The fist number of L. refer to the standard CtP edition published in 1987. The following number shows correspondence between letters in the new 2010 edition. Note that on this website CtP is available only 1987 edition with minor additions.

[L. 73 | 80] 30 November 1963

I have finished Russell's Nightmares and must confess that they did not come up to expectation. No doubt it was my fault for expecting too much, knowing how unsatisfactory I find his philosophical views; but I had hoped that, at least, when he was not writing normal philosophy, he would be entertaining. Alas! I found his wit insipid, and his serious passages almost intolerable—there was something of the embarrassment of meeting a Great Man for the first time, and finding him even more preoccupied with trivialities than oneself.

In his Introduction, Russell says 'Every isolated passion is, in isolation, insane; sanity may be defined as a synthesis of insanities', and then he proceeds to give us examples of isolated insanities—the Queen of Sheba as Female Vanity, Bowdler as Prudery, the Psycho-Analyst as Social Conformity, and so on. Amongst these, as you noted, is the Existentialist as Ontological Scepticism. Here, Russell's satire is directed partly against what Sartre has called 'a literature of extreme situations'; and this, for an Englishman, is no doubt a legitimate target, since the English do not admit that there are such things—though, of course, this makes the English a target for the satire of the rest of Europe, particularly the French.

But what Russell is not entitled to do is to group the insanity of doubting one's existence along with the other insanities, and this for the simple reason that it precedes them. One may be vain or modest; one may be prudish or broadminded; one may be a social conformist or an eccentric; but in order to be any of these things, one must at least be. The question of one's existence must be settled first—one cannot be insanely vain if one doubts whether one exists at all and, precisely, Russell's existentialist does not even succeed in suffering—except when his philosophy is impugned (but this merely indicates that he has failed to apply his philosophy to itself, and not, as Russell would have us believe, because he has failed to regard his philosophy in the light of his other insanities). The trouble really is, that Russell does not, or rather will not, admit that existence poses a problem at all; and, since he omits this category from all his thinking nothing he says concerns anybody in particular.

It is noteworthy that the one nightmare that did amuse me, that of the Metaphysician, does in fact represent Russell's own personal nightmare—a fear of discovering existence (for existence and the negative—'not'—go hand in hand). But Russell has long ago firmly repressed this fear by harsh logical measures, and it only shows its head when he is off his logical guard. Once upon a time, Russell said 'Whatever A may be, it certainly is'; but that was in 1903. Since then Russell has learned sanity (his own brand), and has declared (in 1919) 'It is of propositional functions that you can assert or deny existence'. In other words, Russell holds that you can assert 'lions exist', and that this means '"X is a lion" is sometimes true', but that if you say 'this lion exists' you have said something meaningless. From this it follows that Russell regards the assertion 'I exist' as a meaningless utterance, and this allows him to regard the existentialist as a lunatic.

It is no doubt true that the assertions, 'I exist', 'I do not exist', and so on, are meaningless, but only in the eyes of one who is no longer a puthujjana. And, even then, they are not meaningless in Russell's sense. According to one of the Commentaries, the Buddha once said that 'all puthujjanas are mad', and from this point of view the puthujjana's doubts about his existence are insanity. But this is not Russell's point of view, since he is still a puthujjana.

Together with existence, Russell has removed the word 'not' from Logic (even if he does not go so far as his metaphysician Bumblowski, who has expelled it from his ordinary language). Russell came to the conclusion (I speak from memory) that to say 'A is not B', where A and B are individual things, is illegitimate; what one should say is '"A is B" is false'. Thus, instead of exists and not, Russell has true and false; but whereas the first pair applies to things, the second pair applies to facts—it is only of propositions that you can assert the truth or the falsity. (For the significance of this replacement of things by facts—it is the foundation of positivism—I would refer you to note (f) of the Preface to the Notes.) I may say that I enjoyed Russell's idea of a special department of Hell for those philosophers who have refuted Hume—this is one of the few points about which I agree with Russell (but does it not make nonsense of Russell's whole philosophy of the acceptance of 'scientific common sense'? Russell would be only too happy to be able to refute Hume).

I was interested by the Mathematician's Nightmare, but for quite a different reason. There, you will remember, Professor Squarepoint has a vision in which all the numbers come to life and dance a ballet. Amongst these numbers there is one that refuses to be disciplined, and insists on coming forward. It is 137,[1] and this number is the cosmic number that Sir Arthur Eddington found to be at the base of physics. Now it so happened that I used to be interested in Eddington's interest in this apparently rather undistinguished number, perhaps even because it is so undistinguished in every other respect. And it happened that my interest in this number enabled me, indirectly, to write FUNDAMENTAL STRUCUTRE. Although, now, I have entirely lost my interest in 137, and although it plays no part in my description of Fundamental Structure, yet it is not difficult to trace it in the Notes. In §I/9, I say that the structure of a thing of certain complexity is represented by {mosimage}. This is arrived at by purely phenomenological description (i.e. in the reflexive description of experience as such). Now, Eddington (I reproduce his arguments as far as I remember them) says that this figure represents the structure of a 'particle' (in nuclear physics).[a] Now, so long as Eddington sticks to the figure above as the structure of a 'particle' he remains (whether he knows it or not) within the field of phenomenology (which requires an 'observer' as well as an 'observed'—like the 'subject' and 'object' in phassa). But Eddington is a quantum physicist, and must treat his results with scientific objectivity (which eliminates the 'observer' or 'subject'—see the last footnote to the Preface), and so he must do away with himself. How does he do it? Answer: by putting another 'particle', similar to the first, to take his place. Eddington then quietly retires, leaving a relationship between two identical 'particles'. To find out the nature of this relationship we simply have to multiply the two 'particles' together. Since each 'particle' has 10 o's and 6 x's, simple arithmetic gives us 100 oo's, 36 xx's, and 120 xo's (or ox's). For some reason that I now forget, we ignore the unlike pairs (xo's and ox's), and consider only the oo's and xx's. Added together these come to 136. And this, so it seems, is the number of degrees of freedom of the electron. But there is a snag: since the two particles we multiplied together are absolutely indistinguishable in all respects, we can never know, in any calculation, whether we have got them the right way round or not. So one extra degree of freedom has to be added to compensate for our uncertainty. The total number is therefore 137. (I am afraid, perhaps, that these pages may be something of 'The District Judge's Nightmare'; but there's nothing in them of any importance whatsoever.)

In any case, thank you for sending the book, which both satisfied my curiosity and exercised my critical faculty.


[73.a] I do not allow the validity of the arguments he uses to derive this figure; such, for example, as the postulate that a given particle A has an equal chance of existing or of not existing. This strange assumption, which has currency neither with Russell nor with me, has as its immediate consequence the remarkable conclusion that exactly the same number of things exist as do not exist. (Whatever one may think of this, it is apparently good currency in quantum theory, if we are to judge from the following utterance by Dirac: 'We may look upon these unoccupied states as holes among the occupied ones.... The holes are just as much physical things as the original particles....' [PQM, p. 252] But it must be remembered that quantum theory is an ad hoc system made to account for the observed facts and produce results. So long as it does this [and it does it only rather imperfectly] nobody bothers about whether it is intelligible or not.) [Back to text]

Editorial note:

[73.1] 137: A fraction very close to 1/137 is known to physicists as the 'fine structure constant'. It and its factors are involved in considerations of the weak nuclear and electromagnetic forces and as such it is an important constant for quantum physicists in describing basic electron-electron scattering. [Back to text]