17. Materialism and the Autonomous World 3

What does World 3 look like from a materialistic point of view? Obviously, the bare existence of aeroplanes, airports, bicycles, books, buildings, cars, computers, gramophones, lectures, manuscripts, paintings, sculptures and telephones presents no problem for any form of physicalism or materialism. While to the pluralist these are the material instances, the embodiments, of World 3 objects, to the materialist they are simply parts of World 1.

But what about the objective logical relations which hold between theories (whether written down or not), such as incompatibility, mutual deducibility, partial overlapping, etc.? The radical materialist replaces World 2 objects (subjective experiences) by brain processes. Especially important among these are dispositions for verbal behaviour: dispositions to assent or reject, to support or refute; or merely to consider — to rehearse the pros and cons. Like most of those who accept World 2 objects (the "mentalists"), materialists usually interpret World 3 contents as if they were "ideas in our minds": but the radical materialists try, further, to interpret "ideas in our minds" - and thus also World 3 objects - as brain-based dispositions to verbal behaviour.

Yet neither the mentalist nor the materialist can in this way do justice to World 3 objects, especially to the contents of theories, and to their objective logical relations.

World 3 objects just are not "ideas in our minds", nor are they dispositions of our brains to verbal behaviour. And it does not help if one adds to these dispositions the embodiments of World 3, as mentioned in the first paragraph of this section. For none of these copes adequately with the abstract character of World 3 objects, and especially with the logical relations existing between them.⁴

As an example, Frege's Grundgesetze was written, and partly printed, when he deduced, from a letter written by Bertrand Russell, that there was a self-contradiction involved in its foundation. This self-contradiction had been there, objectively, for years. Frege had not noticed it: it had not been "in his mind". Russell only noticed the problem (in connection with quite a different manuscript) at a time when Frege's manuscript was complete. Thus there existed for years a theory of Frege's (and a similar more recent one of Russell's) which was objectively inconsistent without anyone's having an inkling of this fact, or without anyone's brain state disposing him to agree to the suggestion "This manuscript contains an inconsistent theory".

To sum up, World 3 objects and their properties and relations cannot be reduced to World 2 objects. Nor can they be reduced to brain states or dispositions; not even if we were to admit that all mental states and processes can be reduced to brain states and processes. This is so despite the fact that we can regard World 3 as the product of human minds.

Russell did not invent or produce the inconsistency, but he discovered it. (He invented, or produced a way of showing or proving that the inconsistency was there.) Had Frege's theory not been objectively inconsistent, he could not have applied Russell's inconsistency proof to it, and he would not have thus convinced himself of its untenability. Thus a state of Frege's mind (and no doubt also a state of Frege's brain) was the result, partly, of the objective fact that this theory was inconsistent: he was deeply upset and shaken by his discovery of this fact. This, in turn, led to his writing (a physical World 1 event) the words, "Die ArithmetikistinsSchwankengeraten"("Arithmetic is tottering"). Thus there is interaction between (a) the physical, or partly physical, event of Frege's receiving Russell's letter; (b) the objective hitherto unnoticed fact, belonging to World 3, that there was an inconsistency in Frege's theory; and (c) the physical, or partly physical, event of Frege's writing his comment on the (World 3) status of arithmetic.

These are some of the reasons why I hold that World 1 is not causally closed, and why I assert that there is interaction (though an indirect one) between World 1 and World 3. It seems to me clear that this interaction is mediated by mental, and partly even conscious, World 2 events.

The physicalist, of course, cannot admit any of this.

I believe that the physicalist is also prevented from solving another problem: he cannot do justice to the higher functions of language.

This criticism of physicalism relates to the analysis of the functions of language that was introduced by my teacher, Karl Buehler. He distinguished three functions of language: (1) the expressive function; (2) the signal or release function; and (3) the descriptive function (see Buehler [1918]; [1934], p. 28). I have discussed Buehler's theory in various places,⁵ and I have added to his three functions a fourth — (4) the argumentative function. Now I have argued elsewhere³ that the physicalist is only able to cope with the first and the second of these functions. As a result, if faced with the descriptive and the argumentative functions of language, the physicalist will always see only the first two functions (which are also always present), with disastrous results.

In order to see what is at issue, it is necessary to discuss briefly the theory of the functions of language.

In Buehler's analysis of the act of speech he differentiates between the speaker (or, as Buehler also calls him, the sender) and the person spoken to, the listener (or the receiver). In certain special ("degenerate") cases the receiver may be missing, or he may be identical with the sender. The four functions here discussed (there are others, such as command, exhortation, advice — see also John Austin's [1962] "performative utterances") are based on relations between (a) the sender, (b) the receiver, (c) some other objects or states of affairs which, in degenerate cases, may be identical with (a) or (b). I will give a table of the functions in which the lower functions are placed lower and the higher functions higher.

The following comments may be made on this table:

(1) The expressive function consists in an outward expression of an inner state. Even simple instruments such as a thermometer or a traffic light "express" their states in this sense. However, not only instruments, but also animals (and sometimes plants) express their inner state in their behaviour. And so do men, of course. In fact, any action we undertake, not merely the use of a language, is a form of self-expression.

(2) The signalling function (Buehler calls it also the "release function") presupposes the expressive function, and is therefore on a higher level. The thermometer may signal to us that it is very cold. The traffic light is a signalling instrument (though it may continue to work during hours where there may not always be cars about). Animals, especially birds, give danger signals; and even plants signal (for example to insects); and when our self-expression (whether linguistic or otherwise) leads to a reaction, in an animal or in a man, we can say that it was taken as a signal.

(3) The descriptive function of language presupposes the two lower functions. What characterizes it, however, is that over and above expressing and communicating (which may become quite unimportant aspects of the situation), it makes statements that can be true or false: the standards of truth and falsity are introduced. (We may distinguish a lower half of the descriptive function where false descriptions are beyond the animal's (the bee's?) power of abstraction. Also a thermograph would belong here, for it describes the truth unless it breaks down.)

(4) The argumentative function adds argument to the three lower functions, with its values of validity and invalidity.

Now, functions (1) and (2) are almost always present in human language; but they are as a rule unimportant, at least when compared with the descriptive and argumentative functions.

However, when the radical physicalist and the radical behaviourist turn to the analysis of human language, they cannot get beyond the first two functions (see my [1953 (a)]). The physicalist will try to give a physical explanation — a causal explanation — of language phenomena. This is equivalent to interpreting language as expressive of the state of the speaker, and therefore as having the expressive function alone. The behaviourist, on the other hand, will concern himself also with the social aspect of language — but this will be taken, essentially, as affecting the behaviour of others; as "communication", to use a vogue word; as the way in which speakers respond to one another's "verbal behaviour". This amounts to seeing language as expression and communication.

But the consequences of this are disastrous. For if all language is seen as merely expression and communication, then one neglects all that is characteristic of human language in contradistinction to animal language: its ability to make true and false statements, and to produce valid and invalid arguments. This, in its turn, has the consequence that the physicalist is prevented from accounting for the difference between propaganda, verbal intimidation, and rational argument.

It might also be mentioned that the characteristic openness of human language — the capacity for an almost infinite variety of responses to any given situation, to which Noam Chomsky, particularly, has forcefully drawn our attention — is related to the descriptive function of language. The picture of language — and of the acquisition of language — as offered by behaviouristically inclined philosophers such as Quine seems, in fact, to be a picture of the signalling function of language. This, characteristically, is dependent upon the prevailing situation. As Chomsky has argued [1969] the behaviourist account does not do justice to the fact that a descriptive statement can be largely independent of the situation in which it is used.

⁴  For a fuller discussion of this, see section 21, below.

⁵  For example, my [1963 (a)], chapters 4 and 12; [1972 (a)], chapters 2 and 6.

⁴ The dancing bees may perhaps be said to convey factual or descriptive information. A thermograph or barograph does so in writing. It is interesting that in both cases the problem of lying does not seem to arise — although the maker of the thermograph may use it to misinform us.