Falsafa as Ethics of Belief

Islamic Philosophy and the Ethics of Belief Anthony Robert Booth Chapter 1: Falsafa as Ethics of Belief [1st Draft. Please do not cite. Comments Welcome. a.r.booth@sussex.ac.uk] A Note on Translation, Transliteration, Dates, and References: Throughout this manuscript I have transliterated Arabic words into English, with the use of diacritical marks, and have put the words in italics. However, I used Anglicised versions of certain well-known Arabic words and names. For instance, I used ‘Koran’ instead of ‘Qur’ān’ or ‘Hadith’ instead of ‘Hādīth’, and the names ‘al-Kindi’, ‘alFarabi’, ‘al-Ghazali’ instead of ‘al-Kindī’, ‘al-Farābī’, ‘al-Gḥazālī’. I used the Latinate names ‘Averroes’ for ‘Ibn Rushd’ and Avicenna for ‘Ibn Sīnā’. I used diacritical marks for the transliteration of less well-known (than the above) Islamic Philosophers, but did not use them for the word ‘Falasifa’ and its cognates, since I used it so often. When quoting and referencing works in Medieval Islamic Philosophy, I used the full name of the work in English translation, and in italics, followed by page numbers; for those referred to in the original Arabic, I have used the Arabic name of the work, also in italics, and with a footnote on the translation. All works, in translation or in the original, are referenced at the end of the book, before my list of “secondary” sources (by which I mean all non-Medieval works). All dates used refer to the Common Era calendar. 1. Knowledge in Islam. In his impressive book Knowledge Triumphant, the illustrious scholar of Islam - Franz Rosenthal - argues that the central leitmotif of the Islamic civilization is the concept of knowledge 1 . Understanding the Islamic notion of knowledge, according to Rosenthal, is necessary if one is to understand Islam, the civilisation it gave birth to, and the particular historical course it has taken. Indeed, according to Rosenthal, “Islam means Knowledge”. The position is at least prima facie defensible, as Rosenthal suggests, when one considers that the Koranic term for the time preceding the Koranic revelation is jāhilīyah, a term usually translated as ‘ignorance’2 – and this, naturally enough, seems to suggest that the key difference between the time preceding the Koranic revelation, and the time following it, is that the latter (and not the former) was a time when people in this world – through the instruments of 1 For a similar take cf. Bakar 1998. Though Goldziher (1889) has – controversially – argued that it is better translated as 2 Though Goldziher (1889) – controversially – argued thatthe it state is better translated as the US “barbarity”. Jāhilīyah is thehas term that Sayyid Qutb describes of the West, and “barbarity”. Jāhilīyah is the term that Sayyid Qutb describes the state of the West, and the US in particular, in the 20th Century. Qutb – unfairly dubbed “the philosopher of terror” by certain US journalists – will be discussed in Chapter 3. 2 1 revelation – came to have knowledge3. To the modern epistemologist this claim will raise several questions, however: are we here talking about the claim that people living in pre-Koranic times merely had true beliefs about the world, or that they had only justified, true beliefs (falling short of knowledge) about the world? And if the latter, what does Islam tell us is the crucial difference between justified, true belief and knowledge? Are we to think that those living in jāhilīyah live in complete ignorance of any proposition? Or are they merely ignorance of certain, important propositions? If so, which ones? Do they lack merely propositional knowledge, or also practical? Are they living in “deep” ignorance (where there was no way they could have acquired relevant evidence) or living in culpable ignorance? Rosenthal’s work, although presented as a work of epistemology, does not really answer these questions. What we get, instead, is a historical study on how the Arabic concept ʿilm (knowledge) was taken by the various intellectual movements of medieval Islam, as well as just how central this concept figured in all of these movements. As he puts it: “there is no other concept that has been operative as a determinant of Muslim civilization in all its aspects to the same extent as ʿilm” (Rosenthal 2007, p. 2). Rosenthal even advances the bold thesis that understanding the centrality of the concept of knowledge in Islam is crucial to an understanding the turn the Islamic world has taken in recent times, and how it was been germane to a kind of fundamentalism4. Contrary to Rosenthal, however, I want to try to show in this present study that the Medieval Islamic Philosophers’ epistemology in fact gives us the resources for understanding extremist belief, and what is uniquely wrong with it. I also want to try to suggest that, at least as far as the Medieval Islamic Philosophers (Falasifa) were concerned, the core philosophical issue was not about knowledge in particular, but about what constitutes justified belief. This is of course traditionally a concern in epistemology, but unless one makes the substantive assumption that nothing but evidence (or epistemic reason) can justify belief, then we can consider that concern to outrun topics in epistemology. As such, I will try to argue, a fruitful way of thinking about the Falasifa’s intellectual project is to consider them to be primarily engaged in the ethics of belief. That is, they were concerned with understanding the epistemic (and in many cases the non-epistemic) conditions of justified belief, and, particularly, when belief is blameworthy (and sometimes – as in the case of apostasy – punishable by death). I will first try to show how understanding the views of the Falasifa in terms of the debate in Western Philosophy on the ethics of belief proves fruitful, and then show how Falsafa can be said to occupy a unique and defensible position, that ought to be of considerable interest to those working on the issues sans phrase (that is, in a way that ignores whether the issues “belong” to a particular tradition). 3 A famous Hadith (recalled by Abu Dhara) suggests that the central legacy of Prophecy is knowledge: “The learned are the heirs of the Prophets, and the Prophets leave neither dinar nor dirham, leaving only knowledge, and he who takes it takes an abundant portion” cf. www.hadith.com. 4 “From the Qur’ranic attitude toward knowledge, it would be possible almost to predict the course that Muslim theology, mysticism, jurisprudence and the like were to take, as well as the fate that had to befall the liberating influences set in motion by the reception of the Classical heritage in the ninth century. The triumph and defeats of Muslim civilization are foreshadowed in Muhammad’s understanding of “knowledge”” (Rosenthal 2007, p. 32). 2 One of the many very interesting bits of linguistic data that Rosenthal’s book offers us is that the arabic concept ʿilm differs to the English ‘knowledge’ in several important respects. For instance, in Modern English ‘knowledge’ does not admit of a plural formation, where ʿilm in Arabic does, as per: ʿulum. More significantly, I think, is the fact that the Arabic ʿilm can admit of degrees/can be graded, where ‘knowledge’ in Modern English cannot, at least not if used about propositional knowledge. Indeed, some contemporary epistemologists have taken as data the fact that ‘knowledge-how’ is gradable, where propositional knowledge is not. That is, the former but not the latter locution seems felicitous: - I sort of know how to play the guitar. - I sort of know that Brighton is in England. And this bit of data is then taken to be part of a broader set of evidence for the claim that practical knowledge cannot be reduced to propositional, theoretical knowledge. In English the notion of justification (as applied to beliefs) does seem to be gradable, however. The following locutions seem perfectly felicitous, for instance: - James is more justified than Andrew in believing that he is good at rugby. - I have better justification for the belief that there is a Higgs Boson particle than I had ten years ago. - James is sort of justified in believing that his date will arrive on time. That the English notion of justification but not knowledge is gradable, and that the Arabic notion of ʿilm is gradable, I think gives us prima facie reason to think that when the Falasifa wrote about ʿilm, they had in mind a concept closer to epistemic justification than knowledge. Again, the fact that we can felicitously speak of ‘justifications’ and ‘reasons’ (where talk on ‘knowledges’ sounds odd) gives us defeasible reason to prefer (or at least consider as a plausible alternative) an interpretation where the Islamic Medieval Philosophers were primarily interested in the conditions for justified belief, more than they were with the concept of ‘knowledge’. The primacy of the issue of apostasy, as we shall see, further vindicates this last claim, for according to even the most notoriously strict accounts of apostasy in Medieval Islam (for instance those of the Imam al-Ghazali) one can fail to be an apostate when one fails to know the claims of Islam. Most importantly, however, my hypothesis is given most support, I hope now to show, by the views of the Falasifa themselves. The gradability of the Muslim concept ʿilm is mirrored in their epistemological theories, where, in Aristotelian fashion, they painstakingly enumerate the various differentiea ‘knowledge’ is subject to, and, in Neo-Platonic fashion, the hierarchy that dominates that taxonomy5. I will argue paying more attention to this 5 This hierarchical system seems to cohere with on found more broadly in the Islamic intellectual tradition at large; as Seyyed Hossein Nasr puts it: “The disorder which rules over the modern educational curriculum in most Islamic countries today is to a large extent due to the loss of the hierarchic vision of knowledge as one finds in the traditional Islamic education system. In the Islamic intellectual tradition, there existed a hierarchy and inter-relation 3 countenanced gradability concerning the subject ‘knowledge’ (I claim better considered to be ‘justified belief’) may provide the keys to understanding a unified view with respect to the Falasifa’s take on the relationship between faith and reason, with respect to well-known problems with respect to interpreting their individual views, and with respect to their political philosophy. 2. The Ethics of Belief in the West Let us grant the claim that it is initially plausible that the Falasifa were at least somewhat concerned with issues centred on the ethics of belief. To show that it is fruitful to think of them like this, let us now consider the more modern western debate on this issue. The debate has become a complex one, and, as with any topic in contemporary Western Philosophy, replete with hyper-specialised jargon and hairsplitting distinctions. The way of conceiving the debate will necessarily look bluntedged to the specialist, but I hope will be a detailed enough approximation to be useful given the broader ends of this particular book. It seems to me that, traditionally, the ‘ethics of belief’ has been taken in the western philosophical world (in the last two or three centuries) to denote a debate between essentially two groups of philosophers considering the question regarding what is justified belief. The first – Evidentialists – think that nothing but evidence, or more broadly epistemic reason (reason to think that a proposition p is true) can justify a subject S’s believing that p. Not only do Evidentialists think that evidence that p is necessary for S’s belief that p be necessary, they also think that no consideration other than evidence is relevant to whether S is justified in believing that p. They adhere to: Evidentialism S’s belief that p is justified iff S has epistemic reason to believe that p. The second group I’m going to less standardly term: Non-Evidentialists. This group of Philosophers simply deny the truth of Evidentialism, and think the following: Non-Evidentialism S’s belief that p is justified iff S has, for reasons of sort alpha, beta, gamma, delta, either one or a combination of alpha, beta, gamma, delta reason to believe that p. Non-Evidentialists do not always deny that S’s having epistemic reason to believe that p is necessary for S to be justified in believing that p, but they do always deny that it is sufficient. That is, they deny that having epistemic reason to believe that p always entails that S is justified in believing that p. The archetypal Evidentialist is usually considered to be W.K. Clifford, a late 19th century British Mathematician6 and Philosopher, who famously wrote in his work The Ethics of Belief: between various disciplines which made possible the realization of unity in multiplicity not only in the domain of religious faith and experience but also in the realm of knowledge” (Nasr 1998, p. xi). 6 It is said that Einstein developed theories in Geometry that Clifford had first suggested. [refs?] 4 It is wrong, always, everywhere, and for anyone, to believe anything upon insufficient evidence. (Clifford 1877, p?) The view seems to follow the view of Locke (and other British empiricists7) however: Faith is nothing but a firm assent of the mind: which if it be regulated, as is our duty, cannot be afforded to anything, but upon good reason; and so cannot be opposite to it. He that believes, without having any reason for believing, may be in love with his own fancies; but neither does he seek truth as he ought, nor pays the obedience due to his maker, who would have him use those discerning faculties as he has given him, to keep him out of mistake and errour. (Locke, Essay Concerning Human Understanding, iv, xvii, 24).8 The view as formulated by Locke and Clifford raises many questions; among them is: what sort of ought is the ‘ought to believe’ referred to here, what sort of wrong is it to believe upon insufficient evidence? Is it a moral wrong? Or a purely epistemic wrong? Or is it just wrong tout court or all things considered? I have argued elsewhere (Booth 2012, 2014) that the obligation involved in ‘ought to believe’ is an all things considered ought, such that Locke and Clifford must be talking about all things considered oughts if their view is to make sense9. Clifford, certainly, seems to have ruled out that he is talking about merely epistemic ought, given his opening discussion of the moral wrong involved in a ship captain’s taking his crew out to sea on the belief that the ship is sea-worthy, where such a belief is based on insufficient evidence.10 For now, I am going to assume that the question here is about more than what makes a belief epistemically justified, since I think that the Falasifa, given their concerns with apostasy (and its concomitant death-sentence), must have been concerned with more than with merely what one ought to believe from an epistemic point of view. The opposing view, I have labelled Non-Evidentialism. I’ve given this view a broader term in order to encompass both the positions known as Pragmatism and Fideism. The contemporary opponent to Clifford was famously William James, who in his essay “The Will to Believe” makes the case that belief (and in particular, religious belief) may be made rational by non-epistemic factors, such as the determinants of self-interest. Of course, James was a famous proponent of a more general view known as Pragmatism (where, for instance, the idea of truth as a correspondence relation between proposition and the world is rejected). But in this context, ‘pragmatism’ is not taken to be so comprehensive a doctrine, but rather just the view that belief can be justified by pragmatic or prudential considerations. An example often invoked in this context is Pascal’s Wager; the idea, very briefly, put forward by the 17th century eponymous French Philosopher and Mathematician, that prudential considerations may make belief in God rational, given the utility of believing there is a God if there is a God, compared to the minimal negative utility if there is no God11. Fideism is the 7 For Hume this approach famously led to atheism, cf. Hume 1779. For discussion on Locke on the ethics of belief see Wolterstorff 1996. 9 For criticism see Stapleford 2015. 10 For discussion on Clifford’s view, see Haack 2001. 11 See Pascal 1662. For more recent discussion cf. Jordan 2006. For a recent discussion of the debate between Clifford and James, cf. Aikin 2014. 8 5 view, commonly associated with Soren Kierekegaard (1847), that certain beliefs may be made rational – that is, can be justified – when they are the product of faith12. Paradigmatically, belief in God is considered justified on this basis - but the Fideist need not restrict her domain solely to religious beliefs. What is important to her is that the belief be held in an act of self-constitution rendered possible by our radical human freedom, in a “leap of faith”. Such acts may involve more garden-variety acts of doxastic self-constitution, as per when, knowing that most marriages end in divorce, someone comes to believe at the altar – and against the evidence – that he will be forever faithful to his beloved.13 Both the Fideist and the Pragmatist positions seem to be predicated on the idea that we can believe at will. This idea, while defended by luminaries such as Descartes14, is currently, almost comprehensively, considered to be manifestly false, at least as a contingent, psychological matter of fact about ourselves15 (though some think that the very nature of belief rules out the possibility of our believing at will16). Indeed, the falsity of the proposition that we can (directly, at least) believe at will has been taken to be a data point in favour of an argument against a particular “internalist” account of epistemic justification. The idea here is that internalist accounts of epistemic justification must be committed to a “deontological”17 understanding of the latter – viz. that S’s being justified in believing that p should be understood to mean that S has dispatched her epistemic duties or obligations in believing that p, and that she’s thus blameless in believing that p. But if belief is not under our voluntary control, then the idea that we are subject to epistemic obligations seems to violate the principle that “ought implies can” – that A can be obliged to phi only if phi-ing is under A’s voluntary control. Curiously, the data point that belief is not under voluntary control, has, in the contemporary literature, been also used to defend a different kind of “evidentialism” according to which there are no obligations pertaining to belief other than epistemic obligations. The thought here being that if there were non-epistemic obligations to believe, then belief should be sensitive to non-epistemic considerations when, in deliberation, S considered whether to believe that p. But when we – in full awareness – deliberate as to where to believe that p, only evidential considerations (epistemic reasons) can sway us. So there are no nonepistemic obligations to believe18. This kind of ‘evidentialism’ differs from Clifford’s (the one we’ve been considering), since the latter is consistent with the claim that as a 12 Modern Fideists include C. Stephen Evans (1998) and John Bishop (2007, 2013). For further examples of this kind see Maurusic 2013. 14 Insert quote? For discussion see Weatherson 2011 (for criticism of the latter, see Booth & Peels 2012). It is worth noting that Descartes was an anti-evidentialist in a different way than by endorsing either Fideism or Pragmatism. That is, qua foundationalist, he seemed to endorse the idea that certain propositions ought to be believed, not because the evidence mandates them, but rather because no belief would otherwise be justified. Such “properly basic” beliefs, sometimes considered to be beliefs in “hinge propositions” (cf. Wittgenstein 1969) have been considered by modern Philosophers of Religion to be essential in considering the kind of warrant we necessary for believing in God (cf. Plantinga 2000; for criticism see Philipse 2012). 15 An outlier here is Peels (forthcoming). 16 Notably Shah and Velleman 2005; for criticism of the latter, see Booth forthcoming. 17 The term here is used slightly differently from how it is used in modern Ethics, for a comparative analysis, cf. Booth 2008. 18 Cf. Shah 2006; for criticism of the latter, cf. Booth 2008. 13 6 matter of moral obligation one must believe in accord with one’s evidence. Jamesian Pragmatism is then the view that there may be morally salutary consequences to believing against the evidence, where the modern ‘pragmatist’ is concerned with defending the idea that a debate between James and Clifford is a coherent one19. However, if Jamesian Pragmatism, or the view I’ve called “Non-Evidentialism” above is shown to be true, then the modern pragmatist line must also be. The Falasifa were from the outset explicitly advocating a specifically Islamic philosophy, and conceived of their philosophical project to be furthering the theistic enterprise. Sayed Hossein Nasr has indeed suggested that they should be viewed as engaged with “theosophy”20. As such, their view concerning doxastic obligation was certainly not the modern “evidentialist” one, since they all held that it was – at least in certain cases - morally impermissible to fail to believe certain Koranic propositions, such as the proposition that there is no God but God and that Mohammed is the messenger of God. The central question they were concerned with, it seems to me, is the one the classical (Cliffordian) evidentialist sought to answer: viz. ought we to believe these propositions just in case they are supported by sufficient evidence, where the ‘ought’ referred to is not a mere epistemic ought. Whether they thought that ‘ought’ was then a moral ought is an open question. Again, I have argued elsewhere (Booth 2012) that any position must regard that ought as an all things considered ought, rather than merely a moral or epistemic ought, on pain of inconsistency. Briefly, one cannot determine whether one is blameworthy for holding a particular belief by considering reasons belonging to one sortal of ought, such as moral ought, since different kinds of ought can come into competition with one another. If different kinds of ought can compete, as per prudential ought pushing someone in one direction and epistemic ought in another (e.g. it is inconvenient for a politician to believe that the German Parliament will take a hard stance on Greece’s leaving the Eurozone, but the evidence for that proposition is overwhelming), then the existence of one conflicting kind of ought can excuse the failure to comply with the competing ought (the politician is excused from failing to dispatch her prudential obligation, given the existence of the competing epistemic one). As such, it is only a judgement that takes on board all the differing pushes and pulls of competing kinds of reason that can determine whether one is blameworthy with respect to having a particular belief. I think that, in being acutely aware of the need to address the balance between the demands of theoretical and practical reason, the Falasifa end up with a position much like the one I have just canvassed. I will try to demonstrate that in due course, but in the meantime it will suffice to make do with the working hypothesis that at the very least they did not think that the ought in question here is purely an epistemic ought, and that as such the issues they were concerned with resemble the issues underlying the historical debate between Clifford and James, rather than the modern debate. However, as I will also try to show, they come up with an interesting argument in favour of something akin to Jamesian Pragmatism (or Non-Evidentialism more 19 Cf. Foley 1993, Booth 2014, Reisner 2009. “If by philosophy we mean secularized philosophy as currently understood in the West, that is, the attempt of people to reach the ultimate knowledge of things only through the use of their rational and sensuous faculties and cut off completely from both the effusion of grace and knowledge made available through prophecy and revelation as well as the light of the Divine Intellect, then such an activity is peripheral in the Islamic intellectual universe” Nasr 2006, p. 32. 20 7 broadly), and if their argument succeeds therefore also vindicates the Modern Pragmatist line. There is an issue facing both the Evidentialist and Non-Evidentialist positions I have just described when it comes to accounting for theistic belief. The issue has to do with the role of Prophesy – or divine scripture more generally – in determining correct religious belief. If we take an Evidentialist view, then we must, on pain of special pleading, consider a theistic belief justified just in case it is believed on the basis of sufficient evidence21. In other words, a religious belief’s correctness is determined by its truth conditions. But if this is the case, what is Prophesy’s unique epistemic role? Does Evidentialism entail that we can reason our way, by investigating the world, into correct religious belief (if there is such a thing)? If it does, it looks like we can arrive at correct religious belief without consulting Holy Scripture at all. Perhaps we can think that the latter provides us with sufficient evidence for the former, evidence that would not otherwise be available to us, humans. But then how are we to ascertain for any putative piece of genuine Holy Scripture x that x is a genuine piece of Holy Scripture? In other words, how to we determine - in a non-question-begging, independent way – the probative value of the putative piece? If we are able to find a way, then the idea that the evidence given to us from Scripture is only attainable through the study of Scripture seems to be under threat. And if there is no way, then we seem to have a problem reconciling Evidentialism with the standards of correct religious belief (again, if there is such a thing). This might then motivate taking the non-Evidentialist position, such that we can (without special pleading) consider correct religious belief to be determined by something other than its truth-conditions. But this idea is faced with parallel problems. First, it would seem to entail thinking it legitimate that we consider Holy Scripture not to be offering any evidence in favour of the truth of the proposition (e.g.) that God exists (rather (e.g. pragmatic) reasons in favour of being in the belief-state of believing that God exists). And this might well be considered Blasphemy, and inconsistent with what the very Holy Scripture in question decrees. Second, even if Prophesy is to give us non-epistemic reason to believe in God, we nonetheless need a way of determining genuine Prophesy from non-genuine Prophesy. And, again, the possibility of doing so threatens to undermine the idea that Prophesy teaches us things we could not otherwise have learnt. This (or a close analogue of this), along with the issue regarding what it is to be a responsible believer, I think, is the central problem of Medieval Islamic Philosophy. Islam, qua piece of Scripture, comprises both the Holy Koran as well as a series of Hadith. The latter are a series of reports concerning the practises and sayings of the Prophet Mohammed, of central importance to the Muslim religion, since Muslims are enjoined to treat the manner in which Mohammed lived his life to be exemplary and something to be emulated as much as possible. These Hadith were collected mostly after the death of Mohammed, and partly during the Abbasid Caliphate, also the time when the Islamic civilisation undertook a translation project of Greek Philosophy into Arabic. Perhaps then it will be considered no historical surprise that the Falasifa were acutely concerned with the issue around how to ascertain genuine testimony about 21 There are some important concerns regarding how best to think about sufficient evidence, as well as how to think about what it is to believe on the basis of such evidence. Unfortunately, I do not have the space to discuss these, but see Turri 2011 for useful discussion. 8 Prophetic activity. However, as I hope to show, thinking of them as engaged with this and related issues around the ethics of belief will help bring interpretative clarity as to their enterprise, as well as bring interesting arguments to bear on the Modern debate. I now turn to giving a sketch of their relative positions seen through this prism. 3. The Ethics of Belief in Islamic Philosophy 3.1 Imān vs Islam A Famous Hadith from Caliph Umar22 goes as follows: The Prophet once came across a stranger who asks him: What does Islam mean? The Prophet answers: “Islam means that you should testify that there is no God but God and that Muhammad is the Messenger of God, that you should observe the prayer, pay the zakat, fast during Ramadan, and make the pilgrimage to the Kaba if you have the means to go.” The stranger agrees, but asks: what is meant by belief (Imān)? The Prophet replies that it means that you should believe in God, his angels, his Books and his Prophets. [The stranger turns out to be Archangel Gabriel]. There are lots of ways in which one could take the broader meaning of this Hadith. One, I think very natural, way of taking it, is as saying that acting merely in accordance with the prescriptions one finds in the Koran (such as testifying that there is no God but God, or performing the pilgrimage to the Kaba) is not sufficient (though necessary) for “walking the straight path”, for being a good Muslim. These deeds need to be performed with certain accompanying beliefs – these are propositions we ought to believe.23 Perhaps a converse way (a way that I have not seen suggested elsewhere) is that Islamic belief just is a matter of performing the relevant required actions. As such, we might take the passage to suggest that the Islamic view of belief (at least as far as religious belief is concerned) is a thoroughly dispositionalist one, where ‘S believes that p’ just means that S has a set of behavioural dispositions associated with a belief that p. For instance, believing that there is a cake in the fridge just means that (for someone who likes cake like me) that I will be disposed to open the fridge and eat the cake when hungry.24 A yet another way of reading the passage is as telling us that being a Muslim is to be in a state of knowledge (since knowledge = Islam) and as such is not a matter merely of faith (where Imān is taken to translate as faith, rather than belief, or where ‘belief’ is taken to be synonymous with ‘faith’ when used in religious contexts). This, broadly speaking, looks to commit one to an evidentialist position, according to which all beliefs, including religious beliefs, are subject to only epistemic standards of correctness. 25 The related, obverse 22 Cf. retrieved from www.hadith.com Though the requirement may be belief-in, say, God, rather than belief-that God exists. 24 For a modern defence of this view, cf. Schwitzgebel. For a related more general ‘pragmatic’ account of the contents of our mental states cf. Stalnaker 1984. 25 Although we can make the caveat that religious knowledge is discoverable via special evidence, not the evidence acquired by the senses as with ordinary, quotidian ‘knowledge’. 23 9 interpretation would be that it tells us that faith, Imān (not knowledge) is the central obligation involved in being a good Muslim. This latter view then looks like a commitment to Fideism, the view that certain beliefs (in this case, particular religious beliefs) are a matter of faith, and thus that their justification outruns evidential considerations. As I mentioned in the preceding section, Fideism is a variety of NonEvidentialism. Franz Rosenthal has put together overwhelming evidence that the prevailing Muslim view is that ‘Islam’ and ‘Imān’ are synonyms, and what they denote is identical. This may give the prima facie impression that the prevailing Muslim view in the ethics of belief is a non-Evidentialist one. But as Rosenthal himself argues, that would be a mistake. First, because one can find passages where one finds explicit endorsement for the idea that the two concepts need to be kept apart26. Second, and this is an idea that figures rather pervasively in the study of Medieval Islam, it looks likely that there were factions within Islam in this period who did endorse the distinction, but felt unable to say so explicitly (though we may discern their endorsement by engaging in their work exoterically). As Rosenthal puts it: Those who felt that faith was to be kept separate from knowledge were the ones who were inclined to favour rationalistic or esoteric knowledge and to play down the importance of formal faith. Since “faith”, however, was a sacred term of religion, it was only prudent for such thinkers to avoid public discussion of the term outside their own groups whenever possible, and not to use it when belief was under discussion as an element of epistemology. Thus, we rarely find clear-cut statements of views on the distinctiveness of knowledge and faith. (Rosenthal 2007, p. 108). If Rosenthal is right about this, and I think he is, we can here begin to see a possible bifurcation of views in the Islamic self-understanding that parallel the distinction in the Western ethics of belief between Evidentialist and Non-Evidentialist understandings of what it is that renders belief justified. In the next three sections, I further explore this possibility. 3.2 Islamic Evidentialism The Abbasid Caliphate ruled the Islamic empire from 750 up to 1258 when it was broken up by the Mongol invasions (the Mongol sack of Baghdad took place in 1258). The Islamic empire was enormous during this time, encompassing most of the Middle East, Persia, and North Africa (Islamic Spain was mostly an independent Caliphate during the Abbasid tenure). The Islamic empire had grown at an astonishing speed from 622 to 750, first under the Prophet Mohammed, then the Rashidūn (“rightlyguided”27) Caliphs, and then under the Umayyad dynasty. There was little scope for The Koran (17:36/38) says, for instance: “Do not go after that of which you have absence of knowledge, for hearing, sight, and heart, all these are held responsible”. 26 This is mainly found in (perhaps surprisingly) in the writings of the Ikhwan (cf. Rosenthal 2007, p. 105), an early 20th century militia who made an important contribution to establishing Ibn Saud as the ruler of his new Kingdom (Saudi Arabia). 27 The “rightly-guided” Caliphs, are those four or three who immediately followed Muhammad, and were either close companions of his, or his genetic descendants. Here, of course, is where the split between Sunni and Shia Islam occurs – the former acknowledging the Prophet’s nephew ‘Ali as a legitimate Caliph, and maintaining that the Caliphate should 10 further significant territorial gain for the Abbasids (having vanquished the Umayyads), and their attention seemed to be turned inwards, towards glorifying their dynasty from within. They made Baghdad their capital (it had been Damascus under the Umayyads) which they sought (and succeeded) to turn into the world’s foremost centre of learning and culture – they built there the famous Bayt al-Hikmāh (the House of Wisdom), a vast library, and a centre for the translation and study of the Greek scientific and philosophical works the Islamic world had come into contact with (often originally in Syriac) as they made territorial gains in the Levant and, in particular, on taking Alexandria, with what was left of its wonder of the world library. And the Abbasids during this time – especially the early Abbasid Caliphs Hārūn alRashīd (786 – 809), al-Māʿmun (813-833), and al-Muʿtaṣim (833 – 842) – sponsored and gave extensive patronage to individual scholars, translators and philosophers, many of whom they invited to their courts28. Thus they gave rise to a period often referred to as the “Islamic Golden Age”, or sometimes the “Islamic Renaissance”. Muʿtazilite theology became the state-sponsored, official interpretation of Islam and was quite vigorously enforced.29 Muʿtazilite theology was an especially rationalistic theology, in the sense that it promoted the role of epistemic reason in religious faith.30 As Montgomery Watt (1984) notes, European scholars in the late 19th century came to have a very sympathetic attitude to this school, about which there had been until then little knowledge – in a 1865 account, the Swiss scholar Henrich Steiner had even qualified them as “the free-thinkers of Islam” (cf. Watt 1984, p. 46). Such European scholars saw in the Muʿtazilites an appreciation of the values of free-will and personal moral responsibility, which looked to be in accord with 19th century European values, and so offered the possibility of a rapprochement between the worldviews of Islam and the West. However, as Watt also points out, modern Western scholars of Islam have discovered that this view was overly idealistic, and that the Muʿtazilites far from being “free thinkers” were in fact pretty dogmatic Muslims, zealous in their vision of the faith (cf. test/inquisition point again). Nevertheless, they were instrumental in bringing Greek thought into the Islamic Civilization, as Watt puts it: While it may be difficult for the Western scholar of the last quarter of the twentieth century to share the enthusiasm for Muʿtazilism of the scholars of a century earlier, it certainly made an outstanding contribution to Islamic thought by the assimilation of a large number of Greek ideas and methods of argument. This was essentially the achievement of the great Muʿtazilites of the Golden Age (Watt 1982, p. 54). Further, there is something indeed ‘rationalistic’ (in the sense above) about what came to be their core theological beliefs. These consisted of five essential tenets, the first two of which seem to have been given lexical priority: (i) (ii) God is Unity, One (Tawḥīd). God dispatches Justice (ʿadl). have been hereditary in the first instance (‘Ali the only hereditary candidate was not immediately installed as Caliph), while the Sunnis deny this. 28 For an account of why this might have been politically expedient, c.f. Gutas 1998. 29 Footnote on ‘the test’, cf. Adamson 2015. 30 For a good account of Muʿtazilite doctrine, cf. Watt 1985, chapter 8. 11 (iii) (iv) (v) There is a “promise and the threat” of Paradise and Hell. One should adopt an “intermediary position”. One must “command the right and forbid the wrong”. Principles (iv) and (v) require a little bit of further explication. Principle (v), simply, is about the need (indeed the obligation for Muslims) to impose justice via any means (consistent with justice) necessary, including the use of force. Principle (iv) is rather more complex. The Arabic name ‘Muʿtazila’ translates as ‘withdrawers’ or ‘those who withdrew’. This refers to an (arguably folkloric) account of how it was that the Muʿtazilites came to be known as people who had a particular response to a theological question. The question was something like the following: can someone who has sinned be appropriately considered a true believer? This, of course, is importantly related to the issue of how to interpret the difference (or if there is a difference) between imān and Islam as it appears in Caliph Umar’s Hadith. The issue being about whether one’s actions must accord with one’s belief that p if one is said to believe that p. As I hinted at earlier, answering the guiding question above in the negative might well come with a commitment to what modern, Western epistemologists call dispositionalism about belief, or even a broader more pragmatic account of the content of our mental states. This sort of view would then support the idea that faith and religious knowledge (here thought of as adherence to religious practice) are identical, such that what determines correct (at least, religious) belief outruns evidence - it is a matter of faith. Being a true believer is a matter of acting in accord with Koranic obligation, and such actions we can choose to commit, or choose not to commit. As such, belief can be considered voluntary (a function of our actions, or our dispositions to action), adequately the object of blame and praise, and, further, has non-evidential correctness conditions – viz. compliance with strictures about how to behave. The Muʿtazilites, however, did not take this line, and thought that there was no determinate answer to the question whether someone who has sinned can be considered a true believer. They literally ‘withdrew’ from debates on this question, and proposed an ‘intermediate’ answer - that is, they withheld judgement with respect to the question. The position had some very serious political implications at the time. First, it meant that the Muʿtazilites did not have to take sides on certain pressing contemporary arguments, such as whether Uthmān has really a heretic31, or whether to accept that ʿAli was properly a Rashidūn Caliph. As such, until 850 when the more orthodox Sunni Abbasid al’Mutawakkil became Caliph, the Abbasid Empire had been one where a Sunni/Shia quarrel had been relatively marginal. Second, it allowed the early Abbasid establishment to freely be seen to sponsor and give patronage to scholars whose views might be considered to have been to close to blasphemy, or even plain apostasy for comfort. During this time one could not be legally tried for apostasy if one explicitly claimed allegiance to Islam, even where one’s actions seemed to suggest otherwise, or even if one had other beliefs that seemed to contradict it. Significantly, this was not the view of the Hanbalites, who were persecuted under the Abbasids, though through the re-invigorating work of Ibn-Taymmiyya were to become the thinkers of choice for the 18th century Wahhabist movement (and whose views were to become dogma in the modern-day Saudi-Arabia). However, regardless 31 The Kharijites, among others, accused the Caliph Uthmān of having sinned due to his not having meted out to other sinners the punishments that are prescribed in the Holy Koran. 12 of the progressive political implications of their view here, since the Muʿtazilite doctrine recommends the suspension of judgement with respect to the identity of faith and religious knowledge, or of belief and action (and dispositions to action), one cannot yet firmly qualify them as Evidentialists proper. However, their emphasis on God’s Justice, Principle (ii), I think does this work. I think the emphasis on this principle is what made the 19th century European scholars on Islam so partial to the Muʿtazilites. For the Muʿtazilites here seem to be endorsing the ideal of personal freedom and moral responsibility, and, especially, the Kantian ideal that “’ought’ implies ‘can’”, that we are only morally responsible over things within our voluntary control. A Just God punishes those who have chosen to sin, and rewards those who have chosen to do good, and abide by God’s law. If Zayd is forced to sin, and could not have done otherwise but sin, then it is an unjust God that punishes him32. Now, as we discussed earlier, the same principle seems to apply to our beliefs as well as our actions. The issue relates directly, I think, to how we are to think of the difference between infidels (kufrs), and apostates (riddahs), and on whether and how to punish them. Infidels do not believe the religious principles of the Koran, but do they do something morally objectionable in cases where they could not but have failed to have that belief – that is, when there was no way that they could have had access to the Koranic revelation, or to testimony of the Koranic revelation. The Muʿtazilites – and here is where the 19th century perception of them qua “free thinkers” parts company with reality – took a hard line on this, and thought that in many cases the infidels were indeed doing something morally wrong in failing to have Koranic belief [ref/examples]. What must underlie this – compatible with ought implies can - is the thought that through their use of reason, that is, by considering the evidence around them, they would have come to have beliefs that at least resemble (if are not completely identical to) Koranic belief in all but name. And they could have come to acquire these beliefs even when they had no way of accessing (directly or indirectly) the Koran. People who fail to have Koranic belief are simply failing to exercise their capacity of reason, and since whether or not we use our capacity of reason is within our control, we can be – at least indirectly – held to blame for having beliefs that we would not have believed, had we exercised reason33. It is important to note that there is a thematic connection here, with respect to beliefs that at least resemble proper Koranic beliefs, between infidels and apostates. Consider whether those who introduce “innovation” [bidʿah] into religion – what Islam considers heresy – are also apostates. The modus operandi for the Muʿtazilites on this issue, at least politically speaking, was to here take a more lenient stance, and claim the matter was in many instances indeterminate, subject to suspension of judgement. At least the early Muʿtazilites seemed to take this line, as we briefly discussed, with respect to Shia Islam. The key determinant of apostasy here being about whether a putative 32 As many scholars have noted (cf. Adamson 2015, Frank 1983), a good way to think of the position here is as taking the first horn of the Euthyphro dilemma: does God love the good because it is good, or is the good, good, because God loves it? As we will shortly discuss, an opposing theological group in Islam – the Ashʿ ārite theologians – take instead the second horn, as well as espouse a compatibilism with respect to moral responsibility and a kind of determinism. 33 This is because there are two ways in which blame supervenes on obligation: we are blameworthy simply for breaking an obligation, and we are also blameworthy when a state of affairs occurs that would not have occurred had we not broken an obligation. (Cf. Alston 1989, and Zimmerman 1996). 13 apostate has done all that he can with respect to arriving at correct belief: if he seems to have done, by having beliefs that at least resemble the correct Koranic ones, then – though he may still be guilty of lesser crimes, one cannot charge him with apostasy (the matter is too indeterminate for us humans to judge, to be resolved only at day of judgment), and he thus escapes the death-penalty. Again, the keystones here are that ought implies can, and that the teachings of prophesy do not tell us anything that cannot be learnt – at least in approximation – via the use of reason alone. And part of the relevant indeterminacy here also concerns the issue regarding whether one can sin and be a true believer. The view that we cannot – at least in certain cases – for certain judge whether someone believes as she does in good faith, that is, has done all that is within her control in order to arrive at what she thinks is correct belief, seems to cohere better with a view of belief according to which the latter is a private, occurent mental state34. This all adds up, I think, to a reasonable case for taking (at least the pre-850) Muʿtazilites to be committed to the view I earlier called Evidentialism. The Muʿtazilites case for Evidentialism, premised as it is on the idea of God’s Justice, as I have tried to suggest, looks to be vulnerable to certain philosophical troubles. Among them are at least the following: First, it does not look like the Muʿtazilites left much scope for indeterminacy when it came to their own core doctrines – there was hardly much leniency given with respect to what they considered incorrect answers to their inquisition. However, notice that the guiding question of their inquisition was not: do you accept the principle of God’s Unity? But rather the more indirect and intellectual one regarding whether the Koran was created or not. Understanding how this latter question relates to the issue regarding the Oneness of God requires some thought, the application of the instrument of reason. Arguably, this could be said to demonstrate the Muʿtazilites primary concern with the latter as the proper determinant of faith, rather than blind commitment to rote-learning, even if the consequences of their inquisition were obviously going to militate against that end. Second, one might wonder what the fate – in terms of heavenly rewards and punishments – the Muʿtazilites could think were appropriate for those who die before being given the opportunity to enact any free choice (doxastic or otherwise)… Third, what role could the Muʿtazilites think that Prophesy could have, once they seem to have ruled that one could in principle reason one’s way into correct belief? Perhaps they could hold that it role is to guide reason (in a resonant manner perhaps to how Socrates guides the slave-boy of Plato’s Meno to correct opinion35), but this will look deflationary, to say the least, to those for whom revelation has an essentially indispensible role36. But, the matter is a vexed one for any kind of Evidentialist - the more literally minded Hanbalites seemed to think that the word of Prophesy was to be 34 Though, in particular when one allows dispositions to action to play a role in determining the contents of belief, it may well be sometimes difficult to judge what a subject believes just by investigating their actions, even if belief is just a function of how one acts, or is disposed to act. 35 Ibn Tufayl (a 12th century al-Andalus philosopher) seems to make this point in his philosophical novel Ḥayy ibn Yaqẓān. For an interesting discussion of the latter (taken in comparison with Daniel Defoe’s Robinson Crusoe) see Leaman 2009. 36 Abū Bakr al-Rāzī (854 – 925) was a philosopher and medic who seemed to bite the bullet and to accept that prophesy has such a minimal role. He was very unpopular for this and his contemporary critic - Abū Hātim al-Rāzī - calls him a heretic in his work The Proofs of Prophecy. 14 taken as literal truth, as factively representing the world, but an aspect of the world (indeed the vital aspect of the world), that cannot be cognized through the use of reason alone.37 But if this is the case, then it looks like the possibility of any moral, normative judgment on the beliefs of infidels is obviated. Whatever the philosophical tenability of their views, the Muʿtazilite prevalence as the official theology in Islam during the early Abbasid Caliphate is certainly a central determinant of the rise of Falsafa in the Muslim world. The very first of the Falasifa seems to have been al-Kindi (801 – 873), “the Philosopher of the Arabs” and a clear proponent of Evidentialism. The nickname “Philosopher of the Arabs” was given to him in reference to his good Arabic pedigree, a descendent of the Kindah tribe, one of whose chieftains was a companion of the Prophet Muhammad. Al-Kindi was born and educated in Basra, and worked in Baghdad under the patronage of the Abbasid Caliphs – especially al-Māʿmun and al-Muʿtaṣim (becoming tutor to the latter’s son, and so enjoyed a rather privileged position at court). Al-Kindi’s main concern, qua first Islamic philosopher, was to demonstrate the utility and legitimacy of the study of philosophy within a culture essentially identified with Prophesy.38 As such, he was especially concerned with the relationship between faith and reason, and thus with the question: what justifies belief? – the guiding question, of course, of the ethics of belief. His position here I think is neatly characterized in the following passage, in a work entitled On First Philosophy (essentially a work in Philosophical Theology): We ought not be ashamed of appreciating the truth and of acquiring it wherever it comes from, /even if it comes from races distant and nations different from ours. For the seeker of truth nothing takes precedence over the truth, and there is no disparagement of the truth, nor belittling either of him who speaks it or of him who conveys it. (The status of) no one is diminished by the truth; rather does the truth ennoble all. (al-Kindi On First Philosophy, p. 58). Two important things are implicit in the passage above. The first is that al-Kindi’s Muslim contemporaries were inclined to view philosophy as a foreign (Greek) import, and thus potentially anti-Islamic. Thus the reference here to “nations different to ours” and “races distant”. Indeed, the word given to Philosophy Falsafa is an Arabized version of the Greek word Philosophia, and this would have further enhanced the sense that philosophy was not properly Islamic, than had a direct Arabic translation of ‘Love of Wisdom’ been used instead. [re-write this sentence, add reference to Adamson 2007]. The second is that all beliefs, including religious beliefs, are to be assessed relative to their truth conditions. That is, we have good epistemic reason to hold religious beliefs, especially for al-Kindi, religious beliefs as they are expressed by the Muslim faith. This is what surely underlies the confidence that Muslim belief can withstand the study of non-Islamic works, since nothing that is true in these works can contradict Koranic belief if the reason for Koranic belief is purely epistemic. If one takes Koranic belief in this way, and moreover one takes the Koran to mandate 37 Where Prophetic knowledge seemed to contradict ordinary human knowledge, or itself, then bi-lā –kayf applies: the contradiction is only seeming to our human minds, which cannot see comprehend how there is really none, and such that we should trust that there is none. 38 Sayyed Hossein Nasr qualifies the central problem of Falsafa as that of doing “Philosophy in a Land of Prophesy” (Nasr 2006). 15 the seeking of truth then all true propositions must be considered compatible with Koranic belief. This is of course an Evidentialist position, since the idea is the Koran prescribes that for any belief it is correct just in case it is true. Since, for al-Kindi, we can take it to be obvious that Islamic belief is true, this Evidentialist line amounts to a defence for the study of philosophy (even qua foreign import), since in so far as the latter discovers the truth it is Islamic, or at least perfectly compatible with Islam. As with Muʿtazilite Evidentialism, al-Kindi’s Evidentialism looks to be vulnerable to some problems. It makes no further progress as regards the issue of the unique function of Prophesy39. Further, one may wonder whether the position pays due heed to the possibility – well known to anyone who has read Plato’s Meno - that at least part of what gives true beliefs their epistemic value has to do with how those beliefs were acquired, and whether they have due “resilience”. These problems are addressed in the works of the more famous of the Medieval Islamic Philosophers (namely al-Farabi, Avicenna, and Averroes), in what follows I will argue constitutes a unified position that I term Moderate Evidentialism. 3.3 Moderate Evidentialism The philosophers al-Farabi (872 – 950), Avicenna (980 – 1037), and Averroes (1126 – 1198) are probably the most well known of the Islamic philosophers. I think that they have a view on the ethics of belief, and that their view constitutes a broadly unified view (thought there are certainly important differences between them). AlKindi before them, had gone some way toward making the study of Falsafa (with all its foreign connotations) to an Islamic audience - but he had done more than this. He had also really set the contours of the overall intellectual agenda Falsafa was going to take, at least in two important ways. First of all, he was keen to present the inherited philosophies of Plato and Aristotle as constituting a coherent, harmonious picture, such that Falsafa as a whole was more palatable to his Islamic contemporaries.40 This made him particularly germane to the “Platonising Aristotelians” of later antiquity. Second, and partly as a corollary of the point above, he emphasized a link between Aristotle’s Posterior Analytics and De Anima as regards how we come to have abstract, mathematical knowledge and also then with the emanationism found in NeoPlatonic works such as those of Proclus and Plotinus.41 I will explain the connections 39 Al-Kindi does address the issue by claiming that prophetic knowledge differs from the knowledge of ordinary humans in that the former and not the latter is attained instantly, “without study, effort, or human methods” (This from On the Quantity of Aristotle’s Books translated by Adamson in Adamson 2007, p. 43). This hardly solves the problem, however, since it renders the function of prophesy to be a matter of sheer expediency. That is, the attainment of knowledge faster and the less effort than otherwise. This means that humans might well have gained the knowledge that is said to be in the Koran independently of its having been revealed to us, but it would have taken us longer, and it would have been a more painful process. 40 This point is made by Adamson 2007, p. 28: “it would have been exceedingly inconvenient for [al-Kindi] to admit that Greek thought itself was riven by disagreement. Rather he needed to present it as a single, harmonious body of true doctrines”. 41 This may also have been a historical accident, since al-Kindi would have had access to the Theology of Aristotle – a pseudo work, circulating at the time, containing abridged Plotinian passages passed off as Aristotelian, cf. Fahkry 2004, p. ???. 16 in detail in the next chapter, but this set the scene for a unique body of philosophical views, quite distinct in fact from those of any Greek philosopher. Further, it set the scene for a particular, psychological (almost naturalist) project of accounting for the mechanisms that make Prophesy possible. This was a project taken on by all of alFarabi, Avicenna and Averroes, and means that for all three of the great Falasifa the question of Prophesy came down to accounting for the unique (but human) qualities that set Prophets apart. I think the key to understanding their position is al-Farabi’s notion of certainty, which contains strict K-K (higher-order knowledge) requirements, and which understood in terms of Aristotelian demonstration will contain a knowledge-wh element. Again, I will present how I think of al-Farabi’s account of certainty in detail and defend it against rival interpretations (especially one found in Black 2006, and Strauss 1959) in the next chapter. Here my aim is merely to present the outline of the account, and how it amounts to a position in the ethics of belief. There is a lot of controversy as to how to understand the various positions here, and so there is no canonical way in which to take them, but here is how I understand things (as I say, I will defend the account further in the next chapter): For al-Farabi, only Prophets can have proper certainty (yaqīn) distinguished from mere psychological assent. This is because only their knowledge can satisfy the demand he makes that in order to have certainty that p it must be necessary that one could not but have been certain that p – in other words, I interpret, that certainty that p must have been pre-ordained by God. But it is also because only Prophets can really know why and how the operations of the intellect can deliver us direct knowledge of first principles such as the principle of noncontradiction. The rest of us humans are precluded from such knowledge. This is because the Prophet has a far superior faculty of imagination than ordinary persons. There is a certain unity or coherence to all being (a doctrine that later in Islamic philosophy becomes very important, often termed waḥdat-al-wujūd – “the unity of being”42) that this superior faculty of imagination brings to relief, and lets the Prophet cognize how and why our intellects are able to intuit (hāds) abstract truths. And this superior faculty of imagination has two further roles: (i) it gives the Prophet a certain rhetorical ability, in that he can understand how, say, complex propositions are related to simpler ones, and thus explain the former in terms of the latter; (ii) it gives the Prophet a certain practical ability, in that he can, in grasping how the world is connected up, also understand how best to make certain ideas become reality in our world. Thus the very property that enables the Prophet to attain certainty, is also the very property that gives the Prophet certain practical abilities. This has certain important corollaries: (a) that human perfection, as exemplified by the Prophet’s condition of certainty, is a composite of both practical and theoretical perfection; (b) that the Prophet is the ideal political leader. The last two claims amount to a position I’m calling Moderate Evidentialism in the ethics of belief. Combined, they give us an answer to the issue canvassed earlier as regards the unique role of Prophesy for an Evidentialist account of what we ought to believe. We do need some Aristotelian assumptions (as found in the Nicomachean Ethics) to work as background assumptions, namely: [naturalism] human good is human perfection in accord with human function and nature, such that knowing 42 The doctrine looks to be a corollary of the idea that the physical world is an emanation from the One. 17 human nature is necessary for knowing the good; [particularism] there are no rules for working out what one ought to do (consistent with the good) for any individual case – this is a matter for practical wisdom (Greek: phronesis) the bearer of which can intuit how to apply general moral knowledge to particular situations, and can only be attained through experience. Now, for al-Farabi, us humans, as imperfect creatures43, are prone to lack phronesis, and only the Prophet with his superior capacities is able to really have infallible phronesis – partly because only he has proper certainty about human nature, and partly because only he has the requisite practical knowledge about how things are best carried out. Further, because there are no general moral rules available to us, we must know about human nature, such that we can then gain phronesis. In other words, we cannot act in accordance with good simply by following a set of laws that would ensure that we did, so we need to have at least something resembling knowledge of human nature. Since no human can attain certain knowledge of human nature, this explains the need for Prophesy, qua rhetorical device for teaching us moral truths that we would not otherwise believe. But since the knowledge we gain in this way is far from reaching the level required for certain knowledge, putting it into practice will not really amount to proper, infallible phronesis, such that weakness of the will (Greek: akrasia) becomes possible for us. This gives Prophesy a further role – to give us a manual as to how to live such that we are in control of our desires. The rules the manual gives us on how to live, however, can only function in tandem with the theoretical knowledge it teaches, such that - in isolation from certain bits of theoretical knowledge - they cannot be taken to constitute generalist moral laws. The reason I call this view Moderate Evidentialism is that it is Evidentialist in the sense that the correct rule for the Prophets, those form whom proper certainty is a possibility, is: believe that p just in case that p. This means that prophets ought never to believe propositions for which they have no (in this case utterly irrefutable) evidence. The Prophets thus have no false beliefs, and Prophetic knowledge contains no falsehoods. But the position is Anti-Evidentialist in this sense: for ordinary humans there may be times when one ought to believe propositions for which there is no evidence. This is because of the practical role of Prophesy: it tells us in part how to live our lives such that we can control our desires. Having certain beliefs may be necessary for our acting in the way prescribed. Having such beliefs is thus obligatory, but not strictly for epistemic reason. Further, proper epistemic reason, in the sense that it guarantees certain knowledge, is unavailable to ordinary humans – we must learn to make do with imperfect knowledge, or, to put it another way, merely justified belief. The solution to the problem of the unique role of Prophesy for the Evidentialist is thus solved like this: it is possible for us, ordinary humans, to reason our way to justified belief in religious and moral matters, but it will necessarily stop short of proper certainty, and so such knowledge cannot constitute proper phronesis either such that we are subject to akrasia, and require a manual to teach us how to control our desires – and in satisfying this latter requirement, Prophesy has a unique role. Can we not teach ourselves, however, the contents of the manual? No, since it is the product of a practical capacity that outstrips the capacity of any ordinary person. 43 “Of course as human beings we can be healthy minds on one day and sick souls on the next...” William James Pragmatism, p. 294. 18 As I have already mentioned, I think that al-Farabi’s position is very close to that of Avicenna and Averroes, both of whom I also qualify as Moderate Evidentialists. Avicenna spends relatively little space discussing ethics and political philosophy. He does discuss these subjects, however, in The Healing (Metaphysics) where he expounds the al-Farabian idea that the proper political leader is what he calls the “Prophet law-giver” amid an Aristotelian account of the human good. Avicenna also gives an account of Prophecy in the Psychology part of The Healing (or his De Anima).44 His account is broadly similar to al-Farabi’s, and heavily based on the connection made by al-Kindi between the “Active Intellect” of Aristotle’s De Anima, an Emanationist cosmology/metaphysics, and an account of how we can come to know abstract objects45. I will discuss all of this in the next chapter, but for now it may be worth noting that the main difference between al-Farabi and Avicenna on Prophecy, is that for the latter, new knowledge can be the product of Prophecy, whereas for the former Prophecy can only really facilitate understanding knowledge (in my interpretation, one should really say items of justified true belief, rather than knowledge) that one already has (and this understanding renders the justified true beliefs items of certain knowledge, since they now satisfy strict KK conditions).46 Averroes, I think, consistently, and vehemently, denies this Avicennan innovation. In general, his line on Avicenna is to think that he has departed from true Aristotelianism, and that made Falsafa vulnerable to the charges levied in al-Ghazali’s The Incoherence of the Philosophers (Tahāfut al-Falāsifa), which we will shortly discuss. It looks like there is a discrepancy, however, between what Averroes thinks in his earlier works (in particular his Epitome of the Parva Naturalia) and what he expounds in his later works (in particular in his Incoherence of the Incoherence – an explicit reply to al-Ghazali’s Tahāfut). In the Epitome Averroes quite explicitly denies that Prophecy can deliver new theoretical knowledge, and nor can it give anyone a figurative representation of theoretical truths, and so furnish the Prophet with any special rhetorical ability. Nor can Prophesy deliver us with rules for human behavior that lead to eudemonia (the human good, consistent with our function), indeed its function is limited to giving certain humans the ability to predict the future sometimes [footnote on dreams]. In the Incoherence of the Incoherence, however, Averroes explicitly says that “a religion through intellect alone would necessarily be inferior to religions derived from intellect and revelation” and that “knowledge received from revelation comes to perfect the branches of knowledge of the intellect” (Incoherence of the Incoherence p. 584). Further, he repeatedly says that certain knowledge cannot be attained by some persons “either because of inborn nature, habit, or lack of the means of study” and thus that for them “God has coined images and likenesses of 44 For discussion cf. McGinnis 2010, chapter 8. For more cf. Marmura 1963. 46 As Herbert Davidson puts it: “Avicenna likewise recognizes, and attaches the name prophecy to, knowledge that results when the emanation from the active intellect – or another supernal being – acts on the human imaginative faculty. But as an extension of his view that man receives intelligible thought directly from an emanation of the active intellect, he, unlike Alfarabi, recognizes, and names as prophecy, genuine theoretical knowledge imparted by the active intellect to the human intellect without the human intellect’s having to employ standard scientific procedures” (Davidson 1992, p. 117). Fazlur Rahman makes the distinction by calling revelation of new knowledge “intellectual revelation” and revelation bringing understanding to existing knowledge “imaginative revelation” (cf. Rahman 1958, p. 36) 45 19 things… that can only [otherwise] be learned by demonstration [through the use of theoretical science, broadly speaking]” (Decisive Treatise p. 59). This discrepancy has in turn resulted in there being a number of interpretations as regards what Averroes thought was the proper relationship between philosophy and religion, and thus his broader view on the Ethics of Belief. Philosophers such as Leo Strauss and Ralph Lerner (Lerner 1974, a work dedicated to Leo Strauss) think that really Averroes is a kind of proto-secularist who thinks that revelation is of very limited value, but that he was unable to make the case for this explicitly, but if one reads him closely enough one can see that he communicates this exoterically in his writings; Erwin Rosenthal (Rosenthal 1958) on the other hand seems to think that really for Averroes, Sharia Law trumps any claim of reason; Oliver Leaman (Leaman 1988, 1980) maintains a kind of Kindian vision of Averroes, where reason and faith are both equally efficacious, but different, paths to the same truth; Herbert Davidson also takes a quasi- Straussian line and thinks that the term ‘Prophet’ for Averroes, really denotes “nothing than the human author of Scripture; and the term revelation would mean a high level of philosophic knowledge” (Davidson 1992, p. 351). The Prophet on this reading is really a philosopher, who methodically works out “coolly and deliberately” (and not through an inspired imaginative faculty) how best to present his philosophical knowledge into language more suitable for those untrained in, or simply unable to do, philosophy. I think that Davidson is broadly right, but that we must remember how the enhanced imaginative faculty can in Prophesy (and dreams) give insight according to Averroes as regards the future (recall that there is no discrepancy about this issue between the later and early Averroes). It does so by granting us greater understanding of how the things we already know cohere into a bigger picture, and so also how best to turn our ideas into reality. Here, to illustrate, Averroes uses the example of a doctor who can apply his theoretical knowledge of medicine (parsed in universals) to come to bear on particular medical problems in the real world [ref!!]. It is this kind of ability that the enhanced imaginative faculty gives rise to, so that Prophecy turns out to look like practical knowledge (phronesis). But alone it is not phronesis proper, since it is not based on proper theoretical knowledge (unlike the doctor’s theoretical knowledge of medicine). As such, it cannot lead to eudemonia on its own – hence the need for theoretical science working with revelation according to Averroes. So ultimately Averroes also ends up with a Moderate Evidentialism, even if it is slightly different in detail to al-Farabi’s. The central difference, it seems to me, between al-Farabi and Averroes, is that for the latter ordinary humans (that is, non-prophets) can attain the best kind of knowledge (the philosophers can do so), while for al-Farabi only the prophets can attain proper certainty. This means that al-Farabi can better account for revelation’s unique function than can Averroes. For the latter, Prophesy can at best give us practical knowledge about how to best live our lives (and as such there may be beliefs that are justified via pragmatic, and not epistemic reason, if having those beliefs are constitutive for someone to have the requisite ability), but this is knowledge that is in principle available to everyone, albeit with toil, through experience. That Averroes’s position fails to adequately explain Prophesy’s uniqueness, explains why several theorists have taken him to be a more extreme Evidentialist than he actually was. 20 This interpretation chimes together well, I think, with what Averroes says about the Unity of Truth, and his well-known claim in his Decisive Treatise that “truth does not contradict truth”, or more completely: Now since this religion is true and summons to the study which leads to knowledge of the Truth, we the Muslim community know definitely that demonstrative study does not lead to [conclusions] conflicting with what Scripture has given us; for truth does not oppose truth but accords with it and bears witness to it. (Averroes Decisive Treatise p. 50). The reason that philosophical truth and Prophesy are in accord, or as Averroes puts it that “demonstrative truth and scriptural truth cannot conflict” is that the former is either the work of philosophers presenting their findings in the way most amenable to all, or else it is practical knowledge gained through inspiration (through dreams or heightened faculty of imagination). When the subject matter of scripture are theoretical truths, then they are either apparently in accord with theoretical truths known through philosophy, or they are not. If scripture looks to be in apparent contradiction with demonstrative truth (philosophy), we must conclude that the work of scripture is to be understood allegorically, since it too is the very product of philosophy. And the reason there is a need for scripture to be parsed allegorically at times “lies in the diversity of people’s natural capacities and the difference of their innate dispositions with regard to assent” (Decisive Treatise p. 51). This is why in his Middle Commentary on Aristotle’s Categories he claims that ordinary language predication on God’s properties are neither necessarily equivocal or univocal with respect to religious language predication on the God’s properties. Because religious language can be allegorical, it may simply be pro hens equivocal with respect to ordinary language. [Footnote explaining a bit, reference to “core-dependent homonymy” – Shields 1999]. When, on the other hand, scripture gives us practical knowledge, that is, commandments on how to live our lives “the situation is different…everyone holds that the truth about these should be disclosed to all people alike and to establish the occurrence of unanimity about them we consider it sufficient that the question [at issue] should have been widely discussed and that no report of controversy about it should have been handed down” (Decisive Treatise p. 53). In other words, Averroes is saying here that provided we have no reason to doubt the veracity of the claim of inspiration or of clairvoyant dreaming47, because these give us practical pieces of knowledge and so cannot in principle contradict items of theoretical knowledge, then there can be no further controversy about them. Practical knowledge depicted in revealed texts may look like (has the guise of) theoretical knowledge, but really it is not – memorizing a manual on how to swim is not to know-how to swim [footnote on Anscombe]. This means, incidentally, that the socalled “theory of double truth” attributed to Averroes by the 13th century European Averroists (such as Siger de Brabant, Dante, and Boethius of Dacia48) where the claims of science and philosophy could be considered to be contradictory with those of religion at the expense of neither, is rather wrong-headed, since, for Averroes, 47 “…true visions include premonitions of particular events due to occur in future time…this warning foreknowledge comes to people in their sleep from the eternal Knowledge which orders and rules the universe” (Decisive Treatise p. 55) [my emphasis, to highlight the practical nature of such knowledge for Averroes]. 48 For a helpful account of the doctrines of the various Latin Averroeists and how their views can be considered to constitute a unified view, see Marebon 2007. 21 scripture either says something that when properly understood is in fact the same truth as that of demonstration, or else it tells us something which can in principle not contradict the theoretical truths of philosophy proper.49 3.4 Islamic Anti-Evidentialism Ashʿarite theology is one of the most important, if not the most important, theological movements in Sunni Islam. It begins on its founder, al-Ashʿarī, defecting from Muʿtazilism and embracing Sunni Hanbalite theology, and its reverence for revelation over reason. As I have mentioned, the folkloric story is that he did so on considering the story of the “three brothers” – in short, one brother lives a moral life in accord with the Koran and ends up in Heaven, another fails to live such a life and ends up in Hell, and a third dies as a child and ends up in a sort of purgatory, an afterlife short of Paradise. Since the child has not had the chance to either show him or herself to be worthy of an afterlife in Heaven, the story is meant to be a reductio ad absurdum of the Muʿtazilite emphasis on God’s Justice and human Freedom of the Will. Unlike Muʿtazilism, Ashʿarite theology espouses a kind of voluntarism about the source of moral normativity. That is, the former, to put it in terms of the Euthyphro dilemma, takes God to love the good because they are good, such that there is an independent, source of normativity that exists outside of God’s will. As such, Muʿtazilite theology is often thought of as a kind of moral realism. The Ashʿarites reject this and take the other horn of the Euthyphro dilemma, and think that the good are good because God loves them. That is, God’s will is the ultimate source of moral normativity, hence the label ‘voluntarism’ [footnote on Hobbes]. Of course, taking each horn of the dilemma comes with having to deal with certain problems (else we could hardly call the problem in question even a prima facie dilemma). The Muʿtazilites have to deal with having ceded that there is an authority in moral matters that God merely responds to, and does not have power over, as well as the issue regarding how to think of the moral status of children, animals, and those others whose action fall somewhat short of being under full voluntary control, but who nevertheless have moral status. The Ashʿarites face the long-standing issue of whether they can account for God’s justice, an objection which is particular form of a more general realist strategy sometimes known as the “open question argument” [footnote G.E.Moore, Prichard & Aristotelian naturalism, more in line with mutazilite realism (naturalism still prone to open 49 Richard Taylor’s recent essay in places supports this interpretation: “Hence, there is no conflict between the truth which is found in the conclusion of a demonstration and any truly understood text of the Religious Law. There is no issue of “Double Truth” as was found in discussions of the Medieval West. Truth can be essentially and necessarily grasped as truth, and not only in an accidental way, by the practitioner of demonstration when practicing the art of demonstration correctly. Such a practitioner is the philosopher.” (Taylor 2000, p. 5). See Black 1996 for an account of the differences conceived by Philosophers between practical and theoretical knowledge in the Medieval East. 22 question argument, but did Mutazila realise this?).] (the Ashʿarites define the good in terms of what God wills). The Ashʿarites follow the Hanbalites in thinking, as Montgomery Watts puts it, that “faith (imān) consists of word and act, that is, profession of belief and fulfilment of the prescribed duties” (Watt 1985, p. 67). Put differently, they hold a sort of behaviourist, or dispositional, account of the contents of belief. This bears directly on the question regarding the criteria for ascertaining Apostasy, and meant the abandonment of the Muʿtazilite decree that professing belief in the Ṣhahāda was sufficient for being cleared of Apostasy. This once again enables a Sunni confrontation with Shia Islam, since even if proponents of the latter profess belief, their putatively divergent practises may be taken to militate against the proposition that they actually believe. This is in accord with Hanbali doctrine, as is their rejection of metaphorical readings of the Koran towards a more literalist understanding of the latter (where putative contradictions are passed of as beyond human comprehension – bi-lā kayf). However, in their voluntarism, they differ from the Hanbalites. The latter see that the works of Prophesy are statements of fact, propositions representing the deep nature of reality, and are as such to be taken an infallible testimony as regards these facts, that is, as constituting evidence that the relevant propositions are true. One cannot reason one’s way to having the appropriate beliefs because this evidence is uniquely available through this special testimony. This is in accord with Evidentialism, as I defined it earlier. The Ashʿarite line in the ethics of belief constitutes what I’m going to call Anti-Evidentialism, on the other hand. This is because, parallel to their line on the Euthyphro dilemma, they think that what we ought to believe is determined by what God wills. That is, they take a voluntarist line about doxastic normativity, such that one believes religious propositions because God decrees it, and not because there independent evidence that shows the propositions to be true. Thus, the view is much stronger (and entails) non-Evidentialism, since it claims that all beliefs are to believed for non-epistemic reasons, not only that it is permissible to sometimes believe for non-epistemic reasons. To summarise: Anti-Evidentialism: S’s belief that p is justified iff S has non-epistemic reason to believe that p. The relevant non-epistemic reason for Ashʿarites is clearly God’s Will, such that a S’s belief that p is justified just in case God wills that S believes that p. This is compatible, incidentally, with the idea that God commands that one believe on the basis of good evidence, since God’s Will is nonetheless doing all the normative work. A second difference between Ashʿarite and Hanbali theology is that proponents of the former, and not the latter, thought it legitimate to engage in rational methods/philosophy in support of revelation, or as Watt puts it “the support of revelation by reason” (Watt 1986, p. 65). This is in fact a consequence of their AntiEvidentialism and belief that in the uniqueness of the role of Prophesy. The combination of belief in the uniqueness of Prophesy and Evidentialism – as per Hanbalite doctrine – seems to entail a sort of exceptionalism about the evidence (through testimony) one receives in reading the Koran, where that evidence is unavailable by any other means. And this entails that there is no other way of properly engaging with religious matters than by consulting revealed texts. Rejecting Evidentialism (of the stricter variety – i.e. not Moderate Evidentialism) frees one from 23 having to accept this conclusion. This is not necessarily, as the Ashʿarites show, to place reason above revelation, since the use of the former is merely an instrument to demonstrate the necessity of the latter - – indeed in the last of three dreams that are said to have immediately preceded al-Ashʿarī’s conversion, Mohammed is said to have appeared to him and told him (angrily) not to reject rational methods altogether, despite having in the previous dreams recommended him to have more strictly followed the paths set out in the Hadith and Koran. A third difference is the Ashʿarite commitment to a proto-compatibilist account of free will. The Ashʿarite subscribed to an Occationalist metaphysics, according to which God intervenes at every point in the causal order of things – i.e. underlying every event, God is the proximate (immediate) cause. If God is the proximate cause of every event, and actions are events, if follows that all human actions are predetermined by God. And so the issue regarding God’s justice returns: if I had no control over whether or not I acted according to God’s law, is it really just fof God to punish me for violating God law? The Ashʿarite compatibilist answer is something like this (fabulously resonant with some of the work by the contemporary American philosopher Harry Frankfurt (Frankfurt 1971) and perhaps a precursor to Thomas Hobbes’s Classical Compatibilism): part of what it is to be human, and not an animal, is that we have higher-order desires and beliefs. That is, we can reflect on our own desires and sometimes come to endorse them, and sometimes come not to endorse them. My desire for gorging myself on dates all the time, for example, might not be a desire that I endorse, given that I think that I ought not to be so gluttonous. To act as one wants then is to act in such as way that one’s higher and lower order desires are in sync. When my base desires take over and I do gorge myself on dates, I no longer act as I want, but rather am just an animal and a mere product of the chain of cause and effect. The same can be said about our higher-order and lower order belief – when I have a belief that I think I ought not to have, I am no longer the agent of my own beliefs50. My action or belief s is free then just in case I have not been prevented from acting or believing as I wanted. Since God’s being the proximal cause of my action or belief x does not prevent me from acting or believing as I wanted, it is no obstacle to my acting freely. And, as such, no obstacle to my being held responsible for either my actions of my beliefs. 3.5 Moderate Anti-Evidentialism As I have already mentioned, one the most pressing problems facing Islamic AntiEvidentialism concerns how to discern true from counterfeit Prophets and Prophecy. This issue is an especially relevant one for Islam, given that it counts among its sources of Prophecy not just the Koran as revealed to Mohammed, but also the Hadith – a collection of sayings and accounts of the practices of the Prophet collected after the death of Mohammed. The traditional (and Hanbalite) view was that the Prophet was to be distinguished by his performing miracles, testimony of which was passed 50 Although cf. Shoemaker 2009 for an argument against this possibility. I argued that one’s higher and order beliefs can actually come apart in my Ph.D. thesis (Booth 2006), and for an newer, interesting defence see Hunter 2011. 24 on through tradition (tawātur). And, further, by the Prophet’s perfect moral character and behavior. This issue is especially problematic for the Ashʿarites, given their AntiEvidentialism, since they had to be forced to reject the traditional view, and in particular the idea that the sign of true Prophecy is at least partly to be found in the perfection of the Prophet’s character and behaviour. The latter is to admit that there is a source of moral normativity that exists independently of God’s Will, anathema to the Ashʿarite world-view. And, further, it looks like the Ashʿarite view has a problem with appealing to miracles also. Not only because of the inherent difficulties in identifying proper miracles (as opposed to sheer trickery)51, but also because the appeal to the performance of miracles as a means of verifying the true Prophet from the impostor, is to appeal to good evidence for the veracity of the Prophet’s claim. And this is to accept that God’s Will is not the ultimate source of doxastic normativity (grounding what we ought to believe), and as such to renege on Anti-Evidentialism. Perhaps, in light of this, the Ashʿarites could restrict their voluntarism to matters concerning the normativity of action, and not extend it to belief. But this generates inconsistencies with other with other Ashʿarite doctrines, such as the idea that belief is really an action, or to be evaluated in terms of behaviour or behavioural dispositions, and thus obviating the Ashʿarite confrontation with Muʿtazilite theology. It is this problem that makes later Ashʿarite thinkers such as Fahr al-Dīn al-Rāzi (1149 – 1210) and, especially, the great Imam al-Ghazali (1058 – 1111) diverge somewhat for the traditional view. Al-Ghazali is often claimed to be the most important Muslim after Muhammad, and is the author of a fatwa levied against the Falasifa (it occurs at the end of his The Incoherence of the Philosophers). I will here focus on al-Ghazali’s view here [footnote, since al-Razi claims to be following his views; though very much like falasifa’s], although it has been notoriously difficult to say exactly what his overall take on Prophecy was. For instance, in his well-known book on Prophesy in Islam, Fazlur Rahman writes: “Al-Ghazālī is a most difficult author, if not an outright impossible one, to understand in any coherent manner” (Rahman 1958, p. 94). It looks, prima facie, a consequence of al-Ghazali’s famous “crisis of faith” as regards both rational theology (kalām) and philosophy (detailed in his autobiographical work Deliverance from Error) that some of his writings will appear contradictory. For instance, he seems to condemn the Falasifa vehemently in such works as The Incoherence of the Philosophers but yet seems to rely on their views in other works such as Fayṣal al-Tafriqa However, recent scholarship, especially by Frank Griffel (Griffel 2004, 2012, 2015), has shown that al-Ghazali’s view here is in fact tractable and quite coherent, or (at least) consistent. I think his view amounts to what I will call Moderate Anti-Evidentialism: the view (as we shall see) that for all propositions p, except one very special proposition, p should be believed because God wills it. As with the general Anti-Evidentialist line here, the view is that even if God commands us to believe in accord with our evidence, it isultimately – God’s will that is the source of doxastic normativity. The view here is that one ought to believe this special proposition just because the evidence supports it, such that the evidence itself is normative. I think that al-Ghazali’s position is summed up by this claim of his: 51 Footnote of Avicenna and al-Ghazali on miracles. Both think they are possible (analogue of mind-body problem, but for al-Ghazali, the prophet can change the essential properties, not just the accidents, of material things) but not easy to identify. Cf. Griffel 2004, p. 115. 25 The obligation to believe with one’s tongue and heart is an act, like other acts, that can be founded on probabilistic evidence. (Al-Ghazali Moderation in Belief p. 210). Two things are going on here, I think. The first is that as per Ashʿarite AntiEvidentialism, the normativity of belief is held to be no different to that of action. The second is the claim that both obligation to belief and action then can be founded on evidence. That is, the having of evidence that p can make it – just on its own – the case that one is obliged to perform certain actions (or have certain beliefs, which are for al-Ghazali a certain kind of action). Now, al-Ghazali wants of course to maintain that Prophesy has a unique role, and yet maintain that the legitimacy of the Prophets has an evidential base. He attempts to do this, it seems, by holding that a good portion, though not all (e.g. not theoretical knowledge about what happens in the afterlife) the theoretical claims of Prophesy are available by consulting independent evidence. Further, no practical knowledge is available independent of knowledge of Prophecy. This means that we can ascertain the Prophet’s legitimacy by (a) seeing a sort of verisimilitude between the theoretical claims of the Prophet and what our evidence indicates; (b) in practical matters, seeing that following the practical rules of the Prophet leads to their desired effect, namely to “purify our hearts”; (c) we see that the Prophet not only asserts truths, but does so sincerely (his assertions are not true accidently). In (b), however, what we gain is really theoretical knowledge about the effect of the Prophet’s works. As Griffel (Griffel 2004) shows, al-Ghazali’s account with respect to (c) is heavily based on Avicennan psychology, where different modes of mental representation are enumerated, and sincerity is thought to consist in coherence between what is mentally represented to A and what A asserts. Once we have verified the Prophet’s trustworthiness (ṣidq) on these evidential grounds, we then ought to believe and act on the basis of what the Prophet tells us is the will of God. That is, the source of both the normativity of belief and action is ultimately the will of God. Now, Griffel thinks that this account neatly gets the Ashʿarite position out of jail, since it “does not violate the [Ashʿarite] principle that there is no normative practical knowledge independent from revelation” (Griffel 2004, p. 142). However, it seems a bit ad hoc to maintain that the Ashʿarites thought that only practical normativity, and not doxastic normativity, has its ultimate source in revelation. Further, it is clear, as the quote above from Moderation in Belief shows, that al-Ghazali wants to (rightly, in my view 52 ) conflate doxastic and practical normativity. So I think al-Ghazali’s position is really to concede that there is one piece of normative knowledge that can be gained independently from revelation. This normative blind-spot is belief in the proposition that the Prophet is a genuine Prophet. The evidence for this proposition alone makes it the case that one ought to believe it. But once one believes this proposition, the normativity of all other beliefs (as well as actions) is ultimately determined by the will of God. This is the case even if the Prophet mandates that one ought to believe in accord with our evidence, since what makes this latter obligation normative is the fact that God wills it - the evidence alone is not normative. This is why I think the position is a kind of Moderate Anti-Evidentialism – it affirms that all but one propositions ought to be believed just in case God wills it. The position accords well with al-Ghazali’s “universal rule” for knowing when to interpret the 52 For a modern defence of this claim see Rinard (forthcoming). 26 Koran allegorically or literally. He claims in several works that a passage of the Koran – when taken literally - says something that is in plain contradiction to certain theoretical knowledge (“demonstrative proof”), then it should be taken allegorically. As al-Ghazali puts it in Moderation in Belief, a work attempting to moderate the competing claims of reason and Prophesy: “Regarding what reason deems impossible, if it is reported in the revelation, it must be interpreted metaphorically. It is inconceivable that the revelation contains what is conclusively contrary to reason” (p. 210). The reason it is inconceivable is that revelation’s containing propositions contradicting certain theoretical knowledge would put into jeopardy our grounds for believing in the veracity of the Prophet, as per in (a). In his The Incoherence of the Philosophers he charges the Falasifa with apostasy as regards certain of their beliefs. These are the Falasifa’s beliefs, according to al-Ghazali, that amount to holding that the Koran must be taken allegorically, but where the Falasifa fail to have the demonstrative proof that shows that they should. This amounts to unbelief and apostasy for al-Ghazali, rather than mere mistake, or innovation, and subject to the death-penalty, since it hamstrings (under the interpretation just canvassed) the support for the proposition that the Prophet is a genuine Prophet. Mistakenly claiming that a theoretical truth in the Koran is to be interpreted allegorically and not literally would: (i) block the possibility of demonstrating that the proposition as literally interpreted is supported by our ordinary non-prophetic evidence; and (ii) violate the normative principle that we ought to believe (for all propositions bar the proposition that the Prophet is legitimate) on the basis of God’s Will, and not on the evidence alone. Note that al-Ghazali here is not seeking to prove that there is such a thing as genuine Prophesy, or proving that it is necessary. Rather, he is merely stating the conditions under which he thinks Prophesy is possible, given the constraints that it must have a unique role, and that its legitimacy must be ascertainable by humans53. As such, holding an exceptionalism as regards the proof of a Prophet’s legitimacy need not be considered ad hoc. However, one may wonder whether there is something indeed ad hoc about al-Ghazali’s Moderate Anti-Evidentialism when combined with a test for determining when revelation is to be treated allegorically. Certainly, at least, this test seems to rely on the legitimacy of the Koranic revelation to have already been ascertained. Otherwise, for any putative work of Prophesy, its legitimacy could never be denied, since any claim that appears in that work of Prophesy that is contrary to the certain knowledge we already have will be ruled as allegorical. We may also wonder whether the criterion outlined in (c) above isn’t unstable, given al-Ghazali’s other commitments. His view of belief as a kind of action seems to belie the idea that there is some internal representation that must converge with one’s assertions if one is to be deemed sincere. One’s assertions would be just a function of one’s beliefs, under his quasi-behaviouristic account of belief. However, I think this can be solved by pointing out that even for al-Ghazali’s mentor al-Juwaynī, belief that p already could consist of a propensity to utter (silently) to oneself that p in certain situations, as he calls it: “interior speech in the self” (cf. Griffel 2004, p. 134). This means that the possibility of there being a discrepancy between one’s beliefs and one’s assertions opens up. 53 I have not here discussed his attempt to prove the necessity of Prophecy cf. his Fayṣal alTafriqa, which again seems to follow Avicenna, cf. Griffel 2004. 27 4. Concluding Remarks. The last few issues raised for al-Ghazali, seem to be problems primarly for alGhazali, given certain other commitments of his. It seems to me that independently of his other commitments, his Moderate Anti-Evidentialism looks like a coherent and interesting position, worthy of being philosophically examined in its own right. It and the Falasifa’s Moderate Evidentialism together constitute, I hope to have made the case for, a very distinctive view on the ethics of belief. Though, as I hope also to have shown, Islamic thinkers have also held more orthodox views close to those of later European thinkers on the ethics of belief. In the next chapter I wish to examine the more distinctive doctrines just mentioned, and make the case for each in further detail, showing how they differ to what is arguably also a Moderate Evidentialism championed by St. Thomas Aquinas. Further, I will ask whether al-Ghazali’s Moderate Anti-Evidentialism is equivalent to, or at least compatible with, the Falasifa’s Moderate Evidentialism54. I will argue that they are not, and that their differences have important implications, and underscore important differences, in their respective political philosophies. These political implications, and their relevance to contemporary issues, will be examined in chapter 3. Tamara Albertini (Albertini 2011) was suggested that Islamic Philosophy is “too close” to the Western tradition, and should be evaluated on its own terms. Failure to do so has resulted in a distorted picture of what Islamic Philosophy is and was about, her prime example being what the Latin Averroeists did with the Ibn Rushd’s Philosophy, warping it into a so-called theory of “double truth”. As support for her thesis she cites, for example, El Ghannouchi’s claim that Averroes’ Decisive Treatise is a poor translation of the Arabic Fasl al-Maqal, and that translating him this way makes him out to be “pedantic and pretentious” as if he had wanted to present “the definitive and decisive solution to the secular question of the accord between religion and philosophy”(Albertini 2011, p 247; Albertini’s translation of El Ghannouchi). Interestingly, however, my own take on Ibn Rushd’s has ended up being very similar to hers. This, while attempting to understand his philosophy using the conceptual tools of contemporary European philosophy, and so shows that appropriating his philosophy in this way can yield a robust enough understanding of Ibn Rushd’s work. Of course there is an issue with translation – neatly summarised by the old Italian adage “Traduttore: Traditore” but taking this problem too seriously risks making oneself vulnerable to a reductio ad absurdum wherein no one can communicate with another, given that there is no guarantee that the way one person understand a given word will be the exactly the same as another person understands it (even in the same language). Further, it risks – as I hope this chapter and the remainder of the book testify – the withholding of important insights to a body of scholars. REFERENCES Arabic Works cited (not in Translation): 54 Analogously, Dougherty and Tweedt (2015) argue that the different positions in the ethics of belief – as treated in the contemporary West – overlap in significant ways. 28 Al-Ghazali: Fayṣal al-Tafriqa: Dunyā, S. (ed.) (Cairo: 1961). Arabic Works cited in English Translation: Al-Kindi: On First Philosophy - fī al‐Falsafa al‐Ūlā (Translated by Ivry, A.L. 1974: Al-Kindi’s Metaphysics (New York: SUNY Press). Al-Farabi: The Perfect State – Mabādi’ ārā ahl al madīnat al-fādilah (Translated by Walzer, R. 1985: On the Perfect State (Oxford: Oxford University Press). Avicenna: The Healing (Metaphysics) – Al-Shifā (al-Ilāhiyyāt) (Translated by Marmura, M.E. 2008: The Metaphysics of the Healing (Provo: Brigham Young University Press). The Healing (Psychology) - Al-Shifā (al-Nafs) (Translated by Rahman, F. 1959: Being the Psychological Part of Kitab al-Shifā (London: Oxford University Press). Averroes: Middle Commentary on Aristotle’s Categories – Talkḥīs kitāb al-maqulāt (Translated by Butterworth, C.E. 1998: Averroes’ Middle Commentaries on Aristotle’s Categories and De Interpretatione (Indiana: St Augustine’s Press)). The Decisive Treatise – Kiqtab Fasl al-Maqal (Translated by Hourani, G. F. 1961: Averroes: On the Harmony of Religion and Philosophy (Exeter: Gibb Memorial Trust)). The Incoherence of the Incoherence – Tahafut al-tahāfut (Translated by Van Den Bergh, S. 1954: Averroes’ Tahafut al-Tahafut (London: Gibb Memorial Trust)). Epitome of the Parvia Naturalia - Talḫīṣ al-ḥiss wa-l-maḥsūs (Translated by Blumberg, H. 1961: Averroes’ Epitome of the Parvia Naturalia (Cambridge, MA: Medieval Academy of America). Al-Ghazali: The Incoherence of the Philosophers – Tahāfut al-Falāsifa (Translated by Marmura, M.E. 1997: Al-Ghazālī: Tahāfut al-Falāsifa (Provo: Brigham Young University Press). 29 Moderation in Belief – Al-iqtiṣād fī al- iʿ tiqād (Translated by Yaqub, A.M. 2013: Al-Ghazali’s Moderation in Belief (Chicago: Chicago University Press)) Deliverance from Error - al-Munqidh min al-Dalāl (Translated by McCarthy, R. 1980: Freedom & Fulfillment (Boston: Twayne Publishers) Secondary Sources: Albertini, T. 2011: “Ibn Rushd or Averroes? Of Double Names and Double Truth” in Joy, M. (ed.) After Appropriation: Explorations in Intercultural Philosophy and Religion (Calgary: Calgary University Press). Aikin, S. 2014: Evidentialism and the Will to Believe (London: Bloomsbury). Alston, W. 1989: “The Deontological Conception of Epistemic Justification” Adamson, P. 2007: Al-Kindī (Oxford: Oxford University Press). Bakar, O. 1998: Classification of Knowledge in Islam: A Study in Islamic Philosophies of Science (Cambridge: Islamic Texts Society). Black, D. 2006: “Knowledge (ʿilm) and certitude (yaqīn) in al-Fārābi’s Epistemology” Arabic Sciences and Philosophy 16.1 pp. 11 – 46. ----------. 1996: “Practical Wisdom, Moral Virtue, and Theoretical Knowledge: The Problem of the Autonomy of the Practical Realm in Arabic Philosophy” in B.C. Bazan, E. Andujar, L.G. Sbrocchi (eds.) Moral and Political Philosophies in the Middle Ages (Ottowa: Legas). Bishop, J. 2007: Believing by Faith (Oxford: Oxford University Press). Booth, A.R. & Rik Peels 2011: “Why responsible belief is blameless belief” Journal of Philosophy. Booth, A.R. 2008: “A New Argument for Pragmatism?” Philosophia: Philosophical Quarterly of Israel. Booth, A.R. 2008: “Deontology in Ethics and Epistemology” Metaphilosophy. Booth, A.R. 2012: “All things considered duties to believe” Synthese. Booth, A.R. 2014: “Epistemic Ought is a Commensurable Ought” European Journal of Philosophy. Clifford, W.K. 1887: “The Ethics of Belief” Contemporary Review. Davidson, H. 1992: Alfarabi, Avicenna, and Averroes, on Intellect (Oxford: Oxford University Press). Dougherty, T. & Chris Tweedt 2015: “Religious Epistemology” Philosophy Compass 10(8) pp. 547 – 559. Evans, C.S. 1998: Faith Beyond Reason: A Kierkegaardian Account (Edinburgh: Edinburgh University Press) Fakhry, M. 2004: A History of Islamic Philosophy (New York: Columbia University Press). Foley, R. 1993: Working Without a Net: A Study in Egocentric Epistemology (Oxford: Oxford University Press). Frank, R. 1983: “Moral obligations in classical Muslim theology” Journal of Religious Ethics 11 pp. 205 – 23. Frankfurt, H. 1971: “Freedom of the Will and the Concept of a Person” The Journal of Philosophy 68(1) pp. 5 – 20. Goldziher, I. 1889: Muhammedanische Studien I (Halle: Niemeyer). Griffel, F. 2004: “Al-Ghazālī’s concept of prophecy: The introduction of Avicennan 30 Psychology into Ashʿarite Theology” Arabic Sciences and Philosophy 14 pp. 101 – 144. ------------. 2012: Al-Ghazali’s Philosophical Theology (Oxford: Oxford University Press). ------------. 2015: “Al-Ghazālī at his Most Rationalist: The Universal Rule for Allegorically Interpreting Revelation” in Georges Tamer (ed.) The Impact of al-Ghazālī: Papers on His 900th Anniversary (Leiden: Brill). Gutas, D. 1998: Greek Thought, Arabic Culture: The Graeco-Arabic Translation Movement in Baghdad and Early Society (London: Routledge). Haack, S. 2001: “The ethics of belief reconsidered” in M. Steup (ed). Knowledge, Truth, and Duty: Essays on Epistemic Justification, Responsibility and Virtue (Oxford: Oxford University Press). Hume, D. 1779: Dialogues Concerning Natural Religion. Hunter, D. 2011: “Alienated Belief” dialectica 65(2) pp. 221 – 240. James, W. 1907: Pragmatism ------------. 1896: “The Will to Believe” in His The Will to Believe and Other Essays Jordan, J. 2006: Pascal’s Wager: Pragmatic Arguments and Belief in God (Oxford: Oxford University Press). Kierkegaard, S. 1847: Concluding Unscientific Postscript. Leaman, O. 2009: Islamic Philosophy: An Introduction (London: Polity). -------------. 1988: Averroes and his Philosophy (Richmond: Curzon Press). -------------. 1980: “Ibn Rushd on Happiness and Philosophy” Studia Islamica 52 pp. 167 – 181. Lerner, B. 1974: “Introduction” in his translation Averroes on Plato’s Republic (Ithaca: Cornell University Press). Marebon, J. 2007: “Latin Averroism” in Anna Akasoy, James Montgomery, & Peter Porman (eds.) Islamic Crosspollinations. Interactions in the Medieval Middle East (Exeter: Gibbs Memorial Trust). Marmura, M. E. 1963: “Avicenna’s Psychological Proof of Prophecy” Journal of Near Eastern Studies 22(1) pp. 49 – 56. Maurusic, B. 2013: “Promising Against the Evidence” Ethics 123 (2) pp. 292 – 317. McGinnis, J. 2010: Avicenna (Oxford: Oxford University Press). Nasr, S.H. 1998: “Foreward” in Osman Bakar Classification of Knowledge in Islam (Cambridge: Islamic Texts Society). -----------. 2006: Islamic Philosophy from its Origin to the Present: Philosophy in the Land of Prophecy (New York: SUNY). Pascal, B. 1670: Pensees. Peels, R. forthcoming: “Believing at Will is Possible” Australasian Journal of Philosophy. Philipse, H. 2012: God in the Age of Science: A Critique of Religious Reason (Oxford: Oxford University Press). Plantinga, A. 2000: Warranted Christian Belief (Oxford: Oxford University Press). Rahman, F. 1958: Prophecy in Islam: Philosophy and Orthodoxy (London: Allen & Unwin). Reisner, A. 2009: “The Possibility of Pragmatic Reasons for Belief and the Wrong Kind of Reasons Problem” Philosophical Studies 145(2) pp. 257 – 272. Rinard, S. (forthcoming): “No Exception for Belief” Philosophy and Phenomenological Research. Rosenthal, E.I.J. 1958: Political Thought in Medieval Islam (Cambridge: Cambridge 31 University Press). Rosenthal, F. 2007: Knowledge Triumphant: The concept of knowledge in medieval Islam (Leiden: Brill). Schwitzgebel, E. 2002: “A Phenomenal, Dispositionalist Account of Belief” Nous 36 pp. 249 – 275. Shah, N. 2006: “A New Argument for Evidentialism” Philosophical Quarterly 56 pp. 481 – 498. Shields, C. 1999: Order in Multiplicity: Homonymy in the Philosophy of Aristotle (Oxford: Oxford University Press). Shoemaker, S. 2009: “Self-Imitation and Higher-Order Belief” Erkenntnis 71 pp. 35 – 51. Stalnaker, R. 1984: Inquiry (Cambridge, MA: MIT Press). Strauss, L. 1959: “How Fārābī read Plato’s Laws” in his What is Political Philosophy? And Other Studies (Illinois: Free Press). Taylor, R. C. 2000: “’Truth does not contradict truth’: Averroes on the Unity of Truth” Topoi 19 pp. 3 – 16. Turri, J. 2011: “Believing for a Reason” Erkenntnis 74(3) pp. 383 – 397. Watt, M. 1985: Islamic Philosophy and Theology (Edinburgh: Edinburgh University Press). Weatherson, B. 2008: “Deontology and Descartes’ Demon” Journal of Philosophy 105 pp. 540 – 569. Wittgenstein, L. 1966: On Certainty Wolterstorf, N. 1996: John Locke and the Ethics of Belief (Cambridge: Cambridge University Press). Zimmerman, M. 1996: The Concept of Moral Obligation (Cambridge: Cambridge University Press). 32