COLLECTIVE BEHAVIORcause and being susceptible to influence by simple
acts. Heirich specifies determinants of the process
by which such common perceptions are created
and the process by which successive redefinitions
of the situation take place. Under organizational
conditions that create unbridged cleavages be-
tween groups that must interact regularly, conflict
escalates through successive encounters in which
cleavages become wider, issues shift, and new
participants join the fray, until the conflict be-
comes focused around the major points of struc-
tural strain in the organization.
Also studying collective conflict as a cumula-
tive process, Bert Useem and Peter Kimball (1989)
developed a sequence of stages for prison riots,
proceeding from pre-riot conditions, to initiation,
expansion, siege, and finally termination. While
they identify disorganization of the governing body
as the key causative factor, they stress that what
happens at any one stage is important in determin-
ing what happens at the next stage.
Clark McPhail (1991), in an extensive critique
of all prior work, rejects the concept of collective
behavior as useless because it denotes too little
and fails to recognize variation and alternation
within assemblages. Instead of studying collective
behavior or crowds, he proposes the study of
temporary gatherings, defined as two or more per-
sons in a common space and time frame. Gather-
ings are analyzed in three stages, namely, assem-
bling, gathering, and dispersing. Rather than
positing an overarching principle such as conta-
gion or norm emergence, this approach uses de-
tailed observation of individual actions and inter-
actions within gatherings and seeks explanations
at this level. Larger events such as campaigns and
large gatherings are to be explained as the ‘‘repeti-
tion and/or combination of individual and collective
sequences of actions.’’(McPhail 1991, p. 221). These
elementary actions consist of simple observable
actions such as clustering, booing, chanting, col-
lective gesticulation, ‘‘locomotion,’’ synchroclapping,
and many others. The approach has been imple-
mented by precise behavioral observation of peo-
ple assembling for demonstrations and other
preplanned gatherings. A promised further work
will help determine how much this approach will
contribute to the understanding of those fairly
frequent events usually encompassed by the term
collective behavior.
MACROLEVEL OR STRUCTURAL
THEORIESMicrolevel theories attempt first to understand
the internal dynamics of collective behavior, then
use that understanding to infer the nature of
conditions in the society most likely to give rise to
collective behavior. In contrast, macrolevel or struc-
tural theories depend primarily on an understand-
ing of the dynamics of society as the basis for
developing propositions concerning when and
where collective behavior will occur. Historically,
most theories of elementary collective behavior
have been microlevel theories, while most theories
of social movements have been structural. Neil
Smelser’s 1993value-added theory is primarily struc-
tural but encompasses the full range from panic
and crazes to social movements.Smelser attempted to integrate major elements
from the Blumer and Turner/Killian tradition of
microtheory into an action and structural theory
derived from the work of Talcott Parsons. Smelser
describes the normal flow of social action as pro-
ceeding from values to norms to mobilization into
social roles and finally to situational facilities. Val-
ues are the more general guides to behavior; norms
specify more precisely how values are to be ap-
plied. Mobilization into roles is organization for
action in terms of the relevant values and norms.
Situational facilities are the means and obstacles
that facilitate and hinder attainment of concrete
goals. The four ‘‘components of social action’’ are
hierarchized in the sense that any redefinition of a
component requires readjustment in the compo-
nents below it, but not necessarily in those above.
Each of the four components in turn has seven
levels of specificity with the same hierarchical
ordering as the components. Types of collective
behavior differ in the level of the action compo-
nents they aim to restructure. Social movements
address either values, in the case of most revolu-
tionary movements, or norms, in the case of most
reform movements. Elementary collective behav-
ior is focused at either the mobilization or the
situational facilities level. Collective behavior is
characterized formally as ‘‘an uninstitutionalized
mobilization for action in order to modify one or
more kinds of strain on the basis of a generalized
reconstitution of a component of action.’’ (Smelser
1963 p. 71). The distinguishing feature of this
action is a shortcircuiting of the normal flow of