1 The allusion is to Kierkegaard’s conversation with Christian VIII in which the King asked him for his views on how a King should conduct himself. Kierkegaard said such things as: “First, it would be a good thing for the King to be ugly.” (Christian VIII was very good-looking.) “Then he should be deaf and blind, or at least behave as if he were, for this solves many difficulties…. And then, he must not say much, but must have a little standard speech that can be used on all occasions, a speech therefore without content.” (Francis Joseph used to say: “It was very nice, and it pleased me very much.”—“Es war sehr schön, es hat mich sehr gefreut.”)
2 The case arose from my work with children. One of the boys for whom I was responsible had fallen from a climbing frame and had suffered a fractured skull. I was acquitted because I could prove that I had demanded for months that the authorities should remove the climbing frame, which I regarded as dangerous. (The authorities had tried to put the blame on me; a procedure about which the judge had some strong words to say.)
2a The old house still exists. The entrance was Freisingergasse 4 until approximately 1920; afterwards it became Bauernmarkt 1.
3 See Otto Weininger, Geschlecht und Charakter (Vienna: Braumüller, 1903), p. 176: “All blockheads, from Bacon to Fritz Mauthner, have been critics of language.” (Weininger adds that he should ask Bacon to forgive him for associating him in this way with Mauthner.) Compare this with Tractatus, 4.0031.
4 Cp. n. 57 to Chap. 12 of O.S. [1945(c)], p. 297; [1950(a)], p. 653; [1962(c)], [1963(1)], and later editions, p. 312.
5 Roger Martin du Gard, L’Été 1914; English translation by Stuart Gilbert, Summer 1914 (London: John Lane, The Bodley Head, 1940).
6 The problem has recently reached a new stage through Abraham Robinson’s work on the infinitely small; see Abraham Robinson, Non-Standard Analysis (Amsterdam: North-Holland Publishing Company, 1966).
7 The term “essentialism” (widely used now) and especially its application to definitions (“essentialist definitions”) were, to my knowledge, first introduced in section 10 of The Poverty [1944(a)]; see esp. pp. 94–97; [1957(g)] and later editions, pp. 27–30; and in my O.S., Vol. I [1945(b)], pp. 24–27; and Vol. II [1945(c)], pp. 8–20, 274–86; [1950(a)], pp. 206–18, 621–38; [1962(c)], [1963(1)], and later editions: Vol. I, pp. 29–32; Vol. II, pp. 9–21, 287–301. There is a reference on p. 202 of Richard Robinson’s Definition (Oxford: Oxford University Press, 1950) to the 1945 edition of my O.S. [1945(c)], Vol. II, pp. 9–20; and what he says, for example, on pp. 153–57 (cp. the “utterances” on p. 158), and also on pp. 162–65, is in some respects very similar to what I say in the pages of my book to which he refers (though his remark on p. 71 about Einstein and simultaneity does not agree with what I say in [1945(c)], pp. 18f.; 108f.; [1950(a)], pp. 216f., 406; [1962(c)] and [1963(1)], Vol. II, pp. 20, 220). Compare also Paul Edwards, ed., The Encyclopedia of Philosophy (New York: Macmillan Company and Free Press, 1967; London: Collier Macmillan, 1967), Vol. II, pp. 314–17. “Essentialism” is there discussed at length under the main entry Definition (reference is made in the Bibliography to Robinson).
7a (Added in proofs.) I have recently made a change in terminology from the first, second and third worlds to world 1, world 2, and world 3, upon the suggestion of Sir John Eccles. For my older terminology, see [1968(r)] and [1968(s)]; for Sir John’s suggestion, see his Facing Reality (New York, Heidelberg and Berlin: Springer-Verlag, 1970). The suggestion came too late to be incorporated into the original text of the present book except in one or two places. (Added 1975: I have now revised this to some extent.) See also n. 293 below, and also The Self, esp. Chap. P2.
8 Annual Philosophical Lecture, British Academy, 1960 [1960(d)], [1961(f)]; republished in C.&R. [1963(a)]; see esp. pp. 19 f. See also p. 349 of my “Epistemology Without a Knowing Subject” [1968(s)], now Chap. 3 of my [1972(a)]. (The table reproduced here is a slight modification of the original one.)
9 Cp. the 3d ed. of C.&R. [1969(h)], p. 28, the newly inserted point 9. (Point 9 of the earlier editions is now numbered 10.)
10 Not even Gottlob Frege states it quite explicitly, though this doctrine is certainly implicit in his “Sinn und Bedeutung”, and he even produces there arguments in its support. Cp. Peter Geach and Max Black, eds., Translations from the Philosophical Writings of Gottlob Frege (Oxford: Blackwell, 1952), pp. 56–78.
11 Cp. my article “Quantum Mechanics without ‘The Observer’ ” [1967(k)]; see esp. pp. 11–15, where the present problem is discussed. (This particular equivalence, incidentally, is questioned there.)
12 One could hardly write in a prose translation (Parmenides, fragments 14–15): Bright in the night with an alien light round the earth she is erring, Always she wistfully looks round for the rays of the sun.
12a See Benjamin Lee Whorf, Language, Thought, and Reality (Cambridge, Mass.: M.I.T. Press, 1956).
13 Gottlob Frege suggests — mistakenly, I think — in “Der Gedanke”, Beiträg zur Philos. d. deutschen Idealismus, 1 (1918–19), 58–77 (excellently translated by A. M. and Marcelle Quinton as “The Thought: A Logical Enquiry”, Mind, n. s. 65 [1956], 289–311), that only of the emotional aspects of speech is a “perfect (vollkommene) translation almost impossible” (p. 63; p. 295 of the translation), and that “The more strictly scientific a presentation… the more easily is it translated” (ibid.). Ironically enough, Frege continues to say quite correctly that it makes no difference to any thought content which of the four German synonyms for “horse” (Pferd, Ross, Gaul, Mähre — they are different only in emotional content: Mähre, in particular, need not in every context be a female horse) is used in any formulation. Yet this very simple and unemotional thought of Frege’s is, it appears, untranslatable into the English language, since English does not seem to have three good synonyms for “horse”. The translator would, therefore, have to become a commentator by finding some common English word which has three good synonyms—preferably with strikingly different emotional or poetic associations.
14 Cp., for example, section 37 of my L.d.F. [1934(b)], [1966(e)] and later editions; and also of L.Sc.D. [1959(a)] and later editions. The example I had in mind was gravitational redshift.
15 For this idea, and the quotation, see section 6 of my L.d.F. [1934(b)], p. 13; [1966(e)], p. 15; “Sie sagen um so mehr, je mehr sie verbieten.”; L.Sc.D. [1959(a)] and later editions, p. 41: “The more they prohibit the more they say.” The idea was adopted by Rudolf Carnap in section 23 of his Introduction to Semantics (Cambridge, Mass.: Harvard University Press, 1942); see esp. p. 151. There Carnap attributes this idea to Wittgenstein “due to an error of memory”, as he himself puts it in section 73 of his Logical Foundations of Probability (Chicago: University of Chicago Press, 1950), p. 406, where he attributes it to me. Carnap writes there: “The assertive power of a sentence consists in its excluding certain possible cases”. I should now stress that these “cases” are, in science, theories (hypotheses) of a higher or a lower degree of universality. (Even what I called “basic statements” in L.Sc.D. are, as I stressed there, hypotheses, though of a low degree of universality.)
16 The subset of the informative content which consists of basic statements (empirical statements) I called in L.Sc.D. the class of the theory’s “potential falsifiers”, or its “empirical content”.
17 For non-a belongs to the informative content of a, and a to the informative content of non-a, but a does not belong to its own informative content (unless it is a contradiction).
18 The proof (which in the particular form given here was shown to me by David Miller) is quite straightforward. For the statement “b or t or both” follows from “a or t or both” if and only if it follows from a; that is, if and only if the theory t follows from “a and non-b”. But because a and b contradict one another (by hypothesis), this last statement says the same as a. Thus “b or t or both” follows from “a or t or both” if and only if t follows from a; and this, by assumption, it does not.
19 J. W. N. Watkins, Hobbes’s System of Ideas (London: Hutchinson, 1965), pp. 22 f.; second ed., 1973, pp. 8 f.
20 (This note originally formed part of the text.)
All this can be stated even if we confine ourselves to just one of the two ideas of content so far discussed. It becomes even clearer in terms of a third idea of content, that is, the idea of the problem content of a theory.
Following a suggestion of Frege’s, we may introduce the notion of a yes-or-no problem or, briefly, a y-problem: given any statement a (say, “Grass is green”), the corresponding y-problem (“Is grass green?”) may be denoted by “y(a)”. One sees at once that y(a) = y(non-a): the problem whether grass is green is, qua problem, identical with the problem whether grass is not green, even though the two questions are differently formulated, and even though the answer “Yes” to one of them is equivalent to the answer “No” to the other.
We can define what I propose to call the problem content of a theory t in either of two equivalent ways: (1) it is the set of all those y(a) for which a is an element of the logical content of t; (2) it is the set of all those y(a) for which a is an element of the informative content of t. Thus the problem content is related to the two other contents in identical ways.
In our previous example of N (Newton’s theory) and E (Einstein’s), y(E) belongs to the problem content of N, and y(N) to that of E. If we denote by K (= K1 and K2 and K3) the statement which formulates Kepler’s three laws, restricted to the two-body problem, then K1 and K2 follow from N but contradict E, while K3 and therefore K contradict both N and E. (See my paper [1957(i)], [1969 (k)], now Chap. 5 of [1972(a)]; and also [1963(a)], p. 62, n. 28.) Nevertheless, y(K), and y(K1), y(K2), y3), all belong to the problem content both of N and of E, and y(N) and y(E) both belong to the problem contents of K, of K1, of K2, and of K3.
That y(E), the problem of the truth or falsity of Einstein’s theory, belongs to the problem content of K and to that of N illustrates the fact that there can be no transitivity here. For the problem whether the theory of the optical Doppler effect is true—that is, y(D)—belongs to the problem content of E, but not to that of N or that of K.
Although there is no transitivity there may be a link: the problem contents of a and of b may be said to be linked by y(c) if y(c) belongs to that of a and also to that of b. Obviously, the problem contents of any a and b can always be linked by choosing some appropriate c (perhaps c = a or b); thus the bare fact that a and b are linked is trivial; but the fact that they are linked by some particular problem y(c) (which interests us for some reason or other) may not be trivial, and may add to the significance of a, of b, and of c. Most links are, of course, unknown at any given time.
21 Gottlob Frege, Grundgesetze der Arithmetik (Jena: H. Pohle, 1903), Vol. II, section 56.
22 Clifford A. Truesdell, “Foundations of Continuum Mechanics”, in Delaware Seminar in the Foundations of Physics, ed. by Mario Bunge (Berlin, Heidelberg, New York: Springer-Verlag, 1967), pp. 35–48; see esp. p. 37.
23 Gottlob Frege, “Über Begriff und Gegenstand”, Vierteljahrsschrift f. wissenschaftliche Philos., 16 (1892), 192–205. Cp. p. 43 of Geach and Black, eds., Philosophical Writings of Gottlob Frege, pp. 42–55 (see n. 10 above).
24 See n. *1 to section 4; [1959(a)] and later editions, p. 35; [1966(e)] (e)] and later editions, p. 9; and also my two Prefaces.
25 The problems dealt with here are discussed (though perhaps not fully enough) in the various Prefaces to L.d.F. and L.Sc.D. It is perhaps of some interest that the fact that I criticized there in some detail the whole approach of language analysis was not even mentioned when this book was reviewed in Mind (see also my reply to this review in n. 243 to section 33, below), though this journal was an obvious place in which to mention, and to answer, such a criticism; nor has the criticism been mentioned elsewhere. For other discussions of problems connected with the topic of this digression, see the references in n. 7 in the preceding section 6, and my various discussions of the descriptive and argumentative functions of language in C.&R., [1963(a)] and later editions; and also [1966(f)], [1967(k)], [1968(r)], and [1968(s)] (the first of these now forms Chap. 6 and the last two Chaps. 3 and 4 of [1972(a)]).
An interesting example of a key word (ephexēs in Plato’s Timaeus 55A) which has been misinterpreted (as “next in order of magnitude”, instead of “next in order of time” or perhaps “in adjacent order”) because the theory was not understood, and which can be interpreted in two different senses (“successively” in time, or “adjacent” applied to plane angles) without affecting Plato’s theory, may be found in my paper “Plato, Timaeus 54E–55A” [1970 (d)]. For similar examples, see the 3d ed. of C.&R. [1969(h)], esp. pp. 165 and 408–12. In brief, one cannot translate without keeping the problem situation constantly in mind.
26 See section IV of Chap. 19 of my O.S., [1945(c)], [1950(a)], and later editions, for the ambiguity of violence; and also the Index under “violence”.
27 See, for comments on all this, The Poverty [1944(a) and (b)] and [1945(a)], and [1957(g)], and esp. Chaps. 17 to 20 of my O.S. [1945(c)], [1966(a)]. The remarks on the workers of Vienna which follow here in the text repeat in the main what I said in my O.S., in nn. 18 to 22 to Chap. 18, and n. 39 to Chap. 19. See also the references given in n. 26 above on the ambiguity of violence.
28 G. E. R. Gedye, Fallen Bastions (London: Victor Gollancz, 1939).
29 Cp. [1957(a)], reprinted as Chap. 1 of C.&R., [1963(a)] and later editions.
30 Cp. Ernst Mach, The Science of Mechanics, 6th English ed. with an Introduction by Karl Menger (La Salle, III.: Open Court Publishing Co., 1960), Chap. 2, section 6, subsection 9.
31 The formulation in italics was first suggested, and its significance discussed, in [1949(d)], now translated as the Appendix to [1972(a)]; see also [1957(i) & (j)], [1969(k)], now Chap. 5 of [1972(a)].
32 Albert Einstein, Über die spezielle und die allgemeine Relativitätstheorie (Braunschweig: Vieweg, 1917); see esp. Chap. 22. I have used my own translation, but the corresponding passage occurs on p. 77 of the English translation referred to in the next footnote. It should be noted that Newton’s theory lives on as a limiting case in Einstein’s theory of gravitation. (This is particularly clear if Newton’s theory is formulated in a “general relativistic” or “covariant” way, by taking the velocity of light as infinite [c = ¥]. This was shown by Peter Havas, “Four-Dimensional Formulations of Newtonian Mechanics and Their Relation to the Special and the General Theory of Relativity”, Reviews of Modern Physics, 36 [1964], 938–65.)
33 Albert Einstein, Relativity: The Special and the General Theory. A Popular Exposition (London: Methuen & Co., 1920), p. 132. (I have slightly improved upon the translation.)
34 L.d.F. [1934(b)], p. 13; [1966(e)] and later editions, p. 15; and L.Sc.D. [1959(a)] and later editions, p. 41; see n. 15 to section 7 above.
35 Cp. Hans Albert, Marktsoziologie und Entscheidungslogik (Neuwied and Berlin: Herman Luchterhand Verlag, 1967); see esp. pp. 149, 227 f., 309, 341 f. My very clumsy term, which Albert replaced by “immunization against criticism”, was “conventionalist stratagem”.
(Added in proofs.) David Miller has now drawn my attention to n. I on p. 560 of Arthur Pap, “Reduction Sentences and Dispositional Concepts”, in The Philosophy of Rudolf Carnap, ed. by Paul Arthur Schilpp (la Salle, III.: Open Court Publishing Co., 1963), pp. 559–97, which anticipates this use of “immunization”.
36 Cp. Chap. 1 of my C.&R., [1963(a)] and later editions.
37 For a much fuller discussion, see sections 2, 3, and 5 of my Replies.
38 See C.&R., [1963(a)] and later editions, , esp. the Appendix, pp. 248–50; Chap. 11, pp. 275–77; Chap. 8, pp. 193–200; and Chap. 17, p. 346. The problem was first discussed by me in section 15 of L.d.F. [1934(b)], pp. 33 f.; [1966(e)] and later editions, pp. 39–41; L.Sc.D., [1959(a)] and later editions, pp. 69 f. A fairly full discussion of certain metaphysical theories (centred on metaphysical determinism and indeterminism) is to be found in my paper “Indeterminism in Quantum Physics and in Classical Physics” [1950(b)]; see esp. pp. 121–23.
39 See pp. 37 f. of C.&R. [1963(a)] and later editions.
40 See [1945(c)], pp. 101 f.; [1962(c)] and later editions, Vol. II, pp. 108 f.
41 See Imre Lakatos, “Changes in the Problem of Inductive Logic”, in The Problem of Inductive Logic, ed. by Imre Lakatos (Amsterdam: North-Holland Publishing Co., 1968), pp. 315–417, esp. p. 317.
42 There does not seem to be any systematic time-dependence, as there is in the learning of meaningless syllables.
43 Cp. C. Lloyd Morgan, Introduction to Comparative Psychology (London:
Scott, 1894), and H. S. Jennings, The Behaviour of the Lower Organisms (New York: Columbia University Press, 1906).
44 My view of habit formation may be illustrated by a report about the gosling Martina in Konrad Lorenz, On Aggression (London: Methuen & Co., 1966), pp. 57 f. Martina acquired a habit consisting of a certain detour towards a window before mounting the stairs to the first floor of Lorenz’s house in Altenberg. This habit originated (ibid., p. 57) with a typical escape reaction towards the light (the window). Although this first reaction was “repeated”, “the habitual detour… became shorter and shorter”. Thus repetition did not create this habit; and in this case it even tended to make it slowly disappear. (Perhaps this was something like an approach towards a critical phase.) Incidentally, many asides of Lorenz’s seem to be in support of my view that scientists use the critical method—the method of conjectures and attempted refutations. For example he writes (ibid., p. 8): “It is a very good morning exercise for a research scientist to discard a pet hypothesis every day before breakfast.” Yet in spite of this insight he seems still to be influenced by inductivism. (See, for example, ibid., p. 62: “But perhaps a whole series of countless repetitions… was necessary”; for another passage with clearly methodological intent see Konrad Lorenz, Über tierisches und menschliches Verhalten [Munich: R. Piper & Co., 1965], p. 388.) He does not always seem to realize that in science “repetitions” of observations are not inductive confirmations but critical attempts to check oneself—to catch oneself in a mistake. See also below, n. 95 to section 15, and text.
45 According to The Oxford English Dictionary, the phrase “rule of trial and error” originated in arithmetic (see Trial 4). Note that neither Lloyd Morgan nor Jennings used the term in the sense of random trials. (This latter use seems to be due to Edward Thorndike.)
46 Drawing a ball blindly from an urn does not ensure randomness unless the balls in the urn are well mixed. And blindness regarding the solution need not involve blindness regarding the problem: we may know that our problem is to win a game by drawing a white ball.
47 D. Katz, Animals and Men (London: Longmans, 1937), p. 143.
47a An exception is the school of Otto Selz (killed by the Nazis) and his pupil Adriaan D. De Groot. (See n. 305a, below.)
48 Jane Austen, Emma (London: John Murray, 1816), Vol. III, end of Chap. 3 (Chap. 39 of some later editions). Cp. p. 336 of R. W. Chapman, ed., The Novels of Jane Austen, 3d ed. (Oxford: Oxford University Press, 1933), Vol. IV.
49 For the development of games, see Jean Piaget, The Moral Judgment of the Child (London: Routledge & Kegan Paul, 1932), esp. p. 18 for the dogmatic first two stages and the critical “third stage”; see also pp. 56–69. See further Jean Piaget, Play, Dreams, and Imitation in Childhood (London: Routledge & Kegan Paul, 1962).
50 Something like this view may be found in Søren Kierkegaard, Repetition (Princeton: Princeton University Press; Oxford: Oxford University Press, 1942); cp., for example, pp. 77 f. See my L.Sc.D., new Appendix *x.
51 Joseph Church, Language and the Discovery of Reality (New York: Random House, 1961), p. 36.
52 Ibid.
53 This seems to be the obvious explanation of the tragic incident of Helen Keller’s alleged plagiarism when she was still a child, an incident which made a great impression on her, and perhaps helped her to sort out the different sources of the messages which all reached her in one and the same code.
54 W. H. Thorpe writes in a passage (to which Arne Petersen has drawn my attention) in his interesting book Learning and Instinct in Animals (London: Methuen & Co., 1956), p. 122 (2d rev. ed., 1963, p. 135): “By true imitation is meant the copying of a novel or otherwise improbable act or utterance, or some act for which there is clearly no instinctive tendency.” (Italicized in the original.) There can be no imitation without elaborate instinctive tendencies for copying in general, and even for the specific kind of imitating act in particular. No tape recorder can work without its built-in (as it were innate) ability for learning by imitation (imitation of vibrations) and if we do not provide it with a substitute for the need or drive to use its abilities (perhaps in the form of a human operator who wants the machine to do some recording and playing back), then it will not imitate. This seems to be true, then, of even the most passive forms of learning by imitation of which I can think. It is of course quite correct that we should speak of imitation only if the act to be imitated is not one which would be performed by animal A from instinct alone, without its having been first performed by another animal B in the presence of A. But there will be cases in which we have reason to suspect that A may have produced the act—perhaps at a somewhat later stage—without imitating B. Should we not call it a true imitation if B’s act led to A’s performing the act (much) earlier than it would have done otherwise?
55 C.&R., [1963(a)] and later editions, Chap. 1, esp. pp. 42–52. I refer there on p. 50, n. 16, to a thesis “Gewohnheit und Gesetzerlebnis” [On Habit and Belief in Laws] which I presented (in an unfinished state) in 1927, and in which I argued against Hume’s idea that habit is merely the (passive) result of repetitive association.
56 This is somewhat similar to Plato’s theory of knowledge in Meno 80D–86C but of course also dissimilar.
57 I feel that here is the place, more than anywhere else, to acknowledge the help I have received throughout this essay from my friends Ernst Gombrich and Bryan Magee. It was perhaps not so difficult for Ernst Gombrich for, although he does not agree with all I say about music, he at least sympathizes with my attitude. But Bryan Magee emphatically does not. He is an admirer of Wagner (on whom he has written a brilliant book, Aspects of Wagner [London: Alan Ross, 1968; New York: Stein & Day, 1969]). Thus he and I are here as completely at loggerheads as two people can possibly be. It is of lesser moment that in his judgement my sections 13 and 14 contain well-known muddles, and that some of the views I attack are Aunt Sallies. Of course, I do not quite agree with this; but the point I wish to make here is that our disagreement has not prevented him from helping me immensely, not only with the rest of this autobiographical sketch but also with these two sections that contain views on which we have seriously disagreed for many years.
58 It is a long time since I gave up these studies and I cannot now remember the details. But it seems to me more than probable that there was a certain amount of parallel singing, at the organum stage, which contained thirds and fifths (reckoned from the bass). I feel that this should have preceded fauxbourdon singing.
59 See D. Perkin Walker, “Kepler’s Celestial Music”, Journal of the Warburg and Courtauld Institutes, 30 (1967), 228–50. I am greatly indebted to Dr Walker for drawing my attention to the passage which I quote in the text. It is from Kepler, Gesammelte Werke, ed. by Max Caspar (Munich, 1940), Vol. VI, p. 328. The passage is quoted in Latin by Walker, Kepler’s Celestial Music, pp. 249 f., who also gives an English translation. The translation here is my own. (I translate: ut mirum amplius non sit = there is no marvel greater or more sublime; ut luderet [= that he should enact] = that he should conjure up a vision of; ut quadamtenus degusterat = that he should almost [taste or touch or] reach.) Incidentally, I cannot agree that Plato’s harmony of the spheres was monodic and consisted “only of scales” (cp. Walker, Kepler’s Celestial Music, n. 3 and text); on the contrary, Plato takes the greatest care to avoid this interpretation of his words. (See for example Republic 617B, where each of the eight Sirens sings one single tune, such that from all the eight together “there came the concord of one single harmony”. Timaeus 35B–36B and 90D should be interpreted in the light of this passage. Relevant is also Aristotle, De sensu vii, 448 a 20 ff. where the views of “some writers on concords” are examined who “say that sounds do not arrive simultaneously but merely seem to do so”.) See also on singing in octaves Aristotle’s Problems 918 b 40, 919 b 33–35 (“mixture”; “consonance”) and 921 a 8–31 (see esp. 921 a 27 f.); and cp. the harmony of the heavens in De caelo 290 b 12–24 and in Plato’s myth of Er (Rep., 617B).
60 I have alluded to this story in Chap. 1 of C.&R. [1963(a)] and later editions, end of section vi, p. 50.
61 It was only years later that I realized that in asking “How is science possible?” Kant had Newton’s theory in mind, augmented by his own interesting form of atomism (which resembled that of Boscovich); cp. C.&R., Chaps. 2, 7, and 8, and my paper “Philosophy and Physics” [1961(h)].
62 For this distinction (and also for a more subtle one) see C.&R. [1963(a)], Chap. 1, section v, pp. 47 f.
63 Albert Schweitzer, J. S. Bach (Leipzig: Breitkopf und Härtel, 1908); first published in French in 1905; 7th ed., 1929. See the English ed. (London: A. & C. Black, 1923), Vol. I, p. 1. Schweitzer uses the term “objective” for Bach and “subjective” for Wagner. I would agree that Wagner is far more “subjective” than Beethoven. Yet I should perhaps say here that, though I greatly admire Schweitzer’s book (especially his most excellent comments on the phrasing of Bach’s themes) I cannot at all agree with an analysis of the contrast between “objective” and “subjective” musicians in terms of the musician’s relation to his “time” or “period”. It seems to me almost certain that in this Schweitzer is influenced by Hegel, whose appreciation of Bach impressed him. (See ibid., pp. 225 f., and n. 56 on p. 230. On p. 225 [Vol. 1, p. 244 of the English ed.] Schweitzer recounts from Therese Devrient’s memoirs a charming incident involving Hegel which is not very flattering to him.)
64 The first of these [1968(s)] was an address delivered in 1967 and first published in Logic, Methodology and Philosophy of Science, Vol. III, pp. 333–73; the second [1968(r)] was first published in Proceedings of the XIVth International Congress of Philosophy, Vienna: 2nd to 9th September 1968, Vol. I, pp. 25–53. These two papers are now Chaps. 3 and 4 respectively of [1972(a)]. The third paper [1967(k)] cited in the text is in Quantum Theory and Reality. See also my L.d.F. and L.Sc.D., sections 29 and 30 [1934(b)], pp. 60–67; [1966(e)] and later editions, pp. 69–76; [1959(a)] and later editions, pp. 104–11; my C.&R. [1963(a)], esp. pp. 224–31; and my paper “A Realist View of Logic, Physics, and History” [1970(1)] in Physics, Logic and History, now Chap. 8 of [1972(a)].
65 See my O.S., Vol. I [1945(b)], pp. 26, 96; Vol. II [1945(c)], pp. 12 f.; [1950(a)], pp. 35, 108, 210–12; [1962(c)], and later editions, Vol. I, pp. 32, 109; Vol. II, pp. 13 f.
65a (Added 1975.) The same holds for expressionist or emotive theories of morals, and of moral judgements.
66 See also the last section of my paper “Epistemology Without a Knowing Subject” [1968(s)], pp. 369–71; [1972(a)], pp. 146–50.
67 Cited by Schweitzer, J. S. Bach, p. 153.
68 Arthur Schopenhauer, Die Welt als Wille und Vorstellung [The World as Will and Idea], Vol. II (1844), Chap. 39; the second quotation is from Vol. I (1818 [1819]), section 52. Note that the German word “Vorstellung” is simply the translation into German of John Locke’s term “idea”.
69 The German is: “eine cantable Art im Spielen zu erlangen”.
70 Plato, Ion; cp. esp. 533D–536D.
71 Ibid., 534E.
72 Plato, Ion, 535E; cp. 535C.
73 See also my paper “Self-Reference and Meaning in Ordinary Language” [1954(c)], which now forms Chap. 14 of C.&R. [1963(a)]; and text to n. 163 of my Replies in P. A. Schilpp ed., The Philosophy of Karl Popper (La Salle: Open Court, 1974). Arguments purporting to show that self-referring jokes are impossible may be found in Gilbert Ryle, The Concept of Mind [London: Hutchinson, 1949], for example, on pp. 193–96; Peregrine Books ed. [Harmondsworth: Penguin Books, 1963], pp. 184–88. I think that Ion’s remark is [or implies] “a criticism of itself” which according to Ryle, p. 196, should not be possible.
74 Plato, Ion 541E–542B.
75 See my O.S. [1945(b) and (c)] and later editions, nn. 40 and 41 to Chap. 4, and text.
76 Ernst Gombrich referred me to “In order to make me weep you yourself must suffer first” (Horace, Ad Pisones, 103 f.). Of course it is conceivable that what Horace intended to formulate was not an expressionist view but the view that only the artist who has suffered first is capable of critically judging the impact of his work. It seems to me probable that Horace was not conscious of the difference between these two interpretations.
77 Plato, Ion 541E f.
78 For much of this paragraph, and some criticism of the previous paragraphs, I am indebted to my friend Ernst Gombrich.
It will be seen that the secularized Platonic theories (of the work of art as subjective expression and communication, and as objective description) correspond to Karl Bühler’s three functions of language; cp. my [1963(a)], pp. 134 f. and 295, and section 15.
79 See E. H. Gombrich, Art and Illusion (London: Phaidon Press; New York: Pantheon Books, 1960; latest edition, 1972), passim.
80 It will be seen that my attitude towards music resembles the theories of Eduard Hanslick (caricatured by Wagner as Beckmesser), a music critic of great influence in Vienna, who wrote a book against Wagner (Vom Musikalisch-Schönen [Leipzig: R. Weigel, 1854]; trans. by G. Cohen from the 7th rev. ed. as The Beautiful in Music [London: Novello and Co., 1891]). But I do not agree with Hanslick’s rejection of Bruckner who, though venerating Wagner, was in his way as saintly a musician as Beethoven (who is now sometimes wrongly accused of dishonesty). It is an amusing fact that Wagner was greatly impressed by Schopenhauer—by The World as Will and Idea—and that Schopenhauer wrote in the Parerga, Vol. II, section 224 (first published in 1851, when Wagner was starting work on the music of The Ring), “One can say that Opera has been the bane of music”. (He meant of course recent opera, although his arguments sound very general—much too general in fact.)
81 Friedrich Nietzsche, Der Fall Wagner [The Case of Wagner] (Leipzig, 1888) and Nietzsche contra Wagner; both translated in The Complete Works of Friedrich Nietzsche, ed. by Oscar Levy (Edinburgh and London: T. N. Foulis, 1911), Vol. VIII.
82 Arthur Schopenhauer, Parerga, Vol. II, section 224.
83 Karl Bühler, Die geistige Entwicklung des Kindes (Jena: Fischer, 1918; 3d ed., 1922); English translation, The Mental Development of the Child (London: Kegan Paul, Trench, Trubner & Co., 1930). For the functions of language, see also his Sprachtheorie (Jena: Fischer, 1934); see esp. pp. 24–33.
84 A word may perhaps be said here on Aristotle’s hygienic theory of art. Art no doubt has some biological or psychological function like catharisis; I do not deny that great music may in some sense purify our minds. But is the greatness of a work of art summed up in the fact that it cleanses us more thoroughly than a lesser work? I do not think that even Aristotle would have said this.
85 Cp. C.&R., pp. 134 f., 295; Of Clouds and Clocks [1966 (f)], now Chap. 6 of [1972(a)], sections 14–17 and n. 47; “Epistemology Without a Knowing Subject” [1968(s)], esp. section 4, pp. 345 f. ([1972(a)], Chap. 3, pp. 119–22).
86 Leonard Nelson was an outstanding personality, one of the small band of Kantians in Germany who had opposed the First World War, and who upheld the Kantian tradition of rationality.
87 See my paper “Julius Kraft 1898–1960” [1962(f)].
88 See Leonard Nelson, “Die Unmöglichkeit der Erkenntnistheorie”, Proceedings of the IVth International Congress of Philosophy, Bologna; 5th to 11th April 1911 (Genoa: Formiggini, 1912), Vol. I, pp. 255–75; see also L. Nelson, Über das sogenannte Erkenntnisproblem (Göttingen: Vandenhoeck & Ruprecht, 1908).
89 See Heinrich Gomperz, Weltanschauungslehre (Jena and Leipzig: Diederichs, 1905 and 1908), Vol. I, and Vol. II, part 1. Gomperz told me that he had completed the second part of the second volume but had decided not to publish it, and to abandon his plans for the later volumes. The published volumes were planned and executed on a truly magnificent scale, and I do not know the reason why Gomperz ceased to work on it, about eighteen years before I met him. Obviously it had been a tragic experience. In one of his later books, Über Sinn und Sinngebilde—Verstehen und Erklären (Tübingen: Mohr, 1929), he refers to his earlier theory of feelings, esp. on pp. 206 f. For his psychologistic approach—which he called “pathempiricism” (Pathempirismus) and which emphasized the role of feelings (Gefühle) in knowledge—see esp. Weltanschauungslehre, sections 55–59 (Vol. II, pp. 220–93). Cp. also sections 36–39 (Vol. I, pp. 305–94).
90 Karl Bühler, “Tatsachen und Probleme zu einer Psychologie der Denkvorgänge”, Archiv f. d. gesamte Psychologie, 9 (1907), 297–365; 12 (1908), 1–23, 24–92, 93–123.
91 Otto Selz, Über die Gesetze des geordneten Denkverlaufs (Stuttgart: W. Spemann, 1913), Vol. I; (Bonn: F. Cohen, 1922), Vol. II.
92 Oswald Külpe, Vorlesungen über Logik, ed. by Otto Selz (Leipzig: S. Hirzel, 1923).
93 A similar mistake can be found even in Principia Mathematica, since Russell failed, in places, to distinguish between an inference (logical implication) and a conditional statement (material implication). This confused me for years. Yet the main point—that an inference was an ordered set of statements—was sufficiently clear to me in 1928 to be mentioned to Bühler during my (public) Ph.D. examination. He admitted very charmingly that he had not considered the point.
94 See C.&R. [1963(a)], pp. 134 f.
95 I now find a similar argument in Konrad Lorenz: “… modifiability occurs… only in those… places where built-in learning mechanisms are phylogenetically programmed to perform just that function.” (See Konrad Lorenz, Evolution and Modification of Behaviour [London: Methuen & Co., 1966], p. 47.) But he does not seem to draw from it the conclusion that the theories of reflexology and of the conditioned reflex are invalid: see especially ibid., p. 66. See also section 10 above, esp. n. 44. One can state the main difference between association psychology or the theory of the conditioned reflex on the one hand, and discovery by trial and error on the other, by saying that the former is essentially Lamarckian (or “instructive”) and the latter Darwinian (or “selective”). See now for example the investigations of Melvin Cohn, “Reflections on a Discussion with Karl Popper: The Molecular Biology of Expectation”, Bulletin of the All-India Institute of Medical Sciences, 1 (1967), 8–16, and later works by the same author. For Darwinism, see section 37.
96 W. von Bechterev, Objektive Psychologie oder Psychoreflexologie (originally published 1907–12), German ed. (Leipzig and Berlin: Teubner, 1913); and Allgemeine Grundlagen der Reflexologie des Menschen (originally published 1917), German ed. (Leipzig and Vienna: F. Deuticke, 1926); English ed., General Principles of Human Reflexology (London: Jarrolds, 1933).
97 The title of my (unpublished) dissertation was “Zur Methodenfrage der Denkpsychologie” [1928(a)].
98 Compare with this paragraph some of my remarks against Reichenbach at a conference in 1934 ([1935(a)] reprinted in [1966(e)], [1969(e)], p. 257); there is a translation in L.Sc.D., [1959(a)] and later editions, p. 315: “Scientific theories can never be ‘justified’, or verified. But… a hypothesis A can… achieve more than a hypothesis B…. The best we can say of a hypothesis is that up to now… it has been more successful than other hypotheses although, in principle, it can never be justified, verified, or even shown to be probable.” See also the end of section 20 (text to nn. 156–58), and n. 243 to section 33, below.
99 Rudolf Carnap, Der logische Aufbau der Welt, and Scheinprobleme in der Philosophie: das Fremdpsychische und der Realismusstreit, both first published (Berlin: Weltkreis-Verlag, 1928); second printing, both books in one (Hamburg: Felix Meiner, 1961). Now translated as The Logical Structure of the World and Pseudoproblems of Philosophy (London: Routledge & Kegan Paul, 1967).
100 Victor Kraft, Die Grundformen der wissenschaftlichen Methoden (Vienna: Academy of Sciences, 1925).
101 See p. 641 of Herbert Feigl’s charming and most informative essay, “The Wiener Kreis in America”, in Perspectives in American History (The Charles Warren Center for Studies in American History, Harvard University, 1968), Vol. II, pp. 630–73; and also n. 106 below. [Upon inquiry Feigl suggests that Zilsel may have become a member after his—Feigl’s—emigration to the United States.]
102 Herbert Feigl says (ibid., p. 642) that it must have been in 1929, and no doubt he is right.
103 My only published papers before I met Feigl—and for another four years after—were on educational topics. With the exception of the first [1925(a)] (published in an educational journal Schulreform) they were all ([1927(a)], [1931(a)], [1932(a)]) written at the invitation of Dr Eduard Burger, the editor of the educational journal Die Quelle.
104 Feigl refers to the meeting in “Wiener Kreis in America”. I have briefly described the opening move of our discussion in C.&R. [1963(a)], pp. 262 f.; see n. 27 on p. 263. See also “A Theorem on Truth-Content” [1966(g)], my contribution to the Feigl Festschrift.
105 During that first long conversation, Feigl objected to my realism. (He was at that time in favour of a so-called “neutral monism”, which I regarded as Berkeleyan idealism; I still do.) I am happy at the thought that Feigl too became a realist.
106 Feigl writes, “Wiener Kreis in America”, p. 641, that both Edgar Zilsel and I tried to preserve our independence “by remaining outside the Circle”. But the fact is that I should have felt greatly honoured had I been invited, and it would never have occurred to me that membership in Schlick’s seminar could endanger my independence in the slightest degree. (Incidentally, before reading this passage of Feigl’s I did not realize that Zilsel was not a member of the Circle; Victor Kraft records him as one in The Vienna Circle [New York: Philosophical Library, 1953]; see p. 4.)
107 See my publications listed on p. 44 of my paper “Quantum Mechanics Without ‘The Observer’” [1967(k)].
108 The manuscript of the first volume and parts of the manuscript of that version of L.d.F. which was cut by my uncle still exist. The manuscript of the second volume, with the possible exception of a few sections, seems to have been lost. (Added 1976.) The extant (German) material is at present being prepared by Troels Eggers Hansen for publication by J. C. B. Mohr in Tübingen.
109 See in particular now my [1971(i)], reprinted with minor alterations as Chap. 1 of [1972(a)]; and also section 13 of my Replies.
109a See sections 13 and 14 of my Replies.
110 See John Passmore’s article “Logical Positivism” in Encyclopedia of Philosophy, ed. by Paul Edwards, Vol. V, p. 56 (see n. 7 above).
111 This letter [1933(a)] was first published in Erkenntnis, 3, Nos. 4–6 (1933), 426 f. It is republished in translation in my L.Sc.D., [1959(a)] and later editions, pp. 312–14, and in its original language in the second and later editions of L.d.F. [1966(e)], [1969(e)], etc., pp. 254–56.
112 J. R. Weinberg, An Examination of Logical Positivism (London: Kegan Paul, Trench, Trubner & Co., 1936).
113 For a much fuller discussion of this legend, see sections 2 and 3 of my Replies.
113a (Added 1975.) I suppose that this phrase was an echo of John Laird, Recent Philosophy (London: Thornton Butterworth, 1936), who describes me as “a critic although also an ally” of the Vienna Circle (see p. 187; also pp. 187–90).
114 Cp. Arne Naess, Moderne filosofer (Stockholm: Almqvist & Wiksell/Gebers Förlag AB, (1965); English translation as Four Modern Philosophers (Chicago and London: University of Chicago Press, 1968). Naess writes in n. 13 on pp. 13 f. of the translation: “My own experience was rather similar to Popper’s…. The polemic [in an unpublished book of Naess’s]… written… between 1937 and 1939 was intended to be directed against fundamental theses and trends in the Circle, but was understood by Neurath as a proposal for modifications which were already accepted in principle and were to be made official in future publications. Upon this assurance I gave up plans to publish the work.”
114a For the impact of all these discussions, see nn. 115 to 120.
115 Cp. C.&R. [1963(a)], pp. 253 f.
116 Rudolf Carnap, “Über Protokollsätze”, Erkenntnis, 3 (1932), 215–28; see esp. 223–28.
117 Cp. Rudolf Carnap, Philosophy and Logical Syntax, Psyche Miniatures (London: Kegan Paul, 1935), pp. 10–13, which correspond to Erkenntnis, 3 (1932), 224 ff. Carnap speaks here of “verification” where before he (correctly) reported me as speaking of “testing”.
118 Cp. C. G. Hempel, Erkenntnis, 5 (1935), esp. 249–54, where Hempel describes (with reference to Carnap’s article “Über Protokollsätze”) my procedure very much as Carnap had reported it.
119 Rudolf Carnap, Erkenntnis, 5 (1935), 290–94 (with a reply to Reichenbach’s criticism of L.d.F.). C. G. Hempel, Deutsche Literaturzeitung, 58 (1937), 309–14. (There was also a second review by Hempel.) I mention here only the more important reviews and criticisms from members of the Circle.
120 Hans Reichenbach, Erkenntnis, 5 (1935), 367–84 (with a reply to Carnap’s review of L.d.F., to which Carnap in turn briefly replied). Otto Neurath, Erkenntnis, 5 (1935), 353–65.
121 Werner Heisenberg, “Über quantentheoretische Umdeutung kinematischer und mechanischer Beziehungen”, Zeitschrift für Physik, 33 (1925), 879–93; Max Born and Pascual Jordan, “Zur Quantenmechanik”, ibid., 34 (1925), 858–88; Max Born, Werner Heisenberg, and Pascual Jordan, “Zur Quantenmechanik II”, ibid., 35 (1926), 557–615. All three papers are translated in Sources of Quantum Mechanics, ed. by B. L. van der Waerden (Amsterdam: North-Holland Publishing Co., 1967).
122 For a report of the debate see Niels Bohr, “Discussion with Einstein on Epistemological Problems in Atomic Physics”, in Albert Einstein: Philosopher-Scientist, ed. by Paul Arthur Schilpp (Evanston, III.: Library of Living Philosophers, Inc., 1949); 3d ed. (La Salle, III.: Open Court Publishing Co., 1970), pp. 201–41. For a criticism of Bohr’s contentions in this debate, see my L.Sc.D. [1959(a)], new Appendix * xi, pp. 444–56, L.d.F. [1966(e)] and [1969(e)], pp. 399–411, and [1967(k)].
123 James L. Park and Henry Margenau, “Simultaneous Measurability in Quantum Theory”, International Journal of Theoretical Physics, 1 (1968), 211–83.
124 See [1957(e)] and [1959(e)].
125 See [1934(b)], pp. 171 f., [1959(a)], pp. 235 f., [1966(e)], pp. 184 f.; [1967(k)], pp. 34–38.
126 Albert Einstein: Philosopher-Scientist, pp. 201–41 (see n. 122 above).
127 See esp. [1957(i)], [1969(k)], now Chap. 5 of [1972(a)]; [1963 (h)]; [1966(f)], now Chap. 6 of [1972(a)]; [1967(k)]; and [1968(s)], now Chap. 3 of [1972(a)], in which also is reprinted, as Chap. 4, [1968(r)], where a fuller treatment can be found.
128 Arthur March, Die Grundlagen der Quantenmechanik (Leipzig: Barth, 1931); cp. the Index of [1934(b)], [1959(a)], or [1966(e)].
129 The results given here are partly of a later and partly of an earlier date. For my latest views see my contribution to the Landé Festschrift, “Particle Annihilation and the Argument of Einstein, Podolsky, and Rosen” [1971(n)].
130 Cp. John von Neumann, Mathematische Grundlagen der Quantenmechanik (Berlin: Springer-Verlag, 1931), p. 170; or the translation, Mathematical Foundations of Quantum Mechanics (Princeton: Princeton University Press, 1955), p. 323. Thus even if von Neumann’s argument were valid, it would not disprove determinism. Moreover, his assumed “rules” I and II on pp. 313 f. (cp. p. 225 f.)—German edition p. 167 (cp. p. 118)—are inconsistent with the commutation relations, as was first shown by G. Temple, “The Fundamental Paradox of the Quantum Theory”, Nature, 135 (1935), 957. (That von Neumann’s rules I and II are inconsistent with quantum mechanics was clearly implied by R. E. Peierls, “The Fundamental Paradox of the Quantum Theory”, Nature, 136 [1935], 395. See also Park and Margenau, “Simultaneous Measurability in Quantum Theory” [see n. 123 above].) John S. Bell’s paper is “On the Problem of Hidden Variables in Quantum Mechanics”, Reviews of Modern Physics, 38 (1966), 447–52.
131 C. S. Peirce, Collected Papers of Charles Sanders Peirce, ed. by Charles Hartshorne and Paul Weiss (Cambridge, Mass.: Harvard University Press, 1935), Vol. VI; see item 6.47 (first published 1892), p. 37.
132 According to Schrödinger, Franz Exner made the suggestion in 1918: see Erwin Schrödinger, Science, Theory, and Man (New York: Dover Publications, 1957), pp. 71, 133, 142 f. (originally published as Science and the Human Temperament [London: Allen and Unwin, 1935]; see pp. 57 f., 107, 114); and Die Naturwissenschaften, 17 (1929), 732.
133 von Neumann, Mathematical Foundations of Quantum Mechanics, pp. 326 f. (German edition p. 172): “… the apparent causal order of the world in the large (… [of the] objects visible to the naked eye) has certainly no other cause than the ‘law of large numbers’ and it is completely independent of whether the natural laws governing the elementary processes are causal or not”. (Italics mine; von Neumann refers to Schrödinger.) Obviously this situation has no direct connection with quantum mechanics.
134 See also my [1934(b)], [1959(a)], and later editions, section 78 (and also 67–70); [1950(b) and (c)]; [1957(g)], Preface; [1957(e)], [1959(e)]; [1966(f)], esp. section iv ([1972(a)], Chap. 6); [1967(k)].
135 This is the view which I have upheld consistently. It can be found, I believe, in Richard von Mises.
136 Alfred Landé, “Determinism versus Continuity in Modern Science”, Mind, n.s. 67 (1958), 174–81, and From Dualism to Unity in Quantum Physics (Cambridge: Cambridge University Press, 1960), pp. 5–8. (I have called this argument “Landé’s blade”.) Added 1975: See now also John Watkins’s paper “The Unity of Popper’s Thought”, in The Philosophy of Karl Popper, ed. by Paul Arthur Schilpp, pp. 371–412.
137 Cp. [1957(e)], [1959(e)], and [1967(k)].
138 Why should particles not be particles, at least to a first approximation, to be explained perhaps by a field theory? (A unified field theory of the type, say, of Mendel Sachs.) The only objection known to me derives from the “smear” interpretation of the Heisenberg indeterminacy formulae; if the “particles” are always “smeared”, they are not real particles. But this objection does not seem to hold water: there is a statistical interpretation of quantum mechanics.
(Since writing the above I have written a contribution to the Landé Festschrift [1971(n)] referred to in n. 129 above. And since then, I have read two outstanding works defending the statistical interpretation of quantum mechanics: Edward Nelson, Dynamical Theories of Brownian Motion [Princeton: Princeton University Press, 1967], and L. E. Ballentine, “The Statistical Interpretation of Quantum Mechanics”, Reviews of Modern Physics, 42 [1970], 358–81. It is most encouraging to find some support after a lone fight of thirty-seven years.)
139 See esp. [1967(k)].
139a This sentence was added in 1975.
140 W. Duane, “The Transfer in Quanta of Radiation Momentum to Matter”, Proceedings of the National Academy of Sciences (Washington), 9 (1923), 158–64. The rule may be written:
See Werner Heisenberg, The Physical Principles of the Quantum Theory (New York: Dover, 1930), p. 77.
141 Landé, Dualism to Unity in Quantum Physics, pp. 69, 102 (see n. 136 above), and New Foundations of Quantum Mechanics (Cambridge: Cambridge University Press, 1965), p. 5–9.
142 See esp. [1959(a)], [1966(e)], new Appendix * xi; and [1967(k)].
143 Albert Einstein, “Zur Elektrodynamik bewegter Körper”, Annalen der Physik, 4th ser. 17, 891–921; translated as “On the Electrodynamics of Moving Bodies” in Albert Einstein et. al., The Principle of Relativity, trans. by W. Pennett and G. B. Jeffrey (New York: Dover, 1923), pp. 35–65.
144 Einstein, Relativity: Special and General Theory (1920 and later editions). The German original is Über die spezielle und die allgemeine Relativitätstheorie (Brunswick: Vieweg & Sohn, 1916). (See nn. 32 and 33 above.)
144a (Added 1975.) This positivist and operationalist interpretation of Einstein’s definition of simultaneity was rejected by me in my O.S. [1945(c)], p. 18, and more strongly in [1957(h)] and later editions, p. 20.
145 See Einstein’s paper of 1905, section 1; in Principle of Relativity, pp. 38–40 (see n. 143 above).
146 By wrongly applying the very intuitive transitivity principle (Tr) to events beyond one system one can easily prove that any two events are simultaneous. But this contradicts the axiomatic assumption that within any inertial system there is a temporal order; that is, that for any two events within one system one and only one of the three relations holds: a and b are simultaneous; a comes before b; b comes before a. This is overlooked in an article by C. W. Rietdijk, “A Rigorous Proof of Determinism Derived from the Special Theory of Relativity”, Philosophy of Science, 33 (1966), 341–44.
147 Cp. Marja Kokoszyńska, “Über den absoluten Wahrheitsbegriff und einige andere semantische Begriffe”, Erkenntnis, 6 (1936), 143–65; cp. Carnap, Introduction to Semantics, pp. 240, 255 (see n. 15 above).
148 [1934(b)], section 84, “Wahrheit und Bewährung”; cp. Rudolf Carnap, “Wahrheit und Bewährung”, Proceedings of the IVth International Congress for Scientific Philosophy, Paris, 1935 (Paris: Hermann, 1936), Vol. IV, pp. 18–23; an adaptation appears in translation as “Truth and Confirmation”, in Readings in Philosophical Analysis, ed. by Herbert Feigl and Wilfrid Sellars (New York: Appleton-Century-Crofts, Inc., 1949), pp. 119–27.
149 Many members of the Circle refused at first to operate with the notion of truth: cp. Kokoszyńska, “Über den absoluten Wahrheitsbegriff” (see n. 147 above).
149a (Added 1975.) See especially L.Sc.D. [1959(a)] and later editions, points 4 to 6 on pp. 396 ff. (= L.d.F. [1966(e)], points 4 to 6 on pp. 349 f.).
150 Cp. Appendix iv of [1934(b)] and [1959(a)]. After the war, a proof of the validity of the construction was given by L. R. B. Elton and myself. (It is, I am afraid, my fault that our paper was never published.) In his review of L.Sc.D. (Mathematical Reviews, 21 [1960], Review 6318) I. J. Good mentions a paper of his own, “Normal recurring Decimals”, Journal of the London Mathematical Society, 21 (1946), 167–69. That my construction is valid follows easily—as David Miller has pointed out to me—from the considerations of this paper.
151 Karl Menger, “The Formative Years of Abraham Wald and His Work in Geometry”, The Annals of Mathematical Statistics, 23 (1952), 14–20; see esp. p. 18.
152 Karl Menger, ibid., p. 19.
153 Abraham Wald, “Die Widerspruchsfreiheit des Kollektivsbegriffes der Wahrscheinlichkeitsrechnung”, Ergebnisse eines mathematischen Kolloquiums, 8 (1937), 38–72.
154 Jean Ville, however, who read a paper in Menger’s Colloquium at about the same time as Wald, produced a solution similar to my “ideal random sequence”: he constructed a mathematical sequence which from the very start was Bernoullian, that is, random. (It was a somewhat “longer” sequence than mine; in other words, it did not become as quickly insensitive to predecessor selection as mine did.) Cp. Jean A. Ville, Étude critique de la notion de collectif, Monographies des Probabilités: calcul des probabilités et ses applications, ed. by Émile Borel (Paris: Gauthier-Villars, 1939).
155 For the various interpretations of probability, see esp. [1934 (b)], [1959(a)], and [1966(e)], section 48; and [1967(k)], pp. 28–34.
156 See the Introduction before section 79 of [1934(b)], [1959(a)], [1966(e)].
157 Compare to all this n. 243 to section 33, below, and text; see also section 16, text to n. 98.
158 See [1959(a)], p. 401, n. 7; [1966(e)], p. 354.
159 Some of this work is incorporated in the new appendices to L.Sc.D., [1959(a)], [1966(e)], and later editions.
160 I have read only two or three (very interesting) books about life in the Ghetto, especially Leopold Infeld, Quest. The Evolution of a Scientist (London: Victor Gollancz, 1941).
161 Cp. [1945(c)] and later editions, Chap. 18, n. 22; Chap. 19, nn. 35–40 and text, Chap. 20, n. 44 and text.
162 See John R. Gregg and F. T. C. Harris, eds., Form and Strategy in Science. Studies Dedicated to Joseph Henry Woodger (Dordrecht: D. Reidel, 1964), p. 4.
163 Many years later Hayek told me that it was Gottfried von Haberler (later of Harvard) who in 1935 had drawn his attention to L.d.F.
164 Cp. Bertrand Russell, “The Limits of Empiricism”, Proceedings of the Aristotelian Society, 36 (1936), 131–50. My remarks here allude especially to pp. 146 ff.
165 At the Copenhagen Congress—a congress for scientific philosophy—a very charming American gentleman took great interest in me. He said that he was the representative of the Rockefeller Foundation and gave me his card: “Warren Weaver, The European of the Rockefeller Foundation” (sic). This meant nothing to me; I had never heard about the foundations and their work. (Apparently I was very naive.) It was only years later that I realized that if I had understood the meaning of this encounter it might have led to my going to America instead of to New Zealand.
166 My opening talk to my first seminar in New Zealand was later published in Mind [1940(a)], and is now Chap. 15 of C.&R., [1963 (a)] and later editions.
167 Cp. [1938(a)]; [1959(a)], [1966(e)], Appendix * ii.
168 Cp. H. von Halban, Jr, F. Joliot, and L. Kowarski, “Liberation of Neutrons in the Nuclear Explosion of Uranium”, Nature, 143 (1939), 470 f.
169 Karl K. Darrow, “Nuclear Fission”, Annual Report of the Board of Regents of the Smithsonian Institution (Washington, D.C.: Government Printing Office, 1941), pp. 155–59.
170 See the historical note in The Poverty of Historicism [1957(g)], p. iv; American ed. [1964(a)], p. v.
171 This connection is briefly described in my British Academy lecture [1960(d)], now the Introduction to C.&R. [1963(a)]; see sections II and III.
172 See L.d.F. [1934(b)], pp. 227 f.; [1959(a)], p. 55, n. 3 to section 11; [1966(e)], p. 27. See also [1940(a)], p. 404, [1963(a)], p. 313, where the method of testing is described as an essentially critical, that is, faultfinding method.
173 Quite unnecessarily I used more often than not the ugly word “rationalist” (as in “rationalist attitude”) where “rational” would have been better, and clearer. The (bad) reason for this was, I suppose, that I was arguing in defence of “rationalism”.
174 See O.S., Vol. II, [1945(c)] and later editions, Chap. 24 (Chap. 14 of the German ed. [1958(i)]).
175 Adrienne Koch used “Critical Rationalism” as the title of the excerpts from O.S. that she selected for her book Philosophy for a Time of Crisis, An Interpretation with Key Writings by Fifteen Great Modern Thinkers (New York: Dutton & Co., 1959) [1959(k)].
176 Hans Albert, “Der kritische Rationalismus Karl Raimund Poppers”, Archiv für Rechts-und Sozialphilosophie, 46 (1960), 391–415. Hans Albert, Traktat über kritische Vernunft (Tübingen: Mohr, 1968; and later editions).
177 In the 4th ed. of O.S. [1962(c)], [1963(l) and (m)], and in later editions, there is an important Addendum to the second volume: “Facts, Standards, and Truth: A Further Criticism of Relativism” (pp. 369–96) which has been, so far as I know, overlooked by almost everybody.
178 I now regard the analysis of causal explanation in section 12 of L.d.F. (and therefore also the remarks in The Poverty and other places) as superseded by an analysis based on my propensity interpretation of probability [1957(e)], [1959(e)], [1967(k)]. This interpretation, which presupposes my axiomatization of the probability calculus (see, for example, [1959(e)], p. 40; [1959(a)], [1966(e)], Appendices *iv and *v), allows us to discard the formal mode of speaking and to put things in a more realistic way. We interpret
to mean: “The propensity of the state of affairs (or the conditions) b to produce a equals r.” (r is some real number.) A statement like (1) may be a conjecture, or deducible from some conjecture; for example, a conjecture about laws of nature.
We can then causally explain (in a generalized and weaker sense of “explain”) a as due to the presence of b, even if r does not equal 1. That b is a classical or complete or deterministic cause of a can be stated by a conjecture like
where x ranges over all possible states of affairs, including states incompatible with a or b. (We need not even exclude “impossible” states of affairs.) This shows the advantages of an axiomatization like mine, in which the second argument may be inconsistent.
This way of putting things is, clearly, a generalization of my analysis of causal explanation. In addition, it allows us to state “nomic conditionals” of various types—of type (1) with r < 1, of type (1) with r = 1, and of type (2). (Thus it offers a solution of the so-called problem of counterfactual conditionals.) It allows us to solve Kneale’s problem (see [1959(a)], [1966(e)], Appendix *x) of distinguishing between accidentally universal statements and naturally or physically necessary connections, as stated by (2). Notice however that there may be physically nonnecessary connections, which nevertheless are not accidental, like (1) with an r not far from unity. See also the reply to Suppes in my Replies.
179 See also The Poverty [1957(g)], p. 125. Reference should be made to J. S. Mill, A System of Logic, 8th ed., Book III, Chap. XII, section 1.
180 See Karl Hilferding, “Le fondement empirique de la science”, Revue des questions scientifiques, 110 (1936), 85–116. In this paper Hilferding (a physical chemist) explains at considerable length my views, from which he deviates in allowing inductive probabilities in the sense of Reichenbach.
181 See also Hilferding, “Le fondement empirique de la science”, p. 111, with a reference to p. 27 (that is, section 12) of the 1st ed. of L.d.F. [1934(b)].
182 See The Poverty [1957(g)], pp. 140 f. and 149 f., further developed in Chap. 14 of O.S. [1962(c) and (d)], [1963(l) and (m)]; [1966(i)]; [1967(d)]; [1968(r)] (now [1972(a)], Chap. 4); [1969(j)]; and in many unpublished lectures given at the London School of Economics and elsewhere.
183 See [1957(g)], sections 31 and 32, esp. pp. 149 and 154 f.
184 See Vol. II of [1962(c)], [1963(1) and (m)], pp. 93–99, and esp. pp. 97 f.
185 See [1950(a)], pp. 170 f.; [1952(a)], Vol. I, pp. 174–76.
186 See [1957(g)], section 30–32; [1962(c)]; and more recently [1968 (r)] and [1969(j)].
187 It was this situation which in 1945 led to the publication of a pamphlet Research and the University [1945(e)], drafted by me in cooperation with Robin S. Allan and Hugh Parton, and signed, after some minor changes, by Henry Forder and others. The situation changed in New Zealand very soon, but meanwhile I had left for England. (Added 1975: The story of this pamphlet is told by E. T. Beardsley in A History of the University of Canterbury, 1873–1973, by W. J. Gardner et al. [Christchurch, N.Z.: University of Canterbury, 1973].)
188 See esp. [1947(a)] and [1947(b)]. I was led to this work, partly, by problems of probability theory: the rules of “natural deduction” are very closely related to the usual definitions in Boolean algebra. See also Alfred Tarski’s papers of 1935 and 1936, which now form Chaps. XI and XII of his book Logic, Semantics, Metamathematics, trans. by J. H. Woodger (London and New York: Oxford University Press, 1956).
189 [1950(b) and (c)].
190 [1946(b)]; Chap. 9 of [1963(a)] and later editions.
191 The minutes of the meeting are not quite reliable. For example the title of my paper is given there (and it was so given on the printed list of meetings) as “Methods in Philosophy” instead of “Are there Philosophical Problems?”, which was the title ultimately chosen by me. Furthermore, the Secretary thought I was complaining that his invitation was for a brief paper, to introduce a discussion—which in fact suited me very well. He completely missed my point (puzzle versus problem).
192 See C.&R. [1963(a)], p.55.
193 See p. 167 of Ryle’s review of O.S. in Mind, 56 (1947), 167–72.
194 At a very early stage of the course he formulated, and showed the validity of, the metalinguistic rule of indirect proof: If a logically follows from non-a, then a is demonstrable.
195 Now in Tarski, Logic, Semantics, Metamathematics, pp. 409–20 (see n. 188 above).
196 Ibid., pp. 419 f.
197 See [1947(a)], [1947(b)], [1947(c)], [1948(b)], [1948(c)], [1948(e)], [1948(f)]. The subject has now been advanced by Lejewski. See his paper “Popper’s Theory of Formal or Deductive Interference”, in The Philosophy of Karl Popper, ed. by Paul Arthur Schilpp, pp. 632–70.
198 The mistake was connected with the rules of substitution or replacement of expressions: I had mistakenly thought that it was sufficient to formulate these rules in terms of interdeducibility, while in fact what was needed was identity (of expressions). To explain this remark: I postulated, for example, that if in a statement a, two (disjoint) subexpressions x and y are both, wherever they occur, replaced by an expression z, then the resulting expression (provided it is a statement) is interdeducible with the result of replacing first x wherever it occurs by y and then y wherever it occurs by z. What I should have postulated was that the first result is identical with the second result. I realized that this was stronger, but I mistakenly thought that the weaker rule would suffice. The interesting (and so far unpublished) conclusion to which I was led later by repairing this mistake was that there was an essential difference between propositional and functional logic: while propositional logic can be constructed as a theory of sets of statements, whose elements are partially ordered by the relation of deducibility, functional logic needs in addition a specifically morphological approach since it must refer to the subexpression of an expression, using a concept like identity (with respect to expressions). But no more is needed than the ideas of identity and subexpression; no further description especially of the shape of the expressions.
199 [1950(d)].
200 [1950(b) and (c)].
201 See Kurt Gödel, “A Remark About the Relationship Between Relativity Theory and Idealistic Philosophy”, in Albert Einstein: Philosopher-Scientist, pp. 555–62 (see n. 122 above). Gödel’s arguments were (a) philosophical, (b) based on the special theory (see esp. his n. 5), and (c) based on his new cosmological solutions of Einstein’s field equations, that is, on the possibility of closed four-dimensional orbits in a (rotating) Gödel universe, as described by him in “An Example of a New Type of Cosmological Solutions of Einstein’s Field Equations of Gravitation”, Reviews of Modern Physics, 21 (1949), 447–50. (The results (c) were challenged by S. Chandrasekhar and James P. Wright, “The Geodesics in Gödel’s Universe”, Proceedings of the National Academy of Sciences, 47 [1961], 341–47. Note however that even if Gödel’s closed orbits are not geodesics, this does not in itself constitute a refutation of Gödel’s views; for a Gödel orbit was never meant to be fully ballistic or gravitational: even that of a moon rocket is only partially so.)
202 Cp. Schilpp, ed., Albert Einstein: Philosopher-Scientist, p. 688 (see n. 122 above). Not only do I agree with Einstein, but I would even go so far as to say this. Were the existence (in the physical sense) of Gödel’s orbits a consequence of Einstein’s theory (which it is not), then this fact should be held against the theory. It would not, to be sure, be a conclusive argument: there is no such thing; and we may have to accept Gödel orbits. I think, however, that in such a case we ought to look for some alternative.
203 Harald Høffding wrote (in Den menneskelige Tanke [Copenhagen: Nordisk Forlag, 1910], p. 303; in the German translation Der menschliche Gedanke [Leipzig: O. Riesland, 1911], p.333): “Knowledge, which is to describe and explain the world for us, always itself forms part of the existing world; for this reason new entities may always emerge to be dealt with by it…. We have no knowledge going beyond experience; but at no stage are we entitled to look upon experience as complete. Thus knowledge, even at its highest, provides us with nothing more than a segment of the existing world. Every reality, we may find, is itself again a part of a wider reality.” (I owe this passage to Arne Petersen.) The best intuitive idea of this incompleteness is that of a map showing the table on which the map is being drawn, and the map as it is drawn. (See also the reply to Watkins in my Replies.)
204 See my paper [1948(d)], now [1963(a)], Chap. 16 and, more fully, [1957(i)] and [1969(k)], now [1972(a)], Chap. 5.
204a (Added 1975: See now my [1974(z2)].)
205 There is an interesting and hard-hitting article by William Kneale, “Scientific Revolution for Ever?”, The British Journal for the Philosophy of Science, 19 (1968), 27–42, in which he seems to sense something of the position outlined above, and to criticize it. In many points of detail, however, he misunderstands me; for example, on p. 36: “For if there is no truth, there cannot be any approximation to truth….” This is true. But where did I ever suggest that there is no truth? The set of true theoretical statements of physics may not be (finitely) axiomatizable; in view of Gödel’s theorem, it almost certainly is not. But the sequence of our attempts to produce better and better finite axiomatizations may well be a revolutionary sequence in which we constantly create new theoretical and mathematical means for more nearly approaching this unattainable end.
206 See C.&R. [1963(a)], p. 114 (n. 30 to Chap. 3 and text), and the third paragraph of section 19 of the present Autobiography.
207 In a letter to me of June 15, 1935, Einstein approved of my views concerning “falsifiability as the decisive property of any theory about reality”.
208 See Albert Einstein: Philosopher-Scientist, p. 674 (see n. 122 above); also relevant is Einstein’s letter on p. 29 of Schrödinger et al., Briefe zur Wellenmechanik, ed. by K. Przibram (Vienna: Springer-Verlag, 1963); in the English translation, Letters on Wave Mechanics (London: Vision, 1967), the letter appears on pp. 31 f.
209 See my paper “What is Dialectic?”, now Chap. 15 of C.&R. [1963(a)]. This is a stylistically revised form of [1940(a)], with several additional footnotes. The passage summarized here in the text is from C.&R., p. 313, first new paragraph. As shown by n. 3 of this chapter (n. 1 of [1940(a)]), I regarded this description (in which I stressed that testing a theory is part of its criticism; that is, of EE) as summarizing the scientific procedure described in L.d.F.
210 Compare with this the problems “Which comes first, the hen (H) or the egg (O)?”, and “Which comes first, the Hypothesis (H) or the Observation (O)?”, discussed on p. 47 of C.&R. [1963(a)]. See also [1949(d)], now in English as the Appendix to [1972(a)]; esp. pp. 345 f.
211 See, for example, [1968(r)], esp. pp. 36–39; [1972(a)], pp. 170–78.
212 Schrödinger defends this view as a form of idealism or panpsychism in the second part of his posthumous book, Mein Weltbild (Vienna: Zsolnay, 1961, Chap. 1, pp. 105–14); English translation, My View of the World (Cambridge: Cambridge University Press, 1964, pp. 61–67).
213 I am alluding to Winston Churchill, My Early Life (London, 1930). The arguments can be found in Chap. IX (“Education at Bangalore”), that is, on pp. 131 f. of the Keystone Library edition (1934), or the Macmillan edition (1944). I have quoted from the passage at length in section 5 of Chap. 2 of [1972(a)]; see pp. 42–43.
214 The quotation is not from memory but from the first paragraph of Chap. 6 of Erwin Schrödinger, Mind and Matter (Cambridge: Cambridge University Press, 1958), p. 88; and of Erwin Schrödinger, What Is Life? & Mind and Matter (Cambridge: Cambridge University Press, 1967; two books issued in one paperback volume), p. 166. The views which Schrödinger defended in our conversations were very similar.
215 [1956(b)].
216 Incidentally, the replacement here of “impossible” by “infinitely improbable” (perhaps a dubious replacement) would not affect the main point of these considerations; for though entropy is connected with probability, not every reference to probability brings in entropy.
217 See Mind and Matter, p. 86; or What is Life? & Mind and Matter, p. 164.
218 See Mind and Matter, or What is Life? & Mind and Matter, loc. cit. He used the wording “methodology of the physicist”, probably to dissociate himself from a methodology of physics emanating from a philosopher.
219 What is Life?, pp. 74 f.
220 Ibid., p. 78.
221 Ibid., p. 79.
222 See my [1967(b) and (h)].
223 See, for example, “Quantum Mechanics without ‘The Observer”’ [1967(k)]; “Of Clouds and Clocks” [1966(f)] ([1972(a)], Chap. 6); “Is there an Epistemological Problem of Perception?” [1968(e)]; “On the Theory of the Objective Mind” [1968(r)], “Epistemology Without a Knowing Subject” [1968(s)] (respectively Chaps. 4 and 3 of Obj. Kn. [1972(a)]; and “A Pluralist Approach to the Philosophy of History” [1969(j)].
224 Tarski has often been criticized for attributing truth to sentences: a sentence, it is said, is a mere string of words without meaning; thus it cannot be true. But Tarski speaks of “meaningful sentences”, and so this criticism, like so much philosophical criticism, is not only invalid but simply irresponsible. See Logic, Semantics, Metamathematics, p. 178 (Definition 12) and p. 156, n. 1 (see n. 188 above); and, for comments, my [1955(d)] (now an addendum to Chap. 9 of my [1972(a)] and [1959(a)], [1966(e)], and later editions, n. *1 to section 84.
225 This holds even for the validity of some very simple rules, rules whose validity has been denied on intuitive grounds by some philosophers (esp. G. E. Moore); the simplest of all these rules is: from any statement a, we may validly deduce a itself. Here the impossibility of constructing a counterexample can be shown very easily. Whether or not anybody accepts this argument is his private affair. If he does not, he is simply mistaken. See also my [1947(a)].
226 I have said things like this many times since [1934(b)], sections 27 and 29, and [1947(a)]—see [1968(s)]; ([1972(a)], Chap. 3), for example; and I have suggested that what I have called the “degree of corroboration of a hypothesis h in the light of the tests or of the evidence e”, may be interpreted as a condensed report of the past critical discussions of the hypothesis h in the light of the tests e. (Cp. nn. 156–58 to section 20 above, and text.) Thus, I wrote, for example, in L.Sc.D. [1959(a)], p. 414: “… C(h,e) can be adequately interpreted as degree of corroboration of h—or of the rationality of our belief in h, in the light of tests—only if e consists of reports of the outcome of sincere attempts to refute h…”. In other words, only a report of a discussion which is sincerely critical can be said to determine, even partially, the degree of rationality (of our belief in h). In the quoted passage (as opposed to my terminology here in the text) I used the words “degree of rationality of our belief”, which should be even clearer than “rational belief”, see also ibid., p. 407, where I explain this, and make my objectivist attitude sufficiently clear, I think (as I have done ad nauseam elsewhere). Nevertheless the quoted passage has been construed (by Professor Lakatos, “Changes in the Problem of Inductive Logic”, in Problem of Inductive Logic, ed. by Lakatos, n. 6 on pp. 412 f. [see n. 41 above]) as a symptom of the shakiness of my objectivism; and an indication that I am prone to subjectivist lapses. It is, I think, impossible to avoid all misunderstandings. I wonder how my present remarks about the insignificance of belief will be construed.
227 See esp. my [1971(i)], now Chap. 1 of [1972(a)].
228 What I have called the “fashionable view” may be traced back to J. S. Mill. For modern formulations see P. F. Strawson, Introduction to Logical Theory (London: Methuen & Co., 1952; New York: John Wiley & sons, 1952), pp. 249 f.; Nelson Goodman, Fact, Fiction, and Forecast (Cambridge, Mass.: Harvard University Press, 1955), pp. 63–66; and Rudolf Carnap, “Inductive Logic and Inductive Intuition”, in Problem of Inductive Logic, ed. by Lakatos, pp. 258–67, particularly p. 265 (see n. 41 above).
229 This seems to me a more carefully worded form of one of Carnap’s arguments; see Carnap, “Inductive Logic and Inductive Intuition”, p. 265, the passage beginning: “I think that it is not only legitimate to appeal to inductive reasoning in defending inductive reasoning, but that it is indispensable.”
230 lbid., p. 311.
231 For Carnap’s “instance confirmation” see my C.&R. [1963(a)], pp. 282 f. What Carnap calls the “instance confirmation” of a law (a universal hypothesis) is equal in fact to the degree of confirmation (or the probability) of the next instance of the law; and this approaches 1/2 or 0.99, provided the relative frequency of the observed favourable instances approaches 1/2, or 0.99, respectively. As a consequence, a law that is refuted by every second instance (or by every hundredth instance) has an instance confirmation that approaches 1/2 (or 0.99); which is absurd. I explained this first in [1934(b)], p. 191, that is [1959(a)], p. 257, long before Carnap thought of instance confirmation, in a discussion of various possibilities of attributing “probability” to a hypothesis; and I then said that this consequence was “devastating” for this idea of probability. I am puzzled by Carnap’s reply to this in Lakatos, ed., Problem of Inductive Logic, pp. 309 f. (see n. 41 above). There Carnap says about instance confirmation that its numerical value “is… an important characteristic of the law. In Popper’s example, the law which is in the average satisfied by one half of the instances, has, on the basis of my definition, not the probability 1/2, as Popper erroneously believes, but 0.” But although it does have what Carnap (and I) both call “probability 0”, it also has what Carnap calls “instance confirmation 1/2”; and this was the issue under discussion (even though I used in 1934 the term “probability” in my criticism of the function which Carnap much later called “instance confirmation”).
232 I am grateful to David Miller for pointing out to me this characteristic of all Hintikka’s systems. Jaakko Hintikka’s first paper on the subject was “Towards a Theory of Inductive Generalization”, in Logic, Methodology and Philosophy of Science, ed. by Yehoshua Bar-Hillel (Amsterdam: North-Holland Publishing Co., 1964), Vol. II, pp. 274–88. Full references can be found in Risto Hilpinen, “Rules of Acceptance and Inductive Logic”, Acta Philosophica Fennica, 21 (1968).
233 According to Carnap’s position of approximately 1949–56 (at least), inductive logic is analytically true. But if so, I cannot see how the allegedly rational degree of belief could undergo such radical changes as from 0 (strongest disbelief) to 0.7 (mild belief). According to Carnap’s latest theories “inductive intuition” operates as a court of appeal. I have given reasons to show how irresponsible and biased this court of appeal is; see my [1968(i)], esp. pp. 297–303.
234 Cp. Fact, Fiction, and Forecast, p. 65 (see n. 228 above).
235 See [1968(i)]. For my positive theory of corroboration, see the end of section 20 above, and also the end of section 33, esp. n. 243 and text.
236 See [1957(i)] and [1969(k)], now reprinted as Chap. 5 of [1972 (a)]; and [1957(l)].
237 See [1959(a)], end of section 29, and p. 315 of the translation of [1935(a)], there in Appendix *i, 2, pp. 315–17; or [1963(a)], Introduction; and see below, n. 243 and text.
238 I gave a course of lectures on this particular problem—criticism without justification—in the Institute of Advanced Studies in Vienna in 1964.
239 See esp. [1957(i)] and [1969(k)], now Chap. 5 of [1972(a)]; of [1963(a)]; and Chap. 2 of [1972(a)]. See n. 165a to my Replies.
240 See [1934(b)], p. 186; [1959(a)], p. 252 (section 79).
241 Cp. [1958(c)], [1958(f)], [1958(g)]; now Chap. 8 of [1963(a)].
242 The term “metaphysical research programme” was used in my lectures from about 1949 on, if not earlier; but it did not get into print until 1958; though it is the main topic of the last chapter of the Postscript (in galley proofs since 1957). I made the Postscript available to my colleagues, and Professor Lakatos acknowledges that what he calls “scientific research programmes” are in the tradition of what I described as “metaphysical research programmes” (“metaphysical” because nonfalsifiable). See p. 183 of his paper “Falsification and the Methodology of Scientific Research Programmes”, in Criticism and the Growth of Knowledge, ed. by Imre Lakatos and Alan Musgrave (Cambridge: Cambridge University Press, 1970).
243 Incidentally, realists believe, of course, in truth (and believers in truth believe in reality; see [1963(a)], p. 116)—they even know that there are “as many” true statements as there are false ones. (For what follows here, see also the end of the section 20, above.) Since the purpose of this volume is to further the discussion between my critics and myself, I may here perhaps refer briefly to G. J. Warnock’s review of my L.Sc.D. in Mind, 59 (1960), 99–101 (see also n. 25 to section 7 above). Here we read, on p. 100, about my views on the problem of induction: “Now Popper says emphatically that this venerable problem is insoluble…”. I am sure I have never said so, least of all emphatically, for I always flattered myself that I actually solved this problem in the book under review. Later we read, on the same page: “[Popper] wishes to claim for his own views, not that they offer a solution of Hume’s problem, but that they do not permit it to arise.” This clashes with the suggestion at the beginning of my book (esp. sections 1 and 4) that what I have called Hume’s problem of induction is one of the two fundamental problems of the theory of knowledge. Later we get quite a good version of my formulation of that problem: “how… can [we] be justified in regarding as true, or even probably true, the general statements of… a scientific theory”. My straight answer to this question was: we cannot be justified. (But we can sometimes be justified in preferring one competing theory to another; see the text to which the present note is appended.) Yet the review continues: “There is, Popper holds, no hope of answering this question, since it requires that we should solve the insoluble problem of induction. But, he says, it is needless and misguided to ask this question at all.” None of the passages I have quoted are meant to be critical; rather, they claim to report what I “say emphatically”; “wish to claim”; “hold”; and “say”. A little later in the review the criticism begins with the words: “Now does this eliminate the ‘insoluble’ problem of induction?”.
Since I am at it, I may as well mention that this reviewer concentrates his criticism of my book upon the following thesis which I am putting here in italics (p. 101; the word “rely” here means, as the context shows, “rely for the future”): “Popper evidently assumes, what of course his language implies, that we are entitled to rely [for the future] upon a well-corroborated theory”. But I have never assumed anything like this. What I assert is that a well-corroborated theory (which has been critically discussed and compared with its competitors, and which has so far “survived”) is rationally Preferable to a less well-corroborated theory; and that (short of proposing a new competing theory) we have no better way open to us than to prefer it, and act upon it, even though we know very well that it may let us down badly in some future cases. Thus I have to reject the reviewer’s criticism as based on a complete misunderstanding of my text, caused by his substitution of his own problem of induction (the traditional problem) for mine (which is very different). See now also [1971(i)], reprinted as Chap. 1 of [1972(a)].
244 See Ernst Mach, Die Prinzipien der Wärmelehre (Leipzig: Barth, 1896), p. 240; on p. 239 the term “general philosophical” is equated with “metaphysical”; and Mach hints that Robert Mayer (whom he greatly admired) was inspired by “metaphysical” intuitions.
245 See “A Note on Berkeley as Precursor of Mach” [1953(d)]; now Chap. 6 of [1963(a)].
246 See Schrödinger et al., Briefe zur Wellenmechanik, p. 32; I have used my own translations, but the letter can be found in English in the English ed., Letters on Wave Mechanics, pp. 35 f. (see n. 208 above). Einstein’s letter is dated August 9, 1939.
247 Cp. Erwin Schrödinger, “Die gegewärtige Situation in der Quantenmechanik”, Die Naturwissenschaften, 23 (1935), 807–12, 823–28, 844–49.
248 (Italics mine.) See Einstein’s letter referred to in n. 246 above, and also his very similar letter of December 22, 1950, in the same book, pp. 36 f. (translation, pp. 39 f.). (Note that Einstein takes it for granted that a probabilistic theory must be interpreted subjectively if it refers to a single case; this is an issue on which he and I disagreed from 1935 on. See [1959(a)], p. 459, and my footnote.)
249 See especially the references to Franz Exner’s views in Schrödinger, Science, Theory and Man, pp. 71, 133, 142 f. (see n. 132 above).
250 Cp. my paper “Quantum Mechanics without ‘The Observer’ ” [1967(k)], where references to my other writings in this field will be found (especially [1957(e)] and [1959(e)]).
251 Van der Waerden’s letter is dated October 19, 1968. (It is a letter in which he also criticizes me for a mistaken historical reference to Jacob Bernoulli, on p. 29 of [1967(k)].)
252 Since this is an autobiography, I might perhaps mention that in 1947 or 1948 I received a letter from Victor Kraft, writing in the name of the Faculty of Philosophy of the University of Vienna, asking whether I would be prepared to take up Schlick’s chair. I replied that I would not leave England.
253 Max Planck questioned Mach’s competence as a physicist even within Mach’s favourite field, the phenomenological theory of heat. See Max Planck, “Zur Machschen Theorie der physikalischen Erkenntnis”, Physikalische Zeitschrift, 11 (1910), 1186–90. (See also Planck’s preceding paper, “Die Einheit des physikalischen Weltbildes”, Physikalische Zeitschrift, 10 [1909], 62–75; and Mach’s reply, “Die Leitgedanken meiner wissenschaftlichen Erkenntnislehre und ihre Aufnahme durch die Zeitgenossen”, Physikalische Zeitschrift, 11 [1910], 599–606.)
254 See Josef Mayerhöfer, “Ernst Machs Berufung an die Wiener Universität, 1895”, in Symposium aus Anlass des 50. Todestages von Ernst Mach (Ernst Mach Institut, Freiburg im Breisgau, 1966), pp. 12–25. A charming (German) biography of Boltzmann is E. Broda, Ludwig Boltzmann (Vienna: Franz Deuticke, 1955).
255 See n. 256 and n. 261 below.
256 See E. Zermelo, “Über einen Satz der Dynamik und die mechanische Wärmetheorie”, Wiedemannsche Annalen (Annalen der Physik), 57 (1896), 485–94. Twenty years before Zermelo, Boltzmann’s friend Loschmidt had pointed out that by reversing all velocities in a gas the gas can be made to run backward and thus to revert to the ordered state from which it is supposed to have lapsed into disorder. This objection of Loschmidt’s is called the “reversibility objection”, while Zermelo’s is called the “recurrence objection”.
257 Paul and Tatiana Ehrenfest, “Über zwei bekannte Einwände gegen das Boltzmannsche H-Theorem”, Physikalische Zeitschrift, 8 (1907), 311–14.
258 See, for example, Max Born, Natural Philosophy of Cause and Chance (Oxford: Oxford University Press, 1949), who writes on p. 58: “Zermelo, a German mathematician, who worked on abstract problems like the theory of Cantor’s sets and transfinite numbers, ventured into physics by translating Gibbs’s work on statistical mechanics into German”. But note the dates: Zermelo criticized Boltzmann in 1896; published the translation of Gibbs whom he greatly admired in 1905; wrote his first paper on set theory in 1904, and his second only in 1908. Thus he was a physicist before he became an “abstract” mathematician.
259 Cp. Erwin Schrödinger, “Irreversibility”, Proceedings of the Royal Irish Academy, 53A (1950), 189–95.
260 See Ludwig Boltzmann, “Zu Hrn. Zermelo’s Abhandlung: ‘Über die mechanische Erklärung irreversibler Vorgänge’”, Wiedemannsche Annalen (Annalen der Physik), 60 (1897), 392–98. The gist of the passage is repeated in his Vorlesungen über Gastheorie (Leipzig: J. A. Barth, 1898), Vol. II, pp. 257 f.; again I have used my own translation, but the corresponding passage can be found in L. Boltzmann, Lectures on Gas Theory, trans. by Stephen G. Brush (Berkeley and Los Angeles: University of California Press, 1964), pp. 446 f.
261 Boltzmann’s best proof of dS/dt ≥ 0 was based upon his so-called collision integral. This represents the average effect upon a single molecule of the system of all the other molecules of the gas. My suggestion is that (a) it is not the collisions which lead to Boltzmann’s results, but the averaging as such; the time coordinate plays a part because there was no averaging before the collision, and so entropy increase seems to be the result of physical collisions. My suggestion is further that, quite apart from Boltzmann’s derivation, (b) collisions between the molecules of the gas are not decisive for an entropy increase, though the assumption of molecular disorder (which enters through the averaging) is. For assume that a gas takes up at one time one half of a box: soon it will “fill” the whole box—even if it is so rare that (practically) the only collisions are with the walls. (The walls are essential; see point (3) of [1956(g)].) I further suggest that (c) we may interpret Boltzmann’s derivation to mean that an ordered system X becomes almost certainly (that is, with probability 1) disordered upon collision with any system Y (say, the walls) which is in a state chosen at random, or more precisely, in a state not matched in every detail to the state of X. In this interpretation the theorem is of course valid. For the “reversibility objection” (see n. 256 above) would only show that for systems such as X in its disordered state there exists at least one other (“matched”) system Y which by (reverse) collision would return the system X to its ordered state. The mere mathematical existence (even in a constructive sense) of such a system Y which is “matched” to X creates no difficulty, since the probability that X should collide with a system matched to itself will be equal to zero. Thus the H-theorem, dS/dt ≥ 0, holds almost certainly for all colliding systems. (This explains why the second law holds for all closed systems.) The “recurrence objection” (see n. 256 above) is valid, but it does not mean that the probability of a recurrence—of the system’s taking up a state in which it was before—will be appreciably greater than zero for a system of any degree of complexity. Still, there are open problems. (See my series of notes in Nature, [1956(b)], [1956(g)], [1957(d)], [1958(b)], [1965(f)], [1967(b) and (h)], and my note [1957(f)] in The British Journal for the Philosophy of Science.)
262 See [1956(b)] and section 30 (on Schrödinger) above, esp. the text to nn. 215 and 216.
263 See above, section 30. I lectured on these matters to the Oxford University Science Society on October 20, 1967. In this lecture I also gave a brief criticism of Schrödinger’s influential paper “Irreversibility” (see n. 259 above); he writes there on p. 191: “I wish to reformulate the laws of… irreversibility… in such a way, that the logical contradiction [which] any derivation of those laws from reversible models seems to involve is removed once and for ever.” Schrödinger’s reformulation consists in an ingenious way (a method later called the “method of branch systems”) of introducing Boltzmannian arrows of time by a kind of operational definition; the result is Boltzmann’s. And the method, like Boltzmann’s, is too strong: it does not (as Schrödinger thinks) save Boltzmann’s derivation—that is, his physical explanation of the H-theorem; instead, it provides, rather, a (tautological) definition from which the second law follows immediately. So it makes any physical explanation of the second law redundant.
264 Die Prinzipien der Wärmelehre, p. 363 (see n. 244 above). Boltzmann is not mentioned there by name (his name appears, with a modicum of praise, on the next page) but the description of the “move” (“Zug”) is unmistakable: it really describes Boltzmann’s wavering. Mach’s attack in this chapter (“The Opposition between Mechanistic and Phenomenological Physics”), if read between the lines, is severe; and it is combined with a hint of self-congratulation and with a confident belief that the judgement of history will be on his side; as indeed it was.
265 The present section has been added here because it is, I believe, significant for an understanding of my intellectual development, or more especially, for my more recent fight against subjectivism in physics.
266 See Leo Szilard, “Über die Ausdehnung der phänomenologischen Thermodynamik auf die Schwankungserscheinungen”, Zeitschrift für Physik, 32 (1925), 753–88 and “Über die Entropieverminderung in einem thermodynamischen System bei Eingriffen intelligenter Wesen”, ibid., 53 (1929), 840–56; this second paper has been translated as “On the Decrease of Entropy in a Thermodynamic System by the Intervention of Intelligent Beings”, Behavioural Science, 9 (1964), 301–10. Szilard’s views were refined by L. Brillouin, Scientific Uncertainty and Information (New York: Academic Press, 1964). But I believe that all these views have been clearly and decisively criticized by J. D. Fast, Entropy, revised and enlarged reprint of 2d ed. (London: Macmillan, 1970), Appendix 5. I owe this reference to Troels Eggers Hansen.
267 Norbert Wiener, Cybernetics: Or Control & Communication in the Animal & the Machine (Cambridge, Mass.: M.I.T. Press, 1948), pp. 44 f., tried to marry this theory to Boltzmann’s theory; but I do not think that the spouses actually met in logical space—not even in that of Wiener’s book, where they are confined to strictly different contexts. (They could meet through the postulate that what is called consciousness is essentially growth of knowledge, that is, information increase; but I really do not wish to encourage idealistic speculation, and I greatly fear the fertility of such a marriage.) However, the subjective theory of entropy is closely connected both with Maxwell’s famous demon and with Boltzmann’s H-theorem. Max Born, for example, who believes in the original interpretation of the H-theorem, attributes to it a (partially?) subjective meaning, interpreting the collision integral and the “averaging” (both discussed in n. 261 to section 35, above) as “mixing mechanical knowledge with ignorance of detail”; this mixing of knowledge and ignorance, he says, “leads to irreversibility”. Cp. Born, Natural Philosophy of Cause and Chance, p. 59 (see n. 258 above).
268 See, for example, sections 34–39 and 43 of L.d.F. [1934(b)], [1966(e)], and of L.Sc.D. [1959(a)].
269 See esp. [1959(a)], new Appendix *xi (2), p. 444; [1966(e)], p. 399.
270 For measurement and its content-increasing (or information-increasing) function see section 34 of [1934(b)] and [1959(a)].
271 For a general criticism of thought experiments see the new Appendix *xi of my L.Sc.D. [1959(a)], esp. pp. 443 f.
272 Like the assumption that the gas consists of one molecule M, the assumption that, without expenditure of energy or negentropy, we can slide a piston from the side into the cylinder, is freely used by my opponents in their proofs of the convertibility of knowledge and negentropy. It is harmless here, and it is not really needed: See n. 274 below.
273 David Bohm, Quantum Theory (New York: Prentice-Hall, 1951), p. 608, refers to Szilard, but operates with many molecules. He does not, however, rely on Szilard’s arguments but rather on the general idea that Maxwell’s demon is incompatible with the law of entropy increase.
274 See my paper, “Irreversibility; or Entropy since 1905” [1957 (f)], a paper in which I referred especially to Einstein’s famous paper of 1905 on Brownian movement. In that paper I also criticized, among others, Szilard, though not via the thought experiment used here. I had first developed this thought experiment some time before 1957, and I lectured about it, on the same lines as in the text here, in 1962, on Professor E. L. Hill’s invitation, in the physics department of the University of Minnesota.
275 See P. K. Feyerabend, “On the Possibility of a Perpetuum Mobile of the Second Kind”, in Mind, Matter, and Method, Essays in Honor of Herbert Feigl, ed. by P. K. Feyerabend and G. Maxwell (Minneapolis: University of Minnesota Press, 1966), pp. 409–12. I should mention that the idea of building a flap into the piston (see Fig. 3 above), to avoid the awkwardness of having to slide it in from the side, is a refinement that Feyerabend made to my original analysis of Szilard’s thought experiment.
276 Samuel Butler has suffered many wrongs from the evolutionists, including a serious wrong from Charles Darwin himself who, though greatly upset by it, never put things right. They were put right, as far as possible, by Charles’s son Francis, after Butler’s death. The story, which is a bit involved, deserves to be retold. See pp. 167–219 of Nora Barlow, ed., The Autobiography of Charles Darwin (London: Collins, 1958), esp. p. 219, where references to most of the other relevant material will be found.
277 See [1945(a)], section 27; cp. [1957(g)] and later editions, esp. pp. 106–8.
278 I am alluding to Schrödinger’s remarks on evolutionary theory in Mind and Matter, especially those indicated by his phrase “Feigned Lamarckism”; see Mind and Matter, p. 26; and p. 118 of the combined reprint cited in n. 214 above.
279 The lecture [1961(j)] was delivered on October 31, 1961, and the manuscript was deposited on the same day in the Bodleian Library. It now appears in a revised version, with an addendum, as Chap. 7 of my [1972(a)].
280 See [1966(f)]; now Chap. 6 of [1972(a)].
280a See [1966(f)].
281 See section 33 above, esp. n. 242.
282 See L.Sc.D., section 67.
283 For the problem of “degrees of prediction” see F. A. Hayek, “Degrees of Explanation”, first published in 1955 and now Chap. 1 of his Studies in Philosophy, Politics and Economics (London: Routledge & Kegan Paul, 1967); see esp. n. 4 on p. 9. For Darwinism and the production of “a great variety of structures”, and for its irrefutability, see esp. p. 32.
284 Darwin’s theory of sexual selection is partly an attempt to explain falsifying instances of this theory; such things, for example, as the peacock’s tail, or the stag’s antlers. See the text before n. 286.
285 For the problem of “explanation in principle” (or “of the principle”) in contrast to “explanation in detail”, see Hayek, Philosophy, Politics and Economics, Chap. 1, esp. section VI, pp. 11–14.
286 David Lack makes this point in his fascinating book, Darwin’s Finches (Cambridge: Cambridge University Press, 1947), p. 72: “… in Darwin’s finches all the main beak differences between the species may be regarded as adaptations to difference in diet.” (Footnote references to the behaviour of birds I owe to Arne Petersen.)
287 As Lack so vividly describes it, ibid., pp. 58 f., the absence of a long tongue in the beak of a woodpeckerlike species of Darwin’s finches does not prevent this bird from excavating in trunks and branches for insects—that is, it sticks to its taste; however, due to its particular anatomical disability it has developed a skill to meet this difficulty: “Having excavated, it picks up a cactus spine or twig, one or two inches long, and holding it lengthwise in its beak, pokes it up the crack, dropping the twig to seize the insect as it emerges.” This striking behavioural trend may be a nongenetical “tradition” which has developed in that species with or without teaching among its members; it may also be a genetically entrenched behaviour pattern. That is to say, a genuine behavioural invention can take the place of an anatomic change. However this may be, this example shows how the behaviour of organisms can be a “spearhead” of evolution: a type of biological problem solving which may lead to the emergence of new forms and species.
288 See now my 1971 Addendum, “A Hopeful Behavioural Monster”, to my Spencer Lecture, Chap. 7 of [1972(a)], and Alister Hardy, The Living Stream: A Restatement of Evolution Theory and Its Relation to the Spirit of Man (London: Collins, 1965), Lecture VI.
289 This is one of the main ideas of my Spencer Lecture, now Chap. 7 of [1972(a)].
290 The theory of geographic separation or geographic speciation was first developed by Moritz Wagner in Die Darwin’sche Theorie und das Migrationsgesetz der Organismen (Leipzig: Duncker und Humblot, 1868); English translation by J. L. Laird, The Darwinian Theory and the Law of Migration of Organisms (London: Edward Stanford, 1873). See also Theodosius Dobzhansky, Genetics and the Origin of Species, 3d rev. ed. (New York: Columbia University Press, 1951), pp. 179–211.
291 See [1966(f)], pp. 20–26, esp. pp. 24 f., point (11). Now [1972 (a)], p. 244.
292 See [1970(1)], esp. pp. 5–10; [1972(a)], pp. 289–95.
292a The present and the next paragraphs of the text (and the corresponding notes) were inserted in 1975.
292b See Sir Alister Hardy, The Living Stream (cp. n. 288 above), esp. Lectures VI and VII. See also W. H. Thorpe, “The Evolutionary Significance of Habitat Selection”, The Journal of Animal Ecology, 14 (1945), 67–70.
293 After I had completed my Autobiography in 1969, I took up John Eccles’s suggestion that what I had earlier called “the third world” should be called “world 3”; see J. C. Eccles, Facing Reality (New York, Heidelberg and Berlin: Springer-Verlag, 1970). See also n. 7a above.
294 This argument for some thing’s reality—that we can take “cross bearings” which agree—is, I think, due to Winston Churchill. See p. 43 of Chap. 2 of my Obj. Kn. [1972(a)].
295 Cp. p. 15 of [1967(k)]: “… by and large I regard as excellent Landé’s suggestion to call physically real what is ‘kickable’ (and able to kick back if kicked).”
296 Take, for example, Einstein’s misunderstanding of his own requirement of covariance (first challenged by Kretschmann), which had a long history before it was finally cleared up, mainly (I think) due to the efforts of Fock and Peter Havas. The relevant papers are Erich Kretschmann, “Über den physikalischen Sinn der Relativitätspostulate, A. Einsteins neue und seine ursprüngliche Relativitätstheorie”, Annalen der Physik, 4th ser. 53 (1917), 575–614; and Einstein’s reply, “Prinzipielles zur allgemeinen Relativitätstheorie”, ibid., 55 (1918), 241–44. See also V. A. Fock, The Theory of Space, Time and Gravitation (London: Pergamon Press, 1959; 2d rev. ed., Oxford, 1964); and P. Havas, “Four-Dimensional Formulations of Newtonian Mechanics and Their Relation to Relativity” (see n. 32 above).
297 See [1968(r)], [1968(s)]; see also “A Realist View of Logic, Physics, and History” [1970(1)], and [1966(f)]. (These papers are now respectively Chaps. 4, 3, 8, 6 of [1972(a)].)
298 The talk of “substances” arises from the problem of change (“What remains constant in change?”) and from the attempt to answer what-is? questions. The old witticism that Bertrand Russell’s grandmother plagued him with—“What is mind? No matter! What is matter? Never mind!”—seems to me not only to the point but perfectly adequate. Better ask: “What does mind?”
299 The last two sentences may be regarded as containing an argument against panpsychism. The argument is, of course, inconclusive (since panpsychism is irrefutable), and it remains so even if it is strengthened by the following observation: even if we attribute conscious states to (say) all atoms, the problem of explaining the states of consciousness (such as recollection or anticipation) of higher animals remains as difficult as it was before, without this attribution. (See The Self, Chap. P3, section 19.)
300 See my papers “Language and the Body-Mind Problem” [1953(a)] and “A Note on the Body-Mind Problem” [1955(c)]; now Chaps. 12 and 13 of [1963(a)].
301 Wittgenstein (“The riddle does not exist”: Tractatus, 6.5) exaggerated the gulf between the world of describable (“sayable”) facts and the world of that which is deep and which cannot be said. There are gradations; moreover, the world of the sayable does not always lack depth. And if we think of depth, there is a gulf within those things that can be said—between a cookery book and Copernicus’s De revolutionibus—and there is a gulf within those things that cannot be said—between some piece of artistic tastelessness and a portrait by Holbein; and these gulfs may be far deeper than that between something that is sayable and something that is not. It is his facile solution of the problem of depth—the thesis “the deep is the unsayable”—which unites Wittgenstein the positivist with Wittgenstein the mystic. Incidentally, this thesis had long been traditional, especially in Vienna (and not merely among philosophers). See the quotation from Robert Reininger in L.Sc.D., n. 4 to section 30. Many positivists agreed; for example, Richard von Mises, who was a great admirer of the mystic poet Rilke.
302 David Miller suggests that I called in world 3 in order to redress the balance between worlds 1 and 2.
303 See sections 10 and 15 above.
304 After writing this I became acquainted with the second volume of Konrad Lorenz’s collected papers (Über tierisches und menschliches Verhalten, Gesammelte Abhandlungen [Munich: R. Piper & Co. Verlag, 1967], Vol. II; see esp. pp. 361 f.). In these papers Lorenz criticizes, with a reference to Erich von Holst, the view that the delimitation between the mental and the physical is also one between the higher and the lower functions of control: some comparatively primitive processes (such as a bad toothache) are intensely conscious, while some highly controlled processes (such as the elaborate interpretation of sense stimuli) are unconscious, so that their result—perception—appears to us (wrongly) as just “given”. This seems to me an important insight not to be overlooked in any theory of the body-mind problem. (On the other hand, I cannot imagine that the all-absorbing character of a bad toothache caused by a dying nerve has any biological value as a control function; and we are here interested in the hierarchical character of controls.)
305 R. W. Sperry (“The Great Cerebral Commissure”, Scientific American, 210 [1964], 42–52; and “Brain Bisection and Mechanisms of Consciousness”, in Brain and Conscious Experience, ed. by J. C. Eccles [Berlin, Heidelberg and New York: Springer-Verlag, 1966], pp. 298–313) warns us that we must not think that the separation is absolute: there is a certain amount of overspill to the other side of the brain. Nevertheless he writes, in the second paper mentioned, p. 300: “The same kind of right-left mental separation [reported about patients manipulating objects] is seen in tests involving vision. Recall that the right half of the visual field, along with the right hand, is represented together in the left hemisphere, and vice versa. Visual stimuli such as pictures, words, numbers, and geometric forms flashed on a screen directly in front of the subject and to the right side of a central fixation point, so that they are projected to the dominant speech hemispheres, are all described and reported correctly with no special difficulty. On the other hand, similar material flashed to the left half of the visual field and hence into the minor hemisphere are completely lost to the talking hemisphere. Stimuli flashed to one half field seem to have no influence whatever, in tests to date, on the perception and interpretation of stimuli presented to the other half field.”
305a (Added 1975.) See the most interesting book by A. D. De Groot, Thought and Choice in Chess (The Hague: Mouton, 1965; New York: Basic Books, 1966).
306 Wolfgang Köhler, The Place of Value in a World of Facts (New York:
Liveright, 1938). I have substituted “Values” for “Value”, to indicate my stress on pluralism.
307 See for this the end of the reply to Ernst Gombrich in my Replies.
308 Schiller’s two elegiac distichs may be translated:
Scruples of Conscience
Friends, what a pleasure to serve you! But I do so from fond inclination.
Thus no virtue is mine, and I feel deeply aggrieved.
Solution of the Problem
What can I do about this? I must teach myself to abhor you,
And, with disgust in my heart, serve you as duty commands.
309 See the Addendum “Facts, Standards, and Truth” in O. S., 4th ed. [1962(c)] and later editions, Vol. II.