Documente Academic
Documente Profesional
Documente Cultură
50 articles dealing with stages of group development over time are separated by
group setting, as follows: therapy-group studies, T-group studies, and natural-
and laboratory-group studies. The stages identified in these articles are separated
into those descriptive of social or interpersonal group activities and those
descriptive of group-task activities. Finally, 4 general stages of development
are proposed, and the review consists of fitting the stages identified in the
literature to those proposed. In the social realm, these stages in the develop-
mental sequence are testing-dependence, conflict, cohesion, and functional roles.
In the task realm, they are orientation, emotionality, relevant opinion exchange,
and the emergence of solutions. There is a good fit between observed stages and
the proposed model. Further study of temporal change as a dependent variable
via the manipulation of specific independent variables is suggested.
The purpose of this article is to review the the area of small-group development will be
literature dealing with the devlopmental cited, and a framework within which this phe-
sequence in small groups, to evaluate this nomenon can be better understood and fur-
literature as a body, to extrapolate general ther investigated will be presented. This
concepts about group development, and to framework will also serve to integrate the
suggest fruitful areas for further research. variety of studies cited in a meaningful way.
While small-group processes have been
given great attention in recent years by be- CLASSIFICATION MODEL
havioral scientists, the question of change in
process over time has been relatively neg- The classification approach adopted for dis-
lected. Perhaps the major reason for this is tinguishing between and within developmental
the overwhelming tendency of the small-group studies is a threefold one. The delineations
researcher to run groups for short periods of are based on (a) the setting in which the
time and thus avoid the "problems" created group is found, (b) the realm into which the
by temporal change. Laboratory studies of group behavior falls at any point in time,
developmental phenomena are quite rare. The that is, task or interpersonal, and (c) the
majority of articles dealing with sequential position of the group in a hypothetical de-
group development come from the group- velopmental sequence (referred to as the stage
therapy setting and human relations training- of development). It is this last delineation
group setting, neither of which features strict that allows not only for the separation and
experimental control nor manipulation of in- ordering of observations within each set-
dependent variables. Moreover, the only ma- ting, but for the development of additional
jor theoretical statements of group develop- hypotheses as well.
ment which have appeared are those of Bales
Setting
(19S3), Schutz (1958), and Bach (1954).
In an attempt to bring the facts and the Classification according to setting allows for
issues into sharper focus, existing research in the clustering of studies based on their simi-
1 larity of features, for example, group size,
From Bureau of Medicine and Surgery, Navy
Department, Research Task MR005.12-200S.01, Sub- group problem area, group composition, dura-
task I. The opinions and statements contained herein tion of "group life," etc. More similarity
are the private ones of the writer and not to be between observations made in the same set-
construed as official or reflecting the views of the ting than in different settings is expected.
Navy Department or the naval service at large.
The author is grateful to Irwin Altman for his in- In the group-therapy setting the task is to
valuable ideas and suggestions, and to Robert Nye help individuals better deal with their per-
for his efforts in helping to review the literature. sonal problems. The goal is individual adjust-
384
DEVELOPMENTAL SEQUENCE IN SMALL GROUPS 385
ment. Such groups contain from 5 to IS The laboratory-task setting features groups
members, each of whom has some debili- brought together for the purpose of studying
tating personal problem, and a therapist, and group phenomena. Such groups are small (gen-
the group exists for 3 months or more. The erally under 10 members), have a short life,
developmental data for such groups consist and may or may not have leaders. In this set-
of the observations of the therapist and ting, groups are given a task or tasks which
those professional observers that are present, they are to complete. Quantitative data are
usually as trainees. Such data are highly collected and analyzed based on multiple-
anecdotal in nature and reflect the clinical group performances.
biases of the observers. Furthermore, such The last two settings have been combined
accounts are usually formulated after the fact due to the small number of studies in each
and based on the observation of a single (the dearth of group development studies in
group. Since the bulk of the literature re- the industrial area is notable), and also be-
viewed comes from this setting, its generality cause theoretical statements are reviewed
must be limited by the limitations of the which are generalized to cover both areas. All
setting and the mode of data collection. studies will be classified into one of the three
In the human relations training-group setting categories according to best fit.
(T-group) setting, the task is to help indi-
viduals interact with one another in a more Realm: Interpersonal versus Task
productive, less defensive manner, and to be Within the studies reviewed, an attempt will
aware of the dynamics underlying such inter- be made to distinguish between interpersonal
action. The goal is interpersonal sensitivity. stages of group development and task be-
Such groups contain ordinarily from IS to haviors exhibited in the group. The contention
30 members, usually students or corporation is that any group, regardless of setting, must
executives, and one trainer or leader, and address itself to the successful completion of
endure from about 3 weeks to 6 months. a task. At the same time, and often through
The most striking differences between the same behaviors, group members will be
therapy- and training-group settings are in relating to one another interpersonally. The
the areas of group composition, task, goal, and pattern of interpersonal relationships is re-
duration of group life. Such differences can ferred to as group structure and is interpreted
account for different findings in the two set- as the interpersonal configuration and inter-
tings. The most striking similarity is in the personal behaviors of the group at a point in
manner of data collection. Data in the time, that is, the way the members act and
training-group setting are highly anecdotal, relate to one another as persons. The content
subjective, collected by the trainer and his of interaction as related to the task at hand
co-workers, and often based on the observa- is referred to as task activity. The proposed
tions of a single group. Again, this serves to distinction between the group as a social en-
limit the generality of these findings. tity and the group as a task entity is similar
The natural-group setting is distinguished to the distinction between the task-oriented
on the basis that the group exists to perform functions of groups and the social-emotional-
some social or professional function over integrative functions of groups, both of which
which the researcher has no control. Members occur as simultaneous aspects of group func-
are not brought together for self-improve- tioning (Bales, 19S3; Coffey, 1952; Deutsch,
ment; rather, they come together to do a job. 1949; Jennings, 1947).
Such groups may be characterized either by In therapy groups and T groups, the task is
appointed or emergent leadership. Presidential a personal and interpersonal one in that the
advisory councils and industrial groups repre- group exists to help the individuals deal with
sent examples of natural groups. Similar limi- themselves and others. This makes the inter-
tations to generalization based on the manner personal-task distinction a fuzzy one. A fur-
of data collection and number of groups ob- ther problem with this distinction occurs be-
served applies in this setting as in the previous cause the studies cited do not distinguish be-
settings. tween the two realms and often talk about
386 BRUCE W. TV C KM AN
interpersonal development at one point in the in dealing with the task and how this infor-
sequence and task development at another mation is to be obtained. In orienting to the
point. The distinction will be maintained, how- task, one is essentially defining it by discov-
ever, because of the generic difference between ering its "ground rules." Thus, orientation, in
the reaction to others as elements of the general, characterizes behavior in both inter-
group task versus the reaction to others as personal and task realms during this stage. It
social entities. Failing to separate stages by is to be emphasized that orientation is a gen-
realm obscures the continuity of the develop- eral class of behavior which cuts across set-
mental process. While the two realms differ in tings; the specifics of orientation, that is,
content, as will be seen, their underlying dy- what one must orient to and how, will be
namics are similar. setting-specific.
The second phase in the development of
Proposed Developmental Sequence group structure is labeled as intragroup con-
flict. Group members become hostile toward
The following model is offered as a con- one another and toward a therapist or trainer
ceptualization of changes in group behavior, as a means of expressing their individuality
in both social and task realms, across all group and resisting the formation of group structure.
settings, over time. It represents a set of Interaction is uneven and "infighting" is com-
hypotheses reflecting the author's biases mon. The lack of unity is an outstanding fea-
(rather than those of the researchers) and ture of this phase. There are characteristic
the perception of trends in the studies re- key issues that polarize the group and boil
viewed which become considerably more ap- down to the conflict over progression into the
parent when these studies are viewed in the "unknown" of interpersonal relations or re-
light of the model. The model of development gression to the security of earlier dependence.
stages presented below is not suggested for Emotional response to task demands is
primary use as an organizational vehicle, al- identified as the second stage of task-activity
though it serves that function here. Rather, development. Group members react emotion-
it is a conceptual statement suggested by the ally to the task as a form of resistance to the
data presented and subject to further test. demands of the task on the individual, that is,
In the realm of group structure the first the discrepancy between the individual's per-
hypothesized stage of the model is labeled as sonal orientation and that demanded by the
testing and dependence. The term "testing" task. This task stage will be most evident
refers to an attempt by group members to when the task has as its goal self-understand-
discover what interpersonal behaviors are ing and self-change, namely, the therapy- and
acceptable in the group, based on the reac- training-group tasks, and will be considerably
tions of the therapist or trainer (where one is less visible in groups working on impersonal,
present) and on the reactions of the other intellectual tasks. In both task and interper-
group members. Coincident to discovering sonal realms, emotionality in response to a
the boundaries of the situation by testing, one discrepancy characterizes this stage. How-
relates to the therapist, trainer, some power- ever, the source of the discrepancy is different
ful group member, or existing norms and in the different realms.
structures in a dependent way. One looks to The third group structure phase is labeled
this person, persons, or standards for guidance as the development oj group cohesion. Group
and support in this new and unstructured sit- members accept the group and accept the idio-
uation. syncracies of fellow members. The group be-
The first stage of task-activity development comes an entity by virtue of its acceptance
is labeled as orientation to the task, in which by the members, their desire to maintain and
group members attempt to identify the task in perpetuate it, and the establishment of new
terms of its relevant parameters and the man- group-generated norms to insure the group's
ner in which the group experience will be used existence. Harmony is of maximum impor-
to accomplish the task. The group must decide tance, and task conflicts are avoided to insure
upon the type of information they will need harmony.
DEVELOPMENTAL SEQUENCE IN SMALL GROUPS 387
which will be referred to as Prestage 1. Thorpe 1953), and (/) intellectualization (Clapham
and Smith (1953) and Osberg and Berliner & Sclare, 1958; Wender, 1946).
(1956), in therapy with hospitalized narcotic This stage is also characterized by more
addicts, describe an initial stage of resistance, direct attempts at orientation toward the task
silence, and hostility followed by a testing as illustrated in (a) a search for the meaning
period where patients attempt to discover of therapy (Cholden, 1953), (b) attempts to
what behaviors the therapist deems accept- define the situation (Powdermaker & Frank,
able. Shallow, Ward, and Rubenfeld (1958), 1948), (c) attempts to establish a proper
who worked with institutionalized delinquents, therapeutic relationship with the therapist
described two such stages of resistance and through the development of rapport and con-
hostility preceding the testing stage, while fidence (Dreikurs, 1957; King, 1959; Wolf,
Martin and Hill (1957) theorized about a 1949), (d) mutual exchange of information
stage of isolation and "unshared behavior" (Grotjahn, 1950), and (e) suspiciousness of
preceding one of stereotypic responding to and fearfulness toward the new situation
fellow group members and a dependent orien- which must be overcome (Corsini, 1957).
tation toward the therapist.
Three of the four studies identifying a Pre- Stage 2
stage 1 were specifically based on observa- Group Structure: Intragroup Conflict. Thir-
tions of groups of antisocial individuals (drug teen of the 26 studies of group therapy re-
addicts and delinquents) who probably must viewed identified a stage of intragroup con-
be won over to the situation and their initial flict (in 11 cases as a second stage and in 2
extreme resistance overcome before the normal as a first stage). Abrahams (1949) identifies
sequence of therapy-group development can an interaction stage typified by defensiveness,
begin. This would account for Prestage 1. competition, and jealousy. Bion (1961) dis-
The remaining studies did not identify an cusses a figkt-fligkt period in which members
initial stage of testing-dependence but dealt conflict with the therapist or attempt to psy-
either with task development (to be discussed chologically withdraw from the situation.
below), or offered as an initial Stage 1 which Grotjahn (1950) identifies a stage of increas-
is postulated here as a second stage. Finally, a ing tension, while Parker (1958) talks about
study by Parker (1958) described an initial a crisis period where friction is increased,
stage of cohesive organization in which sub- anxiety mounts, rules are broken, arguments
groups are formed, rules followed, and har- ensue, and a general structural collapse oc-
mony maintaineda description which is dif- curs. Powdermaker and Frank (1948) discuss
ficult to fit into the testing-dependence cate- a second stage featuring sharp fluctuation of
gory. relationships, sharp reversals of feelings, and
Task Activity: Orientation and Testing. "intense but brief and brittle linkages."
During the initial stage, task development is Schindler (1958) talks about a stage of psy-
characterized by indirect attempts to discover chodramatic acting-out and localization of con-
the nature and boundaries of the task, i.e., flicts in the group, while Shellow et al. (1958)
what is to be accomplished and how much describe a stage characterized by ambivalence
cooperation is demanded, expressed specifi- toward the therapist which is expressed
cally through (a) the discussion of irrelevant through the formation of conflicting factions
and partially relevant issues (Bion, 1961; in the group. Stoute (1950) describes a sec-
Coffey, Freedman, Leary, & Ossorio, 1950; ond stage beginning with derogation and
Martin & Hill, 1957; Osberg & Berliner, negativity, while Thorpe and Smith (1953)
1956), (b) the discussion of peripheral prob- describe a stage beginning with disintegration,
lems (Stoute, 1950), (c) the discussion of distance, defenses out of awareness, and dis-
immediate behavior problems (Abrahams, rupted communication. King (1959), in ac-
1949), (d) the discussion of symptoms (Bach, tivity-group therapy, describes a second stage
1954; Taylor, 1950), (e) griping about the of benign regression characterized by extreme
institutional environment (Mann & Semrad, acting-out and unacceptable behavior. Martin
1948; Shellow et al., 1958; Thorpe & Smith, and Hill (1957) theorize about a stage of
DEVELOPMENTAL SEQUENCE IN SMALL GROUPS 389
polarization featuring the emergence of sub- sciousness is developed and establishment and
groups following a stage of interpersonal ex- maintenance of group boundaries is empha-
ploration. sized. Bion (1961) discusses the basic as-
Coffey et al. (1950) identify an initial stage sumption of pairing in which the emphasis is
of defensiveness and resistance where mem- on cohesion, but the unit is the pair as op-
bers clash with one another. However, these posed to the whole group. Coffey et al. (1950),
authors also see "pecking orders" being es- Corsini (1959), and Taylor (1950) describe a
tablished during this period; perhaps their stage following the stage of intragroup hos-
initial stage includes Stages 1 and 2 as postu- tility in which the group becomes unified and
lated in this review. Mann (1953) describes is characterized by the existence of a com-
an initial phase of "working through of hos- mon goal and group spirit. Parker (1958) and
tility" followed by a stage of "working Shellow et al. (1958) see the stage of crisis
through of anxieties." The hostility phase is and factions being followed by one featuring
characterized by disruption and fragmentation consensual group action, cooperation, and
which are reduced gradually in the anxiety mutual support. Mann and Semrad (1948),
phase. Grotjahn (1950), and Powdermaker and
The remaining studies fail to identify this Frank (1948) describe a third stage charac-
stage. Some of them jump from Stage 1 di- terized by group integration and mutuality.
rectly to Stage 3, while others deal with task Noyes (1953) describes a middle stage of
development as concerns the first two stages of group integration, while Stoute (1950) and
therapy-group development. Thorpe and Smith (1953) see the stage of in-
Task Activity: Emotional Response to Task tragroup hostility grading into a period of
Demands. The outstanding feature of this sec- unity, support, and freedom of communica-
ond task stage appears to be the expression of tion. Martin and Hill (1957) theorize about a
emotionality by the group members as a form stage featuring awareness that the group is an
of resisting the techniques of therapy which organism preceding the final stage of develop-
require that they "expose" themselves and of ment. Abrahams (1949) describes the devel-
challenging the validity and usefulness of opment of "we-consciousness" in the third
therapy (Bach, 1954; Barton, 1953; Cholden, stage, while Mann (1953) sees the third
1953; Clapham& Sclare, 1958; Mann, 1953; stage as one of personal mutual exploration
Mann & Semrad, 1948; Martin & Hill, 1957; and analysis during which the group attains
Stoute, 1950; Wender, 1946). Furthermore, unity.
mention is made of the fact that this is a The notion that the group becomes a simu-
period of extreme resistance to examination lation of the family constellation (that is,
and disclosure (Abrahams, 1949; Barton, through transference members react to one
1953), and an attempt at analysis of this re- another as members of their family), with
sistance is made (Wolf, 1949). Others em- the unity and cohesion generally accepted in
phasize ambivalence toward the therapist that structure, fits as a close parallel to the
(Shellow et al,, 1958), the discussion of sensi- stage of development of group cohesion being
tive areas (Powdermaker & Frank, 1948), postulated. Beukenkamp (1952) describes the
psychodrama (Schindler, 1958), and resis- middle stage of reliving the process of the
tance via "putting one on" (Thorpe & Smith, family constellation where the group becomes
1953). a familylike structure, while King (1959)
utilizes a similar description (that is, family
Stage 3 unity in the group) for the final stage in ac-
Group Structure: Development of Group tivity-group therapy. Wender (1946) and
Cohesion. Twenty-two of the 26 studies re- Wolf (1949) both describe a stage preceding
viewed identified a stage in which the group the final stage in which the group becomes the
became a cohesive unit and developed a sense new family through the displacement of parent
of being as a group. Bach (1954), Barton love.
(1953), and Clapham and Sclare (1958) Studies that fail to identify this stage are
identify a stage during which ingroup con- those that deal primarily with task develop-
390 BRUCE W. TUCKMAN
ment or those that integrate it as part of the peutic force producing encouragement and in-
final stage. tegrating problems with roles. Martin and
Task Activity: Discussing Oneself and Hill (1957) identify the group as an integra-
Other Group Members, Many researchers ob- tive-creative-social instrument in its final
served probing and revealing by group mem- stage which facilitates problem solving, diag-
bers at a highly intimate level during this pe- nosis, and decision making. Osberg and Ber-
riod and labeled it as (a) confiding (Clapham liner (1956) describe the self-starting stage
& Sclare, 1958; Coffey et al., 1950; Thorpe & where the group environment supports analy-
Smith, 1953), (b) discussing personal prob- sis, while Mann (1953) discusses a final stage
lems in depth (Corsini, 1957; Mann & Sem- of personal mutual synthesis.
rad, 1948; Osberg & Berliner, 1956; Taylor, Other therapy researchers failing to specif-
1950), (c) exploring the dynamics at work ically delineate this final stage in social de-
within the individual (Dreikurs, 1957; Noyes, velopment have tended to lump the third and
1953), and (d) exploring the dynamics at fourth stages together and not make the dis-
work within the group (Bach, 1954; Martin tinction between the development of cohesion
& Hill, 1957; Powdermaker & Frank, 1948). and the "use" of cohesion (via functional
Beukenkamp (1952) observed that recalled roles) as a therapeutic force. Such descrip-
material was related to the family; Abrahams tions were included in the section on the third
(1949) observed the process of common idea- stage. The small number of investigators
tion; and Shellow et al. (1958) and Wolf identifying this final stage is most likely due
(1949) emphasized patients' discussion of to the high visibility of task functions occur-
topics related to transference to the therapist ring during this time period which obscure
and to other group members which took place and minimize social processes occurring simul-
during this period. taneously.
Task Activity: Emergence of Insight. There
Stage 4 seems to be overwhelming agreement among
Group Structure: Functional Role-related- the observers of therapy-group development
ness. Only 12 of the therapy studies are at all that the final stage of task development is
explicit in their identification of this stage. characterized by attainment of the desired
Almost all of the therapists discuss the final goal, insight into one's own problems, an un-
stage of development of the therapy group in derstanding of the cause of one's abnormal
task terms as the therapeutic stage of under- behavior and, in many cases, modification of
standing, analysis, and insight. The group is oneself in the desired direction (Beukenkamp,
seen as serving a therapeutic function, but the 1952; Bion, 1961; Clapham & Sclare, 1958;
nature of this therapeutic function is not Coffey et al., 1950; Corsini, 1957; Dreikurs,
spelled out. This is a stage of mutual task in- 1957; King, 1959; Noyes, 1953; Schindler,
teraction with a minimum of emotional inter- 1958; Stoute, 1950; Thorpe & Smith, 1953;
ference made possible by the fact that the Wender, 1946; Wolf, 1949). Others (Abra-
group as a social entity has developed to the hams, 1949; Bach, 1954; Barton, 1953;
point where it can support rather than hinder Cholden, 1953; Grotjahn, 1950; Shellow et
task processes through the use of function- al., 1958; Taylor, 1950) place more emphasis
oriented roles. on the processes of attempting to develop in-
Bach (1954) and Bion (1961) both refer sight and change during this last period as
to the group in its final stage as the work opposed to the development of such insight
group. As such it serves a function supportive and change itself.
of therapy. Wender (1946) and Abrahams Two additional therapy-group studies are
(1949) see the group as creating a therapeutic worthy of inclusion, both of which utilized a
atmosphere in the final stage, while Wolf technique for collecting and analyzing data
(1949), Stoute (1950), and Corsini (1951) which was highly dissimilar to the approach
describe this stage as one of freedom and used in the other therapy-group studies,
friendliness supportive of insightful behavior namely, interaction-process analysis (Bales,
and change. Both Coffey et al. (1950) and 1950). Psathas (1960) found that groups
Dreikurs (1957) see the group as a thera- phase from orientation to evaluation to con-
DEVELOPMENTAL SEQUENCE IN SMALL GROUPS 391
trol, based on an analysis of early, middle, being of central concern and emphasize the
and late sessions. Talland (1955) failed to orientation aspects of the first stage, there is
observe this phase movement based on an overlap with the scheme proposed herein.
analysis of the first eight sessions. Moreover, orientation as a first stage fits the
hypothesized initial stage for task activities;
STAGES OF DEVELOPMENT IN perhaps the observation in the Tulane studies
TRAINING GROUPS (1957) of a member orientation as an initial
Stage 1 stage is better classified in the task-activity
Group Structure: Testing and Dependence. area.
Nine of the 11 training-group studies reviewed Task Activity: Orientation. Bradford
that deal with the development of group (1964b) identifies an initial stage of learning
structure identify an initial stage character- how to learn which is characterized by accep-
ized at least in part by testing and depen- tance of the group's goal and orientation to
dence, with the emphasis on the dependent as- the techniques to be used. Herbert and Trist
pect of this stage. (1953) label their initial stage as discovery,
Herbert and Trist (1953), Bennis and in which the members orient themselves to the
Shepard (1956), Bradford and Mallinson consultant or trainer who serves an interpre-
(1958), and Bradford (1964a) describe the tive and educational role. Stock and Thelen
initial group phase as one characterized by (1958) discuss an initial stage characterized
the strong expression of dependency needs by by little "work" and a variable amount of
the members toward the trainer, and attempts "emotionality," during which time the mem-
at group structuring to work out authority bers are concerned with defining the directions
problems by the quick acceptance of and de- the group will pursue.
pendence on such structure and arbitrary As can be seen, initially interpersonal prob-
norms. Thelen and Dickerman (1949) discuss lems are dealt with via dependence, while task
initial stage establishment of a leadership problems are met with task-orienting behav-
hierarchy catering to the dependency needs of ior (i.e., what is to be accomplished and how).
the members. Hearn (1957) sees group mem-
bers making an attempt to structure the un- Stage 2
known and to find their position in the group Group Structure: Intragroup Conflict. Ten
in the earliest group stage. Here again, struc- of the 11 studies identify intragroup conflict
ture reflects the expression of dependency as a second stage, while the remaining study
needs. (Whitman, 1964) describes an initial stage
Miles (1953) describes a first stage charac- encompassing both dependence and hostility,
terized by establishment oj the situation in that order.
through interpersonal exploration and testing, Barron and Krulee (1948) and Bradford
while Semrad and Arsenian (1961) identify (1964a) discuss a second stage characterized
an initial phase during which group members by group cleavage and conflict. Both studies
"test" the central figure and "test" the situa- identify the emergence of polarities during
tion. this stagemembers favoring a more active,
Whitman (1964) describes a beginning less defensive approach versus those who are
stage in which the chief "vectors" are de- more passive and defensive and seek "safety"
pendency and hostility. It would appear that via structure. Thelen and Dickerman (1949),
Whitman has identified a first stage which Hearn (1957), the Tulane studies (1957),
combines the first two stages proposed in this and Bradford and Mallinson (1958), as well,
article. identify a similar polarization and resultant
The two studies that do not yield an exact conflict, frustration, and disruption during the
fit to the proposed scheme are those of Barron second stage.
and Krulee (1948) and the Tulane Studies in Herbert and Trist (1953) describe a second
Social Welfare (1957) which identify an ini- stage characterized in part by resistance,
tial period characterized by the emergence of while Miles (1953) identifies anarchic re-
leadership and orientation, respectively. Inso- bellion during this stage of anxiety, threat,
far as these authors see the authority area as and resistance. Semrad and Arsenian (1961)
392 BRUCE W. TUCKMAN
identify rivalry for the position of central fig- affection bonds; in the latter, the first co-
ure and emotional struggles in this period, hesion stage features the emergence of struc-
while Bennis and Shepard (1956) see a simi- ture, roles, and "we-feeling," while the second
lar power struggle in which counterdependents features increased group identification on a
seek to usurp the leader, resulting in a con- conscious level and vacillation in role accep-
flict between counterdependents and depen- tance. Whitman (1964) talks about a middle
dents. phase, following conflict, described as the de-
There appears to be general agreement that velopment of a new group culture via the
the dependency stage is followed by a stage of generation of norms and values peculiar to
conflict between warring factions representing the group as an entity. Bradford and Mallin-
each side of the polarized issue: dependence son (1958) describe Stage 3 as one of reor-
versus independence, safe retreat into the fa- ganization, in which reforming and repair take
miliar versus risky advance into the unfa- place and a flexible organization emerges.
miliar, defensiveness versus experimenting. Bradford (1964a) describes a third stage
Task Activity: Emotional Response to Task in which the group norm of "openness"
Demands. Bradford (1964b) identifies a sec- emerges, and a fourth stage in which the
ond stage in which individuals learn how to group generates additional norms to deal with
give help which requires that they remove self-revelation and feedback. Furthermore,
blocks to learning about themselves, reduce Bradford (1964b) identifies a third stage as
anxiety, and express real reactions. Stock and one of developing a group climate of permis-
Thelen (1958) see emotionality occurring in siveness, emotional support, and cohesiveness
considerable excess of work during this pe- in which learning can take place. This descrip-
riod. The Tulane studies (1957) describe the tion would appear to subserve both interper-
second stage as one of experimental aggres- sonal and task realms.
siveness and hostility where individuals ex- Bennis and Shepard (1956) describe a
press themselves freely. third stage in which resolution of authority
Thus, self-change and self-denial necessi- problems occurs, and a fourth stage charac-
tated by the learning task is reacted to emo- terized by smooth relations and enchant-
tionally, as is the imposition of the group on ment as regards the interpersonal sphere of
the individual. Often the two (representative group functioning. Finally, Barron and Krulee
of the two realms) are difficult to separate. (1948) identify the third stage as increasing
member responsibility and changing faculty
Stage 3 role in which a definite sense of structure and
Group Structure: Development oj Group goal orientation emerge in the group.
Cohesion. All of the relevant T-group develop- Task Activity: Discussing Oneself and
ment studies see the stage of conflict and Others. Herbert and Trist (1953) identify a
polarization as being followed by a stage char- second stage labeled as execution, in which
acterized by the reduction of the conflict, the group settles down to the description of
resolution of the polarized issues, and estab- a single basic problem and learns to accept
lishment of group harmony in the place of "the examination of what was going on inside
disruption. It is a "patching-up" phase in of itself as a regular part of the task . . . ."
which group norms and values emerge. Stock and Thelen (1958) describe a third
Hearn (1957), Miles (1953), and Thelen task phase in which the group shows a new
and Dickerman (1949) identify a third stage ability to express feelings constructively and
characterized by attempts to resolve conflict creatively. While emotionality is still high, it
and the consequent development of group co- now contributes to work.
hesion and mutual support. Semrad and Ar- While the social function of the third stage
senian (1961) and the Tulane studies (1957) is to cause a unique and cohesive group struc-
each describe two phases in their temporal ture to emerge, the task function is to at-
sequences which would be included in Stage 3. tempt to use this new structure as a vehicle
In the case of the former, their first cohesion for discovering personal relations and emo-
phase is characterized by group cohesion pro- tions by communicating heretofore private
cesses and their second by the development of feelings.
DEVELOPMENTAL SEQUENCE IN SMALL GROUPS 393
losophy becomes pragmatic, that is, the uni- (1959) present evidence which indicates that
fied-group approach is applied to the task. sex of the participants does not affect the pat-
Schroder and Harvey (1963) postulate a tern of phase movements.
final stage of positive interdependence, char- Finally, Smith (1960) has observed that
acterized by simultaneous autonomy and mu- experimental groups show early concentration
tuality (i.e., the members can operate in any on matters not related to the task, and, only
combination, or as a unit), and an emphasis later in the development sequence, concen-
on task achievement which is superordinate to trate on task-relevant activities. Again, this
social structure. finding suggests a strong similarity between
Theodorson (1953) sees the group as de- task development in laboratory groups and
veloping into a subculture over time, along in therapy and training groups, since, in the
with the development of member responsibil- latter settings, constructive task-relevant ac-
ity to the group. tivity appears only late in the developmental
Schutz (1958) does not identify a fourth sequence.
stage; rather, he sees his three postulated DISCUSSION
stages in continually cycling over time.
Task Activity: Emergence of Solution. The The literature that has been reviewed can
third and final phase observed by Bales and be criticized on a number of grounds. First,
Strodtbeck (1951) and Bales (1953) is one it may be pointed out that this literature
in which major emphasis is placed on prob- cannot be considered truly representative of
lems of control (as reflected by categories: small-group developmental processes, since
"asks for suggestion" and "gives sugges- certain settings have been overrepresented,
tion"). The purpose of suggestions is to offer primarily the therapy-group setting, and
solutions to the task based on information others underrepresented, primarily the natu-
gathered and evaluated in previous develop- ral-group and laboratory-group settings. This
mental periods. This then represents an ana- shortcoming cannot be rectified within the
logue to final stages in therapy- and training- existing literature; rather, it must serve as a
group task development where the emergence stimulus for further research in the latter
of insight yields solutions to personal prob- group settings. Furthermore, the inequality of
lems. setting representation necessitates caution in
These authors do not identify a period of generalizing from this literature. Generaliza-
task development in laboratory groups com- tion must, perforce, be limited to the fact that
parable to the second task stage in therapy- what has been presented is mainly research
and training-group development which fea- dealing with sequential development in ther-
tures the expression of emotional material. apy groups.
Again, because therapy and training tasks are A second source of criticism concerns the
personal ones, this will be reflected in the extent of experimental rigor characteristic of
content of discussion, specifically by the mani- the majority of studies cited in this review.
festation of resistance prior to dealing with Most of the studies carried out in the therapy-
the personal task at a level of confidence and group, training-group, and natural-group set-
honesty. This task stage does not appear to tings are based on the observation of single
be quite relevant in laboratory discussion groups. Furthermore, these observations are
groups, and its existence has not been re- qualitative rather than quantitative, and as
ported by Bales and Strodtbeck (1951) or such are subject to the biases of the observer,
Bales (1953). ordinarily the therapist or trainer. This is not
Philp and Dunphy (1959) have further to suggest that the therapy-group setting is
substantiated the findings of Bales and Strodt- not appropriate for studying group processes,
beck (1951) and Bales (1953) by observing but that the study of such processes should
the same phase-movement pattern in groups be more subject to methodological considera-
working on a different type of discussion tions. A good instance of the application of
problem.8 Furthermore, Philp and Dunphy
ment, namely, orientation to evaluation to control.
8
As mentioned earlier, Psathas (I960), working However, Talland (19SS) failed to get this phase
with therapy groups, observed the same phase move- movement in therapy groups.
396 BRUCE W. TV CKM AN
such considerations is the study of Psathas tional responding in the task sphere. These
(1960) conducted in the therapy-group set- behaviors serve as resistance to group influ-
ting. Psathas coded group protocols using ence and task requirements and may be la-
Bales' (19SO) scheme of interaction-process beled as storming.
analysis. After satisfactory reliabilities were Resistance is overcome in the third stage
obtained, the data could be considered as in which ingroup feeling and cohesiveness de-
highly quantitative and objective, and could velop, new standards evolve, and new roles are
then be subjected to statistical analysis. Ap- adopted. In the task realm, intimate, personal
proaches of equal rigor are recommended for opinions are expressed. Thus, we have the
other studies conducted in the therapy-group stage of norming.
setting and other settings as well. Finally, the group attains the fourth and
A final criticism concerns the description final stage in which interpersonal structure be-
and control of independent variables. Since comes the tool of task activities. Roles become
most of the studies in the therapy-, training-, flexible and functional, and group energy is
and natural-group settings used a single channeled into the task. Structural issues have
group, the control and systematic manipula- been resolved, and structure can now become
tion of independent variables was impossible. supportive of task performance. This stage
In the absence of the manipulation of inde- can be labeled as performing.
pendent variables and the consequent dis- Although the model was largely induced
covery of their differential effects within stud- from the literature, it would seem to with-
ies, these effects can only be approximately stand the test of common sense as well as
discerned by comparing studies. However, being consistent with developmental theory
many independent variables are likely to vary and findings in other areas. It is not unrea-
from study to study, for example, group com- sonable to expect "newness" of the group to
position, duration, etc., and little light will be be greeted by orienting behavior and resultant
shed on the effects of these variables on the unsureness and insecurity overcome through
developmental process. Therefore, no conclu- dependence on an authority figure, as pro-
sions about the specific effects of independent posed in the model. Such orienting responses
variables on developmental phenomena will be and dependence on authority are character-
drawn, and further work along these lines is istic of the infant during the first year (Ilg &
encouraged. Ames, 1955), the young child when first ap-
In order to isolate those concepts common prehending rules (Piaget, 1932), and the pa-
to the various studies reviewed (across set- tient when first entering psychotherapy (Rot-
tings), a developmental model was proposed. ter, 1954).
This model was aimed at serving a conceptual After the "newness" of the group has "worn
function as well as an integrative and organi- off," the members react to both the imposi-
zational one. The model will be summarized tion of the group and the task emotionally
here. and negatively, and pose a threat to further
Groups initially concern themselves with development. This proposal is mirrored by
orientation accomplished primarily through the rebelliousness of the young child follow-
testing. Such testing serves to identify the ing his "obedient" stages (Ilg & Ames, 1955;
boundaries of both interpersonal and task be- Levy, 1955).
haviors. Coincident with testing in the inter- Such emotionality, if overcome, is followed
personal realm is the establishment of de- by a sense of "pulling together" in the group
pendency relationships with leaders, other and being more sensitive to one another. This
group members, or preexisting standards. It sensitivity to others is mirrored in the de-
may be said that orientation, testing, and de- velopment of the child (Ilg & Ames, 1955;
pendence constitute the group process of Piaget, 1932) and represents an essential
forming. aspect of the socialization process (Mead,
The second point in the sequence is charac- 1934).
terized by conflict and polarization around Finally, the group becomes a functional in-
interpersonal issues, with concomitant emo- strument for dealing with the task. Interper-
DEVELOPMENTAL SEQUENCE IN SMALL GROUPS 397
sonal problems lie in the group's "past," and thing or may reflect an unusual set of inde-
its present can be devoted to realistic ap- pendent conditions. Parker was observing a
praisal of and attempt at solutions to the task ward population of about 25, rather than a
at hand. This interdependence and "marriage small weekly therapy group. It may be that
to reality" is characteristic of the "mature" the hypothesized first stage is somewhat in-
human being (Erikson, 1950; Fromm, 1941) appropriate for larger, living-together groups.
and the "mature" 9-year-old child (Ilg & While the suggested sequence appeared to
Ames, 1955)* hold up under widely varied conditions of
The suggested stages of group development group composition, duration of group life, and
are highly visible in the literature reviewed. specific group task (i.e., the sequence held up
The fit is not perfect, however. Some of the across settings), it must be assumed that there
studies identify some but not all of the sug- is a finite range of conditions beyond which
gested stages. In some of these cases, two of the sequence of development is altered, and
the suggested stages have been welded into that the studies reviewed did not exceed this
one by the observer. For instance, Barton assumed range to any great extent. Setting-
(1953) describes three stages; the first and specific differences and within-setting differ-
second fit the first two conceptual stages ences may affect temporal change as regards
closely, while Barton's third stage is descrip- the specific content of the stages in the de-
tive of the third and fourth conceptual stages velopmental sequence, the rate of progression
insofar as it is characterized by both the through the sequence, or the order of the se-
emergence of cohesiveness and the working quence itself. In the therapy-group setting,
through of problems. In other cases, one or for instance, task information in the third
more of the hypothesized stages have been stage is considerably more intimate than it is
clearly missing, and thus not recognized in in the laboratory-group setting, and this stage
the group or groups being observed. For may be attained at a later chronological time
instance, Powdermaker and Frank (1948) in therapy groups than in laboratory groups.
identify three stages that fit the first three Certainly duration of group life would be
conceptual stages fairly closely, but they do expected to influence amount and rate of de-
not identify any fourth stage. Perhaps cases velopment. The laboratory groups, such as
like this can be accounted for on the basis of those run for a few hours by Bales and Strodt-
independent variables such as duration of beck (1951), followed essentially the same
group life. course of development as did therapy groups
A few studies identify more than four run for a period of a year. The relatively
stages. Some of these additional stages repre- short life of the laboratory group imposes
sent a greater degree of differentiation than the requirement that the problem-solving
that of the model and are of less generality stage be reached quickly, while no such im-
(i.e., highly specific to the independent condi- position exists for the long-lived therapy
tions of the study). For instance, therapy- group. Consequently, the former groups are
group studies with delinquents and dope ad- forced to develop at a rapid rate. The possi-
dicts identify a stage prior to conceptual bility of such rapid development is aided by
Stage 1 in which the antisocial group mem- the impersonal and concrete nature of the
bers must be won over to the point where laboratory task. Orientation is still required
they will take the therapy seriously. due to the newness of the task but is mini-
Some of the studies identify a stage that is mized by task rules, players' manuals, and the
clearly not in the model. Parker (1958) de- like, that help to orient the group members.
scribes a first stage of cohesive organization. Emotionality and resistance are major features
This divergence from the model may reflect a of therapy-group development and represent
different way of describing much the same personal and interpersonal impediments to
group development and solution attainment as
* A more detailed model of individual development
(similar to the group model proposed here), along a function of the highly emotionally charged
with many citations of supporting literature, may be nature of the therapy-group task. The imper-
found in Harvey, Hunt, and Schroder (1961). sonal laboratory task features no such impedi-
398 BRUCE W. TV CKM AN
FROMM, E. Escape from freedom. New York: Rine- ROTTER, J. B. Social learning and clinical psychology.
hart, 1941. New York: Prentice-Hall, 1954.
GOLEMBIEWSKI, R. T. The small group. Chicago, 111.: SCHINDLER, R. Bifocal group therapy. In J. Masser-
Univer. Chicago Press, 1962. Pp. 193-200. man & J. E. Moreno (Eds.), Progress in psycho-
GROTJAHN, M. The process of maturation in group therapy. Vol. 3. New York: Grune & Stratton,
psychotherapy and in the group therapist. Psychia- 1958. Pp. 176-186.
try, 19SO, 13, 63-67. SCHRODER, H. M., & HARVEY, 0. J. Conceptual or-
HARVEY, 0. J., HUNT, D. E., & SCHRODER, H. M. ganization and group structure. In O. J. Harvey
Conceptual systems and personality organization. (Ed.), Motivation and social interaction. New
New York: Wiley, 1961. York: Ronald Press, 1963. Pp. 134-166.
HEARN, G. The process of group development. Au- SCHUTZ, W. C. FIRO: A three dimensional theory of
tonomous Groups Bulletin, 1957, 13, 1-7. interpersonal behavior. New York: Rinehart, 1958.
HERBERT, ELEONORE, L., & TRIST, E. L. The institu- Pp. 168-188.
tion of an absent leader by a student's discussion SEMRAD, E. V., & ARSENIAN, J. The use of group
group. Human Relations, 1953, 6, 215-248. processes in teaching group dynamics. In W. G.
ILC, FRANCES L., & AMES, LOUISE B. Child behavior. Bennis, K. D. Benne, & R. Chin (Eds.), The plan-
New York: Harper, 1955. ning of change. New York: Holt, Rinehart, &
JENNINGS, HELEN H. Sociometric differentiation of Winston, 1961. Pp. 737-743.
the psychegroup and sociogroup. Sodometry, 1947, SHELLOW, R. S., WARD, J. L., & RUBENFELD, S. Group
10, 71-79. therapy and the institutionalized delinquent. Inter-
KINO, C. H. Activity group therapy with a schiz- national Journal of Group Psychotherapy, 1958, 8,
phrenic boyfollow-up two years later. Interna- 265-275.
tional Journal of Group Psychotherapy, 1959, 9, SMITH, ANTHONY J. A developmental study of group
184-194. processes. Journal of Genetic Psychology, 1960, 97,
LEVY, D. M. Oppositional syndromes and opposi- 29-39.
tional behavior. In P. H. Hoch & J. Zubin (Eds.), STOCK, DOROTHY, & THELEN, H. A. Emotional dy-
Psychopathology of childhood. New York: Grune namics and group culture. Washington, D. C.: Na-
& Stratton, 1955. Pp. 204-226. tional Education Association, National Training
MANN, J. Group therapy with adults. American Laboratories, 1958.
Journal of Orthopsychiatry, 1953, 23, 332-337. STOUTE, A. Implementation of group interpersonal
MANN, J., & SEMRAD, E. V. The use of group therapy relationships through psychotherapy. Journal of
in psychoses. Journal of Social Casework, 1948, Psychology, 1950, 30, 145-156.
29, 176-181. TALLAND, G. A. Task and interaction process: Some
MARTIN, E. A., & Hnx, W. F. Toward a theory of characteristics of therapeutic group discussion.
group development: Six phases of therapy group Journal of Abnormal and Social Psychology, 1955,
development. International Journal of Group Psy- 50, 105-109.
chotherapy, 1957, 7, 20-30. TAYLOR, F. K. The therapeutic factors of group-
MEAD, G. H. Self, mind and society. Chicago: Uni- analytic treatment. Journal of Mental Science,
ver. Chicago Press, 1934. 1950, 96, 976-997.
MILES, M. B. Human relations training: How a THELEN, H., & DICKERMAN, W. Stereotypes and the
group grows. Teachers College Record, 1953, 55, growth of groups. Educational Leadership, 1949,
90-96. 6, 309-316.
MODLIN, H. C., & PARIS, MILDRED. Group adaptation THEODORSON, G. A. Elements in the progressive de-
velopment of small groups. Social Forces, 1953, 31,
and integration in psychiatric team practice. Psy- 311-320.
chiatry, 1956, 19, 97-103.
THORPE, J. J., & SMITH, B. Phases in group develop-
NOYES, A. P. Modern clinical psychiatry. (4th ed.) ment in treatment of drug addicts. International
Philadelphia, Pa.: Saunders, 1953. Pp. 589-591. Journal of Group Psychotherapy, 1953, 3, 66-78.
OSBERG, J. W., & BERLINER, A. K. The developmental TULANE STUDIES IN SOCIAL WELFARE. The use of
stages in group psychotherapy with hospitalized group methods in social welfare settings. New Or-
narcotic addicts. International Journal of Group leans: Tulane Univer. School of Social Work,
Psychotherapy, 1956, 6, 436-447. 1957.
PARKER, S. Leadership patterns in a psychiatric ward. WENDER, L. The dynamics of group psychotherapy
Human Relations, 1958, 11, 287-301. and its application. Journal of Nervous and Men-
PHILP, H., & DUNPHY, D. Developmental trends in tal Disease, 1946, 84, 54-60.
small groups. Sodometry, 1959, 22, 162-174. WHITMAN, R. M. Psychodynamic principles under-
PIAGET, J. The moral judgment of the child. New lying T-group processes. In L. P. Bradford, J. R.
York: Harcourt Brace, 1932. Gibb, & K. Benne (Eds.), T-group theory and
POWDERMAKER, FLORENCE, & FRANK, J. D. Group laboratory methods. New York: Wiley, 1964. Pp.
psychotherapy with neurotics. American Journal 310-335.
of Psychiatry, 1948, 105, 449-455. WOLF, A. The psychoanalysis of groups. American
PSATHAS, G. Phase movement and equilibrium tend- Journal of Psychotherapy, 1949, 3, 16-50.
encies in interaction process in psychotherapy
groups. Sodometry, 1960, 23, 177-194. (Received July 6, 1964)