Islamic Wittgensteinian Fideism?

European Journal of Analytic Philosophy EuJAP | Vol. 18 | No. 2 | 2022 https://doi.org/10.31820/ejap.18.2.5 ISLAMIC WITTGENSTEINIAN FIDEISM? Edward Ryan Moad1 1 Qatar University Original scientific article – Received: 11/04/2022 Accepted: 19/11/2022 ABSTRACT This paper examines recent deployments of Wittgenstein’s thought, by Mustafa (2018) and Asad (2020), in defense of the Islamic “traditionalism” of Ibn Taymiyyah and the Hanbali school. I will briefly summarize the key features of Wittgenstein’s thought crucial to this, and then examine their ramifications. I argue that Wittgenstein’s position actually undermines any claim to interpretive authority, whether of the “rationalist” or salafi “traditionalist” sort. Secondly, the approach to religious language most commonly associated with Wittgenstein—so-called “Wittgensteinian Fideism” may pose bigger problems for traditionalists than the influence of classical philosophy or “rationalist” theological responses to modern skeptical challenges. Keywords: Wittgenstein; Ibn Taymiyyah; Fideism; Hanbali; Salafi. © 2022 Edward Ryan Moad Correspondence: edwardrm@qu.edu.qa (SI4)5 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 Introduction This paper examines two recent deployments of Wittgenstein’s thought in the defense of Islamic “traditionalism”, associated in each case with Ibn Taymiyyah and the Hanbali school, as opposed to the “rationalism” of peripatetic falāsifa and scholastic theologians, or mutakallimūn. For Abdul Rahman Mustafa, reading Wittgenstein alongside Ibn Taymiyya teaches that interreligious communication is best achieved “not through the intermediary of classical philosophy”, but by “attention to the variety of ways in which they have arrived at the meanings of the most important words in their theological lexicons” (Mustafa 2018, 466). Talal Asad, meanwhile, turns to Wittgenstein to “clarify some ideas about what is called “religion” in English”, but more specifically, to “explore and understand for myself what aspects of the Islamic tradition might mean” (Asad 2020, 403-404). Any application of Wittgenstein’s thought by such accomplished scholars of Islam is relevant to the dialogue between Islamic theology and analytic philosophy, not only because he is a major analytic philosopher, but also because of the theological implications of the approach to religious language associated with some of his closest students. Dubbed “Wittgensteinian Fideism”, this approach understands religious statements as expressions of a particular form of life rather than assertions of independent fact liable to truth or falsehood (Hyman 1997, 150-157). Consequently, to criticize speakers as “irrational” for expressing or affirming religious statements without or in spite of the evidence is to misunderstand their use in religious life and practice. I will summarize the relevant features of Wittgenstein’s thought, and then examine how Mustafa and Asad deploy it in their respective analyses. Mustafa finds relevant similarities between Wittgenstein’s position, and arguments regarding the nature of meaning by which Ibn Taymiyya defended the Hanbali insistence on the interpretive authority of the first Muslim generations (the salaf) from perceived threats posed by “rationalists”. Aside from some minor points, I make no claim regarding the accuracy of Mustafa’s account of the Hanbali position. While I do raise some internal issues with that position, given his account, my main argument is that applying Wittgenstein does not support it, as Mustafa suggests, but undermines any claim to interpretive authority, whether of the “rationalist” or “traditionalist” sort. This is most clear from his thoughts on rule following, whose connection to those on meaning I explain in the first section. 6 Edward Ryan Moad: Islamic Wittgensteinian fideism? While the influence of Wittgensteinian Fideism is absent from Mustafa’s analysis, it is prevalent in Asad’s. Once clarified, I argue, his construal of Islamic traditionalism effectively is Wittgensteinian Fideism. I argue that this construal is dubious and the philosophical argument he offers for the position is fallacious. I do not make any claim about whether his argument or ‘Wittgensteinian Fideism’ is or is not genuinely Wittgensteinian. In each case, my main point is that once clarified, the implications of defending Islamic traditionalism through Wittgenstein render the project selfdefeating. 1. Wittgenstein Wittgenstein’s career consists of an “early” phase, based on the Tractatus Logico-Philosophicus, and a “later” phase associated with the posthumously published Philosophical Investigations. The former espoused the so-called “picture theory” of meaning, according to which meaningful language consists of propositions that “picture” the world by expressing thoughts logically representing facts (Pears 2003, 811-815). Facts are relations between discrete individual objects, and the “world” is the totality of the facts. As any picture must have something in common with what it represents, a meaningful proposition must have something in common with the facts it represents. That is its logical structure, reflecting the structure of the relations it asserts to hold between objects. The analytic tools of modern symbolic logic, applied to ordinary language utterances, reveal the logical structure implicit therein. In principle, this can proceed until we arrive at “atomic” propositions asserting relations between simple names referring to discrete individual objects. These propositions are true or false, depending on whether the objects they name relate to each other in the way they assert. The truth-value of any statement in ordinary language is a function of the truth-values of its constituent atomic propositions. Ordinary language utterances correlate to the world via the complex thoughts they implicitly express (Kenny 2007, 54-58 and 132-137). Consequently, meaningful utterances are of two sorts. The “scientific” ultimately consist of atomic propositions, each true just in case the relation postulated between its terms correlates to the relation holding between the objects they name. “Logical propositions” are tautologies, true regardless of the truth or falsehood of their constituent atomic propositions. Utterances not ultimately analyzable into atomic propositions are therefore meaningless “pseudo-propositions”. These include moral, theological, and 7 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 metaphysical statements. Inasmuch as the latter aspire to describe the totality of what is, they are futile, since the description itself is part of that totality, and thus cannot “picture” it. The role of philosophy is simply to analyze ordinary propositions and distinguish them from pseudopropositions—not to resolve philosophical questions but merely dissolve them (Wittgenstein 1921, 6.5-6.521). This inspired the logical positivism of the “Vienna Circle”, formed around one or another version of the “verification principle”. Accordingly, to give the meaning of a proposition is to describe the conditions that would verify its truth (Ayer 1936, 35). For any non-tautological proposition, those conditions must be empirical. Otherwise, it is meaningless. The positivists’ project was to apply the verification principle to distinguish meaningful and “genuinely scientific” propositions from such pseudo-scientific and metaphysical “nonsense” (Kenny 2007, 58-60). On the underlying premise that meaning is a function of truth understood as correspondence between language and the world, what is verified (or otherwise) by the empirical conditions in question is that the proposition represents an independently existing reality. Thus, for positivists, a statement like “there is no God but God”, is meaningless, for in the absence of empirical verification conditions, there is no intelligible state of affairs to which we may understand it as either succeeding or failing to correspond (Martin 1999, 204-205). The defender of the meaningfulness of such a statement has three options. First, she may accept the verification principle in some form while arguing that it is possible to provide empirical verification conditions for the statement. Second, she may reject the verification principle while arguing that the meaningfulness of the proposition, understood in terms of its correspondence to an independently existing reality, need not entail empirical verification conditions. Third, she may argue that the meaningfulness of her statement is not a matter of its truth (so understood) at all. This amounts to claiming that a statement like “there is no god but God”, properly understood, does not assert the independent existence of anything on which its truth depends. A position of this sort emerged among several close friends of Wittgenstein.1 Hence the name given by a persistent critic: “Wittgensteinian Fideism” 1 These include Rush Rhees (d. 1989), Peter Winch (d. 1997), Roy Holland (d. 2013), DZ Philips (d. 2006), and Norman Malcolm (d. 1990). 8 Edward Ryan Moad: Islamic Wittgensteinian fideism? (Nielson 1967, 191). The literature on this position is immense.2 Despite disagreement over the degree to which the view actually reflects Wittgenstein’s, its influence in his name on contemporary philosophy of religion is significant. Tilghman, for example, cites Wittgenstein in his textbook, after explaining the misunderstanding he sees involved in thinking religion is about believing in the truth of certain propositions about supernatural facts. “It leads the religious outsider, the non-believer, to dismiss religion and religious people as foolish”, he writes. It has led some religious people to try to support their beliefs with arguments and evidence, but, since the arguments are invariably bad and the “evidence” cannot stand up to scholarly and scientific standards, they end up making themselves foolish. (Tilghman 1994, 209) As Asad puts it, the assumption held by secular critics of religion that worshipper and worshipped must correspond to completely separate identities makes possible the claim that since God does not exist, the believer’s “desire for God” is no more than a desire for a non-existent person, and his or her dread-awereverence is merely the exposition of an emotion directed at nothing. (Asad 2020, 426) That is, rejecting the “secular assumption” that the believer worships something separate from himself will allow us to concede that God does not exist without forcing us to concede that the believer is fundamentally in error. Wittgenstein set the stage for this by abandoning the effort to explain the variety of ways language acquires meaning under any single theory (Pears 1996, 815-826). Specifically, he opposed the assumption that a mental procedure of linking words to objects rigidly fixes their meanings. The positivists wanted to base meaningful language on so-called “protocol statements” directly describing given experience, the terms of which are fixed by such a process of ostensive definition (for example, christening a color as “red”). Since these given experiences are subjective and “private”, 2 See Carroll (2016), Gomulka (2021), Graham (2014), Holland (1956), Keeling and Mario (1977), Malcolm (1964, 1993), Nielson and Philips (2005), Philips (1965, 1970, 1976), Rhees (1997), Richter (2001), Winch (1987), and Wisdom (1945). 9 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 they faced the question how we can ever understand what anyone else means when, for example, they describe something as “red”. The conclusion was naturally drawn that, while nobody can know what the experience of “red” is (and thus, what it means) for anyone else, our ability to communicate depends on our using the terms in a shared, consistent pattern (Kenny 2007, 58-60). On the contrary, Wittgenstein argues, words do not have their meanings fixed intrinsically by any independent mechanism determining their proper use in every case. Indeed, no such determination is intelligible. That is, the problem is not simply that were meaning fixed by a process of ostensive definition I would be unable to know what others mean by the words they use, but that, so conceived, there is simply no fact of the matter as to what either they or I mean at all. Instead of understanding “meaning” as a link between a word and an object, Wittgenstein advocates examining how we use words in the particular situation, for which he coins his trademark term. “I shall also call the whole, consisting of language and the actions into which it is woven, the ‘language-game’” (Wittgenstein 1953, 5). Philosophical Investigations opens with reference to Saint Augustine’s account, in Confessions, of language acquisition as a childhood process of learning the names of objects by observing one’s elders. Wittgenstein then asks us to imagine a shopping order reading “five red apples”, which the shopkeeper fills by looking up a color sample labelled “red”, opening a drawer marked “apples”, and removing one apple at time while reciting the series of cardinal numbers to “five”. But how does he know where and how he is to look up the word “red” and what he is to do with the word “five”?—Well I assume that he acts as I have described. Explanations have to come to an end somewhere.—But what is the meaning of the word “five”?—No such thing was in question here, only how the word “five” was used. (Wittgenstein 1953, 3) The point is not that Augustine’s description of the process is wrong, but that it is just a description of one (very simple) sort of “language-game”, and not an explanation of meaning as such. As an explanation of meaning, the process of labelling objects multiplies rather than resolves questions (“what is the meaning of five?”). If we want the buck to stop somewhere, then it can only stop at how one uses the word in this particular language game. “It is as if someone were to say: “A game consists in moving objects about on a surface according to certain rules (…)”—and we replied: You seem to be thinking of board games, but there are others”, he writes. “You 10 Edward Ryan Moad: Islamic Wittgensteinian fideism? can make your definition correct by expressly restricting it to those games” (Wittgenstein 1953, 3). When we consider restricting this definition to “board games”, we realize that since there may also be other “board games”, we are on a path to saying that “a game consisting in moving objects about on a surface according to certain rules consists in moving objects about on a surface according to certain rules”. That would be an uninformative string of words, even if we could informatively define each of them. Thus, a language game may involve moving objects about, but not according to certain rules. For to say that words have no meaning independent of and prior to their actual use, is to say there are no independent rules about how to move them around. There is therefore no rigid definition, either of any particular languagegame or of what counts as one. This is clear where Wittgenstein acknowledges that his hypothetical examples are extremely simple. If you want to say that this shews them to be incomplete, ask yourself whether our language is complete; - whether it was so before the symbolism of chemistry and the notation of the infinitesimal calculus were incorporated in it; for these are, so to speak, suburbs of our language. (And how many houses or streets does it take before a town begins to be a town?) (Wittgenstein 1953, 8) This follows from his empiricist framework. There is no ideal form of language by which to discriminate the “complete” from the “incomplete” among actual particular language games. Each language game just is what it is, with its own history and development. “Our language can be seen as an ancient city”, he writes, “a maze of little streets and squares, of old and new houses, and of houses with additions from various periods; and this surrounded by a multitude of new boroughs with straight regular streets and uniform houses” (Wittgenstein 1953, 8). It follows, not only that there is no independent standard by which to evaluate language games (as to their truth, completeness, etc.), but also none by which to evaluate, for any linguistic practice, its continuity with the particular language game in which it is made. The implications are considerable given, as he remarks, that “to imagine a language-game is to imagine a form of life” (Wittgenstein 1953, 8). This was our paradox: no course of action could be determined by a rule, because every course of action can be made out to accord with the rule. The answer was: if everything can be made out to accord with the rule, then it can also be made out 11 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 to conflict with it. And so there would be neither accord nor conflict here. (Wittgenstein 1953, 81) Wittgenstein developed this paradox of “rule-following” in a series of preceding thought experiments.3 Consider a student, following instructions to count by twos, who begins to count by fours after reaching 1000. When asked why he “broke the rule”, he claims not to have. This is the rule he had been following all along: count by twos to 1000, and then count by fours. How do we determine the matter of fact here, about the rule that he had been following and that we had instructed him to follow? The first proposal Wittgenstein considers and rejects is that it is a matter of analogy to precedent. Or suppose he pointed to the series and said: “But I went on in the same way”.—It would now be no use to say: “But can’t you see….?”—and repeat the old examples and explanations. —In such a case we might say, perhaps: It comes natural to this person to understand our order with our explanations as we should understand the order: “Add 2 up to 1000, 4 up to 2000, 6 up to 3000 and so on. Such a case would present similarities with one in which a person naturally reacted to the gesture of pointing with the hand by looking in the direction of the line from finger-tip to wrist, not from wrist to finger-tip. (Wittgenstein 1953, 75) How can we judge between competing claims that one rather than another way of proceeding is appropriately analogous to the precedent, so to count as following the rule it is supposed to have followed? Wittgenstein considers the suggestion that the judgment would require at each proceeding step a “new insight – intuition” of “the order—as it was meant”. What, he then asks, did you actually mean when you gave the instruction? If you meant that he should count 1002 after 1000, “did you also mean that he should write 1868 after 1866, and 100036 after 100034, and so on—an infinite number of such propositions” (Wittgenstein 1953, 75)? One might respond that all those propositions follow from the meaning of the stated instruction. But that is just what is in question: what, at any stage, does follow from that sentence. Or, again, what, at any stage we are to call “being in accord” with that sentence (and with the mean3 For an influential, though controversial account of this, see Kripke (1982). 12 Edward Ryan Moad: Islamic Wittgensteinian fideism? ing you then put into the sentence—whatever that may have consisted in). It would almost be more correct to say, not that an intuition was needed at every stage, but that a new decision was needed at every stage. (Wittgenstein 1953, 75) That new decision, for Wittgenstein, determines the matter of fact as to what accords with the rule at that stage, prior to which there is no such fact. “And so there would be neither accord nor conflict here”, that is, if we assume that any such determination must be an interpretation of previous decisions. “What this shews”, he writes, “is that there is a way of grasping a rule which is not an interpretation, but which is exhibited in what we call “obeying the rule” and “going against it” in actual cases” (Wittgenstein 1953, 81). Though this might lead us to view each action as an interpretation of the rule, he warns, “we ought to restrict the term “interpretation” to the substitution of one expression of the rule for another”. Thus, the way we “grasp” the rule is not by restating it in a vain effort to show what does and does not accord with it. Nothing beyond or behind the present decision can decide that. And hence also “obeying a rule” is a practice. And to think one is obeying a rule is not to obey a rule. Hence it is not possible to obey a rule “privately”: otherwise thinking one was obeying a rule would be the same thing as obeying it (Wittgenstein 1953, 81). Obeying a rule is therefore a public practice, but in which the collective behavior of a community of language users determines what does and does not accord with it at each stage. The “rule” does not guide the practice, but rather the practice defines the rule. In a sense, we make it up as we go along. “We” however, may be any community, and there are communities of communities. This is how Wittgenstein views the emergence and resolution of differences. Following a rule is analogous to obeying an order. We are trained to do so; we react to an order in a particular way. But what if one person reacts in one way and another in another to the order and the training? Which one is right? Suppose you came as an explorer into an unknown country with a language quite strange to you. In what circumstances would you say that the people there gave orders, understood them, obeyed them, rebelled against them, and so on? 13 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 The common behavior of mankind is the system of reference by means of which we interpret an unknown language. (Wittgenstein 1953, 82) By implication, in the case of two communities both claiming to follow the same religious teaching but behaving differently in response to it, if there is any fact at all about which of the two is “right”, it consists in the practice of a broader linguistic community within which they both operate. That ultimately would extend to the most general human practices of differentiating “obedience” from “rebellion”. Again, this practice indicates no independently guiding norm, like “fitra” or “human nature”. There is only the practice as it stands. “So you are saying that human agreement decides what is true and what is false?” asks his imaginary interlocutor. “It is what human beings say that is true and false; and they agree in the language they use. That is not agreement in opinions but in form of life” (Wittgenstein 1953, 88). 2. Wittgenstein and Ibn Taymiyyah Mustafa suggests that reading Ibn Taymiyya’s account of language alongside Wittgenstein’s is potentially constructive because both are “couched within a broader attack on Aristotelianism” (Mustafa 2018, 465). Specifically, he sees them converging in their opposition to the view that the meanings of words are fixed by their signification of objects, assigned through a process of ostensive definition (Mustafa 2018, 469). Consequently, they both deny the epistemic efficacy of technical definitions and categorical syllogisms, as well as the distinction between literal and metaphorical meaning. This is significant for Ibn Taymiyyah, because his opponents advocate metaphorical interpretations of Islamic religious language where they believe a literal reading contradicts what they consider demonstratively proven. For the falāsifa (and some later mutakallimūn), demonstrative proof is expressed through a series of valid syllogisms based on self-evidently true premises expressed in precisely defined terms. According to Mustafa, Ibn Taymiyyah’s position that “the meaning of language arises out of use” undermines the conventional distinction between literal and metaphorical meaning (Mustafa 2018, 473). That distinction, he explains, was that while our understanding of the metaphorical meaning of a term depends on context (e.g. saying “he is a lion” while pointing to a man), our understanding of its literal meaning is context-independent, being what “first comes to mind” on hearing the 14 Edward Ryan Moad: Islamic Wittgensteinian fideism? term. It is plausible that words only convey meaning within a particular context, for speech always occurs in a particular context. If so, it would be appropriate to conclude that this particular explanation of the difference between literal and metaphorical meaning collapses. On Mustafa’s account, Ibn Taymiyya’s conclusion is stronger. Since words only convey meaning within particular contexts that fix their meaning and make them unambiguous, and since speech cannot exist without context, it follows that all words in speech are real and not metaphorical. (Mustafa 2018, 477) So expressed, this could mean one of two different things. On one hand, it may be simply to deny that ambiguity occurs at all: since speech is always in a particular context which as Mustafa says, fixes the meaning of words and makes them unambiguous. That is obviously false. For the statement in question is itself ambiguous. It may also mean that words do not convey meaning unless the particular context renders them unambiguous. That is not directly to deny that ambiguity ever occurs, but rather to deny that words convey meaning in such cases. Yet this is also false. For ambiguity occurs when the words used convey two or more distinct meanings—just as they do in Mustafa’s statement here, it being unclear which one he intends. Therefore, we may agree that words convey meaning only within a particular context, but not that context always eliminates ambiguity from meaningful speech. Asad, for instance, observes that ambiguity is “an ever present source of linguistic creativity as well as misunderstanding” that “undermines the notion of a permanent “system”. He clarifies Wittgenstein’s use of “system” as conveying the idea that reasoning draws on different purposes, feelings, conditions over time (…) by which the point we are trying to make becomes persuasive to particular audiences at particular times and places. (Asad 2020, 416) His ambiguous suggestion that, “the original intention may not necessarily be a primary concern,” is not just that it may contingently “be a primary concern” (Asad 2020, 416). Likewise, when he says, “ambiguity may reflect” he means that it does reflect “contradictory motives in the reader or hearer and the use he wants to make of what he reads and hears” (Asad 2020 416). For his point is that the “boundaries of the network” within which the soundness of an argument is evaluated and within which its 15 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 terms have their use, “change according to the purposes at hand” (Asad 2020, 416). Here, he cites Wittgenstein: And this multiplicity is not something fixed, given once for all; but new types of language, new language-games, as we say, come into existence, and others become obsolete and forgotten. (Wittgenstein 1953, 11) Consequently, the context dependency of meaning does not eliminate ambiguity, but imposes it. It does not, moreover, render all meaning literal (assuming this is what Mustafa means by “all words in speech are real”). It only renders void a particular explanation of the distinction between the literal and metaphorical. The literal is not the meaning conveyed without context. Perhaps, however, literal meaning differs from the metaphorical in virtue of the sort of context on which its conveyance depends. For if the conveyance of meaning is context dependent in every case, it is also so for the terms “literal” and “metaphorical”. We also use these terms differently in different contexts that, as Mustafa explains, include the changing interpretive norms of linguistic communities (Mustafa 2018, 478). This resonates indeed with the later Wittgenstein, for which reason he would not categorically deny any meaningful distinction between the “literal” and “metaphorical”. He would deny only that we could define such a distinction independently of the particular linguistic community in which the terms are used. Here emerges a key difference between Ibn Taymiyya and Wittgenstein. As Mustafa explains, that linguistic norms of a community can change over time is why, for Ibn Taymiyya, scripture must not be interpreted according to the linguistic conventions prevalent in the time of later interpreters and exegetes but only according to the conventions prevalent amongst the Prophet and his Companions to whom the Qur”ān was revealed. (Mustafa 2018, 478-479) According to Mustafa, Ibn Taymiyya argues that the conventional theory of language is an “alien accretion” introduced into Islamic thought, unsupported by the statements of the Prophet, his Companions, or earliest Arabic grammarians (Mustafa 2018, 470). This raises the question whether we must interpret only the language of the Qur’ān or language itself strictly according to the conventions of the salaf. If the former, since a theory of language is not an interpretation of the Qur’an, why would its “alien” 16 Edward Ryan Moad: Islamic Wittgensteinian fideism? origin alone count against it? On the other hand, is it possible that salafi conventions be universally normative with respect to interpreting the Qur’ān, without also being normative with respect to the nature and function of language itself? If not, is there anything at all, of which anyone may speak, over which their conventions are not the final authority? My aim here is neither to answer these questions, nor to guess at Ibn Taymiyya’s answers. Yet whatever they are, they would determine the scope of what we can call the salafi rule: for Ibn Taymiyya (not Wittgenstein), speak only as the salaf spoke. This is one of three principles comprising, for Mustafa, the “Hanbali intellectual movement”. They are: 1) the “supremacy and self-sufficiency of scripture”, 2) the “harmony of reason and revelation”, and 3) the “interpretative authority” of the salaf (Mustafa 2018, 468). Some correction of this description is in order. The Hanbalis obviously were advocating neither the supremacy of “scripture”, nor the harmony of reason and “revelation”, Their concern is specifically with Qur’ān and a collection of authenticated hadeeth, not “scripture” and “revelation” in general. The supremacy of, and harmony of reason with the Gospels or Vedas was not on the Hanbali agenda. This point is not merely pedantic in the context of invoking Wittgenstein. Secondly, given the premise of harmony between reason and the Qur’ān, what can it mean to assert the supremacy of the latter? Supremacy presupposes the possibility of conflict. If the two are necessarily in harmony, then any conflict is merely apparent, and explicable as resulting from either unsound reasoning, or an unsound interpretation of the Qur’ān, or both. The only meaningful “supremacy” to claim would be that of a sound reading of the Qur’ān, over unsound reasoning. Thus, conjoined with the harmony principle, the Qur’ānic supremacy principle does not distinguish the Hanbali position from anyone other than those who claim either 1) the supremacy of unsound reasoning over the sound interpretation of the Qur’ān, or 2) the equivalence of these two. Neither the falāsifa nor the mutakallimūn make such radical claims. They also held that between sound reasoning and a correct understanding of the Qur’ān, no real conflict is possible. They only differed with each other, and with the Hanbalis, as to how apparent conflicts are correctly resolved; that is, as to what sound reasoning and a correct understand of the Qur’ān entail. The Hanbali objection is that it is incorrect to understand the Qur’ān in the light of what either the falāsifa or mutakallimūn take (wrongly, in their view) to be “sound reasoning”. The basis of that objection rests on what Mustafa describes as the “self-sufficiency” of the Qur’ān. 17 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 That is, no metaphorical interpretation thereof is ever correct, much less required by sound reasoning, as the falāsifa and mutakallimūn claim. Any appearance to the contrary results from unsound reasoning. It is not clear, however, whether it is accurate to describe this as a commitment to the “self-sufficiency” of the Qur’ān, given the third of Mustafa’s Hanbali principles, asserting the interpretative authority of the salaf. For if the Qur’ān’s self-sufficiency means that it requires nothing other than itself to make itself understood, then the interpretative authority of the salaf would be superfluous. Conversely, if we can only understand the Qur’ān correctly through the interpretation of the salaf, then it is not self-sufficient in that sense, for it would require their interpretation. That is so, even if it were self-sufficient in conveying its meaning to them (which may be what Mustafa means here), after which everyone else can only understand it by reference to their interpretation. Consider an argument that “the salaf did not apply any external interpretive framework to understand the Qur’ān but instead let it speak for itself; therefore, we should also”. This is incompatible with the salafi rule. For it obviates their interpretive authority, with one or another of two implications: either the Qur’ān necessarily conveys the same message in every context or not. If the first, then its language has its meaning fixed independently of any particular context, contrary to Ibn Taymiyya’s position (as Mustafa recounts it). If the second, then the possibility remains that it may convey its meaning differently—and nevertheless correctly— from one context to another. The salafi rule is that the Qur’ān only conveys its correct meaning within the particular linguistic context of the salaf. Any understanding outside of that will be mistaken. The salafi rule dictates that we should understand the Qur’an exclusively according to the conventions of a linguistic community; not the one we actually inhabit however, but the one inhabited by the first generation of Muslims. We are to deploy the conventions of the salafi linguistic community (in Wittgenstein’s terminology, the salafi “form of life” or “language game”) as an external standard by which to evaluate, correct, or adjudicate between those of other linguistic communities concerning the Qur’ān. Specifically (and crucially), we must deploy salafi conventions to adjudicate between competing claims to having deployed those conventions. If two or more people claiming to follow the salafi rule interpret the Qur’ān (and consequently behave) differently, then we should apply salafi linguistic conventions to determine which is correct. Consider a faylasuf 18 Edward Ryan Moad: Islamic Wittgensteinian fideism? who claims to be following the way of the salaf by giving public expression only to the language in which the Prophet and his Companions addressed the general public (according as the falāsifa claim, to their capacity to understand), while discussing its “deeper” meanings, if at all, only within exclusive, qualified circles. A mutakallim, perhaps, may claim to be following the salafi rule by trying to speak and act as they would, if they were to speak and act in her own different sociolinguistic context. Just such differences arise between those labeled as “salafis” as distinct from “falāsifa”, “mutakallimun”, “sufis”, etc. Most of them, however, have claimed to be following the way of the salaf. The question between them is not whether that way is normative, but how to follow it (who is doing so correctly), and how that is to be decided. To say that it is by applying the linguistic conventions of the salaf to the interpretation of the Qur’ān simply begs the question at hand: how do we know what does and does not constitute such an application? To say that we know their linguistic conventions through the transmitted reports of their statements only raises the question, how we know which understanding of these statements is correct. The point here is not to push a skeptical position. We are simply reading the salafi agenda through the lens of Wittgenstein. This is his paradox of rule following, explained above, of which the salafi rule is not immune. The Wittgensteinian lesson would be that the appearance of the skeptical challenge here is symptomatic of the attempt to fix the meaning of Qur’ānic language by something external to its actual use in the particular linguistic context of that use. The salafi rule constitutes such an attempt. For the imperative to interpret the Qur’ān only according to the linguistic conventions of the salaf is only meaningful on the supposition that their language game is external to ours. Otherwise, our use of Qur’ānic language would be in order just as it is, and it would be nonsense to question its alignment with theirs. With any attempt to follow (or claim to be following) the salafi rule, we face the same question. That is, not only whether one rather than another use or understanding corresponds to the rule, but also how we can determine that and what in the final analysis constitutes the fact of that matter. For Wittgenstein, no such “final analysis” is forthcoming. There is only the use of language in our particular game. Claiming to follow the linguistic conventions of the salaf, and justifying one’s use of language against detractors on that basis may be part of the language game in which one operates. To conceive those conventions, however, as something fixed externally to one’s own linguistic practice, as a guiding rule thereof, leads 19 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 to the paradox symptomatic of philosophical confusion. There is no fact about what does and does not correspond to salafi conventions outside of what our linguistic community accepts as such through its actual practice. Asad expresses this implication as follows. But when the grammar of concepts is translated as a discursive tradition—as the open-ended passing on behavior and styles of argument in which language and life across generations are intertwined—temporality becomes essential to the ways in which meaning is made and unmade, where “inside” and “outside” are not permanently fixed, because the distinction has to do with what is taken for granted only in and for a particular time. (Asad 2020, 415) My point is not that Wittgenstein is right about this, but only that any notion that his approach to language vindicates the Hanbali movement against their “rationalist” opponents is mistaken. On the contrary, it undermines their project inasmuch as that involves drawing a permanent distinction between what is “inside” and “outside” salafi linguistic conventions. To assert the interpretive authority of any historically defined linguistic community over later communities is to suppose such a distinction. 3. Wittgensteinian Fideism and Islamic Traditionalism Asad invokes Wittgenstein to explain and defend a version, quite different from Mustafa’s, of what he calls the “traditionalist” approach of Ibn Taymiyyah and the Hanbalis. Here, the influence of Wittgensteinian Fideism is apparent, and consequently, the potential implications of invoking Wittgenstein’s philosophy to this end more evident. Representation has a dual sense: making a visible sign stand for something or someone, and speaking authoritatively for another. According to traditionalists, God cannot be represented in either sense. And if he cannot be definitively represented, there cannot be contradictory representations of him in Qur’ānic discourse. (Asad 2020, 419) That nobody can speak authoritatively for God rules out Mustafa’s interpretive authority of the salaf over His speech. Asad’s construal of the “traditionalist” position is thus not only radically different but also dubious in itself: one wonders what the role of a prophet is supposed to be on this 20 Edward Ryan Moad: Islamic Wittgensteinian fideism? premise. That aside, would “traditionalists” have made an inference as fallacious as what Asad attributes to them here? For from the fact that God cannot be definitively represented, it follows simply that there cannot be definitive representations of Him. It does not follow from this that there cannot be contradictory representations. God says He is “the light of the heavens and the earth” (Qur’an, 24:35). His being indefinable does not preclude any contradiction between that statement and its negation. This famous verse continues: The example of His light is like a niche within which is a lamp, the lamp is within a glass, the glass as it were a pearly [white] star lit from [the oil of] a blessed olive tree, neither of the east nor of the west, whose oil would almost glow even if untouched by fire. Light upon light. Allah guides to His light whom He wills. And Allah presents examples for the people, and Allah is Knowing of all things. Another famous verse from Surah al-Ikhlas, that “nothing is like Him”, generates an “apparent contradiction” which rationalists, according to Asad, “find themselves compelled to resolve (…) by treating the language in which they appear as metaphorical”. Here where the verse explicitly says that Allah presents examples (amthāl), that seems natural, and without any implication of apparent contradiction. Yet “since God himself declares his revelation to be “clear of any obscurity” (al-kitāb al-mubīn)”, as Asad describes the traditionalist response, “his words should be understood in the way he has uttered them and not in the way some scholars think he must have meant them” (Asad 2020, 419). Asad’s critique, cited above, of original intention as a primary interpretive concern seems to undermine this response by obviating the very distinction between understanding God’s words in the way he has uttered them, and the way a hearer thinks He meant them. That is, unless understanding God’s words in the ‘way he has uttered them’ is not the same, in Asad’s mind, as understanding them in the way God intended us to understand them. Yet this ‘traditionalist response’ as stated presupposes a necessary difference. What the hearer thinks God meant by them cannot in fact be the way He uttered them. It also assumes, first, that God’s utterance of a metaphor would constitute an obscurity on His part, such that any metaphorical understanding must be different from ‘the way he uttered it’. It would be a scandal to claim that God misspoke and offer something construed as a correction. Yet, one may hold that the Qur’ān is expressed in the “best and only way possible” as Asad rightly says all faithful Muslims believe, and yet that sometimes, the best and only way possible 21 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 to express something to us is metaphorically. Allah guides to His light who He wills. There need be no obscurity here. Yet Asad’s solution to this imagined obscurity is that God’s utterances are not truth claims, as a metaphorical interpretation would suppose. “The fact that these are epithets of God produced by God implies that they have a transcendent force not as propositions that call for coherent evidence”, he says, “but as utterances enacting the change and development of human character” (Asad 2020, 420). This is a false dilemma. Acknowledging, rightfully, the role of Qur’anic recitation in Islamic moral self-cultivation, and that they are not merely theoretical theological postulates, does not require separating that function from the propositional content of the language and exclusively disjoining the two. It is thus no reason for denying, as Asad does here, that they have “transcendent force” as propositions. Whether they “call for coherent evidence” depends on what we mean by that vague qualification. They may call for coherent evidence of a sort disclosed through the practice of recitation, not precluding thereby the availability of other forms of evidence. Asad’s subtle implication, however, is that their being propositional entails calling for “coherent evidence” of a sort contrary or subversive to the spiritual function of their recitation, as if acknowledging that requires denying that they actually say anything about God. That is not simply to reject, as Asad puts it, “the notion that the Qur’ān must be an object of an abstract faculty called “reason” to make sense” (Asad 2020, 421). Nor is it Ibn Taymiyyah’s assertion, to which he equates it, that we can form concepts without definitions. From these, it does not follow, as he implies, that the divine names “are not representations but an essential means of relating to divinity” (Asad 2020, 421). Again, why must their being an essential means of relating to God mean that they say nothing about Him? Absent an answer to this question, Asad’s argument turns on a false dilemma. The now evident, though unnamed “Wittgensteinian Fideism” behind his construal of Islamic traditionalism is clearer in what follows. Asad describes visiting a dying Kabbashi friend (after anthropological fieldwork among the Kababish), who said of his illness: “this is the will of God”. I had heard him use these words in a banal way when he was well. For my friend, the expression didn’t signify an object of possible knowledge or speculation. It was simply a reverential 22 Edward Ryan Moad: Islamic Wittgensteinian fideism? expression of trust (what Wittgenstein would call an “avowal”), a mundane part of his form of life, and of death as integral to it. My point is merely that for traditionalists such apparent contradictions (God is All-Merciful, and yet I am dying from a painful and fatal disease) are not to be resolved by resorting to philosophical resources—by one set of words being translated into another—but by words expressing a particular form of life. (Asad 2020, 422) Again, peripherals obscure Asad’s key assertion, which is that saying that his illness was the will of God was not, for Asad’s friend, an object of possible knowledge. Whether Asad is speculating here is peripheral, though included in the question how he knows what this expression did or did not signify for his friend. Is it because in Kababish tradition (as discovered through fieldwork), taking that paradigmatically Muslim statement as noumenal is a requisite for using it as an expression of reverential trust? The real question is theological and not anthropological. Faith in both God’s mercy and providence, as Asad correctly implies, does not require a philosophical resolution to the problem of evil. Why, however, should we think it requires us (as genuine “traditionalists”) to concede that our ascription of everything—including illness and death—to God’s will, is not after all an assertion about how things are, the truth of which can be known, but instead merely an expression of a particular form of life? Asad’s train of thought starts with remarks Wittgenstein made on Frazer’s The Golden Bough (1890), objecting to “Frazer’s extension of judgments of truth or falsity, of sense or nonsense, from propositions where such judgments are appropriate to situations where they are not”. The “main point”, as Asad takes it, seems unremarkable: “religious practice isn’t necessarily based on a theory about the world; it is first and foremost a way of being” (Asad 2020, 405). Of course, it may or may not, be based on a theory about the world. Nothing about its being a “religious” practice, alone, decides this. That would depend on the particular practice in question, which in either case is “first and foremost a way of being”. At least, Asad’s implied disjunction between a practice being theory based or being a “way of being”, is not explicitly exclusive. Taken in the plausible of the two senses, then, that point is simply that it is possible that a religious practice entails no assertions of fact to which judgments of truth and falsity might apply. If so, then whether or not it does depends on the particular case, which is here that of Islamic practice. 23 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 Asad mentions the hadeeth describing degrees of one’s iman as corresponding to whether one resists something one sees as wrong with his hand, tongue, or heart. Whosoever of you sees an evil, let him change it with his hand; and if he is not able to, then with his tongue; and if he is not able to, then with his heart—and that is the weakest of iman. (Hadeeth 34, al-Nawawi’s Forty Hadeeth) The term “iman”, Asad says here, “is used in a sense that is neither epistemological nor aleatory but dispositional” (Asad 2020, 408). Here, his disjunction is exclusive, and questionably so. Can it not be both epistemological and dispositional? It seems that it is. For as Asad observes, “iman here is described as weak because of the subject’s inability to act to stop something he or she recognizes as wrong” (Asad 2020, 408). Yet recognition of something as wrong is epistemological, and indeed pertains to a matter of fact about the world. Thus, iman, understood here in terms of a practice of resisting in one way or another a recognized evil, is based on a “theory of the world”; that is, an assertion of fact (to which judgments of truth and falsity apply) as to the existence of an evil requiring resistance. This does not exclude it from also being a disposition toward that moral fact. Asad has given no reason here to believe that the recognition of and appropriate disposition to a moral fact are separable, much less mutually exclusive; and so no reason to believe that since iman is dispositional, it is therefore not epistemological. Asad’s concern is the discursive disadvantage at which secular modernity puts Muslims and others outside its preferred cultural paradigm, by demanding that we justify ourselves (and our difference) in terms falsely presented as neutral. Science, for instance, is in this context “an ideological construct whose function is to legitimize political and economic policies as well as to control what it defines as “religion”” (Asad 2020, 408). This observation and related concerns are valid. Their validity, however, obscures the nature and value of Asad’s resistance strategy, exemplified in the fallacious inference that since iman is dispositional it is not epistemological. His implicit premise that the two are mutually exclusive is itself a feature of secular modernity. Reference to Alastair MacIntyre’s critique is a matter of course in this respect (MacIntyre 1981). For Asad, however, what brings him to mind at this point is that MacIntyre “insists that there is a rational basis for choosing between contending traditions—traditions that confront 24 Edward Ryan Moad: Islamic Wittgensteinian fideism? each other from the “outside”” (Asad 2020, 411). Asad’s objection amounts to a worry about how much time one gets to provide an adequate rational response to challenges from a contending tradition, before reluctance to convert qualifies one as irrational. Thus, the proposition that one’s tradition may have a rational basis seems, for him, tantamount to accepting that one must earn one’s right to practice the tradition through some sort of trial by debate—a primitive liberal custom he correctly rejects. Can I not, at any rate, evade the fundamental doubt that my external critic seeks to plant in me by refusing a theoretically mandated defense and resorting instead to the practice that has shaped me in my tradition? Can I not refuse to speak in this moment in its defense, and instead resume my ordinary life? And if I can, why is that “irrational”. (Asad 2020, 412) What he should reject is the notion that having a rational basis for continuing within a tradition is a matter of having won some public debate in its defense. Instead, he seems to concede that it is, asking “when rationality is brought in as a method, doesn’t dialogue collapse” (Asad 2020, 412). What definition of “rationality” must we assume in order to maintain that dialogue can only proceed in its absence? Obviously, if we are to resolve all questions by contest, then once the contest is over there is nothing to discuss, but why should we think that is “rationality”? A passage Asad quotes from Wittgenstein’s On Certainty provides an answer. Giving grounds, however, justifying the evidence, comes to an end;—but the end is not certain propositions striking us immediately as true, i.e. it is not a certain kind of seeing on our part; it is our acting, which lies at the bottom of the languagegame. If the true is what is grounded, then the ground is not true, nor yet false. (Wittgenstein 1969, 204-5) This is a radical statement. Action is the ground of truth. Consequently, truth cannot ultimately guide action. From this, it does not just follow that religious practice is not necessarily based on a “theory of the world”, but that it is not possibly based on a theory of the world. That is a much stronger claim, for which the only argument we have here is that “if the true is what is grounded” then the ground is neither true nor false. To be “grounded” as we see from the context, is to be evidentially justified 25 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 (“giving grounds, however, justifying the evidence, comes to an end”). We thus have it that, if the truth is what has been justified, then the justification is not true or false; at least not the ultimate justification (where it “comes to an end”). This only follows in virtue of the bare assertion here that this “end” is not “certain propositions striking us as true” or any kind of “seeing”. Wittgenstein is starting here from the denial of epistemic foundationalism (and associated self-evident first principles, intuition, and the like) in order to assert an alternative sort of foundationalism. “Our acting” as we do lies at the bottom, not of a system of epistemic justification grounding our “language game”, but of the game itself, of which the various practices of demanding and giving epistemic justification are only some features. In that case, truth is just determination by the outcome of one or another kind of game. That implication should motivate us to question whether we have good reason 1) to deny the self-evidence, or epistemic vision, which is supposed here to leave only “our acting” at the bottom, and 2) whether our acting is in fact what is left there in that case. Whether self-evidence or intellectual intuition are possible, and whether they are admissible or efficacious as evidence in a public process of justifying contested claims, are two different questions; unless, that is, truth just is what is justified in the latter sense (which followed here only from the denial of self-evidence). On the other hand, if we accept that something can be true and nevertheless unprovable by publicly available evidence (that is, that truth is not simply the outcome of certain sorts of language games), then the question remains whether something might be self-evident or “seen” to be true apart, from the admissibility of that as evidence in a public forum. In that case, defending the Muslim from modern-secularists’ accusations of irrationality, simply for remaining Muslim without first beating atheists in a rigged debate, does not require denying that Islam makes any claims intended to represent an independent reality and thus appropriately evaluated for their truth or falsehood in that sense. That there is a Creator existing independently of His creation, including human beings and our linguistic conventions, is traditionally a fundamental postulate of Islam. Reinterpreting that postulate as merely an expression of one particular form of life is a high price to pay to defend it from rational scrutiny. We might first interrogate the empiricist suppositions from which the problems motivating the later Wittgenstein arose before discarding rationality and truth altogether. 26 Edward Ryan Moad: Islamic Wittgensteinian fideism? REFERENCES Asad, Talal. 2020. “Thinking about Religion through Wittgenstein”. Critical Times 3(3): 403-442. Ayer, Alfred J. 1936. Language, Truth, and Logic. London: Victor Gollancz. Carroll, Thomas. 2016. Wittgenstein within the Philosophy of Religion. Basingstoke: Palgrave Macmillan. Gomulka, Jakub. 2012. “Wittgensteinian Philosophy of Religion as a Kind of Apophatic Theology”. Studia Religiologica 45(3): 165-172. Graham, Gordon. 2014. Wittgenstein and Natural Religion. Oxford: Oxford University Press. Holland, R.F. 1956. “Religious Discourse and Theological Discourse”. Australasian Journal of Philosophy 34(3): 203-207. Hyman, John. 1997. “Wittgensteinianism”. In A Companion to Philosophy of Religion, edited by Philip Quinn and Charles Taliaferro, 150157. Oxford: Blackwell. Keeling, Bryant, and Mario Morelli. 1977. “Beyond Wittgensteinian Fideism: An Examination of John Hick”s Analysis of Religious Faith”. International Journal for Philosophy of Religion 8(4): 250-262. Kenny, Anthony. 2007. Philosophy in the Modern World. Oxford: Clarendon Press. Kripke, Saul. 1982. Wittgenstein on Rules and Private Language. Cambridge, MA: Harvard University Press. Malcolm, Norman. 1964. “Is it a religious belief that God exists?” In Faith and the Philosophers, edited by John Hick, 103-110. London: Macmillan. ———. 1993. Wittgenstein: A Religious Point of View? London: Routledge. Martin, Michael. 1997. “The Verificationist Challenge”. In A Companion to Philosophy of Religion, edited by Philip Quinn and Charles Taliaferro, 204-212. Oxford: Blackwell. Mustafa, Abdul Rahman. 2018. “Ibn Taymiyyah & Wittgenstein on Language”. Muslim World 108: 465-491. Nielson, Kai. 1967. “Wittgensteinian Fideism”. Philosophy XLII (161): 191-209. Nielson, Kai and Philips, D. Z. 2005. Wittgensteinian Fideism? London: SCM Press. Pears, David. 2003. “Wittgenstein”. In The Blackwell Companion to Philosophy, edited by Nicholas Bunnnan and E. P. Tsui-James, 811-826. Oxford: Blackwell. 27 EuJAP | Vol. 18 | No. 2 | 2022 Special issue Interactions between analytic and Islamic philosophy/theology 3 Philips, D. Z. 1965. The Concept of Prayer. London: Routledge & Kegan Paul. ———. 1970. Faith and Philosophical Inquiry. London: Routledge & Kegan Paul. ———. 1976. Religion without Explanation. Oxford: Blackwell. Rhees, Rush. 1997. Rush Rhees on Religion and Philosophy, edited by D. Z. Philips and M. von der Ruhr. Cambridge: Cambridge University Press. Richter, Duncan. 2001. “Missing the Entire Point: Wittgenstein and Religion”. Religious Studies 37(2): 161-175. Tilghman, B. R. 1994. An Introduction to the Philosophy of Religion. Oxford: Blackwell. Wisdom, John. 1945. “Gods”. Proceedings of the Aristotelian Society 45: 185-206. Wittgenstein, Ludwig. 1921. Tractatus Logico-Philosophicus, translated by D. F. Pears and B. F. McGuiness (1961). London: Routledge and Kegan Paul. ———. 1953. Philosophical Investigations, translated by G. E. M. Anscombe. Oxford: Blackwell. 28