Limitations of Skinner's concept of an 'operant': A theoretical note

It is argued that even if an operant can be defined formally as the "interrelationships among S^D , R , and S ^rein", specific operants require delimitations of emitted behaviour and stimuli in terms of specific sets of contingencies of reinforcement. It is argued that the class membership of operant responses can be determined neither by reference to S^D , response topography, S^R , nor the supposed interrelationships among these, the contingencies of reinforcement. Functional analysis seems to require knowledge rejected by radical behaviourists. It is argued that operant behaviouristic analyses of human behaviour presuppose a common-sense understanding of actions, which–until explicated and shown to be reducible to a technichal behavioural language–must be taken as the point of departure in these analyses.