-
"Academy ot Management Review, 1988, Vol. 13, No, 1, 53-64,
Configurational and CoactivationalViews of Organizational
Structure
GREGORY K. DOWUniversity of Alberta
Two distinct vie'ws of organizational structure, configurational
andcoactivational, are discussed. The conhgurational view stresses
theauthoritative coordination of work, whereas the coactivational
viewemphasizes recurrent patterns of interaction among
participants. Thesetwo conceptual schemes differ in their stances
toward hierarchy,power, and organizational purposiveness. The
coactivational viewsubstitutes "invisible hand" explanations of
structure for the inten-tionalist explanations advanced by
configurational theory.
Organizations are singularly complex systems,and it is far from
obvious how their internal struc-tures can best be described. Two
distinct viewsof organizational structure can be found in
theliterature: the configurational and the coactiva-tional. The
configurational view emphasizes theintegration of work tasks under
common mana-gerial authority, whereas the coactivational
viewstresses recurrent patterns of interaction amongorganizational
participants. These two concep-tual schemes differ in their stances
toward hier-archy, power, and organizational
purposiveness.Configurational theory advances
intentionalistexplanations of organizational structure, whereasthe
coactivational view is more compatible with"invisible hand" modes
of explanation.
This paper is concerned primarily with descrip-tive uses of
these views, rather than with theprescriptions for organizational
design that canbe derived from them. The prescriptive stancesof the
two views are compared briefly in the clos-ing section of the
paper.
The coactivational view opens up promisingnew avenues of attack
on questions which seemlargely impervious to the tools of
configurationalanalysis. These questions include the relation-ship
between organizational hierarchy and mar-
ket exchange, the ways in which labor-managedfirms differ from
conventional enterprises, andmore generally, how conflict and power
becomeembedded in organizational structure. Althoughit is
impossible to justify these claims fully here,it is hoped that the
coactivational view of struc-ture, when coupled with invisible hand
explana-tions of the sort described in the final section,will point
organization theory in fruitful newdirections.
Configurational View of Structure
In the configurational view, an organization'sstructure involves
two core features: the decom-position of an overall objective into
subtasks, andthe integration of these subtasks to achieve
aneffective organizational performance. Subtasksare coordinated by
placing related activities un-der the supervision of a single
manager. Theseideas imply a purposive view of the organiza-tion;
indeed, in this literature organizations oftenare defined as
purposive human associations(Blau, 1974, chap. 1).
The archetypal image of the configurationalperspective is the
organizational chart with its
53
-
vertical lines of managerial authority. Threetypes of variables
are regarded as central:
The properties of the organizational hierarchy.Classic examples
include the number of levelsat which tasks are grouped together
(height ofthe hierarchy) and the number of subtasks super-vised by
a single authority (span of control).
The principles used to group tasks at each levelin the
hierarchy. For example, tasks can begrouped by intended purpose or
product, theprocesses or raw materials used, the clienteleserved,
the time at which activities are carriedout, or their geographic
location.
The administrative mechanisms used to coor-dinate subtasks.
Examples include the relation-ship between line and staff, the
degree of cen-tralization in decision making, and the ratio
ofadministrators to production workers. Pugh,Hickson, Hinings, and
Turner (1968) provided amore extensive list of variables; one
paradig-matic example of empirical research in this veinis that of
Blau and Schoenherr (1971).Intellectual Grenealogy
The roots of the configurational view are an-cient. Wren (1979,
chap. 2) pointed out examplesof it dating back to the early
military, religious,and state bureaucracies of Egypt,
Mesopotamia,and China. Often, modern configurational anal-yses use
Max Weber's portrait of the ideal bu-reaucracy as the point of
departure (1978, pp.212-226, 956-1005). The configurational view
alsois indebted to the work of management theoristssuch as Fayol
(1930), Gulick (1937), Mooney andReiley (1939), and Urwick (1937,
1943). The hu-man relations school flowered partly as a reac-tion
against the methods advocated by thesewriters, and the
configurational approachdeclined in importance. It was revived in
Wood-ward's (1965) attempt to test the validity of classi-cal
management principles, however. Her work,in turn, presaged the rise
of contingency theoryvariants of the configurational view
(Lawrence& Lorsch, 1967; Perrow, 1967; Pugh, Hickson,Hinings,
& Turner, 1969; Thompson, 1967).
Contingency theory moved away from the "onebest way" approach
that characterized much
early configurational writing. Its advocates ad-vanced the idea
that structure is a result of theorganization's size, technology,
and environ-ment. However, the contingency theorists contin-ued to
view structure in configurational terms,much as the classical
theorists did. For example,Lawrence and Lorsch reformulated the
divisionof labor and task coordination as differentiationand
integration, remarking that these ideas"suggest a return to the
central concern of earlyorganizational theorists; i.e., the optimal
divisionof labor given a general organizational purpose"(1967, p.
2). Although the notion of a single bestway to organize was
rejected by contingencytheory, observed structures were regarded
asappropriate adaptations to technological and en-vironmental
exigencies.
Economists arrived at a configurational per-spective by two
quite different routes. First, thisviewpoint figured prominently in
Chandler's(1962, 1977) work on business history, particu-larly in
his account of the transition of large U.S.firms from functionally
departmentalized struc-tures to multidivisional structures.
Chandler'sanalysis stimulated the development of transac-tion cost
economics (Armour & Teece, 1978;Williamson, 1975, chap. 8) and
provoked an ex-tensive body of empirical research by other
in-dustrial organization economists (Caves, 1980).
Another more mathematically oriented vari-ant of economic
configurationism has developedthat has few, if any, precursors in
common withsociological organization theory. Much of thiswork
traces generally to the planning and mar-ket socialism debates of
the 193Gs and 1940s(Nelson, 1981; Ward, 1967, chap. 2), and
morespecifically to the explosion of interest in theeconomics of
information during the 1970s(Spence, 1975). Popular research areas
have in-cluded identifying the limits to a firm's size(Beckmann,
1977; Calvo & Wellisz, 1978; William-son, 1967), explaining
observed intraorganiza-tional pay differentials (Calvo &
Wellisz, 1979;Mirrlees, 1976; Rosen, 1982), and the optimal
co-ordinating of subunits (Cremer, 1980; Groves &Radner, 1972;
Weitzman, 1974). Each of thesestudies postulates an organizational
objective
54
-
(cost minimization, profit or utility maximization,or Pareto
efficiency), hypothesizes that certainaspects of organizational
structure will be ad-justed to promote attainment of this
objective,and assumes a hierarchical coordination or su-pervision
scheme.Hierarchy and Power in ConfigurationalThought
The configurational emphasis on managerialauthority leads to
certain theoretical blind spots,including a reluctance to interpret
market rela-tions as a form of organization. For many
years,economists have sought to explain why sometransactions are
mediated by market exchangewhereas others are internalized within
firms(Coase, 1937); often they have found it useful tothink of
market and hierarchy as endpoints on acontinuum of possible
organizational structures(Mirrlees, 1976, p. 106; Williamson, 1985,
p. 83).The configurational view, however, postulatesqualitative
differences between market exchangeand hierarchy. Blau (1974, pp.
28-29), for ex-ample, distinguishes organizations
(e.g.,hierarchies) that arise through deliberate designfrom
nonorganizations such as markets that arisethrough merely emergent
forces. This line of de-marcation seems increasingly artificial in
lightof recent efforts to handle transactions withinfirms and
across markets in a unified way(Arrow, 1974; Dow, 1987a, 1987b;
Powell, in press,Williamson, 1985).
A second blind spot involves issues of powerand legitimacy. When
managerial authority isdepicted as being exercised on behalf of a
com-mon goal, unless one regards the organizationalraison d'etre as
unacceptable per se (e.g., or-ganized crime) it tends to be seen as
both legiti-mate and functional. The tacit assumption
thatsubordinates and superiors share in a commonorganizational
purpose leads the theorist to speaknot only of managerial power,
but also of mana-gerial authority (that is, legitimated power).
This aspect of configurational analysis may ac-count for its
failure to shed much light on thepossible merits of the
labor-managed firm (Dow,
1986; Horvat, 1982; Putterman, 1984; Russell,1985). The idea
that workers should exercise ulti-mate control over the managers of
an enterprisemakes little sense if all exercises of managerialpower
are legitimate. Thus, the configurationalview obstructs recognition
that monitoring andcontrolling behavior is a two-edged sword
be-cause subordinates have a stake in curbing self-interested
abuses of power by their superiors(Dow, 1987a).Role of
Intentionality in ConfigurationalThought
Once an organization has been defined as apurposive association,
it is only a short step tothe assumption that its internal
structure will beadjusted to facilitate its purposes.
Explanatoryreasoning then acquires a highly
intentionalistcoloration: Observed structures are understoodas
rational devices for pursuing assumed organi-zational goals, within
a given technological,institutional, and market environment.
Such reasoning is most visible among econo-mists, who routinely
prescribe particular organi-zational forms as a means to allocative
efficiency(Groves & Radner, 1972; Hurwicz, 1985; Weitz-man,
1974). This openly prescriptive stanceshades off imperceptibly into
descriptive theo-ries postulating that efficient structures will
infact be observed (Williamson, 1980, 1981). Indeed,the role played
by Pareto efficiency in economicanalyses parallels closely the role
of organiza-tional survival goals in functionalist sociology.The
difficulties of functionalist social theory are,of course, well
known (Burrell & Morgan, 1979,chap. 4; Elster, 1983, chap. 2;
Giddens, 1979, pp.1-7).
Once an intentionalist view of structure isaccepted,
organization theory easily slips into"arguments from design"
reminiscent of thoseused by natural theologians to explain
theadaptations of living organisms (Mayr, 1982, pp.103-105). The
subtle adaptations of an organiza-tion to its environment are
interpreted as the markof the organization's designers, who
instituted asuitable structure at the outset and reshaped that
55
-
structure over time in response to changingcircumstances. These
designers typically areidentified as founders, entrepreneurs, top
man-agement, or a dominant coalition. Chandler(1962), for example,
believed that structure fol-lows strategy, meaning that the
strategic choicesof corporate managers determine
organizationalstructure. Industrial organization economists(Caves,
1980, p. 64) and sociologists concernedwith strategic choice in
organizations (Child,1972) have expressed similar views.
This analysis is at best an incomplete descrip-tive theory of
organizational structure because itignores the counterstrategies of
organizationalactors who may contest managerial
objectives.Implicitly, the configurational approach presup-poses
that some privileged group has the powerto impose its preferred
structural arrangementsupon other organizational participants.
While thispresupposition often approximates reality, a
sat-isfactory theory of organizational structure shouldregard the
power to impose particular patternsof behavior on others as a fact
requiring expla-nation, not as a given. Further, explanatory
as-sertions based on the premise of organizationalintentionality
tend to distract attention from pat-terns of behavior rooted in
intraorganizationalconflict.
Philosophers of science have endorsed the an-cient maxim that
you can't beat something withnothing. In the words of Lakatos:
"[A]n objectivereason [to abandon an existing scientific
researchprogram] is provided by a rival research pro-gramme which
explains the previous success ofits rival and supersedes it by a
further display ofheuristic power" (1970, p. 155, emphasis
deleted).The coactivational view provides an alternativeto the
configurational view.
Coactivational View of StructureThe term coactivity is borrowed
from ecology,
and it emphasizes the importance of ongoingbehavioral
interactions for the study of organiza-tional structure. In the
coactivational view, anorganization is a communication network
inwhich actors or subunits recurrently process re-
sources and information. The organization'sstructure is inferred
from regularities in the be-havior of these actors as they are
observed overtime (Mackenzie, 1976). Conventional organiza-tional
charts are replaced by flow charts depict-ing the movement of
symbols and resourcesthrough the system (e.g., Cyert & March,
1963;p. 87).
Writing in this spirit, March and Simon de-fined structure as
"those aspects of the pattern ofbehavior in an organization that
are relativelystable and that change only slowly" (1958, p.170). A
slight modification of this definition is mostuseful for present
purposes:
The structure of a social system is identical withthe recurrent
patterns of resource and informa-tion flow generated by the
decision rules of itscomponent actors or subunits.
Such structures can be represented mathe-matically using a
variety of block-modeling,simulation, or stochastic process
techniques. Itmight be said that organizational boundaries aretaken
as given under this definition, and it istrue that some universe of
actors must be speci-fied in advance. However, if boundaries can
beidentified behaviorally (e.g., as regions of low-frequency
interaction separating regions of high-frequency interaction), then
the location of suchboundaries within the prespecified system of
ac-tors can be regarded as endogenous.
The decision rules that generate an organiza-tional structure
specify with reasonable accuracyhow each actor responds to messages
from otheractors or from actors in the external environment.The key
task of organizational theory, in thecoactivational view, is to
explain why decisionrules are what they are; that is, why a
givenstructure of interaction emerges in a specific so-cial
setting. The coactivational view highlightsthe following
variables:
The coding systems and observed behaviorsof each actor. Decision
rules are not observeddirectly. Each actor's decision rules must be
in-ferred by studying the relationship between in-coming messages
or resources and his/her be-havioral responses (Barley, 1983;
Gregory, 1983).
56
-
Technological and institutional constraints onthe acquisition of
resources and information byorganizational actors. These
constraints definethe avaiiable channels of interaction, and,hence,
iimit the range of possible structures. Ii isvital to distinguish
these exogenousiy given ruiesof the game from the decision rules of
the actors,which are the dependent variables to be ex-plained. In
practice, this distinction can be dif-ficult to make because
seemingly exogenous"technical" constraints are sometimes imposedon
an actor through the strategic behavior ofother actors (Dow, 1985;
Noble, 1984). Rules ofinteraction that are institutional givens in
oneanalysis may be treated as endogenous struc-tures in another,
but any analysis must positsome framework within which feasible or
per-missible interactions take place.
The response of the external environment toactions by
organizational participants. All orga-nizations exchange resources
and informationwith their environments. Actors in the environ-ment
generally reward some actions and pun-ish others. Environmental
states often can be ma-nipulated through strategic behavior to the
ad-vantage of particular organizational
participants.Intraorganizational decision rules cannot be
ex-plained adequately without referring to this webof incentives,
information exchanges, and powerrelations that link actors with the
external en-vironment.
Intellectual GenealogyAmong U.S. writers, the coactivational
view
received its first clear formulation in the works ofCyert and
March (1963), March and Simon (1958),and Simon (1976), a group
often labeled theCarnegie school, although Barnard (1938) andthe
human relations school also were significantin its development. The
fundamental differencebetween the Carnegie school and classical
the-ory was the former's elevation of individual deci-sion making
to a paramount analytic position.At an early stage, Simon argued
that a scientifi-cally relevant description of an
organization"designates for each person in the organizationwhat
decision that person makes and the influ-
ences to which he is subject in making each ofthese decisions"
(1976, p. 37). The social-psychological focus implicit in this
approachforced organization theorists to consider the linkbetween
the decisions made by individual ac-tors and the structure of the
organization, a themelater developed at length by Weick (1979).
A second crucial break with previous workcame with the emphasis
on dynamic, rather thanstatic, modeling of organizational
phenomena.March and Simon revised classical managementtheory by
taking into account "contingent activi-ties." These are activities
that depend either oninformation obtained from others in the
organiza-tion or on signals about the external world, andthey come
"with time subscripts attached" (1958,pp. 26-29). Ultimately, this
line of thought led tocomputer simulations of organizations
(Cohen,1981, 1984; Cohen, March, & Olsen, 1972;Crecine, 1969;
Cyert & March, 1963) as well asmodels using the mathematics of
stochastic pro-cesses (Dow, 1987b; Padgett, 1980).
The coactivational view entered politicalscience through the
studies of federal and localbudgeting processes conducted by
second-generation researchers at Carnegie (e.g.,Crecine, 1969, and
Gerwin, 1969) who built onthe earlier work of Wildavsky (1964).
Since then,the insights of these authors have been absorbedinto the
mainstream of theoretical and empiricalwork on public sector
budgeting (Danziger, 1978,chaps. 6-7). TTie Carnegie school also
in-fluenced the study of foreign policy decision mak-ing (Allison,
1971, chaps. 3-4; Steinbruner, 1974,chap. 3).
Important economic contributions to the coac-tivational view
were made by Nelson and Win-ter (1982, chap. 5); they argued that
decision-making routines provide the basis for organiza-tional
continuity over time, and that evolutionaryforces select from among
alternative structureson the basis of the profitability of the
firm. Someelements of transaction cost economics, espe-cially
Williamson's (1985, pp. 20-22) emphasison the decision procedures
used by transactingparties in adapting to changing
circumstances,also have a coactivational flavor (Dow, 1987a).
57
-
Marschak and Radner's (1972) theory of teamsexplicitly modeled
organizations as communica-tion networks in which the flow of
messages isgoverned by consistent decision rules. Radner(1972, pp.
177-188) also proposed a game-theoretic extension of the theory of
teams to ac-count for intraorganizational conflict, and hegave the
bounded rationality precepts of the Car-negie school a precise
mathematical formula-tion (Radner, 1975a, 1975b). Of related
interestare the efforts of game theorists to anchor con-ventional
solution concepts either to underlyingprocesses of adaptive
learning (Crawford, 1985;Schotter, 1981) or to natural selection
(Axelrod,1984; Friedman & Rosenthal, 1985).
Examples of coactivational analysis from out-side the United
States include Giddens' theory ofstructuration, in which the
structural propertiesof collectivities are "rules and resources
recur-sively implicated in the reproduction of socialsystems"
(1979, p. 64), and Crozier and Fried-berg's conception of formal
structure as "a provi-sional codification of a state of equilibrium
amongopposing strategies of power" (1980, p. 61). Theseideas are
particularly relevant to questions ofpower and intentionality.
Hierarchy and Power in Coactivational ThoughtGiddens provided
the foundation for a coacti-
vational interpretation of power relations amongorganizational
actors: "The notion of human ac-tion logically implies that of
power, understoodas transformative capacity: 'action' only
existswhen an agent has the capability of intervening,or refraining
from intervening, in a series ofevents so as to be able to
influence their course"(1979, p. 256). As Giddens pointed out, if
oneuses this definition, power may or may not beconsciously
exercised. One's power is deter-mined solely by the potential
impact of one's ac-tions on subsequent events, regardless of
whe-ther or not these repercussions are intended.
Viewed from the coactivational perspective.Actor A's power
hinges on the decision rulesused by other actors and, hence, on
organiza-
tional structure because the decision rules of oth-ers partially
determine the consequences of hisor her actions. Therefore, to
study power rela-tions one must examine the opportunities that
agiven structure of interaction affords each actorto alter the
behavior of other actors and the distri-bution of resources among
them (Mackenzie,1986; Ranson, Hinings, & Greenwood, 1980).
Inparticular, if one actor's behavior is unilaterallyconditioned by
the decisions of another, an au-thority structure can be inferred
(Dow, 1987b);complex hierarchies are characterized by nestedpower
asymmetries of this type.
Systematic investigation of power relationsfrom a coactivational
perspective requires knowl-edge of the decision rules of the
actors, the avail-able channels of resource and informationflow,
and the nature of the exchanges betweenthe organization and its
environment. Since anactor's power is derived from his or her role
inan organizational structure, a theoretical accountof the decision
rules that constitute this structuresimultaneously explains the
power relations em-bedded in it. This broad view of power allowsone
to map the power relations that arise underunconventional forms of
administration (thelabor-managed firm) or in settings devoid of
for-mal hierarchy (relations of bargaining or marketexchange).
Economists often are uncomfortable with theidea of
intraorganizational power because it runscounter to the usual
assumption that market com-petition eviscerates any potential
bargainingpower an organizational actor might seek toexercise.
However, this conventional assumptionbecomes untenable if some
actors control idio-syncratic physical or informational resources
thatcannot be replicated on comparable terms inthe external market.
The advocates of transac-tion cost economics have argued
convincinglythat instances of asset idiosyncrasy are common-place;
indeed, they also help to explain the veryexistence of business
firms in a market economy(Williamson, 1985). Accordingly,
intraorganiza-tional power relations are partially insulated
from
58
-
the discipline of the market, even where the ex-ternal market
has a competitive structure (Dow,1985, in press).Role of
Intentionality in Coactivational Thought
That power need not be exercised intention-ally raises larger
questions: How much intention-ality lies behind the decision rules
of an individ-ual actor, and how much purposiveness shouldbe
imputed to the organization as a collectivity?At the level of the
individual actor, any of thefollowing views would appear to be
compatiblewith the coactivational view.
Hypercognition (the von Neumann actor). De-cision rules are
consciously chosen strategies thatmaximize the expected utility of
the agent, giventhe rules of the game and the strategies beingused
by other actors. Predicted structures arederived by applying a
suitable game-theoreticsolution concept, such as Nash equilibrium
(Dow,1987b; Radner, 1972).
Bounded cognition (the Simon actor). Actorsare "intendedly
rational, but only limitedly so"(Simon, 1976, p. xxviii); they
cannot anticipate allof the consequences that flow from using a
givendecision rule. The problem of calculating conse-quences is
truncated by satisficing, adopting amyopic planning horizon, and
using simplesearch heuristics to modify unsatisfactory deci-sion
rules.
SubcogniWon (the Skinner actor). Decision rulesare shaped by
operant conditioning mechanismsof which the actor may be entirely
unaware.Structures arise spontaneously as a result of re-ciprocal
behavior modification within interactinggroups.
ITie loaded term rationality was avoided inthe preceding
paragraphs because here, the the-oretical issue at stake is simply
the degree towhich foresight determines decision rules. In
par-ticular cases, the choice among hypercognitive,boundedly
cognitive, or subcognitive models willbe governed by criteria of
theoretical tractability,empirical adequacy, and personal
taste.
While coactivational theory can maintain aneclectic stance
toward intentionality at the indi-vidual level, many writers
sympathetic to this
approach have gazed upon ideas of collectivepurpose with
profound suspicion. Giddens, forexample, declared that "social
systems have nopurposes, reasons or needs whatsoever; oniy hu-man
individuals do so. Any explanation of so-cial reproduction which
imputes teleology to so-cial systems must be declared invalid"
(1979, p.7, emphasis in original). Crozier and Friedbergheld that
"we cannot speak of an organization'srationality and objectives as
though they existedin themselves, apart from and above the
indi-viduals and groups who embody them in theirstrategies and
behavior" (1980, p. 46).
Therefore, it is somewhat surprising to find thatCyert and March
regarded their theory as "pre-dicated on the essential
purposiveness of thefirm" (1963, p. 44). However, the Carnegie
tradi-tion derives organizational goals by treating theaspiration
level of each subunit as a constrainton the choice of
organizational policies, so thatany policy satisfying all such
constraints is saidto satisfy the "organizational" aspiration
level(Cyert & March, 1963, chap. 3; Simon, 1964). Be-cause
organizational goals are merely an aggre-gate of disparate
individual goals, real purpo-siveness at the organizational level
of analysis isabsent.
In any event, the technique of superimposingdiverse subunii
aspiration levels breaks downfor more complex goal structures.
Indeed, Ar-row (1963) showed that when individual prefer-ence
orderings are unrestricted there is no gen-eral way to aggregate
these goals to obtain aconsistent collective preference ordering or
so-cial welfare function. If individual goals can beexpressed in
terms of a common medium ofexchange, such as money, tlien it is
arguablyan "organizational goal" to maximize the size ofthe
collective pie prior to its division amongindividuals. However,
even in this special casedistributional issues can be settled only
by re-course to shared norms of justice or the use ofpower.
In short, the notion of collective purpose is ex-cess baggage
for coactivational theory. Aban-doning this concept in no way
precludes interest-ing explanations of organizational
structure.
59
-
because game-theoretic equilibrium concepts, in-teractive
learning processes, and evolutionaryselection mechanisms all remain
as viable ex-planatory frameworks. In none of these frame-works are
organizational structures directly cho-sen by any specific actor or
coalition, much lessby the organizafion as a whole. Even if
organi-zational behavior seems to be rationalizablein terms of some
imputed collective goal(McFadden, 1975, 1976), such goals can be
re-garded as entirely epiphenomenal; they are areflection of
structure, rather than its cause.
Visible and Invisible Hands
All theories of organizational structure are anattempt to
understand patterns of social interac-tion. The two perspectives
described in this pa-per support dramatically different styles of
ex-planatory theory. For example, the two perspec-tives involve
significantly different commitmentsregarding the subjective or
objective nature ofsocial reality, and disagree over the relative
im-portance of social conflict and integration (Burrell&
Morgan, 1979), For present purposes, however,the key area of
contrast involves a distinctionbetween intentionalist "visible
hand" and inter-actionist "invisible hand" explanations of
socialstructure,
Configurational theory posits a visible hand ofauthority which
shapes structure fn the serviceof shared objectives, A structure is
explained byshowing that it enables an organizational de-signer to
accomplish certain tasks in a givenenvironment. In coactivational
theory, on theother hand, structure is seen as a joint product
ofvarious interlocking decision rules or strategiesadopted by
individual actors. Thus, the coactiva-tional view explains
structure by explaining howthese rules arose and why they remain
viablefor the actors who employ them.
Due to conflicts of interest, observed patternsof interaction do
not reflect fully the desires ofany single actor. Therefore, it is
necessary todecipher the processes operating at fhe sociallevel of
analysis which mediate (perhaps onlyincompletely and temporarily)
among the con-
flicting aims of individual actors. Thus, the coac-tivational
view demands an invisible hand ex-planation of organizational
structure.
The concept of an invisible hand is derivedfrom Adam Smith's
(1974) idea that market forcescould yield a coherent allocation of
resourceswithout deliberate guidance from any economicactor. In the
present context, no connotations ofeconomic efficiency or social
harmony are in-tended. Simply, the point is that often one
canexplain patterns of social behavior as the un-planned outcomes
of processes of social interac-tion (UUmann-Margalit, 1978),
An invisible hand approach at the social levelof analysis is
logically compatible with an inten-tionalist view of individual
behavior (Schelling,1978); modem microeconomic theory provides
themost highly elaborated instance of such a blend.Also, although
invisible hand theories can havefunctionalist overtones, as in
economic modelsof perfect competition and in some interpreta-tions
of evolutionary biology (Elster, 1983, chap,2), they need not.
Thus, theories of this type offeran escape from the quicksands of
functionalistorganization theory (Burrell & Morgan, 1979,chap,
5), The game-theoretic concept of Nashequilibrium, for example,
constitutes an invisi-ble hand model of social interaction in the
sensethat the equilibrium position of the game is notdirectly
dictated by any single player. In pris-oners' dilemma situations, a
Nash equilibriumcan prove socially disastrous (Axelrod, 1984),
The clash between visible and invisible handperspectives is
illustrated through controversiesabout corporate culture, A
configurational ap-proach encourages the belief that top
managerscan and should engineer an effective organiza-tional
culture, much as they engineer the coordi-nation of work tasks more
generally. This propo-sition is endorsed to varying degrees by
thecontributors to a volume of readings on the topic,titled Gaining
Control of the Corporate Culture(Kilmann, Saxton, Serpa, &
Associates, 1986),On the other hand, the coactivational
viewstresses the idea that shared beliefs and values,like the
patterns of behavior fhey support, emergefrom processes of social
interaction in which top
60
-
management is only one player among many.Culture in this view is
not an object open todeliberate, unilateral choice by any single
parti-cipantas Riley (1983, p, 418) puts it, culture"should not be
analyzed as something an organi-zation has, it should be analyzed
as part of whatthe organization is,"
At a deeper level, the coactivational viewquestions whether any
tacit organization-wideconsensus on fundamental values and
beliefsactually exists, just as it doubts the axiom ofshared
organizational purposes, Schein poinfedout that within a given
organization one mayfind "a managerial culture, various
occupation-ally based cultures in functional units, group cul-tures
based on geographical proximity, workercultures based on shared
hierarchical exper-iences, and so on" (1986, p, 7),
Organizationsoften are multicultural (Gregory, 1983), and theymay
exhibit conflicts between an "official" mana-gerial culture and
various other cultures. There-fore, it is important to stress that
coactivationalstudies of structure do not hinge on the existenceof
a common culture; as Barley stated, "be-havioral regularity is a
necessary, but not suf-ficienf, condition for fhe explication of a
culturalunderstanding" (1983, p, 398), Even in a socialsystem
wracked by open warfare or class conflict,the structure of
behavioral interaction betweenthe warring camps could still be
analyzed.
The development of configurational ideas bythe classical
managemenf theorists coincided
with the emergence of what Wren (1979, chap,16) terms a top
management viewpoint in organi-zation theory, Configurational
prescriptions tendto be of value primarily to managers of
formalwork organizations in which either (a) no signifi-cant
intraorganizational conflicts exist, or (b)power asymmetries are
substantial, so that man-agers can shape the behavior of
subordinateswith little or no concern for the reciprocal exer-cise
of power by actors with opposing interests.
The coactivational affinity for invisible handexplanations
limits the prescriptions that can bederived from it because
prescription inevitablyinvolves intentionality. However, the
outlines ofa prescriptive coactivational theory can besketched. Any
organizational actor, whetherCEO or production worker, has an
interest indiscovering how changes in the behavior of otheractors
or the environment can be induced byfeasible modifications in that
actor's own deci-sion rules. This search for strategies capable
ofredirecting organizational events is precisely asearch for
previously unrecognized sources oforganizational power. Thus,
coactivational pre-scriptions can help specific actors make their
la-tent power manifest, in order to advance theirinterests. The
resulting struggle between the in-visible hands which have shaped
our histori-cally given structures and the visible hands work-ing
to transform them provides much of thedrama to be found in the
modern world.
References
Allison, G, T, (1971) Essence o/decision. Boston: Little,
Brown.Annour, H., & Teece, D. (1978) Organizational structure
and
economic performance. Bell lournal of Economics, 9,106-122,
Arrow, K, (1963) Social choice and individual values (2ndod,).
New Haven, CT: Cowles Foundation, Yale University,
Arrow, K. (1974) The iimits of organization. New York:
Norton.Axelrod, R. (1984) The evolution of cooperation. New
York:
Basic Books.Barley, S, R, (1983) Semiotics and the study of
occupational
and organizational cultures. Administrative Science Ouar-terly,
28, 393-413,
Barnard, C, (1938) The functions of the executive. Cambridge,MA:
Harvard University Press,
Beckmann, M, J, (1977) Management production functionsand the
theory of the firm, lournal of Economic Theory, 14,1-18,
Blau, P. M, (1974) On the nature of organizations. New
York:Wiley,
Blau, P, M., & Schoenherr, R. A. (1971) The structure
oforganizations. New York: Basic Books.
Burrell, G., & Morgan, G. (1979) SociotogicaJ paradigms
andorganizationai analysis. London: Heinemann,
61
-
Calvo, C, A,, & Wellisz, S, (1978) Supervision, loss of
control,and the optimum size of the firm, lournal of
PoliticalEconomy. 86, 943-952,
Calvo, C, A, & Wellisz, S, (1979) Hierarchy, ability,
andincome distribution, lournal of Political Economy,
87,991-1010,
Caves, R, (1980) Corporate strategy and structure. Journal
ofEconomic Liferafure, 18, 64-92,
Chandler, A, (1962) Strategy and structure, Cambridge, MA:MTT
Press,
Chandler, A, (1977) The visible hand. Cambridge, MA:Belknap,
Child, J, (1972) Organization structure, environment,
andperformance: The role of strategic choice. Sociology,
6,1-22,
Coase, R, (1937) The nature of the firm. Economica, 4,
386-405,Cohen, M, D, (1981) The power of parallel thinking.
Journal
of Economic Behavior and Organization, 2, 285-306.Cohen, M, D,
(1984) Conflict and complexity: Goal diversity
and organizational search effectiveness. American Politi-cal
Science Review, 78, 435-^51.
Cohen, M, D,, March, J, G,, & Olsen, J. D, (1972) A
garbagecan model of organizational choice. Administrative Sci-ence
Quarteriy, 17, 1-25,
Crawford, V, (1985) Learning behavior and mixed-strategyNash
equilibria. Journal of Economic Behavior and Orga-nization, 6,
69-78.
Crecine, I. P, (1969)Governmen(a]probiem-soJving, Chicago:Rand
McNally,
Cremer, ], (1980) A partial theory of the optimal organizationof
a bureaucracy. Bell lournal of Economics, 11, 683-693.
Crozier, M,, & Friedberg, E, (1980) Actors and
systems.Chicago: University of Chicago Press,
Cyert, R,, & March, I, (1963) A behaviorai (heory of (he
firm.Englewood Cliffs, NJ: Prentice-Hall.
Danziger, I, N. (1978) Making budgets: Public resource
allo-cation. Beverly Hills, CA: Sage,
Dow, G, (1985) Internal bargaining and strategic innovationin
the theory of the firm, /ournai of Economic Behaviorand
Organization, 6, 301-320,
Dow, G, (1986) Control rights, competitive markets, andthe labor
management debate. Journal of ComparativeEconomics, 10, 48-61,
Dow, G. (1987a) The function of authority in transaction
costeconomics, /ournai of Economic Behavior and Organiza-tion. 8,
13-38,
Dow, G, (1987b) Self-enforcing authority structures.
Unpub-lished manuscript. University of Alberta, Department
ofEconomics,
Dow, G, (in press) Information, production decisions,
andintra-firm bargaining. International Economic Review.
Elster, J, (1983) Explaining technical change. New York:
Cam-bridge University Press.
Fayol, H, (1930) Industrial and general administration (J,
A,Coubrough, Trans,). London: Pitman.
Friedman, J. W., & Rosenthal, R. W. (1985) A positive
ap-proach to noncooperative games, (Working paper 85-5),Chapel
Hill: University of North Carolina, Department ofEconomics,
Gerwin, D, (1969) A process model of budgeting in a publicschool
system. iVfanagemen( Science, 15, 338-361.
Giddens, A. (1979) Cen(rai probiems in social theory.
London:Macmillan.
Gregory, K, L, (1983) Native-view paradigms: Multiple cul-tures
and culture conflicts in organizations. AdministrativeScience
Quarterly, 28, 359-376,
Groves, T,, & Radner, R, (1972) Allocation of resources in
ateam, /ournai of Economic Theory, 4, 415-441,
Gulick, L, H. (1937) Notes on the theory of organization. In
L,H, Gulick & L, Urwick (Eds.), Papers on the science
ofadministration (pp. 3-45). New York: Institute of
PublicAdministration.
Horvat, B, (1982) The political economy of socialism. Armonk,NY:
M. E, Sharpe.
Hurwicz, L. (1985) Information and incentives in
designingnon-wasteful resource allocation systems. In G. R.
Feiwel(Ed,), Issues in contemporary microeconomics and wel-fare
(pp, 125-168), Albany: State University of New YorkPress,
Kilmann, R, H,, Saxton, M. J., Serpa, R., & Associates
(1986)Gaining control of the corporate culture. San
Francisco:Jossey-Bass.
Lakatos, I, (1970) Falsification and the methodology of
scien-tific research programmes. In I. Lakatos & A,
Musgrave(Eds.), Criticism and the growth of knowledge (pp,
91-195).New York: Cambridge University Press,
Lawrence, P,, & Lorsch, J, (1967) Differentiation and
integra-tion in complex organizations. Administrative
ScienceOuarteriy, 12, 1-47,
Mackenzie, K, D. (1976) A theory of group structures (2
vols,).New York: Gordon and Breach.
Mackenzie, K, D, (1986) Virtual positions and power. Man-agement
Science, 32, 622-642.
62
-
March, I,, & Simon, H, A, (1958) Organizations, New
York:Wiley,
Marschak, J,, & Radner, R, (1972) The economic theory
ofteams. New Haven, CT: Yale University Press,
Mayr, E, (1982) The growth of biological thought. Cambridge,MA:
Belknap,
McFadden, D, (1975) The revealed preferences of a govern-ment
bureaucracy: Theory, Beii Journal of Economics. 6,401-416,
McFadden, D, (1976) The revealed preferences of a govern-ment
bureaucracy: Empirical evidence. Bell Journal of Eco-nomics. 7.
55-72,
Mirrlees, I, A, (1976) The optimal structure of incentives
andauthority within an organization. Bell Journal of Econom-ics, 7,
105-131,
Mooney, J, D,, &Reiley, A, C, (1939) The principles
o/organi-zation. New York: Harper,
Nelson, R, (1981) Assessing private enterprise: An exegesisof
tangled doctrine, Beii Journal of Economics. 12, 93-111,
Nelson, R,, & Winter, S, (1982) An evoiutionary theory
ofeconomic change. Cambridge, MA: Belknap,
Noble, D, (1984) Forces of production. New York: Knopf,Padgett,
J, (1980) Bounded rationality in budgetary research,
American Political Science Review. 74, 354-372,Perrow, C, (1967)
A framework for comparative organiza-
tional analysis, American Socioiogicai Review, 32,
194-208,Powell, W, (in press) Neither market nor hierarchy:
Network
forms of social organization. In B, M, Staw & L, L,
Cum-mings (Eds,), flesearch in organizationai isehctvior,
Green-wich, CT: JAI Press,
Pugh, D, S,, Hickson, D, J,, Hinings, C, R,, & Turner,
C,(1968) Dimensions of organizational structure. Administra-tive
Science Ouarteriy, 13, 65-105,
Pugh, D, S,, Hickson, D, I, Hinings, C, R,, & Turner, C,
(1969)The context of organization structure. Administrative
Sci-ence Ouarterly, 14, 91-114,
Putterman, L, (1984) On some recent explanations of whycapital
hires labor. Economic inquiry, 22, 171-187,
Radner, R, (1972) Normative theories of organization:
Anintroduction. In C, B, McGuire & R, Radner (Eds,),
Decisionand organization (pp, 177-188), Amsterdam:
North-Holland,
Radner, R, (1975a) Satisficing, Journal of Mathematical
Eco-nomics, 2, 253-262,
Radner, R, (1975b) A behavioral model of cost reduction.
BellJournal of Economics, 6, 196-215,
Ranson, S,, Hinings, B, ,& Greenwood, R, (1980) The
structur-ing of organizational structures. Administrative
ScienceOuarterly. 25, 1-17,
Riley, P, (1983) A structurationist account of political
culture.Administrative Science Quarterly. 28, 414-437,
Rosen, S, (1982) Authority, control, and the distribution
ofearnings, Beii/ournai of Economics, 13, 311-323,
Russell, R, (1985) Employee ownership and internal gover-nance.
Journal of Economic Behavior and Organization,6, 217-241,
Schein, E, H, (1986) Organizationai cuiture and leadership.San
Francisco: Jossey-Bass,
Schelling, T, C, (1978) ilficromotJves and macroJbehcrvior,
NewYork: Norton,
Schotter, A, (1981) The economic theory of social
institutions.New York: Cambridge University Press,
Simon, H, A, (1964) On the concept of organizational
goal.Administrative Science Quarterly. 9, 1-22.
Simon, H, A, (1976) Administrative behavior (3K1 ed,). NewYork:
Free Press,
Smith, A, (1974) The wealth of nations. New York:
Penguin,Spence, A, M, (1975) The economics of organization: An
introduction, Beii/ournai o/Economics, 6, 163-172,Steinbruner,
J, D, (1974) The cybernetic theory of decision.
Princeton, NJ: Princeton University Press,
Thompson, J, D, (1967) Organizations in action. New
York:McGraw-Hill,
Ullmann-Margalit, E, (1978) Invisible hand
explanations,Syntiiese, 39, 263-291,
Urwick, L, (1937) Organization as a technical problem. In L,H,
Gulick & L, Urwick (Eds,), Papers on the science of
ad-ministration (pp, 49-88), New York: Institute of
PublicAdministration,
Urwick, L, (1943) The elements of administration. New
York:Harper and Brothers,
Ward, B, (1967) The socialist economy. New York:
RandomHouse,
Weber, M, (1978) Economy and society (Vols, 1-2) (G, Roth
&C, Wittich, Eds,), Berkeley: University of California
Press,
Weick, K, (1979) The social psychology of organizing. NewYork:
Random House,
Weitzman, M, (1974) Prices vs, quantities, Review of Eco-nomic
Studies, 41, 477-492,
Wildavsky, A, (1964) The politics of the budgetary
process.Boston: Little, Brown,
Williamson, O. (1967) Hierarchical control and optimum firmsize,
/ournai of Political Economy. 75, 123-138,
Williamson, O, (1975) Markets and hierarchies. New York:Free
Press,
63
-
Williamson, O, (1980) The organization of work. Journal
ofEconomic Behavior and Organization. I, 5-38,
Williamson, O, (1981) The modern corporation: Origins,evolution,
attributes. Journal of Economic Literature, 19,1537-1568,
Williamson, O, (1985) The economic institutions of capital-ism.
New York: Free Press,
Woodward, J, (1965) Industrial organization: Theory and
prac-tice. London: Oxford University Press,
Wren, D, (1979) The evoiu(ion of management thought (2nded,).
New York: Wiley,
Gregory K. Dow (Ph.D., University of Michigan) isAssociate
Professor of Economics in the Departmentof Economics, University of
Alberta. Correspondenceregarding this article should be sent to
Gregory K.Dow, 8-26 Tory Building, Department of
Economics,University of Alberta, Edmonton, Alberta T6G
2H4Canada.This paper has been improved substantially by com-ments
and criticism from Charles Perrow, Oliver Wil-liamson, and Paul
DiMaggio. The author also ac-knowledges the pervasive influence of
Sidney Winterand Richard Nelson, though these colleagues bear
noresponsibility for the unintended consequences of
theirremarks.
64