EFFECTIVENESS AS TRUTH CRITERION IN BEHAVIOR ANALYSIS

Behavior and Philosophy, 2003 by Tourinho, Emmanuel Zagury, Neno, Simone

ABSTRACT: In this article we examine some of the relations between behavior analysis and the pragmatic philosophy of William James. We point out that the adoption of effectiveness as a truth criterion is common to both systems, which warrants a closer examination of its implications. We present the Skinnerian concepts of knowledge, science, and truth as compatible with aspects of pragmatic philosophy; at the same time they emphasize the role of social contingencies posed by the scientific verbal community in the control of scientific verbal behavior. We argue here that effectiveness, both in Skinner's radical behaviorism and in James' pragmatic philosophy, does not have the status of cither independent or sufficient criterion to validate knowledge claims. We call attention to some generic and imprecise uses of that criterion in behavior-analytic literature and propose that a more consistent usage would involve subordinating effectiveness to a coherence criterion.

Key words: radical behaviorism, pragmatism, effectiveness, prediction and control

Every theoretical system adopts or sets up criteria by which it judges different knowledge claims about its subject matter. Effectiveness has been elaborated in philosophy and science as one of these criteria, and we will examine it here with reference to B.F. Skinner's radical behaviorism and William James' pragmatism. Our goals are a) To characterize Skinner's use of effectiveness as truth criterion, relating it to James' pragmatism; b) To indicate that, in James' pragmatism, effectiveness is subordinated to prior assumptions or beliefs; c) To illustrate uses of the effectiveness criterion in behavior-analytic literature, which are inconsistent with some prior assumptions of behavior analysis; d) To argue that to be consistent with both Skinner's radical behaviorism and James' pragmatism, prior assumptions should be considered when applying effectiveness as truth criterion.

Skinner and the Use of a Pragmatic Truth Criterion

Prediction and control are cited by Skinner (e.g., Skinner, 1953/1965), and before him by Watson (1913), as the objectives of science, hence as the purpose of investigation of behavioral regularity. They are a strict specification of what Skinner means by the effectiveness criterion, through which behavior analysts differentiate between acceptable and unacceptable explanations for behavioral phenomena. Such a criterion means, in general, that explanations that assist the scientist in dealing with behavior in a productive way, or that support the expert in solving the same kind of problems efficiently, will be considered valid. Adoption of that criterion brings behavioral science closer to a pragmatic approach to the scope, logic, and status of scientific assertions (cf. Baum, 1994; Lamal, 1983; Leigland, 1999; Skinner, 1974/1993; Tourinho, 1996; Zuriff, 1980), even though it may be discussed critically (cf. Staddon, 1993, 2001).

Skinner proposes that behavioral science adopt the effectiveness criterion in The Operational Analysis of Psychological Terms (Skinner, 1945). Rejecting the empiricist criterion of agreement based on public observation adopted by "methodological behaviorists," Skinner states that:

The ultimate criterion for the goodness of a concept is not whether two people are brought into agreement but whether the scientist who uses the concept can operate successfully upon his material-all by himself if need be. What matters to Robinson Crusoe is not whether he is agreeing with himself but whether he is getting anywhere with his control over nature. (Skinner, 1945, p. 293)

Rejection of empiricism dates from the initial period of Skinner's theoretical system. References, in 1931, to Ernst Mach as the original author of the notions of "explanation and causation" (Skinner, 1931/1961a, p. 337) and of "a method of criticism. . .with respect to scientific concepts" (Skinner, 1931/1961a, p. 321), used in proposing the concept of reflex as descriptive of behavioral phenomena, suggest that Skinner was already working with functional criteria in defining his scientific project. According to Leigland (1999), "it is clear that Skinner's views on the goals of science, as well as his general view of truth, could be described as strongly pragmatic in character" (p. 483). Zuriff (1980) finds in Skinner's work instances of both a correspondence and a pragmatic theory of truth. The latter, however, is the one that could be taken as "consistent with and characteristic of Skinner's system" (p. 343). Zuriff adds:

Although a version of the correspondence theory of truth appears in Skinner's works, it is neither the dominant theory nor consistent with the basic tenets of his radical behaviorism. A theory more prominent in his work and one more congruent with his philosophy of science is, in essence, a behaviorist version of the pragmatic theory of truth promoted by James, (p. 344)

What should be observed here, and will be addressed further on, is that Skinner derives a particular notion of "explanation" from Mach: the description of functional relationships, applied in his scientific project to the study of the organism's interaction with the world around it. As Smith (1989) observed, "the Machian views of explanation and causality were directly adopted by Skinner early in his career and have continued to figure prominently in his remarks on science" (p. 271).