Introduction
Since the 1960s, metaphysics has flourished in Anglo-American philosophy. Far from wanting to avoid metaphysics, philosophers have embraced it in droves. There have been critics, to be sure; but the criticisms have received answers and the enterprise has carried on.
Matters have been different outside the so-called ‘analytic’ philosophical tradition, and particularly so in theology throughout the past century. Witness, for example, Kevin Hector's recent book, Theology without Metaphysics, which takes as a starting point the idea that metaphysics is a thing to be avoided, a succubus from whose embrace we must struggle to extricate ourselves.Footnote 1 In the opening chapter Hector writes:
Modern thought has engaged in a recurrent rebellion against metaphysics. . . . This recurrent rebellion against metaphysics indicates that although we moderns may want to avoid metaphysics, we have a hard time doing so. It would appear, in other words, that metaphysics is a kind of temptation: we want to resist it, but find it difficult to do so. (p. 2; emphasis in original)
As it arises in theology, the temptation towards metaphysics is supposed to have its origin in our natural propensity to speak positively and substantively about God. Cataphatic theology, so the reasoning goes, is inherently metaphysical. So our propensity to engage in it constitutes a temptation towards metaphysics. In turn, the concern about metaphysics is that it results in both idolatry and violence – idolatry because it shifts our attention away from God and onto a simulacrum of our own creation, and violence because it denies the otherness of God and forces God into creaturely categories. Would-be theologians are thus offered a dilemma: apophatic theology on the one hand, idolatrous and violent theology on the other.
In Theology without Metaphysics, however, Hector seeks a middle ground. He offers a broadly Wittgensteinian theory about the nature and deployment of human concepts and predicates with the goal of showing how both can be applied to God in a non-metaphysical way. In this way, he hopes to show that cataphatic theology is not inherently metaphysical, and that one can therefore engage in it without falling into idolatry or violence.
In what follows, I will argue that Hector has not succeeded in forging a path between the horns of the aforementioned dilemma. I shall begin by attempting to identify the proper target of the ‘idolatry’ and ‘violence’ objections and to provide a clear statement of each. I will then explain Hector's proposal and show how it is supposed to provide a non-metaphysical way of doing substantive, cataphatic theology. In the third and final section, I will highlight five difficulties which beset Hector's view.
Objectionable metaphysics
Despite his provocative title, Hector is explicit about the fact that he does not mean to argue that theology can or should be done without metaphysics of any sort. Rather, his aim is simply show how theology can be done without falling into metaphysics of a particular objectionable sort. Just what sort is that?
There are several ways in which one might distinguish metaphysics from other forms of theorising. One might characterise it by its subject matter, its methods, its presuppositions or some combination of these. Hector and his interlocutors (to some extent following Heidegger) seem to treat objectionable metaphysics as a mode of inquiry defined primarily by its guiding assumptions.
Hector identifies objectionable metaphysics with what he calls essentialist-correspondentist metaphysics (‘ECM’ for short). ECM, according to Hector, is
the attempt to secure human knowledge by identifying the fundamental reality of objects – their being as such – with our ideas about them. . . . [W]hat sets [it] apart [from other forms of metaphysics] is precisely an understanding of the being of beings – their essence – as that which must correspond to the ideas of a human knower. (p. 8)
Note that it is one thing to say that the essences of material objects do or must correspond to the ideas of human knowers, and quite another to identify those essences with human ideas. Since the ‘correspondence’ requirement is plausibly entailed by the ‘identity’ requirement, it is perhaps best to suppose that it is the correspondence requirement at which he intends to take aim. Doing so provides Hector with a more expansive target. Thus, we might say that objectionable metaphysics is just ECM, construed as the view that the being, or essence, of any object that falls within the purview of our theorising corresponds (in some sense) to human ideas.
I am not convinced, however, that ECM is the proper target of the idolatry and violence objections as Hector characterises them; nor am I convinced that it is the characterisation of metaphysics that those who raise such objections themselves have in mind. So I would like to suggest an alternative characterisation.
For Marion and Heidegger, both of whom loom large in Hector's text, metaphysics involves putting the ‘being of beings’ – i.e. the being of things in the world, as opposed to God, or the ultimate ground of being – on a par with Being itself, and assuming that the latter grounds but can be accounted for or explained in terms of the former.Footnote 2 The idea that Being itself can be explained in terms of the ‘being of beings’ amounts, in practice, to the supposition that human concepts – concepts shaped by the experience of beings in the empirical world – can be used to characterise God.Footnote 3
In a similar vein, Levinas, who does not feature prominently in Hector's discussion, but who is nevertheless strongly associated with the idea that metaphysics is violent,Footnote 4 takes metaphysics to be the ‘promotion of the Same before the other, the reduction of the other to the Same’.Footnote 5 The Same, for Levinas, is the thinking subject; so the idea here is that metaphysics presumes (objectionably) that the objects of our study are relevantly just like us – that they can be classified according to our own category system, understood in terms of our own familiar concepts acquired by way of our own perspectivally conditioned experience of the world.
My own inclination, then, would be to characterise ‘objectionable metaphysics’ as follows. To engage in objectionable metaphysics is to conduct inquiry in a kind of ‘self-centred’ mode, one which tacitly privileges one's own conceptual scheme and cognitive capacities as the standard by which the world is to be understood. It is, in particular, to approach a phenomenon that we wish to study or understand as if the following three claims are true of it. First, it is a being or object like me and like other things with which I am acquainted. Second, it is similar enough to me and these other things to be understood in terms of the same fundamental categories (univocally applied). Third, to whatever extent I fail to understand it, the failure is not due to defects or limitations in such concepts as I have or could acquire by further experience but rather to my failure to have a sufficiently wide range of experiences to have the concepts I need in order adequately to grasp the phenomenon I am studying. (I will generally speak of these three claims as ‘presuppositions’ involved in objectionable metaphysics, though I do not really mean to suggest that they are always or even typically explicitly or self-consciously assumed.) The violence and idolatry objections, as I understand them, apply most saliently to an approach towards God which treats God as just another object of inquiry to which human concepts can univocally be applied.
Let us begin with the idolatry objection. Idolatry for Marion is not the creation or worship of religious idols. It is a broader concept. To fall into idolatry is to direct one's thought towards a simulacrum of the phenomenon about which one aims to think or theorise rather than towards the phenomenon itself. This happens when one fails to allow things to ‘give themselves’ in their own way, or to appear on their own terms, presuming (tacitly, via our intention to locate them within our own conceptual schemes) that they will appear only in ways which conform to our concepts. Hence Tamsin Jones’ characterisation of idolatry in Marion:
[Idolatry] is the constraining of any phenomenon within limits alien to the way it gives itself or shows itself. Defining the phenomenon according to one's own subjective conceptual limitations is . . . idolatrous.Footnote 6
Accordingly, Hector construes idolatry (of the relevant sort) to involve the subjection of God to human conditions for the experience of the divine; and he takes Marion to think that concepts are themselves a kind of ‘human condition’ (p. 16). But if that is right, then it looks as if idolatry will be a danger anytime we take it for granted that God can be understood within the confines of human conceptual schemes. The proper target of the idolatry objection, then, is not ECM, but rather objectionable metaphysics. I have construed it, most saliently (though not exclusively), as the view that concepts apply to God.
Now to the violence objection. Hector characterises it in two separate places: once in discussing the work of Heidegger, and again in discussing the work of Caputo. As Hector understands Heidegger, the concern is that metaphysics (ECM in particular)
ends up equating an object's fundamental reality with that which fits within the bounds of [human] categories. The danger is obvious: if one thinks that one's preconceived ideas correspond to an object's fundamental reality, one may be tempted to force the object to fit one's conception of it, whether because one fails to see anything beyond one's conception or, worse, because one tries to make it conform to that conception. (p. 11)
Later, characterising Caputo, he writes:
Language is violent, according to Caputo, in as much as it seeks to fit objects within its horizon, to pin them down, and to hold them within its grasp. This being the case, Caputo reasons that ‘there really is nothing we can say about God that is not violent in the sense that it does not cast God in certain terms, that it does not subject God to a certain horizontality, and so set up something anterior to God, with a kind of ontological violence’. (pp. 20–1)
(The notion of ‘ontological violence’ in play is left unexplained in both Hector's text and Caputo’s, but traces to Levinas. More on this below.) Thus, the core of the violence objection, as Hector sees it, is this: applying concepts to God is violent because it ‘cuts God down’ to creaturely size, force-fitting God into ‘antecedently defined’ human categories (p. 49).
The objection thus construed seems to depend on an understanding of concepts according to which nothing can strictly and literally satisfy a concept F unless it exactly resembles, with respect to its F-ness, some other creature that strictly and literally satisfies F (cf. Theology without Metaphysics, pp. 49–50). One way, but not the only way, of motivating this idea is to suppose that concepts are just human ideas which arise out of experience, and that the extensions of concepts are classes of objects which exactly resemble in some particular respect other objects which lie within our experience. If this sort of view is right, then to say (for example) that God is wise is to say that God is at best paradigmatically wise in the creaturely way. It is surely hyperbolic to say that this violently ‘cuts God down’ to creaturely size; but it is easy enough to see that it would at least result in a distorted vision of God.
Thus far the violence objection, as Hector characterises it. But I am not sure that this fully captures the fundamental worry that the objectors have in mind. Here I think it is illuminating to consider the way in which Levinas (who is not among Hector's primary interlocutors) associates metaphysics with violence. For Levinas, as we have seen, metaphysics is ‘the promotion of the Same before the other, the reduction of the other to the Same’.Footnote 7 It is, in other words, a tacit denial of the otherness of the other, a failure to allow the other to ‘appear’ on his/her/its own terms. In less colourful and evocative language: one is engaged in metaphysics to the extent that one takes as a methodological starting point the idea that the others one encounters are similar enough to oneself that they can be accurately understood and characterised in terms of one's own concepts and categories, without regard for the possibility that they might in fact transcend those categories or somehow otherwise elude characterisation within one's own familiar conceptual scheme. In doing this, one (conceptually speaking) forces the other into a pre-cut mould; and this is where the association with violence comes in.
Violence, for Levinas, ‘does not consist so much in injuring and annihilating persons as in interrupting their continuity, making them play roles in which they no longer recognize themselves, making them betray not only commitments but their own substance, making them carry out actions that will destroy every possibility for action’ (Totality and Infinity, p. 21). One might, of course, challenge this characterisation of violence; but, taking it as read, it becomes easy to see what the connection between violence and metaphysics is supposed to be. In metaphysics, one denies the alterity of the other; one ‘tries to integrate the other into [one’s] project of existing as a function, means, or meaning’.Footnote 8 In so doing, one thereby risks ‘reduc[ing] the other to his countenance’, which, in turn, risks making the other to play a role in which she no longer recognises herself. In this way, the ‘reduction of the other to the Same’ amounts to an exercise of power over the other. As Burggraeve puts it:
In [reducing the other to myself] I approach the other not according to his otherness itself, but from a horizon or another totality . . . I look the individuality of the other, so to speak, up and down, forming a conception of him not as this-individual-here-and-now but only according to the generality of a type, an a priori idea, or an essence. . . . The ‘comprehending’ I, or ego, negates the irreducible uniqueness of the other and tries to conceive of him in the same way as he does the world. Comprehensive knowledge is thus also no innocent phenomenon but a violent phenomenon of power. By my ‘penetrating insight’ I gain not only access to the other, but also power over him. (p. 36)
These remarks apply all the more strongly when the relevant other is God, who is supposed to be radically other.
Caputo seems to have something quite similar in mind. Consider, for example, the following excerpt from the same essay from which Hector draws:
You see the idolatrous functioning of the metaphysical concept: the concept seizes God round about, measures the divine by humanly comprehensible standards, holds the look of the mind's eye captive, and cuts off the infinite incomprehensible depths of God. Lacking infinite depths, the metaphysical look is accordingly not sent off into the distance but is reflected back onto itself. A metaphysical concept of God, let us say that of the causa sui, is an image of the metaphysician. It is not inspired but constructed, not infinite but finite, not an excess but an incision into the divine.Footnote 9
The concern about idolatry is in the foreground; but Caputo's broader concern is violence, and the key ideas in this passage resonate strongly with those in the quotations from Levinas and Burggraeve in the previous paragraph.
Note, however, that the differences between these various ways of casting the objection pertain not to the target (which is, again, simply the presupposition that God can be understood in terms of human concepts) but rather to the reasons for thinking that metaphysics treats the objects of our theorising violently. Moreover, it appears that the Levinasian reason for thinking that metaphysics is violent is quite similar to Marion's reason for thinking that it is idolatrous. In both cases, the fundamental problem is that the metaphysical presupposition occludes the (perhaps quite radical) otherness of our objects of study.
So the violence and idolatry objections, as I understand them, are mainly directed not at ECM but at objectionable metaphysics as I have characterised it; and they apply most saliently to the supposition that human concepts apply (univocally) to God. Assuming this is what puts us at risk of shifting our attention away from God and onto a simulacrum, it is what most clearly manifests our tacit denial of the radical otherness of God, and it is what constitutes our attempt to ‘force’ God to fit human categories.
Responding to the objections, then, will involve establishing each of the following theses:
(1) Assuming that our concepts apply to God does not risk shifting our attention to a simulacrum.
(2) Assuming that our concepts apply to God does not deny the radical otherness of God.
(3) Assuming that our concepts apply to God does not ‘force’ God to fit into human concepts and categories.
Although Hector does not cast his project in these terms, his philosophy of language is crafted so as to guarantee each of these theses. In the next two sections, I shall summarise the main components of his view, and then highlight what I take to be its central problems.
Hector's proposal
Hector's main goal is to show that language is non-metaphysical in the following sense: concept application does not force objects into predetermined creaturely categories, nor does it presuppose that concepts express or correspond to the essences of the things which satisfy them. Key to accomplishing this goal is the deployment of a robust theory of concept use, meaning, reference and truth evaluation. Here I shall focus primarily on what Hector has to say about concept use, but I shall begin by laying out the main components of the overall theory.
Hector summarises his philosophy of language helpfully as follows:
I propose a[n] . . . account of language, according to which (a) to use a concept is to intend one's usage as going on in the same way as certain precedents and to claim this same precedent status for one's own usage; (b) the meaning of a concept is a product of the normative trajectory implicit in a series of such precedents, which entails that a concept's meaning changes, if only slightly, each time a candidate use is recognized; (c) to refer to an object is to link up with a chain of precedents that carries on the normative commitment implicit in an initial picking-out, in such a way that one inherits (and renders further inheritable) that commitment; and (d) to judge some proposition to be true is to see it as going on in the same way as one's other commitments and to use it to judge still other propositions. One can thus arrive at an account of concepts that do not ‘contain,’ of meaning without ‘meanings,’ of reference without ‘presence,’ and of truth without ‘correspondentism’. We arrive, in other words, at an account according to which language might be fit for God-talk. (p. 38)
Note that, thus far, we have only the claim that language might be fit for God-talk. To show that it is fit for God-talk – that theology can be done without idolatry or violence – Hector adds a further component to his theory. This further component is a story about how ordinary concepts come to apply to God; and central to this story is a view about the work of the Holy Spirit.
Drawing on the work of Schleiermacher, Hector develops a view according to which the Spirit of Christ works so as to shape the meanings of our concepts in a way which makes them applicable to God. The details of this theory are complex; but for present purposes the following rough summary should suffice. According to Hector, Jesus taught his disciples how to follow him; he also recognised them as competent at doing so and as competent judges of what counts as following him. They, in turn, did likewise for others, and so on down through the history of the church. This is important because, by virtue of his own divinity, Christ himself faced no problem in applying concepts to God; he was able to ensure that the ‘normative trajectories’ implicit in his own use of the concepts he applied to God had God as their ‘fulfilment’. (Whatever else it might imply, having God as the ‘normative fulfilment’ of a concept guarantees that the concept is applicable to God.) Furthermore, Hector thinks, learning to follow Christ is, in part, a matter of learning what it is to ‘go on in the same way as Christ’ in one's own God-talk. Learning to ‘go on in the same way’ as someone in one's own use of concepts, in turn, involves becoming attuned to the normative trajectories implicit in the ways the other person uses her concepts. It is this sort of attunement that, with the aid of the Holy Spirit, has been passed down by and through the church so that we twenty-first-century theologians can go on in the same way as Christ and can reliably recognise others as doing so. Thus, Christ's own ways of using concepts are among the normatively authoritative precedent uses for our own, and they are linked to our own God-talk via a chain of further precedents, all of which go on in the same way as Christ’s.
So, in short, we can apply concepts to God because Christ's concepts applied to God, and at least some of the concepts which we apply to God are continuous with Christ’s. This does not tell us just how many of our concepts apply to God; and, as we shall see, real sceptical monsters lurk in the neighbourhood. But I shall leave that concern to be developed later.
The question now is why exactly this view allows us to say that concepts apply to God without falling prey to the idolatry and violence objections as they were characterised earlier. Hector takes the answer to lie in part (b) of the executive summary of his position:
the meaning of a concept is a product of the normative trajectory implicit in a series of such precedents, which entails that a concept's meaning changes, if only slightly, each time a candidate use is recognized.
One might question the entailment claim: is it really impossible for there to be, over time, a perfectly stable normative trajectory in the usage of some concept? Has, for example, the meaning of the concept of pi changed in (say) the last twenty minutes as the concept has been introduced or employed in a myriad elementary school classrooms? But never mind this for now. What is important is just the claim that concept meaning does change over time with use; and that meaning is a function of the concept's inherently unstable and therefore not entirely predictable normative trajectory. The result is that concepts are open-ended and vaguely bounded in their extensions. In other (perhaps more familiar) terms: concepts are vulnerable to semantic drift, so much so that it is indeterminate to what objects they might eventually apply.
From this, then, it is supposed to follow that concepts do not necessarily correspond to fixed, unchanging essences of things. Nor do they necessarily correspond to or constitute ‘predetermined’ or even well-defined categories. According to Hector,
if a concept is the product of a series of precedent uses, and if this series changes every time a new use is recognized as carrying it on, it follows that concepts are continually being reconstituted. This means . . . that it makes no sense to talk about the meaning of a concept being fixed by its application to creaturely reality . . . (p. 97)
This is partly because it makes little sense to talk about ‘the meaning’ of a concept, and partly because concept meanings are not fixed. Furthermore, it follows on this view that applying a concept to God does not imply that God exactly resembles any creaturely paradigm; for, on this picture, the extension of a concept is not determined by anything like exact resemblance to paradigms. Thus, according to Hector, applying concepts to God no longer threatens to ‘cut God down to size’, forcing God into merely creaturely categories. Nor is it plausible to think of concept application on this view as contributing essences to things. After all, if what it means to say (for example) that God is wise might change over time, it is hard to see why one should think that calling God wise makes God into something which is essentially (and thus unchangeably) characterisable by the present content of the concept wisdom. Likewise, Hector might say, once we appreciate the fact that concepts work in this way, concept application does not even threaten to distort our vision of God; nor does it threaten to ‘make [God] to play a role in which [God] no longer recognizes [God-self]’; nor does it ‘reduce [God] to [God’s] countenance’. For it will be built into the very idea of applying concepts to God that much is being left out and that the sands are always vulnerable to some shifting. Thus the violence objection is addressed.
Nor does it ‘subject God to human conditions’. According to Hector, ‘the norms by which one assesses one's [theological] concept use are not external to God – and neither are the concepts themselves. On the present account, one learns how to use theological concepts by submitting one's performances to Christ, and one is able to submit them to Christ through the power of the Spirit. . . . One conceptualizes God through God, which is to say that one conceptualizes God by grace alone’ (p. 96). This then does away with the concern about idolatry.
Problems
As noted earlier, there are further components to Hector's overall theory which merit close attention and serious discussion. But I think that the material thus far presented provides a clear enough picture of where the view is going for us to see some of the problems it will face. In this section, I shall discuss five problems, each of which strikes me as rather serious.
Let us begin by noting that one possibility not at all ruled out by Hector's theory is that there are, after all, sharply bounded (non-vague) properties which ‘correspond to’ or serve as the content of our concepts, and that the instability of our concepts over time has simply to do with the fact that concepts take different properties as their precise contents at different times, depending on vagaries of usage.Footnote 10 Indeed, we might even suppose that there are determinable properties (like being wise) which always serve as the ‘normative fulfilment’ of the concepts (like ‘wisdom’) to which they correspond, but that the precise content of the concept at any given time is some determinate of the determinable (for example, Socratic wisdom, or creaturely wisdom, or some other more specific sort of wisdom). On this picture, everything that Hector says about how concept meaning depends on ‘going on in the same way’ as precedent uses, and about the instability of concept meaning, might still be true; but it will nonetheless be a picture according to which concepts, despite their instability, may still correspond to fixed properties, or categories.
Now consider this fact in light of Hector's stated goal of ‘free[ing] us from the metaphysical picture of concepts, according to which one's use of a concept corresponds to an essence-like idea or “meaning” and so [also] free[ing] us from the sense that concept use fits objects into pre-determined categories’ (p. 48). What is a predetermined category? One answer is that it is nothing more or less than a sharply bounded property which objects might or might not objectively, mind-independently exemplify. But if this is right, then by not ruling out the possibility described in the previous paragraph, Hector's view fails to free us from the problematic ‘metaphysical picture of concepts’. Concepts will still correspond to determinate properties, and so, in applying them to objects, we will still be sorting things into predetermined categories. Note, too, that they will still be characterisable as familiar, human categories; for the properties to which our concepts correspond would, in order to secure the correspondence, have to be humanly graspable.
Moreover, if predetermined categories are nothing other than mind-independent, non-vague properties, Hector's talk of ‘freeing us’ from the metaphysical picture of concepts does not make sense. For to say that objects objectively and mind-independently exemplify certain properties is, among other things, to say that human conceptual activity is not part of what makes it the case that they exemplify those properties. Thus, simply by virtue of believing in mind-independent property exemplification we are already freed from a picture according to which human concept usage fits objects into predetermined categories. Objects either fit into those categories or not under their own steam, as it were. Human concept usage does not generate the facts about how objects fit into categories; it simply tries to represent those facts.
Another alternative – and the more plausible one, in light of what has just been said – is that the predetermined categories that Hector has in mind are mind-dependent essences. Suppose, for example, that being a statue is not an objective property of a gold statue, but rather depends in some way upon our thinking of it as a statue rather than (say) as a mere lump of gold. Hector's view, of course, implies that the concept ‘statue’ might shift in meaning over time; so the view allows for the fact that what counts now as a statue might not count in the future (or might not have counted in the past) as such. But it doesn't follow from this that applying the concept now to this particular piece of gold doesn't fit the piece of gold into a predetermined category. Indeed, on this picture, our thinking of the piece of gold as a statue (rather than a mere piece of gold) is part of what makes it a statue, and it is part of what gives the statue persistence conditions like being unable to survive being recast in the shape of a cube which are not otherwise had by a mere piece of gold. So here too it seems that Hector's theory of concepts fails to achieve his goals.
Second problem: Suppose we grant that Hector's theory succeeds in showing us how concepts (in general) can apply to God. Still, it is not clear that the theory shows us how ordinary concepts can apply to God. In showing how we can use his model to illuminate theological concept use, Hector writes:
one's concepts are applied correctly to God . . . if they are faithful to God's revelation in Christ, and they count as such if they go on in the same way as precedent applications which carry on the normative trajectory that stretches back to Christ. . . . To count as using (Christian) theological concepts . . . one's use must be recognizable as trying to carry on the normative trajectory implicit in this chain [that stretches back to Christ] (p. 95)
This is all fair enough; but the thing to notice is that the criterion for the use of Christian theological concepts is not a criterion for the use of ordinary concepts. Thus, unless it can be shown that, as a matter of contingent historical fact the normative trajectory implicit in the use of some ordinary concept C is recognisably identical to the normative trajectory implicit in some chain of precedent uses which stretches back to Christ, we have no reason for thinking that C is a ‘Christian theological concept’, and so no reason for thinking that C properly applies to (the Christian) God.Footnote 11
To illustrate, consider the concept ‘wisdom’ as it is used in the predication ‘God is wise’. Suppose we grant that Christians who say that God is wise are recognisably trying to carry on a normative trajectory implicit in a chain of precedent uses of the term ‘wise’ which stretches back to Christ. Suppose further that we grant that, by virtue of this, ‘wisdom’ applies to God. Still, it does not follow that any ordinary concept applies to God; for it does not follow that the theological concept ‘wisdom’ is identical to (or even intelligibly related to) the ordinary concept which goes by the same name. To show that wisdomtheological = wisdomordinary, we would have to show what Hector has not: namely, that the normative trajectory implicit in non-theological uses of the concept ‘wisdom’ is recognisably the same as the normative trajectory implicit in the chain stretching back to Christ of precedent uses of the theological concept. I do not know how one could show this; but without showing this, it seems that Hector's philosophy of language inadvertently gives victory to the apophatic theologians. Ordinary concepts do not apply to God. Rather, we can speak of God only in a special theological language, employing concepts which, for all we can tell, may not even be intelligibly related to our ordinary ones.Footnote 12
Third problem: Thus far I have been assuming that Hector's theory of concept usage is viable, at least in broad outline. But this too is questionable. For Hector, concept usage depends upon judgements to the effect that one concept is the same as another; it also depends upon the existence and recognition as such of precedent uses. Thus:
To use a concept is to recognize the concept use of others as having a certain normative authority over one, and to claim this same status for one's own use. To count as using a particular concept, one's usage must be recognizable as using the same concept as others, which is to say that it must go on in the same way as uses that are recognized as precedential – that is, as uses of the concept in question. (p. 62)
But it is easy to see why the requirement that there be recognised precedent uses will pose a problem. Assuming one has to use a concept in order to introduce it, the requirement implies that concepts can never be introduced. It also implies (absurdly) that behind every instance of concept use is an infinite series of precedent uses (since any precedent use will count as the use of a concept only if it goes on in the same way as some precedent). Of course, one can fix this problem by stipulating that one can also use concepts without precedent at the moment of introduction. But then one faces a dilemma: either concept introduction is always violent (and theological concept introduction always idolatrous) or else Hector's elaborate story about how concept use depends on normative trajectories implicit in chains of precedent uses is ultimately irrelevant to thwarting the idolatry and violence objections.
But what of the claim that concept usage depends upon judgements to the effect that one concept is the same as another? Here we face a dilemma. Either the judgements must all be explicit, or not. Suppose they must all be explicit. Then we face a vicious regress. Why? Because making the judgement ‘A is the same concept as B’ involves applying the concept ‘sameness’ to A and B. In order to apply that concept to A and B, however, I have to judge that my concept ‘sameness’ is the same concept as one involved in precedent applications of the term ‘sameness’. To make this judgement, however, I have to use the (or, better, a) sameness concept again; and to do that, I have to judge it to be the same as precedent uses, thus using another sameness concept, and so on.
Better, then, to say that the judgements need not be explicit. And, indeed, this is what Hector does seem to say (pp. 64ff.). Hector speaks of the ‘implicit recognition’ that one instance of concept usage goes on in the same way as others, and he also seems to want to allow that one can ‘implicitly try’ to go on in the same way as others (which is good, because otherwise trying to go on in the same way as others would generate the same regress of concept application) (pp. 81–2). But now we face in spades the problem of determining when a person (or even we ourselves) intends to go on in the same way as someone else. Suppose I say that God is wise, but I am neither explicitly (i.e. consciously) judging myself to go on in the same way as Christ nor explicitly intending to go on in the same way as Christ. We might ask what makes it the case that I am doing these things. But even if we waive that worry, Hector's conditions for concept usage require not only that I be doing them, but also that I be recognisably doing them. But if I have not even acknowledged to myself, so to speak, the relevant judgements or intentions, how could others recognise them in me? The cases where I am implicitly trying to go on in the same way as Christ might be indistinguishable in practice from the cases where I am implicitly trying to go on in the same way as someone else who is working with a subtly but genuinely different concept of wisdom.
Fourth problem: According to Hector, a candidate performance counts as going on in the same way as precedent performances if and only if it is recognisable as such by those who know how to undertake such performances (p. 68). But we might well ask what ‘recognisable’ means. If the idea is that one performance goes on in the same way as another if and only if someone who knows how to undertake such performances is capable of forming the belief that it goes on in the same way, then concepts are a lot more open-ended than Hector has led us to believe. Indeed, it would seem that anything goes; for, after all, given world enough and time, together with the fact that those who know how to undertake various kinds of performances are always capable of forming deviant or mistaken beliefs about what goes on in the same way as precedent performances, it would seem that any concept is capable of morphing in virtually unlimited ways. On the other hand, the idea might simply be that there are fixed (and recognition-independent) standards for ‘going on in the same way’ as precedent performances, and that satisfying these standards somehow guarantees that people who know how to undertake the relevant performances can see whether the standards are satisfied. If this is right, however, then every concept will have fixed boundaries after all, for present uses of a concept count as such, on this view, only if they go on in the same way as all of their precedents. But if concepts have fixed boundaries, then they start to look like fixed, predetermined categories after all.
Finally, a fifth problem: Consider the concept of using (or applying) a concept. Marion and Caputo have a concept (call it ‘C1’) of using or applying a concept according to which applying a concept to God is idolatrous and violent. Hector has a concept (call it ‘C2’) of using or applying a concept according to which it is not the case that using or applying a concept to God is idolatrous or violent. However, if Hector is to be taken as responding to Marion and Caputo rather than simply talking past them, C2 must be identical to C1. Or, at the very least, the two concepts must overlap significantly in their meaning. But in order for that to be the case, Hector must be ‘going on in the same way’ as Marion and Caputo; and it is not immediately evident that he is. Admittedly, neither is it evident that he is not. The term ‘going on in the same way as’ is rather vague. He is clearly not going on in a way identical to that in which Marion and Caputo are using the term. But, as discussed in n. 11, ‘going on in the same way’ in fact requires only that he be going on in a sufficiently similar way.
How similar is sufficient? It is hard to say; but there is at least some evidence that Hector's way of going on is not sufficiently similar to Marion's and Caputo’s. For one thing, we know that his concept of a concept is different from theirs. Hector's idea of concepts, elaborated so thoroughly in his book, is intended as a rival to the idea of concepts which gives rise to the idolatry and violence objections. Insofar as Hector has also developed his own theory of concept usage (also intended to contribute to defusing the objections), we have good reason to think that his concept of using a concept – and so his concept of applying a concept – is different from theirs. If that is right, then we should probably deny that Hector intends to go on in the same (relevantly similar) way as Marion and Caputo. Indeed, it seems clear that he means to urge all of us to go on in a relevantly different way from them in our own usage of words like ‘concept’ and ‘apply a concept’. If I am right about this, then, by the terms of his own theory, Hector is not so much responding to Marion and Caputo as he is simply talking past them.
Conclusion
I have argued in this article that, despite its ingenuity, Hector's theory of concepts and concept application is not adequate to the task of rescuing cataphatic theology from the idolatry and violence objections. Assuming those objections do not fail for other reasons, the dilemma thus remains: apophatic theology on the one hand, or idolatrous and violent theology on the other.
Despite what I have argued in this article, however, my broader sympathies do lie with Hector and other cataphatic theologians. I agree that the dilemma is a false one; I believe that cataphatic theology can be done without idolatry or violence. This is not the place to develop my own response to those objections in detail. However, I would like to close with a few very brief remarks reporting my own view of where the objections go wrong.
Both objections turn on the idea that applying concepts to worldly phenomena somehow distorts our vision, so much so that it is no longer the phenomena themselves that we have in view but rather idols, and so much so that we do a kind of violence to the phenomena by reducing them to what is familiar and humanly graspable, and by privileging our own particular perspective on them, rather than allowing them to appear on their own terms and attending to them in ways which respect their alterity.Footnote 13 The idea seems to be that, in providing what we take to be an objective theoretical description of something in terms which are familiar and intelligible, we will somehow inevitably take ourselves (incorrectly) to have a full, complete and perfectly accurate grasp of something which in fact can never be fully, completely, or perfectly accurately grasped; or, worse, we will inevitably make it the case that the thing is as we conceive of it rather than simply allowing it to be what it is in and of itself.
The latter, of course, is a danger only if the anti-realists are correct and human conceptual activity plays a role in constructing the world.Footnote 14 To the extent that the idolatry and violence objections depend on this sort of anti-realism, I think that that is where they go wrong. The answer, then, is not to reconceive the nature of human concepts and human concept application as Hector tries to do, but rather to reject the underlying picture of how concepts relate to objects and their properties.
As to the former – the idea that concept application somehow presupposes that our grasp of worldly phenomena is more complete and accurate than it ever could be – I think that the answer is again not to reconceive the very nature of human conceptualisation but rather to cultivate an appreciation for the richness of the properties to which many of our concepts refer. Heidegger says (with many others in his wake) that ‘the difficulty lies in language’ (Identity and Difference, p. 73). My own view, on the other hand, is that such problems as there are in this neighbourhood lie more in our own contingent attitudes towards the objects about which we theorise and the things that we say about those objects. To say that God is wise, for example, is (as I see it) to say no more or less than that God has a certain rather familiar property – wisdom – a property also had by many human beings. In saying this, we might presume many further things: that wisdom is a universal, God is a particular, and God exemplifies wisdom; that divine wisdom has exactly the same qualitative character as human wisdom; that our grasp of wisdom as we experience it in humans is so thorough as to allow us to deduce many other things about God from the simple thesis that God is wise; that wisdom is part of God's essence, and that in grasping wisdom (and maybe a few other familiar properties) we have a thorough and secure understanding of the totality of the divine essence, and so on. We might presume these things, but we need not do so. All of these claims and more are up for grabs as we theorise further about what we mean when we say that God is wise, about the extent to which that claim can be thought of as literal or metaphorical, and so on. Divine wisdom must bear some intelligible relation to human wisdom in order for the predication to be apt; but we might well be ignorant of exactly what that relation amounts to. So long as we steadfastly maintain this understanding of the limits on what it means to say things like ‘God is wise’, it is hard to see how the idolatry and violence objections can have any real purchase. So, in the end, I think that we can have theology without idolatry or violence. My difference with Hector simply concerns the reason why.Footnote 15
