^{page 129 note 1} A rigorous, fully general formulation of this Law would need to be more elaborate. But this will suffice for our purposes.

^{page 129 note 2} De Trinitate, book I, chapter 7. I have quoted the translation by McKenna, Stephen (Washington, D.C.: The Catholic University of America Press, 1963).Google Scholar

^{page 129 note 3} My exposition of the Trinitarian's dilemma differs in only important respect from Martinich's, A. P. in his ‘Identity and Trinity’, journal of Religion, LVIII (1978), 170–82:Google Scholar he formulates (I–3) using ‘God’

where I have used ‘divine being’ or ‘divine;’. ‘God’, however, is afflicted with an unfortunate ambiguity: it can function either as a singular term or as a predicate. Thus a substitution-instance of ‘x is God’ may be construed either as an identity statement or as a predication. It will shortly become clear that we must treat (2) and (3) as predications, not as identity statements (see pp. 3–4). Therefore it is better to formulate (2) and (3) with a term that can function only as a predicate. (Indeed, Martinich himself interprets ‘The Father is God’ and ‘The Son is God’ as predications instead of identity statements, for he symbolizes these statements throughout his paper as ‘Gf’ and ‘Gs’ respectively.) Martinich shows that if we accept the standard, non-relativistic interpretation of the identity operator, we can derive Modalism from (1–3) and Arianism from (I), (2) and (4) within first-order predicate logic (Ibid. pp. 172–4). Though he does not offer a similar derivation of Bitheism from (2–4), it is very easy to construct one.

^{page 130 note 1} I believe that G. E. L. Owen coined this pair of terms. See his ‘Aristotle and the Snares of Ontology’, in Bambrough, Renford (ed.), New Essays in Plato and Aristotle (London: Routledge and Kegan Paul, 1965), p. 72.Google Scholar

^{page 130 note 2} This claim is easy to confuse with David Wiggins' D, the Thesis of the Sortal-Dependency of Individuation: see his Sameness and Substance (Oxford: Blackwell, 1980), pp. 15, 48, 53 et al. Strictly speaking, D asserts only that no affirmative identity statement can be true unless it has a true sortal expansion. Thus D leaves open the possibility of affirmative identity statements that are false even though there is no answer to the question, ‘x is not the same what asy?’. Sydney Shoemaker offers two plausible examples of such identity statements: ‘I am identical with my body’ and ‘My typewriter is identical with the number Two’. See ‘Wiggins on Identity’, in Munitz, Milton K. (ed.), Identity and Individuation (New York University Press, 1971), p. 108.Google Scholar Shoemaker also points out that Wiggins himself does not always clearly distinguish D from the claim that identity predications are polygamous (Ibid. pp. 106–8).

^{page 131 note 1} Peter Geach is often credited with originating R, which he has defended in numerous works, such as Reference and Generality (Ithaca: Cornell University Press, 1962), pp. 39–40, 151, and 157,‘Identity’, Review of Metaphysics, XXI (1967), pp. 3–12, and ‘Ontological Relativity and Relative Identity’, in Munitz, Milton K. (ed.) Logic and Ontology (New York: New York University Press, 1973).Google Scholar However, Geach believes that Aquinas accepted R – see Three Philosophers (Oxford: Blackwell,1961), p. 118. And Martinich has cited some remarks of William of Ockham that strongly suggest this thesis – see ‘Identity and Trinity’, p. 175, note 5, and ‘God, Emperor, and Relative Identity’, Franciscan Studies XXXIX (1979), 181, note 3. For references to the work of other modern philosophers who have accepted (R), see Wiggins, ibid., p. 16, note 1. (R) comes in stronger flavours than the version in the text, but we may safely ignore them.

^{page 131 note 2} Several present-day philosophers have exploited recent research on (R) in order to formulate and to defend (RT). Geach is very sympathetic to (RT) (see Three Philosophers, pp. 118), but the most prominent and persistent advocate of this view is Martinich, who has developed it in the articles I have cited above. Peter van Inwagen will shortly publish an imposing logical formalization of (RT): ‘And Yet They Are Not Three Gods But One God’, forthcoming in Morris, Thomas V. (ed.), Philosophy and the Christian Faith (University of Notre Dame Press).Google Scholar However, van Inwagen's aim is far more modest than the Relative Trinitarian's. Van Inwagen only wants to propose a way of stating the doctrine of the Trinity that can be shown in a system of relative-identity logic to be free from formal inconsistency. Moreover, he does not claim that his formalization of RT is faithful to Christian orthodoxy – he explicitly leaves that as a matter for further discussion. Van Inwagen's aim is rather modest because a proof within a system that a proposition is free from formal inconsistency does not guarantee that the proposition is possibly true, or even that the proposition is not obviously impossible (and, of course, van Inwagen is fully aware of this point). For example, no formal contradiction can be deduced in first-order quantifier logic from ‘There is a highest integer’; all the same, this statement is necessarily false, and patently so. At the very least, the Relative Trinitarian claims that the logical and metaphysical principles which he uses to evade the Trinitarian's dilemma are reasonable – and that his statement of Trinitarian doctrine is orthodox.

^{page 132 note 1} To be sure, Christian tradition does not equivocate on our ordinary concept of a person when it contends that each member of the Trinity is a person. But this point does not instantly refute a Relative Trinitarian who claims that the members of the Trinity are distinct persons only in an equivocal sense – for he will simply reply that, in any non-equivocal sense of the term, Father, Son, and Holy Spirit are one and the same person. Of course, this reply does not dispose of every objection to the orthodoxy of his view. For example, it is difficult to square his conception of the Trinity with the many Scriptural passages in which Jesus uses a personal pronoun such as ‘you’ or ‘he’ to refer to God the Father – as in the High Priestly Prayer of the seventeenth chapter of John.

^{page 132 note 2} For discussions of this problem, see Young, Frances, ‘Can There By Any Evidence’, in Goulder, Michael (ed.), Incarnation and Myth: The Debate Continued (Grand Rapids: Eerdmans, 1979), p. 62Google Scholar, and Morris, Thomas V., The Logic of God Incarnate (Ithaca: Cornell University Press, 1986), pp. 187–204.Google Scholar

^{page 133 note 1} I am not the first philosopher to attack (RT). Wiggins' versatile book, Identity and Spatio–Temporal Continuity (Oxford: Blackwell, 1967), contains a lengthy critique (pp. 19–25), reprinted virtually unchanged in his Sameness and Substance, pp. 37–42. And at least two other writers have explicitly rejected RT, though neither adds anything substantive to Wiggins' discussion: Brown, David, in The Divine Trinity (London: Gerald Duckworth, 1985), pp. 289–90Google Scholar, and Morris, Thomas, in The Logic of God Incarnate, pp. 26–9.Google Scholar I believe that this paper improves upon the previous criticisms of RT in at least two major respects. In the first place, it acknowledges that Relative Trinitarians are not compelled to maintain that there are any examples of (R) other than the members of the Trinity: (RT) can hold, without resorting to ad hoc manoeuvres, that except for Father, Son, and Holy Spirit, no objects are relatively identical (see pp. 14–15). Thus this paper does not overestimate the damage done to (RT) by the failure of philosophers to think of plausible non-theological examples of (R). Second, and more important, my paper tries to undermine the theological motives for (RT), especially the motive of preserving Christian tradition, by arguing that (RT) cannot satisfy them.

^{page 134 note 1} (LL) must not be confused with a metalinguistic identity principle commonly known as the Substitutivity of Identicals; and it is especially important not to confuse (LL) with such crude formulations of the Substitutivity of Identicals as this:

(SI) Singular terms denoting the same object can replace each other in any context without changing the truth-value, which, as Alvin Plantinga succinctly says, is ‘a “principle” that does not hold for such excellent examples of language as English’ (The Nature of Necessity(Oxford: Clarendon Press, 1974), p. 15). For example,

(A) Giorgione = Barbarelli

and

(B) Giorgione was so called because of his size are both true, but

(C) Barbarelli was so called because of his size is false. But although there are counterexamples aplenty to (SI), it does not follow that there are any counterexamples to (LL). (A–C) is a case in point. In order for these sentences to falsify (LL), each occurrence of the predicate expression ‘was so-called because of his size’ must denote the same property, and it must be a property that Giorgione has and Barbarelli lacks. But unless we specify what Giorgione and Barbarelli were called, this predicate is insufficiently determinate to express a property. And once the predicate becomes sufficiently determinate to express the same property in both (B) and (C), it is clear that no interpretation of (A–C) will make it a counterexample to (LL). If the predicate expresses the property being called ‘Giorgione’ because of his size, then both (B) and (C) are true; if it expresses being called ‘Barbarelli’ because of his size, then both (B) and (C) are false. (I owe most of the points in this paragraph to Richard Cartwright's salutary discussion of the relation between (LL) and the Substitutivity of Identicals: ‘Identity and Substitutivity’, in Munitz, (ed.), Identity and Individuation, pp. 119–34).Google Scholar I have two reasons for devoting so much space to distinguishing (LL) from (SI). First, they are often conflated – apparently even Leibniz himself did not clearly distinguish them. Second, Martinich has tried to defend (RT) against the charge that it violates (LL) by claiming that (LL) succumbs to clear exceptions and that a number of eminent philosophers have rejected it; but in fact his putative counterexamples to (LL) are effective only against unsophisticated versions of the Substitutivity of Identicals such as (SI), and he offers no evidence that any of the eminent philosophers he cites rejected (LL) rather than (SI). See ‘Identity and Trinity’, pp. 179–80.

^{page 135 note 1} The Relative Trinitarian might object that (RT) is certainly not incompatible with a different same-f counterpart of (LL), namely:

(LL′) For any x and any y, if for any sorta] f, x is the same f as y, then for any property P, x has P if and only if y has P.

And with (LL′) in hand, the Relative Trinitarian can avoid at least two of the logical difficulties I mention in this section: difficulties II and III. As it turns out, I concede these points a bit later. But I also argue that RT fails anyway.

^{page 135 note 2} This argument is a generalization of Martinich's proof of the incompatibility of (RT) and (LL) (Ibid., p. 178). Both of these arguments, in turn, are very similar to Wiggins' well-known derivation of the incompatibility of (R) and (LL) (for one version, see Sameness and Substance, pp. 18–20).

^{page 135 note 3} I have taken the first three difficulties from Wiggins, Ibid. pp. 21–2, and the fourth from Inwagen's, van ‘Ontological Arguments,’ Nous, 1 (1977), pp. 389–90.Google Scholar

^{page 136 note 1} Ibid. p. 21.

^{page 136 note 2} For a summary of the usual kinds of putative counterexamples to (LL), and a debunking of them all, see Morris, Thomas V., Understanding Identity Statements (Aberdeen: Aberdeen University Press, 1984), pp. 89–93.Google Scholar For a very thorough criticism of a great many alleged examples of (R), see Wiggins, ibid., pp. 30–6. Geach develops some general arguments for (R) 1n ‘Identity’; Perry, John rebuts them in ‘The Same F’, Philosophical Review, LXXX (1970), 181–200.CrossRefGoogle Scholar

^{page 137 note 1} Ibid.

^{page 137 note 2} Be careful to notice two points about an incarnational property. First, it is not a property that God the Son would have lacked had He not become incarnate as Jesus of Nazareth – it is a property He would have lacked had He not had any sort of incarnation, human or otherwise. We want to say that such properties as having a body are among the incarnational properties of God the Son – but for all we know, God the Son would have exemplified such properties even if He had never become incarnate as Jesus (He might, for example, have had incarnations on other worlds.) Second, the possession of an incarnational property does not logically depend upon being incarnate: I leave open the possibility that (for instance) God the Son assumed an incorporeal, finite nature and had at least some of the same limitations which He possessed during His earthly life. (Suppose, arguendo, that angels are incorporeal. Then God the Son might have assumed angelic nature, and might thereby have become limited in knowledge.) I claim only that the possession of an incarnational property counterfactually depends upon being incarnate. I suspect that my concept of an incarnational property does not pick out exactly the range of properties that Christian tradition ascribes to God the Son but not to God the Father. One of these memories is certainly nescience: in some sense, it is true that God the Son has had this property; but God the Father is eternally omniscient (at least so far as we know). But perhaps even if God the Son had not become incarnate, He would have assumed angelic nature and become nescient; or perhaps He might have become nescient without assuming the nature of another kind of being at all, as Robert Nozick suggests in the following passage. ‘Could not a good God be motivated to deceive itself temporarily, even if not another? … whether from playfulness or whatever motive, such a good God would temporarily deceive itself, perhaps even into thinking it is a human being living in a material realm.’ (Philosophical Explanations (Cambridge, Mass.: Harvard University Press, 1981), p. 202, note.*) That is what I mean when I say that a general characterization of an incarnational property proves to be surprisingly elusive. But no matter: none of the substantive results of this paper depend upon a perfectly accurate account of this concept.

^{page 138 note 1} Translated by Kelly, J. N. D. in Early Christian Doctrines (London: Adam and Charles Black, 1958), p. 329.Google Scholar See also Augustine, ibid. book I, chapter 14.

^{page 138 note 2} As Wiggins points out in Sameness and Substance, p. 40.

^{page 139 note 1} Note that (RT) must be careful to insist that irreducibly reduplicative predicates either do not express properties at all, or express properties that their subjects exemplify only qua some particular sortal. Thus (RT) ought to say, for example, that being born of the Virgin Mary qua human is not a property that Christ has simpliciter; if it is a property at all, Christ has it only qua human.

^{page 140 note 1} For example, Charles Marks believes that the concept of unity of consciousness might play no explanatory role in an adequate scientific psychology – not only would such a psychology find it unnecessary to invoke disunity of consciousness in order to explain the split-brain patient's lack of behavioural integration, it would also find it unnecessary to invoke unity of consciousness in order to explain the integrated behaviour of the normal human being. See his Commissurotomy, Consciousness, and Unity of Mind (Cambridge: The MIT Press, 1981), p. 40.

^{page 140 note 2} After Derek Parfit, whose celebrated thought-experiment, ‘My Physics Exam’, has closely inspired my own imaginary case. ‘My Physics Exam’ first appeared in ‘Personal Identity’, The Philosophical Review, Lxxx (1971), pp. 3–27; it can also be found in Reasons and Persons(Oxford University Press, 1984), pp. 246–7.

^{page 140 note 3} For arguments against multiple-persons interpretations of this case, see Parfit, , Reasons and Persons, p. 248Google Scholar, and Marks, Ibid. p. 32. I owe my account of the ‘key ring’ experiment to Marks', book –see Pp. 4–5.Google Scholar Thomas Nagel can be interpreted as claiming that according to our common, pre–analytic conception of personal identity, this case does not involve a determinate number of persons. See his Brain Bisection and the Unity of Consciousness’, Synthese, XII (1971), pp. 396–413.

^{page 141 note 1} The asterisk warns the reader that a mind* is not a mind. Minds*, bunot minds, can be subjects of states that are not psychological, or not purely psychological. A mind* can be the subject of an action, an involuntary bodily movement, and even an ‘M–property’ – that is, a property expressed by a predicate that can apply to an object that is never conscious. Unless we want to swallow an extremely Cartesian conception of persons, we will want to acknowledge the truth of such sentences as ‘Jesus Christ was located in Galilee qua human’ – and therefore we will want to allow Christ's human mind* to be a subject of M–properties. Of course, English has a customary term for the bearer of such a wide range of properties: ‘person’. But that term will not suit my purposes, and I know of no other such term in idiomatic English. ‘Mind*’ is as good a neologism as any.

^{page 141 note 2} Indeed, at least one philosopher and one theologian have claimed that Jesus Christ has two minds*. The philosopher is Morris – see The Logic of God Incarnate, pp. 102–7 – and the theologian is Brian Hebblethwaite – see The Propriety of the Doctrine of the Incarnation as a Way of Interpreting Christ’, Scottish journal of Theology, XXXIII (1980), pp. 201–2. Morris himself defends the standard Christology by adducing brain commissurotomy as an analogy – though he does not discuss any particular cases. However, I do not wish to attribute any of the other claims in this paragraph either to Morris or Hebblethwaite. In fact, I am certain that Morris rejects at least one of them – see the article cited in the next footnote.

^{page 141 note 3} Which shows that there is a significant dissimilarity between Derek and the Christ of (RT's) Christology. Both of Derek's minds* are embodied. But the Relative Trinitarian is obliged to hold that God the Son's divine mind* is not. If it were, then RT could only preserve its conviction that Father and Son are the same deity by claiming, heretically, that God the Father is embodied in the body of Jesus during the Incarnation. Thus (RT) is committed to the view that Christ not only has two minds*, but that one is embodied while the other is not. And I wonder whether this view avoids Nestorianism. It might be argued, for example, that we regard Simster and Dexter as the same person only because they both have the same body. Incidentally, the same problem plagues any Christology which maintains that qua divine, God the Son is absolutely immutable. For if anything is a change, so is becoming embodied. I regrettably suppress further discussion of this problem – not because I believe that it poses no serious difficulties for (RT) or the doctrine of divine immutability, but because I lack the space to do it justice. For a relevant – though very brief – discussion, see Morris, Thomas, ‘St Thomas on the Identity and Unity of the Person of Christ’, Scottish journal of Theology, XXXV (1982), p. 429.Google Scholar I hope that someone will devote a thorough study to the problem.

^{page 142 note 1} Or at least there seem to be two if I have not misunderstood Wiggins' somewhat opaque discussion of (RT) at pp. 40–2 of Sameness and Substance. The other philosopher, Thomas Morris, is more explicit: ‘/To deny (LL)/ just to defend one construal of the doctrine of the Trinity seems to be nothing more than, in the words of Geach himself, something like bending logic ad hoc to meet the needs of theology’ (The Logic of God Incarnate, p. 29).

^{page 143 note 1} Note that since x = y if and only if for any sortal f, x is the same f as y, the first conjunct of the bound formula in (LL*) is logically equivalent to (LL′) (see note 1 on p. 7). Hence I have conceded the points in that footnote to the Relative Trinitarian.

^{page 143 note 2} Among the many Scholastics who use the term is Aquinas: see, for example, Summa Theologiae (I a), (32), (3). A notio is not only a non-incarnational property, but an ad intra characteristic: that is, a property that a Trinitarian Person exemplifies even in possible worlds with no created beings. Not all nonincarnational properties of a Trinitarian Person are ad infra: an obvious example is being creator of the heavens and the earth.

^{page 144 note 1} Especially in the works of the Cappadocians. See, for instance, Basil of Caesarea, Adversus Eunomium, 1, 20 (PG 29, 557b)Google Scholar and II, 22 (PB 29, 621b), Gregory of Nyssa, Contra Eunomium, I, 26 (PG 45, 369a)Google Scholar and 1, 42 (PG 45, 464b–c). See also the eighth-century theologian John of Damascus, De Fide Orthodoxa, 1, 8 (PG 94, 824b).Google Scholar

^{page 144 note 2} For a familiar illustration of why (LL) needs this requirement to rule out obvious counter-examples, see note 1 on p. 6.

^{page 144 note 3} Indeed, Christian theologians occasionally give the impression that we cannot know anything more about the notiones, at least in this life – or even that it is impious to seek further knowledge. –Define the unbegottenness of the Father and I will describe the nature of the begetting of the Son and the procession of the Spirit, and let us both go mad, prying into God's mysteries’ (Gregory of Nazianzus, Orationes, XXXI, 7, PG 36, 140–1).Google Scholar

It is also not uncommon to find theologians who contend that the distinction between the generation of the Son and the procession of the Spirit is a mystery even to God's holy angels. See, for example, Augustine, , Contra Maximinum, II, 14 (PL 42, 770)Google Scholar, and Didymus the Blind (died c. 398), De Trinitate, II, 1 (PG 39, 448c–d).

^{page 145 note 1} Except for one of the Father's notiones – being unbegotten, which is freedom from asymmetrical dependence upon anything. But this is also a genuine property.

^{page 145 note 2} For example, Maurice Wiles claims that although the Cappadocians sometimes suggest that the Father is the source of the divinity within the Godhead, ‘this is in flagrant contradiction to the main drift of their position. More probably they should be understood to mean that the Father is the source of the hypostatic existence of the other two, that is, their existence as distinguishable hypostaseis or persons’ (The Making of Christian Doctrine Cambridge University Press, 1967, p. 136). Wiles does not cite any supporting texts in his book, but in private correspondence he has called my attention to several passages in Gregory of Nyssa: Contra Eunomium, 1, 35 (PG 45, 405 b–408 a) and 1, 39 (PG 45, 449a – 451 c), and the last paragraph of Quod Non Sint Tres Dii (PG 45, 133b–136a).

In The Philosophy of the Church Fathers (Cambridge, Mass.: Harvard University Press, 1956), Wolfson, Harry seems to attribute the same view to Augustine (pp. 350–7)Google Scholar – though I find Wolfson's interpretation of Augustine less perspicuous that Wiles' interpretation of the Cappadocians.

I believe that neither Wiles nor Wolfson makes a convincing case for his interpretation. But since a detailed defense of my scepticism would be very lengthy, and irrelevant to the main concerns of this paper, I shall presume that Christian tradition permits us to predicate notiones to the Persons only under certain sortais.

^{page 146 note 1} Note the parenthetical adjective. Nothing in this argument depends on the exceedingly dubious claim that ‘exists’ designates a property which an object might have or lack.

^{page 147 note 1} For example, this claim is ruled out by several articles of the Creed of Constantinople, which, as Kelly, J. N. D. rightly says, ‘is one of the few threads by which the tattered fragments of the divided robe of Christendom are held together’. (Early Christian Creeds, Second Edition (London: Longmans, Green (1960), p. 296.)Google Scholar

^{page 148 note 1} See, for example, Brown, , The Divine Trinity, pp. 101–58, 283.Google Scholar The doctrine of the notiones also faces a weighty philosophical objection. For it can be argued that this doctrine robs God the Son and God the Holy Spirit of Their divinity and Their co-equality with God the Father. Consider the following claim:

(13) Necessarily, whatever is asymmetrically dependent for its existence upon some other being is not divine.

(13) has considerable intuitive appeal. And it also cannot be reconciled with even a modally eviscerated doctrine of the notiones. Even if, for example, the Son were only contingently begotten of the Father, (13) would still be incompatible with the Son's divinity. Of course, modal intuitions are defeasible – they can be overthrown by reasons to the contrary. But (13) seems rather straightforward, especially when we compare it with the evidence typically adduced in favour of the doctrine of the notiones. Perhaps (13) is mistaken; but we need to be shown what is wrong with it. I believe that no one has yet presented a convincing reason for denying it.

It is interesting to note that both David Brown and Leonard Hodgson, who are certainly no foes of Christian orthodoxy, reject the doctrine of the notiones because it conflicts with 13 (or perhaps because it conflicts with (13) minus the modal operator – I am not sure). See Brown, Ibid. p. 283, and Hodgson, , The Doctrine of the Trinity (London: Nisbet, 1943), pp. 100–3.Google Scholar

^{page 149 note 1} Ibid. pp. 159–215. Moreover, the Relative Trinitarian can maintain that the Christian Church ought to continue to confess that Jesus Christ is ‘the only–begotten Son of the Father’, and ought to continue to use this and similar phrases in prayer and worship. For like Marcellus of Ancyra, RT can construe the Son of God to be begotten of the Father only qua human – it can claim that Christ's human nature, though not his divine nature, is created and sustained by God the Father. Since the notio of the Son thus becomes another of His incarnational properties, the manoeuvre we perfected in Section III will void any attempt to prove that since the Son is begotten and the Father is not, they are distinct qua divine or simpliciter. It is not so easy to reinterpret the doctrine of the procession of the Holy Spirit. But in the passage cited at the beginning of this footnote, Brown claims that the Spirit, but not the Father or the Son, ‘indwells’ human beings in a particular way (see especially pp. 200–4 – the details need not concern us here). Since this ‘indwelling’ occurred at Pentecost, and since the Church traditionally considers Pentecost to be the fulfillment of Jesus' promise that the Father would send the Holy Spirit (John 14: 26), the Relative Trinitarian may even be able to construe the procession of the Holy Spirit as a purely ‘economic’ property of that person. Of course, (RT) still has to solve a truly monumental problem: it seems evident that the Spirit ‘indwells’ human beings qua divine, and therefore that the Spirit is a different deity from the Father and the Son.

^{page 149 note 2} So far as I know, Bertrand Russell originated this handy term. See his ‘The Philosophy of Logical Atomism’, in Marsh, Robert Charles (ed.), Logic and Knowledge (London: George Allen and Unwin, 1956), p. 277.Google Scholar

^{page 150 note 1} For example, (RT) must fall foul of many of the same New Testament passages that embarrass the Modalist. The claim that God the Father became incarnate in Jesus of Nazareth is very difficult to reconcile with such texts as t Corinthians 15. 24–8, Philippians 2. 6–11, and the numerous passages in the Gospel of John which speak of the Son as having been sent by the Father (e.g. 5. 19–23, 8. 17–18, and 14. 24–5).

^{page 151 note 1} Be careful not to confuse (ST) with the standard formulation of the Social Theory, according to which all and only those who accept the following claim are Social Theorists:

(14) The Trinity consists of three distinct, fully divine persons – in a sense of ‘person’ that is not equivocal with respect to ordinary applications of this term.

Numerous Social Theorists have adopted this formulation: see, for instance, Brown, ibid. 244. (who calls this version of the Social Theory ‘The Plurality Model’), and Plantinga, Cornelius J. Jr, ‘Gregory of Nyssa and the Social Analogy of the Trinity’, The Thomist, L (1986), p. 325 note I.CrossRefGoogle Scholar And so have numerous detractors of the theory, such as Baillie, Donald M. (God Was In Christ (London: Faber and Faber, 1948), pp. 133–40)Google Scholar and Welch, Claude (The Trinity in Contemporary Theology (London: SCM Press, 1953), pp. 268–74).Google Scholar But (14) is too weak to prevent a Social Theorist from adhering to Relative Trinitarianism: (14) is formally consistent with the contention that there is some sortal other than ‘person’ under which Father, Son, and Holy Spirit are numerically the same. Indeed, van Inwagen's statement of (RT) includes (14): as he phrases it, Father, Son, and Holy Spirit are distinct persons but one and the same being (‘And Yet They Are Not Three Gods But One God’, pp. 12–13, 32 ff.). Since anyone who would defend the Social Theory must express it in a way that will survive the demise of (RT), and (14) offers no such guarantee, we ought to forsake this formulation and replace it from now on with (ST).

At least one other writer has already reformulated the Social Theory as (ST) – Morris, Thomas, who in The Logic of God Incarnate variously describes the theory as belief in three distinct divine ‘individuals’ (pp. 210, 213)Google Scholar, ‘objects’ (p. 216), ‘beings’ (p. 217), and ‘ultimate subjects’ (p. 218). Unlike ‘persons,’ none of these terms are true sortais – for instance, to say that Father and Son are distinct individuals is to say that they are distinct under every genuine sortal, i.e. absolutely distinct.

I wish we could call (ST's) conception of the Trinity ‘Tritheism,’ for this name succinctly indicates that (RT) cannot be reconciled with Relative Trinitarianism. Unfortunately, as everyone knows, the Church has always used ‘Tritheism’ as a term of opprobrium, and if I applied it to my theory I would doubtless be suspected of flaunting a position that I knew to be heretical. I leave it to the reader to decide whether I am flaunting (ST); but as the new few pages show, I consider (ST) orthodox.

^{page 152 note 1} See, for instance, Zahn, Theodor, Marcellus von Ancyra (Gotha: Friedrich Andreas Barthes, 1867)Google Scholar, Harnack, Carl Gustav Adolf von, Lehrbuch der Dogmengeschichte, Fourth Edition (Tuebingen: J. C. B. Mohr, 1886–9, 3 volumes), n, p. 232 note 4Google Scholar, and Kelly, J. N. D., Early Christian Creeds, pp. 243 and 254.Google Scholar

^{page 152 note 2} For a much fuller description of the evidence, see Prestige, G. L., God in Patristic Thought (London: Heinemann, 1936), chapter IoGoogle Scholar, Kelly, , Early Christian Doctrines, pp. 234–7Google Scholar, and especially Christopher Stead, Divine Substance (Oxford: 1Clarendon Press, 1977), pp. 190–266.

^{page 153 note 1} See Barth, , The Doctrine of the Word of God, Thomson, G. T. (trans.) (Edinburgh: T. & T. Clark, 1936), pp. 401 ff.;Google ScholarRahner, , The Trinity (New York: Herder and Herder, 1970), pp. 101, 104, 113 et al.;Google ScholarCalvin, , Congrégation de la Divinité de ChristGoogle Scholar, C. R. 47, 473, quoted in Barth, Ibid. p. 410. Other One Person Theorists include Baillie (ibid.). Welch (ibid.), Ott, Heinrich (God (Atlanta: John Knox Press, 1974), p. 60)Google Scholar, and Hill, William J. (The Three-Personed God (Washington, D.C.: The Catholic University of America Press, 1982), pp. 217–24).Google Scholar

^{page 154 note 1} True, the Quicumque insists ‘that we worship one God in Trinity and Trinity in unity, without either confusing the persons or dividing the substance’ (verse 3; here and throughout this note I use the translation of Kelly, J. N. D. in The Athanasian Creed (London: Adam and Charles Black, 1964), pp. 17–20)Google Scholar, that ‘we are forbidden by the Catholic religion to speak of three Gods or Lords’ (v. 20), and that there are not three ‘eternals’, ‘increates’, ‘infinites’ or ‘almightics’ (verses 11, 12 and 14). But catholic contemporaries of the Quicumqueoften use the same terminology when the only opponent that they clearly have in mind is a Germanic form of Arianism, which held that although the Son is inferior to the Father and the Spirit inferior to the Son, all three are worthy of worship and therefore ‘Gods’. See the texts cited in Kelly, ibid. 79, notes 1 and 2; and see also page 81. So perhaps the authors of the Quicumque did not intend to reject belief in three fully divine individual substances. Certainly the Social Theorist will avow that ‘in this Trinity there is nothing before or after, nothing greater or less, but all three persons are co-eternal with each other and coequal’ (vv. 25–6), and ‘their glory is equal, their majesty coeternal’ (v. 6).

So far as I know, the earliest conciliar statement of the Church that clearly rules out (ST) was made by the fourth Lateran Council in 1215 when it condemned, joachim of Fiore (for an English translation of this condemnation, see Deferrari, Roy J., (trans.) The Sources of Catholic Doctrine (St Louis: Herder, 1957), numbers 431 and 432, pp. 170–2).Google Scholar Now a very conservative Roman Catholic criterion of orthodoxy might require the faithful to accept the teaching of the Fourth Lateran Council; but clearly not every reasonable standard of orthodoxy will.

^{page 154 note 2} See Cornelius Plantinga, Ibid. pp. 325–52.

^{page 155 note 1} William Wainwright has assembled the most impressive collection of arguments against the coherence of the Social Theory: see his ‘Monotheism’, in Wainwright, and Audi, Robert, (eds), Rationality, Religious Belief, and Moral Commitment (Ithaca: Cornell University Press, 1986), pp. 289–314.Google Scholar Wainwright does claim that his proofs of the conceptual impossibility of multiple deities do not preclude the existence of the Trinity (pp. 309–11). But as he points out, one of his arguments does preclude the existence of a Trinity in which each Person taken singly is the causally necessary and sufficient condition of the existence of contingent beings (p. 310) – and, though I cannot explain my reasons here, I think that the Social Theory is wildly implausible unless it advocates this conception of the Trinity. (I explain myself at considerable length in ‘And Yet There Is Not One God, But Three Gods’, in preparation.)

^{page 155 note 2} At the moment, the lengthiest defence of the coherence of the Social Theory is Thomas Morris's: see The Logic of God Incarnate, pp. 213–18. I believe that Morris's arguments block any simple proof of the incoherence of the Social Theory, but, as he himself admits (p. 213), his treatment of the issues is rather programmatic. In ‘And Yet There Is Not One God’ I defend the coherence of the Social Theory in much greater detail, and try to embed it in a satisfying general account of the principal divine attributes.

^{page 155 note 3} I owe a considerable debt of gratitude to Swinburne, Richard, Swift, Jennifer, Hughes, C. T., and the Religious StudiesGoogle Scholar referee for reading earlier drafts and making many helpful suggestions.