Review

Reviewed Work(s): A Theory of Cognitive Dissonance by Leon Festinger

Review by: Albert Pepitone
Source: The American Journal of Psychology , Mar., 1959, Vol. 72, No. 1 (Mar., 1959),
pp. 153-155
Published by: University of Illinois Press

Stable URL: https://www.jstor.org/stable/1420234

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

University of Illinois Press is collaborating with JSTOR to digitize, preserve and extend access
to The American Journal of Psychology

This content downloaded from

72.255.37.19 on Wed, 06 Dec 2023 06:33:02 +00:00
All use subject to https://about.jstor.org/terms
 BOOK REVIEWS 153

philosophical-psychoanalytic circles. A major shortcoming is its failure to explore

the evidence offered to support the notion of the unconscious, or its potential
fruitfulness relative to other hypotheses. Although Flew states that he had no inten-
tion to do this, the lack of such wider investigations leaves many such discussions
relatively meaningless to the general psychologist.
Psychologists are much more familiar with the papers by Cronbach and Meehl
reprinted in this volume, and they would seem to need no further comments here.
Finally, a brief comment on Scriven's final chapter, which attempts to draw a
distinction in principle between the physical sciences and behavioraI sciences. Scriven
believes that the number of critical variables to be controIled even in a laboratory
experiment in the behavioral sciences is so great and creates such complexity that
we cannot expect ever to formulate laws of the simplicity of Newton's or Einstein's.
"This is a claim," he believes, "based on the empirical evidence, not on any a priori
necessity." If this is so, of course, the only test for such a notion can be made only
by repeated trials. From a most mundane point of view, we psychologists would
appear to have no choice but to act as if such general laws were discoverable; for
to act otherwise, would guarantee our failure to test Scriven's hypothesis. Moreover,
it is a serious question whether there do not exist already general psychological
laws of sufficient number and quality to warrant some cataloging. Perhaps psycholo-
gists feel unnecessarily inferior.
It is probably not an accident that psychoanalysis receives the lion's share of
the attention in this book, for psychoanalysis is really the only system which has
attempted an explanation of peculiarly human behavior, and of all of it. Apparently,
the philosophers of science take it as a sign of the need for careful analysis that
psychoanalysts rarely find behavior puzzling and generalIy have ready explanations
for almost anything. Or is it, perhaps, that here is another field in which exact data
need not be considered; indeed, the theory requires a method which cannot produce
exact data. Almost all of the authors include at least one sentence in their paprs
stating their awareness of this fact, and they continually reassure us that they are
not passing on the correctness or incorrectness of psychoanalytic theory. Few say
anything about the tenability of the theory's requirement that a method which can-
not produce exact data be employed for its investigation. None seems to recognize
that the methods he would require for the testing of this theory are precisely the
objective methods universally unacceptable to the psychoanalysts. If this is an ad
hominem argument, it hardly seems avoidable, since were it not for the presence
of institutes of psychoanalysis many of the current discussions might not be neces-
sary at all. Psychoanalytic theory would be evaluated alongside other psychoIogicaI
theories. We do not have institutes for Newtonian physics or Einsteinian physics.
One strongly suspects that the psychoanalysts have enticed some philosophers, and
one can only hope that they will emerge unconfused and ready again to lend their
aid in the scientific study of behavior.
University of Pennsylvania JULIUS WISHNER

A Theory of Cognitive Dissonanre, BY LEON FESTINGER, Evanston, IIl., Row-

Peterson, 1957. Pp. v, 291.

There is some dissatisfaction with learning-theory as an explanation for behavior.

Theoretically, the trouble seems to be that in placing the explanatory burden on
past associations and reinforcements, the organism's present cognitions and motiva-

This content downloaded from

72.255.37.19 on Wed, 06 Dec 2023 06:33:02 +00:00
All use subject to https://about.jstor.org/terms
154 BOOK REVIEWS

tions do not receive the attention they deserse as determinants of behavior. The
difficulty is reflected empirically in the fact that .ast experience is found to be
neither directly nor simyrly relatel to present responses. Witness the increasingly
bewiIdering data l)n [.artial reinforcelllent. As a reae-tion to tIlis difEculty, there has
been ( 1 ) a noticeclble elllrhasis upon .Siscrilllinations decision-lllaking, and other
contemporaneous processes; (2) a vig)1ous anarchistic approach in which no
theoreticaI assumptions whatever are made concerning the effects of experience or
'reinforcement schedulesR upon response-production; and (3) a resurgence of interest
in cognition and cognitively related motivation. Festinger's theory of cognitive dis-
sonance falls within this last category.
In brief, it is proposed that much behavior reflects the organism's attempt to re-
duce 'dissonance' roughly an inconsistency between cognitions. There is no pre-
sumption on Festinger's part that this formulation is fundamentally new; indeed,
he likens the reduction of dissonance to classical, homeostatic processes set off by
drives like hunger, and to modern conceptions of the cognitive processes such as
Heider's theory of balance Considering the range of phenomena, however that are
comprehended and predicted in terms of dissonance-reduction, Festinger has de-
veloped an idea of considerable power and originality which is well supported by
experimental evidence.
Since the idea that people 'strain toward consistency' is not new, why is a whoIly
new term introduced? Will not the concept of conflict serve just as well? Festinger
would presumably argue that, not only is 'conflict' a term with muItiple meanings,
but it is a notion that is basically different from 'dissonance.' On the latter point,
dissonance arises after a decision, i.e. after a conflict of action-tendencies is re-
solved. While plausible, this argument is based on a specific assumption concerning
the locus of conflicts. Moreover, there is no necessary reason for assuming that
decision resolves conflict, just as there is no reason to suppose that eating neces-
sarily satifies hunger. The manifest confusion surrounding conflict may be a better
justification for a new concept.
The conceptual definition of dissonance is, however, vague. Festinger says that
"two cognitive elements are in a dissonant relation if, considering these two aIone,
the obverse of one element would follow from another." The catch is, of course,
that obverse and folfow from are not clearly defined, hence it is not clear concept-
ually when a dissonance exists.
The magnitude of dissonance is defined as the weighted (according to their im-
portance) proportion of dissonant elements. Apart from the rather formidable prob-
lem of measurement (e.g. how is a cognitive eIement defined?), there are other
didiculties in this formulation. For example: (1) The intensity of dissonance, i.e.
the degree of 'obverseness' does not seem to be handled by the proportion of dis-
sonant elements in the formulation. (2) The role of uncertainty in producing cog-
nitive dissonance is not clear. The cognitive elements implicit in dissonant reIations
are assumed to be perfectly reliable. Yet most cognitions are uncertain, at least to
some degree. (3) It seems plausible to assume as Festinger does- that punish-
ment affects the degree of dissonance via changing the importance of the dissonant
elements. Punishment also affects the state of the organism directly as pain, dis-
comfort, etc. Is punishment as a motivational variable different from punishment
defined in terms of cognitive importance?
Attempts at the reduction of dissonance take three forms: (1) changing one or

This content downloaded from

72.255.37.19 on Wed, 06 Dec 2023 06:33:02 +00:00
All use subject to https://about.jstor.org/terms
 BOOK REVIEWS 155

more elements involved in the dissonant relationship; (2) adding more cognitive
elements that are consonant with existing cognitions; and (3) decreasing the im-
portance of the elements involved in the dissonant relationship. In general, Festinger
handles this problem in terms of the quantities involved and the circumstances sur-
rounding the dissonant situations. As one surveys the experiments as a whole, how-
ever, it is not always clear why each dissonant relation had the unique solution that
he adopted. Obviously, predictions of the mode of dissonance-reduction, based on
the quantities of the cognitive elements involved, require more effective methods of
measurement than are now available.
In connection with the central role of cognition in the theory of dissonance, it is
not clear how the theory can apply to the case of unconscious processes, e.g. being
forced to associate with someone toward whom there is urzronsSious hostility. One
possible advantage of conflict as an explanatory device is that it is not restricted to
levels of consciousness.
A final question concerns the degree of generality held by the theory of dissonance.
By implication, the painful, or at least negative character of dissonance is universal.
At the same time, there presumably are wide individual variations in the 'tolerance
for dissonance.' One wonders whether the pressure toward consistency is, at least to
some extent, a reflection of certain values in Western society. For example, most
Americans have an abhorrence of being cheated. Any purchase consequently may be
accompanied by gnawing doubts of a mildly paranoid variety. How do they handle
such uncertainty? Many brag about their material advantages, and, where possible,
derogate the articles they decided against.
The criticisms in this review do not detract from the fact that Festinger's theory
of dissonance generates subtle, non-obvious predictions which cover a wide variety
of phenomena that can be confirmed experimentally. This alone is a sizable achieve-
ment.
University of Pennsylvania ALBERT PEPITONE

The Concept of Developmestt. An I5CUA in the Study of Husnan Behclvior. Edited by

DALE B. HARRIS. University of Minnesota Press, 1957. Pp. x, 287, $4.75.

The present-day passion for symposia sometimes makes strange bed-fellows. But
surely there have been few stranger assortments than the conglomeration of men,
ideas, and viewpoints represented in The Conrept of Development. The reader must
at the outset solve two problems: first, in what sense is development a "concept"; and
second, how does such a concept become an "issue" in the study of human behavior.
The most probable answer to the first question is that by "development" is meant
what the biologists call "epigenesis," the coming into being, through development,
of new qualities not represented in previous stages. In the study of human behavior,
this concept is apparently intended to integrate researches concentrated on different
levels of organization. Or at least, this is a likely interpretation; to identify a unifying
theme in this collection of miscellaneous essays is rather more diScult than seizing
a droplet of quicksilver between the finger tips.
The symposium is divided into five sections, of which the first comprises a general
introduction. In the most interesting paper in this section, "Determinism and Devel-
opment," Ernest Nagel makes a well-reasoned defense of moral responsibility in the
face of determinism; it is however unfortunate that the author devotes a consider-
able part of the space allotted to him to biological viewpoints that are rather out-of-

This content downloaded from

72.255.37.19 on Wed, 06 Dec 2023 06:33:02 +00:00
All use subject to https://about.jstor.org/terms