Find the concept of descriptive
87
THE CONCEPT OF REINFORCEMENT: EXPLANATORY OR
DESCRIPTIVE?
François Tonneau
University of Guadalajara
ABSTRACT: The history of psychology has seen recurrent controversies on the circularity
of reinforcement explanations, and behavior analysts disagree among themselves as to
whether the concept of operant reinforcement is explanatory or descriptive. Some behavior
theorists argue that the concept of reinforcement is merely descriptive, whereas others
maintain that reinforcement explanations are acceptable provided extra precautions are
taken. The issue of the circularity of reinforcement also has become embroiled in a more
general problem, that of understanding what a scientific explanation is. Here I argue that
the issue of demarcating scientific explanation from description takes two forms, and that
once these two forms are distinguished most controversies vanish. Like the majority of
scientific concepts, the concept of operant reinforcement is both descriptive and
explanatory, and reinforcement explanations are never circular.
Key words: description, explanation, reinforcement, circularity
Skinner (1938) once portrayed his system as “positivistic” and as one that
“confines itself to description rather than explanation” (p. 44). The notion that
operant concepts do not provide any explanation of animal action has not fared
well, however, and a number of behavior analysts (including a later Skinner, 1953,
1957) have argued that their discipline can and should explain behavior (e.g.,
Hineline, 1990). The issue concerns more than philosophical clarification for its
own sake, and it has moved from the philosophical terrain to affect how
researchers think of operant reinforcement. Is the concept of reinforcement
descriptive, explanatory, or both? And if the concept of reinforcement is
descriptive, are not explanations in terms of reinforcement necessarily circular?
The issue of the circularity of reinforcement explanations has surfaced recurrently
in the history of psychology (e.g., Meehl, 1950; Paniagua, 1985; Schnaitter, 1978)
without reaching any clear resolution.
Among behavior analysts, Catania (1984, 1992) has attempted to defuse the
objection of circularity by denying that the concept of reinforcement is
explanatory. According to Catania, “reinforcement, stimulus control, elicitation,
and so on are not explanatory terms but rather are names of phenomena” (1984, p.
714) and the concept of reinforcement is “descriptive rather than explanatory”
(1992, p. 72). “Reinforcement” simply “names a relation between responses and
the environment; it does not explain the relation” (Catania, 1992, p. 72). If correct,
AUTHOR’S NOTE: Please address all correspondence to François Tonneau, Departamento
de Neurociencias, Universidad de Guadalajara, Sierra Mojada 950 Peatonal 3, Col.
Independencia CP 44340, Guadalajara, Jalisco, México; Email:
[ندعوك للتسجيل في المنتدى أو التعريف بنفسك لمعاينة هذا الرابط].