The Verifiability of Value Judgments

Abstract

IT IS a historical fact that the advancement of the (or the Geisteswissenschaften, as the Germans say) stands no comparison with the brilliant success of the natural sciences. The reasons in terms of which this discrepancy has been explained are varied. A well-known explanation is that in terms of the peculiarity of the subject matter of the social sciences. In the spirit of Bacon's maxim, Scientia propter potentiam (which was in the French enlightenment-I forget by which philosopher or savant-expanded into Savoir pour prevoir,prevoir pourpouvoir), one is inclined to measure the achievements of a science by its ability to predict (at least this holds with respect to dynamic sciences). Now, predictability depends on the fulfilment of two essential conditions: (a) the knowledge of laws asserting that if an event of the kind A happens, then an event of the kind B will, with a finite degree of probability, happen; (b) the possibility of acquiring adequate factual knowledge of the so-called conditions-the state at a given time-of the system whose behavior is to be predicted. The subject matter of the social sciences involves at some point or other human behavior. But reflective behavior, so the apology runs, is by its very nature unpredictable. For suppose a law of behavior is proclaimed, according to which a given individual or group may be expected to act in a specified way at a specified time. Then the individuals about whose behavior a prediction is made may, by a decree of free willand perhaps motivated by the malicious design to prove the futility of a dynamics of human behavior-decide to refrain from acting the way they are expected to act. For example, if Marxists proclaim that the revolution of the proletariat and the ensuing era of universal communism is necessitated by the dialectic law of history, according to which, as the Hegelian jargon runs, every negation must be negated, and such a prediction should penetrate to capitalist ears, the capitalistic nations will presumably take measures to falsify the prediction and to prevent expropriation. The planets are no conscious agents (unless, indeed, Kepler's opinion that they are guided, in their regular courses, by spirits, should still be upheld); hence they never had a chance tostudy the laws of planetary motion; that is why astronomers are so successful in their predictions. Part of the initial conditions of human behavior, however, are motives; and human individuals may be motivated by the desire to falsify a prediction of their behavior; hence the very prediction of human behavior may, as it were, create new forces which modify the initial state upon which the prediction was based. If an analogous predicament obtained in astronomy, if, fantastic as it sounds, every prediction brought a new planet into existence, successful predictions would likewise be in principle impossible; for it would be in principle impossible to base a prediction upon a complete knowledge of initial conditions. The reason why I have outlined this way of accounting for the superiority of