TOWARD A CONSTRUCTIONAL APPROACH TO SOCIAL PROBLEMS: ETHICAL AND CONSTITUTIONAL ISSUES RAISED BY APPLIED BEHAVIOR ANALYSIS

We shall derive a moral implication from certain practices as they relate to change in only one part of the spiral. We start arbitrarily with a patient who is in anguish because he has continually gone through hell to get certain reinforcers (it is not necessary to assume awareness of the causal relations, cf. Holz and Azrin, 1961). If this is the case, the reinforcers must be critical (N.B., the Warden box in experimental psychology). And the reason the patient's requirements for them seem insatiable is because he has been deprived of them. He may have been deprived because of the cost of their delivery to the referent system (among other reasons). In this case, he may have coerced the system into delivery (when you don't deliver, see what I do). The system delivers grudgingly and does not deliver next time. The patient ups his coercion (goes beyond what he did before), and the system yields. And so on. The coercive pattern escalates because, in effect, the system is shaping escalation. Indeed, using positive reinforcers, this is how it shapes the increasingly complex behaviors it desires. The costly consequence is not potent sui generis-few reinforcers are. Deprivation has made it potent, and satiation will make it impotent. The outlay may be expensive only initially. Procedurally, this means that the referent system need not apply deprivation procedures to any other available events to make them consequential. They might best remain inconsequential. To add these additional deprivations is not only procedurally unnecessary but morally outrageous, considering the desperation already imposed on the subject.

The referent behaviors are the object of the social contract with the institution. A constructional consultant may help the institution fulfill it.

3. Change procedures: - Since this is not a treatise on the hows of intervention, I shall confine my discussion of procedures to their ethical implications.

In a conventional p.i. text, the change procedures involve frames. Typically, each frame is an uncompleted sentence, with the student required to fill in the missing phrase. The progression of frames and behavioral requirements and the bracketing entry and terminal repertoires constitute the program. Each frame is a mini-program. There is no conflict between means and ends. Both are constructional.

A program may eliminate an option and thereby make available a range of options which were not exercised hitherto because of interference by the option eliminated. This is then described as a "crippling" pattern. When freedom is increased in this manner (liberation is a better term), the procedures, though often effective, do introduce means which are incompatible with the end, as I have analyzed it. This is not necessarily bad, but I would prefer, if possible, to use a constructional programing approach throughout. The issue is more than logical consistency. While eliminative programs often proceed without coercion or punishment (e.g., Wolpe's desensitization procedures, 1958), aversion therapy is defined by use of aversive stimuli which are often intense. Accordingly, the issue is to develop an effective approach which can produce the same outcomes, in terms of symptom relief that any eliminative procedure produces, through constructional means which add to the subject's repertoire. I have some feelings about subjecting someone who is already desperate to the dense aversive delivery often imposed.68 Accordingly, over the past few years in our own work we have consistently tried to restrict ourselves to constructional analysis of problems now treated as pathologies to be eliminated.