theoretical research programs 4985

For instance, a distributive justice theory that
originally dealt only with situations involving
quantity goods later expanded to deal with qual
ity goods.
When combined with a few other new ideas,
the core ideas may generate a new theory
designed to explain different phenomena. T2 is
a proliferant of T1 if it expands the core ideas (or
adds new ones) to address a new explanatory
domain or new theoretical problems in the
domain of T1. In this case, both theories are
likely to be viable and important explanatory
tools, since they address different problems or
phenomena. Proliferations significantly expand
the range of sociological problems or phenom
ena to which a programs theories can be
applied. For example, the concept of a source
evaluator was added to expectation states theory
to help explain how significant others affect
actors self evaluations and subsequent group
Sometimes two very closely related theories
are developed from the same core ideas to
develop more precise knowledge about how a
process operates. T1 and T2 are variants of each
other if they address the same or very similar
explanatory domains, but propose slightly dif
ferent mechanisms to account for how the pro
cess operates within that domain. Variants
allow the theorist to consider and resolve small
differences in understanding. Often, the result
is a specification of different conditions under
which each mechanism operates. For example,
variant accounts of bargaining processes sug
gest that the use of threatening tactics either
deters others from engaging in punitive beha
vior or leads to a spiral of conflicting threats
that increases the likelihood of punitive beha
vior. Such variant theories may then be closely
compared to determine which account provided
the more effective explanation.
A superficially similar kind of relation appears
when T1 and T2 use different sets of core ideas to
address the same or overlapping explanatory
domains. In this circumstance the theories are
competitors. However, because of the significant
differences in theoretical structure, it is often
much more difficult to resolve explanatory dif
ferences between competitors than between
variants. For example, one theory for the stabili
zation of mental illness invokes a labeling pro
cess, another focuses on psychophysiological
processes, and these have competed with each
other without full resolution since they were
first articulated in the 1960s.
Finally, ideas in one theory may be combined
with ideas from another theory to provide a
deeper or more complete account of phenomena
or domains that previously were treated sepa
rately. Theory T3 is an integration of T1 and T2
if it consolidates many of the ideas from the
earlier theories, articulating ways in which they
may be related. The manner in which integra
tion is accomplished depends on the nature of
the relation between T1 and T2. If the inte
grated theories were variants, a common out
come is likely to be conditionalization. That is,
T1 is identified as operating under one set of
conditions, T2 under another set. Thus, when
considering deterrence versus conflict spiral
accounts of bargaining behavior, research has
shown that deterrence occurs as long as the
stakes in the bargaining remain relatively low.
As the stakes increase, threat tactics prompt like
responses, thereby generating the conditions for
conflict spiral.
If T1 and T2 are proliferants, the integrating
theory is likely to describe ways in which the
different processes considered by the two earlier
theories are interrelated. Accounts of the two
processes may remain distinct, but connections
between them are specified. For instance,
reward expectations theory partially integrates
status characteristics theory and the status value
theory of distributive justice, the latter two the
ories previously having been proliferants.
Reward expectations theory accomplishes this
by specifying how expectations for task perfor
mance and expectations for reward allocation
may form simultaneously in status situations.
Integrating competitors is most challenging.
In this case the integrating theory usually must
specify a new set of core ideas with which to
describe the phenomena within its domain.
Thus, Guillermina Jasso combined ideas from
both multiple prior theories of distributive jus
tice in developing her own. Key to her integra
tion was the concept of a justice evaluation
function, an idea that did not exist in either
earlier theory.
Much more thorough accounts of the types of
relation that may occur in theoretical research
programs are available in Wagner (1984) and
Wagner and Berger (1985). Those sources also