Fact and Fantasy in Freudian Theory (RLE: Freud)

Paul Kline

• Publisher: Taylor & Francis
• Year: 2013
• eBook ISBN: 9781317975946
• Edition: 1
• Topic: Psychology
• Subtopic: Mental Health in Psychology
• Index: Psychology

Originally published in 1972, this second edition in 1981 was fully revised and updated to cover recent developments in the field at the time. Fact and Fantasy in Freudian Theory was written to answer many questions and criticisms surrounding psychoanalysis.

How much, if any, of Freudian theory is verifiable according to the usual criteria of scientific enquiry? Much work had been carried out at the time to discover which parts of Freudian theory are verifiable and which insupportable by experiment. In this book Dr Kline surveys this vast body of work. He takes, one by one, the central postulates of Freudian psychology and discusses the experiments which have been performed to test them. He scrutinizes each test, examines its methodology and its findings and weighs up its value. For some of the theories, it will be seen, there is no evidence whatsoever; for others, on the other hand, there is impressive and sometimes incontrovertible experimental support – for example, for the theory of repression.

This work will continue to be an invaluable, highly detailed reference work for those involved with Freud's work, and a book of great interest to those concerned with the method of psychological enquiry in general.

1

Freudian theory and scientific method

Eysenck (1953a) answering his own question — ‘What is wrong with psychoanalysis?’ — claims that the answer is simple: ‘Psychoanalysis is unscientific.’ First, therefore, it is necessary to clarify the meaning of the term scientific as applied to a theory.

THE MEANING OF SCIENTIFIC

Marx (1963) in a definition that would seem agreeable to the majority of philosophers of science argues that there are three fundamental elements of all scientific-theory construction:

	1. Observations —	which must be under controlled conditions (that is eliminate the role of extraneous variables).
	2. Constructs —	which must be operational (that is have clearly specified and identifiable empirical referents).
	3. Hypotheses —	which must be testable (that is clearly disconfirmable).

It is certainly true to say that on all these criteria much of Freudian theory is open to criticism.

OBSERVATIONS

The observations on which Freudian theory rests are made by psychoanalysts during therapy. The raw data of psychoanalytic theory consist of the free associations of the patient and the reports of dreams to which further free associations are elicited. These raw data are not even recorded as spoken but are recalled later by the analyst. Thus there is no quantification, no control group, nor indeed any check on the reliability of the analyst’s memory.

CONSTRUCTS

Many of the constructs in psychoanalytic theory are vague and difficult to relate to any external referent: for example, the concepts of eros and thanatos, the life and death instincts. Thus Freud (1920) in ‘Beyond the pleasure principle’ writes that there are two kinds of instinct: ‘Those which seek to lead what is living to death and others, the sensual instincts, which are perpetually attempting and achieving a renewal of life’. It is, obviously, no simple matter to operationalize these terms.

HYPOTHESES

It is inevitably the case that if the constructs of a theory are themselves unclear, any hypotheses of which they form a part must be difficult to refute. Even worse, they may be impossible to test. For example, the anal character (Freud, 1908a) is considered to be, in psychoanalytic theory, the result of repressed anal erotism. Although at present this hypothesis is not testable because there are no adequate measures of anal erotism (an example of a vague construct) or repression, it may be tested in principle at least. Until recently the back of the moon could not be observed, yet the statement or hypothesis that the back of the moon was made of cheese was testable and refutable in principle and hence scientific. However, the Freudian theory of the anal character cannot easily be refuted. Thus if parsimony, a sublimation of the retentive desire, were not found to be part of the anal character, the Freudian theorist can point to Jones (1923) and argue that generosity too may be included, being a reaction-formation against the retentive desire.

These examples make it clear that on these criteria of scientific theories Eysenck (1953a) is correct. Psychoanalytic theory is open to serious criticism. However, psychoanalytic theory is of such a kind that these criticisms are not necessarily entirely destructive.

THE NATURE OF PSYCHOANALYTIC THEORY

Rapaport and Gill (1959) have attempted to impose a four-fold hierarchical structure onto psychoanalytic theory which is exceedingly useful for the clarification of the theory. This structure with the examples given by the authors is set out below.

1. Empirical proposition. Around the fourth year of life boys regard their fathers as rivals.

2. Specific psychoanalytic proposition. The solution of the Oedipal situation is a decisive determinant of character formation and pathology.

3. General psychoanalytic proposition. Structure formation by means of identifications and anti-cathexes explains theoretically the consequences of the ‘decline of the Oedipus complex’.

4. Metapsychological proposition. The propositions of the general psychoanalytic theory which explain the Oedipal situation and the decline of the Oedipus complex involve dynamic, economic, structural, genetic and adaptive assumptions.

This taxonomy is the basis for the argument, suggested above, that the failure of psychoanalytic theory to stand scrutiny against the normal criteria of science is not necessarily destructive. The criticisms raised by Eysenck (1953a) are directed against the top of the hierarchical structure — general psychoanalytic propositions and metapsychological propositions; they do not apply with the same force to the empirical basis — empirical and specific psychoanalytic propositions. Indeed, the problem of the scientific validity of psychoanalysis in terms of this structure turns about this question. Are the empirical propositions and specific psychoanalytic propositions true? In other words, the first concern is with the observations. Indeed it seems pointless to criticize the metapsychology of psychoanalytic theory on the grounds of internal inconsistency, or untestability of the hypotheses, before establishing whether or not the empirical data, which the metapsychology seeks to subsume, are well founded or not. This is the approach adopted in this book. Psychoanalytic theory has been carefully examined in order to set out as precisely as possible the empirical or specific psychoanalytic propositions. The psychological literature has then been reviewed to see whether or not these propositions may be regarded as true. This procedure, it should be noted, has been objected to by Martin (1964), who argues that to restate the theory in this way is to alter it. Nevertheless, if one proceeds with care, this need not be the case.

There is a further characteristic of psychoanalytic theory which is relevant to its viability as a scientific theory. This is that it is not a unified theory but a collection of parts (Farrell, 1961). This means that even if certain psychoanalytic hypotheses are shown to be false it does not destroy the whole of psychoanalysis. Thus, for example, even if dreams do not express wishes, the castration complex might well lie behind homosexuality. Farrell (1951, 1961, 1964) has, of course, long been interested in the problem under discussion — the scientific validity of Freudian theory — and he makes a number of points which deserve mention.

1. He agrees with Eysenck (1953a) that psychoanalytic theory cannot be called scientific because the evidence on which it is based is not of established validity. In addition its terms are not clearly enough defined. These criticisms relate to the observation and constructs of Marx (1963).

2. However, he refuses (1964) to consider psychoanalytic theory as a myth, as suggested by Popper (1959), on the grounds that it is not considered to be, even by its most hostile opponents, purely fictitious, it invokes no supernatural persons and does not make use of ‘popular’ ideas. Nor indeed is it a pre-scientific theory in that unlike, for example, De Rerum Natura it can be used for something — psychotherapy.

3. He thus concludes (1964) that psychoanalytic theory cannot be said to be true; for the data on which it is based are not well established. On the other hand it cannot be said to be false — because no collection of evidence has, as yet, falsified the entire complex. Finally there is no acceptable alternative to the theory. This last argument has been much stressed by Conant (1947), who argues that theories are not deposed by facts that do not fit them, but by alternative superior theories. In summary, Farrell describes psychoanalytic theory as a premature empirical synthesis offered in advance of the evidence.

This is the view of psychoanalytic theory accepted in this book. The theory is regarded as a huge collection of empirical hypotheses and propositions some of which may be true. Certainly the metapsychology is unscientific but equally certainly the implicit empirical propositions, in the sense that they are testable, are not. Because the theory is not unified, the disproof of some or even most of the propositions is not fatal to the whole theory — a view also proposed strongly by Sherwood (1969).

SOME OTHER VIEWS OF PSYCHOANALYTIC THEORY

It must be made clear that the view of psychoanalysis and its scientific status adopted in this book, the logical positivist viewpoint advocated by Popper (1959) and Kuhn (1970), is only one of many. Thus Cheshire (1975), for example, has shown that many of the apparent inconsistencies within psychoanalytic theory are in fact more apparent than real and that it is not as incoherent or meaningless as its critics would have us believe. This viewpoint is fully accepted by the present writer but is irrelevant to the purpose of this book, which is to establish the empirical foundations of the theory. However, even if the claims of Cioffi (1970) that the theory is formally defective were accepted (refuted in our view by Cheshire, 1975), this would still not be relevant to this book.

However, as Eysenck and Wilson (1973) stress, Ricoeur’s (1970) argument that what is important in psychoanalytic theory is the phenomenology of environmental variables, how they appear to the subject rather than to the observer, seems relevant. For on this view the positivist approach adopted in this book just will not do. Scientific, publicly verifiable quantification can never deal with the subjective constructs claimed by Ricoeur to be central to the psychoanalytic position. Our position here is, however, clear. It may well be the case that Ricoeur is correct. If he is then in our terms psychoanalytic theory is not scientific. Its study becomes not science but philosophy or hermeneutics. If he is correct then all the experiments to be cited in this book will prove negative. The theory will not be supported. It can be argued, then, that psychoanalysis is not scientific. Our approach implicitly puts Ricoeur’s argument to the test. Thus Ricoeur’s claims do not per se invalidate the positivist empirical approach adopted here, although, of course, they could well explain negative findings.

In summary, therefore, it is not argued that the empirical approach to the testing of psychoanalytic theory is the only method of studying that theory. However, it is claimed that this approach is the only one that is capable, in the view of most philosophers of science, of examining the scientific status of psychoanalytic theory and that, of course, is the purpose of this book.

THE TESTING OF THE EMPIRICAL PROPOSITIONS

As Eysenck (1965a) says, the reason why psychoanalysis ‘has never been taken very seriously by people with some regard for the principles of scientific method’ resides in the nature of the data on which psychoanalytic theory is based. The fact is that the psychoanalytic method is not the scientific method (Farrell, 1964). This means that the effort to establish the empirical propositions of psychoanalysis scientifically must largely depend on the work of psychologists. The data from psychoanalytic sessions have not done this, nor are they likely to do so for some time. Thus Crown (1968), a psychoanalyst writing on psychoanalysis and the scientific method, categorically states that analysts are not prepared to use the methods appropriate to the sciences because these distort the data.

ACCEPTABLE DATA FOR PSYCHOANALYTIC THEORY

In this book the objective evidence for psychoanalytic theory is reviewed and discussed. Only investigations where the data had any claims to scientific respectability have been examined. Thus, for example, if tests were used they had to be tests with some evidence for their validity. This has meant that the majority of studies with the Rorschach have not been considered. Eysenck (1961) succinctly puts the case against the Rorschach technique, namely that in studies of its validity there is an inverse correlation between positive results and methodological excellence! Other projective techniques are open to similar criticisms but their results have been used where evidence for the validity of the particular score was presented.

If Freudian hypotheses concerning special groups were under examination, then control groups were regarded as essential. After all, the Freudian claim that paranoid schizophrenia is caused by repressed homosexuality is not proven by demonstrating that homosexuality is repressed in paranoids unless it also be shown that such homosexuality does not occur in non-paranoid schizophrenics. This is a matter of simple logic.

A third constraint on the types of study examined was that there has to be some form of statistical analysis of results so that non-significant fluctuations of scores would not be seized on as evidence for or against the theory. The judgement of the statistical excellence of the analysis naturally involved a consideration of the sampling techniques.

The final consideration given to all studies was whether the results were in fact relevant to psychoanalytic theory. This to some extent constitutes a value judgement but where the empirical propositions were clear the problem was not so difficult. Where they are not, then the results of objective studies can be used to clarify the theory itself. This difficulty of deciding the relevance of an investigation to Freudian theory was partly overcome by examining only those studies where the hypothesis tested was clearly stated and deliberately related by the author to Freudian theory. Alternative and perhaps more parsimonious explanations of the results were also discussed. In summary, for the type of data that can put psychoanalytic propositions onto a scientific basis, the following criteria were considered important:

1. Sampling procedures and use of adequate control groups.

2. Validity of tests used.

3. Quality of the statistical analysis of results.

4. Relevance of the conclusions to psychoanalytic theory and pos...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Original Title Page
• Original Copyright Page
• Table of Contents
• Preface
• Introduction
• 1 Freudian theory and scientific method
• 2 The Freudian theory of psychosexual development
• 3 The objective evidence concerning psychosexual personality syndromes
• 4 Psychometric studies of personality
• 5 The relation of psychosexual personality syndromes to infantile experiences
• 6 The Oedipus and castration complexes
• 7 The structure of the mind
• 8 The mechanisms of defence
• 9 Freudian dream theory
• 10 Psychoanalytic theory of neurosis, psychosis and other emotional and psychogenic disturbances
• 11 The effects of psychoanalytic therapy
• 12 The neo-Freudians and other analytic psychologists
• 13 The status of Freudian theory in the light of the evidence
• Bibliography
• Name index
• General index