Vous êtes sur la page 1sur 20

T he N ew Handbook of

A d v a n c e s in T h e o r y R e s e a rc h , a n d M e th o d s
A G H D30.3 N 49
The new handbook of organlzatlonal
communlcation : advances m theory,
T h o us d r d Qa k s , C a l. ; Lo r .d o n : Sa ge , C2001
Conceptual Foundations
* University of Colorado at Boulder
r*or al! of recorded history, people have
J L studied and discussed communication pro-
cesses within their dominan! organizations. In
many respects, these discussions differ little
from those presen! during (he past three de-
cades of institutional organizational commu-
nication study. They have been concerned
with the systematic mannejs by which com-
munica on practices can be used to help coor-
dnate and control the activities of organiza-
lional members and relations with externa!
constituencies. Our current situaon is one of
rapid social and organizational change putting
great pressure on researchers today to contin-
ually develop useful concepts and studies to
match the complex interactions characterstic
of contemporary workplaces.
Organizational communication research is
itself a rich communicative process. Re-
searchers have developed and used their theo-
res and research activities for many positive
organizational outcomes. But ther work also
accomplishes a variety of inlertwined Ufe pur-
poses, including the distinction of the re-
searcher and the development and advance-
ment of specifc group interesls. Fundamental
assumptions about the nature of the world,
methods of producing knowledgc, and vales
are developed and advanced in the discourse
of rescarchers. Such assumptions are ncces-
siry to produce any kind of undcrstanding and
knowledge and are usually mosl conlested
during period o' rupil chungc. Wliilc funda-
mental assumptions thcmselvcs iir nol opcn
4 *Theoretical and Methodological Issues
lo refutation, they are to exploration. Scholars
righlfully ask of any research program, 'To
solve whal problems?" "To what ends?"
"Whose meanings?" "Whose knowiedge?"
This essay hopes to foster useful discus-
sions regarding how different scholars con-
struct knowiedge and j ust i fy practices about
organiza!ions, and also about the vales,
hopes, and groups' inters! that they support.
To that end, I begin with an overview of how
the term organizational communication is
used-what it delimits, organizes, or draws
our attention to. Following this introduction,
the central par of the chapter will develop a
two-dimensional scheme for directing atten-
t i on to similarities and differences among re-
search programs. I wi ll arge that the most in-
teresling differences among social research
programs can be displayed through looking at
{ I ) i l i < - i \ of n i k - i . u i i n n particular rcscarch-
etJ 1 . 1 vi n v n l nllitT groups, charactcrized a s
/ / . V I I M i ; ci f versus flite/a priori concep-
t mi i '.. mu ( .') i l i c i novf s i hr researc h ac l i vi t y
n u i l i r p i u i n u i k r l o w i i n l c l o su rc o r i n dc termi -
n. i i v i " i l i . n i n i r i . i i ( u n . f l i ; i i ; u-| f r i ml as con-
i < i ni n M - r k n ^ . v e i M i s tlissensus sc ek i n g.
I l i r' . c t wu ' l i n i i i r . i n . when pu l l ogel hc r pro-
v u l r u i w n l i y ( w u i i i n i i i x c harac teri zi n g di f-
t - i - i M - M - . i i t l i | > ni } '.N i ms. I wi l l di sc u ss
r ml i o lum "idt-iil lype" research programs
I . I . . . I M . ni M I t l i r , ( '.'"( I l -H K i l l y, 1 wi l l conclude
hy i . ' . l mj ' ut me rcsearcher's choice pro-
i'csses in Ihc contcmporary conlexl and look-
ing al f ut ur o research agenda.
What is organizational communication? The
possibility of a shared answer to that ques-
tion seems to be implied in producing a hand-
book of organizational communication or in
detailing conceptual foundations for organi-
zational communication studies. Clear and
simple answers can be given. I could jusl pro-
vide a defnition, compare it with altemative
defi ni li ons, and get on with a review. Defini-
tions are nice; they set clear boundaries and
j ust i fy my looking at the things I ara inter-
ested in, and excluding the rest. But such def-
initions are inevitably arbitrary, usually pro-
vide political advantage for some group, and
can as easily produce blinders as insight. N ot
only is debate possible over altemative defi-
nitions but also over the act of defming
(Deetz, 1 992, chap. 3; Smith, 1 993; Taylor,
1 993). Ultimately, the question "What is or-
ganizational communication?" is misleading.
A more interesting question is, "What do we
see or what are we able to do if we t hi nk of
organizational communication in one way
versus another?" Unlike a defnition, the at-
tetnpt here is not to get it right, but to under-
stand our choices. Rather than killing the bird
("defnition" defnitio, to kill or make final)
and getting on with the dissection, perhaps
we should watch it fly for a while.
Three very different ways of conceptualiz-
ing "organizational communication" are
available. Each of these provides different "at-
tcntions" and different boundaries regarding
what should be covered in this volume and
this chapter. Such conceptions guide research
and teaching as well as provide an identity to a
group of scholars. First, the focus could be on
the development of organizational communi-
cation as a speciality in departments of com-
munication and communication associations.
Organizational communication study is what-
ever anyone does who is a member of these di-
visions or publishes in particular j ournals
(see, e.g., Smeltzer, 1 993). Like with other
"sociologies of fields" time can be spent
looking at the history of this development,
what members of these divisions have studicd
and published, how many students maj or or
achieve advanced degrees in this speciality,
and how many J obs are available (see
Redding, 1 979). These are not unimportant
concerns and such a conception either explic-
itly or implicitly has bccn used to determine
what is or is not an organizational communi-
cation study in many if not most literature re-
views (see K rone, J ablin, & Putnam, 1 987;
M eyers, Seibert, & Alien, 1 993; Putnam &
Cheney, 1 985; Redding & Tompkins, 1 988;
Conceptu? Foundations
Richetto, 1 977; Wert-Gray, Center, Brashers,
& M eyers, 1 99I).1 From these reviews we of-
ten gain more understanding of people, their
relations, careers, and university poiitics than
we do about the underiying conceptions of or-
ganizations and communication. M oving
from reviews of studies to examining altema-
tive theories in organizational communication
studies and the social problems such studies
address is often difcult. It is not surprising
that these reviews often contain laments about
the disunity of the field. This may well be an
artifact of the organizing principie used.
A second approach to conceptualizing or-
ganizational communication focuses on com-
munication as a phenomenon that exists in or-
ganizations. If such an object can be defined,
then anyone who looks at or talks about that
object is studying organizational communica-
tion. This is the logic behind many textbook
definitions of organizational communication.
Wi t hi n this logic, any number of individuis
from different academic units might study this
phenomenon. In such a case, interdisciplin-
arity might be expected. With this focus one
might ask what is "communication" in the or-
ganization and what is something else, what
are the ways the phenomenon can be usefully
subdivided, what are the variables that aflea it
or it affects, and what theories adequately ex-
plain it. H andbooks like this one usually work
from this type ofconcepton (see K rone et al.,
1 987). They assume that a unied phenome-
non exists, and they form chapters based on
subdivisions of the phenomenon or altemative
sites where it appears. Introductory chapters
like this one typically focus on the variety of
ways that the same phenomenon has been ex-
Unfortunately for such a taek, many of ihe
contemporary theories of organizations and
communication deny that a unitary phenome-
non exists out there. Thus, Ihe phenomenon
organizational communicationis difier-
en! for different theories. "Organizaonal
communication" is not one phenomenon with
many explanations; each form of cxplanation
may conceptualize and explain a different
phenomenon. Fixed subdivisions are always a
kind of theoretical hegemony (where one the-
ory's "organizational communication" is priv-
eged over undiscussed others). When this
happens, theory debate is reduced to method-
ological perspectivalism. When thought of as a
distinct phenomenon, the conception of "or-
ganization" is often reduced to a "site" and the
conception of "communication" often be-
comes narrow with social interaction concep-
tually reduced to empirical acts of informa-
tion transfer, often the lowest common
denominator (or dominator) in organizalional
communication (for discussion, see Axley,
1 984; Putnam, Phillips, & Chapman, 1 996;
Smith, 1 993; Taylor, 1 993).
A third way to approach the issuc is to
think of communicalon as u way u describe
and explain organi/ ations. In ( he sanie wi i y
that psychology, sociology, or cconoum-s nin
be thought of as capuhle ol 'e xpl ni ni i i ^ UI J M M I
zations' processcs, communication t mf.hi . ! .
be thought of as a di st i nct modc of rxplnrt i i
tion Or way of t hi nki ng about oij '.aiu/ nluM r,
(see Deetz, 1 994a; Pearce, 1 989). Coinimini
catin theory can be used to explain (he pro
duction of social structures, psychological
states, member categories, knowiedge, and so
forth rather than being conceptualized as M U
ply one phenomenon among these others in
organizations. The focus would be on the pro-
cess of organizing through symbolic interac-
tion rather than on "communication" within
an "organization" (H awes, 1 974). From such
a perspective the interest is not in theories of
organizational communication but in produc-
ing a communication theory of organizations
(Deetz, 1 994a). H istorically, few scholars in
the academic units of organizational commu-
nication have approached the issue this way.
Until recently, psychological or socal-cul-
tural explanations have been more often used
in most studies. Gradually, since the early
1 980s, scholars in communication depart-
ments as well as a large number of non-U.S.
scholars and some scholars from other aca-
demic units have focused on organizations as
complex discursivo formalions where discur-
6 4 Theoretica! and Methodologica! Issues
sive practices are both "in" organizations and
productive of them. Because of the tendency
to delimit organizational communicaon as a
professional unit or a distinct phenomenon,
until recently non-U.S. and non-communica-
tion scholars were often absent from reviews
(e.g., the various works of Knights, Willmott,
Hollway, Cooper, Burre, Gergen, Power,
Townley, and Alvesson).
In this review, I will accept this third way
of thinking about organizational communica-
tion. The recursiveness of this position means
that the production of the field as an academic
unit and organizational communicaon as a
distinct phenomenon are themselves discur-
sive accomplishments. My analysis will thus
work at a metalevel from which conceptions
of organizations and processes in them by re-
searchers and "organizational members" can
themselves be seen "communicalionally."
This view allows a deeper analysis that can
display how study rcsults are produced rather
Ihun j usl providing here anothcr review of re-
siilts Ironi difieren! rcsearch programs. The
dui l i l y o Muil y ng I mni . i i i inlcraction in a
i- . MI . lo, .mol . u n i l n - ussumplion that hu-
i i t i i n i n i f i . i l non is u core fbrmative fcature of
wt n l i l i nu-. i i i i i i ion compl cales analysis much
I nil uKogi cnl l y uinclicsit.
l .nir1y I I nl l uw ihe inslruclion given by
The sntial icicnccs dcal wiih prcnamed,
] ni - cl ; i s s i hf (l ic.il il ii-s wliich huar proper nontis
and common nouns, iitles, signs and acronyms.
Al the risk of unwiltingly assuming responsi-
bility for the acts of constitution of whose logic
and nccessity they are unaware, the social sc-
ences must take as their object of study the so-
cial operations of naming and the rites of ins-
tution through which they are accomplshed.
(p. 106)
Attention can be drawn to how both they who
study and they who partic pate "in" orga-
nizatioas produce phenomena in the world
such as "organizations," "communication,"
"needs," "motivations," "informaton," "prof-
its," and various personal and social divisions
such as "men" and "women," "workers" and
"management." Following this tack requires
some understanding of a nonrepresentational
or constitutive view of language that cannot
be developed here at any length but ought to
be familiar enough to most readers that a
short development will suff ce (see Deetz,
1992, chap. 5).
In line with modem discourse theory, con-
ceptions are always contests for meaning (see
Epstein, 1988; Weedon, 1987). Language
does not ame objects in the world; it is core
to the process of constituting the inde-
terminant and ambiguous external world into
specific objects. The appearance of labeling
or categorizing existing objects is derived
from this more fundamental act of object con-
stitution through language. The world thus
can be consttuted in many ways depending
on alternative systems of valuing. The most
significant part of this contest for object con-
slitution is the capacity to enact the lines of
distinction producing some things as alike and
others as different. Only secondarily is the
contest over the positive or negative valance
ascribed to the produced things. For example,
feminist writers for years have shown how
male dominance is maintained by the domi-
nant group's ability to define the dimens ons
of difference and position themselves at the
positive end of each dimensin (see Treichler,
1989; Weedon, 1987). Marginalized groups,
following this analysis, are defined as "the
other" thus acquiring an identy and valued
funcn'ons but only as given by the opposition
pole in Ihe dominant group's conceptual map
(e.g., "emotionally supportive" rather than "ra-
tional" or "private" rather than "public").
They acquire a type of autonomy but only in a
language/concepnial game not of their own
choosing. In accepting the state of "other,"
they have little self-defnition and the game is
stacked(seeBourdieu, 1977,1991).
From the communicative metaperspective
taken here, the core process in understanding
alternative research programs is to understand
their discoursehow they perceive, think,
and talk about organizational life. Under-
standing a discourse includes idenFicatin of
Concepta/ Foumtations
the object distinctions diey make, whose lan-
guage is used in tnaking those object distinc-
tions, what and whose vales and interests are
carried with those distinctions, and how the
conflicting descriptions of Ihe world are han-
dled as well as exploring their processes of
self-justification and distinction from alterna-
tive research programs. Further, research pro-
grams differ in the extent to which they recog-
nize and make explicit their own constitutive
activities. Many researchers assume mal they
are merely discovering and naming real-world
objects. To the extent that this is done much of
the micropractice of research is missed.
Trying to produce any organizing scheme of
these discourses accounting for different the-
oretical conceptions, methodological prefer-
ences, and valu commitments is filied with
difficulties. Each research program might
well use different ways of comparing and
contrasting itself with other programs. In fact
a primary way that any research program es-
tablishes itself is in its means of distinction,
bolh in the sense of producing a difference
and giving itself the positive terms (see
Bourdieu, 1991).
Many schemes have been proposed for or-
ganizing and ihinking about alternative re-
search programs. Most of these classify stud-
ies based on subdivisions of the organiza-
tional commuication phenomenon or differ-
ences in research methods. For example,
Wert-Gray et al. (1991) suggest three domi-
nant reas of work: (1) information flow and
channels, (2) climate, and (3) superior/subor-
dinate. And Redding and Tompkins (1988) di-
vide the work into (1) formal channels. (2)
superior/subordnate communication, (3) in-
forma] channels, and (4) measurng and data
collection_[Putnarn and Cheney (1985) sug-
gest (1) channels, (2) communicaon climate,
(3) superior/subordinate, (4) network analy-
sis, and (5) communication media with addi-
tional emerging perspectives. And in perhaps
the most exhaustivo study, Alien, Gotcher,
and Seiben (1993) review 17 reas of work:
(1) interpersonal relations, (2) communication
skills, (3) culture and symbolism, (4) informa-
tion flow and channels, (5) power and infiu-
ence, (6) decisin making and problem solv-
mg, (7) communication networks, (8) com-
munication and managemcnt styles, (9) or-
gani/ation-cnvironment interfacc, (10) Icch-
nology, (1 1 ) language and mcssaRes, (12)
struclurc, (13) unccrliiinly mu intotiiialion mi
equacy, (14) groups, (15) cthics, (16) cross-
cullunil, and (17) climale.iThcsc divisions and
study counls are interestinj; and represen! ways
of thinking aboul the field that are fairly com-
mon. But such arjproaches tend to reify tpica!
divisions thal are the constructed outcomes of
discursivo processes thus treating them as natu-
ral raiher than produced, hiding vales and as-
sumptions, and disowning the way mese divi-
sions preference particular studies of commu-
nication. Let us consider for a moment these
First, the tpica! orientation is itself not a
neutral classificaton too!. It assumes and re-
produces a particular view of communication
and organizations. For example, it assumes an
atomstic orientation to the world like the
19th-century natural science model and ad-
vantages studies that follow that model.
Studies based in holistic assumptions, such as
ethnographic approaches, may get put in a
category like "culture" or "climate." This
makes "culture" into one phenomenon among
olhers in organizations that can be studied.
Not only do cultural studies deny that culture
is one lliiii)' among many in organizations; the
classication hures the important things that
etKnographic researchers said about organiza-
lions' structures and aclivities like channels
and interpersonal relatiom. Only sludies that
explicitly study channcls and intcrpcrsonal re-
lations as Kolatcd phenonu-nn appeur in those
calegorics. ILlhnonraphtc rescnrchcrs rarely
study a lopic, ihcy Mudy ,. particular site.
Whal would wc Iran il wc clnfisificd hy site.
8 4 Theoretical and Methodological Issues
the social problem considered, group alle-
giance, or the moral stance rather than topic?
Topical divisions probably made sense when
the vast majority of researchers believed that
the elements of organizations were atomisiic
rather than holistic, that organizaons were
primarily a thing rather than a process, and
that communication was a phenomenon
among others rather than an approach one
takes to organization studies. As these change
so must our ways of accounting for similari-
ties and differences in organization studies.
Second, the devices of data collection
shape the review in further ways. In some of
these reviews, the data pool is limited to stud-
ies published in "communication" journals
and the manner of display is usually the num-
ber of essays. The classifying processes
match assumptions of the natural science
model thus both normalizing its preferred
manner nf rtiport and ovcrcmphasizing its im-
pucl. llie "field" lontts diffcrcnt in reviews
lluil considcr scholarly book chaptcrs, schol-
urly buoks, iind/or unpublished research re-
pui K tu uimpanics inslcad of Journal anieles,
hui l l n, Iho iTonccpl of "studies" itself tends to
|[cl drl mrd in tcrms of data collection, thus
mu I y lie und conceptual work, which often
liave grcal impucl on the field and its prac-
n < - , . lund tu be left out. And further yet, the
conccpi of llie "communication field" has of-
len Icd lo the oniission of non-North Ameri-
can studies that organize "fields" differently
and important "communication studies" on
topics that are definilionally excluded. For ex-
ample, the discursivo studies in such journals
as Organizacin or Organization Studies,
communication-based studies in Accouning,
Organizations and Society, and interaction
studies in "sociology" are left out, and works
by authors in management schools in such
journals as Management Communication
Quartery are included. The tendency is to
bias the counts toward studies from a psycho-
logical and managerial perspectiva. And fi-
naliy, quantity of studies as a measure favors
narrow quantitative analyses. What if we mea-
sured significance of impact, transformative
potential, or applicability to wider stakeholder
interests? Each of these would provide differ-
ent pictures of "our" studies and contribution
and pressure the field's development in differ-
I think we get further if we look at the prac-
tice of research and researcher commitments
rather than looking at topics as if they could
be freed from the researcher's orientation. As
we become more diverse as a people and as
researchers, a consideration of general re-
search assumptions becomes more instructive.
Reviewers looking at research assumptions
and orientations have tended to focus more on
methodological/epistemologicat differences
than study topics. And rarely have they gone
beyond methodological choices to a full con-
sideration of the way theoretica! and valu
commitments are carried with them.
Reviews that have considered research ori-
entations have fairly high agreement in cate-
gories of classification. Putnam (1982;
Putnam & Pacanowsky, 1983), for example,
describes studies as functionalist, interpretive,
and critical. Redding and Tompkins (1988)
describe them in a parallel fashion as modern-
ist, naturalisc, and critical (a scheme fol-
lowed by Wert-Gray et al.. 1991, in their
methodological orientations). These authors
would probably add "postmodernist" if they
were writing these essays today. I suspect that
these divisions are largely a result of the influ-
ence of Burrell and Morgan's (1979) popular
discussion of sociological paradigms as func-
tionalist, interpretive, radical humanist, and
radical structuralist. Their paradigm descrip-
tions have been very influential n manage-
ment and communication studies, and the in-
fluence is well deserved. While I believe
fundamentally fiawed, their approach serves
as a useful point of departure for further de-
velopment (seeDeetz, 19%).
Importantly, Burrell and Morgan's discus-
sion of paradigmatic differences in the late
197Qs gave legitimacy to fundamentally dif-
ferent research programs and enabled the de-
velopment of different criteria for the evalua-
tion of research. Their exhaustive review was
Conceptu/ Foundotions * 9
not only valuable in itself, but they were able
to provide an analysis that probed deeply into
the assumptions on which different research
programs were based. But harms were also
created. I believe thal there are reasons for this
significanl influence beyond the clarity of pre-
sentation and exhaustive compilation of litera-
ture. Whcn the grid and discussion were pub-
lished in 1979, those of us doing alternative
work readily embraced the grid for it gave
each of us a kind of asylum. While some of us
were uncomfortable with the dimensions and
philosophical analysis, we happily accepted
the new-found capacity to present ourselves to
maitistream critics as doing fundamentally
different, but legitmate, kinds of research and
began to work on concepts and evaluation cri-
teria within our now produced as different and
unitary communies. Many of those doing
more mainstream work also found it appeal-
ing since, as I will arge, the conceptual dis-
tinctions Burrell and Morgan used to produce
the grid were the same distinctions (he main-
stream tradition had used to discuss different
research agendas. Thus, they reaffirm that tra-
dition's conceptual map and provide a "safe"
understanding of the developing alternatives.
Further, the conception of paradigms as dis-
tinct schools of thought with their own prob-
lem statements and evaluative criteria could
be used by the dominant "functionalists" to
protect themselves from growing criticism
(the isolationist strategy noted by Reed,
1985). They too would have a safe and sep-
rate place (see Rodrguez & Ca, 1994).
But as organization science and organiza-
tional communication research have contin-
ued to evolve, problems with the Burrell and
Morgan grid and its adaptations have become
more pressing. While not primarily a result of
the original analysis, the four-paradigm solu-
tion has often led to quick categorizations and
to debates around paradigm commensurabil-
ity and appropriate use of the different para-
digms (Hassard, 1991; Jackson & Crter,
1991; Parker & McHugh, 1991; Willmott,
1993). Some of these problems and debates
arise from the tendency to reify concepts, es-
pecially in educational programs and malcri-
is. The Burrell and Morgan grid can easily
produce four unitary paradigms, rather than
provide two Unes of differentiation that draw
attention to important differences in research
programs. Burrell and Morgan invite reifica-
tion by claims of paradigmatic incommen-
surability, by staying at the level of theory and
reconstructed science, and by accepting
Kuhn's lose conception of paradigms. The
dimensions of contrast can be used as a way of
focusing attention to differences that makc a
difference rather than as a mcans of classi-
fication, but few writers and leachers h:we
done so.
But my ni . mi concern r. mtl pi mul i ^ Mi
commcnsurability or reificiilion bul i;illin i l u-
dimensions of contras) llicmst-lvcs A ilrrpri
a nd more i nl ci cst i n^ uiulc standui} > l . < > . !
temporary rescind prutlu/rs ml di - l > . i h . i .
possible by (uctisiiif, on otlin mun i . n - i i .
mensions. Tlic i]ueslum is r u. i Ai r i d- M i h <
right categories or who M-. in c < l i ' hu An
these differences lluil ninhr .. di l l ri ri i t r1.' 11,
these di me n.s ion s provuk* niMglil i ndi f
differcnccs in rescind pioxtuiiiN1' I 1m| -
ai d i i ' t h i nk i i i j ' . l l i r < )i l l cl i * l ucs nnd . mu . mu.
among diflcrci icsniu h i|>|inwii lies, wilh ihr
aim of making our conflicls and tlisaminns
more productive rather than smiply rc|>hicing
four boxes with four diffcreni boxcs. In iii;my
ways, the various adaptations of Uurrcll and
Morgan have hampered the development of
new research agenda and Icd to Icss than pro-
ductive conceptions in the field.
Burrell and Morgan, and subsequently
many organizado nal communication scholars,
largely accepted the conceptual distinctions
from sociological mctionalism and its sup-
porting philosophy of science. Burrell and
Morgan performed a polilical intervention as
they spoke on behalf of the oppositions, the
negave terms, the "others** in "sociological
functionalism's" conceptual map. For exam-
ple, they accepted the traditional functionalist
"subjective/objective" distinction but pro-
vided a careful development of "subjective"
research. Thus, using the dominant concep-
10 4 Theoretical and Methodological Issues
t i ons, they merely asked, "Who is 'other'?"
and "In what ways are they 'other'TBul they
never questioned whether distinctions based
on such conceptions as "subjective/objective"
were useful at all (see Deetz, 1994a). In con-
trast to their analysis, each "other" (each
raarginalized paradigmatic group like "nter-
pretivists" or "radical humanists") would have
defined its difference from the dominant func-
tionaiist conceptions differently, that is, if
they accepted their "groupness" at all (see
Bernstein, 1983; Natter, Schatzki, & Jones,
1995). This positioning, as I have suggested,
partly accounts for the rapid acceptance of the
Burrell and Morgan's grid i nt o the main-
stream of management science and organiza-
lional communicalion discussions.
Furthcr, t h s move protected functionalist
rcscarchcrs from the most damni ng critiques
(and incs llicy would nol understand, e.g-, the
"arli f acli ml" qual i t y of t hei r "facts") in favor
of Ihci i prcfcrrcd baltlcs (e.g., between their
"ohjrcl vi ly" and othcrs' "subjectivity"). At
Iho same t i me, ilu- niosl i nnovat i ve of the new
n . u . d. r. l u i u i t l i i now i -vi ri l more di f f i cult
I i i i . . \ v l i ; i i t t u - y i l u l s i ncc ihey had t o u s e
n Inni '.i i .i f.f in wli i ch t hei r mcanings did not ft
n ( . . i MU al ilwori.sts and phenomenologists
wlio t li d nol iicccpt "subjcct/object" dualism
hnil le ncccpl Ihc classi fi cat i on as "subjective
l i i i i i i . i n i ' . t . ' i l ' t l i cy werc lo haveahomeat all).
Tlu-y had lo choose bctween misrepresenting
tlicmsclvcs clcarly t brough B u i c l l and Mor-
gan or representing themselves well but being
considercd obscure or bad writers. Thus, the
effect was to normalize the emerging research
paradigms favoring rather traditional direc-
tions even within them. For example, when
Burrell and Morgan, and subsequently
Putnam and others, provided "interpretive"
work with the "subjective" ascription (even if
now positively valued) they, perhaps unwit-
tingly, tended to favor cultural studies that fo-
cused on member's meanings that were more
subject to cultural management and manage-
rial control. At the same time the "objective"
ascription protected "functionalist" studies
from a thorough analysis of their hidden val-
es and sources of subjectivity, as if they
might be too objectivea preferred flaw-
rather than too subjectivea flaw they would
not understand. Similarly, the many critical
theorists with strong suspicions of humanist
philosophies suddenly found themselves ei-
ther conceptualized as radical humanists or in-
visible (lost in some hole in paradigmatic
space). The Frankfurt school's attack on the
subjective domination in science all too often
got lost in the radical humanist conception.
My point is not that Burrell and Morgan and
their followers were representationally wrong
in the presentation of organization and organi-
zational communication studies (for (here are
many representationally "right" schemes and
surely the nearly 20 years since their work has
led to many changes), but their conceptions
continu to foster less interesting and produc-
tive conflicts and developments than are pos-
sible. The processes of differentiaon in main-
stream functionalist sociology must be aban-
doned before more challenging differentia-
tions are possible and alternative research pro-
grams can be given a full complementary role.
By focusing on the constitutive moves of
discourse in organizational research and orga-
nizational praclce rather than in psychologi-
cal, sociological, or economic theories of or-
ganizational behavior, more interesting
differences can be displayed. In my develop-
ment below, I will privilege programmatic dif-
ferentiations rooted in what I will develop as a
"dialogic" perspective. What Burrell and
Morgan called "functionalist" research will
thus be implicitly represented as an "other." In
doing so, both the lines of divisin and the ar-
guments tbat extend from this can be redrawn.
"Functionalist" style work can be reclaimed
as legitmate and useful (though neither cu-
mulative or "trae") in specifiable ways as
reunderstood from dialogic conceptions.
Nondialogic research programs will not be
seen as alt ernat i ve routes to truth, but as spe-
cific discourses that specify and provide an-
swers to specific types of problems. By setting
Conceptual Foundations 4 11
Re lat o n to Dominant
Soci al Discourse
Dialogic studies
Interpretive studies
Premodern, traditional
Critical studies
Late modem.
Elite/A Priori
Norrnaiive studies
Modern, profraiive
Fi gure 1.1. Contrasting Di mcnskms From the Metalhcory of Representa!mul l'rucliccs
SOU RCE: Adapted from Du (1994d).
aside typical research claims of universality
and/or certainty, different research traditions
can provide productively complementary and
conflictual insights i nt o organizational life.
The test of my suggested differentiations is
not whether they provide a better map, but
whether they provide an interesting (or what
Rorty, 1989, developed as "edifying") way to
talk about what is happening in research pro-
A more contemporary look at alternative
communication research programs can be
gained by locating research differences in
what was conceptualized earlier as "dis-
courses"that is, the linguistic systems of
dstinction, the vales enacted in those dis-
tinctions, the orientations to conflict and re-
lations to other groups. Two dimensions of
contrast will be developed here. Later in the
essay, four prototypical discourses or re-
search approachesnormative, interpretive,
critical, and dialogicwill be dcvdopcil
from theseconceplions. Ser l-'i gure 1. 1.
First, (JiffcrcncvN anxi i i ] ', H - MMI I - I I n r i c n l . i
tions can be shown by conlrasli nn "locul/
emergen!" research conceplions wi l h "cliti'/n
priori" ones. This di mensi n focuses on the
origin of concepts and problem statemcnts as
part of the constitutive process in research.
Second, research orienlations can be con-
trasted in the extent to which they work wi t hi n
a dominant set of structurings of knowledge,
social relaons, and identities (a reproductive
practice), called here "consensus" discourse,
and the extent to which they work to disrupt
these structurings (a productive practice),
called here "dissensus" discourse. This di-
mensin focuses on the relation of research
practices to the dominant social discourses
within the organization studied, the research
community, and/or wider community. I see
these dimensions as analytic ideal types in
Weber's sense mapping out two distinct con-
tinua. While categories of research programs
are derivatively produced by the dimensions,
the intent here is to aid attention to meaning-
ful differences and similarities among differ-
ent research adi vi nes rather than classifica-
12 f I heoretlcal and Methodological Issues
TABLE 1 .1 Characterizations of the Local/Emergent-EHte/A Priori Dimensin
Comparativa communities
Mltiple language games
S/stematic philosophy as ethnocentric
Situalionallj' or structural determinisir
Local narrativas
Sensuality and mcaning as central concern:
Situated, practica! knowledge
Tends to be feminine in attitude
Sees the strange
Proceeds from the other
Ontology of "otherness" over method
O le/A Prior/
Privileged communiry
Fixed language game
Groundsd in hoped for systematic philosophy
Tlieory driven
Methodobgical determinism
Grand narrativo of progress and emancipation
Rationality and truth as centra! concerns
General i labie, the ore ticai knowledge
Tends to be masculine in attitude
Sees the familiar
Proceeds (rom the self
Epistemolgica! and procedural issues rule over
substantive assumptioni
The LocaUEmergent-
Elite/A Priori Dimensin
The key questions t h i s di me ns i n ad-
dresses are, where and how do research con-
cepls arise, and t h us , implicitly whose con-
ceptions are used? In the two extremes, either
concepts are developed in relation wi t h orga-
nizational niembers and uansforraed in the re-
search process or they are brought to the re-
search "nteraction" by the researcher and
held static through the research processcon-
cepts can be developed with or applied to the
organi zat i onal members and activities being
studied. This dimensin can be characterized
by a set of paired conceptions that flesh out
contrasts embedded in the two poles, Table
1.1 presents an array of these contrasts. The
choice of and s t a bi y of the language system
are of central importance since the li ngui s-
tic/conceptuat system directs the statement of
problems, the observational process itself in
producing objects and highlighling and hiding
potential experiences, the type of clai ms
madc, the report to external groups, and he
li ke ly generalizations (whether appropriate or
not ) readers wi ll make.
The local/emergent pole draws attention to
researchers who work wi t h an open language
system and produce a form of knowledge
characterized more by i nsi ght into emprica!
events Ihan large-scale empirical generaliza-
t i o n s . Ce nt ra l t o t h e i r work i s t he s i t ua t e d
nat ure of the research e nt e rpri se . Problem
statements, the researcher's attention, and de-
scri p ons are worked out as a play between
communities. The theoretical vocabulary car-
red into the research activity is often consid-
ered by the researcher as s e ns i t i z i ng or a g ui d e
to getting started constantly open to new
m c a n i n g s , translations, and redi fferent i at i on
based on interactions in the research process.
Produced insights into organization processes
ma y be particularistic regarding both time and
place even though the emerging a na ly t i c
f rame is designed to aid in the deeper unde r-
st a ndi ng of other particular settings. Cumula -
tive understanding happens in providing sto-
ries or accounts that may provide insighi i nt o
other sites rather than cumulative universal as-
p i r i n g c l a i m s . T h e r c s c u r d i u l t c i u l ' . i i > i h r l u
ings, i n t u i li o n s , a n d m l t i pl e O M I I S u t ' > "
n a li t y of both t he rcsearclicd mil i c w a u l i n
rather t h a n u s i n g a s i n g l e lo g i c i ) f oh-
jectifica on or puri f i e d r a li o n a li ly . The s t u d y
s gui ded more by concept f orma t i on t r i a n
concept appli cat i on. Distan ation and t h e
"otherness" of the other (the way people and
events exceed categories and classifications of
t h e mj are sought by the researcher to forc
reconception and li n g u i s t i c change. This is
considered more valuable than the identifica-
tion and nami ng of preconceived traits, attrib-
utes, or groupings. O bjectivity, to the extent
that it is considered at all, arises out of the in-
terplay and the constan! a bi t i t y of the re-
searched to object and correct. The researcher
is more a skilled collaborator in knowledge
product i on t h a n an expert observen
The elite/a priori pole draws attention to
the tendency in some types of research pro-
grams to privilege the part i cular language sys-
tem of the researcher and the expertise of the
research community as well as hold t h a t lan-
guage system constan! t h r o u g h o u t he re-
search process. S uch research t e n d s . t o be
heavily theory driven w i t h careful attention to
definitions prior to the research process. The
experiences of the researched become coded
i nt o the researcher's language syslem. De-
ma nds of consistency and/or reliability re-
qui re changes in the concept i onal system to
take place outside of rat her t h a n in he re-
search process.
Whether intentional or not , the conceptual
system of the researcher is considered better
or more clearly represents wh a t "reatly" is the
case than that of everyday people and seeks
generality beyond the various local systems of
meanmg. In privileging a language system,
|here is f urt her a tendency to universalize and
jus t i f y such moves by appeals to f ounda t i ons
or essentialist assumptions. Research claims,
thus, are seen as freed f rom their local and
temporal condilions of product i on. In most
cases, these research approaches folio* an en-
llghtenment hope for pro d u ci n g rat i onal
knuwledge not constrained by tradition or par-
Lonceptual r-ourxjutlorn t IJ
I n i i l . i i l - r l i r l ' . y - lni i t i l l l u i i M - I I I I I i
< I I I | I I L | 'llir | iini1 l l i ni ul i .l p i l n.i i i .1
. i ' . i ' i M j - i , / i I .1 i n , u n r | i l l i i I ' I
m i ' ( n u n M M ' . r i l > , i |i . n M u " , u i w r l ! I " n i ) - 1 I n
mor e " i i o i i i i i i l i v r " v c i - . n n i - . < i ] - n l y J I K H I t i i i n
" objccti v i ly" nu i l v nl m* n n i l i a l i l y l i a s n l i i n l i l i -
sharal l: in} ',ii; i(',c (?.; unr m u i L ' s n u ch mcl l u x l s ,
and tend l o ov cr l ook t l i c pos i li ons n f i h c i r own
commu ni ty or al l i an ce s wi t h other groups .
The mor e " cr i li cal" versi ons q u i ck l y note the
pr esence of v ales and di s tor ti ons i n nor ma-
ti v e work, and h old out the hope for a better ,
pu r er form of knowledge based i n pr ocesses
that i ncl u de more interests and means of anal-
ys i s i n the wor k.
Focusi ng on the origin of concepts and
problems u s i n g a dimensin of "local/emer-
gent-elitefa priori" allows three important
gains. First, it acknowledges linguistic/social
const ruct i oni sm n all research posi t i ons and
directs attention to whose concepts are used in
object production and determinaton of wh a t
is problematic (see Deetz, 1973) . S econd, the
focus on the ori gi n of concepts helps d i s t i n -
g u i s h f u n d a me n t a lly different k i n d s of knowl-
edge. Elite/a priori conceptions lead more to
the development of "theoretical codified"
knowledge, a kind of "book" knowledge or
"k nowi ng about." Local/emergent concep-
tions lead more to the development of "practi-
cal" knowledge, a k i n d of "street wsdom" or
a "knowing how." Third, this dimensin helps
us remember t h a t both the application and dis-
covery of concepts can demnstrate i mpli ci t
or explicit political alliances wi t h di f f e re nt
groups in the organization or larger society.
For example, to the extent that organizational
researchers' concepts a li gn with ma na ge ri a l
conceptions and problem statements and are
applied a priori in studies, the k n o w le d g e
cla i ms are i n t r i n s i ca l l y biased toward these
interests as they are applied w i t h i n the site
co mmu n i t y ( M umby, 1988) . The knowledge
claims become par of the same processes t h a t
are being studied, reproduci ng w o rld v i e w s
and personal identities and fostering particu-
lar interests w i t h i n the org a ni z a t i on {sec
K ni gh t s , 1992) .
U * Theoretical and Methodological Issues
TABLE 1.2 Characterizations of the Consensus-Dissensus Dimensin
Hegemonic order ai natural state
Naturalizaron of present
Integration and hanuony are possible
Research focus es on representation
Mirror (reflecting) dominant metaphor
Validity central concern
Theory as abstraction
U ni fie d icience and triangulation
Science is neutral
Life is discovery
Researcher anonymous and out of time and space
Autonomous/free agent
Sus pidn
Conflicts over order as natural state
Present order is hstoricized and politicized
Order indicates domination and suppressed
Research focuses on challenge and
reconsideration (representation)
Lens (seeing/reading as) dominant metaphor
Insight and praxis central concern
Theory as way of seeing
Positional complementan'!/
Science is poltica!
Life i s struggle and creation
Researcher named and positioned
Historically/socialfy siiuated agent
The Consensus-
Dissensus Dimensin
The "consetisus-dissensus" dimensin
draws attention to ihe relation of research to
existing social orders. Consensos or dissensus
should nol be understood as agreement and
disagreement bul rather as presentaron of
uni ly or of difference, the continuation or dis-
ruplion of any prevailing discourse. See Table
1.2 for conceptual i zat i on of this dimensin.
This dimensin is similar to Burrell and Mor-
gan's use of the tradtional sociolgica! dis-
tinction between an interesl in "change" and
"regulation," but enables some advantages.
Rather than being class based, contemporary
conccrns wih conflict and power focus on the
ways predominan! discourses (though often
disorganized and disjunct) place limitations
on people in general i ncludi ng managers and
mi t the successfu! funct i oni ng of organiza-
tions in meet i ng human needs. The focus is
more on ihe supprcssion of diverse vales and
Ihe presence of destructve control processes
than on conflici among groups. The processes
of domi nat i on loday are less often seen as
macrosociological and more often seen as
arising n normative or unoblrusive controls
(seeBarker, 1993;Etzioni, 1961;TompkJns&
Cheney, 1985) and instantiated as routine mi-
cropractces in the work site itself (Ashcraft &
Pacanowsky, 1996; Deetz, 1994b, 1994c,
1998; Knights & Willmott, 1989). The focus
on discursive rather t han group relations aids
the understanding of domination and the vari-
ous ways importan! organizational stake-
holders are left out of discussions as well as
the ways such forros of decisional skewing are
The consensus pole draws attenlion to the
way some research programs both seck order
and treat order production as the dominan!
feature of natural and social systems. Wilh
such a conception, the primary goal of the re-
search is to display a discovered order wth a
high degree of fidelity or veri simil ilude. The
descriptions hope to "mirror" entities and re-
lations that exist out there in a relatively fixed
state reflecting their "real" character. In the
"normative" versin this realty is treated like
the natural world while in "interpretive" work
t is a social wortd. Language is treated as a
system of representations, to be neutralizad
and made transparent, used only to display the
presumed shared wor!d. Existing orders are
largely treated as natural and unproblematic.
To a large extent through the highiighting of
ordering principies, such orders are perpetu-
ated. Random events and deviance are down-
played in signifcance in looking at norms and
the normal, and attention is usuaily to pro-
cesses reducing deviance, uncertainty, and
dissonance. In most cases where deviance is
itself of attention, it tends to be normalized
through looking at the production of deviant
groups (i.e., other orders). Conflict and frag-
mentation are usuaily treated as system prob-
lems and attention is given to how orders deal
with them in attempts at maintenance.
The dissensus pole draws attention to re-
search programs that consider struggle, con-
flict, and tensions to be the natural state. Re-
search itself is seen as inevitably a move in a
conflicrual site. The existing orders indcate
the suppression of basic conflicts and along
with that the domination of people and their
full variety of interests. Research aims at chal-
lenging mechanisms of order maintenance to
reclaim conflicts and tensin. The nonnor-
mative aspects of human conduct and extraor-
dinary responses are emphasized along with
the importance of largely random and chance
events. Rather than language naming and de-
scribing, researcher conceptions are seen as
striking a difference, de- and redifferentiating
experience (Cooper, 1989; Cooper & Burrell,
1988; Deetz, 1992; Martin, 1990; Weedon,
1987). The "mirror" gives way to the "lens" as
the dominant metaphor for language and the-
ory noting the shifting analytic attempt to see
what could not be seen before and showing
the researcher as positioned and active (Deetz,
1992, chap. 3; Rorty, 1979). For dissensus
style research, Ihe generative capacity (the
ability to challenge guiding assumptions, val-
es, social practices, and routines) of an ob-
servation is more importan! than representa-
tional validity (see Gergen, 1978). The
research is, in Knights's (1992) sense,
"antipositive." Dissensus work does not deny
Concepta/ Foundations * 15
the signifcance of an ordered observed world,
rather it takes it as a powerful (power filled)
product and works to break reifications and
objectifications to show fuller potenlial and
variety than is mmediately apparent. For ex-
ample, consensus orientations in cultura)
studies seek to discover the organizational
culture or cultures. Dissensus orientations
show the fragmenlation inherent in any claim
of culture and the work required for site sub-
jects to maintain coherence in the face of this
as well as subjects' own forras of resistance
(see Calas & Smircich, 1991; Holmer-
Nadesan, 1996; Martin, 1990,1992; Smircich
& Calas, 1987; Trethewey, 1997). Consensus
orientations apply role and identity classifica-
tions and relate them to olher variables;
dissensus orienlalions see identity as mltiple,
confiictual, and in process.
While diese differences can be character-
ized clearly in abstraction, in continuous time
every consensus arises out of and falls to
dissensus, and every dissensus gives away to
emerging (if temporary) consensus. The issue
is not the ultmate outcome desired or likely
but rather which part of this flow through time
is claimed in the research process. For exam-
ple, while critica! theorists clearly seek a so-
cial consensus that is more rational, their re-
search tries to produce this through the
creation of dissensus in place of dominant or-
ders. For example, ideolgica! critique in the
critical theory conception of the negative dia-
lectic is to reclaim conflict and destroy a false
order rather than produce a new one. Thus, I
place them on the dissensus end. Critica! theo-
ries differ from many dialogic or "postmod-
ern" positions in the production of dissensus.
In critical theories, dissensus is produced by
the use of lite understandings and procedures
(as in Habermas, 984, 1987; Kunda, 1992;
Mumby, 1987; or severa! essays in Alvesson
& Willmott, 1992). While in dialogic re-
search, deconstnictive processes are used to
unmask lite conceptions thereby allowing or-
ganizational aclivities to be given new, mlti-
ple, and conflicting descriptions (Calas &
Smircich, 1991; Kilduff, 1993; Laclau &
16 $ Theoretical and Methodological Issues
M o u f f e , 1985;Linste ad, 1993; M artin, 1990).
The di al o gi c o u t c o me re qu ire s a co nstan!
de di f f e r e nt i at i o n and r e di f f e r e nt i at i o n f o r t he
sake o f de myl ho l o gi zi ng and e nriching nat u -
ral langu age and co nse qu e ntly o pe ning to re -
co nside ratio n the mo st basic and ce rtain expe-
rie nce s o f e ve ryday wo rk lif e .
The grid pro du ce d f ro m the se two d i m e n -
sio ns pro vide s a spat i al l y and vi su al l y co nve -
n i e n t , di s c u r s i ve f o u r-space so l u l i o n (he ne e
we sho u l d always be e asil y re mi nde d o f its
ar bi l r ar y and f i c t i ve characte r). I w i l l de -
scribe the se as di f f e r e nt disco u rse s to no te a
way o f articu l ating argu me nts and e ngaging
in research practice s rathe r than a me ans o f
re co nstru ctive s e l f - nami ng . Each di sco u rse
pro vi de s an o r i e nt at i o n to o rgani zado ns, a
way o f c o ns t i t u t i ng pe o ple and e ve nts in
t h e m , and a way o f re po rting o n t he m. I ho pe
t hat t hi s also le ads u s to t h i n k abo u t whi c h
disco u rse is be i ng u se d o r ho w it s j o i ne d
w i t h o the rs rathe r than pi ge o nho t i ng spe cif ic
au t ho rs. Table 1.3 pro vide s ske t chy proto-
t ypi c al de scriptio ns o f each research o rie nta-
tio n related to a do ze n dime nsio ns o f inte re st
s h a p i n g o r g a ni z a t i o na l c o mmu ni c at i o n r e -
se arch pro grams.
Cal l ing these disco u rse s "paradigms"
wo u l d be a mistake f or several re aso ns. First,
each of the se f o u r discourses, whi c h are provi-
s i o nal l y he ld apart f o r vie wing, is f ille d with
i nt e r nal co nf lict and st r i f e i ncl u di ng the o ry
de bate s, mo me nts o f i nco mme nsu r abi l i t y, di l -
e t t anl e s, and t yr ant s. Se co nd, the e dge s are
no t de marcate d. M o sl researchers and teach-
e rs do no t cl u ste r aro u nd a pro to type o f each,
bu t gathe r at the crossroads, mix me tapho rs,
and borrow line s f ro m o the r discourses, do dg-
ing criticism by co-optation. Of te n praclicing
re se arche rs happily move f ro m o ne disco u rse
to ano the r wi t ho u t ac c o u nt i ng f o r t he i r own
localiot:. The y oprate like o the r o rganiza-
l i o nal me mbe rs bo rro wing o n disco u rse s that
su it the ir i mme di at e pu rpo se s and the f ashio ns
of the mo me nt (see Deetz, 1994b), There are
ce rtainl y mo re and less serious plays across
the line s, bu t the issu e is no t cro ssing bu t the
se riousne ss o f the play. Third, the discourses
are not the mse lve s sealed off f ro m each other.
They pose pro ble ms for each o the r and ste al
insights across the line s. Fo r e xample , the
philo so phical f ights between Habermas and
Gadamer, Habe rmas and Lyo tard, Habe rmas
and Lu hmann, and Fo u cau lt and e ve rybo dy
have le f t t he i r traces in each one's wo rk. From
[hese stru ggle s, the vario u s o rganizatio nal
c o mmu ni c at i o n research pro grams based in
these works have gaine d e nriche d conceptions
of power, kno wle dge , agency, and po litical ac-
t i o n (see, e.g., M u mby & Pu t nam, 1992).
Pro visio nal o rde ring o f discourses is no t to
plice the lines, but to pro vide a view of the
social resources f ro m which researchers draw
and an u nde rstanding o f the stock argu me nts
use d in de ve lo ping and j u s t i f y i n g research ac-
tivitie s and claims. The ideal type s aid the u n-
de rstanding o f dif f e re nce s t ha t matte r t hat are
hard to se e in the f l o w o f research activity.
Cl arif ying the te nde ncie s in specific type s o f
research po sitio ns he l ps c l ar i f y debates and
the re latio n o f di f f e r e nt gro u ps to the m. Fo r
e xample , the inte rpre tive , cr i t i cal , and dia-
lo gic critiqu e s of no rmative research are qu ite
dif f e re nt. No rmative researchers who are ac-
cu sto me d to maki ng argu me nts against su b-
j e ct i vi t y and traditio nal ism simpl y miss the
po int of each of these critiqu e s; t he y o f te n re-
duce the m to abstract and co nf u se d pre se nta-
tio ns o f what the y t h i n k "opponents" sho u l d
be saying rathe r t han concrete bu t di f f e r e nt ar-
gu me nts f ro m what the y e xpe cte d.
Fu rthe r, while mo st researchers are no t
pu rists, the ir wo rk carnes assu mptio ns and re -
spo nsi bi l i t i e s t hat are ce ntral to u nde r st andi ng
and e valu ating t he i r wo rk, bu t are rarely e x-
plicit in slu dy reports. Fo r e xample , many
f e mi ni st s' writings carry a ge ne ral sympat hy
wi t h the co nce ptu al and anal ytic po we r o f
dialogic re se arch programs, whi l e the y still
wish to have a political age nda t hat re qu ire s
critica! pre co nce ptio ns that assume social di-
C o n c f y t u a l fbundot/ons i 1 7
TABLE 1 .3 Prototypical Discursive Features
Basle goal
Metaphor of
social relaticns
Lawllke relations
among objects
Nomothetlc sclence
unlfled culluie
Recovery o
Integrarle vales
Cultuial criticism,
Ideology critique
Refoimatlon of
social ordei
Reclaim conflict
Claim a space
tor lost volees
Concern with
Narrativo style
Time Identlty
Scclal fear
Marketplace Communlty Pollty
Inetflctency, dlsorder Meanlnglessness, Domlnallon,
lllegltlmacy consent
Fidellty, Influence,
Information needs
Control, expertise
Social Mlsrecognltlon,
acculturatlon, svstematlc
group afflrmallon dtslortlon
Late modem
Commtlment. Portlclpatlon.
qualitv worlc Ufe expended
Friendly Susplclous
Depersonalizallon Authorlty
visions and gender-based do minatio n to be
general (see Rax, 1990; Fraser & Ni cho l so n,
1988; M u mby, 1996). Such works (e.g., Mar-
tin, 1990,1992) can be classified as dialogic,
but the tnica! and political character of many
of these studies cannot be ju stif ie d easily with
dialogic co nce ptio ns alone . The distinctio ns
developed in this essay can he lp display the
te nsio ns and the resources from which such
researchers draw to conduct and ju stif y their
This can f u rthe r be sho wn u si ng my o wn
wo rk as an example. I o f te n draw on concep-
ons f ro m critical and dialo gic wrilings. For
me, critical theory co nce ptio ns of ideology
and disto rte d co mmu nicatio n provide u se f u l
sensitizing concepts and an analytic frame-
work for looking for micropractices of u nwar-
ranted control, discu rsive clo su re , ide ology,
and skewed representaron in organizational
sites. But rarely are these conceptions closely
tied to the f u ll critical theory agenda. They re-
18 * Theoretlcal and Methodological Issues
qu i re cons i derable rcworki ng i n s pecific sites,
and t hc res u l l s of my s tu di es ai m more at fi nd-
i ng and g i v i ng suppressed pos i ti ons a means
of cxpres s i on i ban reali zi ng an ideal speech
s i l u al i on or reachi ng a purer consensos (see
Dectz, 1995b, 1998). What i s i mportanti s not
wticihcr I am a late-modern critical theorist or
u di al og i c postmodemist, bu t rather the mean-
ing and i mpli cati ons of concepts that I draw
from thes c two competitive research orienta-
ti ons . My degrce of consistency is of less in-
lerest than how I handle the tens i n and
whct her the lwo conceptual resources provide
an i ntercs li ng analys i s or i nterventi on. Some
cl ari l y and general u nders tandi ng i n alterna-
li ve research ori entati ons provi de gu i dance
and accounlability or at least a common stock
of materi al for bu i ldi ng and evalu ati ng new ar-
gu ments in these cases. Furher, exploring
general ori entati ons can hel p reveal assump-
li ons hi dden i n one's own way of worki ng
since they remai n unproblematic in one's own
research commu ni ty.
I n an ideal research program, we mi g ht
identify a complementary relation among re-
search ori entati ons wth each asking di fferent
qu es ti ons at di fferent moments and each, at
thc appropriate moment, ans weri ng to the spe-
ci fi c criteria of a particular ori entati on. This
mi g hl oprate in a rolation among incompati-
ble ori cnlati ons wi t hou t any orientation being
privileged or any orientation being reduced to
a prcli mi nary or supplementary role. For ex-
ample, my work relies mu ch on a conception
of di s cu rs i ve closure, a conception that draws
altcnli on to places where cooperative deci s i n
r na k i ng is hampered by arbitrary li mi ts en-
acted in the discussion (see Deetz, 1992, pp.
1 8 7ff.). As a critical researcher I mus t show
how ihese closures are i ntru s i ons of power re-
lalons u s u al l y based in or supporting social
di vi s i ons that lead to distorted communication
and a false consensus. My s tu dy appeals to
reason, lgica! analyses, and a coherent dem-
ons trati on. As a diatogic researcier 1 see these
closures as the suppression of conflicts and
see my own concerns wi th consensus and ap-
pcals to reas on as s i mpl y di fferent acts of
privilege and polential closure. My analys i s is
now judged by the way indeterminacy is al-
lowed to reemerge and the compelling qu ali ly
of recovered claims and voices. But at another
moment yet, I may well pose normative qu es -
tions: Which means of closure are used most
often? Who uses them? When are they used?
Can peopie be tau ght to avoid them? A s tu dy
des i gned to answer such questions now ap-
peals to standards of defi ni ti on, measurement,
sampling, and qu anti tati ve data analys is . And
further yet, there are interprelive concerns:
What sense do these discursive moves have in
a commu ni ty? To what ends are they us ed?
How are they s elf-unders tood and ju s ti fi ed?
What are their actual consequences in specific
circumstances? Interpretive research s tan-
dards are now relevan!.
One can easily see how such a rotation
throu gh ori entati ons mi ght be constan! and
productive wi t hou t los ing the separaion and
tensin among them. Such tens ions could help
enrich work from each orientation. Yet, to be
hones t, few research programs are treated this
way and most researchers, like myself, follow
their own lines of inters!, commi tments , and
trai ni ng, whi ch either leads to an eclipse of
qu es li ons and concerns from other orienta-
ti ons or at least leaves them for someone else
who is interested in those problems, Taking
seriously other works does not mean that we
fmd other groups' is s ues and procedures as
necessarily interesting or helpful or should
we naively believe that all of them are. But our
claims and the relation between our clai ms
and s tudy procedures s hould be clear so that
objections and confli cts can be on those
grounds rather than on imposed traditional
problem statements and methods. The point is
for the researcher to be clear about what type
of qu es ti ons or claims drives the work at any
particular ti me and how the work addresses
the standards and criteria appropriate to it.
A basic u nders tandi ng of alternative re-
search orientations enables shorthand ac-
counts and helps dis tinguis h i ntenti onal
and/or productive ambiguities from careless
and/or u nprodu cti ve ones. As a reviewer, I am
Conceptu/ F o u n d a t i o n s * 1 9
often frusirated by nonreflective mi xi ng of
metaphors and conceptions in submitted es-
says. Often the claims made would require a
di fferent kind of study based on different as-
s umptions and research activities. Partly, I
thi nk this arises from authors trying to antic-
pate reviewer needs for normative type gener-
alizations while being committed to a nonnor-
mative research orientaton, but it also comes
from inattention to what makes different kinds
of research different. Clearly, a balance must
be struck between (1) reitying research orien-
tations through simplistic grids and subse-
qu ent overcharacterizations and rigid stan-
dards and (2) havi ng each study try to be
totally self-justifying and cut lose from any
community. While I do not think there is any
easy way out of this tensin, having good di-
mens i ons of contras! and good character-
izations helps. A very brief sketch of the four
orientations aids further in highlighting differ-
ences and similarilies in these community dis-
courses along the suggested dimensions of
The Di s co u rs e of
Normative St u d i es ^
Normative research tends to accept organi-
zatons as natu rally existing objects open to
description, prediction, and control. Goals es-
tablished by some specific group, us ually up-
per management, are largely aceepted as the
goals of the organization and most often the
research ei ther i mpci tly or explicitly sup-
ports more efficent accomplishment of Ihese
goals. Because of this , commercial corpora-
tions are u s u ally discussed in economic terms
with issues discussed in relation lo "rational"
economic goals' The researchers producing
this discourse have been described as
functionalists, covering-law theorists, or sim-
ply practicing the variable analytic tradition. I
describe this discourse as "normative" to em-
phasize the centrality of codification, the
search for regularity and normalization, and
the implied prescriptive claims (see Deetz,
1973; Hollway, 1984). This discourse is
largely donnant in North America and in ap-
plied organizational research everywhere.
Anieles published by U.S. researchers em-
ployed by communication departments and
published in "communication" journals have
been mostly of this sort, though the mi x is
changing. Mosl textbooks are written in this
discourse emphasizing topical divisions and
research findings even when they review re-
search established in other tradi'ons.
The discourse is decisively modern in
Gergen's (1992) sense and the knowledge is
considered posive, cumulative, and progres-
sive. A grand narrative of progressive emanci-
pati on from disease, disorder, and material de-
privaon is shaped by a commitment to make
a better world through discovery of funda-
mental processes and increased producton
(Lyotard, 1984). While the organizaron is
usually treated as an existing object produced
for instrumental ends, us ually making money,
some conception of the invisible hand makes
that goal well-integrated with other social
goals of development and widespread avail-
ability of goods and services. Generally, the
research is expressly apolitical and valu neu-
tral, but as already shown, vales reside in
lite conceptions, choice of problems to s tu dy,
and relation to other groups.
Most of this work has implicitly supported
an orderly, well-integrated world, wi th com-
pliant members and regulated confli cts , and
has aceepted without examination existing or-
ganizational goals and member pos i ti ons .
They represen! communication primarily in
information and administration terms (see
Beniger, 1986). Much of the discussion of
communication in lli nformaton" terms as-
sumes a control orientation and theories of
persuasin and information transfer domnate
much of the concern with most frequently
studied topics such as supervisor/subordnate
communication, compliance gaining, net-
works, power, and relations wi th the public.
Normative works appear in three basic vari-
eties each wid distinct assumptions and goals
of their owncovering iaws, systems theory,
and skill developmenL
20 * Theoretfcal and Methodological Issues
Covering Laws
Re s e a r c h modeled on the search f or
l a w l i k e gener ali zat i ons i n organizations u n t i l
f a i r l y recenlly has dominated organizational
c ommu n i c a l i on s t u dy . Research of this ty pe
was most e xpl i c i t l y def ended i n commu ni ca-
t i on s t u di e s by Berger (1977) and recon-
s t r u c t c d and well j u s t i f i e d i n Donaldson
(1985; see also Bar ley & Ku nda, 1992;
Di Ma ggi o, 1995; O' Keef e, 1976). The re-
search practices mirror 19th-centu ry concep-
L i ons of ihe n a t u r a l sciences of t e n i n vol vi n g
i he most recent a dva nc e s i n oper at i onal-
i z a t i on , hy pot hes i zat i on, statistical data re-
du c t i on , and pattern "recognition" processes.
Conceptions of operationalization, "objectiv-
ity ," and l a w l i k e relati ons are mer ely the most
obvi ou s f orm of practice. Conventi onal prac-
tices and methodological ("as i f ' ) determi n-
ism h a ve in mosl cases replaced any s t r ong al-
legiance to the posi t i vi st phi i os ophy of
sci ence I h a t gr ou n ds ma n y of the methods and
assu mpt i ons.
The "objects" constru cted by the practices
of t h i s s c i e nc e are given qu ati es of c ons t a nc y
and pe r ma ne nc e (u niversal across time and
place), as if na t u r e endowed them wi t h spe-
c i f i c a t t r i bu t e s . The c ombi n a t i on of a priori
c on c e pt i on s and the focu s on c ons e ns u s leads
ihe ar t i f acls of these research practices to be
dcscri bed as f acts. This discou rse t y pi f i e s the
dc ve l opme n t of ma n y data retri eval sy stems
a nd i nf or ma li on t echnologi es si nce i nf or ma -
t i on can be treated as fixed tru th clai ms freed
f r om the t i me , place, and procedu res of pro-
du c l i on. Facts become commodities and com-
mu n i c a i i o n can be redu ced to a t r ansmi ssi on/
r elr i eval process (f or di scu ssi on of conse-
qu e nc e s , see Boland, 1987; Coombs, Kn i gh t s ,
& W i l l mo t t , 1992; L y y t i n e n & Hirschheim,
Theory and theory t es t ng are central (o the
logi c of me research and ma n y of the st at i st i -
cal procedu res of da t a r e du c t i on. N or ma t i ve
s t u di e s of i h i s t y pe are explicitly dependent on
i hcory , ( h o u g h i n practi ce t he theoretical con-
ccr ns may he redu ced to a mere ref erence li st
of pr i or s t u di e s and theory t e s t i ng to merely
adding to a li st of relations a mon g variables of
i nterest. One characteristic self -cr i t i ci sm is
the lament over the la c k of development or u se
of theory . Most of the s t u di es work as i /they
were in a dedu cve theory t e s t i n g mode even
when their theoretical commitments are less
t ha n clear. Recently , Su tton and Staw (1995)
demonstrated how references, data, variables,
diagrams, and hy potheses are of t e n u sed to
cover u p the lack of theory and act u al theory
This discou rse is exempli f i ed in s t u di es of
compliance g a i n i n g (e,g., Su lli va n & Tay lor,
1991), strategic message design and persu a-
sin (e.g., Alexander , Penley , & J e r n g a n ,
1991), su per vi si n/ su bor dnat e i n t e r a c t i on
(I nf ant e, Ander s on, Mar t i n, He r i n gt on , &
Kim, 1993; J abli n, 1979; Sias & J a bli n, 1995),
and other places more c omple t e ly described
by Bu rrell and Morgan (1979) in thei r di scu s-
sion of "fu nctionalist." Bu l it is also clearly
present in those advocati ng the ma n a g e me n t
of c u lt u r e (e.g., Deal & Kennedy , 1982;
Schein, 1992) t h r ou gh their conception of cu l-
tu re as an object to be st r at egi cally deploy ed
(as Barley , Mey er, & Gash, 1988, have s h ow n ,
t h i s became very common in the 1980s). Most
of the work on cu lt u r e, climate, or varieties of
total qu a l i t y ma na ge me nt (TQM) i n organiza-
t i ona l c ommu n i c a t i on ar e more n o r ma t i v e
t ha n interpretive owi ng t o t he way c u l t u r e i s
treated as a variable or obj ecti ve ou tcome
w i t h i n a larger strategic move of c u lt u r a l ma n -
agement (see Shockley -Zalabak & Mor le y ,
1994). Many of t hose wor k i ng wt h new con-
c e pt i ons of organi zati ons as "postmodern"
(rather t ha n post-modern approaches; Parker,
1992) have a discou rse pr i ma r i ly s t r u c t u r e d in
a normative f a s hon (e.g., Ber gqu i s t , 1993;
Peters, 1987). Many Ma r xi s t stu di es, espe-
c i a l l y those done i n c ont e xt s of Mar xi s t domi -
na t i on of social discou rse, u se normati ve
themes, bu t the li te grou p t hat gives rise to
t he concepts di f f e r s f r om those s u ppor t i ng
most Eu ropean and North Ame r i c a n s t u di e s .
L enin's embraci ng of s c i e n t i f i c ma na ge me nt
was in no way i n c on s i s t e n t . Strategic manage-
me nt i n vi r t u a lly every way i s h i g h l y depend-
en! on t hi s discou rse (Kni ght s , 1992; Kn i gh t s
& Morgan, 1991). Of ten team, qu a li t y , and
par t i ci pat i on programs are assessed u sng re-
search procedu res grou nded in this perspec-
tive (e.g., Cordn, I nf ante, & Graham, 1988;
Miller & Monge, 1985).
St u dy i ng c ommu n i c a t i on i n the organiza-
ti onal context poses some u n i q u e problems
for this approach. The complexity and inter-
dependence of organizational relationships
challenge the rather atomsc and u nidirec-
ti onal models of both the theories and meth-
ods. And su ch relations are hard to du plcate
in laboratory settings and control f or nu mer -
ou s "extraneou s" f actors. Researchers ha ve
responded to t hi s with mu c h more sophisti-
cated modeli ng and statistical analy sis. Un-
f or t u nat ely , the ou tcomes of this are fairly ab-
stract relations that lead to qu est i ons of
validity and u sef u lness. Fu rther, mu ch of the
research has tu rned to data collection f rom
self-report interviews and su rvey qu estion-
naircs r at her t han di r e c t observation (see
Knapp, Pu tnam, & Davi s, 1988). This has led
to a preoccu pation wi th measu rement devices
and wi t h many stu di es that are more i nstru -
ment than theory dri ven leaving the resu lts
di f f i cu lt to u nderstand or u se in any sy stem-
atic way .
Fi nally , most researchers condu ct su ch
stu dies primarily f or generalization and u se
statistical s i gni f i cance tests, whi ch alow gen-
eralization f rom the research sample to some
popu lation. Bu t the qu estion is often raised as
to what is the appropriate "popu laton" for the
generalization. Many of the stu dies draw a
sample f rom a si ngle organi zati on; presu m-
ably , t hi s w ou ld i ndc a t e that this particu lar
organi zati on is the popu lation abou t whi ch the
generalization is proposed. Bu t most re-
searchers want to generalizo their findings to
or gani zat i ons in general. There the sam-
ple/popu lation relation does not hold. Rarely
has any program of work drawn a sample of
enou gh organi zati ons to warranl the ty pe of
generalizations made allowable within the as-
su mptions of the stu dies themselves. Perhaps
the Aston stu dies and the "commu nication au -
dit" sponsored by the I nternational Commu ni-
Conceptu ol Foundations * 21
c a i on Association modeled after the As t on
stu dies are partly u sed exceptions. The i mpor -
tant poi nt is that many normative sty le stu di es
u se the rhetorical power of the n a t u r a l science
model and pri nci pi es of gener ali zat i on and
verif ication, bu t of ten cannot s u ppor t their
stu dies in organizaons based on it. The dis-
cou rse of ten conceals t hi s (see Su tton & Staw,
1995). Many attempts have been made to
su mmarize f i ndi ngs across stu dies, of ten u s-
i n g meta-analy sis. Su ch stu di es are often con-
tradictory and inconclu sive and even f u r t h e r
remove f i n di n g s f r om theoretical commit-
ments and specific site characteristics (see
Baker, 1991; Miller & Monge, 1985; W agner
fcGooding, 1987; W i lk i ns & Anderson, 1991).
1 Systems Theory
Du ring 1970s and 1980s, mu ch theoretical
attention was gi ven to developing "sy stems"
thnki ng in organi zati ons, especially regard-
i ng the organi zati on-envi ronment relati on
spawned i n part by the i nf lu e nc e of L awrence
and L orsch's (1967) work and the develop-
me n t of contingency theory (see Kat z &
Kahn, 1978; Monge, 1977, 1982; Monge,
Farace, Eisenberg, Miller, & W h i t e , 1984).
More recently , this work has become theoreti-
cally more sophisticated throu gh concepti ons
of self-organzing s y s t e ms and chaos theory
(see Bellman & Roosta, 1987; Contractor,
!994;Senge, 1990; Weick, 1979).
While sy stems approaches c o n t i n u t he
search for order and regu lari ty and u lt i mat ely
increased control by advantaged grou ps,,Ahey
tend to emphasize holi sm over a t omi s m and
dy nami c mu t u al cau sali ty over lawlike u n i d i -
rectional cau sality . Rather t ha n seek i ng s u r -
face-level, predictive variables the f oc u s is on
the deep processes of t r ansf or mat i on t hat pro-
du ce and interpret overt patterns of behav-
iorthe processes of organizing rather t h a n
organizations. As Pettigrew (1990) described:
"What is critical is not j u s t events, bu t the u n-
derly ing logics that give events me a ni ng and
signif icance . . . logics which may e x p l a i n
how and w h y these patterns occu r i n parti cu -
lar chronologcal sequ ence" (p. 273). In some
22 * Theorctical and Methodological Issues
cases, in Weick's work, for example, the focus
is so strongly on emergen! properties, the par-
ticular selting, and interpretive processes that
the research begins to sound much like inter-
pretive studies (see Daft & Weick, 1984). But
still, Ihe work s heavily guided by researcher
conceplions, the search for regularities antici-
pated by the researcher, the interpretation of
pattcrns in the researcher's logic, the view from
the outside, the hope for enduring regularities,
and the assumption of managerial goals. The
assumptions and regularities sought differ,
however, from those sought by covering laws
and even early systems theories.
Contractor (1994) has done an excellent
Job of making these differences clear. Five
conceptions are important.
1. Dynamic inferences: Covering-law theorists
develop hypotheses that posit a direc or in-
direct casual link between variables. These
can be tested usng rather standard slatistical
packages (SPSS). Dynamic hypotheses,
however, posit an underlying logic or rela-
lional mathematical rather ihan quantilalive
connection. Similar hypotheses developed
by covering-law and systems theorists are
the same only when one of two central cov-
ering-law assumptions are empirically pres-
en! (or methodologically produced): (a)
There is no change in the two variables over
time, or (b) the change s exactly equal.
2. Mutual causalty: While covering-law theo-
ries posit unidirectional causality, systems
theory suggests that many variables exist in
mutual or circular causal relations. In such
cases, Ihere can be no separation between
independen! and dependen! variables since
the casual relalion between the variables
runs both ways.
3. Hisloricily: Systems theory suggests ihat
the relation beiween variables is often lime
dependen!, henee universalizing claims or
even generalizations cannol be assumed
across time and place. Thus, variable rela-
lions (e.g., between trust and compliance)
presen! early in the history of an organiza-
lion can become quite difieren! as the
organizaron ages. A generalization about
organizational communicaton must always
reference the time in Ihe organization's hs-
lory during which i! was Irue.
4. Time irreversibility: Most covering-law
models assume thal social syslems work
like closed mechanical systems, henee jf an
increase in the quantity of a variable leads
lo an expecled oulcome then decreasing the
quality of that variable will lead to less out-
come. Rarely, however, are organizational
relations simply transitve or slable like
5. Disconlinuity: Covering-law Iheories as-
sume that changes are usually quantitative
and incremental. Syslems theorists display
the presence of sudden qualitative changes
a! certain thresholds.
Systems conceptions have clearly changed
the way people and scholars think about orga-
nizatons. Much theoretical writing is present.
But generally, the empirical research has been
more disappointing. Part of this arises from
the dominance covering-law conceptions have
had on defining the nature of "empirical" re-
search. Frequently, process conceptions in
systems theory are reduced to conceptions
where covering-law data gathering and statis-
tical analysis are applicable (see Everett,
1994; Monge, Cozzens, & Contractor, 1992).
In many respects, systems conceptions are
more productive in providing interesting and
useful conceptions of complex organizationa!
processes and interventions into them than
they are in generating studies that result in
journa! publications (see Cecchin & Stratton,
1991). The conceplion of useful empirical
work may well be biased in favor of cover-
ing-law style studies.
Communicaton Skills
The normative orientation not only guides
much organizational communicaton research
bul also teaching and cnsulting activities. Ar-
guably, much of the work going on under the
Concepta' Foundations * 23
tle "organizational communication" is more
skill development than research directed. In-
cluded is everything from interpersonal and
basic management skills to public speaking
and public relation skills. While it is not my
intent to provide any review of this work, I
think that it is important to show how text-
books as well as training and development
programs have traditionally been connected
wilh the normave approach to organization
In most cases, the implied pedagogy in the
writings has been didactic and reliant on the
presumption of an expert body of knowledge.
And most of the research on skills has used
covering-law style assumptions to test hy-
potheses and measure effectiveness. Further,
while there is a directive quality to this work,
the skills and the knowledge base from which
they are drawn are usually treated as valu
neutral and as equally available and valuable
for different organizational members. In do-
ing so, the influence and control orientation of
this work are treated as natural and self-evi-
dent, and other human goals and communica-
tion purposes are rarely considered. Usually,
upper management's goals for the organiza-
tion are accepted as given and legitmate.
Even when the skills are promoted primarily
for self-interests, generally those interests are
seen as well integrated with upper manage-
ment's organizational goals. Recently, as
teams, stakeholder participation, and organi-
zational creativity and learning have become
of greater concern there is increasingly criti-
cal attention to understanding skill needs cul-
turally, to the power relations in teaching and
textbooks, and to the needs and perspectives
of alternative organizational stakeholders
(Argyris, 1994; Eisenberg & Goodall, 1993;
Grunig & Hunt, 1984; Sprague, 1992).
i .
The Discourse of
Interpretive Studies
The ntimber and importance of interpretive
studies grew rapidly during the 1980s. For
most interpretive researchcrs, the organization
is a social site, a special type of community
that shares important characteristics wilh
other types of communities. The emphasis s
on a social rather than economic view of orga-
nizational activities. 'Traditional methods of
studying communities are seen as especially
useful. The expressed goal of many interpre-
tive studies is to show how particular realities
are socially produced and maintained through
ordinary talk, stories, rites, rituals, and other
daily activities: Most of the early attention for
organizational communication researchers
was derived from interest in the work of an-
thropologists such as Geertz (1973; see
Pacanowsky & O'DonnelI-Trujillo, 1982),
phenomenological and symbolic interactionist-
inspired work in sociology (Bantz, 1983;
Bormann, 1983; Douglas, 1970; Strauss,
1978), and the growing interest in hermeneu-
tics and qualitative research methods
(Trujillo, 1987).
While theoretical tensions and competitive
traditions have grown along with this work,
like these sources, much of the writings have a
clear preservationist, naturalistic tone. Allow
me to start with the more "naturalistic" as-
sumptions held by interpretive researchers in
their studies of organizational culture before
turning to some of the tensions that have de-
veloped, Like many of the more naturalistic
anthropological studies, interpretive research
often appears motivated to save or record a
life form with its complexity and creativity
before it is lost to modern, instrumental life.
The concern with community is often con-
nected with the maintenance of a traditional
sense of shared vales and common practices
and the presumed simple harmonious inner
life of people who lived in such communities.
Gergen (1992) described the romantic sense
of this discourse with its depth and connection
to the inner life bordering on sentimentality at
times. Because of this I refer to the time frame
as premodem in Table 1.3! This suggesis more
a concern with ihose aspects of life ihat have
not yet been systematized, instrumental ized,
24 * Theoretlcal and Methodologio Issues
a n d b r ou ghl u n der I he con tr ol of modern ism
logics a n d scicn ces t ha n a focu s on the past.
\lsludies in o r g a ni z a t i o ns ar e inter-
pr el i v c to t he ex t cn t tha t they n a ve n ot been
ca ptu r ed by n orma tivo, modemist co-
opta tion s. Most in ter pr etiv ists ha v e ta ken cu l-
tu r e to be a n evoca tive meta phor for orga n iza -
l i on a l l i f e r a ther iha n a v a r ia b le or t hi n g tha t
a n or ga n iza r on ha s (Frost, Moore, Lou is,
Lu n db er g, & Ma r tin , 1985, 1992; Smir cich,
1983). C u l t u r e dr a ws a tten tion to wha t orga n i-
za tion a l members mu st kn ow, believe, or be
a b le to do in order to opra te in a ma n n er tha t
is u n der st a n da b l e a n d a ccepta ble to other
members a n d the mea n s b y which this kn owl-
edge, b elief, a n d action routin es a re pr odu ced
a n d reproduced. The in ter est in cor n mu n ica -
t i on processes is fa r r icher tha n tha t of mea n -
in g tr a n smission presen! in n or ma tiv o wor k.
C ommu n ica tion is con sidered to be a cen tr a l
mea n s b y whi ch the mea n i n g of or ga n iza -
t i o n a l ev en ts is pr odu ced a n d susta in ed
{Don n ellon , Gra y, & Bou gon , 1986).
The ba sic f u n ction of in terpretive work s
"to tra n sa te the in ter ests a n d con cems of on e
people i n t o the in ter ests a n d con cer n s of a n -
other" (Pu tn a m, Ba n tz, Deetz, Mu mb y, & Va n
Ma a n e n , 1993). The needs of tr a n sla tion re-
qu ir e both a ca r efu l u n der sta n din g of the
olher a n d a n a b i l i t y to presen ! tha t u n der -
st a n di n g to on e's own culture. A dou b le her -
men eu t i c (a n in ter pr eta tion of a n in terpreted
wor ld) a n d complex commu n ica tiv e process
( met a c ommu n i c a t i on to the cu ltu r a lly di f f er -
e n t ) is t h u s cen tr a l to in ter pr etiv e work a n d
la r gely a ccou n ts for the situa ted a n d emergen t
n a t u r e of the u n der s t a n di n g pr esen t in its texts
(Ba r ley .1990) .
The in ter pr etiv e resea rcher of ten en ga ges
in some t y pe of pa r ticipa n ! ob ser v a tkm or
olher person a l con ta ct to collect ma ter ia l a n d
wor k ou l u n der sla n din g wi t h the site commu -
n i t y . Stu dies a r e u su a l l y don e in the field a n d
a re based on a prolon ged period of observa -
tion a n d/or dept h i n t er v i ewi n g. The in ter est s
in the f u l l person in the or ga n iza tion , t hu s so-
cia l a n d l i f c f u n ction s b eyon d the r ela tion to
the job are con sider ed.JThe goa ls a r e mu ch
mor e opcn a n d emer gen t tha n in n or ma tiv e
work a n d mu ch less con n ected to issues of ef-
ficien cy a n d pr odu ctiv ity. ;The wor kpla ce is
seen a s a site of hu ma n a ctiv ity , on e of those
a ctvities b ein g "work" prpi. The orga n iza -
tion of the en tir e socia l commu n ity is of in ter -
est. While some wr in g mi ght b e somewha t
impression istic a n d focus on the surfa ce feel-
in gs a n d mea n in gs of either the cu ltu r a l mem-
ber or the resea rcher, gen era lly, these wou ld
be con sidered wea k a n d sha llow stu diesi The$
poin t more often is to u n der sta n d the socia l
con dition s of lif e gi v i n g r ise to such f eelin gs
a n d mea n in gs the deep cu ltu r a l rea d. An a -
l y t i c a tten tion is t hu s of ten dir ected to sy m-
bolism, meta phors, stories, jokes, a dyice a n d
rea son giv in g, n a r r a tiv e for r a s, r ites a n d r itu -
a is, a n d the socia l f u n ction s of these a ctivitiesj
(see, for exa mples a n d r ev iews, Br own , 1985;
Br own in g, 1992; Gooda ll, 1990; K n u f , 1993;
Smi t h&Ei sen b er g, 1987;Tr u jilio, 1987).
' In terpretive studies a ccept mu ch of the rep-
r esen ta tion a l a n d con sen su a l view of scien ce
seen in n or ma tiv e wr itin gs, b u t shif t the r ela -
tion between theoretica l con ception s a n d the
ta lk of the subjects u n der stu dy. 'People a re n ot
con sidered to be objects like other objects, but
a re a ctive sen se ma kers like the researchej.
Theory is given a fa r weaker role here. While
theory ma y pr ov ide impor ta n t sen sitizin g con -
ception s, it n ot a dev ice of cla ssif ica tion or
tested in a n y simple a n d direct ma n n er . The
key con ception s a n d u n der sta n din gs mu st b e
worked ou t with the subjects u n der st u dy . Re-
sea rch su b jects ca n colla bora te in di s pl a y i n g
key fea tures of their wor ld. Bu t like n or ma tiv e
resea rch, the pr essu r e is to get it r i ght , to dis-
pla y u n i f i ed, con sen su a l cu ltu r e in the wa y
tha t it "a ctu a lly" exists. The r epor t is to dis-
pla y con v in cn gly a un ified wa y of l i f e wilh
a ll its complexities a n d con tra diction s (Good-
a ll, 1990; Pa ca n owsky & O'Don n ell-Tr u jillo,
1982; Van Ma a n en , 1988). In DiMaggio's (1995)
con ception s, theor y in in ter pr etiv e wor k is of -
ten a n a r r a tiv e a ccou n t of socia l processes
"with empha sis on empir ica l tests of the pla u -
sib ility of the n a r r a tiv e a s well a s ca r ef u l
a tten tion to Ihe scope of the a ccoun t" (p. 391).
On e gets a sen se in tr a ckin g t hi s wor k over
time tha t it is b ecomin g less pr odu ctiv o in it-
' "
self a n d mor e treated a s a supplemen t to other
kin ds of work. Ba rley et a l. (1988) repre-
sented well how the ea rly n a tu r a listic a n d a n -
thr opologica l in terest in orga n iza tion a l cu l-
tures gr a du a lly wa s eclipsed by a ma n a geria l
in ter est in ma n a gin g cu ltu r e. Hen ee, mu ch of
the discu ssion of cu ltu r e ha s been reduced to
"cu ltu r a l va ria bles," a n d the studies became
mor e n or ma tiv e a n d like the clima te stu dies
tha t preceded ihem. Da ta collection tech-
iqu es a n d con ception s emergen t in the field
ha ve been borrowed a n d trien accepted a pr ior i
in codin g a n d cou n tin g studies, for exa mple,
those cor r ela tin g cu ltu r a l cha r a cter istics with
pr odu ctiv ity mea su r e or a da pta tion to cha n ge
(e.g., Ba stien , 1992; Fa irhurst, 1993). An d
critica l resea rchers of ten r ein ter pr et in terpre-
tiv e stu dies a ddi n g cr itiqu es of mea n in g f or -
ma tion s (Ma r tin , 1992; Mu mb y, 1987). Still,
in ter pr etiv e wor k is a n a ctiv e a n d via ble r e-
search or ien ta tion a s it con tin es to evolve. A
n u mb er of resea rch a pproa ches ha ve been
used to help sort out hidden mea n in gs, often
hidden a s well f r om the site c ommu n i t y pa r-
ticipa n ts owin g to surface f a milia r ity, a n d to
or ga n tze the research process a n d the presen-
ta tion of the stu dy itself,
This essa y wi l ] ma ke n o a ttempt to sort out
the diverse wa ys tha t in terpretive researchers
ha ve collected, a n a lyzed, a n d reponed the ob-
serva tion s on which their wor k is based. Eth-
n ogr a phy a n d simila r con ception s of "n a tura l-
istic" i n q u i r y a lr ea dy discussed r ema in the
purest for r a of in ter pr etiv e wor k, Specific site
studies ha ve been a n a lyzed f ocu si n g on meta -
phor s, symb ols, a n d themes (Pa ca n owsky &
O'Don n ell-Tr u jillo, 1982; Smith & Eisen -
berg, 1987; Smith & Turn er, 1995; Trujillo,
1987). Other stu dies ha ve followed other tra -
dition s in clu din g dr a ma tu r gy (Gooda ll, 1990;
Ma n n i n g, 1992), n egotia ted order (Geist,
1995), str u ctu r a tion (Ba stien , McPhee, &
Bolton , 1995; Poole & McPhee, 1983), a n d
r u les theor y (Scha ll, 1983). Du r in g the past 15
lo 20 years, a rich a rra y of stu dies has been
completed. Together these ha v e displa yed
how or ga n iza tion a l cu l t u r es dev elop a n d
cha n ge, how socia l groups con ceive a n d ha n -
C on cept u o foundotons + 25
dle con f lict, how in stitu tion a l structures a re
cha llen ged a n d/or rein sta ted, how cu ltu r es
dif f er across n a tion a l settin gs a n d ma n a ge-
men t practices, an d so for th (see Pepper,
In terpretive studies a re a lso still ev ov in g.
Gr a du a lly, ma n y researchers doin g in ter pr e-
tiv e wor k ha v e began to qu eson the logic of
displa yin g a con sen sua l un ified cu ltu r e a n d
ha ve a tten ded more to its f r a gmen ta tion , ten -
sion s, a n d processes of con f lict suppression
(Frost et al., 1992; Ma r cu s & Fischer, 1986;
Ma r tin , 1992). In this sense the work ha s be-
come more dia logic in character. An d wor ks
followin g str u ctu r a tion theor y ha v e become
more critica l tha n in terpretive in cha r a cter
(Ba n ks & Riley, 1993; Howa rd & Geist, 1995;
Riley, 1983).
Sin ce the mid-1980s, mu ch of the self-
r eflection in in terpretive work ha s focused on
the rela tion of the resea rch to the site commu -
n i t y a n d the "voice" ta ken in the research re-
por t. Of importa n ce a re both the politics of
represen ta tion a n d the r ole of the report a uthor
(C lif f or d & Ma rcus, 1986; Con quergood,
1991; K a u f f ma n , 1992). Va n Ma a n en (1988)
su mma r ized these r ela tion s a s a ltern a tive
tales. Further with greater a lten tion to the re-
l a t i on to the commu n ity a n d a ction poten tia l
in resea rch, in ter pr etiv e work ha s become
more pa r ticipa tor y (Whyte, 1991). Reason
(1994) described dif f er en t types of pa rticipa -
tory in quiry. These cha n ges ha v e con tin ued to
move mu ch in terpreta tive wor k to b e mor e
dia logic a n d critical in its a ccoun t of itself a n d
the type of wor k don e.
T h e Discourse of
Critical Studies
Critical resea rchers see or ga n iza tion s in
gen era l as social historica l crea tion s a ccom-
plished in con dition s of struggle a n d power
r ela tion s. Orga n iza tion s a re la rgely described
as politica l sites, thu s gen er a l social theories
a n d especia lly theories of decisin ma kin g in
the pubtc sphere a re seen a s a ppropria te.
26 4 Theoretical and Methodological Issues
While organizations could be positive social
inslitutions providing forums for the articula-
lion and resolution of importan! group con-
flicts ovcr the use of natural resources, distri-
bulion of income, production of desirable
goods and services, the development of per-
sonal qualities, and Ihe drection of society,
various forms of power and domination have
Icd lo skewed decisin making and fostered
social harms and significant waste and ineffi-
ciency. Either explicit or implicit in their pre-
sentation is a goal to demnstrate and critique
forms of domination, asymmetry, and dis-
torted communication through showing how
reality can become obscured and misrecog-
nized. Such insights help produce fonims
where the conflicts can be reclaimed, openly
discussed, and resolved with fairness and
Critical research aims at producing
dissensus and providing fonims for and mod-
els of discussion to aid in the building of more
open consensos. Of special concern are forms
of false consciousness, consent, systemati-
cally distorted communication, routines, and
normalizations that produce partial interests
and keep people'from genuinely understand-
ing or acting on their own interests. Of the
four orientations, critica! studies have the
most expiicitly stated valu commitments and
Ihe most explicit attention to moral and ethical
issues. With Ibis, much of the discourse has a
suspicious and iherapeutic tone, but also a the-
ory of agency that provides an activisl tone, a
scn.sc thal people can and should act on these
conditions and ihat improved understanding
as wcll as access to communication forums is
core lo positive aclion. Theory development
i critica! theory often has an "enlightenment"
i|ualily, in DiMaggio's (1995) sense, whereby
cuphemisms are developed or exposed "clear-
ing away conventional notions to make room
for iirlf ul and exciling insights" (p. 391; see
ilso Bourdicu, 1991, for the power of re-
The ccnlral goal of critical tfieory in orga-
n/ ntinnal communication studies has been to
acule a society and workplaces that are free
from domination and where all members can
contribute equally to produce systems that
meet human needs and lead to the progressive
development of all. Studies have focused both
on the relation of organizations to the wder
society and their possible social effects of col-
onization (rationalization of society) and
domination or deslrucon of the public sphere
(Deetz, 1992; DuGay, 1997), and on internal
processes in terms of the domination by in-
strumental reasoning, discursivo closures, and
consent processes (e.g., Alvesson, 1993;
Clair, 1993a, 1993b; Forester, 1989; Mumby,
1987, 1988). As indicated they tend to enter
their studies with a priori theoretical commit-
ments, which aid them analytically to ferret
out situations of domination and distortion.
Critical studies include a large group of re-
searchers who are difieren! in theory and con-
ception but who share important discursive
features in their writing. They include Frank-
furt school critical theorists (see Alvesson &
Willmott, 1992, 1996; Czamiawska-Joerges,
1988; Mumby, 1988), confiict theorists
(Benson, 1977; Dahrendorf, 1959), some
structurationisls {Banks & Riley, 1993;
Giddens, 1984,1991;Howard&Geist, 1995),
some versions of feminist work (e.g., Alien,
1996,1998; Benhabib, 1992; Ferguson, 1984,
1994), some Burkeans (Barker & Cheney,
1994; Tompkins & Cheney, 1985), and most
doing labor process theory (Braverman, 1974;
Burawoy, 1979, 1985; Knights & Willmott,
Critical theorists sometimes have a clear
political agenda focused on the interests of
specifc identifiable groups such as women,
workers, or people of color, but usually ad-
dress general issues of goals, vales, forms of
consciousness, and communicative distor-
ons within corporations. Their inters! in
ideologies considers disadvantaged groups
difficulties in understanding their own po-
litical interest, but is usually addressed to peo-
ple in general, challenging consumerism,
careerism, and exclusive concera with eco-
nomic growth (Alien, 1998; DuGay, 1997).
Compared to Marxism, critical theory is not
antimanagement per se, even though one
tends to treat management as institutionalized
and ideologies and practices of management
as expressions of contemporary forms of
domination. Two principal types of critical
studies can be identied in organization stud-
ies: Ideological critique and communicative
Ideology Critique
Most of the critical work has focused on
ideology critique, Analyses of ideologies
show how specifc interests fail to be realized
because of people's inability to understand or
act on their own interests. Some identified
ideologies are group specifc and others are
held by people in technological-capitab'st so-
ciety in general. Ideological critique is guided
by a priori researcher conceptions and aims at
producing dissensus with the hope that the re-
covered conflicts and explicit concern with
vales will enable people to choose more
clearly in their own interests.
The earliest ideological critiques of the
workplace were offered by Marx. In his analy-
ses of work processes, he focused primarily
on practices of economic exploitation through
direct coercin and structural differences in
work relations between the owners of capital
and die owners of their own labor. However,
Marx also describes the manner in which the
explotative relation is disguised and made to
appear legitmate. This is the origin of ideol-
ogy critique. Clearly, the themes of domina-
tion and exploitation by owners and later by
managers have been central to ideology cri-
tique of the workplace in this century (see
works as varied as Braverman, 1974; Clegg &
Dunkerley, 1980; Edwards, 1979). These later
analyses became less concemed with
class-based coercin and economic explana-
tions through focusing on why coercin was
so rarely necessary and on how systemic pro-
cesses produce active consent (e.g., Burawoy,
1979, 1985; Czarniawska-Joerges, 1988;
Deetz & Mtimby, 1990; Gramsci, 1929-1935/
Concepta! Fotmdatans i 27
1971; Kunda, 1992; Vallas, 1993). Ideology
produced in the workplace would supplement
ideology present in the media and the growth
of the consumer culture and the welfare state
as accounting for workers' and other stake-
holders' failure to act on their own interests.
Four themes recur in the numerous and
varied wrilings about organizations working
from such a perspective: (1) concern wii rei-
fication, or the way a socially/historically
constructed world would be treated as neces-
sary, natural, and self-evident; (2) the suppres-
sion of conflicting interests and universaliza-
tion of managenal interest; (3) the eclipse of
reason and domination by instrumental rea-
soning processes; and (4) the evidence of con-
In reification, a social formation is ab-
stracted from the ongoing conflictual site of
its origin and treated as a concrete, relatively
fixed entity. The illusion that organizations
and their processes are "natural" objects pro-
tects them from examination as produced un-
der specific historical conditions (which are
potentially passing) and out of specific power
relations. Ideolgica! critique is enabled by
the elite-driven search for reifications in ev-
eryday life. The resultan! critique demn-
strales the arbitrary nature of "natural objects"
and the power relations that resuli and sustain
these forms for the sake of producing
dissensus and discovering the remaining
places of possible choice.
Lukcs (1971), among many others (see
Giddens, 1979), has shown that particular sec-
tional interests are often universalized and
treated as if they were everyone's interests,
thus producing a false consensus. In contem-
porary corporate practices, managenal groups
are privjleged in decisin making and re-
search. The interests of the Corporation are
frequently equated with management's inter-
ests. For example, worker, supplier, or host
community interests can be interpreted in
lerms of their effect on corporatethat is,
universalized managerialinterests. As such
they are exercised only occasionally and usu-
ally reactively and are often represented as
28 ^ Theoretlcai and MethodologicaJ Issues
s i m p l y econom ic commodities or "costs"
f or exam p l e, ihe price the "corporation" m u s t
p ay f or l ab or , s u p p lies , or environm ental
cleanu p (Deetz, ] 995b). Central to the u niver-
salization of m anagerial interes t is the reduc-
tion of the m l t i p l e clam s of owners hip to f i-
nancia! ownership. In ideolgica! critique,
m anageri al advanlages can be seen as pro-
duced his torically and actively rep roduced
t hrou gh ideological dis cu rs ive p ractices i n so-
ci et y and in corporations themselves (see
B u l l i s & Tompkins, 1989; Deetz, 1992;
Mu m b y , 1987). Critical theory j oi ns other re-
cent theories in argu ing f or the representation
of the f u l l variety of organizational stake-
holders (see Carroll, 1989; Freeman &
L i edt ka, 1991).
Habermas (1971, 1984, 1987) has iraced
the social/histrica! emergence of technical
rationality over com p eting f orm s of reason,
Habermas described technical reasoning as
i ns t ru m ent al , t e ndi ng to be governed by the
theoretical and hypothetical, and f ocu s ing on
control through the develop m ent of m eans -
ends chains. The natu ra! opposite to t hi s ,
Habermas conceptualizes as a practica! inter-
est. Practica! reasoning focuses on the process
of u nders t andi ng and m u tu al determ ination of
the ends to be sought rather than control and
development of means of goal accomplish-
m ent . Bu t in the contem p orary social s itua-
t o n , the f orm and cont ent of modern social
science and the social constitution of expertise
align wi l h organizatonal s tructures to pro-
du ce the dom i nat i on of technical reas oning
(see Alves s on, 1987a; Fischer, 1990; Mu m b y ,
1988; Stablein & Nord, 1985), To the extent
t hat technical reas oning dom inates , it lay s
claim to the entire concept of rationality , and
alternative f onns of reason appear irrational.
To a large extenl, studies of the "hu m an" side
of organizations (clim ate, Job enri chm ent ,
q u a l i l y of work l i f e, worker p articip ation pro-
gr a m s , and c u l t u r e ) have each been trans-
f orm ed f rom alternative ends into new means
t o be b rou ght u ndcr technical control f or ex-
I c n d i n R Ihe d o m i na nt group interests of the
coi p or at i on (Alves s on, 1987a; Barker, 1993;
Wendt, 1994). The p roductive tens in be-
tween the two becomes s ubm erged to the
ef f icient accom p lis hm ent of of ten u nkno w n
b u l surely "rational" and "legitmate" corpo-
rategoals (Crter &Jacks on, 1987).
Early critical theorists f ocused p ri m ari l y
on bureaucracies and other f orras of direct
control and dom ination. As the work has de-
veloped and these f orm s have declined, m ore
sophisticated concep tions of power have
arisen. Various f orm s of indirect control have
becom e of greater concern (see Edwards ,
1979; Lukes, 1974). Many of the these f orm s
of indirect control involve active "consent" of
those controlled. Consent processes occur
throu gh the variety of s itu aons and activities
in which s om eone actively, t hou gh of ten u n-
knowi ngl y , accomplisb.es the interests of oth-
ers in the f a u l t y attem p t to f u l f i l l his or her
own. People are oppressed but are also en-
ticed into activies that crate comp licity in
t hei r own victim ization (f or exam p les , s ee
B ru ns s on, 1989; Clair, 1993a; Pringle, 1989).
As a res ult, rather than havi ng open dis cus -
s ions , dis cus s ions are foreclosed or mere ap-
pears to be no need f or discussion. The inter-
action processes rep roduce f ixed identities ,
reiations , and knowledge, and the variety of
possible differences are lost. Thus, importan!
discussions do not take place because there
ap p ears to be no reason f or them . Cons ent of -
ten appears in direct f orm s as members ac-
tively s u b ord nat e them s elves to ob tain
money, security, meaning, or identity ; things
t hat should res ult f rom the work process
rather than s u b ordi nad on. In f act, both the
s ub ordination and requ rem ent of it ham p er
ie accom p l i s hm ent of these work goals. Crit-
ical organi zat i onal com m u ni cat i on research
du r i ng the 1980s and 1990s includes a rather
wide body of studies s howing where c u l t u r e
and cu ltu ral engineering may be described as
hegem onic (e.g., Alves s on, 1987b; K ni ght s &
Willmott, 1987; Mu m b y , 1988, 1997; R osen,
1985). Other researchers have s hown how
normative, unobtrusive, or concertive control
processes develop in organizations and sub-
vert employee p articip ation p rogram s (see
Conceptu/ Fou ndot i ons * 29
Barker, 1993;Barker&Cheney , 1994;Barker,
Melville, & Pacanows ky , 1993; Barley &
K u n d a . 1992; Bu llis , 1991; Bullis & Tomp-
kins , 1989; Etzioni, 1961; K u nda, 1992; La-
zega, I992;Schwartzm an, 1989).
Several Hm itations of ideology critique
have been demonstrated Three criticisms ap-
pear m os t com m on. First, ideology critique
appears ad hoc and reactive. It largely ex-
p lains af ter the f act why s om ething d i d n ' t
happen. Second, the elitist is of ten criticized.
Common concepta like false needs and f alse
consciousness p res um e a basic weaknes s in
i ns i ght and reas oning processes in the very
same people it hopes to empower. The irony
of an advcate of greater equalily pronounc-
ing what others should want or how they
should perceive the world "better" is apparent
to both dom inant and dominated grou p s .
Third, some accounts f rom ideology critique
appear f ar too s im p lis tic. These studies appear
to claim a s ingle dom inant group t hat has in-
tentionally worked out a system whereb y
dom ination through control of ideas could oc-
cur and its interest could be secured. Clearly,
dom ination is not so simple. Certainly the
power of ideology critique can be maintained
withou t f alling to these criticis ms , and most
studies today carefully avoid each problem.
Largeiy this has been aided by the develop-
m ent of Habermas's theory of com m uncative
Communcative Acion
While earlier critical studies focused on
distortions of cons cious nes s , thought, and
meanings , Hab erm as 's work s ince the late
1970s has concentrated on distortions in com-
m u ni cat i on processes (Habermas, 1984,
1987). This project retains m any of the f ea-
lures of ideology critique, i nc l u di ng the ideal
of sorting out cons training social ideas f rom
those grounded in reason, but it envisages pro-
cedural ideis rather than substantive critique
and thu s becomes quite dif f erent f rom tradi-
tional ideology critique. It also introduces an
af f irm ative agenda, not based on a utop ia, bul
still on a hope of how we m i ght ref orm ins titu -
tions along the lines of m orally driven di s -
course in s itu ations ap p roxim ating an "ideal
speech s ituation" (see Mum b y , 1988). Organi-
zational com m unication scholars have devel-
oped these ideas to support more participatory
com m unication and decis in m aking in orga-
nizations and to dis p lay power-based lim -
itations on organizational dem ocratization
(Cheney, 1995; Deetz, 1992,1995b; Forester,
1989, 1993; Harrison, 1994). From a partici-
p ation perspective, com m unication di f f i cu l -
tes arise f rom communication practices that
preclude valu debate and conf lict, that s u b -
stitute im ages and im aginary reiations f or
s elf - p res entation and truth claim s , that arbi-
trarily lim it access to com m u nication chan-
nels and f oru m s , and that then lead to deci-
s ions based on arbitrary authorily reiations
(see Deetz, 1992, for development).
Basically, Habermas argued that every
speech act can f u nct i on in com m u nication by
vi rt u e of common p res u m p ons made by
speaker and listener. Even when these pre-
sumptions are not f ulf illed in an act u al s itu a-
u'on, they serve as a base of appeal as f ailed
conversation t u m s to argumentation regarding
the disputed validity claims. The basic pre-
sumptons and validity claims arise oul of f ou r
shared dom ains of reality: language, Ihe exter-
na! world, human reiations, and the i ndi v i d-
ual's interna! world. The claim s raised in each
realm are, res p ectively : i nt el l i gi b i l i t y , tru th,
correctness, and s incerity . Each competenl,
com m uncative act makes f ou r ty p es of
claims: (1) presenting an available u nder-
standable expression, (2) asserting a knowl-
edge proposition, (3) establishing legitim ate
social reiations, and (4) dis clos ing the
speaker's posioned experience. Any of these
claims that cannot be b rought to open dis p ute
serves as the basis for s y s tem atically distorted
com m unication. The ideal speech s itualion is
to be recovered to avoid or overeme s uch dis-
tortions .
The ideal speech s itu ation, t hu s , describes
f our basic g u i d i ng conditions as necessary f or
30 * Theoretical and Methodological Issues
frcc and open participation in the resolution of
conf licting claims. First, the attempt to reach
imderslanding presupposes a symmetrical dis-
trib ution of the chances to choose and apply
speech acts that can b e heard and understood.
This w ould specify the minimal conditions of
sk ills and opportunilies f or expression includ-
ing access to meaningf ut forums, media, and
channels of communication. Second, the un-
derstanding and representaron of the external
w orld needs to be freed from privileged pre-
conceptions in the social development of
"truth." Ideally, participants have the opportu-
nity to express interpretations and explana-
tions w ith conf licts resolved in reciprocal
claims and counterclaims w ithout privileging
particular epistemologies or f orms of data.
The freedom f rom preconception implies an
examination of any ideology that would privi-
lege one f orm of discourse, disq ualif y certain
possib le participants, and universalize any
particular sectional interest. Third, partici-
pants need to have the opportunity to estab lish
legitimate social relations and norms for con-
duct and interaction. The rights and responsi-
b ilities of people are not given in advance b y
nature or b y a privileged, universal valu
structure, b ut are negotiated through interac-
tion. The reif ication of organizational stnic-
tures and their maintenance w ithout possib le
dispute and the presence of managerial pre-
rogatives are examples of potential immoral-
ity in corporate discourse. Finally, interactants
need to b e ab le to express their own authentic
inlercsts, needs, and feelings. This would re-
q uire freedom f rom various coercive and he-
gemonic processes b y which the individual is
unab le to f orm experience openly, to develop
and sustain compeng identities, and to form
expressions presenting them.
The most f requent ob jection to Habermas,
and those who have followed this work , is that
he has overemphasized reason and consensus
and has only a negative view of power, w hich
h.impers b oth the conception of social change
and secing the possib le positivity of pow er
(scc Benhab ib . 1990; Lyotard, 1984). What
Hahcrmas does w ell is to give an arguab le
standard for normative guidance to communi-
cation as a critique of domination, even if his
position is distinctly Western, intellectual, and
male (Fraser, 1987; see Benhab ib , 1992, for a
discusson of these prob lems and w ays of re-
covering the critica! thrust of his work ). The
participative conception of communication
describes the possib ility and conditions for
mutual decisin mak ing and also provides a
description of communication prob lems and
inadequacies. In general, most strategic or in-
srumental communicative acts have the po-
tential of asserting the speak er's opinin over
the attempt to reach a more representative
consensus. In such cases, an apparent agree-
ment precludes the conf lict that could lead to
a new position of open mutual assent. In cases
where the one-sidedness is apparent, usually
the processes of assertion/counter-assertion
and questions/answers reclaim a situation ap-
proximating participation.
Critica! theorists have been very effective
in show ing the invisib le constraints to mutual
decisin mak ing in organizations. In many
work places today, strategy and manipulation
are disguised and control is exercised through
manipulations of the natural, neutral, and
self-evident. Critica! work has b oth demon-
strated the presence of ideological domination
and processes of "discursive closure" and
"systemalically distorted communication"
(see Deetz, 1992, chap. 7). While Hab ermas
has been criticized for focusing too much on
consensus at the expense of conf lict and
dissensus, implicit in his analyses is the recov-
ery of conf lict as an essential precursor to a
new consensus and the perpetual critiq ue of
each new consensus as interaction contines.
The Discourse o/Dialogic Studies
I have chosen the term dialogic rather than
the more ob vious postmodemist to organize
this discourse because it attends to k ey fea-
tures of this work and because of the grow ing
commercial use of the term postmodem, re-
sulting in increased dif f icuhy in distinguish-
ing realist assumptions ab out a changing
world (a postmodem w orld) and a postmodern
discourse, w hich denies realist claims ab out
the world (Jones, 1992; Park er, 1992). The
term also mak es it easier to include older the-
orists such as Bak htin for w hom the term
postmodern seems inappropriate (see Shot-
ter, 1993). Dialogic perspectives are based in
a recen set of philosophical wrings originat-
ing most of ten in France. Of greatest interest
are the writings emphasizing political issues
and conceptions of f ragmentation, textuality,
and resistance. These philosophically based
approaches to organization studies have
emerged out of w ork s of Bourdieu, Derrida,
Lyotard, Kristiva, Foucault, Baudrillard,
Deleuze and Guattari, and Laclau and Mouf f e.
Organizational researchers f ollow ing the gen-
eral themes of this w ork include Haw es
(1991), Martin (1990), Calas and Smircich
(1991), Mumb y and Putnam (1992), Knights
(1992), Burrell (1988), Bhab ha (1990),
Bark er and Cheney (1994), Holmer-Nadesan
(1997), Ashcraf t (1998), and several of the es-
says in Hassard and Park er (1993). As w ith
critical writings, this is a wide group of writ-
ers and positions w ith their own disputes, b ut
their work shares features and moves that can
b e highlighted in treating them logether.
Lik e critical studies, the concern is of ten
w ith asymmetry and domination in organiza-
tional decisin mak ing, b ut unlik e the critical
studies' predef mion of groups and types of
domination, dialogic studies focus more on
micropolitical processes and the joined nature
of power and resislance. Domination is seen
as fluid, situational, and w ithout place or ori-
gin. Even group and personal identities cannot
be seen as fixed or unitary. The attention is to
reclaim conf licts suppressed in everyday ex-
periences, meaning systems, and self-concep-
u'ons. Rather than a ref ormation of the world,
dialogic studies hope to show the partiality
(the incompleon and one-sidedness) of real-
ity and the hidden poinls of resistance and
complexity. In place of an active political
agenda and utopian ideis, attention centers
on the space for a continually transf orming
Conceptu! Foundotiom 4 31
world through recovery of marginalized and
suppressed peoples and aspects of people.
Dialogic research emphasizes dissensus
production and the local/situated nature. of un-
derstanding. f vlany of the conceptions on
w hich this is based are dif f icult and not terri-
b ly w ell k now n b y organizational communi-
cation scholars. Ow ing to this I w ill provide
some greater detail here. Seven themes w ill b e
highlighted: (1) the centrality of discourse,
emphasizing language as systems of distinc-
tions thal are central to social construction
processes; (2) fragmened identities, demon-
strating the prob em of an autonomous,
self -determining individual as the origin of
meaning; (3) the critique ofthe philosaphy of
presence, f ocusing on object indeterminacy
and the constructed nature of people and real-
ity ; (4) the loss of foundations and master
narratives, arguing against integrave meta-
narratives and large-scale theorctical systems
such as Marxism or f unctionalism; (5) the
knowledge/power connecion, examining the
role of claims of expertise and truth in sys-
tems of domination; (6) hyperreality, empha-
sizing the f luid and hyperreal nature of the
contemporary w orld and role of mass media
and inf ormation technologies; and (7) re-
search as resistance and indelerminacy, stress-
ing research as importan! to change processes
and providing voice to that w hich is lost or
covered up in everyday lif e. Each of these has
an impact on conceptions of q uality commu-
nication, processes of decisin mak ing, and
research directions.
The Centrality of Discourse
Most curren! dialogic studies grew out of
French structuralism b y tak ing seriously the
"linguistic turn" in philosophy. In this sense,
dialogic studies developed the French tradi-
tion b y mak ing the same move on structuralist
thought that Hab ermas and others n critical
studies did on ideological critiq ue in the de-
velopment of communicative aclion in the
Germn tradition. Language replaces con-
sciousness as central to experience. Textual/
32 ^ Theoretical and Methodological Issues
d i s c u r s i v c f i c l d s repl aced the structure of
thc u nc ons c i ou s and /or cul tur al s tr u c tu r es
c l ai m ed as u ni v er s al . Both crtical and d i a -
l o g i c theor i s ts used these lo f ig ht a two-f r ont
war ; f i r s t, ag ai ns t normative r es ear c her s and
other obj c c ti v i s ts w i th their science aimed at
c ontr ol l i ng natu r e and peopl e, and s ec o n d ,
ag ai ns t i nter pr eti v e researchers and other hu -
m ani s ts w i t h thei r pr i v i l eg i ng of i nd i v i d u al
ex per i enc e, u n i q u e hu m an r i g hts , and nai v e
v er s i ons of hu m a n f r eed om. As discussed
later, the l i ng u i s ti c t u m enabled a critique of
nor mal i v e research's cl aim of objectivity
thr ou g h ex am i ni ng the processes by w hi c h
objects are social l y c ons ti tuted and the rol e of
l ang u ag e i n that process and s i mu l taneou s l y a
c r i ti q u e of interpretive research thr ou g h dem-
ons tr ati ng the f r ag m entati on of cul tur es and
personal i d enti ti es and r emov i ng the psycho-
l ogical s u bjec t f rom the center of experience.
Focusing on l ang u ag e al l owed a conception of
social c ons tr u c ti oni s m that deni ed the nor ma-
ti v e cl aim of c er tai nty and objective tr u th and
the interpretivists' r el i anc e on experience and
neutral c u l tu r al claims that led them to miss
the s oc i al /l i ng u i s tc pol itics of exper i enc e.
Com m u ni c ati on mus becomes a mode of ex-
pl anati on of org anizations and ad i v i nes asso-
ciated w i t h them rather than a phenom enon lo
be expl ained w i thi n them.
M any or g ani zati onal researchers hav e used
thi s i n s i g ht to produce di scur si ve, c ommu ni -
c ati on-c enter ed anal yses of org anizations.
M any of the mor e em pi r i c al dial og ic studies,
bu t not al ] , hav e f ol l owed Foucaul t's concep-
ti o n of d s c ou r s e. For exampl e, Kni g hts and
W i l l m o t t (1989) and Mi l l s (1994) d em on-
strated the way beng subjected l ed to par ti c u -
l ar f or ms of s u b j u g ati on; Knig hts and Morg an
(1991) used Fou c au l t's discursive practices to
s how thc c ons tr u c ti on of person and worl d in
i he d i s c ou r s e of s tr ateg y; Townley (1993) ap-
pl i ed i l to the discourse of hu m an resource
m anag c m ent; and 1 (Deetz, 1998) hav e shown
ho w s el f -s u r v ei l l anc e and sel f -subordination
r epl ac e ex pl i c i l control systems i n knowl -
c d g e- i n t en s i v e c om pani es . Works f o l l o w i ng
oi her r c l atc d phi l os ophi c al perspectives on
discourse hav e tended to be somewhat more
theoretical (e.g., Burrel l , 1988; Cooper, 1989;
Deetz, 1994d;Hawes, 1991).
Fragmened Identities
The pos i ti on on the per son f ol l ow s d i r ec tl y
f r om the conception of d i s c ou r s e. Enl i g hten-
ment thou g ht centered knowl edg e and under-
s tand i ng i n a c onc epti on of an au tonom ou s
and coherent subject. This c onc epti on leads to
an emphas i s o nw hat was devel oped in thi s
essay asa c ons ens u s d i s c ou r s e i n science
and society. Dial ogic studi es reject the noti on
of the au tonom ou s , s el f -d eter mi ni ng i nd i v i d -
ual as the center of the social universe and in
its pl ace s u g g es t the compl ex, c o nf l i c tu al s u b -
ject w i th an emphasis on f u nd am ental dis-
sensus (see Garsten & Grey, 1997; Henr i q u es ,
Hol l way, Ur w i n, Venn, & W al ker d i ne, 1984;
Mi l l s , 1994; Nu kal a, 1996).
There are two versions of thi s c r i ti q u e of a
secure uni tar y i d enti ty. The f i r s t suggests that
the Western conception of man as a centered
subject has al ways been a m y t h. Fr eu d 's work
i s used to show the g r owi ng awar enes s i n
Western tho u g ht of the d i f f i c u l ti es wi th i t.
People have al ways been f i l l ed w i th c onf l i c ts .
The c onc epti on of a u ni tar y au tonom ou s self
was a f iction used to suppress thos e c onf l i c ts
and privilege m as c u l i ni ty, r ati onal i ty, v i s i n,
and control . To the extent that d om i nant d i s -
courses spoke the person, the person g ai ned a
secure i d enti ty bu t participated i n the repro-
d u c ti on of d omi nati on m ar g i nal i zi ng the other
parts of the self and other groups. The sense of
au tonom y served to cover thi s s u bs er v i enc e
and to g i v e c o nf l i c t a neg ati v e c o nno tati o n.
The pr i v i l eg i ng of consensus and nal u r al i za-
ti o n of a constmcted worl d tended to hide ba-
si c c onf l i c ts and conceptual ize the ones that
did arise as based on m i s u nd er s tand i ng s , in-
complete knowledge, or pr ej u d i c e.
The other d i al og i c c r i ti q u e sug g ests i hat
i d enti ty was rel ativel y stabl e i n homog eneou s
societies and I hei r or g ani zati ons w i t h f ew
av ai l abl e discourses. I n contemporary, heter o-
g eneous, g l obal , tel econnected sociel ies and
g l obal i zati on the avail abl e discourses expand
g reatl y. Si nc e i d enl i ty i s a d i s c u r s i v e pr od u c -
ti on, i n thi s new s i t u at i o n the i n d i v i d u al ac -
q u i r es s o r n an y s i m u l taneo u s i d enti ti es
t hr o u g h dif ieren! c om peti ng discourses that
f r ag m entati o n i s v i r t u al l y i nev i tab l e (see
Deetz, 1995b; Gergen, 1991). As society be-
comes mor e f r ag m ented and /or v i r tu al , the
i d ent i t y- s t ab i l i zi ng f orces f or or g ani zal i ons as
wel l as peopl e are l o s t, Such a po s i t i o n sug -
gests the pos s i b i l i ty of tr em end ou s f r eed om
and o ppo r tu ni ty f or mar g i nal i zed g r ou ps and
suppressed aspects of each person to be c on-
c eptu al i zed and d i s c u s s ed i n mor e heterog e-
neou s societies and c haoti c or g ani zati ons. Bu t
the r n u l t i pl i c i t y of discourses can al s o lead to
w hat Gi d d ens (1991) called "ontol og i c al i ns e-
curities." Such i ns ec u r i ti es r eg ar di ng i d enti ty
can lead to str ateg i es that am to secure a "nor -
mal " i d enl i ty (see K n i g ht s & Morg an, 1991;
Kni g hts & W i l l mott, 1985, 1989). This lose
self is open to m ani pu l ati o n (s i nc e the stable
backg round of a d o m i nant r epr od u c ti v e d i s -
c ou r s e is w eakened ) and can be "jerked"
abou t i n the s ys tem, l ead i ng to a sense of ex-
citement and even "ecstasy" but al so can be
conversin pr one and easil y c ontr ol l ed by sys-
tem f orces (as in Bau d r i l l ar d 's c onc epti on of
s i m u l at i o n , 1988; Deetz, 1994d).
The c o nc epti o n of a f l u i d c onf l i c tu al i d en-
ti ty, however, cral es d i f f i c u l ti es i n develop-
i ng po l i ti c al ac ti on. Fl ax (1990), f or exampl e,
s hows the awkwar d pos i ti on i t leaves wom en
n . I f g end er i s treated as a social c ons tr u c ti on,
one c an s how that the d o m i n an t d i s c ou r s e i n
modern or g ani zati ons has pr od u c ed w om en
and thei r exper i enc e as m ar g i nal and
"other"that is, l aki ng al l Ihe neg ati v e terms
i n the l i n g u i s t i c s ys tem and d i s c ou r s e.
R i d d i ng society of s tr ong g end er ascr i pti ons
and g end er ed i d enti ti es m aki ng g ender irrel -
ev ant to wor ki s a m eani ng f u l ac ti v i ty to
pr ov i d e oppor tu ni ti es f or w om en. Bu t to ac -
c om pl i s h such a mov e i n the contemporary
s i tu aon r eq u i r es women to or g ani ze and
show that g end er is an i ssue across nearl y al l
social s i tu ati ons that is, to f ix a centered
Conceptual Foundo.-ons * 33
i d enti ty. The d i l emma i s hei g htened r eg ar d i ng
thei r experience, f or i f wom en's experiences
arise ou t of an essential d i f f er enc e, they c an-
not be denied as i m por tant and needing to be
taken i nto ac c ou nt, bu t to make the es s enl i al i s t
ar g u m ent of d i s t i n c t f em al e ex per i enc es d e-
nies social c ons tr u c ti oni s m and can easil y be
used to f u r ther s ti g mati ze wom en as "other"
in a society where m en hav e more resources.
Ironically, howev er , this is the type of deep
tensi n and i n a b i l i t y to devel op a s i ng l e c o-
her ent pos i ti on that, r ather than w eak en i n g
dialogic work, gives it its reason for being.
The Critique ofthe
Philosophy ofPresence
Normative social science, as well as most
of us in everyday l i f e, treats the presence of
objects as u npr ob l er nati c and bel ieves that
language is to represen! (re-present) these
thi ng s . W hen asked w hat s om ethi ng i s , we tr y
to d ef i ne i l and l i s t i ts es s enti al attr i bu tes .
Di al og i c s tudi es f i nd such a pos i ti on to be il -
l u s i onar y. Rather, the "elemenls" o f t h e wor l d
ar e f u nd am ental l y i nd eter m i nant and c an be-
c om e m any d i f f er ent d eter m i nant "objects"
thr ou g h d i f f er ent w ays of attend i ng to or en-
c ou nter i ng them . Li ng u i s ti c and no nl i ng u i s ti c
practices d i r ec t attenti on and means of en-
c o u nter i ng the "el ements " of or g ani zati ons ,
t hu s ar e central to "object" pr od u c ti on. Since
the "elements" of or g ani zati ons may be c on-
structed/expressed as m any d i f f er ent "ob-
jects," l i m i ted onl y by hu m an c r eati v i ty and
r ec o nf i g u r ati o n of past u nd er s t and i ng s , m ean-
i ng c an never be f i nal ; objects and m eani ng s
ar e al ways i nc ompl ete and open to r ed eter -
m i nati on. Many d i f f er ent, and f u n d am en t al l y
irresol vabl e, "objectivities" thu s exist i n org a-
ni zati onal l i f e and research. The appear anc e
of compl eteness and c l os u r e l ead s us to over-
look the pol i ti c s i n and of c ons tr u c ti on and the
possibil ities f or u nd er s tand i ng s hi d d c n behi nd
the apparent and obvi ous, thu s a par ti cul ar ob-
j ec v i ty may be privil eged.
Lang u ag e i s c entr al to the pr o d u c ti o n of
objects i n that i t provides the social /historical
34 * Theoretical and Methodological Issues
dis iinctions that provide u nity and difference.
Lang u ag e does not mirror the reality "out
thcrc" or people's mental states, b u t rather is a
way of altending to both the ins iders and ou t-
s iders providing them shape and character
(Shotler, 1993; Shotter & Gergen, 1994). Fur-
thcr, the s y s lems of differences or distinctions
his iorically held by langu age are not fixed bu t
melaphorical, f u ll of contradictions and in-
cons is iencies (Brown, 1990; Cooper & Bu r-
rell, 1988). Meaning, thu s , is not universal
and fixed, bu t precarious, fragmented, and s it-
u ated. Since Ihe research commu niry , like oth-
ers , can only escape this s itu ation throu gh dis -
tortion and clos u res , Ihe conceptual base of
research mu s t also be, als o already suggested,
local and emergen!.
Organizational commu nication researchers
have used thes e conceptions to deconstruct
objects of organizational U fe inclu ding the
bou nded concept of an organization and orga-
nizational rationality its elf (Mu mby &
Pu tnam, 1992). Perhaps among the mos t pro-
du ctive have been those s tu dy ing accou nting
practices . The bottom line, profit and los s , ex-
penses, and s o forth have no reality withou t
s pecific practices creating them (Miller &
O'Leary, 1987; Power, 1994). Others have
looked at knowledge and information
(Boland, 1987; Coombs et al., 1992). And
others yet repon practices (Sless, 1988) and
categories of people (Epstein, 1988). Each of
thes e s hows the conditions necessary for ob-
jecls to exis t in organizational life and opens
these objects to redetermination throu gh initi-
aling dis cu s s ions and negotiations of reality
thal were not possible as long as hidden domi-
nance held sway.
The Loss of Fowdaons
ana Master Narratives
Tradilionally , the power of any social posi-
lion has been gathered from its grou nding or
f ou ndalion. This grou nding could either be to
n mctaphy s ical f ou ndaons u ch as an exter-
nu l w orld in empiricis m, mental structures in
rationalis m, hu man natu ra in hu manis m, or
God in religinor a narrative, a s tory of his-
tory s u ch as Marxis m's class s truggle,
social Darwinis m's survival of the fittest, or
market economy's invis ible hand. Pos itions
based on such fou ndations and narratives are
made to seem secure and inevitable and not
opportunis tic or driven by advantage. Cer-
tainly , much normative organizational re-
search has been based on appeals to an "ob-
ject" world, hu man nature, or laws of conduct.
Critical research has a different fou ndational
appeal to qu alities of speech commu nities in
its morally gu ided commu nicative action.
Dialogic researchers are dis tinctly non- or
antifou ndational.
Again, like in the case of identity, dialogic
researchers take two dif f erent bu t compatible
stances in their critiqu e of grou ndings . First,
some arge that fou ndations and legitimating
narratives have alway s been a hoax. Appeals
to fou ndations have been u s ed (u s u ally u n-
knowingly ) to s upport a dominant view of the
world and its order. As feminis ts , for example,
arge following this pos ition, the historical
narrative has always been /iiVtory. Empiricists'
appeal to the nature of the externa! world cov-
ered up the forc of their own concepts (and
those borrowed from lite groups), methods,
tns tru ments , activities, and reports in con-
structing that world (Harding, 1991). Second,
dialogic researchers note the growing social
incredu lity toward narratives and fou ndational
moves. Lyotard (1984) showed the decline of
grand narratives of "spirit" and "emancipa-
tion." The proliferaon of options and grow-
ing political cy nicis m (or as tutenes s ) of Ihe
pu blic leads to a s u s picion of legitimating
moves. In Ly otard's sense perhaps all that is
left is local narratives that is, ad hoc and s t-
uated attempts at ju s tification withou t appeal-
ing to themes that organize the whole of life.
The concern with integrative narratives has
led to sensitive treatments of how stories in or-
ganizations connect to grand narratives and
how different ones have a more local, s itu a-
tional character (see Martin, 1990). Other re-
searchers have used this opening to display
the false certainty in the master narratives in
Conceptuof Fou ndations * 35
management (Calas & Smircich, 1991;
Ingers oll & Adams, 1986). Jehens on (1984),
for example, showed how narratives of "effec-
tiveness," "expertise," and "excellence" were
used to legitimize managerial control systems.
In one of my own s tu dies (Deetz, 1998), 1
show how narratives of "consultancy" and
"integrated solutions" enabled a dominant co-
alition to maintain control through a financia!
crisis in a professional service company.
Dialogic researchers do not see the decline
of foundations as necessarily leading to pos i-
tive outcomes. Certainly , the decline of fou n-
dations and grand narratives removes the pri-
mary prop of security and certainty that
dominant groups trade for s u bordinaton. Bu t
the replacement is not necessarily freedom
and political pos s iblity for marginalized
groups. Lyotard demonstrated the rise of
"performativity," which while developed as a
measure of means toward social ends be-
comes an end in its elf. The performativity
s tandard provides new forms of control not di-
rected by a vis in of society and social good
bu t s imply more produ ction and cons u mption
(see Crter & Jacks on, 1987). Many "quality"
programs evidence this. Certainly, the loss of
grand integrative narratives has not been
missed by management groups. One cou ld
easily say that the common conceptions of
corporate "vis ions " and "cultures" are strate-
gic local narrative cons tru ctions to provide the
integration and motivation in a pluralis tic so-
ciety formerly provided by the wider social
narratives that have passed away.
A difficu lty in dialogic res earch with the
loss of fou ndations , as in the concept of f rag-
mented identities, is how to genrate a politi-
cal s tance in regard to thes e developments .
Women have confronted this most directly in
debates over whether men and women have
dis tinctly different experiences grou nded in
biolgica] sex. Withou t a grounding, the bas is
fortarge-s cale political action is lacking, and
resistance to domination, even general domi-
nation, becomes local and s itu ationaf. If one
rejects an essentialist fou ndaon and believes
that more than local res is tance is needed, criti-
cal iheory may well provide the best remain-
ing option, but not w ithou t costs (see Fraser &
Nichols on, 1988).
The Knowledge/Power
Within dialogic writings , power is tieated
far differently from mos t other w ritings on or-
ganizations . Foucault (1977,1980, 1988) has
led many in s u gges ting that the "power" of in-
terest is not that w hich one possesses or ac-
qures (Clegg, 1989; Jermier, Knights , &
Nord, 1994). Such power is an ou tcome of
more f u ndamental power relations . Power re-
sides in the discursivo practices and forma-
tions themselves. For example, the discourse
that produces a "worker" both empowers and
disempowers the grou p of individu is pro-
duced throu gh this repres entaron. In par-
ticu lar his torical dis cours es , "workers" and
"managers" are produced out of the open "ele-
ments" of organizational lif e and s imu lta-
neou s ly provided w ith s olidarity and interes ts
as well as conflicts , material and s y mbolic re-
sources, and s elf-u nders tandings . Power thu s
resides in the demarcations and the systems of
discourse that produce and s u s tain s u ch
grou pings . U nions and managers mu tu ally
s us tain the other in their conflicts . It is not the
relative power of each that is of inters ! but
how the dis tinction is reproduced.
One of the mos t u s efu l terms entering into
organization s tu dies from Fou cau lt's w ork on
the knowledge/ power connection has been his
concept of "dis cipline." The demarcations de-
veloped in discourse provide forms of norma-
tive behavior. The combinado of training,
rou tines , s elf-s u rveillance, and experts pro-
vides resources for normalization, then disci-
pline (Deetz, 1998; Knights & Collins on,
1987; Townley, 1993). From s u ch a concep-
tion, normative research and the expertis e pro-
duced from it are cons idered to provide re-
sources for normalization and a veneer of
tru th for arbitrary and advantaging dis cu rs ive
practices (Hollway , 1984, 1991). Thcempha-
s is on dis s ens u s dis cou rs e in dialogic research
36 * Theoretical and Methodological Issues
is aimed al disrupting normalization and pro-
vides compeling power relations (Holmer-
Nadcsan, 1997; Knights, 1992; Trethewey,
In dialogic conceptions, linguistic and
nonlinguistic praclices are considered to open
a r elation to external elements (people and
world) and produce these elements in specific
ways. As discussed earlier, the referen! ("ele-
ments" of the world) has no specific charac-
ler; it is always determinable in more ways
than all ihe deter minations or objects that
nave been made of it through various historia
cal practices. To the extent that this "indeter-
min acy " is k n own (the "otherness" of ele-
ments shows), the domination present in any
system can be disrupted and objects de- and
redifferenated. To the extent that indeter-
minacy is rccognized, the possibility of self-
referentialily in a textual system is avoided.
Otherwise, the determinan! object produced
by the practices is referenced by the practices
and the system remains closed.
The presence of media and informaci n
systems increases the possibility of such clo-
sure and the lack of connection to the external
indeterminacy. The referent can disappear as
an y thin g more than another signa produced
object. Thus properly signs would only refer-
ence other signs; images would be images of
images. The system then becomes purely
self-r efer cntial or what Baudrillard calis a
sitnulation (see Deetz, 1994d, for an exam-
ple). In such a world, in Baudrillard's analy-
sis, signs, r ather than connecting us to the out-
side world and pr ov ding a temporary
deler mination, reference only linguistically
alr eady dctcrmined objectsthe "map" leads
us on ly to earlier "maps" of the world. The
"model" is secn as the thing and "model" be-
hav ior rcplaces responsive action. Signs reach
the struclural limit of representalion by refer-
encing on ly themselves with little relation to
uiiy outsidc or interior. In such a situation, a
particular f iction is not produced by a subject
in opposition to reality, but positions an imag-
inary world and subject in place of any real; it
has no opposite, no outside. Baudrillard
(1983) used ihe example of the difference be-
tween feigning and simulating an illn es s to
show the character of this dialogic representa-
tion: 'Teigning or dissimulation leaves the re-
ality principie intact; the difference is always
clear, it is only masked; whereas simulation
threatens the difference between 'truc' and
'false,1 between 'real' and 'maginary.' Since
the simulator produces 'truc' symptoms, is he
ill or not? He cannot be treated objectively ei-
ther as ill, or not ill" (p. 5).
Hochschild (1983) provided an organiza-
tional example of this (though from a theoreti-
cally differ ent position) in her description of
(he appropriateness of flight attendants' emo-
tions. Our traditkmal conceptions allow a
fairly simple distinction between "real" spon-
taneous emotions that arise in response to per-
ceived situations and "acting" where an em-
ployee fakes the managerially desired emo-
tion. Hochschild shows, however, that the
presence of "deep acting" makes this distinc-
ton misleading. In deep acting, the flight at-
tendants in her studies learn to perceive or at-
tend to the situation in such a way that the
managerially desired emotion spontaneously
arises in the employee. Is it fake or not? In the
concepts here, it is self-referential. The sys-
tem appears open and environmentally adap-
tive bul closes or maniplales the environment
in ways that the system adapts to the system
reproduced environment. This is not unlik e
normative research constructing the world us-
ing the concepts of the same theory it hopes to
Research as Resistance
and Indeterminacy
The role of dialogic research is very differ-
ent from more traditional roles assigned to so-
cial science in both its emphasis on dissensus
production and the local forms of knowledge,
It primarily serves to attempt to open up the
indeterminacy that modern social science, ev-
eryday conceptions, routines, and practices
nave closed off. The result is a kind of
antipositive (or positivist) knowledge that
Knights (1992) described. The primary meth-
ods are deconstmction, resistance readings,
and genealogy.
Deconstruction works primarily to critique
the philosophy of presence by recalling the
suppressed terms that have become devalued
in dominan! systems of distinction. When the
suppressed term is given valu, the depend-
ency of the positive term on the negative is re-
vealed and a third term is recovered that
shows a way of thin k in g or attending to the
world that is not dependen! on the opposition
of the first two (see Calas & Smircich, 1991;
Martin, 1990; Mumby, 1996; Mumby &
Putnam, 1992). The resistance reading dem-
nstrales the construclion activity and
problematizes any fixed relationship. The
positive and the polar constructions are both
displayed as acts of domination. Conflicts that
were suppressed by the positive are brought
back lo redecision (see Weslenholz, 1991).
The conflictual ield oul of which objects are
formed is recovered for crealive redeter-
minationconstan! dediffer entiation and
redifferentiation. Given the power ofcommon
sense and organizational routines, such re-
reads require rigor and imagination. The
rereadings are formed out of a keen sense of
irony, a serious p lay f u ln es s , and are often
guided by the pleasure ene has in being freed
from the dull compulsions of a world made
too easy and too conslr aining. The point of re-
search in this sense is not to gel it r ight but to
challenge gu idin g assumplions, fixed mean-
ings and relations, and reopen the formative
capacity of hu man beings in relation to others
and the world.
In looking at dif f er en t or gan ization al com-
munica ion research programs, clear ly difer-
Conceptual Fbundalions 4 37
ent programs have differ ent goals and as-
sumptions and provide dif f er en t for ms of
evaluation. I hope to hav e displayed differ -
ences that give insights into the diverse dis-
courses in or ganizational commu n ication
studies today, dis p lay in g some of the ways
ihat they are alike and differ ent. The relation
among these alternatives is not addressed
well in exclusionary, pluralistic, supplemen-
tary, or integrative terms, Each or ientation
crales a visin of social problems and tries
to address them. Difieren! or ientations have
specific ways of answer ing the types of ques-
lions ihey pose and do not work terribly well
in answer ing the questions of others.
I, like many others, sometimes wish we
were all multilingual, that we cernid move
across orientations with grace and ease, but
this type of Teflon-coated, multiperspectival
cosmopolitan envisioned by Morgan (1986)
or Hassard (1991) is both illusionary and
weak (see Parker & McHu gh, 1991). Good
scholars have deep commitments, Multiper -
speclivalism often leads to shallow readings
and invites unexamined basic assumptions.
Some scholars are more multilingual than oth-
ers, but doing good wor k within an orientation
still mus be prized first. Ideally, altemative
research programs can complement each
other. Consensus without dissensus is stifling
and f in ally maladaptive. Elite/a pr ior i con-
cepts are necessary and probably inevitable,
but we can make them more temporary and
open loreconfiguralion.
Wilhout a doubt, most organizational com-
munication scholars are becoming both more
knowledgeable aboul allernatives and more
appreciative of the differences. This develop-
ment allows us to get beyond relavey unpro-
ductive theoretical and methodological argu-
ments to more basic and serious questions.
The choice of orientation, to the extent that it
can be freed from tr ain in g and depart-
ment/disciplie politics, can probably be re-
duced to altemative conceptions of social
good and preferred ways of living. This ac-
ceptance grounds theory and method debate
in a moral debate that has been neither terribly
38 4 Theoretcal and Methodological Issues
common or explicit in organizational com-
munication studies. I agree with Gergen
(1992) that organizational research and theory
need to be evaluated as much by a question of
"how shall we live?" as by verisimilitude and
methodological rigor. Studies need to be un-
derstood and evaluated on their own terms,
bul should also appeal to the larger social con-
cerns in which both the needs and means of
accomplishment are contested.
Discussions of responsibility and valu are
still relatively infrequent in organizational
communication research, bul present (see var-
ious essays in Conrad, 1993; Deetz, 1995a;
Deetz, Cohn, & Edley, 1997). Certainly, we
nave lagged behind moral and ethical discus-
sions of organizaron available other places
(e.g., Frederick, 1986; Freeman, 1991; Free-
man & Liedtka, 1991; Gergen, 1995; Jackell,
1988; Maclntyre, 1984; Mangham, 1995).
The justication for much organizational
communication research has been aimed at
improving the functioning of organizations
and management as if they were vatue-neutral
tools without regarding how these tools are
applied or whose vales are advanced. With
such a conception, our research has often fo-
cused on the perfectibility of the tool ralher
than the ends it is used lo advance. To the ex-
tenl that this conception has been useful, orga-
nization studies have enhanced Ihe effecve
use of resources and fulfillmenl of certain hu-
man needs. But many researchers now ques-
lion this "tool" versin of organizations and
research, claiming that researchers paid insuf-
ficient altention to alternative needs and goals,
and the numerous social and political conse-
quences of organizational activities (see Mar-
sden, 1993). Until recently, most organiza-
tional communication researchers accepted a
managerial bias in their conceptions of orga-
nizations and articulations of organizational
The bustness environment has changed in
fundamental ways in the past two decades,
Thcse changcs require rethinking decisin
maki ng in corporalions: Who should make the
dccisions? How should they be made? What
criteria should be used to evalate them? If
companies are to stay economically viable
and their host societies healthy, corporate de-
cisions must be more responsive to rapidly
changing environments and human needs.
Understanding new vales and the rights
and capacities of other organizaonal mem-
bers is initiating reforms of organizational
communication research that are as sweeping
as many contemporary changes in organiza-
tonal Ufe. Certainly, this is seen to some ex-
tent in the growth of teams, other participation
programs, customer focus, and increased dis-
cussion of environmental and social responsi-
More importan! than these new programs,
in my mind however, is a growing shift in the
conception of organizations themselves. This
shift offers the greatest challenge and oppor-
tunity for organizational communication re-
searchers. Generally, the conceptual shift can
be characterized as moving from an "owner/
manager" model to a "stakeholder" model of
organizations (see Carroll, 1989; Deetz,
1995b; Freeman & Gilbert, 1988; Grunig &
Hunt, 1984; Osigweh, 1994). In this model, a
variety of groups in addition to stockholders
and managers are seen as having made an in-
vestmenl and thus having a stake in corporate
decisions. Proponents of such a view arge
that in a democratic society all those affected
by the activities of corporations (all stake-
holders) have some representaon rights. But
beyond the question of rights, direct deci-
sional influence by both internal and extemal
constituencies can lead to greater effective-
ness n meeting the diverse social and eco-
nomic goals. A stakeholder model recognizes
mltiple forms of ownership and enables
widespread participation and thus helps initi-
ate important valu debates.
In traditional models of organizations, Ihe
core processes in organizations were con-
ceived as economic. Communication aided
economic accomplishment, but wherever pos-
sible stakeholder representation was limited
to economic representation. If communica-
tion-based decisin making could be reduced
to an economic calculation, it was. In a stake-
holder model, the core processes involve sev-
eral simultaneous goals. The interaction
among stakeholders can be conceived as a
negotiative process aiding mutual goal ac-
complishment. Communication is the means
by which such negotiation takes place. Con-
ceptions of human interaction, negotiation,
and rationality developed by communication
theorists are uniquely suited to these new
needs. To make a full contribution, organiza-
tional communication researchers would need
to use communication conceptions aimed a(
increasing genuine participation rather than
increased influence and control. This change
is still incomplete.
Many organizaron managers understand
the need to attend to stakeholders today but
have not accepted a stakeholder model. New
communication and decision-making concep-
tions are often used to increase the number of
forums in which stakeholder representation
and debate could occur, but few have in-
creased stakeholder voice (Deetz, 1995b;
Deetz et al., 1997; Cordn, 1988). Attention
to stakeholders in these cases is a strategic at-
tempt to increase loyalty and commitment
and decrease resistance rather than seeking
genuine decisional input. The lack of voice re-
sults from constrained decisional contexts, in-
adequate or distorted information, socializa-
tion and colonization activities, and the
socitation of "consent" where stakeholders
"choose" to suppress their own needs and in-
ternal valu conflicts. Gradually, we are learn-
ing that the problem with traditional organiza-
tions was not smply bureaucracy, but control
systems in a variety of forras. To overeme
these problems, new conceptions of interac-
tion can improve collaborative decisin mak-
ing within corporations. The critical and
dialogic scholars were somewhat earlier in
fally appreciating these changes while man-
agers and managerial-biased researchers have
been more ambivalentoften both advocat-
mg new conceptions and programs and sub-
verting their full implementaon. But both
normative and interpretive researchers can de-
Concefrtua/ Poundotions *39
sign studies that enhance the functioning of
the organization as a site of stakeholder coor-
dination rather than a site of control. Finding
new ways of organizing becomes everyone's
Understanding our alternatives requires
Understanding both the relation of concep-
tions to the various social stakeholders and the
relation of research discourse to dominan! so-
cial theories. Thinking through Ihese relations
provides an opening for discussion. We are
learning the positive effects of human diver-
sity as organizational membersbeyond
"seprate but equal" and integrationand or-
ganizational communication research can
benefit from betler conceptual discussions of
research diversity. In doing so, the ultmate
point is not in arguing it out to get it right, but
to reclaim the suppressed tensions and con-
flicts among the many contemporary stake-
holders to negotiate a life together based in
appreciation of difference and responsive de-
cisin making.
1, Citations are selectivo throughout ihis essay.
Rather than try to be exhaustive and produce a cluttered
lexl with hundreds of references, 1 will refercnce what 1
consider lo be well illustrative orespeciallyuseful devel-
opments and will bias the selection loward authorc who
work Jn communication depaitments. This essay was
compleledin 1996. Cilations toliteraturepublishedafter
that time are more limited.
2. Much of Ihis discussion s adapted from Deeiz
Alexander, E., III. Penley, L.. & Jemigan, I. E. (1991).
The efiect of individua] differcnces on manager me-
dia choice. Management Communication Quanery,
5, 155-173.
Alien, B. J. (1996). Feminist standpoint theory: Ablack
woman's (re)view of organiza liona! socializa!ion.
Communication Sludies, 47, 257-271.