Grammar tells us what kind of object anything is. (Theology as grammar).
Wittgenstein, Philosophical Investigations, 373There is a growing feeling that something is amiss with the philosophy of religion (Scrutton & Hewitt (Reference Scrutton and Hewitt2018) ). New movements of thought call for the subdiscipline to be more ‘humane’ (Cottingham (Reference Cottingham2014) ), more aligned to religious and ethical practice (Ellis (Reference Ellis2017) ), more sensitive to affect (Wynn (Reference Wynn2005) ), and more aware of the political role of religion (Clack (Reference Clack, Scrutton and Hewitt2018) ). Where perhaps once we contented ourselves with attending to a small syllabus of problems, philosophers of religion are increasingly exploring new territory.
As new frontiers in the philosophy of religion go, however, a Wittgensteinian reading of Thomas Aquinas might seem almost uniquely unpromising. Apart perhaps from Hume, no single figure has had more impact on the philosophy of religion in the analytic tradition than Wittgenstein, breeding as he has both disciples and sworn enemies. Aquinas meanwhile has enjoyed a revival on the analytic scene ever since figures such as Geach, Anscombe, and Kenny drew attention to the philosophical depth of his work during the latter half of the last century. Moreover, the reading of Aquinas through the lens of analytic philosophy has been recognized as a distinct intellectual movement and discussed in scholarly literature (Paterson & Pugh (Reference Paterson and Pugh2006) ). Is there really anything more to be said?
I think so, since there is a loose movement of thought which differs considerably from the mainstream of analytical thomism yet is both underrepresented in the contemporary philosophy of religion and has a great deal to offer it. Sometimes called grammatical thomism, this marries the thought of especially the later Wittgenstein to Aquinas’ account of God in the first part of the Summa Theologiae. It does this in a fashion which is stylistically engaging and practically orientated, and it would be to the benefit of philosophy if it had a wider audience. The purpose of the present article is to go some way towards making that a reality. I do not suppose that what follows will win many, if any, wholesale converts to grammatical thomism – and there are clearly moves that could be made against some of the position's central claims, about the nature of the theory of meaning, for instance, or in refusing to affirm that God is a person. In this respect I am inviting the beginning of a discussion enriched by grammatical thomism, rather than supposing myself to be writing the last word. That invitation is, however, long overdue.
The two figures most associated with grammatical thomism are Herbert McCabe and David Burrell, with other thinkers such as Brian Davies and Fergus Kerr also reasonably included (gender balance is not one of grammatical thomism's merits). Their approach to philosophy of religion differs from more recent analytical thomism (although not from figures such as Anscombe, Geach, and Kenny) in its taking Wittgenstein seriously. Not unrelatedly, rather than applying a ‘thomistic’ approach to standard questions in the philosophy of religion, it often casts doubt on the intelligibility or usefulness of those questions themselves (for this reason, if for no other, grammatical thomism deserves revisiting at a time when radical questions are being asked about the philosophy of religion syllabus). I understand this approach in Wittgensteinian terms as therapeutic: we are talked away from certain unprofitable questions in the direction of more fruitful enquiry or (perhaps more typically) ethico-political praxis or worship.
Wittgenstein is a divisive figure in philosophy, and Wittgensteinian approaches to the philosophy of religion particularly prone to be misunderstood.Footnote 1 As we will see below when discussing Murphy, the Wittgensteinian focus on language by the grammatical thomists has led some to confusion about the extent to which they are committed (as they most certainly are) to a language-independent God. It is worth, then, being clear at the outset. There is no commitment to any form of non-cognitivism about theological language on the part of the grammatical thomists. Nor does the fact that they take a therapeutic approach to some questions in the philosophy of religion commit them to the view that all philosophical and theological questions are best approached therapeutically. As it happens, McCabe (for instance) is a systematic theorizer over a wide range of philosophical and theological topics: ethics being one obvious example.Footnote 2 He is perfectly prepared to advance positive claims and to regard philosophical questions as substantive and in need of answer. So it should not be thought that what follows is an attempt to co-opt the grammatical thomists for Wittgensteinian anti-philosophy.Footnote 3 Rather, without prejudice to other philosophical topics, the grammatical thomists think there is something particular about the grammar of God-talk which renders modern-day philosophical questions about God prone to a therapeutic approach.Footnote 4 In his editorial introduction to McCabe's God Still Matters, Davies captures the basic idea, and its implications for a reorientation towards action, elegantly:Footnote 5
Wittgenstein held that philosophy leaves everything as it is. McCabe held that there is a sense in which the philosophy of God leaves everything as it is since God makes no difference to anything: not because God is impotent, but because God is the reason why there is anything at all. Yet McCabe was very much concerned with the difference that people can make. (McCabe (Reference McCabe2002b), xiii)
In what follows I will reconstruct a grammatical thomist approach to the philosophy of religion, showing that by attending to how we might introduce the word ‘God' to our language we can make progress in understanding both the possibilities for and the limits of talk concerning God. I will then address objections to the grammatical thomist approach before demonstrating that approach in action, showing how it can be deployed in the cause of a therapeutic response to the problem of evil. This will lead into a brief discussion of arguably the most exciting prospect offered by grammatical thomism, the use of a philosophical understanding of God to undertake a political critique of theology and religious practice, and so further human flourishing. For most of the article I will draw on the work of McCabe, although I will refer to other thinkers where relevant. McCabe shared with Aquinas a conviction that speaking truly of God was tied up with human flourishing. If the thrust of this article is correct, they were both right in this conviction.
I
Suppose that I want to elucidate for someone the sense of the word ‘God’ as it is used in those religions commonly called Abrahamic. By ‘sense’ here, I mean that component of meaning capable of contributing to truth-apt content; surely an adequate exposition of the meaning of ‘God’ meanwhile would encompass its use in acts of praise and supplication, the making of vows, expressions of joy and profanity, and so on. Perhaps my imagined interlocutor has encountered the word ‘God’ in one or more of these contexts and wishes to know whether it is just a means of expressing emotion or an item of linguistic punctuation, or rather whether those using it suppose it to be capable of being used of something. What might I say to her?Footnote 6
My instinct is to teach her how to use the word in making assertions, by indicating to her the conditions under which canonical sentences containing the word ‘God’ may be asserted correctly.Footnote 7 Amongst these sentences are ‘God exists’ and various predications of God. Similarly, an adequate introduction to the sense of ‘God’ will convey something of the denial conditions for sentences containing the word, and of the circumstances under which its inclusion in sentential contexts constitutes a category mistake.Footnote 8 With one eye to these desiderata and the other to my enquirer's request to know whether the word God corresponds, I decide to begin my elucidation by focusing on the sentence ‘God exists’.Footnote 9
Now I take my lead from Aquinas, who in the Summa Theologiae offers famously five arguments for the existence of God. Not wishing to evaluate these arguments for present purposes, I nevertheless note that each proceeds from some feature of the world to the existence of a creator of the world which, says Aquinas, ‘all speak of as God’ (STh Ia., Q1, Art 3, ob.). This reflects what is surely a focal deployment of the word, to speak of the creator of all that is (other than that creator himself); note that our concern here is with ‘God’ as used with the surface grammar of a proper name, rather than that of a count noun – as in ‘the author of this article has three gods: coffee, Celtic football club and Buffy the Vampire Slayer’.Footnote 10
What I learn from these arguments, which cohere with the use of the word ‘God’ in talking about creation, is that I am entitled to assert ‘God exists’ just in case there is something rather than nothing at all.Footnote 11 Since there is indeed something rather than nothing at all, or else I would not be in a position to assert anything, I can say that God exists. The controversy lies a stage back, of course: many philosophers will follow Russell in denying that the question ‘why is there something rather than nothing?’ stands in need of an answer, and so will question the legitimacy of introducing an expression to designate whatever answers that question (Russell (Reference Russell1957) ). Given that this assertion condition for ‘God exists’ is legitimate, however, there are two important results concerning the logic of God-talk.Footnote 12
The first is quite simply that there is such a thing as the logic of God-talk, that it is not simply nonsense. There are intelligible circumstances in our life in which a sentence containing the word can be uttered with assertoric force, and this sentence stands in inferential relations to other sentences. Given that this sentence has a sense, it cannot be the case that its sole constitutive noun-phrase lacks a sense. This point deserves some emphasis, since recent work by Stephen Mulhall has annexed the work of the grammatical thomists to an approach to religious language influenced by the so-called resolute reading of the early Wittgenstein (Mulhall (Reference Mulhall2015) ). On this reading, somebody like McCabe, whose approach to God-talk has been given philosophical statement in the preceding paragraphs of this section, is gleefully talking nonsense, going through the motions of saying what cannot be saidFootnote 13 in order to provide us with therapy for our idolatrous proclivities, all the time keeping in our minds the riddle of existence (‘riddle’ here having the sense of Wittgenstein (Reference Wittgenstein and Ogden1922), 6.5).Footnote 14 Whatever the merits of this approach to religious language, it is quite clearly not McCabe's. He is emphatic that his intent is to safeguard the religious believer from the charge of speaking nonsense: theology ‘is not concerned with trying to say what God is but in trying to stop us talking nonsense’ (McCabe (Reference McCabe2002d), 215). It is for this reason that he lays before his reader how the word ‘God’ is used, in the context of an argument about creation, so that she can see that the word does indeed have a use, and thereby a sense.
This is not to say that there is nothing odd about the word ‘God’ for McCabe. Preserving the strangeness of our talk about God, consonant with the thomistic insistence that we cannot know what God is, McCabe often says that we do not know what our use of words about God means. This form of expression can be prone to mislead, and suggestive of Mulhall's reading of McCabe. In order to see why it does not offer support to that reading, it is important to be clear about the multiple sense of the word ‘mean’.Footnote 15 When you ask me what an expression means you might, depending on the context in which you ask, expect a number of distinct things by way of a reply. You might be content with a translation or synonym of an expression: ‘what does “boulangerie” mean?’ – ‘bakery’ or ‘c'est une magasin ou on achète le pain’. Alternatively, you might be after the reference of the word: ‘what does “Donald Trump” mean?’ – here you'd be satisfied with either a description of or an ostension at the referent. The position that every expression in a language has a reference such that its meaning can be supplied in this sense is a substantial metasemantic claim, closely linked to a representationalist view of language.Footnote 16 By contrast, that some expressions clearly are meaningful in the operative sense is uncontroversial; my name serves as one example.
Finally, a request for the meaning of an expression might be a request to be brought to a practical understanding of the expression, that is, to know how to use the expression.Footnote 17 In my view, following Wittgenstein and Dummett, this corresponds to the primary sense of meaning, and underwrites the connection between the theory of meaning and the understanding of a language (Dummett (Reference Dummett and Guttenplan1975) ) (Wittgenstein (Reference Wittgenstein, Anscombe, Hacker and Schulte2009), 43). Whether or that is correct, however, there certainly is a sense of ‘meaning’ whereby grasping meaning consists in knowing how to use an expression: ‘listen to what she's saying about Donald Trump, she can't know what the word “genius” means’. It is moreover this notion of meaning which underwrites the denial that an expression is nonsense; nothing is nonsensical that has an intelligible use in the language. In the light of this we can clarify McCabe's position on the meaningfulness of the word ‘God’ and our knowledge of it.
The word ‘God’, for McCabe, is meaningful in this last sense of having a use within the language: the assertion of the canonical sentence ‘God exists’ is licensed in response to the question why there is something rather than nothing at all (equivalently, given that anything whatsoever exists, we may assert ‘God exists’), and various inferential moves may be made from sentences containing the word on the basis of its use in answering that question. We can grasp this meaning of the word, which is to say we can have practical knowledge how to use it. On the other hand, we cannot say anything positive about the reference of the word ‘God’, since the considerations which give rise to the word's use ensure that its referent lies outside any intelligible category of being (a conclusion McCabe arrives at through a consideration of creation and divine simplicityFootnote 18). There is then a clear sense in which we do not know what we mean when we use the word ‘God’ without it being the case that any attempt to deploy the word must issue in nonsense.
The grammatical thomist invites us to consider a way-in to the use of the word ‘God’ which both secures the sense-making nature of the word and, under very minimal assumptions (the existence of anything whatsoever), the truth of canonical sentences containing it, whilst also placing severe constraints on what we are entitled to assert about God. In Wittgensteinian terms, they supply a way of understanding the grammar of the word ‘God’, which provides a basis for subsequent philosophical and theological enquiry and which does duty, in a fashion relatively uncommon in the analytic philosophy of religion, to the stress on divine ineffability so often found in living religion. It may sound too good to be true, and in the next section we will consider charges to the effect that it is.
II
We are provided by the grammatical thomist with an account of how God-talk can be introduced to our language. Two serious worries arise fairly immediately which call into question whether the grammatical thomist strategy is in fact capable of explicating theological language in a manner which supports a theism of a recognizably thomist variety. One of these relates to the question whether the issue of God's existence is trivialized by grammatical thomism, the other to whether we are being offered anything that tells us about God, as distinguished from the word ‘God’. We will tackle them in order.
On the manner of introducing the word ‘God’ laid out in the previous section, it is nearly trivial that any given language user will be entitled to assert that God exists. Any plausible account of the relationship between meaning and truth is going to deliver on this basis that it is true that God exists. All that is required is that anything whatsoever exist. Yet surely this makes a difficult question, whether God exists, far too easy. Reasonable people disagree about the question. And in any case the tradition within which we are supposed to be working, thomism, has distinguished itself by opposition to what may seem like a nearby argument, the ontological argument, and to the claim that God's existence is self-evident (STh I, Q2, Art. 1).
It is no good responding to this worry by attempting to defend grammatical thomism against the charge that it provides an easy route to realizing that God exists. We have seen that it does. A better remedy is to embrace the fact and explain why it is not the problem some might take it to be. In order to do this, we should consider why it being a ready result that God exists might be considered unfortunate. I'll address specifically intra-thomist concerns below. From the perspective of contemporary philosophy and wider culture, however, the concern is surely obvious: the question of God's existence is hotly debated by intelligent people and is pursued vigorously in ongoing philosophical research. It cannot be the case that the correct answer to this question is apparent, and to claim otherwise is a brazen exercise of intellectual hubris.
If the question whether God exists concerned the existence of a particular object in the world, to be pursued by evaluating evidence (whether empirical or argumentative) for and against the hypothesis that God exists, then it would indeed be a problem that grammatical thomism delivers such a decisive verdict in favour of the hypothesis, circumventing the weighty considerations that can be assayed against it and thereby effectively shutting down a lively research project. But it is at this point that the grammatical thomist ought to insist that understanding the question of God's existence in this way is the result of a concomitant misunderstanding of the grammar of God-talk. The word ‘God’ is not supposed to pick out an object in the world, such that it makes sense to conduct a quasi-scientific investigation of whether God exists.Footnote 19 God is not one of the items in the inventory of the world, but is rather the reason there is a world at all. This being so, it is far from clear that it is a problem taking God's existence to be potentially obvious; for in so doing we are not making assertions about the contents of the universe from our armchair, but are rather acknowledging the universe's own existence as real, contingent, and not brute.
If this is right, why do so many people, apparently competent users of the word ‘God’, including religious believers, think that it is certainly not obvious that God exists, and that there is a substantial philosophical problem concerning God's existence? That such people exist is certainly decisive witness to the fact that God's existence is not always obvious (claims of obviousness can always be met with the response, ‘obvious for whom, and when?’). Here a Wittgensteinian rejoinder is the correct one: philosophical problems arise from misunderstanding the grammar of our language (Wittgenstein (Reference Wittgenstein, Anscombe, Hacker and Schulte2009), 109). Conventional philosophical approaches to the existence of God, which for these purposes include not simply the outputs of academic philosophy of religion but also the New Atheists and many of their theistic opponents, take it for granted that the purpose of God-talk is to pick out an entity in the world, whose existence may be regarded as an hypothesis subject to evidential investigation. Those pursuing these approaches may adopt this theoretical approach towards God-talk even if their liturgical or spiritual practice, for example, or the way they talk about God outside the confines of the seminar room doesn't sit comfortably with it. The point of the Wittgensteinian diagnosis of philosophical perplexity is precisely that we may misunderstand our own language, a language we competently make use of when not indulging our philosophical instinct. In the light of this, we can see that the grammatical thomist approach to ‘God’ outlined in the previous section can be viewed as a means of drawing our attention to the correct grammar of the word by forcing us to attend to its defining place within the language-games of Abrahamic religious practice as designating the answer to a question posed by the existence of anything whatsoever, that is the Creator. Once we are properly aware of this grammar, the temptations to make illicit inferential moves from sentences containing the word ‘God’ or, more generally, to take the word to be a proper name for some entity in the world, will hopefully subside. And once we are in this position perhaps the obviousness of God's existence might not strike us as something strange.
What though about the specifically thomistic concern that the existence of God ought not to be self-evident and that, in particular, the ontological argument, which might seem to have an affinity to the grammatical thomist strategy, is unsound?Footnote 20 Taking the elements of the concern in reverse order, Aquinas’ objection to the ontological argument as such is the move from the existence of the idea of God in the mind to the existence of God in reality (STh I, Q2, Art. 1, ad 2). Not only does the grammatical thomist not endorse anything like this move, she is positively opposed to anything like an automatic move from the discursive to the theological. The correct grammar for the word ‘God’ is one thing; whether or not that grammar is applicable is another, and depends on how things are in reality. The fact that, for the grammatical thomist, the requirements placed on reality for the admissibility of God-talk are decidedly minimal does not nullify this point. Moving onto self-evidence in general, here Aquinas’ position is that God's existence is self-evident per se, since God is his own existence, but is not self-evident for us. We, not being in a position to grasp the divine essence, can only know that God exists through God's effects. The grammatical thomist does not demur. An apophaticism about the divine nature is built into her account of the grammar of God-talk, and her account of what licenses the application of this talk appeals to what Aquinas would take to be the most characteristic effect of divine action, namely the existence of non-divine entities. In both cases then, the grammatical thomist is shoring up the key thomistic doctrine rather than modifying it.
Another thomistic thought, shared widely across contemporary philosophy of religion, might however seem threatened by the grammatical approach. Aquinas takes ‘that philosophy which is part of theology’, as he describes much of what we would now consider the philosophy of religion, to be a science which has God as its object.Footnote 21 But doesn't the grammatical thomist focus attention not on God, but on the word ‘God’? Isn't a representation of God being confused for the represented reality? By eliding use and mention, the charge goes, the grammatical thomist subtly shifts attention away from what we thought we were talking about – God – and leaves us instead discussing religious language.
Religious language is a uniquely interesting part of human communication and deserving of philosophical attention. The grammatical thomist, however, like Aquinas himself, takes philosophical investigation to be capable of uncovering truths about God, not simply about words concerning God. It is an objection to the grammatical thomist position therefore if the charge of substituting language for reality can be executed successfully. Two versions of the charge should be noted. The first, directed not against grammatical thomism as such but against the general position (whether in Fregean, Wittgensteinian or Dummettian form) that philosophy has a particular concern with a linguistic or conceptual subject matter finds expression in an engaging and sustained attack from Williamson (Reference Williamson1997).Footnote 22 The other, instanced in Murphy (Reference Murphy2007), takes direct aim at grammatical thomism and charges it with failing to uphold a metaphysical realism concerning God, focusing as it does on words about God.
To turn first to Williamson, he thinks that those who regard philosophical questions as concerning language misidentify the subject matter of the discipline. A good example is provided by Williamson's discussion of the law of excluded middle (Williamson (Reference Williamson1997), 27–28). The truth of `Mars was always either dry or not dry’ cannot be a linguistic matter thinks Williamson. For instance, the sentence cannot be true in virtue of the fact that Mars always was, or was not, within the extension of the word ‘dry’. A translation test makes clear that the question whether Mars was always dry or not dry does not concern the word ‘dry’. Translated into Serbian, the question reads,
Da li je Mars uvek bio suv ili nije bio suv?
This translation, as Williamson notes, clearly does not concern the English word ‘dry’ (ibid., 28). Of itself, I would argue, this does not show that the question does not concern linguistic items; the sense of the word ‘dry’ is shared by that word and equivalent expressions in other languages and could, for all Williamson has shown, be the object of enquiry in the case under consideration. However, I am happy to grant to both Williamson and common sense that the question whether Mars was always dry or not dry is about the planet – whatever precisely being about a subject matter might involve here.Footnote 23 This much may be acknowledged however by those who view linguistic methods as central to philosophy: the point is not that the question is about language, such people would insist, it is rather that reflection on language allows us to see that the question will always receive an affirmative answer without empirical investigation (in the passage at issue Williamson is writing about analytic truth). More generally such a philosopher will insist – one in the Wittgensteinian tradition, and in particular Wiliamson's immediate target Dummett (Williamson (Reference Williamson1997), 278–292) – attention to language is distinctive of philosophical method. It is not that we are not concerned as philosophers with the extra-linguistic world, but rather that our access to that world is mediated via language.Footnote 24 As Dummett puts the matter, language `may be a distorting mirror but [is] the only mirror we have’ (Dummett (Reference Dummett2014), 6). So when he himself goes on to engage in a book-length essay on the philosophy of logic and language, he prefaces it perfectly congruently by declaring his commitment to making progress on the central questions of philosophy:
The layman or non-professional expects philosophers to answer deep questions of great import for an understanding of the world. Do we have free will? Can the soul, or the mind, exist apart from the body? How can we tell what is right and what is wrong? Is there any right and wrong, or do we just make it up? Could we know the future or affect the past? Is there a God? And the layman is quite right: if philosophy does not aim at answering such questions, it is worth nothing. (Dummett (Reference Dummett1991), 1)
The grammatical thomist applies the insights of this broad tradition of linguistic philosophy to matters of philosophical theology. The subject of enquiry for authors such as McCabe is God (were that not the case they would not be engaged in theology, philosophical or otherwiseFootnote 25), but they conduct that enquiry by linguistic means. In considering the grammar of the word ‘God’ we come to understand how reality is such that this word is used intelligibly. This understanding is gained through attention to language, but it does not follow that the subject matter of our understanding is itself linguistic. Of course, for the grammatical thomist, the extent to which we can be said to have an understanding of God at all is very circumspect. Following Aquinas, McCabe holds that we can know that God is but not what God is: the nature of God is hidden from us (STh Ia pr. Q3). Our understanding of God, such as it is, shows us that God cannot be contained within a shared category with other entities, and this fact is manifest to us in the marked difference between the usage of the word ‘God’ and that of other noun-phrases of the language, and in particular in the marked contrasts in the types of inferential move licensed by sentences containing it, in the spirit of the epigraph from the Philosophical Investigations.
To Murphy's objection that they do not move beyond language about God to the underlying reality, then, the grammatical thomist will reply that every reality is linguistically and conceptually mediated and that whilst it is perhaps particularly apparent in the case of the reality of God, of whom we can have no empirical experience,Footnote 26 that we come to an understanding about God through consideration of our theological language no more shows us not to be concerned with the divine reality than the fact that analytic philosophers have often approached the metaphysics of causation through analysis of causal language involves them in a denial of worldly causation. Attention to talk of God brings the philosopher to knowledge of God. Yet there remains a lurking suspicion that Murphy will not be satisfied. Do we really have metaphysical realism about God? If this is supposed to consist in some grasp of God which extends beyond our ability to speak then we do not (although part of what we are entitled to say about God is that God's nature is unknowable, so there is a clear sense in which God in reality transcends the limits of our speech), but then neither can we be metaphysical realists about anything else either. Grammatical thomism cannot be criticized justly for not delivering what was never there to be had in the first place.
III
The grammatical thomist begins philosophizing about religion by inviting us to consider how the word ‘God’ might be acquired through asking the question about creation – why is there something rather than nothing at all? That it may be thus introduced draws our attention to constraints on the grammar of language about God in a way that allows us to view certain apparently urgent questions about God as arising out of misuses of language. Grammatical thomism, in other words, opens up therapeutic possibilities for the philosophy of religion.
This is important since it invites a reorientation of philosophical attention away from questions which have troubled the subject since the revival of interest in religion within the analytic tradition (and, in many cases, since Hume and Kant). In this section I will reconstruct a therapeutic approach to the problem of evil on a grammatical thomist basis. The starting point is one captured well by McCabe writing about the creation question:
To say that we have a valid question (one with an answer) is to say that God exists; for what we mean by ‘God’ is just whatever answers the question. Apart from knowing this, says Aquinas most insistently, all we can do is point, as systematically as we can, to several kinds or categories of things that the answer could not be. For one thing, whatever would answer our question could not itself be subject to the question – otherwise we are left as we were, with the same question still to answer. Whatever we mean by ‘God’ cannot be whatever it is that makes us ask the question in the first place. So perishability, decline, dependence, alteration, the impersonality that characterises material things, and so on – all these have to be excluded from God. (McCabe (Reference McCabe2005e), 41)
The reasoning here parallels Aquinas in the first questions of the Summa Theologiae. We can come to know that God exists through consideration of the created world, but we cannot know what God is. In the absence of a route to knowledge of God's nature, Aquinas turns instead to consider what God is not and develops his account of divine simplicity. The grammatical thomists follow a similar path in the light of analytic philosophy's linguistic turn, asking about the conditions under which we may say that God exists and the concomitant limitations on our language about God. Once we have a proper realization of these many philosophical perplexities, including crucially the problem of evil, will be lulled.
The problem of evil in its various forms concerns whether the presence of evil in the world is compatible with the existence of God. The problem may be thought to be a logical one, where the proposition that evil obtains is taken to be prima facie inconsistent with the proposition that God exists (where God is understood as omnipotent and omnibenevolent), or it may be an evidential one, where evil is taken to count against the existence of God.Footnote 27 Either way, theists have responded with theodicies and defences of various kinds, inadequacies have been found in these, and the theistic responses honed in recognition.Footnote 28
That there are worries about the debate around the problem of evil from a Wittgensteinian perspective is not news. In particular, Phillips has patiently developed the criticism that attempts to effect a theoretical reconciliation between theism and evil are flawed morally and logically (Phillips (Reference Phillips2004) ). Yet the grammatical thomists develop an especially fundamental objection to the problem of evil debate, focused on the tacit assumption that God is a person. This line of attack finds its most sustained expression in the work of Brian Davies (Reference Davies2006; Reference Davies2011).Footnote 29 To understand how a commitment to divine personhood underwrites worries about the problem of evil, observe that the problem (in its various forms) concerns the mutual compatibility of:
(1) God exists.
(2) God is omnipotent.
(3) God is omnibenevolent.
(4) There is evil in the world.
Why suppose that (1)–(4) are not mutually compatible, where compatibility might consist in consistency (for the logical problem) or some less demanding probabilistic property (for the evidential problem)? Because we take the goodness at issue in (3) to be moral goodness and reason accordingly: a maximally morally good agent will eliminate as much evil as they are able, an omnipotent agent can eliminate all evil; God exists, but there is evil – contradiction. Yet if we take seriously the grammatical thomist caution concerning language about God, is it apparent that when we say that God is good the goodness in question is moral goodness? That move is licit if God is a person, since moral goodness is the goodness appropriate to persons,Footnote 30 but are we entitled to assert that God is a person? Certainly we use a personal noun, ‘God’, to designate God and the scriptures of the major monotheistic religions present God in personal terms, not by making the metaphysical claim ‘God is a person’, but rather by presenting God as speaking, getting angry, being in labour,Footnote 31 walking by, making his back visible,Footnote 32 and so on. Here close attention to religious language is the clue to progress; the scriptural language ought not to be ignored (after all, we are supposed to be doing philosophy of religion), but instead the question pressed how language is being used in the salient passages. The thomistic understanding of metaphor can help us understand that the genuine use of personal language to communicate about God and humankind's relationship to God needn't carry a commitment to God being a person. Moreover the later Wittgenstein's emphasis on the heterogeneity of language use will in some cases give us reason to question whether communicating information about God is the purpose of a given passage.Footnote 33 In respect of the use of the personal noun ‘God’ to designate the deity, it is here that proper attention to the grammar of the word will allow us to see that, in spite of surface appearances, it cannot be understood as referring toFootnote 34 a person. As McCabe indicates in the quotation above, because the word is introduced to stand for whatever answers the creation question, inferential moves from the existence of God to the attribution of creaturely properties to God are blocked. Our concept of personhood is, however, shot through with creaturely properties. As McCabe writes elsewhere:
For us the business of being persons is extremely closely tied up with the business of talking, of forming concepts and making judgements but there is no reason at all to transfer all this to God: indeed there are strong reasons for not doing so since this version of personality seems associated with the fact that we are physical beings, part of a larger material whole. (McCabe (Reference McCabe2005b), 9)
More than this, there is positive reason not to ‘transfer all this to God’, since God is the creator and not a creature, and is not subject to change or temporality, not simply an inhabitant of the world interacting with other inhabitants. But if God is not a person then whatever is meant by saying that God is good it cannot be that God is good in the distinctive way that persons are good, that is morally good. Now if God is not morally good then the problem of evil does not get off the ground, and can be seen as grounded in a category mistake.Footnote 35 This defuses the problem of evil as a theoretical problem, but does not of course prevent evils from posing urgent questions for human existence and colouring the way religious believers live out their faith. Once the problem of evil as a problem for philosophy has been defused, the problem of evil as a moral, political and pastoral problem remains. We will see next how the grammatical thomist approach can contribute to our understanding of the interaction between faith and politics and so provide resources for a turn towards a more practical philosophical engagement with the world in the light of a therapeutic deflation of the standing syllabus of philosophical theology.
IV
What does the way we talk of God have to do with politics? There is certainly a substantial tradition in Western thought, encompassing Feuerbach, Marx, Nietzsche, and Freud, which claims that to the extent that we involve ourselves in God-talk we become complicit in limiting human freedom. Religious responses to this tradition are familiar and have generally tried to show how it need not be the case that theology enables oppression and exploitation. Incipient in McCabe, and undervalued in engagements with his work,Footnote 36 is an alternative approach which serves to illustrate the potential value of grammatical thomism in reorientating the philosophy of religion. This takes seriously that talk of God (or the gods) is used to damage human beings and legitimate them being kept in situations which prevent them from flourishing. The apophaticism contained within the grammatical thomist approach is linked to the biblical exodus and the rejection of the ‘gods of the nations’ to motivate an account of God's status as not being amongst the contents of the world, but rather being the creator of it, as integral to God's capacity to liberate:Footnote 37
[I]t is the God of the Hebrews (who in the Jewish interpretation comes to be seen as creator) who is hailed in the decalogue as liberator; it is the gods (parts of history) and the whole religion of the gods that is seen to stand for alienation and dependency. ‘I am the LordFootnote 38 your God who brought you out of slavery; you shall have no gods.’
God the creator, who is not one of the participants in history but the mover of Cyrus and of all history, is the liberator fundamentally because he is not a god, because there are no gods, or at least no gods to be worshiped. This leaves history in human hands under the judgement of God. Human misery can no longer be attributed to the gods and accepted with resignation or evaded with sacrifices. The long slow process can begin of identifying the human roots of oppression and exploitation, just as the way now lies open for the scientific understanding and control of the forces of nature. (McCabe (Reference McCabe2005e), 43)
To reconstruct what is going on here: anthropomorphic conceptions of deity, of the kind human societies have a tendency to worship (why they do this is something on which Hume, Marx, and Freud may be able to shed light) facilitate unjust power relations and human suffering. Because, on these conceptions, God (or the gods) is an inhabitant of the world, God may be identified as causally responsible for particular features of the world (to which we may then become resigned, since they are ‘God's will’). Since God is a person like us, we might attempt to bargain with him over our condition (rather than cooperate at a human level to make it better). More generally, since as another object in the world God's agency is disjoint from our own – if, on any given occasion of φ-ing, we are φ-ing then God is not φ-ing and vice versa – then to the extent that we are pious we will not be prone to take charge of our own destiny, lest our action to change the world display a lack of belief in divine providence.
McCabe's response is that once we see clearly that God is other than the world – as both grammatical thomist natural theology and the biblical theology of creation enable us to do – we ought no longer to be tempted to invoke God as an agent in our social affairs and are set free to understand and transform them ourselves. The diagnosis of anthropomorphism is followed by a remedy, the disenchantment of society: we ought not to blame on divine agency what has been brought about by ourselves. As Marx put the point `the criticism of Heaven turns into the criticism of earth’ (Marx & Engels (Reference Marx and Engels1972), 34). Of course, for the grammatical thomist, everything in the world, and inter alia political states of affairs, movements, and events, is brought about by God as creator. The point is that it is a mistake to suppose that this excludes our own agency. God the creator is not in metaphysical competition with ourselves. Once this point has been taken on board, a new task for the philosophy of religion can be envisaged: examining the ways in which the misuse of language about God is complicit in exploitation and oppression, criticizing this, and demonstrating how a more cautious approach to speaking of the divine can frustrate this complicity. McCabe's own political passion was for the urgent work, undervalued in contemporary analytic philosophy, of tackling class exploitation (McCabe (Reference McCabe2005a) ). Complementary to this, a good deal of feminist thought about religion has drawn attention to the ways language and imagery about God can reinforce gender oppression, and this area seems ripe for engagement along grammatical thomist lines.
In an article such as this I can do no more than gesture in the direction of future work. It is a curiosity deserving of note though that, whilst the texts of the world's religions often display a passionate interest in human lives and societies, the focus of philosophy of religion in the analytic tradition has been on questions concerning God in Godself, or more latterly with specifically doctrinal questions (the coherence of belief in the Incarnation, the atonement, and so on). The grammatical thomist project, if executed more generally than it has been to date, has the potential to show that many of the questions that have troubled philosophy of religion arise out of grammatical misunderstanding and to direct our attention instead to matters of practical concern, where its insistence on the distinction between creator and creatures provides a basis for political critique.
V
‘Christianity is not a doctrine’, wrote Wittgenstein in the notes later collected together as Culture and Value, ‘not, I mean, a theory about what has happened and will happen to the human soul, but a description of something that actually takes place in human life’ (Wittgenstein (Reference Wittgenstein and Winch1980) ). I have no desire to add to the already mammoth literature on what Wittgenstein meant by the religious remarks scattered throughout his output, and I would certainly want to qualify the claim that Christianity (amongst other religions) is ‘not a doctrine’ – it certainly involves doctrines. When all that is said, though, there is a reading of these words on which Wittgenstein is profoundly right. Religion is concerned with human life, and needs constantly to be cautioned against the temptation to bury itself in abstruse speculation at the expense of that life. Grammatical thomism turns our minds back to earth, through showing us how God is the creator, the world God's creation, and ushering our thought to respect the integrity of each. In this it follows Aquinas himself and has a lot to offer the contemporary world.Footnote 39
