Popper, Karl Raimund (1902–1994)
POPPER, KARL RAIMUND
Karl Raimund Popper, the Austrian philosopher of natural and social science, was born in Vienna and was a student of mathematics, physics, and philosophy at the university there. Although he was not a member of the Vienna circle of logical positivists and was in sharp disagreement with many of its doctrines, he shared most of the group's philosophical interests and was in close touch with several of its members, having a considerable influence on Rudolf Carnap. His first book, Logik der Forschung, was published in 1935 in the circle's series Schriften zur wissenschaftlichen Weltauffassung. In 1937 Popper went as senior lecturer to Canterbury University College in Christchurch, New Zealand, and remained there until his move in 1945 to a readership at the London School of Economics in the University of London. From 1949 to 1969 he was professor of logic and scientific method at the London School of Economics, and then became professor emeritus. He was knighted in 1964.
Rejection of Verifiability Theory
The foundation of Popper's wide-ranging but closely integrated philosophical reflections is the bold and original form he first gave in 1933 to the problem of demarcating science from pseudo science in general and from metaphysics in particular. The logical positivists had taken this problem to be one of distinguishing meaningful from meaningless discourse and had proposed to solve it by making empirical verifiability the necessary condition of a sentence's meaningfulness or scientific status—in their eyes one and the same thing. Popper dissented both from their formulation of the problem and from their solution. His view had always been that the important task is to distinguish empirical science from other bodies of assertions that might be confused with it: metaphysics, such traditional pseudo sciences as astrology and phrenology, and the more imposing pseudo sciences of the present age, such as the Marxist theory of history and Freudian psychoanalysis. To identify this distinction with that between sense and nonsense is, he held, to make an arbitrary verbal stipulation. It is also an unreasonable stipulation because the line between science and pseudo science is neither precise nor impermeable. Pseudo science, or "myth," as he sometimes called it, can both inspire and develop into science proper: Indeed, the general progress of human knowledge can be considered as a conversion of myth into science by its subjection to critical examination.
A crucial difficulty for the verifiability theory of meaning was David Hume's thesis that inductive generalization was logically invalid. Being unrestrictedly general, scientific theories cannot be verified by any possible accumulation of observational evidence. Moritz Schlick sought to interpret scientific theories as rules for the derivation of predictive statements from observational ones and not as statements themselves at all, but this attempt came to grief on the fact that theories can be empirically falsified by negative instances. This logical asymmetry in the relation of general statements to observations underlies Popper's view that falsifiability by observation is the criterion of the empirical and scientific character of a theory. He maintained, first, that scientific theories are not, in fact, arrived at by any sort of inductive process. The formation of a hypothesis is a creative exercise of the imagination; it is not a passive reaction to observed regularities. There is no such thing as pure observation, for observation is always selective and takes place under the guidance of some anticipatory theory. Second, even if induction were the way in which hypotheses were arrived at, it would still be wholly incapable of justifying them. As Hume showed, no collection of particular observations will verify a general statement; nor, Popper added, is such a statement partially justified or rendered probable by particular confirming instances, since many theories that are known to be false have an indefinitely large number of confirming instances.
For Popper the growth of knowledge begins with the imaginative proposal of hypotheses, a matter of individual and unpredictable insight that cannot be reduced to rule. Such a hypothesis is science rather than myth if it excludes some observable possibilities. To test a hypothesis, we apply ordinary deductive logic in order to derive singular observation statements whose falsehood would refute it. A serious and scientific test consists in a persevering search for negative, falsifying instances. Some hypotheses are more falsifiable than others; they exclude more and thus have a greater chance of being refuted. "All heavenly bodies move in ellipses" is more falsifiable than "All planets move in ellipses," since everything that refutes the second statement refutes the first but much that refutes the first does not refute the second. The more falsifiable a hypothesis, therefore, the less probable it is, and by excluding more, it says more about the world, has more empirical content. Popper goes on to show that the obscure but important concept of simplicity comes to the same thing as falsifiability and empirical content. The proper method of science is to formulate the most falsifiable hypotheses and, consequently, those that are simplest, have the greatest empirical content, and are logically the least probable. The next step is to search energetically for negative instances, to see if any of the potential falsifiers are actually true.
If a hypothesis survives continuing and serious attempts to falsify it, then it has "proved its mettle" and can be provisionally accepted. But it can never be established conclusively. The survival of attempted refutations corroborates a theory; the corroboration being greater to the degree that the theory is falsifiable. Popper's critics have fastened on this theory of corroboration as the point at which the inductive procedure he ostensibly rejects makes an implicit reappearance. Is there any real difference, they ask, between the view that a theory depends for justification on the occurrence of confirming instances and the view that it depends on the failure of falsifying ones to occur?
Furthermore, his critics claim, there is apparently an inductive inference embedded in Popper's doctrine—the inference from the fact that a theory has thus far escaped refutation to the conclusion that it will continue to do so. Popper could reasonably reply that the formal likeness between confirming and falsifying instances conceals an important difference in approach—that between those who glory in confirmations and those who ardently pursue falsifications. However, a certain disquiet about the inductivist flavor of the positive support that his theory allows a hypothesis to derive from the failure of attempted refutations is expressed in Popper's leanings toward a rather skeptical view of the status of unrefuted hypotheses: "Science is not a system of certain, or well-established, statements.… Our science is not knowledge (epistēmē): it can never claim to have attained truth, or even a substitute for it, such as probability.… We do not know: we can only guess." (The Logic of Scientific Discovery, Ch. 10, Sec. 85, p. 278).
To complete his account of the growth of scientific knowledge, Popper had to explain the empirical basis of the falsificatory operation, that is, he had to make clear the formal character of the observation statements that are logically deduced from theories. It follows from the falsifiability criterion that unrestricted existential statements of the form "There is (somewhere at some time) an X " are unempirical because however many spatiotemporal positions have been examined for the presence of an X, an infinity of further positions remains to be examined. This is not true, however, of circumscribed existential statements reporting the existence of something at a specified place and time. Popper takes the basic observation statements to be of this form, to refer to publicly observable material objects, and to be capable of being straightforwardly affirmed or denied as true or false. Such basic statements are motivated by perceptual experiences, but they do not, as they are held to in the usual empiricist tradition, describe them. They can themselves be empirically tested in the light of the further basic statements that follow from them, together with accepted scientific theories. The infinite regress that this conception involves is not a vicious one: It can be halted by a conventional assignment of truth to basic statements at any point. But this convention is not dogmatic, since it is only provisional; if the basic statements in question are challenged, they can always be exposed to empirical tests.
In his later writings Popper drew many further inferences from his initial body of ideas. One is that knowledge has no foundations or infallible sources, either in reason or the senses. He sees the rationalist and empiricist epistemologies of the modern age as united in a determination to replace one sort of authority—a sacred text or an institution—with another—a human mental capacity. Both kinds of intellectual authoritarianism hold the mistaken opinion that truth is manifest and consequently that error is a sin and its propagation the outcome of some kind of conspiracy to deceive. There is no more comprehensive critique of the quest for certainty in the work of any other modern philosopher.
A second conclusion Popper drew is that the traditional empiricist account of concept formation—essentially Hume's idea that concepts are acquired by perceiving the similarity of sets of particular impressions—is mistaken because it embodies the same inductivist error as Francis Bacon's and J. S. Mill's accounts of scientific knowledge. Resemblance is not passively stumbled upon; rather, we classify things together in the light of antecedent preconceptions and expectations. Popper rejects innate ideas strictly so called but believes that we approach the world of experience with innate propensities—in particular, with a general expectation of regularity that is biologically explicable even if not logically justifiable. The influence of Immanuel Kant is especially evident in this side of Popper's thought. In a sense the proposition that nature contains regularities is for him synthetic a priori: It is neither a logical truth nor an empirical truth (since it is unfalsifiable), but it has a kind of psychological necessity as a general feature of the active human intellect.
Popper's dissent from the usual empiricist and positivist view that private, experiential propositions constitute the empirical foundation of knowledge and his insistence on the provisional and incompletable nature of scientific theorizing together determine his attitude to the subject matter or ontological significance of scientific theory. He rejects the essentialism of the rationalist philosophy of science, which conceives the goal of inquiry to be a complete and final knowledge of the essences of things, on the grounds that no scientific theory can be completely justified and that the acceptance of a new theory creates as many problems as it solves. He is equally opposed to the instrumentalist or conventionalist doctrine of those who, like Ernst Mach, Henri Poincaré, and Pierre Duhem, take the theoretical entities of science to be logical constructions, mere symbolic conveniences to assist us in the prediction of experience. The entities of scientific theory (such as molecules and genes) are not distinguishable in nature from the medium-sized public observables (such as chairs and trees) referred to in basic statements: Both are possible objects of genuine knowledge.
A difficulty arises for Popper's falsifiability criterion from the presence in normal scientific discourse of statements about probability in the sense of frequency. No finite sequence of A 's of which none are B decisively refutes the proposition that most A 's are B. In his first book Popper put forward a modified version of Richard von Mises's view that the probability of the occurrence of a property in an unrestrictedly open class is the limit of the frequencies of its occurrence in finite segments of the open sequence, a version that made probability statements accessible to decisive empirical refutation. Since then he had argued that probability statements, although they may rest on statistical evidence, should not themselves be interpreted statistically but rather as ascribing objective propensities to natural objects.
Determinism and Value
Popper's conviction that the mind is essentially active in the acquisition of knowledge and that its progress in discovery cannot be subsumed under a law and made the subject of prediction led him far beyond the philosophy of natural science, with which his central doctrines were concerned. Scientific knowledge is a free creation; it follows that the mind is not a causal mechanism. He contended that no causal model of the most elementary acts of the mind in empirical recognition and description can be constructed, since such a model would leave out the intention to name that is essential to any real act of description. Although the pursuit of knowledge is guided by an innate propensity to expect deterministic regularity in the world, the existence of knowledge as developed by a series of unanticipatable novelties is the strongest reason for rejecting general, metaphysical determinism.
Popper's theory of mind and knowledge also has ethical implications. Judgments of value are not empirical statements but decisions or proposals. Our valuations are not determined by our natural preferences but are the outcome of autonomous acts of mind—a further link with Kant. Popper's own basic moral proposal was, however, not very Kantian. Popper was a negative utilitarian for whom the primary moral imperative is "diminish suffering."
History and Society
In The Open Society and Its Enemies (1945) and in The Poverty of Historicism (1957), Popper applies his theory of knowledge to humankind and society in the form of an attack on historicism, the doctrine that there are general laws of historical development that render the course of history inevitable and predictable. In The Open Society historicism is examined in three influential versions, those of Plato, G. W. F. Hegel, and Karl Marx. In The Poverty of Historicism, historicism is formally refuted and attributed to two oppositely mistaken views about the nature of social science. The formal objection is that since the growth of knowledge exercises a powerful influence on the course of history and itself depends on the anomalous initiatives of original scientific genius, neither the growth of knowledge nor its general historical effects can be predicted. Some historicists have been motivated by the mistaken idea that a science of society would have a general evolutionary law as its goal. This is a naturalistic error. The evolutionary process is not a lawlike regularity at all; rather, it is a loosely characterized trend whose phases exemplify the laws of genetics, for example. The historicists who have made this error are right in believing that scientific method applies to society, but they have a false idea of what scientific method is. However, among historicists there are antinaturalists who hold that ordinary scientific method does not apply to society, for which laws of a special historicist form must be found. Popper asserts that scientific method applies both to nature and to society, and in the same way—to particular isolable aspects of the whole. Social science can discover laws that make clear the unintended consequences of human action, but there can be no laws of the whole system. It follows that social reform must proceed by piecemeal social engineering, not by total revolutionary reconstructions of the social order. Popper presents the central problem of politics in a characteristically falsificationist way: The question "Who should rule?," he says, should be replaced by the question "How can institutions be devised that will minimize the risks of bad rulers?"
Philosophy and Knowledge
Popper did not believe, as do most analytic philosophers, that philosophy is sharply distinguishable from science, either in its methods—which, like science's, must be those of trial and error, conjecture and attempted refutation—or in its subject matter—which is not only language but also the world to which language refers. Furthermore, there is no uniquely correct philosophical method. Both the examination of actual language and the construction of ideal languages can contribute to the philosophical understanding of particular problems, but they are not universal keys to truth. Popper believed that if philosophy is to be of any general importance, it must stand in a close relation to the work of other disciplines. When it is isolated, as a special autonomous craft, from the general pursuit of knowledge, it degenerates into scholasticism and triviality.
See also Basic Statements; Carnap, Rudolf; Confirmation Theory; Conventionalism; Determinism in History; Duhem, Pierre Maurice Marie; Hegel, Georg Wilhelm Friedrich; Historicism; Hume, David; Induction; Kant, Immanuel; Laws, Scientific; Logic, History of: Modern Logic; Logical Positivism; Mach, Ernst; Marx, Karl; Philosophy of Science, History of; Plato; Poincaré, Jules Henri; Political Philosophy, History of; Probability and Chance; Progress, The Idea of; Schlick, Moritz; Scientific Method; Verifiability Principle.
works by popper
Logik der Forschung. Berlin: Springer, 1935. Translated by Popper, with the assistance of Julius Freed and Lan Freed, as The Logic of Scientific Discovery. New York: Basic, 1959.
The Open Society and Its Enemies. 2 vols. London: Routledge, 1945; 4th, rev. ed., with addenda, London, 1961.
The Poverty of Historicism. London: Routledge, 1957; 2nd ed., with some corrections, 1961.
Conjectures and Refutations; The Growth of Scientific Knowledge. London: Routledge, 1963. Collected essays.
"Logic without Assumptions." PAS, n.s., 47 (1946–1947): 251–292.
"New Foundations for Logic." Mind, n.s., 56 (1947): 193–235; corrections and additions, n.s., 57 (1948): 69–70.
"Indeterminism in Quantum Physics and in Classical Physics," I and II. British Journal for the Philosophy of Science l (1950–1951): 117–133; 173–195.
"On the Theory of Deduction." Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen 51 (1 and 2) (1948).
"Probability, Magic or Knowledge out of Ignorance?" Dialectica 2 (1957): 354–374.
"The Propensity Interpretation of Probability." British Journal for the Philosophy of Science 10 (1959): 25–42.
works on popper
Bunge, Mario, ed. The Critical Approach to Science and Philosophy. Glencoe, IL: Free Press, 1964. Contains 29 articles largely concerned with the whole range of Popper's views. Includes a bibliography of Popper's publications, complete up to the beginning of 1964.
Kaufmann, Walter. From Shakespeare to Existentialism. New York, 1959. Chapter 7, "The Hegel Myth and Its Method," is a sympathetic but powerful criticism of Popper's account of Hegel in The Open Society.
Levinson, Ronald B. In Defense of Plato. Cambridge, MA, 1953. This substantial critique of Plato's modern opponents gives Popper pride of place.
Neurath, Otto. "Pseudorationalismus der Falsifikation." Erkenntnis 5 (1935): 290–294.
Reichenbach, Hans. "Über Induktion und Wahrscheinlichkeit. Bemerkungen zu Karl Poppers Logik der Forschung." Erkenntnis 5 (1935).
Warnock, G. J. Review of The Logic of Scientific Discovery. Mind, n.s., 69 (1960): 99–101.
Anthony Quinton (1967)