Many believe that the Marxist philosophy of history entails that man is not free in a sense in which it seems obvious that he is. In particular it is held to be (1) materialistic, (2) holistic, (3) economistic, and (4) fatalistic. It is claimed, in short, that since the Marxist philosophy of history has these features, man is not capable of shaping his own (social) destiny if it is true. I show for each of these features either that it does not entail what it is believed to entail or that it is not correctly attributed to the Marxist philosophy of history.

The views of linguistic analysts and objectivists are explored with regard to the question of interactionism. It is argued that the admission of a logical difference between explanation by cause and explanation by motive cannot disqualify causal explanations of human action, cannot be construed as challenging the competence of science, and cannot count against interactionism. It is also argued that objectivist programs for eliminating mentalistic concepts either implicitly admit interactionism or cannot distinguish relevantly between interactionism and parallelism.

Recent developments in social science methods have made most of the objectivism-subjectivism arguments in the philosophy of social science obsolete. Developments in experimental methods have made possible a behavioristic treatment of everything cherished as important in human action by the subjectivists; developments in computer and mathematical models have made possible a type of theory which carries out the program of the subjectivists but is not vulnerable to the arguments of the objectivists. What remains of the philosophical argument are two types of theory (for example, game theory and learning theory) which are both useful, both scientific, and frequently equivalent. Choice between them by scientists can be made on empirical grounds rather than on the grounds developed in the philosophical controversy.

The charge that the limit-frequency theory of probability is inconsistent due to incompatibility between the required features of randomness and limit convergence is inapplicable when probability sequences are taken to be empirically (i.e., extensionally) generated, as they must be on a strictly empirical conception of probability.

All past attempts to meet this charge by formulating constructive definitions of randomness that would still allow for a demonstrable limit-convergence have, in their exclusive concern with logically (i.e., intensionally) prescribed sequences, left the logic of extensional classes essentially untouched. In the light of a strict differentiation between intensional and extensional classes a generalized approach is possible under which several closely connected senses of randomness, i.e., the formal, material, restricted and unrestricted senses may be easily distinguished and related to the notion of relevance.

“Adaptation” has several meanings which have often been confused, including relations, processes, states, and intrinsic properties. It is used in comparative and historical contexts. “Adaptation” and “environment” may designate probabilistic concepts. Recognition of these points refutes arguments for the notions that: 1) all organisms are perfectly adapted; 2) organisms cannot be ill-adapted and survive or well-adapted and die; 3) adaptation is necessarily relative to the environment; 4) change in environment is necessary for evolution; 5) preadaptation implies teleology. Such notions are associated with metaphysical ideas, and may affect the thinking of biologists.