The Nature of Sensemakittg

ensemaking is tested to the erctreme when people encounter an event whose

occurrence is so implausible that they hesitate to report it for fear they will
not be believed. In essence, these people think to themselves, it can't be,
therefore, it isn't. Just such an event is the battered child syndrome.
"The battered child syndrome consists of a pattern of injuries (usually to
the head, arms, legs, and ribs) to a child, often a very young one, which the
medical'history' offered by the parents is inadequate to explain. The pattern
of injuries is the result of assaults by parents who then either do not report
the injuries as having occurred or pretend that they are the result of an
accident" (Westrum,l982,p. 3S6). The injuries often can be seen only in X rays,
which explains, in part, why it took so long for this syndrome to be recognized
by the medical community and eventually outlawed by every legislature in the
union.
The battered child syndrome (BCS) was first suggested in 1946 by |ohn
Caffey, a pediatric radiologist, in an article based on six cases where parents
gave 'histories" that were silent about how the injuries, seen in X-ray photo-
graphs, had occurred. Some cases in the article were reported 8 years after they
had first been observed. The author speculated that the accidents may have
been due to parents not fully appreciating the seriousness of the injuries or
 SENSEMAKING IN ORGANIZATIONS

"intentional ill treatment." The article was published in a radiology journal

rather than a pediatric journal, and nothing more happened until the mid-
1950s. Articles appeared in 1953 (3 cases reportedby Silverman), 1955 (12 cases
reported by Wooley and Evans ), and in L957 (again by Caffey), but the medical
profession remained unconcerned about this "professional blind spot"'
Awareness did not change until October 1951 when Frederick Silverman
chaired a panel, 'The Battered Child Syndrome," at the American Academy
of Pediatrics. What made this event significant is that data from a national
survey of T7 district attorneys and 7L hospitals were rePorted, and in this
report 749 cases were identified. The results and an editorial were then
pubtished in the lournal of the American Medical 'Lssociation under the title
"The Battered-Child Syndrome."
Public reaction was prompt, and within a few years, laws in all 50 states
required that suspected cases of BCS had to be reported.By 1967,when better
reporting channels had been established, it was estimated that there were 7,000
cases. This estimate climbed to 60,000 by L972 andto 500,000 by 1976
(Westrum,
1982,p.392).
What makes this an instance of sensemaking? First, someone notices some-
thing, in an ongoing flow of events, something in the form of a surprise, a
discrepant set of cues, something that does not fit. Second, the discrepant cues
are spotted when someone looks back over elapsed experience. The act of
looking is retrospective. Third, plausible speculations (e.g., parents fail to
realize severity of injuries) are offered to explain the cues and their relative
rarity. Fourth, the person making the speculations publishes them in a tangi-
ble journal article that becomes part of the environment of the medical
community for others. He or she creates an object that was not "out there" to
begin with but now is there for the noticing. Fifth, the speculations do not
generate widespread attention right away because, as Westrum noted, the
observations originated with radiologists who have infrequent social contact
with pediatricians and families of children. Such contacts are crucial in the
construction and perception of problems. And sixth, this example is about
sensemaking because issues of identity and reputation are involved. fu Westrum
puts it, passive social intelligence about hidden events is often slow to develop
because there are barriers to reporting the events. Experts overestimate the
likelihood that they would surely know about the phenomenon if it actually
were taking place. He calls this "the fallacy of centrality': because I don't know
about this event, it must not be going on. As Westrum (1932) puts it, "this
fallacy is all the more damaging in that it not only discourages curiosity on
the part of the person making it but also frequently creates in him/her an
The Nature of Sensemaking

antagonistic stance toward the events in question. One might well argue that
part of the resistance of pediatricians to a diagnosis of parent-caused trauma
was an inability to believe that their own evaluation of parents'dangerousness
could be seriously in error" (p. 393). Thus BCS is an instance of sensemaking
because it involves identiry retrospect, enactment, socid contact, ongoing
events, cues, and plausibility, seven properties that will be explored further in
Chapter 2.
There remains the question, what makes these events organizational sense-
making? Although a fuller answer will begin to emerge starting with Chapter
3, its rough outline can be suggested. The setting in which the BCS syndrome
was discovered is organizational in several ways. Pediatricians and radiolo-
gists, working through interlocking routines that are tied together in relatively
formal "nets of collective action" (Czarniawska-|oerges,l992,p.3z),perform
specialized tasks intended to preserve the health of children. Medical person-
nel have shared understandings of their roles, expertise, and stature, but they
also act as shifting coalitions of interest groups. The prevalence of routines,
generic understandings, and roles enables personnel to be interchanged.
Although all of this organizing facilitates coordinated action, it also imposes
an "invisible hand" on sensemaking. This was clear in Westrum's fallacy of
centrality,which is adirectby-product of nets of collective action.If we extend
Westrum's observation, it is conceivable that heavily networked organizations
might find their dense connections an unexpected liability, if this density
encourages the fallacyof centrality. "News"mightbe discounted if people hear
it late and conclude that it is not credible because, if it were, they would have
heard it sooner. This dynamic bears watching because it suggests a means by
which perceptions of information technology might undermine the ability of
that technology to facilitate sensemakittg.The more advanced the technology
is thought to be, the more likelyare people to discredit anything that does not
come through it. Because of the fallacy of centrality, the better the information
system, the less sensitive it is to novel events.
Organizations stay tied together by means of controls in the form of incentives
and measures. This suggests that incentives for reporting anomalies, or pen-
alties for nonreporting, should affect sensemaking. More frequent reporting
of what Westrum (1982) calls 'uncorrected observations and experience"
(p. 38a) should intensify ambiguity in the short run, until others begin to
report similar experiences. As anomalies become shared, sensibleness should
become stronger.
Organizations also have their own languages and symbols thathave impor-
tant effects on sensemaking. The relevance of that to the BCS example is the
 SENSEMAKING IN ORGANIZATIONS

striking difference between the phrase "intentional ill treatment" and the
phrase "battered child." The latter phrase evokes a graphic picture of parents
beating and killing their children. That image can mobilize outrage and action.
The more general point is that vivid words draw attention to new possibilities
(Pondy, Ig78),suggesting that organizations with access to more varied images
will engage in sensemaking that is more adaptive than will orgaflizations with
more limited vocabularies.
BCS has elements of both sensemaking in general and organizational sense-
making. I turn now to a fuller investigation of each.
I

The Concept of Sensemaking

I
The concept of sensemaking is well named because, literally, it means the
making of sense. Active agents construct sensible, sensable (Huber & Daft,
Lg87,p. 15a) events. They "structure the unknown" (Waterman, 1990, P. 41).
How they construct what they construct, why, and with what effects are the
central questions for people interested in sensemaking.Investigators who study
sensemaking define it in quite different ways.Many investigators (e.9., Dunbar,
1981; Goleman, 1985, pp.l97-217) implywhat Starbuck and Milliken (1988)
make explicit, namely, that sensemaking involves placing stimuli into some
kind of framework (p. 51). The well-known phrase "frame of reference" has
traditionally meant a generalized point of view that directs interpretations
(Cantril, !g4l,p. 20).When people put stimuli into frameworks, this enables
them "to comprehend, understand, explain, attribute, extrapolate, and pre-
dict" (Starbuck & Milliken, 1988, P. 5l). For example, people use strategy as
a framework that "involves procurement, production, synthesis, manipula-
tion, and diffusion of information in such awaY as to give meaning, PurPose
and direction to the organization" (Westley, 1990, p.337).
A related conceptualization, grounded in newcomer socialization rather
than in strategy, is found in the work of Meryl Louis (1980). She views sense-
making as a thinking process that uses retrospective accounts to explain
surprises. "Sense making can be viewed as a recurring cycle comprised of a
sequence of events occurring over time. The cycle begins as individuals form
unconscious and conscious anticipations and assumptions, which serve as
predictions about future events. Subsequently, individuals experience events
that maybe discrepant from predictions. Discrepant events, or surprises, trigger
a need for explanation, or post-diction, and, correspondingly, for a process
through which interpretations of discrepancies are developed. Interpretation,
The Nature of Sensemaking

or meaning, is attributed to surprises. . . . It is crucial to note that meaning is

assigned to surprise as an output of the sense-making process, rather than
arising concurrently with the perception or detection of differences" (Louis,
1980, p.2al).
Louis suggests that the activity of placing stimuli into frameworks is most
visible when predictions break down, which suggests that sensemaking is
partially under the control of expectations. Whenever an expectation is dis-
confirmed, some kind of ongoing activity is interrupted. Thus to understand
sensemaking is also to understand how people cope with interruptions. The
joint influence of expectations and interruptions suggests that sensemaking
will be more or less of an issue in organizations, depending on the adequacy
of the scripts, routines, and recipes already in place. For example, an organi-
zation that expects change may find itself puzzled when something does not.
The activities of sensemaking mentioned by Starbuck, Milliken, Westley,
and Louis focus on the placement of stimuli into frameworks, but other investi-
gators include more activities than simplythose of placement. Thomas, Clark,
and Gioia (1993), for example, describe sensemaking as "the reciprocal inter-
action of information seeking, meaning ascription, and action" (p. 240),which
means that environmental scanning, interpretation, and "associated responses"
all are included. Sackman (1991) talks about sensemaking mechanisms that
organizational members use to attribute meaning to events, mechanisms that
"include the standards and rules for perceiving, interpreting, believing, and
acting that are typically used in a given cultural setting" (p. 33). Feldman
(1989) talks about sensemaking as an interpretive process that is necessary
"for organizational members to understand and to share understandings about
such features of the organization as what it is about, what it does well and
poorly, what the problems it faces are, and how it should resolve them" (p. 19).
Whereas both Thomas et al. and Sackman mention "action" in conjunction
with sensemaking, Feldman (1989) insists that sensemaking often

does not result in action. It may result in an understanding that action

should not be taken or that a better understanding of the event or situation
is needed. It may simply result in members of the organization having more
and different information about the ambiguous issue. (p. 20)

Some investigators (e.g., Gioia & Chittipeddi, 1991, p.aa$ view sensemak-
ing as a more private, more singular activity. Ring and Rands ( 1989), for example,
define sensemaking as "a process in which individuals develop cognitive maPs
of their environment' (p. 342). Having made sensemaking an individual
 SENSEMAKING IN ORGANIZATIONS

activity, they use the term understandingto refer to mutual activity' a distinc-
tion that is clearly easier to ProPose than to implement:

We decided thatwhenever the written material or resPonses from individu-

als reflected an intention on their part to simply enhance their own per-
spective on a subject, then such actions were indicative of a sensemaking
pro..rr. ...On the other hand, when these kinds of activities were pursued
in activities that reflected reciprociry we classified them as understanding.
This is, of course, the grey area. The same activity may reflect, at once,
sensemaking and understanding processes. (p. 3aa)

Sensemaking is grounded in both individual and social activity' and whether

the two are even separable will be a recurrent issue in this book, because
it has
been a durable tension in the human condition.Witness this description
from
Emily Dickinson:

Much Madness is divinest Sense-

To a discerning EYe-
Much Sense-the starkest Madness-
'Tis the MaioritY
In this, as All, prevail-
Assent-and You are sane-
Demur-you're straigbtaway dangerour
And handled with a Chain-
(cited in Mailloux, 1990' P. 126)

Sense may be in the eye of the beholder, but beholders vote and the majority
rules.

The Uniqueness of Sensem"kittg

So farI have argued that sensemaking is about such things as placement of

items into frameworks, comprehending, redressing surprise, constructing
meaning, interacting in pursuit of mutual understanding, and patterning. I
can sharpen this picture by suggesting what sensemaking is not' To do so,
I
contrast sensemaking with interpretation because interpretation is often used
as a synonym for sensemaking. Such synonymous usage is not a blunder,
but
it does blur some distinctions that seem crucial if one wishes to understand
the subtleties of sensemaking in organized settings. The activity of interpre-
ThcNetureof Sasomking

tation is discussed just often in law (e.g., White, 1990) or the humanities
as
(e.g., Co[ini, Lgg2) as it is in the social sciences (e.g., Rabinow & Sullivan,
tggZ),which suggests that sensemaking, of which interpretation is a comPo-
nent, has widespread applicability. Most descriptions of interpretation focus
on some kind of text. What sensemaking does is address how the text is
constructed as well as how it is read. Sensemaking is about authoring as well
as reading.
To appreciate this difference, consider some characteristics of interpreta-
tion. In Webster's DiAionary of Synonyms (1951), interpretation is described
as a form of explanation that "requires special knowledge, imagination, sym-
pathy, or the like" in the person who would tty to understand some text that
;presents more than intellectual difficulties as in a
Poem, a dream" (p. 318)'
A more compact definition of interpretation is Mailloux's ( 1990) statement
that interpretation is "acceptable and approximating translation'(p. 121). An
"acceptable'reading is one that has some stature in a community. An "aP-
proximating" reading is one that attempts to caPture something, such as an
intention, that is presumed to be "there." And "translation" is an activity such
as historicizing,allegorizing, or punning that gives form to the approximation.
In short, interpretation literally means a rendering in which one word is
explained by another.
When interpretation is equated with translation, the interpretation points
in two directions simultaneously.It points toward a text to be interpreted, and
it points toward an audience presumed to be in need of the interpretation.
The interpreter mediates between these two sites. However, this mediation is
not without a context, which means that an interpretation is never a "private"
reading. Instead, any reading assumes some status "within the power relations
of a historical community" (Mailloux, 1990, p. 127), meaning that most
interpretations involve political interests, consequences, coercion, persuasion,
and rhetoric.
When interpretation is incorporated into organizational studies, (e.g.
Jeffcutt, 1994), it is often invoked because ambiguity and equivocation are
seen as prominent accompaniments of organizational action (e.g., Chaffee,
1 985; Huber & Daft , I 9S7 ). For example, March
and Olsen (197 6) observe that

most of what we believe we know about elements within organizational

choice situations, as well as the events themselves, reflects an interpretation
of events by organizational actors and observers. Those interpretations are
generated within the organization in the face of considerable perceptual
ambiguity. (p. 19)
 SENSEMAKING IN ORGANIZATIONS

It is the very pervasiveness of this ambiguity and of the strong discomfort

people feel when they face it that leads March (1984, p. lE) to argue that
organizational life is as much about interpretation, intellect, metaphors of
theory, and fitting our history into an understanding of life as it is about
decisions and coping with the environment.
What an interpretive reading consists of is summarized in the introduction
to Porac, Thomas, and Baden-Fuller's ( 1989) study of 17 firms manufacturing
high-qualitycashmere sweaters in theborder region of Scotland. Theyground
an interpretive study in four assumptions:

Activities and structures of organizations are determined in part by micro-

momentary actions of their members.
Action is based on a sequence in which 'individuals attend to cues in the
environment, interpret the meaning of such cues, and then externalize these
interpretations via concrete activities.'
3. Meaning is created when cues are linked with "well-learned and/or developing
cognitive structures."
4. People can verbalize their interpretations and the processes they use to generate
them.

With these materials as background, I can now say more about the unique-
ness of a sensemaking perspective. Porac et al.'s (1989) four assumptions
about the nature of an interpretive study focus on attending to cues and
interpreting, externaluing,and linking these cues. What is left unspecified are
how the cues got there in the first place and how these particular cues were
singled out from an ongoing flow of experience. Also unspecified are how the
interpretations and meanings of these cues were then altered and made more
explicit and sensible, as a result of "concrete activities." The process of sense-
making is intended to include the construction and bracketing of the textlike
cues that are interpreted, as well as the revision of those interpretations based
on action and its consequences. Sensemaking is about authoring as well as
interpretation, creation as well as discovery. As we will see later, even though
Porac et al. view their work as an example of an interpretive studS they
actually address all aspects of the sensemaking process.
Clear descriptions of the nature of sensemaking that pry it apart from
interpretation are found in the work of Schtin (1983b), Shotter (1993), and
Thayer ( 1988). Schdn is especially helpful when he discusses problem setting
as a key component of professional work
The N ature of Sensemaking

In real-world practice, problems do not present themselves to the practi-

tioners as givens. Theymustbe constructed from the materials of problem-
atic situations which are puzzling, troubling, and uncertain. In order to
convert a problematic situation to a problem, a practitioner must do a
certain kind of work. He must make sense of an uncertain situation that
initially makes no sense. When professionals consider what road to build,
for example, they deal usually with a complex and ill-defined situation in
which geographic, topological, financial, economic, and political issues are
all mixed up together. Once they have somehow decided what road to build
and go on to consider how best to build it, they may have a problem they
can solve by the application of available techniques, but when the road they
have built leads unexpectedly to the destruction of a neighborhood, they
may find themselves again in a situation of uncertainty.
It is this sort of situation that professionals are coming increasingly to
see as central to their practice. They are coming to recognize that although
problem setting is a necessary condition for technical problem solving, it is
not itself a technical problem. When we set the problem, we select what we
will treat as the "things' of the situation, we set the boundaries of our
attention to it, and we impose upon it a coherence which allows us to say
what is wrong and in what directions the situation needs to be changed.
Problem setting is a process in which, interactively, we name the things to
which we will attend and frame the context in which we will attend to them.
(Sch6n, 1983b, p.40)

Shotter (1993) likens managing to authoring a conversation and describes

the manager's task as

not one of choosing but of generating, of generating a clear and adequate

formulation of what the problem situation 'is," of creating from a set of
incoherent and disorderly events a coherent "structure" within which both
current actualities and further possibilities can be given an intelligible
"place"-xnd of doing all this, not alone, but in continual conversation with
all the others who are involved. . . . To be justified in their authoring, the
good manager must give a sharable linguistic formulation to already shared
feelings, arising out of shared circumstances-and that is perhaps best done
through the use of metaphors rather than by reference to any already
existing theories. (pp. 150, 152)

Thayer (1988) pulls these strands together in a remarkable analysis of

leadership, the crux of which is the idea that a leader is
IO SENSEMAKING IN ORGANIZATIONS

*mind" the
one who alters or guides the manner in which his followers
world by giving it a compelling "face." A leader at work is one who gives
others a different sense of the meaningof that which they do by recreating
it in a different form, a different 'face,o in the same way that a pivotal painter
or sculptor or poet gives those who follow him (or her) a different way of
"seeing"-and therefore saying and doing and knowing in the world. A
leader does not tell it "as it is"; he tells it as it might be, giving what 'is"
thereby a different
*facel. . . The leader is a sense- giver. The leader always
embodiuthe possibilities of escape from what might otherwise appear to
us to be incomprehensible, or from what might otherwise aPPear to us to
be a chaotic, indifferen! or incorrigible world-one over which we have no
ultimate control. (pp. 250, 254)

Although each of these descriptions begins to Pry apart sensemaking and

interpretation, I want to supplement them with a more personal example
based on how I first got interested in sensemaking.My fascination with this
topic dates back to conversations in the early 1960s with Harold Garfinkel and
Harold Pepinsky. The context was Garfinkel's study of decision making in
juries (published in Garfinkel,L967,pp. 104-I I5; seeMaynard &Manzo,L993,
for an updating of Garfinkel's study). What I found intriguing was Garfinkel's
insistence that jurors did not seem to first decide the harm and its extent, and
then allocate blame, and then finally choose a remedy. Instead, they first
decided a remedy and then decided the "facts,'from among alternative claims,
that justified the remedy. furors essentially created a sequence that was mean-
ingfully consistent and then treated it as if it were the thing that actually
occurred. "If the interpretation makes good sense, then that's what happened"
(Garfinkel, L967,p. 105).
Facts were made sensible retrospectively to support the jurors' choice of
verdict. Garfinkel (1967) summarized decision making in common sense
situations of choice this way:

In place of the view that decisions are made the occasions require, an
as
alternative formulation needs to be entertained.It consists of the possibility
that the person defines retrospectively the decisions that have been made.
The outcome comes before the decision. In the material reported here, jurors
did not actually have an understanding of the conditions that defined a
correct decision until after the decision had been made. Only in retrospect
did they decide what they did that made their decisions correct ones. When
the outcome was in hand theywent back to find the "why," the things that
led up to the outcome. . . . If the above description is accurate, decision
The Nature of Sensemaking

making in daily life would thereby have, as a critical feature, the decision
maker's task of justtfying a course of action.. . . [Decision making in daily
life] may be much more preoccupied with the problem of assigning out-
comes their legitimate history than with questions of deciding before the
actual occasion of choice the conditions under which one, among a set of
alternative possible courses of action, will be elected. (pp. I 14-115)

A crucial property of sensemaking is that human situations are progres-

sively clarified, but this clarification often works in reverse. It is less often the
case that an outcome fulfills some prior definition of the situation, and more
often the case that an outcome develops that prior definition. As Garfinkel
(1967) puts it, actors "in the course of a career of actions, discover the nature
of the situations in which they are acting. . . . [T]he actor's own actions are
first order determinants of the sense that situations have, in which, literally
speaking, actors findthemselves" (p. 115).
A similar emphasis on the idea that outcomes develop prior definitions of
the situation is found in cognitive dissonance theory (Festinge6 1957). Dis-
sonance theory focuses on postdecisional efforts to revise the meaning of
decisions that have negative consequences (Cooper & Fazio, 1984; Scher &
Cooper, 1989; Thibodeau & Aronson, 1992).If' for example, people choose
between alternatives with nonoverlapping attractions, they forgo the attrac-
tions of the nonchosen alternatives and gain the negative features of the
chosen alternative. After making such a choice, people may feel anxious and
agitated (dissonance). To reduce dissonance, people "spread" the alternatives
by enhancing the positive features of the chosen alternative and the negative
features of the unchosen alternatives. These operations retrospectively alter
the meaning of the decision, the nature of the dternatives, and the "history"
of the decision in a manner reminiscent of Garfinkel's jurors. In both cases,
people start with an outcome in hand-a verdict, a choice-and then render
that outcome sensible by constructing a plausible story that produced it (in
Garfinkel's words, "the interpretation makes good sense").
A considerable body of work in organizational studies shows the legacy of
cognitive dissonance, including the ideas of enactment (Abolafia & Kilduff,
1988; Weich Lg77),commitment (O'Reilly & Caldwell, l98l; Salancik, L977),
rationality and rationalization (Staw, 1980), escalation (Staw, 1981), attribu-
tion (Calder, 1977; Staw, L975), justification (Staw, McKechnie, & Puffer,
1983), and motivation (Staw,l977).What is shared by these diverse ideas is a
common set of emphases that can be traced back to dissonance theory. These
include the following:
12 SENSEMAKING IN ORGANIZATIONS

Sensemaking by justification, an idea that reflects an earlier emphasis on disso-

nance reduction by increasing the number of cognitive elements that are con-
sistent with the decision;
Choice as the event that focuses sensemaking and justification, an idea that
retains the emphasis on postdecision behavior;
3. Sensemaking by retrosPect, an idea that retains dissonance theory's emphasis
that postdecision outcomes are used to reconstruct predecisional histories;
4. Discrepancy as the occasion for sensemaking, an idea that restates dissonance
theory's starting point, namely, action that follows from the obverse of cogni-
tions held by the actor;
5。 Social construction of justification, an idea that reflects dissonance reduction
by means of social support and proselytizing;
Action shapes cognition, an idea that is a composite of Items 2,3,tnd4 above.

All six of these strands can be found in dissonance theory in more recent
ideas such as commitment, escalation, and enactment, and there are hints of
these strands in ethnomethgdological accounts of decision making in every-
day life (e.g., Handel, 1982; Heap, 1975; Gephart, 1993). Most important for
our purposes, all six are important in any account of sensemaking.
To see this, think about the wonderfully compact account of sensem"kittg
mentioned by Graham Wallas. "The little girl had the making of a poet in her
who, being told to be sure of her meaning before she spoke, said: 'How can I
know what I think till I see what I say?' " (Wallas, L926, p. 106). This recipe,
which is central in organizational sensemaking (Weick,l979,p. 133), retains
several elements of dissonance theory. The recipe is about justification (my
thoughts justify my earlier words), choice (I choose which words to focus on
and which thoughts will ocplain them), retrospective sensemaking (I lookback
at what I said earlier from a later point in time when the talking has stopped),
discrepancies (I feel a need to see what I say when something doesn't make
sense), social construction of justification (I invoke the thoughts I have been
socialized to label as acceptable), and action as the occasion for sensemaking
(my act of speaking starts the sensemaking process).
Sensemaking to social psychologists meant making sense of actions that did
not follow from betiefs and self-concepts, whereas to ethnomethodologists it
meant reasoning in ways that differed from those rational practices associated
with scientific thinking. Sensemaking, because it was influenced by disso-
nance theory, also meant a focus on conflict, affect, motivation, and instability
as antecedents of change, rather than the current, more austere focus in cognitive
studies on cool formation processing (Markus &Zaionc,l985' p.207).
The Nature of Sensemaking 13

What makes current thinking about sensemaking robust is that both ethno-
methodol ogy (Czarniawska-|oerges, 1992, chap. 5 ; Gephart, I 993 ) and disso-
nance theory (Chatman, Bell, & Staw, 1985; Weick, 1993a) still inform some
of the core ideas. Furthermore, both perspectives share common ideas. The
emphasis in ethnomethodology on accounting for what one does in the
presence of other people to prove social competence and the rationality of
actions is very much like the self-justification of dissonance theory, which is
also directed at real or imagined auditors. What is unusual about the topic of
sensemaking is that it is grounded as much in deductions from well-articulated
theories as it is in inductions from specific cases of struggles to reduce
ambiguity. This is a decided advantage for investigators because there is a core
set of ideas that holds this perspective together and has held it together for
some time. One purpose of this book is to make those ideas explicit.
Although the next chapter will describe important characteristics of sense-
making in more detail,I can now at least summarize how sensemaking differs
from interpretation, with which it is often confused. The key distinction is
that sensemaking is about the ways people generate what they interpret. |ury
deliberations, for example, result in a verdict. Once jurors have that verdict in
hand, they look back to construct a plausible account of how they got there.
During their deliberations they do the same thing, albeit in miniature. Deliber-
ating primarily develops the meaning of prior deliberating rather than sub-
sequent deliberating.Iurors literally deliberate to discover what they are talking
about and what constitutes evidence. They look for meaningful consistencies
in what has been said, and then revise those consistencies. Authoring and
interpretation are interwoven. The concept of sensemakinghighlights the action,
activity, and creating that lays down the traces that are interpreted and then
reinterpreted.
Sensemaking, therefore, differs from interpretation in ways such as these.
Sensemaking is clearly about an activity or a process, whereas interpretation
can be a process but is just as likely to describe a product. It is common to hear
that someone made "an interpretation." But we seldom hear that someone
made "a sensemaking."We hear, instead, that people make sense of something,
but even then, the activity rather than the outcome is in the foreground. A
focus on sensemaking induces a mindset to focus on Process, whereas this is
less true with interpretation.
Even when interpretation is treated as a Process, the implied nature of the
process is different. The act of interpreting implies that something is there, a
iext in the world, waiting to be discovered or approximated (see Daft & Weick,
1984). Sensemaking, however, is less about discovery than it is about invention.
14 SENSEMAKING IN ORGANIZATIONS

To engage in sensem"king is to construct, filter, frame, create facticity (Turner,

L987),and render the subjective into something more tangible.
The contrast between discovery and invention is implicit in the word sense.
To sense something sounds like an act of discovery. But to sense something,
there must be something there to create the sensation. And sensemaking
suggests the construction of that which then becomes sensible. Sensemaking
might even be described as an ongoing effort to create aworld in which object
perception, rather than interpersonal perception, would be more appropriate
(Swann, 1984), dthough it never succeeds in doing so. As Morgan, Frost, and
Pondy ( 1983) put it, "Individuals are not seen as living in, and acting out their
lives in relation fo, a wider reality, so much as creating and sustaining images
of a wider realiry in part to rationalize what they are doing. They realize their
reality by 'reading into'their situation patterns of significant meanin g" (p.24).
Thus, the concept of sensemaking is valuable because it highlights the
invention that precedes interpretation. It is also valuable because it implies a
higher level of engagement by the actor. Interpretation connotes an activity
that is more detached and passive than the activityof sensemaking. Sensemak-
ing matters. A failure in sensemaking is consequential as well as existentid.It
throws into question the nature of self and the world. As Frost and Morgan
(1983) suggest, when people make sense of things, they "read into things the
meanings they wish to see; they vest objects, utterances, actions and so forth
with subjective meaning which helps make their world intelligible to them-
selves" (p.207\.The stakes are seldom as high when interpretations fail. Interpre-
tations can be added and dropped with less effect on one's self-perceptions,
which is not true of efforts to replace one sense of the world with another.
And whenever sense is lost, the loss is deeply troubling (e.9., Asch, L952;
Garfinkel, 1963; Milgram, 1963), whereas the loss of an interpretation is more
like a nuisance.
It is also important to separate sensemaking from interpretation because
sensemaking seems to address incipient puzzles at an earlier, more tentative
stage than does interpretation. When people discuss interpretation, it is
usually assumed that an interpretation is necessary and that the object to be
interpreted is evident. No such presumptions are implied by sensemaking.
Instead, sensemaking begins with the basic question, is it still possible to take
things for granted? And if the answer is no, if it has become impossible to
continue with automatic information processing, then the question becomes,
why is this so? And, what next? Several questions arise and have to be dealt
with before interpretation even comes into play. The way these earlier ques-
The Nature of Sensemaking 15

tions of sensemaking are resolved determines which interpretations are pos-

sible and plausible.
The early emergence of sensemaking is also what sets it apart from decision
making, as Drucker O97D made clear:

The Westerner and the fapanese man mean something different when they
talk of "making a decision." In the West, all the emphasis is on the answer
to the question. Indeed, our books on decision making try to develop system-
atic approaches to giving an answer. Tio the fapanese, however, the impor-
tant element in decision makingis definingthe question. The important and
crucial steps are to decide whether there is a need for a decision and what
the decision is about. And it is in that step that the fapanese aim at attaining
consensus. Indeed, it is this step that, to the fapanese, is the essence of
decision. The answer to the question (what the West considers the decision)
follows from its definition. During the process that precedes the decision,
no mention is made of what the answer might be. . . . Thus the whole
process is focused on finding out what the decision is really about, not what
the decision should be. (pp. 466-467)

To talk about sensemaking is to talk about reality as an ongoing accomplish-

ment that takes form when people make retrospective sense of the situations
in which they find themselves and their creations. There is a strong reflexive
quality to this processes. People make sense of things by seeing a world on'
which they already imposed what they believe. People discover their own
inventions, which is why sensemaking understood as invention, and interpre-
tation understood as discovery, can be complementary ideas. If sensemaking
is viewed as an act of invention, then it is also possible to argue that the artifacts
it produces include language games and texts.
But to argue that the bulk of organizational life is captured by the metaphor
of reading texts is to ignore most of the living that goes into that life. I agree
with Czarniawska-Joerges's (L992,pp.253-254) assessment that the text meta-
phor represents the activity of social construction as a static result, implies
that meaning already exists and is waiting to be found rather than that it awaits
construction that might not happen or might go awry and suggests a unity
that is untenable when there are subuniverses of meaning. "Organizations are
not texts, but a text is a common form of interpretation that we deal with"
( Czarniawska-f oerge s, 1992, p. 123).

Finally, what sensemaking is nor is a metaphor. I say this because Morgan

et al. (1983) describe sensemaking as one of three metaphors (the other two
16 SENSEMAKING IN ORGANIZAT10NS

people who favor an interpretive

are language game and text) that are used by
approach to organizational studies. They argue that all three positions are
..concerned with understanding the genesis of meaningful action, how indi-
thus come to define and share
viduals make sense [sic] of their situations, and
realities which may become objectified in
fairly routinized ways' In short, to
aspects of everyday life are
understand how the objective, taken for granted
process" (Morgan
constituted and made real through th. medium of symbolic
et al. 1983,P.22).
Although texts and language games are metaphors for
interpretation' sense-
making something
making is not. Sensemaking is what it says it is, namely,
sensible. Sensemaking is to be understood literally,
not metaphoricdly' Notice
describe the
that Morgan et al. inadvertently acknowledge this when they
"metaphor' of sensemaking as "how individuals make sense of their
situ-
ations." This error of logical WPing (Bateson, 1972) can be avoided
if sense-
making is separated from the class of interpretive activities it names and
set

above this class as a higher level abstraction that includes them.

Although the
sensemakingmayhave an informal, poetic flavor, that should not
mask
word
the fact that it is literally just what it says it is'
 of Sensemaking

he descriptions of sensem"kittg reviewed so far imply at least seven

distinguishing characteristics that set sensemaking apart from other ex-
planatory processes such as understanding, interpretation, and attribution.
Sensemaking is understood as a process that is

l. Grounded in identity construction

2. Retrospective
3. Enactive of sensible environments
4. Social
5. Ongoing
6. Focused on and by octracted cues
7. Driven by plausibility rather than accuracy

These seven characteristics are described and then (p.76 in Chapter 3) applied
to an important study of organizational sensemaking (Porac et al. 1989). These
seven were chosen to organize the discussion because they are mentioned
often in the literature on sensemaking; they have practical implications (e.g.,

17
18 SENSEMAKING IN ORGANIZATIONS

when identities are destabilized during downsizing' sensemakittg Processes

are threatened and these threats can enlarge); each
is a self-contained set of
research questions that relates to the other six; each
incorporates action and
can be rePre-
context, which are key aspects of sensemaking; and all seven
sented crudely as a sequence (people concerned with
identity in the context
and make plausible
of others engage ongoini events from whic.h theyutract cues
sense retrospectively,allthewhile enacting more
or less order into those ongoing
simultaneous
events). This sequence is crudebecause it omits feedbacklooPs'
processing, and the fact that over time, some stePs may drop out.
These seven characteristics serve as a rough guideline for
inquiry into
is' how itworks'
sensemaking in the sense that they suggestwhat sensemaking
and where it can fail. This listing is more like an obseryer's
manual or a set of
raw materials for disciplined imagination (weick' 1989) than
it is a tacit set
serve the
of propositions to be refined and tested. The listing might eventually
l"tt., porpor., but that is not our intention in developing it here' Instead' I
simply want to put some boundaries around the phenomenon
of sensemak-
ing. nach of the seven characteristics will be discussed briefly.
My intention is
chap-
tolntroduce ideas about sensemaking thatwill reappear in subsequent
ters. Having read this initial chapter, readers should begin
to notice subtleties
and patterns in their own efforts to malce sense. As readers use their
own
experiences to anchor these ideas, they should spot more
data, and more
significant data, to refine the structure Presented here.

l. Grounded in Identity Construction

Sensemaking begins with a sensemaker. "How can I know

what I think until
I see what I say?" has four Pronouns, all four of which point to the Person
a trap'
doing the sensem"kittg.Obvious as that assertion may seem, it contains
acts like a single
The trap is that sensemakeris singular and no individual ever
"a parliament
sensemaker. Instead, any one sensemaker is, in Mead's words,
in
of selves." Nowhere is this truth about human beings better portrayed than
Pablo Neruda's (1968) Poem "We Are Many'"

Of the many men who I am, who we are,

I can't find a single one,
they disapPear among mY clothes,
they've left for another citY.
Sarcn Propertiu of Sensemaking 19

When everphing seems to be set

to show me off as intelligent,
the fool I always keep hidden
takes over all that I say.

At other times,I'm asleep

among distinguished people,
and when I look for mybrave sel[,
a coward unknown to me
rushes to cover my skeleton
with a thousand fine excuses.

When a decent house catches fire,

instead of the fireman I summon,
an arsonist bursts on the scene,
and that's me. What I can do?
What can I do to distinguish mysel8
How can I pull myself together?

All the books I read

are full of dazzling heroes,
dways sure of themselves.
I die with envT of them;
and in films full of wind and bullets,
I goggle at the cowboys,
I even admire the horses.

But when I call for a hero,

out comes mylazy old self;
and so I never know who I am,
nor how many I am or will be.
I'd love to be able to touch a bell
and summon the real me,
because if I really need myself,
I musn't disaPPear.

While I am writing, I'm far away;

and when I come back,I've gone.
I would like to know if others
go through the same things that I do,
20 SENSEMAKING IN ORGANIZATIONS

have as many selves as I have,

and see themselves similarlY;
and when I've extrausted this problem,
I'm going to study so hard
that when I explain mYself,
I'll be talking geograPhY.
Pablo Neruda
From EXIRAVAGARIA, translated by Alastair Reid'
Reprinted by permission of Farrar, Straus, & Giroux

A more prosaic way to say the same thing is to assert that "the" individual
"is a typified discursive construction' (Knorr-Cetina, 1981, P. l0). Identities
are constituted out of the process of interaction. To shift among interactions
is to shift among definitions of self. Thus the sensemaker is himself or herself
an ongoingpuzile undergoing continual redefinition, coincident with pre-
senting some self to others and trying to decide which self is appropriate.
Depending on who I am, my definition of what is "out there" will also change.
Whenever I define self, I define "it," but to define it is dso to define self. Once
I know who I am then I know what is out there. But the direction of causality
flows just as often from the situation to a definition of self as it does the other
way. And this is why the establishment and maintenance of identity is a core
preoccupation in sensemaking and why we place it first on our list.
Erez and Earley (1993), in their presentation of cultural self-representation
theory, view the self-which is represented by all statements that include the
words I, me, mine, and myselfas a socially situated "dynamic interpretive
structure that mediates most significant intrapersonal and interpersonal
processes" (p.26).They argue further that self-concept is to a large extent an
agent of its own creation. The processes that develop and maintain a person's
changing sense of self are positedto operate in the service of three self-derived
needs:

(l) the need for self-enhancement, reflected in seeking and maintaining

as
a positive cognitive and affective state about the self; (2) the self-efficacy
motive, which is the desire to perceive oneself as competent and efficacious;
and (3) the need for self-consistency, which is the desire to sense and
experience coherence and continuity. (p.28)

It is the ongoing fate of these needs that affects individual sensemaking in

organizations. This relationship is beautifully documented in Dutton and
Dukerich's (1991) study of the ways in which the New York Port Authority
attempted to deal with the issue of a growing number of homeless people
Seven Properties of Sensemaking 21

occupying its facilities. The Port Authority, whose identity in the eyes of its
employees was that of a professional, altruistic, can-do agency that acted like
a familywhile delivering quality service, became the object of an increasingly
negative set of images that members felt others held of the agency. Both the
positive identity and the negative image affected members'interpretations of
who they were, what they felt, what they faced, and what they were doing. As
Dutton and Dukerich (1991) put it, "Individuals'self-concepts and personal
identities are formed and modified in part by how they believe others view
the organization for which they work. . . . The close link between an individ-
ual's character and an organization's image implies that individuals are per-
sonally motivated to preserve a positive organizational image and repair a
negative one through association and disassociation with actions on issues"
(p. sa8).
It is this very associating and disassociating with what come to be seen as
threats to images as well as identities, or opportunities to repair and reaffirm
them, that affects a person's view of what is out there and what it means. The
same event such as financing drop-in centers for the homeless or creating rules
and regulations for a bus terminal or educating bus patrons about different
types of homeless people all can be seen either as taking responsibility or
disowning it, as defensive or proactive, as consistent or inconsistent with organi-
zational identity, as a threat or an opportunity. The meaning that is actually
sustained socially from among these alternatives tends to be one that reflects
favorably on the organization and one that also promotes self-enhancement,
efficacy,and consistency. If negative images threaten any of these three repre-
sentations of self, then people may alter the sense they make of those images,
even if this means redefining the organizational identity.If redefinition proves
unworkable, then something other than the organization (e.g., political
affiliation with the religious right) may become the mirror in front of which
individuals primp, evaluate, and adjust the self that acts, interprets, and becomes
committed.
In the context of the image of the mirror, which is the image that introduces
the Dutton and Dukerich study ("Keeping an eye on the mirror") as well as
concludes it ("and whether or not they tike the reflection in the mirror"; P. 551),
it is well to remind ourselves how clear Cooley (1902) was when he first
suggested the idea of a mirror and a looking-glass self in 1902, while he was
at the University of Michigan:

As we see our face, figure, and dress in the fiooking] glass, and are interested
in them because they are ours, and pleased or otherwise with them according
as they do or do not answer to what we should like them to be; so in
 SENSEMAKING IN ORGANIZATIONS

imagination we Perceive in another's mind some thought of our aPPear-

manners, aims, deeds, character, friendS, and so on' and are variously
"n..,
affected by it.
A self-idea of this sort seems to have three principal elements: the imagina-
tion of our aPPearance to the other Person; the imagination of his judgment
of that appearanc€; and some sort of setf-feeling, such as pride or mortifica-
tion. ThJ comparison with a looking-glass hardly suggests the second
element, the imagined judgment, which is quite essential. The thing that
moves us to pride or.shame is not the mere mechanical reflection of
ourselves, but an imputed sentiment, the imagined effect of this reflection
upon another's mind. This is evident from the fact that the character and
*iigttt of that other, in whose mind we see ourselves, makes dl the differ-
enci with our feeling,We are ashamed to seem evasive in the presence of a
straightforward -urr, cowardly in the Presence of a brave one' gross in the
.y.t of u refined one, and so on. We always imagine, and in imagining share,
the judgments of the other mind. (pp. 152-153)

The mirror for Port Authority employees, as for organizational members

in general, can be figuratively, if not literally, a bystandet at a neighborhood
cookout who says, "How can you stand that bus terminal!" . . . or that oil com-
pany, tobacco comPany, investment firm, police department, or theme park
developer? Depending on the "weight and character" of that questioner, the
imagined judgment of that person, and one's own resulting self-feeling, that
small act of sensemaking at the cookout can affect individual interpretations
and actions, which can then diffirse and have much larger organizational effects
(see Tice, tggz,for data showing how the looking-glass magnifies). All of this
comes about because sensemaking begins with a self-conscious sensemaker.
Ring and Van de Ven (1989) make a similar point when they adapt the
work of Turner (1987) to their own studies of transactions as occasions for
innovation.

Sensemaking processes derive from . . . the need within individuals to have

a sense of identity-that is, a general orientation to situations that maintain
esteem and consistensy of one's self-conceptions. Sensemaking Processes
have a strong influence on the mannerbywhich individuals within organi-
zations begin processes of transacting with others. If confirmation of one's
own enacted "self" is not realized, however, sensemaking Proc€sses recur and
a reenactment and representment of self follows. . . . [O]rganizational partici-
pants come to appreciate the nature and purpose of a transaction with others
Lyreshaping or clarifringthe identityof theirown organization. Byprojecting
Seven Properties of Sensemaking

itself onto its environment, an organization develops a self-referential

appreciation of its own identity, which in turn permits the organization to
act in relation to its environment. (Ring &Van de Ven, 1989, p. 180)

Several points are worth noting in this description. First, controlled, inten-
tional sensemaking is triggered by a failure to confirm one's self. Second,
sensemaking occurs in the service of maintaining a consistent, positive self-
conception. And, as Steele (1983) has demonstrated, the chance to reaffirm a
self-concept reduces the discomfort felt when the person confronts discrep-
ancies between belief and action similar to those that animate dissonance
reduction. Third, people learn about their identities by projecting them into
an environment and observing the consequences. Although Ring and Van de Ven
are more focused on confirmation than on learning as the desired outcome,
their argument does not preclude learning. Parenthetically, there is a iarring
shift in the level of analysis in the quotation when confirmation of one's own
enacted self becomes "clarifring the identity of their own organization,"
which then becomes the organization developing "an appreciation of its
own identity." Such slippage is not inherent in discussions of sensemaking.
Chatman et al. (1985) describe one remedy:

When we look at individual behavior in organizations, we are actually

seeing two entities: the individual as himself and the individual as rePre-
sentaiive of his collectivity. . . . Thus, the individual not only acts on behalf
of the organization in the usual agency sense, but he also acts, more subtly,
"as the organizationo when he embodies the values, beliefs, and goals of the
collectivity. As a result, individual behavior is more "macro" than we usually
recognize. (p.2ll)

The final two nuances of the quotation concern reciprocd influence and
the self as text. The fourth nuance is that people simultaneously try to shape
and react to the environments they face. They take the cue for their identity
from the conduct of others, but they make an active effort to influence this
conduct to begin with. There is a complex mixture of proaction and reaction,
and this complexity is commonplace in sensemaking.
Fififi, and perhaps most important, the idea that sensemaking is self-referential
suggests that self, rather than the environment, may be the text in need of
interpretation. How can I know who I am until I see what they do? Something
like that is implied in sensemaking grounded in identity. I make sense of
whatever happens around me by asking, what implications do these events
 SENSEMAKING IN ORGANIZATIONS

have for who I will What the situation will have meant to me is dictated
be?
by the identity I adopt in dealing with it. And that choice, in turn, is affected
by what I think is occurring. What the situation means is defined by who I
become while dealing with it or what and who I represent. I derive cues as to
what the situation means from the self that feels most appropriate to deal with
it, and much less from what is going on out there.
The more selves I have access to, the more meanings I should be able to
extract and impose in any situation. Furthermore, the more selves I have access
to, the less the likelihood that I will ever find myself surprised (Louis, 1980)
or astonished (Reason, 1990), although I may find myself confused by the
overabundance of possibilities and therefore forced to deal with equivocality.
A mutable self may cause problems for 'consistency of one's self-conceptions,"
unless flexibility, mutability, and adaptability are themselves central elements
in that self-conception.

2. Retrospective

Perhaps the most distinguishing characteristic of the present conceptuali-

zation of sensemaking is the focus on retrospect. Retrospect, however, is not
something of importance just for issues of sensemaking. It bears on the larger
issue of organizational structure because, as Starbuck and Nystrom (1981)
have noted, structure is irelf "an artifact of postdiction, observation, and
explanation" (p. 12). The basic argument for making retrospect central was
spelled out in 1969 (Weick, 1969, pp.63-69). This argument is a good example
of the continuing influence of ethnomethodology on the study of organiza-
tional sensemaking.
The idea of retrospective sensemaking derives from Schutz's (1967) analysis
of "meaningful lived experience." The key word in that phrase, lived, is stated
in the past tense to capture the reality that people can know what they are
doing only after they have done it. Pirsig (cited in Winokur, 1990) makes this
point when he says, "Any intellectually conceived object is always in the past
and therefore unreal. Reality is always the moment of vision before intellec-
tualization takes place. There is no other reality" (p.32).
Hartshorne (L962) makes the same point:

Man has discovered that his perceived world is in reality a past world. . . . [A] ny
object outside the body, however close, is at least minutely past by the time
we perceive it. Accordingly, if "memory" is defined as "experience of the
Snen Properties of Sensemaking

past," then all percegtion . . . is a form of memory by this definition of the

word. Moreover, the fact that with near objects the time interval may be
extremely small establishes no distinction from personal memory, for (and
philosophers have an inveterate tendency to forget this), while the obvious
oramples of memory cover appreciable time intervals-a minute, a day, a
year-less obvious but undeniable examples cover but a fraction of a second.
Such immediate, or very short-run, memory is so much with us that we
almost fail to notice it consciouslS and our philosophies are greatly injured
by this oversight. fu I begin the latter portion of a long word, my utterance
of the first part is already in the past. But I do not experience this latter
portion as a fresh start, but rather, as continuation of the earlier portion.
We hear a great ded about the mistakes of memory; however, somewhat as
vision for close objects is the most reliable, similarly trustworthy is memory
for the very short-run past. @. aa2)

Schutz, Pirsig, and Hartshorne are all sensitive to the point that time exists
in two distinct forms, as pure duration and as discrete segments. Pure dura-
tion can be described usingWilliam lames's image of a'stream of experience."
Note that experienceis singular, not plural. To talk about experiences implies
distinct, separate episodes, and pure duration does not have this quality.
Instead, pure duration is a "coming-to-be and passing-awaythat has no contours,
no boundaries, and no differentiation' (Schutz, 1967, p. 47 ).
Readers may object that their experience seldom has this quality of contin-
ual flow. Instead, ocperience as we know it exists in the form of distinct events.
But the only way we get this impression is by stepping outside the stream of
ocperience and directing attention to it. And it is only possible to direct attention
to what exists, that is, what has already passed. In Schutz's (1967) words,

When, by -y act of reflection, I turn my attention to my living experience,

I am no longer taking up my position within the stream of pure duration, I am
no longer simply living with that flow. The experiences are apprehended,
distinguished, brought into relief, marked out from one another; the exPe-
riences which were constituted as phases within the flow of duration now
become objects of attention as constituted experiences. . . . For the Act of
attentiot+and this is of major importance for the study of meaning-Pre-
supposes an elapsed, passed-away experience-in short, one that is already
in the past. (p. 51)

Given this concept of experiencing and orperiences, several things are worth
noting. First, the creation of meaning is an attentional process, but it is attention
 SENSEMAKING IN ORGANIZATIONS

to that which has already occurred. Second, because the attention is directed
baclarard from a specific point in time (a specific here and now),whatever is
occurring at the momentwill influencewhat is discoveredwhen people glance
backruard. Third, because the text to be interpreted has elapsed, and is only a
memory anything that affects remembering will affect the sense that is made
of those memories. Fourth,the sequence,stimulus-response, canbe amisleading
analytical unit as we saw earlier in the example ofjuror decision making. Only
when a response occurs can a plausible stimulus then be defined. This reversal
comes about because we can neyer know the beginning phase. fui action can
become an object of attention only after it has occurred. At the time it is noticed,
several possible antecedents can be posited. The choice of "the' stimulus
affects the choice of what the action 'means." And both choices are heavily
influenced by the situational context.
George Herbert Mead (1956) made essentially the same argument that
Schutz made: 'We are conscious always of what we have done, never of doing
it. We are alwaln conscious directly only of sensory processes, never of motor
processes; hence we are conscious of motor processes only through sensory
processes which are their resultants" (p. 136). Actions are known only when
they have been completed, which means we are ahuays a little behind or our
actions are ahuays a bit ahead of us. To anticipate a later point, if hindsight is
a bias (e.9., Hawkins &Hastie, 1990), then everyone is biased all the time. The
nature of time and sensing guarantee that outcome.
To understand how specific meanings arise retrospectively, think of the act
of reflection as a cone of light that spreads bachrard from a particular present.
This cone of light will give definition to portions of lived experiences. Because
the cone starts in the present, projects and feelings that are under way will
affect the baclq glance and what is seen (Sdrwartz, 1991). Thus "the meaning
'ard
of a lived experience undergoes modifications depending on the particular
kind of attention the Ego gives to that lived experience" (Schutz, 1967,p.73).
*attached
Meaning is not to" the experience that is singled out. Instead, the
meaning is in the kind of attention that is directed to this e:rperience.
To see how this works, assume thatpeople are pragmatic (James, 1890/1950;
Rorty, L982), that 'socid thinking is for doing" (Fiske, 1992,p.877). lny
reflective act originates in a here and now where some projects are visualized,
others are under way, and still others have just been completed. "This whole
function of conceiving, of fxing, and holding fast to meanings, has no signifi-
cance apart from the fact that the conceiver is a creature with particular
purposes and private ends' (Iames, 1890/1950, Vol. 1,p.482). Whatever is
Sann Properties of Sensmnking 27

now at the present moment, under way will determine the meaning of what-
ever has just occurred.
Meanings change as current projects and goals change (Gioia & Chittipeddi,
1991, p. a35). The effects of projects on meanings is visible in Lanir, Fischoff,
and Johnsont (19E8) argument that military command-and-control systems
connect people at the top, whose mindset is strategic thinking and calculated
risk taking, with people at the bottom, whose mindset is more tactical,local,
and entrepreneurial and for whom boldness and the exploitation of surprise
are crucial. Projects at the top and bottom differ dramatically, as do readings
of the "same" events. Gephart (1992,pp. Ll9-120) found this when he studied
an accident investigation in which a top-management logic built from projects
defined by steps and procedures differed from the situational logic of the opera-
tors themselves, who saw the same projects differently. Fiske (I992,p. 88a) has
argued that gods can be partitioned either in terms of speed (whictr encourages
the confirmation of expectancies when elapsed experience is examined) or
accuracy (which encourages more complex examination of elapsed experi-
ence). The influential distinction between threat and opportunity (Dutton &
Jackson, L987; Jackson & Dutton, 1988) as contrasting labels for experience
may influence sensemaking at an even earlier stage than we first thought,
because it is conceivable that they dominate the definition of a project and
therefore influence what is octracted from elapsed experience.
Because people typically have more than one project under way, and have
differing awareness of these projects, reflection is overdetermined and clarity
is not assured. Instead, the elapsed experience appears to be equivocal, not
because it makes no sense at all, but because it makes many different kinds of
sense. And some of those kinds of sense may contradict other kinds. That is
not surprising given the independence of diverse projects and the fact that
their pursuit in tandem can work at cross-purposes.
The important point is that retrospective sensemaking is an activity in
which many possible meanings may need to be synthesized, because many
different projects are under way at the time reflection takes place (e.9., Boland,
1984). The problem is that there are too many meanings, not too few. The
problem faced by the sensemaker is one of equivocality, not one of uncer-
tainty. The problem is confusion, not ignorance. I emphasize this because
those investigators who favor the metaphor of information processing (e.g.,
Huber, Ullman, & kifer, 197 9) often view sensemaking, as they do most other
problems, as a setting where people need more information. That is not what
people need when they are overwhelmed by equivocality. Instead, they need
values, priorities, and clarity about preferences to help them be clear about
 SENSEMAKING IN ORGANIZATIONS

which projects matter. Clarity on values clarifies what is important in elapsed

experience, which finally gives some sense of what that elapsed experience
means.
Investigators need not adopt pragmatism to use the idea of retrospective
sensemaking.Any perspective can be inserted into the here and now as long
as its effects on remembering are traced through to answer the question of
why people make the sense they do of their ongoing activity. If one person can
be preoccupied with something in the here and nor4r, so can others. And
whatever that preoccupation is, it can impose a figure-ground relationship on
elapsed experience, therebyfacilitating sensemaking. "No lived experience can
be exhausted by a single interpretive scheme" (Schutz, L957, p. 85).
If we bring this discussion closer to the present, recent discussions of
sensemaking, especially discussions of hindsight bias, tend to emphasize how
much the bachrard glance leaves out and the problems this can create. The
basic finding that investigators keep returning to (e.g., Hawkins & Hastie,
1990) is that people who know the outcome of a complex prior history of
tangled, indeterminate events remember that history as being much more
determinant, leading "inevitably" to the outcome they already knew. Further-
more, the nature of these determinant histories is reconstructed differently,
depending on whether the outcomes are seen as good or bad. If the outcome
is perceived to be bad, then antecedents are reconstructed to emphasize
incorrect actions, flawed analyses, and inaccurate perceptions, even if such
flaws were not influential or all that obvious at the time (Starbuck & Milliken,
1988, pp.37-38'). Thus, hindsight both tightens causal couplings and recon-
structs as coupled events a history that leads directly to the outcome. Starbuck
and Milliken (1988) put the point this way: "Retrospection wrongly implies
that errors should have been anticipated and that good perceptions, good
analyses, and good discussions will yield good results" (p. 40).
We need to pay attention to the phrase "wrongly implies." It is true that
tight implications, formed in hindsight, are wrong because the future is
actually indeterminate, unpredictable. And it is also true that the past has been
reconstructed knowing the outcome, which means things never happened
exactly the way they are remembered to have happened. Retrospective sense-
making does "erase" (Starbuck & Milliken, 1988, p.37) many of the causal
sequences that made it harder to accomplish the final outcome.
But, if people want to complete their projects, if effort and motivation make
a difference in completing those projects, and if the environment is malle-
able, then a reading of past indeterminacythat favors order and oversimplifies
causdity (Reason, 1990, p. 9l) may make for more effective action, even if it
Seven ProPerties of Sensemaking

is lousy history. Brunsson (1982) makes essentially this argument, as does

Gollwitzer (1990).
To keep the findings regarding hindsight bias in perspective, we need to
remember three things. First, retrospective sensemaking in everyday
life in-
means that
volves relatively short time spans between act and reflection, which
memory traces are typically fresh and rich with indeterminacy, and that
back
people are mindfut of only a handful of projects at the time they look
Lver what has just happened. Both tendencies work against the
likelihood that
distortions will be substantial. Second, retrospection "only makes the
past
(Starbuck
clearer than the present or future; it cannot mala the past transParent"
& Milliken, 1988, pp. 39-40). Although the past may be subject to partial
and rationality
eriuiing, it is not obliterated. And thfud, the feelingof order, clarity,
is an important goal of sensemaking, which means that once this feeling
is

achieved, further retrospective processing stops'

The student of sensemaking is well advised to become more comfortable
with the idea of retrospect because much work in organizational studies
(1978, p' 935) investigations
assumes its operation. For example, Mintzberg's
of strategy making are unusually well attuned to the operation of retrospect'
His definition of strategy as observed patterns in past decisional behavior
(realized strategy = consistent behavior) represents a sophisticated treatment
in
of retrospect. Boland (19s4) gathered a grouP of film-lending executives
1982 to 1985' and
1980, provided them with accounting rePorts prepared for
asked them to imagine itwas luly 21,1985, and then to
discuss what the film
service had become and why. This ercercise in future perfect thinking
was an

attempt to otplore the proposition that it is easier to make sense of events

when they are placed in the Pastr even if the events have not yet occurred'
Boland reported that a major outcome of the experiment was that in trying
to understand what had been done in an imaginary future, participants discov-
The experi-
ered that they had an inadequate understanding of an actual Past.
ment uncovered disagreements about the nature and meaning of past events
that people did not realize had impeded their current decision making'
ftt. ptittt of the Boland work, and the more general concept of future
perfect thinking (Weick, lg|g),is that sensemaking can be extended beyond
a larger
the present. As a result, present decisions can be made meaningful in
context than they usually are and more of the past and future can be
brought
to bear to inform them.
retrospect
Finally, Staw (1975) provides an esPeciallyclean example of how
3 people each
operates. He randomly assigned 60 students to 20 grouPs of
and had them study the 1969 annual rePort of a medium-sized electronics
 30 SENSEMAKING IN ORGANIZATIONS

comPany to estimate its sales and earnings per share for 1970. Groups were
given 30 minutes to complete this financial puzzle task After each group
presented its results, it was randomly classified as either a high-performance
group ('your group has done quite well," estimates for sales are off by only
$10,000) or a low-performance group ('not done too well," estimates for sales
are offby $10 million) and given false performance feedback. After being told
their performance, participants filled out a questionnaire about "what went
on in the group" with respect to group cohesiveness, influence, communica-
tion, task conflict, openness to change, motivation, abiliry and clarity of
instructions. On all measures except those for task conflict, individuals ran-
domly assigned to high-performance groups rated their groups significantly
higher than did those assigned to low-performance groups. Just as with
Garfinkel's jurors, whose verdict was an independent variable that influenced
their account of what was significant in their prior deliberations, Staw's
analysts used their knowledge of their group's performance to construct a
plausible history of the process that produced that outcome. How can I know
what we did until I see what we produced? The dominance of retrospect in
sensemaking is a major reason why students of sensemaking find forecasting,
contingency planning, strategic planning, and other magical probes into the
future wasteful and misleading if they are decoupled from reflective action
and history.

3. Enactive of Sensible Environments

The preceding discussions of identity and retrospect begin to spell out

properties of the "sensing" that is associated with sensemaking. Now I want
to say more about the activity of "making" that which is sensed. This discus-
sion has been anticipated at several points up to now. It was anticipated when
I cited Thomas et al.'s (1993, p. 2) argument that the concept of sensemaking
keeps action and cognition together; when I said that interpretation better
explains how people cope with entities that already exist, whereas sensemak-
ingbetter explains how entities get there in the first place; andwhen I implied
that action is a precondition for sensemaking as, for example, when the action
of saying makes it possible for people to then see what they think
I use the word enactment to preserve the fact that, in organizational life,
people often produce part of the environment they face (Pondy & Mitroff,
L979, P. L7).I like the word because it suggests that there are close pardlels
between what legislators do and what managers do. Both groups construct
Seven Properties of Sasemaking 31

reality through authoritative acts. When people enact laws, they take unde-
fined space, time, and action and draw lines, establish categories, and coin
labels that create new features of the environment that did not exist before.
For example, the numbers 399,400, and 401 meant nothing in particular until
the Michigan legislature recently declared that Michigan Bell could charge for
each telephone call above 400 that a customer made in one month. The 400th
call has now become something tangible, unique, visible, and symbolic as well
as something that is an obstacle for someone on a budget to work around. The
legislators enacted a constraint for their constituents that is iust as real as are
the buttons that those constituents push to make that expensive 400th call.It
remains to be seen whether the legislators have also enacted the conditions of
their own defeat.
Consider other examples of enactment. TWo cops are driving in a squad car
on patrol, and a teenager gives them the finger as they drive by. The coPs can
ignore the kid, stop, or, as is most common, return the gesture. Bill Walsh,
when he coached the San Francisco 49ers football team, used to write out the
first 20 offensive plays the team would use in a game before he even got to the
stadium (Business Week, October 24,1983).In the Persian Gulf in 1987, the
United States put an American flag on a Kuwaiti ship, called it "The Gas
Prince,'and then surrounded it with U.S. combat ships. On October 7, 1980,
at Hartsfield Airport in Atlanta, an air traffic controller put five aircraft in a
holding pattern on a clear day, and between 8:14 a.m. and 8:20 o.trl.r there were
10 near misses among those five aircraft. In each case, people created their
own environments and these environments then constrained their actions.
The cops create an environment they have to deal with once they respond to
the teenager. Bill Wdsh creates the defensive environment his offense will face
once he begins to run off the 20 plays without making any adjustment. The
U.S. government creates a challenge to which they have to respond. The air
traffic controller in Atlanta creates an environment of aircraft that he is
increasingly unable to control.
In these cases, there is rof some kind of monolithic, singular, fixed environ-
ment that exists detached from and externd to these people. Instead, in each
case the people are very much a part of their own environments. They act, and
in doing so create the materids thatbecome the constraints and opportunities
they face. There is not some impersond "th.y'who puts these environments
in front of passive people. Instead, the "they''is people who are more active.
All too often people in organizations forget this. They fall victim to this
blindspot because of an innocent sounding phrase, "the environment." The
word fhe suggests something that is singular and fixed; the word environment
32 SENSEMAKING IN ORGANIZAT10NS

suggests that this singular, fixed something is set apart from the individual.
Both implications are nonsense.
Throughout this book,I assume that action is crucial for sensemaking. In
doing so, I take my lead from Follett (1924), whose work I quote at length
because it is not well known and captures subtleties that most people, includ-
ing myself, often miss.
The centerpiece of Follett's thinking is the idea that people receive stimuli
as a result of their own activiry which is suggested by the word enactment.
With respect to the environment, she notes that "we are neither the master
nor the slave of our environment. We cannot command and the environment
obey, but also we cannot, if we would speakwith the greatest accuracy, saythat
the organism adjusts itself to environment, because it is only part of a larger
truth. My farmer neighbors know this: we prune and graft and fertilize certain
trees, and as our behavior becomes increasingly that of behavior towards
apple-bearing trees, these become increasingly apple-bearing trees. The tree
releases energy in me and I in it; it makes me think and plan and work, and I
make it bear edible fruit. It is a process of freeing on both sides. And this is a
creating process" (Follett, L924, pp. I I 8- I 19).
The metaphor of enactment through intentional grafting and pruning is an
instance of artificial selection in evolutionary theory (Weick, L979, p. L76).
Both ideas, the idea of enactment and the idea of artificial selection, invite
close attention to interdependent activities, process, and continuous change.
They also alert us to the traps implicit in the analytical categories of stimulus
and response.

The activity of the individual is only in a certain sense caused by the stimulus
of the situation because that activity is itself helping to produce the situ-
ation which causes the activity of the individual. In other words, behavior
is a relating not of "subject" and "object" as such, but of two activities. In
talking of the behavior process we have to give up the expression act "on"
(subject acts on object, object acts on subject); in that process the central
fact is the meeting and interpenetrating of activities. What phpiology and
psydrology now teach us is that part of the nature of response is the change it
makes in the activitywhich caused so-to-speak the response, that is, we shall
never catch the stimulus stimulating or the response responding. (Follett,
1924,p.60)

To remain alert to the ongoing codetermination that occurs during sense-

*Some
making, we need to be especially careful of how we portray process.
writers, while speaking otherwise accurately of the behavior process, yet use
Sarcn Properties of Sensemaking 33

the word result-the result of the process-whereas there is no result o/process

but only a moment ir process" (Follett, 1924,P. 50). In other words, thoughts,
cause-effect, stimulus-response, and subject-object are simply descriptions of
moments in a process. To explore a different moment is to reshuffle the
meaning of all those supposed "products" culled from inspection of a different
moment.
Follett (L924) argues that rather than talk about "results," we should talk
about "relatings':

As we perform a certain action our thought towards it changes and that

changes our activity. . . . You say, "When I talkwith Mr. X he always stimu-
lates me." Now it may not be true that Mr. X stimulates ever,'one; it may be
that something in you has called forth something in him. That is why I said
above that we must give up the expression "act on," object acts on subject,
etc. . . . I never react to you but to you-plus-me; or to be more accurate, it
is l-plus-you reacting to you-plus-me. "I" can never influence "you" because
you have already influenced me; that is, in the very Process of meeting, bY
the very process of meeting, we both become something different. It begins
even before we meet, in the anticipation of meeting. We see this clearly in
conferences. Does anyone wish to find the point where the change begins?
He never will. (pp. 62'63)

If we begin to think about sensemaking asrelating, several classic issues in

organizational studiesbecome recast.I willdiscuss reformulations throughout
the book but Follett's discussion of resistance to change provides some closure
on her discussion.
The phrase "resistance to change" is organizational shorthand for the more
general idea of "resistance of environment." Follett (1924) argues that use of
the word resistance creates an unfortunate mindset that limits the way in which
we presume people dedwith the environment. She argues that rather than talk
about resistance, we should talk abou

confrontingthe activity of environmen t Thus we need not make anticipatory

judgmentithere may be opposition, there may be resistance, but this defi-
nition leaves it possible for us to wait until we find them. This would make
a greatchange in social sciences. Here we should have not necessarily the
opposing bui the confronting of interests. [See later discussion of sense-
making as arguing in Chapter 6.1 This confronting would make lPPerent
many incomfatibilities of interests, but does not judge the case beforehand
as to what stralt be done about it. Confront does not mean combat. In other
34 SENSEMAKITI.G IN ORGANIZATIONS

words, it leaves the possibility of integrating as the method of the meeting

of difference. (p. 120)

Concepts of population ecology (e.g., Hannan & Freeman,L977) would collapse

if theywere stripped of the assumptions of resistance and combat. However,
we might then find ourselves observing a richer set of options than simply
living or dying when people confront diminishing resources.
These several quotations from Follett's work combine action with retro-
spect and sensemaking with interpretation. People create their environments
as those environments create them. A deep appreciation of this process, and
of the incomplete rendering that occurs when we freeze moments and prod-
ucts in that process, allows us to address what manyview as an occupational
hazardwhen people studysensemaking.In the felicitous phrase of Burrell and
Morgan (1979), this hazard is 'ontological oscillation" (p.266). They argue
that theorists who workwith the ideas of hermeneutics, phenomenology, and
symbolic interactionism (although surprisingly not those who workwith the
related positions of critical theory and conflict theory) often

stressahighlysubjectiviststancewhich denies the existence of socialstructures

and concrete social reality of any form. Yet the attempt to operationalize
their ideas within an empirical context frequently leads them to admit a
more realist form of ontology through the back door. . . . [O]ntological
oscillation is prevalent in all forms of phenomenological sociology which
attempt to illustrate its basic propositions through the empirical study of
situations drawn from everyday life. (Burrell & Morgan,1979, p. 266)

People engage in oscillation when they attempt to show how the supposedly
hard, concrete, tangible aspects of organizational life are dependent on sub-
jective constructions, but then smuggle in realist assumptions that posit
constraints and objects that exist independent of subjective constructions (see
Shotter's L993,p. 154 illustration of how Morgan himself makes this "error").
I have made analyses that *oscillate." One of my favorite conceptual tools, the
notion of requisite variety, is modeled after a carpenter's tool called a contour
gauge (Weick, L979, p. 190). A contour gauge is a set of sensors that registers
the patterns of a solid surface when it is pressed against that surface. The gauge
allows the carpenter to transfer that pattern to another surface where it can
be traced and duplicated.
I have used the contour gauge, which was inspired by Heider's (1959)
discussion of thing and medium, to argue that it takes a complex sensing
Saryn Properties of Sensenaking

system to register and regulate a comple.x object. That is about as realist as one
can get. And yet within earshot of that analysis is another analysis that asserts
that self-fulfilling prophecies are the prototype for human sensemaking.
People create and find what they expect to find. Does this mean, then, that the
contour gauge presses against and registers the equivalent of fulfilled prophe-
cies? Absolutely. That very mixing of ontologies is what drives Burrell and
Morgan nuts.
But it shouldn't. People who study sensemaking oscillate ontologically
because that is what helps them understand the actions of people in everyday
life who could care less about ontology. Noticing (Starbuck & Milliken, 1988),
manipulation (Hedberg,Nystrom, &Starbuck, 1976), interpretation (Isabella,
1990), and framing (Gofftnan,l974) are all plausible events in sensemaking,
even though they represent different combinations of subjective/objective as
assumptions aboutthenahrreof socialscienceand change/regulation asassump-
tions about the nature of society (Burrell &Morgan,l979,pp.I-37).If people
have rnultiple identities and deal with multiple realities, why should we orpect
them to be ontological purists? To do so is to limit their capability for
sensemaking. More likely is the possibility that over time, people will act like
interpretivists, functionalists, radical humanists, and radical structuralists.
Consider newcomer socialization.If, as Louis (19E0) shows, newcomers at
first are flooded with surprises, then they start as interpretivists. And herme-
neutics helps the newcomer gloss the unexpected. But it isn't long until
opposing interest groups make aplay for the loyalties of newcomers, in which
case those newcomers act more like radical structuralists whose actions are
better understood using conflict theory. Over time, as routines develop and
the meaning of objects becomes fixed by organizationd culture, facticity
develops as things become taken for granted, and functional theories such as
social system theorybecome more useful. What has happened is that interde-
pendent activities of the newcomer and others have evolved, and with them,
the referents and accounts that are given when a moment in the process of
evolving is frozen.
The concept of enactment has a touch of realism in its emphasis on
bracketing and punctuating. To cope with pure duration, people create breaks
in the stream and impose categories on those portions that are set apart. When
people bracket, they act as if there is something out there to be discovered.
Theyactlike realists, forgettingthatthe nominalistinthem uses aprioribeliefs
*find" seams worth punctuating (Starbuck & Milliken,
and expectations to
1988, p. 50). Czarniawska-|oerges (1992\ puts the point this way: "A stone
exists independently of our cognition; but we enact it by a cognitivebtacketing,
 SENSEMAKING IN ORGANIZATIONS

by concentrating our attention on it. Thus 'called to life,'or to attention, the

stone must be socially constructed with the help of the concept of stone, its
properties, and uses. We can base a physical action that might annihilate the
stone on this construction" (p. 34). Tlo cite a different example, "everybuilding
is socially constructed. It consists of bricks, mortar, human labor, building
regulations, architectural design, aesthetic expression, and so on; each of
them, in turn, socially constructed and put together by a socially constructed
concept of a building" (p. 33). A contractor carves out elements, relates them,
and animates a system that makes the fragments sensible.
But there are also invention and construction, activities that seemingly
move away from objects and objectivity to subjects and subjectivity. Here the
emphasis is on the fact that brackets and punctuations shape, modify, and give
substance to whatever other activities the person confronts (e.g., Pondy &
Mitroff, L979,p. 13).This is where sensemaking most clearlybecome a process
that creates objects for sensing or the structures of structuration. As Ring and
Van de Ven (1989) put it, "The process of understanding emerges from the
need of individuals to construct an external factual order 'out there' or to
recognize that there is an external reality in their social relationships" (p. I 8 I ).
In other words, people act in such a way that their assumptions of realism
become warranted.
The socially created world becomes a world that constrains actions and
orientations. "What was once recognized as a socially constructed transaction
takes on the form of an externally specified objective reality, where transacting
parties play out preordained roles and 'action routines' " (Ring & Van de Ven,
1989, p. 185). It is this institutionalizingof social constructions into the way
things are done, and the transmission of these products, that links ideas about
sensemaking with those of institutional theory. Sensemaking is the feedstock
for institutionalization.
The implication that enactment is first and foremost about action in the
world, and not about conceptual pictures of that world (enthinlanent, as Lou
Pondy called it), is clearest in Porac et al's discussion of the tight-knit Hawick
communig of people who manufacture classic cashmere sweaters. These
investigators argue that the Hawick mind develops from the 'tnacting of a
competitive group" (1989, pp. 39S-399).By this they mean that the Hawick
group and its environment jointly constructed one another as a result of
material and technical choices involving suchthings aswhatgoods to produce,
which raw materials to purchase, which customers to target, and so on. These
actions create "market cues," rather than respond to a pregiven environment.
Perception of these enacted cues by Hawick manufacturers alters the mental
Seven Properties of Smsemaking 37

model, which then guides subsequent strategic choices. The enacted world is
tangible because it contains material and technical artifacts given substance
and meaning by the manufacturers of Hawick. The enacted world is also a
subjective, punctuated, bracketed world because it has its "origin" in mental
models of causally connected categories that were part of the strategizing that
carved out artifacts in the first place. People in Hawick enact the environment
that enacts their Hawick identity, and this process represents enactment in
sensemaking. There are subjective interpretations, of orternally situated in-
formation, but that information has become external and objectified by means
of behavior (Porac et al., 1989, p. 398). People discover their own intentions.
If this is ontological oscillation, so be it. It seems to work
There are two cautions in working with the concept of enactment. First,
remember that creating is not the only*(ring that can be done with action.
Blumer ( 1959) was especially clear that, because people had the capability for
reflection, self-indication, and interpretation, "given lines of action may be
started or stopped, they may be abandoned or postponed, they may be confined
to mere planning or to inner life of reverie, or if initiated, they may be trans-
formed" (p. 15).Any one of these outcomes, all of which differ from creation,
can still produce meaning. The idea that action can be inhibited, abandoned,
checked, or redirected, as well as expressed, suggests that there are manyways
in which action can affect meaning other than by producing visible conse-
quences in the world. Abbreviated actions, constructed in imagination and
indicated solely to oneself, can also be made meaningful. The caution, then,
is to be careful not to equate action with a simple response to a stimulus, or
with observable behavior, or with goal attainment. To do so may be to miss
subtle ways in which it creates meaning. The act that never gets done' gets
done too late, gets dropped too soon, or for which the time never seems right
is seldom a senseless act. More often, its meaning seems all too clear.
The second caution comes in the form of a forewarning for sensemaking:
Beware of Cartesian arxiety (Varela, Thompson, & Rosch, 1991, pp. 1a0-1a5).
People seem to need the idea that there is a world with pregiven features or
ready-made information, because to give up this idea of the world as a fixed
and stable reference point is to fall into idedism, nihilism, or subjectivism, all
of which are unseemly. Cartesian anxiety is "best put as a dilemma: either we
have a fixed and stable foundation for knowledge, a point where knowledge
starts, is grounded, and rests, or we cannot escape some sort of darkness, chaos,
and confusion. Either there is an absolute ground or foundation or everything
falls apart" (Varela et al., 1991, p. 1a0).
 SENSEMAKING IN ORGANIZATIONS

There are alternatives to nihilism. One solution lies in accepting that

"groundlessness is the very condition for the richly textured and interdepend-
ent world of human experience. . . . [The world is not fixed and pregiven but]
continually shaped by the types of actions in which we engage" (Varela et al.,
1991, p. laa). If as Follett argued, we confront activities, then action, relation-
ships, trust, faith, experience, and presumptions are not just tools of sense-
making. They are also tools of epistemology and ontology. They create that
which they interpret. To charge people who use them with ontological oscil-
lation is to make too much of too few moments in the prociss of sensemaking.
In this context,I think it is important to weigh the message of deconstruc-
tionists such as Eagleton (1990) or Hassard and Parker (1993). It is certainly
not news to students of sensemaking that multiple meanings abound and that
"texts" can be read in more ways than were intended, to the point where
meanings become interchangeable and power privileges some meanings over
others. The destructive side of deconstructionism is the undermining of the
faith and belief necessary to get sensemaking started. If there are multiple
meanings that collapse under scrutiny, why bother with sensemaking at all?
This is the very issue that concerned James (1885/1956) when he asked the
question, "Is lifeworth living?" (see also p.54 in this chapter). His answerwas,
you can make either yes or no valid.If you assume life is notworth living and
act accordingly, then you will be absolutely right and suicide will be the only
plausible alternative. And if you believe life is worth living, then thatbelief too
can validate itself. The issue turns on faith or the lack thereof, because it sets
self-fulfilling action in motion. Faith is instrumental to sensem"kit g. fames
knew this even if deconstructionists have forgotten it. Ironically, their faith in
the sensibleness of the deconstructionist pose validates it and supports James
rather than Derrida.

4, Social

The word sensemaking tempts people to think in terms of an individual

level of analysis, which induces a blindspbt-we need to catch early on. When
*human thinking
discussing sensemaking, it is easy to forget that and social
functioning . . . [are] essential aspects of one another' (Resnick, Levine, &
Teasley, 1991, p. 3). Many scholars of organizations are mindful of the inter-
twining of the cognitive and the social as in this informative definition proposed
by Walsh and Ungson (1991): An organization is "a network of intersubjec-
tively shared meanings that are sustained through the development and use
Seven Properties of Sensemaking

of a common language and everydaysocial interaction" (p.60).This definition

is social several times over in its references to *network," "intersubjectively
shared meanings," "common languager" and "social interaction.'
Those who forget that sensemaking is a social process miss a constant
substrate that shapes interpretations and interpreting. Conduct is contingent
on the conduct of others,whetherthose others are imaginedor physicallypresent.
The contingent quality of sensem"king is found in Allport's ( I 985 ) description
of socialpsychologyas "an attemptto understand andexplain how the thought,
feeling, and behavior of individuals are influenced by the actual, imagined, or
implied presence of others" (p. 3, italics added). Burns and Stalker (1961),
focusing on organizations, say essentially the same thing:

In workingorgrnizatrons decisions are made either in the presence of others

or with the knowledge tlnt they will have to be implenrented, or understood, or
approved by others. The set of considerations called into relevance on any
decision-making occasion has therefore to be one shared with others or
acceptable to them. (p. 118, itaiics added)

The caution implicit in both quotations is that imagined presence can be

overdone and create a specious socid quality. This is the problem with much
of so-called social cognition.

The emphasis of the work on social cognition is that internal constructions

of knowledge or logic affect our understanding of social interactions; how-
ever, these internal constructions are developed independent of other peo-
ple. . . . [An alternative view is that] our intentions and feelings do not grow
within us but between us. . . . [A]n individual creates novel thoughts in the
contoct of interactions with others, and then communicates them to the
larger community. If viable, the larger community generalizes these ideas
such that theybecome part of the culture. (IGhlbaugh, 1993, pp.80, 99)

When people overlook the social substrate, they manufacture theoretical

obstacles that can be distracting. For example, Ring and Rands (1989), in their
investigation of negotiations between 3M and NASA, equate sensemaking with
individual action and understanding with group action. In doing so, they create
obstacles like this: "There also appears to be a definitional question related to
sensernaking and understanding processes: What is the relationship between
one-way and two-way communication processes and sensem"kiog and under-
standing? Clearly, sensemaking can involve one-way communication links. A
person tells me something, and it aids in the developmentof mycognitive map
 SENSEMAKING IN ORGANIZATIONS

of some phenomenon. I need not respond, but if I do, is the response associ-
ated with processes of sensemaking, understanding, or both?" (p. 36a).
This forced separation of individual and social contributions to sensemak-
ing leads them to focus most of their attention on a face-to-face interaction
where joint understanding was furthered (i.e., a 5-day lab tour of NASA
facilities involving 3M and NASA personnel, p. 351) and to downplay the
importance of the pretour sensemaking built around each anticipating how
the other would react to proposals and proposed identities. But it was these
anticipations, these attempts to make sense using the implied, imagined pres-
ence of the others, that enabled people to make sense during the face-to-face
tour. For example, Smith, the NASA representative, learned during the tour
that 3M's dollar commitment to the space project was less than expected. But
Smith did not suddenlybegin to impose the idea of financial commitment on
the face-to-face meetings. Instead, what happened on the tour fine-tuned the
rehearsing that preceded it. And that rehearsing is just as interactive as the tour
itself. Said differently, social influences on sensemaking do not arise solely
from physical presence. That is the whole point of the phrase symbolic inter-
action (Blumer, 1969).
Sensemaking is never solitary because what a person does internally is
contingent on others. Even monologues and one-way communications pre-
sume an audience. And the monologue changes as the audience changes.

Human beings in interacting with one another have to take account of what
each other is doing or is about to do; they are forced to direct their own
conduct or handle their situation in terms of what they take into account.
Thus, the activities of others enter as positive factors in the formation of
their own conduct; in the face of the actions of others one may abandon an
intention or purpose, revise it, check or suspend it, intensify it, or replace
it. The actions of others enter to set what one plans to do, may oppose or
prevent such plans, and may demand avery different set of such plans. One
has toff one's own line of activity in some manner to the actions of others.
The actions of others have to be taken into account and cannot be regarded
as merely an arena for the expresiion of what one is disposed to do or sets
out to do. (Blumer, 1959, p. 8)

Several tactics in scholarship on sensemaking themselves make more sense

if they are seen as attempts to keep socially conditioned activity in the foreground.
For example, socialization is often the setting in which sensemaking is ex-
plored, as we saw in work of Louis (1980). More recent discussions (Lave &
Wenger, 199 I ) of socialization as a process resembling an apprenticeship retain
Seven Properties of Sensemaking 41

this focus on a social setting. In general, socialization studies represent a

variant of Schutz's ( l96a) analysis of the stranger, which suggests that new-
comers need to learn both how to interpret and how to express themselves in
the natives' vernacular.
Investigators who talk about sensemaking often invoke imagery associated
with symbolic interactionism (Fine, 1993), not so much because this is the
unofficial theory of sensemaking but because the theory keeps in play a crucial
set of elements, including self, action, interaction, interpretation, meaning,
and joint action. As we have already seen, these elements are crucial in the
determination of sensemaking, whether one chooses to combine them the way
a symbolic interactionist does or not. Because symbolic interactionism derives
from the work of Mead, and because Mead was adamant that mind and self
arise and develop within the social process, to use the images of symbolic
interactionism is to insure that one remains alert to the ways in which people
actively shape each other's meanings and sensemaking processes.
People who study sensemakingpay a lot of attention to talk, discourse, and
conversation because that is how a great deal of social contact is mediated.
Gronn (1983) describes "talk as the worli'in educational organizations. March
and Olsen (1976) describe organizations as a "set of procedures for argumen-
tation and interpretation" (p. 25).Shotter (1993), in describing the manager
as author, cautions that he does not mean that the manager writes texts, but
rather that the manager is "a'practical-ethical author,' a'conversational author,'
able to argue persuasively for a 'landscape' of next possible actions, upon
which the 'positions' of all who must take part are clear" (p. 157). And Weick
(1985) argues that a significant portion of the organizational environment

consists of nothing more than talk, symbols, promises, lies, interest, atten-
tion, threats, agreements, expectations, memories, rumors, indicators, sup-
porters, detractors, faith, suspicion, trust, appearances,loyalties, and com-
mitments. . . . Words induce stable connections, establish stable entities to
which people can orient (e.g., "gender gap"), bind people's time to projects
("A1, I'd like you to spend some time on this one"), and signiff important
information. Agreement on a label that sticks is as constant a connection as
is likely to be found in organizations. (p. 128)

Although it is important to conceptualize sensemaking as a social activity'

it is also important to maintain a differentiated view of the forms social
influence may take. This sounds obvious, but it is striking how often people
discuss "shared meaning" or "social construction," as if that exhausts what
42 SENSEMAKING IN ORGANIZAT10NS

there is to say about social sensemaking. However, sensemaking is also social

when people coordinate their actions on grounds other than shared meanings
as when joint actions are coordinated by equivalent meanings (Donnellon,
Gray, & Bougon, I 986), distributed meanings (Rasmussen, Brehmer, & Leplat,
1991), overlapping views of ambiguous events (Eisenberg, 1984), or nondis-
closive intimacy (Eisenberg, 1990). Czarniawska-|oerges (1992) argues that
shared meaning is not what is crucial for collective action, but rather it is the
experience of the collective action that is shared (see p. 188 in Chapter 8). She
cites this example:

My two colleagues went to hear a speech given by a well-known business-

man. One "participated in a most exciting encounter between the wis-
dom of practice and curiosity of theory," whereas the other "took part in
an extremely boring meeting with an elderly gentleman who told old
jokes." They are each, nevertheless, members of the same organization, and
what was common for them was that they went to the same room at the
same hour, sharing only the idea that their bosses expected it. (p. 33)

To understand sensemaking is to pay more attention to sufficient cues for

coordination such as a generalized other, prototypes, stereotnres, and roles,
especially considering that organizations seem to drift toward an "architecture
of simplicity'' (Miller, 1993). People who make sense are just as likely to
satisfice as are people who make decisions. Turner's (I971) analyses of organ-
izational talk revealed that "reasons of expediency, or pragmatic considera-
tions, seem to be the most important rule of naming or defining. Other things
being equal . . . a good name was not necessarily the most accurate, but one
that allowed action. It makes sense. 'Tree" or "stone'is enough to decide whether
to use a saw or a hammer; "fir" or "amethyst," albeit more accurate, do not
improve the pragmatic advantage and may prove more costly in social terms
(what if another person at the saw thought it was a pine and wanted to engage
in debate?). Naming seems to be a satisfying process,like any decision-mak-
ing" (quoted in Czarniawska-Joerges, 1992, pp. 178-179).
Blumer (1969, p.76) summarizes well the reasons to be cautious about
overestimating the extent to which social sensemaking means simply shared
understanding. He notes that investigators often argue that common values
are the 'glue" that holds society together, whereas conflicting values des-
tabilize. Blumer (1969) goes on to observe that this
￨:

li sθ ν
θProp`r"“ orSι s“ た
,4g 43
“ “ "α
conception of human society becomes subject to great modification if we
think of society as consisting of the fitting together of acts to form joint
action. Such alignment may take place for any number of reasons, depend-
ing on the situations calling for joint action, and need not involve, or spring
from, the sharing of common values. The participants may fit their acts to
one another in orderly joint actions on the basis of compromise, out of
duress, because they may use one another in achieving their respective ends,
because it is the sensible thing to do, or out of sheer necessity. . . . In very
large measure, societybecomes the formation of workablerelations. (p.76,
italics added)

Alignment is no less social than is sharing. But it does suggest a more varied
set of inputs and practices in sensemaking than does sharing. And it keeps
lines of action in clear view, which, as we just saw in the discussion of enactment,
is crucial.

5. Ongoing

Sensemaking never starts. The reason it never starts is that pure duration
never stops. People are always in the middle of things, which become things,
only when those same people focus on the past from some point beyond it.
Flows are the constants of sensemaking, somethingthat open systems theorists
sudr as Katz and trGhn (1965) taught us, but whidr we have since forgotten
(Ashmos &Huber, 1987). To understand sensemaking is to be sensitive to the
ways in which people chop moments out of continuous flows and extract cues
from those moments. There is widespread recognition that people are always
in the middle of things. What is less well developed are the implications of
that insight for sensemaking.
Dilthey as paraphrased by Burrell and Morgan (1979) and Heidegger as
paraphrased by Winograd and Flores ( 1986) both are sensitive to sensemaking
as ongoing activity. Burrell and Morgan (1979,p.237) citing Rickman (1976),
note that, when Dilthey adapted the so-called hermeneutic circle to social
phenomena, he recognized that
*there are no absolute starting points, no
self-evident, self-contained certainties on which we can build, because we
always find ourselves in the middle of complex situations which we try to
disentangle by making, then revising, provisional assumptions."
Winograd and Flores (1986) make a similar point in their gloss of Heideg-
ger's idea that people find themselves thrown into ongoing situations and have
 SENSEMAKING IN ORGANIZATIONS

to make do if they want to make sense of what is happening. They describe

situations of thrownness in terms of six different properties:

You cannot avoid acting: Your actions affect the situation and yourself, often
against your will.
2. You cannot step back and reflect on your actions: You are thrown on your intuitions
and have to deal with whatwer comes up as it comes up.
3. The effects of action cannotbe predicted: The dynamic nature of social conduct
precludes accurate prediction.
4. You do not have a stable representation of the situation: Patterns maybe evident
after the fact, but at the time the flow unfolds there is nothing but arbitrary
fragments capable ofbeing organized into a host of different patterns orpossibly
no pattern whatsoever.
5。 Every representation is an interpretation: There is no way to settle that any
interpretation is right or wrong, which means an "objective analysis" of that into
which one was thrown, is impossible.
LangUage is action: Whenever people say something, they create rather than
describe a situation,which means it is impossible to staydetached fromwhatever
emerges unless you say nothing, which is such a strange way to react that the
situation is deflected an)l^'ay (pp.3a-36).

Reflecting on this list, Winograd and Flores remark that "Heidegger recog-
nizedthat ordinary everyday life is like the situation we have been describing.
Our interactions with other people and with the inanimate world we inhabit
put us into a situation of thrownness, for which the metaphor of the meeting
is much more apt than the metaphor of the objective detached scientist who
makes observations, forms hypotheses, and consciously chooses a rational
course of action" (pp. 35-36).
We see many of these same themes of thrownness, ongoing experience,
being in the middle as we move closer to organizations. Langer (1989, p.27),
for example,laments that the world is continuous and dynamic, yet we keep
resorting to absolute categories that ignore large pieces of continuity, thereby
entrapping us in misconceptions. Cohen, March, and Olsen (1972) have
remained sensitive to the reality of continuity, thrownness, and flows in their
insistence that streams of problems, solutions, people, and choices flowthrough
organizations and converge and diverge independent of human intention.
Although they imply that people seldom confuse a problem stream with a
choice dr solution stream, students of sensemaking may be forgiven if they
assume fluidity even in those specifications. The same portion of a flow might
be labeled either a problem or a solution to justify some perceived choice, as
Seven Properties of Sensemaking 45

Starbuck ( 1983) has argued. Eccles and Nohria (1992) describe the context of
managing as the ongoing flow of actions and words in an organization, which
is often punctuated by events such as a product launch, an off-site strategy-
planning exercise, or a budget meeting. These events are important for several
reasons. They

focus and crystallize meanings in organizations. These events also serve as

focal points for the different streams of ongoing activityin the organization.
Although they may often only be ceremonial and not be remembered as
events of any significance, they serve as moments to take stock of ongoing
actions, to spin new stories, to set in motion future actions, to formally
announce beginnings, milestones, and ends, to trigger a change of course,
or just to touch base and reaffirm individual and organizational identities.
(p.48)

If people are in the middle, what are they in the middle of? One answer, as
we saw earlier, is "projects." And if people are in the middle of projects, then
what they see in the world are those aspects that bear on their projects. In other
words, even though people are immersed in flows, theyare seldom indifferent
to what passes them by. This is especially true for interruptions of projects.
The reality of flows becomes most apparent when that flow is interrupted. An
interruption to a flow typically induces an emotional response, which then
paves the way for emotion to influence sensemaking. It is precisely because
ongoing flows are subject to interruption that sensemaking is infused with
feeling.
The relation between sensemaking, emotion, and the interruption of on-
going projects can be understood using ideas proposed by Berscheid (1983)
and Mandler (1984, pp. 180-189). They argue that a necessary condition for
emotion is "arousal" or discharge in the autonomic nervous system. And arousal
is triggered by interruptions of ongoing activity. Arousal has physiological
significance because it prepares people for fight-or-flight reactions. But of
even more importance to both Mandler and Berscheid is the fact that arousal
also has psychological significance. The perception of arousal triggers a rudi-
mentary act of sensemaking. It provides a warning that there is some stimulus
to which attention must be paid in order to initiate appropriate action. This
signal suggests that one's well-being may be at stake.
An important property of arousal is that it develops slowly. Arousal occurs
roughly 2to 3 seconds after an interruption has occurred, and this delay gives
time for an appropriate action to occur. Thus the autonomic system is a back-up
 SENSEMAKING IN ORGANIZATIONS

system that is activated if direct action fails. Once heightened arousal is

perceived, it is appraised, and people try to construct some link benreen the
present situation and 'relevant" prior situations to make sense of the arousal.
Arousal leads people to search for an answer to the question, "What's up?"
Their answers differ depending on socialization (Averill, 1984; Hochschild,
1983; Thoits, 1984).
The variables of arousal and cognitive appraisal are found in many formu-
lations dealing with emotion (e.g., Frijda, 1988), but the unique quality of
Mandler and Berscheid is their focus on the interruption of action sequences
as the occasion for emotion. Standard operating procedures (SOPs) are a good
example of organized action sequences. These procedures tend to become more
tightly organized the more frequently they are executed. The interruption of
an ongoing SOP or project is a sufficient and possibly necessary condition for
autonomic neryous system arousal.
Interruption is a signal that important changes have occurred in the envi-
ronment. Thus a key event for emotion is the "interruption of an expectation."
It makes good evolutionary sense to construct an organism that reacts signifi-
cantly when the world is no longer the way it was.
With these concepts, we can now summarizekey arguments linking emo-
tion with sensemaking. Emotion is what happens between the time that an
organized sequence is interrupted and the time at which the interruption is
removed, or a substitute response is found that allows the sequence to be
completed. Until either event occurs, autonomic arousal increases. When in-
terruption first occurs, there is redoubled effort to complete the original
interrupted sequence. If there are many differentways in which an interrupted
sequence can be completed, then arousal is not likely to build very much. This
suggests that generalists, as well as people who are able to improvise, should
show less emotional behavior and less extreme emotions. People in both of
these categories have more substitute behaviors, so their arousal should not
build to the same high levels that are experienced by people who have fewer
substitute behaviors. Arousal should build more quickly the more tightly
organized an interrupted action sequence is. Finally, the interruption of
higher order, more pervasive plans should be more arousing than the disrup-
tion of lower order plans.
If we apply these propositions to organizations, we start by asking, what is
the distribution of interruption in organizations? Where are interruptions
most liLtly to occur, and how organized are the actions and plans that are likely
to be interrupted? If we can describe this, then we can predict where sense-
making will be especially influenced by emotional experiences. For example,
Swen Properties of Sensemaking

contrary to expectations, systems with newer, less well-organized response

sequenaes, settings with fewer SOPs, and settings that are more loosely cou-
pled should be settings in which interruptions of ongoing projects generate
emotion because interruptions are less disruptive. Settings in which there are
few developed plans should be less interruptible and therefore exhibit less
emotion.
So far we have talked only about the frequency of emotion, not about the
kind of emotion that occurs. Negative emotions are likely to occur when an
organized behavioral sequence is interrupted unexpectedly and the interrup-
tion is interpreted as harmful or detrimental. If there is no mean$ to remove
or circumvent the interruption, the negative emotion should become more
intense, the longer the interruption lasts.
There are at least two possible sources of positive emotion associated with
interruption. First, positive emotion occurs when there is the sudden and
unorpected removal of an interrupting stimulus, such as when a hassling boss
is transferre4 the phone is disconnected, studentsleave cirmpus forthe holidays,
or the records of a collection agency are lost. Second, events that suddenly and
unexpectedly accelerate completion of a plan or behaviord sequense can gen-
erate positive emotions. For example, if you submit a manuscript to a journal,
anticipating that at best you will be invited to attempt several revisions, and
if lour first draft is accepted as is for publication, this is an unexpected interrup-
tion of your plan to write several revisions, but because the interruption
accelerated the completion of a plan, the interruption is a positive experience.
If we now look at emotion in the conte:ct of relationships with other people,
we find that these two sources of positive emotion may change over time.
First, if positive emotions are to occur in a close relationship, then one's
partner must have sufficient resources so that he or she can remove interrupt-
ing stimuli or acceleratethe completion of plans. Howerrer, these acts of removal
and acceleration mustbe unexpected if theyare to generate positive emotion.
Furthermore, an individual must have plans or dreams that he or she cannot
complete alone so that a partner can make a difference. This last condition is
hard to meet in most close relationships because each partner usually drops
plans that cannot be accomptished or accomplishes them by some other means.
If positive emotions are to occur at all, each person needs to keep adding new
plans that cannot be accomplished alone, but they also have to be plans that
the partner cannot predictably accomplish either.
The implications of these propositions about positive emotions for the
development of relationships is sobering. As the other person in the relation-
ship becomes more predictable, and as a partner et(pects that person's help,
there should be fewer occasions for positive emotion to occur.
48 SENSEMAKING IN ORGANIZATIONS

As a relationship develops, it is also often the case that plans come more
and more to include the partner. If this happens, it means that the partner can
always interrupt the completion of plans and cause negative emotions, but
can seldom aid the completion of plans unexpectedly because the help is
always expected and predicted. Thus, in a close relationship, the occasions for
positive emotion decline over time, but the occasions for negative emotion
remain consistently high.
In organizational settings, even though relationships may be short-lived,
they are also often close, intense, and interdependent. In intense, short-lived
organizational relationships, the likelihood of unexpected interruption and
unexpected facilitation are both higher because partners know less about one
another. This could mean that organizationallife generates stronger feelings,
both positive and negative, than is true of other settings (recall the strong
feelings at the Port Authority). People may cope with this volatile emotional
environment by trying to become more self-contained and less dependent on
other individuals, a typical Western resPonse (Markus & Kitayama, 1991).
Autonomous people are hard to interrupt, but they are also hard to help,
which should mean that autonomous people report less emotionality in their
organizational orperience.
If we review the conditions that are necessary to produce positive emotions,
then it looks as if organizational sensemaking should occur largely in con-
junction with negative emotion. None of the conditions for positive affect are
plentiful in most organizations. First, people have little control over the onset
or termination of interruptions. Second, over time people tend to experience
more rather than fewer interrupting stimuli in the form of regulations, deaths,
competitors, takeovers, reorganization, and so on. And third, the achievement
of plans in organizations is more often slowed than accelerated due to, for
example, budget cutting, turnover, resignations, shortages, or currency re-
valuation. Culture may modify all three of these effects, as Van Maanen and
Kunda (1989) show.
To summarize, "emotion is essentially a non-response activity, occurring
between the awareness of the interrupting event and an action alternative that
will maintain or promote the individual's well-being in the face of an event"
(Berscheid, Gangestad, & Kulaskowski, 1983, p. 396). When people perform
an organized action sequence and are interrupted, they try to make sense of
it. The longer they search, the higher the arousal, and the stronger the emotion.
If the interruption slows the accomplishment of an organized sequence, people
are likely to experience anger. If the interruption has accelerated accomplish-
ment, then they are likely to experience pleasure. If people find that the
Seven Properties of Sensemaking

interruption can be circumvented, they experience relief. If they find that the
interruption has thwarted a higher level plan, then the anger is likely to turn
into rage, and if they find that the interruption has thwarted a minor behav-
ioral sequence, then they are likely to feel irritated.
These emotions affect sensemaking because recall and retrospect tend to be
mood congruent (Snyder & White, 1982). People remember events that have
the same emotional tone as what they currently feel. Anger at being inter-
rupted should encourage recall of earlier events where feelings of anger were
dominant. These earlier moments of anger should stand outwhen people look
back over their past experience to discover "similar" events and what those
previous events might suggest about the meaning of present events. Past
events are reconstructed in the present as explanations, not because they look
the same but because they feel the same. The resulting attempt to use a feeling-
based memory to solve a current cognitive puzzte may make sensemaking
more difficult because it tries to mate two very different forms of evidence. It
is precisely that possibility that we watch for when we acknowledge that
sensemaking is ongoing and neither starts fresh nor stops cleanly.

6. Focused on and by Extracted Cues

Itseems like people can make sense of anything. This makes life easy for
people who study sensemaking in the sense that their phenomenon is every-
where. But effortless sensemaking is also a curse for investigators because it
means that they are more likely to see sense that has already been made than
to see the actual making of it. Sensemaking tends to be swift, which means we
are more likely to see products than process. To counteract this, we need to
watch how people deal with prolonged ptzzlesthat defy sensemaking, puzzles
such as paradoxes, dilemmas, and inconceivable events. We also need to pay
close attention to ways people notice, extract cues, and embellish that which
they extract.
James (1890/1950, Vol. 2, pp.340-343) pointed to the importance of ex-
tracted cues for sensemaking in his discussion of the "two great points of
reasoning." The points were, "first, an extracted character [cue] is taken as
equivalent to the entire datum from which it comes." As an example, James
suggests that if hewere offered apiece of cloth he mightrefuse to buyit, saying,
"It looks as if it will fade." If that judgment were made because the person
knew that the color of the cloth was secured by a dye that was chemically
unstable, and that this meant the color would fade, then the notion of the dye,
 SENSEMAKING IN ORGANIZATIONS

which is just one of many parts of the cloth, is the extracted cue from which
the character of the cloth itself is constructed.
The second point of reasoning is that the extracted character "thus taken
suggests a certain consequence more obviously than it was suggested by the
total datum as it originally came" (p. 3a0). The extracted character of the dye
suggested that the cloth would last for a relatively short time, a consequence
that could not be derived from mere inspection of the cloth itself. The ex-
tracted cue highlighted a distinct implication that was invisible in the undif-
ferentiated object.
Extracted cues are simple, familiar structures that are seeds from which
people develop a larger sense of what maybe occurring. The importance of these
cues in organizational analysis was recognized by Smircich and Morgan
(1982) when they said that "leadership lies in large part in generating a point
of reference, againstwhich a feelingof organization anddirection can emerge"
(p. 258). They argue that control over which cues will serve as a point of
reference is an important source of power. To establish a point of reference-
for example, to direct people's attention to the dye in a cloth rather than to
the density of its weave to infer value-is a consequential act.
In the preceding paragraph, I intentionally used the metaphor of "seed" to
capture the open-ended qualityof sensemakingwhen extracted cues are used.
When people act, for example, when they produce sentences, there is a duality
of structure in what they produce (Shotter, 1983, pp.2S-3D. The partially
completed sentence contains both content already specified and the means for
continuation of the sentence. The partial sentence limits the number of ways
in which the remainder of the sentence can be finished, but there is still some
latitude as to which of several possibilities will actually be realized. The
beginnings of the sentence reduce, but do not remove, indeterminacy.
The importance of dl this for students of sensemaking, and for the value
of the metaphor of seed, is that

the production of a sentence is hardly different in character from the

growth of a plant. And it will often be useful to bear this image in mind: for
the relation of, say, a person's intention of saying something to their saylng
it, is much more like the relation of seed to plant, than that suggested by
the currently more popular image of script to its performance. For rather
than being the outer expression of something already specified internally,
the expression of an intention is, as a process of temporal unfolding, a
passage from an indeterminate to a more well-articulated state of affairs.
(Shotter, 1983, p.29)
Seven Properties of Sensemaking 51

A seed is a form-producing process that captures much of the vagueness

and indeterminary of sensemaking. The actions of a seed resemble those of
the documentary method (Garfinkel ,1967 ,P.78;Weick, I993a,pp- 26-29). A
specific observation becomes linked with a more general form or idea in the
interest of sensemaking, which then clarifies the meaning of the particular,
which then alters slightlythe general, and so on. The abstract and the concrete
inform and construct one another. Actions create the conditions for further
action (Shotter, 1993, p. 156), the course of which remains vague ProsPec-
tively, but clearer in retrospect.
Shotter (1983) develops the seed metaphor more fully when he notes,

fust intention may be said to "contain" or "point to" its object, so an

as an
acorn may be said to "contain" or "point to" an oak tree. But an acorn certainly
does not contain an oak tree, or anything like it, even in miniature (prefor-
mationism is not true). It is best seen as the structured medium or means
through which, in interaction with its surroundings, an oak tree forms,
developing itself through its own progressive self-specification. Further-
more, although an acorn specifies the production of an oak tree, and not
any other kind of tree, it does not specifr the tree that grows from it exactly
(not the number of branches, twigs,leaves, etc.), for the tree grows in a quite
unpredictable manner, sensitive to local contingencies. Similarly, an inten-
tion may specifu a whole range of possible expressions, the actual one
realized being formulated (progressively) in interaction with its circum-
stances. (pp.29-30)

What an extracted cue will become depends on context ("local contingen-

cies") in two important ways. First, context affects what is extracted as a cue
in the first place, a process that has variously been described in the organiza-
tional literature as search (Cyert & March, 1953), scanning (Daft &Weick, 1984)'
and noticing (Starbuck & Milliken, 1988). The concept of frame (e.9., Goffinan,
lg74\ is used as shorthand for the structure of context. Second, context also
affects how the extracted cue is then interpreted, a stage that has been a
primary focus of ethnomethodologists in their discussions of "indexicals"
(Iriter, 1980; see also Rittg & Van de Ven, 1989, p. 1S1). We explore briefly each
of these roles of context.
The process of noticing, by which cues are extracted for sensemaking, has
been discussed by Starbuck and Milliken, who distinguish noticing from
sensemaking. To them, noticing refers to the activities of filtering, classifying,
and comparing, whereas sensemaking refers more to interpretation and the
activity of determining what the noticed cues mean.
52 SENSEMAKING IN ORGANIZAT10NS

As they put it, "sensemaking focuses on subtleties and interdependencies,

whereas noticing pict<s up major events and gross trends' Noticing determines
whether people even consider responding to environmental events. If events
are noticed, people make sense of them; and if events are not noticed,
they are
not available for sensemaking" (Starbuck & Milliken, 1988, P. 60)'They prefer
the term noticingto scanningbecause the former term implies a more
infor-
mal, more involuntary "beginning" to the process of sensemaking. scanning,
by contrast, sounds more strategic, more conscious, more deliberate' more
and
under the control of preconceptions, and less open to invention' Fiske
Taylor (1991), reviewing the social cognition literature, conclude that among
the things we notice are "things that are novel or PercePtually figural in
that
context, people or behaviors that are unusual or unexpected, behaviors
are extreme and (sometimes) negative, and stimuli relevant to our
current
goals. . . . Our attention also orients us to situationally or personally
primed
categories. Recently, frequently, and chronically encountered categories
are

more accessible for use, and they profoundly influence the encoding of
stimuli" (pp. 265-265). Kiesler and sproull's (1982) influential essay on
problem sensing anticipated many of these conclusions as is evident in this
description:

people attend to and encode salient material-events that are unpleasant,

deviant, extreme, intense, unusual, sudden, brightly lit, colorful, alone, or
sharply drawn. In the world of organizations, salient information includes
otr.ttti.ipated drains on cash flow, new taxes and regulations (unpleasant
information), predictions of best and worst outcomes (extreme informa-
tion), disruptions of routine and emergencies (intense, unusual, sudden
information), and publicity and iconoclastic executives (colorful informa-
tion). The behavior and outcomes of competitors, of course, are sharply
drawn-a figure against ground. (p. 556)

Atthough both lists are complex and long, they do at least make it clear that
context affects the extraction of cues, and that small, subtle features can have
surprisingly large effects on sensemaking. The importance of context becomes
even clearer when we examine indexicals and what happens to cues after they
are extracted.
Leiter (1980) describes the idea of cues and indexicals this way:

Indexicality refers to the contextual nature of objects and events. That is to

say, without a supplied context, objects and events have equivocal or
multiple meanings. The indexical property of talk is the fact that people
Seven Properties of Sensemaking 53

routinely do not state the intended meaning of the expressions they use.
The expressions are vague and equivocal, lending themselves to several
meanings. The sense or meaning of these expressions cannot be decided
unless a context is supplied. That context consists of such particulars as who
the speaker is (his biography), the relevant aspects of his biography, his
current purpose and intent, the setting in which the remarks are made or
the actual, or potential relationship between speaker and hearer. (p. 107)

Thus, not only do students of sensemaking need to be closely attuned to

the social, they also need to think context (e.g., Mowday & Sutton, 1993). And
this is especially true in organizations, as several authors have noted. Salancik
and Pfeffer (1978,p.233), for example, argue that the social context is crucial
for sensemaking because it binds people to actions that they then must justifr,
it affects the saliency of information, and it provides norms and expectations
that constrain explanations. Mailloux (1990) adds the fact that context incor-
porates politics: "Interpretations can have no grounding outside of rhetorical
exchanges taking place within institutional and cultural politics" (p. 133). To
talk about interpretation without discussing a politics of interpretation is to
ignore context. This is evident also in the discussions by Starbuck and Milliken,
who argue that people in organizations are in different locations and are
familiar with different domains, which means they have different interpreta-
tions of common events. When these conflicting interpretations are aired,
they create political struggles, as Hall ( 1984) has shown clearly in the contested
interpretations at the Saturday Evening Posr during its waning years.
As an example of the political struggles that interact with choices of strategy
and organizational design, Starbuck and Milliken (1988) discuss the different
views of people located at different levels in a hierarchy:

People with expertise in newer tasks tend to appear at the bottoms of

hierarchies and to interpret events in terms of these newer tasks and they
welcome changes that will offer them promotion opportunities and bring
their expertise to the fore. Conversely, people at the tops of organizational
hierarchies tend to have expertise related to older and more stable tasks,
they are prone to interpret events in terms of these tasks, and they favor
strategies and personnel assignments that will keep these tasls central. (p. 53)

But regardless of the cues that become salient as a consequence of context,

and regardless of the way those extracted cues are embellished, the point to
be retained is that faith in these cues and their sustained use as a reference
point are important for sensemaking. The importance lies in the fact that these
54 SENSEMAKING IN ORGANIZATIONS

cues tie elements together cognitively. These presumed ties are then given
more substance when people act as if theyare real (Weick, 1983, pp.228-23D\.
A presumed order becomes a tangible order when faith is followed by
enactment. Many people have said as much. "And so it is with all things. If you
are not h"ppy, act the h"ppy man. Happiness will come later. So also with faith.
If you are in despair, act as though you believed. Faith will come afterwards"
(Singer, L96L,p. Iaa). As I noted earlier in Section 3, James (1885/1956)
described how faith that life is worth living generates the action that then
makes life worth living.
Because extracted cues are crucial for their capacity to evoke action, Proc-
esses of sensemaking tend to be forgiving. Almost any point of reference will
do, because it stimulates a cognitive structure that then leads people to act
with more intensity, which then creates a material order in place of a presumed
order (Weick, 1983). This sequence resembles the sequence often described
as a self-fulfilling prophecy. An extracted cue is used to prophesy the nature
of the referent from which it was extracted. When the person acts confidently,
as if that malleable referent has the character inferred from the cue, the
referent often is shaped in directions consistent with the prophecy. But the
prophecy itself is also "adjusted." Each element, the prophecy and the referent,
is informed by and adjusted to the emerging picture of the other. As a result,
almost any old point of reference will do as a start. This conclusion has been
"immortalized" for me in a story that I haul out almost every chance I get (e.g.,
Weick, 1990) because it captures a truth about sensemaking.
This incident, related bythe Hungarian Nobel Laureate Albert Szent-Gyorti
and preserved in a poem by Holub (1977),happened during military maneu-
vers in Switzerland. The young lieutenant of a small Hungarian detachment
in the Alps sent a reconnaissance unit into the icywilderness. It began to snow
immediately, snowed for 2 days, and the unit did not return. The lieutenant
suffered, fearing that he had dispatched his own people to death. But on the
third day the unit came back. Where had theybeen? How had they made their
way? Yes, they said, we considered ourselves lost and waited for the end. And
then one of us found a map in his pocket. That calmed us down. We pitched
carnp, lasted out the snowstorm, and then with the map we discovered our
bearings. And here we are. The lieutenant borrowed this remarkable map and
had a good look at it. He discovered to his astonishment that it was not a map
of the Alps, but a map of the Pyrenees.
This incident raises the intriguing possibility that when you are lost, any
old map will do. For example, extended to the issue of strategy, maybe when
you are confused, any old strategic plan will do. Strategic plans are a lot like
Seven Properties of Sensemaking 55

maps. They animate and orient people. Once people begin to act (enactment),
theygenerate tangible outcomes (cues) in some context (social), and this helps
them discover (retrospect) what is occurring (ongoing), what needs to be
explained (plausibility), and what should be done next (identity enhance-
ment). Managers keep forgetting that it is what they do, not what they plan,
that explains their success. They keep giving credit to the wrong thing-
namely, the plan-and having made this error, they then spend more time
planning and less time acting. They are astonished when more planning
improves nothing (Starbuck, 1993).
When I described the incident of using a map of the Pyrenees to find a way
out of the Alps to Bob Engel, the executive vice president and treasurer of
Morgan Guaranty, he said, "Now, that storywould have been really neat if the
leader out with the lost troops had known it was the wrong map and still been
able to lead them back." What is interesting about Engel's twist to the story is
that he has described the basic situation that most leaders face. Followers are
often lost and even the leader is not sure where to go. All the leaders know is
that the plan or the map they have in front of them are not sufficient to get
them out. What the leader has to do, when faced with this situation, is instill
some confidence in people, get them moving in some general direction, and
be sure they look closely at cues created by their actions so that they learn
where they were and get some better idea of where they are and where they
want to be.
The soldiers were able to produce a good outcome from a bad map because
theywere active, theyhadapurpose (getbackto camp), and theyhad an image
of where they were and where they were going. They kept moving, they kept
noticing cues, and they kept updating their sense of where they were. As a
result, an imperfect map proved to be good enough. The cues they extracted
and kept acting on were acts of faith amid indeterminacy that set sensemaking
in motion. Once set in motion, sensemaking tends to confirm the faith
through its effects on actions that make material that previously had been
merely envisioned.

7. Driven by Plausibility Rather Than Accuracy

The prefix sensein the word sensemakingis mischievous. It simultaneously

invokes a realist ontology, as in the suggestion that something is out there to
be registered and sensed accurately, and an idealist ontology, as in the sugges-
tion that something out there needs to be agreed on and constructed plausibly.
The sensible need not be sensable, and therein lies the trouble.
 SENSEMAKING IN ORGANIZATIONS

A reasonable position to start from in studies of sensemaking is to argue

that accuracy is nice, but not necessary. Isenberg's ( 1986) studies of manage-
rial thinking show the importance of plausible reasoning, which he describes
this way:

Plausible reasoning involves going beyond the directly observable or at least

consensual information to form ideas or understandings that provide
enough certainty. . . . There are several ways in which this process departs
from a logical-deductive process. First, the reasoning is not necessarily
correct, but it fits the facts, albeit imperfectlyat times. Second, the reasoning
is based on incomplete information. (pp.2a2-2a3)

The earlier example of using a map of the Pyrenees to find a path through the
Alps fits Isenberg's description perfectly.
In a related vein, Starbuck and Milliken (1988) observe that "one thing an
intelligent executive does not need is totally accurate perception" (p. 40). This
is a fortunate conclusion in a way, because evidence is beginning to show that
executives are not always accurate anryay in their perceptions of their organi-
zations and their environments. Sutcliffe (L994), in an important emerging
line of work, has shown that accurate perceptions of environmental variation
and environmental munificence are affected by different managerial and
organizational factors. Accurate noticing of variation is associated with scan-
ning and decentralization, suggesting that the breadth and variety of infor-
mational inputs are crucial. By contrast, depth and integration of team
information processing, as represented by length of team tenure, are associ-
ated with more accurate perception of resource levels. Of particular interest
to people like myself (see p.34 in this chapter and p. 89 in Chapter 4), who
swear by the concept of requisite variety (e.g., complicate yourself if you want
to understand complicated environments), is Sutcliffe's finding that the more
diverse the work history of a top management team, the less accurate is the
team in noticing munificence. Aside from this intriguing pazzle, Sutcliffe
(1994) raises the possibility that inaccurate perceptions, under some condi-
tions, may lead to positive consequences:

Misperceptions may be beneficial if they enable managers to overcome

inertial tendencies and propel them to pursue goals that might look unat-
tainable in environments assessed in utter objectivity. Because environ-
ments aren't seen accurately, managers may undertake potentially difficult
courses of action with the enthusiasm, effort, and self-confidence necessary
Seven Properties of Sensemaking 57

to bring about success. Having an accurate environmental map maybe less

important than having some map that brings order to the world and
prompts action. (p. 1.37a)

Even if acctrracy were important, executives seldom produce it. From the
standpoint of sensemaking, that is no big problem. The strength of sensemak-
ing as a perspective derives from the fact that it does not rely on accuracy and
its model is not object perception.Instead, sensemaking is about plausibility,
pragmatics, coherence, reasonableness, creation, invention, and instrumentality.
Sensemaking, to borrow Fiske's (1992) imagery, "takes a relative approach to
truth, predicting that people will believe what can account for sensory expe-
rience but what is also interesting, attractive, emotionally appealing, and goal
relevant" (p. 879).
The criterion of accuracy is secondary in any analysis of sensemaking for a
variety of reasons. First, people need to distort and filter, to separate signal
from noise given their current projects, if they are not to be overwhelmed
with data (Miller, 1978, chap. 5). Thus, from the standpoint of sensemak-
ing, it is less productive to follow the lead of behavioral decision theorists
(e.9., Kahnemann, Tversky, Thaler) who gloat over the errors, misperceptions,
and irrationalities of humans, and more productive to lookat the filters people
invoke, why they invoke them, and what those filters include and exclude (e.g.,
Gigerenzer, 1991; Smith & Kida, 1991).
Second, sensemaking is about the embellishment and elaboration of a single
point of reference or extracted cue. Embellishment occurs when a cue is linked
with a more general idea. Because "objects" have multiple meanings and signifi-
cance, it is more crucial to get some interpretation to start with than to
postpone action until "the" interpretation surfaces. Given multiple cues, with
multiple meanings for multiple audiences, accurate perception of "the" object
seems like a doomed intention. Making sense of that object, however, seems
more plausible and more likely. Perhaps the most common linkage is that of
a present cue with a "similar" interpreted cue from the past. But pasts are
reconstructions (Bartlett, 1932), which means they never occurred precisely
the way they are remembered. Thus accuracy is meaningless when used to
describe a filtered sense of the present, linked with a reconstruction of the
past, that has been edited in hindsight.
Most organizational action is time sensitive, which means that in a speed/
accuracy trade-off (Fiske, 1992), managers favor speed. Thus a third reason
why accuracy is secondary is that speed often reduces the necessity for accu-
racy in the sense that quick responses shape events before they have become
 SENSEMAKING IN ORGANIZATIONS

crystallized into a single meaning. A fast response can be an influential response

that enacts an environment. Aside from this, we might expect that speed rather
than a "constant close look" would dominate whenever anyone has to adapt
to complex cue patterns. "The cost of close looks is generally too high under
the conditions of speed, risk, and limited capacity imposed upon organisms
by their environment or their constitutions. The ability to use minimal cues
quickly in categorizing the events of the environment is what gives the
organism its lead time in adjusting to events. Pause and close inspection
inevitably cut down on this precious interval for adjustment" (Brun er,1973,
P.30).
We have already seen the importance of an "interval for adjustment" in the
earlier discussion of interruption (p. 45) as the occasion for emotion. And we
see this interval again here. In both cases, the issue is not so much accuracy as
it is the continuation of ongoing projects.
A fourth reason why issues of accuracy do not dominate studies of sense-
making is that, if accuracy does become an issue, it does so for short periods
of time and with respect to specific questions. Swann (1984) has preserved
this point in his distinction between global accuracy and circumscribed
accuracy. Global accuraqy comes into playwhen perceivers are concernedwith
forming widely generalizable beliefs.In the case of interpersonal perception,
for example, "the global accuracy of a belief will be high insofar as it enables
the perceiver to predict the behavior of the target in the presence of all the
perceivers that target encounters (transpersonal accuracy), across all the
contexts that target enters (transcontextual accuracy), or across a fairly long
period of time (extended accuracy)" (swann, 1984, p. a6\. Global accuracy
is what investigators usually attempt to tap when investigating the accuracy
of executive perceptions.
Circumscribed accuracy, as the name implies, is less sweeping and is focused
on prediction of specific encounters in a limited number of contexts for a brief
period. In a rapidly changing ongoing stream of activiry circumscribed
accutacy seems to be the most one can hope for, in the event that accuracy
even becomes a concern.A firm goingbankrupt (D'Aveni &MacMillan, 1990)
worries less about industry trends and the stability of the environment than
it does about debt service, cash flow, and meeting the payroll.
Our repeated reference to the interpersonal, interactive, interdependent
quality of organizational life can be interpreted as a fifth reason why accuracy
is not the sole concern in sensemaking. The criterion of accuracy makes more
sense when investigators study object perception rather than interpersonal
perception. The hallmark of object perception is the assumption of stimulus
Seven Properties of Sensemaking

constancy: "the notion that the targets of perceptual activity possess identities
that are immutable and constant" (Swann, 1984, p. a60). Investigators and
practitioners alike would die for pockets of stimulus constancy in the flow of
organizational life, but they seldom find them. What they do find are mercu-
rial stimuli that mimic the inherent equivocality (Bruner,1973) of interper-
sonal perception.
As we saw earlier, personal identities are shifting and multiple. And when
those shifting identities are embodied in members of the top-management
team (a frequent target for accuracy studies), outside observers who try to
predict the behavior of this team using the model of object perception are in
trouble. What they miss is that when teams try to assess industry trends,
they act more like they are perceiving people than objects (e.9., Hambrick,
Geletkanycz, & Fredrickson, 1993). Executives personalize the question of
what are the trends in this industryand subdivide into questions of for example,
who makes markets,what drives those people,where are theyvulnerable,what
is their track record, and what are they up to. Questions in organizations that
look lilc they involve global accuracy and object perception tend to be translated
into questions of intentions and personalities. Such translation means that
interpersonal perception is the better model than is object perception.
We have talked throughout about the important effect projects have on
sensemaking. That ongoing effect provides the background for the sixth
reason that themes of accuracyseldom dominate discussions of sensemaking.
Accuracy is defined by instrumentality. Beliefs that counteract interruptions
and facilitate ongoing projects are treated as accurate. Accuracy, in other words,
is project specific and pragmatic. fudgments of accuracy lie in the path of the
action. G. Stanley Hall put this point well in 1878. Writing in the October issue
of the journal Mind, he said,

All possible truth is practical. To ask whether our conception of chair or

table corresponds to the real chair or table apart from the uses to which they
may be put, is as utterly meaningless and vain as to inquire whether a
musical tone is red or yellow. No other conceivable relation than this
between ideas and things can exist. The unknowable is what I cannot react
upon. The active part of our nature is not only an essential part of cognition
itself, but it always has a voice in determining what shall be believed and
what rejected. (Hall, cited in Sills & Merton, 1991, p. 84)

Enactment in the pursuit of projects provides the frame within which cues
are extracted and interpreted. This same frame circumscribes the area within
 SENSEMAKING IN ORGANIZAT10NS

which accuracy matters. And the action repertoire that lies behind the framing
itself implies what can and cannot be known. Again, people see and find
sensible those things they can do something about. Capabilities for action
affect what is believed and what is rejected. What is believed as a consequence
of action is what makes sense. Accuracy is not the issue.
The seventh reason why accuracy plays a secondary role in analyses of
sensemaking is that stimuli that are filtered out are often those that detract
from an energetic, confident, motivated response. Accurate perceptions have
the power to immobilize. People who want to get into action tend to simplify
rather than elaborate. This point was made convincingly by Brunsson ( 1982)
when he contrasted action rationality with decision rationality. Biased notic-
ing may be bad for deliberation, but it is good for action. In a world that is
changing and malleable, confident, bold, enthusiastic action, even if it is based
on positive illusions (Taytor, 1989), can be adaptive. Bold action is adaptive
because its opposite, deliberation, is futile in a changing world where PerceP-
tions, by definition, can never be accurate. They can never be accurate because,
by the time people notice and name something, it has become something else
and no longer exists.
Bold action is also adaptive because it shapes that which is emerging (e.g.,
Lanir et al., 1988). Events are shaped toward those capabilities the bold actor
already has. With this twist, accuracybecomes reflexive. The actor who knows
what he or she can do, and who shapes the environment so that it needs
precisely these capabilities, comes close to perfect accuracy. People construct
that which constructs them, except both constructions turn out to be one and
the same thing. Although individuals maybe blind to this dynamic, what they
see as a result of its unfolding looks eminently sensible. Both the construction
and the perception reflect the same set of assumptions about capability.
Because accuracy is automatic, it drops out of consideration.
The eighth, and final, reason why accuracy is nice but not necessary is that
it is almost impossible to tell, at the time of perception, whether the percep-
tions will prove accurate or not. This is so "because perceptions are partly
predictions that may change realiry because different predictions may lead to
similar actions, and because similar perceptions may lead to different actions.
Many perceptual errors, perhaps the great majority, become erroneous only
in retrospect" (Starbuck & Milliken, 1988, P.44).
If accuracy is nice but not necessary in sensemaking, then what is necessary?
The answer is, somethingthatpreserves plausibilityand coherence, something
that is reasonable and memorable, something that embodies past experience
and expectations, something that resonates with other people, something that
Seven Properties of Sensemoking 61

can be constructed retrospectively but also can be used prospectively, some-

thing that captures both feeling and thought, something that allows for
embellishment to fit current oddities, something that is fun to construct. In
short, what is necessary in sensemaking is a good story.
A good story holds disparate elements together long enough to energize
and guide action, plausiblyenough to allow people to make retrospective sense
of whatever happens, and engagingly enough that others will contribute their
own inputs in the interest of sensemaking. Stories will be discussed later
(Chapter 5) in the context of the substance of sensemaking.
The point we want to make here is that sensemaking is about plausibility,
coherence, and reasonableness. Sensemaking is about accounts that are socially
acceptable and credible. Stated differently, "filtered information is less accu-
rate but, if the filtering is effective, more understandable" (Starbuck & Milliken,
1988, p. 41). It would be nice if these acceptable accounts were also accurate.
But in an equivocal, postmodern world, infused with the politics of interpre-
tation and conflicting interests and inhabited by people with multiple shifting
identities, an obsession with accuracy seems fruitless, and not of much prac-
tical help, either. Of much more help are the symbolic trappings of sensemak-
ing, trappings such as myths, metaphors, platitudes, fables, epics, and para-
digms (see Gagliardi, 1990). Each of these resources contains a good story.
And a good story, like a workable cause mop, shows patterns that may already
exist in the puzzles an actor now faces, or patterns that could be created anew
in the interest of more order and sense in the future. The stories are templates.
They are products of previous efforts at sensemaking. They explain. And they
energize.And those are two important properties of sensemakingthatwe remain
attentive to when we look for plausibility instead of accuracy.

Summary

The recipe "how can I know what I think until I see what I say?" can be parsed
to show how each of the seven properties of sensemaking are built into it.

l. Identity: The recipe is a question about who I am as indicated by discovery of

how and what I think.
2. Retrospect: To learn what I think, I look back over what I said earlier.
3. Enactment: I create the object to be seen and inspected when I say or do
something.
62 SENSEMAKING IN ORGANIZATIONS

4. Social: What I say and single out and conclude are determined by who socialized
me and how I was socialized, as well as by the audience I anticipate will audit
the conclusions I reach.
5。 ongoing: My talking is spread across time, competes for attention with other
ongoingprojects, and is reflected on after it is finished,which means my interests
may already have changed.
6. Extracted cues: The 'what" that I single out and embellish as the content of the
thought is only a small portion of the utterance that becomes salient because of
context and personal dispositions.
7. Plausibility: I need to know enough about what I think to get on with my projects,
but no more' which means sufficiency and plausibility take precedence over
accuracy.

The close fit between the recipe and the seven properties remains if one or
more of the Pronouns in the recipe is changed to reflect a collective actor (e.g.,
how can we know what we think until I see what we say?).