Bertrand Russell


Russell Society Home Page

About Bertrand Russell

About the Russell Society

The BRS Library

Society Publications

Russell Texts Online

Russell Resources

JOIN the Russell Society!

Officers and Organization

Contact Us



 Philosophy and Grammar (1936)*

By Bertrand Russell

Review of Language, Truth and Logic, by Alfred J. Ayer, Research Student of Christ Church, Oxford. Gollancz, 9s

THIS book is very different from those of the philosophical school which has dominated Oxford since the time of Bradley. Its outlook, as we are informed in the Preface, is, in the main, that of the Viennese circle, and especially of Carnap. It is a very good thing that this philosophy should be presented in English in a readable and not too difficult form, as it is the youngest and most vigorous offspring of the marriage of empiricism and mathematical logic which took place at the beginning of the present century. Mr. Ayer, in common with the Viennese circle, rejects many of the traditional problems of philosophy as illusory, while he regards the remainder as essentially linguistic. Genuine propositions he divides into tautologies, which are a priori and logical, and empirical hypotheses, which are capable of being rendered probable or improbable by sense-experience. What falls outside these two classes is “metaphysics,” which is “neither true nor false, but literally senseless.”

This point of view, whether true or false, has led, in recent years, to much valuable work, which will remain valid in detail even if the general theory is abandoned. Mr. Ayer writes in a manner which is intelligible to the general educated public, with vigour and clarity; the things that he has to say will be new to most readers who are not professional philosophers, and are, at the lowest estimate, worthy of serious consideration.

The condemnation of “metaphysics” leads to some very sweeping conclusions. For example, the proposition “God exists” is condemned as meaningless; from this follows not only a rejection of theism, but also of atheism, which maintains the equally meaningless proposition “God does not exist,” and of agnosticism, which asserts “whether God exists is doubtful.” This view is maintained on the double ground that there can be no empirical evidence either for or against the theistic hypothesis, and that the hypothesis is neither logically necessary nor logically impossible. Traditional theology has, of course, denied this: the argument from design gives empirical reasons for the orthodox view, and the ontological argument maintains that “God exists” is an analytic proposition, the denial of which is self-contradictory. But the rigorous methods of modem logic have made both these views somewhat difficult to maintain. Mr. Ayer is thus led to a view which is opposed equally to the assertions of the orthodox and to the doubts or denials of the sceptics.

The orthodox and the unorthodox alike will feel a certain reluctance to accept the view that the words “God exists” are a mere meaningless noise, like “Abracadabra.” Whatever may be the logical definition of deity, the word “God” is one which arouses certain emotions, and the question in people’s minds is whether there is an object to which these emotions are appropriate. This question is not disposed of by Mr. Ayer’s arguments.

Propositions about matters of fact, according to this philosophy, are never completely certain, even when they come as near as is linguistically possible to the mere assertion of a present sensible occurrence. Those which come nearest to this ideal are called “experiential propositions,” and others are “verified,” or rendered probable, by having consequences which are found to be experiential propositions. The problem of induction is discussed in this connection, and is dismissed as a fictitious problem on the ground that it cannot be solved. “We are entitled,” says Mr. Ayer, “to have faith in our procedure just so long as it does the work which it is designed to do—that is, enables us to predict future experience.” The trouble is that we never know whether our procedure will predict future experience until it has done so; if we are to know this in advance, we shall need a principle of induction in the sense in which, according to the author, no such principle can be known to be true. The author tells us, however, that it is “rational” to be guided by past experience even though it may mislead us as to the future. This, apparently, is a definition of the word “rational”; at least one must suppose so, since no evidence is offered.

While I find myself in broad agreement with Mr. Ayer’s outlook, which is that of “logical positivism,” I cannot but think that there are difficulties of which he is insufficiently aware. Take, for example, the question of the evidence for the existence of persons and things other than ourselves. “It does not follow,” we are told, “from the fact that each man’s experiences are private to himself that no one ever has good reason to believe that another man’s experiences are qualitatively the same as his own. For we define the qualitative identity and difference of two people’s sense-experiences in terms of the similarity and dissimilarity of their reactions to empirical tests.” The trouble is that, when you think you are observing another man’s reactions, it is only your own that you can really observe. You can separate off a group of your own actual and possible experiences and define it as another person, but this is hardly what we mean when we assert that other persons exist; it is, however, all that Mr. Ayer’s principles permit.

Another difficulty in logical positivism concerns the relation between a sensible occurrence and an “experiential proposition,” which, we are told, “records an actual or possible observation.” Mr. Ayer refuses to discuss the problem of meaning, but in the absence of some discussion of this question it is difficult to see how he can know that a form of words “records” an observation. Does he know anything about the occurrence except the form of words? If not, how does he know that the words describe the occurrence? If yes, what is the nature of this non-verbal knowledge? And when some empirical proposition is verified by an occurrence, what is the relation between the occurrence and the proposition, and how is this relation known? Such questions, of which the above is only a sample, seem to me to show that, though logical positivism is effective in sweeping away a mass of ancient nonsense, it has not yet solved all its problems. A casual reading of Mr. Ayer’s book might give the impression that philosophy is finished, and there are no soluble but unsolved problems left. To a philosopher it is comforting to think that this is perhaps not the case.


*  Bertrand Russell, The London Mercury 33 (March 1936), 541-3