Herbert Feigl: 1902-1988*

Herbert Feigl was born December 14, 1902 in Reichenberg, Austria, and he died in Minneapolis, Minnesota, on June 1, l988. His wife, Maria, followed him in death on February 13, 1989. He is survived by a son, Eric O. Feigl, who is professor of physiology at the University of Washington.

Feigl's parents were Jewish but not religious; indeed, his father was an outspoken atheist. His father was a weaver and later an influential textile manufacturer, who would have preferred the son to become an industrial chemist. His mother was a devotee of the arts and an amateur pianist, who would play Beethoven sonatas over and over at the young son's request. His high school interests were philosophy, science fiction, physics, and chemistry (which was his major). Moritz Schlick became his favorite philosopher after he read the latter's Allgemeine Erkenntnisslehre (General Theory of Knowledge). His chief intellectual hero was Einstein. The description of his first contact with relativity will strike a familiar chord in many philosophers of science.

I had happened, at the age of sixteen or seventeen, upon an article...on the special theory of relativity... My first reaction was the suspicion that the theory must be in error. Promptly I set about to refute it. But in trying to do that I learned a great deal of physics and mathematics and, of course, found out after a few months of diligent work just who was wrong! My attitude then changed completely and Einstein became my number one intellectual hero. (Feigl 1980, p. 2.)

Anti-semitism in German universities led him in 1922 to the University of Vienna, where in 1927 he earned a doctorate in philosophy, with a dissertation entitled Zufall und Gesetz (the full title in English: Chance and Law: An Epistemological Analysis of the Roles of Probability and Induction in the Natural Sciences). Some of the ideas in the dissertation appeared as a paper (Feigl 1930) in the positivist journal Erkenntnis, which has been translated into English (Feigl 1980, ch. 6). In 1929 his first book, Theorie und Erfahrung in der Physik (Theory and Experience in Physics), was published. Chapter 3 of the book has been translated into English (Feigl 1980, ch. 7).

While a student at Vienna Feigl became a member of the "Vienna Circle", an informal group of philosophers, mathematicians, and scientists organized by his teacher Moritz Schlick. Among its members were Rudolf Carnap, Otto Neurath, Hans Hahn, Victor Kraft, Philipp Frank, Karl Menger, Kurt Gödel, Friedrich Waismann, and Gustav Bergmann. Among its frequent visitors were Carl Hempel, Alfred Tarksi, A. J. Ayer, and Albert E. Blumberg. Several of the members (including Feigl) were in frequent contact with Ludwig Wittgentstein and Karl Popper. Under the leadership of Neurath the group would later undertake but only partly complete a monumental Encyclopedia of Unified Science (Neurath, et. al. 1938-69), whose nineteen monographs have become classics in the philosophy of science. While the Vienna Circle was in formation so was the Berlin Society for Scientific Philosophy, led by Hans Reichenbach and Richard von Mises. These two groups and their associates would become the most influential, and controversial, collection of scientists and philosophers since the encyclopedists of the French enlightenment, and would come to be called "logical positivists".

The Vienna group's principal texts were Mach's writings, Russell and Whitehead's Principia Mathematica , and Wittgenstein's Tractatus Logico-Philosophicus. Its general philosophical position was Humean empiricism in modern logical dress: statements were held to be cognitively meaningful if and only if they are (i) analytic, that is, logically true or false, or (ii) synthetic a posteriori, that is, testable in principle by observation. Synthetic a priori statements - factual statements that transcend experience - were held to be meaningless. The positivist meaning criterion generated enormous controversy; for it was widely employed to argue that ethical, theological, metaphysical, and many (if not most) philosophical statements are cognitively meaningless, and that true statements of pure mathematics are tautologies, albeit non-trivial ones.

Feigl came to the United States late in 1930 on an International Rockefeller Fellowship, and spent nine months at Harvard University, a group of whose faculty (Bridgman, Stevens, Perry, Quine, etc.) he regarded as the American equivalent of the Vienna Circle. In 1931 he and Albert E. Blumberg published a paper in the Journal of Philosophy entitled: "Logical Positivism: A New Movement in European Philosophy", apparently christening the movement and introducing it to American philosophers. Like many others, he came to prefer the name "logical empiricism", to distinguish his scientific realist position from the phenomenalism of some of the other positivists. He returned to Vienna in 1931 to marry Maria Kasper, a philosophy student, and bring her with him to the University of Iowa, at Iowa City, where he had been appointed lecturer in philosophy. He became assistant professor at Iowa in 1933, and associate professor in 1938. Although other positivists would follow him in their flight from the Third Reich, Feigl was the first to settle in the United States. (Schlick had visited Stanford University for a term in 1930.) He thus became the new philosophy's first representative - "missionary" in Feyerabend's (1966) phrase - to the United States, a role somewhat comparable to that in Britain of A. J. Ayer, author of the widely read Language, Truth, and Logic (1936).

Feigl was awarded a second Rockefeller Fellowship to work at Harvard and Columbia Universities in 1940, and in that same year he accepted a position as professor of philosophy at the University of Minnesota. His three decades at Minnesota were highly productive, both in publications and in activities that advanced the philosophical position he represented; and they were important factors in the formation of philosophy of science as an academic specialty. In 1949 he and his colleague Wilfrid Sellars edited Readings in Philosophical Analysis , which became a standard text of analytic philosophy and logical empiricism, and after enduring success was published in a new edition twenty three years later. In that same year he and Sellars, with colleagues May Brodbeck, John Hospers, and Paul Meehl as co-editors, founded the journal Philosophical Studies. In 1953 he and Brodbeck edited Readings in the Philosophy of Science , which immediately became and for many years remained the standard anthology in the field.

In 1953 Feigl obtained a grant from the Hill Foundation (later supplemented by grants from the Carnegie Corporation and the National Science Foundation) to establish the Minnesota Center for Philosophy of Science. It was the first institution of its kind in the country, and perhaps first in the world. Philosophers of science throughout the world were brought to the Center to participate in workshops and collaborative research. Many of the fruits of this research were published by the University of Minnesota Press in the series, Minnesota Studies in the Philosophy of Science , whose first three volumes (1956, 1958, 1962) contained seminal work by many of the world's leading philosophers of science, and furnished the Center with a global reputation. The second of these volumes contained Feigl's essay, "The Mental and the Physical", which served as a stimulus to Center research on the mind-body problem and philosophy of psychology generally, and in 1967 was expanded into a book under the same title. He believed that history of science was important to philosophy of science, and he brought historian Roger Stuewer to the University of Minnesota in 1967. In 1972 Stuewer founded the University's program in history of science and technology. On his retirement in 1971 Feigl was succeeded as director of the Center by his associate and former student, Grover Maxwell. Maxwell served as director until his death in 1981. The Center has expanded, and continues to produce and subsidize research in the philosophy of science and to publish Minnesota Studies in the Philosophy of Science.

Feigl was a frequent organizer of formal and informal conferences, and was a member of several professional associations. In 1959 he and Maxwell organized a prototypical session on philosophy of science for the American Association for the Advancement of Science (of which he was then vice-president), and the proceedings were published two years later as Current Issues in the Philosophy of Science. He served on the governing board of the Philosophy of Science Association, and was a founding member (1934) of the editorial board of Philosophy of Science , which later became the official journal of the Association. He was president of the Western Division of the American Philosophical Association during 1961-62.

Feigl was among the University's most successful teachers. He was unpretentious and supportive of students, and he promoted the search for truth instead of intellectual competition. His ability to explain abstruse matters in plain terms for students, colleagues, and laypersons was legendary. So was his facility with philosophical slogans and phrases, for example "The power of positivistic thinking". So also was his ability to improvise at the piano and to play the works of his favorite composers by ear. He was appointed Regents Professor of the University of Minnesota in 1967.

After his retirement in 1971, Feigl encouraged colleagues to visit him at his home in Minneapolis for philosophical conversation and his favorite drink, cranapple juice, until a longstanding disease of the nerves in his legs made the visits impossible. Nonetheless he continued to work and publish. In 1972 a new edition of his and Sellars' 1949 Readings was published. In 1974 he published two papers summarizing and defending his philosophical position (1974a, 1974b). The year 1980 saw the publication of Inquiries and Provocations , a selection of his papers with a complete bibliography, edited with his active collaboration by Robert S. Cohen. He provided the editor with a typically overmodest self-appraisal: "I'm more of a catalyst than producer of new and original ideas...with just a few exceptions perhaps " (Feigl 1980, p. ix). His eyesight began to fail; but he continued to read in specially prepared large type, and be read to. His hearing also failed, but he continued to listen to music with electronic aids, and to instruct those entrusted with his care in the virtues of the classical composers, especially the Austrian. A few months before his death from cancer he attended with great difficulty his last concert - a performance of the Second Symphony of his beloved Mahler by the Minnesota Orchestra. He had said in an autobiographical essay in 1974: "I consider truly great music the supreme achievement of the human spirit...I am inclined to think that music expresses (even more than poetry) what is inexpressible in cognitive and especially in scientific language" (Feigl 1980, p. 5). Such was the faith of one of our most ardent and judicious humanists, a thinker who, under the terms of the logical empiricism he helped fashion and propagate, believed passionately in both art and science.

Feigl began his philosophical career as a logical positivist, soon converted to logical empiricism (realist positivism), and remained faithful to the latter position to the end of his life, with a few adjustments he seems to have regarded as minor. I will discuss his philosophical contributions under four headings.

Induction and Probability. Feigl's doctoral dissertation dealt with the roles of probability and induction in the natural sciences, and the topic was a major concern for the rest of his career. He was among the first - if not the first - to argue that probability does not afford a logical solution to Hume's problem of the unjustifiability of induction, and to propose instead a pragmatic solution to the problem - a "vindication" of induction rather than a "validation". His argument was roughly as follows. The probability of an event is empirically meaningful only if it is taken to be the relative frequency of the event in some reference class. However, no directly observable, finite frequency is exactly the probability, and so we identify it with the limit of the infinite sequence of such frequencies, if one exists. Whether the infinite sequence has a limit, and if so what its limit is, can only be inferred from an observed, finite segment of the sequence; and this inference is of course an induction. Thus, the determination of probabilities presupposes induction and consequently cannot replace induction or establish its reliability.

In spite of this difficulty, Feigl felt that induction could be vindicated by the consideration that it is an essential ingredient of scientific practice, that science as we understand it would be impossible without induction. He believed that in some of his attempts at vindication he had partly anticipated the now familiar Reichenbach-Salmon pragmatic justification of induction (Feigl 1980, pp. 27, 101). Either the sequence of relative frequencies of the event converges or not. If not, no inference from the observed to the limiting frequency (probability) of the event is correct, and science is impossible. If so, then any inductive rule of inference that takes the observed frequency to be in some way representative of the limiting frequency will provide estimates that converge to the probability in the long run, which is the best science can offer and perhaps is good enough for practical purposes. (See Salmon 1967.)

Mind-Body Identity. Feigl's second life-long concern was the mind-body problem. Perhaps his best known work is his development and defense of the thesis that mental events are contingently identical with physical (neurophysiological) events. He had inherited the generally monistic view from Schlick (1918) and reinforced it with the writings of Russell. His version of the view was first presented in detail in his essay entitled "The 'Mental' and the 'Physical'" (Feigl 1958). Its central feature was that the identity of mental and physical events is a theoretical identity that must be justified like any scientific theory on the basis of how well it explains the facts. He compared it with the identification of temperature and mean squared kinetic energy in statistical thermodynamics, which is justified on the ground that it provides the simplest, best-unified theory that entails the observed data. This thesis was initially quite controversial. For the dominant mind-body views at the time were variants of behaviorism: either the operationalist, according to which mental events are completely definable in terms of behavior; or the instrumentalist, on which mental events are taken to be theoretical fictions useful in predicting behavior. However, the Zeitgeist was moving against behaviorism, and toward theoretical realism, and Feigl was soon joined in his defense of the identity theory by other philosophers - most notably, the Australian physicalist J. J. C. Smart (1959). Today the mind-body identity theory is the prevailing view among psychologists and philosophers of psychology, though it remains controversial among philosophers generally, at least as regards its contingency.

Realism. As noted above, Feigl called his general philosophical position "logical empiricism" to distinguish it from instrumentalist and operationalist versions of positivism that were commonly held by positivists in the early phases of the movement (Carnap of the Aufbau and Neurath, for example). These other versions maintain that theoretical terms are either completely definable by observation terms, or else refer to calculational devices and/or useful fictions. Feigl's position, like that of Schlick, Reichenbach, and Popper, was scientific realist: he held that the theoretical terms of scientific theories - terms such as "force", "radiation", "electron", "gene", "pain", etc. - are almost never completely definable by observation terms and yet do refer in approximately true theories to existing, "real" entities. Sometimes he called it "critical realism" to distinguish it from the "naive realism" that holds some physical properties of physical objects can be directly known ("known by acquaintance"). Sometimes he called it "structural realism" to indicate that only the "structural" features of physical entities can be known in the strict sense ("known by description"). He saw no conflict between empiricism and scientific realism.

The liberalized meaning criterion allows for hypothetico-deductive tests (i.e., confirmations and disconfirmations), for inductive interpolations and extrapolations, for analogical conceptions and inferences [for example, inferences to other minds]. Hence I think that a critically realist position is justifiable. (Feigl 1974b, p. 13.)

Empiricism. Feigl's summing up, "No Pot of Message" (Feigl 1974b) - which has been frequently quoted above - is a restatement of vintage Humean empiricism. As regards the two types of meaningful statements, he says:

I remain unconvinced by the clever arguments of Quine that are intended to show that there is no sharp line of distinction between the purely formal truths (e.g., of arithmetic) and the factual truths (e.g., of physics). I retain the distinction between the purely formal and the factual type of meaning. But I do not object if the formal type of meaning is regarded as the 'null case', or extreme lower limit of factual significance. The tautologies, i.e., the logical truths, or statements whose form - once definienda have been replaced by their definientia - boils down to logical truth do not require observations for their validation.(Feigl 1980, p. 12)

As regards the criterion of factual meaningfulness, he says:

The much debated and often revised testability criterion of factual meaningfulness seems to me useful and, even, indispensable. Unless some of the concepts appearing in our statements are connected, no matter how indirectly, with some data of immediate experience, those statements would at best have formal significance but they would be devoid of factual meaning. I think the enormous amount of debate and quibbling that concerned the meaning criterion has been largely a waste of time and energy.(Feigl 1980, pp. 12-13)

He then refers - as he often does in his later writings - to the "liberalized [meaning] criterion". He concedes that "much of traditional metaphysics and even of theology is perfectly meaningful" on this criterion, but he insists that there are no good reasons to believe that absolute space, time, and substance, or a personal god, exists. It appears that the positivist anathemas are no longer held to be meaningless, but merely (probably) false.

Feigl's reference to "the liberalized meaning criterion" may suggest that he had in mind a precise criterion of the kind that Hempel finally despaired of finding (Hempel 1950). But no such criterion is presented in any of his writings, and in a 1969 paper he seems to have wondered whether one is achievable.

...some of the early formulations [of the positivist meaning criterion] were too drastic in that they eliminated difficult questions along with nonsensical ones. This was remedied by later more circumspect and more liberal formulations. Discussion and dispute concerning the very feasibility of an adequate formulation continues. (Feigl 1969, pp.77-78).

I conjecture that although he may have continued to hope that an acceptable, precise meaning criterion would be formulated, he believed empiricism did not require it, for the following reason. Scientists in obviously successful sciences such as physics and chemistry evaluate theories by deriving observable consequences from the theory, verifying or falsifying the consequences by observation, and using the resulting data as evidence for or against the theory. Whatever the specific features of the method employed, no theory is factually meaningful unless it can be evaluated by a method of this general type. Alternatively, if the proponent of a theory - such as evolution or creationism - is unable to identify observation statements that would be evidence for the theory if confirmed and others that would be evidence against it, then that theory has no factual meaning for the proponent. This I suggest is "the liberalized meaning criterion" he deems indispensable. It is the necessary condition for empirical meaning that a working scientist might employ, not the necessary and sufficient condition of a logician.

In "Empiricism at Bay" (1974a) he discusses the modifications in empiricism that may be required. He notes that the discovery, or origin and development, of scientific theories is an important topic, but points out that empiricism is a doctrine concerning the justification of scientific theories and is unaffected by what we may learn about the origin and development of theories. He admits that the distinction between analytic and synthetic statements is difficult to apply to actual scientific theories, but he maintains that it was intended to apply to reconstructions of theories in an ideal language that are intended to throw light on the structure and testing of actual theories, though not on their origin and development. He concedes that most observation statements scientists actually employ as data are theory-laden to some degree, and that the correction of data by theory is more pervasive than some empiricists realized, but he continues to hold that observational data are the basis of all factual knowledge. He reminds us that there are many well-established empirical generalizations in the physical sciences that cannot reasonably be doubted, and that any acceptable theory must explain them. His final conclusion is that "empiricism, though in need of renovation, will remain a fruitful and adequate philosophy of science" (Feigl 1980, p. 285).

Feigl wrote these words at a time when he perceived the fortunes of empiricism to be at a low ebb. The frontispiece of Inquiries and Provocations is a 1973 photograph of him and Hempel labelled "The Last Two Empiricists". I think he was able to remain steadfast because he believed that empiricism is the only adequate philosophy for experimental science. Though he became a philosopher instead of a chemist, he never lost the perspective, and the scientific commonsense, of a practical scientist. He was, in the paradigmatic sense, a philosopher of science.

Ayer, A. J. (1936). Language, Truth, and Logic. London: Oxford University Press. Second revised edition, London : Gollancz, 1946.

Blumberg, Albert E. and Herbert Feigl (1931). "Logical Positivism, A New Movement in European Philosophy". Journal of Philosophy 28: 281-296.

Carnap, Rudolf (1928). Der Logische Aufbau der Welt. Berlin-Schlactensee: Weltkreis-Verlag. Translated by R. A. George. The Logical Structure of the World. Berkeley and Los Angeles: University of California Press, 1969.

Feigl, Herbert (1927). Zufall und Gesetz. Unpublished doctoral dissertation, University of Vienna.

___________ (1929). Theorie und Erfahrung in der Physik. Karlsruhe: G. Braun. English translation of chapter 3 in Feigl 1980, pp. 116-144.

___________ (1930). "Wahrscheinslichkeit und Erfahrung". Erkenntnis 1: 249-259. Translated as "Probability and Experience" in Feigl 1980, pp. 107-115.

___________ (1958). "The 'Mental' and the 'Physical'". In Feigl and Scriven, pp. 370-497.

___________ (1967). "The 'Mental' and the 'Physical'." The Essay and a Postscript. Minneapolis: University of Minnesota Press.

___________ (1969). "The Wiener Kreis in America". In D. Fleming and B. Bailyn (eds.). The Intellectual Migration 1930-1960. Cambridge: Harvard University Press, pp. 630-673. Reprinted in Feigl 1980, pp. 57-94.

___________ (1974a). "Empiricism at Bay?" In Robert S. Cohen and Marx W. Wartofsky (eds.). Methodological and Historical Essays in the Natural and Social Sciences. Boston Studies in the Philosophy of Science, Volume XIV. Dordrecht and Boston: Reidel, pp. 1-20. Reprinted in Feigl 1980, pp. 269-285.

___________ (1974b). "No Pot of Message". In P. Bertocci (eds.). Mid-Twentieth Century Philosophy: Personal Statements. New York: Humanities Press, pp. 120-139. Reprinted in Feigl (1980), pp. 1-20.

___________ (1980). Inquiries and Provocations: Selected Writings 1929-1974. Edited by Robert S. Cohen, Vienna Circle Collection, Volume XIV, Dordrecht and Boston: Reidel.

___________ and Wilfrid Sellars (eds.) (1949). Readings in Philosophical Analysis. New York: Appleton-Century-Crofts.

___________ and May Brodbeck, (eds.) (1953). Readings in the Philosophy of Science. New York: Appleton-Century-Crofts.

___________ and Michael Scriven (eds.) (1956). The Foundations of Science and the Concepts of Psychology and Psychoanalysis. Minnesota Studies in the Philosophy of Science, Volume 1. Minneapolis: University of Minnesota Press.

___________ , Michael Scriven, and Grover Maxwell (eds.) (1958). Concepts, Theories, and the Mind-body Problem. Minnesota Studies in the Philosophy of Science, Volume 2. Minneapolis: University of Minnesota Press.

___________ and Grover Maxwell (eds.) (1961). Current Issues in the Philosophy of Science. New York: Holt, Rinehart, and Winston.

___________ and Grover Maxwell (eds.) (1962). Scientific Explanation, Space, and Time. Minnesota Studies in the Philosophy of Science, Volume 3. Minneapolis: University of Minnesota Press.

___________ Wilfrid Sellars, and Keith Lehrer (eds.) (1972). New Readings in Philosophical Analysis. New York: Appleton-Century-Crofts.

Feyerabend, Paul K. and Grover Maxwell (eds.) (1966). Mind, Matter, and Method: Essays in Philosophy of Science in Honor of Herbert Feigl. Minneapolis: University of Minnesota Press.

Hempel, Carl G. (1950). "Problems and Changes in the Empiricist Criterion of Meaning". Revue Internationale de Philosophie 11: 41-63.

Neurath, Otto, Rudolf Carnap, and Charles Morris (eds.) (1938-69). International Encyclopedia of Unified Science , Volumes 1 and 2. Chicago: University of Chicago Press.

Popper, Karl (1935). Logik der Forschung. Vienna: J. Springer. Translated by the author. The Logic of Scientific Discovery. New York: Basic Books, 1959.

Reichenbach, Hans (1938). Experience and Prediction. Chicago: University of Chicago Press.

Russell, Bertrand (1948). Human Knowledge: Its Scope and Limits. Simon and Schuster. Salmon, Wesley (1967). The Foundations of Scientific Inference. Pittsburgh: University of Pittsburgh Press.

Schlick, Moritz (1918). Allgemeine Erkenntnislehre. Berlin: J. Springer. Second edition, 1925. Translated by Albert E. Blumberg, with an introduction by Blumberg and Feigl. General Theory of Knowledge. New York: Springer, 1974.

Smart, J. J. C. (1959). "Sensations and Brain Processes". Philosophical Review 68: 141-156.

PSA 1988, Volume 2, pp. 15-22
Copyright © 1989 by the Philosophy of Science Association.