Professor Reichenbach's writings have repeatedly called attention to the important rôle which probability statements play in all inquiry, and he has made amply clear that no philosophy of science can be regarded as adequate which does not square its accounts with the problems of probable inference. Recently he has brought together in convenient form many reflections on the methodology of science familiar to readers of his earlier works, and at the same time he has set himself the task of solving many well-known problems of epistemology in terms of his theory of probability. His latest book is therefore of great interest, both because of the light it throws on Professor Reichenbach's own views and because it reveals the power and limitations of one approach to the problems of science. In particular, while it does not add to the details of his theory of probability worked out elsewhere, the applications Professor Reichenbach now makes of it supply fresh clues for judging its import and adequacy. The object of the present essay, therefore, is to expound a number of his views on probability and epistemology, with a view to examining his conclusions and their relevance to the problems he aims to resolve. The discussion will try to determine whether several features of his present views do not follow from assumptions which he has not sufficiently considered; whether his logical constructions do not create new puzzles; and whether a different starting-point should not be taken if the clarification of scientific concepts and procedures, to which Professor Reichenbach's devotion is as unexcelled as it is well known, is to be successfully conducted.

The problem of what constitutes science is of considerable significance for the student of society: his work, both in its methods and its results, so far as it claims to be scientific, is regarded sceptically. Possibly in consequence he has tended recently to support a broad definition of science which identifies it with knowledge. Yet, leaving aside the difficult problems of what knowledge is, or what it is knowledge of, most of us would argue that, while knowledge may either be derived from science or lead to science, the two are not equivalent.

The present paper is concerned chiefly with the problem of scientific prediction. It aims at a factual analysis of the processes leading to prediction, and ventures an appraisal, in the light of this analysis, of some modern and unconventional theories of probability and truth. But although prediction is here chosen as the central issue of discussion, I do not wish to imply that, in its usual sense, it is the only or even the dominant issue of scientific research.

There can be no possible quarrel with the assertion that the phenomena of value are amenable to scientific treatment if the statement is taken in an obvious sense, for the act of valuation is a fact of human experience, empirically observable in the same way as any other fact. There is therefore no reason why we should not deal with certain aspects of the phenomena of value as we deal with other empirical phenomena. To deny this would be to deny scientific status to sociology, anthropology, and psychology. Sometimes one encounters such denials, but they need not be taken seriously, for they are obviously grounded on a very arbitrary and narrow definition of science.

It is the purpose of this paper to carry out a partial syntactical analysis of imperatives. Imperatives form a large body of linguistic expressions, appearing, e.g. in mathematical proofs (“Let f(x) be a continuous function!”), laws, moral injunctions, instruction, etc. For analytical purposes we distinguish between two forms of imperatives, the fiat and the directive. By a directive we mean an imperative which includes an indication of the agent who is to carry it out. For example, “Henry, don't forget to stop at the grocery!” is a directive. By a fiat we mean an imperative which includes no reference to an agent who is to carry it out. For example, “Let there be light!” is a fiat. This is a distinction made in terms of meaning. If, however, proper symbolic devices were introduced for the formalization of imperatives, it could be made in a syntactical manner. Thus we could distinguish between imperatives which possess a certain operator—the directive operator formed by putting a name within square brackets—as “[Henry] (Let it be the case that Henry does not forget to stop at the grocery)!”, and imperatives which do not possess this operator, e.g. “Let it be the case that Henry does not forget to stop at the grocery!” In the following we pay attention only to fiats. This, of course, involves a great limitation of subject matter and excludes topics of great interest.

If the history of civilization is to be an instrument for the advancement of civilization, historians must break new paths toward a wider and deeper understanding of human affairs. Despite the achievements of those who have explored the influence of technics, the power of nationalism, the rôle of mass-psychology, the economic factors in political and social life, our analyses of history are from a scientific standpoint still all too primitive. As John Dewey has said, “It is possible to study a multitude of histories, and yet permit history, the record of the transitions and transformations of human activities, to escape us. Taking history in separate doses of this country and that, we take it as a succession of isolated finalities, each one in due season giving way to another, as supernumeraries succeed one another in a march across the stage. We thus miss the fact of history and also its lesson; the diversity of institutional forms and customs which the same human nature may produce and employ.“

It is probably true that the greater number of contemporary American anthropologists feel that “theory” is a very dangerous kind of business which the careful anthropologist must be on his guard against. This statement represents, in the first instance, merely a crude induction from my experience in talking with professional anthropologists. It is, however, symptomatic that not until 1933 did a book by an American anthropologist include the word “theory” in its title. Only a single book published subsequently is explicitly given over to anthropological theory, and this avowedly concentrates upon the historical development of theories rather than upon a fresh and extended analysis of the more abstract aspects of anthropological thought. But because anthropology still painfully remembers the stomach-ache it got from the too easy generalizations of many nineteenth century “arm-chair ethnologists” is insufficient reason, it seems to me, for that almost morbid avoidance of theory which tends to produce acute indigestion from sheer bulk of unordered concrete observations. Landheer has with some justice commented:

An examination of philosophical conclusions and psychological experimentation upon the nature of the emotions raises numerous complex and controversial problems. The terms employed, viz. “the life of feeling”, “instinct”, “imagination” and “emotion” are integral to epistemology, ethics and aesthetics. In epistemology, the teleological aspect of the emotions is of importance. In ethics, the Stoics gave impetus to the demand that the emotions be controlled, a demand that reached its culmination in the Kantian formalism. In aesthetics, the acceptance of the “sensuous medium” as the material of art has implied “feeling” as its subjective counterpart. Yet, the full implications may not be drawn in these particular fields until analysis of the emotions themselves may relate these terms and give them precise and systematic definition.

When discussions of scientific method shift from natural sciences to social sciences, we tend to feel that we are turning from strict sciences to subject matters which may have achieved some measure of intellectual competence but which lack the rigor, objectivity and principles of organization found in mature science. This sense of a difference between the natural sciences and the social sciences is connected with questions of the meaning and rôle of laws. To the extent to which social and historical events have not been reduced to laws or perhaps cannot be, the disciplines which study such events are considered immature or imperfect sciences. To the extent to which social sciences do not offer predictions, they are felt to have little application to the solution of actual and immediate problems. When a bridge has to be built over a chasm, the laws of the natural sciences can be brought to bear. But when a treaty is to be drawn, where, it is asked, are the social and historical laws which can be applied to the situation?

A short time ago, D. D. Lee wrote in Philosophy of Science: “Grammar contains in crystallized form the accumulated and accumulating experience, the Weltanschauung of a people.“ He thus called our attention once more to a theme which was much discussed during the 19th century but which has been in disrepute for some time in philosophical circles. It is a theme, however, which is not without merit. More than seven years of intensive study of primitive languages have convinced me that the work started by Wilhelm von Humboldt deserves serious consideration even at present, and from the ethnological as well as from the philosophical point of view.