This is a classic paper from the great Oxford philosopher A.N. Prior (1914-69). Here, he cleverly and neatly debunks the idea of a 'space-time' language favoured by mathematical logicians of the period, such as Quine, in which talk of individuals enduring through time is replaced by world-lines or life-histories of things, or 'four-dimensional worms'.

Prior argues that the only way we can formalise talk about enduring and changing individuals as we do in common speech, is to use tenses. For space-time talk cannot capture the fact that the same individual can endure through time. As Prior argues, the the individual who had written Marmion in 1807 (Scott) was certainly identical in 1814 with the individual who was then writing Waverley (Scott again), and after that with the individual who had written Waverley (Scott). We express this by a variation in tense. In tenseless space-time language, by contrast, Scott-in-1807 and Scott-in-1814 are there, 'separated for ever by seven years'. So when can they ever be the same person? Never. In any case, they aren't persons at all, but year-thick slices of a four-dimensional worm, and as they are distinct slices, there is no time whatever at which they are identical.

Prior is one of the great philosophical stylists. Some memorable quotes:

Just the bricks, in fact, for building half-way houses.

... half the time I personally have forgotten what the date is, and have to look it up or ask somebody when I need it for writing cheques, etc.; yet even in this perpetual dateless haze one somehow communicates, one makes oneself understood, and with time references too. One says, e.g. 'Thank goodness that's over!', and not only is this, when said, quite clear without any date appended, but it says something which it is impossible that any use of a tenseless copula with a date should convey. It certainly doesn't mean the same as, e.g. 'Thank goodness the date of the conclusion of that thing is Friday, June 15, 1954', even if it be said then.

Nothing can be surer than that whereof we cannot speak thereof we must be silent; but that does not mean that whereof we could not have spoken yesterday thereof we must be silent today.

The paper was originally published in Philosophy, vol.34 1959, pp. 12-17.

E. D. Buckner April 2006

In a pair of very important papers, namely 'Space, Time and Individuals', (STI) in the Journal of Philosophy for October 1955 and 'The Indestructibility and Immutability of Substances' (IIS) in Philosophical Studies for April 1956, Professor N.L. Wilson began something which badly needed beginning, namely the construction of a logically rigorous 'substance-language' in which we talk about enduring and changing individuals as we do in common speech, as opposed to the 'space-time' language favoured by very many mathematical logicians, perhaps most notably by Quine. This enterprise of Wilson's is one with which I could hardly sympathise more heartily than I do; and one wishes for this logically rigorous 'substance-language' not only when one is reading Quine but also when one is reading many other people. How fantastic it is, for instance, that Kotarbinski^[1] should call his metaphysics 'Reism' when the very last kind of entity it has room for is things - instead of them it has the world-lines or life-histories of things; 'four-dimensional worms', as Wilson says. Wilson, moreover, has at least one point of superiority to another rebel against space-time talk, P.F. Strawson; namely he (Wilson) is does seriously attempt to meet formalism with formalism - to show that logical rigour is not a monopoly of the other side. At another point, however, Strawson seems to me to see further than Wilson; he (Strawson) is aware that substance-talk cannot be carried on without tenses, whereas Wilson tries (vainly, as I hope to show) to do without them. Wilson, in short, has indeed brought us out of Egypt; but as yet has us still wandering about the Sinai Peninsula; the Promised Land is a little further on than he has taken us.

From this point on, then, I shall be quarreling with Wilson, but from what has just been said I hope it will be clear that this is a dispute between allies - I want Wilson (with any Wilsonians there may be) to go further in a direction in which he has already started to go, for I do not think the place where he has left us is or can be a real resting place. From such a place as that, we must either go forwards or go back^[2].

First let me sketch Wilson's position more fully, and mainly in his own words. Early (p.592) in the paper STI, he says that when we pass from a space-time language to a substance language, 'the time determinant is shifted across the copula of empirical sentences from subject to predicate'. Later (p.594) he explains what he means by this. 'In our S-T language', he says, 'we might record a simple matter of fact in a sentence like the following:' (here he gives the Russellian for 'The x such that x occupies u₁, u₁, u₁, t₁, is blue'). 'In substance-language we might say:

'a is blue at time t₁, where the copula "is" is used tenselessly'.

And again (p.597) he says that 'the simplest kind of empirical statement in substance-language is one of the form 'a has the quality Q₁ at time t₁'. 'It is so obvious, so necessary', he goes on, 'that if Philip is drunk, Philip is drunk at some time, that if Scott wrote Waverley, he wrote it during some period - it is so obvious and necessary, that in ordinary language we generally drop the "at some time" and are left with the simple, the too simple, noun-copula-adjective form of sentence'. ('Perfidious ordinary language!' he adds in parenthesis at this point). Much of this is repeated in IIS. At the end of this (p.48) Wilson says that although it may be true that a thing 'changes qualitatively and is numerically the same', e.g. 'If a leaf is green in August and red in September it is still that leaf', yet nevertheless 'a "complete" property of an individual is a compound, temporalized (or dated) property, like being green in August 1955 or being born in 1769, and there is no question of an individual changing in the sense of once having and later lacking one of the compound properties'.

This last conclusion should in itself have been enough to frighten him. From this leaf whose 'complete' properties never change, to the pure 'four dimensional worm' of Quine, Kotarbinksi, etc., is surely a very short step indeed. And this 'substance language' goes wrong at the end because it goes wrong at the start. Wilson's basic sentence form is 'S is P', and his idea appears to be that you get from a space-time language to a substance-language by exchanging 'S-at-t is P' for 'S is P-at-t'; in this way the 't' 'crosses the copula', which is thought of as a sort of bridge between S and P. But if the 'is' in these two forms is tenseless, as Wilson explicitly says that it is, I cannot see what the difference between them amounts to. 'S-at-t is P' presumably means that that part of the four-dimensional worm S which has the time co-ordinate t, is P; what else 'S is P-at-t' could mean, i.e. what it means if it does not mean that S is P in the stretch of it specified by t, I cannot imagine. I can, indeed, see something different in the form 'S was P at t', e.g. 'The leaf was green in August'; but here 'In August' is only intelligible as an answer to the question 'When was the leaf green, not as an answer to the question 'In what way was is (was) it green?', 'What sort of green is (was) it?'. The thing then means, not 'The leaf was green-in-August' but 'The leaf was-in-August green'. The t in fact has not 'crossed the copula' but stopped at the copula; though it seems to me that this bridge theory of the copula is wrong anyway. Write 'The leaf is green' in the modern way as 'φx'; here 'x' denotes the leaf, 'φ' means 'is-green', and there is no copula needed, but if we want one we can put 'S' for 'It is the case that -' before the whole, i.e. in the place where we would put 'N' or a tilde for 'It is not the case that -' if we wanted to construct a negative proposition. Then we can think of 'at t' as neither moving across the copula to the predicate nor staying with the copula on the way to the predicate, but as moving across the predicate to the copula, and changing 'It is the case that -' to 'It was the case in August that -'.

A parallel case will, I think, make the matter clear. A person, call him Owen, who is colour-blind, might see this leaf as green when it is in fact red. The leaf, we might then say, looks green to Owen; i.e. though it is not the case that, it does appear to Owen that, the leaf is green. And some philosophers would want to replace this by 'is green to Owen', abolishing the explicit reference to looking or seeming in the same way as Wilson abolishes the explicit tense. But as in the other case, the supposed abolition is only a disguise. 'To Owen' is intelligible as an answer to the question 'To whom does it look green?'; as a description of the kind of green the leaf is, or of the kind of green leaf it is, 'green-to-Owen' is just nothing at all. Unless, indeed, we mean by calling it 'green-to-Owen' that the leaf is green where it is turned towards Owen - green on the side of it. And we might give 'green-in-August' a similar sense - we might mean that that part of the leaf which is in August (like 'that part of it which faces Owen') is green; but then it cannot be really a leaf that we are talking about, for it is not leaves but their world-lines that have parts of this sort. We are, in short, back in Pharaoh's house, with S-at-t being P.

I do not, however, want to make a difference where there is none; and in particular, I do not wish to deny that there is such a property of having been green in August as well as the property of being green, i.e. being green now; nor even, for that matter, that there is such a property as looking green to Owen as well as that of being green, i.e. being green really. I only insist (a) that what is now in question with this leaf is not a property of being green in August which attaches to it tenselessly, but a property of having been green August which attaches to it now; and (b) that having been green in August is not a way of being green now (I am not writing in August); (c) that neither is it a way of being green timelessly - there is in fact no way of being green timelessly (as Wilson very truly says, Philip cannot be drunk without being drunk at some time; and neither can a leaf be green without being green at some time); and (d) that the internal punctuation of 'having been green in August' is 'having in-August been green', not 'having been green-in-August'. Putting it yet another way: A leaf that was green in August is one sort of formerly-green leaf (because 'in August' is one sort of 'formerly'); but a formerly-green leaf is not one sort of green leaf. Indeed in common parlance being formerly-green and being green are often inconsistent - the 'formerly green' is precisely that which is not green but was green, just as a soi-disant philosopher is precisely someone who is not a philosopher but says he is one. And, of course, a leaf which is merely 'green-to-Owen' is precisely one which is not green but seems to that person to be so^[3].

This is perhaps the most suitable point at which to consider a very strange argument put forward by Wilson in IIS, p.47 on the subject of identity: 'When we say that the individual who wrote Marmion in 1807 is identical with the individual who wrote Waverley in 1814, we are not saying that the individuals are identical in 1807 or in 1814. They are identical outside time, as it were. Dates cannot be significantly associated with the identity sign'. (Italics Wilson's). In this passage perhaps more than any other we see the incompleteness of Wilson's emancipation from space-time language. Certainly to say in 1955 or 6 that X and Y (not 'were' but) 'are' identical in 1807, is to say something that grates upon the ear and the mind intolerably; but that is not what Wilson means, for it is clear from his conclusion that he would object equally to the result of repairing his syntax by due attention to tenses. If we do this, we will say that the individual who had written Marmion in 1807 was not then identical with the individual who had written Waverley in 1814 because at that time (1807) nobody at all had written Waverley, or done either that or anything else in 1814. Was he identical, then, with the individual who was going to write Waverley in 1814? This, I admit, is a tricky one, but only because such indeterminism makes me wonder whether there was yet any such individual; the question is not actually improper. Leaving that for the other date: the individual who had written Marmion in 1807 was certainly identical in 1814 with the individual who was then writing Waverley, and after that with the individual who had written Waverley. But Wilson does not want to say any of these things: he wants to say that the author of Marmion and the author of Waverley are 'identical outside of time', whatever that may be.

I cannot help thinking that Wilson is worried here about the relation between Scott-in-1807 and Scott-in-1814 - there they are, separated for ever by seven years, and yet somehow the same person; but when can they be the same person? - clearly nowhen. But of course Scott-in-1807 and Scott-in-1814 aren't persons at all; they are year-thick slices of a four-dimensional worm (as Wilson says, 'S-at-t' is a description from the space-time language, not from substance-language); and as they are distinct slices, there is no time whatever at which they are identical. Had he really left these 4-D worms behind - Scott-in-1807, Scott-in-1814, and Scott-from-his-birth-to-his-death (equally a no-person, a by-product of mispunctuating sentences like 'Scott, from his birth to his death, lived in the Northern hemisphere') - and learnt again to talk simply about Scott, it is hard to see how this strange talk of identity-outside-time, in an enduring object, could have arisen.

Wilson also says on p.47 of IIT that existence is not datable, but is 'a simple something or other which Napoleon simply has and Pegasus (for example) simply lacks'. This will surely not do; but before saying anything more about it let me interpose a peccavi. I have suggested elsewhere that just as there were no facts about me before I existed (not this fact of there being no facts about me; though of course there is now the past-tense fact of there having been none then), so there will be no facts about me after my existence ends (if it does end). And my ground for saying this was the very weak one that if some facts about a thing imply that it still exists and some do not, nobody can state with any precision where to draw the line between these two classes of facts.^[4]

This situation is not in fact anything like as hard for the logician as I made out. For he can use special variables, f, g, etc., for predicates entailing existence (call them E-predicates) without committing himself as to what these predicates are; and he can lay it down that E-predicates are predicates, substitutable for the usual predicate-variables, φ, ψ, etc., and functions of E-predicates like Nf, ('- does not f') and Pf ('- has fd in the past') are likewise predicates, substitutable for φ, ψ, etc., but these last (Nf, Pf, etc.) are not themselves E-predicates, substitutable for φ, ψ, etc. Then we can say that there are facts about Napoleon still, e.g. the fact that no E-predicates apply to him now; and this not being itself an E-predicate, there is no contradiction in so speaking.^[5]

So there is indeed a sort of 'being' that Napoleon has even after having ceased to 'exist'; he is at least a subject of predicates still, and cannot now ever cease to be that. But there is nothing timeless about this. For one thing, even this 'being' of Napoleon, i.e. there being facts about him, is something that had a beginning (when he had a beginning).^[6]

Turning now to a fundamental; I'm a symbol-man rather than an ordinary-speech man myself, but I can see what the ordinary-speech men are worried about when I find Wilson crying 'Perfidy!' at locutions which in fact constitute a more coherent and smoothed-out substance-language than his own. His chief quarrel with ordinary speech is, as he says, that it omits dates; but is is misleading to treat this as pretending to do without a time-reference. I do not know how it is with Wilson, but half the time I personally have forgotten what the date is, and have to look it up or ask somebody when I need it for writing cheques, etc.; yet even in this perpetual dateless haze one somehow communicates, one makes oneself understood, and with time references too. One says, e.g. 'Thank goodness that's over!', and not only is this, when said, quite clear without any date appended, but it says something which it is impossible that any use of a tenseless copula with a date should convey. It certainly doesn't mean the same as, e.g. 'Thank goodness the date of the conclusion of that thing is Friday, June 15, 1954', even if it be said then. (Nor, for that matter, does it mean 'Thank goodness the conclusion of that thing is contemporaneous with this utterance'. Why should anyone thank goodness for that?)

Wilson seems to have the notion that a tensed copula is analysable into a tenseless one plus a date (which once obtained can be transferred to any other part of the proposition that we fancy); but the above example is sufficient to refute this assumption. The fact is that propositions with dates are just not 'the simplest empirical propositions', but are highly sophisticated propositions; well, medium sophisticated - an essential prelude, though only a prelude, to space-time talk. Just the bricks, in fact, for building half-way houses.

THE LOGIC MUSEUM Copyright (introduction) E.D.Buckner 2006.