Behavior analysts are interested only in special types of behavior. This behavior is not that of subatomic particles or celestial bodies. Nor is it the Brownlan motion of pollen grains in water, the paths of vortexes in turbulent fluid, or the dynamics of autocatalytic chemical reactions. Although these events constitute behavioral events both within the vernacular and other sciences (Kantor, 1959), they are not considered relevant to a science of psychology. Instead, psychological enquires are confined to "the behavior of organisms" (cf. Skinner, 1938). This should not be surprising, given that it was questions about the properties of selfcontained, biologically bounded individuals (e.g., consciousness) that provided the impetus for psychological inquiry at the outset.
Interestingly, however, there exists a tension between this alleged interest in the behavior of organisms and the decided absence of the concept of the organism in the principles and core concepts of behavior analysis. For instance, the organism does not appear in the three term contingency (stimulus-response-consequence). Furthermore, behavioral principles, such as the principle of reinforcement, describe stimulus-response relations that are content free. That is, the principle of reinforcement describes a type of relation which, at the behavioral level, entails a response-consequence contingency and a resulting increase in the rate of responding followed by that consequence. It does not, however, specify any particular consequence or any particular response (Glenn & Madden, 1995). Such a relation is "a generalization that [is] spatiotemporally unrestricted [and which] refers to any entities that happen to have the appropriate characteristics" (Hull, 1984, p. 145).
Despite the foregoing, many behavior analysts continue to espouse the idea that behavior is a biological property of organisms (e.g., Staddon, 1993) and that the experimental analysis of behavior represents a "mating between behavioral philosophy and the methods of classical physiology (Marr, 1984, p. 358), implying that psychologically interesting events are defined topographically and located peripherally (e.g., rat/operandum interface) or centrally (e.g., physiological states) in organisms.
One of several problems with asserting the organocentric nature of behavior analysis is that the term "organism" is not a technical one. Even within biology, the organism has escaped precise definition. For instance, in the Penguin Dictionary of Psychology, Reber (1995) writes: "As there is no currently agreed-upon criterial set of features for determination of that which is living, there is no rigorous definition of that which qualifies as an organism" (p. 521). Reber also notes that biologists began to make progress only when they gave up trying to define the nature of living entities (1995, p. 419). In the words of Hall (1969): "there is in fact 'no such thing' as life (nor any 'such thing as matter). Life and matter become, from this point of view, conceptual artefacts or fictions useful fictions, perhaps, but also troublesome ones" (p. 376).(1) Interestingly, one behavior analyst, Marr (1996), has recently noted that inanimate dynamic systems, such as autocatalytic chemical reactions, often display the qualities of living organisms, such as the unpredictable, the irreducible, the irreversible, and the emergent. Marr suggested that scientists might bridge the conceptual chasm between the quick and the dead by using dynamical systems approaches which offer "the promise of bringing quantitative unity to an enormous range of phenomena, living or dead" (1996, p. 19).
Given that the concept of organism is of ambiguous import within the biological sciences, it behooves us to examine the utility of this concept within behavior analysis. Establishing the role and utility of the organism is important because behavior analysts are committed to explanations of behavior stripped …