In 1925, while I was working with neglected children, the City of Vienna founded a new institute of education, called the Pedagogic Institute. The Institute was to be linked, somewhat loosely, with the University. It was to be autonomous, but its students were to take courses at the University in addition to the courses at the Institute. Some of the University courses (such as psychology) were made compulsory by the Institute, others were left to the choice of the students. The purpose of the new Institute was to further and support the reform, then in progress, of the primary and secondary schools in Vienna, and some social workers were admitted as students; I was among them. So also were some lifelong friends of mine—Fritz Kolb, who after the Second World War served as Austrian Ambassador in Pakistan, and Robert Lammer, with both of whom I enjoyed many fascinating discussions.
This meant that after a short period as social workers we had to give up our work (without unemployment relief, or income of any kind—except, in my case, the occasional coaching of American students). But we were enthusiastic for school reform, and enthusiastic for studying—even though our experience with neglected children made some of us sceptical of the educational theories we had to swallow in huge doses. These were imported mainly from America (John Dewey) and from Germany (Georg Kerschensteiner).
From a personal and intellectual point of view the years at the Institute were most significant for me because I met my wife there. She was one of my fellow students, and was to become one of the severest judges of my work. Her part in it ever since has been at least as strenuous as my own. Indeed, without her much of it would never have been done at all.
My years in the Pedagogic Institute were years of studying, of reading and of writing—but not of publishing. They were my first years of (quite unofficial) academic teaching. Throughout these years I gave seminars for a group of fellow students. Although I did not realize it then, they were good seminars. Some of them were most informal, and took place while hiking, or skiing, or spending the day on a river island in the Danube. From my teachers at the Institute I learned very little, but I learned much from Karl Bühler, Professor of Psychology at the University. (Though students of the Pedagogic Institute went to his lectures, he did not teach at the Pedagogic Institute, or hold a position there.)
In addition to the seminars I gave classes, also quite unofficially, to prepare my fellow students for some of the countless examinations we had to sit, among which were psychology examinations set by Bühler. He told me afterwards (in the first private conversation I ever had with a university teacher) that this had been the best-prepared batch of students he had ever examined. Bühler had only recently been called to Vienna to teach psychology, and at that time was best known for his book on The Mental Development of the Child.83 He had also been one of the first Gestalt psychologists. Most important for my future development was his theory of the three levels or functions of language (already referred to in note 78): the expressive function (Kundgabefunktion), the signal or release function (Auslösefunktion), and, on a higher level, the descriptive function (Darstellungsfunktion). He explained that the two lower functions were common to human and animal languages and were always present, while the third function was characteristic of human language alone and sometimes (as in exclamations) absent even from that.
This theory became important to me for many reasons. It confirmed my view of the emptiness of the theory that art is self-expression. It led me later to the conclusion that the theory that art was “communication” (that is, release)84 was equally empty, since these two functions were trivially present in all languages, even in animal languages. It led me to a strengthening of my “objectivist” approach. And it led me—a few years later—to add to Bühler’s three functions what I called the argumentative function.85 The argumentative function of language became particularly important for me because I regarded it as the basis of all critical thought.
I was in my second year at the Pedagogic Institute when I met Professor Heinrich Gomperz, to whom Karl Polanyi had given me an introduction. Heinrich Gomperz was the son of Theodor Gomperz (author of Greek Thinkers, and a friend and translator of John Stuart Mill). Like his father, he was an excellent Greek scholar, and also greatly interested in epistemology. He was only the second professional philosopher I had met, and the first university teacher of philosophy. Previously I had met Julius Kraft (of Hanover, a distant relation of mine, and a pupil of Leonard Nelson),86 who later became a teacher of philosophy and sociology at Frankfurt; my friendship with him lasted until his death in 1960.87
Julius Kraft, like Leonard Nelson, was a non-Marxist socialist, and about half our discussions, often lasting into the small hours of the morning, were centred on my criticism of Marx. The other half were about the theory of knowledge: mainly Kant’s so-called “transcendental deduction” (which I regarded as question-begging), his solution of the antinomies, and Nelson’s “Impossibility of the Theory of Knowledge”.88 Over these we fought a hard battle, which went on from 1926 to 1956, and we did not reach anything approaching agreement until a few years before his untimely death in 1960. On Marxism we reached agreement fairly soon.
Heinrich Gomperz was always patient with me. He had the reputation of being scathing and ironical, but I never saw anything of it. He could be most witty, though, when telling stories about some of his famous colleagues, such as Brentano and Mach. He invited me from time to time to his house, and let me talk. Usually I gave him portions of manuscript to read, but he made few comments. He was never critical of what I had to say, but he very often drew my attention to related views, and to books and articles bearing on my own topic. He never indicated that he found what I said important until I gave him, some years later, the manuscript of my first book (still unpublished—see section 16 below). He then (in December, 1932) wrote me a highly appreciative letter, the first I had ever received about something I had written.
I read all his writings, which were outstanding for their historical approach: he could follow a historical problem through all its vicissitudes from Heraclitus to Husserl, and (in conversations anyway) to Otto Weininger, whom he had known personally, and regarded as almost a genius. We did not agree on psychoanalysis. At this time he believed in it, and he even wrote for Imago.
The problems I discussed with Gomperz belonged to the psychology of knowledge or of discovery; it was during this period that I was exchanging them for problems of the logic of discovery. I was reacting more and more strongly against any “psychologistic” approach, including the psychologism of Gomperz.
Gomperz himself had criticized psychologism—only to fall back into it.89 It was mainly in discussions with him that I began to stress my realism, my conviction that there is a real world, and that the problem of knowledge is the problem of how to discover this world. I became convinced that, if we want to argue about it, we cannot start from our sense experiences (or even our feelings, as his theory demanded) without falling into the traps of psychologism, idealism, positivism, phenomenalism, even solipsism—all views which I refused to take seriously. My sense of social responsibility told me that taking such problems seriously was a kind of treason of the intellectuals—and a misuse of the time we ought to be spending on real problems.
Since I had access to the psychological laboratory I conducted a few experiments, which soon convinced me that sense data, “simple” ideas or impressions, and other such things, did not exist: they were fictitious—inventions based on mistaken attempts to transfer atomism (or Aristotelian logic—see below) from physics to psychology. The proponents of Gestalt psychology held similarly critical views; but I felt that their views were insufficiently radical. I found that my views were similar to those of Oswald Külpe and his school (the Würzburger Schule), especially Bühler90 and Otto Selz.91 They had found that we do not think in images but in terms of problems and their tentative solutions. Finding that some of my results had been anticipated, especially by Otto Selz, was, I suspect, one of the minor motives of my move away from psychology.
Abandoning the psychology of discovery and of thinking, to which I had devoted years, was a lengthy process which culminated in the following insight. I found that association psychology—the psychology of Locke, Berkeley, and Hume— was merely a translation of Aristotelian subject-predicate logic into psychological terms.
Aristotelian logic deals with statements like “Men are mortal”. Here are two “terms” and a “copula” which couples or associates them. Translate this into psychological terms, and you will say that thinking consists in having the “ideas” of man and of mortality “associated”. One has only to read Locke with this in mind to see how it happened: his main assumptions are the validity of Aristotelian logic, and that it describes our subjective, psychological thought processes. But subject-predicate logic is a very primitive thing. (It may be regarded as an interpretation of a small fragment of Boolean algebra, untidily mixed up with a small fragment of naive set theory.) It is incredible that anybody should still mistake it for empirical psychology.
A further step showed me that the mechanism of translating a dubious logical doctrine into one of an allegedly empirical psychology was still at work, and had its dangers, even for such an outstanding thinker as Bühler.
For in Külpe’s Logic,92 which Bühler accepted and greatly admired, arguments were regarded as complex judgements (which is a mistake from the point of view of modern logic).93 In consequence there could be no real distinction between judging and arguing. As a further consequence the descriptive function of language (which corresponds to “judgements”) and the argumentative function amounted to the same thing; thus Bühler failed to see that they could be as clearly separated as the three functions of language which he had already distinguished.
Bühler’s expressive function could be separated from his communicative function (or signal function, or release function) because an animal or a man could express himself even if there were no “receiver” to be stimulated. The expressive and communicative functions together could be distinguished from Bühler’s descriptive function because an animal or a man could communicate fear (for example) without describing the object feared. The descriptive function (a higher function, according to Bühler, and exclusive to man) was, I then found, clearly distinguishable from the argumentative function, since there exist languages, such as maps, which are descriptive but not argumentative.94 (This, incidentally, makes the familiar analogy between maps and scientific theories a particularly unfortunate one. Theories are essentially argumentative systems of statements: their main point is that they explain deductively. Maps are non-argumentative. Of course every theory is also descriptive, like a map—just as it is, like all descriptive language, communicative, since it may make people act; and also expressive, since it is a symptom of the “state” of the communicator—which may happen to be a computer.) Thus there was a second case where a mistake in logic led to a mistake in psychology; in this particular case the psychology of linguistic dispositions and of the innate biological needs that underlie the uses and achievements of human language.
All this showed me the priority of the study of logic over the study of subjective thought processes. And it made me highly suspicious of many of the psychological theories accepted at the time. For example, I came to realize that the theory of conditioned reflex was mistaken. There is no such thing as a conditioned reflex. Pavlov’s dogs have to be interpreted as searching for invariants in the field of food acquisition (a field that is essentially “plastic”, or in other words open to exploration by trial and error) and as fabricating expectations, or anticipations, of impending events. One might call this “conditioning”; but it is not a reflex formed as a result of the learning process, it is a discovery (perhaps a mistaken one) of what to anticipate.95 Thus even the apparently empirical results of Pavlov, and the Reflexology of Bechterev,96 and most of the results of modern learning theory, turned out, in this light, to misinterpret their findings under the influence of Aristotle’s logic; for reflexology and the theory of conditioning were merely association psychology translated into neurological terms.
In 1928 I submitted a Ph.D. thesis in which, though indirectly it was the result of years of work on the psychology of thought and discovery, I finally turned away from psychology. I had left the psychological work unfinished; I had not even a fair copy of most of what I had written; and the thesis, “On the Problem of Method in the Psychology of Thinking”,97 was a kind of hasty last minute affair originally intended only as a methodological introduction to my psychological work, though now indicative of my changeover to methodology.
I felt badly about my thesis, and I have never again even glanced at it. I also felt badly about my two “rigorous” examinations (“Rigorosum” was the name of the public oral examinations for Ph.D.), one in the history of music, the other in philosophy and psychology. Bühler, who had previously examined me in psychology, did not ask me any questions in this field, but encouraged me to talk about my ideas on logic and the logic of science. Schlick examined me mainly on the history of philosophy, and I did so badly on Leibniz that I thought I had failed. I could hardly believe my ears when I was told that I had passed in both examinations with the highest grade, “einstimmig mit Auszeichnung”. I was relieved and happy, of course, but it took quite a time before I could get over the feeling that I had deserved to fail.