Moore, J. (2000). Words Are Not Things. The Analysis of Verbal Behavior, 17(1), 143-160. doi:10.1007/BF03392961
Abstract:
On a traditional view, words are the fundamental units of verbal behavior. They are independent, autonomous things that symbolically represent or refer to other independent, autonomous things, often in some other dimension. Ascertaining what those other things are constitutes determining the meaning of a word. On a behavior-analytic view, verbal behavior is ongoing, functional operant activity occasioned by antecedent factors and reinforced by its consequences, particularly consequences that are mediated by other members of the same verbal community. Functional relations rather than structure select the response unit. The behavior-analytic point of view clarifies such important contemporary issues in psychology as (a) the role of scientific theories and explanations, (b) educational practices, and (c) equivalence classes, so that there is no risk of strengthening the traditional view that words are things that symbolically represent other things.
[Citing Place (1981a)]  [Citing Place (1981b)]  [Citing Place (1982)]  [Citing Place (1983d)]  
Citing Place (1981a) in context (citations start with an asterisk *):
Section FUNDAMENTAL UNITS OF VERBAL BEHAVIOR
Subsection The Response Unit
* To be sure, the conventional practices of a verbal community may well result in speakers' producing a series of words in sequence, or a sentence. This state of affairs may give the impression that the emergent unit of verbal behavior is the sentence. Place (1981a, 1981b, 1982, 1983) argued to this effect some years ago in a series of articles in Behaviorism. Sentences are then taken to express complete thoughts, and the fundamental unit is taken to have some logical integrity, such that verbal behavior is regarded as a logical process. The impression is mischievously deceptive as well, and readers are referred to Sundberg and Michael (1983) for a rejoinder to Place's arguments. To the extent that verbal behavior occurs in the form called sentences, those sentences are the result of the conventional reinforcing practices of the verbal community with respect to the speakers' verbal behavior, not anything about the fundamentally "logical" nature of verbal behavior per se. As Schnaitter (1999, p. 231) has recently noted, for behavior analysis logical and sequential relations in verbal behavior are on the dependent variable side, as effects of an ordering process, rather than on the independent variable side, as antecedent causes of ordering.
Citing Place (1981b) in context (citations start with an asterisk *):
see citation of place (1981a)
Citing Place (1982) in context (citations start with an asterisk *):
see citation of place (1981a)
Citing Place (1983d) in context (citations start with an asterisk *):
see citation of place (1981a)