Thank you very much for your letters of July 15, the long one of September 25, and the note of October 28 concerning FreundlichPFinlay–Freundlich, Erwin, 1885–1964, dt.-brit. Physiker. Your detailed letter raises many interesting problems and I shall now reply to some of them. In addition I was glad to get a picture of your activities, your personal life and moods etc. I should have written you much earlier if InaPCarnap, Ina (eig. Elisabeth Maria immacul[ata] Ignatia), 1904–1964, geb. Stöger, heiratete 1933 Rudolf Carnap had not developed a backache too which makes it hard for her to type; I hope, you will appreciate this letter all the more.
It was indeed very sad news about Stebbing’sPStebbing, Susan, 1885–1943, brit. Philosophin death. We both liked her very much. And now I have a tragic story to tell you, too. In the middle of January Eva HempelPHempel, Eva, 1908–1944, geb. Ahrends, verh. mit Carl Gustav Hempel died a few days after giving birth to a son, in consequence of a second operation. Now HempelPHempel, Carl Gustav, 1905–1997, dt.-am. Philosoph, verh. mit Eva Hempel is alone with the baby, and you can imagine how terrible a blow it is for him. You know how dependent he was on EvaPHempel, Eva, 1908–1944, geb. Ahrends, verh. mit Carl Gustav Hempel. The present plan is to have the baby at home, with a housekeeper-nurse, but we do not yet know how well that will work out.
My back is still not essentially better. I expected to go this winter again to the Mayo Clinic, perhaps for an operation; but in the meantimeamean time I consulted two good specialists who came to Santa Fe’s new Army Hospital, and they both believe that it is not a ruptured intervertebral disk – as the Mayo people believed – therefore they do not consider an operation and they try instead a conservative treatment. It is still too early to say how much success it will have but there is some progress: I can walk now up to about 40 minutes as against twenty before, and I am slowly increasing it. Fortunately the Rockefeller grant has been renewed for another year – so we can stay here until September and the doctors give me hope that I shall be able to teach again at that time.
And now let us plunge into the philosophical problems.
I don’t think that it is very useful for us to discuss Russell’s bookPRussell, Bertrand, 1872–1970, brit. Philosoph much more. (By the way, I heard that he got a fellowship at Cambridge and that he plans to return to England in the summer.) I am myself very critical with respect to the book; but bybHsl. Einschub your many critical remarks about it (with which I do sometimes not quite agree and some of which I even regard as absurdly exaggerated in their attacks against harmless formulations)cHsl. Einschub der Klammern. you push me against my will into the role of a defender of the book. It is certainly not a formulation of my opinions. When you sometimes criticize books by SchlickPSchlick, Moritz, 1882–1936, dt.-öst. Philosoph, verh. mit Blanche Guy Schlick, TarskiPTarski, Alfred, 1901–1983, poln.-am. Mathematiker und Logiker or PopperPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper, that is quite useful because I agree in the basic points with these books. But that is not the case with Russell’sPRussell, Bertrand, 1872–1970, brit. Philosoph last book. I have discussed it in our correspondence only on your request. I do not know why you always come back to this book with which we both do not agree. I have explained my chief differences with RussellPRussell, Bertrand, 1872–1970, brit. Philosoph in a previous letter. 🕮{}I think it would not be very fruitful to go now again into all the details. But if you have a specific question concerning my view on some point I shall be glad to answer it.
On Semantics. If I see it correctly, you raise objections or doubts in two points:
- 1. Semantics may be alright as a mere calculus; but if it is applied to the language of empirical science, it seems doubtful or at least an open question whether it is fruitful and useful.
- 2. Some of the chief concepts of semantics are metaphysical; therefore empiricists have to reject them as meaningless.
I am not quite clear how you intend to combine these two objections; they do not seem to me to fit well together.
On (1). I am not much interested in semantics as a mere calculus, but chiefly in semantics as analysis of the language of science. I believe that it will here be very useful and important when it will be further developed and applied. I admit that today we have in this direction not more than the very first beginnings. Therefore there is not much point in discussing the usefulness. Examples for the application of semantical concepts see (I)dIm Original hsl. eingeschobene eckige Klammern., p. 61f.; every scientist uses these and similar concepts; remember e. g. that FrankPFrank, Philipp, 1884–1966, öst.-am. Physiker und Philosoph, verh. mit Hania Frank, Bruder von Josef Frank says that certain statements of physicists are “tautological”. You ask whether I could perhaps show you that you yourself use semantical concepts without noticing it. Yes, it would indeed be easy to give hundreds of examples. Here some taken from a few pages of your letter of September 25th: You speak of “contradictions” (p. 9 []); you say that WeberPWeber, Max, 1864–1920, dt. Soziologe, verh. mit Marianne Weber gives “a descriptive story of” FranklinPFranklin, Benjamin, 1706–1790, am. Politiker und Schriftsteller; you speak (p. 10 []) about “general laws”, “area of validity”, “observation statements”; you say that a certain statement “pictures the situation” (a synonym for “is true”); you say (p. 11 []) that you will “deal with” RussellPRussell, Bertrand, 1872–1970, brit. Philosoph; you say (p. 11 []) that one item “implies” another one. (In a previous letter I have referred to the examples on p. 61f. of my book; now you express again your doubts and you ask for examples without saying a word about those examples.) Whether the applications will be really fruitful or not will be shown by further work in this field. A new logico-mathematical theory (e. g. set theory, group theory) or empirical theory (e. g. Marx’PMarx, Karl, 1818–1883, dt.-brit. Philosoph und Nationalökonom conception of history, psychoanalysis) usually meets first many doubts about its fruitfulness; these doubts are justified because the majority of newly proposed ideas are not worth much. However, I think that if a new theory is free of metaphysics and is proposed by serious people, then those who have doubts about its fruitfulness should be tolerant and should wait without polemizing too much; the further development will pass judgment. I say “if free of metaphysics”; hence, above all, we have to decide question (2).
On (2). That the semantical concept of truth is not metaphysical can very easily be shown by the following translation: “The sentence ‘This tree is green’ is true” means not more and not less than “This tree is green”. (If the latter sentence does not occur in your strangely restricted language you may take instead any other sentence which you regard as meaningful.) (By the way, if the term ‘true’ were to occur only in connection with a direct quotation as in the example given, the term would indeed hardly be useful; but that is another question.) This translation shows that the concept of truth is not metaphysical but scientific. Furthermore, the translation makes it clear that the term ‘true’ is not at all meant in the sense of ‘absolutely certain’, 🕮{}‘indubitable’ or anything like that as you sometimes seem to believe. And the translation also shows that ‘true’ has nothing to do with ‘accepted’; you make time and again the mistake of demanding that I should translate my semantical sentences into sentences with the term ‘accepted’.
To your letter p. 11 [], the last two paragraphs. I am in complete agreement with your description of the scientific procedure. I should classify this as belonging to the methodology of science. I do not see what it has to do with semantics. Which assertion of mine concerning semantics seems to you to be in contradiction to your description? – Likewise p. 12 [] you seem to believe that semantics intends to propose a new conception concerning scientific procedure; that is certainly not the case.
To p. 13 []‚ “judicial proceedings”. The question is not whether you might be able by certain tricks to avoid the word ‘true’ in such proceedings. My point is rather that the word ‘true’ in its semantical sense is very frequently used in science, judicial proceedings, etc.
To p. 14 []‚ “Sentences propositionally equal”. O. K., no objection. This is a good semantical concept; I use for it the term ‘logically equivalent’ or ‘L-equivalent’. If you wish to avoid propositions as extra-linguistic entities, you may take instead classes of L-equivalent sentences (this is e. g. done by RussellPRussell, Bertrand, 1872–1970, brit. Philosoph, Inquiry p. 209). However, in this way you do not avoid semantics! – You seem to demand that we should not speak about things which we cannot see. You will hardly expect that I take this seriously?
On finitism. The best reading material is: WeylPWeyl, Hermann, 1885–1955, dt.-am. Mathematiker und Physiker, “Die heutige Erkenntnislage in der Mathematik”, Felix Kaufmann’sPKaufmann, Felix, 1895–1949, öst.-am. Philosoph book, Bridgman’sPBridgman, Percy Williams, 1882–1961, am. Physiker pamphlet on set theory, Wittgenstein’sPWittgenstein, Ludwig, 1889–1951, öst.-brit. Philosoph book, Waismann’sPWaismann, Friedrich, 1896–1959, öst.-brit. Philosoph, verh. mit Hermine Waismann “Theses” (mimeographed in Vienna). With the exception of BridgmanPBridgman, Percy Williams, 1882–1961, am. Physiker (whose pamphlet however is rather weak) these are not authors which you otherwise estimate very highly. My opinion on finitism: it contains a sound basic idea, but none of the attempts made so far is practicable for the constructioneHsl. markierte Umstellung von for the construction is practicable. of a logico-mathematical system which could be used as a basis for the language of science. – You have mentioned several times that you have objections against Russell’sPRussell, Bertrand, 1872–1970, brit. Philosoph symbol of existence; but you do not say what the objections are. Could you tell me any logician or mathematician or scientist who has made a proposal how to do in the language of science without the concept of existence and the concept of ‘every’ (which are definable by each other)?
On pluralism. I believe that in this point I am in agreement with your attitude. I say “I believe” and “attitude”, not “opinion”, because your formulation of pluralism (p. 9 []) is so vague that I am not able to see in it any clear thesis. As you formulate it now, nearly everybody would agree, including SchlickPSchlick, Moritz, 1882–1936, dt.-öst. Philosoph, verh. mit Blanche Guy Schlick, PopperPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper, RussellPRussell, Bertrand, 1872–1970, brit. Philosoph. Since pluralism seems to you very important, and in particular also the question whether I agree with you in this point, please give me a more precise formulation of what is asserted by it. – Perhaps the difference here is not a difference of opinion but of emphasis. We emphasize the importance of the task of systematization in science; you, on the other hand, emphasize the fact that the statements accepted by scientists at a certain time 🕮{}do not form a well-connected system and you point to the dangers involved in overlooking this fact. I think you are right in both points. I should be still more inclined to agree with you were it not for the impression that you exaggerate very much in laying more stress on the dangers of systematization than on its usefulness, importance and indispensability; I doubt whether your overstressed warning is fruitful. I am afraid that it may do more harm than good in the education of young scientists. Do you remember how all of us in Vienna, especially MengerPMenger, Karl, 1902–1985, öst.-am. Mathematiker, verh. mit Hilda Menger, were angry about the influence of the WittgensteinPWittgenstein, Ludwig, 1889–1951, öst.-brit. Philosoph-WaismannPWaismann, Friedrich, 1896–1959, öst.-brit. Philosoph, verh. mit Hermine Waismann ideas concerning mathematics upon the young mathematicians? We might call those ideas “defeatism in mathematics”. They were apt to discourage the students to workfHsl. Einschub. in mathematics. Later I (together with the Warsaw people and GödelPGödel, Kurt, 1906–1978, öst.-am. Mathematiker) criticized Wittgenstein’sPWittgenstein, Ludwig, 1889–1951, öst.-brit. Philosoph “defeatism in speaking about language”; it led the young people to the position: if it is not possible to speak about language in an exact way, then let’s not try to do it. Your warning against the dangers of an oversystematization, an oversimplification, and schematization is quite alright to some extent. But I have the impression that by exaggerating your warning you turn it into a new kind of defeatism. The students may think: if systematization, logicalisation, etc. in science is not advisable and even dangerous, then why take pains to eliminate contradictions and try to fit together the accepted statements more and more into a system. I do not say that this is your position, but your polemic against systematization may have this effect. – I believe, if we are careful to avoid exaggerations on both sides, we shall easily be able to come to an agreement on this point; “we” includes PopperPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper, TarskiPTarski, Alfred, 1901–1983, poln.-am. Mathematiker und Logiker, etc. For, here there is really no serious difference in opinion; it is more a difference in temperament and therefore in the direction of interests. I think, it would be best for the development of science if the people on the one side who see more the turbulent whirl of material in all its colorfulness and vagueness, and those on the other side who love nice structural schemata would not polemize against each other but rather realize that the work of both is necessary for science.
On PopperPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper. I have explained to you my view on Popper’sPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper position in my letter of Jan. 29, 1943: It seems to me good as a first approximation. I think in the question of asymmetry we have to distinguish two assertions of Popper’sPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper:
- In testing a universal law, there is an asymmetry between favorable (confirming) and unfavorable (disconfirming) cases. If somebody proposes a law hypothetically then he usually has already a number of favorable cases, and he is looking around for further cases. The asymmetry is this: an unfavorable case has a much stronger influence upon our judgment concerning the law than an additional favorable case.
- One single unfavorable case refutes the law.
Some objections might perhaps be raised against (2). But still I would think that it is acceptable as a first approximation. On the other hand, your rejection of (1) in your last letter is surprising to me. I think that all empiricists and all good scientists agree in this. Who, in your opinion, does not? (The remark on this point which you quote from a letter of Hempel’sPHempel, Carl Gustav, 1905–1997, dt.-am. Philosoph, verh. mit Eva Hempel seems to me quite right; your reply to it is incomprehensible to me.) – As to the importance of negative cases: DarwinPDarwin, Charles, 1809–1882, brit. Biologe said that he was especially interested in all phenomena which 🕮{}did not seem in accordance with his hypotheses of evolution. – You remember how our opponents called us solipsists because we rejected the thesis of realism as meaningless? We always emphasized that if a sentence is metaphysical and hence meaningless, then the same holds of its negation. Now PopperPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper makes certain assertions concerning universal laws, negative instances etc. You deny these assertions. This is your good right. But then you go on to say that Popper’sPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper assertions are metaphysical. If so, however, your counter-assertions are likewise metaphysical and meaningless. Seriously, I think it would be better if you would at long last abandon your habit of calling people who are empiricists and antimetaphysicians metaphysicians if you do not share theirgHsl. Ersetzung von your. opinions. First, it is absurd; and second, it does not help for a successful discussion.
Negativistic exaggerations. You write that you are reading just now many books by scientists concerning their theories and procedures. I suppose that we agree that when we speak about method and language of science it is important for us to see what good scientists do. Of course, we have sometimes to criticize what they do. But in cases of this kind we ought to be especially cautious. I think that in Vienna we sometimes were too rash in condemning something accepted by all or most scientists (e. g. universal sentences); today I am inclined to be more cautious and to think that if a procedure or concept is accepted by the majority of good scientists, then it may still be that it needs some modification; but it is highly improbable that it is entirely wrong. Therefore I am astonished to see how many procedures or concepts used by the overwhelming majority of good scientists are rejected by you without your offering arguments as weighty as this situation would demand. I regard the following items as examples for this:
the concept of existence (RussellPRussell, Bertrand, 1872–1970, brit. Philosoph).asymmetry of negative and positive cases (see above, PopperPPopper, Karl Raimund, 1902–1994, öst.-brit. Philosoph, verh. mit Josefine Popper, (1)).the simple semantical concepts e. g. ‘true’, ‘designation’, ‘ana- lytic’, ‘consequence’, ‘contradiction’.‘error’-‘fallibility’-terminology, used from the point of view of a behavioristic observer (your letter p. 16 []).
‘fact’.
‘unobserved facts’, ‘unexperienced facts’ (p. 16).
‘more confirmed’ (the topological concept of degree of confir- mation; I admit that the quantitative concept is more proble- matical).
‘infinite class’ (and hence, because defined by this, the con- cepts of limit, differential coefficient, velocity, acceleration, etc.).
terms referring to things which we cannot see (p. 14 []).
‘explanation’ (FeiglPFeigl, Herbert, 1902–1988, öst.-am. Philosoph, seit 1931 verh. mit Maria Feigl says that you object to this).
On Freundlich’sPFinlay–Freundlich, Erwin, 1885–1964, dt.-brit. Physiker syllabus. I have not much to comment because first the syllabus is extremely short, and then you know my old doubts about the suitability of this topic for the first volumes. But I think, it will become an interesting monograph. Since the first two vols. are chiefly methodological it might be well if he would stress this point of view as far as possible. You say that we speak about theories; I should rather say, we speak about methods.
Your new monograph. I hope to get the proofs soon. I did not see the MS because MorrisPMorris, Charles W., 1901–1979, am. Philosoph, verh. mit Trude Morris, in accordance with your request, 🕮{}gave it to the Press in order to avoid delay. I did not object to this because you wanted it this way. But I should have preferred if you had given the coeditors an opportunity to read the MS and to write comments to you before you finished it for the Press, as we have done with all other monographs.
I agree with the choice of HogbenPHogben, Lancelot, 1895–1975, brit. Zoologe und Statistiker for the biology monograph, as MorrisPMorris, Charles W., 1901–1979, am. Philosoph, verh. mit Trude Morris will have written you. When and where will his “Interglossa” appear? I am looking forward to it with great interest.
We did not know about Mrs. Waismann’sPWaismann, Hermine, 1894–1942, geb. Antscherl, verh. mit Friedrich Waismann suicide. What was the reason? Do you know more about him now?
I have heard from Switzerland that my children are well; one daughter has just married and the other is engaged. Heaven knows what the husbands will be like! The boy is alright; no news from or about friends.
From the snapshots you see how we live; from Spring to Fall I was out there on the porch (and the manuscript on probability and degree of confirmation grew well in this wonderful climate); FeiglPFeigl, Herbert, 1902–1988, öst.-am. Philosoph, seit 1931 verh. mit Maria Feigl was with us in September (his picture will follow later, & some others too)hHsl. Einschub..
To you and MiezePNeurath, Marie, 1898–1986, geb. Reidemeister, auch Reidemeisterin, Mieze, MR, Mary, dt.-brit. Pädagogin und Sozialwiss., Schwester von Kurt Reidemeister, heiratete 1941 Otto Neurath our best.
