Int J Philos Relig DOI 10.1007/s11153-014-9483-0 ARTICLE

The epistemology of divine conceptualism Nathan D. Shannon

Received: 9 April 2014 / Accepted: 11 September 2014 © Springer Science+Business Media Dordrecht 2014

Abstract Divine conceptualism takes all abstract objects to be propositions in the mind of God. I focus here on necessary propositions and contemporary claims that the laws of logic, understood as necessarily true propositions, provide us with an epistemic bridge to theological predication—specifically, to the claim that God exists. I argue that when contemporary versions of DC say ‘G/god’ they merely rename the notion of necessary truth, and fail to refer to God. Given that God is incomprehensible, epistemic access to the state of propositions in the mind of God is extremely limited. Keywords

Divine conceptualism · God and abstract objects · God and logic

Divine conceptualism is the attempt to preserve the ontology of propositions1 without compromising a robust theism by taking propositions to be thoughts in the mind of God, giving us, we hope, a workable asymmetrical relation between them.2 Though this theory has able defenders both historic and contemporary, I have a few concerns about it. The principle concern I have has to do with the incomprehensibility of God

1 Divine conceptualism attempts to reduce all abstract objects to ideas or concepts in the mind of God, not just propositions. Here I focus on propositions. 2 One way to achieve this asymmetrical relation between God and propositions is to say that propositions are essentially the objects of thought; they are parasitic on minds. This is true when propositions are thought to be truth-bearers. So if propositions are either true or false, they must be the objects of doxastic intentionality. In the case of necessary propositions, the doxastic intentionality can only be God’s. Thus the truth-bearing quality of necessary propositions is thought to imply the existence of a God. Alvin Plantinga claims that propositions are parasitic on minds, as do James Anderson and Greg Welty (see below).

N. D. Shannon (B) Department of Systematic Theology, Westminster Theological Seminary, 2960 Church Rd., Glenside, PA 19038, USA e-mail: [email protected]

123

Int J Philos Relig

and our epistemic right to speak as we do about propositions existing in the mind of God.3 If propositions are essentially about something, as they must be in order to do the philosophical work we need them to do (bear truth, for one thing), they must be about something other than themselves or other abstracta; they must be about real, concrete particulars.4 So when these propositions are taken to be thoughts in the mind of God, we face the question of how it is that God can have doxastic intentionalities5 which are (1) eternal and necessary; and (2) essentially ‘about’ something other than God, even contingent particulars.6 The so-called traditional theist is primarily inclined to avoid any encroachment of the contingent, created order on the eternal self-existence of God alone. One way to preclude this encroachment has been to say that God eternally has in his mind an exhaustive (probably infinite) library of complete sets of logically consistent propositions, called possible worlds. This is thought to help because before God creates (or ‘actualizes’ one of them), all possible worlds are merely possible; no single world enjoys modal or ontological privilege (no possible world is actual or more real than any other). 7 So there is no sense in which any state of affairs, and 3 The theologian will notice that this is in part a question of theological prolegomena. 4 This is less obvious for logical truths, perhaps, but we end up with the same result once logical truths

become thoughts: like all thoughts, they must be about something. 5 I say doxastic intentionalities because, on (divine) conceptualism, for a proposition to have truth-value

it must be more than a mere idea or mental representation; it must be believed or doubted or considered or affirmed or denied; it must be the object of doxastic intentionality. This is a problem unique to conceptualism in the sense that some epistemological account of truth must be introduced once propositions are recognized as essentially cognitive items or objects of thought; and divine (or non-divine) conceptualism is motivated in part by the idea that propositions are parasitic on minds. Plantinga writes, ‘truth is not independent of mind; it is necessary that for any proposition p, p is true only if it is believed, and if and only if it is believed by God. This is truth de dicto; but it is also true, de re, that every proposition has essentially the property of being true only if believed, and if and only if believed by God’ (Plantinga 1982, p. 68). So to follow through on this notion, it would appear to be inadequate simply to take propositions to be ideas in the mind of God, since ideas alone are not truth-bearers; so we must suppose that God believes or at least thinks the relevant propositions. Thus we expect of him some doxastic disposition or attitude. See Davis (2011), pp. 297–298. James Anderson and Greg Welty claim that the laws of logic are God’s thoughts about his own thoughts. On their account, God thinks the laws of logic: ‘The laws of logic are nothing other than what God thinks about his thoughts qua thoughts’ (Anderson and Welty 2011, p. 337). But presumably God doesn’t merely ‘think’ the laws of logic, if that means having them in mind and nothing more; on their account he must believe them to be true. 6 A strong Thomistic doctrine of simplicity makes a real mess of this, but set that aside. Many of the

historical sources attempt to ease the tension by saying that the objects of God’s thoughts are the objects of his necessary will (himself only) and the objects of his free will, the latter being divisible between those he realizes and those he does not. All things not contrary to God’s nature being possible and realizable by God, God can know all possible things. The problem of God knowing actual things remains, however. 7 This also means that for any proposition P, prior to actualizing a possible world, God cannot know P but only it is possible that P. So I suppose then that we would say God knows the embedded proposition P plus its modal qualifier. Prior to creation, God knows some propositions as possible in the same way that given the actual world God knows only in a modally qualified sense that I have red hair. Some Thomistic theologians will deny that we may draw a distinction between the cognitive state of God ‘before’ creating

Footnote 7 continued the actual world and ‘after’ doing so, since everything that God does constitutes a single, eternal, inseparable act identical with his being.

123

Int J Philos Relig

derivatively any object of doxastic intentionality (any particular proposition or the object of its aboutness), infringes upon the self-existence and self-definition of the creator. The question that may be asked is whether we have not, on even this cautious approach, still leveraged on the actual world the notion of possibility which is constitutive of the possible worlds in God’s mind and even of a few of God’s doxastic attitudes, even prior to creating. It seems impossible not to: we are after all contingent beings, actualized in the actual world. Surely our modal categories are our modal categories. So at least in this elementary, though possibly inconsequential sense, possibility is a function of an actual but contingent state of affairs, rather than a necessary one; perhaps our modal categories are like ships of various sizes tied to one another, none anchored securely. Perhaps our modal categories are linked not to the actuality of the self-existent God, but to the actuality of a single, contingent world—this one—and then imposed upon God, making God in his own nature (he has one, yes) correlative to the created order. I say this might be inconsequential because it may very well be that the logical bedrock which secures our notions of possibility (the laws of logic as we know them) may very well be as basic as can be, in which case this problem would not arise. In other words, we are ok (we have not created God according to our own image) so long as logic as we know it is already the logic of God himself. Anderson and Welty have argued exactly that. They claim that the laws of logic as we know them are eternally and necessarily in the mind of God, they are necessarily, the objects of eternal, divine cognitive intentionality.8 This is pure speculation of course—it is natural theology—but it is easy to sympathize with. It is easy to sympathize with because if we do not assume that possibility as we know it is anchored in the selfcontained divine mind, then (unless logic just is God) the reach of our modal categories is severely limited, cut off at Lessing’s divide, and this solution to the problem of abstract objects—a leading candidate for the theologically inclined—is in trouble. Thus, in my view, this is more or less a leading concern for divine conceptualism, at least for the traditional theist: do we have epistemic rights to put these laws of logic in the mind of God? As no doubt the reader will have noticed, I harbor an openness to the possibility that the laws of logic as we know them do not exist necessarily, in the strong sense in which this is usually taken, but only given a few things (whichever things get us from God’s being uncompelled to create all the way to the actual world). Put more precisely, I think there is rather too much confidence (exaggerated epistemic license, we might say) in the claim that the laws of logic as we know them do in fact exist necessarily, even for God, in the very mind of God. To many, I imagine, my view is downright laughable. I enjoy a good laugh as much as the next guy; but perhaps in what follows I can salvage a bit of self-respect. I am not saying that ‘I can conceive of’ or that ‘I can imagine’ a world in which the laws of logic do not obtain or in which they are false or different. But to say that one cannot do so is no argument either. We would not want to argue from ignorance that

8 Anderson and Welty (2011), esp. pp. 335–336.

123

Int J Philos Relig

if we cannot imagine a thing, then none exists or even can.9 As Keith Yandell writes, ‘Appeals to intuitions in contemporary philosophy are largely fancy ways of saying that one finds a claim clearly true without being able to offer an argument for it’ (Yandell (2011), p. 287).10 So the question is not so much whether we can conceive such a world, the question is rather whether we can demonstrate that if there were no actual world and there was only God, then the laws of logic would still exist as we know them (or at least establish that this is the more rational alternative). Notice that the question we now have is theological, since it cannot be answered without information about the nature of God. I believe that at this point proper workflow requires the particular skills of the exegete and the theologian. Let me put it differently. When we run up against the impeccable credentials of the laws of logic, why conclude that all possible worlds—all possibility whatever, even what God can be and do and think—must prove consistent with these laws? Because they are, to us, demonstrably unimpeachable? I think rather that if the laws of logic are evidently necessarily true, and it is inconceivable that they are false, and there is no comprehending the possibility of their not obtaining, then for that reason there is something to be said about the ultimate incomprehensibility of God relative to these propositions, and if these then all others too. From the nature of the case, we cannot suppose that anything demonstrably true is the key to or the governing principle of the being, activity, or mind of the incomprehensible God. According to our lonely but courageous traditional theist, exegesis of Scripture gives us a God who is one essence in three persons. The Son, the second person, is God of himself as to essence, but as Son (as to his person) he is derivative of the Father. The Son possesses of himself the single, self-identical divine essence non-derivatively; but as to his person he is distinct from the Father and the Spirit.11 The same—essential self-existence and personal distinction—may be said of the Father and the Spirit. And, traditionally, God in his triunity is a se—that is, self-existent and self-defined. What is the logic of the one God in three persons? By what logic is God bound?12

9 Anderson and Welty write, ‘We cannot imagine a possible world in which the law of noncontradiction is false’ (ibid., 326). And, ‘the law of noncontradiction is true not only in the actual world but also in every possible world’ (ibid., 325). The latter statement is trivial if taken on a conceptualist ontology of propositions (as suggested by the former). Suppose there are no possible worlds. In that case, a necessary proposition would be no more necessary than its negation. There are no possible worlds in which ∼P; so P. Equally, there are no possible worlds in which P; so    ∼P. P and ∼P are both necessary. (I borrow this argument from Leftow (2012), pp. 58–59). So there must be at least one possible world for P to be true and ∼P false. And so for a necessarily true proposition such as the law of noncontradiction, to say that it is true in all possible worlds is just to say that there are possible worlds. A possible world just is a logically consistent state of affairs. This triviality encourages us to look for a truth-maker (or to satisfy the ontological truth-condition) for logically necessary truths—which is what Anderson and Welty do. On the other hand, Leftow says, ‘We can clearly conceive circumstances in which the truths of classical logic would not be true’ (ibid., 89). He then references the non-classical, three-valued logics of Kleene and Lukasiewicz and concludes: ‘If this sort of treatment is correct and there are gap propositions, then, PNC [the principle of non-contradiction] is not true (nor false)’ (ibid., 90). More below. 10 Yandell (2011), p. 287. 11 In Calvin’s view, this is the teaching of Scripture. See Ellis (2012). 12 For one approach to this question with particular attention to triunity, see Poythress (1995, 2013).

123

Int J Philos Relig

Minimally, the creator/creature distinction is the idea that God is the original, incomprehensible but fully self-comprehended, self-sufficient I AM, and the creation is dependent upon him, derivative of him, and utterly comprehended by him.13 So on the one hand, there is continuity: the created order in some way displays the nature and character of the creator God and even depends upon him as providential sustainer;14 on the other, there is discontinuity: a clean ontological distinction between God and everything else, such that without divine, voluntary condescension—including voluntary self-revelation—there is no knowledge or ‘fruition’ of God.15 To suppose that the laws of logic as we know them obtain as we know them in the mind of God is to exaggerate by the force of contingent, fallible intuition the continuity between the creator and the creature to such an extent that no discontinuity is recognized. This is not to say that discontinuity is denied; but neglecting discontinuity is fault enough. To the question, ‘what, then, do necessarily true or necessarily existing propositions look like in the mind of God?’ ‘I don’t know’ is an under-valued response. Leftow says the following: Perfect-being theology is inevitably speculative . . . I give many perfect-being arguments . . . As I give them, I have a nagging fear that I am just making stuff up. . . our ideas of what it is to be perfect are inconsistent and flawed, and there is no guarantee that they match up with what God’s perfection really is. Our intuitions about absolute possibility, again, probably are not wholly reliable; probably we sometimes take as absolutely possible what is merely epistemically possible, possibly true ‘for all we know’, where we do not know what we would need to know to rule out the candidate possibility. And perfect-being theology is one sort of metaphysics, and so inherits a share of our worries about whether metaphysical reasoning ever shows us more than the insides of our own concepts.16 Even if, from the theologian’s point of view, Leftow’s biblical hermeneutic of perfect-being theology leaves something to be desired,17 he has a healthy sense of the 13 I use the masculine pronoun, following the inspired writers of Scripture. 14 Scott Oliphint thinks of the creator/creature relationship, and a distinction between God as he is to

himself and God in relation to creation, in terms of an eimi/eikon analogy. See Oliphint (2011), esp. pp. 89–93. 15 This would include the act of creation, so that God’s first act or decision ad extra constitutes condescen-

sion. Westminster Confession of Faith 7.1 reads, ‘The distance between God and the creature is so great, that although reasonable creatures do owe obedience unto Him as their Creator, yet they could never have any fruition of Him as their blessedness and reward, but by some voluntary condescension on God’s part, which He has been pleased to express by way of covenant’. 16 Leftow (2012), pp. 11–12. 17 See ibid., 10–12. The principle theological deficiency in this section in Leftow’s text is hermeneutic

consistency across the whole of the Bible. If we do our theology proper reading the NT through the OT, we may dismiss the Trinitarian complexity introduced in the gospel of John as more information than is necessary. This methodology might slip in undetected when we begin with “Western monotheism” generically conceived. But if we read the OT through the NT (as Christians tend to do with most other issues, and as is suggested by the notion of a completed canon), we must acknowledge that in the Hebrew Scriptures the scale tips toward mystery more than it does in the NT. So the question this raises is whether the less complete OT theology proper is merely less complete or whether it is, read on its own, potentially misleading. Obviously this question cannot be addressed here, nor, I suppose, would it be fair to expect Leftow to settle it before writing his book.

123

Int J Philos Relig

authority and necessity of revelation and the precariousness of finite natural conceptions of God.18 It would appear that attempts at clarifying the notion of necessary truth are caught between the need to know something of God and God’s incomprehensibility. In order to explain or understand why a necessary truth is true, one must have some knowledge of a truth-maker which exists in all possible worlds (and, attractively for the theist, one that has explanatory priority over necessary propositions).19 Why is it the case that 2 + 2 = 4? We understand that the contrary is impossible or inconceivable; the proposition represented by 2 + 2 = 4 cannot fail to be true. It is also widely agreed that this entails an ontology: the necessary existence of the same proposition, and since this proposition is essentially the proposition that it is, we add that this proposition has the property of being true necessarily. 2 + 2 = 4 is true in all possible worlds. As Leftow argues, to say that 2 + 2 = 4 is true simply by virtue of its analyticity or by virtue of its being the proposition that it is, is to deny that its truth accrues via some truth-maker or some relation to something. Necessity indeed entails truth; but necessity alone does not produce a very satisfying account of a proposition’s being true. It provides no account at all, in fact. Such a proof ‘certainly convinces us that its conclusion is true, but I doubt it would tell us why it is true’, Leftow says.20 Nor are accounts of truth mutually exclusive; there may be more than one for a single proposition.21 Truth-maker accounts and necessity accounts in fact have distinct explananda. One explains how or why a proposition is true; the other, which cannot do this, demonstrates that a proposition is true. So a proposition’s being necessarily true offers a relatively thin account for a proposition’s being true. Though it is possible to think of a proposition only in terms of this formal necessity or analyticity, that does not mean that a proposition must be true only in this sense.22 To say that it is would be to deny what I take to be the majority view that necessarily true propositions are truth-bearers. So for a proposition to be non-trivially true, it must be true by virtue of a relation to something other than itself—assuming, again, that necessary propositions have truthmakers. This suggests an asymmetrical relation with a necessarily existing thing, a thing which enjoys explanatory priority over necessarily true propositions. Anderson and Welty for this reason say that the laws of logic are thought by (some) G/god about that G/god’s own thoughts, and they suggest that such propositions ‘subsist’ (my term) 18 He even offers a word of consolation to ‘philosophers impatient with discussions of Biblical texts’ (3).

I should also note that it is possible for the authority and necessity of revelation to contend with the same fallibility concerns as those which pester speculation and empirical investigation, or what I call here ‘finite natural conceptions of God’. Not everyone who holds a high view of revelation feels himself immune to modern critiques of metaphysics. 19 Assuming a number of things: that propositions have an ontology; that a weak ontology along conceptualist lines is false or in adequate; and that necessary propositions have truth-makers. 20 Leftow (2012), p. 50. For Leftow’s argument that ‘necessity does not explain’, see ibid., 48–54. 21 ‘Explanations of distinct explananda are often not in competition, and in particular truthmaker expla-

nation and explanation by necessity or mathematical explanation (if there by such) do not compete’ (ibid., 51). 22 Some philosophers hold that such propositions are trivial. See Cameron (2010). I do not believe this,

but I think it is implied in some accounts of necessity.

123

Int J Philos Relig

eternally in the divine mind. This is a kind of theistic truth-maker theory for necessary truths. As I am suggesting, this move by Anderson and Welty is well-intended. But it seems that when we say ‘mind’ in this case, or when we stipulate this G/god’s contingent thoughts as the intentional objects of his necessary thoughts, we are yet talking about a necessary property of propositions, not about something else (such as a divine being). A ‘mind’ then is whatever a necessary truth requires; but this is no more than we knew from the outset about necessary truths. In this case, when we say ‘G/god’ we are re-naming an aspect of a necessarily true proposition, and, as they say, promoting it to incompetence. To prove the existence of God from the necessity of such propositions is to produce a kind of ontological argument with an appendix. Anderson and Welty’s argument draws attention to (1) the need for a distinction between that a proposition is true (because its negation is impossible or inconceivable) and why a proposition is true (which the former does not explain); and (2) the consequent need for a truth-maker, if ‘true’ is to carry much weight. My concern is that such a proof of the existence of God is really just a renaming of a non-explanation with the most laudatory terminology available. A substantive account—a why account—of a necessary proposition’s being true requires true knowledge of the essence of the thing which makes it true. We must be able to pick it out, if we are to have epistemic rights to it and its truth-maker benefits. And here we find ourselves between two time-honored difficulties: the need to know something and the impossibility of knowing exhaustively. I might take divine incomprehensibility to be the fact that God as he is to himself (ad intra) defies explanation. We may point specifically to the essence of God subsisting in three persons, or to the equal ultimacy of three and one, or to the irreducibility of relative personal distinctions and essential unity in the Godhead, or perhaps to the selfexistence (aseity) of each distinct triune person in the unity of God.23 Any of these will do for now. All that is necessary presently is the fact that God’s essence is in an ultimate sense impenetrable by the creature’s mind. So to affirm ‘one essence in three persons’ is to say something true of God (something revealed), something which identifies him essentially, but also to affirm something which wears its impenetrability on its sleeve. In terms then of a necessary but asymmetrical truth-maker, to say blankly ‘god’ is inadequate because it is empty, or enjoys only the content of the definition we have given it: whatever hospitality a necessary proposition requires. This definition, I am arguing, does not entitle us to the divine moniker. To say on the other hand ‘one essence in three persons’ is to make a confession with content, one which bears the requisite knowledge of the essence of the needed truth-maker but also avoids the pitfall of knowing nothing because claiming to know everything. This exploration into the epistemic grounds for divine conceptualism comes down to this: being conscientious houseguests. We cannot presume to make ourselves at home in the divine mind without reckoning with an identifiable notion of what it is actually like, of what we may properly say and what we must say about the nature of God and his relation to contingent things. God is a certain way, and he is distinguishable

23 The latter is Calvin’s view. See Ellis (2012).

123

Int J Philos Relig

from false gods. Nor is enjoying the luxuries of a partial description of God—as for example ontologically ultimate—adequate, since God is not, for example, ontological ultimacy; as the one God in three persons, he is ontologically ultimate. God is not relevantly similar to the required abstract properties. So we must ask: what is logic from God’s point of view? At this point, I take the fifth, as they say. References Anderson, J., & Welty, G. (2011). Lord of noncontradiction: an argument for god fromlogic. Philosophia Christi, 13(2), 321–338. Cameron, R. (2010). Necessity and triviality. Australasian Journal of Philosophy, 88(3), 401–415. Davis, R. (2011). God and the platonic horde: a defense of limited conceptualism. Philosophia Christi, 13(2), 289–303. Ellis, B. (2012). Calvin, classical trinitarianism, and the aseity of the son. Oxford: Oxford University Press. Leftow, B. (2012). God and necessity. Oxford: Oxford University Press. Oliphint, K. S. (2011). God with us: divine condescension and the attributes of god. Wheaton: Crossway. Plantinga, A. (1982). How to be an anti-realist. Proceedings and Addresses of the American Philosophical Association, 56, 47–70. Poythress, V. (1995). Reforming ontology and logic in light of the trinity: an application of Van Til’si dea of analogy. Westminster Theological Journal, 57(1), 187–219. Poythress, V. (2013). Logic: a god-centered approach to the foundations of western thought. Wheaton: Crossway. Yandell, K. (2011). God and propositions. Philosophia Christi, 13(2), 275–287.

123