Edward Erwin

Current philosophical issues in the scientific evaluation of behavior therapy theory and outcome

Recent philosophic debates about scientific realism, it is argued, are germane to issues concerning behavior therapy theories and outcome hypotheses. In this article, I try, first, to clarify a kind of realism that is relevant in particular to behavior therapy. Second, I try to provide some support for this sort of realism.

Is Confirmation Differential

Consider the following puzzle about confirmation. A surprising empirical phenomenon, P, is the object of considerable scientific investigation. Five rival theories have been proposed to account for P; although we have excellent reasons for thinking that they exhaust the plausible options, we have no reason to prefer any one of the five to any of its competitors. All of the five provide equally good explanations of our experimental evidence and are equally plausible given our background evidence. We have good reason, then, for believing the disjunction:

How valuable are psychotherapy experiments?: The idiographic problem.

The idiographic problem is said to arise because psychotherapy experiments study groups of subjects, whereas the clinician wants useful etiological or treatment information concerning a single, unique client. One response is to place little value on psychotherapy experiments and to argue for nonexperimental methods for obtaining clinical knowledge. It is argued here that these alternative methods are defective and that the idiographic problem can be solved without renouncing experimental methods.

The logic of null hypothesis testing

In this commentary, I agree with Chows treatment of null hypothesis significance testing as a noninferential procedure. However, I dispute his reconstruction of the logic of theory corroboration. I also challenge recent criticisms of NHSTP based on power analysis and meta-analysis.

The Standing of Psychoanalysis

B. A. Farrell [1981] tries to elucidate some of the rational considerations that determine the standing and value of psychoanalysis. He is sceptical about much of the positive evidence, but he also tries to provide some support for Freudian doctrines. I examine his supporting arguments and try to show that they have serious weaknesses.

The rejection of natural science approaches to psychotherapy: language and the world.

Many postmodernists appeal to theses about the relationship between language and reality as one basis for rejecting natural science approaches to psychotherapy. The theses include one or more of the following: (a) There are multiple worlds rather than just one, (b) worlds are constructed rather than discovered, and (c) worlds are not independent of mind and language. I refer to the conjunction of these three propositions as the multiple worlds-constructionist view. After examining the most systematic defense of the multiple worlds-constructionist view. I conclude that it fails, and that each component of the view is false.

Testing Freudian Hypotheses

I will discuss three questions: (1) Are satisfactory experimental tests of Freudian theory possible? (2) What is the status of the current Freudian experimental evidence? (3) Can Freud’s theoretical or therapeutic hypotheses be confirmed without experimental tests?

The Use of Theory in Psychology

Eysenck’s nonrealist account of the role of theory in psychology is new and provocative. In what follows, I compare his view to other kinds of nonrealism in psychology and then state a realist’s response.

BEHAVIOUR THERAPY: A REPLY TO AN INCAUTIOUS DISCUSSION

M. E. Grenanders [1981] discussion of my book on behaviour therapy (Erwin [1978]) raises several issues that warrant comment. One issue concerns the use of reviewers for books that span philosophy and some other discipline. It is probably difficult to find willing reviewers competent in two fields, but the problems are likely to multiply if the reviewer is not competent in either field. It is generally not a good idea, for example, to have a book on behaviour therapy, that treats technical issues in the philosophy of science, learning theory and clinical psychology, reviewed by someone trained in English literature. Consider questions of style. Dr Grenander complains that the style of the book is too carefully hedged. Take one example. In discussing systematic desensitization (Erwin, p. 11), I say that there is still reason to be cautious, but there is more reason to believe, on current evidence, that the technique works than to believe that it does not. To someone trained in English literature, this formulation might appear to be too careful, but to a philosopher or psychologist familiar with the current evidence, with all its ambiguities, such caution would probably seem warranted. I think it reasonable to raise the issue of reviewer selection for interdisciplinary books, but I do not want to press the point very far. A reviewers field of expertise is not always known to a journal editor, and, of course, a review should be assessed on its merits. Judging by the content of the review, however, it is also evident that Dr Grenander has little comprehension of the topics she discusses. This is a severe verdict, but before giving support for it, I want to raise another issue. In quoting or paraphrasing an authors remarks, there are minimum standards of accuracy that must be met or the result will be shockingly inept scholarship. For example, it is unacceptable: to delete a premiss from a quoted argument, give no indication that this has been done, and then criticise the argument as invalid; to delete the initial word of a sentence, capitalise the next word, fail to use ellipses, and thereby give the impression that the entire sentence is being quoted; to quote part of a sentence to show that a view is the authors while leaving out the remaining words that show it is not; to complain that a passage is baffling without telling the reader that what is being quoted are two sentence fragments, taken from two different pages, that make no sense when the intervening sentences are omitted; or to ridicule a view expressed by a partially quoted sentence when the entire original sentence expresses no such view.