3

Crafting an Analytic
Framework I: Three
Pillars of Institutions
v v v

Men make their own history, but they do not make it just as they
please; they do not make it under circumstances chosen by them-
selves, but under circumstances directly encountered, given, and
transmitted by the past.
—Karl Marx (1852/1963: 15)

T o an institutionalist, knowledge of what has gone before is vital

information. The ideas and insights of our predecessors provide
the context for current efforts and the platform on which we necessar-
ily craft our own contributions. However, as should be clear even from
my brief review, the concepts and arguments advanced by our prede-
cessors have been strikingly diverse, resting on varied assumptions
and privileging differing causal processes. A number of theorists have
proposed that we can clarify the arguments by boiling them down to a
few dominant paradigms (see, e.g., Campbell, 2004; Hall and Taylor,
1996). However, as Campbell observes, these “schools” exhibit as many
55
56   INSTITUTIONS AND ORGANIZATIONS

similarities as differences. Hence, my own approach to bringing some

order into the discussion is to propose a broad definition of institutions
that can encompass a variety of arguments, and then attempt to iden-
tify the key analytic elements that give rise to the most important dif-
ferences observed and debates encountered. This chapter and the next
identify and elucidate the three analytical elements that comprise
institutions. Each element is important, and sometimes one or another
will dominate, but more often—particularly in robust institutional
frameworks—they work in combination. But because each operates
through distinctive mechanisms and sets in motion disparate pro-
cesses, I emphasize their differences in my initial discussion.
After introducing the principal distinctions around which the
analysis will be conducted, I bravely but briefly consider their philo-
sophical underpinnings. Varying conceptions of institutions call up
somewhat different views of the nature of social reality and social
order. Similarly, the institutional elements relate to disparate constructs
of how actors make choices, the extent to which actors are rational, and
what is meant by rationality. These issues, while too complex to fully
explore, are too important to ignore.
The companion chapter, Chapter 4, completes the presentation of
the analytical framework and associated issues. It begins by examining
what types of institutional beliefs support the development of organiza-
tions. I then describe the concept of structuration, which can assist us in
the effort to reconcile institutional constraints with individual agency.
Finally, I identify the multiple levels at which institutional analysis
takes place. It is important to recognize that even if an investigation
focuses on a particular level, institutional forces operating at other
levels—both “above” and “beneath” the level selected—will be at work.
Chapters 3 and 4 should be taken as a prolegomenon to the more
problem-focused, empirically based discussions in the chapters to fol-
low. They introduce concepts and definitions that will be employed to
examine particular topics as well as preview controversies and issues
that will be encountered as we review, in Chapters 5 through 8, develop-
ments in institutional theory and research from the 1970s to the present.

™™DEFINING INSTITUTIONS

Let us begin with the following omnibus conception of institutions:

Institutions comprise regulative, normative, and cultural-cognitive ele-

ments that, together with associated activities and resources, provide
stability and meaning to social life.
 Crafting an Analytic Framework I: Three Pillars of Institutions    57

This is a dense definition containing a number of ideas that we will

unpack, describe, and elaborate in this chapter and the next. In this
conception, institutions are multifaceted, durable social structures,
made up of symbolic elements, social activities, and material resources.
Institutions exhibit distinctive properties: They are relatively resistant
to change (Jepperson 1991). As Giddens (1984: 24) states, “Institutions
by definition are the more enduring features of social life . . . giving
‘solidity’ [to social systems] across time and space.” They can be trans-
mitted across generations, maintained and reproduced (Zucker 1977).
Institutions also undergo change over time.
Institutions exhibit stabilizing and meaning-making properties
because of the processes set in motion by regulative, normative, and
cultural-cognitive elements. These elements are the central building
blocks of institutional structures, providing the elastic fibers that
guide behavior and resist change. We examine the distinctive nature
and contribution of each element in a subsequent section of this
chapter.
Although symbolic systems—rules, norms, and cultural-cognitive
beliefs—are central ingredients of institutions, the concept must also
encompass associated behaviors and material resources. Although an
institutional perspective gives heightened attention to the symbolic
aspects of social life, we must also attend to the activities that produce,
reproduce, and change them and to the resources that sustain them.
Institutions are, in Hallett and Ventresca’s (2006) useful metaphor,
inhabited by people and their interactions. Rules, norms, and mean-
ings arise in interaction, and they are preserved and modified by
human behavior. To isolate meaning systems from their related behav-
iors is, as Geertz (1973) cautions, to commit the error of

locking cultural analysis away from its proper object, the infor-
mal logic of actual life. . . . Behavior must be attended to, and
with some exactness, because it is through the flow of behavior—
or, more precisely, social action—that cultural forms find
articulation. . . . Whatever, or wherever, symbol systems ‘in their
own terms’ may be, we gain empirical access to them by inspect-
ing events, not by arranging abstracted entities into unified
patterns. (p. 17)

Similarly, for Berger and Luckmann (1967) institutions are

“dead” if they are only represented in verbal designations and in
physical objects. All such representations are bereft of subjective
reality “unless they are ongoingly ‘brought to life’ in actual human
conduct” (p. 75).
58   INSTITUTIONS AND ORGANIZATIONS

Sociological theorists Giddens (1979; 1984) and Sewell (1992)

underline the importance of including resources—both material and
human—in any conception of social structure so as to take into account
asymmetries of power. Rules and norms, if they are to be effective, must
be backed with sanctioning power, and cultural beliefs, or schemas in
Sewell’s terminology, to be viable, must relate to and are often embodied
in resources. Conversely, those possessing power in the form of excess
resources seek authorization and legitimation for its use. As Sewell
observes, “Schemas not empowered or regenerated by resources would
eventually be abandoned and forgotten, just as resources without cultural
schemas to direct their use would eventually dissipate and decay” (p. 13)
The Giddens/Sewell formulation usefully stresses the duality of
social structure, encompassing both idealist and material features of
social life and highlighting their interdependence, an argument I elabo-
rate in Chapter 4.
Most treatments of institutions emphasize their capacity to control
and constrain behavior. Institutions impose restrictions by defining
legal, moral, and cultural boundaries, distinguishing between accept-
able and unacceptable behavior. But it is equally important to recog-
nize that institutions also support and empower activities and actors.
Institutions provide stimulus, guidelines, and resources for acting as
well as prohibitions and constraints on action.
Although institutions function to provide stability and order, they
themselves undergo change, both incremental and revolutionary. Thus,
our subject must include not only institutions as a property or state of
an existing social order, but also institutions as process, including the
processes of institutionalization and deinstitutionalization (see Tolbert
and Zucker, 1996). Scholars increasingly attend not only to how institu-
tions arise and are maintained, but to how they undergo change. As we
will see, much of the impetus for change occurs through endogenous
processes, involving conflicts and contradictions between institutional
elements, but institutions can also be destabilized by exogenous
shocks, such as wars and financial crises.
Institutions ride on various conveyances and are instantiated in
multiple media. These institutional carriers vary in the processes
employed to transmit their messages. In addition, institutions operate
at multiple levels—from the world system to interpersonal interaction.
We examine these diverse carriers and levels in Chapter 4.
Important differences exist among the various schools of institu-
tional scholars, as is apparent from our review of previous work in
Chapters 1 and 2. In my view, the most consequential dispute centers
on which institutional elements are accorded primacy.
 Crafting an Analytic Framework I: Three Pillars of Institutions    59

™™THE THREE PILLARS OF INSTITUTIONS

Regulative systems, normative systems, cultural-cognitive systems—

each of these elements has been identified by one or another social
theorist as the vital ingredient of institutions. The three elements form
a continuum moving “from the conscious to the unconscious, from the
legally enforced to the taken for granted” (Hoffman 1997: 36). One
possible approach would be to view all of these facets as contributing,
in interdependent and mutually reinforcing ways, to a powerful social
framework—one that encapsulates and exhibits the celebrated strength
and resilience of these structures. In such an integrated conception,
institutions appear, as D’Andrade (1984: 98) observes, to be overdeter-
mined systems: “overdetermined in the sense that social sanctions
plus pressure for conformity, plus intrinsic direct reward, plus values,
are all likely to act together to give a particular meaning system its
directive force.”
While such an inclusive model has its strengths, it also masks
important differences between the elements. The definition knits
together three somewhat divergent conceptions that need to be differ-
entiated. Rather than pursuing the development of a more integrated
conception,1 I believe that more progress will be made at this juncture
by distinguishing among the several component elements and identi-
fying their different underlying assumptions, mechanisms and indica-
tors.2 By employing a more analytical approach to these arguments, we
can separate out the important foundational processes that transect the
domain.
Consider Table 3.1. The columns contain the three elements—three
pillars—identified as making up or supporting institutions. The rows
define some of the principal dimensions along which assumptions
vary and arguments arise among theorists emphasizing one or another
element. This table will serve as a guide as we consider each element.

The Regulative Pillar

In the broadest sense, all scholars underscore the regulative aspects
of institutions: Institutions constrain and regularize behavior. Scholars
more specifically associated with the regulatory pillar are distin-
guished by the prominence they give to explicit regulatory processes—
rule-setting, monitoring, and sanctioning activities. In this conception,
regulatory processes involve the capacity to establish rules, inspect
others’ conformity to them, and, as necessary, manipulate sanctions—
rewards or punishments—in an attempt to influence future behavior.
60   INSTITUTIONS AND ORGANIZATIONS

Table 3.1  Three Pillars of Institutions

Regulative Normative Cultural-Cognitive

Basis of Expedience Social obligation Taken-for-grantedness
compliance Shared understanding
Basis of order Regulative rules Binding Constitutive schema
expectations
Mechanisms Coercive Normative Mimetic
Logic Instrumentality Appropriateness Orthodoxy
Indicators Rules Certification Common beliefs
Laws Accreditation Shared logics of
Sanctions action
Isomorphism
Affect Fear Guilt/ Shame/Honor Certainty/Confusion
Innocence
Basis of Legally Morally Comprehensible
legitimacy sanctioned governed Recognizable
Culturally supported

Sanctioning processes may operate through diffuse, informal mecha-

nisms, involving folkways such as shaming or shunning activities, or
they may be highly formalized and assigned to specialized actors such
as the police and courts. Political scientists examining international
institutions point out that legalization—the formalization of rule
systems—is a continuum whose values vary along three dimensions:

•• obligation—the extent to which actors are bound to obey because

their behavior is subject to scrutiny by external parties
•• precision—the extent to which the rules unambiguously specify
the required conduct
•• delegation—the extent to which third parties have been granted
authority to apply the rules and resolve disputes (Abbott,
Keohane, Moravcsik, Slaughter, and Snidal 2001)

I suggest that regulatory systems are those that exhibit high values
on each of these dimensions while normative systems, considered
below, exhibit lower values on them.
Economists, including institutional economists, are particularly
likely to view institutions as resting primarily on the regulatory pillar.
A prominent institutional economist, Douglass North (1990), for
example, features rule systems and enforcement mechanisms in his
conceptualization:3
 Crafting an Analytic Framework I: Three Pillars of Institutions    61

[Institutions] are perfectly analogous to the rules of the game in a

competitive team sport. That is, they consist of formal written
rules as well as typically unwritten codes of conduct that underlie
and supplement formal rules. . . . [T]he rules and informal codes
are sometimes violated and punishment is enacted. Therefore, an
essential part of the functioning of institutions is the costliness of
ascertaining violations and the severity of punishment. (p. 4)

North’s emphasis on the more formalized control systems may

stem in part from the character of the customary objects studied by
economists and rational choice political scientists. They are likely to
focus attention on the behavior of individuals and firms in markets or
on other competitive situations, such as politics, where contending
interests are more common and, hence, explicit rules and referees more
necessary to preserve order. These economists and political scientists
view individuals and organizations that construct rule systems or con-
form to rules as pursuing their self-interests as behaving instrumen-
tally and expediently. The primary mechanism of control involved,
employing DiMaggio and Powell’s (1983) typology, is that of coercion.
Although the concept of regulation conjures up visions of repres-
sion and constraint, many types of regulation enable and empower
social actors and action, conferring licenses, special powers, and bene-
fits to some types of actors. In general, regulatory processes within the
private, market-based sector are more likely to rely on positive incen-
tives (e.g., increased returns, profits); in their role vis-à-vis the private
sector, public actors make greater use of negative sanctions (e.g., taxes,
fines, incarceration). However, as we will see, public sector actors are
capable of creating (constituting) social actors with broader, or more
restricted, powers of acting.
Force, sanctions, and expedient responses are central ingredients of
the regulatory pillar, but they are often tempered by the existence of
rules that justify the use of force. When coercive power is both sup-
ported and constrained by rules, we move into the realm of authority.
Power is institutionalized (Dornbusch and Scott 1975: Ch. 2; Weber
1924/1968).
Much work in economics emphasizes the costs of overseeing sys-
tems of regulation. Agency theory stresses the expense and difficulty
entailed in accurately monitoring performances relevant to contracts,
whether implicit or explicit, and in designing appropriate incentives
(see Milgrom and Roberts 1992; Pratt and Zeckhauser 1985). Although
in some situations agreements can be monitored and mutually enforced
by the parties involved, in many circumstances it is necessary to vest
62   INSTITUTIONS AND ORGANIZATIONS

the enforcement machinery in a third party expected to behave in a

neutral fashion. Economic historians view this as an important func-
tion of the state. Thus, North (1990) argues:

Because ultimately a third party must always involve the state as

a source of coercion, a theory of institutions also inevitably
involves an analysis of the political structure of a society and the
degree to which that political structure provides a framework of
effective enforcement. (p. 64)

However, North (1990: 54) also calls attention to problems that can
arise because “enforcement is undertaken by agents whose own utility
functions influence outcomes”—that is, third parties who are not neu-
tral. This possibility is stressed by many historical institutionalists,
such as Skocpol (1985), who argue that the state develops its own inter-
ests and operates somewhat autonomously from other societal actors.
In this and other ways, attention to the regulative aspects of institu-
tions creates renewed interest in the role of the state: as rule maker,
referee, and enforcer.
In an attempt to broaden the conception of law as a regulatory
mechanism, law and society theorists insist that analysts should not
conflate the coercive functions of law with its normative and cognitive
dimensions. Rather than operating in an authoritative and exogenous
manner, many laws are sufficiently controversial or ambiguous that
they do not provide clear prescriptions for conduct. In such cases, law
is better conceived as an occasion for sense-making and collective inter-
pretation (Weick 1995), relying more on cognitive and normative than
coercive elements for its effects (see Suchman and Edelman 1997; see
also Chapter 6). In short, institutions supported by one pillar may, as
time passes and circumstances change, be sustained by different pillars.
The institutional logic underlying the regulative pillar is an instru-
mental one: Individuals craft laws and rules that they believe will
advance their interests, and individuals conform to laws and rules
because they seek the attendant rewards or wish to avoid sanctions.
Because of this logic, the regulative pillar is one around which rational
choice scholars gather.
Empirical indicators of the development, extent, and province of
regulatory institutions are to be found in evidence of the expansion of
constitutions, laws, codes, rules, directives, regulations, and formal
structures of control. For example, Tolbert and Zucker (1983) determine
whether municipalities are mandated by state law to adopt civil service
reforms, and Singh, Tucker, and House (1986) and Baum and Oliver
 Crafting an Analytic Framework I: Three Pillars of Institutions    63

(1992) ascertain whether voluntary service organizations are registered

by oversight agencies. Dobbin and Sutton (1998) examine financial
allocations to enforcement agencies as an indicator of regulatory
enforcement.
As noted in Chapter 2, symbolic systems relate not only to sub-
stance but also to affect; they stimulate not only interpretive but emo-
tional reactions. D’Andrade (1984) has pointed out that meaning
systems work in representational, constructive, and directive ways—
providing cognitive guidance and direction—but also in evocative
ways, creating feeling and emotions. Emotions are among the most
important motivational elements in social life. Much recent research on
brain activity and cognitive behavior stresses the interdependence of
cognition and emotion. Long regarded as separate spheres related to
distinctive parts of the brain, this dichotomization now appears grossly
oversimplified and misleading (Dolan 2003; LeDoux 1996). Attention
to emotion in social life by social scientists has largely been associated
with ongoing work on identity (see Chapter 2) and on “institutional
work” (Lawrence, Suddaby, and Leca 2009)—emphasizing the impor-
tance of agency in maintaining and changing institutions (see Chapter 4).
Contradictions between institutional demands at the macro level are
experienced as conflicting role demands at the individual level—identity
conflicts that need to be resolved (Creed, Dejordy, and Lok, 2010; Seo
and Creed 2002). Emotions operate to motivate actors to change insti-
tutions in which they have become disinvested or to defend institu-
tions to which they are attached (Voronov and Vince 2012). Note that
attention to the emotional dimensions of institutions privileges work at
the micro (individual and interpersonal) levels of analysis.
Are their distinctive types of emotions engendered by encounters
with the regulative organs of society? I think so and believe that the
feelings induced may constitute an important component of the power
of this element. To confront a system of rules backed by the machinery
of enforcement is to experience, at one extreme, fear, dread, and guilt,
or, at the other, relief, innocence and vindication. Powerful emotions
indeed!
In sum, there is much to examine in understanding how regulative
institutions function and how they interact with other institutional ele-
ments. Through the work of agency and game theorists at one end of
the spectrum and law and society theorists at the other, we are
reminded that laws do not spring from the head of Zeus nor norms
from the collective soul of a people; rules must be interpreted and dis-
putes resolved; incentives and sanctions must be designed and will
have unintended effects; surveillance mechanisms are required but are
64   INSTITUTIONS AND ORGANIZATIONS

expensive and will prove to be fallible; and conformity is only one of

many possible responses by those subject to regulative institutions.
A stable system of rules, whether formal or informal, backed by
surveillance and sanctioning power affecting actors’ interests that is
accompanied by feelings of guilt or innocence constitutes one prevail-
ing view of institutions.

The Normative Pillar

A second group of theorists views institutions as resting primarily
on a normative pillar (again, see Table 3.1). Emphasis here is placed on
normative rules that introduce a prescriptive, evaluative, and obliga-
tory dimension into social life. Normative systems include both values
and norms. Values are conceptions of the preferred or the desirable
together with the construction of standards to which existing struc-
tures or behaviors can be compared and assessed. Norms specify how
things should be done; they define legitimate means to pursue valued
ends. Normative systems define goals or objectives (e.g., winning the
game, making a profit) but also designate appropriate ways to pursue
them (e.g., rules specifying how the game is to be played, conceptions
of fair business practices) (Blake and Davis 1964).
Some values and norms are applicable to all members of the col-
lectivity; others apply only to selected types of actors or positions.
The latter give rise to roles: conceptions of appropriate goals and
activities for particular individuals or specified social positions. These
beliefs are not simply anticipations or predictions, but prescriptions—
normative expectations—regarding how specified actors are sup-
posed to behave. The expectations are held by other salient actors in
the situation, and so are experienced by the focal actor as external
pressures. Also, and to varying degrees, they become internalized by
the actor. Roles can be formally constructed. For example, in an orga-
nizational context particular positions are defined to carry specified
rights and responsibilities and to have varying access to material
resources. Roles can also emerge informally as, over time through
interaction, differentiated expectations develop to guide behavior
(Blau and Scott 1962/2003: Chs. 1, 4). Normative systems are typi-
cally viewed as imposing constraints on social behavior, and so they
do. But at the same time, they empower and enable social action.
They confer rights as well as responsibilities, privileges as well as
duties, licenses as well as mandates. In his essays on the professions,
Hughes (1958) reminds us how much of the power and mystique
associated with these types of roles comes from the license they are
 Crafting an Analytic Framework I: Three Pillars of Institutions    65

given to engage in forbidden or fateful activities—conducting intimate

physical examinations or sentencing individuals to prison or to death.
The normative conception of institutions was embraced by most
early sociologists—from Durkheim and Cooley through Parsons and
Selznick—perhaps because sociologists are more likely to examine
those types of institutions, such as kinship groups, social classes, reli-
gious systems, communities, and voluntary associations, where com-
mon beliefs and values are more likely to exist. Moreover, it continues
to guide and inform much contemporary work by sociologists and
political scientists on organizations. For example, March and Olsen
(1989) embrace a primarily normative conception of institutions:

The proposition that organizations follow rules, that much of the

behavior in an organization is specified by standard operating
procedures, is a common one in the bureaucratic and organiza-
tional literature. . . . It can be extended to the institutions of poli-
tics. Much of the behavior we observe in political institutions
reflects the routine way in which people do what they are sup-
posed to do. (p. 21)

Although March and Olsen’s (1989) conception of rules is quite broad,

including cultural-cognitive as well as normative elements—“routines,
procedures, conventions, roles, strategies, organizational forms, and tech-
nologies . . . beliefs, paradigms, codes, cultures, and knowledge” (p. 22)—
their focus stresses the centrality of social obligations:

To describe behavior as driven by rules is to see action as a match-

ing of a situation to the demands of a position. Rules define rela-
tionships among roles in terms of what an incumbent of one role
owes to incumbents of other roles. (p. 23)

In short, scholars associated with the normative pillar stress the

importance of a logic of “appropriateness” vs. a logic of “instrumental-
ity.” The central imperative confronting actors is not “What choice is in
my own best interests?” but rather, “Given this situation, and my role
within it, what is the appropriate behavior for me to carry out?”
Empirical indicators of the existence and pervasiveness of norma-
tive institutions include accreditations and certifications by standard
setting bodies such as professional associations (Casile and Davis-
Blake, 2002; Ruef and Scott 1998).
As with regulative systems, confronting normative systems can
also evoke strong feelings, but these are somewhat different from those
66   INSTITUTIONS AND ORGANIZATIONS

that accompany the violation of rules and laws. Feelings associated

with the trespassing of norms include principally a sense of shame or
disgrace, or for those who exhibit exemplary behavior, feelings of
respect and honor. The conformity to or violation of norms typically
involves a large measure of self-evaluation: heightened remorse or
effects on self-respect. Such emotions provide powerful inducements
to comply with prevailing norms.
Theorists embracing a normative conception of institutions empha-
size the stabilizing influence of social beliefs and norms that are both
internalized and imposed by others. For early normative theorists such
as Parsons, shared norms and values were regarded as the basis of a
stable social order. And as Stinchcombe (1997) has eloquently reaf-
firmed, institutions are widely viewed as having moral roots:

The guts of institutions is that somebody somewhere really cares to

hold an organization to the standards and is often paid to do that.
Sometimes that somebody is inside the organization, maintaining
its competence. Sometimes it is an accrediting body, sending out
volunteers to see if there is really any algebra in the algebra course.
And sometimes that somebody, or his or her commitment is lacking,
in which case the center cannot hold, and mere anarchy is loosed
upon the world. (p. 18)

Heclo (2008) also embraces a similar stance. Assuming what he

terms an “inside-out” perspective on institutions—that is, viewing
institutions from the standpoint of those participating in them—he
affirms Selznick’s (1957: 101) distinction “between strictly instrumental
attachments needed to get a particular job done and the deeper com-
mitments that express one’s enduring loyalty to the purpose or pur-
poses that lie behind doing the job in the first place.” Heclo insists:

Deeper than the agent/principal issues is the agent/principle

perspective. It presupposes that as beings (which by existing we
surely are) we humans are moral agents. That is to say, by virtue
of being human, we experience our existence as partaking in
questions of right and wrong. To say human life is to say morally-
implicated life. (p. 79)

The Cultural-Cognitive Pillar

A third set of institutionalists, principally anthropologists like
Geertz and Douglas and sociologists like Berger, DiMaggio, Goffman,
 Crafting an Analytic Framework I: Three Pillars of Institutions    67

Meyer, Powell, and Scott stress the centrality of cultural-cognitive ele-

ments of institutions: the shared conceptions that constitute the nature
of social reality and create the frames through which meaning is made
(again, see Table 3.1). Attention to the cultural-cognitive dimension of
institutions is the major distinguishing feature of neoinstitutionalism
within sociology and organizational studies.
These institutionalists take seriously the cognitive dimensions of
human existence: Mediating between the external world of stimuli
and the response of the individual organism is a collection of internal-
ized symbolic representations of the world. “In the cognitive para-
digm, what a creature does is, in large part, a function of the creature’s
internal representation of its environment” (D’Andrade 1984: 88).
Symbols—words, signs, gestures—have their effect by shaping the
meanings we attribute to objects and activities. Meanings arise in
interaction and are maintained and transformed as they are employed
to make sense of the ongoing stream of happenings. Emphasizing the
importance of symbols and meanings returns us to Max Weber’s cen-
tral premise. As noted in Chapter 1, Weber regarded action as social to
the extent that the actor attaches meaning to the behavior. To under-
stand or explain any action, the analyst must take into account not
only the objective conditions but the actor’s subjective interpretation
of them. Extensive research by psychologists over the past three
decades has shown that cognitive frames enter into the full range of
information-processing activities, from determining what information
will receive attention, how it will be in encoded, how it will be
retained, retrieved, and organized into memory, to how it will be inter-
preted, thus affecting evaluations, judgments, predictions, and infer-
ences. (For reviews, see Fiol 2002; Markus and Zajonc 1985; Mindl,
Stubbart, and Porac 1996.)
As discussed in Chapter 2, the new cultural perspective focuses on
the semiotic facets of culture, treating them not simply as subjective
beliefs but also as symbolic systems viewed as objective and external
to individual actors. Berger and Kellner (1981: 31) summarize: “Every
human institution is, as it were, a sedimentation of meanings or, to
vary the image, a crystallization of meanings in objective form.” Our
use of the hyphenated label cognitive-cultural emphasizes that internal
interpretive processes are shaped by external cultural frameworks. As
Douglas (1982: 12) proposes, we should “treat cultural categories as the
cognitive containers in which social interests are defined and classified,
argued, negotiated, and fought out.” Or in Hofstede’s (1991: 4) graphic
metaphor, culture provides patterns of thinking, feeling, and acting:
mental programs, or the “software of the mind.”4
68   INSTITUTIONS AND ORGANIZATIONS

The stress in much of this work on cognition and interpretation

points to important functions of the cultural-cognitive pillar but over-
looks a dimension that is even more fundamental: its constitutive func-
tion. Symbolic processes, at their most basic level, work to construct
social reality, to define the nature and properties of social actors and
social actions. We postpone consideration of this function until the fol-
lowing section of this chapter, however, because it raises questions
regarding the fundamental assumptions underlying social science
research.
Cultural systems operate at multiple levels, from the shared defini-
tion of local situations, to the common frames and patterns of belief
that comprise an organization’s culture, to the organizing logics that
structure organization fields, to the shared assumptions and ideologies
that define preferred political and economic systems at national and
transnational levels. These levels are not sealed but nested, so that
broad cultural frameworks penetrate and shape individual beliefs on
the one hand, and individual constructs can work to reconfigure far-
flung belief systems on the other.
Of course, cultural elements vary in their degree of institutional-
ization—the extent of their linkage to other elements, the degree to
which they are embodied in routines or organizing schema. When we
talk about cognitive-cultural elements of institutions, we are calling
attention to these more embedded cultural forms: “culture congealed
in forms that require less by way of maintenance, ritual reinforcement,
and symbolic elaboration than the softer (or more ‘living’) realms we
usually think of as cultural” (Jepperson and Swidler 1994: 363).
Cultures are often conceived as unitary systems, internally con-
sistent across groups and situations. But cultural conceptions fre-
quently vary: Beliefs are held by some but not by others. Persons in
the same situation can perceive the situation quite differently—in
terms of both what is and what ought to be. Cultural beliefs vary and
are frequently contested, particularly in times of social disorganiza-
tion and change (see DiMaggio 1997; Martin 1992; 2002; Seo and
Creed 2002; Swidler 1986).
For cultural-cognitive theorists, compliance occurs in many cir-
cumstances because other types of behavior are inconceivable; routines
are followed because they are taken for granted as “the way we do
these things.” The prevailing logic employed to justify conformity is
that of orthodoxy, the perceived correctness and soundness of the ideas
underlying action.
Social roles are given a somewhat different interpretation by cul-
tural than by normative theorists. Rather than stressing the force of
 Crafting an Analytic Framework I: Three Pillars of Institutions    69

mutually reinforcing obligations, cultural-cognitive theorists point to

the power of templates for particular types of actors and scripts for
action (Shank and Abelson 1977). For Berger and Luckmann (1967),
roles arise as common understandings develop that particular actions
are associated with particular actors:5

We can properly begin to speak of roles when this kind of typifica-

tion occurs in the context of an objectified stock of knowledge
common to a collectivity of actors. . . . Institutions are embodied in
individual experience by means of roles. . . . The institution, with
its assemblage of “programmed” actions, is like the unwritten
libretto of a drama. The realization of the drama depends upon the
reiterated performance of its prescribed roles by living
actors. . . . Neither drama nor institution exist empirically apart
from this recurrent realization. (pp. 73–75)

Differentiated roles can and do develop in localized contexts as

repetitive patterns of action gradually become habitualized and objec-
tified, but it is also important to recognize the operation of wider insti-
tutional frameworks that provide prefabricated organizing models and
scripts (see Goffman 1974; 1983). Meyer and Rowan (1977) and
DiMaggio and Powell (1983) emphasize the extent to which wider
belief systems and cultural frames are imposed on or adopted by indi-
vidual actors and organizations.
Much progress has been made in recent years in developing indica-
tors of cultural-cognitive elements. For many years, investigators such
as social anthropologists and ethnomethodologists relied on close,
long-term observation of ongoing behavior from which they inferred
underlying beliefs and assumptions (e.g., Turner 1974). Later, quantita-
tive researchers employed survey methodologies to uncover shared
attitudes and common values (e.g., Hofstede 1984). More recently,
however, a “new archival research” approach has emerged in which
scholars employ formal analytical methodologies such as content,
semiotic, sequence, and network analysis to probe materials such as
discourse in professional journals, trade publications, organizational
documents, directories, annual reports, and specialized or mainstream
media accounts. This type of research has flourished due to the wide-
spread availability of computer-analyzable documents. The best of this
work illuminates “relevant features of shared understandings, profes-
sional ideologies, cognitive frames or sets of collective meanings that
condition how organizational actors interpret and respond to the
world around them” (Ventresca and Mohr 2002: 819).
70   INSTITUTIONS AND ORGANIZATIONS

The affective dimension of this pillar is expressed in feelings from

the positive affect of certitude and confidence on the one hand versus
the negative feelings of confusion or disorientation on the other. Actors
who align themselves with prevailing cultural beliefs are likely to feel
competent and connected; those who are at odds are regarded as, at
best, “clueless” or, at worst, “crazy.”
A cultural-cognitive conception of institutions stresses the central
role played by the socially mediated construction of a common frame-
work of meanings.

A Fourth Pillar? Habitual Dispositions

In a thoughtful essay, Gronow (2008) has proposed a fourth ele-
ment, which he argues constitutes yet another basis for institutions.
Building on the work of the American pragmatists, including Dewey,
James, and others who influenced the work of early economic theorists
Veblen and Commons (see Chapter 1), Gronow suggests that we add
habitual actions to our framework. He suggests that “habitual disposi-
tions are related to actions that have been repeated in stable contexts
and therefore require only a minimal amount of conscious thought to
initiate and implement” (p. 362). Although habits can be relatively
automatic, Gronow points out that pragmatists insisted habits are not
mere dead routines, but can include and overlap with reason and con-
scious choice.
While I welcome the strengthening of connections between prag-
matism and institutional arguments and agree that more attention
needs to be given to activities and practices, habits and routines, I am
not persuaded of the need to add a fourth pillar to the conceptual
framework. Shared dispositions are fundamentally cultural-cognitive
elements closely tied to repetitive behavior. Like other such elements,
they can be only semiconscious and taken for granted by the actors.
Moreover, I take into account the important role of activities and rou-
tines by treating them as of one of the major carriers of institutional
elements (see Chapter 4).

Combinations of Elements
Having introduced the three basic elements and emphasized their
distinctive features and modes of working, it is important to restate
the truth that in most empirically observed institutional forms, we
observe not one, single element at work but varying combinations of
elements. In stable social systems, we observe practices that persist
 Crafting an Analytic Framework I: Three Pillars of Institutions    71

and are reinforced because they are taken for granted, normatively
endorsed, and backed by authorized powers. When the pillars are
aligned, the strength of their combined forces can be formidable.
In some situations, however, one or another pillar will operate
virtually alone in supporting the social order; and in many situations,
a given pillar will assume primacy.
Equally important, the pillars may be misaligned: They may sup-
port and motivate differing choices and behaviors. As Strang and Sine
(2002: 499) point out, “Where cognitive, normative, and regulative sup-
ports are not well aligned, they provide resources that different actors
can employ for different ends.” Such situations exhibit both confusion
and conflict, and provide conditions that are highly likely to give rise
to institutional change (Dacin, Goodstein, and Scott 2002; Kraatz and
Block 2008). These arguments are pursued, illustrated, and empirically
tested in subsequent chapters.

™™THE THREE PILLARS AND LEGITIMACY

“Organizations require more than material resources and technical

information if they are to survive and thrive in their social environ-
ments. They also need social acceptability and credibility” (Scott, Ruef,
Mendel, and Caronna 2000: 237)—in short, they require legitimacy.
Suchman (1995b: 574) provides a helpful definition of this central con-
cept: “Legitimacy is a generalized perception or assumption that the
actions of an entity are desirable, proper, or appropriate within some
socially constructed system of norms, values, beliefs, and definitions.”
Legitimacy is a generalized rather than an event-specific evaluation
and is “possessed objectively, yet created subjectively” (Suchman
1995b: 574). The “socially constructed systems” to which Suchman
refers are, of course, institutional frameworks.
Max Weber was the first great social theorist to stress the impor-
tance of legitimacy. In his formulation of types of social action, he gave
particular attention to those actions guided by a belief in the existence
of a legitimate order: a set of “determinable maxims,” a model
regarded by the actor as “in some way obligatory or exemplary for
him” (Weber 1924/1968, Vol. 1: 31). In his empirical/historical work,
he applied his approach to the legitimation of power structures, both
corporate and governmental, arguing that power becomes legitimated
as authority to the extent that its exercise is supported by prevailing
social norms, whether traditional, charismatic, or rational-legal (see
Deephouse and Suchman 2008; Dornbusch and Scott 1975: Ch. 2; Ruef
72   INSTITUTIONS AND ORGANIZATIONS

and Scott 1998). In his cultural-institutional perspective, Parsons

(1956/1960b) broadened the focus of legitimation to include the goals
of an organization, determining the extent to which they were congru-
ent with the values extant in the society. And as we have seen, these
arguments have been advanced and amplified by neoinstitutionalists,
such as Berger and Luckmann (1967), Meyer and Rowan (1977), and
Meyer and Scott (1983b) to include the legitimation of strategies, struc-
tures, and procedures.
In a resource-dependence or social exchange approach to organiza-
tions, legitimacy is typically treated as simply another kind of resource
that organizations extract from their institutional environment (e.g.,
Dowling and Pfeffer 1975; Suchman 1995b). Scholars emphasizing the
regulative pillar share, at least to some extent, this interpretation as
they stress the benefits or costs of compliance. However, from a strong
institutional perspective, legitimacy is not a commodity to be pos-
sessed or exchanged but a condition reflecting perceived consonance
with relevant rules and laws or normative values, or alignment with
cultural-cognitive frameworks. Like some other invisible properties
such as oxygen, the importance of legitimacy become immediately and
painfully apparent only if lost, suggesting that it is not a specific
resource, but a fundamental condition of social existence.
Berger and Luckmann (1967) describe legitimacy as evoking a
“second order” of meaning. In their early stages, institutionalized
activities develop as repeated patterns of behavior that evoke shared
meanings among the participants. The legitimation of this order
involves connecting it to wider cultural frames, norms, or rules. “Legit-
imation ‘explains’ the institutional order by ascribing cognitive validity
to its objectified meanings. Legitimation justifies the institutional order
by giving a normative dignity to its practical imperatives” (pp. 92–93).
In a similar fashion, Johnson, Dowd, and Ridgway (2006) compare and
contrast the social psychological and the organizational views of legiti-
macy to arrive at a four-stage process: innovation, local validation, diffu-
sion, and general validation. That is, for new actions to be legitimated,
they must be locally accepted, and once they are “construed as a valid
social fact, [they are] adopted more readily by actors in other local con-
texts” (p. 60). As a result of successful diffusion, “the new social object
acquires widespread acceptance, becoming part of society’s shared
culture” (p. 61).6 And emphasizing the cultural-cognitive dimension,
Meyer and I propose that “organizational legitimacy refers to the degree
of cultural support for an organization” (Meyer and Scott 1983a: 201).
This vertical dimension entails the support of significant others:
various types of authorities—cultural as well as political—empowered
 Crafting an Analytic Framework I: Three Pillars of Institutions    73

to confer legitimacy. The reproduction of practices is supported by

structures residing at multiple levels (Colyvas and Jonsson 2011). Who
these authorities are varies from time to time and place to place but, in
our time, agents of the state and professional and trade associations are
often critical for organizations. Certification or accreditation by these
bodies is frequently employed as a prime indicator of legitimacy
(Dowling and Pfeffer 1975; Ruef and Scott 1998). In complex situations,
individuals or organizations may be confronted by competing sover-
eigns. Actors confronting conflicting normative requirements and stan-
dards typically find it difficult to take action since conformity to one
undermines the normative support of other bodies. “The legitimacy of
a given organization is negatively affected by the number of different
authorities sovereign over it and by the diversity or inconsistency of
their accounts of how it is to function” (Meyer and Scott 1983a: 202).
There is always the question as to whose assessments count in
determining the legitimacy of a set of arrangements. Many structures
persist and spread because they are regarded as appropriate by
entrenched authorities, even though their legitimacy is challenged by
other, less powerful constituencies. Martin (1994), for example, notes
that salary inequities between men and women are institutionalized in
American society even though the disadvantaged groups perceive
them to be unjust and press for reforms. “Legitimate” structures may,
at the same time, be contested structures.
Stinchcombe (1968) asserts that, in the end, whose values define
legitimacy is a matter of concerted social power:

A power is legitimate to the degree that, by virtue of the doctrines

and norms by which it is justified, the power-holder can call upon
sufficient other centers of power, as reserves in case of need, to
make his power effective. (p. 162)

It is important, however, to point out that power is not always a

top-down process, but can involve bottom-up phenomena. Power, for
example, can be authorized by superordinate parties (Stinchcombe 1968)
or endorsed by those subject to the power-wielder (Dornbusch and Scott
1975; Zelditch and Walker 1984) who collectively enforce norms sup-
porting compliance. Power can arise out of the mobilization of subordi-
nate groups as they attempt to advance their own values and interests.
While power certainly matters, in supporting legitimacy processes
as in other social activities, power is not the absolute arbiter. Entrenched
power is, in the long run, hapless against the onslaught of opposing
power allied with more persuasive ideas or stronger commitments.
74   INSTITUTIONS AND ORGANIZATIONS

Consistent with the preceding discussion, each of the three pillars

provides a basis of legitimacy, albeit a different one (see Table 3.1).7 The
regulatory emphasis is on conformity to rules: Legitimate organiza-
tions are those established by and operating in accordance with rele-
vant legal or quasi-legal requirements. A normative conception stresses
a deeper, moral base for assessing legitimacy. Normative controls are
much more likely to be internalized than are regulative controls, and
the incentives for conformity are hence likely to include intrinsic as
well as extrinsic rewards. A cultural-cognitive view points to the legiti-
macy that comes from conforming to a common definition of the situ-
ation, frame of reference, or a recognizable role (for individuals) or
structural template (for organizations). To adopt an orthodox structure
or identity in order to relate to a specific situation is to seek the legiti-
macy that comes from cognitive consistency. The cultural-cognitive
mode is the deepest level since it rests on preconscious, taken-for-
granted understandings.
The bases of legitimacy associated with the three elements, and
hence the types of indicators employed, are decidedly different and
may be in conflict. A regulative view would ascertain whether the
organization is legally established and whether it is acting in accord
with relevant laws and regulations. A normative orientation, stressing
moral obligations, may countenance actions departing from mere legal
requirements. Many professionals adhere to normative standards that
motivate them to depart from the rule-based requirements of bureau-
cratic organizations. And whistle-blowers claim that they are acting on
the basis of a “higher authority” when they contest organizational
rules or the orders of superiors. An organization such as the Mafia may
be widely recognized, signifying that it exhibits a culturally constituted
mode of organizing to achieve specified ends, and it is regarded as a
legitimate way of organizing by its members. Nevertheless, it is treated
as an illegal form by police and other regulative bodies, and it lacks the
normative endorsement of most citizens.
What is taken as evidence of legitimacy varies by which elements
of institutions are privileged.

™™BASIC ASSUMPTIONS ASSOCIATED WITH

THE THREE PILLARS

Although the differences among analysts emphasizing one or another

element are partly a matter of substantive focus, they are also associ-
ated with more profound differences in underlying philosophical
 Crafting an Analytic Framework I: Three Pillars of Institutions    75

assumptions. While it is not possible to do full justice to the complexity

and subtlety of these issues, I attempt to depict the differences in broad
outline. Two matters are particularly significant: (1) differences among
analysts in their ontological assumptions—assumptions concerning
the nature of social reality; and (2) differences involving the extent and
type of rationality invoked in explaining behavior.

Regulative and Constitutive Rules

Truth and Reality

Varying ontological assumptions underlie conceptions of institu-

tional elements. Thus, it is necessary to clarify one’s epistemological
assumptions: How do we understand the nature of scientific knowl-
edge? My position on these debates has been greatly influenced by the
formulation advanced by Jeffrey C. Alexander (1983, Vol. 1), who pro-
vides a broad, synthetic examination of the nature and development of
theoretical logic in modern sociological thought. Following Kuhn (1970),
Alexander adopts a postpositivist perspective viewing science as operat-
ing along a continuum stretching from the empirical environment on the
one hand to the metaphysical environment on the other (see Figure 3.1).
At the metaphysical end reside the most abstract general presup-
positions and models associated with more theoretical activity. At the
empirical end, one finds observations, measures, and propositions. The
continuum obviously incorporates numerous types of statements, rang-
ing from the more abstract and general to the more specific and particu-
lar. But, more important, the framework emphasizes that, although the

Figure 3.1  The Scientific Continuum and Its Components

Metaphysical Empirical
Environment Environment
General
Presuppositions

Models

Concepts

Definitions

Classifications

Laws

Complex and Simple

Propositions

Correlations

Methodological
Assumptions

Observations

SOURCE: © Jeffrey Alexander. Used with permission.

76   INSTITUTIONS AND ORGANIZATIONS

mix of empirical and metaphysical elements varies, every point on the

continuum is an admixture of both elements. “What appears, concretely, to
be a difference in types of scientific statements—models, definitions,
propositions—simply reflects the different emphasis within a given
statement on generality or specificity” (Alexander 1983, Vol. 1: 4).
The postpostivist conception of science emphasizes the fundamen-
tal similarity of the social and physical sciences—both are human
attempts to develop and test general statements about the behavior of
the empirical world. It rejects both a radical materialist view that
espouses that the only reality is a physical one, and also the idealist
(and postmodernist) view that the only reality exists in the human
mind. It also usefully differentiates reality from truth, as Rorty (1989)
observes:

We need to make a distinction between the claim that the world is

out there and the claim that truth is out there. To say that the world
is out there, that it is not our creation, is to say, with common
sense, that most things in space and time are the effects of causes
which do not include human mental states. To say that truth is not
out there is simply to say that where there are no sentences there
is no truth, that sentences are elements of human languages, and
that human languages are human creations. (pp. 4–5)

Social Reality

Although the physical and social sciences share important basic

features, it is essential to recognize that the subject matter of the social
sciences is distinctive. In John Searle’s (1995: 1, 11, 13) terminology, por-
tions of the real world, while they are treated as “epistemically objec-
tive” facts in the world, “are facts only by human agreement.” Their
existence is “observer-relative”: dependent on observers who share a
common conception of a given social fact. Social reality is an important
subclass of reality.8
Earlier in our discussion of cultural-cognitive elements, I intro-
duced the concept of constitutive processes. Now we are in a position to
develop this argument. Social institutions refer to types of social reality
that involve the collective development and use of both regulative and
constitutive rules. Regulative rules involve attempts to influence “ante-
cedently existing activities”; constitutive rules “create the very possibil-
ity of certain activities” (Searle 1995: 27). Constitutive rules take the
general form: X counts as Y in context C; for example, an American
dollar bill counts as legal currency in the United States. “Institutional
facts exist only within systems of constitutive rules” (p. 28). In general,
 Crafting an Analytic Framework I: Three Pillars of Institutions    77

as the label implies, scholars embracing the regulative view of institu-

tions focus primary attention on regulative rules; for example, they
assume the existence of actors with a given set of interests and then ask
how various rule systems, manipulating sanctions and incentives,
can affect the behavior of these actors as they pursue their interests.
Cultural-cognitive scholars stress the importance of constitutive rules:
They ask what types of actors are present, how their interests are
shaped by these definitions, and what types of actions they are allowed
to take. They thus differ in their ontological assumptions or, at least, in
the ontological level at which they work.
The anthropologist David Schneider (1976) usefully describes the
relation of constitutive culture to social norms:

Culture contrasts with norms in that norms are oriented to patterns

for action, whereas culture constitutes a body of definitions, prem-
ises, statements, postulates, presumptions, propositions, and per-
ceptions about the nature of the universe and man’s place in it.
Where norms tell the actor how to play the scene, culture tells the
actor how the scene is set and what it all means. Where norms tell
the actor how to behave in the presence of ghosts, gods, and human
beings, culture tells the actor what ghosts, gods, and human being
are and what they are all about. (pp. 202–203)

Constitutive rules operate at a deeper level of reality creation,

involving the devising of categories and the construction of typifica-
tions: processes by which “concrete and subjectively unique experi-
ences . . . are ongoingly subsumed under general orders of meaning
that are both objectively and subjectively real” (Berger and Luckmann
1967: 39). Such processes are variously applied to things, to ideas, to
events, and to actors, and are organized into hierarchical linked
arrangements and elaborate systems for organizing meaning. Games
provide a ready illustration. Constitutive rules construct the game of
football as consisting of things such as gridiron and goal posts and
events such as first downs and offsides (see D’Andrade 1984). Simi-
larly, other types of constitutive rules result in the social construction
of actors and associated capacities and roles: in the football context, the
creation of quarterbacks, coaches, and referees. Regulative rules define
how the ball may legitimately be advanced or what penalties are asso-
ciated with what rule infractions. Thus, cultural-cognitive theorists
amend and augment the portrait of institutions crafted by regulative
theorists. Cultural-cognitive theorists insist that games involve more
than rules and enforcement mechanisms: They consist of socially
78   INSTITUTIONS AND ORGANIZATIONS

constructed players endowed with differing capacities for action and

parts to play. Constitutive rules construct the social objects and events
to which regulative rules are applied.
Such processes, although most visible in games, are not limited to
these relatively artificial situations. Constitutive rules are so basic to
social structure, so fundamental to social life that they are often over-
looked. In our liberal democracies, we take for granted that individual
persons have interests and capacities for action. It seems natural that
there are citizens with opinions and rights (as opposed to subjects with
no or limited rights), students with a capacity to learn, fathers with
rights and responsibilities, and employees with aptitudes and skills. But
all of these types of actors—and a multitude of others—are social con-
structions; all depend for their existence on constitutive frameworks
that, although they arose in particular interaction contexts, have become
reified in cultural rules that can be imported as guidelines into new
situations (see Berger and Luckmann 1967; Gergen and Davis 1985).
Moreover, recognition of the existence of such constitutive pro-
cesses provides a view of social behavior that differs greatly from lay
interpretations or even from those found in much of social science. As
Meyer, Boli, and Thomas (1987) argue:

Most social theory takes actors (from individuals to states) and

their actions as real, a priori, elements. . . . [In contrast] we see the
“existence” and characteristics of actors as socially constructed
and highly problematic, and action as the enactment of broad
institutional scripts rather than a matter of internally generated
and autonomous choice, motivation and purpose. (p. 13)

In short, as constitutive rules are recognized, individual behavior

is seen to often reflect external definitions rather than (or as a source of)
internal intentions. The difference is nicely captured in the anecdote
reported by Peter Hay (1993):

Gertrude Lawrence and Noel Coward were starring in one of the

latter’s plays when the production was honored with a royal visit.
As Queen Elizabeth entered the Royal Box, the entire audience
rose to its feet. Miss Lawrence, watching from the wings, mur-
mured: “What an entrance!” Noel Coward, peeking on tip-toe
behind her, added “What a part!” (p. 70)

The social construction of actors also defines what they consider to

be their interests. The stereotypic “economic man” that rests at the
 Crafting an Analytic Framework I: Three Pillars of Institutions    79

heart of much economic theorizing is not a reflection of human nature,

but a social construct that arose under specific historical circumstances
and is maintained by particular institutional logics associated with the
rise of capitalism (see Heilbroner 1985).9 From the cultural-cognitive
perspective, interests are not assumed to be natural or outside the
scope of investigation; they are not treated as exogenous to the theo-
retical framework. Rather, they are recognized to be endogenous,
arising within social situations, as varying by institutional context and
as themselves requiring explanation.
The social construction of actors and their associated activities is
not limited to persons. Collective actors are similarly constituted, and
come in a wide variety of forms. We, naturally, will be particularly
interested in the nature of those institutional processes at work in the
constitution of organizations and organization fields, processes consid-
ered in later chapters.
In their critique of the pillars framework, Phillips and Malhotra
(2008) argue that because the different elements operate at varying
ontological levels, they cannot be combined into an integrated frame-
work. They propose that “authentic” institutional analysis involves
exclusive attention to the cultural-cognitive elements:

The fact that coercive and normative mechanisms are externally

managed by other actors makes them very different from the
taken-for-grantedness of cognitive mechanisms. Where coercive
and normative mechanisms result in strategic action and often
resistance, cognitive mechanisms function by conditioning
thinking. (p. 717)

But is this true? In a world of words, many of the most important

strategies involve choices as to how to frame the situation, how to con-
struct a powerful narrative, how to brand the product. In contested
situations, some of the most effective weapons available to contenders
involve how to define the actions, the actors, and their intent. Are we
seeking “Black power” or “civil rights”?
Cultural-cognitive elements are amenable to strategic manipula-
tion. They are also subject to deliberative processes under the control
of regulative and normative agents. Thus, members of the legislature
or the judicial branch can change the rights and powers of individual
and collective actors. Recently, the U.S. Supreme Court determined
that corporations have political rights allowing them to exercise free-
dom of speech, including unrestricted expenditure of funds for politi-
cal action committees. And professional authorities regularly create
80   INSTITUTIONS AND ORGANIZATIONS

new institutions: new concepts, distinctions, and typologies that

shape the types of measures we use; the kinds of data we collect; and
the interpretations we make (Espeland and Sauder 2007; Scott 2008b).
In short, there are important differences among the pillars, as I
have labored to explicate. I have attempted to construct what social
theorist Charles Tilly (1984) defines as an encompassing theoretical frame-
work, one that examines theories sharing broad objectives and attempts
not simply to argue that they employ provide differing approaches, but
to explicate the ways in which the approaches vary. I continue to
believe that such a framework provides a fruitful guide for institu-
tional analysis.

Rational and Reasonable Behavior

Theorists make different assumptions regarding how actors make
choices: what logics determine social action. As discussed earlier and
in Chapter 1, Weber defined social action so as to emphasize the impor-
tance of the meanings individuals attach to their own and others’
behavior. For Weber and many other social theorists, “the central ques-
tion that every social theory addresses in defining the nature of action
is whether or not—or to what degree—action is rational” (Alexander
1983, Vol. 1: 72). A more basic question, however, is how is rationality
to be defined? Social theorists propose a wide range of answers.
At one end of the spectrum, a neoclassical economic perspective
embraces an atomist view that focuses on an individual actor engaged in
maximizing his or her returns, guided by stable preferences and pos-
sessing complete knowledge of the possible alternatives and their con-
sequences. This model has undergone substantial revision in recent
decades, however, as over time economists have reluctantly acknowl-
edged limitations on individual rationality identified by psychologists
such as Tversky and Kahneman (1974), as described in Chapter 2.
Validated by a Nobel Prize in economics (in 2002), this work has been
incorporated into the mainstream as behavioral economics. Perhaps over-
stating the matter, a review of Kahneman’s (2011) recent book concludes
that his empirical investigations of human decision making “makes it
plain that Homo economicus—the rational model of human behavior
beloved of economists—is as fantastical as a unicorn” (The Economist
2011: 98). Nevertheless, this model, perhaps wearing a somewhat more
modest cloak, continues to pervade much economic theorizing.
Embracing a somewhat broader set of assumptions, neoinstitu-
tional analysts in economics and rational choice theorists in political
science (e.g., Moe 1990a; Williamson, 1985) utilize Simon’s (1945/1997: 88)
 Crafting an Analytic Framework I: Three Pillars of Institutions    81

model of bounded rationality, which presumes that actors are

“intendedly rational, but only boundedly so.” These versions relax the
assumptions regarding complete information and utility maximization
as the criterion of choice, while retaining the premise that actors seek
“to do the best they can to satisfy whatever their wants might be”
(Abell, 1995: 7). Institutional theorists employing these and related
models of individual rational actors are more likely to view institutions
primarily as regulative frameworks. Actors construct institutions to
deal with collective action problems—to regulate their own and others’
behaviors—and they respond to institutions because the regulations are
backed by incentives and sanctions. A strength of these models is that
rational choice theorists have “an explicit theory of individual behavior
in mind” when they examine motives for developing and consequences
attendant to the formation of institutional structures (Peters 1999: 45;
see also Abell 1995). Economic theorists argue that, while their assump-
tions may not be completely accurate, “many institutions and business
practices are designed as if people were entirely motivated by narrow,
selfish concerns and were quite clever and largely unprincipled in their
pursuit of their goals” (Milgrom and Roberts 1992: 42).
From a sociological perspective, a limitation of employing an
overly narrow rational framework is that it “portrays action as simply
an adaptation to material conditions”—a calculus of costs and benefits—
rather than allowing for the “internal subjective reference of action”
that opens up potential for the “multidimensional alternation of free-
dom and constraint” (Alexander 1983, Vol. 1: 74). Another limitation
involves the rigid distinction in rational choice models made between
ends, which are presumed to be fixed, and means. Sociological models
propose, variously, that ends are modified by means, that ends emerge
in ongoing activities, and even that means can become ends (March
and Olsen 1989; Selznick 1949; Weick 1969/1979). In addition, rather
than positing a lone individual decision maker, the sociological version
embraces an “organicist rather than an atomist view” such that “the
essential characteristics of any element are seen as outcomes of rela-
tions with other entities” (Hodgson 1994: 61). Actors in interaction
constitute social structures, which, in turn, constitute actors. The prod-
ucts of prior interactions—norms, rules, beliefs, resources—provide
the situational elements that enter into individual decision making (see
the discussion of structuration in Chapter 4).
A number of terms have been proposed for this broadened view of
rationality. As usual, Weber anticipated much of the current debate
by distinguishing among several variants of rationality, including
Zweckrationalität—action that is rational in the instrumental, calculative
82   INSTITUTIONS AND ORGANIZATIONS

sense—and Wertrationalität—action that is inspired by and directed

toward the realization of substantive values (Weber 1924/1968, Vol. 1: 24;
see also Swedberg 1998: 36). The former focuses on means-ends con-
nections; the latter on the types of ends pursued. Although Weber
himself was inconsistent in his usage of these ideal types, Alexander
suggests that they are best treated as analytic distinctions, with actual
rational behavior being seen as involving an admixture of the two
types. All social action involves some combination of calculation (in
selection of means) and orientation toward socially defined values.10
A broader distinction has been proposed by March (1981), who dif-
ferentiates between a logic of instrumentalism and a logic of “appropri-
ateness” (see also March 1994; March and Olsen 1989), as noted earlier.
An instrumental logic asks, “What are my interests in this situation?”
An appropriateness logic stresses the normative pillar where choice is
seen to be grounded in a social context and to be oriented by a moral
framework that takes into account one’s relations and obligations to
others in the situation. This logic replaces, or sets limits on, individual-
istic instrumental behavior.
Cultural-cognitive theorists emphasize the extent to which behav-
ior is informed and constrained by the ways in which knowledge is
constructed and codified. Underlying all decisions and choices are
socially constructed models, assumptions, and schemas. All decisions
are admixtures of rational calculations and nonrational premises. At the
micro-level, DiMaggio and Powell (1991) propose that a recognition of
these conditions provides the basis for what they term a theory of prac-
tical action. This conception departs from a “preoccupation with the
rational, calculative aspect of cognition to focus on preconscious pro-
cesses and schema as they enter into routine, taken-for-granted behav-
ior” (p. 22). At the same time, it eschews the individualistic, asocial
assumptions associated with the narrow rational perspective to empha-
size the extent to which individual choices are governed by normative
rules and embedded in networks of mutual social obligations.
The institutional economist Richard Langlois (1986b) proposes that
the model of an intendedly rational actor be supplemented by a model
of the actor’s situation, which includes, importantly, relevant social
institutions. Institutions provide an informational-support function,
serving as “interpersonal stores of coordinative knowledge” (p. 237).
Such common conceptions enable the routine accomplishment of
highly complex and interdependent tasks, often with a minimum of
conscious deliberation or decision making. Analysts are enjoined to
“pay attention to the existence of social institutions of various kinds as
bounds to and definitions of the agent’s situation” (p. 252). Langlois
 Crafting an Analytic Framework I: Three Pillars of Institutions    83

encourages us to broaden the neoclassical conception of rational action

to encompass what he terms “reasonable” action, a conception that
allows actors to “prefer more to less [of] all things considered,” but also
that allows for “other kinds of reasonable action in certain situations”
including rule-following behavior (p. 252). Social action is always
grounded in social contexts that specify valued ends and appropriate
means; action acquires its very reasonableness from taking into account
these social rules and guidelines for behavior.
As briefly noted in our consideration of a fourth pillar, recent
scholars have suggested that contemporary theorizing would be
advanced by resurrecting and updating pragmatism, a theory promul-
gated during the late 19th and early 20th centuries by some of America’s
most ingenuous social philosophers and social scientists, including
Oliver Wendell Holmes, William James, Charles Peirce, and John
Dewey. Among their central tenants were that (1) “ideas are not ‘out
there,’ waiting to be discovered, but are tools . . . that people devise to
cope with the world in which they find themselves,” and (2) that ideas
are produced “not by individuals, but by groups of individuals—that
ideas are social, . . . dependent . . . on their human carriers and the
environment” (Menand 2001: xi). Indeed, as Strauss (1993) reminds us:

In the writings of the Pragmatists we can see a constant battle

against the separating, dichotomizing, or opposition of what Prag-
matists argued should be joined together: knowledge and practice,
environment and actor, biology and culture, means and ends, body
and mind, matter and mind, object and subject, logic and inquiry,
lay thought and scientific thought, necessity and chance, cognitive
and noncognitive, art and science, values and action. (p. 72)

Ansell (2005) suggests that pragmatists favored a model of deci-

sion making that could be characterized as practical reason—recognizing
that people make decisions in “situationally specific contexts,” draw-
ing on their past experiences, and influenced by their emotions as well
as their reason.

™™CONCLUDING COMMENT

While it is possible to combine the insights of economic, political,

and sociological analysts into a single, complex, integrated model of
an institution, I believe it is more useful at this point to recognize the
differing assumptions and emphases that accompany the models
84   INSTITUTIONS AND ORGANIZATIONS

currently guiding inquiry into these phenomena. Three contrasting

models of institutions are identified—the regulative, the normative,
and the cultural-cognitive—although it is not possible to associate
any of the disciplines uniquely with any of these proposed models.
We find researchers in each discipline emphasizing one or another of
the pillars. The models are differentiated such that each identifies a
distinctive basis of compliance, mechanism of diffusion, type of
logic, cluster of indicators, affective response, and foundation for
legitimacy claims.
While at a superficial level it appears that social analysts are
merely emphasizing one or another of the multiple facets of institu-
tional arrangements, a closer examination suggests that the models are
aligned with quite profound differences in the assumptions made about
the nature of social reality and the ways in which actors make choices
in social situations. Two sources of continuing controversy are identified.
First, analysts disagree as to whether to attend primarily to regulative
rules as helping to structure action among a given set of actors with
established interests or to instead give primacy to constitutive rules
that create distinctive types of actors and related modes of action.
Second, institutions have become an important combat zone in the
broader, ongoing disputation within the social sciences centering on
the utility of rational choice theory for explaining human behavior. Are
we to employ a more restricted, instrumental logic in accounting for
the determinants and consequences of institutions, or is it preferable to
posit a broader, more socially embedded logic? There is no sign of a
quick or easy resolution to either of these debates.

™™NOTES

  1. Such an integrated model of institutions is elaborated in Scott (1994b).

  2. Not all analysts share this belief. In a rather abrasive critique of an
earlier presentation of this argument (Institutions and Organizations, 1st ed.,
1995), Hirsch (1997: 1704) pointed out that my approach runs the risk of enforc-
ing a “forced-choice” selection of one element as against another, rather than
recognizing the reality that all institutional forms are composed of multiple
elements. Such is not my intent. I willingly accede to the multiplex nature of
institutional reality while insisting on the value of identifying analytic con-
cepts, which, I believe, will aid us as we attempt to sort out the contending
theories and interrelated processes. Far from wishing to “rule out” or “discour-
age interpillar communication” or to make the “cross-fertilization of ideas
unusual and unlikely,” as Hirsch (1997: 1709) alleged, my intent in constructing
this analytic scheme is to encourage and inform such efforts.
 Crafting an Analytic Framework I: Three Pillars of Institutions    85

  3. In his most recent work, however, North (2005: Ch. 3) greatly expands
his interest in and attention to cultural-cognitive facets of institutions.
  4. Note the similarity of these conceptions to Bourdieu’s concept of habi-
tus, discussed in Chapter 2.
 5. Schutz analyzes this process at length in his discussion of “the
world of contemporaries as a structure of ideal types” (see Schutz 1932/1967:
176–207).
  6. However, as discussed in Chapter 6, it is essential that we not conflate
diffusion and institutionalization.
  7. Related typologies of the varying bases of legitimacy have been devel-
oped by Stryker (1994; 2000) and Suchman (1995b).
  8. Searle’s framework is, hence, a moderate version of social construc-
tionism. This more conservative stance is signaled by the title of his book, The
Construction of Social Reality (1995), which differs markedly from the broader
interpretation implied by the title of Berger and Luckmann’s The Social
Construction of Reality (1967).
9. As succinctly phrased by the economic historian Shonfield (1965: 71):
“Classical economics, which was largely a British invention, converted the
British experience . . . into something very like the Platonic idea of capitalism.”
10. Famously, Weber (1906–1924/1946: 280) captured this combination of
ideas and interests in his “switchman” metaphor:

Not ideas, but material and ideal interests directly govern men’s conduct.
Yet very frequently the “world images” that have been created by “ideas”
have, like switchmen, determined the tracks along which action has been
pushed by the dynamics of interests.