Chapter Title: Group Formation

Book Title: The Resolution of Conflict

Book Subtitle: Constructive and Destructive Processes
Book Author(s): Morton Deutsch
Published by: Yale University Press

Stable URL: https://www.jstor.org/stable/j.ctt1dszxst.7

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Yale University Press is collaborating with JSTOR to digitize, preserve and extend access to The
Resolution of Conflict

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
4. Group Formation

In the preceding chapter, we looked at the individual from the inside.

From this perspective, he is seen to be composed of interdependent
subunits that interact with one another in a manner resembling that of
members in a tightly knit group. 1 Just as from the outside the individ-
ual can be considered a unit and from the inside he can be viewed as
a system of interdependent subunits, so, too, is the group a unit from
the exterior and a system of cooperatively interdependent members
from the interior. In this chapter, we shall be concerned primarily
with the formation of groups or, in other words, with the conditions
that determine whether or not cooperation will be initiated among the
potential members of a group. First we shall treat the question: What is
a group? Then we shall discuss group formation; finally, the relation-
.ship between group membership and personal identity will be con-
sidered.

WHAT Is A GROUP?

An examination of the different usages of the term group suggests that

each combines a greater or lesser number of the following distinguish-
ing criteria: a group consists of two or more persons who ( 1) have one

1. To be sure, the individual as a "group" has many special features. I mention

only a few: its members cannot survive outside the group and, thus, cannot join
other groups; the pressures toward internal consistency are relatively strong; its
members are specialized with the consequences that there are highly developed cen-
tralized subunits for coordinating and integrating the activities of the specialized
components; contact with the outside (in perception or in action) is not directly
available to all members but is highly specialized and is controlled by the central-
ized subunits; the organizational structure, the systems of internal communication,
and the relative power of different subsystems is largely determined by biological in-
heritance.

48

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 49

or more characteristics in common; (2) perceive themselves as

forming a distinguishable entity; (3) are aware of the positive interde-
pendence of some of their goals or interests; (4) interact with one an-
other; and (5) pursue their promotively interdependent goals together.
In addition, writers concerned with persisting social units indicate that
groups endure over a period of time and as a result develop (6) a set
of norms that regulate and guide member interaction; and (7) a set of
roles, each of which has specific activities, obligations, and rights as-
sociated with it.
The term group is commonly used to signify at least the first three
of the criteria listed above. Several different degrees of "groupness"
are often distinguished: "quasi-groups," which have only the first three
characteristics listed; "functioning groups," which have the first five
attributes; and "organized groups," which have all the criteria listed.
This multifaceted usage of the term group is consonant with the intui-
tive notion that minimally a group is an entity consisting of people
who are aware of being psychologically bound together by mutually
linked interests. Thus a group is to be distinguished from an aggregate,
class, category, or type, which consists of people who are classified to-
gether because of some common characteristic. "Group" implies a
psychological or perceived bond, not merely an objective linkage, be-
tween the members' interests or goals. Moreover, the psychological
linkage has some cohesive feature to it-i.e., members of a group see
that in some respects they sink or swim together. This latter statement
is not meant to deny that divisive and disruptive tendencies may exist
within a group; rather, it is meant to indicate that, by definition, a
group does not exist if its cohesive bonds are not strong enough to
contain its disunifying influences.
Our discussion of the concept group implies that not all aggregates
of people who have something in common or whose interests are posi-
tively interrelated will act together in pursuit of their common inter-
ests. We shall consider below some of the conditions that affect
whether or not a category of people will become an interacting orga-
nized group. Surely, it makes a difference whether a collection of peo-
ple are a group or not; in addition, it makes a difference what charac-
teristics they have as a group. How cohesive are they? How many
members are there? What kind of structure or organization do they
have? As members of an organized group, individuals are part of a
cooperative system, and their behavior toward others in the group

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
50 Theoretical Ess.-:1ys

and toward those not part of the system is affected by their member-
ship in that system and by the particular characteristics of the system.
In chapter 2, some of the features of cooperative relations were de-
lineated; individuals who are part of an organized group are more
likely to manifest such features than individuals who do not form one.
Thus members of such a group are more prone to attend to and un-
derstand one another; communication among them is more effective
and more frequent; they are more ready to be influenced by one an-
other, to respond helpfully to others, and to facilitate each other's ac-
tions; their attitudes toward one another are more often friendly and
trusting; they are more apt to develop similar beliefs, values, and cus-
toms and to expect other members to conform to them; they are more
inclined to organize themselves so as to pool their specialized skills
and responses in a coordinated effort to achieve their interrelated
objectives; when conflicts occur, they are more prone to view them as
mutual problems to be solved by collaborative effort and are less
likely to use coercive techniques to influence one another or to ex-
pand the scope of the issues in conflict.
To illustrate the difference between an organized group and an ag-
gregate of people with similar interests, consider tenants in a New
York apartment house who are, as individuals, discontent with many
aspects of their housing. If they recognize that the discontent is shared
and if they interact with the objective of changing conditions in their
housing, the tenants will in the process transform themselves into a
functioning group. If they cooperate effectively, their relations to one
another will change. Instead of remaining strangers who interact
rarely and cautiously, they will develop many of the "symptoms" of
cooperation outlined in the preceding paragraph.

GRouP FoRMATION

In light of the great variety of groups, it is unlikely that there is a sin-

gle answer to the question of how a collection of individuals comes to
be a cooperative system. As Cartwright and Zander (1968, p. 54)
point out, a group may be established deliberately by one or more
people to accomplish some objective, it may be formed spontaneously
by those who become members, or it may be created because other
people treat a collection of individuals as a group. Despite the many
different ways in which a group can originate, it seems possible to

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 51

identify several minimal conditions that must be met before a collec-

tion of individuals will turn into a cooperative system.
First, the individuals must be aware of one another's existence. Sec-
ond, they must have some motives (interests, goals, or values) whose
fulfillment they prefer to seek through cooperative interaction with
the others-i.e., they must choose to cooperate with the others.
Third, cooperative interaction must be initiated and reciprocated in
such a way that the interaction serves to confirm for each of the po-
tential group members his expectation of a mutually desired coopera-
tive relationship. Finally, the individuals must become aware of them-
selves as a distinguishable entity. Factors affecting the realization of
each of these basic conditions are discussed below.

Becoming Known to One Another

There has been little systematic study of the many factors, direct or
indirect, that influence the likelihood that any collection of individu-
als will come to know of each other's existence. Propinquity has been
the factor most thoroughly investigated, and the overwhelming evi-
dence suggests that propinquity plays a very important role in estab-
lishing social relationships (Lundberg and Beasley 1948; Hollingshead
1949; Festinger, Schachter, and Back 1950; Deutsch and Collins
1951). Other ecological factors, such as the availability of communi-
cation and transportation facilities, have an influence similar to that
of propinquity (Zipf 1949). Propinquity and other physical factors in-
fluencing ease of contact may themselves be determined indirectly by
demographic variables-i.e., by the tendency of people of similar
economic, religious, racial, or occupational backgrounds to live,
work, and/or play in close proximity to one another (Form 1951).
The existence of various institutionalized settings such as schools, fac-
tories, churches, hospitals, shopping centers, recreational centers,
bars, and meeting halls will also affect the likelihood that people will
get to know one another. In addition, of course, such personal factors
as age, health, sex, social position, and personality will influence the
individual's ability and willingness to get to know others and to be-
come known. The aged, the disabled, and the poor often are limited
in their physical mobility, and thus if they are widely scattered, it will
be rather difficult for them to have sufficient contact with kindred peo-
ple to develop a basis for group formation and effective group action.

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
52 Theoretical Esst1ys

Choosing to Cooperate
The choice to cooperate or not is in many respects sin,ilar to any
other choice. Hence, such general theories of choice as level-of-aspira-
tion theory (Lewin et al. 1944) or utility theory (Edwards 1954) are
applicable. A basic assumption in such theories is that an individual
will choose, from a set of perceived alternatives, the positively valued
alternative that has the highest "effective attractiveness." As applied
to the choice of cooperating or not, this statement means that an indi-
vidual will want to cooperate when he sees that this is the best or only
way to achieve a goal (or goals) that he is motivated to attain.
It is obvious that many goals are social in nature and intrinsically
require the collaboration of others for their attainment-e.g., to be-
long to a given group, to achieve racial and sexual equality, to have
an intimate relationship, to play a game of tennis, to participate in an
interesting conversation, to sell or buy anything, to obtain someone's
approval or support. Other goals, though not intrinsically social, may
nevertheless be impossible to attain without collaboration of others--
e.g., to have a lobster dinner in Chicago, to travel from the United
States to Europe, to live in a better house than one could construct
oneself. Of course, there are many types of nonsocial goals that can
be attained through either individual or collaborative effort~.g .• to
build a canoe, to solve a problem, to repaint the living room. There
are, moreover, certain types of goals relating to self-expression or
self-activity from which it would be intrinsically impossible to gain
personal satisfaction through the activities of others alone-e.g .• to
make a painting or to eat a lobster. For other types of goals, the activ-
ities of others may be entirely self-sufficient-e.g., to see that a blind
man crosses a street safely, to have a nursery school organized so that
one's children may have an opportunity to engage in supervised activ-
ities with other children.
Whether or not one's goals require or permit cooperation for their
attainment will obviously be an important determinant of the choice
to cooperate or not. If one's goals permit but do not require coopera-
tion, the choice to cooperate will be determined by the effective at-
tractiveness of cooperation as compared to other perceived alterna-
tives. The effective attraction of an alternative is a function of two
sets of factors: those that influence its desirability or undesirability

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formtltion 53

and those that influence the subjective probability that it will lead to
the desired result.
The desirability of cooperation is determined by the gains, plea-
sures, and values one hopes to realize through it; its undesirability by
the costs, pains, and disvalues one expects. Clearly, a choice to coop-
erate is unlikely if it is anticipated that the "negatives" will outweigh
the "positives"-unless, of course, cooperation is chosen as the lesser
of evils. In addition to the nature of one's objectives, several other
major types of determinants influence the. desirability of cooperation.
Personality dispositions affect both one's general inclination to be co-
operative and one's preferences regarding with whom one would like
to cooperate (see Terhune 1970 a, b for a review). It is also evident,
as Mead (1937) has demonstrated in her classical study of Coopera-
tion and Competition Among Primitive Peoples, that societal values
may be more or less oriented toward cooperation. In addition to the
more enduring general influences of personality dispositions and cul-
tural values, the desirability of cooperation is influenced by a host of
such specific external factors as the social and personality characteris-
tics of other group members, the reputation of the group, and one's
anticipated activities and status within the group.
No matter how desirable the various aspects of cooperation may
seem, cooperation is not likely to be undertaken unless there is some
hope that what one wants from it will be attained. This is especially
the case when unsuccessful cooperation is seen to be costly. There are
many determinants of an individual's estimate of his chances of suc-
cess in a given situation. These include his prior experiences in similar
situations, his perceptions of the judgments of others, and his person-
ality. Thus if he has had a prior history of success or if his friends as-
sure him that he will be successful, and if he is inclined to be
optimistic, he is likely to be quite confident. There are, of course, con-
siderable differences among individuals in terms of how much confi-
dence of success each requires before he will initiate cooperation.
Nevertheless, if one cannot trust that the other potential group mem-
bers will have the motivation and ability to cooperate effectively, one
is not likely to enter a cooperative process. Because of the centrality
of trust to the initiation of cooperation, a separate chapter (chapter 7)
will be concerned with this topic. Trust is, however, only one of the
three key problems of mutuality that must be resolved if cooperative
interaction is to be successfully initiated and reciprocated. The others

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
54 Theoretical usays

are coordination and bargaining. Each of these problems is discussed

below in terms of the question it poses for the potential cooperators.
Problems of Mutuality in the Initiation of Cooperation
No individual, by his choice alone, can successfully initiate coopera-
tion. Cooperation is a mutual endeavor, and for cooperative interac-
tion to occur, the choice to cooperate must be reciprocated. Contrast
this with a choice involving inanimate objects. If I choose to have yo-
gurt rather than pastry for dessert, the yogurt does not have to choose
me for my choice to be successfully consummated. Arising out of the
requirement that cooperation be mutual and reciprocated are the
three problems mentioned above: trust, coordination, and bargaining.
Trust. The problem of trust raises the question of whether or not
the potential cooperators can trust one another sufficiently to take the
risks involved in initiating cooperation. Suppose, for example, that I
wish to sell an antique clock and you wish to buy one. There is a pos-
sibility that what I am offering you is not a genuine antique but only
an imitation; analogously, you may attempt to pay me with a bogus
credit card or forged check. For me to sell the clock to you, I must
trust that your credit card or check is valid, and for you to buy it, you
must trust that I have not misrepresented the clock.
Figure 4.1 illustrates the problem of trust in an abstract form. P 1
has to choose between C and D and has to announce his choice be-
fore he knows what P 2 has chosen. The situation is the same for P2 •
Clearly, unless he can trust that the other will choose C rather than
D, each is forced to choose D. If both choose D then both will lose,
despite the fact that both can gain by cooperating and choosing C.
Each may, however, feel that the other might be tempted to gain

c D
c +I,+ I -2, +2

D +2, -2 -1, -1
Figure 4.1. The trust problem. This and the following matrices should be read as fol-
lows: P1 chooses between rows C and D, P2 between columns C and D. Pt 's payoffs are
the first numbers in the cell; P2 's are the second numbers. For illustrative purposes,
consider that the payoffs are in money and that each person would prefer to gain rather
than lose, as well as gain + 2 rather than + 1, or lose - 1 rather than - 2.

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation ss
more by choosing D. Or, even if the other can be trusted, it may be
difficult to resist the temptation oneself.
Matrices; like the one illustrated in figure 4.1, have been employed
in many studies to investigate the conditions affecting "trust" and
"trustworthiness." (See chapter 8 for reports of such research done by
my associates and myself.) It is well to recognize that a relationship of
mutual trust is not always based upon knowledge of the other's per-
sonal trustworthiness. Trust may be rooted in factors that are external
to the particular relationship, such as confidence that existing social
institutions will discourage untrustworthy behavior, confidence in
third parties who can vouch for the other, or a general confidence in
the trustworthiness of people who are viewed as similar to oneself.
Particularistic trust, which is based upon personal knowledge of the
other, is characteristic of small, traditionalistic communities. Univer-
salistic trust, which depends upon the social rather than the personal
characteristics of the other, seems to be a necessary ingredient in the
development and functioning of complex modern societies.
Generalized trust as a norm among individuals in a given society
appears to be essential for the operation of large-scale organizations,
which must rely heavily on interpersonal relationships among strang-
ers. In the modern, organizational society, one must trust that an un-
known mailman will bring your letters conscientiously rather than
throw them away, that an unknown worker in a pharmacy will fill
your prescription correctly, that a bank clerk will not make incorrect
entries into your account, that your physician will have obtained his
degree through competence rather than through political influence,
and so on. When generalized trust is impaired, one can expect a corre-
sponding increase in suspiciousness, irresponsibility, and untrustwor-
thiness. Public opinion data reported by the Survey Research Center
of the University of Michigan in November 1971 indicate that from
1966 to 1971, paralleling the increased involvement of the United
States in the Vietnam war, there was a sharp decline, particularly
among the youth, in the generalized trust toward the government and
other major social institutions. Symptomatic of this decline were more
social disorder and disruption, greater apprehensiveness, and an in-
creased sense of lack of participation in a meaningful community.
The level of generalized trust existing in a community determines
whether or not the different individual resources existing within it can
be shared or combined for mutual benefit. The existence of trust is it-

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
56 Theoretical Essays

self a resource of cooperation, and without it the benefits that can be

derived from organized cooperative activity are drastically curtailed.
The development of revolving credit and mutual loan associations
within a community are likely to be inhibited with a low level of trust.
Without such institutions, however, economic activity and growth will
be cumbersome and, as a consequence, deterred. Similarly, the devel-
opment and effective functioning of all community institutions and
organizations-schools, cultural centers, and community organ-
izations-will be strangled by excessive red tape, meetings, and dis-
ruptions if there is little trust. A university, for example, cannot
function effectively if faculty, students, and administrators are all sus-
picious of one another. If each has to monitor the other constantly be-
cause of insufficient trust, none will have time to perform its own
functions adequately, and its failure to do so will confirm the other's
suspicions.
There is a variety of evidence to suggest that levels of trust are
usually higher in communities of higher socioeconomic status than in
those of lower, and higher among the advantaged groups than among
the disadvantaged groups within a community. It is not surprising that
blacks and the poor are less trusting than whites and the well-off: they
have less reason to be so. However, insofar as the low level of trust is
not only characteristic of the relations with the outgroup but also of
the relations within the group, it serves to perpetuate and enhance the
weakness of the disadvantaged group. If the level of trust within a dis-
advantaged group is low, the group cannot develop a strong commu-
nity with effective institutions that would enhance its collective power
and enable its members to cooperate effectively to overcome their dis-
advantages. Of course, an exploitative group that seeks to maintain its
advantages will often utilize any tactics necessary to deter the develop-
ment of group cohesion among the exploited.
Bargaining. The question of mutuality posed by the bargaining
problem is whether the potential cooperators can resolve differences
in preferences with regard to the terms of their cooperation. The buyer-
seller transaction provides an illustration of such problems. The
buyer and seller have concordant interests in that they both desire that
a sale take place. However, they also have conflicting interests in that
the seller would prefer a sale at a high price and the buyer at a low
one. If the seller's minimum acceptable sale price is lower than the
buyer's maximum acceptable purchase price, a mutually acceptable

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 57

agreement is possible. Whether or not the buyer and the seller reach an
agreement will be determined by how successfully they bargain to-
gether.
Bargaining is not only characteristic of economic transactions but
is a feature of all social relationships. Husbands and wives who want
to spend an evening out together but have dissimilar preferences on
where to go will bargain to determine the nature of their cooperation;
similarly, parents and children who have dissimilar views about "bed-
time" will often engage in elaborate negotiations in the process of
working out an agreement. Students and faculty who may have differ-
ent preferences regarding the timing of various academic events never-
theless will need to come to an agreement to prevent an academic im-
passe.
Figure 4.2 presents the bargaining problem in abstract form. It is
evident that each side is better off if they can agree.

X y
A 0,0 +2, +4

B +4, +2 0,0
Figure 4.2. The bargaining problem.

P 1 would, however, prefer Bx while P 2 would prefer Ay. What deter-

mines whether or not P 1 and P2 will be able to come to a mutually ac-
ceptable agreement in such situations? What are the effects of different
methods of attempting to influence the other to yield? These and
other related questions are the focus of inquiry in a series of studies
reported in chapters 9, 10, and 11. Because of the pervasiveness of
threats and promises m bargaining, a theoretical discussion of these
modes of social influence is presented in chapter 6.
Coordination. Even if potential cooperators trust one another and
agree completely on how to cooperate, they still confront the problem
of coordination. For example, suppose a man and woman meet
briefly, by chance, at a cocktail party in New York. They immedi-
ately are taken with each other and would very much like to meet
again. Unfortunately, they forget to exchange last names, addresses,
and phone numbers before leaving the party. They want to get to-

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
58 Theoretical Essays

gether, but how, when, and where can they meet? They face the prob-
lem of coordination. Suppose each of them decides to meet the other
by going to some hangout for "singles" for the next several nights.
Where should each go and at what time to maximize their chance of
meeting?
An abstract version of the couple's problem is presented in figure
4.3. In this version, P 1 and P2 both want to converge on the same
choice, but they do not know what the other is going to choose. In such
a situation, P 1's best course of action depends on the action he expects
P 2 to take, which depends in turn on P 2's expectation of P 1's action.
Each must try to guess what the other thinks he will guess, and so on.
What permits the convergence of expectations rather than an endless
spiral of "second-guessing"? (See Schelling 1960 for an insightful dis-
cussion.)

p2
A B c
A +4, +4 0,0 0,0
Pt B 0,0 +4, +4 0,0
c 0,0 0,0 +4, +4
Figure 4.3. The coordination problem.

Communication is an obvious way of solving the coordination

problem-e.g., P 1 says to P2 : "You choose column A and I'll choose
row A." However, there are many situations in which communication
is impossible or too costly, and in these cases coordination has to be
accomplished tacitly rather than explicitly. Moreover, even when
communication is feasible, there is often some necessity for selecting
what one communicates from that limited range of possibilities that
are likely to be mutually acceptable. Thus Schelling ( 1960, p. 70) has
pointed out:
Most bargaining situations ultimately involve some range of possi-
bility outcomes within which each party would rather make a con-
cession than fail to reach agreement at all. . . . The final outcome
must be a point from which neither expects the other to retreat; yet
the main ingredient of this expectation is what one thinks the other
expects the first to expect, and so on. . . . These infinitely reflexive

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 59

expectations must somehow converge on a single point, at which

each expects the other not to expect to be expected to retreat.
The problem of coordination encompasses more than the conver-
gence of expectations in situations of tacit communication. Even ex-
plicit communication may be misunderstood; the speaker and the lis-
tener may not interpret what is said in the same ways.
Misunderstandings, of course, are very likely when the speaker and
listener are from rather different cultural backgrounds and are not
fully aware of their differences. The stock conversation of returning
tourists consists of amusing or embarrassing anecdotes based upon
misunderstandings of this sort. Even within a given society, unrecog-
nized cultural differences may give the same act or word a different
significance to blacks and whites, to men and women, to members of
different ethnic groups, to people of different class backgrounds, to
scientists and nonscientists.
Thus, for example, only in recent years has there developed some
awareness that lower-class black children, when they enter school,
often speak a form of English that is different from standard American.
Before this was recognized, many teachers assumed that the speech of
these students was simply an inferior, impoverished version of stan-
dard, middle-class American speech. This faulty view buttressed edu-
cators' low expectations regarding the intellectual potential of black
children and contributed to a vicious cycle in which the black chil-
dren became increasingly alienated from their schools.
Although husbands and wives may speak the same form of English,
it is not uncommon for a couple to misunderstand one another be-
cause their styles of emotional communication differ enormously.
There may be differences in what is attended to: the husband may
consider what is said (the words) to be more important, the wife may
attend more to how it is said (the tone of voice, the circumstances).
There may be differences in amplitude and frequency of emotional
expressiveness, the husband expressing his emotions in low key, infre-
quently, and the wife expressing hers at high intensity, frequently.
Each may judge the other's expressiveness in terms of his or her own
norms, the husband judging his wife as very emotional, the wife view-
ing her husband as unemotional. Lack of recognition, lack of under-
standing, or intolerance of such stylistic differences in communication
can create misunderstandings, resentment, and hurt feelings that may
impair cooperation.

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
60 Theoretical Essays

THE PERCEPTION OF THE GROUP AS A

DISTINGUISHABLE ENTITY

In the previous sections, I have considered how potential cooperators

get to know one another, some of the determinants of the choice to co-
operate, and the problems of mutuality arising from the fact that coop-
eration can not be consummated unless it is reciprocated. This discus-
sion has assumed that a group is a voluntary cooperative system and,
hence, that the development of cooperation and group formation are
clearly related topics. However, a group is more than an assemblage
of cooperating people. It is composed of members who are aware of
themselves as an entity: a group is conscious of itself. In addition, it is
evident that membership in many groups (such as family, racial, and
ethnic groups) is by inheritance and not choice. In such instances, the
development of a positive group consciousness may be the precondi-
tion for the development of cooperation, rather than vice versa. Also,
the development of cooperation and the growth of group conscious-
ness may each foster the other. Thus it is apparent that the onset of
group consciousness may precede, follow, or be concurrent with the
development of cooperative relations within the group.
In addition to the experience of successful cooperation, several fac-
tors seem to play key roles in the development of group conscious-
ness. They include (I) the treatment by others. If others treat mem-
bers of the group in a similar way and identify them as one group for
varous purposes, group consciousness will be promoted; (2) an abrupt
discontinuity between the characteristics of members and nonmem-
bers of the group. Thus if members are rather similar in such matters
as beliefs, values, appearances, proximity, past experience, language,
customs, and style but sharply different in these respects from the re-
mainder of the people in their surroundings, group consciousness will
be heightened; (3) the use of symbols to represent the group. The exis-
tence of a name, insignia, uniforms, differentiated and labeled statuses
within the group, and group ceremonials all intensify group aware-
ness; (4) differences in rates, types, and styles of interaction between
members and nonmembers. The more members interact with one an-
other about topics and in a manner that differs from the way they in-
teract with nonmembers, the more intense will be their group aware-
ness; and (5) a history of cooperating together. The existence of an

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 61

historical time perspective, with a sense of continuity of relatedness,

and an awareness of shared memories of past events contributes to a
heightened consciousness of a group identity. A time perspective that
extends from the past to the future is central to a stable group identity.
The operation of these factors in raising group consciousness can
be illustrated by a hypothetical example. Suppose that in a large lec-
ture class I wished to create two groups: the shorties (those 5'6" and
under) and the longies (those over 5'6" in height). To do this, I would
treat the groups differently: require them to enter and exit the class-
room by different doors, give them different assignments and tests, as-
sign different teaching assistants to work with them, allow them to
visit me in my office only at clearly different hours, and encourage
others to treat them differently. If this were not sufficient, I could as-
sign them to sit in clearly separated areas of the classroom so that the
two groups would be quite distinct visually. In addition, I could refer
to the groups by different names and give them distinctive collective
tasks to perform that would encourage interaction within each group
but interfere with interaction between "longies" and "shorties." I
would do this over a period of time, creating distinctive histories and
memories for the two groups as well as expectations that the future
would be continuous with the present. Such behavior on my part
would create two groups whose members were highly conscious of
their group membership.
Typically, one would expect that the more conscious an individual
is of his membership in a group in a given situation, the more likely is
his behavior to be affected by his group membership. Moreover, as an
individual's group consciousness increases, the probability that his be-
havior will be affected by the group in more situations also increases.
Although the enhancement of group consciousness often raises co-
hesiveness, the member may become more negative toward the
group as its salience increases when his attitude toward it is nega-
tive. Such a situation may occur when a person belongs to a group he
dislikes but cannot leave because of external restraints.

GROUP MEMBERSHIP AND PERSONAL IDENTITY

George Herbert Mead, in his classic work Mind, Self, and Society
(1934), pointed out that the individual's self as well as his capacity for
reflective thought develop in the course of social interaction with the

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
62 Theoretical Essays

members of his family and other groups in the community to which

he belongs. By taking the role of others and responding to his own ac-
tions as they would, the individual learns to anticipate the social ef-
fects of his actions. In addition, he learns that he and others are ex-
pected to behave toward one another in specified ways as a function
of his and their particular personal and social attributes-such as age,
sex, social class, race, religion, ethnic background, and nationality.
Thus a black child learns to behave differently toward black than
toward white children, and he learns to expect whites to behave dif-
ferently toward him than they do toward whites. Similarly, children
learn that certain activities are "feminine" and others are "masculine"
and that disapproval is risked by engaging in behavior that is con-
sidered appropriate for the opposite sex but not for one's own. How-
ever, each child's experience is in some respects unique, and thus the
conceptions among a group of what it is to be a member of that
group will not be identical. Moreover, the meaning of any particular
subidentity, such as "black," is influenced by the total configuration of
social identities of which it is an element. Thus the conception of
"black," like that of "Jew," is affected by the linking of the two attrib-
utes in the configuration "black Jew." Adding other elements to the
configuration, such as "rich," "young," "woman," and "Brazilian,"
further alters and defines the meaning of the initially specified sub-
identity "black."
Although the meaning of any personal subidentity is influenced by
the total configuration of subidentities, it would be a mistake to as-
sume that all elements are equally influential in determining an indi-
vidual's thoughts, feelings, and behavior. It is evident that situational
factors help determine which subidentity will be elicited most strongly
at a given time: different subidentities are likely to be most salient and
most influential in different social situations. The subidentity of
"white" is more likely to be elicited in the presence of blacks than in
the presence of other whites-unless the other whites are discussing
blacks or interracial relations. A New Yorker and a Texan are more
likely to feel a common identity as Americans in China than in the
United States. Thus a subidentity is made salient in a situation by
contrast with the presence of members of other different or antitheti-
cal groups that are used to mark off the boundaries of one's own
group (Herman 1970). It is also made salient by the presence of
threats, danger, discrimination, or other potential harm to oneself be-

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 63

cause of membership in a given group. If derogatory comments or

discriminatory actions are liable to be directed at you or other mem-
bers of your group at any time from almost anybody, then you will be
continuously aware of your membership in this group. A subidentity
is also made salient by the prospect of reward or other potential gain
resulting from membership in a particular group. More generally, the
more eliciting stimuli that are present in a situation-whether those
stimuli be negative or positive in implication-the more salient will
be the identity in that situation.
It is apparent that subidentities differ in their readiness to be
evoked. Some subidentities are more pervasive than others and are
readily aroused in many different types of situations. My subidentity
as a member of my family group enters into many more situations
than my subidentity as a member of the Accabonac Tennis Club. It
connects with more people and with more of my other subidentities,
and thus it is a more pervasive influence on my thoughts, feelings, and
behavior.
Subidentities also differ in how central or important they are to the
individual's self-esteem; the more central a subidentity is, the more
likely it is to be evoked, and the more influential it will be when
evoked. One measure of the centrality of a subidentity is one's readi-
ness to resist its derogation or elimination. Thus I am more willing to
give up being a squash player than a tennis player, and I would aban-
don either of these rather than quit my profession. Similarly, I am
more ready to resist derogation of my ethnic group than my age
group.
The importance of a subidentity to one's self-esteem is determined
by the strength of the different types of bonds binding one to it. Sev-
eral different types of bonds can be distinguished (McCall 1970): as-
cribed bonds, bonds of commitment, bonds due to investment, bonds
of attachment, and instrumental bonds. The first three types of bonds
(ascription, commitment, and investment) are in large measure "re-
straining bonds"; they restrain one from leaving a group even if he de-
sires to do so. The latter two (attachment and instrumental) are "at-
tracting bonds," which pull the individual toward the group.
The strongest restraining bonds are those arising out of certain
ascribed statuses-such as family, sex, racial, ethnic, and national
group membership, all of which one acquires by birth rather than by
choice. Such statuses can rarely be changed. It is the combination of

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
64 Theoretical Essays

their unalterability and their social significance that gives these as-
cribed statuses their personal importance. One's handedness, left or
right, may be as difficult to alter as one's race, but it is by no means
as socially significant. Membership in a family, racial, sexual, ethnic,
or national group affects one's thoughts and actions in many situa-
tions; these effects are pervasive. In addition, by common definition,
membership in such groups typically excludes membership in other
groups of a similar type. That is, if you are black, you are not also
white; if you are male, you are not also female; if you are Jewish, you
are not also Christian. Thus being a member is thought to be more or
less distinctive, and since membership is linked to experiences from
early on in one's life, it is not unusual for one to get emotionally at-
tached to such groups, with the result that these memberships play an
important positive role in determining one's sense of identity.
Bonds of commitment may also tie one to a group and to the iden-
tity connected with it. The commitment may be to other members of
the group or to interested outsiders. Thus a girl who is engaged but no
longer interested in marrying may be reluctant to break the engage-
ment because of her commitment to her fiance or because of the ex-
pected disappointment of her parents and friends. Similarly, one's in-
vestments in a given identity-the amount of time, energy, life's
chances, money, and emotion previously expended in establishing and
maintaining the identity-will generally serve to bind one to continue
it even when one might not otherwise choose to do so. Nevertheless, it
should be noted that people do break up long-standing marriages or
change well-established careers if they expect that continued invest-
ments will be costly and not worthwhile. This is particularly likely if
they are aware of a more rewarding alternative for their future invest-
ments. The restraining bonds of commitment and investment are,
however, usually easier to break than those of ascription.
Bonds of attachment attract one to a group; such bonds develop
when significant personal needs-for security, acceptance, and
meaning-have been fulfilled in the group, and the group is thought
to be largely irreplaceable or nonsubstitutable as a source of fulfill-
ment for these needs. A group is likely to be viewed as irreplaceable
when no readily available alternatives are perceived (as in the case of
the small child in relation to the nuclear family), when the feasibility
of leaving the group to go to another one is small (as is the case of the
citizens of most nations), or when, as a result of an extended history

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
Group Formation 65

of participation in the group, the group has taken on a unique signifi-

cance (as is the case of family and ethnic groups).
Bonds of attachment provide a diffuse, nonspecific form of attrac-
tion to a group and to the idea of expressing one's identity by mem-
bership in the group. In contrast, instrumental bonds arise from the
success of the group in providing dependable rewards for fulfilling
one's specific roles or functions within the group and for being identi-
fied as a member of the group. Instrumental bonds are linked to the
specific success of the group in providing specific satisfactions. How-
ever, the more success the group has in doing this, and the wider the
range of satisfactions it provides, the more likely it is that diffuse
bonds of attachment will also develop.
It is evident that an individual who is getting ample instrumental
satisfactions from his group and is deeply attached to it will not find
himself in conflict, because his investments and ascription will restrain
him from abandoning his identification with the group. To the con-
trary, the more the individual is attracted to a group, the more willing
he will be to make investments in it, to make personal commitments
to it, and to bind himself irrevocably to it. Conversely, the less he is
attracted to a group, the less willing he will be to bind himself so
tightly that it would be difficult to leave it if he should choose to do
so.
Suppose that one is emotionally attached to one's subidentity as a
Jew, woman, or black-and irrevocably bound to it by bonds of as-
cription, commitment, and investment-but that it places one at a
distinct instrumental disadvantage in obtaining many kinds of oppor-
tunities and rewards. How one copes with this situation will be largely
determined by whether one views the disadvantages to be just or un-
just. If those who are disadvantaged by their group identity accept
their disadvantages as being warranted, they are unlikely to challenge
and conflict with those who are profiting from their relatively advan-
taged positions. The sense of being treated unjustly because of one's
membership in a group to which one is strongly attached and bound
is the energizer for much intergroup conflict. The sense of injustice is
felt particularly intensely in interracial, interethnic, and intersex con-
flicts because of the centrality of these group identities to the individu-
al's self esteem. When women or blacks or Jews are devalued as a
group, those who are identified and identify with the groups also are
personally devalued.

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms
66 Theoretical Essays

It is evident that those groups who gain from the disadvantaging of

other groups are usually the more powerful ones. They are more
likely to set the terms of their relationship with other groups and,
through their control of the state and other social institutions, to es-
tablish the legal and other reigning definitions of ''justice." Thus in
addition to their gains from exploitative actions, they commonly have
the reassurance of the official definition of justice and the support of
such major social institutions as the church, the press, and the schools
to deaden their sensitivities to the injustices inherent in their relations
with the disadvantaged group. The disadvantaged may, of course, be
taken in by the official definitions and the indoctrination emanating
from social institutions and, as a result, lose their sensitivity to their
situation of injustice. Even when they are not brainwashed into ac-
cepting their inferior positions as just, a sense of hopelessness about
the possibility of change may keep the disadvantaged quiescent. As I
have pointed out elsewhere (Deutsch 1972), the process of activating
the sense of injustice so that it is a stimulant to intergroup conflict and
social change entails falsifying and delegitimitizing the officially sanc-
tioned ideologies and myths that "justify" the injustices; exposing the
disadvantaged to new ideologies, new models, and new reference
groups that justify and give life to the possibilities of change in their
status; stimulating hope by successful efforts to improve their situa-
tion; and reducing their fear by increasing their relative bargaining
strength.
It would be a mistake to assume that the only form of intergroup
conflict is that between exploiting and exploited groups. In the next
chapter, intergroup conflict is considered more broadly. However, it is
well to recognize that deep passions are aroused whenever the value
and merit of an individual's personal identity are challenged, and this
is quite likely to happen when a group that he is closely identified
with is involved in a conflict with another group over such central val-
ues as justice and superiority-inferiority.

This content downloaded from

174.59.81.122 on Wed, 27 Sep 2023 22:35:43 +00:00
All use subject to https://about.jstor.org/terms