Reports a,ndPapers no.

19 1964
in the Social Sciences

i
 SOCIAL SCIENCF
CLEARING HOUSE DOCUMENTS

The special documents of the Social Science Clearing House are intended to
present descriptive or documentary material to a restricted public of spe-
cialists as and when such material becomes available during the execution
of Unesco's programme in the field of the social sciences.
Whereas material of general interest is presented in various others series,
these special documents will contain texts which are more suited for limited
circulation. They will often consist of reports relating to the Regular pro-
gramme of Unesco or to the special Programme of Aid to Member States;
others will take the form of catalogues,repertories and directories.
The authors alone are responsible for the Reports and Papers,and these
should not necessarily be taken to represent the views of Unesco.
These special documents issued by the Social Science Clearing House in
offsetwill be occasional papers,not published with strict periodicity. To date,
the followinghave been published :
SS/CH 1 Survey on the Ways in which States Interpret their International
Obligations, by Paul Guggenheim (out of print in English, still
available in French).
SS/CH 2 Mission on the Teaching of the Social Sciences in Pakistan
Report by Andre Bertrand.
-
SS/CH 3 Research Councils in the Social Sciences.
SS/CH 4 The Social Sciences in Secondary Schools ; Report of an enquiry
presented to Unesco by the International Federation of Secondary
Teachers (out of print).
SS/CH 5 International Organizations in the Social Sciences. (out of print)
SS/CH 6 Research Councils in the Social Sciences, Addendum 1956 (supple-
menting SS/CH 3).
SS/CH 7 Foundations with Social Science Activities : An International
Catalogue.
SS/CH 8 Assistance to underdeveloped countries ; an annorated bibliography
prepared by Jean Viet (bilingual : English/French) (out of print).
SS/CH 9 Selected documentation for the Study of Race Relations, compiled
by Jean Vier (out of print).
SS/CH 10 Rural Communities ; Problems, Methods and Types of Research,
by I. Chiva (out of print).
SS/CH 11 International Repertory of Institutions conducting Population
Studies (bilingual English/ French).
SS/CH 12 N e w Towns ; a selected annorated bibliography compiled by
Jean Viet (bilingual : English/ French).
SS/CH 13 International Organizations in the Social Sciences ; revised and
augmented edition with an introduction by Jean Meynaud.(out of print)
SS/CH 14 Selected Documentation on Criminology ; compiled by the Incer-
national Society of Criminology.
SS/CH 15 International Co-operation and programmes of economic and social
development; annotated bibliography prepared by Jean Viet (bi-
lingual : English/ French).
SS/CH 16 The cinema and social science; a survey of ethnographic and socio-
logical films by Luc de Heusch.
SS/CH 17 International directory of sample survey centres (outside the U.S.A.)
(bilingual French/ English).
SS/CI-I 18 The social science activities of some eastern european academies
of sciences.

An "Advance InformationSheet" will be prepared for each new issue.

Prices will vary according to the size of each number.

SS.64.XV.
19 A

Printed in the Workshops of U n e s c o

-
place de Fontenoy. Paris 7e France
0 UNESCO 1965
 Attitude change 1
A review and bibliography
of selected research I

by E.'E.Davis
Research Psychologist, Social Science Study Group
for Internationul Problems, Munich
 PREFACE

In its continued efforts to promote human rights, within the framework of its scientific activities designed to
help in overcoming prejudice and discrimination, Unesco has long been aware that studies of attitude change are
significant in the context of educational activities intended to combat discrimination in all its forms. Follow-
ing an international meeting of experts on 'Youth and Race Prejudice' held at Gauting, Germany, in February-
March 1961, the Secretariat, at the suggestion of the late Professor Alfred Metraux, the anthropologist w h o for
many years w a s responsible for Unesco's programme o n race relations, decided that the present review should
be undertakenand entrusted it to Dr. Earl E. Davis in his capacity as consultant to the Unesco Youth Institute
at Gauting. Dr. Davis is o n the teaching and research staff of the Department of Psychology of the University
of Illinois. After specializing in social psychology and mental health, including its pedagogic aspects, he
took up research on attitude change at the international level; he has also co-operated in inquiries -partly
-
educational and part19 psychological conducted under the auspices of the Gauting Institute.
It will be seen that in this review, interest centres more specifically o n research o n attitude change in its
relation to prejudice and discrimination than o n the description of problems arising out of racial and other
discrimination. The review, therefore, constitutes one specialized effort among numerous others designed to
assist interested organizations, research institutes and individual scholars in promoting research as well as
in devising educational measures aimed against prejudice.
A s things stand, this particular field of research on attitude change has been explored mostly within the
framework of social psychology, a relatively young scientific discipline which has, especially in its earlier
period, been developed more intensively in the United States of America than in any other country. E v e n nowa-
days, the majority of socio-psychological studies on attitude change have their origin in the United States and
some other English-speaking countries, such as Australia, Canada, lVew Zealand and the United Kingdom.
Social psychology having spread only in recent years to other linguistic areas, there are relatively few studies
on record originating in other parts of the world.
Although responsibility for the contents of this review rests, of course, entirely with the author, w h o does
not express any official opinion o n behalf of Unesco, the Secretariat considers that the interest in such socio-
psychological research is sufficiently large to warrant the publication of this work, in the hope that it will
encourage specialists in different areas to continue, or perhaps to initiate, investigation in this field.
W h e n further information of significance becomes available, it is intended to supplement this first research
review (completed in manuscript form in December, 1961) in order to give it wider international coverage.
 CONTENTS

Introduction ........................... 7

PART I: Experimental research on attitude change ..... 13

1. Personality-oriented research ........... 15

2. Group-oriented research .............. 22

3. Persuasive communications research ....... 27

4. Theoretical problems of attitude change research . 36

PART 11: Action research on intergroup attitudes ...... 39

0
1. Educational programmes in intergroup relations .. 41

2. Intergroup contact and community studies ..... 46

3. Cultural influences and the r6le of society ..... 52

S u m m a r y and prospects .................... 56

Index of authors ........................ 58

 INTRODUCTION

Purpose of this paper Williams, there are numerous organizations in

other countries, as well as at the international
The importance of intergroup relations in the world level, engaged in such efforts. Such activity, indi-
today, the devastating effects of conflicts between cating that the importance of improving intergroup
various racial,ethnic,religious and political groups relations is widely recognized today, is highly
in the past, and the acute danger which such con- encouraging.
flicts represent for the futtwe peace and security These efforts have obviously presented a chal-
of mankind have been so well described by other lenge to social scientists, both as scientists and
writers and are so apparent to every intelligent as citizens; for just as our technical world would
person today that it is scarcely necessary to dwell be inconceivable without the advances that have
at length upon these points here. It is, then, not been made in the natural sciences, there is every
at all surprisingthatsocial scientists have increas- reason to believe that the social sciences have an
ingly turned their attention to the problem of inter- important r61e to play in the solution of the human
group relations. Of course, intergroup relations problems of our time. The sharp increase in
have always been extremely important,but where- empirical research dealing with these problems
as a few decades ago they were scarcely an object in recent years is evidence of the fact that social
of scientific investigation, empirical research in scientists have taken up this challenge.
this area has proliferated in recent years. The The term 'attitude change' , encountered only
body of publications dealing with this issue has occasionally up to about fifteen years ago, has
reached staggering proportions, thus creating occurred with increasing frequency in the profes-
an obvious need for systematic documentation and sional literature in recent years and has come to
periodic reviews of research. It was the recogni- designate a specialized area of research. One m a y
tion of this need which prompted this review. or m a y not like such phrases, which tend to be-
It would be appropriate n o w to delineate the goals come catchwords of professional jargon, ushering
and limitations of the present paper. First of all, in a fad in research; however, whether we like
it must be emphasized that it does not purport to them or not they are there and usually have their
survey the whole field of intergroup relations. reasons for existing. In the present case, this
Such a task, if at all feasible, clearly goes beyond research trend represents a serious attempt at
the time and space available for this paper, and, in- putting some of the propositions upon which a great
deed,beyondthe competence of the present writer, deal of effort toward improving intergroup relations
since such an undertaking could only be carried is based to the critical test of empirical investiga-
out by a large team of experts from the various tion. Without losing sight of the fact that factors
disciplines involved in the study of intergroup rela- other than attitudes are extremely important for
tions. Nor does this paper attempt to provide a determiningbehaviour,w e submit nevertheless that
complete bibliography or exhaustive review of one social attitudes are highly relevant to the question
of these disciplines,e.g. psychology, sociology, of intergroup relations, and that therefore research
etc. Its scope is rather more limited: it is de- on attitude change warrants more careful consider-
signedto provide a review and selectedbibliography ation and review.
of some of the more important research relevant Emphasis here will be upon recent research,
to attempts at changing social attitudes, with a since earlier research in this area has been sum-
view toward improving relations between various marized elsewhere (e.g. Watson, 1947; Williams,
ethnic, racial, religious and other groups. 1947; MacIver, 1948; Rose, 1948; Allport, 1952;
O n the other hand, no attempt can be made at an Saenger, 1953; Simpson and Yinger, 1958). W e
exhaustive coverage of the numerous efforts being shalltry also to be wary of certain pitfalls inherent
made to improve intergroup relations. There is in such an undertaking. One must avoid the temp-
no doubt that the estimate contained in an earlier tation 'totake these provisional results of research
review by Williams to the effect that 'there are not for what they are, i.e. points for further
hundreds of private and governmental agencies e m - discussion, but to take them as the final answers,
ployingthousands of persons and spending millions
of dollars in a sincere effort to do something about 1. References, designated by the author's name
the problem' (Williams, 1947, p. vii)/I, is truer and date of publication, refer to the bibliography
today than ever before. In addition to the agencies appearing at the end of each chapter or in that
in the United States of America, referred to by of a preceding chapter.

7
Introduction

as evidence that the points in questionhave been esta- Oriented Research; (2) Group-Oriented Research;
blished'/'. Especiallynon-scientists,but often scien- (3) Persuasive Communications Research; and
tiststhemselves,fallintothis error. Nor is it the (4) Theoretical Problems of Attitude Change
purpose of such a research review to present past re- Research.
search as anormto whichfuture research should In Part 11 of the paper attention is turned to ac-
conform; our intentionis rather,through a critical tion research on intergroup attitudes. A s mentioned
analysis of current researchtrends,to indicatepos- above, no attempt was made here to survey all the
sibilitiesfor future research. With these notes of various practical efforts towards improving inter-
caution in mind, it is hoped that scientists and group relations in various parts of the world. The
practioners alike will find this review of use. emphasis here is rather upon a few selected efforts
Plan of the paper illustrative of the attempt to utilize the results of
research for the planning, evaluation and control
A s w e have already indicated, social scientists of action programmes. As representative of ef-
have taken up the challenge presented to them by forts of this type, this section contains chapters
the initiative of the numerous individuals and orga- on: (1) Educational Programmes in Intergroup
nizations engaged in action programmes on inter- Relations; (2)Intergroup Contact and Community
group relations. Early efforts in this area were Studies; and (3) Cultural Influences and the Rble of
largely undertaken without the use of scientific Society.
knowledge for planning and control. Thus many In a concluding chapter an attempt is made to
programmes were carried out without any solid summarize contemporary trends and to point to
foundation in fact and with little possibility of de- certain prospects for future research.
termining whether or not the desired effects were
achieved. This early lack of co-operation between A clarification of terms
scientists and practitioners was usually due both
to a certain scepticism on the part of the practi- Before continuing, it would be advisable to clarify
tioners as to the efficacy of scientific methods for the terms of reference of this paper. A working
solving practical problems and to a certain 'ivory definition of the key term 'attitude'will be given
tower' attitude on the part of many scientists, who and its major aspects briefly discussed. Also a
prided themselves on the 'pure'nature of their en- distinction wiu be made between 'attitudeI and a
deavours. Although such attitudes m a y still occa- number of closely related terms such as 'opinion',
sionally be encountered today, they have become 'stereotype'and 'prejudice'. Finally, since our
the exception rather than the rule. Most prac- purpose in studying attitude change lies in its rele-
titioners are well aware that research, although vance to the problem of intergroup relations,the
offering no patent formulae or'cure-alls: does have important question of the relationship between atti-
its indispensable r81eto play in the planning,evalu- tudes and behaviour will be touched upon.
ation and control of action programmes. A n d most The literature on attitudes is voluminous, the
scientists, in addition to recognizing their obliga- definitions nearly as numerous as the authors who
tion as citizens, realize that research on such have written on the subject. The term 'attitude'
real-life problems is most likely to yield rich would appear to be one of the most widely (and
returns in terms of basic knowledge about human differently) defined terms in the whole of social
behaviour. psychology; and yet, for some reason, social
In the light of this fundamental importance of scientists appear to understand each other when
social science research to programmes of improv- using it, The variety of definitions and the quantity
ing intergroup relations, Part I of this paper is of writing about this term is probably indicative of
devoted to a review of some of the more important the importance of attitudes to social psychological
experimental research on attitude change, O n phenomena, Indeed, Allport (1935)has described
occasion, basic research will be considered, even attitudes as subject matter of social psychology.
though the s.pecific attitudes under investigation W e cannot begin to review all or most of the
are not the social attitudes which are our primary definitions of attitudes here. However, nearly all
concern. This will be done where it is felt neces- of them seem to have at least two factors in com-
sary to refer to basic research in order to arrive mon: First, attitude is an inferred entity, some-
at conclusions which m a y be useful for the special thing which is not measured directly but rather
applied field i n which w e are interested, This sec- deduced from other observable data. Concepts of
tion is designed especially to aid the researcher in this sort have been termed 'interveningvariables'
obtaining an overview and guide to a research area (MacCorquodale and Meehl, 1948) or 'latent vari-
which has become somewhat unwieldy. In consider- ables'(Lazarsfeld, 1950). In a standard review,
ing how best to classify the extensive experimental Green (1954, p. 335) sums up the general nature
work in this area, a myriad of possibilities pre- of such variables as follows: 'In general terms, a
sented themselves. An attempt was made to keep latent variable is used to describe the consistency
the number of categories small and to allow the
material to fall into natural divisions, instead of 1. The Prevention of Prejudice: Report of a Meet-
imposing an artificial scheme upon it. This section ing of Experts,Gauting,Unesco Youth Institute,
is divided then into chapters on: (1) Personality- 1961, p. 5.

8
 Introduction

or co-variation of a number of different responses and (c) rigidity (due to selective perception and
to stimuli of the same general class. The variable interpretation,contrary evidence is often either
is viewed as mediatingthe stimuliand the responses. ignored or seen as confirming the preconceived
The responses are said to co-vary because they are notion). Another concept which is closely related
all mediated by the same hypothetical variable. to that of attitude but which likewise falls more in-
The latent variable is useful because it unifies a to a purely cognitive category is that of opinion.
set of data, namely the observed responses. Sec- Of course, the boundaries are not sharp; opinions
ond, attitudes imply some sort of tendency to act m a y also have affective and conative aspects, but
toward the object toward which they are held. To- for the most part they are more conscious and
gether with external factors in the person's envi- readily verbalized. They are more on the surface,
ronment, they co-determinethe manner in which so to speak, whereas attitudes are deeper lying,
he perceives and reacts toward the world. Thus more ego-involved , general reaction tendencies.
Krech and Crutchfield (1948, p. 152) view attitude Also, opinions are held toward a particular ques-
as 'an enduring organization of motivational, emo- tion or issue, whereas the object of reference of
tional, perceptual and cognitive processes with (social) attitudes are individuals or groups. These
respect to some aspect of the individual's world'. two concepts are sometimes used interchangeably
Allport (1935, p. 810) concludes in his review: 'An in the literature (e.g. Abelson, 1959). However,
attitude is a mental and neural state of readiness although there is some overlapping, we feel that
exqrting a directive influence upon the individual's the major emphases of the two terms are suffi-
. response to all subjects and situations with which ciently different to make a distinction between them
it is related. ' necessary.
Hence, as a working definition we m a y regard The affective component of social attitudes
an attitude as an inferred factor within the individ- refers to the fact that, in addition to beliefs about
ual which involves a tendency to perceive and re- particular groups, such attitudes usually entail
act in a particular manner toward some aspect of feelingstowardthese groups as well. Most defini-
his environment. tions of prejudice emphasize this emotional aspect,
Of course, there are many objects toward which in addition to the features of stereotypy described
an attitude can be held. However, we are interes- above. It is customary to consider affects along a
ted here primarily in social attitudes,i.e. attitudes positive-negative continuum. However, Chein
toward individuals or groups. Campbell (1950, (195 1) distinguishes between 'net positive-negative
pI 31) opeLationally defines social attitude as 'an affect' and 'specific affects', pointing out that re-
-
/enduring/ syndrome of response consistency with search has hitherto concentrated primarily on the
regard to La set of/ social objects'. In their former dimension to the neglect of the latter.
review of prejudice and ethnic relations, Harding, Sodhi and Bergius (1953), in their investigations
Kutner, Proshansky and Chein (1954, p. 1022) also on national prejudices, also criticize the tendency
have social attitudes in mind in their definition: to regard the affective component of attitudes as
'By an attitude we mean a tendency or cluster of varying only along the positive-negative dimension
tendencies to react in various specific ways to an- and propose a more elaborate analysis.
other individual or to a group of other individuals.' The conative component of social attitudes refers
In analysing the psychological processes which simply to the fact that, in addition to thinking and
are involved in attitudes, a distinction is usually feeling a certain way about a social group, there
made in terms of their cognitive, affective and is usually also a policy orientation, i.e. a tendency
conative components (Krech and Crutchfield, 1948; to react in a particular way toward members of
Chein, 1951). Of course, this traditional classifi- this group. It is this facet of social attitudes
cation represents only one of many possibilities in which is measured by such methods as the Bogardus
terms of which attitudes m a y be analysed. H o w - Social Distance Scale (Bogardus, 1925; Triandis
ever, for the sake of convenience, we shall retain and Triandis, 1960) and Moreno's Sociometric
this division for the present brief discussion. Technique (Moreno, 1953).
The cognitive component of social attitudes in- This latter aspect leads to the difficult and much-
cludes the perceptions, beliefs and expectations debated question of the relationship between inter-
that the individual holds with respect to members group attitudes and actual intergroup behaviour.
of various social groups. A special case of such This is in turn but a special case of a fundamental
notions are the stereotypes which are popularly question in the social sciences about which there
entertained concerning various groups. These are is a great deal of difference of opinion, the ques-
attributes normally used to describe an individual tion of to what extent behaviour is determined by
(e,g. , physical features, intellectual abilities, variables within the person, of which attitudes are
character traits, etc.)applied instead to whole one example, and to what extent it is determined
groups. These stereotypes are usually character- by the external factors in the individual's social
ized by (a) over-generalization (the traits are at- and physical environment. Of course, everyone
tributed to all, or nearly all, members of a race, agrees that both are important. Lewin's (1935)
nation, etc.); (b) over-simplification (one, or at classical formula _B = f. (2,E),to the effect that
most a few, characteristics are used to encom- -
Behaviour is a function of the _Personand the Envi-
pass the complexity of a whole race, nation, etc.); ronment, has long since become a truism in the

9
Introduction

behavioural sciences. It is merely a matter of the BIBLIOGRAPHY

relative emphasis placed uponthe two factors. Un-
fortunately, discussions on this issue are usually A B E L S O N , H.I. 1959. Persuasion: how opinions
guided by the theoretical orientations and predilec- and attitudes are changed. N e w York, Springer
tions of the various authors. In fact, stands are Pub. Co.
frequently taken along strictly professional lines: ALLPORT, G.W. 1935. Attitudes. In: C.Mur-
psycho-analysts, clinical psychologists and per- chison (ed.). A handbook of social psycholom,
sonality theorists emphasize the importance of p. 798-844. Worchester,ClarkuniversityPress.
factors within the individual; social psychologists . 1952. The resolution of intergroup
seek to bridge the gap; and sociologists, cultural tension. N e w York, National Conference of
anthropologists and economists emphasize the Christians and Jews.
importance of external factors. In addition to its B O G A R D U S , E .S. 1925. Measuring - social
theoretical importance,this questionis of immense distance. Journal of applied sociology, 9 ,
practical significance in dealing with problems of p. 299-308.
intergroup relations, since major policy decisions CAMPBELL,D.T. 1950. The indirect assess-
are likely to be made on the basis of the answers ment of social attitudes. Psychological Bulletin,
given. Thus Tumin, Barton and Burrus (1958), 47, p. 15-38.
taking issue with assumptions of a close inter- CHEIN, I. 1951. Notes on a framework for the
relationship between prejudice and discrimination, measurement of discrimination and prejudice.
argue that prejudice is scarcely amenable to influ- In: M. Jahoda. M. Deutsch and F.W. Cook.
ence through rational evidence, nor can emotional (eds.). Research methods in social relations,
re-education be considered a feasible public alter- p. 382-90. N e w York, Dryden.
native. They favour, instead, considering means GREEN, B.F. 1954. Attitude measurement. In:
of effectively modifying behaviour, without neces- G. Lindzey (ed.).Handbook of socialpsychology,
sarily altering the prejudices themselves. Tumin Vol. 1 , p. 335-69. Reading, Mass., Addison
(1958, p. 35) states the case quite explicitly: 'If Wesley Pub. Co.
we are as successful in the area of blocking the HARDING,J .;KUTNER , B.; P R O S H A N S K Y , H.;
impulse to discriminate as w e have been in block-
ing the impulse to cheat and steal, we shall proba-
bly have done a great deal. It would be utopian to
-
expect more at least in the near future. The
classical investigation of L a Piere (1934) demon-
=.
CHEIN,I. 1954. Prejudice and ethnic relations.
In: G. Lindzey(ed.1. Handbook of social psycho-
Vol. 11, p. 1021-1061. Reading, Mass.,
Addison Wesley Pub. Co.
KLINEBERG,0.1958. Comments. In:The r61e
strated empirically that attitudes and behaviour of the social sciences in desegregation: a
need not be in accordance with one another, in- symposium, p. 47-48. N e w York, Anti-
deed, that they m a y even be widely disparate. Defamation League.
There have been far too few actual investigations KRECH, D.; CRUTCHFIELD,R.F. 1948.
of this question, although a more recent study by Theory and problems of social psychology.
Kutner, Wilkins and Yarrow (1952) confirms L a N e w York, McGraw Hill.
Piere's results; and other findings, including KUTNER, B.; WILKINS, C.; YARROW, P.R.
those of Tumin and co-workers, referred to 1952. Verbal attitudes and overt behaviour
above, point in this direction. However, Kline- involving racial prejudice. Journal of abnormal
berg, in commenting upon Tumin's thesis as to and social psychology, 47, p. 649-52.
the relationship between attitudes and behaviour LA PIERE,R.T. 1934. Attitudes versus action.
(more specifically between prejudice and discri- Social forces, 14, p. 230-7.
mination) adds a note of caution: 'I would like L A Z A R S F E L D , P.F. 1950. Thelogic and mathe-
to_ suggest ...however, that szying that the two matical foundation of latent structure analysis.
-
/prejudice and discrimination/ vary in different In: S.A. Stauffer, et al. (eds.). Measurement
directions, or that they do not always go to- and prediction, p. 362-412. Princeton, N e w
gether, does not mean that they are not at all Jersey. Princeton University Press,
related. I wonder whether w e should not, in- LEWIN,K. 1935. A dynamic theoryofpersonality.
stead, ask in greater detail how they are related, New York, McGraw Hill.
rather than coming to this negative conclusion' M a c C O R Q U O D A L E , K. ; MEEHL, P.E. 1948.
(Klineberg, 1958, p. 47) O n a distinction between hypothetical constructs
_.

W e shall have more to say about this important and intervening variables. Psychologicalreview,
questionlater. Inthe following chapters an attempt 55, p. 95-107.
will be made to give a balanced view of varying MacIVER, R.M. 1948. The more perfect union.
-
research results not in an attempt to arrive at N e w York, Macmillan,
MORENO, J.L. 1953. W h o shall survive (rev.
simple answers to global questions or patent for-
niulae, for complex problems, but rather in an ed.). Washington, Nervous and Mental Disorder
attempt to obtain an overview of research trends Publ. Co.
which will enable us to ask more specific questions ROSE, A.M. 1948. Studies in the reduction of
and seek answers to them in present or future prejudice (2nd ed.). Chicago, American
research results. Council on Race Relations.

10
 Introduction

S A E N G E R , G. 1953. The social psychology of TUMIN,M.;BARTON, P.;B U R R U S , B. 1958.

prejudice. N e w York, Harper. Educatih, prejudice and discrimination: a
SIMPSON, G.E.;YINGER,J.M. 1958. _Racialand study in readiness for desegregation. American
culturalminorities: an analysis of prejudice and sociological review, 23, p. 41-9.
discrimination (rev. ed.). N e w York, Harper. W A T S O N , G.B. 1947. =on and unity. N e w York,
SODHI, K.S.; BERGIUS, R. 1953. Nationale Harper.
Vorurteile. Berlin WILLIAMS, R.M.,Jr. 1947. The reduction of
TRIANDIS, 1I.C.; TRIANDIS, L.M. 1960. Race, intergroup tensions: a survey of research on
social class,religion and nationality as deter- problems of ethnic, racial and religious group
minants of social distance. Journal of abnormal relations. N e w York, Social Science Research
and social psycholoq, 61, p. 110-18. Council. (Bulletin 57.)
TUMIN,M. 1958. Some problems for sociological
research in desegregation. In: The r61e of the
social sciences in desegregation: a symposium,
p. 33-40. N e w York, Anti-Defamation League.

11
 PART I

EXPERIMENTAL RESEARCH ON ATTITUDE CHANGE

13
 CHAPTER 1

PERSONALITY-ORIENTED RESEARCH

Although research on personalityfactorshas played findings, few social scientists have been unaffected
an important r61e in attempts to explain the nature by it. W e cannot begin to summarize here the re-
and causes of ethic prejudice, one might justifia- sults of the extensive research reported by Adorno
bly ask why a chapter on this subject is included in *and his co-workers concerning the relationship be-
a review of research on attitude change. D o we tween personality structure and prejudiced, ethno-
mean to imply that personality-oriented approaches, centric attitudes; nor can we discuss in detail the
such as, for instance, individual psychotherapy, very useful critical review edited by Christie and
represent a feasible possibility for changing nega- Jahoda (1954), or other more recent reviews re-
tive attitudes on a large scale? Obviously we do porting on the widespread subsequent use of the
not. W h y then do we include this question here? IFt(Fascism)Scale developed by Adorno and his
For one thing, the very fact that these studies co-workers (e.g., Titus and Hollander, 1957;
have played such an important r61e in prejudice re- Christie and Cook, 1958). W e shall be concerned
search would warrant their consideration, at least here only with the implications of these findings
briefly, in any discussion of attempts at changing for the question of attitude change and shall review
prejudiced attitudes;but more specifically,it should a few of the empirical investigations which have
be readily apparent that, although methods which been conducted to test the hypotheses deduced by
are successful in changing an individual's attitudes Adorno - et al. from their findings.
m a y not, in themselves, be applicable on a large In expounding their major hypothesis 'thatthe
scale, such studies do contribute to our knowledge political, economic, and social convictions of an
of the process of attitude change. The decision as individual often form a broad and coherent pattern,
to which methods to use in a particular situation as if bound together by a "mentality" or "spirit",
will depend, of course,.on a number of factors and that this pattern is an expression of deep-lying
among which practical considerations must, of ne- trends in his personality' (Adorno, et., 1950,
cessity, play a key r6le. However, regardless of p. l), the authors do not, of course, regard person-
the methods used, attitudes are being changed with- ality as something inherited or unchangeable. In
in individual human beings and therefore, a know- keeping with modern personality theory, they ex-
ledge of the psychological processes involved is plicitly state that 'personalityevolves under the
indispensable. Personality-oriented research, in impact of the social environment and can never be
additionto throwinglightupon individualdifferences isolated from the social totality within which it
which m a y have to be taken into considerationin occurs' (Ibid,p. 5). The basic frame of reference
certain instances, should also provide us with some within which personality dynamics are considered
insight into the general process of attitude change. is that of psycho-analysis. However, as opposed
The following review of findings should not be to more classical psycho-analytical writers on this
interpreted as an argument in favour of the primary subject, such as Fenichel (1946) and Orr (1946),
importance of personality factors in attitude change. who deal with social circumstances only in an inci-
In the same vein, the following chapters will 'argue' dental manner, A d o r n o s . stress the importance
in favour of the importance of group membership, of social factors and the total organization of the
cultural factors, etc. W e are convinced that a con- society. Concerning the relationship of their ap-
sideration of all these factors, and especially the proach to sociologicalvariables, they state: 'The
specific manner in which they are related and inter- present research seeks to discover correlations
act with one another, is essential to the intelligent between ideology and sociological factors ... the
planning of any action programme on attitude general approach being to consider personality as
change. an agency through which sociological influences
Certainly the most important single publication upon ideology are mediated. If the r61e of per-
dealing with the relationship between personality sonality can be made clear, it should be possible
variables and social attitudes, in terms of its better to understand which sociological factors are
impact upon subsequent research, is The Authori-
~~ ~- the most crucial ones and in what ways they achieve
tarian Personality by Adorno, Frenkel-Brunswik, their effects' (op. cit., p. 6). Thus there is no
Levinson and Sanford (1950). Probably no other necessary incompatibilitybetween the personality-
single work has more affected the social sciences oriented approaches and group or society-oriented
in general, and the area of social prejudice in par- approaches, as some writers have seemed to imply;
ticular, in recent years, than this monumental although, of course, personality theorists and socio-
volume. Whether they agree or disagree with its logists m a y often arrive at different conclusions as

15
Personality-Oriented Research

to the most appropriate plan of action. more harmful than others, and w e are sometimes
What are the implications of the findings reported very glad to be able to control a disease even
in The Authoritarian Personality to the effect that though we cannot cure it'; (g, p 973 et seq.).
individuals who are 'highscorers'on scales of eth- O n the basis of their research findings, these
nic prejudicetend also to be characterized by such authors add a serious note of caution, however, as
traits as authoritarian submission, convention- to the possible effects of external manipulation upon
alism,rigid and stereotyped thinking,lack of insight, the authoritarian individual: 'But if it seems un-
over-concern with power, over-identification with likely that his personality will change, there is
the in-group, repressed hostility, displacement of good reason to believe that his behaviour can be
hostility on to out-groups, projection of unaccept- controlled. Indeed he can be too easilycontrolled,
able impulses on to others, etc.? To speak in and therein lies one of the major troubles. There
psycho-analytical terms, the above-mentioned is little in his make-up to render him resistant to
characteristics are indicative of ego weakness, fascist propaganda or to fascist leadership. ...
excessive use of ego defence mechanisms, failure However regrettable, from the democratic point of
of super-egointernalization and inability of the ego view, this susceptibilityto external control might
to deal with id impulses. Clinically speaking, these be, the fact remains that it offers the best basis
are all indications of pathological tendencies in for preventing his anti-democratic tendencies from
the personality; and although research does not expressing themselves in action. The appeal should
completely confirm the relationship between author- not be to his sympathy or to his conscience, but to
itarianism and traditional psycho-pathological his fear and submissiveness. H e must be convinced
diagnoses (sincein a supporting environment the that arrayed against the overt expression of his
prejudiced individual usually manages to maintain prejudices are the law, overwhelming numbers of
a superficial,though precarious, adjustment),these people, numerous conventional authorities and
are certainlynotthe characteristicsof a healthy per- prestige figures. If those who stand for democracy
sonality. What,then,can be done to change such want to win him to their side, they must do more
individuals? The authors admit that persons with than show him that they have high ideals and real-
such a personality structure would not even be istic plans for social improvements; they must
good risks for psycho-therapy individually, apart convince him that they also have strength. Such a
from the factthat the application of individualpsycho- programme, unfortunately,involves an essential
therapy on a mass scale is not feasible. They paradox: in inducing him to behave in accordance
pointto the possible application of their findings to with democratic principles, one is likely to streng-
education, child-rearing practices and group acti- then his authoritarianism and, hence, his anti-
vities patterned after the knowledge obtained from democratic potential. One could not, therefore,
group psycho-therapy. However, they stress the undertake- so io influence the contemporary beha-
fact that their findings are strictly limited to the viour of/such/ individuals . ..unless one exerted
psychological aspects of the problem, and that as much effort toward ensuring that anti-democra-
psychological means alone are not sufficient to tic leadership did not gain the ascendancy in the
effect major changes. Historical and economic future'(ibid,p. 816).
factors and other major forces operatingwithin the D o subsequent research results support these
society are of great importance. Indeed,they liken hypotheses put forth by Adorno and his co-workers?
the task to 'that of eliminating neurosis, or delin- In an experimental investigation designed to test
quency,or nationalism from the world'and conclude: the hypothesis of a relationship between attitude
'These are products of the total organization of change and authoritarian personalie, Wagman
society and are to be changed only as that society (1955) arrived at some interesting results. O n the
is changed ... The problem is one which requires basis of the Adorno, Frenkel-Brunswik,Levinson
the efforts of all social scientists.' Pending such and Sanford study and the experimental work of
a major solution, however, efforts should be Hoffman (1951) on psychological components of
continued on a broad front. 'It would be most compulsive conformity and of Lasky (1950) on the
unfortunate if a grasp of the true enormity of the relation of experimentally induced changes in
fundamental problem should anywhere lead to a ethnocentric statements to underlying personality
diminution of effort. It is impossible to conceive variables, Wagman predicted that: '(1) Under ex-
of any way of attacking the problem that does not posure to information techniques aimed at cognitive
-
involve a multiplicity of sub-goals to be attained restructuring, relatively non-authoritarian, as
by individuals or by groups. Any act, however compared with relatively authoritarian, persona-
limited in time and space,that serves to counter or lities will show more attitude change (inthe direc-
diminish destructiveness can be regarded as a tion of the information content); (2) under an
microcosm, as it were, of a total effective pro- authoritarian suggestion urging reduction of pre-
gramme ... So it is with various other measures judice towards Negroes, relatively authoritarian
which from our point of view are concerned with the personalities, as compared with relatively non-
treatmentof symptoms or particular manifestations authoritarian personalities, will show more
rather than with the disease itself. Yet we cer- suggestion acceptance; and (3)under an authorita-
tainly do not wish to belittle, or to ask for any rian suggestion urging intensification of prejudice
r2duction in, such activities. Some symptoms are toward Negroes, relatively authoritarian, as

16
 Personality-Oriented Research

compared with relatively non-authoritarian, per- found when the items of the (F)scale were grouped
sonalities will show more suggestion acceptance. ' according to the factors referring to the hypotheses
Using the familiar 'before-after'design, the in- on which the (F)scale is based (Adorno et.,
vestigator administeredthe(F)scale and four other 1950, p. 224 et seq.). Among other things it was
scales designed to measure attitudes towards vari- found that opinion changers scored significantly
ous types of discrimination against Negroes to four higher on measures of authoritarian aggression,
groups of students. Three groups were exposed to anti-intraception and power and toughness.
three different types of material designed to change The authors conclude: 'Opinion change is then un-
attitudes, and the fourth group acted as a control. related to total score on the authoritarian question-
The three attitude change methods were: (a) cogni- naire,but it is clearly related to certain components
-
tive restructuring a booklet was presented pro- of this score. It is furthermore related to all the
viding an 'exposureto information' as to the nature sub-scales that represent variables considered by
of intergroup relations,prejudice, etc.; (b)author- the authors of The Authoritarian Personality to re-
-
itarian suggestion in this experimental variation flect personality trends predisposing a person to
the statements of business and military leaders be authoritarian or not. It is unrelated to the two
supporting a non-discrimination policy were pre- scales (PECand E)that reflect social attitudes
-
sented; and (c)authoritarian suggestion reverse in directly. ... It appears, then, that in a population
of this sort, knowledge of how authoritarian-oriented
this variation statements of business and military
leaders supporting a policy of discrimination were a person is does not help to predict his response
presented. The results of the experiment m a y be to communications from sources such as the scien-
summarized as follows: tific authorities used in the opinion change test,
1. The prediction that under exposure to informa- but that an understanding of the pattern of his rela-
tion techniques, relatively non-authoritarian tionship to authority does' (Linton and Graham,
subjects would show more attitude change (in 1959, p. 86). Katz and Benjamin (1960),in a study
the direction of the information content) was of the effects of white authoritarianism in bi-racial
demonstrated in the expected direction forthree work groups, also found greater compliance on the
of the four attitude change measures. For the part of authoritarian personalities in tasks requir-
authoritarian subjects, attitude change opposite ing group decisions. Authoritarians furthermore
to that intended occurred ('boomerang' effect). rated Negroes significantlyhigher on intelligence,
2. The prediction that under an authoritarian sug- maturity and dominance. These results were inter-
gestion urging a reduction of prejudice toward preted by the authors as being due to the authori-
Negroes, relatively authoritarian personalities tarian's fear of revealing anti-Negro attitudes in a
would show more suggestion acceptance was potentially punitive environment. Another interest-
confirmed in part. For one of the four mea- ing resultwas that authoritarians were significantly
sures the predicted change was significant, less productive in two-man groups requiring co-
whereas for the other three, it was not. operation.
3. The prediction that under an authoritarian sug- Other investigations have studied the relation-
gestion urging intensificationof prejudice toward ship between personality variables closely asso-
Negroes, relatively authoritarian personalities ciated with those described by Adorno et. and
would show more suggestion acceptance was attitude change. SarnoffandKatz (19541,in a paper
demonstrated in the expected direction forthree describing the theoretical framework for a subse-
of the four measures. quent programme of research on the motivational
Furthermore, non-authoritarian subjects reacted basis of attitude change, point out that any attitude
in a Iboomerangl fashion. Thus this experiment m a y be regarded as serving one or more of three
indicates that for authoritarian personalities, atti- major motivational determinants: reality testing;
tudes of racial prejudice seem most modifiable in reward and punishment; and ego defence. Kelman
either an accentuated or diminished direction under (1956) describes three processes in attitude change
an authoritarian suggestion method. A non-autho- resembling the above analysis of motivational
ritarian information method, while effective for patterns: (a) internalization or incorporating the
non-authoritarian subjects, tends to 'boomerang' ideational content of the attitude; (b) compliance
for relatively authoritarian subjects. This finding induced by extraneous reward and punishment; and
confirms the hypotheses put forth by Adorno S. (c) identification through taking the r6le of others.
Other investigations have confirmed these re- In light of the complexity of human functioning,all
sults. Linton and Graham (1959), studying the of these determinants are probably involved in any
relationship between. susceptibility to opinion attitude, although they undoubtedly contribute in a
change and authoritarian attitudes, used a ques- differentialmanner. Therefore, in considering
tionnaire drawn from three of the scales presented the probable effectiveness of any technique of atti-
in The Authoritarian Personality,viz. the Politico- tude change, it is necessary to know which parti-
Economic Conservatism (PEC),the Ethnocentrism cular motivational source is supporting the attitude
(E)and the Fascism (F)scales. Although the opin- under investigation. It is apparent that attitudes
ion change groups were not found to differ signifi- serving the function of reality testing are most
cantly in overall authoritarian trends, as measured likely to be affected by informational approaches
by the questionnaire, significant differences were directed at cognitive reorganization,whereas those

17
Personality-Oriented Research

associated with motivations of reward and punish- through attempting to give insight into the self than
ment are most likely to respond to social sanctions. through providing information. However, subjects
O n the other hand, in seeking to change attitudes highest in ego-defensivenesswere shown to be most
which have an ego defensive function, emotional difficult to change through self-insight procedures.
resistance is likely to be encountered, and the In a further investigation (Katz,McClintock and
personality dynamics involved must be taken into Sarnoff, 1957), the prediction was confirmed that
consideration. medium ego defenders would change most after
Janis and Feshbach (1953) have demonstrated exposure to materials designed to give self-insight.
that communications which arouse a high degree of L o w ego defenders were not using this mechanism
fear or anxiety in the recipient tend to stimulate and were thus little affected by insight into it,
defensive reactions which interfere with acceptance whereas high ego defenders, by reason of their ego
of the communicator'smessage. In a subsequent defensiveness, were resistant to insight.
investigation of personality differences associated Further evidence of such resistance is provided
with responsiveness to fear-arousingcommunica- by Culbertson (1955) in a study on the modification
tions (Janis and Feshbach, 1954). these authors of emotionally held attitudes through r61e playing.
found that subjects who were high in anxiety, as This author found r6le-playing to be a generally
measured by a personality inventory, were consis- effective technique for changing such attitudes;
tently less influenced by strong appeals with respect however he found clear evidence that subjects scor-
to dental hygiene practices than were low anxiety ing high on the (F)scale were less influenced by
subjects. However, Feshbach (1961)has cautioned rble-playing as a means of changing attitudes to-
against over-generalizing these findings,pointing ward Negro integration in housing than were low
out that under certain circumstances different re- (F)subjects. McClintock (1956) arrived at results
sults might be arrived at. Berkowitz and Cotting- in a similar vein, showing furthermore that indivi-
h a m (1960), for instance, have shown that the re- duals high in ego-defensiveness,or high in other-
action to fear-arousing communications depends directedness, tended to accept informational
upon the interest value and relevance of the mate- communications providing rationalizations for the
rial to the subject, Goldstein's (1959) investigation maintenance of prejudicial attitudes towardNegroes,
on the relationship between coping and avoiding be- moving furtherinthe direction of prejudice, where-
haviour and response to fear-arousingpropaganda as individuals low in defensiveness, or low in
also suggests a modification of the Janis-Feshbach other-directedness, tended to reject this influence
results, showing that the reaction to a strong fear and moved against it, i.e. became less prejudiced
appeal depends not only upon the level of anxiety toward Negroes, A n encouraging result of the in-
but also upon the manner in which the subject copes vestigation by Katz, McClintock and Sarnoff (1956)
with such anxiety. In the area of ethnic attitudes, was the existence of a 'sleeper effect' on the part
Cooper and Jahoda (1947) found that highly preju- of those whose score with respect to ego defensive-
diced individuals react defensively, blocking and ness was medium, i.e. a tendency to continue to
distorting materials directed at caricaturing their change in a positive direction over time, as shown
beliefs; and Bettelheim and Janowitz (1950) re- by a follow-up investigation.
ported that many prejudiced individuals reacted to The significance of this whole line of research
'pro-tolerance'propaganda by indicating shifts in for any programme of attitude change is stated
the direction of greater prejudice as a result of succinctly by Katz and Stotland (1959, p. 463):
reading it. Feshbach and Singer (1957)have also 'Our major thesis has been that since attitudes
demonstrated that prejudice m a y be increased under serve different needs and functions, they can be
the impact of communications perceived as arous- changed only through relating the change procedure
ing a personal threat. to the appropriate motive pattern. In general, this
In an investigation conducted within the theore- calls for separating subjects on the basis of their
tical framework of the Sarnoff and Katz (1954) needs and values to begin with and making differen-
paper, Katz , McClintock and Sarnoff (1956) studied tial predictions for various change methods, Thus
the relationship between ego-defence and attitude far the greater bulk of the research on attitude
change. Using a number of procedures, including change has started with the attitude itself and has
projective techniques,they categorized the ego- assumed a c o m m o n motive pattern for all people.
defensive tendencies of their subjects as beinghigh. Our own method ... is to begin with measures of
medium or low. Before-and-afterattitudes toward ego-defensiveness as one of the major sources of
Negroes were assessed by means of specially de- attitudes toward minority groups and to gear in-
vised scales. In one experimentalgroup the attitude fluencesdirected at change to the anticipated motive
change technique was informationalin nature,where- patterns ,' Kelman (1956), whose distinction (des-
as in a second experimental group interpretative cribed above) of three processes in attitude change
materials designed to show the relationship between resembles the Sarnoff and Katz (1954) analysis of
mechanisms of ego defence and anti-Negro attitudes motivational patterns, has likewise stressed the
were presented; a third group served as the con- necessity of considering motivational conditions in
trol and was not exposed to any influence procedure. attempts at attitude change (Kelman and Cohler,
The findings confirmed in general the theory that 1959; Kelman, 1960, 1961a).
affect-ladenattitudes are more effectively influenced Another very important line of research, which

18
 Personality-Oriented Research

deals with individual differences in susceptibility importance of individual differences. Asch (1951),
to attitude change, is represented in the extensive whose work on the effects of group pressure on
work of I. Janis and C .I. Hovland, editors of the attitude change are well known, has also shown in
Yale Studies in Attitude and Communication, and his studies on independence and conformity that
their co-workers. Volume I1 of this series is de- individual differences play an important r6le (Asch,
voted to the subject of Personality and Persua- 1956). H e states: 'There are two directions open
sibility (Janis and Hovland, 1959). The thesis is for an interpretation of the observed individual
developed in this work that there exists a factor of differences. They m a y be the product of m o m e n -
'generalpersuasibility'. The authors describe this tary, episodic circumstances. O r they m a y be the
as 'a predispositionalfactor reflecting an indivi- function of consistent personal qualities which in-
dual's susceptibilityto influence from many diffe- dividuals bring into the experimental situation and
rent sources, on a wide variety of topics, and which have as decisive a bearing on their actions
irrespective of the media employed. In its widest as any of the other conditions w e have studied. ...
sense, general persuasibility should be free of in- Although such contingent conditions probably played
fluences emanating from any particular or specific a part, there are strong grounds for holding that
aspect of the persuasion situation'(Janis and Hov- they accounted for the results only partially. The
land, 1959, p. 225). In summarizing the evidence evidence cited in this chapter, particularly that
of research in support of the existence of such a pertaining to the different modes of reaction, pro-
factor, the authors conclude: 'The present series duces the insistent conviction that the responses
of studies indicate that there is such a factor as to the experimental conditions were functionally
general persuasibility,although there are certain connected with crucially important character qual-
limitations to its generality imposed by the experi- ities' (Ibld.,p. 51). This conclusion has also
mental procedures employed. There is evidence been confirmed by other authors (e.g., Flament,
that persuasibility exists as allcontent free"factor; 1958; Nadler, 1959).
that is, it exists independently of the subject matter What conclusions m a y be drawn from this rather
or appeals presented in any particular communica- compelling evidence stemming from personality-
tion'(m, p. 225 et sea.). In addition to the per- oriented research? Certainly one possible conclu-
sonality factors referred to in the chapter byLinton sion is that individually oriented approaches to
and Graham (1959). to which we referred above, a attitude change might be highly effective on an
number of additional personality characteristics individual basis. Indeed, most investigations have
were investigated, as well as individual differences shown that psychotherapy, which leads to a more
in intelligence, sex, developmental factors, etc. healthy personality generally, also usually results
O n the other hand, the question of whether or not in a significant positive change in negative inter-
persuasibility to mass media communications was group attitudes termed prejudice, even though such
the same as susceptibility to inter-personal influ- therapy was not specifically aimed at this goal.
ence was left open. Katz (19601, in summarizing Thus Baron (1957) found that psycho-analytically
the Janis and Hovland results, points out that oriented psychotherapy resulted in significant
although,in general,there was some tendency for changes in the direction of more liberal attitudes
the acceptance of the influence of one communica- on the part of the patients; and Pearl (1955)like-
tion to be associated with the acceptance of other wise found that psychotherapeutic experience re-
influences, only a small number of the correlation duced ethnocentrism, whereby group psychotherapy
coefficients found were statistically significant. was found to be more effective than individual
Katz concludes: 'Though there m a y be some gener- treatment. With children, Axline (1948) found
al susceptibility to influence, it is apparently not play group therapy to be an effective means of
a potent factor and accounts for a small amount of reducing race conflict inyoung children. Kelman
variance in attitude change' (N p. . ,This
204). (1952) has suggested a conceptual analysis in terms
author emphasizes, as in previous publications, of which the relationship between group therapy
the necessity of considering the relevant motiva- and adult education and other group work m a y be
tional basis of attitudes in determining the reac- fruitfully considered.
tions to attempts at attitude change. Although the possibility of practically applying
Another factor which has been found to be as- such methods is not to be altogether rejected, as
sociated with ethnocentric, prejudiced attitudes, we pointed out at the beginning of the chapter, our
namely that of rigidity (Adornoetal., 1950; Rokeach, main reason for presenting the results of persona-
1948, 1960; Schneider, 1958; etc .),has also been lity-oriented research is not that we necessarily
investigated in relation to attitude change. Ford consider the direct application of such methods on
(1956), for instance, found a positive relationship a large scale to be feasible. Rather we have sought
between dogmatism and resistance to attitude to show that an understanding of the psychological
change, and Boomer (1959)found subjective cer- processes involved in attitude change is indis-
tainty on the part of his subjects to bz positively pensable if we wish to understand the relationship
related to resistance to change. between psychological, social-psychological,
Finally, research of quite another sort, i.e. sociological and other factors. It is only when
studies of the influence of social pressure on at- w e understand such interrelationshipsthat we can
titude change, has also shown evidence for the intelligently plan programmes of action. Also, in

19
Personality-Oriented Research

certain cases it m a y be necessary to take into con- F E S H B A C H , S.; S I N G E R , R. 1957. The effects
sideration individual differences, if w e wish to avoid of personal and shared threats upon social
the danger that our efforts remain without results prejudice. Journal of abnormal and social
or even, possibly, have results opposite to those psychology, 54, p. 411-16.
intended. FLAMENT, C. 1958. Aspects rationnels et gene-
tiques des chahgements d'opinion sous influence
sociale. Psychologie franGaise , 3, p. 186-96.
BIBLIOGRAPHY FORD, L.I. 1956. The relationship between pre-
judice and dogmatism in opinion change. Unpubl.
ADORNO, T.W.; F R E N K E L - B R U N S W I K , E.; Ph.D. dissertation, Purdue University,
L E V I N S O N , D.; S A N F O R D , R.N. 1950. The Lafayette, Indiana.
authoritarian personality. N e w York, Harper. GOLDSTEIN, M.J. 1959. The relationship be-
ASCH, S.E. 1951. Effects of group pressure tween coping and avoiding behaviour and response
upon the modification and distortion of judge- to fear-arousing propaganda. Journal of ab-
ment. In: H. Guetzkow(ed.). Groups, leader- normal and social psychology, 58, p. 247-52.
ship and m e n . Pittsburgh, Carnegie. HOFFMAN, M.L. 1951. A study of certain psycho-
. 1956. Studies of independence and dynamic determinants of compulsive conformity
conformity: (I.) A minority of one against a in a task involving linear judgement. Unpubl.
unanimous majority. Psychological monographs. Ph.D. dissertation, University of Michigan,
70 (whole no. 416). A n n Arbor.
AXLINE, V .M. 1948. Play therapy and race con- JANIS, I.; F E S H B A C H , S. 1953. Effects of fear-
flict in young children. Journal of abnormal arousing communications. Journal of abnormal
and social psychology, 43, p. 300-10. and social psychology, 48, p. 78-93.
BARON, L.K. 1957. Attitude changes during psy- I-
. 1954. Personality differences
chotherapy: individual cnanges m ideology as associated with responsiveness to fear-arousing
measured by the "R" and "PEC" scales under communications. Journal of personality, 23,
the influence of psycho-analytically oriented p. 154-66.
psychotherapy. Unpubl. Ph. D. dissertation, ; H O V L A N D , C.I.&a. 1959. In:
N e w York University. I. Janis and C.L Hovland (eds.)Personality and
BEFXOWITZ, L.; COTTINGHAM, D.R. 1960. persuasibility. N e w Haven, Yale University
The interest value and relevance of fear- Press. (Yale studies in attitude and c o m m u -
arousing communications, Journal of abnormal nication, vol. 11.)
and social psychology, 60, p. 37-43. KATZ, D. 1960. The functional approach to the
BETTELHEIPYI,B.; JANOWITZ, M. 1950. Reac- study of attitudes. Public opinion quarterly,
tions to fascist propaganda: a pilot study. 24. p. 163-204.
Public opinion quarterly, 14, p. 53-60. ; M c C L I N T O C K , C.; S A R N O F F , I.
BOOMER, D.F. 1959. Subjective certainty and 1956. E g o defense and attitude change. H u m a n
resistance to change. Journal of abnormal relations, 9, p. 27-44.
and social psychology, 58, p. 323-8. . 1957. The
CHRISTIE, R.; COOK, P. 1958. A guide to measure of egodefense as relatedto attitude
published literature relating to the authorita- change. Journal of personality, 25, p. 465-74.
rian personality through 1956. Journal of
psycholow, 45, p. 171-99.
- ; S T O T L A N D , E. 1959. A preliminary
statement to a theory of attitude structure and
; JAHODA, M. (eds.). 1954. Studies change, In: S. Kock (ed.1. Psychology: a study
in the scope and method of 'The authoritarian of a science, vol. 111, p. 275-423. N e w York,
personality' Glencoe, Ill. The Free Press. M c G r a w Hill.
COOPER, E.; J A H O D A , M. 1947. The evasion KATZ, I.; B E N J A M I N , L. 1960. Effects of white
of propaganda: h o w prejudiced people respond authoritarianism in bi-racial work groups.
to anti-prejudice propaganda. Journal of Journal of abnormal and social psychology, 61,
psychology, 23, p. 15-25. p. 448-56.
C U L B E R T S O N , F.J .M. 1955. The modification KELMAN, H.C. 1952. T w o phases of behaviour
of emotionay-held attitudes through r6le play-
ing. Unpubl. Ph.D. dissertation, University .
1956. Three processes of acceptance
of Michigan, Ann Arbor. of social influence: compliance, identification
FENICHEL, 0. 1946. Elements of a psycho- and internalization. American psychologist,
analytic theory of anti-Semitism. In: E. 11, p. 261 (abstr.).
Simmel (ed.). Anti-Semitism: a social . 1960. Effects of rble-orientation
disease, p. 11-32. N e w York, International and value-orientation on the nature of attitude
Universities Press. change. Paper read at the meetings of the
F E S H B A C H , S. 1961. The consequences of fear- Eastern Psychological Association, N e w York
arousal in public health education. Paper read City, 15 April.
at the Fourteenth International Congress of . 1961 a. Processes of opinion change.
Applied Psychology, Copenhagen. Public opinion quarterly, 25, p. 57-78.

20
 Personality-Oriented Research

KELMAN,H.C.;COHLER,J. 1959. Reactionsto ORR, D.W. 1946. Anti-Semitism and the psycho-
persuasive communication as a function of pathology of everyday life. In: E. Simmel (ed.).
cognitive needs and styles. Paper read at the Anti-Semitism: a social disease, p. 85-95.
meetings of the Eastern Psychological Associa- N e w York, International Universities Press.
tion, Atlantic City, 3 April. PEARL,D. 1955. Psychotherapy and ethno-
L A S K Y , J.J. 1950. Relations of experimentally- centrism. Journal of abnormal and social
induced changes in ethno-centric statements psychology, 50, p. 227-9.
to underlying personality variables. Unpubl. ROKEACH,M , 1948. "Generalized mental rigi-
Ph.D. dissertation, University of Michigan, dity as a factor in ethnocentrism" Journal of
Ann Arbor. abnormal and social psychology,43,p. 259-278.
LINTON,H.; G R A H m ,E. 1959. Personality . 1960. The open and closed mind,
correlates of persuasibility. In: C .I. Hovland Basic Books, N e w York.
and I. Janis (eds.). Personality and persuasi- S A R N O F F , I. ; KATZ, D. 1954. The motivational
bility, p. 69-101. NewHaven, Yale University basis of attitude change. Journal of abnormal
Press. (Yale studies in attitude and com- and social mvchology, 49, p. 115-24.
munication, vol. 11. ) SCHNEIDER,E. 1958. 0problema psicologico
M c C L I N T O C K , C .G. 1956. Personality factors da opiniao e da atitude. Boletin do hstituto de
in attitude change. Unpubl. P h .D. dissertation, Psicoloffia,8, p. 11-21.
University of Michigan, Ann Arbor. TITUS, H.E.;HOLLANDER,E.P. 1957. The
NADLER, E.B. 1959. Yielding authoritarianism California E' scale in psychological research:
and authoritarian ideology regarding group. 1950-1955. Psychological bulletin, 54,p. 47-64.
Journal of abnormal and social psychology, WAGMAN,M. 1955. Attitude change and authorita-
58, p. 408-10. rianpersonality. Journal ofpsychology,40,p.
3-24.

21
 CHAPTER 2

GROUP-ORIENTED R E S E A R C H

Since the determinantsof attitude change are complex social psychology, that the importance of social
and highly interrelated,any attempt at classifying influences upon the individual was fully realized
research resultsinto various categoriesmust neces- (e.g. F.H.Allport, 1924; Bechterew anddelange,
sarily be arbitrary. W e have seen,for instance in the 1924). A s Lewin (1935) has shown, this develop-
last chapter that certain group-oriented research has ment in psychology parallels a similar earlier
relevance to the question of individualdifferences or development in physics in the transition from
personalityfactors. O n the other hand, in the very Aristotelian to Galilean modes of thought: In Aris-
extensive research on interpersonalcommunications totelian physics, the tendency of a light object to
as a means of changing attitudes,the group situation go upwards and of a heavy object to fall downwards
in which such communicationsare presented is of was explained on the basis of inherent propensities
great importance. I nconsidering an investigation of within the objects themselves; since Galileo, we
the effectsof group discussiononattitude change fol- know that the behaviour of a body is a product of
lowingthe presentation of a film,both group pro- the total field forces acting upon the body at any
cesses and the effects of mass communication media given time and that, for instance, even the weight
are involved. Furthermore in the extensive literature of an object is not absolute but is relative to the
on conformity to social norms and the effects of object's surroundings. Similarly, we know today
social pressure, a distinction is often not made that an individual considered in a vacuum, i.e.
between social pressure in the direction of confor- apart from the social influences of his environ-
mity to the norms of smaller primary or secondary ment, is a fiction.
reference groups, or to the norms or expected It is neither possible,nor indeed necessary, to
behavioural patterns of a community or the society sketch further here the vast body of well-known
as a whole. Our attention here will be directed research which has been amassed to demonstrate
primarily to the influence of the group, in the nar- this point. W e shall consider rather some of the
rower socialpsychological sense of the term, upon more important research bearing directly on the
attitude change. question of group influence upon attitude change.
The recognitionof the importance of the influence Amongthe most classical studies of the influence
of the group and the socialenvironment generallyupon of the group situation upon the formation and change
the individualmaybe considered one of the most im- of attitudes and norms are those of Sherif (1936,
portant landmarks in the progress of the scientific 1937), using the autokinetic phenomenon. When
study of human behaviour. In the Nineteenth century, a single small source of light is presented in an
when scientificpsychology m a y be said to have had its otherwise completely dark room, it appears after
beginnings,the emphasis of research was primarily a short time to move in an erratic and irregular
upon general laws to explain the mental (primarily fashion, since there is nothing in the room in
congitive) functioning of human beings, with little reference to which one can locate it. If asked to
regard being paid to individualdifferences. Toward estimate the distance through which the light
the end of the Nineteenth century the emphasis appearedto move, the subject is at a loss for any
shifted to individual differences; and the study of objective standard of comparison. Under these
characterology, and later personality theory, circumstances,Sherif demonstrated that subjects
flourished. As a reaction to the previous trends, tend to establish a subjective range of extent (a
individuality was over-emphasizedby some schools, scale), and a point within that range (a standard
to the neglect of, or even denial of, lawfulness in or norm), whichis peculiar to the individual. Thus
human behaviour. This differing emphasis between the subject establishes for himself a frame of re-
the individual and the general has been charac- ference in an otherwise unstable perceptual field.
terized by G.W.Allport (1937) as the ideographic When this same experimentis conducted in a group
versus the nomothetic approach (adapted from the situation, whereby each individual can hear the
terminology of Windelband). orally given judgement of the other members of
Early approaches to social psychology often the group, a group norm tends to become esta-
reflectedthis same trend,explainingsocial behav- blished within a short period of time, to which the
iour in terms of inherent tendencies within the individual members of the group conform in their
individual, such as instinct, sentiment, etc. (e.g. judgements. If, furthermore,the other members
McDougall, 1908). It m a y be safely said that it was ofthe group are confederates of the experimenter,
not until the 1920's. as modern empirical methods who gives predetermined judgements,the experi-
began to be applied to an appreciable extent in mental subjects will usually conform to this
22
 Group-Oriented Research

pre-established norm. A n interesting finding of conducted, but the general result was that a large
this latter experimental situation is that the con- percentage of individuals yielded to the pressure
vergence toward the pre-determined norm does not of the unanimous majority, even though this major-
disappear when the subject is alone in a subsequent ity Wac,very clearly in error. In other words,
session. In fact, on the whole, convergence toward the pressure of majority opinion was so great that
the prescribed norm is greater in the session alone the individual responded with judgements contrary
than in the initial session when the social influence to that which he felt to be correct. In many sub-
was present. jects there occurred a process whereby they ration-
That the phenomenon described by Sherif is of a alized this discrepancy, thus making it possible
general nature and not dependent upon the particu- to conform to the majority opinion.
lar experimental situation is demonstrated by the W e have discussed these results of Sherif and
fact that similar findings have been found by ex- Asch in such detail for very good reason: if the
perimenters using different stimulus modalities, -
effect of such experimentally created groups i.e.,
such as cutaneous perception of warmth (McCord, persons who were unknown to the subject previously
1948), duration of perceptual phenomena (Sinha, and who are of no particular significance to him-
1952). estimates of size (Bovard, 1951; Mausner, upon the norms and attitude of the individual is so
1954) and of number (Bovard, 1953; Sodhi, 1953) compelling, how much more compelling must the
and aesthetic judgement (Mausner, 1953). Sodhi effect on the individual be of groups who have very
(1953)has furthermore shown that this tendency definite emotional and practical significance for
towards convergence is a function of the level of him? W e might logically expect that the effect of
difficulty in making judgements; as the level of such reference groups upon the individual will be
difficulty increases, convergence towards a com- even greater.
m o n norm is more marked. One of the most impressive investigations of
The implications of the above results for the the effect of reference groups upon attitude change
formation and change of social attitudes generally is the well-known 'Bennington Study' (Newcomb,
are quite obvious. Sherif and Sherif (1956,p. 488) 1943, 1948). In this investigation, the entire
in discussing these implications conclude: 'The student body of a small liberal arts college in the
experiments ...showin a simple waythe basicpsy- United States was studied (morethan 600individuals).
chological process involved in the establishment Bennington College was noted for the generally
of social norms. The psychological basis of estab- liberal atmosphere which prevailed with respect
lished social norms such as stereotypes,fashion, to a wide range of social attitudes. The students
conventions,customs and values is the formation came mostly from families of upper-middle or
of c o m m o n reference points or anchorages as a upper class socio-economic status whose social
product of interaction among individuals. Once attitudes were rather conservative. Systematic
such anchorages are established and internalized study showed a progressive change in social atti-
by the individual they become important factors in tudes from conservative to liberal during the four
determining or modifying his reactions to the situ- years of study at this school. Furthermore, the
ations that he will face later alone. ...' change was proportionate to the extent to which
Also among the most compelling evidence for the college community, which was the membership
the influence of the group upon the individual atti- group, became the reference group for the students.
tude is the well-known work of Asch (1951, 1952). A follow-up investigation (Newcomb, 1950) showed,
These experiments were concerned with conditions moreover, that the longer the residence in this
under which individuals either resist, or are in- 'closely knit, integrated community' the greater
duced to yield to, group pressures when the latter was the change in the liberal direction and the
are perceived to be contrary to fact. The following greater was the persistence of the changed attitudes
experimental technique was used in this series of (ibid.,p. 206 et seq.).
studies. In a group situation the subjects had the These findings, and particularly Newcomb's
task of judging rather simple, clearly structured, distinction between membership groups and refer-
perceptual relations: they were to match the length ence groups, have very important practical impli-
of a given line with one of three unequal lines, The cations and have been confirmed by subsequent
judgements were given orally so that each subject investigations. Siegel and Siegel (1957) studied
was aware of the judgements given by the other attitude changes among women university students
members of the group. In fact, however, all m e m - which occurred over time when reference groups
bers of the group except one, the experimental sub- and membership groups were identical and when
ject, were confederates of the experimenter and they were disparate. When divergent membership
had instructions to respond at designated times with groups with disparate attitude norms were socially
- -
wrong and unanimous judgements. The magni- imposed on the basis of a random event, attitude
tude of the errors made by the majority was con- change in the subjects over time was a function of
siderablygreater thanthat which could be attributed the extent to which the subjects came to take the
to chance. Thus the experimental subject found imposed, initially non-preferred , membership
himself in a situation of being a minority of one in group as their reference group.
the midst of a unanimous majority. A number of An earlier study by Sims and Patrick (1936)had
variations of this experimental technique were shown that this factor of conformity to the norms

23
Group-Oriented Research
of the group of which one is a member, and which they develop in school and in university settings,
one comes to a greater or lesser extent to take as m a y well become reference groups and have a
a reference group, m a y result in attitude change positive effect in changing attitudes, thus counter-
in a negative direction as well. These authors acting the 'culturallag' which is so apparent in our
studied the attitudes of white students from society today. Nowak (1960)has demonstrated this
Northern homes in the United States attending a fact very clearly in his study of egalitarian attitudes
Southern college. The first-year students in this of Warsaw students. This author found that socio-
group held attitudes only slightly less favourable economic status and occupational categories acted
toward the Negro than those of Northern college as a reference group for the students (the higher
students in their first year. However, by the third the reference group is located in the stratification
and fourth year of residence in the Southern college, of income, the less egalitarian the attitudes are);
the students from Northern homes had shifted their on the other hand, he also found a weakening over
attitudes towards Negroes in a negative direction time of the association between egalitarianism and
very nearly to the position of their Southern class- parents' income. The author postulated here the
mates. operation of a process of socialization or adapta-
M o r e recent research has confirmed the fact tion to the standard and norms of the respective
that the influence of the reference group can change school environment and desocialization from the
the individual's attitude in either a positive or a family environment. H e found the r61e of face-to-
negative direction. Kelley and Woodruff (1956)con- face groups to be especially important in attitude
ducted an experiment with a group of students at a change. For instance the degree of correlation
teachers' college which demonstrated this point. between egalitarianism and socio-economic back-
By reason of their training these students were ground was less among those students living in
committed to positive attitudes toward modern, as dormitories,i.e., in more face-to-facecontact
opposed to traditional, teaching practices. These with their fellow students.
attitudes were a part of their firmly held group In addition to school and university groups,
norms. The experimenters presented the students membership in a number of different groups m a y
with a recorded speech which argued in favour of give these groups, to a certain extent, the charac-
the use of traditional rather than modern teaching ter of reference groups and thus change the atti-
practices, thus going counter to the established tudes of their members. Brophy (1946). for instance,
group norms. Several times during the speech found large differences in attitudes toward Negroes
a main point made by the speaker was followed by among white seamen belonging to three different
recorded applause. The experiment was conducted trade unions who differed with respect to their
in a series of small group sessions, whereby some anti-discrimination policies. Gundlach (1950)
of the groups were told that the speech had been investigated the attitudes of white female factory
recorded at a meeting of the faculty of their school, workers in N e w York City who belonged to a 'left
whereas other groups were told that the speech was wing' trade union with a militant anti-discrimina-
recorded at a general public meeting in another tion policy and found them to have a surprisingly
community. Thus the former thought that their low degree of prejudice toward Negroes. The fac-
own teachers (who were part of their reference tory workers investigated by Gundlach were also
group) were applauding points contrary to their significantly less prejudiced than white housewives
norm, whereas the latter thought that the applause in a non-segregated N e w York City housing project
was coming from people w h o m they did not know. (Deutsch and Collins, 1951). In another investiga-
The results showed that those students who were tion, Watson (1950) studied 45 N e w York City resi-
told that the applause was coming from their own dents who reported that there had been a marked
faculty changed their attitudes more in favour of change in their attitudes toward Negroes or Jews.
traditional teaching practices than did those stu- In intensive interviews with these subjects, it was
dents who thought that the applause was coming found that more than half of them reported that the
from an unknown group. change had been preceded by their entry into some
A large number of further investigations have new institutionalized group whose standards were
shown that membership in peer groups on the part different from their previous ones.
of schoolchildren and university students m a y have What happens when attempts are made to induce
an increasingly greater effect upon changing the attitudes which are contrary to the group norms?
attitudes of the subjects as these membership Kelley andVolkart(1952)have shown that attempts
groups increasingly become reference groups for to change group-anchored attitudes usually meet
the individuals. The conclusion that m a y be drawn with resistance, the amount of resistance usually
from these studies is that although negative inter- depending upon the degree to which the group in
group attitudes most frequently develop as a result which the attitudes are anchored is considered by
of adopting the norm of the family unit as the pri- the individual to be a reference group. In a further
mary group, and programmes of action might well investigation,Kelley(1955) demonstratedmore fully
direct attention to parents as instruments of atti- the positive relationship between 'salienceof m e m -
tude change, we need not be discouraged by the bership'and resistanceto change of group-anchored
obvious difficulties which such a programme of attitudes. These results have likewise been
action presents; for peer groups, especially as confirmed by Furuya (1958) in a different cultural

24
 Group-Oriented Research
setting; however, this author's findings suggested frustrating relations with one another. A s a
a more complex hypothesis, i. e. that pressure in result,feelings of in-group solidaritydeveloped,
a group situation to change group-anchored norms but also unfavourable stereotypes came to be
m a y succeed if the pressure is great enough, but used with regard to out-group members. It
there then results a stabilization of the existing might be said that negative intergroup attitudes
attitudes which are resistant to later change. This had developed. (It must be noted, however,
point has also been demonstrated byKelman(1952). that such attitudes do not necessarily develop;
Another important line of group-oriented re- here frustrating situations were deliberately
search on attitude change has been that associated created experimentally.)
with the notion of 'groupdecision' (Lewin, 1953, 3. Stage of reduction of intergroup friction: as a
1958). In these studies, which had their origin means of improving intergroup relations and
during World W a r I1 with the objective of changing changing negative attitudes and stereotypes, the
the food habits of American families to include two groups were brought into situations in which
items usually not found desirable by them, for in- desirable super-ordinate goals existed which
stance beef hearts, kidneys, etc. , a 'lecture m e - necessitated co-operative inter-action between
thod' was compared with a 'discussionmethod'. In the two groups. Following this, a decrease in
the lectures, various appeals were used: informa- intergroup tension and a change in the negative
tion was provided, instruction was given on attrac- attitudes and stereotypes resulted.
tive ways of preparing these cuts of meat and The hypotheses were confirmed that 'when groups
mimeographed recipes were distributed. In the in a state of friction are brought into contact under
discussion group, information was also given, but conditions embodying super-ordinate goals, which
the groups were then stimulated to discuss theissue are compelling,but which cannot be achieved by
and arrive at a group decision. A follow-up inves- the efforts of one group alone, they will tend
tigation to see whether or not the subjects had to co-operate toward the c o m m o n goal, and co-
actually included these food items in their meals operation between groups necessitated by a series
showed that whereas only 3 per cent of the subjects of situations embodying super-ordinate goals will
exposed to the lectures had been influenced to in- have a cumulative effect in the direction of reduc-
clude these food items in their menus, 32 per cent tion of existing tension between groups'. (ibid.,p.
of the individuals in the discussion groups had done 318 et seq.) . The findings of this ingenious study
so. More recent investigations(Levine and Butler, have obvious implications for intergroup relations
1953) have confirmed these results, showing the and attitude change on a larger scale.
compelling influence of 'group-decision' on the In conclusion, we have seen rather compelling
attitudes and behaviour of the individual. evidence that an individual's attitudes do not form
However, Murphy's (1953) comparative studies and persist in a vacuum but are dependent to a
in the setting of the Indian culture have pointed to large degree upon the attitudes and norms of the
the possibility that these results are dependent groups which form his frame of reference. In the
upon cultural norms. In experiments attemptingto following chapter we shall see further evidence of
change caste attitudes of secondary school students social influence upon the attitudes of the individual
in India using various methods, lectures employing through the effect of interpersonal communications
emotional appeals brought about greater changes and mass communication media.
than the discussion method. Further cross-cultural
research is necessary in order to determine the
significance of these differences. BIBLIOGRAPHY
One of the most interesting examples of an ex-
perimental approach to the development and change ALLPORT, F .H . 1924. Social psychology. N e w
of intergroup attitudes is the so-called 'Robbers York, Houghton Mifflin.
Cave Study' by Sherif and co-workers (Sherif, ALLPORT, G.W. 1937. Personality: a psycho-
Harvey, White, Hood and Sherif, 1954). The inves- logical interpretation. N e w York, Henry Holt .
tigation was carried out in Robbers Cave State Park A S C H , S.E. 1951. Effects of group pressure upon
in Oklahoma, U.S.A.; subjects were 22 boys of the modification and distortion of judgement.
about 11 years of age,comingfrom the established, In: H. Guetzkow (ed.). Groups, leadership
middle socio-economic class and having above- and men. Pittsburgh, Carnegie Press.
average IQ. All were healthy, socially well-adjusted , 1952, Social psychology. N e w York,
boys. The experiment was conducted in three Prentice Hall.
stages: .1956. Studies of independence and
1. The stage of intergroup formation: the subjects conformity. (I) A minority of one against a
were divided into two groups (of matched pairs) unanimous majority. Psychologicalmonographs,
and allowed to form group structures and norms 70. no. 9 (whole no. 416).
of their o w n naturally and developed in-group BECHTEREW, V.M.; DE LANGE,M. 1924.
feelings. The two groups adopted the names Die Ergebnisse des Experiments auf d e m
'Rattlers'and 'Eagles',respectively. Gebiete der kollektiven Reflexologie. s-
2. The stage of intergroup friction and conflict: schrift fur angewandte Psychologie, 24, p.
the two groups were brought into competing and 305-44.

25
Group-Oriented Research

BOVARD, E.W.Jr. 1951. Group structure and M A U S N E R , B. 1954. The effect of prior rein-
perception. Journal of abnormal and social forcement on the inter-action of observer
psychology, 46, p. 398-405. pairs. Journal of abnormal and social psy-
. 1953. Conformityto socialnormsin chology, 49, p. 65-8.
stable and temporary groups. Science, 117, p. M c C O R D , F. 1948. The formation of group
361-3. norms. Journal of social psychology, 27,
BROPHY,I.N. 1946. Theluxuryofanti-Negropre- p. 3-15.
judice. Public opinion quarterly,10, p. 456-66. M c D O U G A L L , W. 1908. Introduction to social
CARTWRIGHT. .~ D.:ZANDER. A. (eds.). 1953. psychology. London, Methuen and Co.
Group dynamics: research and theory. Evanston, MURPHY,G. 1953. In the minds of men. N e w
Illinois, R o w Peterson. York, Basic Books.
D E U T S C H ,M.;COLLINS, M.E. 1951. Inter-racial NEWCOMB, T.M. 1943. Personality and social
housing: a psychologicalevaluation of a social change. N e w York, Dryden.
experiment,Minnesota,University of Minnea- . 1948. Attitude development as a function
polis Press. of reference groups. In: M. Sherif (ed.). I&
FURUYA, T. 1958. An experimental study of outline of social psychology. NewYork, Harper.
resistance to the change of attitudes. Japanese
~ ~ ~~~
. 1950. Social psychology. N e w York,
journal of psychology, 28, p. 260-8. Dryden Press.
GUNDLACH.R.H. 1950. Effects of on-the-job NOWAK,S. 1960. Egalitarian attitudes of Warsaw
experience with Negroes upon the racial students. American sociological review, 25,
attitudes of white workers in union shops. p. 219-31.
American psychologist, 5 , p. 300 (abstr.). SHERIF, M. 1936. The psychology of social
KELLEY,H.H. 1955. Salience of membership and norms. N e w York, Harper.
resistanceto change of group-anchored attitudes. . 1937. An experimental approach to the
H u m a n relations, 8, p. 275-90. study of attitudes. Sociometry, 1 , p. 90-8.
;, VOLKART,E.H. 1952. The resistance ; HARVEY, O.J.; WHITE, B.J.; HOOD,
to change of group-anchored attitudes. W.R.; SHERIF, C .W. 1954. Experimental
American sociological review, 17, p. 453-65. study of positive and negative inter-group
; W O O D R U F F , C.L. 1956. Members' attitudes between experimentally produced
reactions to apparent group approval of a groups: Robbers Cave study. Norman, Okla.,
counter-norm communication. Journal of University of Oklahoma (multilithed).
abnormal and social psycholopy,52, p. 67-74. ; SHERIF, C.W. 1953. Groups in
LEVINE,J.; BUTLER,J. 1953. Lecture versus harmony and tension. N e w York, Harper.
group decision in changing behaviour. In: D. ___ . 1956. A n outline of social
Cartwrightand A. Zander (eds.).Group dynamics: psychology, (rev. ed.). N e w York, Harper.
research and theory. Evanston, Illinois, R o w SIEGEL, A.E.; SIEGEL, S. 1957. Reference
Peterson. groups, membership groups, and attitude
LEWIN,K. 1935. Adynamic theory of personality, change. Journal of abnormal and social
N e w York, M c G r a w Hill. psychology, 55, p. 360-4.
. 1953. Studies in group decision. In: B. SIMS. B.M.; PATRICK,J.R. 1936. Attitude
Cartwright and A. Zander (eds.). Group dyna- toward the Negro of Northern college
I -
mics: research and theory. Evanston, Illinois, students. Journal of social psychology, 7,
R o w Peterson. p. 192-204.
. 1958. Group decision and social change. SINHA, D. 1952. An experimental study of a
In: E.E.Maccoby, T.M.Newcomb, and E.L. social factor in perception: the influence of
Hartley (eds. ).Readings in social psychology,3rd an arbitrary group standard. Patna Universitx
ed., p. 197-211. NewYork, Holt, Rheinhart and journal, Patna.
Wilson. SODHI, K.S. 1953. Urteilsbildung im Sozialen
M A U S N E R , B. 1953. Studies in social inter-action. Kraftfeld, mttingen, Verlag fur Psychologie .
(111) Effect of variation in one partner's pres- W A T S O N , J. 1950. Some social and psycho-
tige on the inter-action of observer pairs. logical situations related to change in attitudes.
Journal of applied psychology, 37, p. 391-4. Human Relations, 3, p. 15-56.

26
 CHAPTER 3

PERSUASIVE C O M M U N I C A T I O N S RESEARCH

A s the major textbooks of social psychology cor- groups of subjects, an experimental and a control
rectly point out, communication is the cornerstone group.
of man's social behaviour. It is primarily the As earlier surveys (Murphy, Murphy and N e w -
ability to communicate on a symbolic plane through comb, 1937; Williams, 1947)have shown, most
the use of language that differentiatesHomosapiens such experiments did show some change of attitude
from other species and makes the human being a in the direction of the communication, as measured
socialanimal. Any attempt to persuade, change the by the instruments. However, the studies showed
attitude of, or otherwise influence our fellow m a n little agreement as to the 'best'method of changing
must necessarily make use of communication. Thus attitudes through communication; and not a small
communication would seem to be basic to our whole number of them showed no significant change, or
subject matter, Indeed, logically speaking, the even change in the opposite direction to that which
research which we shall cover here cuts across was intended. The reason for these variable re-
nearly all of the various disciplines. On the one sults is very probably to be found in the fact that
hand, attempts at attitude change through inter- the design of such research did not take into con-
personal communicationsinvolvethe personality of sideration knowledge of interpersonal relations,
the recipient of the communication and of the com- group processes,communicationtheory,and other
municator himself, as well as the relationship factors involved in the process of attitude change,
between the communicator and his audience and the about which w e know considerably more today.
dynamics of the group process manifest in the N o attempt will be made to review these pre-
experimental situation. On the other hand, as we vious studies. The emphasis here will be upon
have alreadypointed out, a distinctionis often not more recent research on the effectiveness of
made between interpersonalcommunications,where communications in attitude change, which has taken
there is face-to-face contact between the com- some of these more differential factors into con-
municator and his audience, and communications sideration. Early research will be referred to only
through mass media where such contactislacking. occasionally where it is necessaryto do so in order
The question of mass media as a means of attitude to elucidatethe issues involvedin current research.
change raises, in turn, broad issues which must Without attempting any complete analysis of the
be considered from the viewpoint of the society or various aspects involvedinthe effects of communi-
culture as a whole. Nevertheless, an inspection cation upon changing attitudes, we shall consider
of the literature on attitude change very clearly differentially the effects of the following factors:
shows a large body of research centred around the communicator; the communication; and the
what m a y be called persuasive communication, W e recipient of the communication. Ln the following
shall therefore attempt to sketch here the major chapter w e shall review some theoretical problems
recent trends in this area which has grown so involved in the process of attitude change through
rapidly in very recent years. communication.
A s has been pointed out by a number of writers,
earlier research on changing attitudes through The Communicator
communication was characterized by a 'scarcityof
well-planned experiments in the bulk of research Although it would appear a priori rather obvious
reports' (Sherif and Sherif, 1956, p. 555). This that the factor of who says or presents a c o m m u -
research actually has a rather long history, as far nication (or in the case of mass media, to which
as the relatively young science of psychology is source the communication is attributed) must be
concerned. A survey of the literature reveals of great importance, this factor was surprisingly
a fair amount of such research as early as the neglected in earlier research. This neglect m a y
1 9 2 0 ' ~with
~ the number of studies increasinginto well have been responsible for some of the discrep-
the 1930's. The general form of these studies has ancies in the findings of previous studies. Thus,
been to use some standard instrument for attitude for instance, Hovland (1958), in reviewing research
measurement, such as a Thurstone or Likert-type on the r61e of primacy and recency in persuasive
scale, in a 'before-after'design, with the influence communication (which w e shall discuss at greater
procedure (writtenmaterials, speeches, discus- length below), suggests that the discrepancy be-
sions, debates, motion pictures, courses of study, tween the findings of an earlier investigation by
etc.) presented in between. Experiments which Lund (1925) and those of a more recent study
were reasonably well-planned usually used two by Cromwell (1950), using the same experimental

27
Persuasive Communications Research

design, m a y be in part attributable to the effects impartial had a considerably greater effect upon
of the communicator; whereas in Lund's study the changing attitudes than did the one whose motives
experimenter was also the instructor of the sub- were suspect. A number of recent investigations
jects, thus increasing the motivation to learn the have confirmed these findings as to the important
first-presented communication, Cromwell, in his effect to 'communicatorcredibility' upon attitude
investigation, used a neutral experimenter (in an change (e.g., Fine, 1957); Ludlum, 1958; Choo,
attempt to minimize the effect of this factor), thus 1960; etc.). Pastore and Horowitz (1955) have
arriving at results which did not confirm Lund's also demonstratedthe influence of the motive attrib-
findings as to the importance of primacy in the uted to the communicator on the acceptance of
presentation of the material. a statement.
More recent experimentalinvestigationsdesigned Onthe other hand, a number of studies indicate
specifically to study the influence of the c o m m u - that the effects of high-credibility sources are not
nicator or source of the communication have con- very enduring. Hovland and Weiss (1951) and
firmed the significance of this variable. Hovland Kelman and Hovland (1953) found in follow-up
and Weiss (1951). using college students as experiments that the differential effectiveness of
subjects, found that by varying the source of com- high-credibility sources had disappeared after a
munication,the same communication had a greater period of afew weeks. Hovland,Janis and Kelley
effect when the subjects regarded its source as (1953) conclude: 'The main implication of the pre-
trustworthy and reliable than when they considered sent results is that the credibility of the c o m -
the source to be untrustworthy. In one case the municator m a y under certain circumstances ...
students were told that the author of the com- be important only with respect to the amount of
munication was a prominent scientist in their o w n immediate opinion change produced' (p. 39). How-
country, and in the other the source was identified ever, as these authors point out, the effect of the
as being a State-controlled newspaper of another communicator m a y be much more lasting when a
nation which was not regarded by the students as very close association exists between the source
being trustworthy (only the subjective evaluation of and content of the communication. When, for in-
the subjects is important; the actual reliability of stance,the source ofthe communicationconstitutes
the sources is not under discussion here). These some form of reference system for the subject,
results have been confirmed in another study by and when the content of the communication seeks
Kelman and Hovland (1953), using high school stu- to change attitudes in the direction of conformity
dents as subjects. These subjects were presented to norms of this reference system, the effect of
with a recorded educational radio programme, the source is likely to be very great. This is es-
whereby three different speakers or 'communica - pecially the case when the source of the communi-
tors' took an identical stand on the issue of how to cation is an important reference person for the
treat juvenile delinquency,all advocating a policy subject (e.g., a communication from a parent call-
of greater leniency. The three speakers could be ing upon a child to change his behaviour in some
characterized as 'positiveI , 'negative',and 'neu- manner or other has very direct implications in
tral'. The 'positive'communicator (identified as terms of the relationship between the child and the
a highly respected, well-informed judge in a juve; parent and is likely to be followed) or a reference
nile court) had considerably more effect on chang- group. An example of the latter is the study by
ing the subjects' attitudes than did the 'negative' Kelley and Woodruff (19561, discussed in the last
communicator (identified as a rather 'shady' chapter, in which attitude change on the part of
and not very law-abidingindividual); the 'neutral' students was much greater in the case where the
communicator (identified as a m e m b e r of the radio source of the communication is perceived as being
audience who was chosen at random) had an effect approved by the student's o w n faculty than when it
which was intermediate between the other two. was approved by persons unknown to them. In this
Another study by Hovland and Mandell (1952) has case probably both prestige factors and the influ-
shown that the subject's evaluation of the c o m m u - ence of reference group norms were involved.
nicator in terms of his impartiality significantly In a study on the effects of 'feed-back'from the
affects the degree of attitude change achieved by experimenter on conformity behaviour, Jones,
the communication. In this investigation students Zell and Torrey (1958)have shown that the beha-
were presented with a general discussion of the viour of the communicator in the course of theinter-
American monetary system, leading to a conclu- action situation (for instance, in terms of approval
sion in favour of devaluation of the currency. In or disapproval) will have an important effect upon
two variations of the experimental procedure, an the subject in terms of his conformity in the direc-
introduction to the communicator was given which tion of the communication'scontent.
in one case led to a suspicion of his motives (i.e. , A n interesting study showing the intimate rela-
he was introduced as the head of a large importing tionship between communicator and the content of
firm who would profit from devaluation), whereas the message is that of Tannenbaum (1956). This
in the other case the communicator 's impartiality author found that not only did subjects tend not to
was emphasized(i.e., he was introduced as a well- accept statements attributed to persons toward
known professor of economics). The results showed w h o m theyhad negative feelings,but vice versa, if
that the communicator who was regarded as being the subject originally favoured the view presented,

28
 Persuasive Communications Research

his attitude toward the source to which the view was with their views and ignore those which are not in
attributed became more favourable. For example, accordance with them. H y m a n and Sheatsley (1947),
subjects with negative feelings toward labour in their analysis of 'some reasons why information
leaders tended to reject arguments in favour of campaigns fail', have shown that with the same ex-
legalized gambling when these were attributed to the posure to various mass media the amount of infor-
labour leaders. On the other hand, subjects who mation actually absorbed concerning certain issues
originally favoured legalized gambling developed was directly related to the amount of interest in
more positive attitudes towards labour leaders upon them. Thus information on an issue in which the
reading a communication favouring their view which person was not interested was not absorbed. More
was attributed to labour leaders. recent evidence of such selectivity has been pro-
vided by Mills, Aronson and Robinson (1959). A s
Communication Flowerman (1949)has poiqted out,informationwhich
runs counter to established group norms is also not
Shifting our focus of attention now from the c o m - likely to have much effect on recipients who are
municator to the communication itself, we are led members of these groups no matter how 'factual'
to ask not only who said it but also what was said or 'scientific'it might be. Inthe last chapter w e dis-
(or communicated) and how. Specifically we shall cussed experimental evidence demonstrating that
consider: (a) the content of the communication; 'group anchored' attitudes are resistant to change
(b) the nature of the appeal; (c) the manner of pre- (KelleyandVolkart,1952; Kelley, 1955).
sentation; and (d)the medium of communication.
Nature of appeal. A major problem in considering
Content of the communication. That the effect what type of appeal should be used in a communi-
of a communication upon the attitude of the recip- cation designed to produce attitude change has been
ient is dependent among other things upon the that of the relative effectiveness of 'emotional'
actual content of the communication is rather versus 'rational'approaches. ln the past, many
apparent. Especially in the area of mass c o m m u - programmes of action designed to change attitudes
nications media, concerning which a vast body of have been based on the assumption that negative
research results has grown up in recentyears which intergroup attitudes are merely a result of 'false
we cannot begin to review in detail here, the ques- information', and that the most effective way of
tion of the content of the communication has re- changing negative stereotypes and prejudice is the
ceived considerable attention. For a more detailed dissemination of 'factualinformation'.
description of the technique of content analysis in There is no question about the desirability of
communication research, the reader is referred distributing information based upon scientific facts
to the excellent review by Berelson (1954). To as widely as possible, and undoubtedly such efforts
mention just a couple of illustrative examples as have an important r61e to play in changing negative
an indication of the importance of this line of intergroup attitudes. O n the other hand, w e rea-
research, Berelson and Salter (1946) analysed lize today that the matter is not quite as simple as
short stories in widely circulated American maga- all that. Learning theorists know that neurosis is
zines and found that members of the 'majority'were based upon certain 'false'or maladaptive, learned
typically shown to be 'nice people', heroes, etc., associations. However, psychotherapists know
whereas members of minority groups were most fre- that presenting a neurotic with a book or pamphlet,
quently shown in negative rbles; and Shuey (19531, or giving him a lecture on the nature of neurosis,
in an analysis of pictures in magazine advertise- is not, in itself, likely to result in a change of the
ments, found members of a discriminated minority neurotic behaviour. If matters were so simple, the
consistently presented in inferior social and occu- process of psychotherapy and the entire field of
pational rdles (indisproportiontothe actual distri- mentalhygiene would be a very easy matter indeed.
bution of occupations according to census figures). Without necessarily implying that this analogy
A large number of other investigations have like- directly parallels the problem of prejudice or neg-
wise shown how such content of mass communica- ative intergroup attitudes, we know that presenting
tion media helps to perpetuate popular stereotypes a highly prejudiced person with a book or pamphlet,
and thus retard progress toward complete equality. or giving him a lecture designed to 'correct'his
Apart from this type of analysis, it is necessary 'false'ideas, will not necessarily be successful in
also to consider the effect of the content of a c o m - achieving the desired goals. For one thing, he is
munication in relationship to the recipient and his very likely not to read the printed material or at-
existing attitudes. An important limitation of the tend the lecture, or if he is forced to do so, he is
effects of mass communication upon attitude change likelyto misperceive or failto perceive the infor-
m a y be seen in the selectivity of the recipients. mation presented to him or in some other manner
Lazarsfeld, Berelson and Gaudet (1944)have shown, engage in 'resistance' (cf. Cooper and Jahoda,
for instance, that during a political campaign those 1947).
people who have a very definite opinion on a parti- O n the other hand, in the course of recent his-
cular issue or candidate selectively react to com- tory we have often enough been witness to the sad
munications in such a manner as to attend to news- fact that highly emotional appeals on the part of
papers, speeches, etc. which are in accordance bigots and despots have been very successful in

29
Persuasive Communications Research

changingthe attitudes of large numbers of people and out inequities or to admonish the majority group
of movingthem to engage in actions which were any- for their discriminatory treatment of minorities.
thing but conducive to healthy intergroup relations. A s Feshbach (1961, p. 12) has pointed out, 'the
In attempting to change negative intergroup atti- effectiveness of an emotional appeal will in part
tudes, which appeal, then, should w e use? Should depend upon the particular behaviour that the c o m -
w e resort to the emotional appeals of the rabble municator is interested in encouraging'. Referring
rouser? H o w can w e get round the obstacle of to evidence put forward by Taylor (1956) showing
resistance to purely rational appeals? Past exper- that the arousal of anxiety tends to interfere with
imental research on this question has produced complex learning, but to facilitate simple learning,
contradictory results. F o r instance, Hartmen Feshbach concludes: 'In order for the strong
(1936) and Menefee and Granneberg (1940) have appeal to have maximal effects,the response to be
reported experimental data showing that c o m m u - encouraged should not only be a simple behavioural
nications with 'emotional' appeals were relatively act but should also be a relatively immediate one ...
m o r e effective than those using purely 'rational' if time is permitted to elapse between the presen-
appeals. O n the other hand, Knower (1935) arrived tation and the performance of the desired behaviour
at results showing that 'emotional' appeals m a y be there will be a greater opportunity for defensive
less effective than 'rational' ones. O n e reason for behaviour to occur. ... Hence, where there is a
such conflicting results is very probablyto be found concern for the audience's future as well as i m m e -
in the lack of unambiguous definition as to what diate behaviour, a minimal anxiety arousing appeal
constitutes 'rational' or 'emotional' appeals. M o r e again takes precedence over a strong anxiety arous-
recent research by Janis and Feshbach (1953, 1954). ing communication' (Feshbach, 1961, p. 12).
Feshbach and Singer (1957) and Feshbach (1961) Although these results should not be over-genera-
have dealt with the problem of one particular type lized, they must be taken into consideration in
of emotional appeal, namely fear-arousing or deciding upon the type of appeal to be used in c o m -
threat appeals. Janis and Feshbach(l953) demon- munications designed to produce attitude change.
strated that communications designed to change The precise nature of the communication which
attitudes by m e a n s of threat m a y elicit defensive would be most appropriate will depend, of course,
responses which reduce the intended effect of the upon the particular concrete situation in which it
communication. Feshbach and Singer (19571, is to be employed and upon the specific goals that
referring to the earlier studies and pointing also to are to be achieved.
clinical evidence showing that threat m a y elicit
hostility as well as avoidant behaviour, examined Manner of presentation. O n e of the mpst classical
'conditions under which threatening situations questions concerning the relationship between the
elicit hostility and effect the expression of social effectiveness of a communication and its manner
prejudice'. The results of their investigation of presentation has revolved around the question of
confirmed the hypothesis that communications per- the relative supremacy of 'primacy' or 'recency'.
ceived as a personal threat result in an increase In an earlier investigation, Lund (1925) presented
in social prejudice. At the s a m e time, however, college students with mimeographed communica-
their investigation showed that communications tions supporting first one side and thenthe opposite
interpreted as a shared threat resulted in a decrease side of a controversial issue. The results of this
in social prejudice. These results are suggestive experiment showed that the side of the issue pre-
of those of Sherif, Harvey, White, Hood andSherif sented first had the greatest effect upon the atti-
(1954), who demonstrated that activity directed tudes of the students. O n the basis of these results,
toward super-ordinate goals requiring the co- Lund enunciated a L a w of Primacy in persuasion,
operation of groups led to a reduction of aggressive stating as a general principle t h
t the side of an
tension between the groups. issue presented first will have a greater effective-
These results, providing empirical verification ness than the side presented subsequently. H o w -
of hypotheses derived from clinical work as to ever, Cromwell (1950), in a later investigation,
defensive reactions to fear or threat, must be found a significantly greater change produced in
seriouslytaken into considerationbefore any attempt the direction of that side of the issue presented
is m a d e to change intergroup attitudes by m e a n s of last, thus obtaining a recency effect rather than a
such appeals. Extrapolating from these results, primacy effect. In an attempt to investigate further
it m a y be hypothesized that communications arous- this discrepancy, Hovland and Mandell (1957)
ing guilt feelings, which is often the case in efforts conducted an investigation in which every effort
designed to reduce prejudice or discrimination, was m a d e to follow Lund's procedure as closely as
m a y , under certain circumstances, also result in possible. The results of this replication were
defensive reactions and thus not have the desired again unclear, with s o m e groups showing primacy
effect. Indeed Haefner (1956), examining the ef- and other groups showing recency effects. In view
fects of both fear-arousal and guilt-arousal, found ofthis lack of clear-cut results,Hovland andMandell
a lesser degree of acceptance of the content of (1957) replicated the Lund experiment once again,
communications when strong feelings gf fear or using this time topics of prequmably greater cur-
guilt were elicited. Of course, this evidence does rent interest. Again an absence of evidence for
not m e a n that no attempt should be m a d e to point primacy effects under the given conditions w a s

30
 Persuasive Communications Research

found, thus raising serious doubts as to the gener- more effective if placed immediately after the set-
ality of the results obtained by Lund. inducing problems and just prior to the test prob-.
It would appear, then, that under certain condi- lems I ,(E. , p . 64). Once again the implications
tions primacy effects are operative and under other of these results for improving intergroup attitudes
conditions recency effects. Further experimenta- are apparent. It is generally agreed that the goal
tion was designed to assess the influence of certain in changing negative intergroup attitudes is not that
variables on the relative prevalence of recency or of substituting rigid, stereotyped notions of a
primacy in persuasive communications. Hovland negative sort with equally rigid and stereotyped
-
et al. (1957), in summarizing the results of a notions of a positive sort, but rather to decrease
series of experiments dealing with this problem, the degree of stereotyping. An extrapolation from
have pointed out a number of factors that are prob- Luchins' results would suggest that general admo-
ably operative. Hovland, Campbell and Brock nitions against stereotypy or prejudging, in tde
(1958) studied the effect of public 'commitment'to hope that these would have an effect in some later
a position after hearing only one side. When the situation when the subject is confronted with stereo-
individual is induced to take an action on the basis typed material and tempted to accept same, would
of hearing only the first side of an issue, i.e. be less effective than presenting the subjects with
publicly commit himself to the position, a primacy concrete examples of stereotyping, or of the
effect is likely to occur. These authors presented Einstellungs effect, and then 'exposing'or explain-
one side of a controversial issue to one group of ing this tendency, cautioning against stereotyping
subjects and then asked them to write their opinion and prejudging.
on the issue for publication in a magazine to be Viewed in this light, no blanket answer can be
read by their peers. Control subjects likewise given to the question of .primacy ,versus recency
wrote their opinions, but these were anonymous since a more detailed analysis of the problem is
and no mention was made of possible publication. necessary. Hovland, Janis and Kelley (1953) have
Subsequently,and without prior announcement,the stressed the necessity of considering differentially
other side of the issue was presented to both groups the r6le of such factors as attention, learning and
and attitude measures were again obtained. The acceptance. In a summary of research dealing
results show that the public expression of opinion with the effects of order of presentation in persua-
tended to 'freeze'the subjects' views and make sion, Hovland (1957, p. 154 et seq.) concludes:
them resistantto influence by the second side of the 'The combined findings from all of the different
issue. The authors hypothesized that this effect studies reported suggest that the side of an issue
is mediated through social rewards and the need presented first is likely to have a disproportionate
for social approval on the part of the subjects. influence on opinions under the following conditions:
Hovland and Mandell (1957)further investigated the (1) when cues as to the incompatibility of different
possibility that private 'commitment'after hearing items of information are absent, (2) when the con-
only one side of the issue would lead to similar re- tradictory information is presented by the same
sults. However, no significant differences were communicator, (3) when committing actions are
found between the experimental and control subjects. taken only after one side of the issue has been pre-
Another factor which m a y explain the 'primacy' sented, (4) when the issue is an unfamiliar one,
effect observed in a number of investigations is the and (5) when the recipient has only a superficial
Einstellungs effect, or set, created by first im- interest in the issue (low cognitive need). '
pressions. Asch (1946) has shown this to be the In addition to the primacy-recency issue, a
case in formingimpressions ofpersonality. Luchins question which has been the object of many inves-
(1958 a), in a more recent study, has also demon- tigations is that of whether it is better to present
strated the existence of this effect. The impli- one side or both sides of a controversial issue.
cations of these findings for intergroup attitudes This question is not an easy one to answer. A
generally are apparent. The persistence and re- number of studies have shown that when both sides
sistance to change of stereotypes concerning groups of a question which is important to the recipient
m a y well be explained partly on the basis of this are communicated to him by two sources perceived
Einstellungs effect, wherein other factors, such to be approximately equal in reliability,there is
as the law of Pragnanz m a y contribute to explain a tendency for the subject to simply line up with
this phenomenon (although, as we have noticed pre- the side which corresponds to his original opinion
viously, there are wide individual differences in on the issue (e.g. Jarrett and Sheriffs, 1953). O n
these tendencies). O n the more positive side, the other hand, Lumsdaine and Janis (1953) have
Luchins, in further experiments, studied possibi- shown definite advantages to presenting both sides
lities of minimizingthe impact of firstimpressions, of an issue. These authors exposed two groups of
that is to say reducing the Einstellungs effect students to persuasive communications on a partic-
(Luchins, 1958 b). The results show that effortsto ular issue. One group was presented with a one-
minimize Einstellungs effects were successful to sided argument, whereas the other one was pre-
a certain extent. 'When an explicit admonition sented with a communication arriving at the same
against the development of a set or Einstellung was conclusion but including arguments on the other
given at the outset ... somewhat less Einstellungs side of the question as well. Both groups were
effect was obtained. Such an admonition was even influenced in the direction of the persuasive

31
Persuasive Communications Research

communication. However, upon later exposure conclusions arrived at by Flowerman and Tumin
to 'counter-argument', the group which had been is the time (approximatelyone decade) intervening
presented with the one-sided argumentwas shifted between the two judgements. Flowerman is cer-
back under the influence of the opposing argument, tainly correct in saying that mass communication
whereas those who had been presented with both media whose content runs counter to established
sides to begin with were considerably less influ- group norms is not likely to have much effect.
enced by this counter argument. They had already However, a decade later mass media expressing
been 'immunized', so to speak. Abelson (1959, disapproval of segregation and strong approval of
p. 2) summarizes the implications of these and equal rights appears to be expressive of (orperhaps
similar findings as follows: W h e n the audience is helped to create) a group norm of the American
generally friendly, or when your position is the society,as a whole. This consideration makes
only one that will be presented, or when you want Tumin's findings quite understandable.
immediate, though temporary, opinion change, The question of the relative effectiveness of the
present one side of the argument. When the au- various media of communication, especially the
dience starts out disagreeing with you, or when mass media, is a field of its own which we cannot
it is probable that the audience will hear the other begin to review extensively here. W e shall merely
side from someone else, present both sides of the indicate very briefly some of the results which more
argument. ' or less correspond to what would be expected.
Still another question which arises in connexion One generalization that can be made from pre-
with the manner in which communications should vious research and which is most likely still valid
be presented is whether the communication should is that oral presentation of material is more
draw a conclusion for the audience or let them ar- effective than printed presentation in changing at-
rive at their o w n conclusions. Hovland and Mandell titudes (Wilke, 1934; Knower, 1935, 1936; Hovland,
(1952)found that drawing a conclusionis more effec- 1954). The effectiveness of films in comparison
tive than not doing so when the content presented with other mass media has also been extensively
is (a) complex, hence not clearly structured, and investigated. Hoban and van Ormer (1951),intheir
(b) not intimately related to intense attitudes of the review, arrive at the conclusion that in the field
subjects. The findings of Hovland,Janis and Kelley of education, effective films are approximately
(1953)indicate,however,that the reverse is likely equal to an instructor in terms of communicating
to be the case if the subject of the communication the desired material. Television has also been
is ego involved, i.e., if attitudes are aroused compared with other media with respect to its ef-
which have personal meaning for the subjects. fectiveness and, as could be expected, the results
favoured television. One laboratory experiment
Medium of communication. The question of the by Goldberg(l950) has showntelevision to be supe-
effectiveness of the mass media of communication rior to radio in terms of effectively communicating
has been a subject of considerable disagreement. the desired content. However, as Hovland (1954)
Most studies and reviews have indicated that the has pointed out, further research must be awaited
various mass media do have some effect in chang- in order to determine whether this superiority
ing attitudes or opinions in the intended direction is intrinsic or due to the relative newness of tele-
(Peterson and Thurstone, 1933; Hovland, L u m s - vision as a mass communication medium.
daine and Sheffield, 1949; Hovland, 1954; etc.).
O n the other hand, many studies have shown no The recipient of the communication
change or even changes in the 'wrong'direction
(boomerang effect), indicating that the matter is That the effectiveness of a communication,will
not as simple as was previously thought, and point- depend in large part upon the recipient himself is
ing out the need for consideration of more differen- self-evident. In Chapter 1 we dealt at length with
tial factors, as w e have attempted to indicate here. the r61e of personality differences in relationship
Flowerman (1949), disagreeing with the conclusions to attitude change. Such personality variables
arrived at by Rose (1948), de-emphasizes the value as authoritarianism, ego-defensiveness,rigidity,
and effectiveness of mass propaganda as a means etc., were found to be significant in determining
of reducing prejudice and supports, instead, tech- the individual's susceptibility to change. Thus,
niques, based upon group structures and interper- for instance, Zajonc (1960) showed that cognitive
sonal relationships. O n the other hand, Tumin's uncertainty was intimately related to cognitive
findings with respect to readiness for desegre- change. McDavid (1958)investigatedthe difference
gation in the South of the United States indicate a between source-oriented and message-oriented
positive relationship between exposure to mass subjects,showingthatthe message-oriented group
media and readiness for desegregation. This was generally less susceptible to influence.
effect is found even when the factor of educational Furthermore, recent research (e.g. Middleton,
legel is held constant. This author states: 'Within 1960) has confirmed earlier findings that more
any educational group, m y data suggest the more prejudiced individuals were less susceptible to
exposure to mass media the more ready for de- persuasion than less prejudiced individuals,partic
segregation is the person' (Tumin, 1958, p. 36). ularly with respect to the effects of mass com-
One probable reason for the difference in the 'municationsmedia. Manis (1960)has also shown

32
 Persuasive Communications R e search

that the interpretation of communications is a func- B E R E L S O N , B. 1954. Content analysis. In: G.

tion of the recipient'sattitude. Lindzey, Handbook of social psychology, Vol. I.
W e have also referred to research showing that p. 488-522. Cambridge, Addison-Wesley.
the relationship of the recipient to the group is of ; S A L T E R , P.J. 1946. Majority and
significance in determining the effect of c o m m u - minority Americans: an analysis of magazine
nications upon him. Thus, for instance, Mitnick fiction. Public opinion quarterly, 10, p. 168-90.
(1958)found that the influence of a film communi- CHOO,T. 1960. Communicator credibility and
cation upop ethnocentric attitudes was considerably communication discrepancy as determinants of
greater when the content of the film was later dis- opinion change. Unpublished Ph.D. diss .,
cussed in a group situation. Boston University Graduate School.
.In addition to individual differences among recip- COOPER,E.; DINNERMAN,H. 1951. Analysis
ients, social and cultural differences among recip- of the film 'Don't be a sucker': a study in
ients of persuasive communications are unques- communication. Public opinion quarterly, 15,
tionably of great importance. Some studies have p. 243-64.
indicated social class differences in the reaction ; J A H O D A , M. 1947. The evasion of
to persuasive communications (Ramseyer, 1939; propaganda: how prejudiced people respond to
Wiese and Cole, 1946), and many studies have anti-prejudice propaganda. Journalof psychology,
shown differences among differing educational 23, p. 15-25.
groups (e.g. Tumin, Barton and Burrus, 1958). CROMWELL, H. 1950. The relative effect on
A question of particular importance,but one on audience attitude of the first versus the,second
which there has been almost no research, is that areumentative sDeech of a series.- Sryeech-
of cross-cultural differences in reactions to c o m - monographs, 17, p. 105-22.
munications attempting to change attitudes. A whole F E S H B A C H , S. 1961. The consequences of fear-
body of sociological and cultural anthropological arousal in public health education. Paper read
evidence points to the fact that such differences at the Fourteenth International Congress of
exist and are probably very important. Unfortu- Applied Psychology, Copenhagen.
nately, most of the work done so far has been con- ; SINGER, R. 1957. The effects of
ducted in the United States of America, withlittle personal and shared threats upon socialpre-
opportunity for cross-validating the results by judice. Journal of abnormal and social
studies in other cultural settings. A s Hovland psychology, 57, p. 411-16.
(1954, p. 1089) has very aptly commented: FINE, B.J. 1957. Conclusion-drawing,c o m m u -
'Too many of the generalizations about reactionsto nicator credibility and anxiety as factors in
communication are based exclusively on American opinion change. Journal of abnormal and social
audiences (this is not to mention the extent to which psychology, 54, 369-74.
they are based mainly on American college sopho- FLOWERMAN,S.H. 1949. Mass propaganda in
mores I). A s the world becomes smaller and social the war against bigotry. Journal of abnormal
scientists begin to draw practical conclusions from and social psycholoa, 42, p. 429-39.
their studies on the importance of communication GOLDBERG,H.D. 1950. Liking and retention of
by communicating more with each other, it is a simulcast. Public opinion quarterly, 14,
hoped that this shortcoming can be overcome. ' p. 141-2.
W e have pointed to the importance of the c o m - HAEFNER,D.P. 1956. Some effects of guilt-
municator,the communication and the recipient of arousing and fear-arousing persuasive c o m -
the communication as factorsin determiningthe ef- munications on opinion change. Unpublished
fects of persuasive communications. This division Ph.D. diss. , University of Rochester,Rochester,
was somewhat artificial sincethere is a high degree N e w York.
of interrelation between these factors. Indeed, HARTMEN, G.W. 1936. A field experiment on
many experimentalfindings which are importantfor the comparative effectiveness of 'emotional'
the theory of attitude change couldnot be discussed and 'rational'political leaflets in determining
under any one of the above chapters or topics, since election results. Journal of abnormal and
they dealt primarily with the manner in which these social psychology, 31, p. 99-114.
various factors are interrelated. In the next chap- HOBAN,C.F.; V A N ORMER, E.B. 1951.
ter,therefore,we shall discuss some of these find- Instructional film research (rapid mass
ings with respectto the theory of attitude change. learning) 1918-1950. Washington, D.C.,
Department of Commerce, Office of Technical
Services. (Technical report no. S D C 269-7-19).
BIBLIOGRAPHY HOVLAND,C.I. 1954. Effects of the mass media
A B E L S O N , H.I. 1959. Persuasion: H o w opinions
of communication. In: G. Lindzey (ed.). w-
book of socialpsychology,vol. 11, p. 1062-103.
and attitudes are changed. N e w York, Springer Cambridge, Addison-Wesley.
Publishing Co. . 1958. The r61e of primacy and recency
A S C H , S.E. 1946. Forming impressions of inpersuasive communication. In:E.E.Maccoby,
personality. Journal of abnormal and social T.M. Newcomb and E.L.Hartley (eds.). Read-
psycho-, 41, p. 258-90. ings in social psychology, p. 137-49.
33
Persuasive Communications Research

HOVLAND,C.I.(ed.). 1957. The order of pre- KNOWER,F.H. 1935. Experimental studies of

sentation in persuasion, N e w Haven, Yale changes in attitudes. (I) A study of the effect
University Press. (Yale studies in attitude and of oral argument on changes of attitude. Journal
communication,vol. I. ) of social psychology, 6 , p. 315-47.
; CAMPBELL,E.H.; BROCK, T. 1958. . 1936. Experimental studies of changes
The effects of 'commitment'on opinion change in attitudes. (11) A study ofthe effect of printed
following communication. In: C .I. Hovland argument on changes in attitude. Journal of
(ed.). The order of presentation in persuasion, abnormal and socialpsychology,30, p. 522-32.
N e w Haven, Yale University Press. (Yale L A Z A R S F E L D , P.F .; B E R E L S O N , B.; G A U D E T ,
studies in attitude and communication, vol. I.) H. 1944. The peoples' choice. N e w York,
; JANIS, I.L.; KELLEY,H.H. 1953. Duel, Sloan and Pearce.
Communication and persuasion: psychological LUCHINS, A.S. 1958a. Primacy-recency in im-
studies of opinion change. N e w Haven, Yale pression formation. In: C.I. Hovland (ed.).
University Press. The order of presentation in persuasion. N e w
; L U M S D A I N E , A.A.; S H E F F I E L D , Haven, Yale University Press. (Yale studies
F.D. 1849. Experiments on mass communica- in attitude and communication,vol. I.)
-
tion. Princeton, Princeton University Press. . 1958 b. Experimental attempts to
; MANDELL,W. 1952. Anexperimen- minimize the impact of first impressions. In:
tal comparison of conclusion-drawingby the C .I. Hovland (ed.). The order of presentation
commumicator and by the audience. Journal of in persuasion. N e w Haven, Yale University
abnormal and social psychology,47, p. 581-8. Press. (Yale studies in attitude and c o m m u -
. 1957. Is there a law of nication, vol. I.)
primacy? In: C .I. Hovland (ed.). The order LUDLUM,T.M. 1958. Effects of certain tech-
of presentation in persuasion. N e w Haven,Yale niques of credibility upon audience attitude.
University Press. (Yale studies in attitude and Speech monographs, 25, p. 278-84.
communication,vol. I. ) LUMSDAINE,A.A.; JANIS, I.L. 1953. Resis-
/ ; WEISS, W. 1951. The influence of tance to 'counter-propaganda'produced by one -
/ source credibility on communication effective- sided and two-sided 'propaganda'presentations.
ness. Public opinion quarterly, 15, p. 635-50. Public opinion quarterly, 17, p. 311-18.
HYMAN,H.H.; S H E A T S L E Y , P.B. 1947. Some LUND,F .H. 1925. The psychology of belief. (IV)
reasons why information campaignsfail. Public The law of primacy and persuasion. Journal of
opinion quarterly, 1 1 , p. 412-23. abnormal and social psycholoa, 20, p. 174-96.
JANIS, I. ; FESHBACH,S, 1953. Effects of fear- MANIS, M. 1960. The interpretation of opinion
arousing communications. Journal of abnormal statements as a function of recipient attitude.
and social psycholopy, 48, p. 78-92. Journal of abnormal and social psychology, 60,
1954. Personality differ- p. 340-4.
ences associated with responsiveness to fear- M c D A V I D , J .W. Jr. 1958. Conceptual processes
arousing communications. Journal of in social conformity. Dissertation Abstract,
personality, 23, p. 154-66. 18, 319 p.
J A R R E T T , R.F.; SHERIFFS,A.C. 1953. Pro- MENEFEE, S.C.; GRANNEBERG,A.G. 1940.
paganda, debate and impartial presentation as Propaganda and opinions on foreign policy.
determiners of attitude changes. Journal of Journal of social psychology, 1 1 , p. 393-404.
abnormal and social psychology, 48, p. 33-41. MIDDLETON,R. 1960. Ethnic prejudice and
JONES, E.; ZELL,H.; TORREY,R. 1958. susceptibility to persuasion. American socio-
Some effects of feed-backfrom the experimenter logical review, 25, p. 679-86.
on conformity behaviour. Journal of abnormal MILLS, J.; A R O N S O N , E.; ROBINSON,H.1959.
and social psychology, 57, p. 207-13. Selectivitv and exDosure to information Journal- ~ ~~~ ~

KELLEY,H.H. 1955. Salience of membership of abnormal and social psychology. 59, p. 250-3.
and resistance to change of group-anchored MITNICK. L.L.:McGINNIES. E. 1958. Influenc-
attitudes. Human relations, 8,p. 275-90. ing ethnocentrism in small discussion groups
; VOLKART,E.H. 1952. The resis- through a film communication. Journal of ab-
tance to change of group-anchored attitudes. normal and social psychology, 56, p. 82-90.
American sociological review, 17, p. 453-65. MURPHY, G.; MURPHY, L.D.; NEWCOMB,
; WOODRUFF,C.L. 1956. Members' T.M. 1937. Experimental social psychology,
reactions to apparent group approval of N e w York, Harper.
counternorm communications. Journal of P A S T O R E , N.; HOROWITZ,M.W. 1955. The
abnormal and social psychology, 52, p. 67-74. influence of attributed motive on the acceptance
KELMAN,H.C.; HOVLAND,C.I. 1953. 'Re- of a statement. Journal of abnormal and social
instatement' of the communicator in delayed psychology, 51, p. 331-2.
measurement of opinion change. Journal of P E T E R S O N , R.C.; T H U R S T O N E . L.L. 1933.
abnormal and social psychology, 48, Motion pictures^ and the social attitudes of
p. 327-35. children. New York, Macmillan.

34
 Persuasive Communications Research

R A M S E Y E R , L.L. 1939. Measuring 'intangible' TUMIN, M. 1958, S o m e problems for socio-

effects of motion pictures. Educational screen, logical research in desegregation: In: The r81e
18, p. 237-8. of the social sciences in desegregation: a
ROSE, A.M. 1948. Studies in reduction of pre- symposium. N e w ' Y o r k ,Anti-Defamation League.
judice, 2nd ed. , Chicago, American Council of ; BARTON, P.; B U R R U S , B. 1958.
Race Relations. Education, prejudice and discrimination: a
S H E R I F , M.; H A R V E Y , O.J.; WHITE, B.J.; study in readiness for desegregation. American
HOOD. W.R.; S H E R I F , C.W. 1954. Experi- sociological review, 23, p. 41 -9.
mental study of positive and negative inter- WILKE, W.H. 1934. An experimental comparison
group attitudes between experimentally pro- of the speech,the radio andthe printed page as
duced groups: Robbers Cave Study. N o r m a n , propaganda devices. N e w York (Archives of
Oklahoma, University of Oklahoma (multi- psychology, no. 169).
lithed). W I L L I A M S , R.M. Jr. 1947. The reduction of
; SHERIF, C.W. 1956. A n outline of inter-group tensions: a survey of research on
socialpsychology,(rev.ed.) N e w Y o r k , Harper. problems of ethnic, racial and religious group
S H U E Y , A.M. 1953. Stereotyping of Negroes relations. N e w York, Social Science Research
and whites: an analysis of magazine pictures. Council. (Bulletin no. 57. )
Public opinion quarterly, 17, p. 281-7. W I E S E , M.J.; COLE, S.G. 1946. A study of
TANNENBAUM , P. 1956. Initial attitude toward children's attitudes and the influence of a
sources and concept as factors in attitude commercial motion picture. Journal of psycho-
change through communication. Public opinion =, 21, p. 151-71.
quarterly, 20, p. 413-25. Z A J O N C , R.B. : M O R R I S S E T T E . J. 1960. The
TAYLOR, J.A. 1956. Drive theory and manifest r81e of uncertainty in cognitive change. Journal
anxiety. Psychological bulletin, 53, p. 303-20. of abnormaland socialpsychology, 61, p. 168-75.

35
 CHAPTER 4

THEORETICAL P R O B L E M S OF ATTITUDE CHANGE RESEARCH

A s the previous chapters have clearly shown, re- similar vein, merely demonstrate a fact to be ob-
search on attitude change cuts across many disci- served in daily life with discomforting frequency,
plines and touches upon many fundamental questions i.e. that discrepancy often exists between attitudes
of human behaviour. It is, therefore, not surpris- and behaviour and, for that matter, between various
ing that such research raises many questions which logically related attitudes held by the same indivi-
are of fundamental theoretical importance in the dual. Obviously then, factors other than the
social sciences. W e cannot begin to describe in tendency toward logical consistency or 'cognitive
detail here the very involved theoretical issues consonance' are operative. What is the experi-
touched upon by such research. However, some mental evidence in favour of the assumption that
of these issues, arising from the results of empi- cognitive dissonance m a y produce attitude change ?
rical research, are not only of theoretical signifi- Under what conditions will cognitive dissonance
cance but also have implications which are of great most likely be created? Under what conditions are
practical importance in making policy decisions as opposing tendencies operative, and what are these
to programmes of action. W e shall, therefore, tendencies? W e can only indicate here very briefly
refer very briefly here to some of the theoretical some recent research on these questions.
problems raised by recent research findings and McGuire's (1960a) research on cognitive consis-
their practical implications. tency and attitude change is most interesting, both
In addition to investigating the many factors as- for its theoretical and its practical implications.
sociated with attitude change, some research has Whereas most techniques of attitude change involve
also addressed itself to the question of the actual presenting the subject with a persuasive message
process itself, that is to say to the theory of atti- from an external source, McGuire, proceeding
tude change. One of the more recent theories from the postulate of cognitive consistency, demon-
which has stimulated a great deal of research is strated that the mere temporally contiguous elici-
Festinger's (1957) 'theory of cognitive dissonance'. tation of opinions which are logically inconsistent
According to this theory, when a person is con- with each other will bring a tendency toward con-
fronted with a communication (e.g. an attempt at sistency to bear upon them and thus produce atti-
changing his attitude), the content of which is at tude change. Using what he terms the 'Socratic
variance with his present attitude, cognitive dis- method', McGuire elicited the subjects' opinions
sonance is created, i.e. psychological tension regarding the subjective probabilities and desira-
having drive characteristics and seeking reduction. bilities of 48 propositions presented in random
This thesis of a tendency on the part of the indivi- order but in fact constituting 16 sets of syllogisms
dual to maintain logical consistency among his (major proposition, minor proposition, conclusion).
cognitions (and between his cognitions, affects and The subjects' responses deviated significantly from
more gross behaviour) actually has a long history. purely logical expectations, showing a certain de-
Space does not permit us even to list the large gree of inconsistency among attitudes. In a second
number of authors in recent times whose theories elicitation of the subjects' opinions, a significant
contain such a notion, let alone to refer back to the decrease in logical inconsistency was shown, thus
precursors of this idea in the writings of earlier demonstrating that the Socratic method of simply
thinkers. Just by way of illustration, we might asking the person to state contiguously his opinions
mention Sumner (1907) who speaks of a 'strain on logically related issues will result in attitude
toward consistency', Lund's (1925) 'need for con- change. McGuire (1960a, 1960b) has furthermore
sistency', or Newcomb's (1953) 'straintoward shown that persuasive communications directed at
symmetry'. More recently Heider's (1946,1958) an explicit issue tend to change, also, the subjects'
'balance theory' has influenced a good deal of opinions on logically related derivative issues in
research. a consistent direction, i.e., in the direction of
If such tendencies do exist, and a large body of reducing logical inconsistency. There were even
experimental evidence points clearly to the fact indications that the impact of the persuasive mes-
that they do, how then can we explain the many in- sages had continued to seep down into the derived
stances to be observed in everyday life which are issues with time. Thus in the session immediately
in contradiction to this tendency, The classical following the messages, the opinions on derived
investigation of L a Piere (1934) and the more recent issues had shown only 52 per cent of the logically
confirmation by Kutner, Wilkins and Yarrow (19521, required amount of change,whereas one week later
as well as a large number of research results in a 91 per cent of the required amount of change had

36
 Theoretical Problems of Attitude Change Research

occurred. Of course, the changes both with and he would avoid if possible, he experiences no
without persuasive messages were not as great as dissonance. The question is, then, would a fait
would be required for perfect logical consistency. accompli, i.e. an event outside the person's control,
McGuire regards 'wishful thinking' as acting in create dissonance or not. Brehm's (1959) experi-
opposition to logical thinking. ment demonstrates a situation in which a fait
One of the derivations of Festinger's theory is accompli does appear to have increased cognitive
that publicly making a statement or engaging in an dissonance; in this case, when the event would
act which is not consonant with one's private opin- have led to the opposite behaviour had it been
ion can create cognitive dissonance and thus, under predictable at a prior choice point. In a further
certain circumstances, lead to attitude change. experiment, Brehm andCohen(1959) dTmonstrated
Janis and King (1954, 1956) have shown that when that a chance event m a y affect the magnitude of
subjects are induced to engage in r61e playing dissonance and consequentattitude change,but only
activity which is at variance with their private under conditions of high choice. Experimentally
opinion,there will be a tendencyfor such behaviour manipulating the variable of choice, these authors
to result in attitude change, whereby improvised found that in a low choice situation, i.e. where
r6le playing was shown to be more effective than the subject felt he had little alternative but to do
non-improvised r6le playing. The authors explain what was required of him, no significant attitude
these results in terms of mental rehearsal and change occurred. Onthe other hand,the subject's
thinking up new arguments. However, in terms of attitude toward the fait accompli became signifi-
dissonance theory a slightly different explanation cantly more positive under conditions where they
is possible (Festinger,1957, chapter 4): If a per- feltthey had a high degree of choice. The implica-
son privately holds opinion 'XI,but as a result of tions of these results for attitude change in actual
pressure brought to bear on him has publicly stated social life are apparent.
that he believes 'notXI,his cognition of his private W e have discussed a few experiments showing
belief is dissonant with his cognition concerning his that cognitive dissonance occurs when a person's
actual public statement. However, the knowledge private opinion is seen to be at variance with the
that he has said 'not X' consonant with the opinion of others, or with his publicly stated
reasons,pressures,promises of rewards and/or opinion or behaviour, and have seen that such
effective punishment which induced him to say 'not cognitive dissonance m a y lead to attitude change.
XI. It thus follows that,everything else being held What other factors are at work to prevent attitude
constant, his total magnitude of dissonance would change, and what other ways are there of reducing
decrease as the number andimportance ofthepres- cognitive dissonance besides changing one's atti-
sures which induced him to say 'notX'increases. tudes? W e have already mentioned McGuire's con-
Hence, if his overt behaviour was brought about cept of 'wishfulthinking'as a tendency working
by offers of reward or threats of punishment, againstcognitive consistency. Similarly,Rosenberg
the magnitude of dissonance should be maximal and Abelson (1960) have shown that, in addition to
when these promised rewards or threatened punish- the motivation toward achieving 'balance' between
ments are just barely sufficient to induce him to attitude components (i.e. between affective and
. cognitive components of an attitude), there is also
say 'not XI. Beyond this point, as the promised
rewards or threats of punishment became larger, sometimes a 'hedonictendency',i.e. a striving
the magnitude of dissonance becomes smaller. And for the maximization of potential gain and minimi-
other things being equal,amount of attitude change zation of potential loss. In the chapter on personal-
willbe related to magnitude ofcognitive dissonance. ity-oriented research we discussed a number of
The results of an earlier experiment by Kelman factors, such as ego-defence mechanisms, etc. ,
(1953) have been described by Festinger and Carl- which m a y resist the acceptance of another attitude
smith (1959) as being consistent with this theory. even though logical evidence favours it.
Kelman found that a large reward produced less Festinger and his co-workers have also shown
subsequent opinion change than did a smaller re- that there are a number of possibilities of reducing
ward with a group of schoolchildren. Festinger cognitive dissonance other than changing one's own
and Carlsmith(19591,in an experiment specifically attitudes. Thus in addition to such mechanisms
designed to test these derivations of dissonance as blocking,repression,selective perception, and
theory, arrived at results which clearly showed selective attention, etc. , the subject m a y reject,
that (a) if a subject is induced to do or say some- discredit or otherwise oppose the communicator or
thing contrary to his private opinion, there will be source ofthe communication. Allyn and Festinger
a tendency for him to change his opinion so as to (1961) found, for instance, that unanticipated
bring it into correspondence with that which he has persuasive communications were more effective
done and said,and (b)the larger the pressure used than anticipated communications. The explanation
to elicit the overt behaviour (beyond the minimum in terms of dissonance theory would be that the
neededto elicit it), the weaker will be this tendency. prepared subjects reduced the cognitive dissonance
Brehm (1959) has investigated some of the by rejecting the communicator, whereas the un-
conditions under which cognitive dissonance takes prepared subjects reduced the dissonance by
place. As various studies have shown,if a person changing their opinions.
is completely forced to behave in a manner which The foregoing has been an attempt merely to

37
Theoretical Problems of Attitude Change Research

suggest some of the complex issues involved in HEIDER,F. 1946. Attitudes and cognitive organi-
recent research on the theory of attitude change. zation. Journal of psychology, 21, p. 107-12.
Judging by current research trends, this area of . 1958. The psychology of interpersonal
research has reached a stage where attempts are relations. N e w York, Wiley.
being made to integrate previouslyisolated and in HOVLAND,C.I.;R O S E N B E R G , M. 1960. - Atti-
part discrepant,results into consistent theoretical tude organization and change. N e w Haven,
systems which would permit the deduction of Yale University Press. (Yale studies in attitude
further hypotheses for experimental investigation. and communication,vol. 111. )
Owing to limitations of space it was unfortunately JANIS, I.L.;KING,B.T. 1954. The influence
notpossible even to mention,let alone describe in of r61e playing on opinion change. Journal of
detail,most ofthe major theories which have shown abnormal and social psychology,49, p. 211-18.
themselves to be quite fruitful in stimulating re- KELMAN, H.C. 1953. Attitude change as a
search. Of particular interest are the theoretical function of response restriction. H u m a n
formulations and experimental results of Kelman relations, 6 , p. 185-214.
(1960,1961a.1961b)who has presented amodel permit- . 1960. Effects of rble-orientation
ting the explanation of many of the experimental and value-orientation on the nature of attitude
results which we have discussed so far and suggested change. Paper read at the meetings of Eastern
areas of further experimentation. Unfortunately, Psychological Association, N e w York.
however, it is scarcely possible to describe this . 1961 a. Process of opinion change.
theoreticalframeworkin afew sentences (see espe- Public opinion quarterly, 25, p. 57-78.
cially Kelman, 1961 a). Anderson (1959)has also . 1961 b. Theinduction of action and
suggested a (mathematical)model of attitude change. attitude change. Paper read at the Fourteenth
Although this chapter must of necessity remain International Congress of Applied Psychology,
incomplete, it is hoped that it has shown, in a Copenhagen.
suggestive manner, some of the recent experimen- KING,B.T.;JANIS, I.L. 1956. Comparison of
tal trends pertaining to a theory of attitude change. the effectiveness of improvised versus non-
For the practitioner this research has implications improvised r81e playing in producing opinion
~~

of which the practical significance is apparent. change. Human relations, 9, p. 177-85.

For the researcher,the few studies cited here can KUTNER, . B.:. WILKINS. C. and YARROW. P.R.
merely point out some current research trends. 1952. 'Verbal attitudes and overt behaviour
involvingracialprejudice' Journal of abnormal
and social psychology, 47, p. 649-52.
BIBLIOGRAPHY LA PIERE,R.T. 1934. 'Attitudesversus Action',
Social Forces, 14, p. 230-37.
ALLYN, J.; FESTINGER, L. 1961. The effec- LUND, F.H. 1925. The psychology of belief.
tiveness of unanticipated persuasive c o m m u - Journal of abnormal and social psychology,
nications. Journal of abnormal and social 20, p. 63-81.
psychology, 62 (l), p. 35-40. M c G U I R E , W.J. 1960 a. Cognitive consistency
A N D E R S O N , N.H. 1959. Test of a model for and attitude change. Journal of abnormal and
opinion change. Journalof abnormal and social social psychology, 60, p. 345-53.
psychology, 59, p. 371-81. . 1960b. Direct and indirect per-
BREHM , J.W. 1959. Increasing cognitive disso- suasive effects of dissonance-producing
nance by a fait accompli. Journal of abnormal messages. Journal of abnormal and social
and social psycholopy, 58, p. 379-82. psychology, 60, p. 354-8.
; COHEN,A.R. 1959. Choice and NEWCOMB, T.M. 1953. An approach to the
chance relative deprivation as determinants study of communicative acts. Psychological
of cognitive dissonance. Journal of abnormal review, 60, p. 393-404.
and social psychology, 58, p. 383-7. R O S E N B E R G , M.; A B E L S O N , R.P. 1960.
FESTINGER, L. 1957. A theory of cognitive Analysis of affective-cognitive consistency. In:
dissonance. Evanston, Illinois, R o w Peterson Hovland, C .I. and Rosenberg, M. (eds.).
Co. Attitude organization and change. N e w Haven,
; CARLSMITH,J.H. 1959. Cognitive Yale University Press. (Yale studies in
consequences of forced compliance. Journal of attitude and communication, vol. 111. )
abnormal and socialpsychology,58, p. 203-10. S U M N E R , W.G. 1907. Folkways. Boston, Ginn.

38
 P A R T I1

ACTION RESEARCH ON INTERGROUP ATTITUDES

39
 CHAPTER 1
EDUCATIONAL PROGRAMMES IN INTERGROUP RELATIONS

W e shall now turn our attention to the question of educator, and especially between the pupil and his
specific programmes of action designed to change peer group. T o the extent to which the peer group
attitudes in the direction of improving intergroup becomes a frame of reference for the child, it will
relations. Of course, a complete review of the be important in forming and changing his social
countless efforts being made throughout the world attitudes. The importance of communications re-
towards this end cannot be given here. Our purpose search for a better understanding of the educational
is rather to discuss briefly the value of such at- process is self-evident,since any attempt to edu-
tempts generally,to describe afew selected, illus- cate the child involves a form of 'persuasive'c o m -
trative examples ofaction research and,especially, munication, be it in reference to particular facts
to attempt to point the way to a greater integration or in the development or change of social attitudes.
of experimental research and action research or What is the empirical evidence as to the effec-
action programmes designed to improve intergroup tiveness of education in changing intergroup atti-
relations. tudes in a positive direction? Earlier research
Certainly education is one of the most promis- had shown highly inconsistent findings: some stud-
ing areas in which social attitudes m a y be changed ies showed positive change, others showed no
with a view to improving intergroup relations. The significant change, while still others showed change
young human being is much more amenable to in a direction opposite to that which was desired
change than are adults; indeed, as countless stud- (boomerang effect). Thus, for instance,Schlorff
ies have shown, the very young child is totally (1930)found that weekly classes over a period of
without the prejudices of adults and acquires these one semester dealing with the history and status
only in the course of the so-called socialization ofNegroes resultedin a significant,positive shift in
process, first within the family, and later in the ranking Negroes on the part of ninth-grade school-
school setting. In the parent-child relationship,w e children as compared with a control group. (The
face the dilemma that in order to bring up children rank assigned to Negroes by the experimental
in a manner which would be conducive to better subjects was, however, still at the lower end of
intergroup relations, it would be necessary first the scale.) On the other hand, Young (1927),
of all to change the parents. However, education studying the results of a course on race relations,
has now become an accepted and important part of and Droba (1932), analysing the effects of a course
every child's life; and, when one considers thatin on the American Negro in a college curriculum,
nearly all countries of the world every child, from
I both found no significant change in attitudes follow-
the age of 6 (or often from the age of 4 or 5) until ing this attempt at educational influence. Likewise,
well into adolescence or even into young adulthood, Manske (1935)found that of 22 high school classes
spends the majority of his time in school or in exposed to ten lessons on the American Negro,
other activities associated with formal education, only two classes shifted their attitudes in the direc-
the potential influence of this area upon formingor tion of their teacher's stand, whereas eight classes
changing the child's attitudes is at once apparent. showed changes in a direction opposed to the
The educationalprocess is very complex,involv- influence desired by the teachers; the remaining
ing essentially all the areas of experimental re- classes were unchanged. Later research results
search which we have discussed in previous chapters. have also shown some variability. However,
The results of personality-oriented research are Harding, Kutner, Proshansky and Chein (1954, p.
important in this connexion, since not only will the 1047), in reviewing studies of the effects of spe-
reactions of individual students depend upon differ- cific educational measures, conclude that 'reports
ences in their personalities, but also the school- of significant favourable changes in attitude out-
child is still in the process of developing his number the reports of insignificant changes or no
personality; and the school experience can play a change by about two to one'.
major r61e in developing particularly such aspects Of course, a number of investigations have
of the personality as are related to the child's shown that whether education has a positive or a
social interactions. Especially important is the negative influence m a y also depend upon the orien-
application of group-oriented research since, as tation of the educators and the peer group. W e '
we know, the educational process does not consist have already cited an earlier study by Sims and
merely of imparting information,but,is,rather, a Patrick (1936)in which attendance of a Northern
highly complex phenomenon of group dynamics in- group at a university where intolerance was the
volving the intricate relationshipbetween pupil and norm increased attitudes ofintolerance onthe part

41
Education Programmes in Intergroup Relations

of these students. These and other investigations a two-year college experience and found conclusive
have led some authors to conclude that it is possible evidencethat studentswho had completed two years
to educate young people in any desired direction in a California college had changed significantly
whatsoever, either tolerant or intolerant. Although in the direction of less ethnocentric attitudes, as
there is no question that educational influences, if compared with control subjects who had not had
dictated by nationalistic, intolerant attitudes, can college experience during this time. A more re-
influence the individual in the direction of inter- cent review by Williams (1958) has reported a
group hostility and conflict, rather than in the greater influence of education in improving inter-
direction of healthy intergroup relations , never- group attitudes to the extent that the broader norms
theless there seems to be some evidence in favour 'ofthe society are congruent with such attempts.
of certain limitations to this negative process. The The study by Nowak (1960),to which we referred
views of the early rationalists, although in need previously, showing an increase in egalitarian
of considerable modification in light of modern attitudes on the part of Warsaw students, exem-
knowledge, m a y , nevertheless, have a certain plifies this point.
amount of validity. Adorno, - et al. (1950, p. 10 Tumin,Barton and Burrus (1958),in summariz-
et seq.) have expressed similar views. They state: ing the effects of education on prejudice, discrimi-
'There is one explanation for the existence of an nation and readiness for desegregation in the South
individual's ideology that has not so far been con- of the United States,have concluded that as formal
sidered: that it is the view of the world which education increases there tend to occur noticeable
a reasonable man, with some understanding of the shiftsfrom: (a) nationalism to internationalism, in
rBle of such determinants as those discussed above, political point of view; (b)traditionalism to secu-
and with complete access to the necessary facts, larism, in general social philosophy; (c) common-
will organize forhimself. This conception, though sense to science, as acceptable evidence; (d)
it has been leftto the last, is of crucial importance punishment to reform, in penological theory;
for a sound approach to ideology. Without it we (e) violence and direct action to law, as agents
should have to share the destructive view, which of policy; (f) rigidity to permissiveness, in child
has gained some acceptance in the modern world, rearing; (g) patriarchy to democracy, in spouse
that since all ideologies, all philosophies,derive relationships; (h) anaesthesia to creativity, in
from non-rational sources, there is no basis for patterns of recreation. This summary would seem
saying that one has more merit than another. A to indicatethat education causes afar-reaching and
number of studies supportthis view. The investiga- deep-going change in social attitudes generally in
tions of McGuire (1960 a,b) to which we referred a direction more likely to be conducive toward
above, have shown for instance that attempts to healthy, constructive intergroup relations. Of
influence an individual in the direction of logical course,sub-culturalgroup norms m a y work against
inconsistency will meet with great resistance, these tendencies, causing resistanceto change.
whereas attempts to influence him in terms of For instance,Young, Benson and Holtzman (1960)
greater consistency will more likely meet with investigated changes in attitudes toward the Negro
success. Of course, it can be argued that there among students at a university in the South of the
are ideologies and frames of reference in terms United States between the years 1954 (just shortly
of which intolerance and intergroup tension are after the United States Supreme Court decision
logically consistent. However, it would appear concerning segregation,but before actual desegre-
that the norms of mankind as a whole, as theyhave gation had taken place) and 1958. These authors
developed in the modern world through greater found no significant change in group means in atti-
social awareness and closer interaction brought tudes toward the Negro during this period, in spite
about by modern means of communication and ofthe many controversialevents which had occurred
technology, are inconsistent with hostile inter- (e.g. the Little Rock incident, etc.). A popular
group attitudes. The latter m a y be regarded as hypothesis was that,althoughthere m a y have been
left-oversofmankind's dark (althoughunfortunately no change in mean scores, the breach between
not too distant)past. Thus educational influences, advocates of segregation and advocates of desegre-
with the background of this prevailing norm, are gation had become greater as a result of the
likely,by providing more information and greater controversial nature of the issue. However, the
understanding, to work in the direction of im- findings of these authors failed to support this
proving intergroup relations. hypothesis, since there was no change either in
With respect to the general effect of formal mean score or variance of scores. The only signi-
education,most reviews (e.g.Williams,1947; Rose, ficant change noted was uncovered when m e n and
1948) have concluded that with increased formal women in the two samples were analysed separa-
education negative intergroup attitudes decrease. tely; whereas the women students appeared to be
One of the most extensive investigations support- slightly more tolerant of the Negro in 1958, the
ing this conclusion was a nation-wide representa- m e n had moved in the opposite direction.
tive survey made by the National Opinion Research These results point once againto the importance
Centre in 1944 in the United States of America of firmly anchored group norms; but more
(Samelson, 1945). In a more recent study, Plant generally they suggest the necessity of basing any
(1956) investigated attitude changes associated with educational programme upon modern knowledge

42
 Education Programmes in Intergroup Relations

concerning attitude formation and change. A s education as a means of improving intergroup rela-
one author has pointed out in a study designed tions is justified if education is considered in the
to develop a resource unit for the training of classical sense as being merely the imparting of
secondary teachers in problems and issues information. However, there is no reason why
involving minority groups (Pitkin, 1950), it is modern knowledge cannot be imparted to the educa-
necessary for the educator to become a capable tor so that he can more effectively fulfil his task.
diagnostician. It is necessary for him to take This does not mean that every teacher must be a
into account individual differences in personal- trained psychologist, sociologist, etc. , but the
ity and existing attitudes on the part of his essential principles of psychodynamics, group
pupils, differences with respect to background, dynamics and sociology might well be included in
group membership and reference groups, per- his training. Although this is not the primary pur-
sonal and group dynamics involved in attitude pose of the present paper, it is possible to trans-
change, and a large number of further factors late the results of research so as to make them
associated with conditions under which attempts understandable to educators without any great tech-
at forming or changing the attitudes of his nical knowledge of social science research. And
pupils in the direction of more positive inter- - of course it is possible to change the curricula of
group relations are likely to be successful or teachers' training institutions so as to include a
not. wider coverage of subjects directly related to the
Although, as we have pointed out previously, it problem of improving intergroup relations. As
is highly desirable to disseminate factual knowledge Watson (1956, p. 309) has pointed out, 'We shall
as widely as possible, it is highly unlikely that this have to talk not about modest additions to our pre-
method alone will be very effective in changing sent programme but about basic transformations
existing negative intergroup attitudes. This general- of our curriculum,our process of teacher selection,
ization holds, by the way, not only for attempts at and the whole conception of public education ...
changing negative intergroup attitudes, but for the W e shall be thinking ... about the need of all
educational process as a whole. Modern education- educational leaders for reorientation and for skills
al theory has realized that the mere presentation of statesmanship. The disparity between the dimen-
of information is not enough to achieve the goals of sions of our previous programmes and our present
education, but that the total personality and espe- problems is formidable.I
cially the group situation involved in the learning All this points to a greater need for close
process is of primary importance. Kurt Lewjn, co-operation between 'pure' and 'applied'science,
one of the foremost pioneers in action research on between researchers in the field of intergroup rela-
group dynamics, and his co-workers have done tions and educators and other practitioners who,
much to point the way toward this new orientation in the final analysis, must apply these research
in education. Lewin has emphasized the necessity findings in actual practice. Of course, such
of seeing education as a group process; thus, the co-operation has enormously progressed in recent
individual's sense of participation in the group at- years and the curricula of teachers'training insti-
mosphere is of prime importance in acquiring new tutions today include many courses in the social
ideas. Lewin (1948, p. 59) states: 'Much stress sciences. In a number of school systems, too,
is laid on the creation, as part of the re-educative action programmes based on the social sciences
process, of an atmosphere of freedom and sponta- have been implemented (e.g., Vickery and Cole,
neity. Voluntary attendance,informality of meet- 1944; Brameld, 1946; MacIver, 1948; Cook, 1950;
ings, freedom of expression in voicing grievances, Cook and Cook, 1954; Ogawa, Kihara, Kubaya and
emotional security and avoidance of pressure, all Tanaka, 1957; for a recent review of programmes
include this element ... If re-education means the in the United States, see Simpson and Yinger, 1958).
establishment of a new super-ego,it necessarily This process is, however, proceeding at a dif-
followsthat the objectives sought willnot be reached ferent rate in different cultures, depending upon
as long as the new set of values is not experienced the stage of development of social science and upon
by the individual as something freely chosen. ' In a number of other factors. The necessity of alter-
a previous chapter we have referred to the earlier ing the curriculum of a teachers 'training institu-
studies of Lewin on 'groupdecision' as a means of tion is dramatically shown in an investigation by
attitude change. Kelman's (1953) investigation Tausch (1961). Using a test to measure the degree
which we cited earlier (showingthat a high degree of understanding of children on the part of teachers
of restriction results in conformity but not attitude and teacher-training candidates, this author found
change, whereas a low degree of restriction results a widespread lack of understanding and frequent
in more attitude change in the long run), has impor- attitudes of criticism and blame; indeed, the
tant educational implications which are in line with differencesin this respect between students in the
Lewin's formulations. first semester, in the fourth semester, and expe-
All this, of course, means that heavy demands rienced teachers with an average of 15 years'
must be made on the educator if he is to be suc- teaching experience were not significant. A further
cessful in changing attitudes in the direction of investigationshowed that a speciallydesigned course
improving intergroup relations. The pessimistic in educational psychology, lasting one semester,
views of certain authors as to the effectiveness of resultedin a significant increase in the understanding
43
Education Programmes in Intergroup Relations

of children's behaviour on the part of teacher- is indispensable to our society. A n ideology of

training candidates, whereas the usual courses in freedom for the teacher is also involved. The re-
general psychology and pedagogy did not lead to any sult is that the teacher has some autonomy, some
significant changes. A n investigation of the effec- -
power to attack prejudice if he wants to and some
tiveness of specific training in handling problems teachers want to' (ibid., p. 758 et seq.).
of intergroup relations would probably show similar In conclusion, it m a y be said that we possess
differences. Tausch used knowledge derived from the knowledge (although,of course, we need more),
non-directive group psychotherapy in her training but this knowledge must be made readily available
experiment (Rogers, 1951), demonstrating that and applied. There are difficulties involved in
such principles are applicable in the educational doing this, but they are not insurmountable. Thus
field. Wieder (1951), in a 'comparative studyof seen, educational programmes have a very definite
the relative effectiveness of two methods of teach- r81e to play in changingnegative intergroup attitudes
ing a thirty-hour course in psychology in modifying and improving intergroup relations.
attitudes associated with racial, religious and eth-
nic prejudice', likewise showed the applicability
of group therapy procedures. Whereas the tra- BIBLIOGRAPHY
ditional lecture-discussion method did not signifi-
cantlymodifyintergroup attitudes,a second method ADORNO, T.W.; F R E N K E L - B R U N S W I C K ,E.;
involving principles of group therapy, non-directive LEVINSON, D.;SANFORD, R.N. 1950. The
methods and socio-drama was highly successful in authoritarian personality. N e w York, Harper
doing so. The non-directive and rele-playingtech- Bros.
niques contributed to personal gxowth as reflected BRAMELD, T. 1946. Minority problems in the
in increased self insight, greater self acceptance public schools, N e w York, Harper Bros.
and a decrease in attitudes associated with racial, COOK,L.A.(ed.). 1950. College programs in
religious and ethnic prejudice. inter-group relations. Chicago, American
W e m a y conclude, then, that we do possess the Council on Education.
knowledge, which if applied on a wide scale,could ; COOK,E. 1950. Inter-group educa-
lead to the accomplishment of our objectives of tion, N e w York, McGraw Hill.
improving intergroup attitudes and intergroup rela- DROBA,D.D. 1932. Education and Negro
tions. Of course, the problem arises of how to attitudes. Sociology and social research, 17,
impIement these methods on a large scale. O n this p. 137-41.
point, some authors have expressed a rather pes- HARDING,J.; KUTNER, B.; P R O S H A N S K Y , H.;
simistic view. Merton (1957, p. 183) for instance CHEIN,I. 1954. Prejudice and ethnic rela-
writes: 'The appeal to education as a "cure-all"for tions. In: G . Lindzey. Handbook of social
the most varied social problem.. . is none the less psychology, vol. 11. Reading, Mass., Addison
illusory ...forhow would this programme of racial Wesley Pub. Co.
education proceed? W h o is to do the educating? KELMAN,H.C. 1953. Attitude change as a
The teachers in our communities? But in some m e a - function of response restriction. Human Rela-
sure ...the teachers share the same prejudices tions, 6, p. 185-214.
they are urged to combat ... Education m a y serve LEWIN,K. 1948. Resolving social conflicts.
as an operational adjunct but not as the chief basis N e w York, Harper Bros.
for any but excruciatingly slow change in the pre- MacIVER, R.M. 1948. The more perfect union.
vailing pattern of race relations. ' Simpson and N e w York, Macmillan.
Yinger (1958) agree with this difficulty but point MANSKE,A.J. 1935. The reflection of teachers'
out: 'Despitethe close connexion between formal attitudes in the attitudes of their pupils. Ph.D.
education and the rest of society, there is a m e a - diss. , Teachers' College, Columbia University,
sure of autonomy in the school system. This auto- N e w York.
nomy is easy to exaggerate, but it is a strategic MERTON, R.K. 1957. Social theory and social
error to dismiss it too lightly, as Mertonberhaps structure,(rev.ed.)Glencoe, Ill. , The Free
does , .. Those professionally connected with edu- Press.
cation, because of their functional r6le in society, NOWAK, S. 1960. Egalitarian attitudes ofwarsaw
are somewhat more concerned with the pursuit of students. American sociological review, 25,
truth, a little less likely to be provincial. In our p. 219-31.
society they are also inclined to be somewhat more OGAWA, T.;KIHARA,K.;KUBAYA, S.;TANAKA,
liberal than the average, although some are timid, K. 1957. A comparative research in the collec-
and others emotionally identify with the upper
classes. Those who control the school systems,
tive attitude of primary school children.
Fac. Educ. Nagoya, 3, p. 390-1.
z.
moreover (the school boards and trustees), are not PITKIN.V.E. 1950. A resource unit for the
inclined to determine every action of the teacher, training of secondary teachers in problems
because they are to some degree dependent on him and issues involving minority groups, with
as the conserver and pursuer of knowledge, as the special reference to Negro-white relationships.
expert and trainer of experts, who, despite the Unpublished Ph.D. diss., N e w York Univer-
disdain in which he is held by the "practical" m a n , sity, N e w York.

44
 Education Programmes in Intergroup Relations

PLANT,W.T. 1956. Attitude change associated with VICKERY , W.; COLE,S.G. 1944. Intercultural
a two-yearcollege experience. Unpublished Ph.D. education in American schools. N e w York,
diss .,StanfordUniversity,Stanford,California. Harper Bros.
R O G E R S , C.R. 1951. Client-centered therapy:
Its current practice,implications,and theory.
- -
W A T S O N . G. 1956. Education and inter-group
relations. Colombia Teachers College record,
Boston, Houghton. 57, p. 305-9.
ROSE,A.M. 1948. Studies in reductionof prejudice, WIEDER,G.S. 1951. A comparative study of
2nd ed. Chicago, American Council on Race the relative effectiveness of two methods of
Relations. teaching a thirty-hour course in psychology
S A M E L S O N ,B. 1945. The patterning of attitudes and in modifying attitudes associated with racial,
beliefs regardingtheAmericanNegro: an analy- religious and ethnic prejudice. Unpublished
sis ofpublic opinion. Unpublished Ph.D. diss., Ph.D. diss., N e w York University, N e w
Radcliff College, Cambridge, Massachusetts. York.
S C H L O R F F ,P.W. 1930. A n experiment in the mea- WILLIAMS,R.M. Jr. 1947. The reduction of
surement and modification of racial attitudes inter-group tensions: a survey of research
in schoolchildren. Ph.D. diss., N e w York on problems of ethnic, racial and religious
University, N e w York. group relations. N e w York, Social Science
SIMPSON, G.E.;YINGER,. J.M. 1958. Racial Research Council. (Bulletin no. 57 .)
and cultural minorities, (rev.ed.) N e w York, . 1958. Racial and cultural relations.
Harper Bros. In: J.B. Gittler (ed.). Review of sociology.
SIMS, B.M.;PATRICK,J.R 1936. Attitude toward N e w York, Wiley.
the Negro of Northern and Southern college stu- YOUNG,D. 1927. Some effects of a course
dents. Journal of socialpsycholom,7, p. 192-204. in American race problems on the race
T A U S C H , A.M. 1961. Experimentalle Undersu- prejudice of 450 undergraduates at the
chungen ilber Art und Ausmass des Verstand- University of Pennsylvania. Journal of
nisses von Erziehern gegentiber Kindern. Paper abnormal and social psychology, 22,
read at the Fourteenth International Congress p. 235-42.
of Applied Psychology,Copenhagen. YOUNG,R.K.; B E N S O N , W.M.; HOLTZMAN,
TUMIN,M.; BARTON, P.; B U R R U S , B. 1958. W.H. 1960. Change in attitudes toward
Education, prejudice and discrimination: a the Negro in a Southern university. Journal
study in readiness for desegregation. of abnormal and social psychology, 60.
American sociological review, 23, p. 41-9. p. 131-3.

45
 CHAPTER 2

INTERGROUP CONTACT AND COMMUNITY STUDIES

In addition to the school as a focus of attention for And is 'contact' involved when, for instance, a
changing attitudes detrimental to healthy intergroup diplomat or some other selected representative of
relations, a variety of action programmes outside a nation or other group makes an 'official'visit to
the school have been conducted in an effort to im- another nation or another group?
prove relations between various racial, religious The conflicting results of earlier studies are
and other groups. Examples of these are adult certainlytobetraced backtothese varying defini-
education campaigns; programmes for providing tions of what is involved in contact, and to a lack
contact between the members of various groups, of specification of the conditions under which con-
both among children and adults; and projects car- tact takes place. In one early study,for instance,
ried out at the local or community level. Adult Dodd (1935) studied the effects of brief contacts
education campaigns usually involve either mass between members of different ethnic groups in the
communications media or inter-personal persua- Near East. The groups were taken to visit each
sive communications such as lectures, discussion other 's religious edifices; when the students'atti-
groups,etc. W e have discussed at length,in previous tudes were re-tested following these experiences
chapters,the problems associated with the use of no significant changes in their attitudes towards
mass media and other persuasive communications each other were found. If anything,an increase in
as well as discussion and other group-oriented the intensity of 'in-group'attitudes could be noted.
techniques in changing intergroup attitudes. In In another early study,however,Smith (1937)found
this chapter we shall discuss primarily action significant positive changes in attitudes among
programmes associated with contact between white graduate students who spent two weekends
members of various groups and community projects. in Harlem (New York) getting to know leading
members of the Negro community. Mussen (1950)
Contact and attitude change investigatedthe effect of contact between white and
Negro boys in a four-week,unsegregated summer
One of the most popular and widespread notions camp. In comparing before and after results of
concerningthe improvement of intergroup relations attitude measurement, it was found that about 25 per
is that 'contact'is,in any case,a good thing. People cent ofthe white boys showed more prejudice follow-
should 'get to know each other better', and once ing the contact and about 25 per cent showed less
they have done so, all of their preconceived notions prejudice;the middle 50 per cent showed no signi-
should disappear and they should become 'good ficant changes. A more detailed analysis of the
neighbours'. This is an inviting proposition in subjects showed the positive or negative changes
which there is no doubt a good deal of truth. to be associated with initial attitude and personality
Certainly an inestimable amount of effort is being factors (the prejudiced became more prejudiced,
expended in the world today to improve intergroup the less prejudiced became less prejudiced, and
and internationalunderstandingwhich is based upon the middle group remained unchanged) as was
the simple assumptionthat this propositionis true. predictable by the hypotheses of the 'authoritarian
However, inline with the thesis stated in the intro- personality' (Adorno,3. , 1950). In a more
duction to this paper that 'goodwill'alone is not recent study, Festinger and Kelley (1951) studied
enough,it is logicalthat we ask ourselves whether the effects on attitude change of social contacts
action research confirms this supposition. among residents in a new housing project. The
As with many other questions,there is no simple, subjects of the investigation were characterized
global answer to the question of whether contact by social isolation and unfavourable attitudes to-
is an effective means of improving intergroup atti- ward their fellow residents and toward residents
tudes and intergroup relations. First of all, of the town as a whole atthe beginning of the study.
'contact'isoftentaken to mean quite a number of Following the contact situations brought about by
different things. Certainly contact is involved when the investigators,attitude changes were once again
members of various groups live together, work such that the initially favourable subjects changed
together or engage in other activitiestogether over in a positive direction, whereas those whose atti-
a long period of time. O n the other hand, a diffe- tudes were initiallyunfavourable showed either no
rent degree (and kind) of contact is involved when positive change or, in some cases, an increase
groups or individuals,who are otherwise separated in their unfavourable attitudes.
from each other,make brief visits to each other's Our present knowledge,based on the experience
homes, places of work, places of worship,etc. ofa large number of action programmes the results

46
 Intergroup Contact and Community Studies

of which have been controlled by scientific methods previous negative attitudes and prejudices were
of attitude measurement, permits us to predict quite strong. W e have referred previously to the
more accurately what type of contact, and under investigation by Brophy (1946)in which the effects
what conditions contact is most likely to lead to of white and Negro merchant seamen working to-
positive attitude change and improvement in inter- gether were investigated. Among white seamen
group relations. who had never shipped with Negroes, 33 per cent
In general,it m a y be said that contacts between were found to be relatively unprejudiced, as mea-
members of various groups are most likely to lead sured by a ten-item attitude scale. For those white
to attitude change in a positive direction when seamen who had shipped once with Negroes, the
(a) the contacts are on a basis of equal status; percentage of non-prejudiced attitudes increased
(b) co-operativeinteraction or super-ordinategoals to 46, for those who had shipped twice to 62; and
are involved; and (c) the contact is of significant of those who had shipped five or more times with
duration. Negro seamen, 82 per cent were found to be rela-
Allport andKramer (1946)found that contact on tively unprejudiced. Also Harding and Hogrefe
an equal status basis between members of various (1952)found that white department store employees
racial and religious groups was likely to lead to a changed their attitudes toward Negroes, at least
positive change in attitude,whereas contacts on an in the occupational sphere, after the stores had
unequal status basis was likely to lead either to begun to employ Negro co-workers.
no change or to a negative change in attitudes. In spite of friction that has occurred in certain
MacKenzie (1948)likewisefound that equal status instances, sometimes even to the point of violence,
contacts between Negro and white university stu- as a result of the integration of housing among
dents was most likely to lead to positive attitude previously segregated groups, a number of studies
change. have shown that living together in close proximity
The 'Robbers Cave' experiment, referred to has a positive effect upon attitudes. The well-
in Part I, Chapter 2 (Sherif,Harvey, White,Hood known study by Deutsch and Collins (1951) has
and Sherif,1954), demonstrated that contact alone demonstratedthis point very clearly. These authors
was not sufficient to improve attitudes between the studied the attitudes of white housewives toward
two experimentally created groups. Only when Negroes in two integrated, interracial public
super-ordinate goals were involved, i.e. the two housing projects, as compared with the attitudes
groups had a task to perform which required their of similar subjects in two segregated housing pro-
co-operative effort, was there a positive change jects. The residential contact between different
in attitude and a breakdown in negative out-group ethnic groups in the integrated housing projects
stereotypes. A number of other 'reallife' inves- constituted anon-competitive, equal-status contact
tigations have shown that contacts involving joint situation of long duration and, as was expected,
co-operative effort lead to positive attitude change. resulted in substantial favourable changes in
One of the most impressive of these studies was intergroup attitudes. The authors found that 53 per
an investigation of the effect of integrating Negro cent of the housewives in the integrated housing
infantry platoons into otherwise white infantry projects favoured a policy of interracial integra-
companies in the United States A r m y toward the tion for city housing projects in general whereas
end of World W a r 11. U p until the time of this only 5 per cent of the white housewives in the
integration, it had been the general policy of the segregated projects favoured such a policy. This
United States A r m y to keep Negro troops out of experience ininterraciallivingalsomade the house-
combat assignments. In the spring of 1945, how- wives more willing to accept Negroes as fellow
ever, Negro platoons were assigned to some white workers on a job,as fellowmembers in an informal
infantry companies in eleven combat divisions social club, as school mates for their children,
operating in Europe. A study conducted later and more ready to accept a Negro as mayor of
(Star, Williams and Stouffer, 1958) showed that their city. By reason of the unique experimental
of the white enlisted m e n in companies to which design of this investigation,the authors were able
Negro platoons had been assigned, 64 per cent to show that the changes in attitude took place along
thought that this was a good general policy for the a number of dimensions, e.g. significant changes
army to follow,whereas in divisions containing no took place in beliefs about Negroes (cognitive c o m -
Negro combat platoons, only 18 per cent of the ponent), feelings toward Negroes (affective c o m -
listed m e n thought that Negro platoons would be a ponent) and policy orientation toward Negroes
good idea; 62 per cent of the m e n in these latter (conative component). Although there m a y have
divisionsthoughtthat they would dislike very much been some difference in initial attitude among the
serving in a company with aNegroplatoon,where- white housewives in the integrated, as compared
as only 7 per cent of the m e n who had had such with the segregated,housing projects, the authors
experience said they would dislike it very much. were able to show by differential analysis that such
A number of other studies have shown that when initial differences could have accounted to only a
members of previously segregated groups are very slight degree for the differences in attitude
brought together in occupational situations (i. e. found after several years of living in the project.
situations usually involving co-operative efforts), Thus there is no doubt that the actual contact situa-
positive attitude change occurs, even though tion itself was largely responsible for the attitude

47
 Intergroup Contact and Community Studies

change. A n interestingfurther result ofthis inves- concerning various aspects of the problem. Even
tigation was'related to the question of the extent to a brief review of research results in this area
which chaQges in attitude toward one ethnic group would exceed the space available here, and fur-
tend to be accompaniedby similar changes in atti- thermore, a number of excellent reviews of such
tude toward other groups. The interview schedule research exist (e.g., Tumin, 1957, 1960;
included social distance questions with regard to Suchman, Dean and Williams, 1958).
Chinese and Porto Ricans as well as to Negroes. W e should like to mention here just one aspect
The former groups were almost entirely absent of this problem which is most likely to be appli-
from the projects, so that any differences in atti- cable to school segregation along racial, religious
tude toward them in the integrated projects, as or other lines which exists in other parts of the
compared to the segregated ones, must be attri- world. It has long been known and demonstrated
buted either to the effects of the somewhat greater in a number of investigationsthat such segregation
ihitial liberalism of the tenants in the integrated has harmful effects upon children belonging to the
project or to the effects of afunctional relationship minority group. They tend to incorporate the pre-
between attitudes toward these groups and attitudes judices of the majority group and develop feelings
towards Negroes. Whereas attitudes toward Porto of insecurity,lack of self-esteem,etc. However,
Ricans were only slightly more favourable in the as Stevens (1958)has pointed out,such segregation
integrated projects (no more than would be expected is harmfulto the children belonging to the majority
from the differences in education and political liber- group as well. Referring to segregated schools in
alism of the subjects), attitudes toward Chinese the South of the Unitedstates,Stevens states: 'The
were substantiallymore favourable in the integrated white child in our society who learns to be pre-
projects, especially with respect to willingness to judiced has been subjected to social situations
have Chinese tenants in the same building. These containing inherent inconsistencies. He is taught
differences were toolarge to be explained in terms justice and fair play by the same persons and insti-
of sampling error or in terms of initial differences tutions who punish him for playing with a child of
in attitude. The authors conclude that among their a different colour, or who encourage him to feel
subjects there was some functional relationship a superiority which the child often senses has an
between attitudes toward Negroes and attitudes unrealistic basis. The fact that the reactions of
toward Chinese,but little or no functionalrelation- childrento social and parental pressure to commit
"
shipbetween attitudes towardNegroes and attitudes specific prejudiced acts are generally sub-clinical
toward Porto Ricans. The necessity of further in nature is not indicative of an absence of harm-
studies to clarifythese results was stressed. Other fulness. Most negative experiences of children
investigations(e.g. Wilner,Walkley and Cook, 1952; which contribute to later maladjustment also have
Sussman, 1957; Irish, 1952) have confirmedthese sub-clinical immediate reactions'(ibid., p. 31
findings, i.e. that contact of the type which is et seq.).
brought about by living together in integrated Stevens is not alone in his judgement. As
projects definitely tends to bring about positive Deutscher and Chein showed in a survey of more
changes in intergroup attitudes. than 500 social scientists,83 per cent of the social
One verymajor area in which 'contact'between scientists who replied to their questionnaire held
members of various groups m a y take place and be the opinionthat racial segregation had detrimental
of significance for forming or changing intergroup psychologicaleffects on members of the privileged
attitudes is the school itself. In addition to the group. Although, as far as we know, there is a
more formal aspects of learning in the school, lack of empirical evidence, we must hypothesize
or of specific programmes designed to change that school segregation along racial, religious or
intergroup attitudes in a positive direction (such other group lines in many parts of the world has
as special courses, educational films, etc .), detrimental effects,not only on the minority child-
the school is a major experiential field in which ren, but on the majority children as well. Thus
children can learn healthy, democratic intergroup 'contact'is not only a method which m a y be useful
behaviour through actual example, i. e. , through in changing negative intergroup attitudes,but m a y
coLoperative interaction between various groups be regarded as a conditio sine qua non for the
within the school. The segregation of schools along development of healthy, democratic intergroup
racial,religious or other group lines denies child- attitudes in a pluralistic society.
ren the opportunity for such healthy experience Another area in which 'contact'has been used
and is, in fact, contrary to the idea of democratic extensively as a means of changing intergroup,in
interaction among school groups. One of the most this case international, attitudes has been that
crucial examples of social change in this respect of international exchange programmes. There
taking place in the world today is the desegregation has been a tremendous amount of effort and a fair
of public schools in the South of the United States amount of research in this area, so that a special
of America, where the previous segregation along paper would be necessary to cover this topic alone;
racial lines was declared unconstitutional by the we shall make only a few tangential references
United States Supreme Court in 1954. This here.
process of change is still going on and a huge body Such programmes have included international
of scientific research results has accumulated s u m m e r camps, or other meetings among youths

48
 Intergroup Contact and Community Studies

of different nations; exchanges of students involv- emphasizing the need for a more careful experi-
ing stays of one year or more in other countries; mental design.
visits of lesser duration by youthleaders, c o m m u -
nityleaders,business leaders,governmentofficials, Community studies
etc. Bjerstedt (1958),for example,in investigating
the effects of international children's camps upon One of the most serious limitations on the action
national stereotypes,found indications of reduced programmes discussed thus far, such as educa-
prejudice following such experience; however,the tion, personal and group contacts, etc., is that
author raises certain objections to the manner in they usually reach a rather limited number of
which such work has hitherto been conducted. individuals; and what is more important, efforts
Danckwortt (1956)in studyingthe effects of interna- in this direction often are at variance with larger
tional work camps for adolescents has also pointed sub-cultural or group norms. W e have referred
out that such contact m a y or m a y not lead to posi- previously (Part I, Chapter 2)to the importance of
tive attitude change and improved international group norms for the formation and change of inter-
understanding,depending upon the manner in which group attitudes. MacIver (1948) has s u m m e d up
the programmes are carried out. A large number this point very succinctly: 'However prejudice
of programmes and organizations are dedicated to m a y first arise, a major determinant of its perpe-
internationalyouth exchange both on an individual and tuation is simply the tendency of the members of
group basis. A description of a number of such any group to take on the coloration of the established
programmes is contained in a report of a study semi- mores' (ibid.,p. 198).
nar held at the Unesco Youth Institute, Gauting/l, Studies which have the community as a focus
In additionto youth exchange programmes,inter- take cognizance of this fact in trying to 'swing'
national exchange programmes involving university whole communities, as it were, to change esta-
students studying for a year or more in another blished norms and practices in the direction of
country have received considerable attention. improving intergroup attitudes and establishing
Recent examples of such studies are Coelho (1958), equality for all groups. This usually involves
studying the reactions of Indian students in the enlistingthe aid of politicians, communityleaders,
United States, and Kelman (1960) on Scandinavian local newspapers and radio stations, and other
students in the United States. Smith (1956) has 'opinion makers'to bring about changes on a com-
reviewed a number of such studies carried out in munity-wide basis. Of course this is easier said
1955. Gullahorn and Gullahorn (1960) have pro- than done, since precisely those individuals whose
vided a very interesting discussion of 'the r6le of aid must be enlisted m a y be among the most
the academic m a n as a cross-culturalmediator'. prejudiced, or m a y have vested interests in main-
Further exchanges of persons of a wide variety taining the status quo. O n the other hand, as
have been studied. Examples of such studies are Simpson and Yinger (1958,p. 769) have pointed
those of Watson and Lippitt (1955), who studied out, the belief, in equality is in many cases 'widely
areas of 'defensiveness' and 'sensitivity' on the shared but ...its violation has often been overlooked
part of German leaders visiting the United States, because many people did not believe that discrimi-
and Riegel (1953). who studied especially the nation existed, or they felt no sense of personal
'residualeffects'of exchange of persons, i.e. what responsibility, or tliey saw no way in which they
happens to the visitors after they have returned to could take effective action'.
their own country. Anumber of other studies have One of the most interesting and promising of
been made of this question. Danckwortt (1959) has community-oriented programmes that has been
compiled afairly completebibliography of research developed in recent years is the so-called 'com-
on exchange of persons up to 1959. munity self-survey technique ' developed by C .S.
An example of an attempt to integrate this type Johnson and associates at Fisk University(Sel1itz
of research into the broader area of attitude change and Wormser, 1949; Wormser and Sellitz, 1951).
is contained in a special recent issue of the Public Inthis method, help is enlisted of prominent citizens
Opinion Quarterly, edited by D. Katz, in which of the community whose opinions carry weight,
Jacobson, Kumata and Gullahorn (1960) review and the aimisto gatherfacts and prepare a report
a number of studies constituting 'cross-cultural upon the actual situation in their own community.
contributions to attitude research'. As a final This avoids the charge of 'interference from
general remark to this type of study, it must be without'. Also, as Haring (1949) has pointed out,
concluded that most studies of the effects on atti- even individuals who are themselves prejudiced
tude change resulting from international exchange can scarcely deny facts gathered by such a group.
programmes have a long way to go in order to In line with the principles of group dynamics,
meet the standards of recent experimentalresearch participation in such activity itself is very likely
pertaining to the question of attitude change. Of to have an effect on the participants' attitudes.
course,there are considerable difficultiesinherent
in such studies, since in investigating the effects 1. A n Analysis ofthe Impactof InternationalTravel
of such global experiences as a year's stay abroad and Exchange Programmes on Young People:
it is difficult to isolate specific influences on Report of a Study Seminar, 9-13 M a y 1960,
attitude change. Recent studies in this area are Gauting/Munich, Unesco Youth Institute, 1960.

49
Intergroup Contact and Community Studies

Simpson and Yinger (1958, p. 768 et seq.) describe DODD,S.C. 1935. A social distance test in the
this effect: 'The fact that they are working together Near East. American journal of sociology,
on a c o m m o n task, affects their attitudes. They are 41, p. 194-204.
more likely to accept data which they themselves FESTINGER, L.; KELLEY,H.H. 1951.
have gathered, for the facts become their facts; Changing attitudes through social contact.
their unity as a functional group obscuresthe ethnic Ann Arbor, University of Michigan, Institute
lines that divide them; and the group factor the- for Social Research.
development of in-group support to new ideas and GULLAHORN,J.T.; GULLAHORN,J.E. 1960.
-
feelings also makes for acceptance. ' The r61e of the academic m a n as a cross-
Without question further work along these lines cultural mediator. American sociological
would prove fruitful. Although basic research on review, 25, p. 414-17.
various aspects of attitudes and attitude change is HARDING,J.; HOGREFE,R. 1952. Attitude of
- -
highly useful indeed indispensable one must not
lose sight of the fact that action programmes
white department store employees toward
Negro co-workers. Journal of social issues,
directed at social change must take into considera- 8, p. 18-28.
tion the total social situation within which attitudes HARING,J. 1949. Some basic principles of self-
or behaviour are to be found. Thus sociological surveys. Journal of social issues,5, p. 21 -9.
approaches are indispensable for the success of IRISH, D.P. 1952. Reactions of Caucasian
major programmes of action. The work of Tumin residents to Japanese-American neighbors.
and his associates (Tumin, Barton and Burrus, Journal of social issues, 8, p. 10-17.
1958) is a promising example of this type of ap- J A C O B S O N , E.; KUMATA, H.; GULLAHORN,
proach. O n the other hand it cannot be stressed J.E. 1960. Cross-cultural contributions to
too strongly that an interdisciplinary approach is attitude research. Public opinion quarterly,
essential. It is to be hoped that we shall continue 24, p. 205-23.
to see the development of scientists and practi- KELMAN, H.C. 1960. Effects of cross-cultural
tioners who, in addition to their competence and experience on national images: a study of
expertise in their o w n field, have an appreciation Scandinavian students in America. Paper read
of the contribution made by research findings in at the Sixteenth International Congress of
ancillary fields and are capable of working together Psychology,Bonn.
in interdisciplinary teams. MacIVER, R.M. 1948. The more perfect union.
N e w York, Macmillan.
MacKENZIE, D.K. 1948. The importance of
BIBLIOGRAPHY contact in determining attitudes toward Negroes.
Journal of abnormal and social psychology,43,
ADORNO, T.W.; F R E N K E L - B R U N S W I C K , E.; p. 417-41.
LEVINSON, D.; S A N F O R D , R.N. 1950. , M U S S E N , P.H. 1950. Some personality and
The authoritarian personality. New York, social factors related to changes in children's
Harper Bros. attitudes toward Negroes. Journal of abnormal
ALLPORT, G.W.; KRAMER, D.M. 1946. Some and social psychology, 45, p. 423-41.
roots of prejudice. Journal of psychology, 22, REIGROTSKI, E.; A N D E R S O N , N. 1960. National
p. 9-39. stereotypes and foreign contacts. Public opinion
B J E R S T E D T , A. 1958. Reduction of 'barrier quarterly, 23, p. 515-20.
tendencies during experience of international RIEGEL,O.W. 1953. Residual effects of
co-living. Acta psychologica, 13, p. 329-46. exchange of persons. Public opinion quarterly,
BROPHY,I.N. 1946. The luxury of anti-Negro 17, p. 319-27.
prejudice. Public opinion quarterly, 10, SELLITZ, C.; W O R M S E R , M.H.(eds.). 1949.
p. 456-66. Community self-surveys: an approach to social
COELHO,G.V. 1958. Changing images of change. Journal of social issues, 5, no. 2.
America: a study of Indian students' SHERIF, M.; HARVEY, O.J.; WHITE,B.J.
perceptions. Glencoe, Illinois, Free Press. HOOD,W.R.; SHERIF, C.W. 1954.
DANCKWORTT,D. 1956. Educational problems Experimental study of positive and negative
in work camps for adolescents. Report on a inter-group attitude between experimentally
sociological survey conducted in the inter- produced groups: Robbers Cave study.
national youth camps of IJGD in Germany. Norman, Oklahoma, University of Oklahoma
Hamburg (mimeo.). (multilithed).
. 1959. Literatur zur Analyse des SIMPSON, G.E.;YWGER, J.M. 1958. Racial
internationalen Austausches von Personen, and cultural minorities: an analysis of
Koln, SozialwissenschaftlicherStudienkreis prejudice and discrimination (rev. ed.),
far Internationale Probleme . N e w York, Harper Bros.
D E U T S C H , M.; COLLINS, M.E. 1951. s- SMITH, B.L.;SMITH, C.M. 1956. International
racial housing: a psychological evaluation of communicationand public opinion: a guide to the
a social experiment. Minneapolis, University literature. Princeton, N e w Jersey, Princeton
of Minneapolis Press. University Press.

50
 Intergroup Contact and Community Studies

SMITH,F.T. 1937. A n experiment in modifying TUMIN,M. 1957. Segregation and desegrega-

attitudes towardthe Negro. In: Murphy, Murphy tion: a digest of recent research. N e w Y o r k ,
and Newcomb (eds.). Experimental social Anti-Defamation League.
psychology. N e w York, Harper Bros. . 1960. Segregation and desegrega-
SMITH, M.B. (ed.). 1956. Attitude andadjustment tion: a digest of recent research, 1956-1959
in cross-culturalcontact: recent studies of (supplement). N e w York, Anti-Defamation
foreign students. Journal of socialissues, 12, League.
no. 1. ; BARTON, P.; B U R R U S , B. 1958.
S T A R , A.; WILLIAMS, R.M.Jr.; STOUFFER, Education, prejudice and discrimination: a
S.A. 1958. Negro infantry platoons in white study in readiness for desegregation.
companies. In: E.E. Maccoby, T.M.Newcomb American sociological review, 23, p. 41-9.
and E.L. Hartley. Readings in social psycho- W A T S O N , J.; LIPPITT,R. 1955. cearning
-, 3rd ed. N e w York, Holt. across cultures. Ann Arbor, University of
STEVENS,R. B. 1958, Some emotional aspects of Michigan Press.
desegregationsin schools for Negro and white WILNER, D.M.;WALKLEY, R.T.; COOK,
children. In: The r6les of the social sciences in S.W. 1952. Residential proximity and
desegregation: a symposium. N e w York, international group relations in public hous-
Anti-Defamation League. ing projects. Journal of social issues, 8,
S U C H M A N , E.A.; DEAN, J.P.; WILLIAMS,R.M. p. 45-9.
Jr. 1958. Desegregation: some propositions WILSON, E.C .; BONILLA,F. 1955. Evaluating
and research suggestions. N e w York, Anti- exchange of persons programs, Public opinion
Defamation League. uarterly, 19, p. 20-30.
SUSSMAN, M.D. 1957. The r6le of neighbourhood WO%MSER. M.H.: SELLITZ. C. 1951. H o w to
associationin private housing for racialminori- conduct a community self-surveyof civil rights.
ties. Journal of social issues, 13, p. 31-7. N e w York, Association Press.

51
 CHAPTER 3

CULTURAL I N F L U E N C E S AND THE ROLE OF SOCIETY

That the cuPture or society as a whole is of great changes are possible? Bogardus (1958)investigated
importance for the formation, maintenance and/or changes in 'social distance' among various racial
change of social attitudes is a self-evidentpropo- and ethnic groups in the United States over a thirty-
sition to which we have referred on a number of year period. Although some groups showed an
occasions. But to what extent are practical pro- amazing degree of stabilityin the amount of social
grammes of action with such a broad orientation distance attributed to them by the respondents over
feasible? H o w can culture,backed by centuries of the thirty-yearperiod, and the total social distance
tradition,be changed or whole societies modified? scoreshadnot decreased appreciably,on the other
Of course, in a modern democratic pluralistic hand, marked changes were to be found among a
society it is scarcely possible for one or a few number of groups with respect to social distance.
individuals to make radical changes in the whole These were primarilypositive changes (less social
society overnight. Onthe other hand,it is beyond distance)in attitude toward those groups with res-
question that groups of individuals, through con- pect to whom there had been concerted efforts in
certed efforts and the application of modern know- the United States duringthe past decades to reduce
ledge, can and should work toward speeding up prejudice and discrimination,andnegative changes
socialchange and overcomingcertain 'culturallagsI. toward certain other groups due to major political
While one must guard against over-optimism, and military events. These data show quite con-
we cannot agree with those pessimists who contend clusively that concerted programmes to improve
that no major changes are possible. This point of intergroup attitudes have along-term success, and
view, by the way, is often held by representatives that major political and other events can also have
of quite different disciplines. O n the one hand, marked effects upon attitudes. Gilbert (1951),
certain psychologically-orientedresearchers hold in a replication of the original study by Katz and
the extreme view that prejudice is an expression Braly (1933) on stereotypes, used a comparable
of deep-lying, irrational tendencies which are group of subjects and found that there was appre-
essentially unchangeable; this fits in with the ciablyless readiness on the part of the students to
'you-can't-change-human-nature' view, which has generalize about ethnic groups than had been the
long been the rationalization of sceptics, and which case 18 years previously. Not only did this later
has oftenfound some credence with the layman. O n generation of students assigntraits muchless fre-
the other hand, certain students of social structure quently to various groups,but a significantnumber
have arrived at similar conclusions; proceeding of the subjects spontaneously voiced a 'protest
from the observationthat certain social phenomena against the unreasonable task of making genera-
have their causes and function, and pointing out -
lizations about people especially those they had
that 'all'societies at alltimesthat we have studied hardlyever met'(Gilbert, 1951, p. 251). W e have
have manifested these phenomena, they argue that referred previously to the conclusions arrived at
-
the status quo can scarcely be changed 'What is byTumin, Barton andBurrus (1958) as to the per-
and always has been will logically continue to be. ' vasive effects of formal education upon attitudes.
One well-known researcher in the area of pre- In a slightly different(though very pertinent) area,
judice, in propounding this point of view during a Bronfenbrenner (1958) has shown that major and
discussion of the problem, went so far as to state significant changesin child-rearing practices have
that teachers could scarcely be expected to help in taken place in the United States over a twenty-five-
overcomingprejudice since the 'r6le'of a teacher, year period, and furthermore that these 'shiftsin
his 'function'in the society, was to transmit the -
the pattern of infant care especially on the part
existing cultural values, and since these values -
of middle-class mothers show a striking corres-
included prejudice,the teacher must transmit such pondence to the changes in practices advocated
prejudice in order to fulfil his 'function'and IrSle' in successive editions of U.S. Children's Bureau
in the society. T o the extent that these remarks bulletins and similar sources of expert opinion'.
were merely designed to point to certain diffi- The above-mentionedstudies were merelyillus-
culties encountered in an attempt to change social trative of the fact that major changes in attitudes
norms, they certainly have their value. O n the and behaviour on the part of a society as a whole
other hand, to the extent that they are taken to over a period of time are possible. Numerous
mean that major social changes are impossible, other investigations could be cited to prove this
we must disagree rather sharply. point. These changes are probably a cumulative
What empirical evidence have we that social result of the combined efforts of many individuals

52
 Cultural Influences and the R61e of Society

and organizations, using a number of different effect upon the opinion of the population.
methods and approaches. The question which In contrast to an opinion which held certain
interests us here, however, is to what extent are currency a number of years ago, most social
approaches on a society-wide basis feasible. W e scientists and others concerned with intergroup
can't discuss in detail here the broad issues of relations are agreed today that legal measures
social theory implicit in this question. However, have a very important r6le to play in intergroup
we do wish to point to some of the methods which relations. As Raab and Lipset (1959, p. 41 et seq.)
are feasible at this level. For example, there is have pointed out, 'theimportance of law is not that
no doubt that the public statements of top govern- it prohibits or eliminatesprejudiced behaviour and
ment officials and other leading personalities whose -
attitudes but that it changes the social situations
opinion is respected by the populace can have a and community practices wnich breed prejudiced
widespread effectupon the attitudes and behaviour behaviour and attitudes. Since these situations and
of a whole population. Another somewhat more practices are the prime learning influences with
binding method by which society m a y make its will respect to prejudice, law then must be perceived
felt upon individual members m a y be seen in legal as a prime educational weapon in combating prej-
measures. Also of crucial importance for the udice ... Laws hasten the elimination of preju-
everyday life, and hence for the attitudes and diced "cues" in the community'. A number of other
behaviour, of individuals are certain economic authors have stressed the importance of legal meas-
factorsprevalentinthe society as a whole. Finally, ures as a means of effectively changing undemo-
we should like to mention certain broad issues cratic intergroup attitudes and behaviour (e.g.
associated with the use of mass communications Myrdal, 1944). Maslow and Robinson (1953) and
media, since these constitute one of the primary Berger (1952, 1954) have given excellent accounts
vehicles by means of which the society conveys its of the r61e of legislation in intergroup relations in
cultural norms and expectations to its members. the United States. Of course, one of the most im-
Numerous investigations have shown that the portant examples in this respect in recent years
statements of importantpoliticalleaders and other was the United States Supreme Court decision of
influential personalities m a y affect the attitudes 1954 declaring racial segregation in schools to be
and behaviour of the populace (e.g., Hovland, unconstitutional. Although much work still remains
Lumsdaine and Sheffield, 1949; Hovland and Weiss, to be done before the full implementation of this
1951). One recent study has demonstrated this decision has been accomplished,it must not be
point quite vividly. Following the epidemic of forgotten amidst the headline-making instances of
anti-Semitic incidents which began with the dese- violence following this decision, that the majority
cration of the Jewish Synagogue in Cologne, on of localities have accepted and implemented this
Christmas eve 1959, and spread not only through decision,thus effectively shattering long-established
western Germany but also through various other patterns of inequality. Itis not without some pride
countries, the Institute for Social Research in that social scientists note that this important
Frankfurt a m Main conducted an investigation of decision was based to a significantextent upon the
the attitudes of Frankfurt residents toward these expert testimony of qualified social scientists who
incidents (Schonbach, 1961). Attitudes toward demonstratedthe harmfulness of such segregation,
various issues related to the incidents were assessed and effectively countered the pseudo-scientific
by means of interviews. Among other things, it arguments infavour of segregation. Closely related
was determined whether or not the respondents to direct legal measures are administrative pro-
felt these incidents to be politically significant cedures relating to intergroup relations. Racial
or largely attributable to pranks, etc. B y coinci- integrationin the United StatesArmed Forces was,
dence,just slightly less than half of the interviews for instance, conducted by administrative order.
had been conducted when, on the evening of 16 Also in the area of public housing much has been
January 1960, Federal Chancellor Adenauer c o m - accomplished by administrative procedures. A s
menting upon these incidents in a radio and Deutsch an2 Collins (1951) have shown, such
television speech said 'But in most cases they measures can have a very positive effect upon inter-
-
/the incidents/ seem to have been pranks without group relations. However, as McEntire (1957)
political foundation. This is also,as I was told, has shown, much remains to be done in this area.
the general opinion of the heads of the security of- The r61e of economic factors has been stressed
fices of the Lander'/l. In comparing the answers by a number of authors. Harding, Kutner,
of those subjects who were interviewed before Proshansky and Chein (1954)point to the inconsis-
Dr. Adenauer's spe'ech with those who were tency among the numerous correlational studies
interviewed afterwards, it was found that where- of the relationship between prejudice and socio-
as only 7 per cent of those interviewed before economic status. However, most authors stress-
the speech regarded the incidents as pranks, 18 ing economic factors have pointed rather to the
per cent of those subjects interviewed after the basic economic structure of the society than to the
speech held this opinion (this difference is statis-
tically significant at the 5 per cent level). Thus 1. Die antisemitischen und nazistischen VorfUle,
it m a y be concluded that the public statement of Weissbuch und Erklarung der Bundesregierung,
an important political leader had a significant Bonn, Bundesregierung, 1960, p. 40.

53
Cultural Influences and the R61e of Society

economic status of individuals.Opinions vary here with the principles of peaceful and harmonious
very widely, with certain classical psychologists relations between all groups?
minimizingthe importance of economic factors and
many sociologists and economists emphasizing
these factors. Indeed, social scientists with a BIBLIOGRAPHY
classical Marxian orientation regard the import-
ance of economic factors as crucial and even many
psycho-analytically oriented writers have stressed ADORNO, T.W.; FRENKEL-BRUNSWICK,E.
the importance of the total socio-economic struc- LEVINSON, D.; S A N F O R D , R.N. 1950.
ture of the society (e.g. Horney, 1937; Adorno The authoritarian personality. New York,
-
et al., 1950; F r o m m , 1955). To sum up, most
social scientists would agree thatthese factors are,
Harper Bros.
ALLPORT,G.W. 19 59. ABC 's of scapegoating,
quite important and Allport's (1959) statement prob- (3rded .)N e w York, Anti-Defamation League.
ably expresses the view of the majority of them: BERGER, M. 1952. Equality by statute. N e w
'Economic insecurity breeds the frustration and York, Columbia University Press,
fear that are part of the soil of scapegoating. Ruth- . 1954. _Racial equality and the law.
less economic competition often means that one Paris, Unesco.
man's success requires another'sfailure. The "ins" BOGARDUS , E.S. 1958. Racial distance changes
insistthat if anyone failsit shallbe the "outs". The in the United States during the past thirty years.
basic remedies would seem to be: (a) raising the Sociology and social research, 43, p. 127-34.
standard of living of all, thus eliminatingthe need BROFENBRENNER, U. 1958. Socialization and
for competition for survival among marginal groups; social class through time and space. In: E.E.
(b) establishing social and educational securityfor Maccoby, T.M.Newcomb and E.L.Hartley
the individual,thus enhancing his feelings of status (eds.). Readings in social psychology. New
and lessening feelings of inferiority and apprehen- York, Reinhart and Winston.
sion; and (c) proper vocational guidance and place- DURR,C.J. 1954. Freedom of speechfor whom?
ment to help prevent feelings of dissatisfaction and In: D.Katz, D. Cartwright, S. Eledersveld
jealousy amongmembers of both majority and minor- and A.M. Lee (eds.). Public opinion and
ity groups' (= .
, p. 36). propaganda. N e w York, Dryden Press.
Finally, as a major source of social influence FROMM, E. 1955. The sane society. New York,
we should like to mention briefly the mass com- Reinhart and Co.
munications media. W e have discussed previously GILBERT,G.M. 1951. Stereotypes,persistence
(Part I, Chapter 3) experimental findings relating and change among college students. Journal of
to the effects of mass communications media. abnormal and socialpsychology,46, p. 245-54.
However, from the broader perspective of mass HARDING,J , ; KUTNER, B.; P R O S H A N S K Y , H.;
media as an instrument of society to change the CHELN,I. 1954. Prejudice and ethnic
attitudes of individuals in a desired direction, it relations. In: G. Lindzey . Handbook of social
is important to consider the question of the intent psychology, vol. 11. Reading,Mass,, Addison
of mass media communications; or, in other Wesley Pub. Co.
words, the question of who controls these media. HORNEY,K. 1937. The neurotic personality of
Of course, this is a very controversial question, our time. N e w York, Norton and Co.
with far-reaching political implications which we H O F L m C .I.; L U M S D A I N E , A.A.; SHEFFIELD,
shall not discuss here. W e do wish to point out, F.D. 1949. Experiments on mass communi-
however, that this question cannot be ignored in cations, Princeton,University Press.
the long run. Indifferentcountries,different forms ; WEISS, W. 1951. The influence
of control are exercised over the mass communica- of source credibility on communication effec-
tion media. In some, these media are considered tiveness. Public opinion quarterly, 15,
as 'business'and, in line with the principles of p. 635-50.
'free enterprise', are owned or controlled by pri- KATZ, D.; BRALY, K.W. 1933. Racial
vate individualsor groups of individuals. A number stereotypes of one hundred college students
of authorshave pointed out certain dangers which Journal of abnormal and social psychology,
m a y be involved when these media become con- 28, p. 280-90.
centrated in the hands of vested interests (e.g. LEE,A.M. 1954. Freedom of the press. In:
Durr, 1954; Lee, 1954). In other countries, such D.Katz, D. Cartwright, S. Eldersveld and
media are State controlled. In still other coun- A.M. Lee (eds.). Public opinion and
tries, they are controlled by independent com- propaganda. N e w York, Dryden Press.
missions composed of representatives of the many M A S L O W , W.; ROBINSON, J.B. 1953. Civil
different interest-groups within a pluralistic rights legislation and the fight for equality,
society. In all cases,the question must be asked: 1862-1952. University of Chicago law review,
to what extent do these media serve the interests 20, p. 363-413.
of society as a whole (i.e.the whole human society), M c E N T I R E , D. 1957. Government and racial
and how can they best be employed to change discrimination in housing. Journal of social
attitudes in a direction which is in accordance issues, 13, p. 60-7.

54
 Cultural Influences and the R61e of Society

MYRDAL, G. 1944. A n American dilemma. S C H O N B A C H , P. 1961, Reaktionen auf die antisemi-

The Negro problem and modern democracy. tische Welle im Winter 1959-1960, Frankfurter
N e w York, Harper Bros. Beitrage zur Sozioloae , Sonderheft 3. Frankfurt
RAAB, E.; LIPSET, S.M. 1959. Prejudice a m Main.
and society. N e w York, Anti-Defamation TUMIN,M.;BARTON,P.;B U R R U S , B. 1958. Edu-
League. cation,prejudice and discrimination: a study in
readiness for desegregation. American sociolo-
gical review, 23, p. 41-9.

55
 S U M M A R Y AND P R O S P E C T S

Although the scientific study of intergroup relations W e have alreadymentionedthe views of Tumin and
is approximately a half century old, and practical co-workers in this respect (Tumin, Barton and
efforts to improve intergroup relations have been Burrus, 1958). Rose (1956) has taken an even
made in increasing number and intensity in the past more extreme view,assertingflatlythat 'prejudice
three or four decades, the bulk of scientific re- has little to do with intergroup relations.. . The
search directed at the question of 'attitude change' laws, ... which govern one are independent of those
has developed within the past 15 years, and espe- governing the other. One is socio-genic and the
cially within the past decade. As we clearly stated, other psycho-genic, and while they m a y both in-
it was not our intentto reviewthe whole vast field of habit the same individual at the same time,they also
racial or other intergroup relations/l ; nor could m a y not'. This author goes onto saythat whereas
we attempt a complete review of the many action 'there are undoubted advantages to interdisciplinary
programmes to improve intergroup relations being co-operative research..., there are sometimes
conducted throughout the world. Especially in the also advantages to division of labour, to the culti-
field of education we should like to have discussed vation of one's o w n specialized garden' (ibid., p.
a few excellent examples of action research in 176). This questionis not only of great theoretical
greater detail; fortunately,however, this has been importance, but also of even greater practical
done very adequately in other publications (e.g. , significance. If behaviour has nothing (or practi-
James and Tenen, 1953; Clark, 1955; Bibby, cally nothing) whatsoever to do with attitudes or
1959)/2. Our aim,rather,was primarily to review other interveningvariables within the individual,but
and point to the practical implications of this is dependent merely upon external circumstances
recent research on attitude change. and m a y be manipulated at will; if indeed our aim
A critical review of the experimental and action in improving intergroup relations is merely to
research in this area unfortunately leads to the manipulate behaviour,with no concern whatsoever
same conclusion arrived at by previous authors: for attitudes,what,then,will prevent a manipula-
with afew notable exceptions,there is far too little tionintle opposite direction at some future point?
(and in the large number of cases, next to no) co- Are civil courage and inner convictions to be
operationbetween experimentalresearch and action considered mere fictions? Before embarking on a
programmes, between scientists and practitioners. programme with manipulation of behaviour as the
Especiallyinlight of the rapid progress in experi- sole aim, it m a y be well to consider again the
mental research and attendant theory in this area words of warning given by Adorno, et., (1950)
in very recent years, such co-operation becomes to which we have referred previously (cf. Part I,
more important than ever before (and,unfortuna- Chapter 1 , p. 12 et seq.). Instead of askingthe
tely, at the same time more difficult), if the gap question whether behaviour or attitude is more
between the two is not to become unbridgeable. As important, it would probably be wise to heed
w e have mentioned before, such co-operation is Klineberg's (1958)word of advice and ask, instead,
mutually advantageous, since not only will the ap- just how these two are related.
plication of recent knowledge make practical efforts Another questionto which present research has
more effective,but at the same time our knowledge given no satisfactoryanswer is the extent to which
of human behaviour is most likely to be enriched present findings can be generalized and applied to
by tackling such challenging problems. O n the different cultures. Unfortunately, most of the
basis of current researchtrends,we would predict published research on attitude change thus far has
that research in the near future will be character- been limited to the cultural setting of the United
ized by ever greater sophisticationof experimental
design and integration of the problems associated 1. For a very useful guide to studies in the more
with attitude change into the broader framework of general area of race relations, see J. Viet,
theories of human behaviour. 'Selected documentation for the study of race
One of the major unresolved issues in the field relations'. Reports and Papers in the Social
of intergroup relations is the question of the rela- Sciences, 1958, No. 9 , Paris, Unesco, 1958.
tionship between attitudes and behaviour. W e have 2. F o r a description of some practical classroom
touched upon this questionbefore and unfortunately techniques, see the Unesco pamphlet Education
space does not permit a detailed theoreticaldiscus- for InternationalUnderstanding: Examples and
sion of it here. Present trends would seem to indi- Suggestionsfor Classroom Use, Paris, Unesco,
cate that the 'behaviour'group is gaining ground. 1959.

56
 Summary and Prospects

States of America, which is reflected by the pre- on the so-called Asch phenomenon, are applied to
ponderance of American studies in this paper. the question of internationaltensions. It is our pre-
Although w e m a y have missed a few significant diction and our hope that this area will increasingly
studies conducted in other countries, since the receive the attention of social scientists and that
facilities at our disposal were far from adequate, research on attitude change will more and more
it is very probable that the proportion represented be conducted within the framework of this larger
in this paper is indicative of the true state of affairs, perspective.
The question of racial and other intergroup relations
has received research attention in other coun-
tries/l. However, as far as we can determine on BIBLIOGRAPHY
the basis of publishedliterature,and on the basis
of extensive correspondencewith colleaguesthrough- ADORNO, T.W.;FRENKEL-BRUNSWICK,E.;
out the world, in most other countries the stage of LEVINSON, D.;S A N F O R D , R.N. 1950. The
research on the problem of attitude change has not authoritatianpersonality. NewYork, H a r p e y
yet been fullyreached. The few cross-cultural stud- Bros.
ies that have been conducted have yielded interest- BIBBY,C. 1959. Race, prejudice and education.
ing results (seefor instance,Jacobson,Kumata and London, Heinemann.
Gullahorn, 1960),but there is a pressing need for BRONFENBRENNER,U. 196 1. The 'mirrorimage
more such studies. Meanwhile,great caution must ofSoviet-Americanrelations. Journal of social
be exercised in generalizingfromthe results found issues, 17.
within one cultural setting. Schtinbach (1961)has CLARK,K.B. 1955. Prejudice in your child.
illustrated this point very clearly by indicating Boston, Beacon Press.
significant differences between his study of anti- GALTUNG,J. 1960. Anti-Semitism in the making:
Semites in Germany, to which we referred in a study of American high school students. Oslo,
the last chapter, and the results of a similar Institute for Social Research (mimeo).
American investigation (Galtung,1960). J A C O B S O N , E .;KUMATA, H.;GULLAHORN,J.E.
A s a final concluding remark concerning pros- 1960. Cross-cultural contributionsto attitude
pects for future research on attitude change, w e research. Public opinion quarterly,24, p. 205-23.
should like to add a note of caution and at the same J A M E S , H.E.O.,TENEN,C. 1953. The teacher
time an expression of hope. The note of caution was black. London, Heinemann.
concerns the danger of resolving intergroupten- KLINEBERG,0.1958. Comments. In: The r61e
sions within a particular nation or other larger of the social sciences in desegregation: a
group by an appeal for unity based on animosity symposium, p. 47-8. N e w York,Anti-Defama-
toward some other nation or larger group, be it on tion League.
political, religious or other grounds. W e feel that R O S E , A.M. 1956. Inter-group relations versus
this represents a major danger in the world today. prejudice: pertinent theory for the study of
W e can follow the development of mankind from social change. Social problems, 4,p. 173-6.
the time when families or clans constituted 'in- S C H O N B A C H , P. 1961. Reaktionen auf die anti-
groups ' and battled against each other, through semitische Welle im Winter 1959-1960. Frank-
successive stages up to the point of the develop- furter Beitrage zur Soziologie, Sonderheft 3,
ment of nation-States, and we have seen that as Frankfurt a m Main.
the antagonizingunitsbecome larger the results of SHERIF, M.;SHERIF, C.W. 1956. A n outline
conflicts between them have become more devas- of social psychology,(rev. ed.) N e w York,
tating. Hence today there are antagonisms in the Harper Bros.
world between whole nations or blocs of nations, TUMIN,M.; BARTON, P.;B U R R U S , B. 1958.
and the resolution of tensionsbetween these gi-oups Education, prejudice and discrimination: a
is the paramount task facing mankind today. A s study in readiness for desegregation. American
SherifandSherif(1956)have pointed out, the reso- sociological review,23, p. 41-9.
lution of smaller antagonisms by the creation of
largerones m a y produce a certaintemporarysuccess,
but is likely in the long run to lead to even larger 1. For a review of recent research on racial rela-
conflicts. A n example of an approach to this broader tions in various parts of the world, see the two
question is a recent article by Bronfenbrenner specialissues of the InternationalSocialScience
(1961), in which some of the results of group- Bulletin, vol. X,no. 3, 1958. and vol. XIII,
oriented research, especially the experiments no. 2, 1961.

57
 AUTHORS INDEX

ABELSON,H.I. (1959) 9, 10, 32, 33 BREHM, J . W .

ABELSON,R.P. see ROSENBERG,M. COHEN,A.R. (1959) 37, 38
ADORNO,T.W.; (1950) 15, 16, 17, 19,20, BROCK, T. see HOVLAND, C.I.
FRENKEL-BRUNSWIK,E.; 42, 44, 46, 50,54. BRONFENBRENNER ,U(1958) 52, 54
LEVINSON,D.; & 56, 57 (1961) 57
SANFORD,R.N. BROPHY,I. N (1946) 24, 26, 47, 50
ALLPORT, F.H. (1924) 22, 25 BURRUS,B. see TUMIN. M.
ALLPORT, G.W. (1935) 8, 9, 10 BUTLER,J. see LEVINE,J.
(1937) 22, 25
(1952) 7, 10
(1959) 54 CAMPBELL, D.T. (1950) 9. 10
ALLPORT, G.W.& CAMPBELL, E.H. see HOVLAND,C.I.
KRAMER, D.M. (1946) 47, 50 CARLSMITH,J. H. see FESTINGER ,L.
ALLYN, J. & CARTWRIGHT,D. 8~
FESTINGER,L. (1961) 37, 38 ZANDER,A. (1953) 26
ANDERSON,N.H. -
(1959) 38 see also CHEIN, I. -
(1951) 9, 10 see also
REIGROTSKI,E. HARDING,J.
ARONSON,E. see MILLS, J. CHOO,T. (1960) 28, 33
ASCH, S.E. (1946) 31, 33 CHRISTIE,R. &
(1951) 19, 20, 23, 25 COOK, P. (1958) 15, 20
(1952) 23, 25 CHRISTIE,R. &
(1956) 19, 20, 25 JAHODA,M. (1954) 15, 20
AXLINE,V.M. (1948) 19, 20 CLARK,K.B. (1955) 56, 57
COELHO,G.V. (1958) 49, 50
COHEN,A.R. see BREHM,J.W.
BARON, L.K. (1957) 19, 20 COHLER , J. see KELMAN,H.C.
BARTON,P. see TUMIN.M. COLE,S.G. see VICKERY,W.
BECHTEREW, V.M. WIESE, M.J.
&DE LANGE,M. (1924) 22, 25 COLLINS, M.E, see DEUTSCH. M.
BENJAMIN,L. see KATZ,I. COOK,E. see COOK, L.A.
BENSON,W.M. see YOUNG, R.K. COOK,L.A. (1950) 43, 44
BERELSON,B. -
(1954) 29, 33 see also COOK, L.A. &
LAZARSFELD, P.F. COOK,E. (1954) 43, 44
BERELSON, B. & COOK, P. see CHRISTIE,R.
SALTER,P.J. (1946) 29, 33 COOK,S.W. see WILNER, D. M.
BERGER, M. (1952) 53, 54 COOPER,E.
(1954) 53, 54 DINNERMAN,H. (1951) 33
BERGIUS,R. see SOHDI,K.S. COOPER,E. &
BERKOWITZ. L 8~ JAHODA,M. (1947) 18, 20, 29, 33
COTTINGHAM,D.R. (1960) 18, 20 COTTINGHAM,D.R. see BERKOWITZ,L.
BETTELHEIM,B. & CROMWELL, H. (1950) 27, 30, 33
JANOWITZ,M. (1950) 18, 20 CRUTCHFIELD,R.F. seeKRECH,D.
BIBBY,C. (1959) 56, 57 CULBERTSON,F.J. M.(1955) 18, 20
BJERSTEDT,A. (1958) 49, 50
BOGARDUS,E.S. (1925) 9, 10
(1958) 52, 54, DANCKWORTT,D. (1956) 49, 50
BONILLA,F. see WILSON,E.C. (1959) 49, 50
BOOMER,D.F. (1959) 19, 20 DEAN,J. P. see SUCHMAN,E.A.
BOVARD,E.W. Jr. (1951) 23, 26 DEUTSCH,M. &
(1953) 23, 26 COLLINS, M.E. (1951) 24, 26, 47, 50, 53
BRALY, K.W. see KATZ, D. DINNERNLAN, H. see COOPER,E.
BRAMELD, T. (1946) 43, 44 DODD,S.C. (1935) 46, 50
BREHM, J. W. (1959) 37, 38 DROBA, D.D. (1932) 41, 44
DURR, C.6. (1954) 54

59
FENICHEL,0. (1946) 15, 20 HOVLAND,C.I.;
FESHBACH,S. (1961) 18, 20, 30, 33 CAMPBELL, E.H.
see also JANIS 1953 & BROCK, T. (1958) 31, 34
FESHBACH,S. & HOVLAND,C.I.;
SINGER,R. (1957) 18, 20, 30, 33 JANIS,I. (L.) &
see also JANIS,I. KELLEY. H.H. (1953) 28, 31, 32, 34
1953, 1954 HOVLAND,C.I.;
FESTINGER,L. (1957) 36, 37, 38 LUMSDAINE,A.A. &
see also ALLYN,J. SHEFFIELD,E.D. (1949) 32, 34, 53, 54
FESTINGER,L. & HOVLAND,C.I.
CARLSMITH,J.H. (1959) 37, 38 MANDELL, W. (1952) 28, 32, 34
FESTINGER,L. (1957) 30, 31, 34
KELLEY,H.H. (1951) 46, 50 HOVLAND,C.I. &
FINE,B.J. (1957) 28, 33 ROSENBERG,M. (1960) 38
FLAMENT, C. (1958) 19, 20 HOVLAND,C.I. 8z
FLOWERMAN,S.H. (1949) 29, 32, 33 WEISS, W. (1951) 28, 34, 53, 54
FORD,L.I. (1956) 19. 20 HYMAN, H.H. &
FRENKEL-BRUNSWIK ,E. see ADORNO,T.W. SHEATSLEY,P.B. (1947) 29, 34
FROMM. E. (1955) 54
FURUYA,T. (1958) 24, 26
IRISH,D.P. (1952) 48, 50

GALTUNG,J. (1960) 57
GAUDET,H. see LAZARSFELD,P.F JACOBSON. E. ;
GILBERT,G.M. (1951) 52, 54 KUMATA, H. &
GOLDBERG,H.D. (1950) 32, 33 GULLAHORN,J. E. (1960) 49, 50, 57
GOLDSTEIN,M.J. (1959) 18, 20 JAHODA,M. see CHRISTIE,R.
GRAHAM,E. see LINTON,H. COOPER,E.
GRANNEBERG,A.G. see MENEFEE,S. C. JAMES,H.E.0. &
GREEN, B.F. (1954) 8, 10 TENEN,C. (1 953) 56, 57
GULLAHORN,J.E. see GULLAHORN,J. T. JANIS,I. &
GULLAHORN,J.T. & FESHBACH. S. (1953) 18, 20, 30, 34
GULLAHORN, J.E. (1960) 49, 50 (1954) 18, 20, 30, 34
GUNDLACH. R.H. (1950) 24, 26 see also HOVLAND,C.I.
KING, B.T
LUMSDAINE,A.A.
HAEFNER,D.P. (1956) 30, 33 JANIS,I. ,
HARDING,J. & HOVLAND,C.I. et al. (1959) 19, 20
HOGREFE,R. (1952) 47, 50 JANIS,I. (L.)&
HARDING,J. ; KING,B.T. (1954) 37, 38
KUTNER, B. ; JANOWITZ,M. see BETTELHEIM,B.
PROSHANSKY,H. JARRETT,R.F.&
& CHEIN, I. (1954) 9, 10, 41, 44, 53, SHERIFFS,A.C. (1953) 31, 34
54 JONES,E. ; ZELL,H.
HARING,J. (1949) 49, 50 & TORREY,R. (1958) 28, 34
HARTMEN, G.W. (1936) 30, 33
HARVEY,0. J. see SHERIF,M.
HEIDER,F. (1946) 36, 38 KATZ , D. (1960) 19, 20
(1958) 36, 38 see also SARNOFF,I.
HOBAN. C.F. & KATZ, D. &
VAN ORMER, E.B. (1951) 32, 33 BRALY,K.W. (1933) 52, 54
HOFFMAN, M.L. (1951) 16, 20 KATZ , D. ;
HOGREFE,R. see HARDING,J. McCLINTOCK,C.
HOLLANDER,E.P. see TITUS,H.E. & SARNOFF,I. (1956) 18, 20
HOLTZNLAN,W.H. see YOUNG,R.K. (1957) 18, 20
HOOD, W.R. see SHERIF, M. KATZ, D. &
HORNEY,K. (1937) 54 STOTLAND,E. (1959) 18, 20
HOROWITZ,M.W. see PASTORE,N. KATZ, I. &
HOVLAND,C.I. (1954) 32, 33 BENJAMIN,L. (1960) 17, 20
(1957) 31, 34 KELLEY,H.H. (1955) 24, 26, 29, 34
(1958) 27, 33 see also
see also JANIS,I. HOVLAND, C.I.
KELMAN,H.C. FESTINGER,L.

60
KELLEY,H.H. & MAC KENZIE,D.K. (1948) 47, 50
VOLKART,E.H. (1952) 24, 26, 29, 34 MANDELL, W. see HOVLAND,C,I.
KELLEY,H.H. & 1952, 1957
WOODRUFF, C.L. (1956) 24, 26. 28, 34 MANIS, M. (1960) 32, 34
KELMAN, H.C. (1952) 19, 20, 25 MANSKE,A.J. (1935) 41, 44
(1953) 37, 38, 43, 44 MASLOW, W.&
(1956) 17, 18, 20 ROBINSON,J.B. (1953) 53, 54
(1960) 18, 20, 38, 49,50 MAUSNER, B. (1953) 23, 26
(196l)(a) 18, 20, 38 (1954) 23, 26
(1961)(b) 38 McCLINTOCK,C.
G. (1956) 18, 21
KELMAN. H.C. see also KATZ,D.
COHLER , J. (1959) 18, 21 1956-57
KELMAN,H.C. & McCORD, F. (1948) 23, 26
HOVLAND,C.I. (1953) 28, 34 McDAVID,J.W. Jr. (1958) 32, 34
KIHARA , K. see OGAWA, T. McDOUGALL,W. (1908) 22, 26
KING,B.T. McENTIRE,D. (1957) 53, 54
JANIS,I. L. (1956) 37, 38 M c GINNIES, E. see MITNICK,L.L.
see also JANIS,I.
L. McGUIRE, W.J. (1960)(a) 36, 38, 42
KLINEBERG , 0 . (1958) 10, 56, 57 (1960)(b) 36, 38, 42
KNOWER,F.H. (1935) 30, 32, 34 MEEHL, P.E.
(1936) 32, 34 MAC CORQUODALE,K.
KRAMER, D.M. see ALLPORT, G.W. MENEFEE, S. C. &
KRECH,D. & GRANNEBERG,A. G. (1940) 30, 34
CRUTCHFIELD,R.F.(1948) 9, 10 MERTON, R. K. (1957) 44
KUKAYA,S. see OGAWA, T. MIDDLETON. R. (1960) 32, 34
KUMATA, H. see JACOBS0N.E. MILLS, J. ;
KUTNER,B. ; ARONSON. E. &
WILKINS, C. ROBINSON,H. (1959) 29, 34
YARROW, P.R. (1952) 10, 36, 38 MITNICK, L.L.&
see also HARDING,J. McGINNIES,E. (1958) 33, 34
MORENO, J.L. (1953) 9, 10
MORRISSETTE , J. see ZAJONC,R.B.
DE LANGE,M. see BECHTEREW ,V. M. MURPHY, G. (1953) 25, 26
LA PIERE,R.T. (1934) 10, 36, 38 MURPHY, G. ;
LASKY,J.J. (1950) 16, 21 MURPHY, L.D.&
LAZARSFELD,P.F. (1950) 8, 10 NEWCOMB, T.M. (1937) 27, 34
LAZARSFELD,P.F.; MURPHY, L.D. see MURPHY, G.
BERELSON,B. & MUSSEN, P.H. (1950) 46, 50
GAUDET,H. (1944) 29, 34 MYRDAL, G. (1944) 53, 55
LEE, A.M. (1954) 54
LEVINE,J. &
BUTLER,J . (1953) 25, 26 NADLER , E.B. (1959) 19, 21
LEVINSON,D. see ADORNO,T.W. NEWCOMBE, T.M. (1943) 23, 26
LEWIN,K. (1935) 9, 10, 22, 26 (1948) 23, 26
(1948) 43, 44 (1950) 23, 26
(1953) 25, 26 (1953) 36, 38
(1958) 25, 26 see also MURPHY,G
LINTON. H. NOWAK, S. (1960) 24, 26, 42, 44
GRAHAM,E. (1959) 17, 19. 21
LIPPITT, R. see WATSON, J. OGAWA,T.;
LIPSET,S. M. see RAAB, E. KIHARA,K.;
LUCHINS, A.S. (1 958)(a) 31, 34 KUKAYA,S. &
(1958)(b) 31, 34 TANAKA,K. (1957) 43, 44
LUDLUM, T.M. (1958) 28, 34 ORR,D. (1946) 15, 21
LUMSDAINE,A.A.
& JANIS, I. L. (1953) 31, 34
see HOVLAND,C.I. PASTORE,N.
LUND, F.H. (1925) 27, 30, 34. 36. 38 HOROWITZ,M.W. (1955) 28, 34
PATRICK,J. R. see SIMS,B.M.
PEARL, D. (1955) 19, 21
MAC CORQUODALE,K. PETERSON,R.C.
& MEEHL, P.E. (1948) 8, 10 THURSTONE,L.L. (1933) 32, 34
MAC IVER,R.M. (1948) 7, 10, 43, 44, 4 9 , 5 0 PITKIN,V.E. (1950) 43, 44

61
PLANT,W.T. (1956) 42, 45 SMITH,C.M. see SMITH,B.L.
PROSHANSKY.H see HARDING,J. SMITH,F.T. (1937) 46, 51
SMITH, M.B.(ed.) (1956) 49, 51
SODHI,K.S. (1953) 23, 26
RAAB, E. & SODHI,K.S. &
LIPSET, S.M. (1959) 53, 55 BERGIUS, R. (1953) 9, 1 1
RAMSEYER,L.L. (1939) 33, 35 STAR,A. ;
REIGROTSKI, E. & WILLIAMS,R.M.Jr.
ANDERSON,N. (1960) 50 & STOUFFER,S.A.(1958) 47, 51
RIEGEL,0. W. (1953) 49, 50 STEVENS,R.B. (1958) 48, 51
ROBINSON,J. B. see MASLOW, W. STOTLAND,E. see KATZ, D.
ROBINSON,H. see MILLS, J. STOUFFER,S.A. see STAR,A.
ROGERS, C.R. (1951) 44, 45 SUCHMAN,E.A.;
ROKEACH,M. (1948) 19, 21 DEAN,J.P. &
(1960) 19, 21 WILLIAMS,R. M.Jr. (1958) 48, 51
ROSE,A.M. (1948) 7, 10, 32, 35, 42, SUMNER,W.G. (1907) 36, 38
45 SUSSMAN,M.D. (1957) 48, 51
(1956) 56, 57
ROSENBERG,M. 8.t
ABELSON,R.P. (1960) 37, 38 TAMAKA,K. see OGAWA, T.
see HOVLAND,C.I. TANNENBAUM, P. (1956) 28, 35
SAENGER,G. (1953) 7, 11 TAUSCH,A. M. (1961) 43, 45
SALTER,P.J. see BERELSON,B. TAYLOR,J. A. (1956) 30, 35
SAMELSON,B. (1945) 42, 45 TENEN, C. see JAMES,H.E.0.
SANFORD,R.N. see ADORN0.T.W. THURSTONE,L.L. see PETERS0N.R.C.
SARNOFF,I & TITUS,H.E. /k
KATZ, D. (1954) 17, 18, 21 HOLLANDER,E.P. (1957) 15, 21
see also KATZ,D. TORREY,R. see JONES,E.
SCHLORFF , P.W. (1930) 41, 45 TRIANDIS,H.C. &
SCHNEIDER , E. (1958) 19, 21 TRIANDIS, L.M. (1960) 9, 1 1
SCHONBACH, P. (1961) 53, 55, 57 TRIANDIS, L.M. see TRIANDIS,H.C.
SELLITZ,C. TUMIN,M. (1957) 48, 51
WORMSaR, M.H. (1949) 49, 50 (1958) 10, 1 1 , 32, 35
see also WORMSER.M.H. (1960) 48, 51
SHEATSLEY,P.B. see HYMAN, H.H. TUMIN,M. ;
SHEFFIELD,F.D. see HOVLAND,C.I. BARTON,P. &
SHERIF. C.W. see SHERIF. M. BURRUS. B. (1958) 10, 11. 33, 35, 42,
1953-54-56 45, 50, 51, 52, 55,
SHERIF,M. (1936) 22, 26 56, 57
(1937) 22, 26
SHERIF,NI. ;
HARVEY,0.J. ; VAN ORMER, E.B. see HOBAN, C.F.
WHITE, B.J.; VICKERY,W. &Z
HOOD, W.R. COLE,S.G. (1944) 43, 45
SHERIF,C.W. (1 954) 25, 26, 30, 35, 47, VOLKART. E.H. see KELLEY,H.H.
50
SHERIF,M. &
SHERIF,C.W. (1953) 26 WAGMAN, M. (1955) 16, 21
(1956) 23, 26, 27, 35, 57 WALKLEY, R.T. see WILNER,D.M.
SHERIFFS,A.C. see JARRETT ,R. F. WATSON,G. (1947) 7, 11
SHUEY,A.M. (1 953) 29, 35 (1956) 43, 45
SIEGEL,A.E. i% WATSON,J. (1950) 24, 26
SIEGEL,S. (1957) 23, 26 WATSON,J.
SIEGEL,S. see SIEGEL,A.E. LIPPITT,R. (1955) 49, 51
SIMPSON,G.E. 2k WEISS. W. see HOVLAND,C.I.
YINGER,J.M. (1958) 7, 1 1 , 43, 44, 45 WHITE,B.J. see SHERIF,M.
49, 50 WIEDER,G.S. (1951) 44, 45
SIMS,B.M. & WIESE,M.J. &
PATRICK,J.R. (1936) 23, 26, 41, 45 COLE,S.G. (1946) 33, 35
SINGER,R. see FESHBACH,S. WILKE,W.H. (1934) 32, 35
SINHA,D. (1952) 23, 26 WILKINS. C. see KUTNER, B.
SMITH,B.L.& WILLIAMS,R.M.Jr.(1947) 7, 1 1 , 27, 35, 42,
SMITH,C.M. (1956) 50 45

62
 (1958) 42, 45 YARROW,P.R. see KUTNER,B.
see also STAR,A. YINGER,J. M. see SIMPS0N.G.E.
SUCHMAN,E.A. YOUNG,D. (1927) 41, 45
WILNER,D.M. ; YOUNG,R.K.;
WALKLEY, R.T. 8~ BENSON. W.M. &
COOK,S.W. (1952) 48, 51 HOLTZMAN, W . H . (1960) 42, 45
WILSON,E.C. 8~
BONILLA,F. (1955) 51
WOODRUFF, C.L. see KELLEY,H.H. ZAJONC,R.B. 8~
WORMSER, M.H. & MORRISSETTE,J. (1960) 32, 35
SELLITZ, C. (1951) 49, 51 ZANDER,A. see CARTWRIGHT,D.
see also SELLITZ,C. ZELL,H. see JONES,E.

63
 T e should like to draw the attention of our readers
to the Unesco Coupon Scheme, unesco
which m a y enable them to purchase materials
listed in this periodical.
Because i t is often difficult to send money
from country to country for the purchase of books,
coupons
films, laboratory equipment, etc.,
Unesco has created a sort of international currency,
the UNESCO COUPON.
These coupons enable schools,
universities, teachers and students
in m a n y parts of the world
to buy freely the materials 1 UNESCO COUPONS
which they need for study and research. can be used to buy :

UNESCO COUPONS are on sale books, periodicals,

in most Member States photocopies,
which have currency restrictions. microfilm copies,
art reproductions,
If you do not know where to obtain coupons, charts, globes, maps,
please consult sheet music,
your National Commission for Unesco gramophone records,
or write to Unesco, Paris, the following categories of
for the information. educational films :
(a) screening prints and
duplicate prints,
T h e T R A V E L COUPON S C H E M E , (b) original negatives and
which is designed duplicate negatives ;
as well as
to overcome the currency barriers
(c) 1Gmm. raw film for printing
to educational and cultural travel, the above categories of films;
is an extension of the Unesco Coupon Scheme various categories of scientific
described above. material for education and
Travel Coupons are a form research, such as :
of international travellers' cheque. optical instruments,
balances and weights,
They provide the foreign exchange
laboratory glass-ware,
needed to enable travellers such as students, electrical measuring instruments,
teachers and research workers analytical and clinical testing
to spend time abroad for study or research. apparatus, etc.

For lull details, which m a y be obtained

please rejer io ihe jollowing explanatory leajlets : upon request / m m the:
UNESCO COUPONS

wiih separate addenda on BOOK COUPONS, FILM COUPONS

COUPONS FOR SCIENTIFIC MATERIAL Unesco
Coupon Office
INTRODUCING THE UNESCO TRAVEL COUPON Place de Fontenoy
(containing ihe addresses oj the naiional agencies responsible
jor the allocation and issue oj coupons. and the banks where
they m a y be cashed)
Paris
France
-
 UNESCO PUBLICATIONS:NATIONAL DISTRIBUTORS

AFGHANISTAN: Panuzai. Press Department, Royal G U A T E M A L A : Comisih Nacional de la Unesco. N E W Z E A L A N D : Government Printing Ofit :,
Afghan Ministry of Education, KABUL. 6.. Calk 9.27,zona 1. GUATEMALA. 20 Molesworth Streci (Private Bag), WELLINGTO
ALBANIA: N. Sh. Botimeve Naim Frashen. TUUNA. HAITI:Lihrairie'A IaCaravelle',36. ru Roux. B.P.Ill, Government Bookshops: AUCKLAND(P.0. Box 534-
ALGERIA : Institut pedagogique national, II. rue PORT-AV-hJNCE. CHRISTCHURCH (P.0. B O X 1721); DUNEDIN (p.
B a t c h a , ALGER. H O N D U R A S : Librerla Cultura, apartado postal 568. Box 1104).
ARGENTINA: Editorial Sudamericana S.A.. Alsina TEGUCIGALPA D.C. N K A R A G U A : Libreria Cultural Nicaragiiense. cal
500. BUENOSAmES. HONG K O N G : Swindon Book Co., 64 Nathan Road, IS de depliembre y avenida Bolivar. apartadc
AUSTRALIA: Tradco Agencies, I09 Swanston St.. KOWLOON. n.O 807. MANAGUA.

G.P.0. BOX 2324 V, MELBOURNE C.1 (Victoria); H U N G A R Y : Kultura, P.O.Box 149,BUDAPEST62. NIGERIA: C M S (Nigeria)Bookshops. P.O. Box 174,
United Nations Association of Australia, Victorian ICELAND: Snaebjijrn Jonsson 4 Co.. H.F.. H a h s r - LAGOS.
Division, 8th Floor, McEwan Hense, 343 Little straeti 9, REYKJAVIK. N O R W A Y : A.S. Bokhlornet, Lille Grensen 7. OSLO
Collins SI., MELBOURNE C.1 (victoria). INDIA: Orienl Longmans Ltd.: 17 Chiltaranjan Ave. For 'Thr Courier': A.S. Narvesens Litteralurjeoeste.
AUSTRIA: Verlag Georg Fromme 4 Co.. Spenger- CALCUTTA13; Nicol Road. Ballard Estate. BOMBAYI; Stortingsgl. 2. Poslboks 115. OSLO.
gasse 39.WIEN 5. Gunfoundry Road,H Y D E R A B A D I ; 36A Mount Road, PAKISTAN: T h e Wesl-Pak Publishing Co. Ltd..
BELGIUM: Editions "Labor", 342 rue Royale. MADRAS 2; Kanson House, 1/24 Asaf Ali Road, Unesco Publications House, P.O. Box 374, 5 6 N
BnuxELLes 3; N.V. Standaard Boekhandel. Belgielei N E W DELHI I. Guiherg Industrial Colony. LAHORE.
15 I. ANTWERPEN. Sub-depots: Indian National Commission for CO- P A R A G U A Y : Agencia de Lihrerias de Salvador
For 'The Courier' onrl slidcs: Louis de Lannoy. (( Lc operation with Unesco, Ministry of Education. NEW Nizza. Yegros. entre 25 de M a y o y Mcnl. Estigarrihia.
Courier de 1'Unesco n. 112,rue du Trhne,BnUxmLES. DELHI 3; Oxford Book and Stationery Co.,17 Park ASUNCI~N: Albo Induslrial Comercial S.A..Seccidr
BOLIVIA: Librerla Universitaria. Universidad San Street, CALCUTTA 16, and Scindia House, NKW Libreria, Gral. Dim 327.ASUNCIbN.
Francisco Xavier, apartado 212. SucnE; Libreria DELIII. PERU: Dislrihuidorn LNCA S.A.. Emilio Altahu?
Banet. Loayza 118, casdla 1057. LA PAZ. INDONESIA: P. N. Fadjar Bbakti, Djalan Nusantara 460 Lince. LIMA.
BRAZIL: Fundaqao Getlilio Vargas. 186 Praia de 22. DJAKARTA. PHILIPPINES: The M o d e m Book Co., 508 Riza
Bolafogo.RIO DE JANEIRO, GB ZC-02. IRAN: Commission nationale iranienne pour I'Unesco. Avenue. M A N I I A .
BULGARIA: Raznolznos. 1 Tzar Assen,SOPIA. avenue d u Musee, Tl?HhAN. POLAND: Oirodek. Rozpowszcchniania Wydawnictv
B U R M A : Burma Translation Society. 361 Prome IRAQ: McKenzie's Bookshop. AI-Rnshid Street. Naukowysh PAN,Palac Kultury i Nauki,WARszAWA.
Road, RANGOON. BAGHDAD. PORTUGAL: Dim Rr Andra de Lda., Livraria Portugal.
CAMBODIA: Lihrairie Albert Portail. 14. avenue IRELAND: The National Press, 2 Wellington Road, riid do Carnio 70. LISWA.
Boulloche, PHNOM-PENH. Ballsbridge. DUBLIN. PUERTO RICO: Spanish English Pubhcarions. tlea-
C A N A D A : The Queen's Printer. OTTAWA(Ont.). ISRAEL: Blumstein's Bookstores, 35 Allenby Road nor Roose\e!t115. apartodo 1912, HATO REY.
CEYLON: Lake House Bookshop. P.O. Box 244, and 48 Nahlat Benjamin Street,TELAvlv. RUiMANIA: Carrlmex, Sv. Aristide Briand 14-IP
Lady Lochore Building.100 Parsons Road,COLOMBO 2. ITALY: Libreria Commissionaria Sansoni S.p.A. P 0.Box 134-135.BuctlnEsll.
CHILE: Editorial Unwersiiaria S A.. avenida via Lamarrnora 45, casella postale 552, FIRENELE; SENtGAL: La Maisoo d u livre. 13. avenueKnunw
B. O ' H i g g h 1058, casilla 10220. SANrIAGO. Libreria Internazionale Rizzoli, Largo Chighi. DAKAR
For 'The Courler': Comisidn Nacional de la Unesco ROMA; Libreria Zanichelli, Portici del Pava- SINGAPORE: Sec Malaysia.
en Chile, alameda B. O'Higgins 1611. 3.n' piro. glionc.BOLOGNA;Hoepli,via UlricoHoepli 5, MILANO; SOUTH AFRICA: Van Schaik's Bookstore (Pty) Ltd.
SANTIAGO. Librairie frangaise,piazza Caslello 9,TORINO. Libri Building. Church Srreet. P.O. Box 72a
CHINA: Tbe World Book Co. Ltd.. 99 Chungking JAMAICA:Snngster's Book Room, 91 Harbour Street. PRKIORIA.
South Road. Section I, TAlPeH (Taiwan/Formosa). KINGSTON. S O U T H E R S RHODESIA: The Book Centre.Gordoi
COLOMBIA: Libreria Buchholz Galeria, avenida Jim& JAPAN: Maruzen Co. Ltd.. 6 Tori-Nichome. Nihon- Avenur. SAtlSBURY
nez de Quesada 8-40. BOGOTA; Ediciones Tercer bashi. P.O. Box 605, Tokyo Central. TOKYO. SPAIN: Libreria Cientifica Medinaceli, D u q u c d *
Mundo. aparlado abreo 4817. BOWTA; Comitb JORDAN: Joseph I. Bahous & Co.. Dar-ul-Kutub. Medinaceli 4. F ~ A D R I14 D.
Regional de la Unesco. Universidad Industrial de Salt Road, P.O.Box 66,AMMAN. For 'The Courirr': Ediciones Iherqamericanas S.A
Santander. BUCARAMANGA; Distrilibros Lld.. Pi0 K E N Y A : ESA Bookshop, P.O. Box 30167. NAIROBI. calle de Ofiaie 15, M A D R ~ D .
Alfonso Garcia,calleDonSanchon.Os 36-119 y 36-125. K O R E A : Korean National Commission for Unaco. S U D A N : A: Dashir Hoo4sliop. P.O. Box 1111.
C A n T A G E N A ; J. German Rodriguez N.. oficina 201. P.O.Box Central 64,SEOUL. KHARTOIIM.
Edificio Banco de Bogota, apartado nacional 83. L E B A N O N : Lihrairie Dar AI-Maref Liban. S.A.r.L.. S W E D E N : A I B C. E. Fritzcs Kunel. Hovbokbandel
GmAnDoT; Escuela Inreramericana de Bihlioteco- immeuble Esseily, 3 e etage, place Riad El-Solh, Fredsgatan 2.SrocKuoLM 16.
login, Universidad de Antioquia. MEDELLIN; Libreria B.P. 2320. BEYROUTH. For 'The C-ourtu : Svenska Uneacortidet, Vasagatan
Universilaria, Universidad Pedagbgica de Colombia, LIBERIA:Cole 4 Yancv BookshoDs Ltd..P.O.Box 286 15-17.STOCKHOI ci C.
TUNJA. MONROVIA. SWITZERLAND: turopa Verlag. RHmistrasse 5.
C O N G O : La Lihrairie. Institut politique cangolais. LIBYA: Orient Bookshop, P.O. Box 255. TntmLI. ZURICH; Payot. 40. rue d u Marchb. GENbVe.
B.P. 2307, L~OPOLDVILLE. LIECHTENSTEIN: Eurocan Trust Reg.. P.O.B. 124. SYRIA: Librairie iniernationale Avicenne. boile pos-
COSTA RICA: Trejos Hennanos. S.A.,apartado 1313. SCHAAN. lale 2356, DAMAS.
SAN Josl. L U X E M B O U R G : Lihrairie Pau IBruck 22 Grand-Rue, T A N G A N Y I K A : Dar es Salaam Bookshop. P.0
For 'TheCourier': Carlos Valerin Sknz &Co. Ltda.. LUXEMBOURG. BOX 9030. DAn ES S A L A A M .
'El Paiacio de las Revistas'. aparlado 1294, SAN Josh. M A D A G A S C A R : Commission nationaie de la R6pu- THAILAND: Suksapan Panit. Mansion 9, Rajdsmnen.
C U B A : Cubartimpex.apanado poslal 6540.L A HABANA. blic malgache. Minislkre de I'education nationale. Avenue BANGKOK.
CYPRUS: Cyprus National Youth Council. P.O. TANANARIVE.For The Courier ': Service des (Euvres T U R K E Y : Librnirie Hachette. 469 Istiklal Caddes
Box 539. NICOSIA. post el piri-scnlaires, Ministtc de I'educalion natio- Beyoglu. ISTANBUL.
CZECHOSLOVAKIA: SNTL. Spalena 51. PRAHA 1 d e , TANANARIVB. A : Uganda Bookshop,P.O.Box 145,KAMPAL.,
(Permanentdisplay) Zahranicni lilerarura. Bilkova 4. MALAYSIA: Federal Publications Ltd., Times Meihdunarodnaja Kniga. MOSKVA G-200.
PRAHA 1. House. Rivcr Vallcy Road, SINGAPORE. UNITED A R A B REPUBLIC: Librairie Kasr El Nil.
D E N M A R K : Ejnar Munksgaard Ltd.. Prags Boule- M A L T A : Sapienza's Library. 26 Kingsway. VALLE~TA. 38, rue Karr El luil, CAIRO.
vard 47, KOBENHAVN S. Sub-drpdt: La Renaissance d'Egypte, 9 Sh.Adly Pashz.
DOMINICAN REPUBLIC: Libreria Dominicaoa.Mer- MAURITIUS: Nalanda Co. Ltd..30 Bourbon Slreet. CAIRO(Egypt).
cedes 49. apartado de correos 656,SANTO DOMIPIGO. PORT-LOUIS. UNITED K I N G D O M : H.M. Stationery Office. P.O.
E C U A D O R : Casa de la Cultura Ecuatoriana, Nljcleo MEXICO: Editorial Hermer. lgnacin Marisca 41, Box 569 London, S.E.I. Government bookshops:
del Guayas. Pedro Moncayo y 9 de Octubre. casilla MEXICO D.F. London,'Belfast.Birmingham, Cardlff, EdlnhurgL
de correo 3542. GUAYAQUIL. Manchester.
EL SALVADOR: Libreria Cultural Salvadorefia. M O N A C O : British Library,30. boulevard des Moulins.
MONTE-CARLO.
UNITED STATES O F AMERICA: Unesco Publi-
SAN SALVADOR cations Cenler (NAIP), 317 East 34th St., New-
ETHIOPIA: Interaalional Press Agency. P.O. Box 120. M O R O C C O : Libmirie A u x belles imnpes ", 281, ave- Yonx.N.Y.10016onderc~plJorperiodicrrls: Columbk
h D l S ABABA. nut hlohunmed V. RABAr (CCP 47.69.) University Press,2960 Droadway,N E W YonK 27,N.Y
FINLAND: Akateeminen Kirjakauppa. 2 Keskuskatu For '71w Courier' (fur teoc+ers) :Cornmission natio- U R U G U A Y : Representacion de Edilonales, plam
HELSINKI. nale rnitrociline pour I'Unesco.20,Zenkat Mourabitine. Cagancha 1342. I.er piso. MONTEVIDEO.
FRANCE: Librairie de I'Unesco. place do Fontenoy. R A U A T(CCP 307.61). VENEZUELA: Librerla Poiilecnica.calle Villaflor,locai
PARlS-7e.CCP 12598-48. M O Z A M B I Q U E : Snlema and Carvalho Ltda., caixa A, PI lado General Electric,Saband Grande.C A R A C A S ;
F R E N C H W E S T INDIES: Librairie J. Bocage. rue poslal 192. BEIRA. Libreria Crun del Sur. Centro Comercial del Estr
Lavoir, B.P. 208. FORT-DE-FRANCE(Martinique). NETHERLANDS: N.V. Marlinus Nijhoff, Langc local 1 I aparlado 10223, Sahona Grande, C A R A C A S
G E R M A N Y (FED. REP.): R. Oldenhourg Verlag. Voorhout 9, ' S - G R A V E N H A G C . Oficina 'Publicaciones de la Unesco. Gobernador z
Unesco-Vertrieh liir Deutschland. Rosenheimer Candilitc n." 37, aparlsdo poslal n.O 8092. C A R A C A S ;
StraSSe 145. M W C H E N 8. NETHE.RLANDS ANTILLES: G.C.T. Van Dorp Libreria Selecta. avenida 3. n 23-23.MERIDA.
G H A N A : Methodist Book Depot Lld.. Atlantis House, and Co. (Ned. Ant.) N.V.. WILI EMSTAD (Curncao. VIET-NAFI: Librairle-papklerlr XuPn-Thu, 195-193
Commercial Street. P.O.Box 100,CAPE COAST. N.A.). rue Tu-Do,B.P. 283. SAIGON.
GREECE: Lihrairie H.Kauffmann.28. rue du Stade. N E W CALEDONIA: Reprex, avenue de la Vicloire, YUGOSLAVIA: Juposlovenska Knjiga, Teraziic 27
ATHkNES. Irnmeuble Painbouc,NOIJMPA. BEOCRAn