Bounded Decision Rationality and the Logic of Appropriateness 185
ambiguity. Classic conceptions of decisionmaking assume an objective reality,
an understandable world amenable to description and understanding. Th is is the
positivist assumption that there are knowable patterns of order, from DNA to
the solar system, and that there are also knowable patterns of human behavior,
including how we make decisions. In decision patterns, there is also the assump-
tion of causality, a structure of connections between causes and eff ects, problems
and solutions. Th e job of the decision theorist is to describe this causality. Finally,
decisions are instrumental, choices designed to bring about or cause preferred
states. Th ese three assumptions, objective reality, causality, and intentionality, are
all conditioned by the bounds of rationality. Much of the development of rational
decision theory from the logic of consequences and the logic of appropriateness is
based on these classic conceptions.
But some is not.
Ambiguity is at the center of an alternative understanding of rationality and
of institutional decisionmaking. Ambiguity is a lack of clarity or consistency in
interpretations of reality, causality, and intentionality. Ambiguous situations and
purposes resist categorization and therefore systematic analysis. Ambiguous out-
comes are fuzzy. In the ambiguous decision world, alternatives are hazy, objec-
tives are contradictory, and reality is not so much to be discovered as it is to be
invented.
How shall we understand this decision world?
In the alternative understanding of rationality, the institution is less under-
stood as a world of decisionmaking and better understood as a world of sense
making (Harmon 1989). Consider these diff erences.
In formal decision theories, preferences are assumed to be observable, con-
sistent, stable, and exogenous. In the sense-making perspective, preferences may
or may not be revealed, and, when they are revealed, are contradictory, volatile,
ambivalent, and both exogenous and endogenous. Planning and visioning pro-
cesses are designed to make sense of preferences and seek agreement regarding
preferences, and such processes can do that. But planning and visioning pro-
cesses are also symbols that signal messages to the environment about what the
organization is doing or might do (Cohen and March 1976; Weick 1979; Har-
mon and Mayer 1986). Plans are also advertisements to attract support or invest-
ments, games to test levels of support, and excuses for interaction (Weick 1979).
“In practice, decisionmakers oft en seem to take an active role in constructing and
shaping their preferences. Th ey make decisions by considering their eff ects on
future preferences” (March 1994, 189–190).
In formal decision theories, individual and organizational identities are
assumed to be knowable, rational, and self-interested. In the sense-making
perspective, identities are ambiguous. Identities are multiple and driven by ex-
pectations. An identity, such as that of a police offi cer, is ambiguous because of
imprecise, unstable, inconsistent, and exogenous expectations. How, then, do in-
dividuals make decisions in their institutional roles? Mostly they enact identities