1). Why might one think otherwise? At this stage, the answer hardly needs repeating.

Kalām and Cognition  57 My central claim has been that the plausibility of (1) issues from a ­ ersistent cognitive defect. This suggestion obviously bears on a larger issue p that I won’t pursue here, but that I hope receives attention from o ­ thers. The first premise of the kalām argument unmistakably resembles the Principle of Sufficient Reason (PSR), namely, that all truths (facts, states of affairs, etc.) have an explanation. It is, apparently, a restricted version of the PSR—one that concerns existential facts of a certain sort and their causal explanation. Interestingly, the PSR seems to be enjoying a revival (Pruss 2006; Della Rocca 2010; Amijee 2021). Could its plausibility issue from the same ­overgeneralization effect? Shouldn’t this not-so-far-fetched possibility by itself make us more skeptical of the PSR’s enduring pull? I’ll leave these questions hanging, but it’s worth considering one influential author’s remarks on the matter. ­ nreflectively According to Pruss, “Many of those who accept the PSR do so u because they take the PSR to be self-evident. I do not think there is any good argument against the propriety of doing so” (2012, pp. 26–27). Unreflective acceptance is a hallmark of System 1 cognition and generic overgeneralization. That should give us pause. Pruss continues: “If the PSR were false, we would expect a profusion of events that would not appear to fit into any kind of nomic causal order” (p. 32). This statement resembles the ­exaggerated rhetoric we observed in Craig’s defense of the kalām argument, and it’s no more reasonable. Those of us who happen to be suspicious of the PSR would agree that, generically, truths are explicable (GPSR), and perhaps even ­necessarily so. But what follows from this? As before, I don’t think Pruss is deliberately exaggerating. He’s far too careful a philosopher. His rhetoric is intelligible on the assumption that he’s assigning the PSR an interpretation that’s equivalent to the GPSR. And rejection of the GPSR would have the consequences he describes. Pruss defends unreflective acceptance of the PSR by analogy: We are perfectly within our epistemic rights to accept the Law of Excluded Middle (LEM), namely the claim that for all p we have p or not-p, because of the self-evidence of LEM, without needing any ­further argument for it. (p. 27) But there seems to be an important difference between cosmological ­principles like the PSR and logical truths like the LEM. Logical truths hold by dint of their form. As Quine memorably put the point, “Logic chases truth up the tree of grammar” (1986, p. 35). And the logically relevant form of a ­representation is traditionally understood to be transparent to res cogitans. What this transparency amounts to is notoriously unclear, but one might plausibly interpret the notion so that the traditional understanding of logical form underwrites a presumptive entitlement to the LEM. I know of no good

58  Mahrad Almotahari reason to think that cosmological truths (in particular, the PSR) should be similarly transparent to the mind. I believe the proposition that there’s a First Cause is an empirical h ­ ypothesis whose truth ought to be evaluated by a global assessment of t­ heoretical costs and benefits. What I’ve argued should be understood in the light of this ­general method. The price of denying the principle of universal causality (that everything that begins to exist has a cause for its existence) is nowhere near as costly as it initially appears. If it seems incompatible with the o ­ bservable causal order of the world, that’s only because it’s been c­ onfused with (11). This sort of confusion is prevalent and explicable: it results from generic overgeneralization. Notes 1 I’d like to thank Fedor Benevich, Hannah Erlwein, Sam Fleischacker, Sarah-Jane Leslie, Alex Malpass, Hashem Morvarid, Mahmoud Morvarid, Brian Rabern, Patrick Todd, and Muhammad Saleh Zarepour for their helpful feedback. 2 The argument in al-Iqtiṣād is formulated in a way that reveals Avicenna’s ­influence. Perhaps it’s a bit misleading, then, to classify al-Ghazālī’s argument as an early version of Craig’s. Nothing important for my purpose turns on this question. Thanks to Fedor Benevich for bringing the issue to my attention. 3 The papers collected in Copan and Craig (2018a,b) demonstrate just how difficult these questions really are. 4 See Leslie et al. (2011), Johnston and Leslie (2012), and Leslie (2017) for ­discussion. I’ll have a lot more to say about this work momentarily. 5 White (2010) is an illuminating contribution to this debate. 6 Several recent discussions of the second premise that I find particularly i­nteresting are Malpass and Morriston (2020), Malpass (2022), Zarepour (2021), and ­Zarepour (2022). 7 Book 1, Chapter 1, p. 6 of the Marmura translation; though the translation in ­Erlwein (2017, p. 44) is a bit more lucid: “It is not self-evident (bayyinan binafsihi), even if close to being self-evident to the mind, that originated things have some kind of principle for them”.  In correspondence, Muhammad Saleh Zarepour suggests an alternative: “It is not self-evident, even if close to being self-evident to the mind, that temporally originated things have some kind of ­originating principle”. 8 Remarkably, in Ishārāt V.9, Avicenna seems to maintain that causes whose effects are eternal (al-ibdā‘) are “nobler in rank than” (Inati 2014, p. 139) or “superior to” (Hassan 2020, p. 73) causes whose effects begin to exist. This suggests that, from Avicenna’s point of view, the kalām argument represents The First Cause as less than maximally perfect.  Though, in correspondence, Muhammad Saleh Zarepour has suggested an alternative reading of Ishārāt V.9. 9 See Shihadeh (2008, pp. 206–207). Erlwein (2017) expresses a somewhat ­different opinion. She suggests that al-Rāzī wasn’t agnostic about the principle’s truth, but maintains that he “evidently does not belong to those who consider the validity of the principle of causation a self-evident matter…” (p. 44). 10 See, also, Chittick (1998, pp. 16–20) and Hallaq (1991, pp. 49–56) for relevant discussion. A mutakallim (pl. mutakallimūn) is a practitioner of kalām, the ­Islamic tradition of systematic theology. 11 According to al-Ghazālī, the initial premise of the kalām argument seems to be analytic: “The one who is not moved by [the claim, ‘The occurrence of every

Kalām and Cognition  59 ­ ccurrent has a cause’,] is, perhaps, not moved because it is unclear to him o what we intend by the term ‘occurrent’ and the term ‘cause’. If he understood them, his mind would necessarily believe that every occurrent has a cause” (alIqtiṣād, p. 28). As this passage indicates, al-Ghazālī treats belief in the premise as a ­precondition for understanding it. A natural thought in the offing is that, if one can’t so much as understand a claim without believing it, then the belief is, in some epistemically relevant sense, blameless (cf., Boghossian 2003). But it’s highly doubtful that Avicenna, al-Rāzī, and other skeptical mutakallimūn failed to ­properly understand the initial premise of the kalām argument (cf., Williamson 2003); wherefore, one might criticize al-Ghazālī by simply quoting al-Ghazālī: “If you know this [premise] through the necessity of reason, how is it, then, that those who oppose you do not share this knowledge…? And these [individuals] certainly do not stubbornly defy reason…” (Tahāfut, p. 17). Ironically, this ­passage from the Tahāfut is part of al-Ghazālī’s “refutation” of the belief (defended at length by Avicenna) that the world is eternal. 12 Not all generics begin with a bare plural noun. For example, the following ­sentences have generic readings: the tiger is a mammal; a madrigal is polyphonic; candy rots teeth. (Compare these readings with the most natural interpretations of the following: the tiger is sleeping on our front lawn; a madrigal is popular; candy is in that jar.) For my purpose, we can ignore these cases. I’ll focus entirely on generics beginning with a bare plural noun. 13 Generics tolerate exceptions is itself a generic that tolerates exceptions: e.g., ­triangles have three sides and conjunctions entail their conjuncts. 14 See, among others, Carlson and Pelletier (1995),  Leslie (2007, 2008), Nickel (2016), and Sterken (2015a, b, 2017). 15 Though, at a deeper level of syntactic analysis, linguists have posited a covert operator: Gen. For further discussion, see the debate between Liebesman (2011), Leslie (2015), Sterken (2016), and Collins (2018). 16 The contributions of many other theorists are acknowledged in this passage. I’ve removed these citations to streamline my presentation. See, among others, ­Carlson and Pelletier (1995), Cohen (1999), and Gelman (2003). 17 Negative exceptions are individuals that lack the property of being F but don’t possess an alternative to Fness. Alternatives are incompatible properties. For ­example, giving birth to live young is an alternative to laying eggs, but carrying malaria isn’t an alternative to carrying the West Nile virus. “There may be reasons to be skeptical of attempts to draw a metaphysical distinction between positive and negative properties. The distinction at hand, however, is not intended as a metaphysical distinction, but rather a psychological one. What matters is whether we take the counterinstances as negative or positive” (Leslie 2008, p. 34). 18 For alternative opinions, see Cohen (1999), Nickel (2016), and Sterken (2015c). 19 One reaction to Leslie’s argument is that tokens of type B are ambiguous ­between a false generic interpretation and a true capacity reading (Asher and Pelletier 2012; Nickel 2016). Insofar as we intuit the truth of mosquitoes carry the West Nile virus, we take it to mean that mosquitoes have the capacity to carry the West Nile virus. And if that’s right, then our intuitive evaluation of the mosquitosentence doesn’t really impinge on the semantics of generics. 20 Saul (2017, p. 4, fn. 2) expresses sympathy for Sterken’s criticism. 21 The discussion that follows is taken from Almotahari (2022). 22 Sterken also relies on claims about disagreement, but I think these claims raise fundamentally the same issues as (4). Contradictions are, after all, disagreements with oneself. 23 Won’t Sterken’s argument overgeneralize? That is, won’t it imply that even ­paradigm generics (not just tokens of type B) are false? Birds lay eggs, but ­typically they don’t sounds contradictory, after all. Sterken is aware of this worry

60  Mahrad Almotahari and addresses it (2015a, p. 86). She also doesn’t rule out the possibility of a more expansive error theory. 24 Strong generics don’t exhibit this behavior. Consider: i ??Tigers have stripes; in fact, they typically do. ii #Tigers typically have stripes; in fact, they do. To my ear, (i) is worse than (10a). Overwhelmingly, my informants corroborate this judgment. It shouldn’t be surprising that we don’t get the kind of contrast in (i) and (ii) that we observe in (9) and (10). Tigers have stripes is about as strong as tigers typically have stripes. 25 The rhetoric in Pruss and Rasmussen (2018, pp. 40–47) is somewhat similar, but in a few key respects the presentation is much more careful. 26 Oppy (1999) recommends a different sort of restriction. Translating into the ­idiom of premise (1), the restriction is that everything that begins to exist, with the exception of the universe, has a cause for its existence. Koons (2001) criticizes this recommendation on the grounds that it’s gruesomely unnatural: Oppy’s principle stands to premise (1) as emeralds are grue stands to emeralds are green. I don’t think Koons is right, but in any case there’s nothing gruesome about genericity; in fact, one way in which we might maximize the naturalness of our classification scheme is by formulating laws as generics (Nickel 2010). 27 See van Inwagen (1996) for an argument to this effect. The argument has inspired many responses. 28 Cf., Pruss and Rasmussen (2018, pp. 40–47). 29 Thanks to Brian Rabern and Patrick Todd for raising this worry in conversation. 30 Cf., Koons (1997), in which the general premise of the argument is a restricted generalization: “normally, a wholly contingent fact has a cause” (p. 197). And this restricted generalization is understood as a “default or defeasible rule” of inference: “…in the absence of evidence to the contrary, we may infer, about any particular wholly contingent fact, that it has a cause” (p. 196). Given the ­characteristic psychological role of generics as default inference licensors (Khemlani et al. 2012), the general premise Koons relies on is functionally equivalent to (11). To this extent, his formulation of the cosmological argument is superior but less conclusive. After all, the origins of the universe seem to be highly abnormal conditions.

References Al-Ghazālī, Abū Ḥāmid. 2000. Tahāfut al-falāsifa. M.E. Marmura, tr., The ­Incoherence of the Philosophers. Brigham Young University Press. ———. 2013. Al-Iqtiṣād fī al-I’tiqād. A.M. Yaqub, tr., Moderation in Belief. Chicago, IL: University of Chicago Press. Almotahari, Mahrad. 2022. ‘Weak Generics’, Analysis, forthcoming. Amijee, Fatema. 2021. ‘Explaining Contingent Facts’, Philosophical Studies 178: 1163–1181. Asher, Nicholas and Pelletier, Francis Jeffry. 2012. ‘More Truths about Generic Truth’, in A. Mari, C. Beyssade and F.D. Prete, eds., Genericity. Oxford: Oxford University Press. Bailey, Andrew M., Rasmussen, Joshua, and Van Horn, Luke. 2011. ‘No Pairing Problem’, Philosophical Studies 154: 349–360. Boghossian, Paul. 2003. ‘Blind Reasoning’, The Aristotelian Society Supplementary Volume 77: 225–248.

Kalām and Cognition  61 Carlson, Greg and Pelletier, Francis Jeffrey. 1995. The Generic Book. Chicago, IL: University of Chicago Press. Chittick, William. 1998. The Self-Disclosure of God: Principles of Ibn al-‘Arabī’s Cosmology. Albany: State University of New York Press. Cimpian, Andrei, Brandone, Amand C., and Gelman, Susan A. 2010. ‘Generic ­Statements Require Little Evidence for Acceptance but Have Powerful I­ mplications’, Cognitive Science 34: 1452–1482. Cohen, Ariel. 1999. Think Generic: The Meaning and Use of Generic Sentences. CSLI. ———. 2012. ‘Generics as Modals’, Recherches linguistiques de Vincennes 41: 63–82. Collins, John. 2018. ‘Genericity sans Gen’, Mind and Language 33: 34–64. Connolly, Andrew C., Fodor, Jerry A., Gleitman, Lila R., and Gleitman, Henry. 2007. ‘Why Stereotypes Don’t Even Make Good Defaults’, Cognition 103: 1–22. Copan, Paul and Craig, William Lane. 2018a. The Kalām Cosmological ­Argument, Volume 1: Philosophical  Arguments for the Finitude of the Past. London: Bloomsbury. ———. 2018b. The Kalām Cosmological Argument, Volume 2: Scientific Evidence for the Beginning of the Universe. London: Bloomsbury. Craig, William Lane. 2018. ‘The Kalam Argument’, in J.P. Moreland, Chad Meister, and K.A. Sweis, eds., Debating Christian Theism. Oxford: Oxford University Press. Craig, William Lane and Sinclair, James D. 2012. ‘The Kalām Cosmological ­Argument’, in W.L. Craig and J.P. Moreland, eds., The Blackwell Companion to Natural Theology. Hoboken, NJ: Wiley-Blackwell. Della Rocca, Michael. 2010. ‘PSR’, Philosophers’ Imprint 10: 1–13. Erlwein, Hannah C. 2017. ‘Proving God’s Existence? A Reassessment of al-Rāzī’s ­Arguments for the Existence of the Creator’, Journal of Qur’ānic Studies 19: 31–63. Fakhry, Majid. 1957. ‘The Classical Islamic Arguments for the Existence of God’, The Muslim World 47: 133–145. Gelman, Susan A. 2003. The Essential Child: Origins of Essentialism in Everyday Thought. Oxford: Oxford University Press. Goldman, Alvin I. 2015. ‘Naturalizing Metaphysics with the Help of Cognitive ­Science’, in K. Bennett and D.W. Zimmerman, eds., Oxford Studies in ­Metaphysics, vol. 9, 171–213. Oxford: Oxford University Press. Goldman, Alvin I. and McGlaughlin, Brian P. 2019. Metaphysics and Cognitive ­Science. Oxford: Oxford University Press. Hallaq, Wael B. 1991. ‘Ibn Taymiyya on the Existence of God’, Acta Orientalia 52: 49–69. Hassan, Laura. 2020. Ash‘arism Encounters Avicennism: Sayf al-Dīn al-Āmidī on Creation. Piscataway, NJ: Gorgias Press. Hollander, Michelle A., Gelman, Susan A., and Star, Jon. 2002. ‘Children’s ­Interpretation of Generic Noun Phrases’, Developmental Psychology 36: 6. Horn, Laurence. 2021. ‘Ceteris Paribusiness: On the Power of Salient Exceptions’, in F. Macagno and A. Capone, eds., Inquiries in Philosophical Pragmatics. Springer. Inati, Shams. 2014. Ibn Sina’s Remarks and Admonitions: Physics and Metaphysics. New York: Columbia University Press. Johnston, Mark and Leslie, Sarah-Jane. 2012. ‘Concepts, Analysis, Generics and the Canberra Plan’, Philosophical Perspectives 26: 113–171. Jönsson, Martin and Hampton, James. 2006. ‘The Inverse Conjunction Fallacy’, Journal of Memory and Language 55: 317–334.

62  Mahrad Almotahari Kahneman, Daniel. 2012. Thinking, Fast and Slow. London: Penguin. Karczewski, Daniel, Wajda, Edyta, and Radosłow, Poniat. 2020. ‘Do All Storks Fly to Africa? Universal Statements and the Generic Overgeneralization Effect’, Lingua 246: 1–15. Khemlani, Sangeet, Leslie, Sarah-Jane, and Glucksberg, Sam. 2007. ‘Do Ducks Lay Eggs? How Humans Interpret Generic Assertions’, in D.S. McNamara and J. G. Trafton, eds., Proceedings of the 29th Annual Conference of the Cognitive Science Society. Nashville, TN: Cognitive Science Society. ———. 2012. ‘Inferences about Kinds: The Generics Hypothesis’, Language and Cognitive Processes 27: 887–900. Kim, Jaegwon. 2003. ‘Lonely Souls: Causality and Substance Dualism’, in T. O’Connor and D. Robb, eds., Philosophy of Mind: Contemporary Readings. New York: Routledge. Koons, Robert. 1997. ‘A New Look at the Cosmological Argument’, American ­Philosophical Quarterly 34: 193–211. ———. 2001. ‘Defeasible Reasoning, Special Pleading, and the Cosmological ­Argument’, Faith and Philosophy 18: 192–203. Lazaridou-Chatzigoga, Dimitra, Stockall, Linnaea, and Katsos, Napoleon. 2017. ‘A New Look at the “Generic Overgeneralization” Effect’, Inquiry 1–27. ———. 2019. ‘Contextualising Generic and Universal Generalizations: ­Quantifier Domain Restriction and the  Generic Overgeneralization Effect’, Journal of ­ ­Semantics 6: 617–664. Leslie, Sarah-Jane. 2007. ‘Generics and the Structure of the Mind’, Philosophical ­Perspectives 21: 375–403. ———. 2008. ‘Generics: Cognition and Acquisition’, Philosophical Review 117: 1–47. ———. 2015. ‘Generics Oversimplified’, Noûs 49: 29–54. ———. 2017. ‘The Original Sin of Cognition: Fear, Prejudice, and Generalization’, Journal of Philosophy 114: 393–421. Leslie, Sarah-Jane and Gelman, Susan A. 2012. ‘Quantified Statements are ­Remembered as Generics: Evidence from Preschool Children and Adults’, C ­ ognitive Psychology 64: 186–214. Leslie, Sarah-Jane and Johnston, Mark. 2012. ‘Concepts, Analysis, Generics and the Canberra Plan’, Philosophical Perspectives 26: 113–171. Leslie, Sarah-Jane, Khemlani, Sangeet, and Glucksberg, Sam. 2011. ‘Do All Ducks Lay eggs? The Generic  Overgeneralization Effect’, Journal of Memory and ­Language 65: 15–31. Lewis, David. 1974. ‘Radical Interpretation’, Synthese 23: 331–344. ———. 1979. ‘Scorekeeping in a Language Game’, Journal of Philosophical Logic 8: 339–359. Liebesman, David. 2011. ‘Simple Generics’, Noûs 45: 409–442. Mackie, J.L. 1982. The Miracle of Theism. Oxford University Press. Malpass, Alex. 2022. ‘All the Time in the World’. Mind 131: 786–804. Malpass, Alex and Morriston, Wes. 2020. ‘Endless and Infinite’, Philosophical ­Quarterly 70: 830–849. Nickel, Bernhard. 2010. ‘Ceteris Paribus Laws: Generics and Natural Kinds’, ­Philosophers’ Imprint 10: 1–25. ———. 2016. Between Logic and the World: An Integrated Theory of Generics. ­Oxford University Press.

Kalām and Cognition  63 Oppy, Graham. 1999. ‘Koons’ Cosmological Argument’, Faith and Philosophy 16: 378–389. Paul, L.A. 2010. ‘A New Role for Experimental Work in Metaphysics’, Review of Philosophy and Psychology 3: 461–476. Pruss, Alexander R. 2006. The Principle of Sufficient Reason: A Reassessment. ­Cambridge: Cambridge University Press. ———. 2009. ‘Recent Progress on the Cosmological Argument’, in K. Timpe, ed., Arguing About Religion. New York: Routledge. ———. 2012. ‘The Leibnizian Cosmological Argument’, in W.L. Craig and J.P. ­Moreland, eds., The Blackwell Companion to Natural Theology. Wiley-Blackwell. Pruss, Alexander R. and Rasmussen, Joshua L. 2018. Necessary Existence. Oxford University Press. Saul, Jennifer. 2017. ‘Are Generics Especially Pernicious?’ Inquiry 1–18. Shihadeh, Ayman. 2008. ‘The Existence of God’, in T. Winter, ed., Classical Islamic Theology. Cambridge University Press. Sorensen, Roy. 2012. ‘The Sorites and the Generic Overgeneralization Effect’, ­Analysis 72: 444–449. Sterken, Rachel Katherine. 2015a. ‘Generics, Content, and Cognitive Bias’, Analytic Philosophy 56: 75–93. ———. 2015b. ‘Leslie on Generics’. Philosophical Studies 172: 2493–2512. ———. 2015c. ‘Generics in Context’, Philosophers’ Imprint 15: 1–30. ———. 2016. ‘Generics, Covert Structure and Logical Form’, Mind & Language 31: 503–529. ———. 2017. ‘The Meaning of Generics’, Philosophy Compass 12: 1–13 Tardif, Twila, Gelman, Susan A., Fu, Xiaolan, and Zhu, Liqi. 2012. ‘Acquisition of Generic Noun Phrases in Chinese: Learning about Lions Without an “-s”’. Journal of Child Language 39: 130–161. Tversky, Amos and Kahneman, Daniel. 1982. ‘Evidential Impact of Base Rates’, in D. Kahneman, P. Slovic, and  A. Tversky, eds., Judgment Under Uncertainty: ­Heuristics and Biases. Cambridge University Press. Van Inwagen, Peter. 1996. ‘Why Is There Anything at All?’ Proceedings of the ­Aristotelian Society 70: 95–120. Wajda, Edyta and Karczewski, Daniel. 2020. ‘Do All Eagles Fly? The Generic ­Overgeneralization Effect: The Impact of Fillers in Truth Value Tasks’, Studies in Logic, Grammar and Rhetoric 61: 147–162. White, Roger. 2010. ‘You Just Believe That Because…’, Philosophical Perspectives 24: 573–615. Williamson, Timothy. 2003. ‘Understanding and Inference’, The Aristotelian Society Supplementary Volume 77: 249–293. Zarepour, Muhammad Saleh. 2021. ‘Infinite Magnitudes, Infinite Multitudes, and the Beginning of the Universe’, Australasian Journal of Philosophy 99: 472–489. ———. 2022. ‘Counting to Infinite, Successive Addition, and the Length of the Past’, International Journal for Philosophy of Religion 92: 167–176.

4 God’s Existence and the Uniqueness Claim Nazif Muhtaroglu

4.1 Introduction Consider someone who claims that there are Tasmanian tigers. It would be adequate to show just one Tasmanian tiger to back up this claim. This is how an existential claim about a concept is justified. On the other hand, if ­someone claims that there is only one Tasmanian tiger, just showing one instance of this species would not be enough. In addition, one must show that there is no other Tasmanian tiger. This is how a uniqueness claim about a concept is justified. Likewise, when we consider arguments for the existence of God, we must clarify what needs to be shown. Is the term “God” understood as a general term much like the species of Tasmanian tigers or is it taken as a proper name? If it is the former, then it is better to write it as “god” to underline its difference from the proper name “God.” “God” as a proper name must refer to a single/unique being if it refers at all. As a result, justifying the existence of a god and God would have different satisfaction conditions. On the one hand, the existence of a god would require justifying that there is at least one instance that falls under the concept of god. On the other hand, the ­existence of God would also require that there is at most one god. The literature on the existence of God roughly since Kant seems largely insensitive to the distinction between god and God as outlined above. In Critique of Pure Reason, Kant classifies arguments for the existence of God into three main classes: ontological, cosmological, and teleological (Kant 2000: A 591/B619-A630-B658). Then, he offers substantial criticisms of these arguments, as known very well. Yet, the formulation of these ­arguments does not even purport to establish that there is only one god. Kant’s main task is to undermine our cognitive capacity to know that there is at least one god understood as an ultimate being beyond the phenomenal realm. Kant’s move is understandable to a certain extent. If we cannot justify that there is at least one god, then there is no point to show that this god is unique. Needless historical to say, Kant’s criticisms were very influential. Arguably, this ­ and ­dialectical context that challenges the existence of any god could have clouded what requires for a full argument for the existence of God. In the DOI: 10.4324/9781003327714-7

God’s Existence and the Uniqueness Claim  65 endeavour to respond modern challenges, the main focus had been shifted to divine existence, and the issue of divine uniqueness largely slipped attention. Nonetheless, for monotheists, God is a unique being, there is no more than one god. It is still a burden on the monotheists to complement their arguments for God with additional steps to justify the uniqueness claim. Recently, there has been an increasing awareness of the uniqueness of God (Van Inwagen 2006: 31–35, Nagasawa 2017: 8–12). Furthermore, specific arguments have been proposed to establish divine uniqueness (Wainwright 1986: 289–314, Zagzebski 1989, Geisler & Corduan 2003: 189–190, Swinburne 2004: 97, Baillie & Hagen 2008, O’Connor 2012: 93, Rizvi 2018, Zarepour 2022). In this chapter, I propose to reconstruct and discuss an argument for divine uniqueness from the Islamic kalām tradition. It is known as “mutual hindrance by two omnipotent beings” (burhān al-tamānuʿ). There are different versions and formulations of this argument. Al-Ghazālī’s (d. 505/1111) version has been picked up in the contemporary literature. There are some reconstructions of his argument (Wainwright 1986: 301–305, Al-ʿAllaf 2003: 12–13, Baillie & Hagen 2008, Gwynne 2009: 176, Chowdhury in Taşköprüzāde 2020: 201–203). Wainwright (1986) and Baillie and Hagen (2008) ­discussed it by considering some objections. I focus on the modal v­ersion of this ­argument, which I call “the argument from possible conflict among gods,” as proposed by Saʿd al-Dīn al-Taftāzānī (d. 792/1390). Al-Taftāzānī’s argument enables us to see the logical structure of this type of argument more clearly. After I update it by the possible world semantics, its logical structure and force will be seen even more clearly. So, the chapter unfolds as follows. Since this argument relies on some premises coming from the arguments aiming to establish that there is at least one god, first I shall outline the general structure of the kalām arguments for the existence of God. This background is missing in the existing literature centred around al-Ghazālī’s argument. Yet, this background enables us to situate al-Taftāzānī’s argument in its proper context and appreciate its function within a web of interrelated arguments. Next, I present al-Taftāzānī’s argument and update it by the possible world semantics. Finally, I consider three significant objections to it and respond to them by extending the existing literature.1 4.2

An Outline of the Kalām Way of Arguing for the Existence of God

To my knowledge, there is no ontological type of argument that aims to establish the existence of God from the concept of god in the works of kalām scholars. Their arguments typically rely on an inference from the existence and various features of the universe to the existence of God. In this reasoning, one “proceeds from a visible effect to an invisible cause” (istidlāl bi l-shāhid ʿalā l-ghā’ib).2 This type of arguing for the existence of God has basically three main steps: (1) argument(s) for the existence of an eternal or necessary being, (2) arguments to establish that the eternal or necessary being

66  Nazif Muhtaroglu in question is a personal or volitional agent having some attributes such as omnipotence, omniscience, and will essentially, (3) argument(s) for the uniqueness of this personal being who is eternal or necessary. All these steps collectively constitute an argument for the existence of God (Allāh). Thus, it would be more proper to consider these steps as part of a meta-argument.3 In the first step, some scholars, especially the pre-Ghazālian ones (10th c.– 12th c.), initially try to establish that the universe has a beginning. Next, they argue that there must be an eternal (qadīm) being responsible for the ­origination as the ultimate cause. It has also been noted that being eternal entails that it is a necessary being (wājib al-wujūd) as well.4 With ­multiple ways to argue for the origination of the universe, this type of a­ rgument is generally known as “the kalām cosmological argument” as popularized by William Lane Craig (Craig 2000). Yet, it would be better to modify this title as “the kalām cosmological argument from temporal origination” (dalīl al-ḥudūth) because there is another widespread type of cosmological ­argument in kalām. Scholars, especially that come later after al-Ghazālī, propose ­arguments that start with contingent beings. Without needing to establish the ­origination of the universe, they infer the existence of a necessary being from the ­contingent beings. Let us call this type of argument “the kalām cosmological ­argument from contingency” (dalīl al-imkān).5 I would like to underline one premise in these cosmological arguments. This premise states the principle of causality. It is expressed in various ­formulations, but the common idea behind them is to consider anything temporal to be an effect of an efficient cause. Thus, anything that comes into being must have a cause. Some scholars directly infer from temporal beings that they need a cause to be existent. Some others think that the need for a cause is understood through their contingency. In other words, temporal beings are contingent beings. Since they are contingent, they must have a cause. Contingency implies alternatives. A contingent being exists but was not existent. Thus, both existence and non-existence are possible states and validly apply to contingent beings. Then, we must ask the question why a contingent being exists rather than not. The answer specifies a cause as a reason for the actual state. As we will see, this analysis of contingency plays an important role in understanding the nature of divine attributes. Al-Taftāzānī uses both types of cosmological arguments, that from ­temporal origination and that from contingency (Al-Taftāzānī 1950: 36–37, 2019: 509–511). He notes that the latter comes from Hellenic philosophers (falāsifa) (Al-Taftāzānī 2019: 509) and gives special attention to the former (Al-Taftāzānī 1950: 36–37). He describes the creator of the universe as a ­necessarily existent being as well as an eternal being (Al-Taftāzānī 1950: 36, 39). In the second step, kalām scholars aim to show that the eternal or necessary being in question has certain attributes that are peculiar to personal beings.

God’s Existence and the Uniqueness Claim  67 The most important properties in this list are power, will, and knowledge, which are intimately related to each other.6 These attributes are regarded as eternal and thus essential to the eternal or necessary being established at the first step. And they are also regarded as sufficient to characterize it as a personal being and a volitional agent (al-fāʿil al-mukhtār). In order to make this point, scholars usually appeal to certain features of the universe. They typically use teleological type of arguments to show that eternal or necessary being possesses knowledge. The temporal origination of the universe, the variety of finite beings and processes in the universe as well as some other features of contingency are used to show that eternal or necessary being has will. And the causality of this ultimate being could be regarded as power since it is under the guidance of will and knowledge. Another significant aspect of this step is that scholars try to show that the attributes in question do not have a finite scope. In other words, they think we cannot ascribe a limited power to a necessary being. There is no degree in divine power. Divine power applies to all possible beings. There is no limit to divine knowledge. Divine knowledge covers all things to be known. Divine will is related to all things that could be willed. In a nutshell, kalām scholars hold that divine attributes have maximal scopes. The way they argue for the maximality of the scopes appeals to a very fundamental principle related to the aforementioned analysis of contingency. They hold that anything that admits alternatives or degrees requires an explanation. Since admitting degree or alternatives is a mark of contingency, they must be determined by something else. Since an eternal or a necessary being is an ultimate being, it is not under determination by any external being. Thus, divine attributes must have maximal scopes. Otherwise, they would have to be determined by something else. In addition, the scopes of divine attributes, though maximal, are not exactly the same. For instance, divine will and power apply only to ­possibilities but not impossibilities.7 Impossibilities in this context are ­understood as metaphysical or logical contradictions. However, divine knowledge has ­ a more extensive scope. It covers possibilities as well as necessities and ­impossibilities. God knows what is possible, necessary, and impossible. Yet, according to the kalām scholars, it does not make sense to say that God can will or create impossibilities such as making somebody both existent and non-existent under the same conditions. Al-Taftāzānī considers the necessarily existent creator to be living, p ­ owerful, knowing, hearing, seeing, desiring, and willing. He appeals to some kind of teleological argument to justify this claim. He also notes that the creator in question cannot be devoid of these attributes since the lack of them would be defects that would contravene the perfection of the c­ reator (Al-Taftāzānī 1950: 41). He considers this creator to be omniscient and omnipotent because lack of them, i.e. ignorance or inability, in any degree requires a cause that determines the degree in question. In such a case, the creator would be under influence, which is contrary to his perfection (Al-Taftāzānī 1950: 48).

68  Nazif Muhtaroglu In the third step, kalām scholars try to show the uniqueness of the creator. The first step aims to establish that there is at least one creator that is eternal or necessary. The second step ascribes certain attributes to this creator. And the third step aims to show that there is at most one such creator. Once this step is taken, the uniqueness claim would be justified. There are various arguments given to this end. But al-Taftāzānī considers the argument from mutual hindrance (al-tamānuʿ) to be the most noted one among the kalām scholars (Al-Taftāzānī 1950: 37). In the next section, I present al-Taftāzānī’s presentation of this argument. Before this, I would like to underline some crucial aspects of the kalām way of arguing for the existence of God. What has been covered in these three steps is a multitude of arguments given for the existence of God. In this sense, the kalām argumentation could be seen as a precursor of the cumulative type of arguments we see today. Yet, the individual parts of the contemporary cumulative arguments are ­usually regarded as standalone arguments for God, and with others, they are ­considered to have a stronger force for this conclusion. On the one hand, kalām scholars did not consider the main arguments at each step ­separately as individual arguments for the existence of God. Only considered together these arguments could be regarded as a cumulative reason for God. The arguments presented in three steps of the meta-argument in question are integrated to each other. They cannot be taken separately and treated as an argument for God. In this sense, the kalām type is different from the ­contemporary versions of cumulative arguments. On the other hand, the s­ ub-arguments in each step could be independent of each other. For instance, one could use either the kalām cosmological argument from temporal o ­ rigination or from c­ ontingency or both within the first step. In this sense, the grouping of ­sub-arguments resembles the contemporary version of cumulative a­ rguments. Second, teleological arguments in kalām are not usually regarded as standalone arguments for God. When formulated in the second step, they are used to justify divine knowledge or will. What is more fundamental is the arguments given in the first step. Only after the first step, other arguments are considered. Now let us examine al-Taftāzānī’s argument for divine uniqueness. 4.3

Al-Taftāzānī on Arguments from Mutual Hindrance among Gods

Al-Taftāzānī presents two versions of mutual hindrance among gods (MHA). The first version relies upon the assumption that gods would conflict in ­actuality. In other words, the wills of gods would oppose over actual state of affairs. According to al-Taftāzānī, it is customary in actuality that there is mutual hindrance among rulers as one of them wants to get the upper hand. The first version of MHA depends on this idea and takes for granted that gods would inevitably conflict over something actual and try to hinder each other. We can call this version “the argument from actual conflict among gods.” Although al-Taftāzānī considers this argument to be convincing (ḥujja iqnāʿīyya), he admits that this version is open to objections. He notes the

God’s Existence and the Uniqueness Claim  69 ­ ossibility that gods may agree on the present visible order (Al-Taftāzānī p 1950: 38). There is a stronger argument that blocks this possibility. That ­version of MHA relies only on the possible conflict among gods. That is to say, gods do not have to oppose each other in actuality but they can oppose each other. We can call this version “the argument from possible conflict among gods.” Now, let us see how al-Taftāzānī formulates this version of the argument in detail (Al-Taftāzānī 1950: 38). Al-Taftāzānī considers somebody called “Zayd.” Now, assume that there are two necessary beings who are omnipotent. Being omnipotent means that their powers apply to all possible situations. Their powers are controlled by their will, and thus their wills apply to all possible cases. When an ­omnipotent being wills such a possible situation to take place, it must happen. One such being, let us call it “N1,” could will Zayd to move (e.g. walking or running) and the other, call it “N2,” could will it to lie down to rest. Either Zayd’s moving or being at rest are possible and fall within the scope of power or will. However, Zayd’s moving and being at rest completely at the same time and under the same conditions are contrary to each other and thus not possible. Al-Taftāzānī considers the following alternative options in this case. (1) Zayd both moves and lies at complete rest. Since this is impossible, this option is ruled out. That is to say, both of the wishes of N1 and N2 could not be realized. (2) Zayd moves. Then the wish of N1 is realized but the wish of N2 is not realized. If N2 wills something possible but could not make it happen, then N2 is not omnipotent, which implies that N2 is not a god. (3) Zayd lies at complete rest. Then the wish of N1 is not realized but the wish of N2 is realized. If N1 wills something possible but could not make it happen, then N1 is not omnipotent, which implies that N1 is not a god. In any case, there is only one omnipotent being in the second or third scenarios. There could not be more than one. Thus, given that there is at least one necessary being that is also omnipotent, then there is only one such being. Hence, God is unique. A fortiori God exists. We can add more alternatives to those listed above in a possible conflict between two omnipotent wills. We can also imagine the following scenarios: (4) Zayd moves but not because N1 wishes it, for some other reason. N1’s power is blocked by N2 in this scenario. (5) Zayd lies at complete rest but not because N2 wishes it, for some other reason. N2’s power is blocked by N1 in this scenario. (6) Zayd neither moves nor stays put. He is annihilated. If these scenarios could be accepted, then we need to affirm that two omnipotent beings can co-exist. I find these scenarios problematic but will evaluate them in later sections in which I respond to objections systematically. 4.4

Analysing and Updating Al-Taftāzānī’s Argument from Possible Conflict among Gods

What al-Taftāzānī aims with this argument is basically to show how a ­possible conflict could arise between two omnipotent wills. The cases he ­considers are possible. Since these states are possible, any omnipotent will

70  Nazif Muhtaroglu must have access to them because such a will has access to all possibilities. Gods in question are also essentially omnipotent because it is impossible for them to lose their power. We may also express the argument in terms of states of affairs. N1 wills a state of affairs to obtain, let us call it “S.” N2 wills S not to obtain, i.e. Not-S. S and Not-S are contradictory in the sense that the propositions expressing them are contradictory. Now, we can formulate this argument in terms of possible world semantics. Here, I do not commit myself to any view of the ontology of possible worlds. They may be real as this world or could be taken as heuristic devices to clarify our modal intuitions. In any case, the choice of the ontological model would be irrelevant to the argument in question. So, my reconstruction of al-Taftāzānī’s argument (RTA) could be reformulated in the following way: Target conclusion: There is at most one necessary being who possesses omnipotent will. 1 Assume that there are more than one such being. Call them N1 and N2 (Assumption for reductio). 2 N1 is omnipotent (Conclusion of other arguments). 3 N2 is omnipotent (Conclusion of other arguments). 4 S is a possible state of affairs, likewise ∼S is a possible state of affairs. 5 S falls under the scope of N1’s power. In other words, N1 could wish S to obtain in all possible worlds and could realize S (from 2). 6 ∼S falls under the scope of N2’s power. In other words, N2 could wish ∼S to obtain in all possible worlds and could realize it (from 3). 7 Let W be the possible world in which N1 wishes S to obtain and could realize it, but N2 wishes ∼S to obtain and could realize it. 8 S and ∼S could not obtain in the same possible world, thus in W (S and ∼S leads to a contradiction in W). 9 Assuming that both N1 and N2 are omnipotent leads to a contradiction. Thus, both N1 and N2 cannot be omnipotent. 10 Both N1 and N2 are omnipotent (2; 3). 11 (9) and (10) are contradictory. 12 (1) is false (due to the contradiction at 11). The purpose of this argument is to show that there cannot be more than one omnipotent being who could realize any possible state of affairs in any ­possible world if he wishes it. The crucial point here is that the will and power of these beings access all possible worlds, not just the actual world. In order to establish the uniqueness claim about the being that have such properties, we go to a possible world in which accepting more than one omnipotent being produce contradictions. By reductio, we deny whatever leads to c­ ontradictions. Hence, the scenario of multiple omnipotent beings is ruled out. Reformulating the argument in terms of possible world semantics

God’s Existence and the Uniqueness Claim  71 is more convenient in responding to the objections to come. Let us now ­consider some objections to the argument and respond to them. 4.5 Objections 4.5.1  The Possibility of Agreement between the Omnipotent Volitions

The first objection I consider draws attention to the possibility of the ­agreement between the omnipotent volitions in question. N1 and N2 could also agree on certain things. For instance, they both wish S to obtain in a ­possible world. Furthermore, they could both agree to realize the same ­universe up to its all details. Perhaps, our actual universe is one such p ­ ossible world, which is uniformly willed by N1 and N2. Democratic intuitions incline us to seriously consider this option. Why should we focus on the possibility of conflict? In a nutshell, the argument for divine uniqueness we have presented may not rule out the possibility of the case in which our actual universe is governed by both N1 and N2. In a response to this objection, there are two crucial points to make. First, the possibility of the agreement between the omnipotent wills is irrelevant to RTA. Second, even if we accept the possibility of the agreement between the omnipotent wills, it is not clear how the powers of different necessary beings would collaborate to realize the same thing.8 As an extension of the first point, let us accept that the wills of N1 and N2 perfectly agree on everything in a possible world called “W1.” But this case does not pose any challenge to the argument in question. Whether there are such possible worlds is irrelevant. The argument aims to find a ­counterexample or a counter case to the statement that an omnipotent being can make all ­possible states of affairs to obtain. Such a counter case is found in a possible world W in which two omnipotent beings have contradictory wishes. This is sufficient to refute the possible co-existence of two ­omnipotent beings. The line of reasoning here is similar to the following example. Assume that somebody claims that all swans are white. If you want to refute this claim, it would be sufficient to show just one black swan to exist. Although the person who makes the initial claim shows you white swans, even billions of them, appealing to white swans becomes irrelevant when a single black swan showed up. The vulnerability of the claim stems from its generality. It makes a universal claim about swans. Thus, it is easy to refute it just by one counterexample. Likewise, what we are considering in RTA is not beings with a limited range of willing or some finite power, but necessary beings with omnipotent wills that access all possibilities. Once we find one single counter case in which two omnipotent wills cannot will what is accessible to them individually, then we can refute the assumption about the existence of two omnipotent beings. This is what we do with RTA.

72  Nazif Muhtaroglu 4.5.2 What If the Omnipotent Beings Have Necessarily Compatible Volitions?

If it is impossible that the omnipotent beings could ever conflict, then it is ­possible to avoid the conclusion of RTA. Two omnipotent beings could co- exist without having a possible conflict between their wills. In response to this objection, first I need to point out that al-Taftāzānī would reject this scenario outright. Like other Ashʿarites, he holds that God is absolutely free, and divine will is not determined by any norm except ­logical and metaphysical ones, as mentioned before. It is impossible that gods of such character have necessarily compatible wills. It is possible for them to will conflicting tasks to realize. However, we may take this objection ­seriously and check as to whether the scenario in question is intelligible. To appreciate the gravity of this objection, let me start with an example. Assume that at home, I have music recordings of various types: rock, jazz, country, and classic. I have basically four main options. Yet, having access to these recordings does not mean that I choose to listen to all of them. I make specific choices. In other words, at each choice, I eliminate three options and decide on a single one. As it happens, the decided option usually turns out to be classical music. Thus, there are two distinct levels pertaining to my will: (1) what I can choose and (2) what I actually choose. Even if my options are many and diverse in the first level, one may wonder whether my actual choice in the second level is somehow determined to a single option. Divine will could be analysed in a similar way in terms of two levels or dimensions. The first level concerns to what divine will is potentially related, which is basically all the logically or metaphysically possible states of affairs. The second level concerns to what divine will is actually related. Given this distinction, we can say that the wills of two omnipotent beings access all the logical or metaphysical possibilities potentially. But the question is as to whether it is the case that their wills are necessarily rendered compatible in choosing what to realize in actuality. Perhaps their wills are determined somehow to be compatible. The way the wills of omnipotent beings are determined to be compatible is either contingent or necessary. If it is contingent, then the possibility of conflict between their wills is not lost. This possibility is sufficient for RTA to work. If the wills of omnipotent beings are determined to be compatible necessarily, we need to investigate the conditions under which this scenario holds so that we can evaluate them one by one. First, let us consider external reasons for such a result. Let us reconsider N1 and N2 as two independent, omnipotent necessary beings. There is no higher ontological category than necessary beings. They are ultimate beings. Thus, it is not the case that some greater beings could determine their will. They are also independent beings, different from each other. Given that, I conceive three scenarios: (1) N1 determines the will of N2, or (2) N2 determines the will of N1, or (3) N1 and N2 come to an agreement through a pact or something.

God’s Existence and the Uniqueness Claim  73 The first and second scenarios explicitly state that one of the necessary beings is under enforcement and determination of the other. This fact renders one of the necessary beings in question incapable of sustaining its independence. In other words, one is weaker than the other. Thus, in these scenarios, there cannot be two omnipotent beings. In the third scenario, coming to an agreement does not imply any necessity on the sides of the agreement. It is possible that one side could break the agreement. Thus, even if they come to agree on the same volitional activity, this would not be regarded as necessary. In short, we cannot explain why N1 and N2 necessarily have compatible wills by reasons external to these necessary beings. Second, let us consider whether there can be any internal reason that could explain why N1 and N2 necessarily have compatible wills. The most plausible reason would appeal to divine perfection to ground an omnipotent will that is determined internally. So, in this scenario, a necessary being is assumed to be absolutely perfect. As a result of its perfection, it wills the best possible world order to realize in actuality. For instance, the necessary being, in Ibn Sīnā’s view, knows the most perfect order. His knowledge in turn determines his will and power towards that order, and the universe conceived as manifesting the most perfect order emanates from the necessary being accordingly (Avicenna 2005: 294–295). Likewise, Leibniz’s God creates the best possible world because God knows it as the best possible world and His perfection determines His will to create it (Leibniz 1977: 156). In these scenarios, divine nature delimits divine will towards a single option. As a result, two omnipotent beings that have such a nature necessarily have the same volition over the same thing. If there is no possibility of conflict between divine wills, then RTA does not work. In response to such an objection (Wainwright 1986: 304–305) states that the possibility of conflict is intimately related to distinct personalities. In other words, having a will that could possibly conflict with another will is a necessary condition for being a distinct person. If two wills cannot ­possibly conflict, then the distinct personality of beings having those wills seems to be threatened. Apart from Wainwright’s worry, al-Taftāzānī would not accept such a scenario in which divine will is limited by divine nature since he believes in absolute freedom of divine will. Nonetheless, I think RTA works even if we accept that omnipotent beings have perfect natures, and their ­perfection determines their wills. In what follows, I shall show that the possibility of conflict between divine wills is still valid even under this assumption. I start with an analogy. There is a practical rule called “lefty-loosey righty- tighty.” It states simply this. If you want to tighten a screw, bolt, or nut you must turn it right, i.e. clockwise. If you want to loosen any of them, you ­ etermined must turn it left, i.e. counterclockwise. However, this direction is d that way arbitrarily, as it was accepted upon convention, though we use the tools in question in accordance with it. If a screw or bolt or nut were designed to be loosened if turned left (clockwise) rather than right (counterclockwise), it would not lose anything from its f­ unctioning perfectly. You may

74  Nazif Muhtaroglu imagine a multitude of cases in which the design of machines or tools have ­certain arbitrary features that are irrelevant to their specified ­functionality. In other words, their primary and intended use is not lost through secondary ­arbitrary settings. Likewise, our universe also has ­arbitrary features which are ­irrelevant to its perfection. Al-Ghazālī makes this point in criticizing the cosmological system of the Neo-platonic philosophers (falāsifa). In their system, the world is at the ­centre of the universe and there are entwined spheres revolving around it. Moon, the sun, and the planets are attached to these spheres. The highest sphere moves from east to west and what is beneath it in the opposite direction. But al-Ghazālī asks why the highest sphere moves from east to west rather than the other way. How is this feature related to the perfection of this ­cosmological system? He comes to the conclusion that there would be no difference in terms of what is achieved in both cases (Al-Ghazālī 1997: 25). In other words, the world system would not be less perfect if the direction changed from the current trajectory to the opposite. In sum, we can ­identify arbitrary features in any model of cosmological systems. If everything in the universe became two times bigger – assuming that the proportions and links among objects are preserved– there would not be any change in the p ­ erfection of the universe. We would not even notice that difference. As a result, even if two omnipotent necessary beings necessarily will the most perfect order, their will could still differ from each other because of the arbitrary features inherent in this perfect order. In other words, there could be equally perfect but distinct possible universes. For instance, N1 could wish to create a universe with the perfect order (f), which N2 must also wish but with a tiny difference: N2 wishes that objects of the universe are three times bigger than those willed by N1. Thus, we would have a case in which N1 and N2 would will the same order, but their volitions would oppose each other. As a consequence, the suggestion that the omnipotent beings have ­necessarily ­compatible volitions is untenable. Under any circumstance, there is a p ­ ossibility of conflict between their volitions. This is sufficient for RTA to work. 4.5.3 Objection 3: Impossible Tasks Do Not Fall Under the Scope of Divine Power

If two omnipotent beings disagree on something, they can block each other’s power like two fighters stop each other from doing something. Thus, the task on which they disagree becomes impossible for them to realize. Since ­impossibilities do not fall within the scope of divine power, as round squares do not fall, their omnipotence is not threatened. In such a scenario, two omnipotent beings co-exist even if there is conflict between their volitions. Briefly stated, this is the third and last objection to RTA. The roots of this type of criticism go back at least to Moses Maimonides. The inability of one deity to influence the province of the other is not regarded as a defect by Maimonides since it is considered to be intrinsically

God’s Existence and the Uniqueness Claim  75 impossible (Maimonides 1956: 139). Recall that kalām scholars exclude impossibilities from the scope of divine power. Knowing that, Maimonides wants to treat the impossible task stemming from the conflict between two omnipotent ­volitions similar to logical or metaphysical impossibilities which are ­admittedly outside the scope of any kind of power. A task like creating a body of squared circle shape would be a category mistake. It does not imply any defect in the power of anybody. Likewise, as Maimonides’s argument goes, the inability of an omnipotent being to influence another omnipotent being is considered to be equally impossible. Such a task does not fall within their powers. Neither is it a problem. In the contemporary literature on the philosophy of religion, Werner (1971) and Mele and Smith (1988) have proposed a similar objection. For instance, Mele and Smith (1988: 285) have engineered the following idea. They draw attention to the tendency to remove impossibilities such as ­spherical cubes from the scope of omnipotence. They consider a case in which one ­omnipotent being (N1) wishes to lift a stone whereas the other omnipotent being (N2) wishes to keep it stationary. In their view, either omnipotent being would be attempting something logically impossible in such a case since it is impossible to override the volition of an omnipotent being. This i­ mpossibility does not count against their omnipotence because impossibilities do not fall within their power. As a result, an actual or possible conflict between the volitions of omnipotent beings does not necessarily imply that the conclusion al-Taftāzānī derives from this case. Rather they could co-exist despite having conflicting wills. In response to this objection, first, I must note that the impossibility ­concerning the task between two conflicting omnipotent volitions is not exactly the same type of impossibility related to logical contradictions or denial of metaphysical truths. Lifting a stone or keeping it stationary are inherently possible tasks, which could be performed by omnipotent beings under normal conditions. They are not like creating a statue of squared circle. Maimonides considered the inability to influence the province of ­ another omnipotent being to be intrinsically impossible. Mele and Smith ignore the difference between different types of impossibilities. Werner (1971: 66) ­considers a being who is unable to make round square objects to be as weak as the one who is unable to frustrate a being whose will cannot be frustrated. Thus, he ignores the difference between the two types of impossibilities in question. Yet, doing something within the province of somebody else is like doing it in your own province. Lifting a stone in the garden of somebody is like ­lifting it in your garden. These tasks are inherently possible. The impossibility in question stems from relating the task to another powerful agent. In other words, it is a relational impossibility. In the story told, the task of lifting the stone or keeping it stationary emerges from the peculiar context in which there are two omnipotent beings whose wills are in conflict. In short, the case in question is not intrinsically impossible. As such, the concept of power

76  Nazif Muhtaroglu is defined only via intrinsic possibilities or impossibilities, not in relation to extrinsic ones. Since an omnipotent being has the maximal scope of power, this maxim is valid in understanding what omnipotence means. Nonetheless, one can present certain cases that do not imply any defect in power although they are intrinsically possible. One can argue that p ­ eople can perform certain physical acts that are counted as immoral, but their ­characters may not allow them to perform such acts. In such cases, their limitations could not be regarded as defects in power. Likewise, if we assume that an omnipotent being has a certain nature which does not allow certain possible tasks, then we cannot treat its non-performance as a defect in power. Note that the omnipotent being in question has the power to realize those tasks. It is not the case that tasks contrary to the nature of the omnipotent being do not fall within the scope of power. It is contrary to its nature. That is why such tasks never come from such an omnipotent being. However, the impossibility pointed out by Maimonides, Werner, Mele and Smith is not like this one. It does not concern the nature of omnipotent beings, and it is ­extrinsically characterized as an outcome from the conflict of two o ­ mnipotent volitions. Thus, it would be a fallacy of equivalence not to consider this type of impossibility to be a threat to the concept of omnipotence. In addition, al-Taftāzānī would again object to the idea of considering divine nature as limiting divine will and power. He would not consider any type of impossibility to be legitimate except the logical contradictions or denials of metaphysical truths. Nonetheless, I think the third objection is problematic even if we accept types of impossibilities defined extrinsically as Maimonides, Werner, Mele, and Smith suggest. Let me start with Baillie and Hagen’s counter argument first. Baillie and Hagen (2008: 32–33) argue that Mele and Smith’s analysis of omnipotence undermines the very idea of omnipotence. Baillie and Hagen consider a possible world in which two omnipotent beings co-exist. Let us call them “W2,” “N1,” and “N2,” respectively. Now, suppose that in W2 N2 wishes the opposite of whatever N1 wishes. In W2, the wills of N1 and N2 are always in conflict. Thus, there is a possible world, say W3, relative to W2, in which N2 thwarts N1 whenever N1 wishes to bring about some states of affairs. It should be clear that in W3, N1 could not bring about anything because it will be blocked by N2. As a result, Mele and Smith’s analysis of omnipotence implies that there is an omnipotent being (N1) in a ­possible world (W3) that is not able to exercise any power. According to Baillie and Hagen, this being is more properly called “omni-impotent” rather than “omnipotent” as it lacks the power to do anything. Second, in scenarios of a possible conflict between two omnipotent beings, one significant issue is to determine the status of the states of affairs in ­question. What happens to the stone or Zayd? Again, N1 wishes the stone to move whereas N2 wishes it to remain stationary. Mele and Smith (1988: 285) think that nothing can move the stone if N2 wishes to keep it stationary

God’s Existence and the Uniqueness Claim  77 and nothing can keep it stationary if N1 wishes to move it. They derive these ideas from the concept of omnipotence. Thus, in a case of conflict between N1 and N2, one finite being cannot move the stone or keep it stationary. Yet, Mele and Smith allow the possibility that the stone goes out of ­existence in such a case because of the struggle between two omnipotent wills. Neither omnipotent volition will be realized in such a case, but something will ­happen to the stone. However, if N1 and N2 disagree as to whether to create or destroy a stone, then Mele and Smith admit that this solution would not work for such a scenario in which N1 and N2 aim contradictory states of affairs. Since these states of affairs, i.e. the existence and non-existence of the ­ appen. stone, are mutually exclusive, there is no other alternative that could h To explain the status of the stone in this case, Mele and Smith appeal to uncaused events. They hold that the stone may either move or remain at rest without a cause. To Mele and Smith, in possible worlds in which uncaused events are possible, N1 and N2 “jointly create a situation in which the stone’s behaviour may be uncaused” (Mele & Smith 1988: 286). As a response to Mele and Smith, let us first assume that there are ­possible worlds in which events could take place without a cause. Note that their explanation for the conflict concerning contradictory states of affairs works only within such possible words in which uncaused events are allowed. Yet, what about the possible worlds in which uncaused events are not allowed? If N1 and N2 disagree over to create a stone or destroy it in a world in which all events are caused, Mele and Smith’s explanation fails. What RTA needs to work is at least one such possible world. Whether there are possible worlds in which uncaused events are allowed is irrelevant to the argument. Furthermore, neither al-Taftāzānī nor any kalām scholar would allow any possible world with uncaused events. In their view, it is a necessary ­metaphysical principle that any event has a cause. Denial of this principle is an inherent impossibility. Denial of this principle allows beings to pop up from nothing and creation ex nihilo without a creator. Recall that al-Taftāzānī’s argument for divine uniqueness is part of his general argument for the existence of God. Kalām scholar appeal to such arguments only after ­showing that there is at least one omnipotent being. If the metaphysical necessity of the principle of causality is denied, this conclusion could not be justified. There is no need to deny the multiplicity of omnipotent beings if one cannot establish that there is at least one. Thus, al-Taftāzānī’s argument or RTA for divine uniqueness would be pointless, if uncaused events are allowed. Unlike Mele and Smith, Werner (1971: 63) does not aim to explain the status of the state of affairs when two omnipotent beings will exhaustive and mutually exclusive (i.e. contradictory) alternatives. Werner simply does not consider this scenario to be possible. Thus, he admits that two o ­ mnipotent beings with contradictory wills are logically impossible. Nonetheless, he thinks that this fact does not compel us to reject the possibility of co-existing omnipotent beings with merely contrary wills.

78  Nazif Muhtaroglu Werner’s suggestion to admit the co-existing omnipotent beings with merely contrary wills while rejecting two omnipotent beings with contradictory wills sounds weird. If two omnipotent beings could have contrary wills, they could also have contradictory wills. Admitting one alternative requires to admit the other alternative. We are considering volitions, not powers to execute them. And the omnipotent beings in question could will anything that is intrinsically possible. State of affairs such as creating a stone or destroying it are intrinsically possible independently of what any other agent wills. 4.6 Conclusion As we have seen, the existence of God cannot be shown unless His ­uniqueness is justified. Arguments for divine uniqueness are an essential part of ­arguing for God’s existence. Although this point has been largely ignored, it is crucial to revive it. Muslim kalām tradition provides substantial arguments for divine uniqueness. In this chapter, I reconstructed al-Taftāzānī’s ­argument from the possible conflict among gods by possible world semantics. I defended it against three main objections. First, the scenario as to whether gods could cooperate is irrelevant to the argument. Second, the scenario in which gods have necessarily compatible wills is problematic, and there is always a ­possibility of conflict between them. Finally, the impossibility emerging from the conflict between divine wills could not be regarded as a legitimate case that do not fall within the scope of omnipotence.9 Notes 1 This chapter is an extension of Muhtaroğlu (2019) and Muhtaroğlu (2020) in which I presented a reconstruction of al-Taftāzānī’s argument from the possible conflict among gods and respond to several criticisms. 2 This is the translation of Joseph van Ess. For more information about this type of inference see Van Ess (1970: 34–35). 3 See Muhtaroğlu (2016) for a more detailed analysis of such a meta-argument for God’s existence in the example of al-Bāqillānī. 4 For more details on the link between eternality and necessary existence in the work of kalām scholars see Wisnovsky (2004: 90–95). 5 As an example of the contingency argument based on the contingency of the ­universe see Al-Jurjānī (2015: vol.3, 14–28). 6 Most scholars hold that power (qudra) implies will (irāda); and will implies knowledge (ʿilm). 7 Concerning the kind of impossibilities that do not fall within the scope of ­divine power, kalām scholars predominantly agree that logical contradictions and ­metaphysical impossibilities are outside the scope. In their view, metaphysical impossibilities are understood as violations of truths about the basic categories of reality. For instance, a substantial body such as a book could not become an accident such as blackness since substances and accidents are two fundamental categories of reality that are irreducible. Logical or metaphysical impossibilities are directly excluded from the scope of any type of power or will. Yet, there is a controversy as to whether divine nature delimits divine will or power. Ashʿarites

God’s Existence and the Uniqueness Claim  79 hold that divine will is absolutely free and was not determined by any norm. As a result, they adopted a radical version of divine command theory, according to which divine will is not subject to any norm including moral norms. Muʿtazilites and Māturīdites, on the other hand, hold that there are objective norms known by God which in turn delimits divine will and power. I need to point out that al- Taftāzānī is an Ashʿarite scholar. 8 The second point was extended by kalām scholars as another argument for divine uniqueness, which is known as burhān al-tawārud. I examined that argument in Muhtaroğlu (2020) and will not address it here. 9 I would like to thank Shoaib Ahmed Malik and Mohammad Saleh Zarepour for their valuable comments on this chapter.

References Al-ʿAllaf, M. (2003) Mirror of Realisation: God Is Precept and the Universe a ­Concept. St. Louis, MO: IIC. Al-Ghazālī. (1997) The Incoherence of the Philosophers (Tahāfut al-Falāsifa), A parallel English-Arabic text. Trans. and ed. by M. Marmura. Provo: Brigham ­ Young University Press. Al-Jurjānī, Al-Sayyid al-Sharīf. (2015). Sharh al-Mawāqif. (Mevâkıf Şerhi). [Turkish Arabic Parallel Text]. Edited and translated by Ömer Türker. 3 Volumes. Istanbul: Türkiye Yazma Eserler Kurumu. Al-Taftāzānī. (1950) A Commentary on the Creed of Islam. Trans. E. E. Elder. New York: Columbia University Press. Al-Taftāzānī. (2019) Al-Maqāsıd (Kelam İlminin Maksatları). Trans. into Turkish and edited by İ. Eyibil & A. Kaylı. İstanbul: Türkiye Yazma Eserler Kurumu Başkanlığı Yayınları. Avicenna. (2005) The Metaphysics of Healing (Al-Shifā’: Al-Ilahiyāt). A p ­ arallel English-Arabic text. Trans. and ed. by M. Marmura, Provo: Brigham Young ­ ­University Press. Baillie, J. & Hagen, J. (2008) “There Cannot Be Two Omnipotent beings”. ­International Journal of Philosophy of Religion, 64: 21–33. Craig, W. L. (2000) The Kalām Cosmological Argument. Eugene, OR: Wipf and Stock Publishers. Geisler, N. & Corduan, Winfried. (2003) Philosophy of Religion. Second Edn. ­Eugene, OR: Wipf and Stock Publishers. Gwynne, R. W. (2009) Logic, Rhetoric and Legal Reasoning in the Qur’an God’s Arguments. London: Routledge. Kant, I. (2000) Critique of Pure Reason. Trans. and ed. by P. Guyer & A. W. Wood. Cambridge: Cambridge University Press. Leibniz, G. W. (1977) Monadology and Other Philosophical Essays. Trans. P. Schrecker and A. M. Schrecker. Eighth Edn. Indianapolis, IN: The Bobbs-Merrill Comp., Inc. Maimonides. (1956) The Guide for the Perplexed. Trans. By M. Friedlander. New York: Dover Publication. Mele, A. & Smith, M. P. (1988) “The New Paradox of the Stone,” Faith and ­Philosophy, 5: 283–290. Muhtaroğlu, N. (2016) “Al-Bāqillāni’s Cosmological Argument from Agency,” ­Arabic Sciences and Philosophy, 26/2: 271–289.

80  Nazif Muhtaroglu Muhtaroğlu, N. (2019) “Tanrının Tekliğine Dair Modal Argüman,” Felsefe Arkivi, 51: 179–185. Muhtaroğlu, N. (2020) “Tanrı’nın Tekliğine Dair Tartışmalar,” Tasavvur Tekirdağ İlahiyat Dergisi, Aralık/December 6/2: 1153–1164. Nagasawa, Y. (2017) Maximal God: A New Defence of Perfect Being Theism. ­Oxford: Oxford University Press. O’Connor, T. (2012) Theism and Ultimate Explanation: The Necessary Shape of Contingency. Malden, MA: Blackwell. Rizvi, S. (2018) “God Is One but Unlike Any Other: Theological Argumentation on Tawḥīd in Islam,” in Monotheism and Its Complexities: Christian and Muslim Perspectives. Edited by L. Moasher & D. Marshall, 121–135. Washington, DC: Georgetown University Press. Swinburne, R. (2004) The Existence of God. Second Edn. Oxford: Oxford University Press. Taşköprüzāde, Ahmad b. Mustafā. (2020) A Treatise on Disputation and Argument (Risālat al-Adāb fī ‘Ilm al-Baḥth wa’l-Munāzara). Trans. by S. Z. Chowdhury with introduction and notes. London: Dar al-Nicosia Publications. Van Ess, J. (1970) “The Logical Structure of Islamic Philosophy,” in Logic in ­Classical Islamic Culture. Edited by G. E. von Grunebaum, 21–50. Wiesbaden: Otto Harrassowitz. Van Inwagen, P. (2006) The Problem of Evil. Oxford: Oxford University Press. Wainwright, W. (1986) “Monotheism,” in Rationality, Religious Belief, and Moral Commitment: New Essays in the Philosophy of Religion. Edited by R. Audi & W. J. Wainwright, 289–314. Cornell: Cornell University Press. Werner, L. (May 1971) “Some Omnipotent Beings,” Critica, 5: 55–72. Wisnovsky, Robert. (2004). “One Aspect of the Avicennian Turn in Sunnī Theology,” Arabic Sciences and Philosophy, 14/01: 65–100. Zagzebski, L. (1989) “Christian Monotheism,” Faith and Philosophy, 6/1: 3–18. Zarepour, Mohammad Saleh. (2022). Necessary Existence and Monotheism. ­Cambridge: Cambridge University Press.

PART III

Attributes of God

5 Divine Power and Divine Knowledge Edward Omar Moad

In their debate with the Muslim peripatetic falāsifa over the divine a­ ttributes, the mutakallimūn of the Ash‘ari school were commonly portrayed as ­defending a concept of divine nature in the likeness of human nature, drawn from an unsound analogy between the temporal and eternal; or as Ibn Rushd puts it, of making God an ‘eternal man.’ This chapter will examine an aspect of this debate as it appeared between al-Ghazālī (d. 1111) and Ibn Rushd (d. 1198), famously, in Ghazali’s Tahāfut al-Falāsifa (‘Incoherence of the Philosophers’) and Ibn Rushd’s Tahāfut al-Tahāfut (‘Incoherence of the Incoherence’). Starting from a point therein at which Ibn Rushd levels the ‘eternal man’ charge, we clarify what, for him, constitutes an unsound analogy between the temporal and eternal, along with the relevant elements of Ghazali’s own position in order to evaluate the veracity of the accusation. In the course of that inquiry, we review arguments from each side ­regarding eternal life, will, and power, all of which Ibn Rushd reduces to eternal knowledge which, according to him, is identical to God’s essence. Ultimately, we find Ghazali innocent of the analogy charge on the first three counts, but wanting a stronger defense on the fourth. That is, Ghazali’s ­conceptions of divine life, will, and power do not depend explicitly on the ‘eternal man’ ­analogy. Like Ibn Rushd, he identifies eternal life and knowledge, the ­difference being that he identifies life as such with knowledge, while Ibn Rushd identifies temporal life with animal functions. Ghazali’s account of divine will does not rest on analogy to human will, but instead on a different model of modality. Yet, his argument for divine knowledge, we find, is open to the analogy charge, absent an independent account of eternal knowledge explaining how it ­follows, as he claims, from the order we observe in the world and how it is dependent on and not, as Ibn Rushd claims, the cause of its object. 5.1

God and the ‘Eternal Man’

When in the 11th discussion of his Tahāfut al-Falāsifa Ghazali argues that the falāsifa have no rational basis for asserting that God knows anything other than Himself, he primarily targets Ibn Sīnā and his followers who departed from Aristotle on that. Having proven that every temporal existent DOI: 10.4324/9781003327714-9

84  Edward Omar Moad (the world) comes to be through God’s will, he says, the Ash‘aris could infer that ‘the existent of necessity exists previously in His knowledge, for the object willed must be known by the willer’ (1954: 255). Yet, the falāsifa, insisting that things proceed from God by causal necessity rather than by divine will, have no more reason to believe God knows them than they have to believe that fire knows about its smoke. Ibn Rushd responds that this is a false dilemma, both horns of which involve temporal concepts inapplicable to God. For God, unlike nature, acts with knowledge, yet without the desire or deliberation process affecting temporal voluntary agents like us (1954:  264). He then makes his case that while superficially persuasive, when the Ash‘ari theory is tested ‘it becomes clear that they only made God an eternal man, for they compared the world with the products of art wrought by the will, knowledge, and power of man’ (1954: 256). Yet, they did so selectively, he says, for they resisted the comparison’s implication of God’s corporeality, as incompatible with His eternity, resulting in a concept of God as an immaterial man. This, according to Ibn Rushd, is a ‘mere metaphor, the deficiency of which becomes evident’ in light of the difference between the nature of the temporal and that of the eternal. ‘It cannot be,’ he says, ‘that there should exist one single species which is differentiated by eternity and non-eternity as one single genus is differentiated through the various differences into which it is divided’ (1954: 256). That is, ‘eternal’ and ‘temporal’ cannot be variations of any real single type, and thus there is no one kind of thing named by the term ‘will’ within which ‘eternal will’ and ‘temporal will’ may be distinguished. They have nothing in common by which to underwrite even the selective comparison involved, for instance, in arguing that eternal will, like the will of our e­ xperience, entails knowledge but differs with respect to temporality and corporeality. It is thus not possible ‘to apply a judgment about the empirical world to the invisible,’ for the terms involved are ‘so equivocal that they do not permit a transference from the visible to the invisible’ (1954: 256). While we may use the same term in each case, the meaning of ‘eternal will’ in reality shares nothing with that of ‘temporal will’ such that we can infer that, if something is true of the latter it is also true of the former. We should carefully distinguish this from denying that we can infer ­anything at all about the eternal from the temporal. Ibn Rushd himself makes many such inferences. What he is rejecting here is specifically any such inference based on the presumption of similarity (no matter how nuanced) between them. The proof for the existence of the eternal starts with that of the ­temporal, though here also (according to Ibn Rushd and as opposed, in his reading, to Ibn Sīnā) ‘existence’ applies to both only equivocally (1954: 223). That is, existence is not a genus shared by temporal and eternal things. We infer the existence of the eternal by the necessity, following from the dependent nature of temporal existence, of a radically different sort of e­ xistence. We do not reason that since existence as we know from experience is dependent,

Divine Power and Divine Knowledge  85 then the existence of the eternal must also be dependent. For the same ­reason we should not infer that since will, as we experience it, entails knowledge, eternal will must also entail knowledge; or that since knowledge in our ­experience entails life, eternal knowledge entails life. Life, Ibn Rushd explains, considered a distinct attribute and condition of knowledge as it is for us, is ‘the potentiality of motion in space through will and sense perception’ (1954: 256). The Ash‘ari agree that motion in space is impossible for God and yet ascribe to Him sense perception without sensory organs. Hence, he says, they face a dilemma. To describe God as living on grounds that life is a condition of knowledge is to ascribe to Him life of the sort we know to be such a condition; that is the animal life with all its ­material organs and motion. The alternative, he claims, is to identify divine life with divine knowledge, as do the falāsifa, discarding thereby the ­procedure of inferring His life as a distinct condition of His knowledge. Yet, in the Al-Iqtiṣād fī al-I‘tiqād (‘Moderation in Belief’), Ghazali’s ­exposition of Ash‘ari theological argument, he does exactly that, but not just for eternal life and knowledge. ‘It is necessary that whoever is a knower and powerful is also living,’ he argues, ‘since we do not mean be “living” a­ nything other than someone who is aware of himself and knows his essence and that which is other than himself’ (2013: 105). The real difference between them on this point is not, therefore, that Ghazali attributes animal functions to God by failing to identify life and knowledge, but instead that Ibn Rushd sees them as essential to temporal life, while Ghazali does not. For Ibn Rushd, likewise, ‘the meaning of “will” in man and in animal is a desire which rouses movement and which happens in animal and man to perfect a deficiency in their essence,’ which is obviously impossible for God. In postulating an eternal will, he argues, the Ash‘ari imagine a sort of desire that is the cause of an act and yet remains in the same unchanged state before, during, and after the act takes place – an absurd ‘eternal version’ of something essentially temporal. Therefore according to the philosophers the meaning of ‘will’ in God is nothing but that every act proceeds from Him through knowledge, and knowledge insofar as it is knowledge is the knowledge of opposites, either of which can proceed from Him, he explains. ‘And the Knower is called excellent by the fact that there always proceeds from Him the better of the opposites to the exclusion of the worse’ (1954: 257). His reasoning is presumably free of any objectionable analogy between the temporal and eternal, and can therefore serve as a model of what he takes to be a valid inference from the one to the other. Indeed, it turns on a radical difference between eternal and temporal knowledge that is a ­recurring theme for Ibn Rushd. ‘That is because existence is the cause and reason of

86  Edward Omar Moad our knowledge,’ he explains in Faṣl al-Maqāl (‘Decisive Treatise’), ‘while ­eternal knowledge is the cause and reason of existence’ (2001: 40–41). He thus ­identifies God’s knowledge with His power, precisely as Ghazali (who insists the two are distinct) defines the latter; that is, the attribute through which the act proceeds (2013: 84). ‘Is what is understood by our saying “powerful” the same as what is ­understood by our saying “knower” or is it different from it?’ Ghazali ­challenges the falāsifa in the Iqtiṣād. ‘If it is exactly the same, then it would be as if we said that God is powerful, which is sheer redundancy’ (2013: 131). Modern analytic readers will recall that, though identical, ‘Hesperus is Phosphorous’ is not necessarily redundant (Frege 1892). Ibn Rushd need not then conclude that ‘knowledge’ and ‘power’ always mean the same for us, despite their unity in the eternal; and insistence on that need not be based simply on defining the word ‘knowledge’ as ‘the attribute through which the act proceeds.’ Since for Ibn Rushd the fact of their unity is a s­ ignificant philosophical discovery, we must at least initially understand each term ­ differently. ‘If they are different, then that is what is desired,’ as far as concerns Ghazali. ‘Hence, you have affirmed two concepts: one of them is expressed as power and the other as knowledge’ (2013: 131). Perhaps affirming two concepts does not preclude discovering, as Ibn Rushd contends, that the eternal power through which things come to be is, properly understood, knowledge. Yet, that raises the question how we know what ‘knowledge’ properly understood is, on which turns that of whether his argument depends on the same sort of analogy between the temporal and eternal, of which he accuses the Ash‘aris. 5.2

Divine Will and the Metaphysics of Modality

For now, we remain on the question of eternal will, which for Ibn Rushd ­connotes merely the fact that everything proceeds from eternal knowledge, to the exclusion of its known opposite. In this sense, it plays no role in ­determining which of the two proceeds, for the fact (itself going by the name ‘excellence’) is that only the better of the two ever does. In another sense of the term that Ibn Rushd subtly introduces, it does. Contrary to Ghazali’s construal, he explains, the falāsifa do not hold that God’s act proceeds ­ ­necessarily from knowledge, for if so, then opposites would co-exist. The fact that only one of the opposites proceeds from Him shows that there is another attribute beside knowledge, namely will, and it is in this way that the affirmation of will in the First must be understood ­according to the philosophers. (1954: 264) These (‘will’ and ‘excellence’) are not, however, divine psychological ­faculties or personality traits, but features of the world according to two points of

Divine Power and Divine Knowledge  87 Aristotle’s physics, and one which Ibn Rushd draws from his metaphysics. The first two are that each thing comes to be from its opposite (‘will’), and that it does so to manifest the good as determined by its telos (‘excellence’). The third is that each thing ‘proceeds from knowledge.’ The ‘attribute ­present beside knowledge,’ due to which one opposite comes to be at a time, is simply the specific nature of the pre-existing matter, in accordance and fulfillment of which change occurs. Aristotelian physics, some may object, is rife with anthropomorphic ­analogy. Yet, if so that is just an analogy between one temporal and another, and does not amount to the sort of analogy between temporal and eternal with which Ibn Rushd imputes the Ash‘aris. We may defer questions over what we are likely to find lurking under his idea that things proceed ‘from knowledge.’ For there is at least nothing here as explicitly analogical as ­arguing that ‘our acts proceed from a desire; therefore God’s proceed from an eternal desire,’ which is what we would expect to find Ghazali saying, given that Ibn Rushd is responding here to him. Instead, Ghazali begins with the observation of the very fact which Ibn Rushd names as ‘eternal will.’ Some things exist, to the exclusion of their opposites. The specification of one over another requires a determining ­factor. This, he argues, can be neither God’s essence, power, nor knowledge. Since His essence is the same with respect to both existent and non-existent possibilities, it alone cannot explain why some rather than others exist. God’s power cannot be the determinant, for it extends equally to both existent and non-existent possibilities (He can enact all of them, including those He does not). As for knowledge, we might expect Ghazali to argue likewise. If, as Ibn Rushd said, God knows all the existents as well as their non-existence ­opposites, then His knowledge of them cannot be what determines which exist. Instead, Ghazali aims at the core of the falāsifa theory of eternal knowledge. ‘For knowledge depends on what is known,’ he says, ‘attaches to it as it is, and does not affect or change it’ (2013: 105). From this denial of the falāsifa proposal that eternal knowledge is the cause of its object, it follows that nothing about God’s mere knowledge of all possibilities will affect the actualization of some rather than others. Instead, His knowledge of which do and do not exist depends on the independently determined fact of that matter, which therefore requires a distinct attribute, ‘will,’ as the cause. ‘If it were possible for knowledge to be sufficient on its own without the need for will,’ he argues, ‘then it would be sufficient on its own without the need for power’ (2013: 106). Not, however, if eternal knowledge and power are one, as Ibn Rushd contends, which is just what is in question. Contrary to Ibn Rushd’s construal, however, Ghazali’s definition of will involves none of the temporality and desire associated with temporal will. Will is simply ‘an attribute whose function is to distinguish one thing among its counterparts’ (2013: 109). Thus, he answers the challenge posed by questioning why the will distinguishes one over another, and asserting that, since

88  Edward Omar Moad either is equally possible in relation to will, it also needs a determinant. On the contrary, Ghazali says, the buck stops with will, which simply is that which distinguishes one from another equivalent possibility in the absence of any further determinant. The endgame of this challenge, as deployed by the falāsifa, is their position that everything necessarily is as it is. For, as Ibn Rushd put it, only the best proceeds from the Knower and ‘eternal will’ is just His knowledge of this fact. Any other possible world would be inferior. In the final analysis, only the actual world is possible. Ghazali’s objection to this, posed first in the Tahāfut, is that there are ­features of the actual world that could have been otherwise, without any compromise in the perfection of its order. The heavenly spheres, he argues, might all rotate in the opposite direction without compromising their ­perfection. They might have rotated on different axes. All points on a sphere being, according to falsifa cosmology, equivalent, every two opposite points can have served just as well as any other. Nothing about the requirements of an optimally perfect order can serve as a determinant explaining why the actual state of affairs obtains, rather than the other possibilities equivalent in these respects. The fact that it exists at all necessarily indicates the existence of a principle that determines one from among alternative possibilities that are themselves equivalent in relation to any other possible determinant. That is, he says, eternal will (1954: 20–25). Ibn Rushd, recall, reduced God’s ‘will’ to the fact that things proceed from Him to the exclusion of their opposites, ‘either of which can proceed from Him.’ For him, however, that either can proceed means that both do, though at different times. For something is possible, for Ibn Rushd, just in case it exists at some time and not another. If it never exists, it is impossible; and if it always exists, it is necessary. Following Aristotle, this more recently dubbed ‘temporal-frequency’ model of modality (from which follows the ‘principle of plenitude’) accompanies the theory of potency, according to which the possibility of the possible is a function of the specific potency of an actually existing thing. That is, what exists sometimes, does so through a change in a previously existing thing possessing a nature by virtue of which it is able to undergo that change under certain conditions (Hintikka 1973). ­ ecessity Thus, the ‘alternative possibilities,’ which for Ghazali indicate the n of his ‘eternal will’ are for Ibn Rushd not possibilities at all. For Ghazali’s imagined heavens, rotating in opposite directions or on different axes are just not the existing heavens, and since the natures of the existing heavens do not include the potential for such a change, it never happens, and is ­therefore not possible (1954: 25–34). Faced with the question how things can proceed from eternal knowledge alone, if that knowledge includes equally the ­existents and their opposites, he would simply reply that ­inasmuch as both are ­possible, both exist in their time. Knowledge is of the real, and the absolute ­nonexistent (that is, for Ibn Rushd, the impossible) is not an object thereof.

Divine Power and Divine Knowledge  89 Ghazali’s notion of eternal will, therefore, rests not (at least directly) on an analogy to temporal will, but on a distinct conception of possibility, ­according to which something is possible just in case it can be coherently supposed to exist. In the Tahāfut, he represents the falāsifa as arguing, that had the cosmos a beginning its existence must have been a prior possibility necessarily inhering in some pre-existing matter – a previous state of the cosmos, which thus has no beginning. ‘The objection is that the possibility of which they speak is a judgment of the intellect. Anything whose existence the intellect supposes, provided no obstacle presents itself to the supposition, we call possible, and if there is such an obstacle, we call it impossible, and if we suppose that it cannot be supposed not to be, we call it necessary,’ he argues. ‘These are rational judgments which need no real existent which they might qualify’ (1954: 60). That is, the sense of ‘possibility’ in which it is self-evident that the ­cosmos must be possible prior to any hypothetical beginning is distinct from the sense in which the possibility of something requires pre-existing matter. Ghazali’s own explanation, echoing Ibn Sina’s distinction between the ‘necessary in itself’ and ‘necessary through another,’ is that ‘it is possible for a thing to be both possible and impossible, in the sense that it is possible by virtue of its essence and impossible by virtue of something other than itself (2013: 87). For Ghazali, that something is not pre-existing matter, but divine will. The possible by virtue of its essence is that the existence of which we can suppose, considered in itself. The impossible in that sense, ‘consists in the simultaneous affirmation and negation of a thing, or the affirmation of the more ­particular with the negation of the more general, or the affirmation of two things with the negation of one of them…’ (1954: 329). This allows for the conception of ‘synchronic’ alternative possibilities that never obtain, and thus a departure from the essentially ‘diachronic’ Aristotelian modality, whereby possibilities are ultimately future actualities, as determined by past actualities (Kukkonen 2000). Ghazali’s conception of eternal will is the principle determining the existence of some rather than other alternative possibilities of this sort, and which is therefore no feature of any contingent actuality. The real issue between him and Ibn Rushd here is not, then, over whether God is an ‘eternal man,’ but over which conception of possibility represents real metaphysical possibility. ­ odality That is, in contemporary terms, the debate over whether ‘nomic’ m is metaphysical, as carried on, for example by Shoemaker (1980) and Bird (2005) on one side and Fine (2002) and Sidelle (2002) on the other. Unlike them, the falāsifa and mutakallimūn both shared the position that the ­existence of the temporal and contingent proves that of an eternal and necessarily existing first cause (though they differed over the nature of the effect, the cause, and the proof). In their context, rival conceptions of m ­ odality consequently lead to divergent conclusions about whether (and in what ­ sense) we may understand the first cause as possessing ‘will.’ Among parties

90  Edward Omar Moad who reject drawing conclusions about God by analogy to humans (or any temporal beings), this should be the ideal procedure, as opposed to drawing conclusions about metaphysics from premises about the divine nature drawn by analogy to the temporal. 5.3

Divine Knowledge and Power

That raises the question how, according to Ghazali, we know that a willer necessarily knows the object of will? This, recall, is the proof of God’s ­knowledge of other than Himself, which Ghazali claims the falāsifa forfeit by virtue of their necessitarianism. Yet, the concept of eternal will he has offered is simply of a principle whose function is to distinguish one from among equivalent possibilities. If it were to distinguish the ‘best’ ­alternative, as Ibn Rushd has it, then we might see how the necessity of knowledge follows. Yet, when we define ‘will’ as functioning precisely where no such evaluation is possible, then it is unclear whether knowledge has any possible, much less a necessary role. This leaves reason to suspect it is a role found in our experience of temporal will, and hoisted by analogy onto God. Human acts of will seem to involve some knowledge of what we want to do, but that does not constitute evidence that an eternal principle distinguishing between equivalent possibilities involves knowledge. Quite the opposite, given that the known object in our experience is one of desire. This motivates a closer look at how Ghazali takes divine knowledge as proven (and what it is). In the Tahāfut, Ghazali connects the ability to prove divine knowledge to his rejection of the falsifa doctrine of cosmic pre-­ eternity. In the process, they argued that since all moments of time prior to that hypothetical cosmic genesis are the same, there is no conceivable reason that it should begin in one rather than any other, and thus impossible that it would. Hence, for Ghazali, the function of eternal will as he conceives it. The price of denying that, he argues, is that the falāsifa have no valid proof that God knows anything, since, as he asserts here without elaboration, the willer necessarily knows what he wills. When, however, we turn to Ghazali’s positive defense of Ash‘ari creed in the Iqtisād, we find no such assertion. Instead, he proves divine knowledge prior to proving divine will. ‘It is known that He knows that which is other than Himself, since what is called “the other” is His well-designed handiwork and His exquisite and well-ordered act,’ he argues, ‘and this proves the knowledge of the Maker as well as it proves His power, as previously shown (2013: 104). That proof, given for divine power, is: 1 Every well-designed product is produced by a powerful agent. 2 The world is a well-designed product. Therefore, the world is produced by a powerful agent (2013: 83).

Divine Power and Divine Knowledge  91 This is clearly his core argument for divine power, will, and knowledge ­altogether. For it concludes that the world is produced by a powerful agent, whereas his earlier proof of God’s existence concludes that the world has a cause. In the Tahāfut, where Ghazali accuses the falāsifa of obfuscation in their claim to affirm that God is the agent of the world, he insists that the term ‘agent’ connotes a cause that wills and knows what it wills (1954: 89). Thus, Ghazali takes the agency of God (entailing not merely His being the cause of the world’s existence but all three attributes in question) as proven by the fact of the world’s being well-designed (defined as its ‘excellent organization, excellent composition, and exactness’), known, he asserts, by ­observation (2013: 83). When he seems to defend the proof’s first premise, however, the e­ xpression changes. ‘How do you know the other principle, which is that the agent who produces a well-designed, well-ordered product is powerful?’ (2013: 83). This is not the other principle, but a new one. For this takes as given that the producer is an agent, questioning only whether it is powerful. That is, specifically, whether power is an attribute distinct from its essence. For Ghazali’s answer is that, were the world to proceed through God’s essence, then it would also be eternal. Since he takes himself to have already proven (in the proof of God’s existence) that the world is not eternal, he concludes that it proceeds through a distinct attribute – power. ‘For “power,” according to linguistic usage, designates an attribute through which the agent becomes prepared for the act, and the act comes to fruition’ (2013: 84). He does not mean that God ‘becomes prepared’ after having been ­unprepared, for God’s power is eternal. The problem this appears to pose for the temporality of the world, he resolves by the attribute of will, through which the actual temporal limits of the world are determined from among the alternative possibilities thereof. While will determines ‘when’ the world exists, God is capable of creating it ‘before’ its existence. Yet, in this context, we must understand ‘before’ non-temporally. For according to Ghazali, there is no time prior to the existence of world. As the measure of its motion, time began with the world (Moad 2015). Thus, the eternity of God’s power amounts to its ontological independence of the world and its time – the fact that the world’s possibility is independent of its actuality. If in this way, the function of divine will renders the temporality of the world compatible with the eternity of divine power, then why would it not also make it compatible with the proposition that the world proceeds from God’s eternal essence? For the proof Ghazali gave here is just that if it had then, God’s essence being eternal, the world would also be. The answer is that Ghazali has not here expressed the root concern, which is not merely the temporality, but the contingency of the world. If it proceeds from God’s essence then it would necessarily be as it is in every respect. That is, there would be no alternative possibilities. The attribute of power as a distinct attribute follows from the fact that there are. It follows, one might say, from the fact that the world could have been different, or not have existed at all,

92  Edward Omar Moad and not simply from the fact that, once upon non-time-before-time, it did not (or one might say that these two statements mean the same). Will, to clarify, follows (and remains distinct from power) not merely from the fact of alternative possibilities, but that one among equivalent a­ lternatives actually exists. For power too, as we see, Ghazali has not based his proof on analogy to temporal, human power. On the contrary, what proves that a creature’s power is distinct from its essence is that one may gain or lose a power while remaining itself. The falāsifa would object that this applies only to its accidental powers, and not the essential power that constitutes its intrinsic nature. Without power to burn, for example, fire will not be fire anymore; otherwise, what does it even mean to be fire (1954: 318–319)? Yet, even if so, to conclude from this, that God’s power is essential in the same sense, is another unsound argument by analogy from the temporal to the eternal. That, according to Ibn Rushd, is the mistake that al-Fārābī and Ibn Sīnā made in concluding that only one thing can proceed from God. Their ­principle that ‘from one only one can proceed’ applies to the empirical agent. Fire qua fire has only the one act of heating, for any other (such as producing smoke, or a mark) arises not from its singular essence, but from something else to which it is related (such as wood, or the shape of a heated metal brand). Yet, to apply this principle to God is invalid. For the first agent in the divine world is an absolute agent, while the agent in the empirical world is a relative agent, and from the absolute agent only an absolute act which has no special individual object can proceed. (1954: 108) Thus, he absolves the ‘true philosophers’ from the error that leaves al-Fārābī and Ibn Sīnā open to Ghazali’s critique, that their assimilation of God’s action to the natural agent is as unsound an analogy as that of assimilating it to the temporal voluntary agent. This is an example of internal critique among the falāsifa that, along with similar examples among the Ash‘aris, illustrate the oversimplification involved in construing the difference between ­‘philosophers’ and ‘theologians’ as simply over anthropomorphism (2013: 55–62). If the order of the world proves that it has an agent, then exactly how do each of the specific attributes constitutive of agency follow? Ghazali does not explain this in detail. In the case of will, we may suppose, the premise left implicit is that any complex order exhibits the specification from among alternative possibilities that requires divine will. The case of knowledge is not as simple, despite the appearance of Ghazali’s argument. ‘For if one sees well- arranged lines that proceed with precision from a scribe, and then doubts that the scribe is knowledgeable of the art of writing, he would be

Divine Power and Divine Knowledge  93 foolish to have such doubts,’ he writes. ‘It has thus been established that God knows Himself and what is other than Himself’ (2013: 104). This initially appears to turn on an analogy between the temporal and the eternal: since a scribe knows his art, God does also. Alternatively, we could read it as an analogy between temporal things: the writing proceeding from a scribe, and the well-ordered things of nature. Since we would be foolish doubt that the scribe is knowledgeable of his writing, the implied conclusion on this reading is that it would also be foolish to doubt that the well-ordered world proceeds from a knowledgeable agent. We have two good reasons to read it this way, the first being that the principle of charity ­mitigates against ­interpreting a scholar of Ghazali’s caliber as drawing a direct analogy between a scribe and God when a better interpretation is available. The second, more substantial reason is that the premise rests not on the condition of seeing a knowledgeable scribe, but of seeing ‘well-arranged lines that proceed with precision from a scribe.’ The appropriate analogue to that is the rest of the order we observe in the world that, just as the order in ­intelligible writing, indicates a knowledgeable source. In each case, the nature of what proceeds leads us to infer the presence of knowledge in that from which it proceeds (scribe or otherwise), not the presupposed fact that one of two analogous things is knowledgeable. While Humean objections to this sort of argument rightly come to mind, we are immediately interested here with issues specific to the context at hand. One of those is that according to Ghazali (on a standard interpretation), in one sense neither writing nor anything else ‘proceeds from’ a scribe. As Yaqub (the author of this translation) notes, the expression ‘proceeds from’ (taṣdur ‘alā) carries here the sense of being acquired by rather than ­created by the scribe, in line with the Ash‘ari doctrine of ‘acquisition’ (kasb) whereby temporal agents only acquire, but do not create their acts (2013: 104, n. 48). In the Tahāfut our scribe turns up in a different, yet illuminating, ­context. Objecting to Ghazali’s denial of causal necessity, the falāsifa mention some absurd implications, challenging him whether they would not then be ­possible. One of these is a ‘zombie’ scribe. As to God’s moving the hand of a dead man, and raising this man up in the form of a living one who sits and writes, so that through the movement of his hand a well-ordered script is written, this is not impossible as long as we refer events to the will of a voluntary being, and it is only to be denied because the habitual course of nature is in opposition to it. (1954: 330) In this proof for divine knowledge, then, perhaps Ghazali means that it would be ‘foolish to doubt’ that the scribe is knowledgeable, given that ­zombie scribes do not appear in the habitual course of nature (though we must now ask how we could know that). In any case, what follows necessarily (as

94  Edward Omar Moad opposed to what would merely be foolish to doubt) from the orderliness of the writing, and not from the fact the scribe’s hand is moving, is its procedure from an agent (a knowing and willing power), and not that the agent is in fact the scribe whose hand is moving. Ghazali clearly considers the identity of the agent in the scenario as ­irrelevant to this proof, whether it be the scribe, God, or both in some way. If so, and the orderliness of the writing alone proves it’s procedure from an agent, apart from any consideration of its appearing to proceed from a knowledgeable scribe, then the fact that we do not observe any such appearance in the case of nature’s order does not alone constitute a difference ­relevant to the analogy’s soundness. Ghazali might ask how we know the scribe, or any supposed observable agent, is knowledgeable, if not by the orderly nature of what proceeds from her? That alone, then, is the sign of knowledge. Ghazali’s argument for divine knowledge, so extrapolated, turns out not only to be free of any unsound analogy between the temporal and eternal but also potentially defensible against Humean-style objections to analogies between nature and artifacts. Alternatively, we might analyze the argument as involving no analogy, and asserting simply that the knowledgeable agent follows necessarily from the observation of order. The absence of any observable agent in most cases (the ‘natural’ rather than ‘artificial’) often obscures that fact. Since, as argued, we only ever infer the knowledge of the observed agent from the orderliness of the act, removing only the observation of a supposed agent does not remove the evidence of agency. Comparing the case where we observe scribes to cases with no such observation merely clarifies the irrelevance of this difference to the direct evidence of the presence of agency in both. Standard analytic procedure is to interrogate the nature of knowledge and order to pin down precisely how order in the product proves knowledge in the producer. Ghazali does not define knowledge here, other than to say that ‘the only meaning of knowledge is that it is what necessarily reveals its object’ (2013: 109). That does not explain why an ordered product is necessarily revealed to its producer. Neither does his assertion that ‘knowledge depends on what is known, attaches to it as it is, and does not affect or change it’ (2013: 108). That, however, is a pillar of Ibn Rushd’s charge that the Ash‘aris have conceived God as an ‘eternal man.’ For only temporal knowledge depends on its object, according to him, while eternal knowledge produces it. Hence, again, its identity with, as opposed to distinction from divine power. That puts the ball in Ghazali’s court. He seems to have two options. One is to bite the bullet and defend the soundness of inferring by analogy that, since our knowledge depends on its object then so does God’s. The other is to provide an account of divine knowledge free of any such analogy, as he did for divine will, showing both how divine knowledge depends on its object,

Divine Power and Divine Knowledge  95 and how it follows from the order of its object. Ibn Rushd, meanwhile, does provide an account, according to which divine knowledge follows from the order of its object, in virtue of its being, in one sense the cause of its object, but in another sense (in which the order of the object is understood as its very being) ultimately identical to it. He provides brief, variant summaries of it in the third and sixth discussions, respectively, of the Tahāfut (1954: 126–130, 216–218). While the space of this chapter is not adequate to fully explore or evaluate this argument, we touch on it in the hope of inspiring closer study. The account is of course rooted in the Aristotelian hylomorphic system, according to which objects of perception are either sensible or intelligible, with the natures of things being of the latter sort (1954: 216). Temporal knowledge is a process whereby the mind renders that intelligible nature, or form, abstract from the matter of the individual thing known thereby. Change is a process whereby matter exchanges one form for another: an essential form when one thing of a different nature replaces another and an accidental form when something changes while maintaining its essential nature. That the distinction between essence and attributes amounts only to this is, according to Ibn Rushd, why the falāsifa insist that it applies only to the temporal. Thus, ‘the First could not be composed of attribute and subject, and since everything free from matter was according to them intellect, it was necessary that the First should be intellect’ (1954: 218). Ibn Rushd’s proof of divine knowledge is, in brief, that the order ­manifest in the very intelligibility of things in the world is itself divine knowledge. A key premise here is the Aristotelian principle that actual intellect is identical with its object (De Anima III: 4). As Ibn Rushd explains in his Long Commentary on De Anima, ‘if it has its own form, then that form will impede its receiving other extraneous forms because they are other than it’ (2009: 301). The intelligible form, purely abstract from matter (as opposed to merely ­realized in material intellects like ours) is that perfect knowledge of itself, which constitutes its innermost nature, and not a distinct object thereof. For this reason, we cannot positively represent this reality in our material intellects, though rational inference leads to it. Thus, Ibn Rushd’s language a­mbiguously ­distinguishes and unites divine knowledge and essence. The First possesses a quiddity that exists absolutely, and all other e­ xistents receive their quiddity only from it, and this First Principle is the existent which knows existents absolutely, because existents become existent and intelligible only through the knowledge this principle has of itself; for since this First Principle is the cause of the existence and intelligibility of existents, of their existence through its quiddity and of their intelligibility through its knowledge, it is the cause of the existence and intelligibility of their quiddities. (1954: 219)

96  Edward Omar Moad Ghazali’s response to the identification of God’s essence and knowledge is just to call it foolish, ‘for knowledge is an attribute and an accident which demands a subject.’ According to Ibn Rushd, however, it is Ghazali who is confused, for it has been proven that there is among attributes one that has a greater claim to the term ‘substantiality’ than the substance existing by itself, and this is the attribute through which the substance existing by itself becomes existing by itself. (1954: 220) The debate here is unsatisfyingly inconclusive, and this represents an ­ pportunity for further study. Ideally, that would involve a more detailed o reconstruction of Ibn Rushd’s theology and its basis in his reading of Aristotle’s psychology and metaphysics, along with, if possible, discovery of the ­alternative account (or elements thereof) of divine knowledge that are missing here, either from Ghazali’s or a wider reading of Ash‘ari work. Absent that, contemporary scholars working in that tradition might c­ onstruct an original account, or perhaps question the very premise that the God-man analogies are categorically unsound. 5.4 Conclusion The conclusion of this chapter is meager. That is, that the conflict between the falāsifa and the Ash‘ari over the divine attributes, at least as it culminates in the Tahāfut debate, turns on this question over the nature of knowledge and is not, therefore, merely about whether analogies of any sort are sound between God and humans. Yet, the challenge for Ghazali, of defending the Ash‘ari position from the charge of resorting to such analogies, requires more than simply rejecting the identification of God’s knowledge with His essence, by asserting that knowledge is an attribute. It calls instead for an alternative account of both temporal and divine knowledge more substantial than we have discovered in these sources. References Al-Ghazālī, Abū Ḥāmid. (2013). Al-Iqtisād fī-al i’tiqad (‘Moderation in belief’). Translated by Aladdin Yaqub. Chicago, IL: University of Chicago Press. Bird, A. (2005). ‘The Dispositionalist Conception of Laws’, Foundations of Science, 10: 353–370. Fine, K. (2002). ‘The Varieties of Necessity’, in Conceivability and Possibility, T. Gendler and J. Hawthorne (eds.), 253–282. Oxford: Clarendon. Frege, Gottlob. (1892). ‘Über sinn un bedeutung (‘On Sense and Reference’)’, In Zeitschrift für Philosophie un philosophische Kritik, 100: 25–50. Translated by M. Black (1948) in The Philosophical Review 57: 207–230.

Divine Power and Divine Knowledge  97 Hintikka, Jaako. (1973). Time and Necessity: Studies in Aristotle’s Theory of ­Modality. Oxford: OUP. Ibn Rushd, Abū al-Wālid. (1954). Tahāfut al-tahāfut (‘Incoherence of the i­ ncoherence’). Translated by Simon Van Den Bergh. Cambridge: Cambridge University Press. Ibn Rushd, Abū al-Wālid. (2001). Kitāb faṣl al-maqāl (‘Decisive treatise’). Translated by Charles Butterworth in Decisive treatise & epistle dedicatory. Provo: Brigham Young University Press. Ibn Rushd, Abū al-Wālid. (2009). Long Commentary on the De Anima of Aristotle. Translated by Richard Taylor. New Haven, CT: Yale University Press. Kukkonen, Taneli. (2000). ‘Possible Worlds in the Tahāfut al-Falāsifa: Ghazali on Creation and Contingency’, Journal of the History of Philosophy, 38/4: 479–502. Moad, Edward. (2015). ‘Al-Ghazalis Position on the Second Proof of the ­Philosophers for the Eternity of the World, in the First Discussion of the Incoherence of the ­Philosophers’, Sophia, 54/4: 429–441. Shoemaker, S. (1980). ‘Causality and Properties’, in Time and Cause, P. van Inwagen (ed.), 109–135. Dordrecht: D. Reidel Publishing Company. Sidelle, A. (2002). ‘On the Metaphysical Contingency of Laws of Nature’, in ­Conceivability and Possibility, T. Gendler and J. Hawthorne (eds.), 309–336. ­Oxford: Clarendon.

6 The Doctrine of Divine Simplicity Considered in Light of Averroes’s Conception of Divine Knowledge Bakinaz Abdalla The 12th-century Andalusian philosopher, Abu al-Walid Ibn Rushd (Averroes), stands out as one of the most rigorous Muslim philosophers whose faithfulness to Aristotelianism has promoted questions about his actual attitude to the teaching of Islam. While Averroes emphatically states that the truth does not contradict the truth, referring to the truth of philosophy and the truth of religion (Faṣl al-maqāl 1969, 31–32), the purported harmony between Aristotelianism and the teachings of Islam is hard to delineate from his writings. It takes no effort to recognize that the Averroist-Aristotelian conception of God differs from the one imparted by Scripture. Despite the differences, both stress oneness as the essential feature of God. In Averroes’s writings, divine oneness tawḥīd/waḥdāniyyah comes hand in hand with simplicity. This chapter will show that the doctrine of divine simplicity (henceforth DDS) collapses in the course of Averroes’s endeavour to establish God’s knowledge of existents and its causal relation to the world. As a result, Averroes is rendered vulnerable to ­critiques of theism dwelling on similar instances of conceptual incoherence. The ­contemporary opponent of religion Richard Dawkins raises one such critique, taking into consideration the assumption that any cause of a diverse universe would involve multiplicity, and therefore would require ­further explanation for its existence. Dawkins’s critique implies the failure of the DDS, typically ­considered the crucial explanation of God’s ontological independence. After discussing the DDS, its Islamic and Averroist contexts, and highlighting problematic issues with Averroes’s conception of God’s knowledge, I will present possible solutions, drawing insights from contemporary ­discussions of DDS. 6.1

Tawḥīd, Simplicity, and Ontological Independence

The fundamental Islamic tenet of tawḥīd (divine oneness) is affirmed through the shahādah, the Muslim declaration of belief, ‘I bear witness that there is no god but God (Allah) and that Muhammad is His messenger,’ and ­several Quranic verses explicitly stating that God is one. The expression of the shahādah indicates that one’s affiliation with Islam hinges on a fundamental DOI: 10.4324/9781003327714-10

Doctrine of Divine Simplicity  99 requirement, to believe—irrespective of the criteria for belief—or simply verbally confess that there is no other god besides God. Other features of God, like His individuality as the Creator, dissociation from partnership or kinship, and superiority to all creatures are integral to the Islamic tenet of tawḥīd, as can be seen from verses that pair God’s ­oneness with these divine features. Consider, for example, the concluding verses of Surat al-Ḥashr. He is Allah—there is no god worthy of worship except Him: Knower of the seen and unseen (22). He is the Most Compassionate, Most Merciful. He is Allah—there is no god except Him: the King, the Most Holy, the All-Perfect, the Source of Serenity, the Watcher of all, the Almighty, the Supreme in Might, the Majestic. Glorified is Allah far above what they associate with Him in worship! (23) He is Allah: the Creator, the Inventor, the Shaper. He alone has the Most Beautiful Names. Whatever is in the heavens and the earth constantly glorifies Him. And He is the Almighty, All-Wise (24). (English Translation by Mustafa Khattab) In these verses, God reveals His nature to mankind by means of the most essential tool for communication among human beings: language. Considered in light of the face value theory of meaning (for this theory see Scott 2013: ix), the verses form a cluster of descriptive utterances with specific propositional contents. In addition to the core message of the shahādah, namely that (1) there is no god but God (Allah), the verses also convey the (particularly religious) fact that (2) Allah alone has the most beautiful names/attributes. Each of these names informs something about God. Accordingly, (3) God is All-wise, (4) God is Almighty, and (5) God is Most Compassionate. Viewed in this manner, there is arguably no conflict between the doctrine of tawḥīd, as expressed in utterance 1, and utterances 2, 3, 4, and 5. However, things are not so simply conceived in the Islamic philosophical and theological traditions. The concept of tawḥīd has expanded far beyond numerical oneness to incorporate internal oneness, or unity. In philosophical discourse, the latter is also known as the doctrine of divine simplicity, the doctrine that God is devoid of any form of composition or complexity— throughout this paper ‘the doctrine of divine simplicity’ (DDS) is used synonymously with ‘internal unity,’ the second conceptual component of tawḥīd. With tawḥīd conceived as essentially involving numerical oneness and internal unity, simplicity, it has become a complex doctrine, one that hardly reconciles with the belief of God as having a multiplicity of beautiful names/ attributes. Consequently, the apparent coherence of the Islamic conception of God disappears. Medieval Islamic philosophy and theology demonstrate a strong ­adherence to the DDS,1 because it was believed to undergird the Quran’s denial of any resemblance between God and creatures, as expressed in verse 11 of surah 42: laysa ka methlihi shayyʾun (Adamson 2003; Yaqub 2010). As thought,

100  Bakinaz Abdalla tawḥīd ­involves belief not only in the numerical oneness of God, but also His ­uniqueness. ­Unlike all creatures, God is distinguished by the unique ­characteristic of internal unity which means that His essence is free of any ontological constituents or parts, including conceptual parts that specify ­differences between genera and species. Nor does God’s essence involve any division, such as the division between essence and existence, potentiality and actuality, form and matter, and essence and attributes (al-Ghazālī [Maqāṣid] 2000, 104–108; al-Kindi [Rasāʾil Falsafiyyah] 1999, 201–207; Avicenna [al-Shifāʾ] 1960, 43–47). Furthermore, Muslim philosophers perceived absolute simplicity to be ­intimately related to God’s ontological independence. This is best reflected in Avicenna’s proof of God’s existence, known in the Islamic tradition as Burhān al-ṣidiqīn (the Proof of the Righteous). Proceeding from a division of ­existents into (1) that which is necessary in virtue of itself and (2) that which is ­contingent in virtue of itself, but necessary in virtue of another (Avicenna [al-Najāh] 1938, 227–229), Avicenna argues that the latter is not self-­sufficient by its very essence; it requires a cause that preponderates its ­existence over non-existence. Considering all contingent existents as a chain, all existents must terminate in a being that is necessary in itself (given the impossibility of infinite regress). That being does not require a cause or a principle for its e­xistence, but its existence is entirely due to itself. The trait of necessity entails a number of traits that permit identifying the necessary existent with God (see Adamson 2013, 170–189). Among these are non-compositeness, i.e., internal unity or simplicity, and external oneness. It is argued, since the necessary existent is uncaused, it has no parts, including conceptual parts; otherwise, it would be ontologically dependent on its constituents. Furthermore, it follows that the necessary existent is one. For supposing there are many necessary existents, say necessary existent A and necessary existent B, the essences of A and B would involve, in addition to the trait of ­necessary existence, respective differentiating features. But given that the necessary existent, as argued, lacks composition or division of conceptual parts, there cannot be a necessary existent that involves such differentiating features. This proves the numerical uniqueness of the necessary existent (Avicenna [al-Najāh] 1938, v.2 227-229).2 Overall, medieval Muslim philosophers acknowledged a logical ­connection between God’s uniqueness and ontological independence, on the one hand, and absolute simplicity, on the other hand. This connection was also e­ ndorsed in Christian and Jewish philosophical speculations about God (see e.g., Aquinas [Summa Theologiae] 1981, I.3.4; Maimonides [The Guide of the Perplexed] 1963, I: 50–60). Yet the DDS, despite its centrality to fundamental theistic beliefs, may, under certain considerations pose a challenge to the belief in the existence of a self-sufficient God. In contemporary discourse, no philosopher denies the subtle aspects of the DDS. Elenore Stump, for example, describes the DDS as ‘the strangest and hardest’ of all the doctrines ‘regularly discussed by medieval theologians and accepted by them as part of orthodox religious

Doctrine of Divine Simplicity  101 ­ elief’ (Stump 1997, 250). Many attempts have been made to explain away b its ­difficulties. Equally, criticism has been advanced by both proponents and opponents of theism. Considering the contentious nature of the DDS, its value for ­establishing fundamental beliefs about God diminishes. The danger may not be so much inherent in the DDS per se, but in the fact that philosophers made it the basis for explaining God’s self-sufficiency and primacy as cause, thereby placing these essential qualities of God on a shaky foundation. Among contemporary theist philosophers, Alvin Plantinga stands out for his criticism of the DDS, primarily due to its inconsistency ‘with the claims of Christian theism at the most basic level’ (Plantinga 1980, 53). Plantinga highlights several problems. One problem is that the central claim of the DDS, namely that God is identical with all his properties, suggests that he is ‘a self-exemplifying property,’ which raises outrageous theological implications: ‘No property could have created the world; no property could be omniscient, or, indeed, know anything at all.’ If God is a property, then he isn’t a person but a mere abstract object; he has no knowledge, awareness, power, love, or life. So ‘taken the simplicity doctrine,’ in Plantinga’s view, is ‘an utter mistake’ (ibid., 47).3 Aside from theological concerns, opponents of religion might seize upon the problematic aspects of the DDS to challenge not only the notion of ontological independence but also the fundamental principle of theism: the existence of God. The long-standing assumption of classical theists regarding the ontological dependence of ‘every’ composite existent on its constituents, and which has necessitated adherence to the DDS, could be turned against the conception of the Deity as Uncaused Cause, should it be contested that the supposed Deity involves complexity. Richard Dawkins follows a similar strategy; his target is not the DDS per se but, more radically, the existence of God. In the God of Delusion, Dawkins introduces what he calls the ‘God ­hypothesis’ before repudiating it: ‘There exists a superhuman, supernatural intelligence who ­deliberately designed and created the universe and everything in it, ­including us’ (2006, 31). That God, Dawkins aims to conclude, ‘almost certainly’ does not exist. The crux of his argument is: ‘any God capable of designing a universe’ must be ‘supremely complex.’ Such a god ‘needs an even bigger explanation than the one he is supposed to provide’ (2006, 147). In this ­connection, Dawkins states that the idea that there exists an intelligent designer/creator of the universe is ‘self-defeating.’ Dawkins’ argument, though primarily aiming at making a case for ­atheism, contributes to intensifying a perennial complexity associated with the DDS. If it is true that God is absolutely simple, how could the manifoldness of the world be attributed to Him? Apparently, to say that God (1) is simple and (2) is the cause of a manifold world are inconsistent claims. However, a theist would be reluctant to abandon either claim. For her, God remains the cause of all existents. Furthermore, His simplicity is not a matter of dispute, as without simplicity His ontological independence would be compromised, and consequently the God-creature distinction would be undermined.

102  Bakinaz Abdalla To reconcile the v­ arious ­aspects of the ­concept of God many Muslim philosophers appealed to the Neo-Platonic scheme of emanation. This scheme seemed to them to satisfy the rule of uniqueness or one (qāʿidat al-wāḥid)), which dictates that a one-to-one relation must exist between cause and effect (Amin 2020; Kogan 1981, 1985, 249). Central to the emanation scheme is the principle ­maintaining that from the one, only one thing proceeds (henceforth ­principle 1). From this principle it follows that God, being simple, gives rise to only one entity: the first intellect. Unlike the First Principle/Intellect, the first ­emanated intellect, while being immaterial, is conceptually complex because it thinks its source and itself. Thanks to this conceptual multiplicity a more complex entity emanates from the first intellect which, in turn, gives rise to further complexity. Multiplicity increases downward across the chain of emanated beings, terminating with the Active Intellect, the tenth celestial intellect in the sequence (for this scheme see al-Ghazālī [Maqāṣid] 2000, 108; Avicenna [al-Najāh] 1938, v.2 251, 273–278). This account has faced objections due to its internal incoherence and the failure to convincingly explain how the the physical world and its ­manifoldness have resulted from an immaterial simple principle. To avoid these ­drawbacks, Averroes presents an alternative account that, as claimed, ­explains the manifoldness of the world, including the physical realm, in a flawless coherence with the simple nature of God. However, as we shall see, the ­purported harmony collapses as we consider other aspects of God’s ­relationship to the world, thereby falling prey to the multiplicity claim in which Dawkins’s argument against theism is rooted. As it seems, God can hardly maintain actual causal relation to or knowledge of the world ­unless multiplicity in His essence is assumed. Within the classical theoretical framework of the DDS, this assumption, in turn, jeopardizes God’s ontological independence. The present chapter aims to address and rectify this issue. 6.2 Averroes’s Conception of Divine Omniscience and Its Implication for Divine Simplicity and Ontological Independence In his Tahāfut al-tahāfut (the Incoherence of the Incoherence 1954, h ­ enceforth TT), Averroes devotes extensive discussion to problematic aspects of the Deity’s waḥdāniyyah. From the discussion, we can tell that waḥdāniyyah signifies the two components of tawḥīd: numerical oneness and internal ­ unity, i.e., non-compositeness. A good deal of the contention pertains to the second component, which will be the focus herein. For the sake of easiness, I henceforth translate waḥdāniyyah and derivative terminologies as divine simplicity, unless a specific variation needs be highlighted. Many problematic aspects of the doctrine of divine simplicity had been introduced by al-Ghazālī in his critiques of the philosophers in Tahāfut al-falāsifah (the Incoherence of the Philosophers). Al-Ghazālī had

Doctrine of Divine Simplicity  103 systematically critiqued the philosophers of Islam, exposing what he believed to be inherent inconsistency in their opinions, their failure to remain faithful to the principles of Islam along with their philosophical commitments, and, as a consequence, their vulnerability to the charge of unbelief. The scheme of emanation with its emphasis on principle 1 stands out as a major source of incoherence. As al-Ghazālī argues, this principle is violated at every stage of the emanation scheme. If, in accordance with the assumption of the ­philosophers, there is a one-to-one correspondence between cause and e­ ffect, the first emanated intellect would be inherently simple. What, then, is the ground of multiplicity in the first emanated intellect, with which ­multiplicity assumingly starts, and the subsequent emanated intellects? The ­response from Muslim philosophers provides ground for further criticism. Their ­argument posits that multiplicity stems from the dual conceptions of the first ­emanated intellect, as it apprehends itself and its Cause (the First Intellect). However, al-Ghazālī contends that this reasoning compels the philosophers to admit that a similar instance of multiplicity applies to the First Intellect or that the first emanated intellect is just as simple as the First Intellect. This contention is framed in the following question: is the first emanated intellect’s consciousness of its ­essence and cause is i­dentical with itself, or is it an a­ dditional property? If that c­ onsciousness is identical with the first intellect, then, it is simple, just like the First Intellect. If not, then, the First Intellect too must be complex, since it apprehends itself and other intellects (al-Ghazālī’ss Tahāfut al-falāsifah quoted in TT, 117–120). Averroes’s articulation of the issue in consideration combines critiques of both the philosophers and al-Ghazālī, the former for subscribing to ­non-­demonstrative theories that are alien to Aristotle, and the latter for occasionally misrepresenting the philosophers to create a ground for criticism. The principle maintaining that from the one, only one thing proceeds is among the non-demonstrative elements espoused by the philosophers. Due to the ­adherence to this principle, many errors occurred in the domain of m ­ etaphysics. To some extent, Averroes is convinced (at least insofar as TT is concerned)4 that such a principle cannot effectively establish a causal relation of God to the separate intelligences, on the one hand, and the entirety of the divine realm and the physical world, on the other hand, while maintaining divine simplicity. In that he does seem to align conceptually with al-Ghazālī. Yet, Averroes is not pro al-Ghazālī all along the line. While the latter thinks that philosophers cannot reconcile God’s causation of the world with His simplicity, Averroes, on the contrary, argues that simplicity is ­crucial for explicating God’s causal relation to the world and its manifoldness. ­Contrary to al-Ghazālī’s criticism, Averroes contends: from the One, ‘plurality arises’ (henceforth principle 2) (TT, 109). The philosophers, followed by al-Ghazālī, errored because they thoughtlessly analysed the God-world relationship against the yardstick of the scheme of causation in the empirical world. C ­ onsequently, they naturally concluded that, similar to a human agent from which only one effect proceeds at a time, from the One, only one thing proceeds. In

104  Bakinaz Abdalla contrast to this misconception, Averroes denies the p ­ resence of any actual ­similarities between divine and empirical agents, insisting that comparison between them is invalid except by way of equivocation—we will come to the notion of equivocation later. On this basis, he proceeds to establish principle 2, from the One, a plurality arises. This principle represents the authentic ­Aristotelian account of the world, as Averroes states. Everything whose existence is only effected through a conjunction of parts, like the conjunction of matter and from, or the conjunction of the elements of the world, receives its existence as a consequence of this ­conjunction. The bestower of this conjunction is, therefore, the bestower of existence. And since everything conjoined is conjoined through a unity in it, and this unity through which it is conjoined is conjoined must depend on a unity, subsistent by itself, and be related to it, there must exist a single unity, subsistent by itself and this unity must of necessity provide unity through its own essence. This unity is ­distributed in the different classes of existing things according to their natures, and from this unity, allotted to individual things, their existence arises, and those unities lead upwards to the Frist Monad (al-wāḥid al-ʾawwal)[…] by means of this Aristotle connects sensible existence with intelligible, ­saying that the world is one and proceeds from one, and that this Monad is partly the cause of unity, partly the cause of plurality […] It is evident therefore that there is a unique entity from which a single power emanates through which all beings exist. And since they are many, it is necessarily from the Monad, in so far as it is one, that plurality arises or proceeds. (TT, 108–109) This account presents a holistic picture of the world in which all existents are viewed as parts of a unified whole and their coming-to-be as an outcome of the unity permeating the world. Ontological dependence is an essential ­feature of composite things, given that their existence is intricately tied to both their constituting parts and the agents that confer unity upon these parts. Furthermore, the ­unification of the parts defines the process by which composite beings come to be. And, this process is effected through immediate causes/composers that, owing to their composition, also ontologically depend on their constituting parts and composers. All these composite causes and effects, which form the whole being, ultimately depend on a Principle that itself requires no composer for its e­ xistence. Thanks to the unity emanating from that Principle, the parts ­making up individual composite existents, in turn, the parts making up the world as a whole, are conjoined. And thanks to this Principle, multiplicity of things arises. While this account seems well-suited to reconcile the beliefs in God as absolutely simple/non-composite and as the source of multiplicity in the world, it faces a ­challenge when we consider another essential element of Averroes’s conception of God’s relation to the world: divine knowledge.

Doctrine of Divine Simplicity  105 Two interrelated claims lie at the heart of Averroes’s theory of divine knowledge: (1) God knows all existents and events in the world, such that no atom escapes His knowledge, (2) God’s knowing of existents is compatible with His perfection of which simplicity, immutability, and self-sufficiency are essential characteristics. To be sure, these two claims are not innovative and could be seen as central to speculations of most Muslims philosophers about God. Nonetheless, Averroes establishes them in a distinctly unique manner, with the aim of sidestepping implications associated with other philosophical accounts. As is often the case with many of Averroes’s metaphysical discussions, Avicenna’s thought lies in the background. Therefore, it is important to briefly delve into Avicenna’s theory of the concept of divine knowledge. Out of commitment to the standards of divine perfection, Avicenna argues that God knows particulars in a universal way (Al-Ghazālī [Maqāṣid] 2000, 118; Avicenna [al-Najāh] 1938, v.2, 247). This statement has led to ­various interpretations, and in some instances, it has been met with accusations of implicit denial of divine knowledge altogether (for different ­interpretations see Acar 2004; Adamson 2005; Belo 2006; Marmura 1962; McGinnis 2010, 172–177). For present purposes, suffice it to note that Avicenna, like most Aristotelians, held biases against knowledge of particulars, as opposed to universals, given its plurality and association with transitorysensory objects. These features stand at a stark contradiction with divine perfection. Accordingly, ascribing knowledge of particulars to God entails ­multiplicity in Him. Moreover, to say that God knows particulars as they are implies ­reliance on the known objects, which, of course, impairs the ontological independence of God. Despite these implications, Avicenna would not resort to denying God’s knowledge of particulars—obviously for theological reasons. Instead, he insists that God knows particulars in a universal way. Regardless of the exact meaning of this statement, Avicenna’s intention to distance God from particular knowledge and link Him to a nobler type—or mode—of knowledge is evident. By ruling out God’s reliance on particular objects, Avicenna makes it possible to speak of the self-sufficiency of divine knowledge. To this effect, he explains that God apprehends only His own essence. By virtue of this act of self-apprehension, God knows all existents emanating from Him, insofar as He is their Principle (Avicenna [al-Najāh] 1938, v.2, 247–249). Al-Ghazālī points out several problems with this theory, one of which is its failure to fulfil its primary intended objective: establishing the ­compatibility of divine simplicity and omniscience. In light of al-Ghazālī’s criticism, divine simplicity is compromised through the very act of self-apprehension, for when God knows His essence as the ‘principle’ of other things, ‘plurality’ is ­introduced to His knowledge through inclusion and concomitance. Moreover, God’s knowing His essence and ‘the relation’ of ‘being-a-principle’ involves conceptual plurality (TT, 202). Averroes detects errors and ­misrepresentations in al-Ghazālī’s criticism, but he concurrently concedes that Avicenna’s theory

106  Bakinaz Abdalla of divine knowledge—and, generally speaking, the underlying elements of the Avicennian metaphysical scheme—involves ambiguities and lends itself to criticism. In response to the point in consideration, Averroes clarifies that the cause of error is that human and divine intellects and their modes of k ­ nowledge are evaluated based on the same theoretical standards, while, in reality, they are utterly distinct. Thus, unlike human apprehension, ‘­plurality’ does not necessarily result when God apprehends things. Yet while it is ­philosophically acceptable to ascribe knowledge of existents to God, describing that knowledge as universal is improper. This is so since universal ­knowledge is yet-tobe instantiated in actuality, which means that it is ‘­deficient’ (nāqiṣ). Such knowledge is improper with respect to God. How, then, does God know existents, while maintaining divine perfection and transcendence to the modes and objects of human perception? Averroes introduces an account which he attributes to the philosophers—that Averroes subscribes to that account is evident from the fact that its key premises are approvingly reemployed throughout TT and other works. Let’s reformulate this account of divine knowledge in the following four steps. For this purpose, I mainly draw on Averroes’s response to al-Ghazālī concerning the incompatibility of the Avicennian account of divine knowledge with the doctrine of divine simplicity in TT (third discussion, 117–124 and sixth discussion 202–206). The first and second steps of the argument explain the relation of the human intellect to the perceived object. The third and fourth steps proceed from the conclusions reached about human intellect and knowledge to argue that God possesses a transcendent mode of knowledge that is not only self-sufficient and all-inclusive but also casual of all existents. We will see that this account, despite the disclaimer, violates divine simplicity. A 1 What is intelligible (maʿqūl) in a thing is its core reality, its form (ṣūrah). 2 The core reality of a thing is perceived by abstracting its form from matter. 3 When attached to a material object the form is not an intellect. 4 The form becomes an intellect only when it is abstracted from matter in the process of apprehension.5 5 According to the Aristotelian principle of cognitional identity, the knower, the known, and apprehension are one and the same thing (for the principle of ‘cognitional identity’ see Black 2019). 6 Thus, the human intellect is its apprehension and the apprehended intelligible forms—the human intellect is (ontologically) identical with the forms only insofar as they are immaterial, abstracted from matter. 7 Accordingly, the fully activated intellect is nothing but the apprehension of the order and core nature (i.e., intelligible forms) of all existents.

Doctrine of Divine Simplicity  107 B 8 The human intellect has limitations with respect to apprehending the totality of all existents and their order. 9 But since the nature of things and the cosmic order follow the law of the intellect, there must necessarily be a type of knowledge that stands as a cause for the arrangement and order residing in all existents. 10 The intellect possessing this knowledge is not subject to the limitations associated with the human intellect. 11 It is an intellect, separate from matter, by its very nature. C 12 The immaterial (separate) intellect is nothing but the a­ pprehension of the order and core nature (i.e., intelligible forms) of the existents. 13 An essentially separate intellect is ontologically superior to material existents and the natural order. 14  Thus, the knowledge that a sperate intellect possesses cannot be ­posterior to or dependent on particular-material objects. D 15 The perfection and constitution of a separate intellect consist in its knowledge which, according to (12), is the totality of the intelligible forms of all existents. Also, given the principle of cognitional identity, this knowledge is the essence of the separate intellect. 16 What is superior does not derive its existence and perfection from what is lower in rank. Separate intellects, although they share the same epistemic 17  content, differ hierarchically with respect to the mode of ­ knowledge. 18 Each separate intellect derives its knowledge from the intellect that ­precedes it in hierarchy. 19 The higher intellect stands as a cause for the lower intellect by virtue of the knowledge it supplies for its existence and perfection. 20 The chain of causation culminates in an intellect that demands no ­supplier of knowledge. 21 This First Separate Intellect, the Deity, apprehends the intelligible forms of all existents by apprehending His essence. 22 Concurrently, given (9 and 10), the First Intellect is that which gives rise to all existents. The crux of this compacted account is that the Deity apprehends all existents, not by relying on existing things, but by apprehending His essence (cf. Ibn Rushd’s Metaphysics 1984a, 194–195). Ascribing any of the human cognitive

108  Bakinaz Abdalla processes or modes of knowledge to God is absurd since God, being immaterial, possesses none of the physical organs needed for a­ cquiring ­knowledge of particulars and the universal knowledge derived thereof. Nonetheless, God is not ignorant of existents. His apprehension of His essence not only leads to His knowledge of all existents but also forms the foundation of the ­cosmic-natural order and arrangements that underlie all things (cf. Ibn Rushd’s Metaphysics, 197). In this respect, the self-apprehending Deity is said to be the Cause of existents. Note that there is no duality in the divine act of apprehension. Unlike the Avicennian model, God does not apprehend His essence and the relation of being-the-cause of existents, but His essence. Also, God apprehends Himself as a being by Himself, not in relation to a cause, unlike other intellects whose mode of self-knowing is relational, that is in relation to a cause. This mode results in conceptual multiplicity in ­intellects other than the First (TT, 122. For discussions of Averroes’s use of the idea of relation see Brenet 2020). On the face of it, Averroes achieved a threefold accomplishment: evaded violating divine simplicity, while preserving God’s omniscience and s­ upremacy as the cause of multiplicity in the world. Each of these accomplishments is questionable, notwithstanding. First, it is not clear how knowing the core natures of things, i.e., their intelligible forms, accords the Deity knowledge of the particulars of creatures, thus establishing God’s omniscient (for this point see Druart 1995; Flynn 1978–1979). Second, the way the epistemic and ontological elements work together is obscure (Kogan 1985, 246). Least to say, Averroes provides no convincing explanation of how things materialize through the very act of self-apprehension. Let’s put these matters aside and stick to the main point of the present study: divine simplicity. As it appears, Averroes could easily fall prey to the point raised by Dawkins concerning the compositeness, and consequently the ontological dependence, of the cause of multiplicity in the world. To clarify, let’s consider premise 21, that God apprehends the core natures of existents independently of external knowledge by apprehending His essence. Does not this mean that God’s essence contains a multiplicity of forms? If S apprehends its essence E, and if by apprehending E S ­apprehends x, y, and, z, does not this mean that E contains x, y, and z? Indeed, Averroes states, approvingly citing the philosophers, that the essence of the First ‘contains’ ‘all intellects,’ i.e., the intelligible forms of existents (TT, 121). This statement seems to eliminate divine simplicity, and by implication, renders the First Intellect ­ontologically dependent on the totality of forms that make up His essence. And the e­ limination of ontological independence results in the elimination of the ­primacy of the First Intellect as cause. To evade these outrageous implications, Averroes simply affirms that God’s apprehension of the forms of existents does not entail multiplicity. The reason specified in TT is that God knows the forms in an absolutely noble manner such that He constantly knows them actively, simultaneously, and indivisibly. Moreover, God knows the forms, not as being different entities,

Doctrine of Divine Simplicity  109 but as being identical with Himself. These features, as argued in TT, remove the implication of multiplicity (TT, 280). In the commentary on Metaphysics, Averroes points to a potential ‘doubt’ concerning the object of intellection of the First Intellect: since the object of God’s knowledge is composed of intelligibles, it ‘necessarily have parts different one from another, not similar one to another’ (Ibn Rushd’s Metaphysics 1984a, 102). This implies that God’s knowledge, which is His essence, is composed of many parts.6 In response to this potential critique, Averroes contends that the cognitional identity between the First Intellect and the intelligible forms is devoid of any features of multiplicity. Unlike human cognition, the First Intellect is identical with its object of knowledge in every possible respect. If the intellect and its object were united in all respects, then it would not follow that its intelligibles are many, for the cause of multiplicity is the difference between the intellect and the intelligible in us […] The intelligible of that which is absolutely free from matter, and the intellect are absolutely the same thing. (Ibn Rushd’s Metaphysics 1984a, 196) This explanation seems insufficient to permit ascribing both omniscience and simplicity to the Deity. Arguably, Averroes cannot have it both ways. Let’s grant that the First Intellect is absolutely identical with the intelligible forms, would not this negate their individuality as distinct objects of knowledge? Can the First Intellect perceive distinctions in the intelligible natures of x, y, and z—let alone produce them as distinct beings—if they are all identical with His essence in every possible respect?7 Alvin Plantinga criticizes Thomas Aquinas on his theory of divine attributes which affirms absolute identity between the divine essence and attributes to secure divine simplicity. This state of identity, as Plantinga contends, suggests that God has only one ­property, whereas we know that he has many unidentical properties (Plantinga 1980, 47). The same critique applies to Averroes, which results in a diminishing of the value of the epistemological and cosmological ­purports of his account of divine knowledge. For it is not clear how the identity between God and the ­intelligible forms, assuming that the intelligible forms in the divine essence actually carry substantial knowledge of creatures, substantiates that God is omniscient and that His knowledge, as argued, causally underlies all ­existents, the order, and arrangements in the world. The bottom line is that it is hard to smoothly combine absolute ­simplicity and divine omniscience as explained by Averroes. To secure the latter, Averroes would have to admit that the intelligible forms in God’s essence are somehow epistemically differentiable. In other words, he would have to ­tolerate speaking of some sort of internal composition in the divine essence. But could such an attitude be adopted without impairing God’s ontological independence? Put differently, is it possible to speak of a complex causal Principle of existents that requires no further explanation? In the following,

110  Bakinaz Abdalla I proposes to appeal to negative and positive answers from contemporary discussions. The negative answer insists in preserving the DDS and ­advocates a model of mysterianism as a way out of its conflict with the omniattributes. The positive answer suggests a ‘weaker version of ontological ­independence’ that only requires the First Being to be metaphysically ‘prior’ to Its i­nternal parts. Within the framework of the latter, a theory of divine priority is p ­ roposed as an alternative to absolute simplicity to corroborate the self-­sufficiency and ultimacy of the Deity as cause. I will explain how both answers could be accommodated into some of Averroes’s metaphysical views, offering a ­perspective for resolving the lack of harmony in his conception of God and His relation to the world. 6.3

The ‘Mysterian Move’8

In God Without Parts, James Dolezal offers a systemic defence of the DDS, a project that is motivated by the persuasion that the failure to sustain the DDS impairs God’s absoluteness and ontological independence. Upon considering arguments against the DDS, Dolezal notes that their c­ ommon denominator is ‘the strong commitment to ontological univocism.’ ‘Each critic speaks as if God and creatures were beings in the exact sense, ­reducing the Creatorcreatures distinction to a difference of degrees’ (Dolezal 2011, 29). Dolezal’s response to critics of the DDS, thus, involves establishing God’s utter distinctiveness. This approach informs his treatment of the ­tension between divine absoluteness and freedom, a resilient sub-problem of the DDS. In general terms, the DDS goes hand in hand with a fundamental view about the nature of the Deity: God is pure act, meaning that His essence involves no unrealized potentiality. This view, however, conflicts with God’s absoluteness which dictates that God possesses absolute freedom. Yet the idea of freedom implies the ability to choose to do something or its contrary. Therefore, ascribing freedom to God implies that His essence involves unrealized possibilities and potentialities. Needless to say, this contradicts the principle of simplicity. In response to this difficulty, Dolezal abandons neither simplicity nor absolutism, admitting that understanding how these incoherent metaphysical premises concurrently apply to God eludes our present state of perception. It should be readily confessed that the exact function of free will in God who himself is free act is beyond the scope of human knowledge. Just as we cannot apprehend God as ipsum esse subsistens, we cannot apprehend the identity between God as eternal, immutable, pure act, and his will for the world as free and uncoerced. Though we discover strong reasons for confessing both simplicity and freedom in God, we cannot form an isomorphically adequate notion of how this is the case. (ibid., 210)

Doctrine of Divine Simplicity  111 This explanation alludes to a distinction between knowledge-that and ­knowledge-how. Through sound reasoning we can acquire knowledge that something exists and that it possesses certain qualities, but we may fail to fathom the whyness and howness of its qualities.9 The latter state of ignorance does not necessarily undermine the former state of knowledge. Averroes’s conception of God could be understood along the line of a similar mysterian outlook. Support for such an outlook can be collated from the idea of ‘equivocation’ (tashkīk), that Averroes uses to distinguish between God’s and human’s attributes and features. What Averroes means by equivocation is open to different explanations. According to one opinion, Averroes means that the distinction between the features of God and those of creatures is hierarchical. According to another opinion, equivocation is complete, which means that the God-creature distinction pertains to type, not merely to the rank (for these distinctions see Druart 1995; Flynn 1978–1979). The latter opinion renders God an absolute other whose nature and attributes share nothing with those of human beings except naming. Assuming that Averroes has in mind complete equivocation, it should become acceptable to appeal to agnostic assumptions to mitigate the inconsistencies perceived in God’s nature. To say that God’s essence contains the intelligible forms of all existents does not necessarily preclude His simplicity; for God’s apprehension of His essence is qualified by absolute cognitional identity, a unique quality that negates the forms’ divisibility and plurality. Moreover, this mode of cognitional identity does not reduce all forms to one thing; otherwise Averroes’s doctrine of divine omniscience would be nothing more than lip service. Both aspects are valid with respect to God, but we do not understand how.10 Indeed, Averroes puts a similar insight into focus in several places in TT to establish a foundation for his systemic rebuttal of al-Ghazālī and his defence of the philosophers. Most of the errors falsely attributed by al-Ghazālī to the philosophers stem from the shortsighted assumption that what applies to human beings also applies to God. Averroes contends that this assumption is false. It is impossible, according to philosophers, that God’s knowledge should be analogous to ours, for our knowledge is the effect of the existents, whereas God’s knowledge is their cause, and it is not true that eternal knowledge is of the same form as temporal. He who believes this makes God an eternal man and man a mortal God, and in short, it has previously been shown that God’s knowledge stands in opposition to man’s, for it is His knowledge which produces existents, and it is not the existents which produces His. (TT, 285) If we extend the idea of complete equivocation to its limits, then the ­incoherence between divine simplicity and the variable aspects of God that

112  Bakinaz Abdalla entail multiplicity may not be as troublesome as it initially seems. Problems arise when the nature of the Deity is assessed against the rules governing human nature, behavior, and actions. This approach is invalid, as God and creatures do not belong to the same category of being. Nonetheless, ascribing complete equivocation to Averroes does not fit ­perfectly. One difficulty that flies in the face of the mysterian perspective is that some of the information that Averroes deduces about God dwell on premises relevant to the human intellect. Recall parts A and C of the argument for divine knowledge and its causal function in the world where Averroes makes conclusions about the content and nature of separate intellects, including the First Intellect, based on a comparison with the human intellect. This argument would collapse if we were to endorse complete equivocation. Moreover, in a concluding remark to the discussion of divine knowledge, Averroes advices that anyone seeking to acquire further knowledge about this issue should become professionally familiar with the qualities of the human intellect (TT, 123). This remark implies an analogical link between the First Intellect and the human intellect, appearing to reduce the distinction between them to a matter of hierarchy and rank rather than of type. In short, if the divine and human intellects turn out not to be entirely ­different in type, it is hard to see how Averroes can avoid being paradoxical by simply stating that the fact that God’s essence comprises the intelligible forms of all existents does not entail multiplicity. To sustain divine omniscience and its causal function, it seems that divine simplicity would have to be compromised. In what follows, I explain how Averroes could navigate this path without undermining God’s ontological independence. 6.4 Divine Priority as an Alternative to Absolute Divine Simplicity The difficulties associated with the doctrine of divine simplicity spurred some contemporary philosophers to propose the doctrine of divine priority as an alternative. This doctrine takes its cue from an understanding of the wholepart dependence relationship as extending from the whole to the parts, not vice versa (Schaffer 2010). In this respect, the whole is perceived to be more ontologically ‘fundamental’ than the parts. A typical illustrative case is the organic human being where the concrete unified whole is that which gives being to and explains the parts. A similar mereological view lies at the core of monism; in particular, what Schaffer labels as ‘priority monism’ (ibid.). According to the widespread understanding, monism is the view that only one thing exists. Schaffer argues that this view, which he labels as ‘existence monism,’ misrepresents ‘historical monism’ whose central tenet, corresponding to priority monism, ‘is not that the whole has no parts, but rather that the whole is prior to its part’ (32–34). Drawing on Shaffer’s conception of priority monism, Gregory Fowler (2015) formulates the doctrine of divine priority (DDP) as a replacement of the DDS while maintaining its traditional motivation.

Doctrine of Divine Simplicity  113 Fowler commences by pointing out the formulation of priority m ­ onism: For all x, if x is a proper part of U (where U signifies the mereological fusion of all concrete objects), then x depends on U for its existence (ibid., 120–121). This formulation is set forth with a specific type of ontological dependence in mind, illustrated by the following example: ‘Consider Socrates and his singleton {Socrates}. It is plausible that {Socrates} depends in some way on Socrates for its existence but that Socrates does not depend in the same way on {Socrates} (ibid., 122).’ The intended type of dependence is not causal, so that there would be a reciprocal causal relationship between Socrates and {Socrates}. Nor is it counterfactual, so that the non-existence of {Socrates} would entail the non-existence of Socrates and vice versa. Instead, it is ‘metaphysical dependence.’ That {Socrates} metaphysically depends on Socrates for its existence can differently be expressed as follows: ‘the ­existence of {Socrates} is grounded in Socrates, Socrates is prior to the existence of {Socrates}, {Socrates} exists in virtue of the fact that Socrates exists, and so on’ (ibid., 119–122; cf. Schaffer 2010, 35). Characteristically, then, this dependence relationship, unlike causal and counterfactual relations, is ‘irreflexive’ and ‘asymmetric.’ With this conception of metaphysical dependence in mind, Fowler sets out to argue against the general assumption, which has necessitated the DDS, that any composite being is dependent on its parts. Upon refuting this assumption, Fowler presents the DDP which explains the relation between God, His parts, and His properties, in light of the structure of priority monism—note that this is not meant to be an endorsement of priority monism, since ‘priority monism and the claim that God exists a se are incompatible’—as follows: ‘For all x, if x is a proper part of God or x is a property of God, then x depends on God for its existence’ (Fowler 2015, 122). Given the asymmetric and irreflexive character of the dependence relation of all x to God, it does not necessarily follow that God ontologically relies on His proper parts or properties. On these considerations, the DDP satisfies ‘the traditional motivation’ of the DDS: sustaining God’s ontological independence. The merit of DDP lies in that it can accommodate the supposition that God has parts and properties without compromising His ontological independence. As a result, it can spare us the conflicts that the DDS might have with various theological doctrines. Averroes’s conception of God and His relation to the world could ­benefit from the DDP if we can establish that its underlying premises have some basis in his works. This is not to suggest that Averroes departed from Aristotelianism in a manner that anticipated contemporary metaphysics. Indeed, the conceptual parallel between Averroes’s viewpoints and those central to priority monism and DDP is grounded in Aristotelianism. Both Schaffer and Fowler invoke Aristotle’s view of the whole-part relationship in ‘integrated wholes,’ e.g. living beings, as opposed to mere aggregations, e.g. a heap. In integrated beings, the parts are defined and acquire their particular characteristics by virtue of their functional integration in the whole. For

114  Bakinaz Abdalla example, a hand is what it is by virtue of its function in the whole organic body. In a deceased body, the hand may retain the same configuration, but since it no longer fulfils its original function in relation to the whole body, it can no longer be considered a hand in anything more than name (Aristotle On the Parts of Animals 1.1; cf. Metaphysics VII 10935b, 20–25). One note that Fowler takes to support the metaphysical priority of the whole to the part reads as follows [I]f the parts are prior to the whole, and the acute angle is a part of the right angle and the finger a part of the animal, the acute angle will be prior to the right angle and finger to the man. But the latter are thought to be prior; for in formula the parts are explained by reference to them, and in respect also of the power of existing apart from each other the wholes are prior to the parts. (Metaphysics 1034a, 27–34) By application, Fowler explicates the DDP: Suppose that God has proper parts. It is plausible that each of these proper parts is such that being the sort of proper part it is involves performing a certain function in God. So if God is a living thing, it is plausible that each of God’s proper parts is a functional proper part of God. Thus, given the revised Aristotle-inspired view, if God is a living thing, it is plausible that each of God’s proper parts depends on God for its existence. (Fowler 2015, 129) Turning to Averroes, let’s begin by considering his view of the whole-part ­relationship. It is worth noting that Averroes generally regards the parts as prior to the whole. However, Averroes also highlights that the ­priority of the parts to the whole is not absolute. In Tafsīr mā baʿd al- ṭabīʿah (Long Commentary on Metaphysics), particularly in his commentary on the ­passage from Aristotle mentioned above, Averroes explains that p ­ riority and ­posteriority concerning either the whole or the parts vary depending on the ontological mode of the intended referent. Take for example the f­ormula ­circle. This formula refers either to a concrete circle or to a circle in the absolute sense. A concrete-artificial circle ontologically depends on the ­ ­composition of its matter and form, and on the unity of its segments, i.e., diameter, radius, chord, arc, and semi-circle, etc. In this sense, the parts are prior to the whole (Tafsīr mā baʿd al- ṭabīʿah 1938, 1035a, 9–12, 898). A different whole-part relationship holds in the absolute conception of the circle. Theoretically, the circle has many segments. These segments, however, are not necessary to think of the circle. This is not the case with the segments; they cannot be thought of or defined except with reference to the circle (ibid., 1035a, 35-b3: 902).

Doctrine of Divine Simplicity  115 The example of the absolute circle clarifies the dependence of the parts on the whole and the priority of the whole to the parts. While Aristotle ­attributes this type of dependence relationship to absolute formulae, as opposed to ­particular things composed of matter and forms in which the parts are prior to whole (Tafsīr mā baʿd al- ṭabīʿah 1034b, 32–33, 893), there are occasions where he, as Averroes explains, ‘weakens’ (uḍʿif) the priority of the parts to the whole in the latter type of things. This is suggested by Aristotle’s clarification that, in some hylomorphic compounds, it is impossible for the parts to exist independently of the integrated whole. Such parts cannot be separate from the form [unifying the compound entity] or precede or succeed it ontologically. Examples of these are [any of] the animal’s parts in relation to the [organic] animal. For if the soul leaves [the body of the animal], none of these parts is named by its actual name anymore except by way of equivocation. An example of this is the finger whose relation to the living and the dead body is different. If the body is dead, signifying it with the word finger occurs only by way of equivocation. The same is true with respect to its condition before sensation had permeated it. […] It is impossible for parts like theses to exist separately from the [unifying] form [of the compound entity]. […] All parts of the animal are in this regard similar to the actual part in which the principle of the soul lies, regardless of whether that part is the heart, as some believe, or the mind, as others say. This disagreement does not negate the assumption that that primary part would [also] be signified by its name only equivocally if it became lifeless, despite the fact that it is the locus of life. (ibid., 1035b, 5–20: 910–911. My translation) Averroes’s exposition of Aristotle’s view of the whole-part relationship s­ uitably coheres with what Fowler suggests regarding the functional i­ ntegration of the parts in the whole. We see that the parts in integrated beings depend on the whole in such a way that confirms the ontological priority of the whole. There is nothing in Averroes’s exposition indicting that this dependence ­relationship is necessarily counterfactual. If anything, the case of absolute circle suggests that the relation of the parts to the whole is asymmetric and irreflexive. Thus, the parts of the circle cannot be defined except with reference to the circle, but not vice versa. Reference to this type of asymmetrical dependence relationship is made in TT as well. Averroes specifies: the two parts in any compound must be either (1) ‘mutually a condition for each other’s existence,’ (2) ‘neither of them is a condition for the existence of the other,’ or (3) ‘exclusively one is the condition for the other’ (TT, 199). The third type accommodates with the dependence relationship scheme on which the DDP dwells. Now that we have identified parallels in Averroes’s thought to the ­underlying premises of the DDP, I believe it can be used as a viable a­ lternative to absolute divine simplicity in order to mitigate the implications of the

116  Bakinaz Abdalla account of divine knowledge, in particular its central premise that the divine essence includes the forms of all existents. Supposing that each intelligible form is part of God, then, applying Fowler’s formulation of the DDP (For all x, if x is a proper part of God or x is a property of God, then x depends on God for its existence), each form can be said to ontologically depend on God. And considering that the dependence relationship here is asymmetric, it does not follow that God ontologically depends on the forms. To further bolster the ontological dependence of forms on God we can make the following point based on an analogy drawn by Averroes in TT between the world and a living animal. In its totality, the world is like a living animal, the benchmarks of this is the cohesion and functional interdependence of the internal parts and their dependence on the whole. Consider the celestial spheres. These are f­ undamental components of the world whose particular characteristic, the thing by virtue of which they are what they are, consists in the constant m ­ ovements they produce and through which all beings in the sublunary world materialize. Just like the parts of a living animal, the celestial spheres are connected one to another and ‘refer to one body and to one end’—‘namely, the world in its totality’ (TT, 129). It is only within the context of the orderly unity holding the world together that these celestial spheres can carry out the kinetic function sustaining them and their role within the universe. Out of this context, this function becomes null, and so does their very reality. Consider comparing the relationship between the intelligible forms and the divine essence to the relationship between the parts of the world and the world as a unified whole. This approach is legitimate since we know that the forms in God’s essence constitute a transcendent image, if so to speak, of the world (see also TT, 262). In other words, there is a structural correspondence between the intelligible forms within the divine essence and the parts of the world. If that is true, wouldn’t the relationship between the parts of the world and their dependence on the whole correspond to the relationship between the intelligible forms within God’s essence and their dependence on the totality of the essence? Answering this in the affirmative is reasonable. If the parts in the concrete world are dependent on the whole, it seems illogical to assume a reverse dependence relation within the source, the divine essence and all what it includes of forms. In concluding this section, I must emphasize that I do not argue for Averroes’s espousal or anticipation of the DDP. However, the foregoing discussion has, I hope convincingly, shown that Averroes would not dismiss such a doctrine in favor of absolute divine simplicity. At least, not if he aimed to ­philosophically support God’s knowledge of existents and its causal function in the world without compromising the fundamental ontological independence of God. 6.5

Concluding Remarks

This chapter has made an attempt to reassess Averroes’s conception of God and His casual relation to the world in light of relevant contemporary

Doctrine of Divine Simplicity  117 ­ iscussions. Oddly enough, Averroes’s account of God’s knowledge, with d its central premise that God’s essence comprises the intelligible forms of all existence, undermines God’s simplicity, notwithstanding the disclaimer. Seen in light of the commonly accepted link between divine simplicity and ontological independence, Averroes seems to heedlessly sacrifice both. Dawkins’s antitheist argument, stemming from the assumption that a complex god would be ontologically dependent and require further explanation, could not have found stronger support. One way to evade this predicament is to ­minimize the seriousness of Averroes’s account of divine knowledge, relegating it to a mere rhetorical-theological role. Else, one could simply accept the claim that the divine essence’s inclusion of the intelligible forms of all existents does not conflict with God’s simplicity, assuming that Averroes’s references to equivocation are meant in the complete sense, which would introduce a sense of mysterianism. We can then excuse his paradoxical claim on the grounds that the human intellect has no capacity to fathom how the divine nature operates. But as we have seen, it is not certain that Averroes adopted complete equivocation. Since Averroes’s accounts of God’s knowledge of all existents and responsibility for the world’s plurality are fundamentally rooted in the ­ ­conception of the divine essence as comprising the intelligible forms of all existents, I thought it would be suitable to avoid compromising it and to try to solve the problem of ontological independence differently by appealing to the doctrine of divine priority. Despite the presence of supportive textual evidence of the whole-part relationship in Averroes’s works, the question remains open as to whether Averroes would ever intentionally consider a flexible attitude to divine simplicity. It is worth noting that simplicity is not an exclusive ­characteristic of God, alone such that we can claim that His unique ontological independence is secured solely on its basis. Indeed, Averroes maintains that simple intellects are also simple. Considering the customary connection between simplicity and ontological independence, separate intellects should also be deemed ontologically independent, just like God. Averroes, however, denies this, affirming that separate intellects are ontologically dependent on the First Intellect, the Deity. The latter is distinguished from other separate ­intellects by the fact that His knowledge is entirely self-sufficient, whereas each ­ separate intellect apprehends, besides its essence—which, like the essence of the Deity, comprises the forms of all existence—its relation to the Deity as being its effect (TT, 120–122). That Averroes ascribes the quality of ­simplicity to ­separate intellects— we may add here that Averroes considers celestial spheres to be simple (on the simplicity of the spheres, see De Substantia Orbis 1986, 102; TT, 145)—weakens the connection between simplicity and ontological independence that customarily informed ­discussions of God in Averroes’s i­ ntellectual circle. This is so since neither sperate intellects nor celestial spheres are considered ontologically independent.

118  Bakinaz Abdalla Additionally, the DDS appears to occupy little theological significance for Averroes, as can be detected from his engagement with t­heological discussions of tawḥīd and the nature of divine attributes in his Manāhij al-adillah fi ʿaqāīd al-millah. Characteristically, Averroes bypasses philosophical intricacies and resorts to a humble proof of tawḥīd drawn from the Quran that merely proves the numerical uniqueness of God. The proof dwells on the premise that two or more agents do not ­produce and sustain one and the same action/effect. Since the universe is an integrated unit and is not corrupted, it must be produced and sustained by one agent (Averroes, Manāhij al-adillah 1964, 156–159). Nothing beyond believing in numerical uniqueness and ‘denying godhood’ (nafy al-ilāhiyyah) from any being other than God (ibid., 159) seems required to be a faithful adherent of the doctrine of tawḥīd. In the Islamic intellectual tradition, discussions of the doctrine of tawḥīd are typically intertwined with discussions of divine attributes, both of which furnish a suitable framework for defending divine simplicity. Accepting a realist explanation of the nature of divine attributes renders God’s essential features, e.g. knowledge and will, ontologically independent, which undermines the uniqueness of God’s ontological independence. One ­common way to evade this predicament is to identify God’s attributes with His essence. This identity between God’s attributes and essence, as we saw formerly, lies at the core of some formulations of the DDS. In his assessment of Islamic discourse on divine attributes, Averroes condemns excessive ‘innovative’ (al-bidaʿ) attempts to eliminate the attributes (ibid., 165–166), pointing out that the attributes-essence-identity mode falls beyond human comprehension, and that’s why Scripture ascribed many attributes to God. This is not to say that Averroes would accept a realist conception of the attributes, if he would permit a thoroughgoing investigation of the issue. Averroes proceeds from pragmatic considerations: If we have no capacity to understand how God could be identical with His attributes–even if we are convinced that this mode of identity is true—it is unwise to popularize the issue and provoke doubts about the Scriptural account of God’s attributes (ibid., 166). All in all, Averroes’s articulation of tawḥīd in his Manāhij al-adillah keeps distance from the ­intricate issues involved in theological discussions, an attitude that suggests openness to the simple scriptural expression of tawḥīd which merely denies the existence of other gods beside God. Whether God turns out to be absolutely simple or involving immaterial parts, like the intelligible forms of all existents, makes no difference for the adherent of tawḥīd, so long as she ascribes the production and governance of the world solely to the agency of God and brings no other agent into the dominion of His governance, whether by way of verbal confession, belief, or practical worship. Whether Averroes would approve this lenient understanding of tawḥīd on philosophical basis or relegate it to a matter of religious belief is a question that requires drawing attention to his conception of the relationship

Doctrine of Divine Simplicity  119 between religion and philosophy. Some of Averroes’s treatises, including his Manāhij al-adillah are known in modern scholarship for their exclusively theological and pragmatic purposes (see Kukkonen 2002), and therefore they are mostly marginalized as unsuitable sources for identifying Averroes’s well-considered positions which can only be learned from the commentaries on Aristotle. Accepting this view unconditionally, however, could turn religion and philosophy into separate domains. What, then, can we make out of Averroes’s affirmative statement that the truth does not contradict the truth (Faṣl al-maqāl 1969, 31–32) which endorses the unity of the truth (see Taylor 2001; cf. Arnaldez 2000, 79–89; Belo 2013, 21–49)? Indeed, Averroes ­highlights that the Quran’s and philosophy’s accounts of tawḥīd are ­coherent, the only difference being the profundity of the manner through which philosophers understand the unity holding together the u ­ niverse’s parts, and consequently the oneness of God (Manāhij al-adillah 1964, 157). If Averroes ­ ifference indeed believes in the coherence of religion and philosophy, the d between them with respect to tawḥīd would come down to a m ­ atter of ­methodology rather than concept. Irrespective of Averroes’s actual stance on the relationship between religion and philosophy, it remains true that his philosophical analysis of the nature of God suffers from instances of incoherence that could threaten God’s ontological independence and His causation of all existents if absolute simplicity were deemed the sole warranty for these qualities. Fortunately, as argued in this chapter, some elements in Averroes’s thought allow a partial reduction in the commonly-held link between absolute simplicity and ontological independence, thus preventing his conception of God from facing such outrageous consequences. “The publication of this chapter was possible thanks to a generous funding from the John Templeton Foundation. The opinions expressed in the paper are those of the author and do not represent the Foundation.” Notes 1 Since this chapter is not meant to be a survey, I do not focus on addressing ­differences among philosophers. 2 The process of deriving divine uniqueness from simplicity is sometimes reversed, as Avicenna does in his works. For this approach, see Zarepour (2022, 48–53). 3 It is worth highlighting that some medieval Muslim theologians, including alGhazālī, raised fundamental objections that anticipate Plantinga’s criticism of the DDS, see G. McGinnis (2022) and Muhtaroğlu (2020). I am grateful to Dr. Zarepour for bringing these sources to my attention. 4 Averroes espouses this principle in some of his early works, see e.g. Talẖīṣ mā bʿad al-ṭabīʿah (1994, 160–164). 5 That any actual form is intellect is a premise that Averroes repeatedly uses in his works (Taylor 1998, 520). 6 This doubt finds roots in Plotinus’s thought where the idea that the true One is beyond intellect serves to safeguard divine simplicity (Cohoe 2017). Echoes of this idea can be found in the writing of al-Kindi (Adamson 2018, 258). 7 Averroes himself admits this implication. However, the solution he presents is no more helpful in securing divine omniscience; see Ibn Rushd’s Metaphysics (1984a, 197–198).

120  Bakinaz Abdalla 8 I borrow this phrase from Vallicella (2019). 9 This can be linked to the different types of factual and explanatory demonstrations discussed by Arab Aristotelians: demonstration-that, demonstration-how, and demonstration-why; see al-Fārābī, Book of Demonstration (2007, 68), and Averroes, Sharḥ al-burhān (1984b, 348). 10 In medieval philosophy, the idea of equivocation was widely used in discussing the divine nature. Faithful students of Averroes from Jewish circles adapted it to agnostic views about the Deity that were advocated by Maimonides. For further discussion; see Manekin (2002).

References Primary Sources al-Fārābī, Abu Naser. ‘Book of Demonstration.’ In Classical Arabic Theology: An Anthology of Sources, eds. and trans. Jon McGinnis and David C Reisman, ­Indianapolis, IN: Hackett Pub. Company, 2007. Al-Ghazālī, Abu Hamid. Maqāṣid al-falāsifah, edited with introduction by Mahmud Bigou. Damascus: Al-Ṣabbaḥ, 2000. Al-Kindi, Abu Yusuf. Rasāiʾl falsafiyyah by al-Kindi, edited by Moahmed Rida and ­ rabic-­Islamic Faut Sezgin. Frankfurt: Publications of the Institute for the History of A Science, 1999. Aquinas, Thomas. Summa Theologiae. Allen, TX: Christian Classics, 1981. Aristotle. The Basic Works of Aristotle, edited and translated by Richard McKeon. New York: Modern Library, 2001. Averroes. Tafsīr mā baʿad al-ṭabīʿah, edited by Maurice Bouyges. Beyrouth: ­Imprimerle Catholique, 1938. ———. Tahāfut al-tahāfut [The Incoherence of the Incoherence], translated from the Arabic with introduction and notes by Simone Van Den Bergh. London: E.J. Gibb Memorial, 1954. ———. Manāhij al-adillah fi ʿaqāīd al-millah, edited with an introduction by Mahmoud Qasim. Cairo: Maktabat al-anglo al-maṣriyyah, 1964. ——— Faṣl al-maqāl wa taqrīr mā bayna al-ḥikmati wa al-sharī‘ah min itiṣāl, a refined ­edition with introduction by Mohamed Emarah. Cairo: Dār al-ma‘ārif, 1969. ———. Ibn Rushd’s Metaphysics: A Translation with Introduction of Ibn Rushd’s Commentary on Aristotle’s Metaphysics: Book Lam, edited and translated by Genequad, Charles. Leiden: E.J. Brill, 1984a. ———. Sharḥ al-burhān wa talkhīs al-burhān li Ariṣṭu, edited by Abd al-Rahman Badawi. Kuwait: Al-majlis al-waṭanī liltathqīf, 1984b. ———. Averroes’ De Substantia Orbis. Critical Edition of the Hebrew Text with English Translation and Commentary by Arthur Hyman. Cambridge, MA: ­ ­Medeival Academy of America, 1986. ———. Talẖīṣ mā baʿd al- ṭabi‘ah, edited by Jirar Jahami. Beirut: Dār al-fikr, 1994. Ibn Sina (Avicenna). Al-Najāh, ed. Mohyi al-Din Kurdi. Cairo: Al-maktabah almurtaḍawiyyah, 1938. ———. Al-Shifa-- Ilāhiyyat, with an introduction by Ibrahim Madkur. Cairo: Almaṭābiʿ al-amīriyyah, 1960. Maimonides, Moses.  The Guide of the Perplexed, translated with an introduction by Shlomo Pines. Chicago, IL: University of Chicago Press, 1963.

Doctrine of Divine Simplicity  121 Secondary Sources Acar, R. ‘Reconsidering Avicenna’s Position on God’s Knowledge of Particulars.’ In Interpreting Avicenna: Science and Philosophy in medieval Islam, eds. J. McGinnis and D. Reisman, 142–156. Leiden; Boston, MA, 2004. Adamson, Peter. ‘Al-Kindi and the Muʿtazila: Divine Attributes, Creation, and ­Freedom.’ Arabic Sciences and Philosophy, vol. 13 (2003), 45–77. ———. ‘On Knowledge of Particulars.’ Proceedings of the Aristotelian Society, New Series, vol. 105 (2005), 257–278. ———. ‘From the Necessary Existent to God.’ In Interpreting Avicenna: Critical ­Essays, edited by Peter Adamson, 170–189. Cambridge: Cambridge University Press, 2013. ———. ‘The Simplicity of Self-Knowledge after Avicenna.’ Arabic Sciences and ­Philosophy, vol. 28, no. 2 (2018), 257–277. Amin, M. Wahid. ‘From the One, Only One Proceeds. The Post-Classical Reception of a Key Principle of Avicenna’s Metaphysics.’ Oriens vol. 48 (2020), 123–125. Arnaldez, Roger. Averroes: A Rationalist in Islam, translated by David Streight. ­English Language edition, Indiana: University of Notre Dame Press, 2000. Belo, Catarina. ‘Averroes on God’s Knowledge of Particulars.’ Journal of Islamic Studies, vol. 17, no. 2 (2006), 177–199. ———. Averroes and Hegel on Philosophy and Religion. Farnham: Ashgate, 2013. Black, Deborh. ‘Constructing Averroes’ Epistemology.’ In Interpreting Averroes, ­edited by Adamson, Peter and Matteo Di Giovanni, 96–115. Cambridge; New York: Cambridge University Press, 2019. Brenet J.-B. ‘Relation as Key to God’s Knowledge of Particulars in the Tahāfut Al-Tahāfut and the Damīma: A Cross-Talk between Averroes, Al-Ġazālī and ­Avicenna.’ Arabic Sciences and Philosophy, vol. 30, no. 1 (2020), 1–26. Cohoe, Caleb M. ‘Why the One Cannot Have Parts: Plotinus on Divine S­ implicity: Ontological Independence, and Perfect Being Theology.’  The Philosophical ­Quarterly, vol. 67, no. 269 (2017), 751–771. Dawkins, Richard. The God of Delusion. London: Bantom Press pbk, 2006. Dolezal, James E. God without Parts: Divine Simplicity and the Metaphysics of God’s Absoluteness. Eugene: Pickwick Publications, 2011. Druart, Thérèse-Anne. ‘Averroes on God’s Knowledge of Being Qua Being.’ In S­ tudies in Thomistic Theology, edited by Paul Lockey et al., 175–209. Houston, TX: ­Centre for Thomistic Studies, University of St. Thomas, 1995. Flynn, J. G. ‘St. Thomas and Averroes on the Knowledge of God.’ Abr-Nahrain, vol. 18 (1978–1979), 19–32. Fowler, Gregory. ‘Simplicity or Priority?’ In Oxford Studies in Philosophy of ­Religion, edited by Jonathan Kvanvig, 114–138. Oxford; NY: Oxford University Press, 2015. Kogan, Barry. ‘Averroes and the Theory of Emanation.’ Mediaeval Studies, vol. 43 (1981), 384–404. ———. Averroes and the Metaphysics of Causation. Albany: State University of New York Press, 1985. Kukkonen, Taneli. ‘Averroes and the Teleological Argument.’ Religious Studies, vol. 38 (2002), 405–428. Manekin, Charles H. ‘Maimonides on divine knowledge: Moses of Narbonne’s a­verroist reading.’ American Catholic Philosophical Quarterly, vol. 76, no. 1 (2002): 51–74.

122  Bakinaz Abdalla Marmura, Michael E. ‘Some Aspects of Avicenna’s Theory of God’s knowledge of Particulars.’ Journal of the American Oriental Society, vol. 82, no. 3 (Jul–Sep. 1962), 299–312. McGinnis, John. Avicenna. Oxford: Oxford University Press, 2010. McGinnis, Jon. ‘Simple Is as Simple Does: Plantinga and Ghazālī on Divine ­Simplicity.’ Religious Studies vol. 58, no. S1 (2022), S97–S109. Muhtaroğlu, Nazif. ‘Plantinga and Ash‘arites in Divine Simplicity.’ Kader vol. 18, no. 2 (2020): 488–499. Plantinga, Alvin. Does God Have a Nature? Milwaukee: Marquette University Press, 1980. Schaffer, Jonathan. ‘Monism: The Priority of the Whole.’ The Philosophical Review, vol. 119, no. 1 (2010), 31–76. Scott, Michael. Religious Language. Basingstoke: Palgrave Macmillan 2013. Stump, Eleonore. ‘Simplicity.’ In Companion to Philosophy of Religion, edited by Philip L. Quinn and Charles Taliaferro, 250–256. Oxford: Blackwell, 1997. Taylor, Richard. ‘Averroes on Psychology and the Principles of Metaphysics.’ The Journal of the History of Philosophy, vol. 36 (1998), 507–523. ———. ‘Truth Does Not Contradict Truth: Averroes and the Unity of Truth.’ Topoi: An International Review of Philosophy, vol. 19, no. 1 (2001), 3–16. Vallicella, F. William. ‘Divine Simplicity.’ Stanford Encyclopaedia of Philosophy (2019). Yaqub, Aladdin M. ‘Al-Ġazali’s Philosophers on the Divine Unity.’ Arabic Sciences and Philosophy, vol. 20 (2010), 281–306. ­ vicennian Zarepour, Mohammad Saleh. Necessary Existence and Monotheism: An A Account of the Islamic Conception of Divine Unity. Elements in Religion and ­Monotheism. Cambridge, 2022.

PART IV

Miracles

7 The Metaphysics of Miracles Avicenna on Natures and Prophetic Powers Emann Allebban

What must the world be like to host both causal relations and their v­ iolations? If miracles1 are to be understood as violations of nature, or as exceptional interventions on behalf of the divine, then it appears difficult to make sense of such events within a natural account of the cosmos and its causal structure. Must one choose between a world with natures, or a necessary causal structure, versus a world caused by a willing and omnipotent God? Is the notion of natural causation incompatible with any notion of a God that interacts with the world and its constituents? Moreover, are all miracles created the same, as it were, or is there room for varying notions of the disruptive nature of miracles? That God is involved in the world is central to many Islamic theological worldviews and readings of scripture. Such a belief emphasizes not only God’s creative power but God’s connection with us in unique ways, e.g., responsiveness to prayers,2 divine intervention or immanence in, say, saving Abraham from fire. But how is one to understand God’s interaction with the world in relation to a causal view of the cosmos? For the falāsifa, the proponents of Greek philosophy in the Islamic world, who believed that the natural order involved necessary and unique causal relations, such interventions ought to be interpreted away. We read God’s protecting Abraham from fire metaphorically, because fire necessarily burns those things that can be burned. That, at least, appears to be the implication of their worldview. Moreover, a philosopher like Avicenna seems not just committed to necessary causal relations, but to a God who acts necessarily. Are miracles, in any sense, possible within such a cosmic system? It is for these reasons and others that an occasionalist like al-Ghazālī is ­ etaphysics taken to state that there is no room for miracles within Avicenna’s m and natural philosophy. But, as the following shows, the situation is more ­complex both in terms of Avicenna’s perspective and Ghazālī’s representations of the latter. This chapter explores a largely overlooked strategy in Avicenna, one which does not focus on God’s role, but rather on how certain ­“powers” of a human soul can radically violate causal relations, including taking control of the celestial bodies. Avicenna thinks about miracles not in terms of God’s intervention but with respect to the nature and power of a certain DOI: 10.4324/9781003327714-12

126  Emann Allebban kind of soul, namely, the prophetic human soul. He believes that the cosmos comprises an extraordinarily powerful agent, a human nonetheless, who has the ability to function as a kind of “soul for the world.” This agent is likened to the heavenly soul governing the sublunar world, the Active Intellect, but has powers over the sublunar world, and even the superlunar, in ways that the Active Intellect does not. Avicenna seems to go so far as to suggest that this agent can override the necessary natural powers in the sublunar world such that they are subordinate to the prophetic soul more than to their own nature. By working miracles into his system in this way, Avicenna’s prophetas-world-soul seems to allow him to maintain God as a necessary cause within the emanative structure of the cosmos. But the philosophical tradeoff appears costly. The prophetic human soul can render necessary natures contingent in a way that outreaches the heavenly souls and even God. It is unclear how such a view is supported by Avicenna’s s­ystem, but it raises important questions about the latter’s view of the ­necessary structure of the cosmos. My entry point into these issues will not be the usual one – i.e. Ghazālī’s reprimand of the philosophers for denying miracles (muʿjizāt). I begin instead with Ghazālī’s lesser-known endorsement (or provisional assumption) of the falsafa account of the three ways prophets are able to enact miracles. Contrary to received views, Ghazālī’s critique is not that the philosophers plainly deny miracles, but that their exclusion of certain categories of events from the realm of possibility over others has no consistent basis. Ghazālī is concerned with the exclusion of three prophetic miracles in particular: the changing of the staff into a serpent, the reviving of the dead, and the cleavage of the moon. Examining concrete cases and the tension between the miracles that the falāsifa affirm and the miracles they allegedly deny will tell us more about the fine-grained concepts at issue. I begin with Ghazālī’s account of the philosophers on prophetic ­miracles with the kalām background of his critique. I then turn to Avicenna’s account of miracles in the Ishārāt, the De Anima and Metaphysics of the Shifāʾ. I conclude with the broader implications of this new reading of a classical Islamic debate.3 7.1 7.1.1

Ghazālī on the Philosopher’s Miracle Three Sources of Miracles

In the introduction to the Natural Sciences of the Incoherence of the Philosophers, Ghazālī summarizes the philosophers’ account of three ways prophets enact miracles,4 as explained through their more general theory of action. The first is through the prophet’s superior imaginative faculty (al-quwwa al-mutakhayyila) which allows the prophet, while awake, to see future events. The second is through the prophet’s theoretical rational faculty (al-quwwa al-ʿaqliyya al-naẓariyya), which allows the prophet to

The Metaphysics of Miracles  127 know ­intelligibles “in the quickest of times” and without instruction, where it would usually take others time and instruction to apprehend the middle term, if given the conclusion, or to apprehend the conclusion, if given the middle term. Ghazālī is not explicit here on what types of miracles this would allow for, but it seems what he has in mind is the prophet displaying expert knowledge of a wide range of different sciences or facts, such that it would be extraordinary for a human to have such an immediate grasp of a variety of subject matters. In addition to the speedy or effortless acquisition of knowledge, a prophet’s theoretical faculty may also play a role in terms of effecting some act through that knowledge, say, of healing the blind through knowledge of a medicine.5 The last category of miracles that Ghazālī attributes to the philosophers is the most relevant. The category concerns not simply an individual’s knowledge, or cognitive faculties and acts, but the effecting of natural change in the world. Events in the latter category occur by way of the prophet’s practical faculty (al-quwwa al-nafsiyya al-ʿamaliyya), which allows the prophet’s soul to affect not just the movements of his own body but those of other natural bodies as well. As Ghazālī explains, “If, then, it is possible for the corporeal parts of his body to obey it [the practical faculty of his soul], it is not impossible for corporeal parts other [than his] to obey it.”6 This practical faculty of the soul would then “reach a point [in strength] whereby natural things are influenced by it and come under its control (tatasakhkhar).”7 He says this would allow for the prophet to, for instance, cause various meteorological and geological events: rain, wind, a storm, or even an earthquake. What I want to focus on in this account by Ghazālī is that he says the philosopher’s account cannot accommodate all miracles: This, then, is their doctrine regarding miracles. We do not deny ­anything they have mentioned and that this belongs to prophets. We only deny their confining themselves to [these kinds] and their denying the ­possibility of the changing of the staff into a serpent, the r­ evivification of the dead, and other [miracles]. It is then necessary to plunge into this question to affirm miracles and [to achieve] something else— namely, to support what all Muslims agree on, to the effect that God has power over all things.8 This distinction between possible and impossible miracles raises several ­questions: Why are some events permitted and others not, and how does Avicenna, or the falāsifa under Ghazālī’s reading here, draw that line? It is not immediately clear what the difference is between reviving the dead, t­urning wood into an animal, and cleaving the moon versus a human ­causing an earthquake or storm, by a sheer act of will. They are all extraordinary events that break from “custom”, the regularly ordered pattern of events, which is a kalām criteria for something to count as a miracle.9 On the falāsifa’s view, there must be something specific to turning wood into a snake such that even a prophet’s superior practical faculty cannot will it to be so.

128  Emann Allebban The received view is that Ghazālī thinks the philosophers must deny ­ iracles because they affirm a necessary connection between cause and m effect. This view is based on one of Ghazālī’s most cited passages, the beginning of Discussion 17 of the Incoherence, where he states that the connection between a cause and effect is not necessary but rather decreed by God. As he writes just before in the introduction to the Natural Sciences, “Whoever renders the habitual courses [of nature] a necessary constant makes all these [aforementioned miracles, e.g. changing of the staff into a serpent, revival of the dead, and splitting of the moon] impossible.”10 It seems if he can show the weaknesses in the view that observed causes and effects are necessarily connected,11 then the turning of wood into an animal is well within the power of God.12 Ghazālī begins Discussion 17 as follows: The connection between what is believed to be a cause in terms of habit and what is believed to be an effect in terms of habit is not necessary, according to us. Rather, [with] any two things, where ‘this’ is not ‘that’ and ‘that’ is not ‘this’ and where neither the affirmation of the one entails the affirmation of the other nor the negation of the one entails negation of the other, it is not a necessity of the existence of the one that the other should exist, and it is not a necessity of the nonexistence of the one that the other should not exist — for example, the quenching of thirst and drinking, satiety and eating, burning and contact with fire, light and the appearance of the sun, death and decapitation, healing and the drinking of medicine, the purging of the bowels and the using of a purgative, and so on to [include] all [that is] observable among ­connected things in medicine, astronomy, arts, and crafts. Their ­connection is due to the prior decree of God, who creates them side by side, not to its being necessary in itself, incapable of separation. On the contrary, it is within [divine] power to create satiety without eating, to create death without decapitation, to continue life after decapitation, and so on to all connected things. The philosophers denied the ­possibility of [this] and claimed it to be impossible.13 As has been pointed out,14 Ghazālī is repeating this point made earlier by al-Bāqillānī: Concerning that over which [the philosophers] are in such a stir, namely that, they know by sense perception and necessarily that burning occurs from fire’s heat and intoxication from excessive drink, it is tremendous ignorance. That is because what we observe and perceive sensibly when one drinks and fire comes into contact is only a change of the body’s state from what it was, namely, one’s being intoxicated or burnt, no more. As for the knowledge that this newly occurring state is from the action of whatever, it is not observed.15

The Metaphysics of Miracles  129 Ghazālī attributes to the philosophers the view that there is a necessary ­connection between cause and effect, and he seems to say here that the ­necessity goes both ways: it’s not simply that the philosophers hold that upon contact with fire, burning results, but that if there is burning, there must be contact with fire: “It is not a necessity of the existence of the one that the other should exist, and it is not a necessity of the nonexistence of the one that the other should not exist.” The same is the case, he explains, with the other examples he mentions, such as drinking and thirst quenching. Ghazālī explains that if, according to the philosophers, there is this necessary connection between cause and effect – such that fire must burn upon contact with a flammable object (and assuming all the right conditions are in place) – then natural courses of events are not just customary or regular but necessarily and causally so.16 Ghazālī later attributes the necessity to the falsafa view of natures. If it is in the very nature of fire to burn upon contact with a flammable object, i.e. if the powers of things are contained within their very nature, which they are for Aristotelian natures, then neither a prophet, nor even God, can break that relation. (Setting aside, for now, whether God can, for instance, intervene and clothe Abraham with talc, or whether God can eliminate, change, or somehow pause some particular fire’s nature and hence its causality for some duration of time). However, it’s not clear how far causal necessity takes us in understanding why the philosophers must exclude reviving the dead or changing a staff into a snake, but allow a human to will an earthquake into existence, not much differently from how one wills their legs to stand and walk to get a drink. In other words, wouldn’t a human causing an earthquake or storm also involve a break of the necessary connection between whatever typically causes them? The philosophers under critique here would be affirming the existence of an effect (rain) without its cause. Granted, it is not yet clear what precisely the prophet is doing in such cases – drawing warm air up that then cools, i.e. preparing the air to become water? Or is the prophet “giving” the form of water to air? But whatever the prophet’s causal role, the rain falls without at least one of its usual and allegedly necessary causes. Since the philosophers affirm certain categories of miracles, the philosophers cannot be said to deny miracles simply because they are committed to necessary causation. Thus the question Ghazālī is asking in Discussion 17 of the Incoherence, and which we ought to ask, is not why philosophers deny miracles in some blanket fashion, but rather why they deny the specific ones they do. Reframing the debate in this way makes it clear that the usual story is not a sufficient explanation of the philosophers’ denial of some miracles and acceptance of others. It is not enough to simply say that given their causal theory, the ­philosophers in question cannot allow for a break between cause and effect. In affirming some miracles, they clearly do allow cases where you have an effect without its (usual) essential or natural cause. Indeed, although much has been made of Ghazālī’s famous critique that observation can only affirm regularity and not causality between natural things, little of the argument that follows

130  Emann Allebban

Figure 7.1  The main moves of Ghazālī’s argument in Discussion 17.

in Discussion 17 is about this question of necessity between cause and effect (Figure 7.1). What emerges is that, for Ghazālī, the question of will is central to his argument and, for him, necessary for any theory of causation that allows for the miraculous.17 Ghazālī structures his critique in terms of two possible causal theories that the philosophers may hold with regard to cause–effect pairs. In the first causal theory, a philosopher holds that fire is the agent (fāʿil), and it is an agent by nature (ṭabʿ) not by choice (ikhtiyār). Thus, when brought into contact with cotton or anything flammable, and assuming all other conditions are in place, the fire can only act in accord with its nature – i.e. burn. How would this explain allowing certain miracles and ­disallowing others? If fire is the one and only agent of burning, and if fire acts from its nature, then prophets cannot make it so that fire does not burn, since fire has no ability to choose in its action – to sometimes burn and to sometimes not burn, all else being equal. However, this theory would still allow for prophets to cause rain because, in the latter case, there is nothing that is being asked to act contrary to what its nature prescribes. That is, the nature of water is such that water is receptive to both cooling and warming, but a prophet cannot, say, make water burn or talk. Thus, while it still sounds extraordinary that a human soul can cause rain or an earthquake, such events do not ask of things in the world to act contrary to their natural powers.18

The Metaphysics of Miracles  131 The second causal theory a philosopher may hold in Ghazālī’s discussion is that it is the Giver of Forms or God who is the agent of burning, not fire or other sublunar causes, who act only as preparers. However, the Giver of Forms, like fire, acts by nature and not by choice, so it cannot help but cause the burning if all the appropriate sublunar conditions are in place. I take it that this theory too would allow for some miracles for the same reason that the first theory does. Presumably a prophet can still cause rain in this causal theory, because, again, none of the entities in question are being asked to act in ways that are outside of their nature. But it would not allow for the Giver of Forms to stop emanating the suitable forms, since the Giver of Forms, like fire, acts by nature and not choice. Upon contact of fire and cotton, and assuming all other conditions are in place, burning will still result. Just as the prophet cannot override the nature of fire and cause it to have, say, a cooling effect, so too the prophet cannot override the nature of the Giver of Forms, who causes forms appropriate to the sublunar conditions, in this case burning when fire and cotton come into contact. This is how Ghazālī explains the prophet’s enacting miracles, If, then, it is possible for the corporeal parts of his body to obey it, it is not impossible for corporeal parts other [than his body] to obey it. His soul then looks toward the blowing of a wind, the falling of rain, a storm of lightning, or a quaking of the earth for the annihilating of a people— all this being contingent on the occurrence of coldness, heat, or motion in the atmosphere— causing the cold or heat in his soul, from which these things are generated without the presence of a ­manifest natural cause. This, then, would be a miracle for the prophet; but it would only occur in an atmosphere disposed to receive [such action], and it would not extend to having wood change into an animal and to having the moon, [which is] not receptive of cleavage, be split.19 As Ghazālī explains, the issue for the philosophers is that however p ­ owerful this prophetic agent is, he cannot make some matter become something it is not disposed to become, or make something act in ways outside of its nature. This serves as the dividing line between miracles that are possible and ­miracles that are not. But why would it be impossible for things to act or be acted upon in ways that are outside of their natural propensities? This relates to the philosophers’ views on potentiality and hylomorphism. Ghazālī focuses on the agent’s ability to choose as being a blocking factor, as is clear from the two positions he attributes to philosophers as to why they cannot allow certain miracles. In both, he says, in their voice, that burning must result because the agent of the burning acts by nature and not by choice, whether the agent of burning is the fire as in position 1 or the Giver of Forms as in position 2. For things in nature to act or be acted upon in ways contrary to their nature, they or some other cause must will it to be

132  Emann Allebban so. And surely this alone cannot be enough, for nature would still serve as a limiting factor in that it would (i) demand the need for some level of power in the agent to override, impede, or change it and (ii) limit the possibilities to some degree for any agent, even if omnipotent. The latter is another consideration in answering why it is impossible for things to act or be acted upon in ways that are outside of their natural propensities. Ghazālī anticipates it toward the end of his argument and as a part of one of his suggestions as to how miracles may be enacted under position 2 of the philosophers. He clarifies that, in stating that a staff may become a serpent, one is not thereby saying that it is both earth and not earth, which would be impossible, but rather that there is one form leaving and another coming.20 7.1.2

A Modified Position for Miracles

I conclude my discussion of Ghazālī with an outline of two ways he thinks even the impossible miracles may be accommodated within the philosopher’s theory. Before turning to those, it is clear that Ghazālī sees no recourse under the first position of the philosophers, as all of his responses to that position reject the view that fire (alone) is the agent of burning. His first argument is that observation does not prove that fire is the cause of burning, only that it occurs “side by side” with burning. Then, he argues that the philosophers themselves do not think that fire is the cause of burning; it is rather God or the intermediary angels that are the cause while sublunar things only prepare matter for the reception of forms. This effectively brings him to the second position of philosophers, namely that the celestial principle, i.e. the Giver of Forms, is the agent of burning and sublunar causes are only preparers, and that the Giver of Forms causes by nature and not choice, in the way that light proceeds from the sun. In response to this theory, Ghazālī first rejects the notion that God or the celestial principles act by nature and not choice, referring the reader to his discussion of the world’s creation where he treats this at length.21 He also responds to an objection to his own view. The ­objection is that a willing God as cause undermines the possibility of knowledge since nature would then have no predictable course but is rather entirely up to the whim of the Creator. In response, Ghazālī advances an occasionalist theory of human knowledge, namely that God causes in us knowledge that the habitual course of nature will continue when it does, and then when God disrupts it with a miracle, He does not create in us that same knowledge that such breaks in nature do not happen. Thus, if Ghazālī is successful in defending the view that God is the cause of the burning and that He does so out of choice, upon contact of fire with cotton, then God could just as well choose not to do so in some circumstance and thereby produce a miracle. This brings us to the more interesting and relevant part of his discussion where he concedes, for sake of argument,22 the philosophers’ causal theory

The Metaphysics of Miracles  133 but suggests two ways in which the problematic miracles may still be accommodated in a way, he thinks, is consistent with the philosophers’ views. The first is that God creates a quality, i.e. in the fire or in the prophet in the case of the Abrahamic miracle, so that the prophet does not get burned. As he writes, Thus, there would come about either from God or from the angels a quality in the fire which restricts its heat to its own body so as not to go beyond it. Its heat would thus remain with it, and it would [still] have the form and essence of fire, but its heat and influence, however, does not go beyond it. Or, there occurs in the body of the individual a quality which will not change him from being flesh and bone [but] which will resist the effect of the fire. For we see [that] a person who covers himself with talc and sits in a burning furnace is not affected by it.23 He concludes that this is well within the power of God to do and ought not be denied. The second way to allow for the problematic miracles under the ­philosophers’ causal theory is that God cycles matter through its transformations quicker than usual, i.e. in the case of the changing of the staff into a serpent or raising the dead. Ghazālī is attempting to provide a model for understanding such a miracle in a way that does not involve the problematic view of matter becoming something it is not disposed to become, or something acting in ways contrary to its nature. It is not that wood is directly becoming an animal, but rather it cycles through the usual transformations much quicker than usual. As he explains, Thus, earth and the rest of the elements change into plants, plants— when eaten by animals— into blood, blood then changes into sperm. Sperm is then emitted into the womb and develops in stages as an animal; this, in accordance with habit, takes place in an extended ­ period of time. Why, then, should the opponent deem it impossible that it lies within God’s power to cycle matter through these stages in a time shorter than we are acquainted with?24 Ghazālī explains that the causes of the various dispositions of matter include “strange and wondrous things”25 so we ought not be quick to deny the ­possibility of matter transforming quickly. For instance, he discusses the case of spontaneous generation, whereby life is generated from the earth, such as the worm as well as the mouse, snake, and scorpion. He also discusses the talismanic arts, where the master is able to, for instance, rid a town of the bedbug, snake, or scorpion through her knowledge of the stars and the ­special properties of minerals. Ghazālī cites these examples to show that we do not fully understand the causes of the dispositions of matter, but we know at least that the causal story is complicated – that we cannot fully ascertain it and that it involves the superlunar world, as is evident through these “strange

134  Emann Allebban and wondrous” examples that he gives. As a result, he thinks we shouldn’t be quick to deny the possibility of rapid matter transformation. In offering these two ways that miracles may still be possible under the ­philosophers’ causal theory, it is not entirely clear whether Ghazālī is ­suggesting that under this modified theory one must at least admit that: (1) God is acting by will and not nature and (2) sublunar causes are still only ­preparers and not causes of burning as well. With regard to the latter, my sense is that he is here continuing with the view that fire is not the real cause of burning, but only a preparer, as he had long ago dispensed with p ­ osition 1 of the philosophers that it is fire alone that is the agent of burning. In the modified theory it seems he is conceding, for the sake of argument, some degree of secondary causation but trying to show how it would still be ­compatible with miracles. What is less clear to me is whether he is also conceding, for the sake of argument, the philosophers’ God: a God who acts by nature and not by choice. Marmura thinks Ghazālī is not conceding this, that in proposing this possible solution he takes it as at least necessary to hold that God is a ­willing cause.26 I can see how this would make sense, since it seems in order to create the blocking quality or to transform matter more quickly, God would need to will it to be so as a kind of exception. If God were causing by nature, such exceptions would not happen, as the habitual course of nature would just continue. (To be clear, for the philosophers, it is the prophet and not God who is doing this, which is, I take it, the main advantage of developing a theory of how a prophet can enact miracles: you don’t need to introduce choice or will in the divine in order to produce such an exception.) However, I take pause, because this assumes that God could not have an exceptional act be part of the emanative order from eternity. That is, in God’s timeless knowledge, God knew such circumstances would arise where a prophet would need to be saved from fire or prove his prophecy, so as to “manifest the good.”27 If the good emanates from God as a consequence of God’s nature, then an emanation of the quality could be part of that and not necessarily a direct intervention in time as in an occasionalist conception of divine action. I also take pause because, at least argumentatively, if Ghazālī is not conceding for the sake of argument the philosopher’s God who acts by nature and is instead making these suggestions under the assumption that one accepts a God who acts from choice, then his suggestions can no longer be read as trying to make space for miracles under the philosophers’ causal theory. And if that is the case, it would seem odd to discuss the ­solutions at all, if, ultimately, the philosophers would not accept the foundational premise of Ghazālī’s choosing God.28 Granted, Ghazālī’s suggestion would still be much more palatable to the philosopher than an occasionalist framework: it accommodates some degree of secondary ­causation and a world of natures. At this juncture, many questions remain unanswered. One glaring question is what, according to Avicenna, is the prophet’s causal role in effecting miracles. Is the prophet (1) preparing the matter in question to receive the desired form; (2) providing the desired form for properly prepared matter;

The Metaphysics of Miracles  135 (3) both; or perhaps even (4) neither – i.e. the prophet is imposing a form on matter that is not disposed to receive it. The answer to these questions would affect not only the plausibility of any theory of miracles in the falāsifa’s hylomorphic context, but it would also affect the question at hand of why certain miracles are allowed and others not. However, notice that under the first three options, one may still be able to tell the same story as to why a staff becoming a serpent or fire not burning is impossible while prophet-induced rain is possible. The prophet’s acts must fall within the established range of natural propensities, as defined by things’ natures, and cannot extend to causing something to act contrary to its nature or take on functions that are outside its nature. Thus, even a prophet cannot bring a decapitated body to life, i.e. have it grow, move, and think. Of course, if the last option is the case, i.e. the prophet is imposing a form on a matter that is not receptive to it, then this story would not apply, since such a model would include miracles of the staff-to-serpent and decapitated-to-life sort. The exception to this is if one accepts Ghazālī’s proposal to understand such miracles as involving a blocking quality or an accelerated transformation of matter, rather than forcing a form on matter that is not suitable for it. However, as has been seen, the accelerated transformation of matter would seem to involve the superlunar world as well, which would raise a different set of problems. Would the philosopher then need to hold that a prophet has causal influence on not just sublunar happenings but on the heavenly bodies as well? At some point, this does raise the question of whether philosophers should just give up and admit that it is God, the omnipotent, who causes such miracles instead of adding more and more extraordinary superpowers to a human. Of course, Ghazālī does not spell out these different ways of understanding the prophet’s causal roles in effecting a miracle, but perhaps we can take Avicenna’s model as an example, which I turn to next. 7.2

Avicenna on Miracles

I begin with the final book of the Ishārāt, in X:25, where Avicenna states the following: You may receive reports about the gnostics that seem to run counter to observed custom, and then you begin to disbelieve [such reports]… Hold yourself and do not be hasty in rejecting these things, for such [occurrences] have causes in the mysteries [asrār] of nature. Perhaps it will be possible for me to relate some of these causes to you.29 Avicenna here recounts several such reports of feats performed by the ­gnostics. He mentions the gnostic making rain fall, healing the sick, causing earthquakes, saving people from floods, subduing beasts, and other such occurrences that he states are not “clearly impossible.” This might be a passage that Ghazālī has in mind in his depiction of the philosophers’

136  Emann Allebban account of miracles. That is, here Avicenna mentions those miracles that Ghazālī attributes to the ­ philosophers – causing rain, a storm, or an ­earthquake – and Avicenna characterizes them as having causes within nature. It is perhaps the absence of an explicit ­mention of the other miracles (cleaving the moon, turning a staff into a serpent, and raising the dead), in addition to Avicenna’s repeated remarks that these events have causes in nature and that one ought not rush to deny them,30 that leads Ghazālī to conclude that Avicenna must not allow miracles like a staff becoming a serpent. However, here, Avicenna is only recounting those feats performed by the gnostics, while the excluded miracles are particular events that are attributed to the prophets Muhammad, Moses, and Jesus, respectively. Avicenna’s failing to mention them here need not mean that he thinks they are impossible. Importantly, Avicenna elsewhere does explicitly include the changing of a staff into a serpent as within the miracles prophets have the power to enact.31 Avicenna goes on in the next chapter of the Ishārāt to sketch this cause in nature that he alluded to, namely the advanced practical faculty of some souls, Thus, do not think it far-fetched that some souls possess a fixed habit whose influence reaches beyond their bodies and, that due to their power, these souls operate as a kind of soul for the world. As these souls influence their bodies in the world due to a dispositional q ­ uality, they also influence other bodies in the world …This is particularly so since, as you have already learned, not everything that is heated (musakhkhan) is the hot [i.e. the effect of the form of hotness], nor everything that is cooled is the cold. Therefore, do not deny that some souls have this power such that they act on bodies other than their own and that such bodies react to these souls as the bodies of these souls do.32 In this passage, Avicenna emphasizes that the scope of causal influence of certain souls extends beyond their bodies.33 However, the rest of the ­passage is less clear. He notes that such souls are similar to “a kind of soul for the world” – by which he may be drawing a parallel to the Active Intellect. This raises many questions in itself. His statements regarding the principles of bodies and qualities are a bit murky. For example, his phrasing regarding not everything that is “heated is the hot” likely refers to instances of heat that are not essential or caused by a form. The latter is how the Ashʿarite Fakhr al-Dīn al-Rāzī (d. 1210) interprets Avicenna’s view, giving the example of motion causing heat.34 Avicenna seems to make a similar reference in a passage in his De Anima of the Shifāʾ,35 Kitāb al-Nafs IV 4, in discussing the soul’s effect on the body, We say that on the whole it is within the purview of the soul that through it changes can occur in the temperament of bodily matter without any

The Metaphysics of Miracles  137 corporeal action or affection. Thus heat and cold are produced without there being a hot or cold body. To be sure, when the soul thinks up an image and [that image] becomes strongly set in the soul, the bodily matter speedily accepts a form or quality corresponding to it.36 His point here is to say that heat or cold in the body can be produced by the immaterial soul, i.e. without the actual contact or influence of a physical source of heat and cold. Avicenna goes on to give examples: the form of health in a doctor’s mind causes healing in the body and the form of chair in the carpenter’s mind causes an actual chair in the external world. As another example, he discusses the effects of positive thinking: a sick man who strongly believes that he is well produces actual healing in the body, while a healthy man who strongly believes he is sick causes actual sickness in the body. In such cases, “the efficacy of imagination is greater than that which any doctor could achieve by instruments and media.”37 Having run through these examples of the immaterial soul producing material effects on the body, Avicenna discusses the ability of a powerful and noble soul to affect not just its own body, but other bodies. In the case of the Universal Soul (al-nafs al-kulliya), which belongs to the heavens and the [sublunar] world, these [forms] may influence the nature of everything (al-kull), while in the case of the individual soul, it may affect a particular natural thing. Often a soul can influence other bodies just as it influences its own body as in the case of the evil eye and the “effective thought” (al-wahm al-ʿāmil). Indeed, when the [human] soul becomes powerful, noble, and similar to the [celestial] principles, matter (ʿunṣur) that is in the [sublunar] world obeys it, and is affected by it, and that which is conceived [as a form;  yutaṣawwaru]  in [the soul] becomes existent in the matter. This is because the human soul, as we shall show, is not impressed (munṭabiʿa) in matter (mādda). To the contrary, it governs matter by its intention (himma) toward it. If such a connection with body makes it possible for the soul to transform a ­bodily element in a manner contrary to what its nature requires,38 then it is no wonder that a really noble and powerful soul should exert its influence beyond its own body, if it is not deeply immersed in its inclination to this body and has at the same time both a dominating nature and powerful habitus.39 Here, Avicenna suggests that certain souls are so perfected that they are not limited to their own body, becoming so powerful40 as to be able to influence bodies other than their own. Of course, our souls can influence other bodies too, insofar as forms in our mind cause some physical effect in the world, as in the carpenter example. But we cannot have such effects without our body or some physical tool as a medium. Here, Avicenna is proposing that, for some souls, they may simply will such effects in bodies other than their

138  Emann Allebban own in the same way we will our own bodies to, say, move. This needs much unpacking and raises a host of questions. What is the scope of influence of such a soul – in terms of what bodies it can influence and in what way(s)? In his commentary on the Ishārāt passages, Rāzī suggests that Avicenna limits the scope of prophetic agency to the sublunar realm since the celestial natures are not susceptible to generation or corruption.41 This would make sense, since the celestial bodies are unchanging for Avicenna. However, astonishingly, Avicenna does seem to suggest that a prophet’s influence can extend to the celestial realm. In his commentary on the pseudo-Aristotelian work Theology of Aristotle, Avicenna writes, It is not impossible that the celestial bodies should be employed by souls other than their own in certain ways. Especially, when a soul has perfected its power in its own body, it may, when need or expediency so demand, employ, in its place, a higher and more noble body than its own.42 Here, Avicenna says these souls’ domain of influence can extend to the ­celestial realm in terms of, so to speak, becoming the commanding soul of the celestial bodies – in some way displacing or overpowering the celestial soul. The “need” or “expediency” he speaks of may well be when a prophet needs to cause some miracle, as necessary for the good. The question to ask here is, having gained control over the celestial body, what does the prophet’s soul do with it? Recall that one of Ghazālī’s solutions of including the “impossible” miracles within the causal theory of the philosophers is that the matter is cycled through its usual transformations much quicker than usual. This allows for the staff to become a serpent or the dead to be brought back to life without having to think that, for instance, wood directly turns into an animal. Recall as well Ghazālī’s invocation of the role of celestial causes in determining the various dispositions of matter. This might be one context that a prophet taking over a celestial body would be needed. The details of course are obscure, and I won’t attempt to sketch the causal steps involved in quickly changing the matter of the staff through the influence of the celestial body, but I suggest it only as one possible way of understanding why Avicenna thinks a prophet can and seemingly sometimes does take over control of a celestial body.43 More generally, in Avicenna’s system, we know celestial bodies play a key role in sublunar processes like generation and corruption, so it would make sense that Avicenna is trying to secure their involvement in some way in order to allow for miracles, but presumably not through their own souls, which cause necessarily as does God. I end with a concluding passage from Shifāʾ Kitāb al-Nafs IV 4. Having discussed how an immaterial soul can affect a material body, and how some souls’ influence extends beyond their own bodies, Avicenna now states, So such a soul may heal a sick person, may cause a wicked person to get sick. It is befitting for [this soul] to destroy natures and to reinforce

The Metaphysics of Miracles  139 natures and for elements to transform for it so that the non-fire becomes fire, and the non-earth becomes earth. By its will (irāda), it generates rainfall and  fertility  just as it generates an eclipse and plague, all of which accord with what is rationally necessary. In general, it is possible that its will is followed by the existence of what connects to the changing of an element with respect to its contrary [qualities] because the element by nature obeys [the soul] and in [the element] is generated what is represented in the soul’s will. This is because the element in general is obedient to the soul and its obedience to the soul is stronger than its obedience to the contrary qualities in it, and these are also among the special characteristics of the prophetic faculties.44 Here Avicenna seems to go much further than what he claimed in the Ishārāt. The Ishārāt passages do not suggest that prophets or gnostics can change fire into non-fire or intervene in the causal chain of generation and corruption. At least, one might think that if the prophet is involved in substantial changes, it would only be as a kind of preparer of matter. But here Avicenna affords them a larger role, i.e. destroying and reinforcing natures and transforming elements, to the extent that the elements obey the prophet’s soul more than their own natures. This is consistent with the earlier passage where he says that “matter (ʿunṣur) that is in the [sublunar] world obeys it, and is affected by it, and that which is conceived [as a form; yutaṣawwaru] in [the soul] becomes existent in the matter.”45 Receiving forms from the soul could have just been, say, bringing about alteration or other accidental changes in something. But reading this in conjunction with this last passage, and the other passages together, it seems Avicenna is asserting the prophet’s ability to cause substantial change in the sublunar world. The prophet’s soul is not just able to take on the role of celestial bodies, but of a celestial soul as well – ­functioning as a kind of Giver of Forms to sublunar things. The prophetic faculty seems to mash together a willing agent like an ordinary human and an Active Intellect as being the cause of the form and substantial change. The parameters remain vague. However the details fall, I conclude with a few broader questions his theory raises Avicenna holds that an immaterial thing has the ability to influence the material. It is notoriously difficult to understand and ­formulate the limits and conditions of a causal relation between immaterial and ­material objects. One such question that concerns both modern and ­premodern interpreters is what limits an immaterial soul to a specific portion of changing ­matter. Avicenna, surprisingly, doubles down and uses this obscurity to explain miracles. This of course raises many questions concerning the ­foundations and consistency of his view, including the following. Another problem is how his theory of prophecy affects our understanding of the necessity of causal relations and his essentialism more broadly. Recall that Ghazālī thinks the philosophers’ commitment to (observed) causal relations being necessary features of the world, and not simply regularities, precludes the possibility of a class of miracles – namely those that

140  Emann Allebban involve a break in certain causal relations. His opening passage of Discussion 17 depicted such a causal theory as involving a kind of two-way necessity, which he rejects.46 However, I suggested there must be more to the story, as this would not explain why philosophers, according to him, allow some miracles and deny others. When we look closer at his argument, it becomes clear that the critical issue is whether matter can take on a form that it is not disposed to receive, or whether something can act in ways contrary to its nature. That is, it concerns Avicenna’s view of natural essences and their proper active and passive properties. Ghazālī suggested ways of including miracles even if things in nature do have genuine causal powers. This notwithstanding, Avicenna’s theory of the prophetic soul affecting bodies other than its own seems to constitute a break of causal relations held to be necessary. At least, it introduces a new agent in the world that has a radical range of possible effects. That is, with the prophetic soul in the world, one cannot hold that, given some effect, the (essential) cause must exist. According to Ghazālī’s Ashʿarite view, it is possible to have quenching of thirst without drinking, satiety without eating, burning without fire, light without the sun, death without decapitation, healing without medicine, and purging of the bowels without a purgative. The difference in Avicenna is that such effects are within the prophet’s power, while Ghazālī attributes such feats directly to the divine. It remains an open question as to what the scope of Avicenna’s prophetic power is. The larger question in all of this is how the above view of the causal powers of the prophetic faculty meshes with Avicenna’s ideas about essentialism and the necessary causal order.47 A critical question regarding essentialism raised by Rāzī is that the prophetic faculty or soul that can effect such change is not an essential cause or form. That is, a prophet is essentially a human and these special prophetic powers that are the source of essential change in the world fail to rise even to the level of a true form. As Rāzī suggests, it would be strange that such a powerful intervention and ability to change natures is grounded in something that is not itself a basic form or essence. Even more, it seems that the (non-essential) prophetic faculty can intervene or change essential properties and causal relations. The causal role of the prophetic faculty raises questions about our understanding of the role of forms and debate regarding the contribution of the Active Intellect in substantial generation.48 It remains to determine where the prophet, in virtue of these miracles he enacts, falls in Avicenna’s schema of physical versus metaphysical efficient causes and their respective contributions. Miracles are an especially interesting case because they tread the lines between natural philosophy, theology, and the occult. Avicenna brings these together, drawing on his philosophical and mystical works to form his theory of prophecy. Importantly, in the end, he considers his theory to be a natural one – oft-repeating that miracles and other feats have a natural cause.49

The Metaphysics of Miracles  141 7.3 Conclusion Returning to the guiding question of this chapter, namely which miracles are allowed, and which are not, it seems Avicenna might actually have a far wider inclusion of miracles than even Ghazālī thinks. It remains to be determined the extent to which Avicenna can include such events and how consistent it is with his broader metaphysical commitments. Avicenna has attempted to introduce an agent into his necessary world with the kind of will and choice that Ghazālī thinks is needed for miracles to be possible, but just not where we might expect. It is not one of the heavenly causes, or even God, who produces these miracles. The willing agent is instead a kind of radically powerful superhuman. Rāzī is right to ask here50 whether Avicenna is, in so doing, affirming a whole other species in affording these extraordinary powers to prophets. Notes 1 I would like to thank my own miracle, Maryam Aya, who was born as I was ­writing this chapter. 2 For example, we find in the Qur’an regarding omnipotence, 7:54, “Do not ­creation and command belong to Him?” (trans. Nasr et al. 2017: 427); r­ egarding responding to prayers and proximity to the human experience, 40:60 “And your Lord has said, ‘Call upon Me, and I shall respond to you’” (trans. Nasr et al. 2017: 1151–1152) and 2:186 “When My servants ask thee about Me, truly I am near. I answer the call of the caller when he calls Me” (trans. Nasr et al. 2017: 81–82). 3 I must set aside treatment of the epistemological questions about miracles – e.g., how one can distinguish a miracle from other exceptional occurrences and what justifies such a belief. 4 Ghazālī’s summary of these three ways mirrors Avicenna’s account in Risālat alFiʿl wa-l-infiʿāl. See Avicenna (1935) qtd. in Alper (2013: 177–178). 5 See Davidson (1992: 116–122) for an overview of these first two types of “­cognitive” miracles of prophets. See also Marmura (1963: 51–52). 6 Al-Ghazālī (2002: 165) (trans. Marmura with modification). 7 Al-Ghazālī (2002: 164) (trans. Marmura with modification). 8 Al-Ghazālī (2002: 165) (trans. Marmura with modification). 9 Breaking custom not just once but several times so as to not be confused with a kind of lucky fluke, as expressed here for example by the Ashʿarite al-Bāqillānī (d.1013) in his treatise Explaining the Distinction between (Prophetic) Miracles and (Saintly) Marvels, Trickery, Soothsaying, and Magic, “The miraculousness is to be found in the violation of custom present in empowering them to produce a considerable amount of such things and to acquire it in a way in which it is not customary for them to be so empowered. So the miraculousness consists in their being empowered to produce a quantity that is contrary to custom.” (1958: 35; trans. McCarthy in Renard 2014: 174). Bāqillānī’s point in the larger context here is that the miraculous nature of the Qur’an is greater than these other miracles that may be written off by some as accomplished through magic, simple trickery or someone’s mastery of certain techniques that hide the processes of change. As he writes, “So it cannot be denied that one may suppose that a person who does away with leprosy in a brief time has come to know a remedy with the property of doing away with that in the twinkling of an eye.” (1958: 27; trans. McCarthy in Renard 2014: 170).

142  Emann Allebban 0 Al-Ghazālī (2002: 163) (trans. Marmura with modification). 1 11 He describes the causal theory in question as follows, “…this connection between causes and effects that one observes in existence is a connection of necessary ­concomitance, so that it is within neither [the realm of] power nor within [that of] possibility to bring about the cause without the effect or the effect without the cause.” (Al-Ghazālī 2002: 163; trans. Marmura). 12 Some read Ghazālī’s arguments concerning the preservation of miracles as decidedly “theological.” See for instance Barry Kogan who writes, “His project, in the final analysis, is theological,” in contrast to the philosophers who “hope to insure the possibility of scientific knowledge… Their aim is essentially epistemological” (1981: 114). Michael Marmura similarly writes, “…despite Ghazālī’s epistemological argument in the 17th Discussion, his motive throughout remains theological: to show that c­ ertain miracles deemed impossible by the Islamic philosophers are in fact possible” (1981: 87). However, concerning the theological, al-Ghazālī himself in the Deliverance from Error cautions against basing one’s faith in prophecy in such ways: “By this method, then, seek certainty about the prophetic office, and not from the transformation of a rod into a serpent or the cleaving of the moon…for if your faith is based on a reasoned argument involving the probative force of the miracle, then your faith is destroyed by an ordered argument showing the difficulty and ambiguity of the miracle.” (2019: 62–63 trans. Watt). 13 Al-Ghazālī (2002: 166) (trans. Marmura with modification). 14 McGinnis (2015: 120–121). 15 Al-Bāqillānī (1957: 43) (trans. McGinnis 2015: 120). 16 For a discussion of this and what necessary connection means for Avicenna, ­especially his comparison of causal necessity with mathematical necessity, see ­Dutton (2001: 30–35). 17 For more on Ghazālī’s view of the agent and will, contextualized within the broader kalām and falsafa dispute on whether only intelligent beings could be agents, see Normore (2015: 194–199). 18 Of course, the prophet’s theory of prophetic miracles raises a different problem: how is causing atmospheric events within the range of the natural powers of a ­human? Is it qua human that a prophet effects these changes? Or do prophets have fundamentally different natures or essences than the rest of humanity? These are questions Rāzī asks and which I return to in brief in the next section. 19 Al-Ghazālī (2002: 165) (trans. Marmura with modification). 20 Al-Ghazālī (2002: 176). 21 See Al-Ghazālī (2002: 55–77). 22 There is an exegetical question of what Ghazālī is really doing here – whether only entertaining the philosophers’ view for sake of argument and remaining an Ashʿarite occasionalist; or endorsing it; or maintaining an agnostic position. Frank Griffel, for instance, thinks Ghazālī remains undecided on whether God acts alone or there is indeed secondary causation (2020). It is beyond the scope of this piece to attempt to settle this, but see Michael Marmura (1981), ­Marmura (1995) and Richard Frank (1992). Here I will limit myself to laying out the ­options to explore their philosophical merit, especially in conversation with Avicenna’s theory in the next section. 23 Al-Ghazālī (2002: 171) (trans. Marmura with modification). 24 Al-Ghazālī (2002: 172) (trans. Marmura with modification). 25 Al-Ghazālī (2002: 173) (trans. Marmura). 26 Marmura (1981: 106–107). 27 As al-Ghazālī explains, “But the time meriting its [i.e., the miracle’s] appearance, however, is when the prophet’s attention is wholly directed to it and the order

The Metaphysics of Miracles  143 of the good becomes specifically [dependent] on its appearance so that the order of the revealed law may endure. All this gives preponderance to the side of [the] existence [of the miracle], the thing in itself being possible [and] the principle [endowing it being] benevolent and generous. But it does not emanate from him except when the need for its existence becomes preponderant and the order of the good becomes specified therein. And the order of the good becomes specified therein only if a prophet needs it to prove his prophethood in order to manifest the good.” (2002: 172, trans. Marmura with modification). 28 Not to over-simplify a contested question of what it means to assert a willing God. See Rufus and McGinnis (2015: 160–195) and McGinnis (2022: 97–109). 29 Avicenna (1985: 150) (trans. Inati 1996: 104 with modification). 30 In addition to the passage in X:25, see also X:5 and X:7. 31 Avicenna (1935) qtd. in Alper (2013: 178). 32 Avicenna (1985: 153–154) (trans. Inati 1996: 105 with modification). 33 Avicenna takes it as possible that the immaterial soul can cause physical changes in the body. For a discussion of Avicenna on the soul’s control of and influence on the body, with corresponding passages from Shifaʾ Kitāb al-Nafs, see Rahman (1958: 48ff). 34 Al-Rāzī (2005/2006, vol. 2: 658). 35 Avicenna has a lot of interesting things to say about prophecy in the ­Metaphysics of the Shifāʾ X 1-5, but his comments there are directed towards showing the ­necessity and purpose of prophethood, and the prophet’s role in divine providence. He does not treat the metaphysics of miracles or address the present ­concerns, beyond just saying that a prophet must perform miracles so that people will recognize him as such. 36 Avicenna (1975: 176) (trans. Rahman 1958: 49 with modification). 37 Avicenna (1975: 177) (trans. Rahman 1958: 50 with modification). 38 I.e., for the soul to control one’s body in doing, say, rational acts. 39 Avicenna (1975: 177) (trans. Rahman 1958: 50 with modification). 40 To be clear, for Avicenna, although prophetic miracles are explained fully by ­reference to human powers, he doesn’t think that just any human has these ­powers, at least not to the degree needed to perform such miracles. As he writes, “Moreover, this individual who is a prophet is not one whose like recurs in every period. For, the matter receptive of a perfection like his occurs only in few b ­ odily compositions,” [Shifāʾ Metaphysics X 3 (1)]. As we will see, this will raise a ­question for later commentators on the nature of the species and the differences between kinds of human souls. To the above point, a couple of chapters later, he adds, “… whoever combines theoretical wisdom with justice is indeed the happy man. And whoever, in addition to this, wins the prophetic qualities becomes ­almost a human god. Worship of him, after the worship of God, exalted be He, becomes almost allowed. He is indeed the world’s earthly king and God’s deputy in it,” [Shifāʾ Metaphysics X 5 (11)]. Of course, Avicenna outlines further distinctive properties of the prophet, other than excelling in these faculties and virtues, but that is a discussion for another time. 41 Al-Rāzī (2005/2006, vol. 2: 657). 42 Avicenna in Badawi (1978: 72) (trans. Rahman 1958: 85 with modification). Rahman also cites a passage from Farābī’s Fuṣūṣ al-Ḥikam that similarly states that the created world obeys the prophet’s soul in the way that one’s own body obeys one’s soul. 43 Another context in which taking over a celestial body might be necessary is when a prophet causes an eclipse, which he is said to do according to Avicenna in the next passage. 44 Avicenna (1975: 177–178), emphasis mine.

144  Emann Allebban 5 Avicenna (1975: 177). 4 46 “It is not a necessity of the existence of the one that the other should exist, and it is not a necessity of the nonexistence of the one that the other should not exist.” (2002: 166 trans. Marmura). 47 Modern scholars discuss Avicenna’s view that a thing becomes necessary through its cause, and if not then it’s not really the cause and one must then posit a further cause. But this only shows that something becomes necessary in existence vis-à-vis its cause, not that there can’t be other causes for that thing nor that the connection between cause and effect is necessary such that if there is fire then there is burning. That needs further argument and citing Avicenna’s oft-repeated arguments on the contingent in itself becoming necessary through its cause does not quite get us there. In light of his theory of prophecy and the passages we’ve looked at, there is now even more reason to complicate things. 48 I have argued elsewhere for a new reading of Avicenna’s theory of ­metaphysical and physical efficient causation, vis-à-vis the roles of the Active Intellect and ­sublunar natural causes in substantial change and the generation of form. See ­Allebban (2018). 49 For instance Rahim Acar argues that Avicenna has a “naturalistic” account of miracles (2017: 165–167). 50 Al-Rāzī writes, “Since he mentioned that it is possible to break causal regularity (kharq al-ʿāda) by means of a specified property possessed by the individual’s soul, and since it is his position that rational human souls are of the same species, that is, wholly the same in essence – even if he never mentions in his books any doubt over it much less a proof for it, it is necessary for him to affirm that that soul [i.e., the one with special powers], which is distinguished from other souls by this power, is not distinguished in virtue of an essential property, but rather that it must [be so] in virtue of an accidental property.” (2005/2006, vol. 2: 660). He goes on to consider three ways this accidental category can be understood: as true mixture resulting from differences in the material body or the receiving matter; as an externally/accidentally occurring mixture; as acquired through spiritual effort. See also Avicenna (2005: 365) where Avicenna affirms that the prophet “must be a human” yet possess a “special characteristic not present in other people,” namely the ability to perform miracles.

References Acar, Rahim. 2017. “A Naturalistic Explanation of Miracles: The Case of Ibn Sīnā.” Toronto Journal of Theology 33 (1): 161–173. Allebban, Emann. 2018. Conservation and Causation in Avicenna’s ­Metaphysics. Montreal: McGill University. https://escholarship.mcgill.ca/concern/theses/mc87ps877. Alper, Ömer Mahir. 2013. “The Epistemological Value of Scriptural Statements in Avicenna: Can Religious Propositions Provide the Premises of PhilosophiDemonstrations?” In Philosophy and the Abrahamic Religions: Scriptural cal ­ ­Hermeneutics and Epistemology, edited by Torrance Kirby, Rahim Acar, and Bilal Baş, 175–190. Newcastle upon Tyne: Cambridge Scholars Publishing. Al-Bāqillānī. 1957. Tamhīd. Ed. R. J. McCarthy. Beirut: Libraire Orientale. Al-Bāqillānī. 1958. Kitāb al-Bayān ʿan al-Farq bayna al-Muʿjizāt wa’l-Karāmāt wa’l-Ḥiyal wa’l-Kahāna wa’l-Siḥr wa’l-Nārajāt, ed. Richard McCarthy. Beirut: ­al-­Maktaba al-Sharqiyya.

The Metaphysics of Miracles  145 Al-Ghazālī. 2002. The Incoherence of the Philosophers, 2nd Edition. Translated by Michael E. Marmura. Brigham Young University - Islamic Translation Series. Provo, UT: Brigham Young University. ———. 2019. Deliverance from Error and the Beginning of Guidance. Translated by W. Montgomery Watt. Kuala Lumpur: Islamic Book Trust. Al-Rāzī, Fakhr al-Dīn. 2005/ 2006. Sharḥ Al-Ishārāt Wa-l-Tanbīhāt. Edited by ʿAlī R. Najafzāde. 2 vols. Tehran: Anjumān-e Āthār va Mafākhir-e Farhangī. Avicenna. 1935. “Risālat Al-Fiʿl Wa-l-Infiʿāl.” In Majmūʿ Rasāʾil Al-Shaykh al-Raʾīs, edited by ʿAbdallāh b. Aḥmad al-ʿAlawī, 1–11. Hyderabad: Dāʾirat al-Maʿārif. Avicenna. 1975. Kitāb Al-Shifāʾ/Ṭabīʿiyyāt (6): Al-Nafs. Edited by G. Qanawātī and S. Zāyid. Cairo: Al-Hayʾah Al-Miṣrīyya Al-ʿĀmma Lil-Kitāb. ———. 1985. Al-Ishārāt Wa-al-Tanbīhāt, Maʾa Sharḥ Naṣīr al-Dīn al-Ṭūsī Wa Fakhr al-Dīn al-Rāzī. Edited by S. Dunyā. Cairo: Dār al-Maʿārif. ———. 2005. The Metaphysics of The Healing. Translated by Michael E. Marmura. Islamic Translation Series. Provo, UT: Brigham Young University. Badawi, Abd Al-Rahman. 1978. Aristu Inda Al-Arab Dirasah Wa-Nusus Ghayr Manshurah. Cairo: Wakalat Al-Matbu’at. Davidson, Herbert Alan. 1992. Alfarabi, Avicenna, and Averroes on Intellect: Their Cosmologies, Theories of the Active Intellect, and Theories of Human Intellect. New York: Oxford University Press USA. Dutton, Blake D. 2001. “Al-Ghazālī on Possibility and the Critique of Causality.” Medieval Philosophy and Theology 10 (1): 23–46. Frank, Richard M. 1992. Creation and the Cosmic System: Al-Ghazâli & Avicenna. Heidelberg: Carl Winter Universitätsverlag. Griffel, Frank. 2020. “Al-Ghazali.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Summer 2020. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/sum2020/entries/al-ghazali/. Inati, Shams. 1996. Ibn Sina and Mysticism: Remarks and Admonitions: Part Four. London and New York: Kegan Paul International. Kogan, Barry S. 1981. “The Philosophers Al-Ghazālī and Averroes on Necessary Connection and the Problem of the Miraculous.” In Islamic Philosophy and ­ ­Mysticism, edited by Parviz Morewedge, 113–132. Delmar, NY: Caravan Books. Marmura, Michael E. 1963. “Avicenna’s Psychological Proof of Prophecy.” Journal of Near Eastern Studies 22 (1): 49–56. ———. 1981. “Al-Ghazālī’s Second Causal Theory in the 17th Discussion of His Tahāfut.” In Islamic Philosophy and Mysticism, edited by Parviz Morewedge, 85– 112. Delmar, NY: Caravan Books. ———. 1995. “Ghazālian Causes and Intermediaries.” Edited by Richard M. Frank. Journal of the American Oriental Society 115 (1): 89–100. McGinnis, Jon. 2015. “The Establishment of the Principles of Natural Philosophy.” In The Routledge Companion to Islamic Philosophy, edited by Richard C. Taylor and Luis Xavier Lopez Farjeat, 117–130. London: Routledge. McGinnis, Jon. 2022. “Simple Is as Simple Does: Plantinga and Ghazālī on Divine Simplicity.” Religious Studies 58 (S1): 97–109. Nasr, Seyyed Hossein, Caner K. Dagli, Maria Massi Dakake, Joseph E. B. Lumbard, and Mohammed Rustom, eds. 2017. The Study Quran: A New Translation and Commentary. Reprint edition. New York: HarperOne.

146  Emann Allebban Normore, Calvin. 2015. “Metaphysics in the Orbit of Islam.” In Aristotle and the ­Arabic Tradition, edited by Ahmed Alwishah and Josh Hayes, 177–199. ­Cambridge: Cambridge University Press. Rahman, Fazlur. 1958. Prophecy in Islam: Philosophy and Orthodoxy. London; New York: Routledge. Renard, John, ed. 2014. Islamic Theological Themes: A Primary Source Reader. ­Oakland: University of California Press. Rufus, Anthony, and Jon McGinnis. 2015. “Willful Understanding: Avicenna’s Philosophy of Action and Theory of the Will.” Archiv Für Geschichte Der ­ ­Philosophie 97 (2): 160–195.

8 Of Miracle of Miracles An Introduction to the Qur’anic View of Miracle Rouzbeh Touserkani

8.1 Introduction The Qur’an contains no single word or concept that stands for ‘miracle.’ In kalām, or Muslim scholastic theology, terms such as mu‘jizah, kharq al-ādah, and karāmah are found. However, none is the same as miracle1 nor, more importantly, are these terms derived from the Qur’an. Even though it is often said that the Qur’an uses the word āyah (which, in this chapter, I shall use in the form of the ‘āyah-sign’ construct) for ­miracle, this view oversimplifies the problem. As I shall demonstrate, the ­multi-­layered usage of āyah-sign in the Qur’an renders this common claim out of the question. In this chapter, I aim to show that the complex, multifaceted concept replacing the idea of miracle in the Qur’an can be of interest to ­contemporary philosophy of religion. The chapter is structured as follows. In Section 8.2, I try to trace the origins of the existing gap between the philosophical and the scriptural approaches to defining miracles and, in Section 8.3, I narrow down that divergence to the specific instance of the speculations in kalām versus the Qur’anic treatment of miracle. In Section 8.4, I analyse the Qur’anic world view and several of its key passages, emphasizing lexical and n ­ arrative ­structures to formulate the Qur’anic outlook on miracles. In Section 8.5, I use that outlook to offer a new definition for miracle. In Section 8.6, I p ­ rovide an analytic-philosophical reading and, by way of conclusion, discuss certain problems, including Hume’s argument, that are debated in c­ontemporary philosophy of religion. 8.2

Philosophy and Scripture

Probably the main cause behind the gap between the philosophical and the scriptural approaches to defining miracles concerns a fundamental difference between the language of philosophy versus scriptures and how these b ­ odies DOI: 10.4324/9781003327714-13

148  Rouzbeh Touserkani of text articulate the same concepts differently. The language of philosophy has been dominated, since the Pre-Socratic schools in ancient Greece, by what I term proposition-centrism. Given that philosophy strives towards logical validity and intersubjective understanding, historically it has tended towards propositions. Propositions lend themselves to formal relationships and truth values easier than either lexicon, from which they are constructed, or narratives, which take propositions as their building blocks. Naturally, philosophical lexicon has rarely been treated with the same rigour as forms of argumentation have been. However, the recent developments in linguistics and the philosophy of language have demonstrated that lexicon is hardly ever less important than propositions. For instance, without any evidential basis or impetus, we could think of brown sheep and brown goats observed on the Iranian plateau as broshoat and, next, we could start enquiring into their evolutionary origins. However, thinking of brown sheep and brown goats on the Iranian plateau as one and the same ‘legitimate’ evolutionary category without any evidential basis may very well be misleading. In other words, we could identify any subset of ­phenomena with a new term, but the more important question is whether this manner of unwarranted identification adds any value to our analyses. I shall argue that, from the Qur’anic perspective, the concept of mu‘jizah in kalām is, in its own way, similar to broshoat. The consequence of proposition-centrism is that terms are defined using statements (definitions). This is the opposite of the typical approach in c­ omplex literary texts. In scriptures, a word takes on a new specialized ­meaning from how it is used in various contexts, through its ­interrelationships in the text, and through the narratives in which it plays a role. In addition to insufficient attention paid to lexicon, a typical ­philosophical text does not rely on narrative structures, whereas in scripture it is the ­narrative that substitutes lived experience and supplies a degree of ­empirical evidence. For example, miracle stories define what a miracle is without ­providing a clear-cut definition composed of a series of propositions.2 Perhaps the main cause behind the paucity of work on a ‘philosophy of scripture’ boils down to these drastically different approaches to using ­language. Given that philosophy is invested in its own approach, reducing scripture to propositions can be likened to a translation process, which in turn needs its own tools and methods to close-read scriptural texts—­something that has only been done in the last 200 years. 8.3

Qur’an and Muslim Scholars

Another cause of the philosophy-scripture rift can be traced to paradigmatic differences. If every discipline needs a paradigm for its rules of reasoning and the weight it grants to facts and questions, then philosophical and scriptural paradigms are by nature drastically different, if not incommensurable.

Of Miracle of Miracles  149 This is clearly demonstrated, for instance, by how far questions in kalām diverge from issues that have obvious primacy over them in the Qur’an. In the ­following example, I try to explain the manner in which Muslim scholastic theologians (mutikallimūn) and philosophers have typically drawn upon the Qur’an in their works. One of the topics debated by mutikallimūn concerns the cause of mu‘jizah: does God and/or a metaphysical agent directly carry them out and prophets only seem to be performing them, or do the latter indeed author instances of mu‘jizah through the power vested in them? This question goes back to the dawn of falsafah, or Muslim ­philosophy, and kalām. On one side of the debate, Muslim philosophers, under the influence of Greek philosophical teachings, sought a rational interpreta­ tion of waḥy ­(revelation), nubuwwah (prophecy), and mu‘jizah. In building upon the works of al-Kindī and al-Fārābī (Avennasar) on prophecy, Ibn Sīnā (Avicenna) presents a well-known philosophical analysis of the prophets’ mu‘jizah, in which he attributes them to a prophet’s al-‘aql al-qudsī, or ‘holy intellect.’3 Also, rationalist theologians who lean towards the Muʿtazilah school (Mutazilites) have mostly accepted the prophets as the performers of mu‘jizah. On the other side, the attribution of mu‘jizah to God and divine agents is accepted, naturally enough, by the dogmatically inclined Ashʿariyyah (Asharites). However, there has been a third group of religious thinkers, ­including Imam al-Ghazālī and Imam ar-Rāzī, who tried to answer this question by referring to the Qur’anic text, providing a more nuanced solution. The unqualified question of the cause of mu‘jizah implicitly carries the presupposition that they were all the same in terms of their cause. By analysing the Qur’anic text, we find that this is not the case at all. In fact, it seems that in some instances, such as the turning of Mūsā’s (Moses) staff into a serpent, not only did the prophet not interfere, but he was not even aware of what was about to happen. In other instances, such as ‘Īsā’s (Jesus) resurrection of the dead, the will of the prophet (with divine permission) is directly involved. Therefore, here we are dealing with a spectrum of mu‘jizah: on one end are those instances the direct cause of which is divine, and on the other end are those caused partially by the involvement of the prophet in question. This exemplifies how some Muslim scholars, in particular mutikallimūn, were obviously aware of the Qur’anic textual nuances and at times even cast doubt upon their predecessors’ long-held assumptions. At the same time, they rarely ventured as far as critiquing the terminology constructed and widely used in kalām to ask, for instance, if the concept of mu‘jizah might be flawed and in need of reconfiguration, or if isolating prophetic miracles from other types of miracle should be a justified move. It is actually my argument that this specious differentiation between the two categories of miracle has given rise to the unwarranted debate around its cause. More broadly, I do not intend to resolve questions typical of falsafah and kalām as it is this type of second-hand treatment of original Qur’anic concepts that I want to avoid.

150  Rouzbeh Touserkani Instead, I seek the framework in which the Qur’anic text itself talks about miracles. 8.4

Qur’an and Miracles

In the Qur’an, there are multiple miracle stories such as Mūsā’s parting of the sea and ‘Īsā’s resurrection of the dead. The Qur’an uses the words āyah and, with lesser frequency, bayyinah, burhān, and sulṭān whilst narrating these stories. To piece together the Qur’anic view of miracle, we need to analyse the word āyah and the other concepts it collocates with, e.g. nuzūl and bayyinah, whose prerequisite is understanding some principles of the Qur’anic worldview.4 The first principle is that all phenomena are āyah-signs of God. God has created the universe, provides for it, and is the only authority to be followed. In a nutshell, the Qur’an seeks to instil the idea that all phenomena constitute āyah-signs of the attributes of God: from the creation of the heavens and the earth, the succession of day and night, all the way to a ship sailing on the seas, we perceive āyah-signs pointing to divine creation and providence. However, the meaning of āyah in the Qur’an ranges even wider than the above. Every phenomenon can be perceived as an āyah because it signifies something beyond itself. According to this Qur’anic view, not only are all ­created phenomena signs of the ultimate truth (tawḥīd), but some can point to other truths such as qiyāmah, or resurrection of the dead. What is more, the Qur’an does not always use āyah in a strictly religious context; e.g. the Ark of the Covenant becomes an āyah-sign for the reign of Ṭālūt (Saul). In short, divine āyahs are a superset of miracles, being closer to the c­ oncept of ‘sign’ in contemporary semiotics. Therefore, it is highly inaccurate to consider āyah and miracle synonymous: the former is a general concept, of which a subset is āyahs of God, and miracles still form a subset of the latter. The second principle pertains to the hierarchical nature of existence which in a similar fashion conditions human understanding. According to this principle, as human beings physically grow up, they also gradually move up through a hierarchy of intellectual (cognitive as well as emotive) ­development. Some fail to attain higher levels as the dominant traditions of their social surroundings or their own earthbound existence dull their power of perception.5 On the other extreme, there are those rare individuals who ascend so high that they can witness malakūt, or the veiled workings of the universe on the grand scale. One criterion to differentiate people based on their level of understanding is how much they are able to grasp the meaning of āyah-signs. According to the Qur’anic worldview, although all phenomena are āyah-signs of tawḥīd, most people, due to their undeveloped capacities, do not comprehend them as they truly are. In order to guide these people towards the ultimate truth, God, whose grace goes beyond people’s merit, sends down even more obvious āyah-signs (a process known as nuzūl or irsāl in the Qur’an).

Of Miracle of Miracles  151 Extraordinary āyah-signs such as miracles of prophets, which seem to v­ iolate what is considered ‘natural,’ are unveiled so that people of ­undeveloped capacities may notice their more explicit meanings. If some do not register human birth as indicative of a higher truth, the emergence of a camel from a rock may awaken them; if they do not perceive God’s might, they may finally see it once the sea is parted before them. Thus, not only through divinely sent prophets but also by means of public and private signs that appear in the world people are shown the way to the ultimate truth. If we qualify the concept of a ‘revealed sign’ with the Qur’anic word bayyinah—i.e. ‘clear,’ ‘self-evident as to be irrefutable’—we arrive at a more narrowed-down category of ‘self-evident revealed signs’ (SERESIGNs), the examples of which bear a great deal of similarity to instances of ­miracle. On one hand, all the prophets’ miracles in the Qur’an are described as SERESIGNs. On the other hand, the term ‘revealed’ (or ‘sent-down’) indicates that these phenomena, being caused by divine intervention, occur against the natural course of events (i.e. similar to what we expect from miracles). What is crucial is the new perspective that the Qur’an adopts in relation to miracles: first and foremost, SERESIGNs signify a truth beyond themselves, and then they are sent down (in terms of the hierarchy of truths) from God to a person or people, which, as a consequence, go against the normal run of life. Finally, they are self-evident and irrefutable. There is yet a third principle of the Qur’anic worldview which ­dramatically alters the scene: the khalq vs amr duality. If what matters most about an āyahsign is its capacity for signification and if natural language is the main means of signification for human beings, then speech or production of language can stand for the process of nuzūl/irsāl: a speaker utters a truth translated into speech/writing, guiding their audience through language towards that truth itself. Therefore, any divine utterance signifying a truth is a ‘revealed sign,’ which places divine revelations in the same category as miracles. It is by the same logic that the Qur’an refers to all inspired words of God, i­ncluding itself, as revealed signs, and in the Islamic tradition, Qur’anic units are ­properly called āyahs rather than ‘verses’ (or bayt, plural abyāt, in Arabic). Thus the Qur’an posits a deep relationship between revelation and ­miracle, rendering revelation as the dual of miracle in the broader khalq vs amr binary that structures the Qur’anic world. This third principle requires a fuller ­treatment, but it would be outside the scope of the present chapter. To summarize, God has self-revealed through the creation; for those who have not adequately developed their understanding to comprehend these less direct natural signs, God reinforces that first revelation through the sending down of scriptural signs as well as miracles. Apart from positing this deep revelation–miracle relationship, there is another aspect to SERESIGNs that is critical: they bring great risk to their audience. This is demonstrated in multiple Qur’anic accounts, perhaps most notably in the narrative of Prophet Ṣāliḥ (Qur’an 17:59).

152  Rouzbeh Touserkani Ṣāliḥ is divinely appointed to invite his tribe away from their ancestral gods to tawḥīd. They reject him. In order to prove the authenticity of his mission (probably at the provocation of the tribe), it is implied that a camel miraculously emerges from the face of a rock.6 What happens is so self-­evident that it leaves no room for denial. From that point on, the camel lives amongst the tribe and is a reminder to them of Ṣāliḥ’s mission as well as the falsity of their own belief system. The trauma of the camel’s appearance together with the implication that it drinks copious amounts of the water available to the tribe become too much to bear. Some of Ṣāliḥ’s adversaries kill the camel, intending to murder the prophet too. It is at this point that divine punishment is visited upon them. Beyond schematizing what happened to the tribe of Ṣāliḥ, the narrative also dramatizes the observer’s mental struggle when faced with a SERESIGN. An irrefutable fact is suddenly presented from the outside, leaving the observer with their (typically inadequately developed) faculties torn between what they literally see in front of them and the deeply inculcated, authoritative beliefs they hold onto and which the sign has appeared to challenge. This glaring contradiction makes living with it intolerable and, therefore, has to be resolved one way or the other. Most people would then act as the tribe of Ṣāliḥ did, continuing to hold onto habit and authority. They would doubt what they experienced first-hand and gradually come to deny it, going against their instincts if necessary (Qur’an 91:10) which, from a religious point of view, is the only way to perdition (Qur’an 8:42). Thus the perilous nature of receiving a SERESIGN may be the obverse of the divine hiddenness problem.7 8.5

Towards a New Definition of Miracle

There has been a growing consensus amongst philosophers about the inadequacy of the traditional definitions of miracle such as those by St ­ Thomas Aquinas and David Hume—‘an event that exceeds the productive power of nature,’ according to Aquinas, or ‘a violation of the laws of nature,’ as Hume has it (McGrew 2019). This awareness has recently led to several attempts to formulate a new definition. In this section, I shall make my own attempt based on the discussions in the preceding sections.8 Definition: Event M is a miracle for observer O if the following conditions are true: I   M signifies a truth with religious significance. II  M takes place outside the natural course of events by divine intervention. III  O is in possession of sufficient and justifying evidence for Conditions I and II. Event M is a miracle for set A of observes if it is so for every member of A.

Of Miracle of Miracles  153 8.5.1

Forness of Miracles

According to the above definition, for each miracle, there must be an observer other than the miracle worker. In fact, if we posit that a miracle is sent by a personified deity for a purpose like the observer’s guidance, then the ­existence of an audience to receive it shall be necessary, rendering its ­omission an ­unjustified simplification. In what follows I shall try to show that this ­omission may result in false-positive instances of miracles. Theo the Mischievous: Theo, a mischievous deity, has created a world solely comprised of celestial bodies which move in space according to the Newtonian Laws of Motion except when Theo directly intervenes and modifies their spatial course. From time to time, Theo makes such an ­ ­intervention, causing them to temporarily abandon their natural trajectory. These violations of the natural laws may be interpreted as mischievous— or autotelic displays of power or something else—but, at any rate, they are not miracles. A miracle by its very nature wants to be revealed. There is no secret miracle, i.e. no miracle takes place with no observer. Let us turn to Christ’s Resurrection as an example. Secret Resurrection: No one witnessed the resurrection of Jesus as it ­happened, but he announced it to his disciples afterwards when he appeared to them, which is considered part of his miraculous resurrection. If the ­resurrection had not been announced, and the Apostles and other Christians had assumed Jesus had never walked the earth again, would this unnoticed ­resurrection have counted as a miracle? With the above definition, it is clear that an unnoticed resurrection would have been considered an act of God but not a miracle. It is also unlikely that Christian theologians would recognize the possibility of a secret resurrection. 8.5.2

Conditions I and II

Condition II is actually a rephrasing of the traditional definitions of a m ­ iracle, and I do not intend to discuss it here. Condition I states that a miracle must have religious significance. A considerable number of philosophers agree on this point. In fact, this is the most important revision offered for the d ­ efinition of miracle by contemporary philosophers of religion.9 It is obvious that natural scientists would not think of a ­phenomenon violating the established laws of nature as a miracle or attribute it to a ­ ­metaphysical agent. Such an occurrence would be interpreted as an o ­ bservational error or an anomaly and, in case of replicability, a fact ­contradicting existing theories. Therefore, under what circumstances is it j­ustified to attribute such a violation to divine intervention? I submit that only through meaning would such attribution be considered justified. The unnatural phenomenon must have a religious meaning and signify a religious truth (e.g. the Resurrection of Jesus or the parting of the sea for the Israelites); or it must fulfil a request with religious significance (e.g. when the

154  Rouzbeh Touserkani Biblical/Qur’anic prophets would perform an extraordinary act in fulfilment of their audience’s demands). 8.5.3

Condition III

This condition is a continuation of ‘forness’ condition and complements it. If there is an observer who does not comprehend a miracle—an unnatural phenomenon has occurred, the phenomenon was a divine act, or it signified a religious truth—then their presence or absence makes no difference, and it is as if the miracle took place with no observer.10 The following two thought experiments can be illuminating. The Story of Ibn Ḥayyūm: O is a good-hearted, young believer who, under peer pressure, has decided to join ISIS together with his friend, Ibn Ḥayyūm (who happens to know Hume’s argument in ‘Of Miracles’ (Hume 1999). At the meeting place, he is faced with an uncanny-looking man who claims to be an angel with a mission to dissuade him from joining an evil movement, and he has the power to do anything asked of him in order to prove the authenticity of his claim. O has a huge collection of different coins that he has brought along and a list that enumerates them all. He writes heads or tails next to each coin on the list and gives them all in a bowl to the man, asking him to make the coins face up in accordance with O’s list. It is at this moment that Ibn Ḥayyūm arrives. The man throws the contents of the bowl to the ground and leaves. Coins hit the ground according to the list, persuading O of the man’s claim. However, Ibn Ḥayyūm, who has merely witnessed the tossing of the coins without being aware of O’s condition, thinks they are heads or tails in a completely random way. O explains what happened, but Ibn Ḥayyūm refuses to believe O’s testimony since he is aware of Hume’s argument. Ascribing O’s decision not to join ISIS to fear, he leaves alone and is killed shortly after. By our definition, the event is a miracle for O but not for Ibn Ḥayyūm since his only evidence is through human testimony. In this example, even if O provides his list of heads or tails and Ibn Ḥayyūm is convinced that something extraordinary has taken place, the latter can still question its ­significance because the meaning of the event is bound up with the story that O has narrated. Based on the traditional definition, this tossing of the coins may ­constitute a violation of natural laws, hence qualify as a miracle. Now in the same thought experiment, assume O is fond of high jump and has completed a 2.30 personal record. Looking at the uncanny man’s physique, O thinks he cannot clear a 2.30 bar and asks him to jump over a 2.50 bar (the current world record being 2.45), just to be on the safe side. The man leaps over the 2.50 bar easily. This new event may be even more convincing for O, who is far more enthusiastic about sports than the bookish Ibn Ḥayyūm. In this ­version of the story, it is highly disputable to say any law of nature whatsoever has been violated and, as a consequence, it is not a miracle. Let us investigate this further by another example.

Of Miracle of Miracles  155 The Story of BP: BP is a pious young man who has led a life of m ­ editation. For the very first time, he accepts an invitation to a feast where the a­ bundance of food, wine, and women arouses in him a passion for life. On his return, he has a feeling that living the godly life means appreciating it to the full rather than abstaining from worldliness. Building on previous doubts about the possibility of a personal God, he starts to entertain deistic beliefs. At that moment as his carriage is crossing a bridge, its wheel gets into a pothole and its door is flung open. BP is about to fall headlong into the bottom of the river dozens of feet below, but it is as if a hand catches him and pushes him back into his seat. After his experience, BP is suddenly uninterested in all the thoughts he was thinking shortly before as he decides not to look back on the pleasures of the flesh but become firmer in his asceticism. BP had enough evidence of a miracle. Who, but an omniscient God, could have heard his inner struggle and who, but an omnipotent God, could have resolved it by saving him from drowning, a kind of death that he had been afraid of all his life (even you, the reader of these lines, did not know that fact until now)? Only God could have arranged such a powerful scenario in a blink of an eye. Now let us review the story from the bridge operator’s viewpoint: a ­carriage got into the oft-encountered pothole and the door handle caught in the lucky passenger’s coat, preventing his fall. There is not a vestige of any natural law violation. Furthermore, what BP does not know but is a fact well known to the operator is that at least ten carriages get into the same pothole every day, a carriage door is flung open roughly once a week, and there has been on average a near fall every month. Nothing he witnessed that night has any obviously religious meaning. Would any of these facts and statistics sway BP to come to a different conclusion? This is a miracle for BP alone, the experience and meaning of which cannot be conveyed to others who were not in his shoes on that night nor may the event and its evidence be even adequately explicable to others. Therefore, even though there is no secret miracle, there are private ­miracles—i.e. a miracle to the evidence of which only one person is privy. Given the various potential ways and methods for justified belief in miracles, it may not be possible to enumerate necessary and sufficient conditions for the actualization of justifying evidence for miracles. I would like to point out at the end of this section one of the most important justified categories of such evidence. It is claimed any evidence implying that an omniscient and omnipotent agent has caused an event is sufficient to prove that divine intervention has taken place. But any possible event only requires finite knowledge and power for its actualization: how can we deduce the infinitude of the cause by ­analysing the effect? Perhaps we can rely on mathematics to provide an adequate answer. Nearly a century after calculus was invented on the basis of the vague notion of the infinitesimals, Augustin-Louis Cauchy (1789–1857) was able

156  Rouzbeh Touserkani to find a firm footing for the concepts and methodology of calculus through a precise definition for limit. According to his definition, when we want to prove that as the variable approaches a, the amount of a function tends to infinity, it is sufficient to show that for any large number M, we can always bring the variable nearer a in order to make the function larger than M. Even though for this problem exact mathematical proof exists, we cannot find such proofs outside the world of mathematics. This fundamental limitation notwithstanding, it is still possible to show that through numerous attempts we can minimize the probability of an error to be smaller than any possible number so that we are in possession of justifying evidence. For instance, a miracle worker claims to have a divine mission and be able to perform any feat. The audience can challenge them to do something so difficult that, without violating any natural laws, it is next to impossible. If they succeed, we can repeat this process as many times as necessary to push the probability of the falsity of the miracle worker’s claim towards zero, thus producing some justifying evidence for a miracle.11 8.6

Of Hume’s Argument

Philosophical definitions do not simply articulate our intuitions; rather like in mathematics, they help us argue and produce proofs. In this section, I shall address Hume’s argument against the unreliability of human testimony which seems to be the most important philosophical argument concerning miracles. First, I demonstrate that Hume’s argument does not depend on the traditional definition of miracle, but it can be used against our new ­definition. Next, I venture to find a more realistic substitute for Hume’s argument through its critique, showing that Hume’s argument can actually be used in kalām and in service of the Islamic view of miracles. I shall begin with a lemma which I regard as the core of Hume’s argument and a straightforward corollary of Bayes’ Rule. Lemma H. Consider that from O’s viewpoint the probability of error for Channel C is PC and the probability of the occurrence of M is PM. Consider PC is sufficiently larger than PM. If O receives information about the occurrence of M through C, then O is not justified to accept it. It seems Hume implicitly bases his argument against the admissibility of human testimony for miracle on the above lemma. First, he defines ‘a ­miracle [as] a violation of the laws of nature,’ observing that its probability is extremely low. Then he offers reasons for the high probability of error in human testimony, concluding that admitting human testimony for miracles is not at all justified. In fact, the only inference Hume draws from defining miracle as a violation of the laws of nature is its negligible probability. Therefore, Hume’s argument would apply to any other definition—including our own—which deems the occurrence of a miracle as extremely improbable.12 However, if we

Of Miracle of Miracles  157 grant the high improbability of the occurrence of a miracle, is Hume’s argument still acceptable? There have been so many critiques of Hume’s ‘Of Miracles’ that even ­listing them would not fit in this chapter, and I do not intend to repeat their points. Instead, I aim to show that the standard reading of Hume’s ­argument (which I do not sympathize with personally) is an extreme simplification of real-life situations. Therefore, a more realistic retelling of it requires a ­setting more complex than the simple model behind Lemma H. This model, say ℳ, ­consists of only a recipient O, one channel C, and a message M. Illustrating my point with several thought experiments, I intend to clarify in which respects the simplifications occur and provide suggestions to revise our model from each of those viewpoints. 8.6.1

We Do Not Receive Information from One Channel Only

In a real-life model, there are a high number of communication channels that compete with one another. The Narrative of Three Channels: Hūmān rings up his friend, C′, asking about the results of the ongoing football tournament. All the results reported by C′ are unremarkable except for one: the hosts, who are also one of the best teams in the world, have been defeated seven to one. Since it is not easy to believe this extraordinary result, Hūmān calls C″ whom he regards as a more reliable source of information. What C″ reports is all well and good, but Hūmān believes the results narrated to him by C′ exactly because they contain one incredible result. Flashback: Hūmān is watching sport news. The anchor says, ‘The Cup experienced an astonishing scoreline last night….’ There is a blackout right at this point and then Hūmān calls C′. The reason why some people take miracle narratives seriously can be due to their certitude, via other means, that such occurrences are possible. For instance, BP who is convinced he has had first-hand ‘experience’ of a miracle can reasonably expect that throughout history miracles on much grander scales have taken place as well. Therefore, he knows that, with high probability, at least some miracle stories are true. Thus, he sets himself only the task of distinguishing those historically true narratives. A more realistic model of human beings faced with information is as ­follows: we receive information about events from various channels, each of which has its own error probability. In addition to human testimony, this information can be obtained via several routes to knowledge—including perception, memory, introspection, and a priori reasoning—all of which being subject to error. From a Bayesian perspective, each event has a probability of occurrence at any moment. This probability is refreshed every time new information is received depending on the error probability of the channel through which that information arrives. In short, the model behind Lemma

158  Rouzbeh Touserkani H is unrealistic not only because it assumes the existence of only one channel but also because it allows only for the binary values of acceptance/rejection. It is conceivable that every person sets a specific threshold of credibility, p0, for themselves, i.e. they would deem credible any event whose probability of occurrence surpasses p0. According to this more complex model, say ℳ′, it is possible that some information about the occurrence of an improbable event received from an unreliable channel help the probability of an event exceed p0.13 In this model, human testimony about the occurrence of a miracle is not a special category; rather, it is information concerning an event with low probability via a channel with high probability of error. Thus, it is possible that by setting one’s threshold of credibility high enough, they would hardly ever—but not never—reach a justified belief in the occurrence of a miracle (or any other low-probability event). 8.6.2

We Are Not Passive Listeners of Testimonies

The example below sheds light on another shortcoming of ℳ. The Story of the Jerichoan Man: Hūmūs is a man living in Jericho ­during Jesus’ mission. One day his friends tell him about Jesus and his miracles, including his resurrecting of the dead. However, as per Hume’s argument, he is not convinced. A few days later he is told that Jesus is in the area and is p ­ erforming new miracles, to which again he pays no attention. The last round of testimonies he ignores is that Jesus is performing miracles not only in Jericho but right outside his window. It is at this point that divine ­punishment is ­visited upon him because he did not fulfil his epistemic duties. This example demonstrates that, contrary to the setting of Lemma H, those who hear of a miracle via human testimony are not confined to a passive epistemic position. Like Hūmūs, they are able to investigate a miracle and its human testimony from different perspectives. In model ℳ′, we can rectify the problem by introducing new thresholds in addition to p0: for instance, p1 (p1 ≤ p0) can represent a threshold exceeding which would require investigating the occurrence of the miraculous event. We can continue approximating real-life situations and adopt more ­complex models. For example, the error probability of a channel is variable and depends on the content of the information: a lifelong atheist providing human testimony for a miracle is more credible than a cleric whose ­profession is to proseltyze religious ideas. For another example, following Pascal’s Wager, we can make p1 dependent on the practical importance of the event. However, Hume’s argument needs revision from another dimension which, in my opinion, is far more serious than the above shortcomings. Before ­proceeding further, let me emphasize that Hume’s argument, at least in its more realistic form, never proves that belief in the testimony of miracles is always unjustified, but rather proves the difficulty of justifying such belief.

Of Miracle of Miracles  159 8.6.3

We Can Interact with Channels and Messages

Similar to the traditional models of information theory, the previous models ignore the interaction between the recipient and the channel. In fact, both models fail to consider that it is possible to test the fallibility of a channel. In the story of Ibn Ḥayyūm, for instance, imagine that the angel (channel) testifies to the truth of Jesus’ resurrection (message). O tests the angel several times and is convinced of his authenticity. In this scenario, O is faced with a miracle-working channel and, therefore, it is reasonable to accept his testimony as true. Let us consider one final thought experiment. Miraculous Email: On 3 January 2022 you notice an email received on 30 December 2021 from an anonymous sender. The email lists incredible events that will occur on the first of every subsequent month. By the end of the year you witness, in extreme incredulity, that all the listed predictions have come true and conclude that the email itself is miraculous; you have missed only one event—the event supposed to happen on 1 January—which, incidentally, is not verifiable. Is it reasonable to accept that the event on 1 January has occurred? The answer is a definite yes, i.e. it is justified to believe that the event on 1 January has also taken place. Overall, we may summarize the conclusion of the previous two examples as follows: Thesis. It is hardly ever justified to accept a message which contains testimony of a miracle unless the content of that message or its channel is proved to be miraculous.14 This conclusion, which is a revised form of Hume’s argument, would in fact be congenial to the normative understanding of miracles in Islam. Seeing the Qur’an as a miracle is one of the cornerstones of Islamic belief: a m ­ iraculous text that testifies to earlier miracles. Many Muslim thinkers regard the Qur’an as not just a miracle but the only miracle of Prophet Muhammad. The thesis above states that a Muslim’s faith, through the Qur’an, in the miracles of the earlier prophets is justified (i.e. if a Muslim’s own faith in the miraculous nature of the Qur’an is justified).15 Muslims even go further and consider the Qur’an as an ­intersubjective miracle because of its taḥaddī. Moreover, they think of the Qur’an as ­incorruptible in its overall significance (and not necessarily its minutiae). In fact, Hume’s argument offers new prospects to modern Islamic kalām; that in the absence of prophets (i.e. miraculous channels), the existence of a miraculous message that can be grasped intersubjectively is necessary. For instance, an argument can be made that this text has picked up the function that was left off with the ending of prophethood after Muhammad’s passing. I regard the limiting of the concept of i‘jāz (the infinitive of the noun mu‘jizah) to only what earlier prophets brought forth in order to prove the authenticity of their mission to blame for the consequent lack of attention

160  Rouzbeh Touserkani paid to the broader concept of miracle in kalām and certainly as one of its theoretical weaknesses. In this chapter, I have tried to show how mu‘jizah, a concept widely discussed in kalām, is not properly grounded in the Qur’anic world view. No doubt, it is tempting to think that other such concepts may be subject to the same judgement. Therefore, a reappraisal of many ­topics in philosophy of religion, such as the problem of evil, divine hiddenness ­problem, religious language, and religious pluralism can benefit substantially from thematic and discursive studies of the Qur’an that aim to unearth its original perspectives without getting stuck in the historical quagmires of ­traditional kalām, opening the way to a ‘Qur’anic kalām’ and, more broadly, a philosophy of scripture. Notes 1 If we define a miracle as an interruption of the order of nature by divine ­intervention, then a mu‘jizah is a special miracle which indicates the truthfulness of a prophet. 2 We should note that in recent years major steps have been taken in a­nalytic philosophy to appreciate the roles that lexicon and narrative play in the ­ ­construction of texts. Movements like conceptual engineering (Cappelen 2018) and ­narrative paradigm (Fisher 1987) are especially interesting. Also, the increasing use of thought experiments from the 1980s on can be seen as a sign of the growing importance of narratives in analytic philosophy. Nozick’s experience ­machine, John Searle’s Chinese room, and Hilary Putnam’s Twin Earth are just a few well-known examples. 3 Ibn Rushd (Averroes) has spoken cautiously about this theory because he does not find it compatible with Aristotelean philosophy. 4 I owe the method I have applied in this section to Toshihiko Izutsu (Izutsu 1966). 5 Following Alvin Plantinga’s comments on the noetic effects of sin, the literature of analytical philosophy has in recent years opened up to the cognitive effects of the way people live their day-to-day lives. The reader is encouraged to consult Azadegan (2013), amongst others. 6 That is, the Qur’an only hints at the miraculous nature of the camel, but the details of its emergence have been elaborated on in ḥadīths. 7 The following seems to be the Qur’anic response to the divine hiddenness ­problem: (1) God is not totally hidden, because all phenomena are divine signs; on the contrary, it is the inability of most people to identify divinity in its signs. (2) God does not totally self-reveal since, as I hope to have shown in this chapter, divine revelation through SERESIGNs can be dangerous to their audience. Another Qur’anic aspect of the problem of divine hiddenness concerns the ­historicity of SERESIGNs and how they seem to have peaked only during the Mosaic prophethood. 8 In this section, I had two paths ahead: to introduce the new definition as a new concept or introduce it as a new definition of miracle. I have chosen the latter, which is more challenging. Readers who find this definition inconsistent with their intuition about the concept of miracle may accept it as a definition for a new ­concept: SERESIGN or miracle-for. 9 For example, John Hick claimed that ‘religious significance’ is one of the features agreed upon by most thinkers over the centuries (Hick 1993). R. F. Holland by his famous example of ‘the child and the locomotive’ claimed that a religiously significant coincidence may qualify as miraculous without violating natural laws (Holland 1965). Also, the great American philosopher Charles Sanders Peirce emphasized that a miracle could be nothing more than ‘a great wonder that serves to function as a sign‑event, a semeion’ (Tritten 2014).

Of Miracle of Miracles  161 10 Note that criteria such as the presence of an observer and the existence of s­ ufficient evidence do not make our definition purely epistemological. To define ‘reflection’ ontologically, we would say that an object should be placed in front of a mirror and the light coming from the object should be reflected from the surface of the mirror. 11 This is similar to the method suggested in the Qur’an to prove it is a miracle: the concept which has come to be known in kalām as taḥaddī, i.e. the call to all competition to produce a match for the Qur’an. 12 This reading is corroborated by another part of Hume’s chapter on miracles where he considers an Indian prince’s refusal to believe the extreme ‘effects of frost’ to have arisen from just reasoning. 13 I wish to reiterate that my use of probabilistic models should not be construed as my advocacy for Bayesian epistemology, just as offering computational ­models in physics does not translate into the belief that physical laws are necessarily mathematical. In my estimation, Hume’s argument is inherently probabilistic and, therefore, it is only appropriate to utilize probabilistic models in order to express and/or revise it. 14 This thesis is actually a rewrite of Hume’s famous statement: ‘No testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous than the fact which it endeavors to ­establish’ (Hume, 174). 15 I hold that the following statement reflects a similar line of thinking on Hume’s part (perhaps as a master of ‘innocent Dissimulation’): ‘There runs, however, through the whole of these a ridiculous comparison between the miracles of our Saviour and those of the Abbe; wherein it is asserted, that the evidence for the latter is equal to that for the former: As if the testimony of men could ever be put in the balance with that of God himself, who conducted the pen of the inspired writers’ (Hume, 180).

References Azadegan, E. (2013) ‘Divine Hiddenness and Human Sin: The Noetic Effect of Sin’, Journal of Reformed Theology, 7: 69–90. Cappelen, H. (2018) Fixing Language: An Essay on Conceptual Engineering, 1st ed., Oxford: Oxford University Press. Fisher, W. R. (1987) Human Communication as Narration: Toward a Philosophy of Reason, Value, and Action, Columbia: University of South Carolina Press. Hick, J. (1993) Philosophy of Religion, 5th ed., London: Prentice Hall. Holland, R. F. (1965) ‘The Miraculous’, American Philosophical Quarterly, 2: 43–51. Hume, D. (1999) An Enquiry Concerning Human Understanding, Ed. T. L. Beauchamp, Oxford: Oxford University Press. Izutsu, T. (1966) Ethico-Religious Concepts in the Qur’an, Montreal: McGill ­University Press. McGrew, T. (2019) ‘Miracles’, in Zalta E. N. (ed.) The Stanford Encyclopedia of ­Philosophy, ­accessed 15 May 2022. Tritten T. (2014) ‘Per Posterius: Peirce, Hume, Miracles and the Boundaries of the Scientific Game’, Argument: Biannual Philosophical Journal, 4(2): 247–266.

PART V

Faith and Its Obstacles

9 Faith, Knowledge, and Certainty An Islamic Philosophy Perspective Hamid Vahid

In the Islamic tradition, the quest for faith is often intertwined with the search for knowledge and certainty. The underlying thought is that once a believer lowers her confidence in a belief in response to her evidence, she can no longer espouse it as faith. But the traditional accounts of both knowledge and certainty leave a lot to be desired. Faith, as it is currently understood, is a multifaceted notion involving both cognitive factors (like belief) and noncognitive factors (like trust). Moreover, as almost all Muslim scholars agree, faith comes in degrees. Accordingly, it would be plausible to think that, in its early stages, faith can coexist with doubt and would involve some amount of risk. At such stages, the cognitive element of faith may be more appropriately described in terms of less committal cognitive states. Thus understood, knowledge and certainty would be more apt as descriptions of the more mature stages of faith. But even this statement risks being misunderstood in the absence of a proper understanding of both knowledge and certainty. This is particularly true about certainty. For while we have a rather firm grip on the concept of knowledge, we are not so sure how to characterize certainty especially when our goal is to understand ‘faith’. This is because the type of certainty that is often associated with faith is psychological rather than epistemic certainty. In this chapter, we will be mostly concerned with how faith is understood in mainstream Islamic philosophy. As it turns out, the common view among Muslim philosophers is that faith is knowledge that is both certain and resistant to counter-evidence. The main problem with this account is whether it can correctly describe the cognitive lives of the ordinary people of faith since it is not clear that such people can measure up to the lofty standards of faith such as knowledge let alone certainty. My goal in this chapter is to dig out resources, drawn from the philosophical traditions in Islam, to resolve such conundrums. This is how the chapter is structured. In Section 11.1, I will present the mainstream conception of faith as maintained in Islamic philosophy. Focusing on the cognitive aspects of faith (knowledge, belief, etc.), I will explain how to bring such aspects in line with the cognitive lives of the ordinary people without sacrificing knowledge as an important ingredient of DOI: 10.4324/9781003327714-15

166  Hamid Vahid faith toward which people strive. Section 11.2 deals with the more difficult problem of certainty. Here, I shall focus on Alfarabi’s articulation of certainty in his short but rich essay, Conditions of Certitude. I will be particularly concerned to compare and contrast my interpretations of Alfarabi’s text with those of other scholars. I shall consider the extent to which Alfarabi’s account of certainty might help fill the gaps in the definition of faith as suggested by Muslim philosophers. I conclude by saying a few words about the feasibility of Alfarabi’s account of certainty in the light of recent discussions of this concept. 9.1 Islamic Philosophy: Faith as Knowledge, Certainty, and Practical Commitment The concept of faith is central to the Islamic view of the relationship between God and humanity. It is no exaggeration to say that faith plays a, or perhaps the central role in the Quran when seeking to characterize humans’ optimal response to God. This explains the enormous efforts that Muslim scholars and philosophers have put into drawing the contours of the concept of faith. This is not, of course, to say that there is collective agreement on how it is to be understood. Nevertheless, it is possible to identify certain essential ingredients of the concept about which Muslim theologians and philosophers are in agreement. This kernel includes such notions as knowledge, belief, certainty, and practical commitment. For example, the Muslim philosopher, Mulla Sadra, maintains that ‘faith is nothing other than knowledge (belief) in God and the hereafter; a belief that is both certain and which one assents to in one’s heart’.1 To clarify, he adds that ‘[e]ternal life can only be secured by having knowledge that God and the Day of Judgment exist, and this knowledge is what genuine faith is’.2 In the same vein, Fayz Kashani has stated that faith is nothing but knowledge because to have faith is to assent to a true proposition which, in turn, requires the veridical representation of its object. The assent and representation constitute knowledge. Knowledge and faith come in degrees, and both are on an epistemic par.3 Mulla Sadra and other like-minded philosophers also take certainty to be a constituent of genuine faith. For example, in his commentary on Sadra, Mulla Hadi Sabzevari states that ‘genuine faith is a belief that is certain’.4 The same claim has been made by other philosophers and theologians like Muhammad Taghi Amoli.5 It seems what lies behind taking certainty as necessary for faith is that the mere identification of faith with knowledge renders faith fragile and amenable to adverse evidence. So, to ensure that faith is robust enough to withstand the onslaught of doubt, it must also be certain. The identification of faith with knowledge is not confined to the Islamic tradition. Christian philosophers, too, have advocated similar views. On a

Faith, Knowledge, and Certainty  167 prominent account, due to Calvin6, faith is knowledge of faith propositions that are revealed by God and sealed to our hearts. According to a contemporary version of this view, propounded by Alvin Plantinga, faith (at least in paradigmatic instances) is knowledge, knowledge of a certain special kind. It is special in at least two ways. First, in its object: what is allegedly known is (if true) of stunning significance… But it is also unusual in the way in which that content is known; it is known by way of an extraordinary cognitive process or belief-producing mechanism. The belief-producing process involved is dual, involving both the divinely inspired Scripture..., and also the internal instigation of the Holy Spirit.7 As we can see, the aforementioned accounts of faith are conclusively concerned with the cognitive aspect of faith like belief, knowledge, and ­ ­certainty. Of course, those philosophers were well aware that there is more to faith than its cognitive dimension. After all, they note, devils and Satan also have knowledge of the highest degree while lacking in faith. Genuine faith, they think, should also involve an impulse to worship God and submit to His will. In modern parlance, the non-cognitive dimension of faith is couched in terms of the subject having a pro attitude toward the propositional object of faith (p) i.e., considering p as desirable and having a positive, conative orientation toward the truth of p. As mentioned earlier, however, I shall be only concerned with the cognitive aspect of faith here, namely, the positive cognitive attitudes (knowledge and certainty) that a believer adopts toward a faith proposition. My main goal is to find out whether and, if so, how such attitudes can aptly describe the cognitive lives of people who profess to live a religious way of life while their cognitive attitudes fall short of the standards that underlie such epistemic achievements as knowledge and certainty. The idea then is to uncover a way of integrating the Islamic philosophers’ view of faith with the reality of how ordinary people conceive of their relationship with God. In this section, I shall focus on knowledge leaving certainty for the next section. Although knowledge or belief may accurately describe the cognitive lives of philosophers, mystics, and the heroes of faith, many are reluctant to see them as an accurate description of the states of mind of the ordinary folk who happen to live in the currently secular and intellectually unsettled climate. Accordingly, attempts have been made to associate the cognitive ingredients of faith with doxastically less committal states such as ‘acceptance’,8 ‘acknowledgment’, and ‘assuming’.9 Acceptance, for example, differs from belief in a number of respects.10 To begin when, unlike belief, acceptance is under one’s control. Moreover, one can accept that p irrespective of whether one believes that p and reasons for accepting can be practical rather than evidential. Most importantly, for philosophers who see belief attribution as requiring a strong evidential position, acceptance is seen as an apt proxy for

168  Hamid Vahid belief because, unlike believing that p, accepting that p does not require a high degree of felt certainty.11 However, the very same reasons that prompted some philosophers to look for a proxy for belief have prompted others to search for a still more minimal substitute for acceptance. What has, in particular, motivated such attempts is the widely shared perception that faith is fundamentally consistent with doubt. Those who go along this route have opted for the attitude of ‘assuming’.12 The dispositional profile of ‘assuming’ seems to permit assuming things one does neither believe nor accept while ruling out disbelieving that p. Despite such differences, however, both believing that p and assuming that p share the same disposition to take a stand on the truth of p by, for example, using p as a premise in theoretical and practical reasoning. Although the preceding accounts of the cognitive aspect of faith seem to be individually rationally motivated, the very fact that they permit a plethora of distinct cognitive states for having propositional faith has the drawback of undermining the internal unity of that concept. Fortunately, this is not an insurmountable problem. The key to resolving this worry is to construe faith as a process whose various stages and strengths are anchored to appropriate cognitive states like belief, acceptance, and assuming. After all, it is a given among all religious traditions that faith comes in degrees. Plantinga, among many others, talks about ‘the paradigmatic instances of faith’13 and ‘ideal faith’ when seeking to identify faith with knowledge. In the Islamic tradition, the Quran talks about the strengthening and weakening of faith: ‘The faithful are only those whose hearts tremble [with awe] when Allah is mentioned, and when His signs are recited to them, they increase their faith, and who put their trust in their Lord’ (8:2). Earlier, we also saw Muslim scholars such as Fayz Kashani testifying to the different strengths of faith. Indeed, the process conception of faith should be especially welcome to Muslim philosophers for the Quran itself describes the cognitive lives of people of faith in precisely such terms: The nomadic Arabs say, “We believe.” Say, ˹O Prophet, You have not believed. But say, ‘We have submitted,’ for faith has not yet entered your hearts. But if you obey Allah and His Messenger ˹wholeheartedly˺, He will not discount anything from ˹the reward of˺ your deeds. Allah is truly All-Forgiving, Most Merciful. (49:14) It would be tempting to construe ‘submitting’ in the above verse as ­‘assuming’ or ‘accepting’, but I would rather not be drawn into the troubled waters of exegesis. Our goal has so far been to see whether we can make sense of the features that Muslim scholars attribute to faith. We have noted that there is a way of accounting for their basic pronouncements about faith in a way that is compatible with the way the ordinary folk conceives of their relationship with God. This, however, still leaves us with ‘certainty’ as another ingredient

Faith, Knowledge, and Certainty  169 of faith. The problem with certainty is that it is a less worked out concept than knowledge. This would make it difficult to see why certainty has been perceived to be essential to faith when knowledge by itself seems sufficient to describe the higher reaches of faith. As we saw, however, one particular strand in the inclusion of certainty in the concept of faith was that, for most Muslim scholars, genuine faith is something that is supposed to be resilient and robust in the face of new counter-evidence. ‘Certainty’, thus, seemed well placed to ensure the stability of faith and allow the faithful to behave in ways that are in conformity with what the object of faith demands. It is quite plausible to see ‘certainty’ as denoting the maximal robustness that the higher stages of faith enjoy. But this requires having a good grip on the concept of certainty. Fortunately, there is a sophisticated account of certainty, due to the Muslim philosopher Alfarabi, on which other Muslim philosophers have invariably drawn, and to which we shall now turn. 9.2

Faith and Certainty: Alfarabi’s Conditions of Certainty

Historically speaking, certainty was a more popular concept among philosophers than it is now. This decline can be traced to its stringent standards involving immunity to doubt and radical revision. After all, very few of our beliefs can be said to have such properties. An important question that is naturally raised in this context is what good ‘certainty’ does for our cognitive life that knowledge cannot. To address such questions, we first need to distinguish between psychological and epistemic certainty. While psychological certainty pertains to the strength of our confidence in the truth of what we believe, an epistemically certain belief is held on the basis of quite strong reasons. When attributing certainty, the psychological and epistemic senses of certainty are often reflected in the following sentence constructions, respectively. i I am certain that I closed the door. ii It is certain that the door is closed. (i) pertains to my confidence in whether I have closed the door while (ii) concerns the adequacy of the grounds of my belief that I have closed the door. Going back to the question about whether knowledge renders the concept of certainty redundant, the following observations suggest a negative answer. Typically, when we say, for example, that ‘we know something for certain’, we are making a stronger claim than if we merely utter ‘we know it’. To utter ‘he knows for certain’ is not to make a redundant utterance. This shows that ‘knows for certain’ picks out an epistemically stronger cognitive state than ‘knows’. In other words, while knowledge eliminates certain error cases, epidemic certainty eliminates a larger set of such cases.14 With these preliminary remarks in the background, let us see if we can make more progress in understanding the nature of certainty, and for that, we can look to Alfarabi for guidance.

170  Hamid Vahid 9.2.1

Alfarabi’s Analysis of Certainty

Following Aristotle, Muslim philosophers normally take logic to be the discipline through which knowledge and truth can be attained. Knowledge is, however, divided into the acts of taṣawwur (conception) and taṣdīq (belief or assent). This distinction, attributed to Alfarabi, is said to have its origin in Aristotle’s Posterior Analytics where Aristotle speaks of the two kinds of pre-existing knowledge that ground all intellectual learning. For Alfarabi, conception refers to the act of understanding (fāhima) whereby one gains access to the essences of things (dhawāt) while belief involves our affirmation or judgment that something is the case. Since our beliefs may or may not match what is in fact the case, there will be a correlation between degrees of assent or belief and their successes in depicting the facts. Certitude or certainty obtains when the success of our act of believing is somehow secured. It has been claimed that the notion of certitude, as it appears in Alfarabi’s work, was conceived in sin since Abu Bishr Matta, the Arab translator of Posterior Analytics mistakenly used the term ‘al-yaqīn’ and its cognates to translate certain epistemic terms that Aristotle employs in the course of defining ‘demonstration’ and demonstrative syllogism.15 It is not, however, my intention to examine the truth of this claim. As is customary, one needs to distinguish between the context of discovery and the context of justification of a claim or thought. Whatever its origins, Alfarabi seems to have legitimate motivations to try to identify the nature of certainty and the functions it is intended to discharge. My aim in this section is to examine the conditions that Alfarabi thinks have to be satisfied if a belief is to be regarded as certain. To do so, I will be guided by two requirements: faithfulness to the text and context of Alfarabi’s words as well as maintaining consistency when interpreting those words. In fulfilling these requirements, I shall not hesitate to use the modern philosophical terminology, when it is needed, while trying to remain faithful to the spirit, though perhaps not the letter, of Alfarabi’s words. As it turns out, however, Alfarabi’s account of certainty proves to be a highly sophisticated one even by the standards of modern analytic philosophy. It is worth remarking that my understanding of Alfarabi’s conception of certainty will be almost entirely based on his most detailed account of this concept in his Conditions of Certitude.16 In this essay, Alfarabi begins with a brief statement of what he takes to be the necessary and sufficient conditions for being certain in regard to some proposition, after which he goes on to explain and clarify what those conditions consist of. Accordingly, I begin by reporting Alfarabi’s own words in regard to each of those conditions followed by my own gloss on what he is trying to get at while being guided by the two methodological principles described above. In doing so, I will be particularly concerned to compare and contrast my interpretations of Alfarabi’s text with those of Deborah Black17 and Anthony Booth.18 I shall conclude by saying a few words about the feasibility of Alfarabi’s account of certainty and examine its bearing on our discussion of the nature of faith.

Faith, Knowledge, and Certainty  171 9.2.2

The Conditions of Certainty

In the Conditions of Certainty, Alfarabi begins by giving a rough account of the conditions that he takes to constitute certainty. He considers the first condition, the belief condition, as the genus of certainty with the other five conditions constituting its differentiae. Quoting Alfarabi’s own words, these conditions are as follows. ‘Absolute certitude is: 1 to believe of something that is thus or not thus, 2 to agree that it corresponds and is not opposed to the existence of the thing externally, 3 to know that it corresponds to it, 4 that it is not possible that it not correspond to it or that it be opposed to it, 5 that it is not possible that there does exist anything opposed to it at any time, and 6 that all of this does not happen accidentally, but essentially’.19 Alfarabi then goes on to discuss each of these conditions in some detail. As can be seen, the above conditions cross-reference to one another, sometimes in a rather convoluted way. Moreover, it so happens that Alfarabi’s subsequent clarifications occasionally differ from what those conditions initially appear to express. So, it is important to keep track of Alfarabi’s thoughts and try to come up with an interpretation that, as far as possible, maximally combines the two virtues of faithfulness and consistency. The first condition (the genus of certainty) is straightforward. A subject, S, is certain that p only if S believes that p. The second condition pertains to truth and seems to require that p be true. Booth, however, construes (2) as (2*).20 (2*) S believes that p for epistemic reasons. But this seems rather far-fetched. There is obviously no reference to reasons in (2), epistemic or otherwise. Booth supports his interpretation by appealing to the occurrence of the expression ‘to agree that it corresponds’ in (2) which seems to involve some sort of attitude. But that cannot be a decisive reason to read (2) as (2*) because, when explaining condition (3), Alfarabi uses a similar set of words which, given the context, cannot have anything to do with reasons, and this is sufficient to rule out Booth’s interpretation. Condition (3), says Alfarabi, ‘is only made a condition for [certitude] because it is conceivable that there should be agreement and that it corresponds to the thing’.21 Judging by its role in this sentence, it is clear that the contentious expression that Booth appeals to is merely intended to convey the agreement and correspondence of beliefs with facts. So what (2) is really intended to state is something along the following lines: ‘[the belief] agree[s] [and] that it corresponds and is not opposed to the existence of the thing externally’. When explaining (2), Alfarabi makes it clear that he understands truth as a relation (iḍāfah) between belief and the fact:

172  Hamid Vahid if the soul’s belief is affirmative, then this thing which is external (external to the belief, that is), is also affirmative, and if the belief is negative, then this thing which is external to the belief is negative. For this is the meaning of truth.22 So, it is safe to assume that Alfarabi’s second condition of a certain belief requires that the belief be true. The third condition is tricky. It requires the subject ‘to know that it corresponds to it’. But if to be certain that p requires one to know that p, then this would make both the belief and the truth conditions (1) and (2) redundant. Let us then see how Alfarabi himself articulates condition (3): it is conceivable that there should be agreement and that it corresponds to the thing, but that believer is not aware that it corresponds, but rather, it is in his view possible that it may not correspond…in which case this is true for him accidentally. Likewise, if it does not correspond… And in this way, there may be both true opinions and false opinions… And for this reason, it is necessary that a human be aware of the correspondence of the belief to the existence or nonexistence of the thing.23 Given Alfarabi’s own description of the third condition, it is tempting to see it as requiring a second-order attitude on the part of the subject. This is, in fact, how Black is inclined to take it as implying: ‘If the third condition for certitude is absent, then, a person may indeed have a true belief that corresponds to some actual state of affairs, but she will not have the requisite second-order belief that this correspondence itself must hold’.24 Black also notes that although Alfarabi initially employs the term ‘ilm’ in his analysis of the third condition, he switches to ‘awareness’ in the course of his explication. She, therefore, claims that an acquaintance model of knowledge, rather than a mere propositional conception, is better suited to Alfarabi’s concerns. She takes this interpretation to be supported by Alfarabi’s subsequent use of the visual analogy where he says that ‘the meaning of “knowledge…” is that the state of the intellect with respect to the intelligible…comes to be like the state of vision with respect to the visible at the time of perception’.25 Black concludes that the visual analogy serves to reinforce Alfarabi’s earlier suggestion that the [third] ­condition consists of a direct acquaintance with or awareness of the extrinsic ­subject of the proposition of which one is certain. Knowledge, like vision, requires direct epistemic contact with the object known at the time when it is occurring… Certitude, inasmuch it is a relation ­dependent upon the existence of its two correlates, the knower and the known, thus requires simultaneous acts of self-awareness of one’s own cognitive states and awareness of the external object of one’s knowledge. It is a second-order act.26

Faith, Knowledge, and Certainty  173 I have several problems with Black’s second-order gloss on Alfarabi’s third condition. Starting with her claim that Alfarabi upholds an acquaintance model of knowledge, the visual analogy in fact suggests an anti-­acquaintance account. Alfarabi makes his remarks in the context in which he is discussing things that can be external to belief which he takes to include mental states such as emotions and beliefs themselves. Here, we are surely in the realm of second-order mental states and introspective self-knowledge. But an acquaintance model of self-knowledge requires both epistemic and metaphysical directness whereas, in the case of visual perception, perceptual belief is only metaphysically direct. Our awareness of our environment is, rather, mediated by sensory experience. So, if anything, what the visual analogy suggests is, to use the modern jargon, an ‘inner-sense’ account of self-knowledge where introspection is likened to perception in virtue of involving a monitoring mechanism that takes mental states as input and yields representations of those states as output.27 The connection between the input–output states is causal and contingent though it can be non-­ inferential. In fact, the inner-sense model turns out to be quite congenial to the explicit reliability streak in Alfarabi’s account of certainty highlighted by his fourth condition. As for Black’s claim that certitude is a second-order act, this is highly controversial. Being certain that p is as first-order as knowing that p is. They both concern our first-order belief that p. Black claims that, on her reading, condition (3) allows Alfarabi to introduce a level of second-order knowledge into the theory of demonstration without obvious regress or circularity. The object which one knows is now established as distinct from the object about which one is certain: Knowledge is usually about the external world…whereas certitude concerns the status of my first order belief.28 It is, however, rather difficult to make sense of this latter claim. Both knowledge and certitude concern first-order beliefs that possess particular features. Knowledge is a belief that is, among other things, true and justified whereas certitude is a belief that is true and maximally justified. They can both be about the world as well as one’s mental states as when one claims that one knows and is certain that p or one knows and is certain that one believes that p. I think the reason why Black goes for a second-order reading of the third condition is that she is inclined to construe certainty in terms of secondorder knowledge. And this is understandable for, assuming that a certain belief is one that is immune to error, there is a sense in which ‘knowing that one knows that p’ is more immune to error than ‘knowing that p’. It should however be noted that the former still falls short of delivering infallibility. For, once knowledge is taken to be fallible knowledge, which it should be so taken, iterating it cannot eliminate fallibility and so cannot give us the fullblooded immunity that we may expect of certainty.

174  Hamid Vahid Booth also follows Black in giving a second-order interpretation of the third condition along the following lines. (3*) S (infallibly) believes that S believes that p for epistemic reasons. He takes this to be an internalist condition on knowledge: Alfarabi insists on there being, as Black says, “direct epistemic contact” between the knowing subject and the object known. That is the subject must have something akin to a direct intellectual perception of the object in question… As such, it seems that Alfarabi maintained that only abstract truths can be the object of absolute certainty.29 The points that I raised against Black’s second-order interpretation equally apply to Booth’s version. But Booth’s interpretation has the further problem of taking the third condition to limit the scope of the objects of absolute certainty to abstract truths. Now, it may well be that, with the satisfaction of all the conditions of certitude, one can only be certain of abstract truths (intelligibles) but, at this stage, we are still concerned with the satisfaction of the first three conditions. So, I do not find Booth’s (3*) to correctly represent Alfarabi’s intents. On my part, I am inclined to understand the third condition as a ‘justification’ condition. Recall Alfarabi’s own account of the motivation behind this condition: ‘it is conceivable that there should be agreement and that it corresponds to the thing, but the believer is not aware that it corresponds, but rather, it is in his view possible that it may not correspond’. Here, Alfarabi seems to be conceiving of a subject who has formed a belief, say, the belief that there is an even number of stars, but without having any reasons or grounds for that belief which is why, from the subject’s point of view, it is possible that ‘it may not correspond’. The belief may, however, turn out to be true by accident. But Alfarabi maintains that ‘the condition of truth in the case of certitude is that it not be accidental’.30 Accordingly, if the predicament of Alfarabi’s conceivable subject has been brought about by his lack of reasons, it would be only natural to take the third condition as demanding that the subject have reasons for his belief and be also ‘aware’ of them. In other words, the subject must be justified31 in his belief in order to rule out its accidental truth. Note that although the subject is required to be aware of his reasons, he is not required to know that they are adequate. So, it would be fair to take the third condition as demanding that the belief be justified in a weakly internalist sense if it is to be certain. Coming to the fourth condition, it states that the belief p is certain only if it is not possible that it be false. Since I have construed condition (3) as requiring that the belief p be justified, it is quite plausible to take the fourth condition as saying that the belief p could not be mistaken. Since conditions (1)–(3) are also included in an account of knowledge that p, it is only natural

Faith, Knowledge, and Certainty  175 that, to provide an account of certain belief, Alfarabi should go modal at this stage, thus, condition (4). The modality should not, however, be understood in a logical or metaphysical sense since, when explaining the fifth condition, later on, Alfarabi himself takes sensibles to fall within the scope of condition (4): ‘for the [fourth] condition may also occur in sensibles and in existential propositions’.32 After all, we can be certain of such paradigmatic contingent propositions as ‘I exist’. So, we must have a better account of the sense of the modality that condition (4) involves. Indeed, Alfarabi himself provides a most interesting slant on the modal aspect of (4) identifying it with the assurance (taʾkīd) and strength (wathāqah) by which conviction and belief (al-iʿtiqād wa al-raʾy) enter into the definition of certitude… And this strength and assurance in the belief itself is an inference/acquisition (istifādah) from the thing which produces [the belief]. This thing is either by nature (bi-tabiah) or the syllogism.33 This shows that the ‘assurance and strength’ of a certain belief is not psychological but epistemic (truth-conducive) having to do with the ‘assurance and strength’ of the process producing the belief. In other words, it is an epistemic assurance that is conferred on the belief by that process. This, as Black says, signifies the causal-reliability character of Alfarabi’s account which includes both perceptual and inferential reliable processes. So we can understand ­condition (4) as demanding that the belief p be guaranteed to be true by the reliable process producing it. There is one final remark before we leave condition (4). When explaining the third condition, we noted that there was some pressure to understand justification in an internalist sense. This would, however, seem to create a tension with the externalist character of Alfarabi’s account as just noted. Things become more complicated if we take condition (4) to fall within the scope of the verb ‘know’ in the third condition. In that case, what (4) says is not just that ‘it is not possible that the belief p be false’ but that we know this to be so which means that we must know that the process producing the belief is reliable rather than the process just being reliable as a matter of fact. The fifth condition, however, requires that there be no time at which the belief p can be false. Alfarabi wants this condition to provide additional assurance and strength for the truth of the belief. While condition (4) involves an assurance that has its source in the process that produces the belief, the fifth condition locates a further source of assurance and strength in the truth-maker of the belief: ‘This too is an additional assurance of the acquisition/inference of the belief from the assurance of the thing which is its subject’.34 Alfarabi explicitly states that the conditions (1–4) can be satisfied in the case of sensibles or empirical beliefs such as the belief that ‘this paper is white’ or that ‘zayd is seated’. But such beliefs can cease to be true at other times. This piece of paper

176  Hamid Vahid may change color or Zayd may decide to stand up. So condition (5) seems to restrict the range of the propositions of which one can be certain to those that have fixed or permanent truth-makers. (Although Alfarabi’s description of (5) in terms of ‘truth over time’ only suggests that the truths be nomologically necessary, he calls such truths ‘necessary intelligibles’ or ‘abstract truths’.) It is trivially true that belief in such propositions can never turn out to be false. Before subjecting Alfarabi’s fifth condition to critical scrutiny, it is necessary to dispel a particular objection that Black levels against it. She observes that condition (5) ‘narrows the range of certitude to ‘necessary intelligibles’ which it seems unlikely that one could know merely accidentally. Still, Alfarabi explicitly claims that accidental certitude may occur in the case of “necessary propositions”’.35 But why is it impossible to fallibly know a necessary proposition or believe such a proposition in an uncertain manner? Certainty, as we have frequently observed, is a maximally justified belief. It is, thus, a feature of our doxastic states. One could have an uncertain belief (fallible knowledge) that p where p is a necessary truth. Although the belief p could not be false, it can still be uncertain because it does not enjoy a strong degree of justification. Having a necessary proposition as the content of such a belief does not guarantee that it is certain, only that it cannot be false. What then is condition (5) intended to achieve? It is customary to distinguish between certainty ‘at a time’ and ‘lasting or stable’ certainty. Philosophers often focus on the former kind of certainty, which may be called ‘synchronic’ certainty i.e., certainty at a time. For one might be certain that p at t1 but not at t2. Things are different with stable certainty – which Descartes calls ‘scientia’. This would be diachronic certainty or certainty over time. It is quite possible that, with condition (5), Alfarabi is after diachronic certainty or scientia. But, if so, he seems to have mislocated its source. Certainty, as we have seen, speaks to the epistemic strength of our attitude toward a particular proposition p where this strength is a function of our evidential situation vis-à-vis p. We can have certain and less-than-certain belief in p. Moreover, certainty comes in degrees. Conditions (3) and (4) both address the subject’s evidential situation taking it to be the required source of assurance and strength. But, with condition (5), Alfarabi no longer seeks to locate that source in the subject’s perspective on the world, but exclusively in the world itself viz., the beliefs’ truth-makers. A clear drawback of this requirement is that one can no longer be certain of nomologically contingent truths like ‘I exist’. As with his previous interpretations, Booth offers a radically different reading of (5).36 (5*) S (infallibly) believes that there can be no time at which there is no epistemic reason for her to believe that p. (5*) is clearly pushing Alfarabi’s sources of assurance from the outside world into the subject’s perspective i.e., her epistemic reasons or evidence. But this is, once again, reinterpreting Alfarabi’s remarks beyond recognition for there are clearly no references to ‘epistemic reasons’ in Alfarabi’s articulation of the fifth condition.

Faith, Knowledge, and Certainty  177 Finally, the last condition, (6), requires that the previous conditions hold essentially and non-accidentally. Alfarabi says that ‘it is not impossible that all these things might arise in a human being by chance, rather than from things whose natural function is to cause them to arise’.37 This seems to suggest that, with the sixth condition, he wants to ensure that the processes that produce certain beliefs are maximally reliable to minimize the possibility of error as far as possible. But how could error still arise despite the satisfaction of the conditions (1)–(5)? Alfarabi mentions a few sources. I – A belief might obtain by induction or by the testimony of people ‘in whom the person has confidence’. Although both of these sources are reliable, Alfarabi compares them with the cases in which the subject acquires beliefs from his own perspective i.e., ‘from his own vision…[or] like the state of someone who observes things at the time when he is considering it and is aware that he is considering it’.38 He clearly thinks the second route is less prone to error than induction and testimony. His intuition is, I think, justified. For when one acquires a belief through, say, testimony, there are two sources of possible error involved: (a) error due to the speaker and (b) error due to the hearer. However, when it is only the subject that forms his beliefs on the basis of his visual experiences, error can only have one source. II – Even when the conditions (1)–(5) are satisfied, error, says Alfarabi, is not entirely ruled out. For emotions can still get a foothold and influence beliefs or even ‘take the place of belief for this person… [H]is zeal and partisanship, or his long-standing anger and affection [might blind him to some relevant evidence]’.39 Alfarabi seems to think that, even with (1)–(5) being satisfied, error might still be creeping in: ‘And for this reason it is necessary to seek the thing from which and concerning which certitude arises essentially and not accidentally’.40 Black, however, thinks that condition (6) is superfluous because the other five conditions taken conjointly (and in some cases even in isolation), seem sufficiently strong to rule out such chance occurrences. For example, the requirement…that one’s belief be produced by the reliable method of acquisition…seems to rule out accidental certitude by stipulating a causal process that by itself guarantees the truth of the relevant belief.41 But this is wrong and the reason why it is wrong is a testament to Alfarabi’s acumen in recognizing the possibility of the phenomenon of accidental reliability. This is, in fact, the motivation behind the more sophisticated contemporary reliabilist theories of justification that go beyond the mere requirement that a justified belief is one that is produced by a reliable cognitive process. Consider, for example, the following scenario.42 An agent has a brain lesion that happens to produce in her the belief that she has a brain lesion. Although

178  Hamid Vahid the belief-forming process is reliable in the actual world, intuitively this is an example of an unreliable process because, in many nearby possible worlds, the brain lesion would not produce such doxastic outputs. In other words, the belief-producing process involving the brain lesion is only accidentally reliable. Reliability is a modal notion. A reliable process must not only produce true beliefs in the actual world but also in the nearby possible worlds. So, pace Black, condition (6) is not superfluous. 9.3 Conclusion It is an adequacy condition on any account of certainty that it can explain the possibility of contingent beliefs that are certain as well as uncertain beliefs in necessary propositions. The fulfillment of this condition immediately rules out those accounts which identify a belief that is certain with one that could not be false. This is indeed what makes it difficult to come up with an acceptable account of certainty for, on the one hand, we want certainty to be a higher form of knowledge and, on the other hand, not so high as to exclude certain beliefs in contingent propositions. For example, according to a well-known account of absolute certainty, a belief p is certain in case it is subjectively and objectively immune to doubt.43 Subjective immunity is guaranteed if the subject is warranted in denying every proposition (r) such that it would reduce the subject’s warrant for p if added to his beliefs while objective immunity is ensured if there is no true proposition (d) such that it would reduce the warrant for p when added to the subject’s true beliefs. But, even with the fulfillment of these conditions, certainty is not guaranteed. For it is always possible to ensure that a belief is immune to objective doubt by manipulating the conditions that fall outside the subject’s ken even when the belief in question receives only moderate support from its grounds. Retuning to Alfarabi’s account of certainty, it is difficult not to be impressed by the fact it is driven by more or less the same motivations such as the immunity to error that inspire the contemporary approaches to certainty. Of the six conditions in his account, only the fifth condition does not sit well with the rest as it seems to limit the scope of certain beliefs to the so-called intelligibles or abstract truths. However, if, as noted earlier, we read this condition as merely intending to provide a diachronic (or stable), rather than a synchronic, notion of certainty, we would end up with the following account. A belief p is certain if and only if it is true, justified (based on reason, in a weakly internalist sense), and is produced by a process that is non-­accidentally reliable. But these conditions do not seem to be sufficient for certainty for they (roughly) correspond to some of the subtlest reliabilist theories of knowledge. We still need an account that places certainty above knowledge but not high enough to deny us having certain beliefs in contingent propositions. This is, however, a shortcoming that Alfarabi’s view shares with the most sophisticated contemporary accounts of certainty. Nevertheless, his analysis of certainty is rich enough to give substance to the Muslim philosophers’ claim that certainty is an ingredient of faith.

Faith, Knowledge, and Certainty  179 Notes 1 Mulla Sadra (1987, p. 543, vol. 1). 2 Ibid. p. 58, vol. 2. 3 Feiz Kashani (1979, p. 6, vol. 1). 4 Mulla Sadra (2007, pp. 7–8, vol. 6). 5 Amoli (1958, pp. 468–469). 6 Calvin (1960). 7 Plantinga (2015, p. 63). 8 See, for example, Alston (1996). 9 For example, Howard-Snyder (2013). 10 For example, Cohen (1992). 11 Alston (1996). 12 Howard-Snyder (2013). 13 Plantinga (2015, p. 63). 14 See, for example, Beddor (2020). 15 Black (2006). 16 A translation of Sharāʾiṭ al-yaqīn in Majid Fakhry (ed.) (Beirut, 1986–1987). 17 Black (2006). 18 Booth (2016). 19 Conditions, 98. 1–4 (Black’s translation with some minor differences). 20 Booth (2016, pp. 45–46). 21 Conditions, 98. 27–28, emphasis mine. 22 Conditions, 98. 13–16. 23 Condition, 98–99. 24 Black (2006, p. 20). 25 Condition, 99 (last paragraph). 26 Black (2006, p. 22). 27 For the inner-sense account of self-knowledge, see, for example, Armstrong (1993). 28 Black (2006, p. 20). 29 Booth (2016, p. 46). 30 Conditions, 99 (second paragraph). 31 It might be objected that ‘justification’ is a modern concept and, thus, alien to the philosophical thinking of that period. But a justified belief is one that is based on reasons and, surely, the concept of reason was available then. So one can view ‘justification’ as a proxy for ‘reason’. 32 Conditions, 100. 14–15. 33 Conditions, 100. 2–10, my italic. 34 Conditions, 100. 12–14. 35 Black (2006, p. 29). 36 Booth (2016, pp. 46–50). 37 Conditions, 100. 29–31. 38 Conditions, 101. 1–3. 39 Ibid., 101. 4–5. 40 Ibid., 101. 17–19. 41 Black (2006, p. 29). 42 Plantinga (1993, p. 199). 43 Klein (1981).

References Alfarabi, A.: 1986–1987, Conditions of Certainty. This is a translation of Sarait ­al-yaqin (by Deborah Black) that originally appears in vol. 4 of al-Mantiq inda AlAlfarabi edited by Majid Fakhry, Beirut.

180  Hamid Vahid Alston, W.: 1996, ‘Belief, Acceptance and Religious Faith’, in Jordan and HowardSnyder (eds.), Faith, Freedom and Rationality, 3–27. Lanham: Rowman and Littlefield Publishers. Amoli, M.: 1958, Dorar Al-fawaed, Qom: Ismaelian Publishers. Armstrong, D.: 1993, A Materialist Theory of the Mind, London: Routledge. Beddor, B.: 2020, ‘New Work for Certainty’, Philosophers’ Imprint, 20(8): 1–25. Black, D.: 2006, ‘Knowledge (‘Ilm) and Certitude (Yaqin) in AlAlfarabi’s Epistemology’, Arabic Sciences and Philosophy, 16: 11–45. Booth, A.: 2016, Islamic Philosophy and The Ethics of Belief, London: Palgrave, Macmillan. Calvin, J.: 1960, Institutes of the Christian Religion, ed., McNeill and trans. Ford Lewis Battles, Philadelphia, PA, Westminster: Ford Lewis Battles. Cohen, L. J.: 1992, An Essay on Belief and Acceptance, Oxford: Clarendon Press. Feiz Kashani, M.: 1979, Ilm-AlUsul fi Usul Al-din, Qom: Bidar Publications. Howard-Snyder, D.: 2013a, ‘Propositional Faith: What It Is and What It Is Not’, American Philosophical Quarterly, 50: 357–372. Klein, P.: 1981, Certainty: A Refutation of Scepticism, Minneapolis: University of Minnesota Press. Plantinga, A.: 1993, Warrant and Proper Function, New York: Oxford University Press. Plantinga, A.: 2015, Knowledge and Christian Belief, Grand Rapids, MI: William Eerdmans Publishing Company. Shirazi, S. (Mulla Sadra).: 1987, Sharh Al-Usul Al-kafi, Tehran: Cultural Studies Institute. Shirazi, S.: 2007, Al-Hikamh Al-Mutaliyah fi Al-Asfar Al-Aqliyyah Al-Arbaah, Qom: Mustafvi Publication.

10 Islam and the Problem of Divine Hiddenness Imran Aijaz

10.1 Introduction In this chapter, I will discuss the problem of ‘divine hiddenness’ and how it presents a challenge to the rationality of Islamic belief. The general thought behind the problem as it is conceived of in contemporary philosophy of religion is that certain facts about God’s ‘hiddenness’ (an important term that I will clarify) are in some way incompatible or incongruent with the existence of God. The specific version of this problem that has received the most attention from philosophers of religion in recent literature is J.L. Schellenberg’s ‘hiddenness argument,’ formulated in his seminal work Divine Hiddenness and Human Reason (Schellenberg 1993). Since the publication of this work, Schellenberg has continued to develop, refine, and defend the hiddenness argument against various objections, while keeping its basic structure intact. In referring to God’s ‘hiddenness,’ Schellenberg and other proponents of his hiddenness argument are not suggesting that God’s actual existence and His act of hiding from us constitutes evidence against the existence of God. Such a suggestion will no doubt strike many theists as odd. Surely, they may think, how can God’s hiding Himself constitute anti-theistic evidence when the idea that ‘God hides’ entails God’s existence? That is, 1 God exists and God is hidden. entails 2 God exists. This construal of divine hiddenness is not, however, what defenders of the hiddenness argument like Schellenberg have in mind. Rather, as Klaas Kray clarifies, ‘divine hiddenness,’ in the sense of interest to defenders of the hiddenness argument, ‘refers to a strategy of arguing that various forms of nonbelief in God constitute evidence for God’s nonexistence’ (Kray 2013). Given the centrality of nonbelief in this strategy, referring to the problem of, or argument from, nonbelief seems preferable to the original nomenclature of DOI: 10.4324/9781003327714-16

182  Imran Aijaz ‘divine hiddenness’ (see Drange 1998, for instance). Since, relative to this strategy, ‘divine hiddenness’ is a façon de parler, I agree with Kray that it is an infelicitous phrase when used to describe attempts to argue for atheism from the existence and distribution of nonbelief in God (i.e., the absence of the belief that God exists). As philosophers of religion and others who discuss such attempts still predominantly refer to them using the label ‘divine hiddenness,’ I will follow suit. To discuss how the problem of divine hiddenness presents a challenge to the rationality of Islamic belief, I will focus mainly on Schellenberg’s hiddenness argument. In essence, Schellenberg’s argument states that a philosophical consideration of the nature of God secures the truth of the claim that if God exists, there is no nonresistant nonbelief. Since there is nonresistant nonbelief, it follows that God does not exist. In my assessment of this argument, I will argue that a Muslim theist can reasonably reject it by contesting the philosophical assumptions and views about God that it pivots on. The next section of this chapter, which takes up most of my discussion of divine hiddenness, is devoted to this objective. Following this section, I will briefly discuss how a simpler statement of Schellenberg’s hiddenness argument can reveal other philosophical difficulties for Muslim theists concerning divine hiddenness. I will conclude with some suggestions regarding how these might be addressed. 10.2

Schellenberg’s Hiddenness Argument

In his most recent critical survey of discussions on divine hiddenness, Schellenberg presents the hiddenness argument in standard form as follows: 1 If a perfectly loving God exists, then there exists a God who is always open to a personal relationship with any finite person. 2 If there exists a God who is always open to a personal relationship with any finite person, then no finite person is ever nonresistantly in a state of nonbelief in relation to the proposition that God exists. 3 If a perfectly loving God exists, then no finite person is ever nonresistantly in a state of nonbelief in relation to the proposition that God exists (from 1 and 2). 4 Some finite persons are or have been nonresistantly in a state of nonbelief in relation to the proposition that God exists. 5 No perfectly loving God exists (from 3 and 4). 6 If no perfectly loving God exists, then God does not exist. 7 God does not exist (from 5 and 6) (Schellenberg 2017a, 1). The hiddenness argument is deductively valid; the truth of its final, atheistic conclusion is secured by the truth of its preceding premises and inferences. Are the premises (1, 2, 4, and 6) true? The reflective Muslim theist can, it seems, reasonably question and reject some of them.

Islam and the Problem of Divine Hiddenness  183 Premises 1 and 2 of the hypothetical syllogism that constitutes the first part of the argument appear to be predicated on an anthropomorphic conception of the Divine, as suggested by claims about having a ‘personal relationship’ with God. While talk of having a personal relationship with God (or Jesus as the Incarnation of God) is relatively common in many Christian circles, it is, by and large, foreign to the Muslim mindset. God, according to Muslim consensus, is not a person.1 And, as far as Jesus is concerned, the Qur’an states that, although he is a great prophet (5:75, 19:30), he is not God (4:171), nor did he ever claim to be God (5:116). Indeed, God is not like anything at all in His Creation, as the Chapter of Purity (Ar. Surah Ikhlās) in the Qur’an states: Say, “He is God, the One. God, the Absolute. He begets not, nor was He begotten. And there is none comparable to Him” (112:1–4) This Qur’anic chapter encapsulates the core Islamic tenets about God that also receive significant attention in Muslim philosophical literature, where they are articulated and elaborated upon in some detail. To consider just one thread of analysis from this material, many classical Islamic philosophers (Ar. falasifah) hold that God is not ‘a God’ at all; that is, God is not a being in a class of beings (regardless of whether that class is inhabited by other beings). In his On First Philosophy, for example, Al-Kindi (d. 873), the ‘philosopher of the Arabs,’ offers a summary account of the ‘True One’ (God) that includes this point: The True One … has neither matter, form, quantity, quality, or relation, is not described by any of the remaining intelligible things, and has neither genus, specific difference, individual, property, common accident or movement; and it is not described by any of the things which are denied to be one in truth. It is, accordingly, pure and simple unity, i.e., (having) nothing other than unity, while every other one is multiple. (Al-Kindi 1974, 112)2 Al-Ghazālī (d. 1111), the great Muslim polymath and champion of Sunni orthodoxy, concurs with the classical Islamic philosophers on several matters of divine ontology, despite his frequent disagreements with them regarding religious matters. In his work on Islamic theology and spirituality, The Revitalization of the Religious Disciplines, his remarks about the nature of God are highly congruent with Al-Kindi’s account cited above: God is not a body shaped nor a substance delimited and determinate. He does not resemble bodies either in being determinate or in being susceptible of division. He is not a substance, and substances do not inhere

184  Imran Aijaz in Him; and He is not an accident, and accidents do not inhere in Him. He does not resemble any existing thing, and no existing thing resembles Him. Nothing is like Him, and He is not like anything. Measure does not limit Him, and boundaries do not contain Him. (Al-Ghazālī 2014, 109–110) Many similar examples from other Islamic thinkers can be given, where the theistic doctrine being affirmed is what Brian Davies calls ‘classical theism,’ a form of theism that he distinguishes from ‘theistic personalism’ (Davies 2004, 1–16). As Davies explains, classical theism is what, for centuries, many theists in the Jewish, Christian, and Islamic traditions took ‘theism’ to mean; on the classical theist understanding of God, roughly speaking, God is the metaphysically ultimate source and foundation of all reality. A central claim of classical theism is that God is not a person. Theistic personalists, most of whom are situated in the more recent historical timeline of theism (­people like Alvin Plantinga, Richard Swinburne, etc.), deny this claim, upholding personhood as essential to God (Davies 2004, 9).3 Although there is an ­ongoing debate between classical theists and theistic personalists regarding which of their respective accounts of God is more plausible, I am not aware of any significant Muslim participation in the controversy. This isn’t surprising, given what I’ve said about the Muslim rejection of the idea that God is a person, a rejection that follows as a strict consequence of Islamic belief. How, then, do Muslims understand Divine Love if not as the love of a person? Although there are variations and differences in Islamic accounts of God’s Love, these generally share a common foundation: God’s Love consists of His conferring goodness on all existing things. In his Treatise on Love, Ibn Sīnā (d. 1037) presents a Neoplatonic account of God’s Love, according to which this Love is responsible for the love in all existing things. He explains this point at the start of his treatise as follows: Every being which is determined by a design strives by nature toward its perfection, i.e., that goodness of reality which ultimately flows from the reality of the Pure Good, and by nature it shies away from its specific defect which is the evil in it, i.e., materiality and non-being,—for every evil results from attachment to matter and non-being. Therefore, it is obvious that all beings determined by a design possess a natural desire and an inborn love, and it follows of necessity that in such beings love is the cause of their existence. (Ibn Sīnā 1945, 212) The basic idea here is that every being is driven by love toward a telos of perfection. This love is implanted in each being by God. It is, says Ibn Sīnā, ‘a necessary outcome of His wisdom and the excellence of His governance to plant into everything the general principle of love’ (Ibn Sīnā 1945, 213). In his Book of Healing, Ibn Sīnā explains that God loves all the order and good

Islam and the Problem of Divine Hiddenness  185 in His creation (Ibn Sīnā 2005, 292). Al-Ghazālī, although very critical of Islamic Neoplatonism, nevertheless agrees with Ibn Sīnā on the general point about God’s Love consisting of His conferring goodness on all creatures. Commenting on the description of God as The Loving (Ar. Al-Wadūd), one of God’s 99 names (attributes) mentioned in the Qur’an (11:90, 85:14), he writes: Al-Wadūd—the Loving Kind—is one who wishes all creatures well and accordingly favours them and praises them. Its meaning is close to ‘the Merciful’, but mercy is linked with one who receives mercy, and the one who receives mercy is needy and poor. So the actions of the Merciful presuppose there being one who is weak to receive mercy, while the actions of the Loving-kind do not require that. Rather, bestowing favours from the outset results from loving-kindness. (Al-Ghazālī 1992, 118–119) The theological understanding of such articulations of God’s Love, focusing on God conferring goodness on His creation, can be found in the Qur’an itself (see e.g., Muhammad 2010, 13, 37–40). Having made clear the points that, for Muslims, God is generally not understood as a person and that, consequently, God’s Love is not the love of a person, we are now well-positioned to see how Schellenberg’s hiddenness argument has little force against Islamic theism. Let’s grant, for the sake of argument, that the hypothetical syllogism (1–3) in the hiddenness argument is sound and consequently that 3 is true: 3. If a perfectly loving God exists, then no finite person is ever nonresistantly in a state of nonbelief in relation to the proposition that God exists. Since this conclusion is arrived at through an inference that rests on the truth of theistic personalism, it is better to state it as follows (I’ll call the revised conditional below 3*): 3*. If a personal, perfectly loving God exists, then no finite person is ever nonresistantly in a state of nonbelief in relation to the proposition that a personal God exists. In response to 3*, a Muslim theist can simply shrug his or her shoulders and note that it passes harmlessly over the orbit of Islamic doctrine. If the hiddenness argument secures the truth of the conclusion that no personal God exists, he or she may think, so be it. Similarly, the Muslim theist need not contest the truth of premise 4 of the hiddenness argument, which, once clarified, states that some finite persons are in a state of nonbelief in relation to the proposition that a personal God exists. Let’s call this clarification 4*:

186  Imran Aijaz 4*. Some finite persons are or have been nonresistantly in a state of nonbelief in relation to the proposition that a personal God exists. The Muslim theist may not only let 4* pass without argument, but he or she may even welcome its truth, thinking it a good thing that people do not believe in the existence of a personal God. Since 3* and 4* are of no threat to the Muslim theist, neither is the conclusion that validly follows from their conjunction. I’ll call this 5* to make explicit the point that it is the existence of a personal God that is being denied: 5*. No personal, perfectly loving God exists. All well and good, the Muslim theist may say to this conclusion. There is no personal God who loves us in the way, say, Aphrodite (Venus) loves Adonis: ‘Thrice-fairer than myself,’ thus she began, ‘The field’s chief flower, sweet above compare, Stain to all nymphs, more lovely than a man, More white and red than doves or roses are; Nature that made thee, with herself at strife, Saith that the world hath ending with thy life. (Shakespeare 1593) This example will, of course, strike many as jarring if juxtaposed with theism, but I intend it as such to stress an important point. The problematic comparison of God to Aphrodite (Venus) doesn’t merely rest on picking a GrecoRoman deity whose psychological profile significantly differs in degree from God’s, so that some other Greco-Roman deity would offer a better approximation of who God is (e.g., Persephone). For the Muslim theist (and classical theists in other religious traditions), such comparisons rest on a fundamental category mistake. While philosophers may be more sensitive to this point and give it due consideration, it is often lost on public advocates of atheism who write for a general audience. Perhaps the best example of this is Richard Dawkins, one of the major figures of the so-called ‘New Atheism’. Commenting on Dawkins’ understanding of God that drives his anti-theistic diatribe in The God Delusion, Terry Eagleton reacts to the view of God in play and also offers a classical theist corrective: Dawkins speaks scoffingly of a personal God, as though it were entirely obvious exactly what this might mean. He seems to imagine God, if not exactly with a white beard, then at least as some kind of chap, however supersized. He asks how this chap can speak to billions of people simultaneously, which is rather like wondering why, if Tony Blair is an octopus, he has only two arms. For Judeo-Christianity, God is not a person in the sense that Al Gore arguably is. Nor is he a principle, an entity, or

Islam and the Problem of Divine Hiddenness  187 ‘existent’: in one sense of that word it would be perfectly coherent for religious types to claim that God does not in fact exist. He is, rather, the condition of possibility of any entity whatsoever, including ourselves. He is the answer to why there is something rather than nothing. God and the universe do not add up to two, any more than my envy and my left foot constitute a pair of objects. (Eagleton 2006) Certainly, I don’t mean to suggest that Schellenberg’s hiddenness argument is motivated by the kind of caricature of God one finds in Dawkins’ writings, or that it is insensitive to the philosophical discussions of God in religious traditions the way that Dawkins’ apologia for atheism is. There is agreement between Dawkins and Schellenberg, however, in taking God to be a personal God, notwithstanding the latter’s erudition in dealing with theistic critics in the philosophy of religion. A category mistake is a category mistake, even if it is sustained by philosophical sophistication and rigor. One can admire as sophisticated, for instance, an articulation and defense of mind-body dualism, but reject it as mistaken because it fundamentally misunderstands what ‘mind’ is (see Ryle 1949, Chapter 1, for a well-known discussion of this example in philosophy). To get from 5* to the hiddenness argument’s final conclusion, Schellenberg presents one last premise: 6. If no perfectly loving God exists, then God does not exist. Again, clarification here is important. By 6, Schellenberg means what I’ll call 6*: 6*. If no personal, perfectly loving God exists, then God does not exist. From 5* and 6*, we have the bridge to the main, atheistic conclusion of the argument: 7. God does not exist. Obviously, 7 will be contested by the Muslim theist, who otherwise, as I have been arguing, won’t have much trouble with the hiddenness argument (up to 5* in its line of reasoning). To rationally reject 7, the Muslim theist will have to deny 6*. One way that he or she can do this in a relatively easy manner is to point out that Schellenberg’s hiddenness argument ignores the significant history of classical theism, including several defenses of it by thinkers in the Islamic tradition (as well as by those in other religious traditions). Since 6* can be rationally resisted by the Muslim theist for whom classical theism provides a philosophically appropriate model to think about God, the hiddenness argument fails as an argument against the truth of Islamic theism. Here, then, is an initial Muslim response to Schellenberg’s hiddenness argument. Can it be sustained against objections? Let’s see.

188  Imran Aijaz In his most recent critical survey of the hiddenness argument (Schellenberg 2017a), Schellenberg is aware that the argument may be criticized for its reliance on anthropomorphism. He writes: Perhaps it will seem that conception of God I have been developing is too anthropomorphic. Let me pause to point out two things. One, even if we think of God as a loving person, it has to be allowed that the mode of God’s loving behavior and the texture of God’s inner life might be rather different from that of any human person. Two, because we are purposely testing one way of thinking about the ultimate, we must deliberately stick to the concept of a person. If we do not, what we say in description of God threatens to collapse back into the more generic ultimistic idea we have elaborated, and the point of the exercise is lost. (Schellenberg 2017a, 2) I will address the two points here in turn. The first point can be granted while continuing to press the objection that Schellenberg is nevertheless relying on an understanding of God as a person. A Muslim theist can concede that the inner life of a divine person, a person like Zeus, say, may very well be rather different from the life of ordinary persons. Zeus’ sense of justice, for instance, may significantly differ from ours, so someone in Euthyphro’s situation (prosecuting his father) should engage in moral deliberations more judiciously than Euthyphro himself in attempting to follow Zeus as a moral exemplar (Euthyphro 5d-6a). Differences between Zeus’ justice and human justice are, however, still differences that exist between persons, and it is this attribution of personhood to God that the Muslim theist will reject. Maintaining that there may be great scope and variety within personhood offers no justification at all for thinking that God is a person, nor does it assuage concerns that applying personhood to God is, at least for the Muslim theist, a category mistake. The first half of Schellenberg’s second point seems unobjectionable. If the hiddenness argument is motivated by an interest in philosophically exploring one possible construal of ultimate reality, the existence of a personal God, as an assumption, then it is, of course, only reasonable to retain this construal for the whole of the thought experiment. But if this is what the hiddenness argument amounts to – an elaborate thought experiment about the Divine affixed to a personal conception of God – then it is difficult to see how it poses any sort of threat to the Muslim theist. He or she can consign the focus of the hiddenness argument, a personal God, to the category of unsuitable candidates for the Station of the Divine, in a way that is consistent with this thought experiment. This category includes a physical god (idol), multiple gods, or a triune god, to name a few. In the Qur’an, the lifeless nature of physical idols and their inability to do anything whatsoever is cited as evidence against their alleged divine status (e.g., 21:52–67). Polytheism can also be ruled out, the Qur’an maintains, because we would see chaos and disorder if multiple gods existed (21:22); we do not, however, see chaos and

Islam and the Problem of Divine Hiddenness  189 disorder in God’s creation, which displays great harmony (67:3). For similar reasons, the Qur’an rejects trinitarianism (23:91). To this list of concepts of the Divine rejected by the Qur’an, one can add a personal God, whose nonexistence can be affirmed on the grounds that it incorporates anthropomorphism (42:11, 112:4). It is the second half of Schellenberg’s second point, however, that is problematic, viz., the claim that if we refrain from construing the ultimate as a personal God, then all we may be left with is a ‘general ultimistic idea,’ or ‘ultimism,’ as Schellenberg calls it. This is ‘the idea of a reality ultimate in the nature of things, in inherent value, and in the value it can bring to creaturely lives’ (Schellenberg 2017a, 2, 7). Ultimism, Schellenberg maintains, is a more fundamental idea than theism, which he equates with the idea of a ‘personal ultimate’ (Schellenberg 2017a, 2).4 Now, as already discussed, the Muslim theist will reject the claim that theism just is (i.e., by definition) the idea of a personal ultimate. Moreover, as a connected point, he or she will note that Schellenberg’s description of ultimism appears indistinguishable from (Islamic) theism, given that both notions lack the idea of a personal ultimate. The Qur’an affirms God’s metaphysical ultimacy (e.g., 57:3, 31:30, 6:62), His axiological ultimacy (e.g., 4:79), and his soteriological ultimacy (e.g., 2:156). These Qur’anic themes are elaborated upon in the Islamic philosophical literature on classical theism. In defending the hiddenness argument, however, Schellenberg understands theism as one possible interpretation of ultimism, an interpretation that can be ruled out by the soundness of the hiddenness argument (Schellenberg 2017a, 2). This is made clear in premise 6 of the argument, a premise that I’ve clarified as 6*: 6*. If no personal, perfectly loving God exists, then God does not exist. Since the Muslim theist can rationally resist this premise, as I have argued, the hiddenness argument fails as an atheistic argument, an argument against the truth of Islamic theism. To contest this claim, one needs to robustly engage with philosophical defenses of classical theism, something which (as far as I am aware) Schellenberg and other defenders of his hiddenness argument have not done. A number of the points I’ve raised in my discussion thus far have been persuasively laid out by Jon McGinnis in his assessment of Schellenberg’s hiddenness argument, clearly demonstrating how it is undone when they are pressed against it (McGinnis 2015). In brief, McGinnis’ critique includes the following key points: (1) Schellenberg’s hiddenness argument purportedly applies to all forms of theism, and especially Abrahamic theism of the Jewish, Christian, and Islamic kinds (McGinnis 2015, 157); (2) many of the assumptions of the argument rest on views about God that would be contested by classical Islamic philosophers (McGinnis 2015, 158); (3) these assumptions, furthermore, appear to be unique to Christianity rather than to Abrahamic theism in general (ibid.); (4) one of the critical assumptions of the argument

190  Imran Aijaz is that relational-personal love is essential to theism, s­ omething that ­classical Islamic philosophers would reject in favor of alternative interpretations of God’s love, e.g., as another way of speaking about Divine Providence (McGinnis 2015, 160); (5) another critical assumption of the argument is that love is, in fact, a perfection (so that it is something one can reasonably expect from God, who is traditionally assumed to be perfect), but this assumption would also be questioned by some of the classical Islamic philosophers (McGinnis 2015, 161, 164–165); (6) one more critical assumption the argument rests on is the notion of God as a person, which many classical Islamic philosophers would criticize (McGinnis 2015, 169); and (7) because they reject the idea of God’s personhood, any account of God’s relation to us understood as a relation between persons, for the classical Islamic philosophers, evaporates (McGinnis 2015, 173). In his recent critical survey of the hiddenness argument, Schellenberg responds to McGinnis’ criticisms, placing them in the category of theological treatments of divine hiddenness, which, he thinks, apply methodological constraints to discussions of this topic (Schellenberg 2017b, 6). Regarding these treatments, Schellenberg raises two questions. First, what motivates the turn to theology? Second, does the turn to theology advance philosophical discussions of divine hiddenness? (ibid.). He then considers four possible ways of answering the first question: First, it may be felt that philosophy has taken over a notion—the hiddenness of God—that is originally theology’s, and that philosophy should pay more attention to theology if it wishes to see hiddenness issues aright. Call this a gatekeeping motivation. Second, people may be supposing that I or other philosophical hiddenness arguers are using some hiddenness-related idea on the assumption that a certain theological tradition accepts it and be questioning the accuracy of this assumption. Call this a fact-checking motivation. Third, it may be thought that some theological response to a related matter satisfactorily responds to a philosophical worry too. Call this the problem-solving motivation. Fourth, someone may think that theology opens up new intellectual vistas in connection with hiddenness themes that philosophy ought to recognize, perhaps holding as well that these render some philosophical hiddenness preoccupation an instance of misplaced attention or concern. Call this the consciousness-raising motivation. Of course, these motivations may often coexist and overlap in various ways. (Schellenberg 2017b, 6) Following this taxonomical account of theological treatments of divine ­ iddenness, Schellenberg proceeds to discuss examples in recent work where h he thinks these motivations are in play. This discussion is intended to assist in crafting an answer to his second question about whether the turn to theology advances philosophical considerations of divine hiddenness (ibid.). In

Islam and the Problem of Divine Hiddenness  191 reply to McGinnis’ assessment of his argument, an assessment he regards as belonging to the category of the ‘fact-checking motivation,’ Schellenberg offers the following remarks: McGinnis’s paper contains some interesting historical information. But it is odd that he thinks views about God must be lifted by a philosopher from some theistic tradition or other instead of, at least in part, constructed from resources more widely available and by independent means. This seems to me to fundamentally misunderstand philosophy, even where its peculiar concern happens to be religious. A philosopher can be thinking about, and hold to be uninstantiated, the idea of a personal God without supposing that what she has in mind has been endorsed by any theology and also without seeing herself as mounting her case against it because it has been thus endorsed. Theologians and their supporters too easily assume—and incorrectly assume, where my argument in particular is concerned—that a philosopher’s main aim is to attack theology. Having frequently had their views attacked by philosophers in the past few centuries, theologians … understandably have developed too narrow a sense of the possible motivations of philosophers, who may indeed—as in my case—be discussing an idea of God while recognizing that for reasons peculiar to its own tasks, theology would not see it the same way … I fully recognize that the deity I have discussed is not always theology’s deity; and so recognizing that what I have had to say about God does not always correspond to what theology has said is merely a small step toward understanding my view rather than the basis for a legitimate criticism of it. (Schellenberg 2017b, 7–8) There is, I will now argue, nothing in these remarks that constitutes an adequate response to McGinnis’ assessment of his hiddenness argument, or to the objections I’ve canvased in my own criticisms of it. As an initial observation, let me note that dismissing McGinnis’ discussion of how a classical Islamic philosopher would respond to the hiddenness argument as nothing more than ‘lifting views about God by a philosopher from some theistic tradition’ is about as charitable as dismissing Schellenberg’s discussion of divine hiddenness as nothing more than ‘lifting views about God by a philosopher from some non-theistic tradition’ (situated in secular, 21st-century Canada, perhaps). In his survey of Divine Love and of relations with the Divine as found in the classical Islamic philosophical tradition, McGinnis focuses on two important Islamic thinkers: al-Fārābī and Ibn Sīnā. Both of these thinkers firmly belong to the tradition of falsafa (philosophy), a discipline that, for centuries since medieval times, has been distinguished in the Islamic tradition from kalam (scholastic or philosophical theology).5 Although both al-Fārābī and Ibn Sīnā regarded themselves as Muslims, they nevertheless engaged with philosophy as philosophers, often in ways that

192  Imran Aijaz were seen as controversial and even blasphemous by some of their Muslim peers.6 As McGinnis notes, these thinkers subscribe to philosophical views about God for which they proffer philosophical arguments. For instance, both interpret the core Islamic doctrine of God’s oneness (Ar. tawhid) as an affirmation of the doctrine of divine simplicity. From this doctrine, one can infer various important conclusions that shed light on how the nature of God should be understood, e.g., the complete absence of any potentiality in God means that He cannot be in need of anything external to His essence (McGinnis 2015, 161). These observations are relevant to discussions of divine hiddenness. In his rejoinder to McGinnis, unfortunately, Schellenberg says nothing about such views and their accompanying arguments. Al-Fārābī's and Ibn Sīnā adherence to the doctrine of divine simplicity, along with their commitment to many other philosophical views about God that predate Islam as a historical religion, is a point we can use to reject as baseless Schellenberg’s suggestion that they did not construct their views about God from ‘resources more widely available.’ The doctrine of divine simplicity, to consider just one philosophical viewpoint about God upheld by both al-Fārābī and Ibn Sīnā, has roots in ancient Greek philosophy and also enjoys significant philosophical support in Judaism and Christianity (Leftow 1998). To be clear, my point here is not that one must automatically defer and acquiesce to philosophical views about God extant in a theistic tradition when considering the hiddenness argument. Rather, it is that these cannot be ignored in an evaluation of the argument’s soundness, particularly if it is intended, as Schellenberg’s argument seems plainly to be, as an argument for atheism, and thus as an argument for rejecting Judaism, Christianity, and Islam. Let’s think about Schellenberg’s alternative proposal for a moment. What does it mean to construct views about God by ‘independent means’? More specifically, here we may ask what precisely the means are independent of. Schellenberg doesn’t say, at least not in the critical survey in which he responds to McGinnis. If by ‘independent means’ he is referring to philosophical reflection not constrained by presuppositions of the familiar kind that animate theology and philosophical theology (e.g., the existence of God, the reality of God’s revelation to us, etc.), then classical Islamic philosophers like al-Fārābī and Ibn Sīnā would have no objection to this methodological requirement when it comes to philosophical reflections about the Divine.7 One cannot, therefore, simply bracket out the Islamic philosophical tradition and its views about God when gauging the force of the hiddenness argument. The fact that Islamic philosophers have a religious affiliation does not mean that their views about God lack a philosophical foundation; a Muslim philosopher may contribute to discussions about the nature of God qua philosopher. Consequently, there is nothing philosophically inappropriate in making an allowance for what a philosopher in a theistic tradition says about God when evaluating the hiddenness argument. Schellenberg’s contention to the contrary, according to which this constitutes a fundamental misunderstanding of philosophy, is simply incorrect.

Islam and the Problem of Divine Hiddenness  193 I do grant Schellenberg’s point that a philosopher can formulate a concept of a personal God, φ, and argue that φ is uninstantiated without supposing that φ is accepted or rejected in any theological tradition. As we’ve seen in the Islamic case, the Muslim theist can agree with Schellenberg that φ is indeed uninstantiated. Where he or she will part company with him is, of course, on claims about the alleged falsity of theism inferred from φ’s non-existence. By ‘atheism,’ Schellenberg isn’t merely stating that φ is uninstantiated (i.e., that personal-God atheism is true); he is also categorically asserting that God does not exist (i.e., that theism is false), as made clear by the final conditional premise of the hiddenness argument: 6*. If no personal, perfectly loving God exists, then God does not exist. This premise will be rejected by the Muslim theist, and he or she can appeal to philosophical considerations to justify its rejection, e.g., considerations found in the philosophical literature on Islamic classical theism. To further explain why Schellenberg needs to take these ­considerations seriously, let me draw attention to another kind of motivation absent from his taxonomy of theological treatments of divine hiddenness: a ­dialectical motivation. In his books and articles on the hiddenness argument, Schellenberg presents it as an argument that justifies atheism. It is offered as a piece of ‘natural atheology,’ which Alvin Plantinga defines as ‘the attempt, roughly, to show that, given what we know, it is impossible or unlikely that God exists’ (Plantinga 1967, vii). In presenting the hiddenness argument, Schellenberg nowhere offers it as something addressed only to individuals who share his non-theistic inclinations and views about God. This is understandable, since, after all, such a restriction would contravene his avowed philosophical approach to divine hiddenness. In working through the hiddenness argument, Schellenberg is not, it seems fair to say, adopting the kind of dialectically limited methodology that, say, a Christian analytic theologian will deploy in order to understand better and articulate his pre-existing convictions about a triune God. Rather, Schellenberg’s hiddenness argument is offered as a philosophical argument to a broad audience, one that includes theists, agnostics, etc. Take two possible members from this audience: a reflective Muslim theist, M, and a reflective agnostic, N, for whom atheism and Islamic theism both have some appeal and thereby make up a ‘living option,’ to use William James’ terminology (James 1896). M, because of her Islamic affiliation, takes seriously and believes in some of the philosophical accounts of God found in works of Islamic philosophy. She believes, let’s say, that God is not a person because that would entail several limitations on God, thereby violating God’s aseity. Suppose now that M is presented with Schellenberg’s hiddenness argument. Upon probing it, she discovers that it blossoms from reflections on the human self, which are extrapolated to the Divine. M reads one of many specimens found in Schellenberg’s writings explaining how these reflections proceed:

194  Imran Aijaz I’d say that what gives our justification prima facie status is basically that its reasoning instantiates an intuitively very plausible and widely accepted way of determining the meaning of divine attributes – namely, extrapolation from mundane examples of the relevant properties … [W]hen discussing the goodness of God, most philosophers assume that we are using an ordinary term in its usual sense. In discussion of the problem of evil, for example, it is widely agreed that a good God, like a good human being, might well have a reason for permitting evil to occur, while – again like a good human being – necessarily resisting what is seen as unjustified evil … [W]hat we know of how these terms are used in human contexts provides a prima facie good reason for believing that talk of God is to be understood in the same way. (Schellenberg 2002, 45–46)8 This method of reflecting on the Divine nature will be rejected by M because it clashes fundamentally with several of her antecedently held philosophical views about God. Unless Schellenberg can offer her compelling reasons to have his understanding of God displace hers, so that this ‘philosophical boom barrier’ may be lifted and the hiddenness argument’s march to atheism proper can proceed, M can reasonably reject the argument. Hence, in this case, the argument fails as a piece of natural atheology.9 To appreciate the dialectical motivation further, let’s turn to the case of N, a reflective agnostic vacillating between atheism and Islamic theism. N, let us stipulate, adopts a third-party perspective on the hiddenness argument, being aware of its advocacy by Schellenberg, its rejection by M, and the different philosophical assumptions and views about God held by each side. Why, N may reasonably ask, should he accept Schellenberg’s philosophical assumptions and views about God over M’s, particularly those that seem committed to an anthropomorphic idea of God? Of course, as part of his agnostic reservations regarding ‘God talk,’ N may also ask M why he should accept her philosophical assumptions and views about God, but that is not relevant here (we are not entertaining whether M is interested in arguing for the truth of Islamic theism). For N, too, Schellenberg’s hiddenness argument fails as a piece of natural atheology, unless reasons can be given to accept its philosophical starting point. Contrary to what Schellenberg claims, a proper understanding of the ­theological context in which a Muslim theist is placed and in which he or she is apprised of the hiddenness argument reveals critical philosophical assumptions that need to be engaged with by the proponent of the argument in order for its dialectical success. As a piece of natural atheology, the hiddenness argument is intended to show to a broad audience (one that includes theists, agnostics, and others) that God does not exist. Since this audience is not necessarily trying to argue that God does exist, the dialectical context of Schellenberg’s hiddenness argument is best described by what Douglas

Islam and the Problem of Divine Hiddenness  195 Walton calls ‘asymmetrical persuasion dialogue’. The more fundamental notion of ‘persuasion dialogue’ involves two participants and the attempt by each to persuade (prove to) the other the truth of a thesis, using rules of inference and concessions agreed upon by both (Walton 2008, 4–5). In the case of asymmetrical persuasion dialogue, however, the obligations of the two participants differ. Walton offers an example: Karl is a committed believer in God who is trying to convince Erik that God exists. Erik is not convinced by Karl’s arguments and raises many doubts. Erik is not an atheist, but calls himself an agnostic. (Walton 2008, 11) In this situation, Karl is attempting to prove a positive thesis to Erik (­God’s existence), but Erik is not attempting to prove a negative thesis to Karl (God’s non-existence). Because of this dialectic asymmetry, Walton explains, Karl has a positive burden of proof, the burden of proving his thesis, while Erik has only a negative burden of proof, the burden of raising questions about Karl’s proof and the acceptability of the arguments he provides (Walton 2008, 11). In a similar fashion, a proponent of Schellenberg’s hiddenness argument carries a positive burden of proof to establish the argument’s conclusion using rules of inference and concessions agreed upon by participants in the discussion. As far as the Muslim theist is concerned, I’ve noted that he or she simply will not provide the proponent of the hiddenness argument the concessions required for the argument to go through. By calling into question the argument’s presuppositions and views about God in ways I’ve covered in my discussion, the Muslim theist can satisfy the requirements for discharging his or her negative burden of proof. And this, in turn, creates an obstacle for the proponent of the hiddenness argument in discharging his or her positive burden of proof. Let me bring together and tie the threads of my discussion of Schellenberg’s hiddenness argument by presenting its proponent with a dilemma. Either (a) the hiddenness argument is a piece of natural atheology aimed at Muslim theists (though, of course, not exclusively aimed at them) or (b) it is not. If (a), then its proponent will need to engage with the philosophical presuppositions and views about God found in Islamic accounts of the Divine (which, as far as I can tell, has not been done; indeed, as we have seen, Schellenberg himself explicitly eschews engagement with McGinnis’ discussion of Divine Love in the Islamic philosophical tradition). If (b), then it may reasonably be ignored by the Muslim theist as something not aimed at his or her views about God (if a body of theology, T, is not relevant to a body of philosophy, P, as Schellenberg suggests in effectively dismissing the ‘turn to theology’ in philosophical discussions of divine hiddenness, then, naturally, P is not relevant to T either). Either way, it seems, Schellenberg’s hiddenness argument, at least as far as the debate surrounding it currently stands, does not succeed in undermining the rationality of Islamic belief.

196  Imran Aijaz 10.3

Some Concluding Reflections

By way of concluding reflections, I will briefly discuss how, despite my c­ riticisms of Schellenberg’s version of the problem of divine hiddenness, it is far from evident that other versions will prove to be just as problematic or unsuccessful in discussions of the rationality of Islamic belief. Let’s consider Schellenberg’s hiddenness argument again, but this time focusing only on its essential core. In an online presentation of the argument, Schellenberg articulates it much more simply than the version I’ve discussed in the preceding section. This simpler version, which I’ll call the ‘simple hiddenness argument’, goes as follows: 1 If God exists, there are no nonresistant nonbelievers. 2 There are nonresistant nonbelievers. 3 (Therefore) God does not exist (Schellenberg 2017c). Like the more detailed version we’ve seen, the simple hiddenness argument is valid. Notice, however, that it says nothing about a personal God or God’s Love understood as relational-personal love. Hence, many of the criticisms I’ve offered in response to the detailed version will not apply here. To motivate the simple hiddenness argument, a proponent can argue that premise 1 is a claim accepted by many Muslims and is endorsed by several thinkers in the Islamic tradition. These thinkers argue that premise 1 is true because there is an abundance of compelling evidence for God’s existence, so there are no nonresistant nonbelievers (i.e., people who inculpably lack the belief that God exists). Instead, all purported cases of ‘nonbelief’ are really cases of refusal to acknowledge the reality of God. Ayatullah Abdul Husayn Dastaghaib provides a succinct summary and endorsement of this position: The … excuse of one who does not know Allah and is unaware of Him, is not acceptable … Allah is not hidden from any sensible person … ­[W]hatever is present in the universe are all proofs of His Absolute Knowledge, Wisdom, Power and Command. Hence there is a Being, infinitely knowledgeable and Wise, Omniscient, Mighty, Omnipotent, and Powerful … People, who doubt, are in fact those who do not want to recognize the truth itself. If they were really seekers of truth, they would have looked at the creatures of the world and sought a lesson from it. If they had looked at the marvel of the wisdom and amazing power of the Creator of the Universe with proper attention, they would never have doubted. (Dastaghaib; emphasis mine) But surely, a proponent of the simple hiddenness argument can object, premise 2 is true. Aren’t there at least some nonresistant nonbelievers (e.g., people in parts of the world who’ve never even heard of ‘God’, reflective Buddhists, agnostics who are unpersuaded by arguments for and against God’s existence,

Islam and the Problem of Divine Hiddenness  197 etc.)? Since premise 2 is eminently plausible, the proponent will argue, the simple hiddenness argument is sound. We have a challenge to the rationality of Islamic belief. How might a Muslim theist reply? Many Muslim thinkers argue that premise 1 is indeed true because we all have knowledge of God. To support this claim, they hold that belief in God is a universal and innate natural disposition (Ar. fitrah), that God sent messengers to all nations calling people to worship God, that some of the philosophical arguments for God’s existence (e.g., cosmological and teleological arguments) are good arguments, and that the revelation of the Qur’an offers strong evidence that God exists. Based on all of this, these thinkers dismiss premise 2 of the simple hiddenness argument as false; there is no nonbelief but only refusal in any alleged case of non-theism (this is Dastaghaib’s view, as we saw); the modus tollens of the simple hiddenness argument is their modus ponens: 1 If God exists, there are no nonresistant nonbelievers. 2 God exists. 3 (Therefore) There are no nonresistant nonbelievers. Let me develop the force of the simple hiddenness argument a bit more to resist this modus ponens conversion. Although I cannot evaluate here the reasons given by Muslim thinkers to defend the claim that we all have knowledge of God, I will observe (and many will concur) that each of these reasons is controversial (see Aijaz 2014 for analysis and evaluation of these reasons). As an alternative way of responding to the simple hiddenness argument, one can grant premise 2 (‘There are nonresistant nonbelievers’) as plausible. I, for one, think that it is plausible (see Aijaz 2013 for a discussion and defense of it). What seems less plausible to me is the truth of premise 1 (‘If God exists, there are no nonresistant nonbelievers’). One might argue that its truth is entailed by Islamic doctrine and that it has been defended by Islamic thinkers. While I acknowledge that some Muslim thinkers have tried to defend premise 1, it is not at all clear to me that God’s existence entails the absence of nonresistant nonbelievers. At a philosophical level, one would need to provide an argument for the conclusion that, given a particular philosophical understanding of God, one should expect the complete absence of nonresistant nonbelief in our world. The prospects for such an argument succeeding are contentious (think of comparable debates about the problem of evil and what is philosophically ‘reasonable’ to expect from God). And as far as claims that the truth of premise 1 is entailed by Islamic doctrine are concerned, this too seems debatable to me. One easy way to show its questionable status, even for Muslims, is to point out how frequently the ‘What happens to those who’ve never heard of Islam?’ question is raised by Muslims as a serious question, and to see the answers given to it by Muslim theologians, which typically do not say ‘Everyone has heard of Islam!’. An example:

198  Imran Aijaz A person who has never heard of Islam or the Prophet [Muhammad] and who has never heard the message in its correct and true form, will not be punished by Allaah if he dies in a state of kufr (disbelief). If it were asked what his fate will be, the answer will be that Allaah will test him on the Day of Resurrection. (Islam Question & Answer 1998) What an answer like this indicates, even if naïve or wrong, is that there may be enough elasticity in Islamic theology to accommodate the existence of nonresistant nonbelief (in Aijaz 2007, I’ve explored some ways in which this might be done). If this is right, then a Muslim theist can begin to sketch out a defensible reply to the simple hiddenness argument that incorporates both philosophical and theological reflections. What a full-fledged Islamic response will be like remains to be seen, as Muslim discussions of divine hiddenness in the philosophy of religion are nascent. No doubt, there is much more to be said, especially given that debates surrounding divine hiddenness do not seem to be dissipating anytime soon.10 Notes 1 In his discussion of God’s personhood, Gary Legenhausen observes that ‘within the fold of Islam (at least among theologians), belief in a personal God is a minority position. The theological and philosophical groundwork for the Muslim claim that God is not a person is found in both Sunni and Shi’ite sources’ (Legenhausen 1986, 307). 2 Al-Kindi and many of the classical Islamic philosophers reject what David Bentley Hart insightfully terms ‘monopolytheism’, ‘a view of God not conspicuously different from the polytheistic picture of the gods as merely very powerful discrete entities who possess a variety of distinct attributes that lesser entities also possess, if in smaller measure; it differs from polytheism, as far as I can tell, solely in that it posits the existence of only one such being. It is a way of thinking that suggests that God, since he is only a particular instantiation of various concepts and properties, is logically dependent on some more comprehensive reality embracing both him and other beings’ (Hart 2013, 127–128). 3 Classical theists in general would not deny the importance and role of personalist language in conveying truths about God. Al-Fārābī, for instance, explains that to cognize the principles of beings (philosophical truths about reality) is to understand them as they really are and to have their essences imprinted in one’s soul. By contrast, to imagine these principles is to have images, representations, or imitations of them imprinted in one’s soul. Since, according to him, most human beings are incapable of intellectual cognition of the principles of beings, religion presents the alternative of imagination: ‘[R]eligion is but the impressions of these things or the impressions of their images, imprinted in the soul. Because it is difficult for the multitude to comprehend these things themselves as they are, the attempt was made to teach them these things in other ways, which are the ways of imitation’ (Al-Fārābī 1967, 40–41). 4 Schellenberg makes this equation explicit in his book The Hiddenness Argument: ‘My claim is that the idea of God, which … is in western philosophy the fairly specific and detailed idea of an all-powerful, all-knowing, all-good and all-loving creator of

Islam and the Problem of Divine Hiddenness  199 the universe, represents one way in which the broader idea of ultimism can be filled out, a way that utilizes the concept of a person. Theism gives to ultimism a face, one might say (quite a familiar one). Another way of putting it is to say that theism is an elaborated or personal version of ultimism. You could also turn the point into an equation: theism = personal ultimism’ (Schellenberg 2015, 20). 5 Ibn Khaldūn (d. 1406) defines kalam as ‘a science that involves arguing with ­logical proofs in defense of the articles of faith and refuting innovators who deviate in their dogmas from the early Muslims and Muslim orthodoxy’ (Ibn Khaldūn 1958, 34). As this definition of kalam makes clear, it is an apologetic enterprise and was seen as such by a number of classical Islamic philosophers. 6 Perhaps the best-known repudiation of their views comes from Al-Ghazālī, who criticized their philosophy on several points. On three of their philosophical positions, (1) the eternity of the world, (2) God’s lack of knowledge of particulars, and (3) denial of bodily resurrection, Al-Ghazālī judged them to be unbelievers (Al-Ghazālī 2000, 226). My point here is that one cannot dismiss al-Fārābī and Ibn Sīnā as dogmatic theologians or pseudo-philosophers who were unwavering in their commitment to whatever religious authority proclaimed to be ‘correct’ theology. 7 John Clayton provides further clarification: ‘Islamic “scholastic theologians” can be said to have held that firm faith leads naturally to rational reflection about the divine nature and existence as revealed in Qur’an; the ecclesiastically more independent-minded philosophers, on the other hand, can be said to have held that only those doctrines which can be rationally demonstrated are finally worthy of belief, even if this procedure sometimes leads to deviation from views clearly taught in Qur’an and in Muslim tradition. Al-Ghazālī and others used theistic proofs in order to confirm traditional doctrine; Ibn Sīnā and Ibn Rushd, however, used them at least in part to correct or to reform traditional doctrine’ (Clayton 2006, 91; emphasis mine). 8 Many similar examples from Schellenberg’s writings can be cited. Here are two more: ‘In examining [the] concept [of divine love], [we must] develop our understanding of it …by reference to what is best in human love’ (Schellenberg 2020, 174); and, ‘[R]eflection on the concept of divine love shows that a perfectly loving God would necessarily seek personal relationship with all individuals belonging to [the category of non-resisting nonbelievers] … In defense of … [this claim], we may point out that the seeking of personal relationship is an essential part of the best human love … Something similar must apply to God’s love for us’ (ibid.). 9 Here, I think it’s essential to discuss the criteria Schellenberg’s hiddenness argument must satisfy in order to count as a successful piece of natural atheology. Comparisons with the criteria for a successful piece of natural theology will be helpful. The goal of a theistic proof, as Stephen T. Davis explains, is to ‘substantiate the theist’s belief in God, give a good reason for it, show that it is credible, show that it is true … [T]heistic proofs … try to demonstrate the rationality of theistic belief to all rational persons (whoever exactly they are)’ (Davis 1997, 6; emphasis mine). Given this aim, and in addition to the conditions of formal and informal validity, there is a further condition that a theistic proof must meet for success: It is crucial that the premises of a successful theistic proof be known to be more plausible than their denials. But known to whom? They must be knowable by any rational person; and ideally they must be known by the people to whom the rationality of belief in the existence of God is to be demonstrated (including, presumably, the person who offers the proof). If the premises of a theistic proof are more plausible than their denials but the

200  Imran Aijaz relevant people do not know that fact, the rationality of theism will not be demonstrated to them.  (Davis 1997, 7; emphasis mine) It is often with respect to this condition, what we can call the ‘premise-­ acceptability’ condition, that many critics of natural theology declare the failure of arguments for the existence of God. J.L. Mackie, for instance, objects to the Leibnizian version of the Cosmological Argument, which relies on the Principle of Sufficient Reason (PSR), by stating that he just does not find the PSR rationally compelling: ‘The principle of sufficient reason expresses a demand that things should be intelligible through and through. The simple reply to the argument which relies on it is that there is nothing that justifies this demand, and nothing that supports the belief that it is satisfiable even in principle … The form of the cosmological ­argument which relies on the principle of sufficient reason therefore fails completely as a demonstrative proof’ (Mackie 1982, 85–87). Unless we are to engage in special pleading for atheism and arguments of natural atheology, it seems to me that the premise-acceptability condition must be enforced in exactly a symmetrical manner whenever assessing the success of a natural atheological argument, including Schellenberg’s hiddenness argument for atheism. 10 I am grateful to John Bishop, Oliver Leaman, and Mohammad Saleh Zarepour for kindly reading through an earlier draft of this chapter and providing helpful comments.

References Aijaz, Imran (2007), ‘Belief, Providence and Eschatology: Some Philosophical Problems in Islamic Theism’, Philosophy Compass, https://doi.org/10.1111/j.1747-9991.2007. 00119.x Aijaz, Imran (2013), ‘Some Ruminations about Inculpable Non-Belief’, Religious Studies, Vol. 49, No. 3, 399–419. Aijaz, Imran (2014), ‘Traditional Islamic Exclusivism – A Critique’, European Journal for Philosophy of Religion, Vol. 6, No. 2, https://doi.org/10.24204/ejpr.v6i2.186 Al-Fārābī (1967), The Political Regime, trans. Fauzi Najjar in ‘Alfarabi: The Political Regime’, in Ralph Lerner and Muhsin Mahdi (Eds.), Medieval Political Philosophy: A Sourcebook, (New York: The Free Press), 31–57. Al-Ghazālī (1992), The Ninety-Nine Names of God, trans. David Burrell and Daher Nazir, Al-Ghazali on the Ninety-Nine Beautiful Names of God (Cambridge: The Islamic Texts Society). Al-Ghazālī (2000), The Incoherence of the Philosophers, trans. Michael E. Marmura in The Incoherence of The Philosophers (Provo, UT: Brigham Young University Press). Al-Ghazālī (2014), The Revitalization of the Religious Disciplines, trans. W. Montgomery Watt in John Renard, Islamic Theological Themes (Oakland, CA: University of California Press). Al-Kindi (1974), On First Philosophy, trans. Alfred Ivry in Al-Kindi’s Metaphysics (Albany, NY: State University of New York Press). Clayton, John (2006), Religions, Reasons and Gods: Essays in Cross-Cultural Philosophy of Religion (New York: Cambridge University Press). Dastaghaib, Ayatollah Abdul Husayn (n.d.), Immaculate Conscience, https://www. al-islam.org/qalbe-saleem-immaculate-conscience-sayyid-abdul-husayn-dastghaibshirazi/fourth-disease-doubt

Islam and the Problem of Divine Hiddenness  201 Davies, Brian (2004), An Introduction to the Philosophy of Religion, 3rd ed. (Oxford: Oxford University Press). Davis, Stephen T. (1997), God, Reason and Theistic Proofs (Edinburgh: Edinburgh University Press). Drange, Theodore M. (1998), Nonbelief & Evil: Two Arguments for the Nonexistence of God, (Amherst, NY: Prometheus Books). Eagleton, Terry (2006), ‘Lunging, Flailing, Mispunching’, London Review of Books, https://www.lrb.co.uk/the-paper/v28/n20/terry-eagleton/lunging-flailingmispunching Hart, David Bentley (2013), The Experience of God: Being, Consciousness, Bliss (New Haven, CT: Yale University Press). Ibn Khaldūn (1958), Muqaddimah, trans. Franz Rosenthal in The Muqaddimah, Vol. 3, 2nd ed. (Princeton, NJ: Princeton University Press). Ibn Sīnā (1945), Treatise on Love, trans. Emil Fackenheim in ‘A Treatise on Love by Ibn Sina’, Medieval Studies, https://doi.org/10.1484/J.MS.2.305873 Ibn Sīnā (2005), Book of Healing, ed. and trans. Michael E. Marmura in The Metaphysics of the Healing (Provo, UT: Brigham Young University Press). Islam Question & Answer (1998), ‘The Fate of Kuffaar Who Did Not Hear the Message of Islam’, https://islamqa.info/en/answers/1244/the-fate-of-kuffaar-whodid-not-hear-the-message-of-islam James, William (1896), ‘The Will to Believe’, https://www.gutenberg.org/files/26659/ 26659-h/26659-h.htm Kray, Klaas (2013), ‘The Problem of Divine Hiddenness’, Oxford Bibliographies Online, https://doi.org/10.1093/obo/9780195396577-0178 Leftow, Brian (1998), ‘Classical Theism. God, Concepts of’, Routledge Encyclopedia of Philosophy (Taylor and Francis), doi:10.4324/9780415249126-K030-1, https://www.rep.routledge.com/articles/thematic/god-concepts-of/v-1/sections/ classical-theism Legenhausen, Gary (1986), ‘Is God a Person?’, Religious Studies, Vol. 22, No. 3/4, 307–323. Mackie, J.L. (1982), The Miracle of Theism (Oxford: Oxford University Press). McGinnis, Jon (2015), ‘The Hiddenness of “Divine Hiddenness”: Divine Love in Medieval Islamic Lands’, in Adam Green and Eleonore Stump (Eds.), Hidden Divinity and Religious Belief: New Perspectives (Cambridge: Cambridge University Press), 157–174. Muhammad, Ghazi bin (2010), Love in the Holy Quran (Chicago, IL: Kazi Publications). Plantinga, Alvin (1967), God and Other Minds: A Study of the Rational Justification of Belief in God (Ithaca, NJ: Cornell University Press). Ryle, Gilbert (1949), The Concept of Mind (London: Hutchinson University Library). Schellenberg, J.L. (1993), Divine Hiddenness and Human Reason (Ithaca, NJ: Cornell University Press). Schellenberg, J.L. (2002), ‘What the Hiddenness of God Reveals: A Collaborative Discussion’, in D. Howard Snyder & P. K. Moser (Eds.), Divine Hiddenness: New Essays, (Cambridge University Press), 33–61. Schellenberg, J.L. (2015), The Hiddenness Argument: Philosophy’s New Challenge to Belief in God (New York: Oxford University Press). Schellenberg, J.L. (2017a), ‘Divine Hiddenness: Part 1 (Recent Work on the Hiddenness Argument)’, Philosophy Compass, https://doi.org/10.1111/phc3.12355

202  Imran Aijaz Schellenberg, J.L. (2017b), ‘Divine Hiddenness: Part 2 (Recent Enlargements of the Discussion)’, Philosophy Compass, https://doi.org/10.1111/phc3.12413 Schellenberg, J.L. (2017c), ‘John Schellenberg: The Hiddenness Argument and the Contribution of Philosophy (1/5)’, YouTube, uploaded by Philosophy at the University of Edinburgh, https://youtu.be/7hBpZ-ystOg Schellenberg, J.L. (2020), ‘Does Divine Hiddenness Justify Atheism?’, Michael L. Peterson and Raymond J. VanArragon (Eds.), Contemporary Debates in Philosophy of Religion, 2nd ed. (Wiley-Blackwell), 30–41. Shakespeare, William (1593), ‘Venus and Adonis’, http://www.shakespeare-w.com/ english/shakespeare/w_venus.html Walton, Douglas (2008), Informal Logic: A Pragmatic Approach, 2nd ed. (New York: Cambridge University Press).

11 An Analysis of Some Problems of Evil from an Islamic Perspective Hajj Muhammad Legenhausen

11.1 Introduction There are a number of distinct problems that fall under the heading of the problem of evil: the logical problem of evil, the evidential problem of evil, the problem of horrendous evils, the problem of natural evils, the problem of the suffering of the innocent, etc. A common kind of response that theologians give to these problems takes the form of a theodicy. Theodicies are attempts to justify the belief that God is perfectly good given that He is omnipotent and created this world with all the evils it contains. Theodicies are arguments to the effect that given the value of some feature of the world, and the fact that this feature cannot be obtained without incurring at least as much evil, the value feature provides God with an excuse for the amount of evil there is. In other words, according to theodicies, there are no gratuitous evils, in the sense that for any actual evil, e, had God created a world without e, the total amount of evil in the world would have been greater or equal to the total amount of evil in the actual world. This is often formulated by saying that the actual world is the best of all possible worlds (although additional complications occur in libertarian free will theodicies, which allow possible worlds better than the actual one in which agents freely abstain from sin); but if there need not be a uniquely best world, it would suffice to say that the actual world is no worse than any other possible world. According to Leibniz, there must be a uniquely best possible world; for otherwise God would have no reason to create one among several worlds tied for first place, and this idea is also endorsed by most Muslim theologians. Although most Muslim philosophers and theologians proposed the thesis that the actual world is the best of all possible worlds, they offered different explanations for the value feature of the actual world that makes it the best.1 Ash‘arites held that what God wills is good because God wills it. So, an Ash‘arite might say that the value feature of the actual world that makes it the best is simply that God chose to create it. The main difficulty with this view is that if the divine choice is arbitrary, which would follow if the choice were not grounded in the value of what is chosen independent of

DOI: 10.4324/9781003327714-17

204  Hajj Muhammad Legenhausen God’s choice, it is difficult to understand how it could be considered a manifestation of divine wisdom. Another kind of explanation for the features of the actual world is given by Avicenna. God’s unique action is the divine emanation and everything else is a necessary consequence of this. God does not create in order to fulfill any need. Creation is interpreted as emanation; and this is a necessary consequence of God’s own perfection. This is the best of all possible worlds because no other worlds are really possible. In this, Avicenna follows Plotinus. As Shams Inati explains: Because there is nothing, good or evil, that is better than what God possesses, God cannot intend anything for Himself or for the world. It is true that God is said to cause the world, but He is said to do so by necessity. God cannot therefore be blamed for the presence of evil in the world.2 In what follows, I will defend an unorthodox view that is neither ­determinist nor Ash‘arite. It might be considered an attempt to reconcile tendencies within the dominant positions that have been taken by Ash‘arites and their opponents. The strategy is to grant that the mere counterfactual possibility that God chose to create world w is not sufficient to justify the claim that the divine choice would have provided w with the value feature that would compensate for any evil in w. At the same time, it will be argued that there are some cases in which the choice of w could be morally permissible even if—aside from the divine choice—w does not possess any value feature that would make it better than any other possible world. The proposal that will be defended is one that denies that God’s moral perfection requires Him to act in a manner that will always be in conformity with human moral standards. God’s beneficence does not require Him to refrain from creating w if there is some w′ such that w′ contains less evil than w. Furthermore, it will be argued that divine beneficence does not require that there be no gratuitous evil, evil that could have been avoided without loss of some other value had God created the world differently. Many years ago, I argued the same basic point on the basis of the ­distinction in John Rawls’ between ‘obligations’ and ‘moral laws’.3 Obligations, in Rawls’ sense apply to God insofar as they are instituted by agreements or covenants between God and humans. Moral laws, as Rawls understands them, are based on human nature, in particular, on what humans would agree to in order to realize their interests. Rawls describes moral laws as: ‘principles that free and rational persons concerned to further their own interests would accept in an initial position of equality as defining the fundamental terms of their association’.4 Since God neither pursues interests nor does He enter into association with mortals on a basis of equality with them, moral laws will not necessarily apply to Him. Here, this argument is extended to show that important aspects of human morality do not apply to God given other accounts of morality, and that, as a result, the evil that exists in the world is

Analysis of Some Problems of Evil  205 not evidence that God is unjust or morally imperfect. The root of the many problems of evil is diagnosed as an anthropomorphism that imposes moral responsibilities on God as if He were a person who could be held accountable for shoddy workmanship in a creation that has resulted in unfathomable misery. Theodicies are generally designed to show that there is something that could make all the misery worthwhile and thereby let God off the hook. Although theodicy has been popular with many Muslim philosophers and theologians, one can find in the tradition of Islamic philosophy the resources needed to deny that God has a moral obligation to create a world with a minimal amount of gratuitous evil. 11.2

The God of Islamic Philosophy

The conception of God in Islamic philosophy is paradigmatically one to which Anthony Kenny used the term ‘the God of the philosophers’.5 Kenny’s term was used derisively. Kenny and those who oppose ‘the philosophers’ argue that the concept of God that the philosophers constructed—a ­timeless, omnipotent, omniscient, and perfectly good being—is so far from the ­personal God of lived religious faith and scripture that it should be rejected as a ­non-existent entity distinct by nature from the true God.6 Furthermore, Kenny argued, the philosophical construction of divinity was logically flawed. Paradoxes abound at every turn among the divine attributes as understood by the philosophers. The rejection of the God of the philosophers has become very widespread, especially among theologians, many of whom see the task of modern theology as one of routing out the vestiges of Neoplatonism. The emerging orthodoxy is one that takes God to be a person, one who is eternal, but not timeless, and one whose knowledge might not extend to some future contingent propositions, especially if their truth is dependent on the free choices of human persons. While this new emerging orthodoxy is particularly dominant in contemporary Christian theology, there are indications that it is gaining ground among Muslim thinkers, as well. Let’s call theological projects that emphasize the personhood of God as a temporal being personalist theologies. The kinds of theologies that developed in the context of classical Islamic philosophy, from al-Kindi (c. 800–870) through the philosophers of the Safavid period and those who continue in this tradition until the present, have been distinctively non-personalist. In his introduction to Rethinking the Concept of a Personal God, Thomas Schärtl writes: …[W]e we can view the difference between traditionally Christian and traditionally Muslim concepts of God as [an] embodiment of rivalling theistic options within a monotheistic framework [i.e., personalist and nonpersonalist], although we would have to add that within the history of each religion the opposite option always remained present in one way or the other – despite certain tendencies to marginalize its legacy or impact.7

206  Hajj Muhammad Legenhausen The roots of the divergence between these two theistic options has roots in the ways in which Christian and Muslim theologians drew upon the legacy of Neoplatonism. The Fathers of the Church did not wholeheartedly take over the hypostases of the Neoplatonists: The One, the Intellect, and the Spirit or world-soul. The Neoplatonists themselves also developed differing views about how these hypostases were to be understood. For example, according to Plotinus (c. 204/205–c. 270), the One is beyond substance (ousia), while in the later writings of Porphyry (c. 234–c. 305), the One appears to be a threefold substance. There were also differences among the Neoplatonists and among the Church fathers about how to understand the relations among the hypostases and about the function of the world-soul or Spirit. Christian and Neoplatonic speculations were developed in parallel; and several Church fathers explicitly compared the views of some Neoplatonists and drew attention to parallels to the doctrine of the Trinity. Clement of Alexandria (c. 150–c. 215) regards God as a thinking ­intelligent being, contrary to Plotinus, for whom the One is superior to any intellect or thing possessing intelligence, and is even superior to being. Origen of Alexandria (c. 185–c. 253), like Clement, was guilty of what the Church would later call ‘subordinationism’, the view that the Son is subordinate to the Father and that the Holy Spirit is subordinate to both the Father and the Son. Origen also held that God the Father is a self-thinking intelligence, a formulation found in Aristotle, although in some of his writings he equivocates, and merely claims that God is either an intellect or is above intellect. Plotinus held that the administration and maintenance of the order of the world was the responsibility of the world-soul or Spirit, while for Origen, these tasks were allotted to the Logos, the Son. The intellect, or nous, is divided into two levels, transcendent and immanent, corresponding to the Son and the Holy Spirit. A Trinitarian view without the taint of subordinationism was only espoused in a definitive manner by the Cappadocian Fathers of the 4th century. Like Clement and Origen, the Cappadocians considered the Father to be a thinking substance, contrary to the One of Plotinus.8 The rejection of subordinationism required the first three hypostases to be given equal standing as divine persons, despite the fact that the worldsoul was considered to be temporal by the Neoplatonists, in contrast to the timelessness of the One and the Intellect; and Basil (330–379) argued against those who had lowered the Spirit to the status of a servant of God that the Spirit was fully divine.9 The problematic status of the Spirit and the breadth of views current in the early church is illustrated by the position taken by the first Syriac Church Father, Aphrahat (c. 280–c. 345), who suggested that the Spirit and the Logos are one, and to be identified with Jesus.10 Aphrahat appears to be unaware of the Council of Nicaea (325) and to have what some have called a binitarian instead of a trinitarian theology.11 Without further examination of the development of competing accounts of the Trinity, the point to take away for our purposes is that there were

Analysis of Some Problems of Evil  207 differing views among the Neoplatonists regarding whether the One ­transcends intellect (nous); and that in the development of Christian thought, God was understood in such a manner as to be a thinking being, in contrast to the One of Plotinus. In Islamic philosophy, to the contrary, despite some inconsistencies, God was considered to above anything that could be or possess cognitive faculties. Just as God can be said to hear without having ears, He can know without need for cognitive faculties. The influence of Plotinus on al-Kindi is described by Peter Adamson: [al-Kindi] argues that neither soul nor intellect is ‘‘the true One,’’ because thinking involves a motion from thought to thought, and intellect grasps a multiplicity of universals…. Thus he follows Plotinus’ rejection of Aristotle’s understanding of God as a pure intellect, on the basis that intellect falls short of absolute simplicity.12 If this means that intellect is clearly separate from and subordinate to God, so that the resulting conception of God is decidedly non-anthropomorphic, ­al-Kindi’s view of the celestial spheres, on the other hand, is anthropomorphic, and he argues that they are alive and rational. Although the history of the treatments of Neoplatonic themes through the development of Islamic philosophy is fascinating, as we continue through Farabi, the Ikhwan al-Safa, Avicenna, and beyond, there is never any restoration of the intellect to the divine position it held in Porphyry and the Church Fathers. Farabi13 and Avicenna14 both attribute intellect to God, but only in the sense in which whatever is free from matter can be called ‘intellect’ and not in the sense of having the power to engage in processes of reasoning.15 At the same time, the Muslim theologians’ view of God differed from the One of Plotinus in having the divine attributes of life, knowledge, wisdom, and will, although these were sometimes interpreted in an apophatic theology. Even when such attributes were given a positive interpretation, intellect was not included among them, and the intellect remained the first creature created by God, and not a hypostasis. For the Muslim philosophers, there follows a series of intellects of which the tenth, in Farabi’s system, is the Active or Agent Intellect, which governs the sublunary world, and corresponds in some ways to the Neoplatonist world-soul and the Christian Holy Spirit. All the intellects higher than the tenth are, like God, atemporal. It is the Active Intellect that straddles the atemporal and temporal realms. 11.3

Divine Agency and Responsibility

The above excursion into the relation between Neoplatonic, Christian, and Muslim theologies shows that there is a fundamental difference between some strands of Christian and Islamic philosophical theology. If God ­contains intellect, divine agency will be understood differently than if God transcends intellect. If God possesses an intellect of His own analogous to the manner

208  Hajj Muhammad Legenhausen in which humans can be said to have the power of intellection, then it will make sense to ask what reasons God might have for his actions, what He intended, and His aims and purposes. If, on the other hand, God is simple, and does not have mental faculties, it will be difficult to explain in what sense agency can be attributed to God at all. This problem was addressed over the course of the development of the reflections of the Muslim philosophers on the nature of the divine will and agency. The main contours of this development are described by ‘Allamah Misbah Yazdi as follows.16 First, any entity that causes some motion or change, was considered to be an agent (fā‘il). The agents were then divided into those whose actions are willful and those agents that bring about changes without willing them. The willful agents were then divided into those that bring about an effect intentionally and the higher agents that require no intention, but act providentially. Divine actions were designated as providential (bil-‘ināyah) rather than intentional (bil-qaṣd). Intentional actions were considered to be those for which the agent required some sort of enticement to act, a reason for acting, that could serve as motivation for the action. God and the ­intellects were considered to be so perfect that nothing could serve as an enticement for them to act, as a motivating reason. They could not act intentionally because the notion of an intention harbors an implicit need or deficiency, a want, that the action would be performed to alleviate. The Illuminationists introduced another kind of agency, agency by agreement (fā‘iliyyah bil-riḍā), which is said to occur when an agent brings to mind something that was implicitly known but of which the agent was not actively conscious. This bringing to awareness would be a mental activity by virtue of which what is brought to mind is in agreement with what was previously known; and this was used as a model for divine agency. God acts by bringing to actuality what was already in agreement with his implicit knowledge. Finally, Mulla Sadra developed the idea of agency by self-­disclosure (fā‘iliyyah bil-tajallī), which he adopted from Ibn al-ʿArabi. God acts through the outpouring of the divine essence whereby the spectrum of the variable intensities of existence appears that are attributed to the divine manifestations, that is, to everything in creation. The relevance of all of this for the problem of evil is due to the manner in which divine accountability is to be understood. For human agents, who are intentional agents, actions are performed in order to achieve ends or p ­ urposes. If an intentional action results in harm, the agent will be b ­ lameworthy unless the agent has an excuse for the harm produced. The excuse can be either ­ignorance or compensatory results. A human being could plead that there was no way the resulting harm could have been foreseen; but this kind of excuse cannot be used to justify divine actions, given divine omniscience. Theodicies generally attempt to justify the evil that results from divine actions by compensatory results. The harm that resulted was collateral damage justified by the greater good of human freedom, the physical order of the universe, soulmaking, or some supposed unknown factor that makes it all worthwhile. According to theodicies, then, there are reasons in the mind of God that

Analysis of Some Problems of Evil  209 justify the creation of the world with all the evils it contains. God’s act of creation is performed in full knowledge of the evil that will result, but the divine reasons for going ahead with the creation justify God’s action despite the resultant evils. Much of the literature in the philosophy of religion on ­ roposed the problem of evil consists of debates about whether or not some p excuse is sufficient to exculpate God.17 If, however, God does not have a mind or intellect,18 and is not a p ­ erson who operates on the basis of reasons,19 the moral accountability of God for the evil in the world would be misplaced. God is not responsible in the sense of being someone whose reasons or intentions might justify his actions, because God has no intentions. He is not an intentional agent, but is, rather, an agent by self-disclosure. The problem of evil resurfaces in Islamic philosophical theology, even among those with the strongest affinities to Neoplatonism, as the problem of how to justify not God’s actions but our attributions (and scriptural attributions) to God of justice and wisdom. Thus, Shahīd Mutahhari’s major work on the problem of evil is entitled Divine Justice.20 God is not wise because He is sensitive to relevant reasons, for this would imply a kind of passibility inconsistent with divine simplicity, atemporality, and perfection. He is wise because the world can be understood as a manifestation of His wisdom, that is, the world displays a wondrous order that in part is reflected in the discoveries of the natural sciences. If the world were significantly more chaotic than it is, we would not be able to understand the world in this way, and the attribution of wisdom would be unjustified. The problem of evil may be reformulated as a threat to divine justice because the evil in the world might be alleged to be inconsistent with understanding the world as a manifestation of divine justice. If, however, the injustice of an agent can only be justified by attributions of culpable negligence or malevolence, and if these are due to defects in character, such as greed and selfishness, then God could be considered just in the sense of being without the flaws that would warrant attributions of injustice to God. This is expressed by the Shi‘i theologian and traditionalist Shaykh Saduq (918/919–991) in a comment on a narration about the loss of a loved one: The reason that the actions of Allah, the Blessed and Exalted, are wise, is that they are far from contradiction, they are free from defect, some depending on others, the need of a thing for its like, having specified forms, and the uniting of each with its own kind. If one even would imagine it otherwise, regarding the revolutions of the celestial spheres, the movements of the sun and moon, the courses of the stars, there would be defect and corruption. Since the actions of Allah, the Mighty and the Merciful, satisfy the conditions of justice, as we said, and they are sound with regard to what was mentioned previously, free from causes of injustice, it is correct that they are done wisely. The reason why injustice is not realized by Allah, the Mighty and Magnificent, and He does not do it, is that it has been proven that Allah, the Blessed and

210  Hajj Muhammad Legenhausen Exalted, is eternal (qadīm), needless (ghanī), most knowing without ignorance, while injustice is not done except from ignorance of its evil or because of a need to do it to profit by it.21 Divine agency is not like human agency, for it is not intentional; and this means that certain kinds of accusations against God will be misplaced. If the cause of great suffering is not an intentional being, such as a volcano whose eruptions cause tremendous harm or a virus that kills millions and leaves many more debilitated, we do not blame the causes or consider them to be cruel. These non-personal causes of harm are not subject to moral blame because they are less than human. I suggest, with some qualifications, that non-personal agents who are superior to human persons, such as God and His angels, are also not to be subject to moral blame if there is some sense in which they can be said to be the causes of suffering.22 It will be objected that this sort of maneuver might free God from moral blame, but it would also make divine praiseworthiness meaningless. If causing harm is no reason to consider God unjust, one might suspect that there is likewise no reason to think God is just. The answer this objection has two parts. First, God may be considered just in the sense that there are no ­factors that could motivate God to act unjustly. Due to the divine perfections, God is free from the selfish desires that cause injustice. Second, God is just because He fulfills the obligations that He places upon Himself, to which we now turn. 11.4

Obligations for Allah

Although the consensus among philosophically inclined Muslim theologians was that this is the best of all possible worlds, the argument was usually presented in the context of divine wisdom rather than that of divine justice. Divine wisdom could only be affirmed if everything in creation occurred because of something that would make it an appropriate choice. It was taken to be self-evident that ‘there can be no preponderance without a preponderant’.23 God could not choose to make the world as it is unless there were something to make it preferable to any alternative. The entire creation is seen as God’s choice, and no detail of that choice could be left to chance without undermining divine wisdom or power. Various compatibilist strategies were then invoked to allow for human and divine freedom of will. God’s moral goodness is not to be safeguarded by arguments that this is the best world that even an omnipotent divinity could create, but by divine obligations (and supererogations). God is good because all benefits are from him, because he has guided humanity toward a good life through his prophets, and because he perfectly fulfills his obligations and is forgiving beyond obligation. The question of divine obligations was a contested one among Muslim theologians. Here, I will only briefly consider an account by the 17th-century Safavid theologian, ‘Abd al-Razzāq Fayyāḍ Lāhījī, who was a

Analysis of Some Problems of Evil  211 student of and son-in-law of Mulla Sadra, before turning to some suggestions by ‘Allamah Tabataba’i. Fayyāḍ considers those who oppose the idea that anything is obligatory for God. These theologians offer two arguments why God cannot have any moral obligations. First, it is argued that if God were subject to the moral law, His sovereignty would be undermined. The law would be higher in authority than the divine will, and this is impossible. Furthermore, the moral law, if it can be known independent of divine commands, is allegedly known by reason. The theologians opposed to divine obligations then argue that if God had obligations, he would be subject to a law determined by human reason; and this would be a blasphemous denial of divine sovereignty. Second, it is argued that if God had obligations, then if he violated them, it would be proper to hold God morally culpable. However, it could never be proper to blame God for anything. It follows that God has no obligations. Fayyāḍ refutes both of these arguments. No great exertion is required to see what is wrong with the second argument. Even if it is impossible for God to do anything blameworthy, it will remain intelligible to consider conditions under which He would be blameworthy. This is because whenever we apply a concept that has an opposite to something, the attribution of both the concept and its opposite will be meaningful. Since it is agreed that God is praiseworthy, and praiseworthiness is the opposite of blameworthiness, the claim that God might do something blameworthy will be intelligible. Although it might be sinful to actually blame God for anything, it is not sinful to imagine that had God done something wrong, He would have been culpable. Divine culpability is intelligible, even if there is a sense in which it is inconceivable that God would ever do anything blameworthy. As for the first argument, Fayyāḍ argues that when we say that God is subject to the moral law, what we mean is that God would never perform actions which He would blame and punish were they done by a human being. It is not that God must respect the moral law because of the law’s authority over Him, but that God simply cannot perform actions that we would blame when performed by humans, because God has none of the human flaws that result in their blameworthy actions. Fayyāḍ gives the example of God’s ­prohibition of oppression for Himself: For example, since God the Exalted forbids oppression, because of its evil, and He blames and punishes its agent, then surely it will be inadmissible and inconceivable that He might engage in oppression Himself. This is the meaning of the obligation to abstain from oppression for God the Exalted.24 I will contest this argument; but before doing so, we should turn to a positive argument that Fayyāḍ gives to show that God is indeed subject to obligations. The argument is an example of perfect being theology. God has all perfections. We consider it to be a perfection if one abides by one’s ­obligations.

212  Hajj Muhammad Legenhausen Hence, we can attribute this perfection to God, as well, and claim that He fulfills His obligations. It follows that He must have obligations. One could object here that fulfilling obligations is only a perfection in creatures who have obligations. If God had no obligations, there would be no imperfection in Him due to not having obligations. Likewise, it might be a perfection in humans to successfully resist temptations, but this does not mean that divine perfection requires God to have temptations and successfully resist them. The problem with the argument Fayyāḍ gives to explain God’s moral obligations as obligations to refrain from what would be blameworthy in humans is that much of what is generally accepted by believers as divine actions would surely be blameworthy if intentionally performed by humans. God sends the angel of death to take the souls of those whose time it is to die. If a human sovereign were to have such an arrangement, it would be considered murder. Indeed, the power of the problem of evil as an argument against the existence of a morally good divinity lies precisely in the fact that it is so difficult to find exculpatory excuses for divine actions when divinity is interpreted anthropomorphically. If a human were to cause the evils (for terrestrial creatures) of natural disasters, the human would be blameworthy; yet God does not refrain from allowing such things to take place. Muslim theologians who oppose the idea that there are obligations for God (wujūb ‘alā Allah) often support their position with the following āyah of the Qur’ān: ‘He will not be questioned about what He does; but they will be questioned’ (21:23).25 On this basis, one might attempt to prevent the problem of evil from being posed, for raising the problem of evil amounts to questioning what God does. If God is literally not responsible for what He does, because no response from Him can be demanded, and if obligation implies responsibility, then God would have no obligations. Since there would then be no divine obligations for God to violate, the evil in the world would not be evidence that God did anything wrong. This might explain why God is not blameworthy, but it only does so at the expense of threatening divine praiseworthiness. Another response to the question of why God is not to be questioned is that all of creation is God’s property in such an absolute manner that He can do whatever He wants with it without warranting any moral criticism. God is free from all obligations of proper treatment of His creatures, according to this view. For the most part, however, the exegetes of the Qur’ān have not i­ nterpreted (21:23) as a denial that God has any obligations. ‘Allamah Tabataba’i reports that some argue that the statement that God will not be questioned only pertains to divine wisdom. Since God is perfectly wise and all creation is a manifestation of divine wisdom, all things, even the things we consider to be evils, occur as elements in a system of causes and effects that display an order that might not be possible if the evils had not occurred. The divine order of the universe and its system of causes and effects function as the ­exculpatory excuses for the divine allowance of evil. God will not be ­questioned, ­according

Analysis of Some Problems of Evil  213 to this interpretation, means that it will be recognized that God is perfectly wise, and so must have His reasons for whatever is in creation. So, it would be foolish to question God about the evil in the world. Evil has no specifically moral justification, since it is simply a natural result of the exigencies of creation. ‘Allamah Tabataba’i and other exegetes have pointed out that (21:23) is not restricted in any way to a particular type of questioning concerning the wisdom of creation. They claim that the āyah absolutely makes any questioning of God inappropriate. Any sort of questioning of God is inappropriate because God has absolute dominion over His creation. Considerations along these lines are developed by ‘Allamah Tabataba’i in his exegesis of (21:23).26 He argues on textual grounds that the unquestionability of God’s actions is not restricted to the issue of divine wisdom. To question a human action with respect to its wisdom, a reason is sought from the agent. The intention for the performance of the action is given by the agent with reference to the reason, e.g., by performing the action some desire is fulfilled. However, with God, there is no practical reasoning, no desires, and no intentions. Hence, questioning God’s actions with regard to their wisdom is absurd. God’s actions manifest His wisdom; but this is not sufficient to explain why these actions are unquestionable. The previous āyah contains a reference to the divine throne, which is taken to be an indication of divine sovereignty. This leads ‘Allamah to the conclusion that the unquestionability of God’s actions is fundamentally due to God’s sovereignty and unconditional ownership of all things. Contrary to ‘Allamah, we can imagine an intelligible questioning of God’s actions both with respect to their wisdom and with respect to God’s right to perform them. With regard to divine wisdom, this is said to be manifest in creation. The previous āyah states: ‘Had there been gods in them [the heavens and the earth] other than Allah, they would surely have fallen apart (lafasadtā)…’ (21:22). If we suppose that the divine wisdom is manifest in all things, at the very least because the heavens and the earth are not falling apart, one might nevertheless point to particular cases in which things are falling apart and ask how this manifests the divine wisdom. Although we may assume there are adequate answers to this having to do with the laws of nature or some metaphysical harmony, one might still ask how God’s wisdom is manifest in such cases. A believer might be convinced of the divine wisdom in all things, and still question it, not for the sake of denying it or for raising doubts, let alone for making accusations against God, but for the sake of understanding how the world manifests the divine wisdom. Notice that this kind of questioning would be appropriate even if we accept ‘Allamah’s position (and that of the tradition of Shi‘i philosophical theology he represents) that God’s wisdom does not consist in doing things for reasons. With regard to the divine right to perform the action of creation as God did, this can be questioned even if divine ownership of all creatures is taken to be absolute. First, one might grant absolute ownership as a matter of power and possession, but question whether the divine owner’s treatment of

214  Hajj Muhammad Legenhausen His property is just. The ownership of slaves, according to Islamic law, did not give the owner a right to certain abuses of the slaves. So, one may recognize that creatures are the slaves of Allah without admitting that absolute ownership gives the owner the right to do anything at all to his possessions. Second, one may grant that God has the right to treat His creatures as He will but continue to question how this treatment manifests the divine attributes of mercy and beneficence. Finally, if the reasons given by ‘Allamah for the unquestionability of God’s actions were correct, then since these reasons are independent of whether God ever sent messengers or made any covenant with human beings, the divine actions would have remained unquestionable in the counterfactual situations in which there were no messengers sent or covenant made. This, however, is contrary to the āyāt of the Qur’ān that explain the covenant and the messengers as removing any potential cause for complaint on the part of humans. In defense of ‘Allamah’s position, one might argue that what is at issue is only the kind of questioning that is made on behalf of a higher authority, where the questioning is done to determine whether punishment is due. Since there is no higher authority than God and no one could punish God, God cannot be questioned in this sense. This would permit a kind of divine moral accountability such that it could be at least counterfactually supposed that God would have been blameworthy in some counterfactual situations. Such situations are explicitly presented in the Glorious Qur’ān itself, as where Allah, the Mighty and Majestic, says that he sent: ‘…apostles, as bearers of good news and warners, so that mankind may not have any argument against Allah, after the [sending of the] apostles; and Allah is all-mighty, all-wise’ (4:165).27 If God had not sent the prophets, peace be with them, then some sort of moral questioning of God would have been expected, despite the divine absolute ownership of his creatures. In general, God is able to place Himself under obligations by the implicit specification of conditions under which He could be subject to justified blame. Thus, when God promises to reward those who do good deeds and have faith, he obligates himself by implying that he would be blameworthy if he were not to keep the promise. When God takes on obligations, he opens the way to the unrealizable potential for moral blame. The unrealizability is due to the divine perfections. The potential is in the conceivability: God would be blameworthy were he not to keep his promises. In earlier reflections, I thought that the only obligations that could be ascribed to God would be those for which there is explicit scriptural grounding; and in the Qur’ān and hadith literature there are plenty of examples in which God ascribes obligations to himself. However, in the counterfactual cases mentioned in the Qur’an, such as the above mentioned (4:165), God may be said to have an obligation to provide guidance if he is to justly punish those who go astray, regardless of whether any such obligation is explicitly mentioned in the Qur’ān or not. It is by no means clear the extent of the obligations that are not binding on God because of his absolute ownership of his

Analysis of Some Problems of Evil  215 creatures and those obligations that may be assumed to apply to God, despite divine sovereignty. While I do not claim to be able to provide a definitive criterion for the discernment of divine obligations, some general considerations about the nature of moral obligation will provide reason for holding that neither does God have an obligation to eliminate gratuitous evil, nor is he morally prohibited from creating a world in which there is gratuitous evil. 11.5

Moral Norms

There are a number of rival theories about the basis of moral norms. Virtually all such theories assume that moral norms apply to humans. Human moral norms do not apply to animals. If gorillas occasionally murder one another, it might conceivably be a violation of a moral norm found in gorilla societies, although this would be disputed; and even if there are moral norms in animal societies, they do not form systems of norms in such a manner that we could say that the same norm applies across species. In theological language, one could say, by analogy, that the divine morality is infinitely beyond the morality of human beings, given that human morality is to a more limited extent beyond the morality or proto-morality that is found in other social animals. In the ethical views of Plato and Aristotle, virtue is defined in terms of the faculties of the soul: courage is attained by controlling one’s fears, temperance by controlling one’s appetites, wisdom by controlling one’s intellect, and justice through the harmony of the other three. But God does not require virtue to control fears or appetites; so, there is no courage or temperance for God, and nothing corresponding to them is to be found in the most sacred Names attributed to Allah in the Qur’ān. God is said to be just and wise, however. So, half of the cardinal virtues are applied to God and half are considered inappropriate. But the entire project that sees the virtues as necessary conditions for eudemonia cannot be applied to God, for God has no need to maintain virtue for the sake of any kind of divine happiness. If we turn to more recent theories of norms, we will also find no reason to think that because a moral norm applies to humans, it will also apply to God. The authors of the book Explaining Norms, Geoffrey Brennan, Lina Eriksson, Robert E. Goodin, and Nicholas Southwood, divide theories of norms into reductive and nonreductive theories. The two major projects for reductive accounts of norms are the social practices view, which holds that all norms may be understood as tradition, conventions, or customs; second, there is the desires view, which understands norms as clusters of desires plus non-normative beliefs. The nonreductive accounts, like the one favored by the authors, take norms to be inexplicable without circularity, and, hence, primitive. The nonreductive view defended in Explaining Norms holds that ‘norms are clusters of normative attitudes plus knowledge of those attitudes’.28 None of the views surveyed requires that God must be excluded from the scope of norms, although it is assumed that the norms under discussion arise in human societies. If norms are to be applied to God, God must be considered to be included in the groups of agents whose relations are governed by

216  Hajj Muhammad Legenhausen norms. This does not imply that God must personally endorse the norms or that he cannot be considered exceptional. We can develop norms in our societies that apply to all members of the society, including its criminal elements, whether or not they recognize these norms. We can also develop norms that make sin reprehensible in a society of sinners who understand these norms as being endorsed by God, and who is considered to be exempt from the kinds of factors that we take to cause sin among humans, such as greed, selfishness, cowardice, and the other moral vices. The norms we adopt might not be widely practiced, or may be practiced very imperfectly. So, norms cannot be reduced directly to practices. In the broadly empiricist tradition, norms are understood as clusters of desires and beliefs that serve to promote coordination and cooperation within the groups these norms govern, which have partially overlapping goals. Of course, God has no interest to be served by norms in this way, although we might ascribe interests to God by extension; for we can say that the fulfillment of divine commands is in God’s ‘interests’, even while, philosophically speaking, we deny that God has the wants or preferences that define human interests. If the language of interests were extended in this way, however, it would not mean that moral norms are to be applied to God for the sake of cooperating in the achievement of common interests with believers, assuming that the believers want to avoid sin and live in a society in which sin is minimized. But God does not play the role of an equal member of a society of moral agents here. The society of believers benefits by the avoidance of and condemnation of sin. The benefit is for human agents, and not for divinity. Through divine commands and heavenly guidance God may encourage respect for norms that facilitate social coordination and cooperation among believers, but this is separate from the ways in which humans interact with God, through worship, observance of divine guidance in scripture and narration, and the spiritual life. According to the ‘norms as normative attitudes view’ defended in Explaining Norms, the function of norms is to set standards by which we hold one another accountable. In societies in which religious and moral norms are blended together, believers hold one another accountable and they expect themselves to be held accountable by God. To imagine oneself as accountable before God reinforces the restraints and encouragements of religious/moral norms beyond the limitations of human blame and praise. Although we do enter into relations of moral recognition with God in which there are implicit counterfactual situations in which God could be blamed, e.g., if he were cruel or broke his promises or lied to us, it is assumed that such situations can never arise because of the divine perfections. Although it is conceivable that we should enter into relations of mutual accountability with God on the basis of general moral principles to maximize welfare and eliminate gratuitous evil, there is no theological basis for this sort of ethics. In covenantal theology, what is agreed upon is that God will guide believers and reward those who have faith and do good works. While God’s just judgment of his creatures might be considered as counting as morally praiseworthy divine works, God is generally held to be above the moral

Analysis of Some Problems of Evil  217 law as understood when it is applied to humans. For example, the idea of holding God accountable for property rights is annulled with the claim that all creatures are God’s property. When the Lord taketh away, however, it is not considered stealing. The concept of stealing is not properly applied to God because the rules that govern property rights are applied exclusively to humans, and the divine ownership of all things functions not to include God in the institution of property but to absolve him of any claims that he has unjustly taken something from someone. We hold one another accountable for what we take, while we take God to be unaccountable for what is taken from us, but praiseworthy for all he has given us. More generally, we hold one another responsible for any harms we do to one another, while God is not to be blamed for the harms done through what insurance companies call ‘acts of God’, such as floods, storms, and volcanic eruptions, and is praiseworthy for all the harms from which we have been spared. By making God exceptional in this way, human moral norms are reinforced in three ways: first, divine moral perfection is upheld as an ideal, despite what would be violations for human agents; second, the moral norms are taken to be endorsed or, in some cases, explicitly commanded by God; third, God cannot be given as an excuse for human violations, e.g., I cannot cause harm with the excuse that all harms and benefits ultimately come from God. In recent philosophy of religion, the issue of the moral accountability of God has been given extensive treatment by Mark Murphy.29 The position I have sketched here is in agreement with Murphy’s in denying that God has any moral obligation to maximize the welfare of his creatures, or to create a world free from gratuitous evils. Brian Davies also argues that God is not morally accountable by human standards of morality; but Davies claims that since God does not act on the basis of reasons, he is not a moral agent at all.30 The position I have sketched falls between those taken by Murphy and Davies. Like Davies, I hold that God does not engage in practical reasoning to decide what to do; but I don’t think that this implies that God is not a moral agent at all. So, I agree with Murphy that God does have moral obligations and that these obligations do not correspond with the moral obligations of humans, although there is overlap. I agree with Davies that the goodness of God is primarily to be explained with reference to the divine perfections, and to the fact that all that is good is from him. But there are sources in the Qur’ān and hadiths that indicate that God has taken on obligations, too. God’s goodness does not consist in his fulfilling his obligations; rather it is because of his goodness that he fulfills his obligations. I am in agreement with Davies’ position that God does not act for r­ easons; and I find support for this in Islamic discussions of the nature of divine agency. Sharon Street has also argued persuasively that there cannot be divine reasons for some evils.31 She argues that on the most widespread understandings of theism, according to which God is a person, the lack of divine reasons for evils implies that God does not exist. I have tried to show that the conception of God developed in Islamic philosophy provides resources to respond to Street. We can agree that if God were an intentional agent, the range of evils in

218  Hajj Muhammad Legenhausen the world would lead to normative skepticism. The conclusion to be drawn, ­however, is not that there is no God, but that he is not an intentional agent. 11.6

The Problem of Evil

Problems of evil arise when gratuitous evil is taken to be inconsistent with— or evidence against—the existence of a divinity who is omnipotent and ­morally good. Theodicies may be understood as attempts to show that there are no gratuitous evils. The evils in the world all occur for reasons that exculpate the Creator. Either the evils in the world are a necessary result of the creation of a material world, or they are compensated by goods that could not be achieved without creating a world with as much evil as there is. The classical theodicies attempted in a general way to show what sorts of reasons God might have had for failing to create a world with any less evil than it has. Skeptical theodicies are those that claim that God must have some unknown reasons for having allowed as much evil to appear in the world as there is.32 For the most part, whether skeptical or not, proponents of theodicies claim that even an omnipotent divinity would not be able to create a world with greater overall value than there is in the actual world. There is no way for even an omnipotent being to reduce the overall amount of evil in the world. The inconsistency between gratuitous evil, omnipotence and divine goodness is removed by theodicies by limiting omnipotence. Even an omnipotent deity could not have done any better than this. Another sort of response does not claim that nothing better could have been created, but this does not make God blameworthy either because God is not a moral agent (Davies) or because he is not obligated to avoid gratuitous evil (Murphy). Contrary to Davies, I have suggested that God may be understood as a moral agent because he presents himself as such through revelation, although, contrary to Murphy, God does not act for reasons and does not engage in practical reasoning. Although many Muslim thinkers have sought to answer the problem of evil by arguing that this must be the best of all possible worlds, I have suggested there are sufficient resources in the Islamic tradition to provide responses to the problem of evil without recourse to this dubious strategy. The legacy of Neoplatonism in Islamic thought provides reason to deny that God is a person and to reject the claim that he uses practical reason to decide upon the divine actions. Nevertheless, there are theological grounds in the Qur’ān and narrations that support the view that God takes obligations upon himself. These obligations consist in the acknowledgement that he would be blameworthy if he did not fulfill these obligations, although we are assured that the divine perfections guarantee that the obligations will be fulfilled. Hence divine goodness consists in both the divine perfections, which are intrinsically good, and in the divine fulfillment of obligations that are indicated through revelation, both specifically and generally.

Analysis of Some Problems of Evil  219 Despite the fact that God may be understood as a moral agent, we cannot conclude that the same moral norms that apply to humans will necessarily apply to God. Moral norms arise in human societies in order to fulfill social needs such as the needs to coordinate behavior and the needs for mechanisms whereby the members of society are able to hold one another responsible. In religious societies, God also plays a role in the social arrangements in which moral norms emerge, but not as one who is subject to these norms, with notable exceptions. The function of divinity in relation to moral norms in religious societies is to provide motivation to respect the law, regardless of possible human praise and blame or punishments. Furthermore, revelation serves to establish conventions needed to resolve some differences in what moral norms are to be maintained. The exceptions arise from the fact that God enters into social arrangements with humans, not as an equal, but as one who makes assertions through revelation that imply divine obligations, directly or indirectly. These obligations do not include any that would require God to maximize human welfare or eliminate gratuitous evils; but God knows best! Notes 1 Still unsurpassed is Ormsby (1984). 2 Inati (2017, 165). 3 Rawls (1971, 114–117), cited in Legenhausen (1988); this is further developed in Legenhausen (1996). 4 Rawls (1971, 115). 5 Kenny (1979). 6 A similar criticism has been issued against Hegel’s philosophy of religion by William Desmond, who calls Hegel’s God, ‘a counterfeit double’. See Desmond (2003). 7 Schärtl, Tapp, and Wegener (2016, 27). 8 The above comparison of trinitarian and Neoplatonic theories is drawn from Lilia (2017). Also helpful for the Neoplatonic background to these discussions, see Remes (2008). 9 Lilia (2017, 164). The position found in the Qur’ān in this regard contradicts Basil’s: ‘They question you concerning the Spirit. Say, “The Spirit is of the command of my Lord, and you have not been given of the knowledge except a few [of you]”’ (17:85). Translations of the Qur’ān are from Qara’i (2003), unless otherwise indicated. 10 Schwen (2010, 84–85). The possibility of an identification of the logos, kalimah, and spirit, rūḥ, with Jesus is also suggested in the following āyah, although there are other uses for these terms in the Qur’ān, as well: The Messiah, Jesus son of Mary, was only an apostle of Allah, and His Word that He cast toward Mary and a Spirit from Him. So have faith in Allah and His apostles, and do not say, ‘Three’ (4:171). (My revision of Qara’i’s translation.) 11 Schwen regards Aphrahat’s stance as ‘basically binitarian—God and Christ, or God and the Holy Spirit—viewing the Logos-Christ and the Holy Spirit as essentially one’. Schwen (2010, 92); cited by Ruzer and Kofsky (2010, 19, n. 62). 12 Adamson (2007, 54). 13 Majid Fakhry writes: ‘It is to be noted at this point that in identifying the First with intellect or thought, al-Fārābī was unwittingly breaking with the Neoplatonic

220  Hajj Muhammad Legenhausen tradition, as represented primarily by its founder, Plotinus, for whom the One is entirely transcendent, lying above being and thought’ (Fakhry, 2002, 81). 14 Avicenna states: ‘The Necessary Existent is pure intellect because He is an essence dissociated from matter in every respect’ (Avicenna, 2005, 284). 15 For more on Neoplatonism in Islamic philosophy, see Fakhry (2002), Kalin (2010), Morewedge (1992), and Netton (1989). 16 Misbah Yazdi (1999, 314–319). 17 For a typology of responses to the problem of evil in Christian and Islamic traditions, I recommend (Hoover, 2003). 18 Legenhausen, Does God Have a Mind? (1393/2014). 19 Legenhausen, Is God a Person? (1986). The position that God is not a person is also defended from a Christian perspective in Davies, The Reality of God and the Problem of Evil (2006, 93ff). 20 Mutahhari (1357/1978). 21 My translation, cf. Ibn Babawayh (2009), Bāb 61, comments on the twelfth narration. 22 I argued this point in Legenhausen (1988), which also used the volcano example. 23 See the discussion by Avicenna in Avicenna (2005, 303f). 24 My translation from Lahiji (1372/1993, 349). 25 (My translation of the Qur’ān.) In the New Testament, Paul writes ‘Who indeed are you, a human being, to argue with God?’ Rom. 9:20; and on the basis of this, Davies argues that God is not ‘an agent subject to moral requirements’ (Davies, The Reality of God and the Problem of Evil, 2006, 96). 26 For the English translation, see Tabataba’i (2021); the original is Tabataba’i (1970). 27 Also see (7:172–173) and the discussion in Legenhausen, Dimensions of Divine Freedom (2021). 28 Brennan, Eriksson, Goodin, and Southwood (2013, 15). 29 Murphy (2017). 30 See Davies, Review of Mark Murphy’s God’s Own Ethics (2019). 31 Street (2014). 32 For an influential collection of essays in which ‘skeptical theism’ is debated, see Dougherty and McBrayer (2014).

References Adamson, P. (2007). Al-Kindi. Oxford: Oxford University Press. Avicenna. (2005). The Metaphysics of the Healing. (M. E. Marmura, Ed.). Provo: Brigham Young University Press. Brennan, G., Eriksson, L., Goodin, R. E., & Southwood, N. (2013). Explaining Norms. Oxford: Oxford University Press. Davies, B. (2006). The Reality of God and the Problem of Evil. London and New York: Continuum. Davies, B. (2019). Review of Mark Murphy’s God’s Own Ethics. The Thomist, 83(1), 111–115. Desmond, W. (2003). Hegel’s God: A Counterfeit Double? Aldershot: Ashgate. Dougherty, T., & McBrayer, J. P. (Eds.). (2014). Skeptical Theism: New Essays. Oxford: Oxford University Press. Fakhry, M. (2002). Al-Farabi, Founder of Islamic Neoplatonism. Oxford: Oneworld. Hoover, J. (2003). A Typology of Responses to the Philosophical Problem of Evil in the Islamic and Christian Traditions. Conrad Grebel Review, 21(3), 81–96. Ibn Babawayh, M. I. (2009). Kitab Al-Tawhid. (A. R. Rizvi, Trans.). Qom: The Savior Foundation.

Analysis of Some Problems of Evil  221 Inati, S. (2017). Ibn Sīnā’s Theodicy, 2nd ed. Piscataway, NJ: Gorgias Press. Kalin, I. (2010). Knowledge in Later Islamic Philosophy: Mulla Sadra on Existence, Intellect and Intuition. Oxford: Oxford University Press. Kenny, A. (1979). The God of the Philosophers. Oxford: Oxford University Press. Lahiji, ’.-R. F. (1372/1993). Gawhar-e Morad. Tehran: Vizarat Irshad. Legenhausen, G. (1988). Notes Toward an Ash`arite Theodicy. Religious Studies, 24, 257–266. Legenhausen, M. (1393/2014). Does God Have a Mind? In Muhammad Baqir Ansari (Ed.), Nafs va badan dar akhlāq va Insānshenāsī (Soul and Body in Ethics and Anthropology), Vol. 7 (pp. 158–202). Qom: Academy of Islamic Sciences and Culture. Legenhausen, M. (1986, Sept./Dec.). Is God a Person? Religious Studies, 22, 307–323. Legenhausen, M. (1996). Lecture 13: Evil, Part 8 of the series The Contemporary Revival of the Philosophy of Religion in the United States. Al-Tawhid, 13(1), 147–161. Legenhausen, M. (2021). Dimensions of Divine Freedom. Journal of Philosophical Theological Research, 21(3), 73–91. Retrieved from http://pfk.qom.ac.ir/article_2030.html Lilia, S. R. (2017). The Neoplatonic Hypostases and the Christian Trinity. In M. Joyal (Ed.), Studies in Plato and the Platonic Tradition (pp. 127–189). London and New York: Routledge. Misbah Yazdi, M. (1999). Philosophical Instructions. (M. Legenhausen, & ’. Sarvdalir, Trans.). Binghamton, NY: IGCS and Brigham Young University. Morewedge, P. (Ed.). (1992). Neoplatonism in Islamic Thought. Albany: State University of New York Press. Murphy, M. (2017). God’s Own Ethics. Oxford: Oxford University Press. Mutahhari, M. (1357/1978). ’Adl ’Ilahi (Divine Justice). (10th printing ed.). Qom: Entesharat Sadra. Netton, I. R. (1989). Allah Transcendent: Studies in the Structure and Semiotics of Islamic Philosophy, Theology and Cosmology. Routledge: London. Ormsby, E. (1984). Theodicy in Islamic Thought. Princeton, NJ: Princeton University Press. Qara’i, S. A. Q., tr. (2003). The Qur’an with English Paraphrase. Qom: The Centre for Translation of the Holy Qur’an. Rawls, J. (1971). A Theory of Justice. Cambridge: Harvard University Press. Remes, P. (2008). Neoplatonism. Stocksfield: Acumen. Ruzer, S., & Kofsky, A. (2010). Syriac Idiosyncrasies. Leiden: Brill. Schärtl, T., Tapp, C., & Wegener, V. (Eds.). (2016). Rethinking the Concept of a personal God. Münster: Aschendorff Verlag. Schwen, P. (2010). Afrahat: Seine Person und sein Verständnis des Christentums. Piscataway, NJ: Gorgias Press. Street, S. (2014). If Everything Happens for a Reason, Then We Don’t Know What Reasons Are: Why the Price of Theism is Normative Skepticism. In M. Bergmann, & P. Kain (Eds.), Challenges to Moral and Religious Belief: Disagreement and Evolution (pp. 172–192). Oxford: Oxford University Press. Tabataba’i, S. H. (1970). Al-Mizan fi al-Tafsir al-Qur’an, 20 vols. Beirut: Mu’asassat al-A’lami li al-Matbou’aat. Tabataba’i, S. H. (2021). 21. Al-Anbiya, Verses 16-33. Al-Mizan Volume 28. (T. Raja, Trans.). Tawheed Institute Australia. Retrieved March 27, 2022, from almizan.org: https://almizan.org/

PART VI

Ethics and Morality

12 Islamic Philosophers’ Idea of Heaven A Moral Analysis Amir Saemi

In “the pursuit of the ideal”, Isaiah Berlin argues that any notion of the ideal is incoherent and immoral. Berlin believes that any notion of the ideal is bound to encounter two problems. [The coherence problem] “The notion of the perfect whole, the ultimate solution, in which all good things coexist, seems to me to be not merely unattainable – that is a truism – but conceptually incoherent.” (Berlin 2013, p. 14) [The moral problem] “In a society in which the same goals are universally accepted, problems can be only of means…. That is a society in which the inner life of man, the moral and spiritual and aesthetic imagination, no longer speaks at all.” (Berlin 2013, p. 15) The coherence problem concerns the coherency of the ideal. The problem arises from the fact that there is a plurality of values and that these values are often incompatible. Any good human life necessarily lacks some values and must have at least some disvalues, and so the ideal life cannot be ­conceived coherently. The moral problem is this: in the assumption of  the  existence of the ideal life, morality would be only instrumental, and therefore not indispensable. Almost all religions portray life in heaven as the ideal life. Heaven stands for the ideal. Berlin’s objections, then, are applicable to the idea of heaven. In this chapter, I will discuss Berlin’s objections vis-à-vis the Islamic notion of heaven advocated by Islamic philosophers such as al-Fārābī and Avicenna. While I mainly discuss the views of al-Fārābī and Avicenna (I mention al-Ghazālī only in passing), it is important to note that al-Fārābī and Avicenna’s view of heaven had a tremendous influence on subsequent Islamic philosophers, to such an extent that their view of heaven is almost shared by nearly every subsequent Islamic philosopher, barring a few exceptions and small variations. Moreover, although I focus primarily on philosophers’ conceptions of DOI: 10.4324/9781003327714-19

226  Amir Saemi heaven, other conceptions of heaven which are similar in key respects to philosophers’ ­conceptions of heaven are susceptible to the same type of difficulties discussed here. Finally, although I consider the coherence problem to be very important, my discussion of the coherence problem in this chapter will be very brief. Instead, the majority of this chapter is devoted to discussing the moral problem. 12.1

Philosophers’ Conception of Heaven

Nicomachean Ethics begins with Aristotle distinguishing between production (poiesis) and chosen activity (praxis). The aim of production is to produce products. So, products are more worthy of productions because the activity of production is done for the sake of its products. Praxis, on the other hand, has no end over and beyond its very activity. Aristotle’s general principle is this: if an action is performed for the sake of something else, then that thing is better than the action. Let’s call actions that have goals beyond themselves telic actions and actions that do not have goals beyond themselves atelic actions.1 Aristotle’s principle implies that telic actions are less worthy of their goals, whereas atelic actions enjoy the possibility of being sought only for their own sake. According to Aristotle, the perfect activity should be atelic, it should be performed for its own sake. Perfect life, or eudemonia, then, for Aristotle, consists in atelic activities. This idea was also shared by al-Fārābī and Avicenna. For example, Avicenna characterizes the final good as follows: The good is that which is sought for its own sake, everything else being sought for its sake. For, if something is sought for the sake of some other thing, then it would be [merely] beneficial, not a true good. (The Metaphysics of The Healing, 8.3.25, p. 270) If only atelic actions can be truly good, the perfect life in heaven would c­ onsist of only atelic activities, activities that have no goal beyond themselves. But what are those atelic activities? Al-Fārābī and Avicenna hold that the happiness for humans is to achieve perfection, and human perfection consists in the perfection of the rational soul. The rational soul attains perfection when it learns all intelligibles. According to al-Fārābī, the soul would not need the body anymore when it learns all intelligibles. In his description of happiness, he says: The presence of the first intelligible in man is his first perfection, but these intelligibles are supplied to him only in order to be used by him to reach his ultimate perfection i.e., felicity. Felicity means that human soul reaches a degree of perfection in (its) existence where it is no need of matter for support, since it becomes one of the incorporeal things and the immaterial substances that remains in that state continuously for ever. But its rank is beneath the rank of the Active Intellect. (The Perfect State, section 13, pp. 206–207)

Islamic Philosophers’ Idea of Heaven  227 When the rational faculty attains to being an intellect in actuality, that intellect … becomes similar to the separate things and it intellects its essence that is [now] intellect in actuality. …. Through this, it becomes such as to be in the rank of the active intellect. And when a human being obtains this rank, his happiness is perfected. (Political Regime, p. 33) Avicenna describes happiness as follows: [In the case of] the rational soul, the perfection proper to it consists in its becoming an intellectual world in which there is impressed the form of the whole; the order in the whole that is intellectually apprehended …. until it completes within itself [the realization of] the structure of existence in its entirety. It thus becomes transformed into an intelligible world that parallels the existing world in its entirety. (The Metaphysics of The Healing, 9.7.11, p. 350) Avicenna holds then that life in heaven consists in the soul’s understating of all truths. There is a slight difference between Avicenna’s view of afterlife and al-Fārābī’s. Al-Fārābī believes that the perfect human being somehow joins the active intellect. Al-Fārābī also seems to believe that imperfect souls who do not join the active intellect would not be immortal. While Al-Fārābī’s view remains unclear regarding the question of a personal afterlife, Avicenna believed that each person had an eternal soul that lived in the afterlife. I am not interested in the differences between al-Fārābī’s view and Avicenna’s regarding immortality. What interests me is that each characterizes the life in heaven as intellectual; the perfect soul is the one in which all knowledge is imprinted. Thus, the activity of the soul in heaven is to know and to understand. Arguably, the activity of understanding is an atelic activity which has no purpose beyond itself; this atelic activity would make for the perfect life. Avicenna holds that understanding is an activity that brings immense pleasure, especially when the body is separated from the soul and the soul’s perception is not distorted by the body. If the rational faculty had brought the soul to a degree of perfection by which it is enabled, when it separates from the body, to achieve that complete perfection that is [appropriate] for it to attain, its example would then be that of the benumbed person made to taste the most delicious taste and exposed to the most appetizing state but [who] does not feel this, but [who thereafter] has the numbness removed, experiencing [as a result] momentous pleasure all at once. This pleasure would not be of the same genus as sensory and animal pleasure, but a pleasure that is similar to the good state that belongs to the pure living [celestial] substances. It is more elevated than every [other] pleasure and more noble. This, then is happiness. (The Metaphysics of The Healing, 9.7.17, p. 352)

228  Amir Saemi As implied by the passage, according to Avicenna, in heaven, a soul’s ­activity is not just one of understanding, it is also one of experiencing pleasure. Experiencing pleasure is also an atelic activity, an activity with no purpose beyond itself. Al-Ghazālī infamously charge Avicenna with disbelief for his view on afterlife. The point of the contention is Avicenna’s denial of physical resurrection in the afterlife. Al-Ghazālī, however, does not reject Avicenna’s view of heaven completely. He thinks the Avicenna’s view of heaven is more or less correct, except that in heaven there can also be physical pleasure, in addition to intellectual pleasure. Most of these things [i.e., Avicenna’s view on the afterlife] are not contrary to the religious law. For we do not deny that there are, in the hereafter, kinds of pleasures superior to the sensory. Nor do we deny the survival of the soul after separation from the body[…]What is contrary to the religious law among [the things they hold] is the denial of the resurrection of bodies, the denial of bodily pleasures in paradise…. (The Incoherence of the Philosophers, Discussion 20) I am not concerned with the debate about whether pleasure in heaven is physical, intellectual, or spiritual. The important point is that it appears that the dominant view among Islamic philosophers concerning life in heaven is that it consists of understanding and/or experiencing pleasure. There are two key elements to this conception of heaven. First, it is morally neutral, i.e., it does not involve any essential reference to the moral life of the agent. Second, it involves only atelic activities. I will argue that any conception of heaven that has these two features faces the coherence problem as well as the moral problem. 12.2

The Coherence Problem

Before getting to the moral problem, let me briefly touch on the coherence problem. According to the coherence problem, the notion of the perfect life is incoherent. There are competing goals in life. If one pursues some valuable goals, one can’t pursue some other valuable goals. Moreover, the pursuit of valuable goals necessarily involves the promotion of some disvalue. So, no life is perfect because it necessarily misses out some on valuable activities and necessarily includes some disvalued activities. For example, suppose you value doing philosophy. Doing philosophy well requires spending a considerable amount of time studying alone. But that deprives one of attending some parties with friends. So, doing philosophy well implies missing out on some parties. Moreover, some of your friends would lose the chance of being with you in those parties and this can make them unhappy to some degree. So, doing philosophy involves compromising some values and promoting some disvalues (i.e., the unhappiness of your friends).

Islamic Philosophers’ Idea of Heaven  229 I won’t discuss whether Berlin’s coherence problem is insurmountable, for the problem is not directly applicable to the Islamic philosophers’ conception of heaven. Berlin’s incoherence problem stems from the fact that life offers a variety of different and competing valuable goals. But the Islamic philosophers’ heaven only involves atelic activities. In their views, the perfect life should not include any sort of goal at all. One worry for this conception is that it isn’t clearly true that telic activities are necessarily worse than (at least some) atelic activities. But setting this worry aside, a major problem of this conception of heaven is that it is so alien from anything that we know of as a good life that is not clear whether it would even be a good life for humans. Consider a normal good human life. It involves various telic activities. It might involve raising a kid, writing a book, spending time with friend, helping others, etc. All these activities are telic. For example, to spend time with friends, we need to pursue mutual projects e.g., going camping together, going to a movie, etc. This requires setting a goal, doing some planning, achieving the goal, and then moving on to the next thing. We have no idea what a life without goals would look like. We don’t even know whether it would be good, boring, or tormenting to live eternally with no goals. A life with no goals would not be a human life and it would not be anything we desire to have. Consider an atelic activity say experiencing pleasure. Experiencing pleasure with no goals would be at best something like having an orgasm. But would it be good to experience an eternal orgasm? Again, a perfect life without any goals would be so far from what we consider a good life that it seems strange to call it an ideal life. Of course, to mitigate the problem of the philosophers’ heaven, we might seek to incorporate some telic activities in it. But then Berlin’s incoherence problem reappears. 12.3

Philosophers’ Conception of Morality

Al-Fārābī and Avicenna have a teleological view of morality. For them, the purpose of morality is to help the soul to achieve the perfect life. As al-Fārābī states, “the voluntary actions which help in attaining felicity are the good actions… these being goods not for their own sake, but goods for the sake of felicity only” (The Perfect State, section 13, p. 207). On this conception of morality, once theoretical reason discovers the true nature of happiness, the goal of practical reason is to figure out a way to achieve it. Al-Fārābī describes the instrumental role of prudence (ta’aqul) as follows. It is particularly characteristic of [theoretical] wisdom that it knows the ultimate reasons for every ultimate being…Wisdom, therefore, is what seizes upon happiness in truth, whereas prudence is what seizes upon what ought to be done so that happiness is attained. These two, therefore, are the two mutual assistants in perfecting the human being[theoretical] wisdom being what gives the ultimate goal, and prudence being what gives the means by which that goal is gained. (Selected Aphorism, p. 35)

230  Amir Saemi The same view is also espoused by Avicenna. Avicenna holds that moral virtues are our means to achieve happiness. As he says, “this true happiness is not fully achieved except by rectifying the practical part of the soul” (The Metaphysics of The Healing, 9.7.20, p. 354). Avicenna shares Aristotle’s view that moral virtues are dispositions that lie on a mean between two extremes. On Aristotle’s view, the mean is determined by phronesis (i.e., practical wisdom or prudence) (Nicomachean Ethics 1107a). Phronesis, roughly speaking, is the virtue of intellect responsible for making judgements about what one ought to do. While it is clear that Aristotle thinks that one cannot exercise moral virtues without employing phronesis, it is less clear what role phronesis would play in one’s exercising virtues, given that Aristotle says “virtue makes the aim right, and practical wisdom the things towards it” (Nicomachean Ethics 1144a). On an influential interpretation, virtue provides one with an indeterminate goal which serves as the starting point of practical reasoning. The practically wise person then employs phronesis to make determinate the indeterminate goal with which she began. On this view, practical reasoning is constitutive reasoning through which the agent determines what specific goals constitute the general end identified by virtue.2 Irwin characterizes the role of phronesis as follows: “the practical intellect is not concerned with means as opposed to ends. Insofar as it is concerned with constituent “means,” it is also concerned with ends” (Irwin 1975, p. 571). McDowell (1998) and Wiggins (1980), have suggested that phronesis is similar to moral perception by which the agent can see what the right action is in each particular circumstance. Al-Fārābī and Avicenna, however, differ from Aristotle in that for them prudence or practical wisdom plays no constitutive role in practical reasoning. It only plays an instrumental role. Its function is to find where the mean is between two extremes given that our goal is to transcend the body and achieve theoretical perfection. He [i.e. Aristotle] has commanded in the books on ethics that the mean between two opposing moral temperaments should be used. …As the disposition toward the mean it is intended that transcend the conditions that tie [us to the body] and preserve the proper state of the rational soul, while so preparing [the rational soul] to go beyond and transcend [the body]. (The Metaphysics of The Healing, 9:7:10,22, p. 354) Al-Fārābī explicitly says that practical reason is subservient to theoretical reason. In his view, “practical reason is made to serve theoretical reason. Theoretical reason, however, is not to serve anything but has as its purpose to bring man to felicity” (The Perfect State, section 13, p. 209). Practical reason cannot discover practical truths independent of theoretical reason. In his view, phronesis (or, as he calls it, prudence) does not even deserve to be called wisdom because it is only for humans, and divine beings don’t need it (Selected Aphorism, p. 34).3 This human “wisdom” is only a servant of

Islamic Philosophers’ Idea of Heaven  231 the real wisdom, which is theoretical wisdom. It is the job of the t­heoretical ­wisdom to “seizes upon happiness in truth”, whereas prudence’s job is to “seize upon what ought to be done so that happiness is attained” (ibid., p.  35); it discovers what human actions are instrumentally necessary for achieving happiness. In this picture of morality, the goal of practical intellect is to help the soul learn theoretical truths by keeping the body in check. Avicenna describes the goal of morality as follows: We have established the true nature of the afterlife and have proved that true happiness in the hereafter is achieved through the soul’s purification. The soul’s purification removes it away from the acquisition of bodily dispositions opposed to the means for happiness. This purification is realized through moral [states] and positive dispositions. Moral states and positive dispositions are acquired by acts whose task is to turn the soul away from the body and the senses and to make continuous its remembrance of its [true] element. (The Metaphysics of The Healing, 10.3.5, p. 369) On a personal level, morality is about managing the body in a way that it enables the soul to flourish. Bodily desires need to be addressed to the extent that they are necessary for the proper functioning of the theoretical intellect. So, individual morality aims at keeping the body in balance. Moreover, the theoretical intellect can’t work properly in a chaotic society. Thus, morality also involves ensuring that society is well structured so that the theoretical soul may achieve perfection. The theoretical intellect, hence, determines the goals of morality at both the individual level and the social level. At both levels, morality instrumentally necessary to achieve theoretical perfection. In contrast to what I have claimed, some scholars of al-Fārābī maintain that perfection for al-Fārābī is at least partly practical.4 Holding that perfection is not purely theoretical, Booth writes that: The central trouble with theoretical perfection is that it seems at odds with what can be called al-Farabi’s motivational externalism—the view that knowing one ‘all things considered’ ought to Φ (and there are no excusing conditions with respect to Φ -ing) does not guarantee that one will be motivated to Φ. Following Aristotle, al-Farabi clearly held that weakness of the will (akrasia) is possible… As such, it appears that al-Farabi held that one could know what happiness is, and yet fail to be motivated to pursue it, such that knowledge of happiness does not entail being happy. (Booth 2016, p. 68) Booth uses his account of al-Fārābī’s certainty to show how perfection for al-Fārābī is partly practical. According to Booth, the prophet’s perfect power

232  Amir Saemi of imagination contributes to his having absolute certainty. For, the prophet’s power of imagination helps him to have a superior understanding of the world which is necessary for absolute certainty. What gives someone the property of superior understanding, the p ­ articular imaginative powers, perforce also gives that person superior practical wisdom. This means that all theoretical knowledge, insofar as it admits of absolute certainty, is a composite of theoretical and practical virtue. (ibid., pp. 70–71) Given this understanding of theoretical knowledge, in Booth’s view, because of his theoretical knowledge, derived partly from his power of imagination, the prophet does not succumb to the weakness of the will and thus a perfect life, at least on earth, is the life of the prophet, which is a composite of theoretical and practical perfection. It is also worth noting that Booth is aware that his account of perfect life cannot easily be extended to the perfect life in heaven. Imagination, which plays an essential role in Booth’s account of prophetic knowledge, requires body and there is no body in the afterlife. For one thing, I am not sure if I quite understand the objection to t­ heoretical account of perfection based on al-Fārābī’s acceptance of ­motivational externalism (i.e., the view that one’s knowledge that one ought to do an action does not guarantee that one will be motivated to do that action). It is true that al-Fārābī holds that “one could know what happiness is, and yet fail to be motivated to pursue it”. But happiness for him is not the knowledge of happiness (or the knowledge of what one ought to do to achieve happiness), it is the knowledge of all theoretical truths. One does not attain perfection when one knows what perfection is. One attains perfection when one acquires all intelligibles. Motivational externalism does not seem to be relevant to whether perfection is a purely theoretical matter. The worry might be that it is possible that one acquires all intelligible without being happy. But this possibility is different from the possibility of akrasia. Moreover, it is quite likely that al-Fārābī would deny this possibility. He could deny it by holding Avicenna’s view that knowledge of all theoretical truths is associated with significant intellectual pleasure (at least once the body is removed). Therefore, al-Fārābī could accept motivational externalism while denying that the person who achieves theoretical perfection could be unhappy. The more important point is that even if Booth and De Boer are right that for al-Fārābī perfection is partly practical, my main point that al-Fārābī’s morality is defined in terms of theoretical perfection still stands. Let’s grant for the sake of argument that Booth is right that absolute certainty requires imagination and that the imaginative faculty is necessarily coextensive with the practical faculty. Let’s also grant for the sake of argument that the prophet needs appropriate practical power not to succumb to the weakness of will. That still does not imply that practical virtues or phronesis do not have instrumental roles. The view of the opponents of purely theoretical perfection is

Islamic Philosophers’ Idea of Heaven  233 consistent with saying that the goal of the practical faculty is to figure out the best way to achieve theoretical perfection and to produce the appropriate motivational attitudes to make this goal possible. Motivational externalism has no implication about whether prudence is instrumental. So even if the perfect life partly consists of having appropriate motivational attitudes, the perfect life could still be a life of pursuing and understanding theoretical truths. Therefore, the current debate about whether al-Fārābī’s conception of perfection is partly practical does not bear on the issue of whether practical wisdom has autonomy in figuring out what the right action is. 12.4

The Moral Problem: The Extension Problem

As noted before, the Islamic philosophers have a teleological understanding of value. On this understanding, if X is done for the sake of Y, then the value of X is explained by the value of Y. Some things have value in themselves, while other things have value in virtue of their role in the promotion of things with intrinsic value. According to the Islamic philosophers, the only thing that has value in itself is felicity (‘saʿādah’) or happiness in heaven. All actions are done for the sake of felicity and thus their values are to be explained in terms of their role in achieving felicity. As discussed before, the Islamic philosophers define felicity in terms of atelic activities such as theoretical activities or experiencing pleasure. This implies a commitment to value monism according to which all moral values can be reduced to the single value of theoretical understanding (or experiencing pleasure derived from theoretical understanding). However, this value monism is very implausible. Not all moral values can be reduced to the value of theoretical understanding (or to the value of experiencing pleasure, or both). Berlin thinks that the notion of the ideal, which he calls it the final solution, is a dangerous idea. For if one really believes that such a solution is possible, then surely no cost would be too high to obtain it: to make mankind just and happy and creative and harmonious for ever – what could be too high a price to pay for that? (Berlin 2013, pp. 15–16) And this is a moral problem for any conception of felicity defined independently of the agent’s moral life. The moral problem for the Islamic philosopher’s conception of morality and felicity is that their view cannot capture the extension of common-sense morality. For example, it may well be possible that intuitively non-obligatory or wrong actions are necessary to bring about the attainment of heaven in which case intuitively non-obligatory or wrong actions are judged to be obligatory by the philosophers’ theory. Or, it may be possible that some intuitively obligatory actions do not bring about the attainment of heaven, in which

234  Amir Saemi case intuitively obligatory actions are deemed non-obligatory by the theory. Here is one way to illustrate this thought. Suppose φ is an intuitively morally wrong action, say a failure to perform one’s duty in an easy rescue case, e.g., not saving the little girl who is drowning in the pond in Singer’s famous example (2016). The problem is that φ’ing (i.e., failing to fulfill one’s duty in an easy rescue case) can bring about, or at least not hinder, the attainment of heaven. Yet, it is morally disturbing and most likely false to think that one’s not saving the little girl in an easy rescue case is right or permissible. The problem is quite general, for a vast number of moral values do not necessarily promote the theoretical perfection. It is too fantastical to think that we can explain all values in terms of the value of intellectual understandings. Therefore, if heaven is defined independently of the moral life of the agent, any “morality” defined in terms of its role in enabling us to achieve heaven would be very different from common-sense morality. One might think that there is a theological way to get around this moral problem – which we can call it the extension problem. The idea can be expressed by appealing to what Plantinga calls the noetic effect of sin (Plantinga 2000). According to Plantinga, a believer’s sinfulness ­undermines his knowledge of God. Assuming that heaven consists of intellectual ­understanding of God, sin – which is a failure to live up to more or less everyday morality – can deprive us of heaven. In other words, God created the structure of the world and our mind in such a way that following common-sense morality is instrumentally necessary to attain heaven and performing intuitively wrong actions would hinder the achievement of heaven. That is, God might guarantee that the morality defined by its instrumental role in achieving heaven is ­co-extensional with everyday morality, and thus the extension problem would be resolved. Put it differently, God could create the causal structure of the world in a way that saving the little girl in the pond is actually instrumentally necessary to go to heaven. This response is very artificial though. Having the same extension as common-sense morality is important because common-sense morality is important. If the point of morality were only to achieve felicity and felicity is essentially independent of the agent’s moral life then having the same extension as common-sense morality should not matter very much. Or, to put it another way, this response makes morality arbitrary in the same way that the divine command theory does. The major problem of the divine command theory is that God has no reason for His commands or prohibitions. And while, according to the divine command theory, His commands define morality, morality remains arbitrary. For God’s commands would be ultimately arbitrary with no normative justification. The same objection applies to the noetic effect of sin. Why should God care to make it the case that the failing to save the little girl is a sin, if the saving of the little girl was not already important? Why should God create a world in which all and only commonsense moral actions are instrumentally necessary for the attainment of heaven if morality has no value in itself? In other words, if a morality is just defined

Islamic Philosophers’ Idea of Heaven  235 by its role in attaining heaven, and if common-sense morality has no independent value, God has no reason to make all and only common-sense moral actions instrumentally necessary to attain heaven. At this point, one might say that the extension problem is not a problem for the philosophers’ conception of heaven as much as it is for their theory of morality i.e., for their value monism. It is value monism that should be abandoned, not the philosopher’s conception of heaven. It is not their conception of heaven in itself that is problematic, but the assumption that morality is defined by its role in achieving heaven. One could retain the philosophers’ conception of heaven while rejecting the instrumental nature of morality. In this picture, the value of morality is independent of its role in attainment of heaven. Yet, sin has the noetic effect, that is, moral behaviors, in fact, facilitate the attainment of heaven. But moral behaviors aren’t valuable simply for facilitating the attainment of heaven; they have independent intrinsic value. Heaven plays no role in fixing the extension of morality; the extension of morality is to be determined independently of heaven. If so, the extension problem may not be a problem for the notion of heaven in itself. 12.5

The Moral Problem: The Intension Problem

While Berlin sometimes expresses the moral problem with the notion of the ideal in terms of the extension problem, there are some other passages in his work which point to another moral problem for the notion of the ideal. We can call this problem the intension problem. While the extension problem concerns the problematic role of heaven in determining the extension of morality, the intension problem is about the morally problematic role of heaven in our rational behavior. The intension problem arises from this fact: if attaining heaven amounts to achieving eternal happiness, it is difficult to understand why attaining heaven should not be of the highest priority for any rational person. But, then, to use Berlin’s language, a society whose goal is to attain heaven “is a society in which the inner life of man, the moral and spiritual and aesthetic imagination, no longer speaks at all”. To put it differently, morality should guide our life and our actions, but if our true goal is to go to heaven, then it is not clear how morality could play such a role. Berlin does not distinguish between the extension problem and the intension problem, but the following passage is a passage in which he seems to be worried about the intension problem (for our theological discussion, I slightly modified (theologized) the passage). Since [God] know[s] the only true path to the ultimate solution of the problem of society, [God] know[s] which way to drive the human caravan; and since you are ignorant of what [God] know[s], you cannot be allowed to have liberty of choice even within the narrowest limits, if the goal is to be reached. (Berlin 2013, p. 16)

236  Amir Saemi To illustrate the intension problem, let’s suppose that moral actions are in and of themselves valuable. Yet, this does not mean that when one has a choice between doing the right thing and pursuing heaven, it is rational to sacrifice one’s eternal happiness for the sake of morality. Our eternal happiness is such a valuable goal that it is hardly rational to forsake it for anything, even morality. We cannot blame people for not forsaking their eternal happiness. But now suppose that there is a conflict between morality and the pursuit of heaven. To have a concrete example, imagine a person who is convinced, perhaps because of his education, that he has a religious mission to set a bomb in a school in which there are innocent children. Although he might feel uneasy if he killed those children, he believes that doing so would bring him closer to heaven. Given this belief, which may very well be justified by the misleading evidence that he had, it looks rational for him not to spare the children’s lives. We can’t blame him for putting his eternal happiness above all other goals. Examples of this sort can show that people can rationally and blamelessly commit deeply immoral acts when seeking heaven. Nevertheless, performing deeply immoral actions is never rational and is almost always blameworthy, or so I believe. And this creates a moral problem with the concept of heaven: A belief in heaven can provide rational excuse for people to commit immoral acts that they shouldn’t be rationally excused for. There are two ways to respond to the intension problem: (i) One might say that the pursuit of heaven never provides a rational excuse to perform deeply immoral actions. Or, (ii) one might say that the intension problem is not peculiar to the notion of heaven: under any plausible moral theory, moral ignorance provides excuse for committing immoral acts. Let’s call the first response “No Immoral Path to Heaven”, and the second, “The Excuse of Moral Ignorance”. Let’s start with No Immoral Path response. First, we should be clear that the intension problem is not about whether actions that are in fact immoral can lead to heaven. The question is whether a person can justifiably believe that performing an action, that is in fact immoral, can lead to heaven. The intension problem arises as long as there is such a possibility. And I can’t see how this possibility can be denied. Imagine someone being taught that religious law is above secular moral law. He may believe that although the act of bombing is against secular moral laws, it is in agreement with the religious law. Even if this view is incorrect, one can justifiably come to believe it due to one’s misleading evidence. According to the Excuse of Moral Ignorance, it is not the belief in heaven that provides the rational excuse for immoral actions. The excuse is provided by moral ignorance. However, various philosophers have argued that moral ignorance does not generally, or ever, remove blame (for example, see Harman 2011). I don’t want to be committed to a strong view here. My point is simply that the prospect of eternal happiness provides an excuse in a way that moral ignorance does not. To illustrate, consider a case of pure moral ignorance; imagine a racist who does not know that racism is wrong. He is still blameworthy for who he is and how he acts. He is a person who

Islamic Philosophers’ Idea of Heaven  237 does not respect others, a person who is unfair and uncaring. Pure moral ignorance, i.e., ignorance of purely moral facts, does not absolve one from responsibility. There are reasons to be kind, caring, and fair toward others, even if one’s flawed character prevents one from appreciating those reasons. The racists would be blameworthy for his flawed character and for not seeing the reasons that he should have seen. But now consider the person who sets the bomb to go to heaven. Even if he sees strong reasons for not killing the children, it would be too much to ask him to abandon his eternal happiness for those reasons. Living happily eternally is so valuable than it can outweigh moral reasons not to carry out the mission. This is especially true when he also believes that the innocent children who die in the mission will also go to heaven as well. And this is the main idea behind the intension problem. The prospect of heaven and eternal happiness for oneself and, the belief that God will take care of our action’s negative effects on others in the afterlife, neutralizes moral objections to immoral actions in a way that moral ignorance does not. ­ erson One might still say that the pursuit of heaven is irrelevant here. The p who performs an immoral action out of religious conviction is a person who thinks that religion must guide us morally, and he performs that action because he thinks that it is the right thing to do. If so, doing something immoral for heaven is exactly like any other case of pure moral ignorance. True, a religious person often regards religious guidance as moral guidance. But this is not necessarily so. As mentioned before, some people believe that religious law should take precedence over moral law. Moreover, when a person commits an immoral act in pursuit of heaven, their motivations are often not purely moral, but rather theological. Discussing the possibility of a scholar’s having an incorrect interpretation of the Qur’an, Averroes mentions a saying of the Prophet: “This is why the Prophet, peace on him, said, ‘If the judge after exerting his mind makes a right decision, he will have a double reward; and if he makes a wrong decision, he will  have a single reward’” (Fasl al-Maqal).5 This saying was used by Averroes to indicate that an erroneous understanding of religion, if done honestly, can be rewarded as well. A person who does an immoral action in pursuit of heaven would be of similar mind. He would think that if he gets it right, he would be doing the right thing and would go to heaven for that, and if he gets it wrong, he still goes to heaven for his pure heart. Because he aims for eternal happiness, his action is not rationally objectionable.6 A religion, like any other practical theory, can mislead us morally. This is not the intension problem. The intention problem is that the belief in heaven plays a morally problematic role in our practical life. On the one hand, a belief in heaven should have significant practical influence on our behavior; after all, it is irrational of us not to be concerned about our eternal happiness. On the other hand, this influence is morally problematic. It weakens a person’s moral voice by providing excuses for actions for which there should be none. In the event that a person’s religious conviction differs from her moral conscience, the belief in heaven suppresses the moral conscience. To be

238  Amir Saemi clear, one can act morally while believing in heaven. The belief in heaven does not necessarily diminish the moral worth of our moral actions. The problem is that when one’s religious beliefs diverge from one’s moral beliefs, the belief in heaven rationally prevents one’s moral beliefs from playing a role in one’s practical deliberation. According to the intension problem, the belief in heaven trivializes the moral mistakes of sincere religious believers, and thus the belief in heaven can rationally silence one’s inner moral voice. Throughout this section, I discussed the intension problem as a problem for a belief in heaven in general. But the intension problem needs not to be a problem for any conception of heaven; it is a problem for philosophers’ conception of heaven according to which heaven is constitutionally independent of our moral life on earth.7 But this is not the only conception of heaven. For example, Kantian “highest good” consists in “happiness distributed [to persons] quite exactly in proportion to [their] morality (as a person’s worth and his worthiness to be happy)” (Kant, p. 111). In this sense, Kantian heaven conceptually is tied to morality. There is also an important conception of heaven in Islamic tradition, according to which, heaven is simply the spiritual embodiment of our earthy good actions in the afterlife. This conception has its root in some Qur’anic verses such as these: “On the Day when every soul finds all the good it has done presented” (3:30) “On that Day… Whoever has done an atom’s weight of good will see it” (99:7). To the best of my knowledge, Sheikh Bahai is one of the first Islamic scholars who explicitly defends the view that heaven is nothing but the spiritual embodiment of our earthly moral actions uses these verses to support his view on heaven. Space constraints prevent me from going into details about this view of heaven. But as far as the intension problem goes, according to this conception of heaven, heaven is conceptually or constitutionally connected to morality. This conception of heaven, thus, does not silence the voice of moral conscience as it provides no rational excuse to act immorally. But there is an important theological worry regarding this conception of heaven, namely, if heaven is constitutionally dependent on morality, then God cannot allow wrongdoers to go to heaven by forgiving them. Hence, this conception of heaven places serious limitations on God’s power of forgiveness. The believer, thus, faces a dilemma: Either God has the power to forgive sincere wrongdoers and grant them heaven or He doesn’t. If He does, then heaven is not essentially connected to morality, hence the intensions problem. If He doesn’t, this would constitute a serious limitation on God’s power, which is theologically undesirable. Therefore, the view that heaven is constitutionally dependent on morality calls for a new perspective on God’s forgiveness. However, the subject of God’s forgiveness requires another essay. Notes 1 For further discussion of the point, see Lawrence (1993). 2 Moss (2011) has a detailed and plausible discussion of Aristotelian notion of phronesis; see also McDowell (1998) and Irwin (1975).

Islamic Philosophers’ Idea of Heaven  239 3 In differentiating different meanings of ‘intellect’, al-Fārābī explicitly addresses NE VI and says that the practical intellect discussed by Aristotle is a very different than theoretical intellect discussed in metaphysical works of Aristotle (see McGinnis and Reisman, 2007, p. 70). 4 See Fakhry (2004), De Boer (1967), Galston (1990), and Booth (2016). 5 See McGinnis and Reisman (2007, p. 319). 6 In his very helpful comments on this chapter, Mohammad Saleh Zarepour correctly notes that this conclusion is not entailed by the hadith. “For one might say that the hadith is only saying that such a judge will have a single reward for exerting his mind and having the right intention, but he will also be punished for the bad decision he has made. See also verses 18:103-4 of the Quran”. Zarepour is correct in that the conclusion is not a logical implication of the hadith; however, it is based on one interpretation of the hadith, in fact, the interpretation Averroes himself held. Moreover, the Quranic verse immediately following the one Zarepour mentions says that the problem with those who thought their actions were good was their disbelief in their meeting with God. The important point here is that many believers hold that the key to attaining heaven is having a pure heart and sincerely seeking it by doing your best to attain it. It is this prevalent belief which leads to the intension problem. 7 If we add value pluralism as well as the noetic effect of sin to the philosopher’s account of heaven, living morally would be causally necessary for the attainment of heaven. Yet, heaven, conceptually and essentially, remains independent of morality.

References Al-Fārābī, Al-Fārābī on the Perfect State. Translated by R. Walzer. New York: Oxford Univesity Press, 1985, pp. 241–243. Al-Fārābī, The Political Writings, Selected Aphorisms and Other Texts. C. Butterworth. Ithaca, NY: Cornell University Press, 2001. Al-Fārābī, The Political Writings: “Political Regime” and “Summary of Plato’s Laws”. Translated by Butterworth. Ithaca, NY: Cornell University Press, 2015. Al-Ghazālī, The Incoherence of the Philosophers. Translated by M. Marmura. Provo, UT: Brigham Young University Press, 2000. Aristotle, Nicomachean Ethics. ed. by Crisp, Roger. Cambridge: Cambridge University Press, 2014. Avicenna, The Metaphysics of The Healing. Translated by Marmura, M. Chicago, IL: University of Chicago Press, 2005. Berlin, Isaiah, The Crooked Timber of Humanity. Princeton, NJ: Princeton University Press, 2013. Booth, Anthony Robert,  Islamic Philosophy and the Ethics of Belief. London: Palgrave, 2016. De Boer, T.J., The History of Philosophy in Islam. Translated by E.R. Jones. New York: Islamic Philosophy Online, 1967. Fakhry, M., A History of Islamic Philosophy. New York: Columbia University Press, 2004. Galston, M., Politics and Excellence: The Political Philosophy of Alfarabi. Princeton, NJ: Princeton University Press, 1990. Harman, Elizabeth. “Does Moral Ignorance Exculpate?” Ratio 24.4 (2011): 443–468. Irwin, T. “Aristotle on Reason, Desire and Virtue.” Journal of Philosophy 73 (1975): 567–578.

240  Amir Saemi Kant, Immanuel, Critique of Practical Reason. Translated by Werner S. Pluhar. Indianapolis, IN: Hackett Publishing Company, 2002. Lawrence, Gavin. “Aristotle and the Ideal Life.”  The Philosophical Review  102.1 (1993): 1–34. McDowell, J. “Some Issues in Aristotle’s Moral Psychology,” in, Mind, Value and Reality. Cambridge: Harvard University Press, 1998: 23–49. McGinnis, J. and D. Reisman, Classical Arabic Philosophy. Cambridge: Hackett Publishing Company, 2007. Moss, J. ““Virtue Makes the Goal Right”: Virtue and Phronesis in Aristotle’s Ethics.” Phronesis 56.3 (2011): 204–261. Plantinga, A. Warranted Christian Belief. Oxford: Oxford University Press, 2000. Singer, Peter. Famine, Affluence, and Morality. New York: Oxford University Press, 2016. Wiggins, D. “Deliberation and Practical Reason,” in A. Rorty (ed.), Essays on Aristotle’s Ethics. Berkeley: University of California Press, 1980, 221–240.

13 Towards Epistemic Justice in Islam Fatema Amijee

The epistemology of testimony has taken a central place in contemporary philosophical discourse. Amongst the questions that it is concerned with is the following: what are the conditions under which a hearer can learn through testimony? For example, if I were to inform you that the nearest gas station is just around the corner, what would it take for you, as a hearer, to come to learn this information based on my testimony? Plausibly—though not uncontroversially—I must at least know that the nearest gas station is just around the corner,1 I must be someone who does not generally lie about such things, i.e. I must be reliable, and it should perhaps be the case that you believe that I am reliable. Against this background of norms governing the acquisition of knowledge through testimony, we get the potential for an epistemic wrong—a wrong that is done to someone in their capacity as a knower. Suppose that in a board meeting what a woman says is ignored even though she is making a contribution that falls under her area of expertise, and even though she is widely regarded as an expert and thus reliable with respect to her claims in the meeting. That is, suppose that her contribution is ignored simply because she is a woman. In this situation, the woman is wronged in her capacity as a knower, for she is perceived and treated as less credible than she in fact is for reasons that have nothing to do with her credibility. An epistemic injustice occurs when a knower is wronged in this way. Following Fricker (2007), let us isolate two distinct kinds of epistemic injustice: hermeneutical injustice and testimonial injustice. Hermeneutical injustice concerns shared resources for social interpretation, whereas testimonial injustice concerns testimony. My focus in this chapter will be on the latter as it arises in the Qur’an, and particularly in its second chapter The Cow (Surah Al-Baqarah). Scholars often point to the following passage as supporting gender inequality with respect to testimony: O you who have believed, when you contract a debt for a specified term, write it down. And let a scribe write [it] between you in justice. Let no scribe refuse to write as Allah has taught him. So let him write and let the one who has the obligation dictate. And let him fear Allah, his Lord, DOI: 10.4324/9781003327714-20

242  Fatema Amijee and not leave anything out of it. But if the one who has the obligation is of limited understanding or weak or unable to dictate himself, then let his guardian dictate in justice. And bring to witness two witnesses from among your men. And if there are not two men [available], then a man and two women from those whom you accept as witnesses—so that if one of the women errs, then the other can remind her. (2:282, emphasis added)2 The above verse purportedly supports a divine norm according to which a woman’s testimony is worth half that of a man’s testimony in the particular case of witnessing financial contracts, and perhaps across the board. Because both Islamic law and social practices in Muslim societies are often ultimately grounded in the Qur’an, the norm in question has had significant uptake. In several Muslim-majority countries, women’s testimony does not have the legal status afforded to men’s testimony. Moreover, the verse is often used to support inequality with respect to testimony not merely in the case of financial contracts, but in all cases where testimony might be required.3 A divine norm that states that a woman’s testimony should be taken as less reliable simply because she is a woman is in direct conflict with the norms that govern testimonial justice. The norms that govern testimonial justice demand that women should not be treated as less credible simply because they are women. Yet, the divine norm in question seems to prescribe just that. Cases where divine instruction conflicts with other independently held norms are by no means restricted to the case of testimony. What God instructs through the Qur’an or through the prophets is often at odds with what we intuitively feel is right or just.4 However, in the case of testimony, God’s putative instruction not only comes into conflict with but also undermines the norms that govern testimonial justice. This is because when the divine norm that says that a woman’s testimony is worth half that of a man’s testimony is endorsed and internalised by Muslim women who then assert it, testimonial justice seems to require—paradoxically—that we take these women to be less reliable compared to men. Suppose, for example, that a Muslim woman comes to believe that her testimony is not worth as much as a man’s testimony on the basis of the Qur’an. She then goes on to live by that edict, and asserts, when necessary, that that is her view. The norms that govern testimonial justice tell us that we should not disbelieve women simply because they are women. Moreover, a Muslim woman’s belief that her testimony is worth less than a man’s seems to concern a matter of her own values, which are grounded in the broader network of her religious beliefs, and are not a matter of descriptive fact (such as the laws of physics or where the nearest gas station is). As such, there does not seem to be room for a factual error on her part, and so no obvious reason that her testimony should be ignored or taken less seriously. But if we believe her, it follows—from the norms that govern testimonial justice—­that we should take her testimony to be worth less than a man’s

Towards Epistemic Justice in Islam  243 testimony, as per her own claim. The norms governing testimonial justice thus seem to generate a puzzle: to follow them is to not follow them. In some of my previous work,5 I discuss how putatively sexist verses in the Qur’an might be reinterpreted in an egalitarian way. But the puzzle noted above poses a distinct challenge that seems to arise even if there is an available interpretation on which there is no tension between the Qur’an and the norms that govern epistemic justice. To see why, suppose that a feminist interpretation of the Qur’an is available—an interpretation on which the Qur’an does not say that a woman’s testimony is worth less than a man’s testimony. The mere availability of a feminist interpretation of the Qur’an, however, is dialectically ineffective against someone who endorses the inegalitarian, sexist interpretation. Unless a feminist interpretation is the correct one and not merely an available one, a Muslim woman can adopt the sexist interpretation without having made a factual mistake (i.e., the mistake of adopting an incorrect interpretation). I have argued elsewhere (see Amijee 2023) that extant attempts to motivate a feminist interpretation of the Qur’an have not succeeded in showing that it is the correct interpretation, and so we cannot appeal to the feminist interpretation to show that a woman who adopts a sexist interpretation has made a factual mistake. Thus, the Muslim woman’s adoption of a sexist interpretation—and in particular, an interpretation on which a woman’s testimony is worth half a man’s testimony—is enough to generate the puzzle.6 As I will argue, the solution to the puzzle lies in recognising that the application of Qur’anic norms to a given situation requires getting some relevant facts right, and so is not purely a matter of one’s values. As such, asserted divine norms derived from the Qur’an are not subject to the deference characteristically demanded by the norms that govern testimonial justice in cases that concern assertion of value alone. I proceed as follows. I first discuss the tension that arises between divine instruction and the norms that govern testimonial justice in more detail. In Section 13.2, I discuss some unsuccessful solutions to the puzzle and draw some lessons from them about the shape of an adequate solution. In Section 13.3, I show how the apparent tension between divine instruction and the requirements of testimonial justice as well as the puzzle dissolves once we reconceive the relationship between the believer and the Qur’an. I conclude in Section 13.4. 13.1

A Puzzle about Testimonial Injustice

The norms that govern testimonial justice tell us that we should not dismiss or take less seriously the testimony of women simply because they are women. But this is not to say that we should always believe what women say. We should not, for example, simply take the word of a woman who professes to have solved Goldbach’s conjecture. We should see whether her claim is supported by the evidence, whether she is generally reliable when it comes to

244  Fatema Amijee difficult mathematical problems, and so on. To take a more mundane case, suppose that a friend (who is also a woman) tells me that the nearest gas station is just around the corner. But I know from past experience that this friend is directionally challenged and has often—though with good i­ntentions— given me wrong directions. In this case, I should not simply take my friend’s word for where the nearest gas station is located, especially if she is not familiar with the area. Indeed, at the outset of the chapter, I mentioned a number of criteria that may need to be met for a hearer to acquire knowledge through a speaker’s testimony. But if that is right, one may wonder whether a puzzle arises after all. Should we not treat the case where a woman asserts that her testimony is worth less than a man’s testimony on par with the case where a woman asserts that she has solved Goldbach’s conjecture? If we treat the two cases similarly, then we can defuse the puzzle by showing either that the woman who asserts that her testimony is worth less than a man’s is not reliable with respect to the comparative worth of her testimony, or that her claim is falsified by the evidence. The puzzle (as I have sketched it) arises in part because the norms of testimonial justice demand that we should not disbelieve women simply because they are women. But it arises also, and only, on the assumption that there are no other legitimate reasons to disbelieve women who assert that their testimony is worth less than a man’s. In order for this latter assumption to hold, the case where a woman asserts that her testimony is worth less than a man’s testimony must be importantly distinct from the case where she gives directions to the nearest gas station or the case where she claims to have solved Goldbach’s conjecture. In order to see where the difference between the two cases lies, let us distinguish claims that are evaluative from those that are purely descriptive. Claims of the first kind include ‘modesty is the highest virtue’, ‘murder is wrong’, and ‘men are more trustworthy than women’. Claims of the latter kind include ‘men run faster than women’, ‘water is H2O’, and ‘birds lay eggs’. Claims of the first kind are claims that concern value because they involve evaluative terms such as ‘wrong’, ‘right’, ‘better’, ‘higher’, and ‘more trustworthy’.7 By contrast, claims of the latter kind are purely descriptive: they involve no evaluative terms. This distinction between evaluative claims and purely descriptive claims need not be sharp for my purposes. It may be that some claims cannot be classified clearly under either category. Moreover, it may be that some claims appear to be evaluative claims but are in fact descriptive. Take, for example, the claim that men are better than women. But better in what respect? Once the sense of ‘better’ is precisified—for example, better with respect to lifting weights—it becomes clear that a claim that appears to be evaluative is in fact purely descriptive: it is just the claim that men can lift heavier weights than women. Similarly, some claims may appear to be descriptive which are in fact evaluative.8 If the case where a woman asserts that her testimony is worth less than a man’s testimony is a case where she makes a claim that concerns only value

Towards Epistemic Justice in Islam  245 (in the above sense) and not fact, then her claim seems immune to being undermined by claims that are purely factual or descriptive. This is because claims that concern only value are not claims about what is the case in the world. Take, for example, the claim that murder is wrong. It is difficult to see how such a claim could be straightforwardly undermined by pointing to a feature of the world or to a worldly state of affairs. A woman’s assertion that her testimony is worth less than a man’s testimony is moreover not a claim that can be easily undermined by a competing value claim. To see why, suppose that A and B disagree over whether capital punishment is morally wrong. In this type of case, there is no obvious reason why A’s views should be given deference over B’s, or vice versa. But in the case of a woman who asserts that her testimony ought to be valued less than a man’s, the woman’s assertion—at least as it applies to herself—ought to be given deference over a competing assertion. It may be that the woman believes falsely that her testimony is worth less than a man’s, but in a disagreement, and absent any other argument or considerations to the contrary, the woman is in a privileged position to make a claim about her own trustworthiness. The puzzle that I have sketched above thus gets off the ground when we take a woman’s assertion of the divine norm to be a claim that concerns only value and take it to be a claim that demands deference. In the next section, I discuss two ultimately inadequate ways to solve the puzzle before introducing my proposed solution. 13.2

Unsuccessful Solutions

13.2.1 Contextualising Qur’anic Norms

A natural candidate way out of the puzzle generated for the norms that govern testimonial justice is to situate the verse that generates the puzzle in its sociopolitical context and to thus contextualise it. If the verse were contextualised, then it would cease to have normative force for us here and now: instead of a universal and timeless prescription to ɸ, we could take the verse to prescribe ɸ for the particular sociopolitical conditions that obtained in 7th-century Arabia.9 The puzzle would then dissolve, for women who assert that their testimony is worth less than a man’s would be making a factual mistake: they would be mistaking a claim that has force only in the sociopolitical conditions found in the 7th-century Arabia for a claim that has force here and now. This sort of mistake would be factual, for it would be a mistake either about what sociopolitical conditions we currently live in, or a mistake about the content of the prescription. How might one contextualise verse 2:282? Several scholars have endorsed the view that the Qur’an draws a distinction between the worth of a man’s testimony and a woman’s testimony in only one verse—the one quoted above. In four other cases described in the Qur’an that involve the giving of evidence, no distinction is drawn between men and women.10 Thus, the prescription

246  Fatema Amijee at 2:282 should not be taken as a general claim—a claim that holds in all cases—about the relative worth of women’s testimony. Indeed, Barlas (2002, pp. 189–190) argues that the testimony of a woman is in fact privileged over the testimony of her husband in a case where a man charges his wife with adultery. Barlas argues that if the testimony given by two women were equal in value to the testimony provided by one man across the board, there would not be more value attached to a woman’s testimony (relative to her husband’s) in the case of adultery. Moreover, proponents of the above line of argument claim that there is a reason why a distinction is drawn between the value of a man’s testimony versus a woman’s testimony at 2:282. They claim that this particular verse concerns financial transactions, and insofar as women in 7th-century Arabia were not involved in such transactions, their testimony was in fact less reliable. El-Ali (2022, p. 236) writes, for example: [Several scholars] point out that the phrase “so that if one of them forgets/errs the other can remind her” is a reference not to the inferiority of women but rather to the fact that women at the time did not engage in borrowing or lending themselves and were moreover mostly illiterate, making them less effective witnesses then men to a debt contract should a dispute arise. The contextualising strategy, as applied to verse 2:282, is often applied to other verses that seem to run counter to modern sensibilities. One such example is verse 4:2–3, which has widely been taken to sanction polygyny: And give to the orphans their properties and do not substitute the defective [of your own] for the good [of theirs]. And do not consume their properties into your own. Indeed, that is ever a great sin. And if you fear that you will not deal justly with the orphan girls, then marry those that please you of [other] women, two or three or four. But if you fear that you will not be just, then [marry only] one or those your right hand possesses. (4:2–3) As Barlas (2002, p. 190) argues, the purpose of polygyny in 7th-century Arabia—a society that had very few social welfare mechanisms—was to secure justice for female orphans. Understood in its context, the verse does constitute a universal and timeless sanction for polygyny but instead prescribes a very specific context-sensitive solution to a very specific problem. I have argued elsewhere11 that the contextualising strategy does not work as an adequate solution to the problem of reconciling the text of the Qur’an with a feminist outlook. I will argue here that the contextualising strategy also fails as a way of dissolving the puzzle for testimonial justice discussed in the previous section. I will briefly state two ways in which the strategy fails,

Towards Epistemic Justice in Islam  247 and then elaborate the reasoning in more detail. First, the rationale given for the distinction between the worth of a man’s and a woman’s testimony seems inadequate. Second, contextualising renders the Qur’an into a historical document and takes away its normative force—a result that would be unacceptable to any Muslim woman who takes at least some of the Qur’an’s prescriptions to have normative force for us here and now. Importantly for our purposes, it is a result that would be unacceptable to the Muslim woman today who asserts, on the basis of the Qur’an, that her testimony is worth less than a man’s testimony. Let us begin with the first point. Even supposing that it is true that most women did not engage in commercial transactions in 7th-century Arabia, it is far from clear that this rationale justifies the view that a man’s testimony is worth more than a woman’s testimony in the restricted case of a financial transaction or more generally. Even if women in 7th-century Arabia did not engage in borrowing or lending, it is certainly not obviously true that all men did: there were men who likely engaged in no transactions (e.g., slaves or simply inexperienced men) or men who engaged in cash-only or barter-only transactions. The testimony of such men would have been unreliable in the same manner as the testimony of women; yet, there is no caveat or provision attached to the verse. Moreover, it is difficult to believe—especially in the absence of evidence— that no women in 7th-century Arabia engaged in borrowing or lending. Women, after all, ran the household and would have come across (directly or indirectly) transactions that pertained to groceries and other household goods. The general prescription in the verse would have surely unjustly discriminated against these women. The same reasoning applies to the point about literacy: not all men were literate (indeed, the prophet Muhammad was illiterate himself) and some women likely were literate. In the absence of evidence that supports the claim that all men in 7th-century Arabia were literate and engaged in transactions that involved lending and borrowing and that women were illiterate and did not engage in such transactions, the explanation that has been offered for the differential treatment of men and women at 2:282 is too quick. One might argue that even if some women engaged in financial ­transactions and were literate and some men did not engage in such transactions and were illiterate, the general prescription is nevertheless justified for it captures the majority of cases. But while such a prescription might be fine as a matter of human law,12 it seems problematic as a matter of God’s law: it opens up the possibility that a literate woman with knowledge of financial matters can have her testimony unjustly taken less seriously as a matter of God’s decree. But God is just, and so, surely cannot permit such a case. Putting aside the question of whether the rationale that is typically given for the prescription at 2:282 when the verse is contextualised is successful or not, there is a deeper problem with the contextualising strategy more generally: the strategy renders the Qur’an into a historical document with a

248  Fatema Amijee message whose normative force is contingent on the time and place in which the message was revealed. The strategy thus undermines the universal and timeless normative force that Muslims generally take the Qur’an to have. If the Qur’an is contextualised, then no prescription in the Qur’an has normative force for us here and now, because the normative force of the Qur’an is conditional on particular sociopolitical conditions, i.e., those obtained in 7thcentury Arabia. And those conditions do not obtain in our present context. One might argue that the above argument against the contextualising strategy is too quick: surely, we should contextualise only those verses that are problematic (from a contemporary point of view) and take the rest to have normative force for us here and now. But to rely on our own contemporary values in determining which verses to contextualise is problematic, for it undermines the objectivity of the message of the Qur’an. The message of the Qur’an is objective in the sense that (most) Muslims take the Qur’an to have a message that does not depend on who is reading it. In this sense, it is unlike a work of art which can have multiple meanings for different people. Even when Muslims disagree about the correct interpretation, they presuppose that there is a correct interpretation and that other candidate interpretations are false. But if there is a correct interpretation, then letting our background values about a particular domain (say, gender) play a role in determining whether a norm that concerns that domain (i.e., gender) should be contextualised is problematic: the set of verses that I think ought to be contextualised (given my values) will differ from the set of verses that someone else with different values thinks should be contextualised. But we cannot both be right. Thus, if the objectivity of the message of the Qur’an is to be preserved, we cannot let background values that pertain to a given domain influence how we interpret verses that are about that domain. It may turn out to be the case that a ‘value-free’ interpretation (in the relevant sense) is not possible. For instance, it may be that all readers of the Qur’an have background values that pertain to gender that are grounded in culture and tradition. If that is right, it would be impossible to isolate the correct set of verses pertaining to gender that should be contextualised, unless we have a standard that is independent of our values for determining which verses have normative force for us here and now and which verses get contextualised. But the contextualising strategy, as usually employed by feminist exegetes, makes no appeal to such a criterion. Absent an independent criterion that allows us to determine which verses to contextualise, there is no reason why some verses should be contextualised, but not others. The contextualising strategy then threatens to render the entire Qur’an into a historical document—a result that would be unacceptable to any Muslim who takes the Qur’an to prescribe norms that bind us here and now. The strategy thus dissolves the puzzle we began with, but only at the cost of failing to preserve the normative force of the Qur’an for contemporary Muslims.

Towards Epistemic Justice in Islam  249 13.2.2 A Factual Disagreement

The candidate solution to the puzzle discussed above relies on its being the case that Muslim women who assert, on the basis of the Qur’an, that their testimony is worth half that of a man’s testimony are making a factual ­mistake—a mistake either about the sociopolitical conditions we currently live in or a mistake about the content of the prescription. But yet another way out of the puzzle would be available if the prescription in verse 2:282 did not concern purely a matter of value, but also pertained to a matter of fact. In Section 13.1, I drew a distinction between claims that concern value and those that concern fact and argued that the puzzle arises in a straightforward manner only if the prescription at 2:282 is a claim of value. This is both because it is not obvious that the claims of Muslim women are owed deference in a matter that concerns facts about the relative trustworthiness of men and women with respect to testimony and because a factual claim can be shown to be false on the basis of evidence. But what if the prescription at 2:282 does not concern only value? If the divine norm prescribed by the Qur’an concerns or is ultimately grounded in a matter of fact, then we can argue that the puzzle does not arise, for we no longer have a case where two value claims—a divine norm and a norm of testimonial justice—conflict.13 We can reconstrue the two putatively conflicting claims as follows: Norm of Testimony: All else being equal, value a woman’s testimony to the same degree as a man’s testimony. Divine Norm: Value a man’s testimony more than a woman’s testimony because men’s minds are more reliable than women’s minds. That the norm prescribed at 2:282 may have a factual basis is suggested by the verse, whose last line reads ‘And if there are not two men [available], then a man and two women from those whom you accept as witnesses—so that if one of the women errs, then the other can remind her’. A proponent of understanding the divine norm as grounded in a fact about the relative reliability of men’s and women’s minds may reconstruct the argument suggested by the verse as follows: P1. Testimony is valuable only to the extent that it is reliable. P2. Men’s minds are more reliable than women’s minds. C. Men’s testimony is more valuable than women’s testimony. If we understand the norm of testimony and the divine norm as suggested above, then the puzzle does not arise: if men’s minds are indeed more reliable than women’s minds, then all else is not equal, and we need not put a woman’s testimony on par with a man’s testimony.

250  Fatema Amijee However, the Qur’an, as God’s word, is infallible.14 Yet, whether men’s minds are more reliable than women’s minds is a matter we can investigate and show the affirmative answer to be false. It is thus problematic to understand verse 2:282 as grounded in a putative fact—the claim is falsifiable and plausibly falsified. But a solution to the puzzle that impugns God’s infallibility would be unacceptable to Muslims, and thus no solution at all. Yet, one might worry: can such an interpretation of verse 2:282 be avoided? I argue that it can. A decision to adopt one interpretation of a verse over another is a political choice: to take a Qur’anic verse as saying that men’s minds are more reliable than women’s minds—and to take this to be a universal generalisation that holds at all times and in all places—requires that we endorse certain background values that tip the scales towards one reading over another. Thus, what appears to be a purely factual claim is based on a value-laden interpretation, where the relevant values deem women inferior to men. An interpretation that takes the prescription at 2:282 to be a value claim that is grounded in a putatively factual claim is thus not forced on us. 13.3

A New Solution

To make way for a new solution to the puzzle, let us return to the problem I highlighted earlier with the contextual strategy. As we saw, the contextual strategy fails as an adequate solution to the puzzle because it threatens to render the Qur’an as a whole into a historical document whose message does not have normative force for us here and now. What we need is a more sophisticated version of the contextual strategy—a version that allows us to draw a principled distinction between those verses in the Qur’an that bind us here and now, and those that do not. As we have seen, in drawing this principled distinction, we cannot rely on our own background values and cherry-pick the verses that we think should have normative force for us. As I argued above, doing so undermines the objectivity of the message of the Qur’an: insofar as Muslims believe that the Qur’an is the infallible word of God, background values can interfere with the ability of the reader to grasp this message. In particular, if I have background values that pertain to gender (e.g., feminist values) that I then bring to bear when interpreting a Qur’anic verse that itself pertains to gender, I lose the entitlement to take myself to be converging upon God’s message. But I will argue that we do have an independent standard available that enables us to draw a principled distinction between those verses that have normative force for us here and now, and those that do not. To get at this standard, we need to rely on a distinction between two types of verses in the Qur’an: verses revealed in Mecca, where the Muslim community was initially smaller and where the prophet Mohammed started his ministry, and the verses revealed in Medina, where the prophet eventually moved in 622 C.E. and where he established a social order. The Qur’an as a whole consists in entire chapters (surahs) revealed in Mecca or Medina. At the superficial level,

Towards Epistemic Justice in Islam  251 the distinction between Meccan and Medinan verses is simply a ­distinction that concerns where each type of verse was revealed. But a closer look tells us that there is also a distinction in form and content. Meccan verses are shorter, whereas Medinan verses are long and involved. Meccan verses also tend to be prophetic and often refer to the day of judgement and to God’s unity. They, moreover, tend to prescribe virtues. Here is a typical example of a Meccan verse from Chapter 93 (Surah Duha) By the morning light.  And the night as it settles. Your Lord did not abandon you, nor did He forget. The Hereafter is better for you than the First.  And your Lord will give you, and you will be satisfied. Did He not find you orphaned, and sheltered you? And found you wandering, and guided you?  And found you in need, and enriched you? Therefore, do not mistreat the orphan. Nor rebuff the seeker. But proclaim the blessings of your Lord. (93:1–11) By contrast, Medinan verses are often legalistic and seem concerned with establishing laws governing the interpersonal domain, including the relationships between men and women. The Muslim community in Medina was larger and had a more complex demographic than the community in Mecca at the beginning of Muhammed’s ministry. The Medinan verses can be seen as responding to a specific need that arose after Muhammed’s move to Medina—the need to provide a foundation for a new social order. Unlike Meccan verses, Medinan verses often deal with the mundane: questions that concern inheritance, marriage, usury, financial transactions, and the like. Here is a typical example of a Medinan verse from Chapter 4 (Surah An-Nisa) Allah instructs you concerning your children: for the male, what is equal to the share of two females. But if there are [only] daughters, two or more, for them is two thirds of one’s estate. And if there is only one, for her is half. And for one’s parents, to each one of them is a sixth of his estate if he left children. But if he had no children and the parents [alone] inherit from him, then for his mother is one third. And if he had brothers [or sisters], for his mother is a sixth, after any bequest he [may have] made or debt. Your parents or your children - you know not which of them are nearest to you in benefit. [These shares are] an obligation [imposed] by Allah. (4:11) Unlike Meccan verses, Medinan verses tend to prescribe norms that pertain to very specific situations. Importantly for my purposes, the distinction between Meccan and Medinan verses maps onto a distinction between norms that tell us how to be and norms that tell us what to do. An example of a norm that tells us

252  Fatema Amijee how to be is ‘be just’ or ‘be merciful’. By contrast, an example of a norm that tells us what to do is the one specified in the above verse 4:11. The verse provides instruction concerning the division of inheritance. It tells us what to do in a very specific circumstance. Let us call a norm that tells us how to be a ‘thick norm’ because it employs (whether explicitly or implicitly) thick evaluative concepts. Examples of thick evaluative concepts include courageous, cruel, kind and forgiving. By contrast, let us call a norm that tells us what to do a ‘thin norm’ because it employs thin evaluative concepts. Examples of thin evaluative concepts include right, permissible, and bad. This distinction between two types of norms, together with facts about the circumstances in which verses prescribing each type of norm was revealed, can now provide us with a standard independent of our values for drawing a distinction between verses that have universal normative force, and those that do not. The verses that prescribe thick norms have universal normative force— they bind us at all times and in all places—whereas those that prescribe thin norms have a normative force that is conditional on specific social and political circumstances. To see why this is so, we need to look at the form of the prescribed norm, i.e., whether it tells us how to be or what to do. Norms that tell us how to be are universal, for how we ought to be does not differ from time to time or place to place. By contrast, what we ought to do in a given situation does depend on the time and place. Yet, on my view, the Meccan and Medinan norms are not unrelated: the thin Medinan norms inherit their conditional force from being appropriate expressions of what to do in a particular time and place, given the unconditional thick norms expressed by the Meccan verses. The normative priority this proposal gives the Meccan verses rests neither on independent values nor on textual interpretation. The normative priority is forced on us by the philosophical distinction between norms that tell us how to be and norms that tell us what to do. The textual distinction between Meccan and Medinan verses simply serves as evidence that the distinction between these norms is marked in the Qur’an. Does my proposal require that we ought to simply ignore the Medinan chapters, and thus the thin norms that they express? Such a proposal would be problematic, for it would render the Medinan chapters irrelevant and thus dispensable for the contemporary Muslim. But my proposal does not entail that the Medinan chapters are irrelevant to us today: the thin norms serve as a partial guide for how to apply the thick norms. Given information about the context in which the Medinan verses were revealed, we can come to understand how the thin norms expressed by those verses were the best way for people in 7th-century Arabia to implement the thick norms expressed by the Meccan norms. And this in turn enables us to better understand those thick norms. The mature moral understanding which this in turn leads to allows us to properly apply the thick norms to the radically different circumstances we now find ourselves. To see this structure between the thick Meccan norms and the thin Medinan norms at work, let us return to the Meccan verse quoted above (93:1–11)

Towards Epistemic Justice in Islam  253 that enjoins us to not mistreat the orphan. Even though at the ­surface level this prescription tells us what to do (‘do not mistreat the orphan’), the norm instructs us to be good to orphans.15 But what does being good to orphans consist in? Verse 4:2–3 tells us how to be good to orphans in the context of 7th-century Arabia: ‘ And give to the orphans their properties and do not substitute the defective [of your own] for the good [of theirs]. And do not consume their properties into your own. Indeed, that is ever a great sin. And if you fear that you will not deal justly with the orphan girls, then marry those that please you of [other] women, two or three or four. In 7th-century Arabia—a society in which a safety net for the poor was virtually nonexistent—the way to act in accordance with the thick norm that tells us to be good to orphans was to permit men to marry up to four orphan girls, who would then be provided for. But this particular way of caring for orphans and the destitute does not generalise to our context. In our context, perhaps the right thing to do would be to donate to charities or pay higher taxes. Let us now return to the case of testimony. What thick norm might be relevant to the thin norm prescribed at 2:282, a verse that implies that a man’s testimony is worth twice that of a woman’s testimony? Of the virtues prescribed by the Qur’an, justice is one of the central ones. The Qur’an says, for example Surely Allah enjoins justice, kindness and the doing of good to kith and kin, and forbids all that is shameful, evil and oppressive. (16:90) The above verse belongs to Surah An-Nahl, a Meccan surah. With the thick norm prescribing justice in the background, we can ask ourselves what thin norms ought to govern testimony today and here, and in particular, whether there ought to be a difference in how a man’s testimony is weighted compared with a woman’s testimony. Determining what the thin norms that govern testimony are today and here will depend on ascertaining what circumstances obtain today and here, and deciding how these circumstances bear on the derivation of thin norms from thick norms. But we can now say that those who insist that a man’s testimony is worth twice a woman’s testimony today are making a mistake either about the circumstances or about how those circumstances bear on the application of a thick norm. That is, they are either wrong about the social and political conditions that we live in, or they are wrong about what the correct thin norm would be today given the thick norm that prescribes justice. As we have seen, my argument appeals to a crucial difference in the form and content of Meccan versus Medinan verses. This difference was exploited

254  Fatema Amijee by Mahmoud Mohamed Taha in his important work The Second Message of Islam (1987). Taha argues that modern Islamic law ought to be grounded in the Meccan message. Following Taha, Abdullahi Ahmed An-Na’im argues that Meccan message is the true message of Islam and can provide the basis for an Islamic reformation (see, especially, Toward an Islamic Reformation: Civil Liberties, Human Rights, and International Law (1990)). While my project is inspired by the work of both Taha and An-Na’im, it takes an approach distinct from theirs in two fundamental respects: (1) While I appeal to the distinction between Meccan and Medinan verses, the more fundamental distinction that does the work in my argument is the distinction between norms that prescribe how to be and those that prescribe what to do. The distinction between the Meccan and the Medinan messages and the fact that these messages map onto the two distinct types of norms explains why there would be a distinction between two types of norms marked in the Qur’an, even if the mapping turns out to be imperfect (i.e., some Meccan verses may prescribe norms that tell us what to do rather than how to be, just as some Medinan verses may prescribe verses that tell us how to be); (2) Unlike Taha and An-Na’im, I do not take the Meccan message to be the true message of Islam while treating the Medinan message as irrelevant to Islam as practiced today. In my view, both messages are relevant and have an equal claim to being God’s word. Yet, they play distinct roles. While it is the thick norms expressed by Meccan verses that bind us here and now, the thin norms expressed by Medinan verses, along with the distinctive relationship (that I have argued) obtains between the Meccan and Medinan verses, provide us with a blueprint for how to implement Meccan norms today. Let us now return to the puzzle we began with. Recall that the puzzle arises on the assumption that a Muslim woman is making a claim only about value (rather than fact) when she asserts that her testimony is worth less than a man’s. But we now have the resources to say that women who endorse, on the basis of the Qur’an, the claim that their testimony is worth less than a man’s (and then assert it) are making a factual mistake. Moreover, the attribution of this mistake to the Muslim woman does not problematically presuppose that the Qur’an as a whole is a historical document (as I have argued is the case with the contextual strategy discussed above). We thus have a solution to the puzzle that does not threaten the place of the Qur’an for the contemporary Muslim. I close this discussion by briefly addressing an objection to my proposed solution to the puzzle. In my view, the norm prescribed at verse 2:282 is the correct thin norm for 7th-century Arabia, yet not the correct thin norm for us here and now. Yet, we might ask whether the norm prescribed at verse 2:282 is indeed the correct norm for 7th-century Arabia. As we have seen above, the contextual justification that is often provided—that it was men who engaged in financial transactions in 7th-century Arabia, and that women, because ignorant about financial matters, could not witness financial transactions as competently as men—does not quite work. But the norm makes sense against

Towards Epistemic Justice in Islam  255 the background of pre-Islamic society—a society in which women had very little status relative to men and would not have been permitted as witnesses at all. The norm prescribed at 2:282 can be plausibly construed as a steppingstone to an ideal norm. It is not itself ideal, but it is the correct norm for the context of 7th-century Arabia. 13.4

Concluding Remarks

My aim in this chapter has been to address a puzzle that arises for the norms that govern testimonial justice when confronted with a divine norm that supports inequality in testimony between men and women. The puzzle arises when a woman who has internalised the divine norm asserts it. Testimonial justice demands that absent any reason to distrust the woman, we should not treat the woman as less credible simply because she is a woman. Yet, to take the woman seriously paradoxically entails that we take her less seriously. My proposed solution to the puzzle requires that we reconstrue the relationship between the Qur’an and the believer. Instead of a simplistic view on which believers are passive recipients of divine instruction, I have argued that the Qur’an leaves room for us to exercise our own moral understanding. In particular, given the template provided by the relationship in the Qur’an between the thick Meccan norms and the thin Medinan norms, it is up to us to discern what the corresponding thin norms might be for our contemporary context. Against this backdrop of a more sophisticated relationship between the Qur’an and the believer, we can see how a solution to the puzzle that I have discussed for the norms of testimonial justice becomes available.16 Notes 1 Among those who endorse the principle that speakers need to know that p when they assert p in order for a hearer to come to know that p on the basis of their testimony are Audi (2006), Burge (1993, 1997), and Schmitt (2006). Lackey (2006, 2008) and McKinnon (2013, 2015) argue against this principle. 2 All Qur’anic verses in this chapter are quoted from the Sahih International translation of the Qur’an. 3 Cf. Saeed (2018, p. 115). 4 See Amijee (2023) for some other instances where there is tension between contemporary egalitarian values and what the Qur’an seems to prescribe. 5 See Amijee (2023). 6 In Amijee (2023) I express skepticism about whether any feminist interpretation has been shown to be the correct one, rather than merely an available one. The proponent of a competing interpretation that is merely available is likely to be dialectically ineffective against the proponent of the standard interpretation (on the assumption that the standard interpretation is hermeneutically sound), and is left without resources for showing that the standard interpretation should be abandoned. 7 We can draw a further distinction between ‘thick’ and ‘thin’ evaluative terms. See Väyrynen (2021). 8 The fact-value distinction has been problematized by Putnam (2002), amongst others. Challenges to the general distinction between fact and value are, however,

256  Fatema Amijee dialectically irrelevant to my project: if all putatively descriptive facts involve an evaluative component, then the puzzle I discuss for the norms that govern testimonial justice arises straightforwardly. 9 See Amijee (2023) for a detailed discussion of the contextualizing strategy for reconciling putatively sexist verses in the Qur’an with a broadly speaking feminist outlook. 10 See Barlas (2002, p. 190). By contrast, Ahmad (2011) writes that there are seven other verses about ‘recording evidence’ in the Qur’an. 11 See Amijee (2023). 12 Take, for example, the laws against drinking and driving. Not everyone who drinks above the drinking limit and drives will be a dangerous driver, for people differ in their tolerance levels to alcohol. Yet the legal limit applies to all individuals across the board. 13 For the purposes of discussing this candidate solution to the puzzle, I presuppose that the distinction between a claim that concerns value and one that concerns fact is in good standing. Yet, because this is a solution that I will ultimately reject, the burden of showing that the distinction is in good standing lies with those who might endorse this solution. 14 For the purposes of this chapter, I take on board the assumption that the Qur’an is God’s infallible word. This is because this is the predominant view held by Muslims today about the authorship of the Qur’an. However, the question of the origins of the Qur’an has long been contested in scholarly circles. For a recent treatment of this issue, see Soroush (2018), who argues that the author of the Qur’an is not God, but Mohammad, and that revelation is an account of Mohammad’s dreams. 15 Instructions that are explicitly of the what-to-do form, such as the inheritance laws stated in verse 4:11, cannot be similarly rephrased as how-to-be norms. 16 For discussion and feedback on earlier drafts, thanks to Dominic Alford-Duguid, Michael Della Rocca, Mohammad Saleh Zarepour, Imran Aijaz, Rosabel Ansari, and audiences at the University of Michigan-Dearborn and the Kolloquium fur Philosophie in der MENA-Region.

References Ahmad, N. (2011). Comment: Women’s testimony in Islamic law and misconceptions. A critical analysis. Religion and Human Rights, 6: 18. Amijee, F (2023). How to be a Feminist Muslim. Journal of the American Philosophical Association, 9(2): 193–213. An-Na’im, A. (1990). Toward an Islamic Reformation: Civil Liberties, Human Rights, and International Law. Syracuse, NY: Syracuse University Press. Audi, R. (2006). Testimony, credulity, and veracity. The Epistemology of Testimony. Eds. Jennifer Lackey and Ernest Sosa. Oxford: Oxford University Press, pp. 25–49. Barlas, A. (2002) “Believing Women” in Islam: Unreading Patriarchal Interpretations of the Qur’an. Austin: University of Texas Press. Burge, T. (1993). Content preservation. The Philosophical Review, 102: 457–488. Burge, T. (1997). Interlocution, perception, and memory. Philosophical Studies, 86: 21–47. El-Ali, L. (2022). No Truth Without Beauty: God, the Qur’an, and Women’s Rights. London: Palgrave Macmillan. Fricker, M. (2007). Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press.

Towards Epistemic Justice in Islam  257 Lackey, J. (2006). Learning from words.  Philosophy and Phenomenological Research, 73.1: 77–101. Lackey, J. (2008).  Learning from Words: Testimony as a Source of Knowledge. Oxford: Oxford University Press. McKinnon, R. (2013). The supportive reasons norm of assertion.  American Philosophical Quarterly, 50.2: 121–135. McKinnon, R. (2015). The Norms of Assertion: Truth, Lies, and Warrant. Houndmills: Plagrave Macmillan. Putnam, H. (2002). The Collapse of the Fact/Value Dichotomy and Other Essays. Cambridge, MA: Harvard University Press. Saeed, A. (2018). Human Rights and Islam: An Introduction to Key Debates between Islamic Law and International Human Rights Law. Cheltenham: Edward Elgar Publishing. Schmitt, F. (2006). Testimonial justification and transindividual reasons.  The Epistemology of Testimony. Eds. Jennifer Lackey and Ernest Sosa. Oxford: Oxford University Press, pp. 193–224. Soroush, A. (2018) Kalām-e Muhammad, ruʾyā-e Muhammad [Muhammad’s Word, Muhammad’s Dream]. N.p.: Soqrat. Taha, M.M. (1987) The Second Message of Islam. Trans. Abdullahi Ahmed An-Na’im. Syracuse, NY: Syracuse University Press. Väyrynen, P. (2021). Thick ethical concepts. Stanford Encyclopedia of Philosophy.

PART VII

Science and Religion

14 Adam, Eve, and Human Evolution Is There a Conflict? Shoaib Ahmed Malik

14.1 Introduction In 1859, Charles Darwin laid down the initial blueprint for the theory of evolution. Since then, it has been substantiated even further, with several modifications along the way. Without getting technical, the theory of evolution, as understood today, has three core propositions (Fowler and Kuebler 2007, 28–29): 1 Deep time – the age of the universe and the earth is in the time span of billions and millions of years (also known as geologic time). 2 Common ancestry – all biological organisms are connected through a ­historical lineage. 3 Causal mechanics – natural selection and random mutations are the ­primary driving forces of evolution.1 By contrast, most Muslims have and still generally believe that Adam and Eve were created miraculously and are the sole progenitors of humanity (Jalajel 2009; Malik 2021). In other words, humans have nothing to do with evolution. Thus, common ancestry and the creation narrative of Adam and Eve are the bases for the apparent conflict between Islam and evolution. To be clear, there are other apprehensions held by Muslims, but, for the purposes of this chapter, we shall only be concerned with the perceived tension between Islamic scripture, i.e., the miraculous creation of Adam and Eve, and common ancestry. The compatibility between Islam and evolution has been a hotly debated topic (Guessoum 2016). Muslim thinkers have generated ideas which fall all over the spectrum. In my recent work, I divide thinkers into four different positions differentiated by what they accept or reject as part of common ancestry (Malik 2021). Table 14.1 captures the gist of these positions. Creationism and the no-exceptions camp are complete opposites of one another. Creationism is the stance that common ancestry is entirely false, while the no-exceptions camp believes the contrary, and thus permits no special pleading for Adam and humans. Human exceptionalism (hereon referred DOI: 10.4324/9781003327714-22

262  Shoaib Ahmed Malik Table 14.1  Classification of Muslim perspectives on evolution Position

Are Non-Humans a Product of Evolution?

Are Humans a Product of Evolution?

Is Adam a Product of Evolution?

Creationism Human Exceptionalism (HE) Adamic Exceptionalism (AE) No exceptions

✘ ✓

✘ ✘

✘ ✘













to as HE) is the position that common ancestry is true except for human beings. In this perspective, Adam is considered the first member of humankind, and, since he was created miraculously, he and his entire progeny are exempted from the process of evolution. This position is held by Yasir Qadhi and Nazir Khan (2018) and they reconcile it as follows. For them, God miraculously created Adam and Eve at the expected moment when humans were supposed to spring forth through evolutionary pressures. Subsequently, the fossil record appears like a seamless continuation of evolutionary processes that includes human history when looked at scientifically. From a theological lens, however, Adam and Eve’s miraculous creations were real metaphysical events as depicted in Islamic scripture. For Qadhi and Khan (2018), this does not lead to a conflict as such, as these two ideas occupy different disciplinary modalities: “The occurrence of such a scenario is theologically plausible and would be impossible to disprove empirically since it is a metaphysical assertion.” Adamic exceptionalism (hereon referred to as AE) agrees with HE insofar that Adam’s miraculous creation is retained. However, rather than assuming that Adam is the first member of humankind, AE states that Adam and Eve are isolated miracles, and there is nothing in Islamic scripture that explicitly and clearly suggests that they were the first members of humankind (exactly what is meant by humankind here will be discussed in more detail later in this chapter). This leaves open the hermeneutic possibility and compatibility of there being pre/co-Adamic humans/hominins on earth when Adam was created miraculously, which does not contradict Islamic scripture. This position’s validity is argued by David Solomon Jalajel (2009, 2018). The differences between HE and AE are visually depicted in Figure 14.1. Unsurprisingly, HE and AE have received their fair share of criticisms from proponents of creationism and the no-exceptions camp. Creationists generally view evolution unfavourably due to a variety of scientific contentions and theological problems, which we needn’t review here. From their perspective, any attempts of reconciliation between Islam and evolution are always going

Adam, Eve, and Human Evolution  263

Figure 14.1 Differences between HE and AE.

­ o-­exceptions to be seen with suspicion on their part. As for proponents of the n camp, some reject the necessity of holding on to ­miraculous readings of Adam and Eve either due to them believing that miracles are impossible, Adam’s miraculous creation is incompatible with the science of evolution, or because scripture offers potential readings where n ­ on-­miraculous readings can be maintained. Take Nidhal Guessoum, for instance (Bigliardi 2014a). He rejects miracles and believes that events which refer to such incidents should either be understood as moral stories or as lucky events. Accordingly, when scripture does conflict with science, the scriptural data should be reinterpreted non-literally as metaphors in light of the sciences (Bigliardi 2014b). It is then no surprise to see that Guessoum (2011, 478) views Jalajel’s thesis as one that “fully contradicts not only the whole scientific perspective on the subject but tons of empirical evidence as well.” In this chapter, I will argue that, contrary to Guessoum’s evaluation, one can still maintain a miraculous reading of Adam and Eve without it contradicting the science of evolution. To this end, I will provide an integrated ramified argument against Guessoum’s thesis, which I will develop in three sections. First, I will explicate Ashʿarism, which is a Sunnī school of

264  Shoaib Ahmed Malik theology, that I adopted in my previous work and continue adopting here. This, I believe, is important to make explicit, as everyone has a metaphysical framework, whether they acknowledge it or not, that acts as their modus operandi in science and religion discussions. In doing so, I discuss the role of miracles and natural laws in the Ashʿarī framework, which I then couple with a discussion on methodological naturalism. This is to establish that miracles that go beyond scientific measurability or possibility are theological or metaphysical possible realities in Ashʿarism, but if one is using science to deny such possibilities, then, from an Ashʿarī perspective, this would be seen either as a commitment to scientism or philosophical naturalism. Second, I look at certain developments in philosophy of science which I argue provide certain philosophical options when interpreting science and highlight certain limitations with historical sciences like evolution. Third, I use the ideas of Jalajel and Joshua Swamidass, a Christian thinker, to show the proper placement of Adam and Eve in the context of theology and science. Finally, before commencing with the next section, I would like to point out that I defended the theological possibility of creationism, HE, and AE in my previous work. In this chapter, however, I am specifically defending AE from the charge of Guessoum. While the first and second sections are equally applicable to all three positions, the third section only applies to AE. This is not to undermine their validity as such, but I am simply defining the scope of this study. 14.2

Theology, Science, and Miracles

14.2.1 Ashʿarī Theology

Several doctrines can be identified with the Ashʿarī school. However, relevant to evolution (and science more broadly) are the following three principles (Malik 2021): 1 God is a necessary being – everything other than God is (radically) contingent while God Himself is a necessary being who has (a) a will (irāda), (b) power (qudra), and knowledge (ʿilm). 2 God’s omnipotence – God’s creative power applies to whatever is logically possible, i.e., He can do anything that is logically possible. 3 Occasionalism – this is the divine action model that characterises how God interacts with the created world as the ultimate efficient cause. Proponents of the Ashsarī school divide all that exists into what is necessary and contingent. God is the sole, necessary being while everything is r­ adically contingent. These contingencies include physical entities, e.g., the universe, humans, and particles, among others, and metaphysical (supernatural) realities, e.g., angels, demons, heaven, and hell, among others. The Ashʿarīs also stress God’s omnipotence. God’s creative powers aren’t curtailed by any moral, metaphysical, or physical necessities, and are only governed by the laws of logic. In other words, God can only create what is logically possible.

Adam, Eve, and Human Evolution  265 In this kind of framework, God can create worlds that are totally chaotic with no laws at all and worlds with different laws from ours. He can equally create worlds that don’t look designed and are simpler than ours or even more complex than our world. Furthermore, Ashʿarīs believe in an occasionalist DAM in which God is the ultimate efficient cause of all phenomena. No created being can have ontological autonomy outside or beyond God’s power. Using animations as an analogy, God wills each moment to define every single detail from one timeframe to another (Jackson 2009).2 How do these relate to miracles? Given these positions, God wills natural regularities in the created world, which is how we can do science in the first place. If those regularities were not there, science would not be possible. However, in Ashʿarism, God can freely will momentary local events that defy natural regularities. Accordingly, Ashʿarīs have no problem with accepting miracles as actual possibilities in creation. God could very well split the seas of Moses, turn his staff into a snake, split the moon, etc. However, given that our focus is on historical miracles, there are nuances to be made here which will be useful to unpack. It could be argued that what were once miracles can now be explained by the currently known sciences. Take Moses’ staff as an example. It could be suggested that a yet unknown force can explain how the seas split. Further, it could be suggested that some historical miracles for which we don’t have a scientific explanation yet could be explained by future scientific developments that we currently don’t know. The motivation for either of these options might come from placing miracles under the rubric of scientific plausibility, which, in turn, might be to make religion or theology more respectable. Both positions are possible within an Ashʿarī framework. God could have willed miracles into effect in the distant past such that they could have a scientific explanation in the future with the advancements of civilisational knowledge. However, it is also equally possible for God to execute miracles that have no scientific explanation. Given that logical possibility is the defining feature of this theological framework, what God can will is a broader set of possibilities than what is scientifically possible. It is why no scholar of this particular theological school ever denied the reality of divine miracles that have been relayed in Islamic scripture. Therefore, while some miracles could be explained by current or future sciences, there is nothing to rule out events that have no scientific explanation in Ashʿarī theology.3 These positions are summarised in Table 14.2. Table 14.2  Possibilities in Ashʿarism Type of Event

Possible in Ashʿarī Theology?

Natural regularities Historical miracles that can be explained by current science Historical miracles that can be explained by future science Historical miracles that can never be explained by science

Yes Yes Yes Yes

266  Shoaib Ahmed Malik 14.2.2 Science and Methodological Naturalism

Some argue that science is a naturalistic enterprise and thus has no room for supernatural entities such as God or events like miracles in the natural world. In other words, science is antithetical when it comes to religious beliefs. But this may seem strange since many theists have no problem doing science and seeing laws in nature as a form of vindication of God’s wisdom (Malik et al. 2022). To get around this, it will be useful to distinguish between the two types of naturalism. Philosophical, metaphysical, or ontological naturalism (hereon referred to a PN) implies the denial of the existence of supernatural entities (Draper 2005, 279). This denial proposes alternative answers for the inquiries which traditionally occupied the domain of religion, e.g., is there a creator? Are there angels? Is there a life after death? etc. From the perspective of PN, the answers to all these questions are negative because there is only the natural world that is restricted by time and space dimensions, and it is the ultimate reference for seeking any answers. Methodological naturalism (hereon referred to as MN) isn’t as restrictive. MN implies that scientists solely focus on natural phenomena, and they do not try to connect it with the supernatural. In other words, under the umbrella of MN, science is strictly the methodological study of the natural world, leaving aside the question of whether there are any supernatural entities (Draper 2005, 279).4 MN, then, is an epistemic or a methodological position. It is then clear that, by adopting MN, one can be a theist and a scientist while this option is ruled out with PN. Accordingly, however, scientists (and perhaps philosophers of science) want to mark the boundaries and capabilities of science is entirely up to them. What matters theologically, however, is God not being bound by scientific measures. As stated earlier, in the Ashʿarī paradigm, God isn’t restricted to what science can or cannot methodologically determine. The laws of nature are nothing but the manifestation of God’s will that run consistently. Thus, it is these regularities that science is determining as the laws of nature, but nothing is stopping God from being able to change those regularities either locally or globally if He so wished. These points are being stressed because it may be said that the adoption of MN entails a complete rejection of miracles. In practice and theory, miracles could be rejected on scientific grounds. However, to the extent that science is in the business of finding regularities, the irregular, the spontaneous, the anomic [miraculous events] is systematically screened out of scientific laws and theories, and experimental methodologies aimed at finding the regularities will overlook it. (Horst 2014, 342) In other words, science is blind to miracles. So, if a historical miracle did indeed occur, science would neither know nor care. What matters in Ashʿarī

Adam, Eve, and Human Evolution  267 theology is that scientific impossibility doesn’t entail theological impossibility; God can do things beyond scientific realities. But if one sees the world only through the lens of science and cannot accept anything beyond that, then this entails that one is either committed to PN or scientism,5 or at least this is how it would be viewed through the lens of Ashʿarism (Malik 2021, 179–211). 14.2.3 Contentions

It could be argued that Ashʿarī theology is antiscientific (against science), unscientific (not science), or pseudoscientific (poor science). Starting with the first one, Ashʿarism does not deny or is against science. Theologians of this tradition have regularly made remarks about how the knowledge of the structure and regularities of the world allows us to get a glimpse of divine wisdom, and thus see science in a positive light (Malik et al. 2022). However, this theological model also recognises the limitations of science in the context of God’s power, which goes beyond what is scientifically possible or measurable. While scientific plausibility has its place in Ashʿarī theology, it is not circumscribed by it. Accordingly, the charge of Ashʿarī theology being unscientific is, to some extent, true, as it does entertain realities that will not come under the rubric of science; God, a transcendent existence, being one of them! However, “to dismiss this rational intellectual endeavour because it transcends what science can empirically verify is not itself a position demanded by science. It is ideological scientism” (Gregory 2008, 509). Finally, Ashʿarī theology could be critiqued as being pseudoscientific if it was being presented as a poor or unscientific alternative or competitor of science, which is not the case; Ashʿarī theology is not a scientific position; it is a theological position, a meta-framework which grounds both scientific realities and miracles that fall within the power of God. 14.3

Philosophy of Science

14.3.1 Scientific Realism

Much ink has been spilt over the question of scientific realism (hereon referred to as SR) in philosophy of science. As a start, SR is an attitude that requires we believe in our current theories and models as being true or approximately true. This includes both the observable and unobservable components of the world. SR entails different kinds of commitments. These comprise semantic, epistemic, and metaphysical (or ontological) dimensions. Semantically, SR entails that we take scientific equations, models, and theories as being literally true. Epistemically, SR requires a commitment that scientific knowledge constitutes genuine knowledge of the world. Metaphysically, SR requires a commitment to a mind-independent existence that can be investigated by

268  Shoaib Ahmed Malik our sciences. The combination of these commitments furnishes us with high degrees of confidence in our scientific theories (Chakravartty 2017). SR is usually contrasted with scientific anti-realism (hereon referred to as SAR), which comes in various forms (as does SR). One version of SAR is instrumentalism, which maintains that scientific theories are seen as mere instruments that give us some functional utility without any necessary correspondence with reality. According to instrumentalism, science is just in the business of providing us with systematised descriptions of scientific phenomena that can give us calculable or predictive power; unobservable entities like electrons, quarks, strings, and extinct species are seen as mere artefacts of our theories and equations rather than carrying any real ontological or epistemic commitments. As one looks even more deeply into the various positions on the matter, the discussion becomes even more challenging (Chakravartty 2017). Does science uncover things (entity realism) or structures (structural realism)? Can there be a compromise of sorts between SR and SAR (natural ontological attitude)? In short, SAR questions any one of the commitments (semantic, epistemic, and metaphysical) or kinds of therein that advocates of SR adopt. It is why the SR-SAR divide should be less seen as a linear continuum and more as a multidimensional, and thus complex, landscape, with many options on the table (Saatsi 2017). There are many substantive arguments for either side (Wray 2018), but it is not our concern to evaluate the SR-SAR debate here. The primary purpose of this section is to expose a nuance that is currently missing in the discourse of Islam and evolution; namely, proponents who claim that Adam’s miraculous creation undermines the science of evolution seemingly presume that SR is the only option on the table when, in fact, the literature proffers many different positions. This is important to note because if some thinkers adopt a version of SAR, they may not necessarily see evolution to be a problem for Islam to begin with. 14.3.2 Historical and Experimental Sciences

Recently, there have been developments in the philosophy of science discourse in which key thinkers have tried to show how the discussion of SR has been largely skewed by physics, a stock example of experimental sciences. Disciplines like physics are about the “here and now” and analyse lawful phenomena that can be experimentally manipulated, are repeatable, and make predictions (Cleland 2020). The divide between SR and SAR stems from the exotic territories and entities that are studied in physics, e.g., the microphysical world of quarks and bosons, that, even though aren’t directly visible, can be manipulated through which we can infer their properties. For some philosophers of science, the emphasis on physics has skewed the SR-SAR debate because it neglects other sciences that don’t have the same foci, terrain, and epistemic advantages as physics (Turner 2007).

Adam, Eve, and Human Evolution  269 Consider historical sciences such as astronomy, archaeology, geology, and evolutionary biology (Cleland 2020). These disciplines don’t have the advantage of experimental sciences because their objects of study are historical, i.e., buried in the past, which are subject to decay and degradation, some of which might be permanently lost to history. Furthermore, historical scientists cannot manipulate their objects of study like practitioners in experimental sciences can. They can’t observe or manipulate extinct dinosaurs that lived several millions of years ago or look at the first protein assemblies involved in the origins of life in the laboratories, for instance. Moreover, historical events aren’t repeatable like experimental sciences are. For these reasons, some have called for a reassessment of the SR-SAR debate, which is largely preoccupied with experimental sciences (Turner 2007); otherwise, we might end up being radically sceptical of all historical sciences if we take experimental sciences as our absolute benchmark. To be sure, the aforementioned limitations don’t entail that historical sciences are paralysed. Historical scientists use a variety of methods to determine information about the past (Currie 2018). These include simulating informed reconstructions of the past (via experiments or computer simulations); detection and attenuation of historical traces; and using multiple lines of inferences and evidence (consilience of induction) to offer the best or the most coherent explanation of historical entities, e.g., existence and properties of extinct species; events, e.g., a meteorite crashing on earth leading to mass extinction; or processes, e.g., evolutionary mechanisms. Philosophers of science who have written on SR and historical sciences seem to be divided on what kind of attitude we should have towards them. Some are optimistic while others are pessimistic (Turner 2007; Currie 2018; Cleland 2020). However, proponents of either side do largely agree about the epistemic differences between historical and experimental sciences, some of which have been summarised in Table 14.3. Whether or not the differentiation between historical sciences and experimental sciences is representative, needs to be modified, or dropped altogether remains to be seen; but if permitted, then I believe that it potentially adds another layer of complexity to the already multifaceted, and perhaps physicscentred, SR-SAR debate. If historical sciences are indeed more epistemically disadvantaged, then they could be more susceptible to antirealist interpretations. Subsequently, how one philosophically evaluates evolution, a historical science, within or in addition to the broader SR-SAR debate may further impact how seriously they deem it to be problematic for their religious beliefs. 14.3.3 Contentions

There are three contentions that I would like to address. First, it could be argued that considerations of the SR-SAR are a distraction from the conversation, in that it doesn’t or shouldn’t impact the science of evolution. Some

270  Shoaib Ahmed Malik Table 14.3 Differences between experimental and historical sciences (adapted but significantly modified from de Santis 2021) Experimental Sciences Epistemic goal

To understand the repeatable regularities in the world (focus on a single, complex hypothesis; control for extraneous factors) Epistemic possibilities (the Direct and indirect test problem of access of of observations (able information) to interfere and repeat observations) Criterion of acceptability Confirmed predictions

Quality of evidence

Evidence remains intact (can be accessed again)

Methodological emphasis

Laws and regularities

Historical Sciences To understand spatially and temporally restrict events (to bring out a coherent and testable causal nexus) Indirect test of observations (uses traces from the past) Common historical causes; use of consilience of multiple independent lines of evidence; inference to the best or most coherent explanation Historical processes degrade and destroy a great deal of information about the past (e.g., fossilisation and extinction) Historical entities, events, or processes

may even contend that I am perhaps using SAR to undermine the science of evolution. Second, some could argue that, if a sceptic adopts a SAR interpretation of evolution while simultaneously having no problems with flying planes, driving cars, the internet, mobile, etc., then it could be seen as a serious case of scientific schizophrenia; such an interlocutor is holding on to SR for some sciences and SAR for others, which is inconsistent. Third, it could be argued that the preceding discussions might be utilised by certain antievolutionists to strengthen their narrative(s). Indeed, Ken Ham (2017, 31), a famous young-earth creationist, utilises the aforementioned discussion of experimental and historical sciences to argue for the validity of creationism Experimental science is what most people think of when they hear the word science. It uses the method of repeatable, observable experiments. To determine how the creation operates or functions in the present so that we can manipulate it to develop new technologies, find cures for disease, or put a man on the moon. Most of biology, chemistry, and physics, and all of engineering and medical research are in the realm of experimental … science … In contrast, historical science includes historical geology, palaeontology, archaeology, and cosmology. These sciences use reliable eyewitness testimony … and observable evidence

Adam, Eve, and Human Evolution  271 in the present to determine or reconstruct the past, unobservable, ­unrepeatable event(s) that produced the evidence we see in the present. Creationists and evolutionists are looking at the same observable evidence. But the young-earth creationists trust the eyewitness testimony of the Creator, and secular evolutions reject that testimony out of a sinful, rebellious heart. As for the first contention, to bypass the conversation of the SR-SAR debate or not is a methodological choice. Some thinkers may want to bypass the conversation altogether, for whatever reason(s), while others may think it is necessary to investigate the SR-SAR debate to make their respective cases and then see if and how it impacts their approach to Islam and evolution. This is all well and good. However, if one is making a stronger claim that the SR-SAR debate is useless and therefore doesn’t need to be entertained, then this case needs to be made rather than simply assumed. The SR-SAR debate is one of the most important discussions in the philosophy of science, with experts arguing for both sides very rigorously at an academic level, as they believe it has significant implications for science (Chakravartty 2017; Saatsi 2017; Wray 2018). To say that they are wasting their time and why would be important to point out. Furthermore, I would like to make it clear that, while some thinkers may resort to SAR to deflate the scientific efficacy or relevance of evolution altogether, this has not been my approach. In my previous work, I make it very clear that I simply take the science of evolution, Neo-Darwinism, as a given (Malik 2021). In other words, I interpreted Neo-Darwinism through SR, as I wanted to see how much full-blown Neo-Darwinian evolution could be accepted under Ashʿarism. So, that charge is inapplicable to my personal approach to Islam and evolution. On the face of it, the second contention of scientific inconsistency may appear to be true. However, this is only the case if science is seen as one monolithic entity, which it isn’t; there are many scientific disciplines, each one having its own scope, methodology, and language. Therefore, one’s position in the SR-SAR debate is not necessarily a single stamp label. Scientists and philosophers of science may simultaneously occupy a range of positions in different sciences. One could be an instrumentalist for one particular science, e.g., quantum mechanics, while also being a realist for another, e.g., evolution; the philosophical interpretation of a particular scientific domain or theory needn’t be the same as what they think of other sciences. In other words, there is an appreciable difference between one’s local position, i.e., one’s stance on a particular theory, and a global one, i.e., a stance on science in general (Henderson 2017). It is then helpful to see that distinguishing between local and global positions on SR offers a useful distinction without forcing oneself into a simplistic stance. Finally, I will not defend Ham’s position in this chapter, as I am neither a Christian nor a young-earth creationist. However, what I will say is that Ashʿarism does take Islamic scripture very seriously. As a general rule,

272  Shoaib Ahmed Malik Ashʿarīs maintain an agnostic stance on unseen events that took place in the past (Jalajel 2018). In other words, they are theologically neutral about historical events. It is why if one believes in dinosaurs or not has no theological implications.6 However, if God has mentioned certain events in the past in scripture, this is understood as God informing us of what He did or did not do in history and therefore become theologically binding matters. Subsequently, under Ashʿarism, Muslims are required to believe in the prophets and their stories that have been revealed in Islamic scripture, with Adam being no exception (Jalajel 2018). If these events contain miraculous events, they are taken at face value due to the metaphysical backdrop of Ashʿarism discussed earlier. In my earlier work, I make it clear that Islamic scripture has nothing to say about the creation of plants and animals, the age of the universe and earth,7 common ancestry, and the mechanisms of evolution (Malik 2021, 87–105). Accordingly, whether they are true or not has no theological implications. The only exception is a theological commitment to (1) the miraculous creation of Adam, (2) the miraculous creation of Eve, and (3) them being universal ancestors of all human beings alive today due to scriptural references.8 Accordingly, the SR-SAR debate is not being used to undermine evolution in my case. 14.4 Adam and Eve’s Miraculous Creations and the Science of Evolution This final section is a discussion about the proper placement of the existence of Adam and Eve in the context of theology and science and mostly relies on the works of David Solomon Jalajel, a Muslim theologian who we looked at earlier in the introduction, and Joshua Swamidass, a Christian computational biologist. Both argue for the possibility of what I have coined as AE. Their overall narratives lead to the same conclusion, with Jalajel focusing on what can’t be determined theologically and Swamidass focusing on what can’t be determined scientifically. To start with, both argue that the creation account of Adam and Eve are theological propositions that are predicated on scripture, not science: Sunni theologians have always looked at Adam and Eve as a matter of the Unseen which comes under the category of the samʿiyyat, something that is known to us exclusively through scriptural sources without any empirical evidence and accepted by believers on the strength of faith. (Jalajel 2018, 15) The hypothesis that Adam and Eve were de novo created arises from theology, not science, and is not a proper scientific claim … We do not have genetic evidence for or against the Virgin Birth either. The only evidence we have for it is the testimony of Scripture. Whether or not we affirm the Virgin Birth is determined by whether we trust Scripture and

Adam, Eve, and Human Evolution  273 what we think it teaches. Scientific evidence, however, does not tell us one way or another. The same may be true of Adam and Eve. (Swamidass 2019, 29) Jalajel stresses that, when talking about humans, ­theologians are not necessarily talking about the same thing as evolutionary biologists. Islamic scripture refers to humans as insān, bashar, and banī Ādam, which are terms that genealogically link all of us back to Adam and Eve. On the other hand, biologists are talking about biological characteristics that attempt to define biological entities such as Homo sapiens and other hominins, e.g., Homo neanderthalensis.9 These aren’t necessarily and categorically the same thing: Theologians and biologists are asking different questions. Biologists, working within their field, have nothing to say about Adam and theologians, working in their field, have nothing to say about how any biological species evolved. To do so is to transgress the limits of their respective disciplines. (Jalajel 2018, 18) Furthermore, Jalajel argues that neither the question of the biological status of Adam and his immediate descendants, nor the question of interbreeding is discussed in scripture, and therefore Muslims cannot make either position (negation or affirmation) a point of theology: In the absence of any unequivocal textual evidence describing Adam’s earliest descendants in detail, there would be no way to gauge the extent or rate of genetic and phenotypic change that has taken place among Adam’s later progeny. Therefore, scriptural evidence cannot be used by theologians to indicate whether Adam’s earliest descendants would have been classified biologically as Homo sapiens or possibly as some earlier human species. This means that theologians would not attempt answers to questions like whether Homo neanderthalensis were from Adam’s descendants, any more than they could argue whether or not creatures that scientists would classify as being biologically Homo sapiens had already evolved on Earth and were populating it before Adam’s arrival upon it … Consequently, it is not a matter of religious belief to accept or reject the scientific account of hominid evolution up to and including organisms that are taxonomically Homo sapiens. It remains purely a scientific question. From a scriptural standpoint, it is not possible to determine precisely when Adam and Eve made their appearance on Earth nor whether they may have met any pre-existing hominid species. In short, Islamic scripture is silent on (1) when Adam and Eve existed, (2) the biological status of Adam and his immediate descendants, (3) the biological entities that co-existed at the time, and (4) whether they interbred or not

274  Shoaib Ahmed Malik (recall Figure 14.1). Therefore, if interbreeding between Adam’s descendants did occur, it would not undermine scripture. Swamidass (2019, 9–13) makes the same arguments from a Biblical angle but also explains what the science of evolution can and cannot tell us. He makes an important distinction between genealogical ancestry and genetic ancestry. Genealogical ancestry is a connecting thread of familial relations, e.g., grandfather, father, son, and so on. Genetic ancestry is the measure of genetic material traces (DNA) from a history of ancestral lines. Swamidass argues that scripture only discusses genealogical ancestry, not genetic ancestry. To assume that scripture is discussing the latter would be a significantly anachronistic and thus erroneous reading (Swamidass 2019, 41). Thus, to assume that genealogical ancestry and genetic ancestry are one and the same thing would be a mistake. Take, for instance, endogenous retroviruses (ERVs). These are viruses that, if they enter a biological host, can structurally affect the genetic material therein. If they do, these genetic changes can then be passed on to offspring. This is significant because one piece of evidence used for establishing the common ancestry of humans and chimpanzees is identical genetic changes caused by ERVs found in both species (Finlay 2013, 21–69). Getting to the point, while we typically identify our fathers and mothers as part of our genealogy, we wouldn’t put viruses in our family

Figure 14.2 An illustration of genetic dissipation.

Adam, Eve, and Human Evolution  275 tree (Swamidass 2019, 36). So, there is an important distinction between the two ancestries. With this distinction cleared, Swamidass argues that developments in the science of human evolution reveal important insights for the discussion. Without getting too technical, three points are of significance. First, he points out that genetic ancestry and genealogical ancestry work very differently. Genetic ancestry is diluted across generations whereas genealogical ancestry increases exponentially (see Figure 14.2). In simpler words, genetic material dissipates over time as one moves forward from a genealogical ancestor, whereas genealogical relationships increase exponentially (Swamidass 2021, 48). Second, and following from the first point, determining the genealogical ancestry of distant ancestors is virtually impossible from genetic ancestry. If a person existed several hundred years ago, he might have several descendants today, but his genetic material will be so thinly dispersed throughout the successive generations that his original genetic material would not be retraceable. According to Swamidass, the original genetic material of ancestors from more than 15 generations ago (250–350 years) is irretraceable: ­ escendants Detectable ancestors must (1) pass DNA to at least some d that live today, and (2) this DNA must be identifiable as coming from them. Genealogical ancestors in the distant past, however, are only rarely genetic ancestors; they usually leave their descendants no DNA. As one study explains, commonly universal ancestors are genetic ghosts who leave DNA to only some of their descendants, not all. Many of our ancestors are genetic super-ghosts ‘who are simultaneously (i) genealogical ancestors of each of the individuals at the present, and (ii) genetic ancestors to none of the individuals at the present.’ Even when they do leave us DNA, we usually cannot identify this DNA with specific individuals in the past. Most of our genomes are identical to other humans, even though we inherit DNA from different sources … This is a critically important point. Since most of our ancestors leave us no DNA at all, let alone identifiable DNA, genealogical relationships are ‘essentially unobservable’ in genetic data past about fifteen generations. (Swamidass 2019, 50–53) Third, the genetic data reveals that there are, in fact, many universal ancestors of all humans today (Swamidass 2019, 49–51). Given these three points, Swamidass argues that all humans today can still believe in Adam and Eve as their universal genealogical ancestors, but they aren’t exclusively their genetic ancestors (Figure 14.3). To argue that Adam and Eve are exclusively all modern-day humans’ universal genetic ancestors would be anachronistic. Like Jalajel, Swamidass believes that scripture is silent on what co-existed at the time of Adam and Eve’s creations. Therefore, if there were co-Adamic entities, they could have interbred with Adam’s descendants, thus fostering a genetic ancestry that goes back to Adam and Eve and one that

276  Shoaib Ahmed Malik predates them, too, all while maintaining universal genealogical ancestry for Adam and Eve (Swamidass 2019, 41). More importantly, this can be maintained even if Adam and Eve were created miraculously (Swamidass 2019, 81): There is no evidence for or against Adam and Eve, ancestors of us all. Genetic ancestry gives us a tunnel-vision view of the past. Genetics is a streetlight, revealing our immediate genealogical ancestry, but not much more. Genetics is a telescope, giving us a powerful view of our history.10 Adam and Eve, however, fall in a gaping blind spot, hidden from our view. If they are real people in a real past, we expect them to be universal ancestors, but we cannot find evidence to prove or disprove them. They could have been de novo created outside our view. These findings are significant two key reasons. First, modern human beings have several universal ancestors of which anyone of them could be Adam and Eve. Given that we don’t know when the scriptural Adam and Eve existed, i.e., Islamic scripture is silent on the question, and what their biological traits were, i.e., Islamic scripture makes no mention of their genotypic

Figure 14.3 A phylogenetic tree showing the possibility of multiple universal genealogical ancestors, of which one could be Adam while still being created miraculously (same goes for Eve).

Adam, Eve, and Human Evolution  277 and phenotypic traits, we cannot specifically pinpoint their existence in the phylogenetic reconstructions.11 Second, since our genealogical relationships are empirically invisible after fifteen certain generations, Adam and Eve are in a scientific blind spot, and so we have no way of scientifically making the case for or against their miraculous creations. As stated by Swamidass (2019, 79) Paradoxically, Adam and Eve could have been created from the dust and a rib, without parents, and at the same time we all could also share common ancestry with the great apes. These two facts are not in conflict. There is no evidence against the de novo creation of Adam. Our genomes would appear as if we shared ancestry with the great apes, because … genomes would record the true story of everyone outside the Garden. The genetic evidence that we descend from Adam and Eve, however, would be lost long ago. For this reason, there is no genetic evidence for or against the de novo creation of Adam and Eve. So, when Guessoum (2011, 478) says that the miraculous creation of Adam and Eve, “fully contradicts not only the whole scientific perspective on the subject, but tons of empirical evidence as well,” not only is this inaccurate, but it is also false. Science can neither prove nor deny Adam and Eve’s miraculous creations given our empirical evidence.12 14.5 Conclusion At the beginning of this chapter, I relayed the opinion of Guessoum, who believes that the miraculous creations of Adam and Eve contradict the science of evolution. For a thorough analysis, I divided my response into three parts. In the first part, I argued that, from an Ashʿarī perspective, miraculous events, which are understood as events that go against the laws of nature, are entirely possible. With the stress on the radical contingency of the world, the characterisation of God’s omnipotence being defined by logical possibilities, and occasionalism as its divine action model, miracles are possible realities. Historical miracles can have scientific explanations later on, but it is equally possible for miracles to have no scientific explanations. When looked at from a scientific perspective, MN seems to be the limiting factor of what makes something scientific. However, Ashʿarism permits realities that go beyond what is scientifically possible or measurable. But to take what is scientifically possible or measurable as the ultimate benchmark for assessing theological claims would be seen either as a commitment to scientism or PN from the lens of Ashʿarism. In the second part, I looked at the discussion of SR-SAR. The purpose of that section was to show that there are many philosophical options on the table that thinkers could entertain when thinking about Islam and evolution. Furthermore, the recent developments in distinguishing between experimental and historical sciences reveal that there could be important asymmetries

278  Shoaib Ahmed Malik between the two, with the historical sciences facing certain epistemic ­limitations that the experimental sciences don’t. As a reminder to the reader, though some could use SAR to undermine evolution, I do not entertain the SR-SAR debate or the differences between historical and experimental sciences to deflate the science of evolution. As I stated earlier, in my previous work, I viewed evolution through the lens of SR to see the full impact of evolution on Ashʿarī theology. I am simply highlighting philosophical nuances that I think need to be acknowledged for carrying forward further critical conversations in the area of Islam and evolution. I also believe that a broader conclusion can be drawn from this. In science and religion circles, localised interfaces between science and religion, e.g., Islam and evolution, are referred to as activities of science-engaged theology (SET). I believe that we equally need a philosophy of science-engaged theology, which is an area that needs to be developed on its own terms and how it could intersect with and impact SET. In the third section, I demonstrated that, through the ideas of Jalajel and Swamidass, Adam and Eve’s miraculous creations cannot be negated through our scientific evidence. Both correctly recognise that Adam and Eve’s miraculous creations are theological propositions. Furthermore, from a theological standpoint, we don’t know when they existed, what their biological properties were, what co-existed at the time of their miraculous creation on earth, and whether their descendants interbred with other potentially co-existing humans/hominins, as scripture is silent on those details. From a scientific angle, our genetic reconstructions reveal that our universal geological ancestors in the distant past cannot be traced due to genetic dissipation. In other words, while we know that Adam and Eve are our universal genealogical ancestors from scripture, their existence cannot be proved nor disproved through our genetic ancestry. These points hark back to the epistemic limitations of historical sciences as opposed to the experimental sciences discussed previously. Putting it all together, there is no scientific evidence for or against Adam and Eve’s miraculous creations. Subsequently, I conclude that, contrary to Guessoum’s observation, Adam and Eve’s miraculous creations do not necessarily undermine the scientific evidence for evolution; there are theological and philosophical considerations that could be entertained when looking at the conversation. But even if one wanted to bypass the discussions of theology and philosophy of science, the scientific evidence against Adam and Eve’s miraculous creations remains surprisingly moot. Notes 1 These specific causal mechanics give evolution a Neo-Darwinian rendition. Currently, there is a debate amongst evolutionary biologists on whether these causal factors are sufficient. The specifics and outcome of this debate are irrelevant for our purposes. 2 More details can be found in Malik and Muhtaroglu (2022).

Adam, Eve, and Human Evolution  279 3 For a full exposition of an Ashʿarī perspective in light of science, see Malik and Kocsenda (2024). 4 There is a well-known distinction between intrinsic methodological and ­pragmatic methodological naturalism in the literature. I have discussed and argued elsewhere that both are compatible with Ashʿarism (Malik 2021, 179–211). 5 I loosely define scientism as an ideology through which everything is measured by science. This should be distinguished from science as a methodology through which we attempt to understand the world. See Stenmark (2001) for a more thorough exposition of scientism. 6 By this I mean that either affirming or negating dinosaurs has no eschatological consequences. The function of theology is to provide its proponents with a set of beliefs for salvific ends. 7 The reference of six days in Islamic scripture is semantically elastic, and therefore could be any six durations of time and not necessarily six period of 24-hour days (Malik 2021, 90–91). 8 Muslim also believe in Jesus’ miraculous creation, but this point is irrelevant to this work. 9 On this point, see Kemp (2011). 10 The streetlight and telescopic metaphors are, in effect, the first two points that were discussed earlier (Swamidass 2019, 35): “Genetic ancestry is not ­genealogical ancestry. Genetic ancestry, nonetheless, gives us a view of genealogical relationships. The genetic view of our past is like a streetlight and a telescope. In some ways, genetic ancestry is like a streetlight, illuminating genealogical relationships in our immediate vicinity. Close relationships can be inferred from patterns in DNA, often with very high confidence. DNA tests can determine if a child’s father is her biological parent with very high accuracy, or if two children are siblings. Just a few centuries back, the genealogical relationships of individual people are outside the streetlight, much more difficult to determine from genetic information. In other ways, genetic ancestry is like a telescope, giving us information about populations in the very distant past. Information in this telescope, for example, causes most scientists to conclude that humans and chimpanzees share common ancestors, with very high confidence. In the intermediate range, genetics reveals an otherwise invisible history of population movements and migration over the last 10,000 years. Telescopes, however, block our peripheral vision. Genetic ancestry tells us about populations in the distant past, but very little about particular individuals within larger populations. Looking through a telescope, I might see the moons of Saturn and its rings, but I will be blind to my wife standing right next to me, and just about everything else” (the italicized are my own emphasis). 11 It should be noted that Swamidass (2019, 46) does make it clear that the earliest common ancestor could exist as early as 3,000 years ago. 12 Interestingly, Swamidass (2019, 79–80) shares how even atheist scientists concede this point: “This finding is a secular claim about evidence, which does not require belief in a de novo created Adam and Eve to affirm. For example, several atheist scientists participated in the workshops, and all agreed there was no evidence against a couple specially created within a larger population. They agreed with this evidential conclusion even though they personally do not think Adam and Eve were real, nor do they believe that God exists.”

References Bigliardi, Stefano. 2014a. “Above Analysis and Amazement: Some Contemporary Muslim Characterizations of ‘Miracle’ and Their Interpretation.” Sophia, 53: 13–129.

280  Shoaib Ahmed Malik Bigliardi, Stefano. 2014b. Islam and the Quest for Modern Science: Conversations with Adnan Oktar, Mehdi Golshani, Mohammed Basil Altaie, Zaghloul El-­Naggar, Bruno Guiderdoni and Nidhal Guessoum. Istanbul: The Swedish Research Institute. Chakravartty, Anjan. 2017. “Scientific Realism.” The Stanford Encyclopedia of Philosophy. Accessed 19th of August 2020. Available at: https://plato.stanford. edu/ archives/sum2017/entries/scientific-realism Cleland, Carol E. 2020. “Is It Possible to Scientifically Reconstruct the History of Life on Earth? The Biological Sciences and Deep Time.” In Kostas Kampourakis and Tobias Uller, eds. Philosophy of Science for Biologists. Cambridge: Cambridge University Press, 193–215. Currie, Adrian. 2018. Rock, Bone, and Ruin: An Optimist’s Guide to the Historical Sciences. Cambridge, MA: The MIT Press. de Santis, Marcelo D. 2021. “Misconceptions About Historical Sciences in Evolutionary Biology.” Evolutionary Biology, 48: 94–99. Draper, Paul. 2005. “God, Science, and Naturalism.” In William J. Wainwright, ed. The Oxford Handbook of Philosophy of Religion. Oxford: Oxford University Press, 272–303. Finlay, Graeme. 2013. Human Evolution: Genes, Genealogies, and Phylogenies. Cambridge: Cambridge University Press. Fowler, Thomas, and Daniel Kuebler. 2007. The Evolution Controversy: A Survey of Competing Theories. Ada, MI: Baker Academic. Gregory, Brad S. 2008. “No Room for God? History, Science, Metaphysics, and the Study of Religion.” History and Theory, 47(4): 495–519. Guessoum, Nidhal. 2011. “Islam and Biological Evolution: Exploring Classical Sources and Methodologies by David Solomon Jalajel.” Journal of Islamic Studies, 22(3): 476–479. Guessoum, Nidhal. 2016. “Islamic Theological Views on Darwinian Evolution.” ­Oxford Research Encyclopaedia of Religion. Accessed 1st of January 2020. Available at: https://oxfordre.com/religion/view/10.1093/acrefore/9780199340378.001.0001/ acrefore- 9780199340378-e-36 Ham, Ken. 2017. “Young Earth Creationism.” In J. B. Stump, ed. Four Views on ­Creation, Evolution, and Intelligent Design. Grand Rapids, MI: Zondervan, 17–48. Henderson, Leah. 2017. “Global Versus Local Arguments for Realism.” In Juha Saatsi, ed. The Routledge Handbook of Scientific Realism. Abingdon: Routledge, 151–163. Horst, Stephen. 2014. “Miracles and Two Accounts of Scientific Laws.” Zygon, 49(2): 323–347. Jackson, Sherman. 2009. Islam and the Problem of Black Suffering. Oxford: Oxford University Press. Jalajel, David Solomon. 2009. Islam and Biological Evolution: Exploring Classical Sources and Methodologies. Western Cape: University of the Western Cape. Jalajel, David Solomon. 2018. “Tawaqquf and Acceptance of Human Evolution.” Yaqeen Institute. Accessed 1st of January 2020. Available at: https://­yaqeeninstitute. org/dr-david-solomon-jalajel/tawaqquf-and-acceptance-of-human-evolution/#. Xgw_HxczbPA Kemp, Kenneth. 2011. “Science, Theology, and Monogenesis.” American Catholic Philosophical Quarterly, 85(2): 217–236. Malik, Shoaib A. 2021. Islam and Evolution: Al-Ghazālī and the Modern Evolutionary Paradigm. Abingdon: Routledge.

Adam, Eve, and Human Evolution  281 Malik, Shoaib A. and Karim Kocsenda. 2024. “What Is a Miracle? An Islamic (Ashʿarī) Perspective on Prophetic Miracles.” Journal of Islamic Philosophy, forthcoming. Malik, Shoaib A. and Nazif Muhtaroglu. 2022. “How Much Should or Can Science Impact Theological Formulations? An Ashʿarī Perspective on Theology of Nature.” European Journal of Analytic Philosophy, 18(2): (SI8)5–35. Malik, Shoaib A., Hamza Karamali, Moamer A. Khalayleh. 2022. “Does Criticizing Intelligent Design (ID) Undermine Design Discourse in the Qurʾān? A Kalāmic Response.” Zygon: Journal of Religion and Science, 57(2): 490–513. Qadhi, Yasir, and Nazir Khan. 2018. “Human Origins: Theological Conclusions and Empirical Limitations.” Yaqeen Institute. Accessed 19th of August 2020. Available at: https://yaqeeninstitute.org/nazir-khan/human-origins-theological-conclusionsandempirical-limitations/ Saatsi, Juha, ed. 2017. The Routledge Handbook of Scientific Realism. Abingdon: Routledge. Stenmark, Mikael. 2001. Scientism: Science, Ethics and Religion. Abingdon: Routledge. Swamidass, S. Joseph. 2019. The Genealogical Adam and Eve: The Surprising Science of Universal Ancestry. Downers Grove, IL: IVP Academic. Turner, Derek. 2007. Making Prehistory: Historical Science and the Scientific Realism Debate. Cambridge: Cambridge University Press. Wray, K. Brad. 2018. Resisting Scientific Realism. Cambridge: Cambridge University Press.

15 Islamic Occasionalism within the Context of Modern Religion and Science Debate Ozgur Koca

15.1

Islamic Occasionalism1

Muslim philosophers, theologians, and mystics elaborated an array of ­theories on causality. A closer study of these theories allows us to identify and explore certain major trends among them. Islamic occasionalism is one of these major trends.2 Islamic occasionalist accounts often claim that finite beings do not have causal efficacy. God creates both cause and effect and attaches them to each other on a self-imposed habitual pattern. The consistency of natural processes is due to these divine habits (ʿādat). The observed consistency in the natural processes, however, does not prove necessity. There is no necessary connection between cause and effect; there is only constant conjunction. The development of occasionalist accounts in the Islamic intellectual history is closely linked to discussions taking place in the early period on the relationship between the divine attributes and God, between the divine will and human freedom, and between the Qur’anic emphasis on divine sovereignty and human accountability. The accentuation of the divine will and freedom leads to denial of any type of necessity in God or in the world. These claims are then supported with a physical atomistic ontology. Muʿtazilite theologians—with the possible notable exception of Jubbāʿī— believe in a necessary relationship between cause and effect (Ashʿarī 1963/1382, 412; Juwaynī 1969, 506). When cotton and fire are brought together, a combustion necessarily occurs. Every object has a “nature” (ʿabʿ) and acts necessarily in accordance with it. When a stone is cast into the sky, it necessarily falls back down (Ashʿarī 1963/1382, 314). Things such as fire and water have specific natures which do not change (Baghdādī 1928, 314; Shahrastānī 1961, I, 75). From a wheat seed, barley never comes out, because wheat has a specific nature that prevents it from becoming barley. A heavy object cannot levitate in the air without any support. God does not create things in contradiction to their natures or without any cause. To say the opposite is to defend absurdity in the divine acts. Natures and causality are then the principle of intelligibility of the world and God. Despite this strong tendency among Muʿtazilites, Ashʿarites reject the idea of natures due to its necessitarian implications. God can prevent fire from DOI: 10.4324/9781003327714-23

Islamic Occasionalism  283 appearing even if we bring fire and cotton together. It is not their natures but God who causes burning (Ibn Fūrak 1987, 271). An early Ash’arite theologian Bāqillānī presents one the clearest expressions of this theory where he criticizes Muʿtazilites for being “people of nature” (ahl al-ʿibāʿ). People argue that they know that there is a necessary relationship between fire and burning or drinking (alcohol) and drunkenness. This, however, is great ignorance. For all we observe here is that when someone drinks alcohol or an object is brought near fire, there will be some changes. That person will be drunk and that object will burn. However, we do not observe who exactly is the agent here. This problem can be understood through a meticulous research and careful thinking. We are of the opinion that this is the act of an Eternal Being… Some also argue that it cannot be known whether this relationship between drinking and drunkenness or fire and burning is due to the natures of entities or to an external agent. (Bāqillānī 1987, 62) The major concern here is to save the divine will and freedom from any implications of necessity. Or, as Ibn Fūrāk asserts, God creates “without a reason (sabab) that makes it necessary or a cause (ʿilla) that generates it” (Ibn Fūrak 1987, 7).3 The rejection of natures invites certain questions. If regularity in the world is not caused by the nature of entities, then how do we explain it? Here, Muslim occasionalists introduce the notion of “habit” (ʿāda). The notion describes divine acts that occur consistently and are repeated regularly. According to the theory of habit, God acts on a freely chosen, self-imposed habitual paths (Ibn Fūrak 1987, 176).4 The regularity and predictability of natural phenomena are due to the habits of God, not to the natures of things. The concept of habit allows Ashʿarite theologians to affirm the regularity of the world. God, however, is not bound even by these self-imposed habits. This would be contrary to God’s absolute freedom. In a world where the notion of necessity is somehow applicable to God, the divine freedom is undermined. Therefore, in some cases, the habits of God could change (Qāḍī ʿAbduljabbār 1962–1965, XV, 202; Juwaynī 1985, 61; Gimaret 1990, 459–463). As Ibn Fūrak writes, God creates satiety after eating and drinking. And He creates thirst and hunger in the absence of food and drink…However, if He wishes, he could do it the other way. This would be a nullification of the habits (naqḍ al-ʿāda). (Ibn Fūrak 1987, 134). The same logic applies to mental causation. Ashʿarītes do not see a necessary relationship between contemplation (naẓar) and knowledge (ʿilm).

284  Ozgur Koca God can create knowledge without contemplation. The relationship is one of proximity. God could also choose not to create knowledge despite the existence of contemplation. Contemplation does not necessarily generate (tawlīd) knowledge. Thus, cognitive causal relations are also possible. God sustains this relationship due to His habit (Juwaynī 1969, 110–115; 1985, 27–28).5 Everything—except some logical impossibilities—is possible with God’s power. The cause-and-effect relationship is then one of possibility and not of necessity. Then, to explain the intrinsically possible nature of causal processes, Ashʿarite occasionalism introduces some key concepts, such as conjunction (iqtirān) and proximity (mujāwara). For Bāqillāni, for example, cause and effect are coupled together and happen regularly in close proximity. This creates a certain disposition in mind that they will always do so (Bāqillānī 1958, 50). This proximity however does not imply necessity. There is only conjunction without necessary relation. Any type of necessity would undermine the divine will. For Juwaynī as well, the relationship between consecutive causal events must be perceived in terms of proximity and not of necessity. God ­creates cause and effect in close vicinity. Coupling cause and effect together is, again, a divine habit (Juwaynī 2003, 219). Hence, the concepts of conjunction and proximity ground the idea that despite observed regularities in the world, causal relations are characterized by possibility rather than necessity. The natural processes are law-like but not strictly law-governed. The intrinsic nature of the world process is based on the notion of possibility. The regularity of natural processes is not necessary but only possible. Thus, such processes can be nullified. It is possible that a stone can levitate in the air, seeing can occur without light, satiety without food, burning without fire when God suspends His habits. Prophetic miracles which violate the laws of nature are, in fact, manifestations of the possible nature of causal relations. It is possible that the divine habit is nullified to verify a prophet in a similar fashion to a king’s unusual gesture to verify the authenticity of his messengers. As Ibn Maymūn (Maimonides) writes, “a king’s habit is to ride a horse through the market place…; but it is possible that he walks through” (2002, 128). In accordance with this theology, Ashʿarites lean toward an atomistic cosmology in which atoms and accidents are re-created at each moment. Ashʿarite atoms do not subsist by themselves and are continuously created in time. God assigns the accident of “subsistence” (baqāʾ) to atoms to continue their existence. The divine command “continue to exist!” assigns the accident of duration and continuity to atoms. Both atoms and the accidents inhering in them are constantly re-created at each moment by the divine will and power (Ashʿarī 1963/1382, 359–360; Juwaynī 1985, 41; Bāqillānī 1987, 38–39; Ibn Fūrak 1987, 205).6 This leads to the idea of the constant re-creation of the world which keeps the moments of the cosmic history discrete. The Ashʿarite world continuously pulsates between existence and non-existence. There is no causal glue

Islamic Occasionalism  285 between consecutive events in the world. The only thing that connects past, present, and future is the divine will and power. The relationship between two consecutive moments is, therefore, not necessary but possible. As a stone moves in the air, its atoms and accidents are re-created anew at each moment in different locations. The fact that this motion can be studied in light of mathematical formulas does not imply necessity, though it suggests that God creates in a predictable manner. If both time and space are continuously recreated, one need not posit a necessary causal connection between two consecutive events. In later periods famous Ghazālī offers one of the most succinct summaries of early Ashʿarite occasionalism: “The connection between what is habitually believed to be a cause and what is habitually believed to be an effect is not necessary according to us” (Ghazālī 1997, 166). Connection between cause and effect “is due to God’s decree, Who creates them side by side, not to its being necessary in itself, incapable of separation” (Ghazālī 1997, 166). Observation shows only concomitance—not any necessary connection between cause and effect. God can create an effect without its habitual cause, such as satiety without eating. Cotton can transform into ashes without contact with fire. Constant conjunction of cause and effect does not prove a necessary connection. One can say effect exists with cause, but one cannot say effect exists by cause. “All the acts of His servants are His creation, connected with His power” (Ghazālī 1992, 21). Moreover, all temporal events, their substances and accidents, those occurring in the entities of the animate and the inanimate, come about through the power of God, exalted be He. He alone holds the sole prerogative of inventing them. No created thing comes about through another [created thing]. Rather, all come about through [divine] power. (Ghazālī 1994, 4.314–315) If a person who is “blind from birth and has a film on his eyes and who has never heard from people the difference between night and day were to have the film cleared from his eyes in daytime,” he would believe that “the opening of his sight is the cause of the apprehension of the forms of the colors.” But when the sun sets and the atmosphere becomes dark, he would then know that “it is the sunlight that is the cause for the imprinting of the colors in his sight” (Ghazālī 1997, 168). Observation cannot locate any necessitating connection between cause and effect. All we observe is constant conjunction. 15.2

Islamic Occasionalism and Science: Merits and Challenges

One of the greatest merits of Islamic occasionalism is that its framework entails several presuppositions that overlap with the guiding principles of scientific inquiry. Namely, an occasionalist can argue, without contradicting the internal logic of the theory, that interactions in the world are regular,

286  Ozgur Koca physical occurrences are lawful, and physical calculations and predictions are ­possible. The consistency of natural processes is a basic concept in Islamic occasionalism, as the notion of habit suggests. Although occasionalism denies the necessary connection, it affirms constant conjunction in a way that can secure the regularity and uniformity of cause-and-effect interactions. The denial of causal necessity in things does not necessarily imply chaotic happenings in the world. The theory explicitly states that God creates on a selfimposed habitual pattern and does not rule over the world in an arbitrary manner. If this is the case, as Griffel observes, to the extent that there is no break in God’s habits, the world of Islamic occasionalism remains indistinguishable from a universe governed by physical laws (Griffel 2016). But what about miracles? How can an occasionalist deal with the possibility of breaks in God’s habits without undermining the guiding principles of natural sciences? Ibn Rushd has already insisted that the moment one denies the necessary connection in causality, one cannot make any reliable predictions about the world. As he puts “denial of cause implies the denial of knowledge, and denial of knowledge implies that nothing in this world can be really known” (1954, 318). In my view, Islamic occasionalism can offer three possible solutions to this challenge. 1. The first is to marginalize extraordinary events to an extent that the ­normal operation of the laws of nature remains the central explanatory framework for the physical world and to appeal to the supernatural, as William Dembski suggested elsewhere, only when there are very strong reasons to believe that “empirical resources are exhausted” (1994, 132). In fact, Ghazālī himself proposes a similar solution. Anticipating Ibn Rushd’s criticism, Ghazālī accepts that if a necessary connection between cause and effect is denied, then it is possible that “if someone leaves a book in the house, this book, on his returning home, could change into an intelligent slave boy or into an animal” (Ghazālī 1997, 170). Provided that this is not a logical impossibility and that God is all-powerful and absolutely free, then, in fact, a book could turn into “a slave boy or an animal.” However, this does not happen, nor should we expect this to happen, for the following reason: God created for us the knowledge that He does not enact these possibilities (mumkināt)… The continuous habit (istimrār al-‘āda) of their occurrences repeatedly, one time after another, fixes unshakably in our minds the belief in their occurrence according to the past habit. (Ghazālī 1997, 170) In this passage, Ghazālī aims to marginalize the extraordinary events without denying their possibility. The consistency of the natural processes “fixes unshakably in our minds the belief in their occurrence according to the past habit.” According to this version of occasionalism, the world is not governed by an “arbitrary king.” This is not the conclusion scholars like Ghazālī mean

Islamic Occasionalism  287 to suggest. Our “unshakable” belief in the regularity in this world depends on the consistency of the cosmic history of the world. If regular occurrences dominated the past, they will also dominate the present and the future.7 What is important for our discussion is that despite his criticism of necessary connection, Ghazālī still bows to the idea of consistency, which we use to regulate our lives. Ghazālī is skeptical only about the metaphysical framework of causality, not the structure of causality. In his system, natural processes remain regular and law-like. Although God can change the laws of nature, He chooses not to change them. Thus, a “common-sense” view of causality is necessary.8 While we fail to demonstrate the necessary connection, we must still run our lives as if there were a necessary connection. One could choose, from a metaphysical point of view, to believe the truth of occasionalist claims, but one would still, from a practical point of view, live life with the predictability of the world in mind. As such, Ghazālī endorses the view that, in an occasionalist framework, despite a qualified skepticism about the nature of causal relations, the idea of consistency of natural processes is still, to borrow a Kantian concept, a necessary “postulate of practical reason.” A similar tendency to accept but marginalize miracles can also be observed among earlier Muslim occasionalists. If, as Bāqillānī argues, God breaks the normal course of events and creates miracles to protect His prophets then miracles are meaningful only in the context of prophetic missions. Since, according to Islamic prophetology, the prophetic missions ended in the 7th century, one should not expect any subsequent breaks in the apparent ­natural order. At this juncture, we may also note the possibility for an occasionalist to make a distinction between metaphysical opinion and scientific method, given that the primary concern of occasionalism is the metaphysical framework of efficient causality, not the structure of efficient causality. This comes from the basic insight of occasionalism that observation reveals only constant conjunction and cannot prove the necessary connection. The dichotomy between necessary connection and constant conjunction always remains even after the best scientific explanation. Accordingly, an occasionalist could believe that the natural sciences provide a great degree of explanatory possibility about the physical world than any other known method without offering metaphysical truth, for the way phenomenal causes exercise influence is beyond our comprehension. In fact, in later occasionalist tradition there is a move in this direction. For example, Ibn Khaldūn—who is widely viewed as an occasionalist9 and the first philosopher to study history and society as an object of ­systematic science—writes “even if [those sciences] are not adequate to achieve the intention [metaphysical certainty] of the Philosophers, they ­constitute the soundest norm of speculation we know” (Ibn Khaldūn 2004, 405). Although physical sciences cannot achieve metaphysical certainty, they can present greater or lesser approximations about the physical world. This

288  Ozgur Koca separation between metaphysical truth and scientific methodology makes it easier for an occasionalist to subscribe to occasionalism as a metaphysical opinion while acknowledging the methodological value of scientific research. It may even be argued that occasionalistic skepticism toward mental constructions that cannot be located in observation might have certain advantages for scientific research. This skepticism in fact leads to a questioning attitude toward some of the mental constructions of Ptolemaic Astronomy and Aristotelian physics. Ali Kushchu, for example, criticizes the idea of celestial spheres as being purely mental construction without any basis in the senses. This might have played a role in stripping scientific activity of certain unnecessary philosophical concepts. These and similar incidents in the history of Islamic occasionalism indicate that a coherent occasionalist study of the world should resort to natural causes when explaining physical phenomena. This form of occasionalism only appeals to the supernatural when there are very strong reasons to believe that “empirical resources are exhausted.” The acceptance of these extremely low-probability events that occur perhaps only once in the universe’s lifetime does not really affect how one studies the world here and now. On the other hand, such occasionalism can also bring together both general and specific divine action proposals, since God’s universal action is “the continuous background” of extremely rare but possible specific divine actions. 2. The second possible answer is to argue that, although Islamic occasionalism accepts the possibility of breaks in God’s habits, it certainly does not require them. Occasionalism sees the divine action behind all causal activity. The way the world operates is the same as the way God creates. Hence, although Muslim occasionalists allow the possibility of an intervention, they do not require it to affirm the divine action. The theory is not designed to seek God in extraordinary events or miracles, but in causality itself. Both Jesus’s walk on water and a ship’s floating on water, in this view, can be called miracles. In fact, the concepts of intervention and miracle are quite marginal according to the internal logic of the theory, the primary goal of which is to explain the normal, law-like flow of natural processes without positing necessary causal connections and without establishing causal intermediation between God and the world. To put it another way, occasionalism aims to view all ordinary events as miracles, in that God’s conjunction of cause and effect is miraculously predictable and frequent. The main goal is to find God “in what we know.” Given that Islamic occasionalism does not need violations of the laws of nature in order to argue for divine action, the theory actually comes close to scientific methodology as it pertains the study of the world. 3. The third possible answer, which is closely related to the second, is to argue that occasionalism can adopt an entirely different view of miracles. Many authors have already proposed a rational and naturalistic view of miracles. In this literature, miracles do not appear as “violations of the laws of nature” but as “extremely rare events that fall under the laws of nature.” They are “consistent with, but transcend, natural processes” (Nichols 2002,

Islamic Occasionalism  289 703–716). In Muslim intellectual history this view of miracles can be traced back to Ibn Sīnā and Ibn Rushd. For example, Ibn Rushd defends the position that while miracles result from extremely rare causal sequences, we may not be able to identify what those causes are. Despite this conviction, he also argues that miracles should not be questioned, at least publicly, because of their function in the religious and ethical education of the masses (i.e. confirming prophethood).10 Can an occasionalist accept the view of miracles as consistent with natural laws? I believe the answer is in the affirmative. There is no need for Islamic occasionalism to depict miracles as violations of natural laws, despite the general tendency among Ashʿarites to see them as such. God could create miracles as violations of natural laws or consistent with natural laws. If Muslim occasionalists’ main idea is that both cause and effect are created by God without intermediation and attached to each other on a self-imposed habitual path, then these premises are preserved in both cases—even when one sees miracles as extremely rare events that are consistent with the known/ unknown laws of nature. So long as God is preserved as the creator of cause and effect, both natural and supernatural views of miracles appear to be in accordance with the logic of occasionalism. If this is true, then Islamic occasionalism could endorse the idea that the world is to be studied as if there are no causal gaps or breaks in God’s habits or violations of the laws of nature. At this point, a critic might argue that if science has shown that the world follows laws in all its processes and phenomena, why can we not just say that God acts indirectly by sustaining the world and its laws and letting things work out according to these laws? It is true that from the perspective of deistic and similar alternative accounts, one can make the case that one should study the world as if it is governed by secondary causality. The aim of this discussion, however, is not to make Islamic occasionalism compete with other proposals that explain the divine creative action. It simply examines whether this theory of causality—which throughout Islamic intellectual history (and in the present) offered a theologically and metaphysically compelling account of the world for many Muslims—can contribute to this discussion of divine creative action. Another criticism might be that Islamic occasionalism is simply mental acrobatics. Occasionalism rests on the notion that the “constant conjunction” (between events) does not constitute “necessary connection;” that natural laws are God’s habits; and that the world is re-created anew at each moment. The premises are not necessarily intuitive. Moreover, when the “conjunction” is always there, it is difficult to argue that there is no connection. Thus, a critic might hold, occasionalism is merely sophistry. Perhaps the only response that Islamic occasionalism could offer here is to say that an argument should not be discarded simply because it sounds far-fetched. As a metaphysical doctrine, its strength can only be evaluated by exploring its internal consistency and adequacy to explain sensual phenomena on its own terms. A critic might also argue that the idea of indistinguishability of divine causation and secondary causation leaves us with no argument for God’s

290  Ozgur Koca existence. Islamic occasionalism postulates that God affects the universe as a whole in all points of time and space. But if all instances of change and continuity in the world are the work of God, then how can we infer the existence of God from natural phenomena? If every event is the result of God’s creative act, then nature cannot be really said to provide a framework for investigating divine activity. In other words, if God is everywhere, then there is no way of saying God is “here.” As Walker writes, if God’s action is applicable to every class of events (natural, miraculous) and, in some cases, to everything, then the words ‘God’s action’ seem to have lost any substantive content either as that which excites wonder, on the one hand, or as individual enough to be construed as the separate class of acts, called ‘God’s acts’, on the other. (1982, 29) Can Islamic occasionalism provide an answer here? A possible response is that the regularity, elegance, and beauty of nature and its laws are enough of an “argument” for God’s existence here and everywhere and all the time. Not specific instances but the totality of the world justify the argument. The critic may also argue that if Islamic occasionalism’s reconciliation rests on the assertion that natural laws are, in reality, reflections of God’s habitual creation, then the theory provides nothing more than a way of thinking about God’s causality and a scientific understanding of causality as being on two parallel tracks. However, it is not appropriate to see this as a way of reconciliation. Are we not just offering a lexical equivalence to materialistic naturalisms? First, this criticism can be raised against all accounts of the world that locate God in all points of time and space, including Plotinus’s One, Tillich’s Ground of Being, Neville’s Creator, and Ibn ‘Arabī’s al-Ḥaqq. Second, I believe the objection is premised on the wrong assumption of what the reconciliatory strategy of Islamic occasionalism actually is. The theory does not offer two parallel tracks (theological and scientific) relating to and explaining the same natural phenomena. Its strategy is to locate scientific discourse in a larger metaphysical framework without negating its legitimacy, authority, and distinctive methodology. Maybe the more proper analogy here would be concentric circles instead of parallel tracks. 15.3 Challenges Despite these merits, there are certain theological and philosophical challenges to Islamic occasionalism. These challenges, I believe, should lead us to question its viability as an option for thinking about the divine causality without undermining the rigor of the scientific methodology. First, Islamic occasionalism can be a viable option only if the reality of free will is established. Is human choice created? If occasionalism envisages that both cause and effect are created by the Divine Power, then human choice, as

Islamic Occasionalism  291 a cause, must also be created by God. In other words, human choice is not an uncaused cause of itself. In this case, we would lose genuine human freedom. Can this problem be solved? Ashʿarite theologians do accept a type of power on the part of the servant. This, however, is an acquisitive power (kasb, iktisāb) and not a creative one. It can only acquire what is already created. Here, again, Ashʿarites’ theology rejects any necessitarian relationship between the acquisitive power and human acts by detaching the former from the latter. Their relationship is one of proximity (mujāwara) rather than necessity. God habitually creates an act after the servant’s tendency toward that act. The servant does not have a causal efficacy other than this disposition toward an act. Human will is an occasion for the divine creative act. Although there is no necessary relationship between human will and created act, God habitually creates an act in accordance with the tendencies of human will. This is how a human will acquires some acts and not others. However, despite the Ashʿarites’ efforts to ground human agency while at the same time not attributing to it any causal efficacy, it is not clear whether this acquisitive power is truly free. Ashʿarī himself appears to suggest that even this acquisitive power is under the control of the divine will and power. As he writes for example, “God creates acquisition for his servants and he is also powerful over their acquisition” (Ashʿarī 1963/1382, 552), and “there cannot be, under the authority of God, any acquisition that God does not will” (Ashʿarī 1967, 103). These passages suggest that acquisition is not an uncaused cause. It remains caused by God. This is probably why Shahrastānī asserts that Ashʿarite theologians fall into determinism when they assert that the servant’s acquisitive power is also caused by God. If human will is not an uncaused cause of itself, then it is not free (Shahrastānī 1934, 78–79).11 As such, Ashʿarite occasionalism, while aiming to introduce the divine causality to all levels of existence and to make the immensity of God concretely present in the world, appears to render human will subservient to the will of God. Some scholars attempt to solve this problem by placing human choice between existence and non-existence. Ṣadr al-Sharīʿa, for example, describes human will as neither existent nor non-existent (lā mawjūdun wa-lā ma‘dum). As such human choice is defined as a relative matter (amrun ‘itibarīyyun) (nd, 332, 337, 349).12 Relative matters, such as rightness or leftness, can only exist in relation to an object. Terms like “rightness and leftness” or “aboveness and underness” are relative matters in that they exist as relational entities. When we build a wall, relative matters such as the rightness or leftness of the wall emerge. These relative matters cannot be put in the same ontological category with the wall, the object existing in concreto. The wall is an existent, but rightness and leftness of the wall are neither existent nor non-­existent. They do not have external existence and exist only as relative matters in the eye of the observer. The basket is on the right side from one perspective and on the left from another. These, for Sharīʿa, are relative matters located between existence and non-existence. They do not have external existence

292  Ozgur Koca and exist as relative matters only in the mind. How does this ­solution escape the seemingly deterministic conclusions of occasionalism? The argument continues that if human will-choice is not truly an existing thing, it cannot be considered created or caused. If human will is beyond the scope of the act of creation, it should be free. This solution is also consistent with the principles of occasionalism, for God still creates every existing entity. In my view, this proposition does not appear to solve the problem, for even if relative matters are beyond the scope of divine power, they still emerge as a result of the application of the divine power in the natural world. Rightness and leftness of a wall emerge as a result of the creation of a wall. Relative matters, as relational entities, are then determined by the creation of other beings to which they are related. This implies that free will, as a relative ­matter, is still a determined quality. Moreover, how can a relative matter, which does not qualify as fully existing, be robust enough to anchor genuine freedom? How can there be a thing which is existent enough to affect our choices but not existent enough to be caused by the divine power? More generally, how are we to make sense of the putative intermediate realm between existence and non-existence? How can there be a third possibility between existence and non-existence (principium tertii exclusi)? Does Ash’arite theology also offer a proper ontology which can sustain the claim for the third possibility between existence and non-existence? As far as I am aware it does not. These are difficult questions that Islamic occasionalist tradition must address in order to establish the reality of free will in a consistent way with the logic of occasionalism. Second, the failure to sufficiently ground the existence of free will within the logic of occasionalism leads to dire conclusions. If there is absolutely no reality to secondary causality, then it follows that God’s habits have, in fact, no real role in the world. The idea of habitual creation is intelligible in a world where there is some reality to secondary causes. If God did e­ verything without taking the created beings’ causal contributions into consideration, then the creation of cause and effect would be entirely arbitrary. There would be only an appearance of habit. In fact, metaphysically speaking God would remake the world out of nothing at every instant. Neither the state of the universe nor any previous acts of God has any bearing on the next instance. Islamic occasionalism thus becomes somewhat inconsistent when it comes to God’s habits. The very idea of habit suggests a kind feedback from the created order, which occasionalist logic does not allow. Thus, when we lose creaturely feedback we also lose habits. God becomes an arbitrary king. There is no reality of human will, or any other will, therefore no responsibility. In fact, there would be no creation at all; we would lose the world, because if beings are making no causal contribution to the world, their real existence would be unintelligible. Third, even if one can sustain the notion of the divine habits questions remain. Namely that, there seems to be no agreement among the Islamic occasionalist to what extent of the breaks in the divine habits are to be

Islamic Occasionalism  293 taken into consideration when we think about the world. As I argued above, the notion of habit can preserve a theological explanation for the regularity in the world only if the breaks in the divine habits, the miracles, are seen as extremely low-probability events. If these events are occurring only a few times during the lifetime of the universe—and perhaps limited only to ­prophetic ­missions—then they may not affect how one studies the world here and now. One should firmly resort to natural causes to explain the physical events. However, if they are not extremely marginal events, but quite common to an extent that they are as important to us as the habitual ­continuity itself, then it will necessarily effect how we understand the world. In this world, the breaks in the divine habits would be as descriptive as the divine habits themselves. And there seems to be no agreement in the Ash’arite ­literature on such marginalization, despite some attempts indicated above. Theological ambiguity around the notion of the breaks in the divine habits leads to a ­tension between the scientific methodology and the occasionalist perception of the divine causality in the world. Fourth, there are also certain theological problems which can briefly be indicated here. Islamic occasionalism implies God’s direct involvement with evil in the world. If God causes every event within creation, then he causes evil. There is obviously a contradiction between the moral perfection, absolute power, and knowledge of God and the abundance of suffering in the world. To respond to this challenge, an Islamic occasionalist might argue that if the reality of free will and autonomy of beings are preserved within the occasionalist framework, then the doctrine could counter this objection. However, as we just discussed, this is a difficult task to accomplish for Islamic occasionalism. Without providing a convincing answer here, Islamic occasionalism remains susceptible to difficult challenges relating to theodicy. Fifth, it can also be argued that in the world of occasionalism there seems to remain no distinction between God and the world. If God is all causality, then what prevents us from completely eliminating events and things that appear to us as cause and effect. William Hasker writes:  It can be argued on both epistemological and metaphysical grounds that the material substance posited by occasionalists is redundant and should be eliminated, leading to a Berkeleyan immaterialism. Epistemologically, occasionalism holds that material substances make no causal contribution towards our perceptions of objects. (2012) To respond to this point, an occasionalist might argue that the doctrine does not eliminate the world so long as it can sustain the autonomy of beings. However this is not an easy task within the framework of occasionalism without subscribing to an ontology that can sustain the idea of an intermediate realm between existence and non-existence.

294  Ozgur Koca Notes 1 Parts of this chapter previously appeared in Koca (2020). Reprinted with permission. 2 There are also different versions of participatory accounts. How Muslim philosophers and mystics developed these accounts is examined in Koca (2020). 3 Also cited in Griffel (2009, 127). 4 The Qurʾanic verses, “there is no change in God’s creation” (30:30) or “there is no change in God’s words” (10:64) are usually understood to be alluding to this. 5 Later Naṣīr al-Dīn Ṭūsī also writes that Ashʿarī and his followers propose a ­possible and habit-based relationship in mental events. God creates a habitual ­relation between thinking and knowledge. However, he may also not create it (Ṭūsī 1980, 60). 6 Also see Perler and Rudolph (2000, 28–62). 7 Ghazālī is not alone in holding this conviction. Many in the later occasionalist Kalam tradition agree with him. See for example Jurjānī (2015, 512). 8 At this juncture, it can also be noted that the Scottish “common sense” school shares a similar sentiment. The most famous member of the school, Thomas Reid, shares much of the skepticism of Hume regarding causality. He agrees that we have no sensory experience of the necessary relation between cause and effect. But this does not make our common sense of causality mistaken. Please see Nichols and Gideon (Winter 2016 Edition). 9 For evaluations of Ibn Khaldūn’s theory of causality, see Moad (2017, 61–82), Alatas (2012, 105–116), Gibb (1933), Al-Azmeh (1981, 79–81), Wolfson (1959). 10 In addition, several modern Muslim thinkers such as Nidhal Guessoum and Mahdi Golshani have adopted naturalistic explanations of miracles. See Bigliardi (2014, 57–60, 81, 145–146). 11 Ibn Maymūn (Maimonides) agrees with Shahrastānī: “The Ashʿarites were therefore compelled to assume that motion and rest of living beings are predestined, and that it is not in the power of man to do a certain thing or to leave it undone.” The Guide for the Perplexed, 284. The fundamental criticism of Ibn Maymūn here is that Ashʿarite attempts to reconcile human acquisition and the divine creation can “exist only in words, not in thought, much less in reality” (69). 12 For a very good analysis of this scholar’s defense of free will in an occasionalist framework, see Muhtaroğlu (2010, 45–62). See also al-Dīn Ibn Humām (1979, 112–113).

References Alatas, Syed Farid. 2012. Ibn Khaldun (Oxford: Oxford University Press). Al-Azmeh, Aziz. 1981. Ibn Khaldun in Modern Scholarship: A Study in Orientalism (London: Third World Centre for Research and Publishing). Ibn Humām, Kamāl al-Dīn. 1979. Kitāb al-Musāyara (İstanbul: Çağrı Yayınları), 112–113. Ashʿarī. 1963/1382. Maqālāt al-Islamiyyīn wa Ikhtilāf al-Muṣallīn, ed. Helmut Ritter (Wiesbaden: Franz Steiner Verlag). Ashʿarī. 1967. The Elucidation from Islam’s Foundation (Kitāb al-Ibāna ‘an Uṣūl al-Diyāna), trans. Walter C. Klein. American Oriental Series vol. 19. (New York: Kraus Reprint Corporation). Baghdādī. 1928. Uṣūl al-Dīn (Istanbul: Dar al-Funun Ilahiyat Fakültesi). Bāqillānī. 1958. Kitāb al-Bayān, ed. Richard McCarthy (Beirut: al-Maktabat al-Sharqiyya).

Islamic Occasionalism  295 Bāqillānī. 1987. Kitāb al-Tamhīd, ed. Imaduddin Ahmad Haydar (Beirut: Muassasa al-Kutub al-Thakafiyya). Bigliardi, Stefano. 2014.  Islam and the Quest for Modern Science: Conversations with Adnan Oktar, Mehdi Golshani, Mohammed Basil Altaie, Zaghloul El-­ Naggar, Bruno Guiderdoni and Nidhal Guessoum. Transactions (Istanbul: Swedish ­Research Institute in Istanbul). Dembski, William. 1994. “On the Very Possibility of Intelligent Design,” in The Creation Hypothesis, ed. J. P. Moreland (Downers Grove: InterVarsity Press), 113–138. Ghazālī. 1992. al-Maqsad al-Asnā (The Ninety-Nine Beautiful Names of God) translated by David B. Burrel and Nazer Daher (Cambridge: The Islamic Texts Society). Ghazali. 1994. “Ghazali’s Chapter on Divine Power in the Iqtiṣad,” translated by Michael E. Marmurah, Arabic Sciences and Philosophy 4, no. 2: 279-315. Ghazālī. 1997. The Incoherence of the Philosophers (Tahāfut al-Falāsifa), a Parallel English-Arabic Text. ed. and trans. M. E. Marmura (Provo, UT: Brigham Young University Press). Gibb, H. A. R. 1933. “The Islamic Background of Ibn Khaldun’s Political Theory,” Bulletin of the School of Oriental Studies, University of London 7, no. 1: 23–31. Gimaret, Daniel. 1980. La doctrine d’al-Ashʿarı¯. Paris: Cerf, 1990. Théories de l’acte humain en théologie musulmane. Paris: J. Vrin Griffel, Frank. 2009. Al-Ghazālī’s Philosophical Theology: An Introduction to the Study of his Life and Thought (New York: Oxford University Press, 2009). Griffel, Frank. 2016. “Al-Ghazali,” in The Stanford Encyclopedia of Philosophy, ed. Edward N. Zalta. . Hasker, William. 2012. “Occasionalism,” excerpted on 9/6/2012. . Ibn Fūrak. 1987. Mujarrad, ed. Daniel Gimaret (Beirut: Dar al-Mashriq). Ibn Khaldūn. 2004. Muqaddima, edited by N. J. Dawood and translated by Franz Rosenthal (Princeton, NJ and Oxford: Princeton University Press). Ibn Maymūn (Maimonides). 2002. The Guide for the Perplexed (Dalālat al-Ḥāʾirīn), trans. M. Friedlander 2. Ed. (Skokie, IL: Varda Books). Ibn Rushd. 1954. Tahāfut al-Tahāfut, trans. Simon Van den Bergh (London: Messrs. Luzac and Company, Ltd). Jurjānī, 2015. Sharḥ al-Mawāqif, trans. by Omer Turker, a Parallel Turkish-Arabic text. 3 vols. (Istanbul: Turkiye Yazma Eserler Kurumu Baskanligi). Juwaynī. 1969. al-Shāmil fī Uṣūl al-Dīn, ed. A.S. al-Nashshar, Faysal Budayr Awn, and Suhayr Muhammad Mukhtar (Alexandria: Munshaʾat al-Maʿarif). Juwaynī. 1985. Kitāb al-Irshād, ed. Asad Tamimi (Beirut: Muassasa al-Kutub al-Thakafiyya). Juwaynī. 2003. al-ʿAqīda al-Niẓāmiyya, ed. Muhammad Zubaidi (Beirut: Dar Sabil al-Rashad). Koca, Ozgur. 2020. Islam, Causality, and Freedom (Cambridge: Cambridge University Press). Moad, Edward Omar. 2017. “Ibn Khaldūn and Occasionalism,” in Occasionalism Revisited: New Essays from the Islamic and Western Philosophical Traditions, ed. Nazif Muhtaroglu (Abu Dhabi: Kalam Research Media), 61–82. Muhtaroğlu, Nazif 2010. “An Occasionalist Defense of Free Will,” in Classical Issues in Islamic Philosophy and Theology Today (New York: Springer), 45–62. Nichols, Ryan and Gideon Yaffe. 2916. “Thomas Reid”, The Stanford Encyclopedia of Philosophy  (Winter 2016 Edition), Edward N. Zalta  (ed.). .

296  Ozgur Koca Nichols, Terence L. 2002. “Miracles in Science and Theology,”  Zygon, 37, no. 3: 703–716. Perler, D. and U. Rudolph. 2000. Occasionalismus: Theorien der Kausalität im arabisch-islamischen und im europäischen Denken (Göttingen: Vandenhoeck & Ruprecht). Ṣadr al-Sharīʿa. nd. Sharḥ al-Talwīh, vol. 1 (Beirut: Dar al-Kutub al-Ilmiyyah). Shahrastānī. 1961. al-Milal wa-l-Niḥal, ed. Muhammad Sayyid Kilani (Beirut: Dar al-Ma’rifah). Shahrastānī. 1934. Nihāyat al-Iqdām, ed. Alfred Guillaume (London: Oxford ­University Press). Qāḍī ʿAbduljabbār. 1962–1965. al-Mughnī fī Abwābi al-ʿAdl wa-l-Tawhīd, ed. Ibrahim Madkur, Taha Husayn, Mahmud M. Khudayri (Cairo: al-Dar al-Misriyya). Ṭūsī. 1980. Talkhīṣ al-Muḥaṣṣal, ed. Abdullah Nurani (Muassasa-yi Muṭālaʻāt-i Islāmī-i Dānishgāh-i McGill, Shu’ba-yi Tihran). Walker, Ian. 1982. The Problem of Evil and the Activity of God (London: New Blackfriars). Wolfson, Henry A. 1959. “Ibn Khaldun on Attributes and Predestination,” Speculum 34, no. 4: 585–597.

Index

abstract: object 101; truth 174, 176, 178 accident 15, 78, 96, 183–184, 284–285 accidental: changes 139; certitude 176– 177; form 95; powers 92; property 144; reliability 177; truth 174 active intellect 11, 102, 126, 136, 139– 140, 144, 207, 226–227; see also Giver of Forms Adam 3, 261–265, 268, 272–273, 275–279 Adamic exceptionalism 262 afterlife 3, 14, 227–228, 232, 237–238 agency 14, 18, 91–92, 94, 118, 138, 207–208, 210, 217, 291 agnostic 58, 111, 120, 142, 193–195, 272 Allah 66, 98–99, 168, 196, 209–210, 212–215, 219, 241, 251, 253 Alfarabi 166, 169–178; see also al-Fārābī analytic philosophy 1–2, 7–9, 11–12, 16–17, 19–20, 24–27, 35–36, 160, 170 anthropomorphic (or anthropomorphism) 87, 92, 183, 188–189, 194, 205, 207 Aquinas 100, 109, 152 Aristotle (or Aristotelianism) 11–12, 14–15, 28, 83, 88, 98, 103–104, 114–115, 119, 170, 206, 215, 226, 230–231 Ashʿarism (or Ash’arite or Asharites) 149, 263–265, 267, 271–272, 277, 279, 283, 292–293 atheism 101, 182, 186–187, 192–194, 200 atom 105, 238, 284–285 Averroes 98, 102–106, 108–120, 160, 237, 239; see also Ibn Rushd

Avicenna 11–12, 14–17, 20, 29, 36–37, 43, 58–59, 73, 100, 102, 105, 119, 125–127, 134–144, 149, 204, 207, 220, 225–232; see also Ibn Sīnā āyah 147, 150, 212–213, 219; āyah-sign 147, 150–151 Berlin, Isaiah 3, 225, 229, 233, 235 causality 58, 66–67, 77, 129, 282, 286–294 causal necessity 84, 93, 129, 142, 286 causation 41, 53, 56, 58, 103, 107, 119, 125, 129–130, 134, 142, 144, 283, 289 certainty 3, 13, 30, 142, 165–171, 173–174, 176, 178, 231–232, 287 continental: approach 11; phenomenology 13; philosophy 8, 19; tradition 9, 12–13, 19 cosmological argument 45, 60, 66, 200; see also kalām cosmological argument creationism 261–262, 264, 270 decolonial 7, 9, 17–18, 20 decoloniality 9, 17–19 Descartes 176 divine: action 134, 208, 212, 214, 218, 264, 277, 288; attributes 66–67, 83, 96, 109, 118, 194, 205, 207, 214, 282; command (theory) 79, 234, 284; hiddenness 3, 152, 160, 181–182, 190–193, 195– 196, 198; knowledge 3, 67–68, 83, 85, 90, 93–96, 98, 104–106, 109, 112, 116, 117; power 3, 67, 74–75, 78, 83, 90–91, 94,

298 Index 128, 285, 290, 292; priority 110, 112, 117; simplicity (or DDS) 3, 15, 98–103, 105–106, 108–113, 115–119, 192, 209 emanation 12, 102–103, 134, 204 empiricism (or empiricist) 12, 37, 216 epistemic injustice 241 existence of God 2, 14, 64–66, 68, 77– 78, 101, 181, 192, 199–200 essence 9, 12, 15–16, 83, 85, 87, 89, 91–92, 95–96, 100, 102–105, 107–112, 116–118, 133, 140, 142, 144, 170, 192, 198, 208, 220, 227 essentialism 12, 139–140 estimation (or wahm) 24, 27–29, 30–32, 36–37, 137, 161 eternal: attributes 67; being 65–67, 283; creator 68; effects 58; essence 91; happiness 235–237; knowledge 84, 86–88, 94, 111; power 86, 91; soul 227; will 84–85, 87–90; world 59 evolution 3, 261–265, 267–275, 277–278 Eve 3, 261–264, 272–273, 275–279 faith 3, 142, 159, 165–170, 178, 199, 205, 214, 216, 219, 272 falāsifa (or falasifah) 66, 74, 83–93, 95–96, 125–127, 183 falsafa (or falsafah) 126, 129, 142, 149 al-Fārābī (or al-Farabi) 1, 3, 24, 27, 29– 30, 32–34, 92, 120, 149, 191– 192, 198–199, 219, 225–227, 229–233, 239; see also Alfarabi freedom: human 208, 282, 291; divine 73, 110, 210, 282–283; of thought 13 free will 110, 203, 290, 292–294 Frege, Gottlob 10, 13–14, 86 Ghazali (Ghazālī or al-Ghazālī) 1, 3, 41–42, 47, 56, 58–59, 65–66, 74, 83, 85–94, 96, 100, 102– 103, 105–106, 111, 119, 125– 136, 138–142, 149, 183–185, 199, 225, 228, 285–287, 294 generic overgeneralization effect 42–43, 46–47, 49, 51–52, 54 generics 43–45, 47–50, 52, 59–60

Giver of Forms 131–132, 139; see also active intellect happiness 215, 226–233, 235–238 heaven (or heavens) 3, 88, 99, 126, 137, 150, 213, 216, 225–229, 232–239, 264 Hegel, Georg Wilhelm Friedrich 9, 219 Hume, David (or Humean) 93–94, 147, 152, 154, 156–159, 161, 294 human exceptionalism 261 hylomorphic (or hylomorphism) 16, 95, 115, 131, 135 Ibn al-ʿArabi 208, 290 Ibn Babawayh 220 Ibn Khaldūn 199, 287, 294 Ibn Rushd 1, 83–90, 92, 95–96, 160, 199, 286, 289; see also Averroes Ibn Sīnā 1, 27–33, 36–37, 73, 83–84, 92, 149, 184–185, 191–192, 199, 289; see also Avicenna Ibn Taymiyya 20 immutable (or immutability) 15, 105, 110 intellectual history 10, 36, 282, 289 intelligible form 95, 106–109, 111, 112, 116–118 al-Jurjānī, al-Sayyid al-Sharīf 78, 294 justification 170, 174–177, 179, 188, 194, 213, 234, 254 kalām cosmological argument (or kalām argument) or 1–2, 41–43, 46, 51–58, 66, 68; see also cosmological argument Kant, Immanuel 9–10, 64, 238, 287 al-Kindi 52, 100, 119, 149, 183, 198, 205, 207 Leibniz (or Leibnizian) 73, 200, 203 Maimonides 74–76, 100, 120, 284, 294 miracle 1, 3, 126–127, 131–133, 135, 138, 141–143, 147–161, 266, 288; see also muʿjizah Mīr Dāmād 15 modality 12–13, 41, 83, 86, 88–89, 175 monopolytheism 198 muʿjizah 147–149, 159–160; see also miracle

Index  299 Mullā Ṣadrā (or Mulla Sadra) 15–17, 20, 166, 179, 208, 211 mutakallim (or mutakallimūn) 43, 58–59, 83, 89 Mutazilite 30, 149

prophethood 143, 159–160, 289 prophetic: faculty 139–140; power 140; vision 37 Prophet Muhammad 3, 136, 138–144, 149, 151–152, 154, 159, 247

naturalism 16, 264, 266, 279, 290 necessary: being 65–67, 69–74, 264; existence 78, 100 necessary existent 15, 100, 220 Neoplatonism 185, 205–206, 209, 218, 220 normative: attitudes 215–216; belief 215; force 245, 247–248, 250, 252; justification 234; priority 252; skepticism 218

quiddity 95 Quran (or Quranic) 1, 3, 98–99, 118– 119, 166, 168, 239

occasionalism 3, 264, 277, 282, 284–293 omnipotence 66, 74–78, 141, 218, 264, 277 omnipotent 3, 65, 67, 69–78, 125, 132, 135, 155, 196, 203, 205, 210, 218 oneness of God 98–100, 102, 119, 192; see also taw al; uniqueness of God ontological (in)dependence 91, 98, 100–102, 104–105, 108–110, 112–113, 116–119 particulars 29; God’s knowledge of 105, 108, 199 personal God 155, 185–189, 191, 193, 196, 198, 205 polytheism 188, 198 principle of sufficient reason (PSR) 57–58, 200 problem(s) of evil 160, 194, 197, 203, 205, 208–209, 212, 218, 220 prophecy 14, 24, 27, 29–34, 134, 139– 140, 142–144, 149 prophet 29–32, 126–127, 129–131, 133

rationalism (or rationalist) 9, 149 rationality 18; Islamic belief 181–182, 195–197, 199; theism 200 al-Rāzī, Fakhr al-Dīn 20, 43, 58–59, 136, 143–144 revelation 15, 29, 149, 151, 160, 192, 197, 218–219, 256 scientific realism 267 scripture 15, 98, 118, 125, 147–148, 160, 167, 205, 216, 261–263, 265, 271–276, 278–279 spirit 8, 29, 167, 170, 206–207, 219 spiritual 11, 16, 144, 216, 225, 228, 235, 238 supernatural 101, 264, 266, 286, 288–289 al-Taftāzānī, Saʿd al-Dīn 3, 65–70, 72–73, 75–79 tawḥīd 98–100, 102, 118–119, 150, 152, 192; see also oneness of God; uniqueness of God theism 98, 101–102, 184–187, 189– 190, 193–194, 199–200, 217, 220 theodicy 203, 205, 208, 218, 293 testimony 3, 30, 32, 154, 156–159, 161, 177, 241–247, 249, 253–255, 270–272 uniqueness of God 2–3, 65; see also oneness of God; tawḥīd universals 18, 29, 37, 52, 105, 207