What is what

A.J. Ayer

  • Sameness and Substance by David Wiggins
    Blackwell, 238 pp, £12.50, May 1980, ISBN 0 631 19090 2

Professor Wiggins’s new book was originally intended to be a revision of his book Identity and Spatio-Temporal Continuity, which appeared in 1967 and had been allowed to go out of print. Like the earlier book, it is concerned with questions of identity, and especially with the identity of things which persist through change, and it advances the same theory of individuation. So much, however, has been added, and so earnest an attempt has been made to clarify what the earlier book had left obscure, that this ranks as an independent work. Though it is less arcane than its predecessor, the density of its argument and the author’s predilection for symbols still make it difficult reading, but the philosophical interest of its subject, and the thorough honesty of its treatment, more than make up for this deficiency. There is no denying the importance of having a proper theory of identity, and even if not all his arguments carry equal conviction, the attempt which Wiggins has made to supply this need deserves to be treated with very great respect.

The identity with which Wiggins is concerned is primarily that of concrete individuals and is therefore encompassed in his theory of individuation. The demand which he makes of a theory of individuation is that it ‘comprise at least three things: first, an elucidation of the primitive concept of identity as sameness; second, some however abstract account of what it is for something to be a substance that persists through change; and third, (supervening in [his] treatment upon the first two things) the beginnings of some lifelike description, however schematic, of what it is for a thinker at one time and then another to single out the same substance as the same substance’. Wiggins’s way of meeting these demands is to engage in what Strawson once called descriptive metaphysics. He sees himself as bringing to light the principles that govern our use of a crucial set of terms, thereby at once illuminating the structure of our thought and the character of a central part of its subject matter.

As in his earlier book, Wiggins takes his stand upon ‘Leibniz’s Law’, which states that if a is the same as b, what is true of a is true of b and vice versa. This principle, which is sometimes known as the Indiscernibility of Identicals, is professed by Wiggins, not as a definition of identity, which he takes to be a primitive notion, but as part of the elucidation of its nature. He also maintains that if a and b are identical, there must be some sortal concept f which assigns both a and b to some natural or artificial kind and accounts for the manner of their coincidence. Thus, on this view, a cannot simply be the same as b: it has to be the same something as b and the character of this something, whether it be a mountain or a motor-car or a donkey or a person, determines how the identity of a or b is constituted in the case in question.

It is to be remarked that this thesis of Sortal Dependency, as Wiggins terms it, does not carry a logical commitment to Leibniz’s Law. Many philosophers who agree with Wiggins that if a and b are the same there must be the same something that they are, also hold that identity is relative. They think that a and b may coincide under the concept f, yet fail to coincide under some other concept g. For instance, I am the same person as won such and such a prize at school over fifty years ago, but not the same boy, for I am no longer a boy. Or again, the river in which I swam yesterday may be the same as that in which I swim today, but the water will not be the same: the current will have taken it elsewhere. Again, the Lord Mayor may be the same man as the Chairman of such and such a company but he is not the same official. It would clearly be easy to devise many similar examples. Wiggins has various ways of dealing with them. In the case of the man and the boy, we need to take care about tenses. I am not now the boy that won that prize but I was once that boy, just as that boy was going to be this man. If one abolishes tenses, then I am timelessly both the person who won the prize in his boyhood and the person who writes this review in his old age. The implication is that what Wiggins calls a ‘phased sortal’ like ‘boy’ can always be subsumed under a sortal covering the whole duration of the object in question – in this instance, ‘person’ or ‘human being’. The case of the two officials is said to exploit an ambiguity from which we escape by distinguishing concepts from what falls under them. There is only one official but he fulfils two offices. As for the river in which I bathe twice without bathing in the same water, it is argued that we have to distinguish between the ‘is’ of identity and the ‘is’ of constitution. A river is water only in the sense that water composes it.

The one alleged counter-example which Wiggins does not find it easy to accommodate is that of the Trinity, the Son being supposed to be the same God as the Father without their being credited with all the same attributes. For example, to hold that the Father was crucified is to succumb to the heresy of Patripassionism. Wiggins is too polite to say that this orthodox conception of the Trinity is nonsensical, but he leaves the reader with the impression that no sense can be made of it. This is not a conclusion with which I should wish to quarrel. If I have a complaint against Wiggins’s treatment of Leibniz’s Law, it is that he passes too quickly over the fact that it breaks down in modal and indirect discourse. Thus, to use well-known examples, one may wish to hold that the number 9 is necessarily greater than 7, but hardly that the number of planets is so necessarily, though there are in fact nine planets. One may believe that Cicero denounced Catilius without believing that Tully did so, if one is ignorant of the fact that Cicero was Tully. Wiggins disposes of such cases by saying that he follows Quine ‘in holding that intensional replacements [for property-variables] are excluded by “the incoherence of bound variables in any but referential positions”.’ This has the acceptable consequence that expressions like ‘being believed by so and so to be such and such’ are not regarded as standing for genuine properties, but it is a point that could have done with some elaboration rather than being merely consigned to a relatively short footnote.

Wiggins is strongly opposed to the idea that the identity of a temporally-extended object can be resolved into a relation, or set of relations, between its momentary phases. He quotes with approval Leibniz’s saying that ‘by itself continuity no more constitutes substance than does multitude or number,’ and with scornful disapproval the view of C.D. Broad that the history of the cliffs of Dover has as good a title to be considered an event as a flash of lightning or a motor accident, the only relevant difference being that it lasts longer than they do and that the adjacent slices of the motor accident are more diversified. His own view is that ‘material objects and events are in some sense duals,’ and he contends that so far from its being possible to take the quality and location of an event as primitive, and treat a persistent object as the sum of a set of suitably qualified and localised events, it is only our knowledge of the principles which determine the persistence of an object of the type in question that enables us to assign the appropriate qualities to the events. I have to say that I do not find this argument convincing. What we are initially presented with in sense-perception is a set of momentary patterns; because these patterns reappear in relatively constant spatial relations to one another, we are led to form the idea of their persisting in the intervals between their manifestations; so eventually they come to be represented as individual objects tracing a continuous path in space-time. Wiggins’s assumption that the conditions of identity for a persistent object are somehow made available to us in advance of its appearance seems an unwarranted concession to the doctrine of innate ideas.

In fairness to Wiggins, it should be added that he does not express opposition to what he calls ‘four-dimensional talk as such’. He allows that physicists have discovered that ‘the three-dimensional Aristotelian way of thinking’ is unequal to the task of explaining physical phenomena ‘at the deepest possible level’, or even to giving an adequate description of ‘the most familiar physical events and processes’. This should not, however, lead us, in his view, to overlook the fact that ‘the old three-dimensional language is the language of much of biological science, the language of the physics laboratory itself, and the language of aerodynamics and other “applications” of physics’. Neither, he maintains, could ‘the pure space-time language’ have come to be used intelligibly unless our ordinary three-dimensional talk of physical continuants had served to introduce it. This may very well be so, but need not amount to more than the uncontested fact that we are all brought up to speak in Aristotelian terms. It would not show that the four-dimensional method of description was logically derivative and that it could not represent a complete world-view. The crucial question, as Wiggins sees, is whether it is possible to translate out all our ordinary discourse into four-dimensional terms, and he is pretty clearly of the opinion that it is not. He is content to leave the matter there, since he holds that we have no proper motive for seeking to identify the disparate entities that respectively figure in the different forms of language. As he sees it, ‘the two kinds of entity simply do not compete for room in the world’. It seems to me, on the contrary, that they must compete, unless their discord is removed by the analysis of one in terms of the other.

Wiggins follows current fashion in finding ample room for de re modalities. He argues that any term like ‘man’ which stands for what we call a ‘natural kind’ has its sense fixed, not by the fact that the things to which it applies possess some set of manifest properties, but rather by their possessing some internal constitution, of the nature of which the user of the term need not even be aware. The idea is that the term is initially understood as applying to some specimens of its extension; when the physical constitution of these specimens is discovered, the term is then taken to apply to anything which has that constitution. It is further held that since the possession of such a constitution is a necessary condition for anything to fall under the term, the necessity can be insinuated into the bestowal of the property that marks out the object as being of its natural kind. So supposing that G is a feature of man’s constitution, Wiggins is ready to assert that anything is ‘necessarily if-a-man-then-G’. There would be no objection to this if it amounted to no more than the result of a decision to withhold the term ‘man’ from anything that lacked the property ‘G’. As a matter of fact, I think it doubtful whether all the terms that are said to stand for natural kinds do have their senses fixed in this fashion, but it is a policy that could he followed without one’s committing oneself to a belief in natural necessity. Wiggins, however, wishes to make it a necessary property of, say, Julius Caesar that he was a man, and consequently also necessary that he had the requisite constitution. Here he no longer carries me with him. There is indeed a problem about the interpretation of proper names, but whatever its correct solution, I believe that necessity plays no part in it except as characterising some logical consequence of an assumption with which the use of the name is linked. If the name ‘Julius Caesar’ is taken to refer to some particular man, then it is taken to refer to that man, but there is no rule that it must always so be taken, nor is anything thereby said about the man himself.

I also think that Wiggins is mistaken in holding that statements of identity like ‘Hesperus is Phosphorus’ or ‘Dickens is Boz’ are necessarily true, if they are true at all. It is surely an entirely contingent fact that the two phases of Venus, which we count as appearances of the Evening and the Morning Star, are spatio-temporally connnected in the way that they are, and an equally contingent fact that ‘Boz’ is a pseudonym for Dickens rather than any other person, or indeed that it is a pseudonym at all. Wiggins relies on an argument, originally propounded by Ruth Barcan Marcus, which has for its premisses that if x and y are identical, they have all the same properties, and that x has the property of being necessarily identical with itself. It then will follow that if y is identical with x, it is necessarily identical with x. Since I hold the conclusion to be false, without denying that it follows from the premisses, I must reject at least one of the premisses, and in fact I should deny that the identity of a thing with itself can correctly be said without further ado to be a necessary property. Undeniably, if two verbal tokens name the same thing they name the same thing, but it is only if they carry the same associations that they do so necessarily, and this may not be the case even if the two tokens are of the same type.

Wiggins deals at length with the problem of personal identity, reviews the possibility, suggested long ago by Locke and revived in our own time by dealers in Science Fiction, of a divergence between physical continuity and the continuity of consciousness, but favours the conservative view that the concept of a person is not to be treated ‘otherwise than as a peculiar restriction of the natural kind concept animal’.

It is in accordance with this view that he allows memory to be relevant to personal identity only as being one important element ‘in the account of what it is for a person to be still there fully alive’. I am inclined to think that this underplays the role of memory, but here as elsewhere Wiggins has strong arguments in support of the position that he holds.