Karl Popper

Joseph Agassi

Tel-Aviv University and York University, Toronto

All rights reserved by the author. This file may be downloaded and/or printed only for the purposes of individual private study. The making of multiple copies, either directly or indirectly, is forbidden.

Department of Philosophy
Tel-Aviv University
joseph agassi

On September 17, 1994, Karl Popper died at the age of 92. He was described as the official opposition of the "Vienna Circle", the philosophical club which in the inter-war period was glamorous and which espoused the then popular doctrine of logical positivism, so-called. His relations with that club were friendly-hostile, to use the term with which he liked to characterize the relations between scientific researchers. He is the last of that generation (unless it is Carl G. Hempel, who, however, sees himself as too young to belong there). The public aspect of Popper's friendly-hostile relations with his Viennese peers was unfortunately more hostile than friendly. Somehow, philosophers have managed to keep the accounts open for too long. The end of the era is the time to close accounts so as to be able to go on with the job at hand, since the intermingling of personal affairs with objective knowledge is unhealthy. Yet most of the leading heirs of the "Vienna Circle" still coast around the issues raised by Popper, and so they can neither overlook him nor quote him correctly. If past evidence is reliable here, then this will alter just about now. Is it?
Popper is known as the philosopher who offered a solution to the problem of induction without denying the validity of David Hume's classical critique of all theories of induction. The problem of induction is often put these days as, is the past a reliable indicator for the future? Or, will the future be like the past? This wording is not very fortunate, as we all agree that the past cannot be revived, that I will never be young again. The traditional way of stating the question was, are generalizations from experience reliable? This is better but still not very happy, as we all know from our childhood experience that some generalizations are most unreliable. A much better restatement of the same idea is, can we rely on the generalizations that rest on many instances and have no instance to the contrary? This is better, of course, but still not satisfactory, as there are generalizations that are erroneous yet it is not easy to find evidence that contradict them, especially those generalizations that rest on stereotypes. Moreover, science regularly suggests theories, such as about atoms and their properties, that are not generalizations at all. The best way to word the problem of induction, then, is to refer to the fact that science consists in learning from experience and developing theories: how can evidence help us learn about the world at large? How is theoretical learning form experience possible? How can evidence support theories?
This wording makes Hume's critique of all theories of inductive support unanswerable, as it is the true observation that there is no valid inference leading from particular evidence to universal theories. All efforts to find such inferences, especially the theories of probable inference, were refuted by Hume already. Popper agrees that learning from experience is not inferences from evidence to theories, that is to say, it is not inference supporting any theory, but, he says, evidence can and does lead to the negation of theories; it leads to their refutation: such inferences are classically a part of deductive logic. The novelty of Popper's idea is in not the assertion that refutations are valid inferences, but in the assertion that they are instructive, that they are indeed what makes science progressive and exciting.
This is the heart of Popper's philosophy of science: learning from experience is not by positive but by negative instances. This is a view that is hard to take seriously, as evidenced from the fact that quite a few of his self-portrayed followers deny that this is his teaching despite the fact that he has said so many times and never took it back. It must be admitted, though, that, in his discussions of the details of his view and in his response to the incredulous comments on it, he did make two minor additional assumptions. These minor assumptions do not amount to a reversal of his view from the idea that learning from experience is by refutation to the traditional view that it is by confirmation; moreover, they reduce the power of his theory even if they make it more credible to some of his audience. At the very least one ought to cite him correctly, and preferably one may discuss separately his theory once in its austere version and once with its ancillary additions.
The most prevalent misrepresentation of Popper's views sidesteps the whole issue just raised and centers round his characterization of science. For, he solved the problem of the demarcation of scientific theories in line with his solution of the problem of induction: a theory is scientific, he said, if and only if, on the supposition that it is false there is a practicable way to try to refute it. This theory was systematically misrepresented by his peers from the "Vienna Circle", especially Rudolf Carnap: whereas Popper studied the demarcation of scientific theories, Carnap alleged that what he (Popper) studied was the demarcation of the language of science. The last time I heard Carl Hempel, the leading heir of the "Vienna Circle", he repeated this misrepresentation, and he refused to accept my correction, even though he knows that Popper repeatedly complained about it. The difference between the two is very clear: the negation of a scientific theory is traditionally viewed as not scientific, whereas the language of science includes the negation of every statement that it includes. Thus the statement that is the negation of Newton's theory is not scientific both by tradition and by Popper's view, that very statement belongs to the same language to which Newton's theory belongs. Hempel refuted Carnap's variant of Popper's view and declared that Popper's view had been superseded. This kind of reasoning was never corrected by anyone except Popper and his close associates.
The views that Carnap and Hempel took as the best were those they had borrowed from Ludwig Wittgenstein. Evidently they presented their own variant of Popper's view as an improvement over the original, one that Popper himself refused to endorse out of excess pride and /or having a blind spot for the greatness of Wittgenstein's contribution. It is therefore impossible to assess Popper's contribution without reference to Wittgenstein's even though logically speaking they are worlds apart. Indeed, the difficulty of admitting the significance of Popper's contribution is that this town is too small for both of them, as they say in westerns just before a showdown. Except that there is going to be no showdown, and the central view of Wittgenstein's that Popper spent so much effort in refuting is by now admitted to be erroneous: both Wittgenstein and Popper disliked metaphysics, but whereas Wittgenstein declared it ungrammatical, Popper vehemently denied that from the start. The latest classic on the "Vienna Circle", Alberto Coffa's To the Vienna Station, though it is on the philosophy of language, and though it praises Popper much more than usual, still manages to skip entirely both the theory that all philosophy is meaningless and the unfair attribution of it to Popper. This will not do.
The assessments of the contribution of dead thinkers is usually a form of ancestor worship: in identifying the ancestors we identify ourselves. This is why it is rather hard to switch ancestors: it amounts to conversion, and a conversion is both an act of treason and an act of self-humiliation, of the admission of having belonged to the wrong party. But we need not take the assessments of the contributions of dead thinkers to be a form of ancestor worship. We may find it more important to think clearly than to play such social games. To the extent that this clarity depends on our clear image of the past, these assessments are made in order to enable us to think clearly, not in order to enable ghosts to lie in peace.
Popper's contributions to social and political philosophy is in line with his philosophy of science and is the more significant of the two -- both politically and intellectually. He said, democracy is the regime that allows the public to correct peacefully their mistakes, such as the election of an undesired ruler. His best known idea is that the theory of historical destiny is not scientific. He never said it is false, only that it cannot be worded in a manner clear enough to be put to the test of experience, since, to cohere with known facts it must for ever remain vague, since, notoriously, future science is in principle unpredictable (or else it will be current science) and the impact of science on society is tremendous. Of course, this is true of the doctrine in all of its generality. One may, of course, develop a doctrine concerning historical inevitability that will be testable; and then it will be empirically easily refuted by this very observation. For example, Karl Marx predicted that capital will become increasingly concentrated, and this will make the socialist revolution both easy and unavoidable. This prediction was based on the observation that industrial physical plants tend to be ever bigger, because of economies of scale. This, observes Popper, was overturned when the electric dynamo was invented. One has the choice, then, between considering Marx' theory as general or specific: the first choice renders it metaphysical and the second renders it empirically refuted.
And so there is the choice between taking Marx' theory to be scientific and refuted, or not scientific. It is not empirically confirmed, Popper adds, unless it is so worded that it can be only confirmed (if true) but not refuted (if false), so that it is then not scientific though its confirmations may give the impression that it is. Marx' followers either ignore Popper or distort his ideas. Those who ignore him, observes Ernest Gellner, do so while acknowledging that Marxism is not empirical, so that they have altered the status of the theory and are now advocating it irrationally.
Every item mentioned here can be elaborated on. Much was added to it by Popper and his followers, and more can be added to that. But the basic point is to acknowledge the little that has been said here. And the most important aspect of it has been left out of this discussion: the views of Popper may well be criticized, and some of his better followers did criticize. But the point made here has to do with historical truth rather than with philosophical truth, as the latter has to be discussed by diverse parties. Those who disagree with the historical description offered here will do the public service by explaining why. Those who agree with it should not conceal their agreement. This will clear the air and allow the debate to continue fruitfully.