|
Article Excerpt
Philosophical theories of the nature and meaning of probability can be distinguished from the mathematics of probability theory. The latter subject is not particularly controversial. However, there are different theories of the meaning of probability, and these theories involve corollary notions about under what circumstances probabilistic calculation is or is not meaningful as a guide to human conduct. The problem of uncertainty then is not strictly a mathematical problem. It involves the meaning of probability and its relation to intentional behavior.
Over the history of thought regarding the nature of probability, there has been a broad range of such conceptions (Hacking 1975). One might include at least the following categories: (i) probabilities are a priori degrees of rational belief; (ii) probabilities are subjective assessments of risk; (iii) probabilities are ex post empirical frequencies; and (iv) probabilities are real propensities, which exist in nature.
Which concepts of probability, and corollary concepts of uncertainty, are appropriate for institutional economics? At least since the 1990s, if not earlier, post Keynesian uncertainty concepts have played a role in institutional economics (e.g., Minsky 1996; Ferrari.Filho and Conceicao 2005; Dequech 1999; 2000a; 2000b; 2001; 2003a; 2003b; 2003c; 2004; 2005a; 2005b).
The following quote from a recent article in the Journal of Economic Issues captures the flavor of this influence:
As Hyman Minsky wrote, to comprehend Keynes 'it is necessary to understand his sophisticated view about uncertainty, and the importance of uncertainty in his vision of the economic process. Keynes without uncertainty is something like Hamlet without the Prince' ... For institutionalists, in a world of incomplete and imperfect information institutions are necessary to force economic agents, with limited insights, to adopt strategies characterized by conventions ... The concept of uncertainty is very important because it allows us to understand not only the instability of contemporary economies but, above all, the relevance of institutions in coordinating them. (Ferrari-Filho and Conceicao 2005: 580)
The present paper agrees with others that post Keynesian uncertainty theory provides a basis for an institutional approach to the problem of uncertainty. It is argued, however, that even when uncertainty is distinguished from probabilistic risk, most, if not all, uncertainty concepts still imply an underlying theory of probability.
John Maynard Keynes came to regard his own early theory of probability as problematic. However, there is deep controversy over whether or not he recanted his theory in favor of a subjectivist position. This controversy probably will never be resolved. As a consequence, some ambiguity remains regarding whether Keynes' theory of probability is necessary for post Keynesian uncertainty concepts. While the present paper is not a critique of Keynes' theory of probability, I argue that recent contributions in post Keynesian uncertainty theory are compatible with a new, essentially institutional, theory of probability, which retains some important features of Keynes' earlier theory but without its difficulties.
Keynes argued that all human knowledge is subjective in the sense that it is known by individuals and therefore is relative to the individual's acquaintance with facts and perception of logical relations. However, he argued that there exists among individuals an objective aspect, due to the existence of a "common intellectual and motivational constitution" (Davis 1991, 98). For Keynes, this common constitution stems both from the constitutions of our minds and from the access that individuals can gain, via logical intuition, to a domain of probabilities, which he conceives as objective logical relations. I argue that Keynes was right to think that a common constitution exists among people, but, while that commonality stems partly from natural selection, human institutions, rather than access to a domain of logical relations, can be held to account for that common constitution.
The sections of the paper are organized into two main parts. The first part of the paper briefly reviews Keynes' theory of probability and closely related concepts of uncertainty. The second part then discusses the role of institutions and an institutional approach to probability and uncertainty.
A Brief Outline of Keynes' Theory of Probability and Related Concepts of Uncertainty
In his Treatise on Probability, Keynes defines probability relations as objective logical relations between propositions, where propositions include evidence and probable conclusions (1921, 3-9). He uses the following symbols to express a probability relation. Let h = the evidential premises, a = the conclusion, and [alpha] = the degree of rational belief. Keynes then writes the probability relation: a/h = [alpha] (1921, 4, Chapter 5). Given the evidence, if the conclusion is tautological, then [alpha] = 1. If it is contradictory, then [alpha] = 0.
Keynes refers to the conclusion as a "primary proposition," whereas the probability relation, a/h, is a "secondary proposition." He also distinguishes between direct and indirect knowledge. Direct knowledge is knowledge of a proposition, which is obtained by "direct acquaintance," where direct acquaintance includes "experience, understanding, and perception" (1921, 12). According to Keynes, we can have direct knowledge of evidentiary premises and of secondary propositions. In the special case where [alpha] = 1, direct knowledge of premises gives us knowledge of the conclusion. In general, however, where < [alpha] < 1, direct knowledge of premises gives us indirect knowledge about the conclusion.
According to Keynes, such indirect knowledge is obtained by argument. The idea is that a conclusion, a, is not probable by itself. It is probable only in relation to knowledge of evidential premises, h, which is included in what Keynes refers to as the "corpus of knowledge," which an individual possesses (1921, 4). Through argument, we then pass to indirect knowledge of the conclusion, which is probable based on the evidence.
Regarding this transition from evidence to probable conclusion, Keynes argues that: "When we know something by argument this must be through direct acquaintance with some logical relation between the conclusion and the premise" (1921, 14). Elsewhere he refers to this transition as relying upon a person's "power of logical intuition" (18). Thus, while the conclusion, a, is not necessarily demonstrative (i.e., ct is not necessarily one or zero), nevertheless given the evidence, h, the logical relation itself, a/h = [alpha], is objective and can be known via direct acquaintance, due to the capacity for logical intuition.
John B. Davis argues that an important key to understanding Keynes' theory of probability is that "Keynes ... saw two dimensions to an individual's thinking--subjective and objective sides ..." (1991, 99). On the subjective side, Keynes strongly emphasizes the element of individual judgment and the uniqueness of individual experience. This aspect implies an element of relativity in all human knowledge. On the objective side, he emphasizes that among individuals there exists a "common intellectual and motivational constitution" (98). By means of "logical intuition," which is a capacity inherent in the "constitutions of our minds" (Keynes 1921, 18), different individuals might perceive these probability relations, which are objective logical relations.
Moreover, Keynes' dual conception of the individual, as having objective and subjective sides, led him to draw a "... distinction between what an individual can think and feel and what an individual ought to think and feel" (Davis 1991, 98). According to Davis, "... he [Keynes] believed value judgments could be objective in the sense of it being possible to say what an individual ought to think and feel in given circumstances ..." (102). For instance, Keynes argues that "[i]n the ordinary course of thought and argument, we are constantly assuming that knowledge of one statement, while not proving the truth of the second, yields nevertheless some ground for believing it. We assert that we ought on the evidence to prefer such and such a belief (1921, 5, emphasis in original).
In this vein, in the Treatise on Probability, Keynes writes "... the term certainty is sometimes used in a merely psychological sense to describe a state of mind without reference to the logical grounds of belief. With this sense I am not concerned" (1921, 15, emphasis in original). That is, Keynes' concern is not with the psychological certainty (or, by implication, uncertainty) of individual persons but rather with what it is rational for them to believe in a given situation:
A proposition is not probable because we think it so. When once the facts are given which determine our knowledge, what is probable or improbable in these circumstances have been fixed objectively, and is independent of our opinion. The Theory of Probability is logical, therefore, because it is concerned with the degree of belief which it is rational to entertain in given conditions, and not merely with the actual beliefs of particular individuals, which may or may not be rational. (1921, 4)
According to Keynes, different people possess different evidence, and their powers of logical intuition may differ as well. On both accounts, probability judgments are said to be subjective. However, Keynes argues that "... in the sense important to logic, probability is not subjective. It is not, that is to say, subject to human caprice" (1921, 4). Thus, while probability judgments are subjective opinions of individuals, probability relations are objective logical relations between propositions. Charles McCann explains the matter this way:
The subjective element in Keynes's interpretation of probability...
|
