University of Southern California Dissertations and Theses

A critical analysis of ideology and discourse in two hotels

(USC Thesis Other)

A critical analysis of ideology and discourse in two hotels

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A CRITICAL ANALYSIS OF IDEOLOGY
AND DISCOURSE IN TWO HOTELS
by
Stephen P. Banks
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL '
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Communication)
July 1987
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90089
This dissertation, w ritten by
Stephen P. Banks
under the direction of h.hs Dissertation
Committee, and approved by all its members,
has been presented to and accepted by The
Graduate School, in partial fulfillm ent of re
quirements for the degree of
Ph.D.
C M
? S7
B 2 I8
33#/ B3,¥t
D O C TO R OF PHILOSOPH Y
Dean of Graduate Studies
DISSERTATION COMMITTEE
Chairperson
11
DEDICATION
This dissertation is dedicated to the memory of my late
brother, Christopher Allen Banks. Ule are all refugees in
the same uiar.
iii
ACKNOWLEDGMENTS
I am deeply grateful to the members of my dissertation
committee for their wise counsel, unflagging
encouragement, and good-humored patience. Eric
Eisenberg’s critical insights and enthusiasm for new ideas
have been models of innovative research and good
conversation for me in the past two years. Walt Fisher
has been a personal example of humane and disciplined
thinking; he has shown me countless ways to improve my
writing, although I don’t always come up to standard. The
reach and depth of Elinor Ochs’ scholarship have been
invaluable to me; probably more important to me has been
her steady personal support and guidance. I am especially
grateful to Patti Riley for encouraging me to explore
research interests that sometimes test disciplinary
boundaries and orthodoxy. All four of these individuals
honor me by bestowing the benefits of their instruction
and the pleasure of their friendship.
Among the many friends and colleagues who have
inspired me or cajoled me or set a positive example for me
when I needed a boost, I am particularly in the debt of
Joan Cashion, Ed Chao, Janet Fulk, Paul Kirk, Peter
Marston, Janet Metzger, Howard Oran, Ron Sept, Dolores
Tanno and, especially, Ruth Smith. My brother, Russell
Banks, was largely responsible for setting me on the track
I lv
toward academia some thirty years ago— "There at the
beginning, there in the middle, and there in the end."
Most important, I wish to thank Anne Holden for her
trust and confidence in me. She has been a sympathetic
sounding-board for my inchoate ideas, as well as a
companion and Friend of unexpected grace.
V
TABLE OF CONTENTS
page
DEDICATION..................................... ii
ACKNOWLEDGMENTS ....................................... iii
LIST OF TABLES AND FIGURES..................... vii
CHAPTER 1 .............................................. 1
Introduction................... 1
Literature Review ................................. El
Organizational Culture ........................ E4
Ethnography of Communication ................. 40
Ethnolinguistic Identity Research ............. 50
Language Attitude Research . . . ............. 56
Power in Organizations................... 5S
Career Trajectories ...................... . . . 64
Research Questions ................................. 65
CHAPTER E .............................................. 73
Sources of D a t a ................................... 73
Settings of the Data Sources................. 74
Rationale for the Setting ..... 81
Discourse Analysis ................................. 84
Bracketing Discourse .................... . . . 86
Recording and Transcribing Techniques ..... 37
Critical Theory Analysis .......................... 101
Critical Theory as Philosophy and Technique . . 10E
Comment on Method............. 114
CHAPTER 3 .................................................ISO
Four Principles of Power and Discourse............. 1E1
Meaning, Ideology, Power, and Discourse .... 1EE
Power is Social ............... 1E4
Discourse Enacts, Creates, and Recreates Power. 1E5
Multiple Features: The "Ulhat Gets Said" .... 1E6
Multiple Features: The "How of Saying" .... 136
Reference and Discourse .......................... 140
The Culture of Reference in Two Hotels............. 144
Reference and Hierarchy ........................ 145
Reference and Relationships .................... 153
Analysis of We/Us/Our . ............... 163
The Nature and Function of "We"..................171
Interpreting "We" in Discourse..................18E
An Illustration of FPPP Analysis................E01
N o t e s .................................................E15
vi
page
CHAPTER 4 .................................................E17
Structuration: A Foundational Social Theory . . . 219
The Individual and Agency........................ 224
Structure and Structuration .................... 229
Power, Domination, and Structure . ........... 233
Discourse, Meaning, and Ideology Revisited .... 236
A Critique of Three Rites ........................ 247
New Employee Orientations ...................... 249
Departmental Training Meetings ............... 263
Lewittalks............... 270
Summary of Rites Critique ...................... 276
Critique of Two Hotels . 270
Rational and Veiling Legitimation ............. 279
Distortions in Structural Properties ......... 204
Ideologies and Career Trajectories ........... 291
CHAPTER 5 .................................................290
Summary.............................................. 290
Research Implications ............................. 303
Organizational Culture ........................ 303
Power in Organizations...........................305
Critical Theory........... 307
Discourse Analysis ............................. 300
Ethnolinguistic Identity ...................... 309
Limitations.......................................... 309
Epilogue ................................. ..... 312
REFERENCES...............................................314
v i i
LIST OF TABLES AND FIGURES
page
TABLE 1. Reference Practices and Hierarchy ..... 159
TABLE E. Reference Practices and Relationships . . . 165
TABLE 3. Referential Domains of "We".............S00
FIGURE 1. System of Domination and Power ........... E35
FIGURE S. Relationships Among Structures and
Interaction in Time-Space ............... E36
FIGURE 3. Four Types of Distortions of
Organizational Rationality ............... E85
C h a p ter One 1
Not only may language play a role in obtaining
employment, but certain linguistic skills may
be a vital prerequisite for advancement and
higher income.
— Stanley Lieberson
Introduction
Organizational studies have became increasingly
concerned in recent years with questions of human
understanding and how sense-making influences
organizational life CCarbaugh, 1985; Eisenberg & Riley, in
press; Poole & IlcPhee, 1983; Putnam & Pacanowsky, 1983).
The concern for meanings held by members of organizations
has been particularly evident in the literature on
organizational culture Ce.g., Frost, floore, Louis,
Lundberg & Martin, 1985; Schein, 1985) and interpretive
and critical analyses of organizational communication
Ce.g., Deetz & Kersten, 1983; Gray, Bougon & Donnelon,
1985). When developing arguments to explain how sense-
making occurs in organizations, culture researchers
typically refer to the symbolic nature of interaction, and
they commonly call on natural language as a representative
symbolic system. Sypher, Applegate, and Sypher C1985, p.
81), for example, state: "The very core of the inter
disciplinary approach to organizational cultures
2
and culture analysis is communication— the understanding
of symbolic action," and Carbaugh C19B5, p. 36) notes that
cultural phenomena are seen by researchers as "constituted
in organizational discourse."
Although language is widely recognized as an
indispensable concept in theories of organizational sense-
making, crucial gaps exist in our understanding of how
language as it is employed in everyday interaction by
organizational members affects the careers of those
members. The gaps in understanding most in need of
filling are those that exist between established
theoretical principles and work done in the applied
context of real work organizations. For example, some
researchers have gone far in their search for explanations
of domination and control in organizations Ce.g., Conrad,
1903; Conrad & Ryan, 1985; Oeetz, 1905; Fowler, Hodge,
Kress & Trew, 1979), but they have not demonstrated by
appeal to data from real organizational interaction how
their concepts and principles account for the linguistic
behavior of actual workers and managers. Proponents of
critical theory, moreover, proclaim emancipation from
hegemony as a central goal of their approach, but the
pragmatics of emancipation— what concrete ways
relationships and behavior must change— are missing from
such work. Likewise, other scholars have produced studies
of situated interaction among organizational members
— ■ 3
Ce.g,, Anderson, 1986; Fisher, 1984; llehan, 1986;
Treichler, Frankel, Kramarae, Zoppi & Beckman, 1984)
without placing their work in a theoretic frame that
accounts for the relation of members’ linguistic behavior
to their organizational destinies.
Other gaps in understanding the sense-making
processes in work organizations involve difficulties in
the development of theoretical concepts. Gareth Morgan,
for example, asserts that scholars need better
appreciation of the link between conscious and unconscious
aspects of symbolism, and to understand "how much of
everyday organizational practice may often be rooted in
the unconscious" Cflorgan, 1985, p. 29). More
pragmatically, William O ’Barr maintains that the model of
speaker selection and floor allocation in routine
conversations can be manipulated "politically" to
establish or promote dominance in interpersonal
relationships CO’Barr, 1984). He asks what other
conversational practices might be manipulated to maintain
dominance patterns. Such practices will rise to the
surface as scholars begin looking at the relationship
between meaning construction and dominance in work
organizations. In addition, 0’Barr’s essay exemplifies
the need to present evidence from actual interaction in
work organizations that tests his ideas and identifies
other ways that linguistic behavior and power are related.
4
Justification oF the Studu
This dissertation presents a project that
demonstrates houi the routine use of language by
organizational members affects their worklife
relationships, influences their career opportunities, and
constrains the discourse possibilities open to them. It
focuses on the ways language is mobilized in the day-to-
day talk of managers and employees. More importantly, it
takes as its central social concern the ways that routine
linguistic practices of powerful members of work
organizations influence the hierarchical positioning and
qualities of worklife of ethnolinguistic minorities. I
refer here to the members of work organizations whose
first language is not the variety CHudson, 1980) that is
treated as a standard in the workplace. Typically,
ethnolinguistic minorities are workers whose first
language is a different language altogether from that
spoken in the workplace; occasionally it is a region- or
class-based dialect that differs noticeably From the
standard Csee, e.g., D’AngleJan & Tucker, 1973; Fishman,
1986a; Trudgill, 1983).
Two major Factors lie behind my interest in the role
language plays in the careers of ethnolinguistic
minorities. First, and perhaps most important, is the
recent appearance of studies that open up the heretofore
ignored relation between discourse and social control in
--------------------------------------------------------------- — — — -----------------------------------------------------------------------------------------------5
the context of work organizations Ce.g., Di Pietro, 1980;
Fisher & Todd, 1986; Hodge, Kress & Jones, 1979;
Kramarae, Schulz & O’Barr, 1984; Hey, 1985; Shapiro,
1984). The now well-established principle that individual
and group identity is enmeshed in language practices
CGiles, 1979; Biles & Johnson, 1981; Gumperz, 1985a; Lamy,
1979) implies that ethnolinguistic minorities are less
equipped to cope with the control and dominance aspects of
discourse than employees whose first language is the
standard. Not only might minorities be less sensitive to
the subtle cues or strategies of discourse that result in
social control, but they also occasionally struggle with
challenges to their sense of personal and group identity
based on having to make choices of language variety in the
work setting CBourhis, 1979).
Clearly, obvious structural barriers to equality of
job opportunity, such as employment selection procedures
and pay practices, have been largely overcame. It is
likely, however, that such barriers are overcome Just
because of their obviousness. More subtle, covert, or
unconscious factors that perpetuate discrimination,
inequality, and dominance have not been systematically
addressed by government bodies or employers. Likewise, it
is passible that such practices continue just because of
their subtlety; and part of their "unseen" nature
6
conceivably is attributable to their being ingrained in
the normalized, day-to-day discourse routines of
organizational members.
Critical studies in linguistics, communication,
political science, sociology, and history have advanced
awareness and understanding of the more subtle aspects of
power, dominance, and language within the academic
community. Now discourse data from actual organizations
need to be conjoined to critical theorizing. And the
discourse data need to be analyzed not only in terms of
dominance and discrimination patterns, but also within a
more general theoretical framework that can account for
the production, reproduction, and change of social
structures (Riley, 1985, p. 49).
A second motivation for this study is the insistent
calls by researchers in organizational communication to
learn more about the link between the use of language and
the architecture of power in work organizations. Gray,
Bougon, and Donnelon (1985), for example, say that
"further articulation and demonstration of how powerful
members construct meaning for others is needed" (p. 94).
Stanley Deetz, discussing organizational culture studies,
states that "a fine Job" is being done of describing how
work is accomplished and how such forms of discourse as
stories and myths arise and are used in organizations.
But such studies, he says, have generally failed "to
7
consider how these same cultural elements hide
contradictions, distort the experience and expression of
member interests, and create and sustain false consensus"
CDeetz, 1905, p. 123).
The principal form of these cultural elements is talk
CStubbs, 1903), and "talk-in-use" is a principal
manifestation of organizational power CEisenberg & Riley,
in press). Consequently, Deetz’s complaint must be
directed more pointedly at the talk that goes on in work
organizations and particularly at the talk that involves
less fully competent users of the language variety which
stands as the local standard. More generally, Karl Uleick
cites the need for establishing explicit connections
between talk and action— on the assumption that talk is
not a form of action— as a plea for research efforts that
link linguistic and other behaviors of individuals and
groups in organizations CUJeick, 1903).
Two subsidiary arguments support the general impetus
to study the intersection of language, power, and the
worklives of ethnolinguist minorities. One is the
practical matter of the expanding multiethnicity in the
workforces of the developed nations. In the United
States, the situation is no longer a matter of accreting
to existing workforces new members from northern Europe
and Canada, members whose linguistic and social
experiences favor quick assimilation. On the contrary,
a
significant immigration gains in the U.S. noui come from
MesoAmerica and Africa, as well as Southeast Asia CGlazer,
1983, p. 323). In 1980, 18,W i ,000 U.S. residents had a
home language other than English CDolson, 1985, p. 135),
and such ethnolinguistic diversity may be expected to
persist in the future CFishman, Gertner, Lowy & Milan,
1986, p. 269). Glazer C1983, p. 317) contends that the
new influx of workers into the American workplace will be
neither easily nor soon assimilated precisely because of
the expanding pattern of multiethnicity.
This pattern, moreover, is not limited to experience
in the U.S. Changing immigration trends in Great Britain,
for example, have resulted in a new "multiethnic and
multilingual" workplace there CJupp, Roberts & Cook-
Gumperz, 1982). Besides the influence of changing
immigration trends, the workplaces of developed nations
also are contending with contributions to a new
multiethnicity resulting from what Joshua Fishman has
called the "ethnic revival" of the recent decade CFishman,
1986b, p. 171). Thus Heller C19B2) writes about the
politicizing of conversation in Montreal since the rise of
Francophone awareness and the dissolution of old
sociolinguistic norms in the past twenty years. Now a
process of negotiating language choice takes place at the
outset of most conversations between nonintimates in
Montreal, and the negotiation requirement is not limited
--------- _ _ _ g
to situations where an English speaker is interacting with
a Francophone: degrees of intimacy or Formality are often
negotiated by reciprocal choices of the French personal
pronouns, tu/vous.
The second subsidiary argument centers on
technology. Technological innovations in the workplace
are associated with changes in workers’ use of language in
at least three ways. First, as technology becomes more
complex there is an apparent increase in lexical
innovation practiced by the users of the technology
CChisholm, 1986; Evered, 1983; Mathiassen & Andersen,
1986). Second, the introduction of computerized
communication systems into the workplace demands higher
literacy and other language skills from workers (Kerr &
Kiltz, 1983). Computer-mediated work processes also may
increase the amount of interaction between workers and
change social networks COlson & Lucas Jr., 1982). Third,
the ongoing process of deindustrial-ization transfers
employment to the service sector of the work world with
concomitant increases in interpersonal interaction
(Forester, 1981; Goldwyn, 1981; Sadler, 1981). The rate
of deindustrialization is expected to increase, and as a
result "information operatives" will become the dominant
Form of labor in the developed countries CStonier, 1981).
. - 10
Description of the Studu
These five major and subsidiary reasons— Ca) gaps
between theory and applied studies of language use in work
organizations, Cb) lack of understanding about how
discourse and power are related, Cc) calls by
organizational researchers to explore the ways that
meaning might be implicated in various aspects of
organizational life, (d) the rapid growth of
multiethnicity in the workforce, and Ce) the inexorable
shift toward technological complexity— stand as the
rationale for this project. This study places the issues
of language use and asymmetrical work relationships in a
framework of organizational culture. It employs discourse
analysis to critique members’ use of language and open up
the nature of power relations that might be submerged in
the web of cultural practices of the organizations under
study.
Two hotels in a high quality, international corporate
chain provided the data analyzed here. Samples of
discourse were obtained from three types of ritual
interactions CTrice, 1984; Trice & Beyer, 19B5; Turner,
1969), which, fallowing Trice’s usage, I will call rites:
new employee orientation sessions, training meetings, and
"Lewittalks," a cover term I use to refer to a species of
meeting wherein the general manager of the hotel solicits
views from non-managerial employees in a private,
n _
confidential conference. The discourse in these rites is
then critically analyzed, primarily using Giddens’ theory
of structuration CGiddens, 1979, 1981, 1984}, and
conclusions are drawn about the nature of power relations
in the hotels and the effects of discourse on the career
trajectories of ethnolinguistic minorities in the
workforce.
This introductory description generates two key
issues that help account for the approach and the
structure of this dissertation: Why culture? and Why
language? The first question brings up an additional
challenge that needs clarification: If one is to better
understand the relationship between organizational
processes and the career chances of minority workers, why
not study the structural barriers in organizational life
with a functionalist theoretical orientation?
Rationale for culture.
In line with Uleick’s C19B5, p.388) observation, I
take organizational culture to be centrally concerned with
meanings held more or less in common by members. Research
that takes into account subjectively held meanings
immediately calls up the necessity to deal with action,
the knowledgeable member CGiddens, 1979, 1981), and an
ethnomethodological approach to experience in the
organization. In his critique of Durkheim’s view of
structure, Giddens C1979, pp. 50-52) describes the
IE
difficulties encountered when the object is favored over
the subject and constraint is conceived as a property of
objective social structures. Giddens generally favors
opening up the interaction between object and subject,
between agent and social institution: "ClIn social theory,
the notions of action and structure presuppose one
another" CGiddens, 1979, p. 53).
Similarly, Linda Putnam C1983) contrasts the
characteristics of functionalist research in
organizational communication with the approach she calls
"interpretivist." She points out that functionalists take
social phenomena as objective facts constrained by pre
existing structures which typically are explained within a
nomological view of research. Interpretivists, on the
other hand, see social phenomena as emerging from social
interaction of members whose relationships constitute
social structures insofar as they are given meanings
subjectively.
Putnam describes two interpretive research models,
the naturalistic and the critical. Naturalistic research
attempts to describe organizational phenomena without an
agenda for change and assumes that meanings are
consensually held by members. Critical research takes a
j judgmental stance on the social structures of
i
organizational relationships by critiquing the
subjectively held meanings of members and the underlying
processes that lead to those meanings CKersten, 1986).____
_____ _ 13
The present study assumes that meanings held by
members are not necessarily consensual, although they
might be more or less held in common at times. Meanings
are generally thought of as windows into the members’
worklives and relationships. Further, by understanding
the potential meanings of crucial interactions,
researchers may extirpate the deep structural levels of
organizational realities and examine the sedimented
structures of power CConrad, 1983). A critical-
interpretive view thus aligns more closely with the
objectives of this study than would a naturalistic
approach or a Functionalist perspective as Putnam
describes them. The degree to which interpretive and
critical theory and methods coincide is discussed more
Fully in Chapter 2.
Culture is appealed to here as a concept For
understand-ing organizational relationships for at least
three reasons. First, in spite of considerable emphasis
on culture as a manipulable variable in organizational
life Ce.g., Deal & Kennedy, 1982; MitroFF & Kilmann,
1985), nearly all organizational culture theory accords a
central role to subjective meaning. Louis, For example,
states that "interpretation is a Fundamental issue in
studying culture in work settings. Interpretive processes
are Fundamental to both the phenomena under study and the
process of studying those phenomena" C19B5, p. 88).
14
Culture more broadly considered, as in
anthropological uses of the term, has traditionally
focused on native views of social life Ce.g., Seertz,
1973; Turner, 1967, 1969). Thus Smircich C19B5, p.65)
urges that organizational analysts see themselves as
anthropologists and argues that researchers who study
organizations should "examine not only ’organization’ but
also the cosmology and metaphorical conventions underlying
language and action."
A third reason for appealing to culture as a
framework for understanding organizational processes is
the presence of a well-established body of research
containing theoretic approaches that bridge individual
behavior and macrosociological categories CSypher,
Applegate & Sypher, 19B5). Culture, as Meryl Louis C19B5,
p. EB) says, is "an interweaving of individual and
community." In addition, culture supplies a vocabulary
for discussing such concerns as meaning, subjective
experience, and symbolism within organizational contexts.
Pertinent examples of this research tradition are
described later in the literature review section, but it
needs to be made clear that the concept of culture is
useful as a heuristic that facilitates examination of
language practices and associated social relationships in
workplaces and not as a portrayal of any organizational
entity or as an end in itself.
---------— .................... 15
Rationale for language.
Van Maanen calls For a "multivocal" approach to the
study of organizational symbols, meaning that symbols of
many different systems should be analyzed conjointly and
simultaneously C1985, p. 119). One reason for not heeding
that call Just yet is that no one symbol system in
organizational settings has been studied sufficiently to
undertake an integration with other symbol systems. Such
an objection, of course, assumes that symbol systems are
bounded and separable. That probably is not the case,
even For an elementary semiotic analysis of organizational
life: In S/2. Barthes C1974, pp. 4-9) refers to the
polysemous nature of all texts, and the very possibility
he holds out for layers of metalanguages CBarthes, I960)
argues toward a concept of organizational life as a
semiotic system with many farms. Nonetheless, language is
a very obvious and salient correlate to any integrated
symbolic system CWittgenstein, 1972), and its relation to
institutional power is still poorly understood CQ’Barr,
1984).
At least four other points argue in favor of studying
language first, forestalling an integration with other
aspects of culture until the place of language is better
understood. First is the often noted pre-eminence of
language as a code Csee Carbaugh, 1985, pp. 30-33) for
penetrating the nature of culture CLouis, 1985; Morgan,
IE-
1985, p. 29). Second, the study of language use engages
the researcher in day-to-day practices of a group and
culture CGeertz, 1973, p. 30). Third, and more
particularly, language use is centrally involved in
understanding power and domination in social
institutions. Foucault, for example, describes power as a
set of manifold relations that constitute institutions;
these relations of power, he says, "cannot themselves be
established, consolidated nor implemented without the
production, accumulation, circulation, and functioning of
a discourse" C1980, p. 93). Other views on power, such as
that of Dale Spender C198Ht), hold that power is
constituted by language use. Finally, starting with the
study of language-in-use is sensible because a
satisfactory set of research practices and logics have
been developed to study the relationships between language
and social variables in other contexts Csee, for examples,
the works discussed below in the subsections on
sociolinguistics and language attitudes).
Language versus discourse.
Discourse as a technical term needs to be
distinguished from language. In this study, discourse
refers to language embedded in use structures above the
sentential level and evident in actual, situated
interaction. Uliddowson C1973) distinguishes discourse
from "text," "sentence," and "utterance" on the basis of
17
the presence or absence of + situated use and +
suprasententiality. Thus-.
C- suprasententialD, C- use! = sentence
C + suprasententiall, C- use! ® text
C- suprasententiall, C+ use! = utterance
C+ suprasententiall, C+ usel = discourse
The mere fact that discourse is bracketed at a level above
the sentence, however, does not mean that analysis is
limited to discussing units of language at that level. An
appropriate item of interest might range from a phoneme to
many sentences or turns of talk. Ulhat is relevant about
bracketing discourse at this level is what it says about
the interactional context. Understanding what the item
under analysis is doing far the interactants requires a
view of their use of language at some remove on both sides
of the linguistic environment of the item; the objective
is to understand the meanings of interactants, to analyze
the patterns of their use of language, and to discover how
those patterns play out in asymmetrical power
relationships.
A key distinction that comes into play here also is
contributed by Uliddowson C1979, p. 711. A static view of
linguistic analysis is based on describing rules of well-
formed sentences, treating sentences as agentless objects
Ce.g., Chomsky, 1957, 1965). Uiddowson calls this
approach, and the tradition of textlinguistics, a
1H_
discourse-as-product view. When human intention and
agency is interjected into the concept of linguistic
analysis, a dynamic, discourse-as-process results.
Analysis in the latter view aims at the purposes and
meanings of linguistic data, including the purposes and
meanings oF the interaction to the participants.
When taking a primary interest in the aims and
understandings of discourse producers, the analyst must be
mindful oF what Brown and Yule C19B3, p. 26) call "doing
pragmatics." By this they mean that the contexts oF
interaction are crucial to the work to be done. Contexts
that inFluence the work are Ca) the discourse context,
including the linguistic environment per se; Cb) the
social context, including socio-cultural knowledge and the
relationship oF the participants; and Cc) the "expanding"
context CBrown & Yule, 19B3, p. 50), or the metarules For
understanding and contextualizing reFerences to space,
time, and identity. The speciFic techniques oF discourse
analysis employed in this study are discussed in detail in
Chapter 3.
Power and asummetrical relationships.
I do not take power to be identical with asymmetrical
relationships, because relationships can be diFFerent
without the constitutive asymmetries leading to one
person’s control aver another. People in work teams, For
example, can Fill diFFerent roles at the same
................. j-g-
arganlzational level and not have a positional authority
asymmetry; they also might be of diFFerent religions,
different ethnicity, or different geographic origins
without entailing asymmetries of power in their
relationship. These kinds of differences may be called
innocuous asymmetries.
Giddens C1984) describes power as created in and
through the structures of domination of an organization.
By that he means that power is axiomatic with the control
of allocative resources— the material aspects, the
procedures and techniques, and the products of
organizational work— and authoritative resources— the
organization of social and personal aspects of
organizational relationships. Thus the asymmetries that
count with respect to power are those that potentially
lead to domination by virtue of control over physical,
procedural, social, or other resources, all of which can
influence an individual’s understanding of the world
expectations of what is normative.
When asymmetrical relationships are discussed in this
study, I will refer to the paradigm case of structures of
domination in organizations— differences in the positional
hierarchy. Because power, privilege, prestige, and level
of compensation increase with the ever-narrowing ascent up
the corporate hierarchy, there is an inherent gradient on
these bases that differentiates among members of the
— .................................................................... EU ..
organization. Persons lower in the positional hierarchy
are with rare exceptions ipso facto lower in power,
privilege, etc. The gradient results not only from the
legitimate authority that attaches to the positional role,
but it also is generated by an wide array of implications
that attach to the social relationships and the givenness
of a situated context into which members enter when they
are employed. Differences of position in the formal
hierarchy, however, remains the most obvious indicator of
power asymmetries in an organization.
I should emphasize at this point that differences of
power per se are not axiomatically suspect and that
structures of domination are not illegitimate as such.
Organizing as a human activity is not possible if all
members of a social collectivity must have identical
authorities, responsibilities, and competencies. In
modern work organizatias, differentiation of task-roles
and integration through managerial structures are
unavoidable because of the demands of technological
complexities, both in production and in the marketplace.
Power and domination became unacceptable, houiBver, when
they prevent individuals from being able to choose to
participate in decisions or other administrative processes
that affect the ways they perform their own work; when
they present unwarranted obstacles to equality of
opportunity for workers to develop their careers as they
E l
wish; or when they erode workers’ personal dignity by
subjecting them to illegitimate and unwarranted control by
other persons.
Literature Review
An investigation of discourse and power relationships
in organizations necessarily reaches into several distinct
research traditions. In the first place it must accord a
central place to studies of institutional power and power
as carried by, reflected in, or constituted by language.
With a specific concern for the careers of ethnolinguistic
minorities, however, it is vitally important to explore
the implications of discourse for individual and group
identity. At the same time, it is important to heed John
Edwards’ caution about the complexity of this
relationship:
Questions of language and identity are extremely
complex. The essence of the terms themselves is open
to discussion and, consequently, consideration of
their relationship is fraught with difficulties.
This is not to say that treatments of the theme are
few; indeed, relationship has been dealt with in a
number of disciplines. This in fact is part of the
difficulty— there has been much isolated discussion,
and some reinventing of the wheel too C1985; p. 1).
A fundamental premise of this study is that identity
is grounded in the meanings actors have for their social
world and for their own experience and physical selves.
Berger and Luckmann C1967) discuss the grounding of
identity as a dialectic between social process and
individual humans’ corporeal selves, a dialectic that
ee
plays itself out continuously from infant socialization
through old age. Two aspects of the dialectic account for
the dual constraints of society and biology on every
individual and for the dual possibilities of personal
growth and societal change.
Externally, it is a dialectic between the individual
animal and the social world. Internally, it is a dia
lectic between the individual’s biological substratum
and his socially produced identity.... In the external
aspect, it is still possible to say that the organism
posits limits to what is socially possible....
Biological factors limit the range of social
possibilities open to any individual, but the social
world, which is pre-existent to any individual, in
its turn imposes limits on what is biologically
passible to the organism. The dialectic manifests
itself in the mutual limitation of organism and
society Cpp. 1B0-1B11.
Internally, the dialectic manifests itself as the
biological individual resisting the structuring of social
influences, particularly in terms of early childhood
socialization. Although Giddens C19791 effectively
challenges the concept of internalization, it is
undeniable that there is a reciprocal interaction between
individuals’ creative shaping of their worlds and the
social worlds’ pressure to constrain individuals’ behavior
and beliefs. As Berger and Luckmann describe it:
Man Csicl is biologically predestined to construct
and to inhabit a world with others. This world
becomes for him the dominant and definitive reality.
Its limits are set by nature, but once constructed,
this world acts back upon nature. In the dialectic
between nature and the socially constructed world the
human organism itself is transformed. In this same
dialectic man produces reality and thereby produces
himself C1967, p.1831. ___________________ _
8 3
Hast important for an appreciation of the role of
language and identity is Berger and Luckmann’s view of how
knowledge is acquired by individuals. They point out that
language typifies experience, enables objectivations by
individuals, and allows participation in mundane living:
"Everyday life is, above all, life with and by means of
the language I share with my fellawmen. An understanding
of language is thus essential for any understanding of the
reality of everyday life" Cp. 373. Moreover, language is
the means of typifying and objectivating one's self
through interaction with others and through listening to
oneself Cpp. 50-59). Thus language is implicated in the
"acting back" upon the individual and his or her identity
by the constructed social world.
Besides drawing on the literatures of institutional
power and identity, this study also must reach out to
several other disciplines to specify the nature of the
problem in a cogent and productive way. The following
sections discuss pertinent work in five research
traditions. No one research program or academic discipline
fully explores the problems of how organizational
relationships and discourse might be related; however, by
aggregating selected work from a range of traditions a
useful starting position can be established. The five
research areas discussed are: Ca) studies of
organizational culture; (b) the ethnography of
2 4
communication; Cc) social psychological studies of
ethnolinguistic identity; Cd) the largely athearetical
communication projects on language and speech stereotyping
Caften referred to as "language attitude research"); and
Ce) studies of power in work organizations and other
institutional settings. Major views and trends in
research on careers are discussed at the end of these
sections to inform later analysis of the impact of power
relationships on workers’ careers.
Organizational Culture
Researchers who have analyzed organizations as
cultures or as multiple cultures CGregory, 1993) offer a
globalizing concept for addressing many of the problems of
concern in this study. "Culture" is conceived of
differently by different commentators, and Smircich
C1903a) provides a taxonomy of the major views of culture
in organizational studies. First is the view that aligns
culture with country or national origin, which Smircich
labels the "cross-cultural" view. Representative of this
approach is work that compares practices in typical
organizations in one country to those of another country,
such as the recent spate of books about the Japanese style
of management and organizing CJohnson & Ouichi, 1974;
Munchus, 1983; Ouichi, 1981; Pascals & Athos, 1981;
Schein, 1981).
Smircich's second category is similar to the First in
that both treat culture as a variable that can be
manipulated by people. The second view, "corporate
culture research," takes its name from the book by Deal
and Kennedy, which exemplifies this approach. Here,
culture is the cumulated symbolic artifacts, including
stories, myths, and ways of speaking that are peculiar to
a particular organization. This view is evident in the
work of Ralph Kilmann CKilmann, 198S, 1985; nitroff &
Kilmann, 1985) and Edgar Schein CSchein, 1983, 1984,
1985).
The latter three views identified by Smircich are
said to conceive of culture generally as a "root metaphor"
for organization. By this term is meant that culture is
not something an organization has but something it is.
Thus, the third category, "organizational cognition,"
takes culture as the set of cognitive structures members
hold in common as rules for sanctioned organizational
behaviors. The fourth view, "organizational symbolism,"
analyzes culture as that which produces patterns of
discourse that typify membership in an organization. And
the fifth view focuses on culture as the manifestations of
"unconscious psychological processes." Representative
research taking each of the last three approaches to
culture, respectively, are: Schall’s C1983) investigation
of normative rules and Dandridge’s C19B5) essay on the
26
life stages of a symbol, reflecting the third view;
Trujillo’s C1983) analysis of talk at Polito Dodge as a
"performance," exemplifying view four, even though the
study is not presented as organizational culture research;
and Krefting and Frost’s C19B5) research into the
metaphorical language of managers, reflecting the
unconscious psquestions that can be asked of organizations
and the type of claims that may be advanced about
findings. In addition, the taxonomy organizes haphazardly
developing branches of culture research in organizational
studies, providing thereby a framework for newer
researchers to use when orienting their own work.
questia
ns that can be asked of organizations and the type of
claims that may be advanced about findings. In addition,
the taxonomy organizes haphazardly developing branches of
culture research in organizational studies, providing
thereby a framework for newer researchers to use when
orienting their own work.
On balance, the taxonomy is not grounded in
fundamental differences of theory or epistemological views
and accordingly has an ad hoc quality about it. Moreover,
the commonalities among the five views of culture are
conceivably more important than their differences. All
the approaches to culture take symbolic systems CEco,
--------- 27
1976, p. 3B-3B and, especially, Chapter 2) as central in
defining concepts; all the approaches take culture as
being embedded in historical frameworks, even if it is not
expressly stated; all the approaches are grounded in
organizational members’ personal understandings of their
social experience; and none of the approaches precludes a
treatment of organizational settings as having a
localized, special nature by virtue of its culture.
Perhaps more important is the difficulty of assigning
many organizational culture studies to only one category.
For example, Sudykunst and his associates CGudykunst,
Steward & Ting-Toomey, 1985) have recently expanded their
work on the social psychology of intercultural contact
into work organizations. An innovative chapter by Ting-
Toomey explores possible connections between culture and
conflict CTing-Toomey, 1985). Her work, however, is both
cross-cultural, in that she embraces culture in the sense
of Edward T. Hall’s "high-context and low-context
cultures," and corporate cultural research, in that she
focuses her analysis on work organizations and bases it on
symbolic codes brought into play in such processes as
bargaining and conflict management styles. In addition,
she develops propositions predicting members’ behavior
based on what they ostensibly have in their own minds in
terms of cultural prescriptions; this orientation seems to
place the study in the third category by Smircich, the
28
organizational cognition view. Krefting and Frost’s
(1985) study, already mentioned as a representative of the
Fifth category, also shows strong elements of the fourth
category— organizational symbolism— because of its close
analysis of metaphors in observed discourse.
Sypher, Applegate, and Sypher (1985) clearly
differentiate functionalist and interpretivist approaches
to culture and communication research. Putnam (1983)
thoroughly discusses the many ways functionalist and
interpretivist theory and research diverge; however, as
Sypher et al. point out, she does not account for the
strong presence of Functionalist writers in the
organizational culture literature. The distinction is
important, as the functionalist view is strongly
represented in research, and it has quite different
assumptions From the interpretivist view. It is thus
necessary to look at the leading examples of functionalist
research before looking beyond them for perspectives to
inform the present study.
Functionalist views of culture.
Among the best known work on organizational culture
in the functionalist perspective are popular books by Deal
and Kennedy (1982) and Peters and Waterman (1982). Both
of these works, while compellingly illustrated with
examples From contemporary organizational life, are
nonetheless short on theoretical foundations and rigorous
29
methods of inquiry. Their value is in their suggestion
that something gestalten-like contributes heavily to the
quality of life and effectiveness of work organizations
and that it involves the symbolic order of things at the
workplace.
A much more disciplined offering is found in Edgar
Schein’s book, Organizational Culture and Leadership
CSchein, 1985). Schein defines culture as:
a pattern of basic assumptions— invented, discovered,
or developed by a given group as it learns to cope
uiith its problems of external adaptation and internal
integration— that has worked well enough to be consid
ered valid and, therefore, to be taught to new
members as the correct way to perceive, think, and
feel in relation to those problems. Because such
assumptions have worked repeatedly, they are likely
to be taken for granted and to have dropped out of
awareness Cp. 9).
This definition tells a great deal about Schein’s view of
culture. It is first of all a cognitive construct that is
somehow transmittable to individuals entering the
organization; it is created by group members for the
purpose of problem-solving; but insofar as it involves
members’ thinking, perceiving, and feeling, it is not
behavioral. Schein goes on to explain that the patterns
of assumptions are easily taken for granted and become
submerged from awareness. Overt behavior arises from the
interaction of situational contingencies in the external
environment and the "cultural predisposition" Cp. 9).
30
Schein envisions culture as evident at three levels:
artifacts and creations, values, and basic assumptions.
Artifacts include technology, artistic productions,
language, and all behaviors of members. He highlights the
difficulties in analyzing artifactual evidence of culture
by pointing to the fact that what constitutes an artifact
could be anything that interests the researcher. A
further problem is positing what any artifact means, how
it relates to other artifacts, and what deeper patterns of
culture might be implicated in it. Schein’s idea of
values is the member’s sense of "what ought to be." But
there is often a difference between what any member says
are prevailing values and what values actually drive the
social behavior of the organization. Thus, it is
important, he warns, to be sure espoused values are
congruent with the underlying assumptions. Such taken for
granted, "preconscious" assumptions concern the nature of
reality, time, space, human activity, human nature, and
human relationships, including one’s relationship to the
external environment. For example, assumptions about the
nature of human activity might contrast a proactive,
"doing" orientation to a reactive, "being" orientation, or
to a harmonizing, "being-in-becoming" orientation.
The basic method for uncovering these assumptions is
an "iterative interviewing" process Cp 113). Schein
proposes that developed hypotheses of the underlying
31
assumptions be confirmed with key insiders Following a
period of sensitization to the artifacts of the
organization. In addition to Using critical incident
questions, the researcher may use formal documents of the
organization, analyze stories, myths, and legends, and
conduct group interviews directed at the content of the
underlying assumptions.
The basic incongruity in this thesis is its reliance
on surface manifestations of a culture that is conceived
fundamentally as a cognitive construct. Behavior,
ironically, is viewed as problematic because it cannot be
determined if behavior is generated by culture or by non-
culture factors: "when we observe behavior regularities,
we do not know whether we are dealing with a cultural
artifact or not" Cp. 9), This apparent paradox
notwithstanding, Schein’s disciplined approach to
gathering and using data From members is a laudable
example to follow. His techniques are systematic and
oriented toward specific activities and questions for
researchers to pursue.
Schein’s view of the place of language in culture
typifies the Functionalist perspective. Although he
acknowledges the "semiotic approach to cultural analysis"
Cp.15), Schein appears not to include language as part of
the visible behavior and artifacts that scholars using the
semiotic approach would investigate. Instead, he
32
addresses language as if it were a tool, an objective
code, that members employ to build mutual understanding
and consensus. When discussing internal integration, for
example, Schein asserts that it is necessary For groups to
establish a common language and conceptual categories in
order to maintain their group cohesion: "Members of a
founding group coming together to create a new
organization need to learn about each other’s semantic
space Ceven if they start with a common basic language,
such as English) in order to determine what they mean
by ... abstractions" Cpp. 65-66). It is thus uncertain Just
what is the relation of language to culture in Schein’s
perspective.
Power is seen only as positional authority, as
complex "pecking orders" Cp. 72), laid up in stratified
allocations among members. The distribution of power is
accomplished by Formal allocation or by negotiation and
renegotiation, and the human tendency to excesses in the
use of power is constrained by culture: "CCUultural norms
regarding the handling of aggression help us deal with
feelings that might run out of control, endangering us and
others...." Cp. 73). Consequently, it seems that both
power and language are external to the concept of culture
as Schein sees it.
Perhaps the strongest clue that Schein’s view is
quintessentially functionalist is his characterization of
33
his own approach as a "clinical perspective" as against an
ethnographic perspective. In an ethnographic approach to
culture, the members of the unit under study "usually have
no particular stake in the intellectual issues that may
have motivated the study" Cp. 21). A clinical approach,
on the contrary, is motivated by a solicitation from the
client, who "must get the cooperation of the
helper/consultant" Cp. 21). The relationship between not
only client and consultant, but also between consultant
and research questions and between consultant and data are
radically different from the relationship that obtains in
an ethnographic situation. According to Putnam C1983, p.
37), functionalist approaches to organizational analysis
often are characterized by a "managerial bias," and they
take managers’interests, problems, and assumptions about
the nature of work relationships as the "givens" of
research projects. Throughout his work, Schein relies on
his own consulting experience as the illustrative data for
his theory of organizational culture, and his views
reflect a wholesale acceptance of a management definition
of problems, situations, relationships, goals, and so
forth.
Culture to Schein, then, is a configuration of
Jointly held beliefs that are pragmatic in their effects
on organizational members, and they are cultural only if
they have these pragmatic effects on members. Meanings
34
are important insofar as members need to share "semantic
space" to establish and maintain their groupness.
Subjectively held meanings are important only to the
extent that they contribute to group consensus on the
nature of reality, time, space, etc. These views are
evident in the work of other major functionalist writers,
such as Deal and Kennedy, George Gordon, Ralph Kilmann,
and Peters and Waterman.
Interpretivist approaches to culture.
A review of recent collections of essays and studies
on organizational culture shows that comparatively fewer
scholars take a clearly interpretivist approach CFrost et
al., 1S85; Gudykunst et al., 1985; Kilmann et al., 1885)
than a functionalist approach. Nonetheless, interpretive
views of organizational culture might become the more
influential mode insofar as they emphasize investigation
of intersubjectivity and more deeply anchored meanings of
power and sense-making. One of the most prominent
spokespersons of this view is Linda Smircich C1983a,
1883b, 19B5).
Smircich sees organizations as cultures, that is, as
"networks of meaning" (13B3b, p. 160) and members as
active agents. Members "impose themselves" on the world,
creating experience through intentional action "that
assumes its meaning and significance within the context of
3 5
intrepretive schemes embodying a particular pattern of
purpose, value, and meaning" Cp. 161).
Culture in this view is an epistemological concept
borrowed from the symbolic and cultural ecology schools of
anthropology Csee Ortner, 1984), and as such it is a
heuristic for framing the ways that organizations are
"commonly held fabricCsl of meaning" CSmircich, 1983b,
p.162). The researcher’s task is to piece together
evidence that reveals the nature of the meaning system in
use by members, similar to the tasks set for
anthropologists CGeertz writes that "culture is most
effectively treated... purely as a symbolic system...by
isolating its elements, specifying the internal
relationships among those elements, and then
characterizing the whole system in some way" in terms of
underlying structure or ideology C1973, p. 173). Smircich
directs the researcher to seek recurrent themes "that show
how symbols are linked into meaningful relationships" Cp.
163) and are implicated in the activities of members.
Symbols, then, connect members’ values, beliefs, and
actions. Researchers need to study not only the
"meaningful ordering" of symbols but also the "cosmology
and metaphorical conventions underlying language and
action" CSmircich, 1985, p. 65).
Interpretivists view language as one of many systems
members are enmeshed in, although it is seen as "often the
36
most accessible symbol system" CSmircich, 1983b, p. 170).
Language is relevant to interpretivists because it reveals
members’meanings and the ways they organize their
perceptions. Meanings are "the essence of the social
world," Cp. 160) and organizations are shared social
realities, constructed by the meaningful interaction and
interpretation of members.
The central position of symbolic orders in Smircich’s
view is reflected in her recent turn toward focusing on
symbols instead of culture and her recommendation that
analysis be conducted as a dialectic between "the two
concepts of symbol and power" CSmircich, 1985, p. 67).
Her view of power is not evident, and she sees the outcome
of interpretive analysis of organizational life to be
enlighten-ment for members: All persons involved need "to
construct or reconstitute knowledge so that events,
situations, and problems are confronted or engaged from
multiple paints of view" C1985, p. 71). While this view
clearly does not favor a management interpretation of
organizational life, it seems to be saying that one
interpretation is as valid as another, and what wise
members need is consciousness-raising in terms of opening
up their range of alternative interpretations.
Elsewhere Smircich reinterprets management theory in
light of interpretivist tenets CSmircich, 1983c). There
she discusses management effectiveness and control as a
37
matter of developing and maintaining "widely shared
interpretations that allow coordinated action to occur"
Cp. 238). The role of the interpretivist consultant in
that context is to help members surface unconscious or
tacit assumptions, beliefs, and norms so as to identify
practices that might trap or block members from greater
self-consciousness.
Sypher et al. C19B5) ask whether all culture research
in organizational studies is necessarily interpretivist.
A similar question might be posed about symbolism and
culture: Can all organizational symbolism research be
subsumed under the label of organizational culture? This
easily could be assumed to be the case by a reading of
Putnam’s and Smircich’s essays. Some researchers,
however, clearly see culture as only one of several frames
or metaphors by which symbolism in organizational contexts
can be understood. Gray, Bougon, and Dannelon C1985), for
example, view the culture metaphor as too static to
account for concerted worker actions that do not
necessarily involve shared meanings. By this they mean
that unexpected outcomes of action CGiddens, 1979) and
retrospective sensemaking CUIeick, 1979) frequently emerge
from Joint actions that involve members who understand the
premises and nature of the action differently and who have
divergent goals.
Organizational symbolism research is fundamentally
38
about sensemaking, and the borderland between it and
organizational culture studies often is simply a matter of
what the researcher chooses to label the product;
additionally, a major component of the organizational
symbolism tradition— critical studies— is typically more
concerned with issues of power and emancipation CConrad,
1983; Stablein & Nord, 1985) from domination than most
culture studies. Confusing the issue unnecessarily is
Deetz’s C1985) coinage of the label "critical-culture
research," for nowhere in his concise and insightful
review of critical studies in organizations does he
develop the connection with culture.
Critical studies are discussed later in the section
on power literature. What is apparent in the literature
of organizational culture per se are several useful
concepts, postures toward research activities, and
attitudes toward concepts and research. All culture
studies direct a focus on meanings held by members of
organizations, whether it is a concern for surface-level
congruence of members’ lexical understandings, as the
cross-cultural view would engage in, or a concern for
systemic nets of meaning held in common as tacit knowledge
by members, as interpretivists would have it. In
addition, culture studies inevitably place organizational
meanings in the broad context of a local historical
framework. Finally, the culture metaphor provides an
39
integrative Framework that leads researchers to examine
organizational life as a community. As Meryl Reis Louis
describes it:
What distinguishes the current formulation— as culture
— is its basis in anthropology. The quest here is not
for the strictly psychological or sociological compo
nents of the phenomenon, as was the case in the past.
Rather, the uniquely integrative and phenomenological
core of the subject, in which the interweaving of
individuals into a community takes place, has finally
become the subject of investigation among organiz
ational scientists. So too has the conceptually
slippery notion of meaning— its ontological status,
emergence,and function 01985* pp. £7-20}.
By taking account of meaning and sensemaking in
organizations and by recognizing that the process of
bestowing meaning on experience is itself symbolic,
scholars are opening up organizational research to the use
of new methods or approaches to understanding that are
borrowed from other disciplines, such as anthropology.
Accordingly, some writers are encouraging the use of
multiple methods in the pursuit of organizational culture
research CCarbaugh, 1985; Eisenberg & Riley, in press;
Sypher et al., 1985}. The recommendation of multiple
methods not only reflects an innovative and practical
posture toward research, it also promotes in turn an
openness to new questions and concepts in the research
activity.
The problems with culture research are significant.
Its chief deficiency, in terms of this dissertation, is
40
its limited consideration of pouier, particularly of power
that might be implicated in the very systems of symbols
that typify the interests of culture research. Similarly,
there is a dearth of discussion of ideology in the
literature. An additional deficiency is the lack of work
that explores the specific ways in which language affects
such cultural dimensions as values, beliefs, and meanings,
employing observed data from real organizational members.
Although Carbaugh C1985) argues persuasively for a
communication orientation toward culture research in
organizations, little work beyond metaphor analysis CSmith
& Eisenberg, in press), drama (Rosen, 1385), and special
vocabulary CEvered, 138S) has been done using the
linguistic experience of members.
The last problem area, the examination of how
language is involved in the communicative and sense-making
experience of members, has been the province in recent
years of a major area of sociolinguistics.
Ethnooraphu of Communication
Stephen Levinson C19B3) devotes considerable effort
to distinguishing the terms "pragmatics" and
“sociolinguistics" without coming to any firm conclusion
about the ambit of either subdiscipline. Both tend to
address issues involving meanings of language, pragmatics
more generally being concerned with meanings intended and
communicated within utterances, and sociolinguistics more
41
generally concerned with the ways in which language is
used in situated contexts for accomplishing social aims.
Levinson seems to favor Gazdar’s formulation of pragmatics-
-it investigates meaning that is independent of semantic
truth conditions of the sentences uttered CLevinson, 1983,
p. 15).
Insofar as such a definition places the concerns of
pragmatics in the realm of "the organization of verbal
means and the ends they serve," pragmatics and
sociolinguistics overlap in Just the places that
constitute what Hymes considers the primary interests of
sociolinguistics, or the "ethnography of communication"
CHymes, 1974, p. 8). Hymes makes it clear that the
central concern is with people using language in social
situations: "The interaction of language with social life
is viewed as first of all a matter of human action, based
on a knowledge, sometimes conscious, often unconscious,
that enables persons to use language" C1974, p. 45).
Consequently, the scope of sociolinguistics discussed here
includes much of what commonly is called pragmatics and
all of what is referred to as the ethnography of
communication.
A major portion of socialinguistic theory and
research is not of direct interest in this investigation.
That portion is the variable-rule tradition associated
with William Labov and his followers Ce.g., Fashold, 1973;
42
Labov, 1366, 1372; Sankoff, 1374, 1300). The variable
rule method seeks to identify relationships between
linguistic variables and social variables by quantitative
analysis of the occurrence of linguistic evidence in
circumscribed contexts of elicitation. Huspek C1386)
points out that there are two fundamental problems with
this method. First, linguistic and situational
constraints are assumed to influence the speakers’
linguistic choices independently. Environment is
considered to mean only the linguistic context; social and
stylistic variables are treated as extraneous to speakers’
motivation to choose one way of speaking versus another.
The second problem involves the assumption that linguistic
constraints are grammatical, that is, they are
phonological or syntactic and are independent of meaning.
This assumption reflects a further assumption that all
speakers are ideally competent members of a homogeneous
speech community. Huspek C1S86) studied lumber mill
workers’ use of "ing/in’" endings in their everyday
conversation. He Found that individual speakers varied
their word endings by differences in the social and
discourse context in which they were speaking. Contrary
to the variable rule researchers, Huspek’s work
demonstrates that speakers employ variants as "expressions
at the social-stylistic level of determinate, contrastive
meanings. Such meanings... provide us with understandings
4 3
more reliable than conjectures on the significance of
statistical correlations between linguistic variant and
social or stylistic variables" C1986, p. 159).
A fully developed alternative to the variable rule
approach to sociolinguistic analysis is offered by John
Gumperz in his theory of "interpretive sociolinguistics"
CGumperz, 1982a). Gumperz focuses on "communication" in
his approach to linguistic analysis, and he places
variation of discourse in the center of his theory:
Any sociolinguistic theory that attempts to deal with
problems of mobility, power and social control cannot
assume uniformity of signalling devices as a precondi
tion for successful communication. Simple dichoto-
mous comparisons between supposedly homogeneous and
supposedly diverse groups therefore do not do justice
to the complexities of communication in situations of
constant social change Cp. 7).
Drawing on his earlier work in language and dialect
contact in various contexts CBlom & Gumperz, 1972;
Gumperz, 1977; Gumperz & Hernandez-Chavez, 1971; Gumperz &
Wilson, 1971), Gumperz demonstrates how speakers employ
conversational practices and knowledge to accomplish their
communicative goals. Among the practices he examines are
speaker switches from one language variety to another
C"code-switching"), prosody, and other paralinguistic cues
that he calls "contextualization conventions." Speakers
must hold in common extensive socio-cultural knowledge in
order to employ conversational implicatures, word-games,
44
double-entendres, etc., and thereby enrich their discourse
strategies.
Gumperz agrees with most other researchers that code
switching is not a randomly employed linguistic behavior.
But he goes beyond analysis based on relating code-
switching to its linguistic environment to consider the
social event the behavior is embedded in and the
conversational aims of the speaker. In many instances,
code-switching is a comment on the social relationship of
the participants in a conversation. It can be used
transparently to convey group solidarity and promote
convergence between speaker and hearer CGiles, 1979;
Tajfel, 1978), it can be used ironically to promote
divergence, or it can be used as a form of implicature, to
convey meaning by partially violating a conversational
rule known to both participants.
Gumperz’s concept of contextualization conventions
is relevant for this study because the concept accounts
for every communicative interchange that occurs. It
subsumes code-switching behavior in that it is "any
feature of linguistic form that contributes to the
signalling of contextual presuppositions" CGumperz, 19B5a,
p. 131). As such, the concept is roughly equivalent to
Gregory Bateson’s notion of the "command" aspect of any
utterance, which signals houi the information in the
"report aspect" is to be interpreted CRuesch & Bateson,
45
1351) by commenting indirectly on the relationship of the
participants.
Contextualization conventions assume that speakers
share an enormous amount of information about how meanings
are formulated and interpreted; moreover, such
understanding is not acknowledged expressly in the course
of producing a conversation: "The meanings of
contextualization cues are implicit....Their signalling
value depends on the participants’ tacit awareness of
their meaningfulness" CGumperz, 19B£a, pp. 131-132).
Gumperz observes, however, that the conventions are
grounded in cultural backgrounds and experiences of
speakers. Consequently, for persons of differing cultural
backgrounds, conversation often is problematic because of
participants’ inability to agree on what kind of activity
is constituted by their conversation. As Gumperz
describes the relationship, "if conversational inference
is a function of identification of speech activities, and
if speech activities are signalled by culturally specific
linguistic signs, then the ability to maintain, control
and evaluate conversation is a function of communicative
and ethnic background" Cp. 167). This observation turned
inside out suggests that ethnolinguistic minorities are
vulnerable to domination by not being able to attribute
appropriate meanings in or control the discourse they
46
might participate in with speakers of the standard or
prestige language variety.
Gumperz asserts that hidden sources of
misunderstanding and subjugation can be brought to the
surface and studied by using conversational analysis,
particularly as applied to situations involving
interethnic interaction. A typical example of what this
approach can accomplish is a study by Jupp, Roberts, and
Cook-6umperz C1982) of South Asian immigrant workers in
Britain. Jupp et al. describe the interethnic work
situation as one where minorities are systematically kept
in low status positions because of assumptions held by
both managers and the minority workers about identity and
communication. Unfair discrimination at the outset
relegates most South Asians, regardless of qualifications,
to the lowest status positions in service and
manufacturing industries. Their language learning often
is not accompanied by complete language socialization;
consequently, their competence in applying
contextualizatian conventions in conversation is limited.
As Jupp et al. describe it,
the assumptions of poor language use and of weak
communicative power on the one hand, and the low-
status job position an the other which, in
interaction with each other, make up the socially
created identity of incompetence or lesser competence
and uncooperative behavior and attitudes. This
identity of the South Asian worker as less competent
or in other ways inadequate is then used to warrant
the maintenance of the lower status of this group
CJupp, Roberts & Cook-Gumperz, 1S02, p. 240).
47
The authors use samples of discourse to illustrate
British workers’ attitudes toward the immigrants and vice
versa; in addition, discourse From an actual job interview
is analyzed to identify precisely how the interethnic
contact goes awry for all parties, based in their failure
to interpret contextualization conventions effectively.
They conclude that it is possible to train participants to
more effectively create contexts and make inferences so as
to avoid problems in interaction.
While Gumperz makes greater use of discourse data in
other studies Ce.g., Gumperz, 1382b), the study Just
described is valuable because it explicitly addresses
disadvantage in the workplace and links it to the parties’
use of language. In addition, Gumperz’s approach
generally deserves attention because it seeks social
meaning in linguistic features that are covert and often
unconscious, such as suprasegmental cues and formulaic
expressions. In addition, his work is valuable for its
insights into the communicative uses of code-switching,
syntactic choices, use of implicature, and other pragmatic
formulations.
□ther work in the ethnographic tradition of
sociolinguistics similarly links social meanings and
linguistic features of discourse, but most of the studies
either do not address discourse in the workplace Ce.g.,
48
□chs, 1984; Scherzer, 1984) or otherwise take a
superficial approach to the discourse Ce.g., Fisher &
Todd, 1986). One significant exception is the work of
Shirley Brice Heath, particularly in her ethnography of
two communities in the Piedmont Plateaus of the
southeastern U.S. CHeath, 1983). Heath convincingly
demonstrates how language habits that discriminate one
group from another are products of and reflections of each
individual’s history, dare importantly, she explores the
relationship between the development of childrens’
language and communication skills and their interpretation
of their world, including its time, space, and social
dimensions. In addition, Heath shows how all parties to a
social situation, as in a school, can create real social
improvements by learning to experience others in an
ethnomethodological frame. In her project, Piedmont
teachers become ethnographers of their students’ learning
activities and then teach the students to apply
ethnographic methods to their own experiences and life
situations. The new sensitivity to their unique language
habits enriches the learning experience for both teachers
and students, and the students’ intellectual and social
lives are improved.
Three shortcomings are evident in most of the
research in the ethnographic tradition of
sociolinguistics. First, the approach to meaning is
49
primarily speaker-oriented, focusing sharply on speaker
intentions and only secondarily an others’
interpretations. Gumperz explicitly takes a "speaker
oriented approach to conversation...and strategies"
(19BSa, p. 35), and his project is a theory that "accounts
For the communicative functions of linguistic variability
and for its relation to speakers’ goals" Cp. S9). What is
needed to complement this view is an overarching social
theory that accounts For the contributions and
interpretations of participants in interaction, in terms
of linguistic Features, irrespective of their role as
speakers or hearers.
Second, studies in this tradition present data in an
ad hoc fashion, without a disciplined reason For selecting
the participants or the segments of discourse that are
presented other than its problematic appearance. For
discourse practices to be representative of a class of
sociolinguistic problems, something must be known about
the form, purposes, and usual expectations of participants
in a selected stretch of discourse, and those forms,
purposes, and expectations should be explicitly described
at the outset of analysis. Without such an understanding
of what the selected discourse is supposed to do, there is
no principled way of bracketing a stretch of discourse for
analysis, and any stretch of talk is theoretically as
representative as any other stretch.
5 0
Third, while admirably grounded in the real world of
interaction, most of the work in this tradition is weak on
theoretical Foundation. Gumperz briefly mentions Alfred
Schutz as an authoritative source for his nation of
typified schematic knowledge CISBEa, p. 22), and he
liberally refers to the early work of Dell Hymes. Uhile
Hymes’ work is innovative, it nonetheless is grounded in
turn in such authorities as Sapir, Frake, Firth, Jakobson,
Cassirer, and even Kenneth Burke; and to cite Hymes
without the others is to miss the primary authorities. As
it is, Gumperz’s work is noteworthy because of its lack of
reference to writers who would provide an epistemological
or ontological orientation for his approach to
sociolinguistic problems.
Ethnolinaulstic Identitu Research
A refinement of my earlier comments on power and
identity would point out the particular problems of
discourse and identity experienced by ethnolinguistic
minorities. According to Jeffrey Ross C1979), language
use is central to group identity for any member of an
organization. In distinguishing ethnic groups from other
types of collectivities, Ross notes that ethnic groups
have the social flexibility to mobilize language
strategies For promoting their collective identity.
Language, he says, "is important, not in itself, but as a
symbol of an underlying image of group purpose and
51
Identity. In this regard, language is probably the most
powerful single symbol of ethnicity because it serves as a
shorthand for all that makes a group special and unique"
C1S79, pp. 9-10). This view is essentially the same as
that of George DeVos C1975), who states, "language
constitutes the single most characteristic feature of a
separate ethnic identity" Cp. 15).
Strategies of language use, however, have been found
to play a significant role in situations of interethnic
contact. Giles and Powesland C1975), for example, have
set forth a theory of speech accommodation that accounts
for both the change and resistance to change of language
use by ethnolinguistic minorities. Based on social
exchange and attraction theories, accommodation theory
says that speakers will adjust their speech toward that
used by their listeners if they wish to be identified mare
closely with the listeners’ apparent group identity
(called "convergence") and will adjust their speech away
from that of their listeners if they wish to differentiate
from the listeners’ group ("divergence"). The basis for
accommodating speech is thought to be a consideration of
social costs and benefits, that is, what one stands to
gain by appearing more like (and thus more attractive to)
the listener versus what one stands to lose by doing so.
Giles and Powesland also hold that accommodation—
either convergence or divergence— varies in the degree to
52
uihich speakers are conscious of their speech adjustment or
resistance to it. Moreover, listeners may be more or less
aware of speakers’ efforts at accommodation, and attitudes
toward speakers might be affected by the degree of
awareness by listeners of speakers’ efforts to adjust
their speech. Street and Giles C1982) maintain that
intergroup language differences are attenuated when
cooperation is high or the salience of intergroup
differences is low.
More important than accommodation per se is the
variety of viewpoints about what ethnolinguistic
minorities might give up when accommodating their language
use. Many scholars Ce.g., Giles, 1977; Giles, Robinson &
Smith, 1980; Giles & St. Clair, 1979) see efforts by
ethnic speakers to adjust their language use toward that
of some other group as a repudiation of their ethnicity
and group identity. Glyn Williams C1979), for example,
presents an interpretation of the struggle faced by
ethnolinguistic workers that emphasizes the interlocking
relationship between language and position in the
organizational hierarchy. Such workers, she says, need to
cross a linguistic boundary by learning to speak like
those in positions of power before obtaining an
opportunity to cross the positional boundaries in the
workplace. Unless the ingroup language of the ethnic
group is also the prestige language of the workplace, the
5 3
worker must learn to change his or her language toward
that of the outgroup Ci.e., those in high level positions)
to gain access to rewards of the workplace, and in the
process the worker increases social distance between self
and ingroup members. Pool C1979) compares language with
other possible correlates of ethnic identity and concludes
by concurring with Williams’ view: "more people always
identify with the group associated with their language
than with the group associated with their residence,
religion, name, or ancestry."
flora recent work suggests that Williams’ view is an
oversimplification of the situation, Margarita Hidalgo
C1986), far example, has found that ethnolinguistic
minorities along the American Southwest border have
differing views of their use of English, depending on
whether their purposes for interaction are basically for
accomplishing a task— an instrumental use of English— or
are for establishing social solidarity— an integrative
use. Hidalgo reports that "studies have consistently
reported that individuals claim a higher instrumental than
integrative motivation toward English" Cp. 196).
Moreover, multilinguilism studies have shown that, despite
high levels of "mother-tongue claiming among ethnic
minorities in various countries, ethnic identity can be
maintained in the absence of current use of mother-tongues
CFishman, 1906b). Edwards C1905, p. 161) stresses in
54
addition that a common language variety, while important,
is not a necessaru factor far groups to maintain their
collective identity.
In spite of the diversity of views of just how and to
what degree language influences individual and group
identity, it is clear that language is implicated in
important ways. Howard Giles and associates CGiles,
Bourhis & Taylor, 1977; Giles & Johnson, 1901; Edwards &
Giles, 1984) have framed the problem in terms of a set of
social-psychological variables that determine
ethnolinguistic vitalitu. A group’s vitality, their
likelihood of maintaining their distinctive identity,
depends on the group’s relative social status (including
the perceived status of the group’s language both from
within and from outside the group), demographic profile,
and institutional support and control in such areas as
education, religion, mass media, etc. Vitality is both an
objective measure of a group’s ethnolinguistic cohesion
and, more importantly for identity theory, a subjective,
cognitive assessment that members make, based on their
understanding of the three factors.
In addition to subjective ethnolinguistic vitality,
three other factors contribute to Biles’ theory of
identity. First is a principle from Henri Tajfel’s theory
of intergroup relations CTajfel, 1970) that groups create
various strategies to distinguish themselves as separate
55
From other groups and as being desireable social
entities. Second is the idea that group boundaries vary
in terms oF their strength, value, and distinctiveness and
may thus be described as "hard" or "soFt" boundaries. The
permeability oF group boundaries, in turn, aFFects
vitality and the chances oF group assimilation by another
group. Last, the theory accounts For the membership in
other social groups that ingroup members maintain. In
general, linguistic diFFerentiation will increase with
high subjective vitality, many strategies For ingroup
status enhancement, "hard" intergroup boundaries, and Few
strong memberships in other groups. Convergence, on the
other hand, would occur most readily in cases oF low
vitality, Few strategies, "soFt" boundaries, and many
strong aFFiliations with other groups.
The social-psychological tradition oF ethnolinguistic
identity has contributed much to the understanding oF the
relationship between language use and patterns oF
assimilation. It has been particularly useFul For
understanding second language acquisition among
immigrating ethnic minorities. Most importantly, it has
sensitized scholars to the need For devoting attention to
the subjective experience oF ethnic group membership in
intergroup contact situations. The perspective has not
been without detractors, however. The strongest criticism
holds that the theory’s variables are ambiguous, multic
56
orrelated, and atheoretical CHusband & Khan, 19BE).
Others contend that the theory is a static model and does
not account Far changes in group membership and linguistic
competencies CLePage & Tabauret-Keller, 19BE).
Furthermore, it sets out to describe interethnic contact
and predict linguistic differentiation based on a theory
of membership in a subordinate group within a social
situation; seriously lacking is a simultaneous account for
the influence of perceptions, beliefs, and attitudes of
superordinate group members. That added perspective would
factor in a number of implications about intergroup power
and dominance that are not available in the social-
psychological approach to ethnic identity.
On the other hand, the social-psychological studies
on language attitudes have much to say about speaker and
hearer attitudes, beliefs, and perceptions. This
tradition helps fill in the blanks evident in
ethnolinguistic identity research.
Language Attitude Research
Many studies have demonstrated that listeners Judge
speakers on a number of traits based on samples of their
speech. Typical of the personal dimensions reflected in
language attitudes are speakers’ competence, integrity,
and attractiveness CGiles, 1971; Lambert, 1967).
Researchers have examined listener reactions to different
varieties of speech, including varieties of British
57
English CGiles, 1971; Trudgill, 1974), Canadian French and
English CLambert, Frankel & Tucker, 1956), Mexican Spanish
and American English CCarranza & Ryan, 1975; de la Zerda &
Hopper, 1979), Standard American English and Black
American Vernacular CHopper & Williams, 1973). In
addition, studies have been conducted in a variety of
practical situations, such as Frederick Williams’ work in
the educational setting Ce.g., F. Williams, 1975), Lind
and □’Barr’s C1979) study of powerful and powerless speech
in the courtroom, and the numerous studies carried out in
job interview situations CHopper & Williams, 1973; Kalin &
Rayko, 1900; Rey, 1977).
□F particular relevance to this dissertation are the
findings of research on language attitudes in work
situations. Unfortunately, virtually all such studies are
concerned with pre-employment interviewing, and the wide
variety of other job-related contexts have not been
investigated. Although this part of the research has been
effectively reviewed by Rudolph Kalin C1982), several
conclusions deserve emphasis here. First, it is apparent
that judgments of speaker traits typically result in
"double discrimination," such that nonstandard accented
speakers are adjudged less appropriate than standard
speakers far high level positions but more appropriate for
low level Jobs, while standard speakers are thought to be
more appropriate than nonstandard speakers for high level
58
positions and less appropriate For low ones Cde la Zerda &
Hopper, 1979; Kalin & Rayko, 1980).
Second, there is little agreement on the degree to
which race and ethnicity are involved in judgments of
speakers’ traits or on the mechanisms by which judges
might believe race or ethnicity are manifested in speech.
Some studies have focused on both racial and ethnic
variables, and judges successfully discriminated among
speakers on those bases CHopper, 1977; Rey, 1977).
Judgments of speakers’ ethnic group membership have
consistently been more accurate than chance CKalin,
1982).
Third, judgments of speakers seem to cluster into two
dimensions, competence Cmade up of factors such as
perceived intelligence, industry, knowledgeability) and
attractiveness Cmade up of such factors as friendliness,
humor, trustworthiness). These two dimensions have been
shown to be predictors of hiring decisions among actual
personnel interviewers CKalin, 1982).
While language attitudes research has brought into
sharper focus the influence of speech perceptions on such
vital issues as educational opportunity, medical care,
legal processes, and employment opportunity, it by and
large has ignored the issues of power and dominance.
Surely, if inappropriate Judgments of applicants’
qualifications are being made by Job interviewers on the
59
basis of their subjective impressions of applicants’
speech, power of a direct and consequential kind is being
exercised. It has not been demonstrated or claimed in the
research reviewed here, however, that the judgments
interviewers make on the basis of speech samples actually
result in decisions to hire or not hire in real
organizations. Nor has it been shown that any other
organizational relationship is in fact affected by the
nature of powerful members’ perceptions of others’ speech
characteristics.
What has been demonstrated by this research is that
listeners are likely to make judgments about the identity
of others partly on the basis of experiencing their
speech. If powerful members of work organizations, like
Job interviewers, make judgments about others’ identities,
it is likely that members of organizations who are in
subordinate positions also make reciprocal judgments about
powerful members’ identities.
Research on Institutional Power
To hold that some members of an organization are
powerful while others are not is to entertain that "power"
is asymmetrically distributed among members. Theories of
power in institutions differ widely in their basic views
of what is meant by the term, but one of the ways they can
be partitioned is whether the concept is seen as a unity
or as a multilayered phenomenon. Traditional views of
60
power in institutions reflect Max Weber’s concept that it
is one actor’s capacity to change another’s behavior even
against the other’s will CGiddens, 1979, p. 88), a view
captured in Robert Dahl’s term, "willed action". Giddens
C1979) points out the similarity between the Weberian
concept and the view of such social critics as Arendt,
Parsons, and Poulantzas. For these researchers, power "is
specifically a property of the social community, a medium
whereby common interests or class interests are realised"
Cp. 89).
Both of the foregoing views of power, however,
operate at the surface level of relations among members
and between members and the supposed structures of the
organization. Clegg, on the other hand, proposes to view
power as a multilevel set of relations whereby power is
manifest at the surface level of organizational life in
the outcomes of exchanges among members. This level of
power is called "surface structure," in an analogy with
transformational syntax CClegg, 1975, p. 78). Underlying
the surface manifestations of power are the deep
structures of power in the form of rules that provide the
rational theory of the organization’s relations within
which the exchanges occur. Among the deep structural
rules are those that are generated by domination. Clegg
envisions domination as a "form of life" in
Wittgenstein’s sense of a way of being in the world that
61
is normalized C1975; 1977). Thus the very ontological
basis for experiencing organizational life and its taken-
for-granted rules are the genesis of power. Clegg states:
"power relations are only the visible tip of a structure
of control, rule and domination which maintains its
effectiveness not so much through overt action, as through
its ability to appear to be the natural convention" (1977,
p. 35).
In his genealogy of power, Foucault also sees power
as ingrained in the normalized relationships of day-to-day
practices. He theorizes that in the seventeenth and
eighteenth centuries there arose a new form of power that
depends on “obtaining productive service from individuals
in their concrete lives" (1980, p. 1E5). Hence, "power
had to be able to gain access to the bodies of
individuals, to their acts, attitudes and modes of
everyday behaviour" (p. 125). Foucault sees power as
"always there," (p. 1H1) infusing the social scene and
circulating through its social network as a potentiality
in all molecular elements of society; it must be
understood from the bottom up, in its "infinitesimal
mechanisms, which each have their own history, their own
trajectory, their own techniques and tactics" (p. 99).
Foucault’s view of power is less static and
structural than Clegg’s, yet it is less systematic, more
diffuse and less involved specifically in the discourse of
68
institutional members. Another perspective on power that
accounts For the dynamics of social process, the minutiae
oF everyday living, and the centrality of discourse is the
theory of Anthony Giddens C1979, 1984). The fundamental
axis of power in organizational relationships for Giddens
is in the transformations of the allocations of resources
and the authorizations of positions into domination
structures. The term "transformation" is intended to
convey the sense that resource allocation and
authorization become embedded in the taken-for-granted
nature of organizational relationships. Power is seen as
the relation of the capacity to transform resources to
structures of domination, the latter becoming in the
process a resource itself. Power thus is dependent on
choice-making human agents, members who have an awareness
of their choice alternatives, or, as Giddens terms it, the
"could have done otherwise" C1979, p. 98).
A more detailed discussion of Giddens’ view of power
and its related concepts is presented in Chapter 4. It is
sufficient here to point out that power is intrinsic to
language and discourse through the concept of
sianification. the process of encoding rules of a system
through which domination may be constituted and
regularized:
...we must once more substitute duality for dualism:
signification, as concerning generative properties of
structure, is linked recursively to the communication
of meaning in interaction. Signification refers to
63
structural features of social systems, drawn upon and
reproduced by actors in the form of interpretive
schemes...Cand! meaning must be treated as grounded
in the ’contexts of use’ of language C1979, p. 98).
Other critics and social theorists have focused on
discourse as the embodiment of power relations in work
organizations and have pioneered the search to articulate
the ways linguistic asymmetries in the workplace are
related to asymmetries of the distribution of other
resources, including rewards, prestige, and honor CClegg,
1975; Kramarae et al., 1984; hey, 1985). UJhat is absent
from these studies, however, is drawing an explicit
connection from in situ discourse to the fates of workers
in organizations who might be system-atically
disadvantaged, and framing that connection within an
overarching theory of social reality. Giddens C1979,
1984), for example, notes that ideology can be studied in
the strategic action in the form of modes of discourse;
from another angle, ideology can be studied from an
institutional perspective in the form of "basic structural
elements which connect signification and legitimation in
such a way as to favour dominant interests C1979, pp. 191-
2)." Yet there is no bringing together of discourse and
structure, no blended analysis of the practices of
signification in their specific instances and
legitimation.
64
Career Trajectories
Much of the literature on career development and
uiorklife fulfillment is centered on employee motivation,
the rewards system, and social network management at work
Ce.g., Lawler, 1377; Porter & tliles, 1374; Raelin, 1384).
Studies in this genre assume cultural homogeneity among
workers. Many studies of employee socialization , such as
those by Van Maanen and Schein (1373), Wilson (1384), and
Eisenberg (1386), refer to newcomers’ socialization into a
given world view and acknowledge the communication basis
of that process. They nonetheless omit discussion of how
the newcomers’ discourse practices and those of the
organization affect employees’ careers. The rash of
popular literature on careers over the past decade or so
(e.g., Harragan, 1377; Houze, 1385; Peter, 1375;
Rothschild, 1384; Sinetar, 1387) typically treats problems
involving discourse as localized codes that have to be
mastered to proceed with career advancement. Career
trajectories are most commonly predicated on such
strategic contingencies as knowing the right people,
identifying the right promotion opportunities, and making
the right plan.
For many workers, unfortunately, planning ends at the
entry level of employment. Structures of domination are
so established that without certain prerequisites,
including access to a linguistic repertoire that reflects
and generates power, — some workers-’ — formal positions and---
65
personal relationships remain Fixed for all time. Every
year at the Lewitt hotels, For example, employees who have
worked For Fifteen and twenty years as maids or stewards
receive awards for their long periods of faithful
service. Some have said in interviews that they have
chosen to remain in their humble positions, suggesting an
implicit belief that they could have chosen otherwise.
Giddens C197S) comments that domination is mast normative
when its consequences are accepted by all organizational
members as a natural state of affairs. For employees who
have no effective way out of powerless social and
occupational positions, the status quo must seem
powerfully a natural state of affairs.
Research Questions
Each of these five areas of scholarship contributes
important findings toward understanding the relationship
between discourse and the worklife chances of
ethnolinguistic minorities in particular and workers in
general. Organizational culture research provides a
framework that at once permits an integrated, macrolevel
view of organizational phenomena; it enables researchers
both to account for differences in world view among
members of different national or ethnic origins and to
delve into the range of symbolic elements of specific
organizational settings. The other four areas reviewed
6 6
contribute a deeper understanding of what discourse is and
how it functions in social interaction.
Several propositions can be set down as a result of
this review. First, organizations can be analyzed as
local networks of meanings held more or less in common but
generally recognized by members. Uncovering these
meanings will help our understanding of the nature of
social relations in organizations. Second, meanings
intended by speakers are formulated through the complex
use of discourse conventions that overlay and frame the
straightforward composition of linguistic elements;
moreover, speakers’ awareness of their use of such
contextualization conventions may range from full
cognizance to complete unconsciousness, and the same might
be so for others involved in interaction with them.
Third, in addition to advocating partisan views of what is
meaningful, important to notice, and of value in the
world, discourse establishes and maintains individual and
group identities of members. Fourth, many workers in
lower positions who speak nonstandard language varieties
have less capacity through discourse available to them to
establish and maintain personal and group identities which
others would associate with success in the organization.
Last, organizational power is enacted and reflected in the
discourse practices in the minute, day-to-day interactions
of members, and only when the covert system of power
6 7
relations breaks down does the more conventionally
recognized set of overt power strategies play a role in
work relationships.
The conceptual problems, omissions, and other
deficiencies identified in the foregoing review constitute
the basic issues addressed in this dissertation. What is
not known but seems to be the next pivotal item on the
research agenda of organizational culture, applied
sociolinguistics, and critical studies is how specific
linguistic variables or practices contribute to the system
of relationships in workplaces. Taking meaning to be a
concept that mediates between linguistic items in use and
the social order inherited or promoted by institutional
powerholders, exploration of how discourse results in
meanings understood and potentially subscribed to by
members is crucial to the overall project.
To refine this general statement of objective to
specific research questions requires explication of three
factors: a) source of the data, b) nature of the
linguistic elements to be analyzed, and c) assessment of
shared understanding. For data to be more than
adventitiously gathered and examined, an appropriate
setting and bounded genres of interaction should be
demarked. The discourse to be examined must be capable of
being bracketed as a sensible unit for analysis. One
obvious way of effectively bracketing the discourse is to
6 8
work with units whose Function is explicitly stated or
generally recognized. Several administrative events in
the organizations that Form the basis oF this study
exhibit discourse between managers and other members and
have speciFic, avowed purposes. Three selected For
analysis are: a) new employee orientation sessions, b)
departmental training meetings, and c) "Lewittalks,“ a
special Forum For the exchange oF views, problems, and
suggestions For improvements involving the general manager
oF the hotel and rank-and-File employees. These three
events are analyzed in more detail in Chapter 2, using
Trice’s concept oF "organizational rites."
A large number oF linguistic elements and discourse
practices have been studied as maniFestations oF power in
discourse. Bumperz has worked with prosody, code
switching, and thematic structures. Spender C198Q, 1984)
has looked at lexico-semantics; Mahan’s research C1986a,
1986b) has involved glosses on arguments Csee GarFinkel
and Sacks, 1970, Far a discussion oF "glossing
practices"), as has Clegg’s C1975); and others working in
the discourse oF medical practice have Focused on turn-
taking and adjacency pairs Ce.g., Fisher & Todd, 1986;
West, 1984). The elements oF discourse studied here,
generally encompassing reFerence and identiFication, are
instances oF naming, categorizing, reFerring to, or
69
pointing out organizational members by other members or by
those same members self-reflexively.
Three arguments Favor examining linguistic
elements of reference as the basis for the discourse
analysis. First, there is an established group of studies
relating pronoun use to power and social status (e.g.,
Brown & Ford, 1961; Brown & 6iliman, 197S; Brown &
Levinson, 1979). Second, problems involving reference
ambiguity CAuer, 1984) and the very nature of reference
CLinsky, 1967; Levinson, 1903) provide speakers with
opportunities to manipulate discourse strategically so as
to maximize the possibilities of multiple interpretations
and inferences For ends desired by speakers. Third, the
naming or categorizing of people and objects in the world
implicitly exercises social power by virtue of introducing
to others what is noticeable, what is important, and how
it is related to other phenomena CBerger & Luckmann, 1967;
Habermas, 1979). van Djik Cin press) explains that
powerful members control discourse topics and attributions
about them more than do powerless members. Moreover,
Kenneth Burke has pointed out that where there is
ambiguity of identification "you have the characteristic
invitation to rhetoric" C1969, p. 25).
The last Factor to consider is assessment of the
degree to which understanding is shared by participants in
70
any instance of discourse under study. As Watson has
noted C1986, p. 96-97), Cicourel’s approach to
ethnomethodology calls For an appreciation of
participants’ understandings within the Framework oF their
own interaction and by reFerence to prospective and
retrospective consideration oF the unFolding discourse.
In other words, the participants themselves explicate
their own understanding oF their discourse, and the
analyst need only judge by the evidence within the
discourse itselF. In this view, meaning is not considered
solely From the perspective oF speakers, nor oF hearers,
nor even oF researchers, although researchers make
judgments oF the meanings discourse has For participants
en passant.
SpeciFic questions to be answered are as Follows:
Research Question #1: What is the nature oF reFerence
in discourse that characterizes the organizational
culture oF the two hotels under study?
Research Question tt£: In what speciFic ways does
reFerence in the discourse oF the organizations
contribute to available meanings that are understood
by members?
Research Question #3: How does discourse reFerence
aFFect the social order and maintenance oF control
desired by powerFul members oF the organizations?
71
Research Question #4:: Ulhat are the implications of
discourse for markers’ career patterns, particularly
for careers of ethnolinguistic minorities in the
workplace?
Toward answering these questions, a number of
preliminary and subsidiary issues must be addressed. This
study is organized so that pertinent stretches of
discourse in the two organizations are examined first;
then the discourse analysis findings are employed as
evidence in a critical discussion of the organizations’
cultures and the production and reproduction of their
social orders. Chapter 2 describes and Justifies the
methods used throughout the study. Also presented there
are the rationale for choosing the context and
participants of the research, the principle applied when
selecting discourse for analysis, and the peculiarities of
the discourse analytic techniques employed. Chapter 3
analyzes the discourse from three types of formal
activities in the hotels and presents arguments to explain
how reference is used to construct social meanings
concerning the relationships among individuals and groups
in the workplace. Chapter 1 presents a critical-analytic
essay on the relation of organizational members’ discourse
to their "life chances" CUJeber, 1946) in regard to
personal career trajectories. Structuration theory is
drawn upon heavily to explicate the ways in which talk in
72
the workplace contributes to institutional control and the
maintenance of the social order. Chapter 5 summarizes the
project, points out its major limitations, and identifies
implications for future research and practice in work
organizations.
73
CHAPTER 2
CUlncertainty ultimately involves choice.
— Gareth Morgan
Sources oF Data
Any effort at research has some particular focus,
whether it be a material object, such as a harbor seal or
a planet, or an intangible experience, such as a dream or
an idea. The focus of research, whether material object
or social experience, typically is examined, described,
analyzed, and otherwise poked and prodded so that some
claim to knowledge may be made in association with the
researcher’s experience of and with that phenomenon. One
central focus in this study is discourse that has occurred
among employees of two large hotels located in two major
urban areas an the West Coast. Another central focus is
the pattern of power relationships among employees of the
hotels. I take it as given that the power relationships
are asymmetrical because of the differential distribution
of authority, prestige, and privileges associated with the
position hierarchy. Ulhat is of interest is how the
asymmetries are reflected in the discourse of members and
how actual instances of discourse might contribute to the
creation and maintenance of power relationships.
74
Uhat Follows is, First, a description oF the hotels
where the discourse samples were obtained and their parent
corporation. A brieF justiFication For conducting this
research in the hospitality industry and in these particu
lar hotels is included. These remarks are Fallowed by a
discussion oF the research methods used, discourse
analysis and critical theory. The section on discourse
analysis includes a rationale which accounts For the
selection oF speciFic stretches oF discourse For
investigation. Critical theory is then assessed both as a
philosophical orientation to research and as a set oF
practices. This chapter concludes with an argument, in
harmony with Gareth Morgan's plea C1985b), For a
mulitFaceted, reFlective approach to research in the
social sciences based on a view oF epistemology that, too,
is multiFaceted and reFlective.
Settings oF the Data Sources
The Following description oF the sites where data
were collected is based on a three month period oF
observation, interviews with division heads, general
managers, and long-term employees, and inspection oF
company documents. Other inFormation was obtained From
standard reFerence materials; however, since the
corporation that operates the two hotels is privately
owned, no detailed inFormation on the Firm’s Financial
perFormance is available.
75
The parent Firm of the two hotels is an international
corporation that owns and manages approximately eighty-
five convention, resort, and transient residence
properties. The corporation is a family-owned subsidiary
of an umbrella holding company that is controlled by the
same family. Many of the properties managed by the hotel
corporation are owned outright by the Firm; others are
operated in joint partnerships; and still others are
operated under contracts For management, some of which
include provision of other financial services. Despite
the lack of financial performance data on the firm or any
of its properties Cas individual hotels usually are
called), some insight about corporate performance can be
gleaned from available information. The management self-
portrait displayed in the corporate public relations media
shows a firm with a history of rapid, trend-setting growth
in the hospitality industry and an expectation of further
expansion into new markets in the future. Currently, the
Firm is developing dozens of new hotels in the medium-size
range of 200 to 300 rooms, targeted at suburban or spur-
line communities and the middle-class vacationing Family
who do not need convention and banquet facilities.
Corporate offices are located in a major Midwestern
city, and regional vice presidents, who oversee operations
in six to ten hotels in each geographical region, report
to an executive vice president at the corporate office.
76
Corparate-uiide policy is made at the headquarters, but the
general managers of the individual hotels have
considerable latitude to innovate in the areas of
marketing, services, employee rewards, administration, and
equipment and decor. Their financial performance,
however, is closely monitored by corporate officers, and
the career paths of managers are likewise of interest to
headquarters executives.
As a condition of obtaining permission to conduct
research in the hotels under study, I agreed to mask the
identity of the firm and hotels and not to identify any
persons by name. The hotel general managers offered to
permit full disclosure of the organization’s identity if
the research conclusions ultimately so not appear inimical
to their corporate interests. Therefore, to help disguise
the identity of the corporation and the individual hotels,
I have given the pseudonym "Lewitt Hotels" to the parent
corporation and will call the two hotels concerned the
Airport Lewitt and the Marina Lewitt. Wherever neologisms
that incorporate the name of the corporation are
encountered, either in the data to be analyzed or in my
own descriptions of the sources of data, I will substitute
"LBwitt" in the place of the corporate name. For example,
the adjective commonly used to refer to an employee who
has been thoroughly indoctrinated to the corporation’s way
of viewing the enterprise and conducting business is to
7 7
add "-ize" to the name of the corporation; thus, a true
neliever is an employee who has been "Leujittized."
Lewitt started in the late 1350’s when a wealthy
family, seeking to expand their corporate interests,
antered the emerging hospitality field by purchasing a
small motor hotel on the west coast. Within a few years,
the family had acquired several other properties in the
region and began a legacy of continued expansion based on
nigh quality service to clients and innovation in hotel
architecture and guest amenities. Most of the current top
aperating executives of the corporation, including the
chairman of the board, began with Lewitt decades ago and
rose through the ranks, typically from low level clerical
nr laboring positions. Many managers stated in interviews
uith the author and research associates that the style of
bhe firm is to stay loose and promote innovation and self-
ieterminism; nonetheless, by now a firmly hierarchical
structure by which the headquarters entity exercises
aontrol over the field properties has emerged. For
sxample, less than two years ago, the corporate structure
jnderwent a substantial centralization by the
astablishment of new regional vice president positions,
narrowing the span of control and adding a managerial
layer between the individual hotels and the executives at
the headquarters.
70
Airport Lewitt is a high-rise structure that was
built twenty-five years ago and purchased by Lewitt twenty
years ago From a hotel corporation then in decline. It
is located at the entrance to a major international
airport and competes with an ever-expanding number of
newer hotels along the thoroughfare, usually called "the
strip." In comparison with the newer hotels on the strip,
Airport Lewitt appears dated, plain, unexciting. Of all
the strip hotels, however, it is located closest to the
airport. In addition, it maintains longstanding
relationships with several major corporations that provide
substantial numbers of transient travellers and convention
guests. The property, which is owned by a consortium of
show business celebrities, earns a healthy profit under
Lewitt"s management and name, and the executives of the
hotel believe they are leaders in the corporation’s
efforts to increase its share of repeat business in the
Face of increasing competition.
Three unions represent units of Airport Lewitt’s rank-
and-file employees. Any new employee entering a position
in a bargaining unit is required to Join and pay dues to
one of the unions. Negotiated contracts with all three
unions exist primarily to set prevailing wages for covered
positions and to identify union-sponsored Fringe benefits
and the employer’s contributions to the benefits. There
is no evidence of labor strife, and there has never been
79
an instance when arbitration was called on to settle a
labor dispute.
The nonmanagerial workforce appears to be drawn
primarily from unskilled and semi-skilled laborers from
immigrant minority groups. Although precise figures are
not kept, informal estimates based on visual surveys by
personnel office staff indicate that more than half the
line employees in maintenance, housekeeping, and kitchen
positions are Black or Hispanic. Most of the Hispanics
are first-generation immigrants from Mexico and Central
and South America. The acceptance of their limited
command of standard English is evident in the publication
of policy memoranda, employee newspapers, forms, and
various workplace signs in both English and Spanish.
Airport Lewitt is organized in a structure that is
typical of most large Lewitt hotels. A general manager
oversees all hotel operations through directors of the
following divisions: Rooms, Food and Beverage, Sales,
Engineering, Personnel, Finance, Executive Chef, and
Catering and Conventions. Those directors and the general
manager comprise the executive committee, who function as
a joint planning, reporting, and coordinating body.
Reporting to each of the division directors are managers
of various departments, such as housekeeping, security,
stewards, room service, etc. Also resident at Airport
Lewitt is the vice president for the region that includes
00
Airport and Five other nearby Leuiitt hotels, but not
Marina Lewitt. The regional vice president, who lives in
the hotel, in Fact perForms many oF the roles and
Functions oF a general manager in addition to his
responsibilities as vice president. At Airport Lewitt,
all members oF the executive committee are White except
the director oF Sales, who is Japanese-American; all speak
standard English except the CheF, a polyglot From Eastern
Europe.
Approximately twenty miles away From Airport Lewitt
is the glittering glass and steel tower oF the Marina
Lewitt. Built three years ago as one oF the showcase
convention hotels owned by the Lewitt corporation, Marina
Lewitt is typical oF the dramatic architectural style and
upper-middle class luxury that has come to characterize
Lewitt projects. A multi-storied atrium restaurant is
located just a glass wall away From pool-side. The upper
levels oF the residential tower overlook a park, marina,
beaches and the bay; meandering around two sides oF the
tower is a broad water course set in bermed, manicured
lawns. The Marina Lewitt is situated in a separate but
adjacent metropolitan area From that oF the Airport
Lewitt; unlike the Airport property, Marina Lewitt is
primarily a convention Facility and so Far has been
Financially successFul in that market. BeFore its third
Full year oF operation, For example, the Marina Lewitt
01
general manager was named corporate "General Manager of
the Year" For 1386, a recognition bestowed by the
executives at corporate headquarters based on overall
hotel performance.
While the management structure at Marina is the same
as that at Airport Lewitt, minus the regional vice
president, the rank-and-file employees are drawn From a
somewhat different resource pool. Most of the entrance
positions in low skill occupations are filled by newly
arrived workers from Southeast Asia, The Philippines, and
Central America. With few exceptions, for example, the
housekeepers are women from Asia and Central America.
Owing to the presence nearby of a state university, many
of the more skilled positions, such as front desk clerk
and bar and restaurant positions, are staffed on a less
than full-time basis with university students. The hotel
is not unionized, and there appears to be no serious
movement under way to initiate a campaign to gain official
recognition.
Rationale for the Settings
The differences notwithstanding, the communities
surrounding both Airport Lewitt and Marina Lewitt hotels
have several significant similarities. Both are within a
major urban belt that is known for its changing
demographic composition. Two decades ago, Anglos
comprised a majority of the population in the region; now
BE
no single ethnic or racial group is a majority; it is
estimated that by the end of the 1990’s the Hispanic
population will be a clear majority CPhillips, 1986).
This evolution is partly due to differences in birth rates
and interstate population migrations within the United
States. But it is more attributable to immigration from
Mexico and Central America— along with Miami and Texan
border cities, the area described here is one of the prime
destinations of immigrating individuals and families from
the south. In the case especially of Marina Lewitt,
however, there also is a rapidly growing community nearby
of newly arrived immigrants from Vietnam, Kampuchea, and
the Philippines. Hotel employment is attractive to these
new additions to the labor force because there are work
opportunities that do not require highly specialized
skills, U.S. citizenship, or high proficiency in standard
English.
This ethnolinguistic diversity is a central reason
for seeking discourse data in these two hotels. Another
reason is the unusual mix of similarities and differences
between the two properties: Although they are different in
their market orientation, history, physical plant
unionization, and presence versus absence of a regional
vice president, they nonetheless share the same parent
organization and higher levels of management, are guided
by the same top level policies, share the same network of
83
managerial personalities and politics, and conduct
business almost in the same neighborhood. It was hoped at
the outset of the study that these similarities and
differences would contribute to interesting qualitative
research conclusions; however, the similarities are much
more important than the differences.
For related reasons, the hospitality industry itself
deserves scrutiny as a source of workplace discourse. In
urban areas, such service industries as hotels and
restaurants are language-rich because of their hiring
practices and labor-intensive because of their task
structure. Certainly, hiring practices and task structure
are bound together in both direct and complex ways, but a
key feature of the hospitality industry is that it is an
apt representative of the deindustrialization trend in the
American workplace. In view of the growth of service
industries and their expected domination of Jobs in the
American economy, it makes sense to study interaction
among employees in that sector. An additional reason
Justifying the attention given the hospitality industry
here is implied in its name. Hotels are places that are
hospitable to strangers and have an inherent ethic of
tolerance toward persons different from their own staffs.
This openness plus the high number of hotels in urban
settings make them accessible workplaces where the
84
researcher can learn about discourse practices and work
relationships.
Discourse Analusis
As a Fundamentally interpretive procedure, discourse
analysis "is, necessarily, the analysis of language in
use" CBrown & Yule, 1883, p. 1). It treats discourse as
the dynamic process of attempts by participants to elicit
speciFic meanings in one another. Thus the Focus oF I
i
discourse analysis is on the linguistic and paralinguistic !
techniques participants employ in creating meanings.
Brawn and Yule characterize discourse analysis as
"primarily... ’doing pragmatics’," and as such "is
I
concerned with what people using language are doing, and J
accounting For the linguistic Features in the discourse as j
the means employed in what they are doing" Cp. 26).
The techniques oF "doing pragmatics" are not
uniFormly subscribed to by persons claiming to do
discourse analysis. In some cases Ce.g., Stubbs, 1883)
discourse may include both spoken and written texts, which
are analyzed in terms oF the exchange structure that is
evident. Functional categories are established For each
turn, such as Initiation, Response, Feedback, InForm,
!
I
etc., and the pattern oF assignment oF turns to categories:
is studied. In other examples, whole stretches oF
discourse are analyzed as strategy-laden interactional
structures CEdmonson, 1881). Still others Ce.g.,
B5
Coulthard, 1977) consider more traditional topics from
conversational analysis to be subsumed under the analysis
of discourse; although severely critical of Sacks,
Schegloff, and Jefferson’s uiork on speaker turn
allocation, Coulthard nonetheless incorporates it and
other rule-oriented concerns into discourse analysis.
The interest of the analyst and the goals of the
analysis are the basic difference between conversational
analysis and discourse analysis. Conversational analysis
has sought regularities in conversational data so that
procedural rules may be inferred and techniques of
interaction management may be better understood
CMcLaughlin, 19B5). On the other hand, discourse analysis
seeks to understand what actual people are doing with
language and paralinguistic features and how they
accomplish their communicative activities. Thus,
discourse analysis is necessarily sensitive to the socio
cultural context and the context of surrounding discourse,
as well as the type of interaction the participants are
creating.
The analysis of discourse presented here does not
seek to establish procedural rules but instead seeks
specific ways that interspeaker and intergroup reference
may be identified, categorized, and analyzed. The
categorization and analysis of items requires looking back
to previous discourse and forward to predictable and
06
probable discourse and to what is actually exhibited by
participants. By moving backward and Forward in the
transcribed interaction and by considering fully the
officially stated purposes of the interaction, as well as
its socio-cultural implications, inferences are drawn that
permit new categories of reference to be set forth. This
iterative process is more easily understood when the logic
for selecting the discourse to be studied is known.
Bracketing Discourse
Two crucial questions confront every discourse
analyst in the early planning of a research effort. The
first is, what is to be considered "relevant data?" CBrown
& Yule, 1083, p. 20). Will only spoken interaction among
members of the organization be admissible? Will various
Forms of written materials be discarded? Should analysis
be limited to only spontaneous conversation among members,
or should researchers engage members in talk and use
elicitation techniques to prompt linguistic evidence from
the members, as Labav C1072, pp. 204-5) has suggested?
Will talk conveyed by various media, such as videotape,
closed-circuit television, telephone, radio, etc., be
admissible, or should only face-to-face interaction be
studied?
The second crucial question is, What is the
appropriate unit of analysis for the study of discourse?
Conceivably, discourse can be studied in any actional or
87
interactional unit From phoneme to entire encounter and in
any conceptual approach From topic and cohesion to
persuasive strategy. A number oF scholars, however, have
proposed discourse models that speciFy some basic unit oF
analysis, and the units proposed are more remarkable For
their diversity than their similarity. For Labov, much oF
whose work Focuses on identiFying repeated patterns oF
phonological Features, the unit oF discourse analysis is
the utterance: "The Fundamental problem oF discourse
analysis is to show how one utterance Fallows another in a
rational, rule-governed manner" CLabov, 1972, p. 252).
Sinclair and Coulthard C1975) assert that the "exchange,"
a unit comprised oF three speaker turns in an ABA Format,
is the Fundamental object oF discourse analysis. Many
other scholars have approached the analysis oF discourse
by use oF Austin’s C1962) theory oF speech acts Ce.g.,
Bach & Harnish, 1979; Fraser, 1975; Searle, 1969, 1975;
Wish, D’andrade & Goodnow, 1980). While not all these
investigators claim that the boundaries oF a discourse
coincide with their chosen unit oF analysis, their
attention to a speciFic unit implies that discourse can be
described and explained by understanding language behavior
at that level.
The current study seeks to understand how a certain
linguistic process, reFerence. contributes to the
Formation and maintenance oF hierarchical relationships in
00
a workplace. fly claim is that to accomplish this under
standing, a satisfactory account of speaker meaning and
hearer meaning at the surface level of discourse is
necessary, as well as an offering of alternative meanings
that may be embedded in deeper, less obvious or noticeable
levels of the same discourse. Consequently, the issues of
what constitutes "relevant data" and what is an
appropriate unit of analysis must be discussed with regard
to these objectives.
Bracketing relevant data.
Goffman C1974) describes "bracketing" as the
conventionalized way of marking off a collectively
organized social activity. Bracketing has a temporal and
a spatial aspect such that events within the brackets are
set off from the temporospatial flow of events outside,
helping to identify the bracketed experience as a
particular kind of activity. Bracketed experience thus
provides a frame for including certain expectations and
themes while excluding others. Mary Douglas Cquoted in
Goffman, 1074, p. 252) refers to such bracketed experience
as rituals and observes that they "alert a special kind of
expectancy."
The concept of ritual has been highly developed and
widely applied by anthropologists CGeertz,, 1973; Turner,
1960). For Turner C1969, p. 52), rituals are elaborate
themes whose performance can take several forms in a
aa
sequential scheme. He refers to the episodes of
ceremonial behavior within a sequence as "rites."
Borrowing from and extending Turner’s use of ritual and
rite, Harrison Trice C1384) sets forth a taxonomy of
I cultural forms in which he differentiates rite from
ritual. Ritual, Trice states, is a set of standardized
techniques and behaviors that manage anxieties of various
sorts but which rarely have practical, intended outcomes.
Thus Haas and Shaffir C19B2) interpret medical students’
attitudes and behavior concerning practice licensing
examinations as rituals because those activities and
responses are viewed primarily as anxiety reducing
techniques. Grades on the practice examinations are seen
as indicators of social and professional competence by the
medical students, even though the practice materials and
l
procedures are contrary to the students’ past and future i
i
professional experiences. The kinds of questions asked ;
and the examination format are not directly related to the 1
diagnostic and treatment tasks the students perform as
■ )
professional workers, nor are they related to students’
past and future methods for demonstrating their
professional competence.
Rites, in Trice’s view, are "relatively elaborate,
dramatic, planned sets of activities" that consolidate I
I
cultural expressions and are carried out as social inter
action, usually involving an audience. flore importantly,
90
rites are contrasted with rituals in that the Former have
multiple social consequences CTrice & Beyer, 1985) while
the latter do not. The social consequences of rites are
both planned and unplanned, bath manifest and latent.
Trice and Beyer C1985) also point out that social
consequences might be both technical and expressive.
Technical consequences are those practical, concrete
outcomes the behavior is expected to accomplish by being
performed. Trice, then, would view the practice medical
examination as a rite in part because it is expected to
enhance students’ scores on the actual licensing
examination. Expressive consequences, on the other hand,
have indirect impacts that are intangible: practice
examinations thus also serve as ways to "convince
legitimating audiences of [[students’3 successful adoption
of the professional role" (Haas & Shaffir, 1982, p. 149).
Trice’s concept of rites is an effective answer to
the "relevant data" question of discourse analysis when
the discourse is viewed as the Joint transactional and
interactional use of language. Brown and Yule C19B3) set
forth two contrasting functions of language in use, the
transactional function and the interactional function.
Transactional language "serves in the exp'ession of
’content’" Cp. 1) and conveys propositional information,
while interactional language expresses social
relationships and emotional states. Brown and Yule align
91
transactional language generally with written texts and
interactional language with face-to-face talk. This
alignment of function with method of production has been
challenged as too narrow in scope and conception CQchs,
1979; Tannen, 1982); nonetheless, discourse analysts must
take into account the two aspects of language use when
selecting discourse to analyze, because participants’
meanings are always relevant to the analysis. If
discourse is viewed as a process in which participants
accomplish communicative work that is fundamentally
anchored in creating meanings, then analysis must consider
not only propositional information but also social
information— the context of the discourse itself, and the
wider context of social, historical, and spatial and
temporal settings in which the discourse occurs CBrown &
Yule, 1983).
Since rites are conventionalized, bracketed social
activities, they have an ostensive function that is known i
to participants. One consequence of their conventional
nature is that certain meanings are ascribed to behaviors,
including speech, in formulaic, predictable fashion. Thus
an instance of two cashiers counting a bank Cor cash
drawer) at shift change in the hotel restaurant has the I
i
ostensive function of balancing the shift’s cash receipts ■
and apportioning fair responsibility for maintaining the
cash balance. In addition, that rite has the expressive
92
Function of placing the leaving member in an off-duty
status and identifying the other as an on-duty member,
with all the rights and obligations that go with those two
different statuses. Knowing just this about the activity
permits substantial inferences to be made about the
meanings members may be expected to attach to their own
discourse and other behavior.
Examining discourse within the framework of rites
also offers the analyst parameters for deciding where the
relevant discourse begins and where it ends. A number of
ways have been proposed to decide where an analyzable
chunk of discourse begins and ends. Formulaic story
structures typically have conventional beginnings C'once
upon a time...") and endings C"and they lived happily ever
after"), but those cues are relevant primarily if one
wishes to study formulaic stories. Moreover, Brown and
Yule C19B3, p. 69) note that "the data studied in
discourse analysis is always a fragment," and the problem
is that speakers often do not provide overt signals
telling others when an appropriate fragment has been
completed.
Not uncommonly, appeal is made to the idea of "topic"
to identify appropriate stretches of discourse for
analysis. But "topic" is a conceptually ambiguous term
CBrown & Yule, 1983, p. 70), is often conceived
differently by each participant in a discourse CEdmonson,
93
19B1, p. 166-7), and in fact occasionally is not shared by
participants at all CCoulthard, 1977, p. 77). Rites, on
the other hand, are performed in bracketed circumstances
which carry clear indicators of beginnings and endings,
such as having special locations for their occurrence and
specially appointed times and prescribed durations. As
such, rites greatly facilitate and enrich the task of the
analyst, because the entity of topic— "what is being
talked/written about" (Brown & Yule, 1983, p. 73)— no
longer is a matter of discovery so that the analyst may
identify relevant discourse, but instead topic becomes an
open question to be resolved within the on-going discourse
analysis.
Trice and Beyer C1984) paint out several other
advantages to using rites:
The sheer magnitude of the events and their pre
planning should make them relatively easy to detect
and anticipate. It is extremely difficult for
organizational researchers to pursue true
ethnographic research....Rites and ceremonials
provide culturally rich occasions for intermittent
observation. Also, because they often occur in
public, with an audience present, researchers are
likely to be able to gain repeated access to such
events and to observe them with minimal research
effects....More important, their social consequences
link rites and ceremonials to other characteristics
of organizations that interest organizational
scholars and practitioners Cp. 655).
The organizational activities selected for study in
this dissertation fulfill Trice and Beyers’ definitions of
rites and ceremonials. The three types of activities
94
selected jare: Cal new employee orientation sessions,
comprised of information-giving, hotel tour, the rules
contract, and in-processing; Cbl departmental meetings,
comprised of announcements, training activities, rewards
and awards, and finale; and Cel Lewittalks, comprised of
introductions and invocation, orientation, problem
dialogue, and closing. Selecting these activities to
study discourse excerpts contributes a final benefit in
the research— enhanced replicability. Discourse obtained
and studied within the bracketed domain of rites may be
obtained by other researchers in similar contexts with a
higher degree of replicability than would be the case when
selecting discourse as a matter of convenience, by speaker
topic or discourse topic, or by random selection,
McLaughlin C19B5, p. E43) notes that with replicable
procedures, "the acquisition of Cnatural discourse! data
i
t
in noncontrolled settings may be the most appropriate way |
j
to proceed." !
i
Selecting the unit of analusis.
Lamenting the absence of discourse data in several
influential studies of interaction, Michael Stubbs C19831
points out that "there is...a lack of recognized and
accepted procedures for collecting, presenting and '
1
analysing conversational data" Cp. 2191. If by "data" is j
meant the pre-existing raw material collected from the
researcher’s experience which then is turned to for j
95
creation of arguments, theories, explanation, or the
development of new questions, then the data far this study
exist at three levels. First, at the level of the texts
that contain transcribed discourse are data comprised of
linguistic items. The specific itBms are described and
justified in Chapter 3, but they are selected for scrutiny
partly because of the scholarly tradition that points
toward them as potential items of interest and partly in
the spirit of "grounded research" by which the data call
attention to themselves by their mere suggestiveness.
Such items are interpreted by consulting prior and
subsequent discourse to discover their possible functions I
and meanings to participants and by identifying related
items in other discourse contexts.
The second level of data is the aggregated sets of
discourse selected for analysis, the corpus in toto. A
major justification for selecting particular stretches of
talk is set forth in the preceding subsection; however,
from the viewpoint of the overall research questions of j
this study, they were selected because they satisfied four >
1
criteria beyond the relevance issues discussed earlier.
First, they involve talk between managers and rank-and-
i
File employees; second, the sets of discourse contain !
talk that is primarily about work matters, including
changes, problems, and plans concerning the hotel’s way of
doing work and organizing for work; third, all formats of
96
the rites provide members opportunities For relatively
Free-flowing talk, which means that any member may
initiate topics and take multiple conversational roles;
and Fourth, the rites selected provide For examining
discourse between managers and newcomers as well as
between managers and members who have been in their
assignments more than Four months. These conditions
characterize the kind oF discourse that is needed For the
study oF how power and dominance become structurated
within the language use oF an organization’s members.
The possible meanings oF the corpus oF discourse were
interpreted by use oF Trice and Beyer’s C1904) Four types
oF multiple social consequences oF rites. Thus the new i
I
I
employee orientation rite was interpreted in regard to its (
maniFest technical consequences, its latent technical !
consequences, its maniFest expressive consequences, and
its latent expressive consequences. Respectively, these
consequences might be: Ca) maniFest technical— to provide
needed ihFormation so newcomers can Function appropriately
in the new setting; Cb) latent technical— to bring
productive capabilities up to standard promptly; Cc)
maniFest expressive— to welcome newcomers Formally; and
Cd) latent expressive— to introduce newco.iers to their
i
status in the social setting. This step of the analysis !
I
does not address the way the inherited, ingrained habits !
I
oF discourse are connected to systemic patterns oF |
97
relationships in the organization; however, it links the
microanalytic level of the linguistic term to the final
step of analysis, critical theoretic analysis by providing
the ostensive purpose of the rite.
The latter step addresses the third level of data,
that of the organization as integrated system of human
relationships. The data in this final step are the under
standings reached through the earlier discourse analysis,
coupled with knowledge of the organizations’ historical
development, dominant values and objectives, and
i
contemporary everyday practices excluding members’ I
discourse. The sources of these kinds of data at the I
i
macroanalytic level are interviews with top managers and
I
key nonsupervisory employees, examination of policy and
information documents, and extensive observation of
workers in the routine performance of their work. These
data at the third level are interpreted through
I
application of critical theory, described below. j
j
Before describing and justifying the use of critical j
theory, however, I will conclude this section an the
sources of discourse with a description of the techniques
used to record and transcribe the rites under study.
Recording and Transcription Techniques
All discourse data were recorded by the author using
a hand-held Sanyo Model M1012 cassette recorder. In each
instance of recording, permission first was obtained from
98
the manager responsible For conducting the activity that
constitutes the rite under study. In the case of
Lewittalks and new employee orientations, employee
participants were told at the beginning of the activities
that a researcher from a university was present in
connection with a university research program to learn
more about worklife at Lewitt hotels. Their permission
was requested for the researcher to attend and record the
activity; invariably, no employee objected to the
researcher’s presence.
McLaughlin C1985) points out the potential risk of
subject reactivity to taperecording equipment in the
collection of discourse data; the transcripts here,
however, indicate that there were no references made to
the presence of the researcher or the recorder once the
introductions had been completed. In other words, no
participant overtly indicated discomfort over the
taperecording activity. Reactivity might also be detected
by speaker reticence; a lack of participation or lack of
discussion about important, controversial, or personal
topics might indicate participant awareness of the
recording equipment. The transcripts show substantial
member participation, so much so that two of the rites
extended beyond their scheduled time, and one of the
others — a Housekeeping Department training meeting— at
one point turned into a free-for-all pillow fight.
99
Moreover, the researcher’s subjective experience of the
activities was one where participants took virtually no
notice of him or the recording equipment. This might be
the case because the recorder is a small, hand-held piece
of equipment, and every effort was made to have an
unobtrusive presence at the activity, such as placing the
recorder on the floor or otherwise out of plain view.
Field notes were written during the course of
attending rites and were used to assist in transcribing
the taperecorded discourse later. The major purpose for
making field notes was to identify speakers; an additional
purpose was to highlight certain interaction for later
scrutiny or to clearly spell out certain terms that might
not be audible or understandable when transcribing. The
author transcribed all tapes with only two exceptions: a
colleague transcribed one of the training sessions, and a
research assistant transcribed another. The author
audited and reviewed the tapes that were transcribed by
other persons to be certain that there is as much
consistency as possible in the transcriptions as regards
accuracy in representing lexical items, level of
paralinguistic features represented, and transcribing
conventions. The conventions used are described in a
footnote to Chapter 3.
Clearly, there are numerous opportunities in this
procedure for researcher biases of various types to
100
influence the final form of the transcribed data that
constitute what is analyzed in the discourse analysis.
Stubbs (1983) discusses the hazards for the transcriber in
the necessary effort of "changing the medium, from aural
to visual" Cp. EES) and concludes that there is no
technique of transcribing that salves all the problems of
researcher bias. One key control is consistency. For
example, to minimize the bias that may result from
selecting a certain degree of paralinguistic detail
instead of a finer-grained level or a coarser one, the
transcriber minimally needs to avoid varying the degree of
detail represented in transcripts. Certainly, close
attention must be paid to an aspect of discourse that has
relevance to the research questions, so that if the
project involves duration and occurrence of filled pauses,
all instances of filled pauses must be sought out in the
discourse and transcribed with equal attention to their
placement and timing in the tran-script.
Insofar as the linguistic feature of interest in this
dissertation is any unit of speech that constitutes an
instance of reference, the issue of level of detail is
less problematic than that of simple accuracy in
portraying what speakers in fact say. Stubbs (1983, p.
EES) warns of the tricks that one’s ears can play and
comments that even with the most sophisticated equipment,
transcribers are subject to "auditory hallucinations" that
101
can pose problems For practice and theory. Some
misinterpreted items are perhaps inevitable in any
transcription; however, if the data subjected to analysis
are extensive enough to offer reduncancies of the items of
research interest, wrong hearings should stand out by
their failure to conform to the general patterns exhibited
in the data. Such should be the case in the transcription
of Lewitt hotel discourse, which amounts to a corpus of
220 pages. Nevertheless, much of the interaction
recorded occurred in meetings with background noise and
often involved speakers with heavily accented speech. AS
a partial safeguard to clearly unwarranted
interpretations, any instances of speech that were not
readily audible to the transcriber are marked
"//inaudible//" or with the uncertain interpretation
enclosed in double slashes and question marks.
Critical Theoru Analusis
To understand how speakers in organizational
interaction establish interpersonal and group boundaries
through their use of linguistic techniques and discourse
strategies is a major step, but only a first step, toward
answering the research questions I have set forth. Issues
involving asymmetries of power bring the analysis and the
analyst face to face with the ideology of the community
and organizational culture and with the relative
distribution of individuals by hierarchical position and
102
class privilege, as well as with the specific histories of
the organization and its members. Thus it is necessary to
go beyond the sociolinguistic analysis of the discourse
and situate those findings in an overarching analysis of
the social structures of the organization. Such an
analysis must address the central issue of how instances
of employee behavior are constrained by the existing
social structures of the workplace and how, reciprocally,
the actual, in situ behavior affects social structures.
For several reasons explained below, critical theory is
employed to create a tentative knowledge about the deeper,
sometimes obscured nature of work relationships and the
ways that discourse is implicated in the creation and
maintenance of those relationships.
Critical Theoru as Philosophu and Technique
A brief review of recent literature in organizational
theory and communication studies demonstrates that there
is considerable confusion about what critical theory is
and how analysis using critical theory is done. Robert
Pryor C19813 embraces a critical theory that is
characterized by dialectical method, human praxis, and
interpretation. Slack and Allor, on the other hand, cite
"the political question of social power, linked with the
epistemolagical question of causality, Casl what
ultimately distinguishes the critical approaches" C1983,
p. 2153. Rogers C19823 distinguishes the critical school
103
by contrasting it to the empirical school. In his view,
the salient characteristics of the latter are quantitative
empiricism, functionalism, and positivism; and the essence
of the critical school is a philosophical emphasis, a
focus on social structures, a Marxist origin, and a
concern with power. Heydebrand C1983) sees critical
theory as one of four "moments," or points, in the
developmental sequence of research that leads to an
understanding of praxis— -in his sense of an ideal genre of
interaction. Forester C19B3), however, sees it as a
system of thought with a particular ontology and
epistemology that facilitates various strategies of
analysis. Similarly, Kersten C1986) describes "the
critical approach... in terms of how it views social
reality, what constitutes valid knowledge, and how such
knowledge is to be obtained" Cp. 759).
The differences among views notwithstanding, it is
apparent that critical theory is both a theorist’s way of
viewing the world and a set of techniques for entering,
understanding, and, ultimately, changing the world. To
describe briefly the central tenets of critical theory as
it is employed in this study, I return to its roots in the
Frankfurt School and particularly to the essays of Max
Horkheimer C1975). First the main features of critical
theory as philosophy are identified and, where
appropriate, their status in contemporary organizational
104
communication research is commented on. Then the various
methods of analysis using critical theory are described,
and the place of discourse in such an approach is
explored. This section concludes with a critique of
critical theory and a brief discussion of methods for
social research.
Realitu and knowledge.
Horkheimer C197S, p. 246) calls critical theory "a
new dialectical philosophy" whose interest is not only an
increase in knowledge per se but more crucially the
emancipation of humans from slavery. By slavery,
Horkheimer means the constraining, distorting, and
demeaning effects of social structures on the social
underclass. Four concepts characterize his view of
critical theory as a philosophy: totality, dialectic,
ideology, and critique.
The Frankfurt School under Horkheimer’s directorship
established an intellectual program that avowedly aimed to
counter scientistic specialization and the isolation of
disciplines. Those scholars embraced a system of thought
that was holistic insofar as it folded psychology into its
philosophical Marxism. The predominant view of material
reality held that there is a real physical world and real,
though intangible, sets of social structures and
relationships. Humans have differential knowledge of both
the physical world and social structures and
105
relationships, depending on the degree to which they are
inhibited by a traditional worldview Cp. 199); however,
all perceptions of natural and social phenomena are
ineluctably tainted by the influence of society as a
whole, because perception itself implicates a history of
the object perceived as well as the history of the
perceiver Cp. 200).
One aspect of totality, then, is the stitching
together of all phenomena so that the individual and the
social are conceptually interlaced, and the psychological
and philosophical are congealed in the interdisciplinary
program of the Institut fur Sozialforschung at Frankfurt
CJay, 1986). Another aspect of totality is its nature as
a program abjective. In this sense, totality represents
the seamless, classless society of truly free,
knowledgeable, self-reflexive individuals who represent
the end-state of critical theory as philosophy. Jay
C1986, p. 119) paints out that this optimistic aspect to
totality eventually turned cynical and evolved into
Adorno’s negative critique of the "antagonistic whole"
CAdorno, 1967, in Jay, 1986, p. 117). All the same, under
the concept of totality, subject, object, and theory are
not categories that might be privileged as to their
objectivity or conceptual uniqueness but instead are
problematic and opaque phenomena inhabiting the same
world. Similarly, there is a unity of theory and
106
practice, of theorist and target of the theory, such that
the researcher will always be a participant in that which
he or she researches.
Although totality has long since submerged from the
explicit Features of critical theory, its residue can be
recognized in the concern of critical theorists with study
of large-scale systems of dominance by recourse to
historical and psychological arguments CGerbner, 1983;
Real, 1904). It also is evident in contemporary critical
analyses that treat reality as both a concrete materiality
and sets of relationships that are constituted by the
practices of individuals CBenson, 1903). In addition,
traces of totality are reflected in the general acceptance
of multiple research methods by critical researchers and
their insistence that there is no confirmatory procedure
or authority for their conclusions CGeuss, 1981; Jay,
1986).
Kersten C1986, p. 771) highlights the implications of
totality For communication studies. 3he notes that while
scholars view communication either as principally
subjective and symbolic or as objectively observable
behavior, these perspectives may be viewed as aspects of
the same reality:
CGJbJective material conditions shape symbolic
interpretations and vice versa. UJhen people
communicate, the basis for their communication is in
the common experience that is based in historically
and socially mediated material reality.
Communication, then, is not an isolated activity,
107
symbolic or behavioral, butrather a part of the
totality of social relations and processes, and
should be studied as such.
The second philosophical concept, dialectic,
constitutes an epistemological technique. The Frankfurt
School sees society as inherently conflictual, and
dialectic is conceived as a range of tensions that are
always present in human society. At its fundamental
level, dialectic involves tensions among persons based on
the division of labor and class distinctions CHorkheimer,
1972, p. 207). For critical thinkers, tensions exist
between interpretations within the existing set of social
constraints and interpretations as pure idealism Cp. 208),
that is, interpretations from the frame of reference of an
ideal social system. Other tensions exist between nature
and society, between thought and practice, and "the
apposition between the individual’s purposefulness,
spontaneity, and rationality, and those work-process
relationships on which society is built" Cp. 210). As a
way of knowing, dialectic refers to the interpretation of
social phenomena and economic categories within the world
view of the existing order and the concomitant search for
their negatives, for contradictions that nullify those
interpretations. Horkheimer Cp. 208) comments that
the critical acceptance of the categories which
rule social life contains simultaneously their
condemnation. This dialectical character of the
self-interpretation of contemporary man is what,
in the last analysis, also causes the obscurity
108
of the Kantian critique of reason. Reason
cannot become transparent to itself as long as
men act as members of an organism which lacks
reason.
In contemporary terms, Pryor C13B1, p. 26) explains
that the apparent world is not what it seems, and the
critical theorist looks beneath the surface order and
stability to discover the contradictions of myriad
diversities.
Seeing that the social world is dialectical
means that the events and objects which make up
the social world in man’s perception and under
standing of them can be apprehended in terms of
contradictory, mutually dependent structures and
formations that negate each other and give rise
to observable objectifications of reality.
The dialectic is thus an interplay between what seems to
be and what lies beneath the apparent reality of social
life; and the underlying reality is often contradictory
and changing. The dialectical goal is to "identify those
human structures and formations which impinge on the
freedom of Reason to develop its potentialities. As such,
dialectical analysis focuses on those elements which block
free and open communication and on the material conditions
which give rise to distorted communication" CPryor, 1901,
p. 27).
The distortions, material conditions, and other obstacles
are part of a society’s ideology, and "the very heart of a
critical theory of society is its criticism of ideology"
109
CBeuss, 1981, p. 3). Beuss describes three senses of
ideology, the descriptive, the pejorative, and the
positive. Ideology in the descriptive sense is the
salient features of a socia-cultural system as it changes
over time. The pejorative sense of ideology is false
consciousness of social relations and institutions.
Positive ideology is a way of life that enables persons to
satisfy some of their wants and needs. The basic program
of critical theory has been the critique of ideology, or
Ideoloaiekritik. which focuses on the pejorative sense of
ideology. Beuss Cp. 13) identifies three ways that human
consciousness may be false: a) in virtue of its epistemic
properties, that is, if the beliefs and motives of such
consciousness are not supported by empirical evidence; b)
in virtue of its functional nature, which may include
permitting or fostering dominance and repression,
obstructing production, or masking social contradictions;
and c) in virtue of its genetic properties, that is
because of its origin and history. It is worth noting
that evidence of ideology in all three senses and of the
three types of consciousness can be found in everyday
discourse. A critique of discourse practices might be the
most direct way to extirpate false consciousness. Such a
project has been started by Michel Pecheux and his
associates CThompson, 1983), but their "automatic analysis
of discourse" has not generated much enthusiasm to date.
110
Jacob Mey C19B5) provides several examples of
Ideologiekritik. although most of the discourse he
examines is text from either print journalism or broadcast
news. His critique of the Danish Radio reports on labor
conflict involving oil distributors during 197B shows how
false consciousness is created in the language used by
reporters and editors. Topic selection, descriptions and
attributive statements, and ideologically biased labels
contributed to epistemic distortions; the media portrayal
of striking truck drivers as irresponsible wage hoarders
fostered the dominant position of employers and masked the
deeper contradictions of the issue; and the issue was
genetically distorted by television newscasters’
interpretation of historical origins of the conflict, in
which they characterized the conflict as contributing to
the existing national economic downslide, rather than
showing the soft underbelly of an ineffective national
policy of economic recovery.
Critique is the link between theory and practice in
the critical approach. It is not only a presentation of
societal contradictions through "an expression of the
concrete historical situation but also a force within it
to stimulate change" CHorkheimer, 1972, p. 215). Because
it acts on human consciousness, critique is a practice
that results, if validated, in change. Moreover, within
the totality of the social system, critique applies
I l l
equally to the theorist: "he exercises an aggressive
critique not only against the conscious defenders of the
status quo but also against distracting, conformist, or
utopian tendencies within his ouin household" CHorkheimer,
1972, p. 215).
Against the background of these four philosophical
principles— totality, dialectic, ideology, and critique—
knowledge must take account of both concrete, historical
specifics and the social totality. In addition, critical
knowledge must be not only descriptive knowledge that is
grounded in empirical observation but also an analytic
knowledge that is grounded in reflection. Benson C1983)
describes knowledge production in the dialectical process
as
like other forms of social production, shaped by
its context and by the way in which knowledge
producers are inserted into the social world.
The producers of social knowledge react to a
real world but not in a merely passive way Cthe
reflection view). Rather, through their
practices, shared within knowledge communities,
they actively shape the knowledge they generate
Cp. 334).
Knowledge is thus specific and historical, applicable
throughout the social totality, and always imprinted with
the ideology of its producer. In the case of critical
knowledge generated from dialectical critique, that
imprint proclaims a goal of emancipation from false
consciousness and the repression of social structures.
11E
Critical knowledge, however, is always tentative, partly
because it is about the dynamics of society and partly
because it always involves reflection. Horkheimer C197E,
p. EE7) confirms the tentative nature of critical theory
when he says it is "the unfolding of a single existential
judgment." More generally, Farrell and Aune C1979)
caution that critical theory is a project that as yet is
unfinished and is itself subject to critique and
modification.
Doing critical theoru.
Kersten C198B, p. 768) notes that it is not possible
to talk about "the" critical method. She describes seven
guidelines that direct critical research, which I
summarize as fallows. First, the choice of method is
determined by the nature of the object of inquiry.
Second, no method is privileged for use, but a pluralistic
approach to research is called for. Third, analysis is a
dialectical interaction between theory and data. Fourth,
observations and conclusions are located within the social
totality. Fifth, research categories, methods, and
concepts change as the social formations being studied
change. Sixth, research has an open and explicit concern
for social change. And seventh, principles of dialectical
method are followed as an aid in the collection and
analysis of data.
113
The last guideline refers to a set of questions
formulated by Sherman C1976, p. 62, cited in Kersten,
1986) that give shape to a dialectical inquiry. Sherman
proposes that critical researchers must ask about
interconnections between observed problems and the social
totality, about its history, conflicts, contradictions,
discontinuities, and negations. What Sherman and Kersten
do not emphasize is the central place in the critical
method of interpretation. Pryor C1981, pp. 30-31)
identifies two dialectical relationships involving
interpretation:
the dialectic between praxis and interpretation
and the dialectic between socially shared
meanings and individual creativity. With regard
to the dialectic between praxis and interpret
ation, critical theory argues that individual
behavior is a continuous stream of action and
that interpretations emerge from action.... Inter
pretation is dialectical; a tension exists
between permanence as manifested in shared
social understandings, institutions, and
structures and change, as implicit in individual
creativity .
Thus, to Pryor, the researcher must "investigate the
social constraints on individual creativity" Cp. 31).
More to the point, however, is the demand for
interpretation on the part of the critical theorist. To
be able to address the questions Sherman sets up as
research guides requires that the researcher interpret
historical data, documents, discourse, cultural artifacts
114
and practices, and other relevant aspects of the socio
cultural context.
Indeed, this type of interpretive work might have
been what Horkheimer had in mind when he described the
critical theorist’s output as an "existential Judgment."
Comment on Method
The lack of generally accepted methods is both a
weakness and a strength of critical theory. It is a
weakness for three reasons. First, without a standard set
of research practices, critical theory inhibits neophytes’
access to the full meaning of the theory and discourages
even its advocates from actually conducting research in
work organizations. With a few significant exceptions
CClegg, 1975; Heydebrand, 19B3; Hehan, 19BBa; Mey, 19B5),
most organizational studies taking a critical perspective,
including those focusing on organizational communication,
do not involve field studies of real organizations.
Second, lack of a firm methodological procedure leads
nonadherents to believe that critical theory has an ad hoc
epistemology CStevenson, 19B3). Third, the theory itself
is drawn into question: critics of the critical theory
approach usually are trained in the traditional research
premises that there is an essential correspondence between
theory and method and there is no relation between science
and value CBabbie, 19B6). A theory that welcomes multiple
research methods and holds social change as an axiom of
115
the research process cannot be evaluated within the
framework of the traditional, logico-positivist paradigm.
□n balance, the lack of specified procedures for
conducting critical research allows critical researchers
the kind of flexibility that is often recommended in
social science research CBabbie, 1986; Bantz, 1983; Miles
& Huberman, 1984). Moreover, viewed within the
assumptions of its own ontology and epistemology, critical
theory must invoke multiple methods to link microlevel
data to the macrolevel context of the social totality. It
seems inevitable that critical studies be developed around
two levels of argument, a primary level at which the data
considered are grounded in the concrete, specific actions
of individuals and a secondary level at which the
arguments concern institutional structures and societal
norms. It seems inevitable, too, that conclusions reached
at the primary level of data analysis be called upon as
evidence in arguments at the secondary level, so that data
and theory, observation and Judgment, individual and
society, quantitative and qualitative dimensions may be
joined under the dialectic of critical theory.
Although critical theory claims to situate analyses
in the social totality, a significant weakness in the
theory, as well as its practice and methods, is its
exclusive reliance on the workplace as the site par
excellence for the study of power relations, dominance,
116
and false consciousness. The latent Marxist assumptions
about the Fundamentally economic and materialist nature of
society often result in analyses that ignore important
relationships that are not primarily production-oriented,
such as those among family members, those between ethnic
or racial groups, religious institutions, etc. Complete
analyses should extend to related "existential Judgments"
about societal forms other than those where productive
forces are believed to be at work, such as the home, the
cafe or opera house, the place of worship, and other
institutions where ideology contributes to individuals’
consciousness.
Certainly, any analysis should be reflected back on
the situation of social science research, especially
studies that focus on discourse and communication. Morgan
C1903b), citing Gadamer, Rorty, and other scholars,
asserts that the quest for knowledge is a particular kind
of human practice which involves expressing ourselves and
our relation to the world. Therefore, research must be
under-stood as much as a human enterprise laden with
ethical, ideological, and political values as an
epistemological project. To evaluate such an enterprise,
Morgan says, we need to include in our research a more
reflective stance; and such critical reflection will "help
researchers orient their activities in ways that attempt
to take full account of the relations within which such
117
action is set, the consequences of that action, and of
alternative actions" Cp. 374).
IF knowing in the social sciences implicates the
researcher as part of that which is known, there is no
hope For achieving impartiality, pure objectivity, or
certainty. Consequently, social knowledge is not only a
matter oF discovering Facts but oF creating arguments and
Judgments. In so doing, insists Morgan, "we may be able to
replace the papular concern with eliminating uncertainty
with the idea that diversity, uncertainty, and perhaps
even contradiction are central aspects oF the process oF
knowing" Cp. 383).
Research in the natural sciences takes as its objects
oF interest phenomena that have no apparent will,
creativity, negativity, or selF-reFlexiveness. In such an
enterprise it might still be reasonable to seek nomothetic
relations among objects oF research, although the new
physics seems to be casting doubt on that view CCapra,
1977; Zukav, 1979). Social science research, however, is
decidedly diFFerent in that the Focus oF interest is human
beings who have an innate capacity For selF-reFlexive
action and creative initiatives. As such, social research
is a closed system that is continuously reForming itselF
within its own boundaries. Any research program that
seeks to establish unambiguously certain and objective
knowledge about causal relationships is inappropriate For
118
the social world. Ulhat is needed instead is research
approached that are sensitive to human agency, social
values, the production and reproduction of social
structures, and the inescapable involvement of the
researcher as system member. Discourse analysis addresses
one of the most fundamental human impulses, and critical
theory provides the requisite reflective stance so that a
more useful and humane knowledge might be generated.
A few examples stand out among the many critical
theory studies of organizations by virtue of their
examination of members’ discourse and the self-reflexive
concern for the researcher’s presence in the study. One (
1
such project is Stewart Clegg’s C1S75) study of power ;
relationships among construction workers at a "Northern
Town" construction site in Great Britain. Clegg analyzes
workers’ and managers’ discourse on the basis of the
argument "themes" and projected "pictures" of what the
organization looks like to the speakers. Using the themes
and pictures that emerge from the discourse, he formulates j
i
a set of rules that describe how the organization is j
constructed from the workers’ perspective. Clegg then
relates the inferred world views to an account of the
firm’s "mode of rationality": "CTJhe exercise of power |
i I
concerns the negotiation not of the taken-for-granted
structure of power, which is typically submitted to
without its overt exercise Clegitimate authority), but the
i
’understanding’ of its ’mode of rationality *** Cp, 100),_____ j
119
The researcher is present and accounted for at many
junctures in Clegg’s study. He frequently examines the
possibility of researcher effects, for example, "nearly
everyone I talked to on site was concerned to tell me that
this was not a ’good’ Job. Different reasons were
advanced by different people as to why this should be" Cp.
154). His assumptions and Judgments are openly
acknowledged and examined as part of the study:
There is no interest for me, or I suspect far
sociology, in pointing the finger at a specific
individual as a powerful person. The sociological
question concerns his identity as an ’historic
individual’— in Simmel’s reformulation of the Kantian
question, the interest resides instead in the
’possibility’ of any such person as a social being.
And so I have sought to answer the question ’How is
such a society Cas that which I have recorded on my
tapes) possible? ...The writing is a slice of me
making my sense for other people of some conceptual
and empirical problems as I found them Cpp. 155-156).
Clegg is self-reflexive without being self-glorifying.
His is an initial effort that can lead future critical
theory work in a productive and humane direction.
120
CHAPTER 3
I am arguing that the sense in which expressions
Cas opposed to speakers} can be said to refer to
things is derivative.
— Leonard Linsky
This chapter presents an analysis of the discourse of
managers and nonmanagerial employees in two Lewitt
hotels. This part of my project answers the first two
research questions and partially answers the third
research question, each of which were set forth at the end
of Chapter 1. To answer the first question, "What is the
nature of reference in discourse that characterizes the
organizational culture of the two hotels under study?," I
present examples from the transcripts that typify
reference practices of the participants in organizational
rites. Differences between the practices of managers and
nonmanagers are pointed out, and arguments are presented
for viewing the organizations as seeking highly
participative, egalitarian relationships among all members
but exhibiting contradictions in the ways that attitudes
toward both hierarchy and personal relationships are
captured in discourse reference.
lei
The second research question asks in what specific
ways reference contributes to members’ possible meanings.
One pivotal linguistic item, the use of "we," is examined
in detail, and a taxonomy of speaker uses of "we" is
proposed. Other related items are discussed briefly as
holding promise for further study.
The third question concerns the ways in which
discourse reference might be related to social order and
control in the organizations. This issue is addressed in
this chapter by a preliminary discussion of the power of
discourse to create possible worlds and the relation of
that power to personal and sectional power in
organizations. The linkages between discourse and power
and, by extension, social order and control are developed
more completely in the critical analysis in Chapter 4.
The fourth research question, which addresses the possible
influence of discourse practices on workers’ life chances
in the workplace, is addressed entirely in Chapter 4.
Four Principles of Power and Discourse
The analysis and conclusions reached in this chapter
are based on four principles about power and discourse,
which need to be discussed before the nature and types of
reference can sensibly be explored and the detailed
analysis of the token of reference,"we," can be
presented. The four basic principles are: Ca) Power and
discourse are linked through meaning and ideology; Cb)
122
power is social because it is created reciprocally in
relationships; Cc) discourse both enacts power and creates
power; and Cd) power is enacted and created in discourse
through multiple features and techniques.
Meaning. Ideoloou. Power, and Discourse
To function effectively in the world, persons must
have knowledgeable orientations to the world. That is,
they must have understandings of the world’s objects,
ideas, actions, and consequences of actions such that
their physical, emotional, and social needs are met.
While such under-standings are built up from subjective
experience and inevitably result in subjectively held
views of reality, possible meanings are generated by
social interaction CBerger & Luckmann, 1967). Moreover,
successful adaptation to social life requires individuals
to hold understandings that are confirmed or validated by
others in their culture and society. Berger and Luckmann
C1967, pp. 20-34) theorize that face-to-face interaction
is the prototypal form of encountering others’ meanings,
and that common footings are established, typifications of
reality are worked out, and objectivations are agreed upon
through an interpersonal negotiating process.
It is through the use of language that meanings
become conventionalized in semantic fields and that
possible meanings become "linguistically circumscribed"
CBerger & Luckmann, p. 41). What constitutes a socially
123
acceptable meaning for an individual thus depends on the
meanings held or advocated by others. In addition, van
Djik Cin press, p. 2) points out that "Cdliscourse and
communication are the primary vehicles for the expression,
enactment, social distribution and legitimation of
ideology and of social cognitions in general."
Foucault C1980, p. 131) makes the point that power is
implicated in the kinds of discourse that are acceptable
as true:
Truth is a thing of this world: it is produced
only by virtue of multiple forms of constraint.
And it induces regular effects of power. Each
society has its regime of truth, its ’general
politics’ of truth: that is, the types of
discourse which it accepts and makes function as
true; the mechanisms and instances which enable
one to distinguish true and false statements, the
means by which each is sanctioned; the techniques
and procedures accorded value in the acquisition
of truth; the status of those who are charged with
saying what counts as true.
It is the last political element that binds ideology to
the power/discourse relationship. What may be accepted by
persons in a society as appropriate, effective knowledge
of everyday life is constrained by the group interests and
interpretations of dominant individuals and institutions,
van Djik (in press, p. 3) maintains that ideology "may be
observed and studied in the sociocultural practices, such
as discourse, which are controlled by such ideologies."
And within discourse itself, says Foucault, are "internal
1 24
rules where discourse exercises its awn control" over
itself (1972; p. 220).
Pouier is Social
Power, as the capacity to constrain others’ possible
worlds and thereby to influence their thoughts, feelings,
or behavior, often is seen as a reciprocal process. This
second principle is reflected in social theories from
Ghandi to Giddens, from Parsons to Bacharach and Lawler.
Person A has power over person B to the extent that B
acknowledges A’s power and permits herself or himself to
be dependent on A. B may choose to satisfy her or his
needs by some other alternative, or, in the absence of
options, choose to forego satisfaction altogether.
Charles Conrad (1983) notes that employees may choose not
to honor formal constraints placed on them by their
employers; he makes the special point that it is a wonder
that employees almost never choose to flout those
constraints. Although formal authority gives employers
the nonreciprocal prerogative to use an ultimate sanction—
removal from employment— nonetheless, their power is based
on voluntary compliance by employees.
Taking a structuration approach to organizational
power, Conrad emphasizes another aspect to the mutual
nature of power. Employees, he says, observe, interpret,
and remember their own decisions about obeying or flouting
company rules; and their choices and interpretations not
125
only reflect the nature of the existing power
relationship, but also contribute to the creation and
reinforcement of power structures that are relied upon as
decision contexts in the future.
Discourse Enacts. Creates, and Recreates Power
Similarly, discourse may be thought of as both an
enact-ment of power and a contribution to the structuring
of power relationships in its very enactment. Giddens
C1979, p. 190) allows for both discourse that is
consciously ideological or manipulative and discourse that
reflects and contributes to relationships of power and
domination by virtue of its historically based and
unconsciously used forms. Thus a manager who tells his
subordinates that he will not tolerate a certain behavior
is consciously enacting the power relationship between
them by virtue of his calling upon his positional
authority and capability to invoke sanctions against
them. Obtaining employees’ compliance in that instance
contributes to their habituation of compliance on the
basis of the manager’s explicit authority. On the other
hand, each interaction between the manager and employees
in which the manager controls the topic initiation of
their discourse is a more subtle contribution to their
relationship of asymmetrical power. In many cases,
neither the manager nor the other participants is
conscious of the differences in privilege to initiate
126
topics in discourse, but those very differences are
powerful enactments of power. The phenomenon of topic
management by more powerful members is well documented in
the doctor-patient literature CFisher & Todd, 1903, 1906;
West, 1904), but is not well established for superior-
subordinate interaction, except when the members happen to
be of different sex Cvan Djik, in press).
Multiple Features: The ‘ ‘What Gets Said"
Control over what constitutes permissible topics in
discourse is only one of several ways that power is
enacted when individuals or groups interact. One major
class of discourse features in which power may be enacted
is comprised of various ways to constrain interaction. By
this I mean that power may be enacted and perpetuated by
virtue of what is possible to say, as opposed to how
discourse is realized. A second major class of power
features, then, is comprised of techniques or habits of
expression that limit or direct hearers’ meaning options.
In the first major class are such features as: Ca)
Control over the genre of interaction; Cb) privileged
speech acts; Cc) control over information; and Cd) control
over temporal and attributional perspectives. Genre
control is evident in discourse when a more powerful
member can characterize the interaction as serious
discussion or as a Joke, as a chat or a lesson, as
official or off-the-record. In the discourse corpus
127
described and analyzed below, numerous examples of genre
control stand out. For example, in the following excerpt,
a department head (EXEC) has been responding to various
complaints from lower level supervisors and employees at a
departmental meeting:
02277:RLB
(22) EXEC
(23) MGR3
(24:) EMP1
(25)
(26)
(27) EXEC:
( 28)
(29)
(30)
(31)
□ EP: EliC
What was wrong\?
It we=the changin’ of the shifts.
No, it was later that night, though. There was
no change.
C
You know, this is all,
this is all excuses. Change of the shi:ft is
slo:w time:, the guy was out there pickin’ at
his nose. Ha, this is=this is all a bunch a
crap. I am here...
The executive labels the content of what is being said as
"all excuses" and a "bunch of crap." Not only is he not
controverted subsequently, but the ensuing discourse shows
that participants take up his view of the situation as
their own.
van Djik (in press) discusses the asymmetrical uses
of speech acts between powerful and less powerful members
of organizations. Typically, such speech acts as
isa
directing, criticizing, threatening, instructing, Firing,
evaluating, and, in some extreme situations, even
questioning are the privilege of the mare powerful
members. van Djik Cp. 24) makes the key observation that
some speech acts which seem to offer options and
initiatives to hearers, such as suggestions, requests, and
recommendations, may also be used as mitigated forms of
commands.
• Control of critical information, the kind of
information upon which organizational planning may be
based and which otherwise might lead to outcomes of
serious consequence, typically is the privilege of a
ruling elite. The grapevine notwithstanding, access to
such critical information is limited, and its display
indirectly tells hearers that the speaker is more powerful
than they. In the following excerpt, the general manager
of the hotel is meeting with invited rank-and-file
employees in a regularly scheduled, confidential meeting,
called a "Lewittalk." The purpose of Lewittalks is
ostensibly to share information, problems, possible
solutions, ideas, questions, and so forth, directly
between line employees and the top executive. Here, the
general manager CMGR) tells the employees what is on the
drawing boards for new hotel construction:
159
03197:RLB:HTK:9
C04) MGR: Ahm, under
C05) discussion still in Lewitt La Jolla C5) th=they
C 06) are building that hotel but Lewitt has not sign
C 07) ed a deal to manage it yet. So they’re under
C 08) construction.=They’re still talking about the
COS) Lewitt Laguna, but=and I think they’re building
CIO) that hotel and we have not sianed a=a manaaement
Cll) agreement for it=it’s possible someone else will
C 15) do that/ . UJe are still talking about the Lewitt
C 13) Santa Barbara/. That’s fifty-fifty still. Cl)
C 14) Ah::m we are still looking for us a hotel in
C 15 ) Anaheim, a major hotel in Anaheim and a major
C IB) hotel in Los Angeles=but we haven’t signed those
C 17 ) deals yet. And that’s about what’s happening in
CIS) California.
The general manager dispenses information of national
scope and financial implications of enormous size with a
casual familiarity that conveys personal intimacy uiith the
details of these major corporate developments. Enhancing
the appearance of her control over the information is the
executive’s use of evaluative statements, such as "That’s
fifty-fifty still," and her authority to close out the
description as complete in lines 17 and 18. The
importance and interest of the information, as well as its
13C
exclusivity, casts the hearers in a less powerful
position, and the practical outcome of this information
display is to force the audience to become questioners
while'the speaker "remains the- dispenser of critical
information. In Fact, the s.ucceed~ing twelve turns of
talk, covering SS lines of transcript, are comprised of
six questions by employees and the six answers by the
executive.
Those sections of the transcript that are comprised
of. information-giving by - emplouess'and questions by the
manager, on the contrary, typicaliy. involve information
about work incidents, technical details, or low level
problems. In many instances the executive’s questions are
fashioned mdrs to lead the informdtion-giver to a
particular insight or conclusion than to elicit truly new
information. In such cases, exemplified by the Following
excerpt, the manager in fact ..is displaying her own
superior information and uses her cum questions to inform
the employees. Here the general manager has asked'how the
employees feel about the employee cafeteria. One employee
states that others in his work unit have not been very
happy with the quality of food in the past couple of
months. The manager CMGR) questions him CCS) and another
employee CIS):
03137 RLB: HTK : 27
C07) MBR: Uihat are some examples?
COS) 08: I’m sorry, I //inaudible-, drops voice//.
C 09) MGR:
131
Wait just a minute, Todd;. Are we talking...
CIO)
C
Cll) 08: I think it
C 12 ) may be due to the Fact that the broiler is off.
C 13 ) MBR: 'Kay. Is=ha=are you talking about largely the
C 14 ) three to eleven shift?
CIS) 08: What’s that?
CIS) MGR: Three to eleven. C4) Yes, Kevin?
C 17 ) 19: I have a Few examples From the beginning of
C 18 ) February where ah employeees eating in the
C 19) cafeteria ah: in cafeteria they ah ah they got
C20) sick. That was I don’t know if it was the Food
C21) the food or something else.
C22) MGR: Ahm: when did they get sick?
C23) 19: The beginning of February.
C24) MGR: How long after they’d eaten?
C25) 19: About two hours/.
C26) MGR: How long does it take For people to get food
C27) poisoning\?
C20) 19: Ah I’d say one of the employees actually took a
C29) couple of hours but I don’t know...
C 30 ) 1
C 31) MGR: Ya=I=I don’t think
C32) we can actually say that the food that was
132
C33) within two hours was the problem, because in
C34) in actuality food poisoning is from six to
C35) eight C...1
The manager goes on to describe the conditions under which
food poisoning occurs and finally dismisses the topic by
saying that there’s been a lot of illness around lately
that is hard to avoid. The next turn is by an employee
who compliments the cafeteria staff for doing a good Job.
Exchanges such as the foregoing reflect not the superior
information of the employees but the manager’s use of
employees’ information to shape and direct the interaction
in ways that turns back on the employees by teaching them
or by informing them of the manager’s views on the topic.
Powerful members also enact power in discourse by
controlling the temporal and attributive aspects of a
discourse. Control over the temporal aspects occurs by
initiating and concluding topics, controlling the sequence
of turns at talking, limiting the duration of turns, and
even scheduling the occurrence of the interaction. Each
of the six rites Ccomprising three broad types) from which
discourse for this study was obtained and transcribed was
scheduled by a manager; in all but two of the rites, the
scheduling was done by or for a hotel general manager, and
in the other two, the scheduling was done by or for a
department head. A typical example of how a participant
133
can control temporal aspects of discourse occurs when the
general manager explains that the cafeteria grill cannot
be kept open during slow hours because with a reduced
staff the grill would have to be left unattended:
03197:RLB:HTK:26
C51> MGR: ...And
C52) because of bonding, that’s not passible. I
C53) mean 1=1 could give you a song and dance about
C54:) that, but it wouldn’t be true. I mean that’s
C01) really the reality.=Yes, Todd.
Here, she states a conclusion which, if it is to be
refuted and the topic is to be maintained in discussion,
invites a challenge to her version of reality and her
credibility and authority. No employee who would dare to
initiate such a challenge, however, has an opportunity to
do so because she immediately directs the conversation to
a designated next speaker at line 01. The absence of any
rhythmical beat between "reality" and "Yes" are evidence
that the speaker is directing the conversation away from
the limited hours of grill operation in the employee
cafeteria. Indeed, the next speaker, Todd, switches
topics by discussing a survey of employee views about food
quality.
134
Control over turn allocation is only one of several
ways that temporal aspects of discourse may be exploited
by powerful participants. Thus van Djik Cin press, p. 36)
notes that generally "the more powerful speaker has Cmore)
control over the allocation or appropriation of turns,
over topic selection and change, and the overall
organization of talk, including initial greetings, the
nature and length of side sequences, and various types of
closings.“
Control over attributive aspects of the discourse
means that some speakers are privileged to make
evaluative statements, particularly about the on-going
talk itself, while others seem not to be so privileged.
The distribution of the privilege to evaluate is not
simply a matter of one group having exclusive rights to
make evaluations and others having no right to make
evaluations; the transcript data indicate that the
privilege is distributed so that evaluations of major
programs, highly placed individuals, and objects, plans,
or activities that have large-scale economic impact are
the privilege of one group, those in high managerial
positions. Evaluations by nonmanagerial employees are in
abundant evidence, but they uniformly concern local,
detailed topics of low economic, social or organizational
impact. In the following example, a new employee CNEf1P4)
at an orientation session offers an evaluative statement.
135
Her comment is followed by the employ-ment manager’s
description of the corporation’s historical development.
12226:LAX:QRN:4
CIO) NEMP2: Ya=I already C2) in the short
C11) time I’ve worked here I’ve enjoyed it.=Every-
C12) body’s so: nice.
C13) EMGR: Oh:. good. Good. Ahm, this next little section
C14) is called=this is my Favorite file, is called
... ...Cdescribes videotape!
C22) ...Lewitt, who started it and ah C2) where it’s
C23) gone since nineteen fifty-seven. It’s quite
C24) surprising.
The evaluation statement on lines 23 and 24 is not
obviously a personal response to the Lewitt story; rather,
since it just precedes the presentation of a videotape
showing the highlights of the corporation’s growth during
its First 25 years, it is a general attribution about the
story that may shape newcomers’ expectations of what they
are about to see.
Another Form of attributive control appears in the
almost exclusive managerial privilege to make meta-
communicative statements*. Commentary that evaluates the
on-going talk occurs throughout the transcript corpus, but
in almost all instances it is a department head or general
136
manager who characterizes the discourse. Three major
exceptions occur when-. Ca) employees comment on each
others’ participation, often by joking or cheering; Cb)
employees characterize the whole conversation to that
point, which is done as an aside, sotto voce; or Cc)
employees attempt to quiet each other by shushing or
asking for quiet.
Multiple Features: The "How of Sauing"
These content-based features of power in discourse—
control over genre definition, speech act types,
information, and temporal and attributive aspects of
interaction— reflect the "what gets said" dimension of
discourse. By this I mean they are features that
constrain what is possible for any particular person to
contribute to the discourse. In a direct way, power is
evident in the distribution of these control features and,
thus, over the range of meanings that are available to
hearers in an interaction. Possible meanings are also
constrained by speakers’ techniques of expression.
Instead of controlling the content of discourse as it is
created by all participants, a powerful person in this
instance controls how he or she produces discourse,
thereby limiting hearers’ meaning options or directing
hearers to understand Cor not to understand) talk in
particular ways.
137
This latter way of constraining hearers’ meanings is
grounded in how the talk is performed, and it depends on
conversational presupposition and implicature. Before
briefly considering the nature and function of presup
position and implicature, however, it is worthwhile
enumerating the major discourse features that can be
allocated to this second category of evidence. Perhaps
most central is the feature of lexical choice, which
reflects the field of discourse, determines the tenor of
discourse, and contributes to the register of discourse
(Gregory & Carroll, 1978D.
Gregory and Carroll describe "field" as the language
user’s purposive role, "what is ’going on’ through
language" Cp. 7). Field goes beyond topic and describes
the social function of the discourse at hand. Tenor may
be thought of as the linguistic reflection of the
relationship between speaker and hearer, apparent in such
qualities as formality and warmth, and such speaker-hearer
roles as lecturer-audience, entertainer-audience,
evaluator-subject, and so forth. Register, which
encompasses field and topic, refers to the language
variety that is used for a particular social purpose, such
as presenting the annual report to stock-holders or
awarding prizes at the company picnic. By choosing to
address hearers as "you guys," for example, instead of
"ladies and gentlemen" Cas frequently is the case in the
138
hotel transcripts), a speaker structures audience
expectations about what kind of activity they are
participating in, uihat kind of relationship obtains
between them and the speaker, and what the speaker is
doing with his or her talk. van Djik Cin press, p. SB)
notes also that lexical vagueness, obscurity, the use of
jargon, and lexical variation all contribute to restricted
comprehensibility and thereby restrict hearers’ access to
communication. The use or avoidance of politeness terms,
epithets, compliments, qualifiers, and hedges also
structure hearer understandings.
Besides lexical choices, syntactic and other choices
also influence the field, tenor, and register of
interaction. Long, complex utterances or sentences and
strict grammaticality imply formality and distance between
speaker and hearer Cvan Djik, in press). Paralinguistic
features, such as pause duration CGood & Butterworth,
1980), intonation (Levinson, 1983), and other prosodic
elements (Gumperz, 1982b) also influence the passible ways
that the literal meanings of discourse can be interpreted
by hearers.
Choices among all these linguistic and paralinguistic
resources are made by speakers against a background
awareness of the potential for conversational
implicature. Implicature refers generally to methods of
creating meanings that go beyond what is literally or
139
conventionally interpretable From the cues available in
the immediate discourse. Levinson C1983, p. 97) describes
implicature as providing "an explicit account of haui it is
possible to mean more than what is ’actually said’...."
Implicature is not a logical entailment or logical
implication of words or utterances; rather, it is "an
unstable, context-specific pragmatic overlay on a core of
stable semantic meanings" Cp. 99). According to Grice
C1975; 1981), implicature works by inferences based on
both the available literal meanings and conventional
knowledge about how conversation is conducted. The
literature on implicature is large and complex and need
not be reviewed in detail here; an excellent summary is
contained in Levinson C1983). What is relevant to this
discussion of power and discourse is the principle that
meanings are generated in a process that resembles more a
dynamic social and linguistic gestalt than a fixed formula
based on a lexico-semantic code. It also is important to
remember Grice’s C1975) caution that implicature usually
is 1 1 intuitively grasped," rather than consciously worked
out.
Like implicature, presupposition has a complex,
largely philosophical literature that is only tangential
to the needs of this study. Levinson C1983) points out
the major differences between the technical use of the
term in the philosophy of language and its ordinary
140
language uses. I will use presupposition in the
nontechnical sense of background assumptions about the
nature of language, the social setting, the relationship
among participants, and the world generally— assumptions
participants invoke in speaking and in understanding one
another. Such assumptions are prompted in hearers by
speakers’ display of the linguistic and paralinguistic
features just discussed. Speakers’ presuppositions are
antecedent to intended meanings, and thus are prompted by
speakers’ perceptions of the discourse field, tenor, and
register, either as it has occurred so far or as speakers
wish it to be.
Reference and Discourse
Implicature and presupposition, along with the
linguistic and paralinguistic resources speakers and
hearers have at their disposal, are important because they
are used to accomplish the work of discourse. That work
generally is for speakers to bring their audiences’
attention to some desired abject, idea, action or other
state of awareness. To speak of how language is used to
direct others’ attention is to speak of reference CLinde,
1379). Peter Strawson C1371, p. 17) describes reference
this way:
One of the main purposes for which we use language
is the purpose of stating facts about things and
persons and events. If we want to fulfil this
purpose, we must have some way of forestalling the
question, ’What Cwho, which one) are you talking
about?’ as well as the question, ’What are you________
141
saying about it Chim, her)?’ The task of fore
stalling the first question is the referring Cor
identifying) task. The task of forestalling the
second is the attributive Cor descriptive or
classificatory or ascriptive) task.
Thus referring identifies rather than describes.
Also, referring is something that speakers do and not
something that words, phrases, utterances or sentences do
CLevinson, 1983), and it is not the same as meaning. On
the differences between referring and meaning, Linsky
says:
...referring to someone is an action; meaning
someone is not an action. As an action it can be
right or wrong for one to perform. Thus it can be
wrong of you to refer to someone; but not wrong of
you to mean someone. It can be important or
necessary that you refer to someone, but not
important or necessary that you mean someone. One
can intend to refer to someone, but not intend to
mean him C19B7, p. ISO).
The discussion of reference in this study is not
intended to confront the major philosophical problems of
logic, reference, and description. For a succinct review
of the major issues involving reference, see Levinson
C19B3). The central works on reference as a concern for
logicians and philosophers can be found in Frege C195E),
Beach C19BS), Kripke C197S), Linsky C19B7), Quine C1961),
Russell C1958), and Strawson C1971). Linsky C1967)
presents an excellent analysis of earlier arguments by
Meinong, Frege,.and Russell. The present discussion also
is not aimed at contributing to the research on reference
14S
in sentence grammars or semantics Ce.g., Cole, 1978;
Hawkins, 1980; LakoFF, 1976). Instead, it conceives oF
reFerence in the practical, everyday sense oF using
language and paralanguage to direct hearers’ attention to
some intended entity the speaker has in mind. It is Just
this sense oF reFerence that Auer C1984) has in mind when
he explores how speakers and hearers turn reFerence into
discourse problems and then devise ways to repair the
discourse.
Donnellan C1966, 1978) speaks oF "speaker reFerence"
and "semantic reFerence." The distinction is based upon
an assumption that "people reFer and expressions reFer"
C1978, p. 47), and that the kinds oF reFerences people can
make are limited to concrete objects, named individuals,
or previously mentioned entities. Semantic reFerence, on
the other hand, is reFerence in the logician’s sense, the
reFerence that expressions "make" without human agency but
which satisFy truth conditions For logical
interpretation. Speaker reFerence occurs only in
connection with deFinite descriptions. DeFinite
descriptions are those For whom the speaker has some
particular person or object in mind, such as "the manager
who started the training meeting." IndeFinite
descriptions, such as "managers who start training
meetings" hallmark semantic reFerence.
143
Donnellan presents persuasive arguments to show that
"uie cannot divorce speaker reFerence from semantic
reference" C197B, p. 68). His view, however, retains the
crucial test of truthfulness to determine if reference
occurs. In terms of reference in ordinary discourse,
these distinctions are not useful because the assumption
of truthfulness, awareness, and probity on the part of
speakers and hearers cannot be assumed. Moreover, Linsky
C1967) points out that the question, "To whom does the
phrase ’so and so’ refer?" is derivative of the question,
"Ulhom are you referring to when you use the phrase ’so and
so’?" If indeed referring is something that speakers do
and not something that linguistic units do, then
DonnelIan’s marriage of speaker and semantic reference is
annulled, and the issue of truth or falsity of utterances
does not determine whether reference has taken place.
More importantly, referring need not be restricted solely
to the act of calling hearers’ attention to entities by
definite descriptions; any communicative intention on the
part of a speaker may call forth referential expressions.
Reference is not limited to the use of descriptive
phrases in the form the/aCn) Cmodifier) noun Cmodifier),
like "new managers who start training meetings." A full
taxonomy of reference types has not been presented as the
centerpiece of scholarly study; however, several lists of
possible types are available or implied in existing work C
144
Chastain, 1975; Geach, 1962; Labov, 1976; Strawson,
1971). In general, reference is accomplished by speakers
when they use one or more of the following linguistic
forms: Ca) names, including proper names of persons and
objects and "quasi-names," i.e., objects which have "grown
capitals" CChastain, 1975) like "the Sultan of Swat"; Cb)
definite descriptions, e.g., "the videotape we saw
yesterday"; Cc) indefinite descriptions, e.g., "videotapes
about the hotel’s history"; Cd) epithets, e.g., "the
bastard"; Ce) deictic expressions, e.g.,
"this/that/these/those/here/there/now," etc.; and Cf)
personal pronouns.
Obviously, a great deal of discourse is not
referential. For example, verb phrases express action or
states of being and relation; and other parts of language
connect reference to the discourse and help to make
discourse comprehensible. The referential forms listed
above, then, comprise the "what this piece of discourse is
about," and the talk that is not reference comprises the
commentary on what the discourse is about.
The Culture of Reference in Two Hotels
These six reference forms— names, definite descrip
tions, indefinite descriptions, epithets, deictic
expressions, and personal pronouns— may be examined in the
discourse of a group to make claims about how the group
typically uses reference. By this I mean that groups
145
might be classifiable on the basis of their habits of
referring in discourse as: Ca) firmly structured and
rigidly divided versus loosely defined and flexibly
interacting in terms of hierarchy; and Cb) formal, distant
and cool versus informal, solidary and warm in terms of
personal relationships. To make claims about hierarchy,
observations are needed of the relative degree to which
definite descriptions reflect positional distances among
members. It also would help to know how the naming of
persons reflects organizational titles, how references to
problems and accomplishments typically are shared by
members at different hierarchical levels, and so forth.
To make claims about relationships, the use of first names
and inclusionary pronouns like "we" and "us," the presence
of epithets, and the expansiveness, discursiveness and
degree of participation in discourse of members at various
organizational levels need to be examined. The following
assessment is based on anecdotal evidence contained in the
transcript corpus, which includes talk by managers at all
organizational levels and nonmanagerial employees in rites
where managers and nonmanagers talk together.
Reference and Hierarchy
Hotel general managers report to regional vice
presidents who typically have their offices in one of the
hotels in each region. Regional vice presidents in turn
report to an executive vice president at the corporate
146
head-quarters in Chicago. Within each hotel, general
managers supervise seven to ten division directors. A
single division might have several departments, each with
a department manager as head, as is the case of food and
beverage division; one division, personnel, has no
subordinate departments at all.
In the larger divisions of the two hotels under
study, there are four levels of management between first
line supervisors and division heads. For example, the
beginning supervisors in housekeeping, called leaders,
report to an assistant executive housekeeper, who in turn
is subordinate to the executive housekeeper. The
executive housekeeper, a department head, reports to the
director of rooms division. Thus, there is a possible six-
level position hierarchy in each hotel.
The position hierarchy is somewhat mitigated,
however, by the fact that many of the nonsupervisory
positions are partially compensated through guest tips.
As a consequence, the Compensation hierarchy does not
align perfectly with the position hierarchy. Among the
most stable occupations in the hotels are bellman and
server in the quality restaurants. Both the employees and
their managers report that the workforce stability in
these occupations is partially due to the high level of
pay based on tips.
147
Awareness of hierarchy is not readily evident in the
transcripts. Position titles are not used to refer to
individuals, either as forms of address or as reference to
persons not present. With the exception of new employee
orientation, in the entire corpus there are only two
instances of reference to persons not present by using
their position title. In one instance, a manager makes a
position reference by saying "Julie B ’s supervisor."
Since Julie B is a division head, it is possible that
the first line manager does not know who Julie B ’s
supervisor is. In the second instance, a division director
refers to the assistant personnel director, a person with
less than three months of employment at the hotel; later
in the same rite, moreover, he refers to the assistant
personnel director’s supervisor by name. Thus it appears
that the two tokens of reference by position title are
anomalous, possibly occurring because the speakers cannot
remember the names of the persons referred to.
As to new employee orientations, the following two
excerpts show how the employment manager CEHGR3 refers to
hotel officials by their position titles, but most
commonly affixes the names of the persons referred to:
148
12226:LAX:ORN:27
C34) EI1GR: F Ive . Heh=h::=heh=heh=heh. H::=it’s I have
C35) bought so many name tags from our timekeeper,
C36) Clara, she Just laughs at me now.
12226:LAX:ORN:18
C17) EMGR: So. Ah: what happens at the awards banquet is
C18) our controller, Tom L , comes up: to the...
In view of the Fact that the employment manager in new
employee orientation is talking with persons who do not
know other hotel employees by name, it is only appropriate
that officials be referred to by position titles for the
reference to be successful.
If persons are not referred to by position, what
naming practices are typical of interaction in the rites?
There is an almost exclusive use of first names by all
participants. Managers at all levels refer to other
managers, both above and below them in the position
hierarchy, by first names. Here, a first line supervisor
CMRGDin the coffee shop refers indirectly to the division
director CMaury), two levels above her:
149
0EE77:RLB:DEP:EMC:EE
C33) EMP11 No?
C34) MGR1: No. Tickets that Maury shows us with orchid
C35) cafe stamps oh maybe Five times on it, you...
The division director was seated nearby during that
exchange. MGR1 also uses his first name when addressing
him directly CMaury = EXEC):
0EE77:RLB:DEP:EMC : 49
COB) EXEC: I’ll take care of it.
C07) MGR1: Maury, it’s serious. Very serious.
Nonmanagerial employees refer to one another by first
names, and they routinely refer to managers by their first
names as well. Here are two examples of reFerence to
supervisors by their subordinates; the first is indirect
reFerence by an employee CEMP1) to a second level manager
CIris), and the second is a direct address by an employee
CEMP0) to her supervisor CMGR1):
0EE77:RLB:DEP:EMC:45
C01) EMP1: No, wasn’t=wha=I Cl) I mean Cl) this man, he
COE) was^he was upset but yet Cl) he ahm Iris went
C03) =’n’ talked to him, but Cl) I put the...
150
0EE77:RLB:DEP:EHC:5/6
C50) HGR1: beautiful looking Cl) mint every day.. The chef.
C 51) C
C5S) EflPB: How is that
C01) ordered, Tess?
COE) MBR1: I’m sorry?
C03) EMP8: Is that ordered when you...
In the following example, an employee benefits
manager CPLE) is briefing a meeting of housekeeping
employees and refers to the assistant personnel director
CHanna) and the personnel director CJulie):
0SE77:RLB:DEP:HSK:8
C16) PLS: ...Okay? And if you don't understand
C17) or.don’t agree with it, come down make an
C18) appointment with Julie or Hanna they’ll be glad
CIS) to discuss it with you.
Thus first names are the rule among employees, between
employees and managers, and among managers. A clear
exception is evident, however, as concerns officials above
the division director level. In both hotels, the general
managers consistently are referred to by their last names
only, even when managers are speaking to managers. The
firmness of this exception is illustrated by the following
151
excerpt, in which a First line manager CM6R1) is
describing to her employees how critical it is to keep
accurate timekeeping records oF work breaks.
02277 : RLB: DEP : EMC -.13/14
(50) MSR1: Just a
(51) minute. This is a leoa:1 Cl) document. This is
C5E) the document that we take to court. This is:
(53) gone over by Mrs.:=I mean Julie B____’s super-
(01) visor. This is very important=it’s not a joke.
In starting to say Mrs. some person, the manager most
likely reFers to the general manager oF the hotel. This
is so because the general manager is the only married
woman in upper management and is the only oFFicial other
than Julie B likely to review schedules and time cards
that are part oF a controversy. By switching mid-
utterance From "Mrs." to "Julie B ’s supervisor,"
however, the manager also might be displaying a
misunderstanding oF the chain oF command: Julie B. reports
to the general manager Csee the discussion oF the use oF
position titles, above). An alternative explanation is
that the manager reFers to a division head’s supervisor,
rather than naming that person, because the position
reFerence implies greater distanciation From the audience
oF employees and greater authority. In either case, this
152
is a graphic example of the naming rule boundary between
division directors and the top executive position of the
hotels.
Rules for naming practices, as Ervin-Tripp C1972)
applies them and they are used here, are descriptive
rather than prescriptive. In Ervin-Tripp’s model of the
address system in a western American academic community,
use of first names entails an addressee who is younger
than the speaker or of lower rank than the speaker or, if
much older than or higher rank than the speaker, has given
a first naming dispensation to the speaker. Applied to
Airport Lewitt hotel, the first name rule must exist
because of an implicit dispensation given to lower level
employees in the use examples set by managers and longer
term employees.
The name rule boundary also exists at the other
hotel. A program manager CPHGR) stands next to the hotel
general manager during a training session where a new
program is being introduced to a large group of employees
and managers. The general manager CMrs. E ) has Just
introduced PfIGR in her new role as the manager of the new
program, repeatedly referring to her as Maryanne.
153
03137:LAX:TNG:GPP:3
C13) GM: Let’s give her a round of applause.
C14} — //applause//
C15) PMGR: At the same time, it’s Mrs. E___’s birthdau
C16) today.
C17) — //cheers and applause//
Besides the rule of first names for addressing and
indirectly referring to employees up to division heads,
there appears to be a practice of not Juxtaposing the
terms "management" and "employees." In fact, management
is not used at all in the carpus. There is abundant
evidence of "employeeCs)" and, in separate discourse
contexts, "managerCs)," but they are not used so as to
contrast employees and managers. Instead, they are used
as generic terms to specify a subgroup of an audience or
other group being referred to. Similarly, employees often
are referred to collectively by their occupational titles
or by local jargon that substitutes for occupational
titles. Thus the transcripts include references by use of
such titles as barback, server, houseman, bellman, clerk,
sommelier, busser, cook, sous chef, PBX operator,
housekeeper, ladies, bartender, hostess, and toaster. On
the other hand, with the noteworthy exceptions of
housekeeping team leader and personnel manager, a manager
is simply a manager. On the basis of reference to
154
occupational title, then, management appears to be a
monolithic unit, uihile rank and File employees are
partitioned by many occupational groupings.
Another form of address reFerence reinforces the
perception of a homogeneous manager group and a hetero-
eneous nonmanager group— the systematic use of "you
guys." The transcripts reveal 44 tokens of reference by a
speaker to hearers using "you guys" or, as direct address,
"guys." In every instance the speaker is a manager, and a
total of eleven managers used this form of reference. The
managerial levels ranged from first line manager to hotel
general manager, and in no instance was the reference
exclusively to managers.
"Guy" is a term that, while not usually considered
vulgar, carries some pejorative connotations. It has been
a slang term for man or boy, and in some parts of the
United States still is; thus it tends to aggregate male
and female audience members of an audience under a label
that is conventionally male. "You guys" carries
connotations of conspiracy, linked etymologically as it is
to the Guy Fawkes affair; and its use clearly sets off
those being referred to as a group with separate interests
from the speaker CWebster*s Third New International
Dictionaru. 1976).
Thus "you guys" typically is heard to mean "you
others are Csome attribute!." I suspect the use of the
155
term by adults has changed in tone, function, and effect
in recent years. Currently it seems to indicate extreme
familiarity by speaker toward hearerCs) and consolidates
the audience in the most informal way possible short of
being insulting. Beneath the apparent familiarity,
however, the older connotations of separateness seem to
lurk, the setting off of the "guys" from the speaker.
More importantly, it is a reference that is reserved for
the use of managers toward nonmanagers; there is no
instance of nonmanagers using the term, either toward
managers or toward other employees. It also is important
to note that its use appears to be unnoticed by either
speakers or hearers; it therefore is deeply ingrained in
the discourse patterns of the workplace.
Deictic expressions reveal little about hierarchy.
The absence of epithets, however, implies that a norm of
politeness exists in the workplace. Nonetheless, unless
epithets were routinely used by some hierarchical sector
against others without reciprocation, little can be
concluded about hierarchy in discourse from epithet
evidence. The remaining evidence is derived from the
speakers’ use of pronouns to refer to programs, problems,
resources, objectives, and organizational processes at the
two hotels. Hierarchy seems to be mitigated by the very
high use of the pronouns "we," "our," and "us." The
following three examples illustrate the apparent Joining
156
of speakers’ interests to audience interests by the use of
first person plural pronouns when discussing,
respectively, programs, problems, and objectives. The
speakers are all managers at the division head level or
above.
0313:LAX: TNG:GPP:1/2
C45) GM: As I say what we’re here to learn about our new
C 46) class one frequent travellers program. A:nd
C47) difference between the new program and the old
C01) is now we are awarding points on total dollars
02277:RLB:DEP:EMC:52
(03) EXEC : I ha=I had a (2) I had couple comment cards
C 04 ) that they came to us. Up to: December thirty-
C05) first, our comment cards in orchid cafe
C06) improved by eight percent. January tenth to
C07) February twenty-second you declined sixteen
C08) Dercent. Cl) Okau? Cl) It’s ouch. It’s buried.
COB) Okay? We can: ’t, we can’t tolerate tha:t, we
CIO) are a Lewitt hotel. (2) We are not the hotel
Cll) down the street. We are Lewitt hotel. We charge
C12) a good Cl) bucks Cl) for people to stay here.
157
02277:RLB:DEPL:HKG:30
C36) RME: what we’re looking to do is get those people,
C37) class one members that we have now alre=already
C38) but we wanna get Cl) new ones. Alright not only
C33) from the Standard up the road here, but some of
C40) the other chains like the Ramadas and the Shera*
C41) tons who Cl) ah we Feel we give a lot better
C42) service to and we’ll be able to get them From
C43) their programs and have them Join Lewitt. And
C44) bring back our occupancy to where it was.
In the next section oF this chapter I consider in
detail the use oF "we" as a representative oF pronoun
reFerence. It is important to note here, however, that
the Frequent use oF First person plural pronouns instead
oF speciFying the persons reFerred to appears at First
consideration implicitly to Join a speaker and hearers. I
will show later in this chapter that beyond the surFace
appearance oF solidarity lie habits oF expression that in
Fact separate groups and perpetuate the dominance patterns
that already exist.
From the Foregoing pieces oF evidence based on
speakers’ use oF reFerring expressions, three conclusions
can be drawn relative to hierarchy. First, there are
routines oF discourse that clearly de-emphasize awareness
oF the Formal position hierarchy. The evidence For this
158
conclusion is the widespread use of First names For
address and indirect reFerence, the absence oF the term
"management," the low level oF reFerence based on position
titles, and managers’ heavy use oF First person plural
pronouns. Second, there is implicit recognition oF a
hierarchical structure that does not coincide with the
position hierarchy. On the basis oF naming practices,
where last names are used only in cases oF reFerring to
general managers, the workForce is segmented into general
managers and those who are not general managers. IF other
naming practices are Factored in, however, the workForce
appears to be segmented into three groups: general
managers, all other managers, and nonmanagerial
employees. This scheme is based on the polarizing use oF
"you guys" by managers at all levels and on the managers’
use oF "employeeCs)" and "managerCs)" as terms to reFer to
identiFiable sectors oF the workForce. Table 1 summarizes
these implications For hierarchy.
A third conclusion is derived From the First two: A
contradiction exists among some oF the discourse practices
exhibited. The most readily observable reFerring
practices oF those discussed— use oF names and use oF
First person plural pronouns— give the impression that
speakers wish to convey an absence oF hierarchical
boundaries; yet, the asymmetrical use oF "you guys,"
occupational titles For nonmanagerial employees, and the
159
terms "emplayeeCs)" and "managerCs)" conveys de facto
boundaries that separate managers From all others. As
such, there is an appearance of a caste-like system of
hierarchy CLenski, 1984, p. 392) in the hotels that
contradicts the avowed egalitarianism and openness of the
organization.
Table 1
Reference practices and hierarchy
REFERENCE TYPE DISCOURSE EVIDENCE
Def inite
descriptions
Indef inite
descriptions
Naming
Pronouns
job titles used for
workers only
"management" not
used; "managers"
versus "employees"
first names used by
all except for
general managers
"you guys" used for
workers only
first person plural
used more by managers
third person singular
used more by workers
IMPLICATIONS
collapses hierarchy
managerial homogen
eity
partitions workers
and managers
collapses hierarchy
segments workforce
into three
partitions workers
and managers
partitions workers
and managers
Reference and Relationships
In interviews with managers and nonmanagerial
employees alike, a singular characterization of
relationships was heard: The people are what make this
hotel an enjoyable place to work. In corporate produced
videotapes for local use in new employee orientations,
160
Lewitt employees repeatedly are described as friendly,
competent, industrious members of a team. Affability is
promoted overtly by top management at both hotels. The
general manager of Marina Lewitt tells newcomers at their
orientation session:
01137 : RLB : C3RN : E
C01) GM: You play a key role in the success of my Job/. If
COE) you come to work with a smile on your face, you
C03) give it freely to the customers and to each other
C04) and you approach your Job in a positive manner...
C13) Ahm, I will look Cl) for your smiles.
Employees are described in corporate brochures as
team members and as members of the Lewitt family. All of
the division and department heads interviewed alleged that
they have an open door policy and that it is the policy of
the hotel to foster open vertical communication throughout
the organization. At the airport Lewitt, the employee
handbook says in part: "It is important for you to know
that we encourage you to communicate your ideas,
suggestions and problems... we encourage each of you to
take advantage of our communication systems, so that we
can work together...." And the Marina Lewitt’s handbook
states in blank verse: "Our Lewitt spirit is/Genuine
161
smiles that welcome/Courteous and helping
attitudes/Dedication to quality and the courage to
achieve/Team pride that shows, and/Enthusiasm that
encourages this spirit in others."
The intention, expressed and implied, is to Faster
relationships that are egalitarian, open, friendly, and
productive, regardless of individuals’ positions in the
hotels. Employees’ reference practices in discourse
described in the discussion of hierarchy above reveal that
the intended relationships are partly realized.
Indefinite descriptions by both managers and nonmanagerial
employees include reference to teamwork and unity:
05577:RLB:DEP:EMC:17
C33) nGRl: Ule are a team. Because if we don’t make it, CS)
C31 !) as a team, Cl) trust me, nobody will.
03197:RLB:HTK:14/15
CSS) 07: Well we got a good unity there...
C03) 07: We all go drinking together.
C04) GM: Well=I, I do think you have a good core of
COS) attachment.
COB) 07: Yeah. We we’re pretty uh we’re all pretty
C07) compatible and we’re all pretty close.
162
First name usage is reciprocal through the division
head position; at the general manager position, the name
use convention becomes asymmetrical. In 48 pages of
Lewittalk transcript recording the discourse of the Marina
Lewitt general manager and 28 nonmanagerial employees, the
general manager was not once called by either First or
last name. At the same time, she called every employee in
attendance by first name at least once, used exclusively
first names, and occasionally modified the term of
address to something even more familiar, such as changing
from Robert to Bobby and from John to Juan. Thus the
uniform egalitarianism appears to break down at the
general manager position, Just as was the case with
indications of attitudes toward hierarchy.
Similarly, deFinite descriptions which use such terms
as employees, union members, managers, and occupational
titles Cas discussed previously) all argue for setting off
managers as a monolithic group and nonmanagerial employees
as a heterogeneous collection of occupational and task
unit groups. Interviews and observation revealed some
intergroup conflict among managers; however, there is no
evidence of disagreements among managers anywhere in the
transcript corpus. On the other hand, many of the
problems presented by nonmanagerial employees in
departmental meetings and in the Lewittalk rite openly
display intergroup conflicts among the lower level
1B3
positions. There is discussion of conflict between cooks
and servers C02277:RLB:DEP:EMC:48), between servers and
bussers C0EE77: RLB: DEP: EMC: 34) , between evening shift
“turndown ladies" in housekeeping and day shift
housekeepers C0EE77:RLB:DEP:HKG:22), between barbacks and
bartenders C03197:RLB:HTK:35), and so forth.
It is the nonmanagerial employees bringing forth
complaints who use the preponderance of third person
pronouns— he, she, it, and they Cand their objective and
possessive forms). The incidence of third person pronouns
used by nonmanagerial employees noticeably rises when
discussion centers an cross-task or cross-occupational
conflicts. Here, servers are complaining that bussers are
undertrained and unavailable to help them:
0EE77:RLB:DEP:EMC:34
C33) EHP6: You know, like Cl) I don’t=I mean
C34) we never see ’em.
C35) EMP11: Yeah, and they take the attitude they’re not
C3B) gonna get tipped anyways.
In another example, a night shift housekeeper complains
that the day shift workers disturb her lost and found
items in the housekeeping closet:
1B4
02277:RLB:DEP:HKG:37
C38) Et1P37: The other day, Cl) everything went up C2) uhm
C39> the next day we come back Cl) an’ in the Cl)
C40) closet //inaudible//. They go and they move
C41) everything, sometime shoes,sometimes C2)other
C42) clothes in there ’n’ they been takin’them out
When these third person pronouns are used to refer to
other nonmanagerial groups of employees, they reflect the
heterogeneity of employees at that organizational level.
According to this discourse evidence, it appears that
controlling interests of the organization— local and
corporate office managers— strive to foster solidarity
among the workforce by instilling and modelling practices
that are oriented toward maintenance of interpersonal
relationships. Affability, informality, and cordiality in
regard to behavior, appearance, goals, and work values are
advocated. These qualities do characterize relationships
among members of task units and occupational groups of
nonmanagerial employees. However, the reference practices
indicate that this is not the case for intergroup
relationships at that organizational level. Persons are
referred to as ' ‘they," setting up opposing groups based on
job classification or task role, and occupational, task
unit, and organizational unit groupings are referred to in
definite descriptions. These referring practices are
165
noticeable because they contrast with those of managers
when referring to other managers, and they contrast with
the absence of references among managers that reveal
interpersonal or intergroup conflict. Table 2 summarizes
the culture of reference and its implications for
relationships.
Table 2
Reference practices and relationships
REFERENCE TYPE
Definite
descriptions
Indef inite
descriptions
Naming
Epithets
Pronouns
Referring practices in the discourse of the two
hotels suggest at least two findings about relationships.
First, interaction between individuals typically is
informal and cordial, polite Cin the sense that no
epithets or vulgarity is evident), and solidary Cin the
sense of belonging to a work team or occupational group).
DISCOURSE EVIDENCE
"managers" versus
"employees"
team metaphor
high use of first
names
absence of epithets
manager use of "we"
worker use of
"they"
IMPLICATIONS
managerial homogen
eity; worker
heterogeneity
workforce solidarity
role differentiation
conveys equality and
informality
norm of politeness
conveys solidarity
partitions workforce
worker heterogeneity
166
Second, the important relationship dynamics appear to be
those between groups, and these dynamics work in two
dimensions. There is obvious heterogeneity among
nonmanagerial groups horizontally, on the basis of
i
occupation, task role, organizational unit, etc; moreover,
there is a clear difference between nonmanagerial
employees and managers on the basis of managers’
appearance of within-group solidarity, differences in
privileges to use others’ names, and indefinite and
definite descriptions that treat employees and managers as
separate groups. Thus a strong effort by management to
foster solidarity, egalitarianism, and openness in all
employment relationships is confined to those within the
quasi-castes of managers and nonmanagerial employees.
To claim there is evidence in organizational I
discourse that hierarchy exists and that solidarity exists |
i
more in the horizontal than the vertical dimension of
organizational relationships might seem to be stating the
obvious.- The explicit claims of managers in interviews,
■ however, are that all members of the organization are
equally important, all are members on the same team, all
treat one another in the same friendly, open, democratic j
spirit. At first glance, the day-to-day Interaction of !
all employees seems to support those claims. The analysis
of discourse here, however, shows that there are
hierarchical differences among employees with attendant
167
differences of privileges to refer to the uiorld around
them in particular ways.
How does it happen that employees do not cross the
discoursal boundaries when there is no overt recognition
of a class or caste system in the organization? Why does
no housekeeper refer to the dozen managers standing at the
front of the conference room as “you guys?" How are the
group boundaries maintained when organizational members in
effect deny that boundaries exist? The transcript
fragments discussed above show how reference practices can
reflect the meanings of speakers and hearers. Because
reference is a pointing-out, a representation CBrown &
Yule, 1993, p. 206) of the world known by the speaker, or
intended by the speaker that the hearer recognize, it is
an offer of reality writ small that speakers present to
hearers. host instances of reference, even when
problematic CAuer, 1984) are transient, routinized, almost
preconscious experiences. Reference does its work subtly
and cumulatively.
As Brown and Yule summarize this process of reference
and discourse representation, speakers formulate their
intentions into discourse on the basis of some assumptions
about their hearers. Reciprocally, hearers formulate
their meanings on the basis of similar assumptions about
speakers. The assumptions are "of a similar general
experience of the world, socio-cultural conventions,
168
awareness of context and communicative conventions" C1983,
p. 2073, including a set of linguistic and discourse
codes. It is the application of these assumptions within
a social context of pre-existing hierarchical positions
that the explanation of how discourse constrains social
relationships begins to emerge.
Pronouns are the form of reference that promise the
most insight into the way that speakers and hearers manage
meaning intentions and interpretation. Brown and Yule
state that pronouns are "the paradigm examples of
expressions used by speakers to refer to ’given’ entities"
C1983, p. 214:3. This is so because they are "lexically
attenuated" CChafe, 1976, cited in Brown & Yule, 19833, or
they conventionally convey less direct information about
what they refer to-, moreover, in contrast to definite and
indefinite descriptions, names, and epithets, pronouns
occur with high frequency and have a nearly limitless
number of possible intended referents. In the latter way
pronouns are similar to deictic expressions, but deixis
has a more concrete relationship to referents than do
pronouns.
To explore the ways referring practices in discourse
influence social relationships, I have chosen the first
person plural pronoun forms to analyze. We/us/our are key
forms because they potentially unite speakers and hearers,
they are noticeably abundant in the discourse of hotel
169
general managers and division and department heads, and
they are frequently used in the policy statements and
videotapes created For the Lewitt hotels.
Analusis of UJe/Us/Our
Unless otherwise noted, I will use instances of "we"
to represent the class of pronouns that includes "our" and
"us." As a shorthand, I will refer collectively to all
three forms of First person plural pronouns as FPPP. "We"
may be examined as a representative of the other two,
however, for two basic reasons. The set of possible
domains of reference for "we" appears to be coextensive
with those for the other two first person plural pronouns,
and the linguistic and social conditions by which hearers
interpret "we" appears to be the same as those drawn upon
when interpreting the other two. The term "possible
domains of reference" is hearer-oriented and means the
possible ways a referring term can be interpreted.
In this section I describe the basic nature and
function of the FPPP set, using Brown and Gilman’s
analysis of second person pronouns as a model. Then I set
forth the commonly accepted explanation of the linguistic
Factors involved in the interpretation of pronouns in
discourse, noting possible additions and modifications in
the case of "we." After describing the possible domains
of reference of FPPPs, I draw on the transcripts to
exemplify eight ways that "we" is used by managers in
170
discourse with nonmanagerial employees. By analyzing the
pragmatic meaning of tokens of "we" in discourse, I add to
the set of linguistic conditions identified as
constraining interpretations, and I show that FPPPs do
work in discourse that goes beyond their lexical meaning.
Lyons C1977) defines lexical meaning as the
conjoining of denotation and "sense," or the
conventionally assigned properties of an entity. Lyons
equates denotation with reference; however, as Brown and
Yule C1983) point out, and as I have indicated previously,
reference is an action of speakers by which they point out
through their use of linguistic and paralinguistic
expressions the entities they are talking Cor writing)
about. Brown and Yule, as well as Auer C19B4), therefore
address reference as an activity that is successful or not
successful:
Successful reference depends on the hearer’s
identifying, for the purpose of understanding the
current linguistic message, the speaker’s intended
referent, on the basis of the referring expression
used. This last point introduces the notion of
’identifying the speaker’s intended referent,’
which is of crucial importance. C1903, p. 805)
Thus to derive a taxonomy of possible domains of reference
brought to bear on discourse situations by speakers, it is
necessary to work backwards toward speaker intentions by
analyzing possible hearer interpretations. Such analysis
must be based on the discourse context and the set of
171
social conditions that encompass the speaker and hearer
and are relevant to the kind of discourse at hand.
The Nature and Functions of “We"
As pronouns, FPPPs stand for some other expression or
expressions that describe entities in a speaker’s mental
world. In this view, the relationship between the pronoun
and another expression is that of antecedent Ci.e., the
other expression) and anaphor Cthe pronoun); and the two
terms are said to be co-referential to some entity in the
world. To the extent that pronouns have antecedents with
which they are co-referential, FPPPs point out certain
entities with specifiable properties. In the English
pronoun system there is only one abjective case FPPP,
"we," and its sense conventionally involves as a paradigm
property referring to the speaker and some other person or
persons.
Cl) I Just saw my office mate, and we’re wearing the
same color socks and slacks.
"We" in this case is co-referential with the combined "I"
and "my office mate" Just because they precede "we" in the
discourse and are situationally captured by the
reference. It would not be reasonable, for example, to
assume that the speaker was referring to the speaker and
172
the hearer, because those persons would know for certain
if they were wearing the same color clothing.
Paradigmatically, reference to the hearerCs) alone
invokes "you," reference to others invokes third person
pronouns, and reference to the speaker alone is conveyed
by "I." Thus an utterance that includes "we" suggests in
a fundamental way that the discourse is about the speaker
and some combination of hearerCs) and other personCs).
Since English has no conventional way to distinguish
speaker-hearer "we" from speaker-other "we," hearers must
infer from the discourse and social contexts which "we" is
intended. Other languages, such as Mandarin Chinese,
disambiguate the senses of "we" by providing two
linguistic forms. To complicate this ambiguity of FPPP
domain of reference, "we" is used in a sense that
conventionally refers solely to the speaker. This is the
case of the "royal we," which Brown and Gilman C197P)
trace to the Roman emperors’ use of nos to refer to
themselves. Presumably the use of FPPPs to refer to the
emperor began during the fourth century when the throne
was split between Rome and Constantinople. The realm,
however, was administered as one unit; consequently, to
refer to one was to refer to two. In addition, the royal
ego was thought Cmost commonly by those on the throne) to
encompass and represent the whole population. In sum,
173
then, "we" basically refers to speaker and hearerCs),
speaker and otherCs), or speaker, hearerCs) and otherCs).
"We" may bind people together in ways that other
pronouns do not. "They" paints out persons or objects
segregated from speaker and hearer, as do the other third
person pronouns. "I" sets the speaker off from the rest
of the world; "you" points out the hearerCs), thereby
setting them off from speaker and others. Among
contemporary English pronouns only FPPPs have the
potential to join speaker and hearer, and it is in those
referring expressions whose domains of reference
unequivocally include speaker and hearer that maximal
solidarity is communicated.
In their classic work on second person pronouns
Chereafter called SPPs), Brown and Gilman (1975) discuss
two axes of social influence implicit in speakers’ choices
of pronouns. One axis is the power semantic, which
captures the set of norms in asymmetrical power relations
that dictate choices of pronouns; the other axis is the
solidarity semantic, which captures the set of social
motivations in close personal relationships that dictate
choices of pronouns. Speakers may choose to use familiar
forms of "you" Ce.g., tu, du, the outmoded English thou,
etc.) or formal forms Ce.g, usted. Sie. you). The two
forms usually are symbolized by T for familiar forms and V
for formal ones. Brown and Gilman reason that while
174
individuals might differ in terms of power Chased on
wealth, status, office, profession, etc.), it also is
passible for persons to differ in other ways Csuch as city
and clan of origin, school attended, religion, political
affiliation, etc.). Solidarity is the name given to the
symmetrical and reciprocal relationships in which persons
enjoy like-mindedness or other similarities. Solidarity
thus is a spectrum that ranges from highly solidary to
completely dissimilar. Both members of a solidary
relationship use T forms in discourse.
The power semantic, on the other hand, is not
reciprocal: One member is more powerful, and the other is
less powerful. Consequently, when the power semantic
operates in the selection of SPPs, one person uses T and
the other uses V. The two axes themselves are not in
equilibrium, often resulting in conflict between the
dictates of each for pronoun selection. A French worker
wishes to address his boss, who is a female about his age
who went to the same university: Does he say T for
solidarity or V for her superiority? Similarly, a
superior person in the power axis who wishes to address an
inferior with whom he or she is highly solidary on some
other basis will find the T/V choice unclear. Brown and
Gilman’s piece was first published in 1960; at that time
they observed that the trend universally appeared to favor
greater choice of T whenever T and V conflict. Paulson
175
C1384:) reports that in Sweden there is rapid change toward
sole use of the T Farm in its most solidary
configuration. Brown and Levinson (1079), however, warn
that the reasons explaining speakers’ choices of pronouns
must include consideration of social structures of various
kinds, including the relationship of the individuals and
their social classes.
There is clearly a sense in which English FPPPs
demonstrate a solidarity semantic, and arguably they can
influence the power semantic, as well. Many social
situations find a speaker addressing hearers about a topic
that involves the hearer and possibly others but not the
speaker. In such situations the speaker may choose to use
either a SPP or a FPPP:
02577:RLB:DEP:EMPC:59
(48) EXEC: Alright when we go to a table with wine we need
(49) go this is our beaujolais villa:ges
The division director speaking here does not serve wine to
guests in the restaurants; he is referring to those who do
that task as part of their assigned tasks. A more precise
(but Fictional) version of his intended message is:
176
C2) Alright, when you go to a table with wine you need to
go, ’this- is our beau.lolais villages. ’
By substituting the FPPP, the director is subtly
suggesting that speaker and hearers have a community of
interest in the task of serving wine; at the same time he
is avoiding directing their attention to the Fact that
they serve wine and he does not, yet he is telling them
how to serve wine. That fact would became more obvious
with the use of "you" in this instance.
Another way that FPPPs can be used to enhance
solidarity is to eschew "they" and substitute "we" to
refer to some larger aggregation of persons than the
situation in fact calls for.
03197:RLB: HTK:10
Cll) MGR*. UJell,t=the way Lewitt works: Cl) is we have C2)
C12) in our company we have eighty-seven hotelsright
C13) now=we probably own thirty of them.
Here a hotel general manager is telling nonmanagerial
employees about the highest level of corporate financial
relationships. The phrase in Cll) and C1E), "we have in
our company," Frames what is to Follow as being a company-
wide situation; thus "we" in C13) most likely refers to
speaker, hearers, and all other employees of the company.
177
The manager could have chosen not to frame the information
that way and said, "they have eighty-seven hotels right
now and they probably own thirty of them." No person
involved in the discourse owns any part of any of the
hotels being discussed, but by using the FPPP, the speaker
creates a community of interest with those others not
present whom the discourse is about. At the same time she
conjoins herself and her audience as part of a more
encompassing unit.
Pronouns that demonstrate the solidarity semantic are
reciprocal, that is, they are forms that may be used by
either person in the relationship. In the case of Lewitt
hotels, employees use both the forms of solidarity "we"
described above.
03197:RLB:HTK:10
Cl) 19: Yeah. Did we have something in the TIME
CE) magazine about properties in Maui?
178
C3) tlgr: Yeh=th=th=that they have ah Japanese ownership
C4) I think.
C5) 19: How does that work?
C6) Hgr: Ah:m, well, Cl) ahm essentially ah Chris N___
C7) was the owner and developer a:nd he sold his
C8) interest to a Japanese company.
C9) 19: Does that mean we don’tCl) still consider it
CIO) as a Lewitt?
At line Cl) an employee C19) asks the general manager if
there was an item in Time magazine about the corporation’s
properties. "We" in this instance joins speaker and
hearer in an implicitly larger aggregation, which,
incidentally, is at a higher organizational level than the
actual relation-ship of the two persons. The same
employee asks the general manager at line C9) how to
interpret the information that the developer of the Maui
hotel sold his interest to a Japanese company. In effect
he is asking the manager for her conclusion, so that he
will know how to interpret her previous answer. "You"
could be substituted for "we" in line C9) without changing
this reconstruction of its meaning.
Two referential domains of "we" are not reciprocal in
the hotel discourse corpus and are exclusively exhibited
by managers: the "we" that refers to the speaker only and
the "we" that refers to hearers and others. CThe only
179
example of nonmanagerial employee use of "we" referring to
hearer alone is cited in line C9) in the previous
excerpt.) Nonetheless, the exclusively managerial FPPPs
are those that are most likely to represent power semantic
references. In the following excerpt, the general manager
at Airport Lewitt CGM) is announcing the designation of a
management trainee to a new position. The general manager
herself created the position and appointed the new
manager.
03137:LAX:TNG:GPP:3
CIS) 611: Ahm: Cl) the reason that we named Maryanne Class
C19) One manager is that1s to sho:w that we think
C20) it’s a really important program. It’s so import-
C21) ant that we g=we’re putting someone in Cl) ’s
C22) who:le lo:b is gonna make=s gonna be: that we do
C23) it right.
Well informed, experienced hearers know that the general
manager appoints program managers to newly created
positions in the hotel; therefore, the "we" in C18) and
C19) are recognizable as alternative forms of "I." This
is the "royal we" that suggests some aggregation of
persons is encompassed and spoken for by the speaker. The
same informed hearer will recognize that the "we" in C22)
does not include the speaker because Maryanne is both
lao
subordinate to the speaker and beholden to her for the new
manager’s position and continued employment.
This use of "we" accomplishes several discourse tasks
simultaneously. On the surface, it tells the informed
hearer that the speaker is indirectly boosting the
authority of the new manager by suggesting that the
general manager is in the group who will be guided by
Maryanne. Very naive hearers might accept this reading as
a matter of fact, i.e., that Maryanne will supervise
everybody in the hotel with respect to the program. That
interpretation, however, is unlikely. More likely is the
interpretation of the "we" in CEE) as meaning you, the
audience. Unlike the polite we/you substitution discussed
above in connection with the solidarity semantic, this
we/you substitution is a veiled imperative.
The differences between solidarity we/you
substitutions and power we/you appear to be of two kinds.
First, the instances of solidarity we/you substitutions
are limited to cases where the referent being pointed to
by speaker’s use of "we" is a single hearer. When a
higher status person uses this form toward lower status
persons, it takes on the connotation of a diminutive.
Second, power we/you substitutions are found in discourse
stretches that involve the speaker ordering, directing, or
enjoining the audience. The whole purpose of the meeting
181
From which the last excerpt is taken was to generate
employee support for the new Class One program.
"Ule" accordingly has implicit power domains of
reference and solidarity domains of reference. The power
domains are: Ca) reference to hearer and otherCs? when a
we/you substitution is associated with speakers’
imperatives; and Cb) reference to speaker alone as an
instance of the "royal we." The solidarity domains of
reference are: Ca) speaker and hearerCs); Cb) speaker,
hearerCs), and otherCs); and hearerCs) only. Other
possible referential domains are discussed below; however,
I would like to emphasize here that these wide-ranging
discourse functions involving solidarity and power might
be related to the form of English SPPs.
Many Indo-European languages, such as Berman, French,
Italian, Portuguese, and Spanish, have developed T/V
forms, but for various historical reasons modern English
has not retained T/V forms. With diminished discourse
resources for marking power and solidarity relationships,
it is possible that English speakers have turned to FPPPs
For the subtle tasks performed by T/V pronouns in other
languages. Certainly such a suggestion must be taken
cautiously, and its confirmation requires study of "we"
referential domains in other languages. It is clear,
however, From the two social axes discussed here that
18S
FPPPs have multiple capacities to enable speakers to
create perceptions of groupness among hearers.
A final observation an the nature and functions of
"we" concerns the concepts of divergence and convergence
used in speech accommodation theory CGiles & Smith,
1979). In general, it appears that the uses of "uie" that
promote solidarity also contribute to speaker-hearer
convergence, in the sense that the referential domain
fosters a same-group identity for both parties. Likewise,
power uses of "we," those which delineate or reinforce
hierarchical boundaries between speakers and hearers, take
on characteristics of divergence. Such uses of "we"
segregate speakers and hearers into separate referential
groups and therefore separate them into distinct social
aggregations; concomitantly, the very use of divergence
tokens of "we" is a reflection of a pre-existing
hierarchical arrangement which permits the logical and
interpretable employment of such tokens.
Interpreting "We" in Discourse
The notion of co-referentiality has been criticized
CSag & Hankamer, 1977; Levinson, 1983) because it entails
a previous expression in discourse which anaphors refer
to, and many utterances contain pronouns which obviously
have no such antecedents. An example from a conversation
overheard among my neighbors is:
103
C2) A: I saw firs. NcPeak today.
B: Oh, are they back From vacation?
A parallel "we" example CFictianal) might be:
(3) A: I saw firs. flcPeak today.
B: Oh, are we going to start playing bridge again?
The classic example by Karttunen is provided in Levinson
C1983):
C4) A: The man who gave his paycheck to his wife was
wiser than the man who gave it to his mistress.
In C£) "they" refers to Mr. and firs. flcPeak, who had
not been previously mentioned in the conversation. But
because fir. and Mrs. are often seen in the neighborhood
together and it is known by the speaker and hearer that
Hr. and firs, recently went on vacation together, both knew
what was referred to by "they." In C3) "we" refers to
persons who have previously played bridge, probably
speaker, hearer, Mrs. flcPeak, and others. The knowledge
of past social conditions informs the hearer whom
specifically "we" refers to here. Clearly the "it" in the
second line of C4) does not refer to the paycheck of the
first line; rather, the idea of paycheck is invoked in the
184
initial noun phrase and called up by the “it" in the
second line to convey the speaker’s idea of the unwise
man’s paycheck.
Cases like these indicate that reference is not
necessarily dependent on previously uttered noun phrases
in the current discourse. In these cases, reference is
accomplished because the pronouns used prompt certain
kinds of knowledge the speaker and hearer most surely have
in common, both knowledge about the real world conditions
of the matter at hand and knowledge about how discourse
works. Brown and Yule C1983) propose five kinds of hearer
knowledge that might be used to interpret pronoun
reference: Ca) antecedent nominal expressions, as in the
co-referential example; Cb) the understood "roles" of
antecedent nominal expressions, as in C3) above, where it
is understood that Mrs. McPeak is associated with the
speaker and hearer by their bridge playing; Cc) predicates
attached to antecedent nominals that provide key
information about the referents; Cd) implicit antecedent
predicates C"A locomotive rolled by and he waved at us"
implies that there is an engineer in the locomotive); and
Ce) "new" predicates attached to the pronoun C"The kids
were standing by the tracks when he tooted the locomotive
whistle" allows "he" to be interpreted by virtue of what
follows it).
105
A review of FPPPs in the hotel corpus establishes
that hearers may use other kinds of knowledge as their
primary resource when interpreting "we." To examine the
conditions under which hearers might turn to knowledge
other than those types set forth by Brown and Yule, a
taxonomic description is needed of the referential domains
of "we" that are evident in the transcripts. For each
type of referential domain the associated discourse
conditions or hearer knowledge required for interpretation
can be identified and compared with the taxonomy offered
by Brown and Yule.
"UJe" domains of reference.
Unlike other pronouns, "we" has the unique property
of potentially referring to any individuals and groups,
including the speaker CS), the hearer or hearers CH), and
others who are neither speakers nor hearers CO). In face
to face discourse it may be assumed that S and H are
identifiable to one another; however, 0 might be
identifiable or not identifiable to either S or H. By
identifiable I mean that H could: Ca) refer to □ by name,
title, or other specific designator like kin terms; Cb)
assign 0 to some specific group in a formal organizational
unit, such as "a member of the executive committee,"
"people who work in housekeeping," or "all financial
managers"; or Cc) label □ by reference to 0’s commonly
known function or interaction with S and H in the past,
106
such as "the outside contractors hired to maintain windows
and elevators" or "the city politicians who visited last
month." Conceivably others might be included in an
instance of "we" and those others are not identifiable to
H.
Given that "we" can be used potentiallu to refer to
any combination of S, H, and □, and 0 might be
identifiable to H or might not be identifiable, what
evidence is there for each possible type of referential
domain of "we"? The most elementary and direct domain of
reference is S + H. Examples of this domain, which I will
mark with a postscript, CUEl), are evident in the
transcripts, both where S is a manager and where S is a
nonmanagerial employee:
02277:RLB:DEP:EMC:35
(18) EMP1: Uait. Are we gonna do p.m.?
(19) -- //Many voices overlapping//
(20) MGR: Ue’re not gonna do any p.m’s.
02277:RLB:DEP:EMC:10
(23) EMP0: Can we just (1) not Just Cl) speak out?
02277:RLB:DEP:HSK:41
(41) EH1: I think we oughta give ’m a big round of
(42) applause.
187
In these examples, there are no antecedent nominal
expressions, antecedent predicates, implied predicates, or
roles understood For antecedents. In each case, however,
there is a "new" predicate which, combined with knowledge
of what role "we" takes in metacommunicative statements
and knowledge of the current interaction so far,
facilitates the interpretation. There are surprisingly
few instances of "we" that are clearly in the referential
domain WEI; however, it is obviously a domain used by both
managers and nonmanagerial employees.
Likewise, both managers and rank-and-file employees
use "we" with another referential domain, S + H + □:
0EE77:RLB:DEP:HSK:15
CIS) EhPlH: We’re in first place.
C17) EHE: UJe’re in first place, but that’s not good.
C1B) C3) So we have a lot to learn, you guys
The assistant department head CEHE) has Just told the
assembled housekeepers in a departmental meeting that the
housekeeping department has had twice as many accidents in
the past year than any other department. About one-third
of the department’s employees are absent from the meeting;
therefore, when the employee CEHP14) and EHE say "we’re"
in reference to the aforementioned housekeeping
department, they are including many employees who are not
188
present to hear their discourse. In this case, an
antecedent nominal expression is the central information
for hearers to knou) For interpreting the FPPP used. In
the cases shown here, the □ undoubtedly is identifiable by
H— S + H + 0 is totally a known entity to all present.
This domain is labelled UJE2.
In other cases, 0 is more problematic:
12226:LAX:ORN:26
C01D EMGR: Please. You only
C02) be the one CD responsible for your time card,
C03) you punch, and you take care of it. Don:’t eh=
C04) you know, don’t punch anyone else’s and don’t
C05) let anybody punch yours. So. CD Ah C2) we
COB) wouldn’t want that to happen.
The "we“ in line COS) is not made understandable by
what precedes it or by the new predicate. It is
reasonable to assume the employment manager CEMGR), who is
explaining benefits and obligations of employment to new
employees, would not want employees to punch the wrong
time cards. Nor would any other member of management.
Moreover, it is in the employees’ interest not to do so.
From EMGRs viewpoint, there probably is no reasonable
person in the world who would want that to happen. The
"we" then becomes an unbounded term, working as a
------------------------------------------------------------------------------------------------------------- TBS'
substitute Far "anybody" or "nobody." But by
personalizing the statement’s argument, that nobody would
want that to happen, it includes the audience in that
evaluative position taken by the speaker and reduces the
need for defending or accounting for the position taken.
This type of "we"— S + H + □ with a parenthetical n
for "not identifiable" next to the 0— is labelled WE3; it
is evident in the transcripts only when S is a manager.
Interestingly, it is the referential domain occasionally
used by adults when stating an evaluative claim to a small
child.
A similar domain of reference is the "we" that refers
to the hearer alone. Mothers may be heard to say to their
children statements like, "Are we ready for bed now?" when
only the child is about to go to bed. Here are two clear
examples of this type, WE4, from the hotel corpus:
1E226:LAX:ORN : 5
C41) EMGR: Ah, I’d like Jane to introduce herself/, a:nd
C42) tell us where she’s working, if you’re able to
C43) do it=ah=heh=heh.
C44) NEMP: Oh, okay. Urn, I Jane G =Hi=and I work in ahm
C45) Mrs. Candy’s. Down in the coffee shop upstairs.
C46) EMGR*. And, we tell something a little unique about
C47) themselves or what they like to do in their
C4B) free time.
C49) NEMP: Ahm I’m a musician and I sing in a ba:nd________
190
03137:LAX:TNG:GPP:1
C45) 6M: As I say what we’re here to learn about our new
C46) Class One Frequent travellers program. A:nd
In the First excerpt, supporting evidence that the
employment manager’s "we” in C46) reFers to Jane is the
Fact that EflGR switches to third person pronouns in the
same utterance. "Ule" is thus not necessary For her
intended meaning in this situation, and she could have
said, "tell something unique about yourselF and what you
do in your spare time." Here, a new predicate is the kind
oF knowledge that hearers need For interpreting the
pronoun. The same kind oF knowledge is invoked by the
second excerpt. In this utterance, the general manager is
stating to an audience oF lower level managers and rank-
and-File employees the purpose oF the training meeting.
The speaker designed and helps conduct the training;
consequently, S is obviously not present to learn about
the program.
UJith the Following marginal case as an exception,
only managers use UIE4 in the hotel corpus.
03197:RLB:HTK:10
(9) 19: Does that mean we don’t Cl) still consider it
CIO) as a Lewitt?
131
This is an excerpt discussed previously, in which an
employee CIS) asks the general manager how to interpret
her answer to the employee’s previous question about a
Japanese company buying one of the Lewitt hotels. Here he
could be saying, "How do you consider the hotel’s status"
or alternatively, "How should we employees consider the
hotel’s status." In either interpretation the employee’s
use of "we" is more solidary than a choice of "you" or a
noun phrase in that slat.
Another possible domain of reference is H + 0, or
UJE5. The speaker uses "we" to designate an aggregation of
persons that does not include S:
02577:RLB:DEP:HSK:15
C17 ) EH2: Ue’re in first place, but that’s not good. (3)
(18) So we have a lot to learn, you guys, we’re not
In this example the manager CEH2) follows a UIE2 with a
UIE5. At (18) the first "we" includes all members of the
housekeeping department who have a lot to learn about
preventing accidents so that the departmental record can
be improved. EH2 is a principal trainer in the
department, however, and it is highly unlikely that she
would be included by any hearer as being among those who
have a lot to learn. There are no immediately adjacent
discourse features that hearers must know for interpreting
192
this "we"; however, Following some audience interruptions,
the manager issues an injunction:
CEE) EHE: Do not be so quickCE)to do things, you know, be
CS3) in a hurry, an’ Cl) clean up some glass, or Cl)
C24) y’know moving a bed or anything like that.
In a sense this stretch of Following discourse is a
new predicate, except that its scope is larger than a
portion of the same utterance containing the pronoun.
This type of referential domain is used only by managers
in the hotel corpus, and in each case its interpretation
is dependent on some extended stretch of following or
preceding discourse that establishes situational premises
on which the interpretation can be based.
There are no clear and unambiguous examples of a
domain of reference that is comprised of H + 0 where
others are not identifiable by hearers. For there to be
such a domain, "we" must refer to "anybody but me" or
"nobody but me" and "me" is the speaker. Likewise, there
is no clear evidence of a domain that is comprised of S +
0 where others are not identifiable. That would refer to
"anybody but you" or "nobody but you." The substitution
of "we" in this case would make more intuitive sense than
H + OCn) because of the primordial sense of speaker
inclusion in "we." The one possible token of this domain
193
is spoken by the Marina Lewitt general manager when
describing the status of a proposed Lewitt hotel in Santa
Monica:
03197:RLB:HTK:9
(20) MGR: A=they have no property there right now, but
(21) there is a developer who would like to build a
(22) Lewitt hotel and we have the site but we’re not
(23) sure that it’s a good site. (3) Any questions
The second "we" in (22) is an evaluative statement
that could include the general manager among those who are
not sure about the adequacy of the site. Whoever else
might be unsure is not apparent at all; certainly it is
not the hearers because this is all new information to
them. An alternative interpretation is that this token is
an instance of the "royal we," referring to the speaker
only. This domain type is discussed more fully below, but
there is sufficient ambiguity here not to propose a
separate domain category of S + 0(n).
□n the other hand, there is ample evidence to support
a category of S + □. In the following excerpt,
interpretation is facilitated by knowledge of an
antecedent nominal expression:
194
03197:RLB:HTK:6
CEO) MGR: Ahm CE) Robert and I workekd together at the
CE1) Lewitt Edgewater far a while Cl) before we both
CEE) moved over.
Here the general manager addresses a group of hearers as
if Robert is not present and overtly states who "we"
refers to. Just in case a hearer might interpret the
subjects of "moved over" as including hearers or as
referring only to the speaker, MGR includes "both" to
disambiguate her utterance.
Other kinds of interpretation knowledge is used in
the WE6 category of S + 0:
0SS77:RLB:DEP:EMC:51
Cll) EXEC: You covered yourself. Now if the food is not
C1E) ready, there is a ti:me on that chit. And your
C13) handwriting is on it. There is no excuse for
the
C14) server or the sous chef or the line cook to
C15) say, you didn’t tell me. Because you put that
C1G) chit out. //inaudible//
C17) I
C10) EMPS: So then we don’t trust each other, then.
195
The division director CEXEC) mentions in his turn of talk
a non-hypothetical "you" and server, sous chef, and line
cook. But previous discourse indicates the issue involves
intergroup problems and procedural snarls between the
kitchen workers and the servers; consequently, the
speaker’s "we" at CIS), while including himself as a
server, refers to floor workers and kitchen workers not
trusting each other. Three turns later, a manager
confirms this interpretation in her response to EfIPE:
CE7) MGR1: Chris, do you understand the reason
CE8) behind whu we have to write it down? Cause theu
CE9) can’t remember fourteen //inaudible//
Clearly, the parties involved in the issue of mutual trust
are those who have to write orders down Cfloor workers)
and those who cannot remember fourteen somethings Ckitchen
workers). Thus interpretation knowledge must include an
extended stretch of the discourse at hand, in this case
preceding the pronoun and confirmed by discourse that
follows.
The most frequent type of "we" in the corpus is WEB,
and it occurs in contexts where S is a manager and is
referring to S and some other unspecified memberCs) of
management. The following four typical examples all occur
in stretches of discourse where there are no clues to
136
interpretation knowledge:
03137:LAX: TNG:GPP:3
C38) GM: ThB next thinp I want=talk aboutCDis what’s in
C33) it For you. C2) ’Kay? We want you to make this
C40) program a success.
01337:RLB: HTK: 6
C28) MGR: They are as traditional
C23) very strong individual travelling business
C30) //inaudible// ahm so we’ve been working ahm to
C31) reaffirm loyalty in that market.
01337:RLB: HTK:12
CIO) MGR: We think
Cll) we can get up to about two fifty per room in
C 12) time Cl) and it will be fully operational ah in
C 13) a few weeks.
02277:RLB: DEP:HSK:23
C40) RME: What we’re looking to do
C 4:1) is get a lot more of the transient business to
C42) come in to=ah Cl) into our particular hotel
197
Although each of these tokens of "we" implies that S
and somebody else are involved as subjects of the actions,
these could arguably could be instances of the royal uie.
An example from the Lewittalk transcript argues against
that interpretation because it has the same form as the
foregoing examples but explicitly contrasts the "we" used
to "I":
03197:RLB:HTK:IB
C05) MGR: N’ 1=1 have to tell you real straightforwardly
COB) that we are looking at four approaches to it
C07) now. UJe’re looking at management Cl) we’re
COB) looking at wage rates Cl) we’re looking at
C09) physical layout Cl) and we’re looking at
CIO) equipment. C2) And because I don’t think
Not only is there contrast between "I" and "we," but
also it is unreasonable to believe that the general
manager looks at the four factors in turnover she
describes with no involvement of other persons. Thus,
there is an S + 0 domain of "we" that is used exclusively
by managers, is associated with discourse about high level
plans, policies, and organizational goals, and has no
apparent interpretation features in the discourse at
hand. The □ in these cases of WEB is not specified by the
speaker and would dispositively be someone who vaguely is
1 90
associated with the speaker at the speaker’s level of work
and shares an undefined community of interest. Because of
these differences between the other types of S + 0 domain
and the one under discussion here, I designate them a
separate category of "we" referential domain, WE7.
Finally, there is a passible domain comprised of
references to S only. All instances of this category,
labelled UE8, are spoken by managers. The common
characteristic of tokens in this category is that they
describe a personal emotional or evaluative response to
some event or fact:
03137:LAX:TNG:GPP:3
C32) GM: ’S a big jo:b, we know that she da it, and we
C33) think it’s an important one. Okay?
03197:RLB:HTK:5
C33) MGR: Ue feel real comfortable that April we’ll be
C34) back to our normal business
Ulhat informs the hearer that the speaker is referring only
to herself only in each of these excerpts is the absence
of explicit mention of other managers in prior utterances
and the personal, subjective types of verbs that predicate
the pronouns— know, think, feel. Verbs of subjective
experience such as these are less likely to characterize
199
group responses than individual responses. Notice that
"I" can be substituted For "we" in C32) of the first
excerpt and in the First slot in C33) of the second
excerpt, but cannot be sensibly substituted for the latter
"we" in C33) of the second excerpt, nor can "my" be
substituted For "our" in C34) .
At least two contributions to the extant views on
pronoun reference and discourse are apparent in this
discussion. First, it is passible to set forth a taxonomy
of FPPPs, as in Table 3, based on their referential
domains, who is privileged to used them in organizational
discourse, and what kind of interpretive knowledge hearers
use when understanding them.
Second, Brawn and Yule’s set of knowledge types,
which hearers use to base their interpretations of
pronouns in discourse, needs to be expanded to accommodate
reference interpretation in cases of "we." The absence of
discourse information that hearers would otherwise use is
an important key to making Judgments based on some broader
sense of implicature and worldly knowledge. This is
particularly true of those categories of "we" that are
used solely by managers. In addition, it is apparent that
some interpretations must rely on stretches of discourse
that are not contiguous to the turn that contains the
pronoun. Hearers must take into account earlier discourse
and, if necessary, suspend their lack of interpretation
BOO
until later discourse provides needed interpretive
information; otherwise, repair of the reference is
necessary (Auer, 1984). It is noteworthy that there are
no examples of repair involving FPPPs anywhere in the
hotel corpus.
Table 3
Referential domains of "we"
TOKEN DOMAIN
TYPE
WEI
WEB
UIE3
WE1 !
WE5
WEB
WE7
WEB
S + H
S + H + 0
S + H +
OCn)
H
H + 0
0
S + OCn)
S
USER GROUP
M & NM
M & NM
M only
M only
M only
M & NM
M only
M only
INTERPRETATION
KNOWLEDGE
new predicates
new predicates
cues absent
POWER
VS
SOLIDARITY
P/S*
P/S*
P/S*
new predicates P
preceding & following
discourse P
antecedent nominal P/S*
expansions; preceding
& following discourse
cues absent P
cues absent P
* P/S indicates that the domain can be either a power or a
solidarity reference, depending on social and discourse
context.
There is an obvious asymmetry in the use distribution
of the various categories of "we" in the transcripts. The
major implications of the findings shown in Table 1
involve how individuals’ understandings are shaped by
speakers’ use of reference and how group designations are
accomplished. These implications and their relation to
ideology are addressed in the critical analysis in Chapter
201
4. This chapter concludes here with an extended
application of the "uie" taxonomy in an analysis of a
typical stretch of managerial discourse.
An illustration of FPPP analysis.
Application of the "we" taxonomy and the principles
of group identity and referential ambiguity is exemplified
by the fallowing analysis. The transcript excerpt is
taken from a housekeeping departmental meeting and
includes as principal speaker the assistant executive
housekeeper CEHED; also speaking is a leader, or first-
line supervisor CLDR3D. The manifest technical
consequences of this rite are exchanging information about
changes of policy and procedures and presenting group
training of various sorts. In the rite, the group
training addressed rights and obligations of employment as
spelled out in the newly issued employee handbook, safety
and security, bed making, placement of shower curtains and
towels, and equipment handling. Policy and procedure
information covered departmental goals, corporate
incentive program for guests, new employee reward
programs, changes in placement of amenities in guest
rooms, purchase of new equipment, etc.
The latent technical consequences include restating
performance expectations, encouraging joint problem
identification, and increasing homogeneity in worker task
behavior. Manifest expressive consequences of this rite
eo2
typically include enhancement gf team Feelings and
increasing intradepartmental solidarity. Possible latent
expressive consequences are sharing excitement and play
with other members of the department, venting anger and
resentment in public, and improving social positions
within informal group hierarchies.
Many of these consequences are easily interpretable
because at various times throughout the rite one of the
three highest managers attending the meeting would
announce what had been happening or what should happen
next. Far example, Just prior to the section to be
discussed below, the assistant executive housekeeper says:
02277:RLB:DEP:34/35
C45) EH2: Okay,ladies.(2) Does anybody have anything they
C46) want to talk about, any problems, concer:ns,
C47) anything like that? C2) Everything’s wonderfu=
C48) =oh Chris has one.
This turn of talk reveals both a manifest and a
probable latent technical consequence of the rite—
information exchange and the opportunity to participate in
ostensible Joint problem solving. In addition, it
contributes to the understanding of expressive
consequences in that it gives the appearance of soliciting
open and full participation by anu member present, and
203
those persons may talk about anything they want to talk
about. The alternative to speaking out, as implicated by
EH2’s comment, is to communicate by silence that
everything is "wonderful" in the department. The person
who raises her hand with a concern is not a rank-and-file
employee but a First line supervisor.
C49) LDR3: UJhat about spotted glasses, there\? I’m getting
C50) a lot a glasses, especially ones on the
C51) credenza, that they’re not being wiped out or
C52) y’know Cl) not being washed...
C53 ) C
C01) EH2: Okay. A=are you meaning from the
COE ) dishwasher?
C03) — //many voices// Yes.
C04) EH2: Okay, let me explain to you guys, and this is
C05) somethin we’ve been working on with stewarding.
C06) Oka-.y? UJe do not wash the glasses in the rooms
C07) //slowly// Cl) okay? That isn’t do:ne because
of
COS) the=ah infectious diseases going around and
C09) because it isCl) now a law that we are required
CIO) to put it through high temperature=the glasses
Cll) are not to be washed in the rooms. C2) 0ka:y?
C12) We did not talk about the glasses, did we? C2)
C13 ) This is another good point, you guys.C2) We are
2 0 4
C 14 ) changing Cl) the position of the glasses,
CIS) because of the robobars you have in your rooms/
C16) C2) okay. Ule are now going to put only two wine
C17) glasses Cl) on the robobar in the bedroom area,
C18) and you’re gonna keep the two tumblers in the
C19 ) bathroom. But you are not putting two tumblers
C20) and two wine glasses and the ice bucket on the
C21) robobar.
Within the last turn of talk shown here, consisting
of a mere 17 lines, "we" occurs seven times. It is
apparent at first examination of the transcript that the
use of "we" is not the same in all seven instances. To
obtain more insight about how the "we’s" are employed and
work in this discourse, however, each one must be
interpreted in view of the discourse, situational, and
expanding contexts. Then the power implications of each
"we" must be elucidated by arguments using the taxonomy
already developed, and the ideological implications of the
excerpt based on the conception of the members’
relationship that is promoted by the use of these
particular tokens of "we."
In lines C04) and C05) the manager tells the hearers
there is something that the hearers do not know, and EHP
can explain it to them. Furthermore it is something that
"we" have been working on with stewarding; it logically
£ 0 5
can not be something the hearers have been working on with
stewarding or they would already know about it. This
token of "we," then, refers not to the intended audience
but to the speaker Cwho does know what it is and can
explain it to them) and perhaps some other person or
persons. Whatever it is, the speaker is more powerful
than hearers with regard to it because: Ca) she knows it
and they do not; Cb) they need to know it and she can
explain it to them; Cc) it is important enough to have
been in the works for a time Ci.e., "we’ve been working
on" conveys continued past action up to the present); and
Cd) it involves interdepartmental work of same kind. The
UIE6 type, S + □, can convey either convergence or
divergence, depending an the situational context and the
expanding context. In this case, the manager uses "we" to
include herself in a planning and decision-making activity
involving another department, and in her very use of "we"
she excludes the intended audience, i.e., the rank-and-
file employees who do not know about the arrangement. As
such, she sets herself and unspecified others apart from
the hearers on the basis of her superior knowledge and her
participation in interdepartmental planning activities.
It appears, then, that this token of WEB is oriented
toward divergence, segregating non-managerial employees as
less powerful members.
2 0 6
In sharp contrast to the First token, the "uie" in
line C06) clearly does not include the speaker.
Supervisors, and particularly the assistant department
head, do not wash glasses under any normal circumstances.
More to the point, the clause, "we do not wash glasses in
the rooms," is not meant as a simple description of
routine behavior; the meaning is made clear by the
paraphrasing in lines CIO) and Cll), "the glasses are not
to be washed in the rooms." Ixlhat EH2 means in C06) is,
"do not wash glasses in the roams," but she chooses to use
a "we" statement rather than the injunction. Why? The
construction chosen softens the directive nature of what
she intends to say by implying that she, too, is subject
to the injunction of not washing glasses in the rooms;
however, the prosodic element of slowing down her speech
rate during her utterance and bracketing the utterance by
"okay?" conveys very strongly the type of indirect order
often given to children and other less than fully
competent persons: "We pick up our glass with two hands,
okay?" This condescending, baby-talk variant of "we"
places the speaker in a parental role and the hearers in
child roles with respect to the use of "we" and thus is a
divergence token, too.
At*line CQ9) EH2 states a second reason why glasses
are not to be washed in rooms— a law prohibits such
practices by requiring that glasses be "put through a high
2 0 7
temperature." The use of "we" has as a referential domain
persons uiho are subject to the law, presumably everybody.
To the extent that she shares responsibility For seeing
that glasses are put through a high temperature in the
washing procedure, the assistant executive ‘ housekeeper
joins the hearers as being subject to the law she
mentions. Consequently, this token of "we" appears to be
conventionally used to refer to S + H + 0 and at this
level of analysis appears to be a convergence use of UIE2.
Note also that the interpretation knowledge called upon is
a new predicate— it is not Just that there is a law
concerning washing glasses, but that it requires exposing
glasses to high temperature, something the speaker can be
implicated in accomplishing.
Similarly, the two "we’s" in line C12) must be taken
to refer conventionally to speaker and hearers, because
EH2 is talking about the discourse so far between those
present. As such, these tokens of U1E1, calling on the new
predicate "did not talk about the glasses" as
interpretation knowledge, convey convergence among the
members present.
The token at line (133 is similar to that at line
(053: It announces a decision that the hearers do not know
about and sets off the speaker as a knowledgeable member
who participates in decision processes not accessible to
the hearers. The "we" token cannot include hearers in the
EOB
domain of reference because of the subordinate clause,
"because of the robobars you have in your rooms" and
because it is not passible sensibly to substitute "you and
I" for "we" in C13) . The "we" at C16) is a repetition of
the "we" at C13) insofar as it too is followed by a
subordinate clause specifying "you" in connection with the
actual carrying out of the decision. In this
interpretation, "we" refers to persons who have made the
decision implied by "we are now going to put," meaning “it
has been decided" or "we have decided to put only two wine
glasses." An alternative interpretation of "we" at line
C16) is that it is a mitigating substitution for "you" in
a directive, similar to the token in COB). This
interpretation appears dependent an hearer knowledge of
previous and following discourse, which is more
characteristic of managers’ use of WEB than of U1E6 or
UE7. To make the point clearer, EHE follows up with the
utterance at lines C19) to CE1): "But you are not putting
two tumblers and two wine glasses..." This utterance
lends weight to the argument that “we" at line C16) in
fact is a substitute for "you," and the utterance should
be heard as an injunction. In any of these
interpretations, both "we’s" in C13) and C16) imply
speaker-hearer divergence, and both cast the speaker in a
decidedly more powerful role and the hearers in a less
powerful role.
209
A Janus-like property of FPPPs in this excerpt both
provides For orderly interpretation of the "we" tokens and
does the crucial work of fostering established power
relationships in the department. In the first instance,
hearers do not interrupt the discourse and ask who "we"
refers to, despite the high probability that there are at
least four referential domains of "we" in the one turn of
talk. The habitual use of "we" in discourse within the
known setting of existing relationships sets up routine
expectations that are recognizable by but below the
awareness of hearers, and on the basis of the emerging
context and speaker/hearer presuppositions, the reference
is accomplished successfully in each case. The "we"
tokens are not subject to challenge because they are
derived from the existing sets of relationships and are
matched to the discourse context effectively by the
speaker. In this sense, interpretation is constrained by
the pre-existing social structures in which the discourse
takes place.
The other face of FPPPs adds to those pre-existing
social structures. Whenever an instance of "we" is used
in a discourse context so that it meshes with speakers’
presuppositions, it reinforces the conditions of the
social and emerging contexts from which the
presuppositions are drawn. Because creating and
recreating social relationships through habitual discourse
BIO
practices is mostly an automatically performed routine
without self-reflexive examination, the use of FPPPs is a
basic component of practical consciousness. In the hotel
setting, the ways "we” is used in manager-worker discourse
is taken for granted. However, because it carries the
capacity to define group membership and differentiates
among groups an the basis of pouter and domination, it is
not an innocuous component.
Both faces of FPPPs reveal their ideological nature.
The referential domains of "we" shape hearers’
understanding of the nature of their social world on the
basis of which domain is evoked by which variant of "we" a
speaker uses in a discourse slot. The very choice of an
FPPP instead of a more precisely referring pronoun, such
as I or you, is itself an ideological choice, even if its
ideological nature is inscribed in the habituation instead
of in conscious choosing. And what about the apparently
non-ideological uses of "we" in lines (09) and CIS)
above? Do they argue against a critical reading of FPPPs
in discourse? Although WEB token in line (09) clearly
refers to speaker, hearer, and others, thus implying
convergence of speaker and audience, it says something
particular to the hearers. It is used in an utterance
about the application of a law to a job task within the
department; it is not used to link members directly to the
job task. As such the speaker says through her use of a
211
UE2 token of "we" that speaker and hearers are equal with
respect to the law. Although it might not be strictly
true in practice within the larger national system of
justice, it is a commonly held belief that all citizens
are equal under the law. To imply otherwise would be
publicly to challenge hearers’ beliefs about their sacred
constitutional rights.
The apparently convergent "we" tokens in line C12)
are an instance of a metacommunicative statement, where
the reference is reflexively made to the current
interaction of speaker and hearers. In this sense, the
speakers is saying to the hearers, "UJe are the same
insofar as we are present and interacting here." It is
likely, however, that the strategic reason for using "we"
instead of "I" is to mitigate the mistake the managers
have made so far in the meeting. The law that requires
high temperature washing of glasses has not been discussed
with employees; obviously the speaker knows about the law,
because she is telling the employees about it now.
Moreover, employees have not been told that there is a new
policy for placing glasses in the rooms and bathrooms, nor
that stewarding will be responsible for changing the water
in rooms, as is revealed in later stretches of the
discourse. These are policy decisions and conditions that
employees should have been told about, and workers’
confusion about how to handle glassware in the rooms is
E1E
understandable in view of their lack of policy
information. Recognizing this, the speaker at CIS) could
have said:
C6) I didn’t talk to you about the glasses, did I?
Instead, she chose to include the hearers in the failure
to talk about the glasses to that point.
If these interpretations of "we" in lines COS) and
(IE) are accepted, all FPPPs in the manager’s turn of talk
are of the divergence type, and all, in spite of a surface
suggestion of convergence, set nanmanagerial employees
apart from managers on the basis of managers’ superior
knowledge, involvement in planning and policymaking, and,
recursively, the right to use FPPPs that define group
membership.
To the degree that tokens of "we" are ambiguous,
hearers predictably rely on the existing social
arrangements, emerging context, and discourse context to
Fill in the slot with an appropriate referential domain of
"we." This dependence on presupposition is itself part of
a supervening presupposition that speakers and hearers
invoke— the knowledge that others will rely on certain
kinds of knowledge to disambiguate references. The
reliance on hearers to make the correct choice of
referential domain based on expected presuppositions is
2 1 3
perhaps the most powerful technique for maintaining
existing patterns of power relationships. I refer to this
technique as a referential enthumeme. because one premise
of the reference is omitted and filled in by the hearer,
involving the hearer actively in the construction of the
discourse’s meaning. The three parts of the enthymeme
are: Ca) "We" is being used to refer to some person or
persons in discourse; Cb) of the eight possible domains of
reference for "we," the hearer can infer one Cin the
absence of interpretation knowledge within the discourse)
from the social, emerging, and discourse contexts, and
that one is the one meant by the speaker; and Cc) the
hearer’s interpretation of the token of "we" therefore has
the domain of reference intended by the speaker.
The referential enthymeme is invoked by instances of
"we" where there is no obvious discourse supply of
interpretive knowledge and the token is sufficiently
ambiguous to require some choice to be made by the hearer,
even if the choice is routinely, formulaically, or
unconsciously made. Certainly, every instance of
reference has some element of ambiguity— virtually all
writers on language, discourse, and linguistics agree that
there are no unique, one-to-one correspondences between
symbols and objects in the world. Each instance of symbol
use requires consideration of the contexts as they have
been discussed above for understanding to take place. It
2 1 4
is in the stretching of the limits of linguistic ambiguity
that speakers farce hearers to rely more heavily on
presuppositions for successful interpretation— i.e.,
finding meanings that coincide with speakers’ intentions—
and bring the referential enthymeme into play. In this
way, existing social structures, including power
relationships and the rights, benefits, and obligations
that they carry, became reinforced through discourse. In
a linguistic sense, dominance is exercised by speakers
when they coerce hearers to rely on presuppositions to
interpret speakers’ ambiguous reference.
The fundamental contradiction demonstrated by
managers’ use of FPPPs in the hotel transcripts is that
which exists between a surface appearance of convergence,
solidarity, and full participation of rank-and-file
employees with managers, on the one hand, and routine
discourse practices involving the use of "we" which
segregate the groups on the basis of asymmetries of power
and privilege. Ulhat is apparent from this examination of
FPPPs is that the modes of practical consciousness fully
recognize the hierarchical power structure in
relationships, and the evidence is clear in the discourse
of the two groups.
E15
Notes
1. Excerpts from the transcripts of recorded discourse
that occurred in the rites described in Chapter S are
numbered as they appear in the transcripts. Each excerpt
is preceded by a code that identifies the rite, the hotel,
and the transcript pages where the excerpt belongs. In
addition, each of these excerpts shows the transcript
lines associated with the discourse under discussion. The
notational conventions used in the transcripts are as
follows:
double slash marks,// //, contain transcriber
comments;
single parentheses,C ), contain speaker pauses, estim
ated to the nearest second;
brackets, C 1, indicate overlapping talk;
equal sign, = , signifies run-on speech with no
pause;
colon, , indicates elongated sound;
underscoring, ____, shows vocal emphasis;
falling slash, \, indicates falling vocal tone;
rising slash, /, indicates rising vocal tone.
In several cases, fabricated sentences are presented
to exemplify a point in the discussion; such sentences are
numbered sequentially and are identified in the text as
being fictional sentences.
E. No conventions exist for naming the surface-level
lexical meanings of discourse as opposed to intended,
inferred, or possible other meanings. Levinson C19B3)
uses both "conventional content" and "sentence-meaning"
for the former concept and notes that non-conventional
meanings are those that are derived by general principles
of inference based on consideration of contextual
factors. Insofar as all discourse meaning depends to some
degree on contextual factors, the use of "conventional
content" may sound misleading. Moreover, Levinson Cp. 17)
asserts that the distinctions between literal meaning and
conveyed or other meanings is not entirely clear. I will
use "lexical meaning" to refer to first-order, literal
meanings such as are found in dictionaries and "implicated
meaning" to refer to meaning that depends on consideration
of contextual factors. In this way, problems of multiple
versions of "sentence meaning" are avoided, as is the
ambiguity of "conventional meaning" and the limitations
implicit in Brice’s term, "speaker meaning."
3. There is little agreement on what counts as reference
types. Only Lakoff C1976) appears to treat epithets as
non-names. Chastain, for example, would be likely to
consider an epithet as a title, and others treat it as a
definite description. Lakoff does not discuss deixis in
216
his hierarchy of noun phrases types. Geach argues against
including relative and indefinite pronouns as referring
items.
4. Descriptions of the organizational context, positions,
and employment practices are based on the author’s three
months of observation and structured interviews with
managers and long-term employees in both hotels. As part
of an on-going project, 21 managers and eight long-term
employees were interviewed. In addition, organizational
documents were examined, including employee handbooks,
training manuals and videotapes, personnel form^,
timekeeping records, safety reports, occupancy reports,
and corporate policy memoranda. This experience of the
organization’s processes and structure has sensitized the
author to detect most instances of atypical practices in
the use of reference.
S 17
C h a p te r 4
To see power is to look backwards in time.
— James P. Carse
If an analysis of organizational discourse, such as
the last section of Chapter 3, is to say something useful
about institutional patterns of domination, further
development of the arguments is needed. To reveal
ideological practices that might be inscribed in the
organization’s routines, for example, a critique of each
routine as a whole is in order. In addition, an adequate
analysis requires a more disciplined account for ideology
than I have given so far, including particularly an
explicit account of the relationship between discourse,
ideology, and institutional structures. At a more general
level, a critique of the organization’s policies,
practices, and relationships must be undertaken, for it is
in the gestalt of the whole organizational setting that
workers’ careers are played out and institutional
domination and legitimation have their effects.
This chapter therefore presents a critical-analytic
discussion of the three rites from which discourse has
already been analyzed and a mare general critique of the
hotels as organizational entities. In each of the
sections devoted to critique, I disclose pertinent
S 1 8
contradictions by pointing out veiling or mystifying
discourse, identify the ideological bases for the
contradictions, and propose simple methods far
incorporating more emancipatory practices into the process
under discussion. In keeping with the premises of this
study, I strive to fix a steady analytic gaze both on
situated discourse and an the issue of meaning-
construction in Giddens’ sense of discursive and practical
consciousness as I critique the rites and the hotels’
practices and policies as a whole.
Before beginning this part of the work, however, I
turn to a description of structuration theory, principally
in the work of Anthony Giddens, to set up a social theory
to serve as a foundation for my discussion of the rites
and the hotels. I have adopted Giddens’ version of
structuration, rather than that of Bourdieu or Bhaskar,
for example, because his views are more accessible, more
systematic, and more comprehensive than those of others.
An exposition of structuration’s main ideas provides three
important contributions to a study of this type: Ca5 It
serves as a necessary foundational theory of society and
the individual actor; Cb) it establishes a basis for
linking social structures and institutional change and for
linking the material presence of the day to day social
*
world and abstract categories used for its description and
analysis— or as Bourdieu C1977, p. 725 terms it, the
2 1 3
dialectic of incorporation and objectification; and Cc) it
provides a vocabulary for a critique of organizations,
actions, and the relationships among members. The last
part of my discussion of structuration is a brief exegesis
on ideology in which the relationship involving discourse,
meaning-construction, and ideology is clarified and its
connection to institutional structures is explored.
Structuration: A Foundational Social Theoru
I have called structuration "foundational" because it
is both an account of individual personality, action, and
reflexiveness and an explanation of how actors’ behavior
is guided by and contributes to social structures. By
structures, I mean here the conventional sense of the term
i
— descriptive categories for repeated institutional
practices and procedures which endure for aggregates of
persons who regularly come into contact with each other.
Giddens calls these "structural properties," and he
defines them as "structured features of social systems,
especially institutionalized features, stretching across
space and time" C13B4, p. 337). He takes "structure" to
be much more intimately bound up with routine practices of
social group members and limits that term to those rules
and resources "recursively implicated" in the members’
repeated articulation of the organization’s
relationships.
eeo
Structuration does not supplant critical theory as a
theoretical basis For critique. On the contrary, it
complements critical theory by supplying an explanation of
individual psychology and group and societal structural
properties, an explanation that is founded on members’
actions. All of Giddens’ explanatory principles and each
of his arguments give primacy to thB routine, everyday
practices of individuals. Most commentators on critical
theory emphasize the Fact that it has strong roots in
Marxist and Freudian theory, and some amalgam of neo-
Marxist and psychoanalytic concepts are usually trotted
forth to serve as a legitimating theoretical Foundation
for critical theory (e.g., Deetz & Kersten, 1363; Geuss,
1981; Kersten, 1986; Pryor, 1981; Smythe & Tran, 1983).
Views on the status of critical theory as a theory,
however, are not uniform. In many discussions, Freud is
avoided Ce.g., Pryor, 1981) or is expressly relegated to a
secondary role (Geuss, 1981). Moreover, some
commentators, like Forester (1985), acknowledge that there
is great variance in the degree to which critical theory
researchers reach down to the Marxist roots, let alone
those of psychoanalytic theory, to inform their work.
Forester (1985, p. 236) observes that critical theorists
in communication research may "reformulate the classical
Marxist critique of ideology in terms of critique of
systematically distorted communication." In other words,
221
the Freudian and Marxist roots might influence current
critical theory less as nurturing concepts and techniques
than as inspirational origins that may merely suggest
contemporary analogues.
In either case, critical theory is not a
comprehensive theory of the individual and society. In
practice, there is not a critical theory, but, as Geuss
points out, multiple critical theories which are
"reflective" forms of knowledge that aim to enlighten and
emancipate individuals in their personal lives.
Structuration complements these critical theories in the
way that some writers have implied psychoanalytic theory
does Cby providing an account for individuals’ personality
and action) and Marxism does Cby providing an account for
the nature of social relations).
To the degree that Freudian and Marxist principles
have been relied upon, they generally are inadequate as
foundations, anyway. Freud’s apparent ambivalence over the
conceptual status of the superego and the "ego-ideal," for
example, has for decades been a theoretical obstacle.
Giddens C1984) analyzes the problems Freud’s concept of
the tripartite self— ego, id, and superego/ego-ideal—
present for social theory; he also concludes that Freud
undervalues humans’ capacity for intentional, conscious
action and overplays the unconscious aspects of behavior;
finally, he notes that Freudian theory is oriented toward
EES
synchronic Features, like static architectures, while a
social theory that aims to integrate human agency needs to
account For actions that take place in real time and
space.
Marx’s ideas may be Found inadequate because they are
too narrow For critical theories’ purposes. They are too
narrow in the sense that they are restricted to a view oF
social relations in which the ownership oF the means and
methods oF production is the central Feature that sets up
class divisions and diFFerentiates exploiter From
exploited. Such a view surely reFlects nineteenth century
economics and sociology and does not anticipate the
immensely widespread practice oF public ownership oF
corporations through stocks, bonds, and other Financial
instruments seen today. Marx’s ideas also are too narrow
in the sense that the modes oF exploitation are assumed to
be uniFormly aligned with class distinctions and operate
consistently against an oppressed group regardless oF the
institutional setting. Ideology in the Freudian-Marxist
roots oF critical theory thus is a set oF
macrosociological belieFs.
Giddens C19B1), however, points out that ideologies
are better viewed as a kind oF knowing than as a kind oF
belieF, and the ways oF knowing may vary widely within
exploited groups. Moreover, ideology must be analyzed in
terms oF members’ experience oF their day-to-day lives
5S 3
where they have some measure of control; moreover,
credence must be given to the kind of power any member of
society has solely by virtue of being human— the power to
create rational accounts For his or her own state of
affairs. Host crucially, Marx’s idea of evolutionism— the
notion that capitalism is the inevitable culmination of
all historical development and that further experience
will see the overthrow of capitalism, the rise of the
socialist state, the eventual dictatorship of the
proletariat, and the ultimate withering away of state
apparatus— such evolutionism says Giddens C1981) is a
misreading of history and puts the whole Marxist
enterprise in jeopardy. Besides the existence of small
societies that have not fallowed Marx’s prescribed
evolutionary pattern, the world’s political and economic
record of the twentieth century seem to refute Marx— so
far. Most trenchant, however, is Giddens’ observation
that oppression, domination, and exploitation are
generated by forms of relationship besides economic ones,
such as oppression because of race, ethnicity, sex, or
other bases of group differences.
Giddens avowedly is not a neo-Marxist C1981);
however, he finds many of Marx’s views "indespensable”
C19B1, p. 8) to structuration theory, particularly Marx’s
ideas on time and capitalist production. Similarly,
Giddens takes much From the psychoanalytic tradition,
224
particularly from the ego psychology of Erik Erikson.
Both these origins of ideas and inspiration are noticeable
in the description of structuration that follows. To
organize and briefly present a complex theory that has
taken several books and a decade to develop CGiddens,
1979, 1901, 1982, 1904, 1986) necessitates abridgement.
The most comprehensive treatment of the theory is found in
Giddens’ The Constitution of Socletu. but important
details are worked out elsewhere. I have organized this
description under three headings: Ca) the nature of human
agents; Cb) structure, system, and structuration; and Cc)
discourse, meaning, and ideology.
The Individual and Agencu
Three major characterizations of the human agent
distinguish the structuration view. First, every member
of society is a knowledgeable and capable agent. Second,
every agent has three layers or aspects of cognition and
motivation, called the "stratification model" of
consciousness. Third, routinization of everyday practices
is indispensable to an understanding of human personality
and action.
Giddens frequently refers to persons as "agents" to
acknowledge a human capacity that often is neglected in
social theory— the capacity to create and carry out plans,
imagine alternatives and choose among them, and affect the
social institutions in which they participate. He aims to
2E5
understand human nature in terms of action, and a
considerable measure of human action is attributable to
creative initiatives by agents. To place action in the
center of an account of human nature requires that people
be vieujed, first of all, as knowledgeable about several
key aspects of social life. Humans must know how and why
people act generally as they do so that their own actions
will integrate smoothly with those of others. Giddens
describes knowledgeability this way:
CAD 11 of us know a great deal about why we behave as
we do, and about the social conventions relevant to
that behaviour. Philosophers are prone to talk of
such knowledgeability in terms of the ’reasons’ that
individuals have for their conduct. However, this
can be misleading, for it suggests that human conduct
involves a string of discrete reasons aggregated
together: and that every act has some definite reason
attached to it....I prefer to speak of the
’rationalization’ of human conduct as an inherent
feature of human behaviour...by the rationalization
of conduct I mean that human agents chronically, but
for the most part tacitly, ’keep in touch’ with the
grounds of their activity, as a routine element of
that activity C1982, pp. 29-30).
This "reflexive self-monitoring" is grounded in
previously acquired knowledge about how humans behave and
adds to the agent’s store of such knowledge with each
instance of its invocation. While it is more or less
continuous, reflexive self-monitoring occurs typically as
an unreflective activity, thus is done "for the most part
tacitly." The tacit awareness of what happens socially
and the monitoring of the self against a backdrop of
practical knowledge constitutes knowing "how to go on," a
2 2 6
phrase Biddens borrows From Wittgenstein C1982, p. 30).
This sense of knowledgeability is similar to Berger and
Luckmann’s idea of "recipe knowledge," which they see as
limited to "pragmatic competence in routine performance"
but occupying a prominent place in agents’ social
knowledge C1967, p. 42). Berger and Luckmann, however,
make the key point that social knowledge is "socially
distributed," that is, it is held differently in different
forms by different people, and no individual’s
knowledgeability exactly coincides with another’s. On the
other hand, Giddens treats knowledgeability as a key to
agency— to know how to go on is to go on in some fashion,
oriented in some unique way to one’s own sense of social
relations.
Besides knowledgeability, agency depends also on the
general human trait Giddens calls "capability." To be
capable is to be able "to have done otherwise."
Capability is thus not a form of knowledge but an attitude
towards action; it is an acknowledgment of the potential
for choice-making and Free will. It also implies that
individuals can give accounts for why they act as they do
and, by extension, for why others behave as they do. Thus
these concepts constitute a major part of the explanation
of agency: knowing how to go on and going on with the
attitude that in any instance of social action one could
have done otherwise. They link action to the self, which
SS7
is a learning, selF-reFlexive being, and they avoid
reliance on any sort of intrapsychic homunculus that tells
an individual what to do Cor as Giddens describes it, a
Freudian "miniagent within the agent as such" C1984, p.
43) ) .
But knowledgeability and capability alone do not
Fully account For agency. Giddens observes that humans
are not uniFormly conscious oF their histories and
actions; moreover, consciousness is not a unitary concept,
either. He proposes a model oF consciousness that has
three levels oF awareness, which may be viewed
simultaneously as a model oF three aspects oF motivation.
The unconscious is a state that has no reliance on memory
or recall; behavior that is ascribed to the unconscious is
behavior without motive, that is, behavior that is
reFlexive. Recently Giddens has written that "the
unconscious can proFitably be regarded as the other side
oF language— what cannot be said in language because it is
the Foundation oF linguistic usage" C1986, p. 539).
Actions that are not unconscious, however, may be
described and justiFied with varying degrees oF
completeness and accuracy by agents. Some actions can be
described and explained because agents are Fully aware oF
what they are doing. This level is what Giddens calls
"discursive consciousness," and it is reFlected in the
kinds oF behavior that might prompt a question about what
2 2 8
the actor is doing. On the other hand, much of human
behavior is so routine, so chronically performed, that it
is unlikely to elicit such a question. The level of
awareness associated with these mundane, deeply ingrained
practices is called "practical consciousness," and it is
intended to apply to such behavior as the use of language
and paralanguage in conversation, and any other highly
routinized social behavior. Giddens describes practical
consciousness this way:
Intentions and reasons which agents have for what
they do are sometimes capable of being expressed in
what they can say about the conditions of their
action. In other words, agents can in some degree—
fluctuating according to historically given social
circumstances— give a discursive account of the
circumstances of their action. But this by no means
exhausts what they know about why they act as they
do. Many of the most subtle and dazzlingly intricate
forms of knowledge embedded in, or constitutive of,
the actions we carry out are done in and through the
practices which we enact. They are done knowledge
ably, but they are not necessarily available to the
discursive awareness of the actor C198G, p. 536).
The concept of practical consciousness reintroduces to
social theory "the skilled knowledgeable subject, whose
activities are geared into the continuities of day-to-day
social life and whose knowledgeability is expressed in
practice" Cp. 538). UJhile there is a clearcut difference
between unconsciousness and the two types of
consciousness, no such hard difference exists between
practical and discursive consciousness: "Between
discursive and practical consciousness there is no bar;
229
there are only the differences between uihat can be said
and what is characteristically simply done" C19B4, p. 7).
Practical consciousness plays a pivotal conceptual role
in bridging individual action and larger scale social
institutions. A central aim of Giddens’ project has been
to overcome the traditional dualism between objectivism in
social theory, which focuses on society and holds large
scale institutions to be of primary interest, and
subjectivism, in which the human agent is the focus of
interest. Structuration finds valid and useful ideas in
each perspective, as well as obvious shortcomings.
However, by engaging objectivism and subjectivism in
social theory not as a dualism but as a dualitu. the
traditional dichotomizing of perspectives and resultant
conflict among theorists is dissolved. More importantly,
by consolidating objectivist and subjectivist views in the
"duality of structure," Giddens offers a much more
powerful explanation of social structures Cin the
traditional sense) and change than either view can alone.
Structure and Structuration
Structure is not to be thought of as a set of boxes and
arrows or, by the usual analogy to architecture, as the
skeletal framework of a building called the social
institution. Rather, structure is the set of rules and
resources of various types that are used by agents in the
course of their social interactions. The rules and
2 3 0
resourcss are "recursively implicated" in the reproduction
of social systems because in their instantiation they also
serve to rearticulate the social systems from which they
are drawn. Giddens therefore links structure directly to
individuals and their actions: "Structure exists only as
memory traces, the organic basis of human
knowledgeability, and as instantiated in action" C1984, p.
377). Structure is coexistent with human action:
Structure is both the medium and the outcome of the
human activities which it recursively organizes. By
the recursive character of social life I mean that
social activity in respect of its structural properties
exists in and through the use of the resources which
agents make in constituting their action, which at the
same time reconstitutes those structural properties as
qualities of the systems in question C1986, p. 533).
Rules are the implicit, repetitive patterns of
interactions persons experience in their day-to-day
lives. Rules are linked to practice this way: "To know a
rule, as Wittgenstein says, is to ’know how to go on,’ to
know how to play according to the rule.... Rules generate—
or are the medium of the production and reproduction of—
practices" C1979, p. 67). To carry out life’s activities
according to rules in this sense does not necessarily
imply that persons must be able to describe or formulate
rules; they need only follow rules as implicit guides to
action.
Resources are of two types, and each type contributes
to the determination of power relationships. Allocative
resources "derive from human dominion over nature"
2 3 1
C19BH:,p. 373) and represent the material objects available
For human exploitation, such as space, buildings,
Furniture, production materials, etc. Authoritative
resources are derived From dominion over other persons and
represent the non-material harnessing oF human
activities.
Structure is realized generally in any social system
as practices oF signiFication, domination, or
legitimation. These three terms indicate houi individuals’
actions "act back" on the social settings in which they
occur: SigniFication draws on rules and resources to
instantiate communication practices; domination draws on
rules and resources to instantiate power relationships;
and legitimation draws on rules and resources to
instantiate sanctions. Separation oF these three aspects
oF "structures" is an analytic artiFice; all three are
present in day-to-day interaction.
The concepts oF agency, the duality oF structure,
rules and resources, and structures still do not account
For the interpenetration oF individuals’ actions with
social systems and institutions. The key concept oF
structuration theory that binds them together is time-
space. Social action does not occur in Freeze-Frame;
every instance oF thinking, talking, or any other "doing"
occurs in real time and in a spatial context. The
temporality oF events is what gives substance to practical
2 3 2
consciousness because it provides for the routinization of
experience. Structure is "out of time," but social
systems— Giddens paints out that cohesiveness or
similarity is unnecessary for individuals to be aggregated
into collectivities— and institutions are sets of human
activities that are bound by time and space. Social
practices that are stretched out over time and space by
repetition and recursivity become inscribed in daily life
as routinized behavior. Those routinized aspects of day-
to-day interaction become the rules and resources of
social activities at the same time as they are
reconstituted as the "structural properties" of social
systems and institutions. Structural properties are
features of social systems that stretch across large units
of time and space, such as hierarchical positions, command
authorities, and so forth.
Time is experienced by agents in three ways: (a) As
an ongoing stream of action in which the present behavior
is embedded but which is typified as the flow of daily
experience, called by Giddens duree: Cb) the finite
temporality of a human lifetime, which is an existential
and historical reference point sometimes called Passim
and Cc) the longue duree. or institutional time in which
chronically repetitive interation stretches out to the
time-space horizon. Time and space are crucial to
structuration theory because they are the basis for
2 3 3
individuals’ acquisition of rules and resources: "Time,
space and repetition are closely intertwined....The
cyclical character of repetition or social reproduction in
societies governed by tradition is geared indirectly to
the experience and mapping of time" C1979, pp. 204-205).
In addition, each individual’s experience of time-space is
like a holographic map, which only approximates the maps
of other individuals; differences and similarities of
values, behavioral responses, knowledge, expectations, and
so forth may be accounted for by reference to the
similarities among the individuals’ personal histories.
Rules and resources used by persons in the activities of
social production and reproduction are available only
because persons experience social interaction in a time-
space context. Recursively, it is only because of time-
space that social systems and institutions may have their
characteristic structural properties attributed to them.
Power. Domination, and Structure
In its most elementary form, power is an individual’s
capacity to transform a social interaction by virtue of
being "able to do otherwise." Power understood as
individuals’ transformative capacity, then, is linked to
the attainment of personal goals in social interaction.
In the process of attaining goals, individuals rely on
their command of resources— their authoritative and
allocative resources— which are asymmetrically distributed
234
within any social collectivity. Domination is created by
the asymmetrical distribution of authoritative and
allocative resources, and domination is a central
empowering Factor that allows one person to transform a
social interaction and prevents others from doing the
same. In other words, resources are drawn upon and
reproduced as power relations in interaction; and the
transformations achieved in interaction are reproduced as
structures of domination.
From this perspective, power is not the same as
resources, it is not something an institution has, nor is
it, as Foucault envisions power, a ubiquitous social
coercive capacity to be cornered by speculators as if it
were a precious metal; instead, power in structuration is
a quality of being human, part of "being able to do
otherwise." However, it is in the comparative exercise of
power in relationships that a difference is made:
Differential capacities to transform social interaction,
based on differential control over resources, results in
asymmetries of domination. Domination in turn reproduces
asymmetrical access to resources, and so on. The system
may be represented as in Figure 1.
2 3 5
Figure 1 System of domination and power CaFter Giddens,
1979, p. 92)
DOMINATION
ALLOCATIVE
RESOURCES
AUTHORITATIVE
TRANSFORMATIONS POWER
The remaining two aspects of structure, legitimation
and signiFication, interact with domination and are
mutually produced and reproduced with it. Available
sanctions in day-to-day interaction operate through norms
that are part oF agents’ knowledgeability to produce
legitimations oF interaction. InsoFar as interaction
reFlects power relationships, sanctioning legitimates
domination; in addition, insoFar as signiFication
articulates the social order, it is an expressive aspect
oF both domination and legitimation. These relationships
are depicted in Figure 2. Giddens says oF the
interdependence oF the three aspects oF structure:
IF signiFication is Fundamentally structured in and
through language, language at the same time expresses
aspects oF domination: and the codes that are _
involved in signiFication have normative Force.
536
Authorization and allocation are only mobilized in
conjunction uiith signifying and normative elements;
and, finally, legitimation necessarily involves
signification as uiell as playing a major part in co
ordinating forms of domination C1979, p. 107,
emphasis in original).
Figure 5
Relationships among structures and interaction in time-
space CGiddens, 1904, p. 59)
STRUCTURES SIGNIFI CAT I ON -*-*-D0M I NAT I ON -*-*«LEG ITI HAT I ON
t j t t
MODES INTERPRETIVE FACILITY NORMS
{
SCHEMES A A
$ ▼ t
INTERACTI ON CONVERSATI ON ------► POWER ---► SANCTIONS
Discourse. Meaning and Ideoloau Revisited
In Chapter 3, power and ideology were linked to
discourse through the concept of intersubjective
understanding that is acquired and expressed in language.
The social cognition view was amended by Foucault’s point
that what can be known is constrained by institutional
conventions of language use; therefore, the premises of
discourse as well as the content of discourse itself set
ideological limits on social actors. That analysis is
extended in this section by application of structuration’s
approach to agency, practical consciousness, and the
contingent nature of all structural properties.
Giddens reminds us that "a quite direct line of
continuity" extends from Saussure to Derrida, from
structuralism to post-structuralism, which locates meaning
2 3 7
in "the play of difference within linguistically
constituted codes" C19B6, p. 540). This sense of meaning,
which occurs outside the binding of time-space to
experience, inevitably involves an idealization of
speakers and hearers; that is, they share unambiguous
codes within a homogeneous speech community.
Paradoxically, the linguistic terms of such an approach to
meaning are necessarily ambiguous, varying in their sense
from one analyst to another. Such an approach to meaning
constrasts sharply with meaning as it is constructed by
participants in actual conversation, where they "are able
to follow what each other say and relate what is said to
the referential properties which are involved" Cp. 540).
The very fact that persons in conversation are able to go
on Cand in fact typically do go on rather smoothly)
demonstrates not only the sharpness of meaning as it
occurs in discourse, in contrast to the fuzziness of
meaning in codes, but also the central role of practical
consciousness in the creation of meaning in social
interaction. Without routinized procedures of
signification and communication, discourse would be
riddled with false starts, hesitations, interruptions,
questions, clarifications, and challenges. It is the
binding of time-space into the knowledgeability of agents
that provides them the capacity to go on. Giddens sums up
the relationship this way:
2 3 8
. . .uie have to accept that it is temporallu and
spatiallu situated conversation, not the text and not
writing, which is most essential to explaining what
language and meaning are. The consequences of this
for social theory, I think, are very important. They
essentially involve rescuing the knowledgeable agent
as the conceptual center for social analysis, and
situating what ’knowledgeability’ is in the context
of the ongoing practices of social life. Social life
does then not appear as a phenomenon external to
agency, but is contingently produced and reproduced
in the moments of social activity stretching across
the time/space context of action C1S88, p. 541).
For social theory and for analyzing organizational
relationships, then, meaning is more usefully understood
as emerging in the actual instances of social
interaction. It is "contingently produced and reproduced"
based on participants’ knowledgeability, and it
interrelates with structures of domination and
legitimation in all instances. This is why
presuppositions transform what is vague and ambiguous in
linguistic and paralinguistic codes to fluent stretches of
apparent understanding in conversation; this is how
implicature transforms what is objectively paradox, irony,
sarcasm, or inappropriateness to fully understandable and
appropriate contributions to conversation.
Ideology works through the signification-domination-
legitimation structures whenever domination of one group
or individual over another is objectionably exploitative.
All relationships are to some degree exploitative, since
interdependence necessarily encompasses mutual
£ 3 9
exploitation of others’ capabilities, resources, talents,
and so forth. Exploitation becomes objectionable when an
individual’s or group’s capacity for self-direction and
control is so diminished that they are denied a Just
opportunity to participate in the determination of their
ouin futures, their fundamental human dignity is
suppressed, or their "life chances" in the Weberian sense
are foreshortened.
The routinization of domination through repeated
articulation of power relationships in discourse serves to
legitimate exploitation, normalizing over time social
systems that favor the sectional interests of one group or
individual over those of others. As Giddens describes the
relationship:
If ideology be understood... as the modes in which
exploitative domination is legitimized, we certainly
must include ’knowing how’ within this category— bath
in regard of how the position of hegemonic groups is
sustained and of how those in subordinate positions
limit or resist their hegemony. ’Ideas’— or Cmore
accurately) signification— are inherently embroiled
in what people do, in the texture of the practices of
daily life. Some of the most potent forms of
ideological mobilisation do not rest upon shared
beliefs Cany more than shared normative commitments);
rather, they operate in and through the forms in
which day-to-day life is organized C1901, p. 68).
Ideology in this view reflects differences,
differences of beliefs about what is real or important or
right and wrong or mighty and weak, differences of taste
or style or interests, differences of goals or motives or
valu
2 4 0
es. Mare importantly, ideology is grounded in differences
of group membership when ramifications of those group
membership differences entail the exploitation of one
group by another. Giddens’ view holds that ideology is a
way of understanding and acting that accepts as normal and
legitimate asymmetrical relations of power and control
among groups.
Giddens’ view also suggests that where there is no
exploitative domination among groups, the situation is
without ideology. But to claim that groups exist in
relation to each other without ideology is to state that
any differences among the groups are innocuous with
respect to the distribution of power and control;
moreover, such a state is likely to occur only when there
are negligible differences among groups in terms of
available resources, the opportunity for self-direction,
and the capacity to achieve one’s goals. Clearly such an
egalitarian neutrality is an idealization. Giddens (1979,
p. 1B9D points out that individuals have unique wants and
interests and in the aggregate may be said to participate
in group wants and interests. Even the powerless, the
disfranchised, the subordinated groups have interests in
the world and share a way of understanding relations of
power and control, regardless of whether or not they
accept the status quo. As to those groups that might
oppose the proffered legitimation of exploitative
2 4 1
domination: Are they non-ideological? Such groups have a
program for a particular understanding, their own
practical consciousness, based perhaps on their own
culture, which takes as normal and routine a set of
relations among groups that differs from the status quo.
Such is often the case among deviant groups like
prisoners, street people, and graduate students.
Thus all versions of practical consciousness may be
deemed ideological. Ideology becomes noticeable,
however, whenever groupness is challenged, is reinforced,
or otherwise is made an issue. Insofar as discourse is
inevitably inscribed with social variables such as class,
ethnicity, age, and education CEdwards, 1985-, Gumperz,
1982a, 1982b; Gcherer & Giles, 1979), there are no
instances of nonideological discourse. Linking discourse,
ideology, and groups, Giddens C1979, p. 188) says, "To
analyse the ideological aspects of symbolic orders... is to
examine how structures of signification are mobilised to
legitimate the sectional interests of hegemonic groups"
Cemphasis in original).
In the analysis in Chapter 3, the structures of
signification that matter are those that say what group
boundaries are and who is routinely to be deemed a member
of what group. The importance of this interest lies in
the fact that the established order of group relations in
the hotels under study, as is the case in most complex
2 4 2
organizations, reveals crucial differences among groups.
A small group of persons have legitimated rights to
participate in decision-making that affects large sections
of the enterprise; the rest of the workforce do not. The
same small group generally receive a much larger
proportion of the rewards of the enterprise, both in money
compensation, perquisites, and benefits and in prestige.
Members in this small group have legitimated rights to
allocate tasks and resources and to evaluate the quality
of both; out-group members are obligated to perform tasks
as assigned and to request resources, and their evaluative
role as regards both is perfunctory. In-group members are
empowered to evaluate the performance, conduct, and even
the appearance of out-group members; this is not a
reciprocal arrangement.
This description obviously refers to asymmetries
between managers and nonmanagerial employees. These
asymmetries devolve upon individual members of one group
and another, but insofar as they apply systematically as a
condition of group membership, they may be thought of as
salient group differences.
Chapter 3 revealed the discursive penetration of
practical consciousness in the routine uses of "we" by
organizational groups; that analysis is an example of
ideology in the discourse of members and provides the
basis for critique at the level of interaction. After the
2 4 3
following discussion of ideology in FPPPs, I present a
critique of the rites in which the discourse excerpts were
embedded; and I conclude with a critique of the
organization at a level where time-space is stretched over
greater expanses— the hotel organization as a whole.
The particular relevance of a study of FPPPs Cfirst
person plural pronouns) to ideology is the peculiar
capacity of "we" to encode group membership. In Chapter 3
I showed that eight different configurations of the
referential domain for "we" are evident in the hotel
corpus; two other configurations— S + 0, 0 not
identifiable, and H + 0, 0 not identifiable— are
conceiveably present in other discourse. The use of these
eight domains by powerful members, in this case by
managers, shepherds hearers’ understandings of group
boundaries and membership. Portraying nonmanagerial
employees as, first of all, belonging to a distinct group
from managers and, secondly, not belonging to a powerful,
privileged group reinforces the domination system that
exists prior to the discourse. It also legitimates the
\
existing order of relationships whenever the implied group
boundaries and memberships occur in discourse without
being challenged.
2 4 4
Some domains of reference are more ideological than
others. By "more ideological" I mean that some domains
set speakers off from hearers more distinctly and
distinctively than others, some align speakers with
vaguely powerful others, and some imply that hearers are
as controllable and as in need of control as children. Of
the eight domains, those that maximally separate
managerial groups from non-managers are WE5, UJE7, and UJEB
Csee Table 3). I cite those domains because they clearly
faster divergence among speakers and hearers, they are
used solely by managers, and the interpretation knowledge
they prompt hearers to use consists of remote stretches of
discourse or presuppositianal knowledge in the absence of
specific linguistic cues.
The relationship between tokens of "we" and ideology
is more subtle than this, though. Most tokens of "we" are
somewhat ambiguous— some considerably more ambiguous than
others. None is without competing alternatives of
referential meaning because referential meaning always is
set in a context of social relations. The degree of
ambiguity varies from one referential domain to another,
and speakers appear to exploit that ambiguity as a way of
leading hearers not only to certain interpretations, but
more importantly, to rely on certain kinds of knowledge—
24 5
their practical consciousness knawledgeability— to
formulate an interpretation. Eisenberg C1984) places
ambiguity of this sort in an analytic framework that
encompasses not only speakers’ goals but also an
overarching theory of organizational communication.
It is with ambiguity that the "referential enthymeme"
enters as a discourse technique speakers employ to lead
hearers to formulate interpretations based on
presuppositions. The paradigm case of the referential
enthymeme's three terms are: Ca) Unless otherwise marked
in the discourse, tokens of "we" conventionally signal
speaker-hearer solidarity; Cb) there is no discourse
marker associated with this token of "we"; CcD therefore,
"we" is understood conventionally, given the nature of the
situation and the speaker-hearer relationship.
The referential domains that have no cues for
interpretation knowledge are those with the most inherent
ambiguity; however, other domains have a high number of
new discourse predicates or a discourse context that
provides many competing alternatives for hearer
interpretation. Those domains, too, are rich in
ambiguity. The powerful might use ambiguous constructions
in their discourse to create impressions of their
alignment with even more powerful groups, to suggest even
greater separation between speaker and hearer, or to
£ 4 6
create a tension within hearers who cannot determine how a
token of "we" is to be taken. The latter case is a matter
of audience mystification by which resistance to dominance
is thwarted because hearers cannot establish an
understanding of circumstances and relationships.
Finally, it is important to recognize the time-space
aspects of discourse, particularly discourse which hearers
experience within practical consciousness. Giddens C1984)
makes the point that practical consciousness, while not
the same as unintentionality or unconsciousness, is
nonetheless a mode of consciousness where understanding is
grounded in the doing of interaction and not in
discursivity. Thus when hearers engage in discourse
within practical consciousness, they are immersed in the
flow of communicative behavior unself-reflexively; in that
condition, an encounter with speaker utterances that would
warrant reflection to achieve an adequate and critical
understanding would require speakers to change their modes
of consciousness toward a more discursively aware state.
In addition, participants in oral discourse typically
are simultaneously speakers and hearers, and the discourse
they produce is not directly recoverable for immediate
analysis subsequently to redirect the interaction. The
luxury of analyzing linguistic tokens in face-to-face
discourse is reserved for those who are not direct
2 4 7
participants; tokens of "we," unless extraordinarily
marked as contradictions, appear, exhaust their evanescent
lives, and fade out in fractions of seconds. As such,
they readily become part of the routine discursive
penetration of social life. Insofar as practical
consciousness is the mode of experience that normalizes
domination through discourse and other forms of
interaction, it is the repository of ideology.
But the rules and resources of practical
consciousness, when extended over vast stretches of time
and space, are the constitutive properties of institutions
CGiddens, 1384). Institutional orders, then, are inscribed
in the day-to-day routines of members in conversation Cand
their other practices); and in terms of the duality of
structure, institutions are reproduced in each instance of
interaction between and among members. Dominant
ideologies that flow from the sectional interests of
legimitated, powerful groups at the institutional level
are manifested in the forms of routine practices that
members exhibit in their day-to-day practices.
A Critique of Three Rites
Each of the rites that contributed discourse to this
study is discussed below. New employee orientations,
departmental meetings, and Lewittalks are analyzed in four
steps: Ca) Identification of "veiling" Cfley, 1985) or
2^8
mystifying discourse; Cb) specification of the dominant
ideologies of the rite, that is, pointing out the
sectional interests that are maintained by the discourse
of the rite; Cc) description of contradictions between
apparent purposes and nature of the rite and what is
revealed at a deeper level of critical analysis; and Cd)
suggestions for changes in the routine practices of
members to emancipate members. The fundamental purpose of
this critique is "to look for the modes in which
domination is concealed as domination... and for the ways
in which power is harnessed to conceal sectional interests
on the level of strategic conduct" CGiddens, 1979, p.
193) .
By emancipation I mean providing opportunities for
individuals to increase their control over decisions that
affect their work processes, their social satisfaction,
and their career opportunities. Emancipation is not meant
to imply that some employee groups are in need of greater
freedom and others are not; in a typical corporate
environment, institutional structures of domination are
such that most members in field installations experience
the frustration of diminished personal control, and where
hegemonic relationships occur, the hegemon is as much
trapped by dependency as is the subordinate CGiddens,
249
1979). Thus unveiling manipulative language and proposing
procedural changes in rites are aimed at emancipating all
members and transforming the rites into more humane
activities.
Neui Employee Orientations
Jacob Hey C1985, p. 63) describes "veiling" as "using
language that claims to deal with reality in an objective,
correct, matter-of-fact way, while hiding reality from
language’s users." Manipulation of hearers through veiling
language use may be accomplished by manipulating the form
or the content of language or both. Although Mey
considers principally language qua language and misses the
richer implications of discourse, his sense of veiling
captures the manipulative aspects of what occurs in
organizational discourse. To use the term "manipulative"
is not to imply that speakers knowingly and intentionally
mislead hearers; it may mean that they participate in a
tradition of discourse routines in which manipulation by
sectional interests is deeply and subtlely inscribed. It
means at least that, but in some instances it means a more
knowing and intentional form of manipulative discourse.
Three elements in the discourse of new employee
orientations stand out as potentially veiling: Labelling
the employment contract, reading the employee handbook,
and describing work requirements in corporate-prepared
videotapes. The employment contract at both hotels is
E50
Framed as an employee handbook. At Airport Lewitt it is a
two-part document, half of which is titled, "What You Can
Expect From Us," and half of which is titled, "What We
Expect From You." At Marina Lewitt the handbook is a
glassy paged, new manual prepared in a slick art-deco
style. Like the shorter and drabber manual at Airport
Lewitt, it contains employee benefits and obligations and
enumerates major conditions of employment and work rules.
"What You Can Expect From Us" lists benefits and
general working conditions, such as when paydays occur,
recognition system, advancement policy, and so Forth.
"What We Expect From You" describes fifteen areas of work
rules, each of which may contain up to a dozen
instructions or injunctions concerning implementation of
the work rules. For example, "Appearance and Dress"
begins with a general rule which is followed by many
detailed instructions:
As a representative of Lewitt, you are expected to
take great pride and care in your personal
appearance, cleanliness and general grooming. Female
employees are asked to be conservative in the use of
make-up and jewelry, and in their dress and hair
style. Non-uniformed female employees may wear
regular street-length dresses, coordinated pants
outfits Cno jean style or jean materials), or long
skirts and dresses, provided they are appropriate for
daytime business wear. Male employees are to be
clean and well-groomed, with hair neatly trimmed at
all times Cetc.J.
In addition, the second half of the Airport Lewitt manual
contains a list of 28 statements describing "conduct
251
considered to be unacceptable." Newcomers might be
mystified about to whom such conduct is unacceptable at
this time and who must consider it unacceptable for
disciplinary action to be initiated. By omitting an agent
in the description of offenses, the offenses take an an
aura of universality that goes beyond the social context
of the place of employment.
Although the introduction to conduct rules at both
hotels is an instance of veiling language in new employee
orientation, a mare obvious example is found at the end of
the Airport Lewitt manual. All newcomers are enjoined to
sign the last page of the manual— the employment manager
stands over the group of newcomers while they sign it
before they are released from the orientation session. A
statement above the signature line on the last page says:
"I have read the contents of this Handbook and am willing
to operate under the set of agreements outlined in its
pages." Although by expressing such willingness the
newcomer is agreeing to the employment conditions set
forth in the manual, the term "agreements" describes the
contents of the manual before the newcomer signs it. In
other words, the veiling language operates here to
characterize the work rules, list of unacceptable conduct,
and other provisions of the contract as something on which
the employer and new employee have reached agreement.
252
Substantiating this interpretation is a line earlier
in the manual that says in part: "Should there be a
situation in which you Fall short of achieving our
mutuallu agreed upon standards of performance and conduct,
you will be creating a situation where corrective action
becomes necessary" Cemphasis added). Consider, however,
that this is typically read and endorsed during the
newcomer’s First day on duty: Not only has there been no
negotiation of performance and conduct standards. there
also has been no opportunity For newcomers to inform
themselves fully about what the pre-existing standards
might be. The endorsement must be done in nearly complete
ignorance of the performance and conduct parameters within
which the newcomer must live at work; moreover, many
newcomers understandably will be lulled into a false sense
of concordance with the employer by the manual’s
description of rules as agreements. My argument is not
that negotiation should have taken place but that the
language characterizing the employment contract Implies a
degree of employee knowledgeability and willing acceptance
that is unwarranted by the circumstances.
Other veiling practices occur in the orientation
sessions, involving both the employment managers who
conduct the programs and the corporate-produced videotapes
that are used at both hotels to punctuate the in-class
activities with general employment information and the
5 5 3
company’s philosophy. The managers either read to the
newcomers or have the newcomers themselves read aloud from
the handbook to the group. The explanation given at
Airport Lewitt for this practice is:
15EEB:LAX:ORN:8
CE6) EMGR: And I’m not Cl) trying to insult
CE7) anyone’s intelligence by Cl) y’know having
C58) people read alou:d^or Cl) y’know things like
CS9) that.=But I think that it works really well if
C30) we do that.
After several newcomers nervously stumble over the
readings they have been assigned, the manager calls on a
Hispanic woman whose command of English is obviously
limited. This newcomer CNEMP4), assisted by another
CNEMP6), struggles through the embarrassing situation:
1S5S6:LAX:ORN:IE
C15) EMGR: Okay. CE) Ah:m Cl) Maria. Why don’t you take
C16) the next one\? Workers’ compensation/.
C17) NEMP4: Ow::, like reading I don’ know the wor:ds.
C18) EMSR: Oh:. Okay.
C10) NEMP4:: Ha=heh“heh=heh.
CEO) EMGR: You could take=d’you-do you wanna try?
C51) NEMP4: M:. //inaudible// this one on insurance?
(22) EMSR: Ah: this one right here.
254
C 23 ) NEMP4: Okay. The ploymen insurance is provided uhm I
(24) don’t say that right=h::...
(25) NEMP6: You on workers’ compensation?
C26) EMSR: Uh huh/.
(27) NEMP6: Well it’s on this. (1) This one.
(28) NEMP4: Ah working compentian?
(29) NEMP6: M hm/ .
(30) NEMP4: Okay.(l)This insurance is paid full by Lewitt
(31) for you medical spenses and continue (1) ah=I
(32) don’t know what what is.
(33) C 1
(34) NEMP6: Sources.
(35) C
(35) EMSR: You’re doing very well.
(37) NEMP4: Sources of income::under certain: cir=
(38) circum=s=s=cir=
(39) C
(40) EMSR: Circumstances, urn hm.
(41) NEMP4: =cumstances if you s::oon be:(4)o::ort on the
(42) yob.
(43) EMSR: Q:ka:y, good. Ahm, see you did fi::ne. (l)Ah:m
(44) work*
(45) C
(46) NEMP4: I was scare=h::=heh.
(47) EMSR: =ers’ compensation is provided by the state
2 5 5
The manipulation is in the claim that "it uiorks well
this way." It works well From the manager’s viewpoint if
it makes the presentation of a laundry list of employment
rights and benefits less tedious for her. It also works
better insofar as it places the language of the work rules
in the mouths of the newcomers, whose internalization of
the rules as their own guides to conduct is in the best
interests of managers. CThe manual says: "We feel it is
important for you to know, understand and accept as your
own, the standards and conditions of your employment with
us....") It does not work well for many of the newcomers.
People like Maria suffer extreme embarrassment, even if
they experience no stage fright when reading before
strangers in their native language. Many newcomers would
recognize the parental attitude implicit in the practice:
The litany here involves newcomers reading passages
Followed by an extensive interpretation and commentary on
each by the employment manager. One major effect of this
process is an exemplification of haw little the newcomers
know and how much the manager knows about "how to go an"
in this work setting; this dependency relationship is
clarified by the constant reminders that newcomers should
seek information from managers whenever they have a
question. In this way, rival interpretations of how to go
on may be minimized and homogeneous practical
consciousness is promoted.
2 5 6
Some of the more obvious veiling language in the
videos concerns the reasons given for various conditions
of employment. For example, the narrator states about
guest service: "Everything you do— smiling, offering a
friendly hello, providing courteous service and expert
assistance— all contribute to guest satisfaction. At
Lewitt, you’re on stage, no matter how you feel, whenever
you’re on duty." Auditors of the video are told it is
their job to see that excellent service is made to look
easy. The reasons given for these requirements are that
satisfied guests return and help Lewitt continue to grow,
"building secure careers for Lewitt employees throughout
the company." The narrator then says that certain
guidelines help employees provide "that special Lewitt
service" Cthereby implicitly helping to build their secure
careers). These guidelines are presented as descriptions
of pre-existing and standardized ways of behaving, for
example, "All employees wear name badges."
When discussing safety requirements, the narrator
says that the reason it is "critical" that employees
follow safety rules is employees will "meet difficult and
demanding guests" and might be "asked to do the
impossible." The relevance to employee safety of meeting
demanding guests who might ask one to do the impassible is
not immediately clear. It might be that when attempting
to meet guest demands an employee must be conscious of
£57
safety requirements; or it might be that safety may be
used as a legitimate reason for not attempting to do "the
impassible," whatever that might encompass. The final
video comment on safety is "Keep your world a safe one."
This implies that safety is for the protection of the
employee; however, the obvious safety concern, in this era
of high-rise hotel fires and personal injury lawsuits, is
for the protection of guests.
Security regulations are portrayed by the narrator as
"protecting everyone— guests and employees." YBt the
rules that are discussed caver: Ca) reporting "incidents"
or suspicions of theft; Cb) limiting employee access to
the hotel; Cc) employee package inspections and passes;
and Cd) identification cards for employees. The narrator
justifies these restrictions on employees: "These
guidelines are the first step in providing a secure
jobsite for you." Rules that prevent employees from
entering or leaving the building except by the back door
Cwhich are not applied to higher level managers), rules
that require package inspections, rules Cnot specified in
the video but spelled out in the manual, which is referred
to repeatedly in the video) that permit management
inspections of employee lockers, and injunctions to report
to security anything the employee feels "just isn’t right"-
-such rules do not speak directly to the security needs of
employees as much as they do to the security needs of the
2 5 8
corporate interests. Those interests are to minimize
capital losses through employee misconduct and to maximize
gross income from new and repeat business, principally
through the agency of satisfying guests. This
interpretation is borne out by the fact that video cameras
are installed in locations for surveillance not of public
or guest areas but of employee work areas. In Airport
Lewitt, security cameras are mounted even in the back
corridors of the hotel where money and high value
materials are rarely handled.
Contradictions in the discourse of new employee
orientations appear, then, in at least the following three
ways. Pre-existing, non-negotiable standards of conduct
and employee benefits are portrayed as "agreements." The
practice of directing newcomers to read aloud employee
manual provisions is portrayed as a preferred way to
proceed. Quest service, safety rules, and security
measures are justified as positive influences on employee
careers.
The sectional interests served by these features of
orientation, however, are overwhelmingly those of
management. To permit actual negotiation with employees
over standards and rules releases management’s hold on
sanctions and the legitimation structures of the
organization as diagrammed in Figure E, above. To trust
newcomers to interpret the employee manual for themselves
259
might release the personnel managers’ hold on
signification in the organization and might break down the
barriers between managers and employees that is based on
managers’ possession of decision and procedural
information and employees’ need to request it. To admit
the self-interest of management in safety and security
rules is to admit that rank-and-file employees are not
trustworthy in general and that the real objective of the
firm is not advancing every employee’s career but
maximizing the company’s growth and profitability.
The ideology that dominates the orientation sessions
is one which says there is a Lewitt way of working and
living in this workplace; it pre-exists and is not subject
to modification by workers; it is made more palatable to
workers by describing it as mutually constructed for the
benefit of workers; and workers are expected to be told
about the Lewitt way of working and behaving and then to
conform to the description provided. Transmission of a
certain practical consciousness is the objective of new
employee orientations at Lewitt Cand perhaps in most other
corporate workplaces). The practical consciousness that
emerges from an examination of the actual practices of
orientation sessions is one where smiling, attractive,
well-behaved workers passively cooperate with an elite
group of managers who dispense information, interpret
E60
rules, decide on policies, evaluate performance and
conduct, and manage subordinates’ careers.
Counteracting this dominant ideology in the practices
of neui employee orientation would mean first of all
removing veiling language and procedures. The Marina
Lewitt handbook is less manipulative in this respect than
the Airport Lewitt manual and may serve as a guide. The
general manager’s welcome in the handbook, for example,
says:
UJe want you to know that your role is important to
the efficiency and success of our facility. In order
for you to become familiar with how the hotel
functions, we are providing you with this Handbook
containing information on company benefits, practices
and regulations.
The videotapes could be similarly adjusted to more frankly
state the bases of policies and rules. More important for
the emancipation of all persons involved, however, is
addressing the practical consciousness that is modeled for
the naive newcomer. To open up the constraining
dependence that is created by the orientation discourse,
worker involvement and choice-making should be increased.
A range of informational resources, orientation
activities, peer conferences, and training materials
should be made available for newcomers to choose from
according to their own assessment of their orientation
needs.
261
No personnel system operates in isolation;
orientation of neui employees must be linked to ongoing
activities of information exchange, performance
evaluation, compensation, career development, and so
forth. In this regard, the entire personnel management
system must be explained, Justified, and critiqued with
newcomers. This approach to orientation of newcomers
obviously requires a much greater commitment of resources
by the firm than is presently the case; however, total
workforce turnover figures of over fifty percent per year
indicate that employee commitment to the firm is lacking.
An investment in increased employee participation might be
justifiable on simple economic grounds.
The major change that is needed, however, is to
unveil discourse that perpetuates the separation of the
workforce into two groups, one a docile, reactive, and
comparatively powerless underclass, and the other a
powerful, active elite. Managers and nonmanagerial
employees are presented in orientation as two sectors of
Lewitt hotels who have clearly different communities of
interest, vastly different legitimated capacities to take
action, and mysteriously different privileges concerning
the ways of interacting in the workplace. However, as I
show in the last section of this chapter, the differences
of stakeholding in fact are not so great as they are
portrayed; and the differences that are evident in
2 6 2
individuals’ and groups’ practices are based more on
adventitious privilege than on necessary allocations of
power and responsibility.
Certainly the answer to unveiling this aspect of
manipulative discourse in orientations is not to increase
uses of FPPPs. The contradictions inherent in veiling
uses of "we" have already been discussed and are taken as
givens in this exploration of new employee orientation
practices. A broad-based democratization of procedures in
the workplace, such as worker participation in
disciplinary boards, rules review committees, selection
panels, and so forth would result in different procedures
and responsible officials as topics of discussion in
orientation sessions.
I have devoted much of my critique of rites so far to
the new employee orientations because that activity
engages employees who are disoriented with respect to
"locating the self in time and space" CVan Maanen, 1976).
In their first encounter with the organization they
typically are in a special state of neutrality, having
little recourse to past workplace knowledge and no
understanding yet of the day-to-day realities of the new
workplace. In this state of "liminality" CTurner, 1969),
newcomers are at the interstices or margins of social
structures, or they are at the "lowest rungs" of the
social order. The consequence of this liminal state is a
2 6 3
special susceptibility to proffered interpretations of the
social structural properties they are entering, including
salient differentiations and inequalities CTurner, 1969,
p. 97). When the proffered interpretations are based on
an ideology of exploitative domination, the patterns of
domination are more easily reproduced in the newcomers’
practical consciousness than would be the case if they
could influence the very process of their own orientation.
Departmental Training Meetings
Some low level democratization of workplace
activities is evident in the second rite studied,
departmental training meetings. The evidence, however,
generally is of the most unstructured, accidental type:
Two women housekeepers have their babies present in the
meeting; a housekeeper leaves the audience area during a
bed-making demonstration and joins in the demonstration to
show what kind of problem the "turn-down ladies" have been
experiencing; a training program designed to inform
employees about a new guest incentive campaign and solicit
employee support for the campaign is translated
simultaneously into Spanish.
The use of veiling discourse, however, is more
evident than any efforts to invite employee participation
in the design and implementation of activities that affect
their work processes. Departmental meetings are described
in the employee handbook and orientation sessions as
2 6 4
opportunities For employees to "communicate with
supervisors and Fellow employees about CtheirJ area oF the
hotel." That opportunity, however, is structured into
meeting Formats as intermittent requests by managers For.
audience questions; in every case observed, involving Five
diFFerent departments, solicitation oF questions occurred
in conjunction with a topic already in the discourse and
initiated by the manager. No Forum is included in the
departmental meetings For open-ended employee discussion;
thus, it is necessary For employees to interrupt the topic
at hand to Freely "communicate with" others about their
areas oF concern.
In all but one oF the departmental meetings observed,
the events comprising the meetings have three traits in
common. First, all meetings are Fully orchestrated by
managers beForehand, and the items on the agenda and the
methods oF presentation are not generally known to rank-
and-File employees until the meeting unFolds. In one
case, even the departmental employees who were called on
to set up props For a meeting activity— the pillow-
stuFFing game in housekeeping— did not know the purpose oF
what they were doing as they arranged Furniture,
pillowcases, and pillows.
Second, each meeting has a game-like motiF oF
presentation, involving either prize giveaways, ad hoc
quizzes, or an entertainer-audience routine. In the last
2 6 5
motif, the managers stand in front of the other employees,
often using a microphone to talk to them, and perform a
loosely scripted role of instruction or information,
usually containing comic remarks and in-jokes. There are
frequent instances of audience applause in mast of the
departmental meetings.
The third trait in common involves content. Policy
changes and new procedures are presented as pre-existing
realities, fait accompli. In the cases where the central
purposes of the meetings did not expressly involve policy
changes, the announcements of changes occurred at the
closing of the meetings. The discourse of one meeting
occupies 62 pages of transcript; on page 61 the division
director announces two policy changes, one minor Cchanging
the color of room service’s napkins) and one of
extraordinary impact— changing the hours of operation of
the cafe where most of the employees attending the meeting
work. During the last two minutes of a two hour meeting
in another department, the manager informs the workforce
about two procedural changes that alter the way they do
their tasks and the way they work with another
department.
Among the various types of reference I discussed in
Chapter 3 is naming. One naming practice evident in the
meeting rite is so noticeable and so captures the concept
that discourse inscribes group identities in practical
£ 6 6
consciousness that it warrants pointing out here. The
largest occupational group in the hotels is that of
housekeeper, or maid. Unskilled men and women are hired
by the housekeeping departments to clean rooms, make beds,
replace amenities and linens, and scrub bathrooms. Hen
are called "housemen" and do no in-room cleaning; they
provide supplies for the women, move furniture, and
perform hallway maintenance, interior window washing, and
other heavy cleaning. Women are formally called
"housekeepers" and perform all the routine cleaning tasks
inside guest rooms. A small group of housekeepers spruce
up guests’ rooms on the evening shift and turn down
guests’ beds.
Throughout the hotels, it is a common practice for
managers to refer to housekeepers as "ladies," and to
those who turn down guests’ beds as "turn-down ladies."
The label "lady" has an archaic, honorific ring to it; but
in the circumstance of applying the term to poorly
educated, mostly immigrant, unskilled workers, "lady"
takes on a decidedly ironic ring. Connotations often
associated with the lexical meaning of that term include
refinement, aristocracy, tradition, class, wealth, social
leadership; even the common phrase, "the lady of the
house," implies she who is in charge of the house. In
this latter sense, "lady" should be expected to describe
women occupying positions in top management, rather than
2 6 7
those in the lowest paid and least prestigious positions.
When used in conjunction with “gentlemen," the term more
appropriately and conventionally refers to women in
general. But that is not the way it is used in the
hotels.
The use of "ladies" by managers, as occurs frequently
during the housekeeping departmental meetings, serves to
aggregate a large group of women under one rubric that has
a mock-honorific undertone. It says to the group of
women, you are recognizeable by what you have in common,
and we are going to refer to you in a way that we do not
refer to any other collectivities of women in this
enterprise, as the "ladies." What the "ladies" have in
common, however,, is not refinement and high status, but
low status, low pay, and a lack of social refinement. By
using this label, managers create a specialness about the
group of women that sets them off from other groups simply
by referring to them. The cruel irony is not in the
choice of "ladies" as the appellation, but in the prideful
acceptance of the term by the group of women themselves;
they refer publicly to themselves as "ladies" and "turn-
down ladies." As such, they participate in identifying
themselves as a unique group within the workforce. Their
sense of the group commonality might be other than low
skills, low pay, and low status; they might indeed feel a
special sense of camaraderie and group interest that does
2 6 8
not include their lowly employment and social status.
Nevertheless, the facts of their position cannot be
ignored, since they are the most obvious and well-known
traits uniting the "ladies" into a group; furthermore, the
use of "ladies" by managers toward employees exactly
coincides with the boundaries of the occupation of
housekeeper.
The discourse of meetings expresses more strongly
than that of orientation sessions the dominant ideology
that employees are expected to be passive, docile,
compliant agents in the workplace, and managers are
active, powerful, and informed. Veiling language is so
deeply ingrained in the hotels’ approach to such
gatherings that they are called meetings, rather than pep
rallies, briefings, demonstration sessions, or similar,
more accurate descriptions. Besides naming practices in
meetings C"you guys," "ladies") and calling the gatherings
meetings in the first place, other contradictions are
evident. All hotel media for employees proclaim meetings
of various types to be opportunities to carry out the open
communication that is vital to effective employee
relations. Policy memoranda in the employee cafeteria and
hallway bulletin boards, employee handbooks, orientation
sessions, and Lewittalks all contain statements that open
communication is a hotel goal, and it is linked to
employee careers by claims, for example, in the Airport
263
Lewitt manual, that by taking advantage of the
communication system, "we can uiork together to provide the
best possible service to our guests and so that you can
achieve you own personal and professional goals."
Departmental meetings, however, are designed so that
genuine and open employee participation generally are
blocked. Meetings reinforce the informed
manager/uninformed employee dichotomy and reproduce the
social structures that allow meetings to be fully in the
control of managers in the first place. The ceremonial
games, prizes, and theatrical performances seem to be a
way of involving employees in the proceedings without
according them a consequential involvement in the design
and implementation of meetings. In interviews, employees
and managers alike consistently describe meetings as "fun"
and "exciting." Fun and excitement in this instance
contradict a stated policy of providing employees "an
opporutnity Csicl to participate and communicate your
suggestions, ideas, and any problems..."
The dominant ideology suggests that managers believe
rank-and-file employees will respond most effectively to
leadership if meetings take the form described here. An
emancipatory act affecting both, however, would involve
all organizational members in developing and conducting
meetings and provide a serious— though not somber— and
constructive opportunity for employees to exchange views
E70
and information. The emancipating effect this change would
have on nonmanagerial employees is obvious; it is perhaps
less obvious that managers would be freed from their
perceived need to transform commercial and technical
information and problems into something fun and exciting
for employee consumption. As with changes in
socialization activities, restructuring meetings implies
that managers must release a measure of exclusive control
over allocative and authoritative resources.
Lewittalks
The last rite observed, Lewittalks, is described in
the Marina Lewitt employee handbook as being a "general
open discussion" between randomly selected employees and
the general manager. The description continues:
The purpose of these meetings is to bring out the
employees’ true thoughts, viewpoints, and
suggestions. The Lewittalk meetings, as contrasted
with other staff and general meetings, allows the
line employee the freedom of expressing his or her
viewpoint in the company of peers, directly to the
General Manager, without the presence of other
management personnel.
The one Lewittalk recorded and transcribed begins
with the general manager’s description of how she views
Lewittalks, i.e., as "predicated on confidentiality" and
that employees "are comfortable talking to me about the
way we do business and your Cthe employees’! concerns and
your suggestions of the compliments you have about the
hotel and the people you work with...." The general
271
manager clarifies her intention not to name names to the
supervisors of the employees who attend Lewittalks, and
she explains that the confidentiality requirement applies
to employees, too. Then she describes her method:
What I do during Lewittalks is I take notes and then
I go back and talk to the executive committee about
them....I don’t present issues to them with ’Irene
said such-n-such and such-n-such.’ I try to cleanse
the issues and talk simply about the problem areas or
the good things that are going on, or the suggestions
for change without identifying your names. It’s
important that you understand that because I must
work through the executive committee for change in
our hotel. That’s the policy of Lewitt to work
through our people for change, and so I have to take
this information back to them. After we have
discussed them in the executive committee, I go back
to you in writing with a letter to let you know what
we’re going to do about any of the points....
Thus she portrays herself as an agent of the
executive committee, an aggregation of her immediate
subordinates who manage the six major subdivisions of
hotel operations, plus the executive chef and the
personnel director. As the Lewittalk proceeds, three
identifiable types of interaction take place. Following
the preliminary remarks and introductions all around the
U-shaped arrangement of tables, the general manager
describes and explains the current state of hotel
business, corporate business, and corporate plans. The
employees listen attentively and occasionally ask for more
information or clarification.
A second type of interaction begins when the general
manager prods employees to contribute to the meeting. In
2 7 2
this type of exchange, employees present questions about
hotel policies and operations and describe problems they
have experienced on the job; the general manager responds
by interpreting hotel and corporate policy, Justifying and
explaining existing practices, and even interpreting
general knowledge as an answer to employees’ complaints
Csee the excerpt in Chapter 3 concerning food poisoning
and food in the employee cafeteria). In this type of
interaction, the manager deflects or defuses the
employees’ concerns by providing new information, policy
interpretation, or changing the subject.
Occurring simultaneously with the second is a third
type of interaction. In this type the general manager
seeks clarification of a problem area, writes a note about
the issue on her notepad, and expresses a commitment to
take up the issue with the executive committee. In the
entire transcript, covering 40 pages and over two hours of
meeting time, the third type of interaction occurred seven
times. On two occasions, the general manager engaged in
informal problem-solving with the employees, soliciting
ideas and evaluating them as problem solutions; however,
no implementation of solutions was decided upon in those
instances.
These three patterns of interaction, then, comprise
the content of Lewittalks: Ca) The manager describes and
explains current business activity; Cb) the manager
2 7 3
responds to employee concerns by explaining existing
policies and Justifying practices; and Cc) the manager
responds to employee concerns by taking issues to the
executive committee. This innovative forum for democratic
involvement of nonmanagerial employees is commendable from
three viewpoints. First, employees have an opportunity to
openly express concerns directly to the organization’s top
official, without substantial worry about the usual
implications of political distortion, career jeopardy, or
interpersonal difficulties that attend raising concerns
through the normal chain of command.
Second, employees receive direct and authoritative
interpretations of the firm’s policies and information
about the hotel’s and the corporation’s future. Third,
employees and manager learn to see one another as
individuals and share personal perspectives and goals.
Lewittalks empower employees by making them more informed
about their workplace and by demystifying the power of top
management through personal experience with the general
manager. Likewise, the meetings benefit management by
facilitating direct probes into the grass roots of the
organization to obtain information about work problems,
personal feelings about specific managers, and areas where
the workforce needs information, training, or other
changes.
2 7 4
While it is a commendable move toward worker
emancipation, the Lewittalk nonetheless reproduces the
structure of domination in several ways besides the
manager’s use of FPPPs. Employees are powerless to
influence the form of Lewittalks or who attends. The
general manager sets out the ground rules as a first order
of business; thus Just at the time she brings order to the
participants’ confusion and mystification about what is to
happen in the meeting, she establishes the metarule that
it is she who determines how the meeting will proceed. At
such a time, it is unlikely any rank-and-file employee
would challenge her position.
Contributing to employee mystification is the basis
of their attendance: Employees are invited by letter to
attend Lewittalks, and they are selected by the personnel
director and employment manager. Employees may not
volunteer to attend and cannot demand to attend the
meetings. They may only elect not to attend once invited;
however, they are directed to send an RSVP to the
personnel office. Moreover, none of the long-term
employees interviewed could explain the reason why some
employees are invited and others are not. Consequently,
the only control employees have over their attendance at
the most egalitarian type of meeting in the hotel is a
passive, negative control— to opt not to attend.
2 7 5
Finally, it is worth noting that the content of the
meetings involves no active problem resolution by the
nonmanagerial participants. The general manager
effectively removes problem-solving from the ambit of the
meeting by insisting that any issues that cannot be
disposed of by her own deflection or defusing must be
referred to the executive committee. Employees are
informed in a letter about the corrective action taken by
management. Employees therefore participate only as
sources of information for management, a channel for grass
roots assessments of how policies and procedures are
working at the shop floor level of the organization. They
also receive official interpretations of "how to go on" in
the workplace and may thus serve as agents for the
official view of how policies should be applied outside
the executive offices.
The Lewittalks therefore seem to empower employees in
an extremely limited fashion. Because they have no
control over the form, participants, content, or problem
resolution in the meetings, they remain passive,
relatively powerless members of the organization.
According to the duality of structure concept, by
participating in the Lewittalks employees contribute to
the reproduction of their own powerless positions in the
organization— acceding to the executive committee right to
resolve problems solidifies the taken-for-granted premise
£ 7 6
that employees do not solve the kinds of problems brought
up in Leuiittalks. Furthermore, no redistribution of
either allocative or authoritative resources from
management to rank-and-file employees takes place as a
consequence of Lewittalks. The central contradiction of
this rite lies in that fact: Lewittalks are touted as the
major empowering forum for employees, the technique
instituted, as the employee handbook says, to provide
"input CthatJ counts" in the "constant critiquing of the
Chotel’sJ work environment." The crucial outcome of
Lewittalks with respect to domination, however, is the
reconfirmation of the rule-making and decision-making
power of upper management and the relative powerlessness
of non-managerial employees.
An emancipatory move would be to permit employees to
designate themselves as participants in Lewittalks.
Members of the workforce at all levels could Jointly work
out format ground rules for the meetings. Ad hoc problem
solving groups could be designated by participants to
address the problems that are brought up in the
Lewittalks. Employees who raise problems they believe
exist between departments or between themselves and other
individuals, or problems involving their own work
processes should be empowered to work through the problems
with other employees and managers concerned.
2 7 7
Summaru of Rites Critique
The contradictions that are common to all three rites
discussed here involve the publicly proclaimed or taken-
For-granted purposes of the rites and their actual
underlying effects that reproduce existing structures of
domination in the hotels. Surely the rites accomplish
legitimate and needed work of the organization— informing,
training, rallying, obtaining information from the
workforce, etc. However, the legitimation structures upon
which their scheduling, design, interactional format,
constraints on participation, and content are premised
turn each rite into a protracted instantiation of the
existing domination patterns. Similarly, the constraints
that flow from existing domination and legitimation
structures frame what can be said and how within the
discourse of the rites. The what and how of discourse
recursively reconstitutes the fundamental relationships of
persons participating in the rites.
In the asymmetrical access to and use of allocative
and authoritative resources that is evident in each of the
rites, group membership in a managerial group of roles and
an undifferentiated nonmanagerial group can be
identified. The groups are identifiable only by virtue of
the presence of the other’s apparent power and
privileges. This defining aspect of dominance is borne
out by Deschamps C1981), who notes that "the dominant
2 7 0
discourse assigns their place to individuals through
defining, locating and ordering each of them in relation
to others." In addition, the group identity of managers
and nonmanagerial employees uias shown in Chapter 3 to be
also both reflected in and reproduced in members’ routine
linguistic practices.
This analysis does not consider the asymmetries in
the distribution of allocative and authoritative resources
among the various levels of management, nor does it
analyze group identities reflected and reproduced within
the discourse of managers alone. This is so because of
the overwhelming differences between the resources
available to managers as a whole versus those available to
nonmanagerial employees. The analysis also coincides with
the group boundaries identifiable by the study of
reference practices discussed in Chapter 3. UJhat needs to
be added at this point is an analysis of how the existing
structures of signification, domination, and legitimation
affect the organization as a whole and the worklives of
individuals in the workforce.
Critique of Two Hotels
The analysis presented so far— both the critical
discussion of organizational rites above and the analysis
of power and ideology in discourse reference in Chapter 3—
focus primarily on interaction per se at the level of
discourse. Both focuses of analysis, however, also
2 7 9
constitute evidence of systematic institutional veiling or
manipulation through communicative action. This section
Focuses on the institutional implications of ideological
veiling in the two hotels and their parent corporation.
It concludes the critical examination of discourse
practices that affect and effect structures of
legitimation and domination, and consequently influence
workers’ careers.
When I refer to the organization, I mean the social
system in Giddens’ C1984) sense, a patterning of social
relations that have instituted reproduced practices across
time-space. In this study, the organization is the
instituted set of reproduced practices involving persons
employed by Lewitt, or alternatively by the Airport Lewitt
or the Marina Lewitt. Where differentiation is necessary
to make a point, I will specify the appropriate
organizational entity.
Rational and Veilino Legitimation
The reproduced practices that constitute the
organization provide the rationale for and rationality of
organizational events. Richard Harvey Brown C197B),
citing Garfinkel’s work, notes that members of corporate
enterprises "tend to legitimize their activities by
accounting for them in terms of rationalistic vocabularies
of motives" Cp. 370). In the rationalizing effort, some
categories of activity become defined as normal and
EBO
rational while others are not. Members of the executive
committee make hotel-wide decisions, rather than low level
supervisors or nonmanagerial employees. This state of
affairs is rationalized through the greater vision and
perspective managers axiomatically have at higher levels
in the organization. Rationalization also may be
accomplished tacitly, through inherited practices. For
managers to call employees by first names or nicknames,
for example, is rational within the set of relationships
created by recurring practices at the two hotels; it is
not rational, however, for nonmanagerial employees to call
executive committee members by their first names,
principally because "it is Just not done."
The legitimation by rationalistic vocabularies of
motives, as I already have demonstrated in the case of the
discourse of certain rites, in effect defines the
organizational realities of relevance and the order of
social relations. Brown C1978), too, links this process
to power and hence to domination: "The study of reality
creation is a study of power, in that definitions of
reality, normalcy, rationality, and so on serve as
paradigms that in some sense govern the conduct
permissible within them" Cp. 371). The sense in which
conduct is so governed is the formation of practical
consciousness and its maintenance by historical practices
and structural properties.
E81
Brown C107B, p. 376) extends his analysis by noting
that dominant interests come not only to control the
contents of reality Ci.e., the means and objects of
material and social production) but also to define the
foundational assumptions of the organization, such as
"rights," "obligations," "legitimacy," and, I would add,
group membership domains. For example, decision-making
may be considered an allocative resource controlled by
management at Lewitt; however, a more deeply sedimented
reflection of the domination pattern is found "in the
design and imposition of paradigmatic frameworks within
which the very meaning of such actions as ’making
decisions’ is defined" CBrown, 1978, p. 376).
Accordingly, Lewitt’s new guest incentive program is
described in the corporate-produced training videotape as
"not Just another program but a whole new way of life" in
which a "new level of guest service" becomes the norm.
UJhat constitutes a corporate way of life and what
constitutes acceptable guest service is conveyed to
workers as a predetermined reality.
Besides these rational legitimations that dictate or
constrain organizational realities, members also
systematically distort or veil conditions for various
reasons and ends. Veiling might be carried out
intentionally, such as when a division director tells a
group of housekeepers that implementing 19 additional
2 8 2
guest services as part of the new incentive program will
be "fun" and "exciting" C02227:RLB:DEP:HSK:33D for the
employees who must provide the extra services. On the
other hand, veiling can be unintentional as when it is
produced as part of the routine, inherited practices of
the organization that are historically institutionalized.
For example, the employment manager tells newcomers in
orientation sessions: "Weekly schedules are posted in your
area at least one week in advance." This statement is
generally not true. A week's advance posting of work
schedules is rarely if ever carried out in practice,
particularly in areas where multiple or split shift
operations are the norm. The employment manager presents
this information, though, either in the naive belief that
it is true or without reflecting on it at all, because it
is part of a routinized practice which was in place before
she was employed.
As an inherited veiling, it is not necessarilu
intentional; if the manager knows about the true practices
of posting work schedules and still claims there is a
week’s advance posting, responsibility for knowingly
maintaining the veiling practice falls upon her
personally. Ironically, only if she knows about the
veiling nature of her statement is there a possibility
that she can interrupt it and address the matter more
accurately. Consequently, inherited veilings are the more
S83
insidious ones because they have less likelihood of being
interrupted by their perpetrators. IF they are already in
place and are presented as taken-For— granted conditions,
they are in eFFect pre-legitimated.
Not only may veilings be analyzed as intentional or
unintentional, they also may be more or less consequential
in the uiays they aFFect receivers. Consider two Lewitt
hotel examples. A basic message is expressed similarly in
the employee handbook, the incentive program videotape,
and by a general manager in a training meeting: Increased
guest satisFaction leads to greater rewards For you, the
employee. The rational vocabulary oF motive here is
grounded in a sort oF employee compensation trickle-down
theory. Increased guest satisFaction yields increased
repeat business, which raises revenues and proFits, which
help the Firm expand, which will provide increased
advancement opportunities For current workers. The
rewards, on the contrary, would most massively and
directly accrue to the Family that owns the corporation;
moreover, increased worker rewards occur only within the
conFines oF locally established pay rates and worker
incentive systems that appear only remotely related to
individual or organizational perFormance. In addition,
currently there is a national managerial turnover rate
that creates more vacancies than reasonably can be Filled;
2 8 4
more promotional opportunities would certainly work
against the firm’s survivability.
A second Lewitt example of veiling is revealed in the
frequently heard claim that I paraphrase as, "we’re one of
85 hotels with the Lewitt name." Most of those hotels are
not owned outright by the Lewitt organization; many are
simple contract for management arrangements, while others
are various kinds of partnerships. Although the veiling
phrase implies greater corporate strength and unity with
respect to other hotel chains than is true, the commonly
held belief that "there are 84 more like us" is relatively
innocuous.
These examples show veilings of different degrees of
consequence. In the first, a distortion of motives might
lead workers to greater sacrifice of their time, effort,
or interests in the unfounded belief that they are
increasing their career chances. In the second, employees
are probably not affected materially, even though they
hold false beliefs about the nature of their
organization. How can a critical analysis of
organizations be approached so that the veiling practices
that make a difference to workers’ interests are separated
out from those that are fundamentally benign?
Distortions in Structural Properties
A scheme for analyzing veilings is available in John
Forester’s C1382, 1885) approach to bounded rationality.
SB5
Forester views "distortions" in much the same way hey
thinks of "veiling"— as discourse that manipulates or
misleads hearers and misrepresents or suppresses
information. Forester analyzes distortions in two
dimensions, the autonomy of the source and the contingent
nature of distortions. Source autonomy refers to
distortions that occur from ad hoc, interpersonal sources
versus those that derive from institutionalized structural
properties of social systems. This dimension is parallel
to the personally created versus inherited veilings
described above. Contingent nature takes account of
whether or not the distortion is inevitable Cor
unforeseeable or socially unpreventable) versus
nonessential or socially preventable. Forester’s analytic
matrix C19B5, p. E43), then, has four sectors as shown in
Figure 3.
Figure 3
Four types of distortions of organizational rationality
AUTONOMY OF SOURCES
Ad hoc Systematic/structural
CONTINSENCY
Inevitable I II
Socially
unnecessary III IV
Sectors I and II contain random and idiosyncratic
misinformation Cl) and inevitable distortions because of
2 8 6
technical specializaticn in the division of labor Cl I).
Sector III addresses instances of willful manipulation,
deception, bluffing, etc., by individuals in an
unpatterned fashion. These three categories are of less
relevance to this study than the avoidable distortions
that derive from the structural properties of social
systems shown in sector IV. Forester describes sector IV
distortions as based on structural legitimation in the
same sense that Giddens refers to the structuration of
rules and resources through legitimation structures.
Distortions in sector IV are grouped into three classes:
Ca) monopolistic distortions of exchange, Cb) monopolistic
creation of needs, and Cc) ideological rationalization of
class or power structure.
Examples of sector IV distortions are readily
apparent in the Lewitt corporation and hotels. I will
illustrate each of the sector IV classes by one or more
examples from the Lewitt practices, and I will conclude by
describing how distortions are linked to individuals’
career trajectories and suggesting initial steps that
should be taken to emancipate all Lewitt employees.
A major distortion of exchange monopolized by hotel
managers is the employment contract. It is expressly
stated in the employee manual CAirport Lewitt) and
handbook CMarina Lewitt). All conditions of employment
are presented as pre-existing rules, benefits, and
207
obligations. Regardless of the management practice of
referring to it as an "agreement," the contract as
presented is sanctified as a predetermined, immutable
articulation of the employer-employee relationship. Much
of the actual relationship in practice is informally
negotiated and adjusted over time; however, the employment
contracts in the manuals at both hotels have clauses which
preclude employee initiatives in any future adjustments of
the relationship. One such limiting clause is a global
prohibition against "refusing to obey a legitimate and
responsible order of a supervisor"; at the tlarina Lewitt,
"orders" are replaced with “requests," making the
prohibition deeper.
Another limiting provision is the employer’s wrap
around clause that prohibits not only the offenses
stipulated, but also "violations of any other established
hotel or departmental regulation." Finally, employees are
required to sign a pledge to abide by all current and any
future policies, procedures, and practices. The terms of
employment, however, are not grounded in social necessity
or even in business necessity. A handy proof of the
nonessential nature of most management work rules lies in
the radical change toward negotiability that occurs when
aggressive unions are elected to represent workers’
interests.
2 8 0
Another monopolistic distortion of exchange involves
the reward system. Beyond wages and tips, the reward
system consists primarily of a series of game prizes
awarded in conjunction with various employee motivation
campaigns. Typically, prize winners need either to be
lucky game contestants or to complete a task that is
unrelated to their Jobs, like collecting tokens.
Consequently, much of the reward system— i.e., that part
commonly designated as the "incentive system"— is
distributed at the whim of management and is patently
unrelated to individuals’ job performance as such. The
totally serendipitous and irrational basis for the awards
obviates any influence by individual employees on the
system’s outcome. At a deeper level of distortion, the
game-like quality preserves the system’s surprise
potential and hides from consciousness the fact that
employees do not participate in the formulation of what
should constitute adequate rewards for being employed at
the hotel .
The second class of distortions in sector IV,
monopolistic creation of needs, refers to one group’s
privilege to institute and foster dependency relations. A
number of dependency relations at Lewitt hotels have
already been described: withholding or limiting
information about corporate plans, local policy changes,
or other matters in which employees have a group interest
2 8 9
and warranted involvement; exclusive managerial rights to
develop solutions to operational problems; and
substituting Fun and entertainment in meetings For
substantive consultation an work matters oF consequence.
In addition, managers create dependency relations through
a Fringe beneFit that is peculiar to the lodging
industry. AFter a year oF employment, Lewitt employees
may have up to twelve nights lodging Free at any Lewitt
hotel, space permitting. Managers and employees alike cite
this provision as a major beneFit oF employment. Greater
beneFits accrue, however, to the Firm. The rooms given
away would otherwise remain vacant; although no room rent
is gained or lost, any guest, paying or not, will spend
money in other venues oF the hotel during a stay.
Moreover, the practice builds employee loyalty, not only
in virtue oF the Fringe beneFit per se. but also by
eFFectively preventing employees From staying in
competitors’ lodgings. Yet this gratuitous program leads
many employees to exclaim their delight in the belieF that
their employer is sharing the Firm’s resources with them.
The central example oF sector IV’s third category—
ideological rationalization oF class or power structure—
involves teamwork. Virtually all the managers interviewed
discussed teamwork as an essential trait oF successFul
operations. The term "team" is ubiquitous in the
employment literature oF the two hotels, the videotapes,
290
and the discourse of managers and employees in the rites
studied. The Airport Lewitt presents a motivational
award monthly, quarterly, and annually to prominent
workers; the award is called the "Spirit Team Award." A
similar award is given at Marina Lewitt, where employees
are placed on the "Lewitt All Star Team." In the
institutional discourse about teams, the concept of a
Lewitt team is one of exclusivity: New members are
privileged to become a member of the team; once on the
payroll, an employee is automatically "on the team" and
has opportunities to be inducted into even more exclusive
team structures, too.
To be on a team implies that there are specific roles
for members to play, and they must play their
predesignated roles properly for the team to win, that is,
for the enterprise to succeed. Success of the enterprise
thus depends on their acceptance of assigned roles. Teams
also have leaders, captains, managers, owners, etc. In
virtue of their mystification about where policy comes
from, how decisions are made, why some persons must punch
a timeclock and use the back door while others do not,
etc., employees can tell intuitively that they are not
leaders, captains, managers of the team. Their assignment
to be subordinate and docile is implicit in the metaphor
of being "a good team player."
2 9 1
The heavy use of the team metaphor also tends to
minimize individual and group differences among members of
the workforce. By implying that managers and rank-and-
file workers alike are members of the same team, the firm
is asserting tacitly that there is an overarching
community of interest joining them that goes beyond their
divisions of authority and responsibility. The discourse
shows, on the contrary, that the sectional interests and
identities are predominant. More importantly, the team
label works against ethnolinguistic identity in that it
attempts to forge an in-group of members comprised of
persons who are constituted in various in- and out-groups
in the larger, surrounding society. Homogeneity of the
workgroup— accepting the premise that all employees
socially are first and foremost a team— forces members to
seek out and remain aware of the supposed commonalities
shared with all others in the workplace. Ironically, this
very fact is a likely reason that over time employees
continue to see themselves as members of subgroups with
their own sectional interests. A further irony is that
workforce homogenization usually is not in the long-range
interests of an organization anyway. It is commonly
acknowledged among organizational researchers that diverse
perspectives, experiences, and solutions to problems
enriches an organization’s resources for survival in
mildly or very turbulent environments.
5 9 2
Ideologies and Career Trajectories
Distortions and veilings of the type exemplified in
the discourse of the hotels are inscribed as received
norms and taken-for-granted social practices. They are
established by recursive use over long stretches of time,
not only in the hotels studied here, but perhaps more
importantly, in typical work organizations generally.
They have become naturalized expectations for members, and
for the most part they have sunk into the social clay so
deeply that they are experienced and known
nondiscursively. They are expressions of the positional
hierarchy and of social class and educational differences,
insofar as those characteristics of workers are correlated
with position in the hierarchy. The distortions and
veilings thus are axiomatic accessories of power,
prestige, and privilege; moreover, they are corruptions of
the signification structures and contribute to the
corruption of legitimation and domination structures.
As features of practical consciousness, distortions
and veilings in organizational discourse inhibit
individual emancipation in two ways. First, workers
cannot discover the real social conditions in which they
are living; practical consciousness tells workers that the
extant structures of domination are natural and
inevitable. Second, by definition practical consciousness
is not open to self-reflective questioning and to
£ 9 3
challenges of the premises on which social relations are
based. The only way emancipation can be begun is by
interrupting the practical consciousness, either by
changing structural properties of social systems from the
outside or by enlightening members of the system with new
and persuasive interpretations of their routine practices.
The former technique is highly unlikely, in view of
the entrenched condition of dominant interests, who,
afterall, share some of the practical consciousness that
structures interaction as perceived by subordinate
individuals and groups. Both managers and workers, for
example, take it as natural and rational for workers to be
called "you guys" but for managers never to be referred to
that way. The latter solution implies that practical
consciousness must be converted to discursive
consciousness. Discursive consciousness is required for
gaining the necessary "transcendent subjectivity" CBrown,
197B, p. 37£3 to either escape the social system’s
constraints or gain control over its authoritative and
allocative resources.
Increasing discursive consciousness, however, depends
on a number of personal skills and attitudes, none of
which is uniformly distributed in the workforce.
Foundational knowledge of a sophisticated kind is needed.
By this I mean a socio-political awareness of the
interests, methods, available leverages, and power of
2 3 4
various individuals and groups in the organization must be
1
i
acquired. Gaining knowledge of this sort is somewhat j
I
dependent on cultural acumen, knowing beforehand what
constitutes forms of power, leverage, interests, and so
forth.
Also needed for increasing discursive consciousness
is ambition. The worker must not only perceive unveiling
as desireable, but also deem it feasible. Many workers
I
express the belief that "it doesn’t matter" how they I
|
explain the actions and discourse of others in the J
I
workplace because the dominant ideology of management will j
I
prevail anyway. This cynicism stands in the way of j
1
increased self-reflexivity and makes emancipation more
difficult to begin.
Emancipation cannot be initiated without certain
!
!
requisite skills, chief among them linguistic skills. j
Sensitivity to the possible meanings in discourse demands 1
skill and experience with the syntax, semantics, and
pragmatics of the discourse that typifies the dominant
group. Socio-cultural knowledge of the possible contexts j
in which discourse occurs is also necessary for a more
self-reflexive reading of the dominant discourse. More
important for emancipation, though, is the linguistic
skill needed to reframe referential practices, reinterpret
stretches of discourse and whole rites, and redefine group
interests and boundaries.
2 9 5
Employees with these three skills and traits— socio
political awareness, ambition for emancipation, and
linguistic skills— are likely to employ them by either
promoting their organizational career trajectories through
rising in the hierarchical order or promoting their
occupational career trajectories through leaving the
domination they are experiencing CBrown, 1982). Employees
without these skills and traits, however, easily become
entrapped in the prevailing ideology and dominant
practical consciousness. Such workers typically expect
only immediate rewards from their employment and envision
no long-term occupational or organizational career
trajectory. Richard K. Brown C1982, p. 126) calls workers
with this coping strategy "careerless," and he notes that
all strategies far coping with work and employment "are
not formulated in isolation from the structure of
opportunities" Cemphasis in original).
Ethnolinguistic minorities are at a severe
disadvantage for increasing discursive consciousness
because their structure of opportunities is impoverished
in three ways. First, they often have little access to
the socio-cultural knowledge necessary for analyzing the
organizational politics and gaining access to allocative
resources from positions without authoritative resources.
Second, they might be so grateful simply to have
employment of any kind that realizing an ambition toward
2 9 6
personal emancipation at work seems ludicrous if it is
sensible at all. Several employees who are recent
immigrants from South Asia and Central America said in
interviews that the best thing for them about working for
Lewitt is the fact that they have a Job. Third, many
ethnolinguistic minorities have sufficient command of
vernacular English to survive in the workplace; however,
they have insufficient command to analyze and reinterpret
the discourse that goes on around them. Many of the
housekeepers interviewed in conjunction with this study
could barely understand the most basic questions about
their employment experience or formulate adequate
responses.
What employers should and can be expected to do to
enhance the discursive consciousness of "careerless"
employees is the subject of another dissertation.
However, a constructive critical theory demands that
suggestions for emancipatory actions be advanced.
Training in standard English usage is essential, and it
should be made attractive and effortless for employees to
take part. Such training should be conducted on duty time
at the workplace where other types of employment training
are done. Full participation in committees, developmental
assignments in various positions throughout the hotel, and
peer involvement in selection, orientation, counseling,
and advancement might be considered. More basically, the
2 9 7
specific determination of emancipatory strategies should
be generated by democratically constituted groups of
employees.
Techniques for initiating emancipation suggested
throughout this chapter fundamentally favor the interests
of any enlightened employer. If lively, creative humans
are desired as employees, the risks associated with their
emancipation must be accepted. John Forester trenchantly
describes the alternative:
When discourse is blacked, the very intersubjectivity
and sociality of human beings is threatened:
Cooperation is endangered, belief can no longer be
grounded, consent cannot be Justified, and attention
is distracted. As discourse is denied to
participants, they are likely to be rendered
dependent, powerless, ignorant, and mystified (1905,
p. 240).
Giddens has commented that the augmenting of
material resources is fundamental to increasing power
(1904, p. 260). Surely, emancipation of workers involves
some measures of this kind of empowerment. But Giddens
also points out that changes in the distribution of
authoritative resources are necessary as well. In the
case of workers who are subject to exploitative
domination, this means giving them more control over
decisions affecting their work, more control over rules
affecting their social interaction, more control over
their careers and lives.
298
Chapter 5
I could drown you in words...
— Stephen Jay Gould
This chapter summarizes the purposes, methods,
findings, and conclusions of the study. In the second
section I present implications for five research areas—
organizational culture, power in work organizations,
critical theory, discourse analysis, and ethnolinguistic
identity. Finally, I discuss limitations and suggestions
for further work in this multifaceted line of inquiry.
Summaru
A study that concludes with a plea for empowerment of
all workers began with a concern for the ways that
discourse in work organizations affects the career
possibilities of employees, particularly among the growing
populations of ethnoliguistic minorities. Along the way
it has been necessary to justify the research, identify
specific questions, explore the nature of power in work
relationships, and ground the study in an overarching
social theory, as well as collect, analyze, and report on
specific data.
One central Justification for this research is the
radical increase in multiethnicity among the workforces of
most so-called advanced nations. The study also is
£ 9 9
warranted by pleas from researchers For a better
understanding of the relationship between language use and
power in organizations. In addition, it is a timely
inquiry because revolutionary changes in worker
relationships are resulting from the introduction of
mediated communication technologies and from increasing
deindustrialization of the American workplace. The
burgeoning academic interest over the past decade in
organizational cultures and symbolism has provided
groundwork for this focused analysis of worker discourse.
The discourse recorded in two hotels comprises the
data which is analyzed within a framework of background
information obtained from observations, interviews, and
examination of archival materials. The focus on the
ritual interaction in the two hotels has enabled me to
frame research questions that are specific to the setting
as a representative case. By focusing on reference in the
discourse I have been able to use the discourse to more
generally examine possible meanings constructed by
participants in the organizational rites. As to research
questions, first it is necessary to ask what the nature of
reference is that characterizes interaction in the hotel
settings. Next I ask how the typical uses of reference by
organizational members contributes systematically to
members’ meanings. Third, I ask how reference affects the
social order and the allocation of power and control in
3 0 0
the hotels. The Final question asks more broadly uihat the
implications of discourse practices are for workers’
career trajectories.
The principal research methods used in this study are
discourse analysis and critical theory analysis, which
uses as part of its evidence the findings obtained in the
discourse analysis. Because simply identifying and
describing patterns of discourse practices is an extremely
weak approach to discourse research, much of the discourse
analysis seeks to explain patterns and forms that are
apparent in the transcripts in terms of speakers’ goals,
auditors’ meanings, and the apparent social and power
relationships among participants. Consequently, the
boundary where what I call discourse analysis ends and
critical theory analysis begins is fuzzy. The critical
theory analysis, however, focuses more directly on the
socio-political nature of relationships in the workplace
and on the distribution of power, privilege, and other
resources than on the discourse qua discourse.
The discussion of findings is situated in
structuration theory and makes extensive use of Siddens’
concepts of rules and resources, the structural principles
of signification, domination, and legitimation, and the
crucial notions of agency and the duality of structure.
By appealing to structuration as a foundational social
theory, I am able to account For systematic patterns of
3 0 1
domination as both a reflection and a consequence of
discourse practices among organizational members.
The examination of discourse reveals practices which
appear highly egalitarian and solidary on the surface but
which carry the markers of a rigidly segmented and unequal
social arrangement. A polite friendliness is evident in
some of the naming practices, definite descriptions, and
*
pronoun use, and this cordiality extends across all levels
of the position hierarchy. However, a more complete
examination of naming reveals that the first name
privilege does not extend toward general managers; workers
are referred to, both directly and indirectly, as "you
guys," yet managers are not so designated; workers are
referred to by occupational or position titles, while this
is not the practice when workers or managers refer to
managers; and while "management" and "employees" are not
contrasted in the discourse, groups of employees are
referred to separately or identified in separate discourse
contexts as belonging to one or the other grouping. Thus
the discourse is found to contain mechanisms that
routinely set up and maintain group boundaries.
A more detailed analysis of first person plural
pronouns CFPPPs), exemplified by an examination of "we" in
the discourse, reveals similar asymmetries of use that
coincide with the hierarchical grouping of managers as a
powerful, privileged group above workers. Eight types of
302
referential domain For "we" are identified, and the kinds
of auditor knowledge associated with each "we" type are
proposed. Differences between managers’ practices and
workers’ practices in the use of "we" types are evident;
managers’ speech is characterized by the types of FPPPs
that maximally separate groups by the pronouns’
referential domains and the types that necessitate
greatest auditor reliance on presuppositions.
Hare generally, this study finds that variation in
the routine uses of "we" contributes to the structuring of
group boundaries and thus of group membership,
encompassing distributions of power, privileges, and other
resources, as well as obligations, within the
organization. A key finding is that this structuring of a
normative world occurs most consequentially in the taken-
for-granted, chronically used routines of conversation.
The routines are historically set in place and are
typically inherited by members of all positions who accept
them uncritically into practical consciousness. By
creating and maintaining group memberships that coincide
with the control of allocative and authoritative
resources, such discourse is ideological. Power in
discourse legitimates structures of domination through the
very discourse it shapes and constrains.
Without substantially equal linguistic resources,
ambition to challenge the dominant ideology, or interest
3 0 3
in co-opting the system of relationships, many
ethnolinguistic minorities can suffer a unique
disadvantage. As a special case of dominated workers,
they are more susceptible to the manipulations of veiling
and distorting discourse because they are less able to
engage intergroup situations with a discursive
consciousness and critically reflect on the socio
political implications of routine interaction. They are
thus more likely to became the "careerless" sector of the
workforce with severely limited opportunities to influence
the decisions that affect their warklives or to control
the day-to-day actions that constitute their jobs.
Research Implications
Organizational Culture
Smircich and Calas Cin press! lament the adoption of
the culture metaphor by "the positivist, technical
interestCsl" in organizational studies and say that
organizational culture literature may now be dead. They
then ask, "How can we revive/revise/revitalize ’culture’
for organizational studies." Without concern for the
accuracy of their assessment, Smircich and Calas’ question
reflects longstanding deficiencies in the culture concept
that might indeed be fatal to the metaphor’s vitality.
One major deficiency has been a lack of foundational
social theory, a deficiency even in the most comprehensive
treatments of organizational culture, such as Schein’s
3 0 4
work and that of Smircich herself. Certainly, accounts
often are given for the social contexts and psychological
mechanisms contributing to an organizational culture
perspective Ce.g., Van riaanen & Barley, 19B5), but they
invariably are an ad hoc pastiche of borrowed concepts.
This study demonstrates, however, that organizational
culture will benefit by adoption of structuration as a
foundational social theory. Virtually all perspectives on
culture involve subjective beliefs and meanings of
organizational members; more generally, culture research
asks what is special about an organization’s way of life
and how did it come to be as it is. Structuration
accounts for the structural properties of any social
collectivity by reference to the knowledgeable action of
members; and, through the concept of the duality of
structure, it accounts for individual action by reference
to structural properties. Because action and structure
are two faces of a single ongoing process, the
individual/organizational, micro/macro dichotomies that
plague culture researchers become nonissues.
As shown in this study, the symbolic aspects of the
production and reproduction of an organization,
particularly its discourse, are central to an
understanding of the key relations of power and
domination. While what is said and how it is said often
3 0 5
is a concern of culture researchers, they have not
established a clear and systematic link between discourse
and domination.
Another deficiency in the organizational culture
tradition is its failure to see everyday understandings of
members as ideological. This study, however, concurs
with Pecheux’s assertion that all discourse is
ideological. Moreover, by directing my approach to the
study of discourse and power in organizational
relationships with a concern for the career trajectories
of ethnolinguistic minorities, the relation of culture to
workers’ emancipation becomes more difficult to ignore or
sweep under the rug of "scientific objectivity." In this
connection, the focus on the routine discourse practices,
rather than on the noticeably odd, problematic, or vivid
discourse features, departs from common organizational
culture approaches. It is in the chronically reproduced
features of interaction, however, that the fundamental
nature, the naturalized premises, become inscribed in
members’ practical consciousness and become evident to the
analyst.
Power in Organizations
The central contribution of this study to research in
organizational power is its utility as an example. Far
too many discussions of power eventually arrive at the
conclusion that power involves communication in one way or
3 0 6
another, hut then examine structure, decision-making,
resources allocation, or other traditional variables
without ever examining discourse per se. Here the value
of looking directly at the talk between members of
powerful, privileged groups and those of subordinated
groups is demonstrated. The noncoercive, indirect means
of maintaining the structures of domination and
asymmetrical control over authoritative and allocative
resources are evident in a close examination of routine
discourse practices.
The implication for power theorists is the apparent
need to investigate behavior that has been neglected in
the quest to identify the desiderata of power. Typically,
the indicators of power are sought in such obvious
phenomena as reallocations of material resources,
determination of who wins and who loses policy battles,
examination of members’ physical environments, assessments
of reputations for having power, and so forth. The
neglected indicator is members’ everyday talk. What needs
to be examined is not so much the arguments and
consequences of that talk, but how linguistic choices are
made and how the discourse is constructed by all
participants.
Critical Theoru
The analyses of discourse in the hotels’ rites and of
the hotels and corporation demonstrate how critique can be
3 0 7
negative and oriented toward discovered contradictions
while simultaneously being constructive and emancipatory.
In Fact, it is difficult to find other critical analyses
that are genuinely emancipatory because negative critique
paradigmatically ends at the step of sensitizing members
to hidden contradictions.
Exposing false consciousness, however, emancipates no
one when the members who are dominated have no cultural or
personal resources for overcoming conditions that create
and perpetuate that false consciousness. Moreover,
without a constructive step, critique presents to the
dominant sectional interests only resistance and a
reasoned denial of their legitimacy. Critique seems to
portray the social world as a set of human oppositions— a
group of good, but exploited, people against a group of
exploitative, bad people. In many systems of exploitative
domination, however, members of the dominant group are as
unconsciously immersed in the system of exploitation as
members who are subordinated. Both groups inherit social
positions, sets of expectations, and discursive practices
that normalize their relations of domination and power.
Interrupting exploitative domination to initiate
emancipation can be accomplished with persuasive
suggestions far alternative practices that are more
egalitarian, humane, and democratic. Part of the work of
critical theory, then, is providing a vision of an
308
improved world, expressed in practical, concrete, and
persuasive enough terms to effect positive change.
Discourse Analusis
Detailed analyses of first person plural pronouns are
rare; only Fowler and Kress’ C1979) brief inventory of
"we" in their discussion of the "grammar of modality" Cp.
200) attempts to link FPPPs with power and domination in
institutional relations. Theirs and other previous work,
however, is Fragmented and is not supported by analysis of
examples taken comprehensively from actual organizational
discourse. The study reported here not only develops a
complete taxonomy of the referential domains of "we" that
is well grounded in the literature on reference generally,
but it also develops conclusions from an examination of
discourse that is obtained and analyzed in a systematic,
replicable manner.
By bracketing the discourse to be studied in terms of
organizational rites, the discourse analysis benefits in
two ways. First, inferences about what the participants
are supposed to be doing with their discourse are easier
to make in comparison with any less disciplined way of
selecting stretches of discourse to analyze. Second,
"doing pragmatics" necessarily involves knowing the socio
cultural context of discourse production; the socio
cultural setting in which the discourse takes place is
easier to assess when the bracketed discourse is a rite.
3 0 9
Ethnolinguistic Identitu
In a small and indirect uiay, this study suggests to
the group identity research tradition that power and
domination relationships at work need to be concidered as
well. The recent work an subjective ethnolinguistic
vitality theory Ce.g., Giles & Johnson, 1901) holds that
multiple group membership influences the strength and
stability of ethnolinguistic group identity. Among the
kinds of groups that must be examined are those that
result from segregating workplace populations in virtue of
members’ power and privileges. The segregations evident
in this examination of the discourse of hotel rites are
into basically two groups— a powerful, paternalistic,
dominant group of managers and a passive, nonprofessional, {
dominated group of rank-and-file employees. It is highly i
probable that ethnolinguistic group identity is maintained
in a social milieu which is influenced by individuals’
relations to the dominant ideology of their workplace.
Limitations
This dissertation is an exploratory effort to bridge
the ideas and research findings of several disciplines to
advance understanding of how discourse and employees’
worklife experiences are related. From the perspective of
any one of those disciplines, the methods, data, or
findings presented here will be found wanting. The
highlighting of specific limitations, however, is also
3 1 0
simultaneously a warrant and outline for future
development of this project.
The settings of the study are both a major strength
and its principal limitation. Although hotels as
workplaces have a number of unusual features, they share a
wide range of characteristics with other places of
employment. Thus my analysis of discourse and power in
these two hotels presents observations and conclusions
that are tempting to generalize to work organizations of
other kinds. Although several of the ideas I propose and
explanations I offer appear to be equally valid in other
contexts— I have in mind the taxonomy of FPPPs in
discourse, the use of names and titles as guides to group
boundaries, and the concept of converting practical
consciousness to discursive consciousness to begin
emancipation— nevertheless, much of this study’s findings
must be viewed as applying uniquely to the two hotels. As
such, many claims here must be only suggestions of ways to
explain the dynamics of organizations in general. Much
more research needs to be accomplished before the
relationship involving discourse practices, organizational
power, and personal career trajectories is fully
understood.
Similarly, the rites chosen for analysis might not be
fully typical of the discourse that routinely occurs
between managers and nanmanagerial employees in other
311
settings within the hotels. Consequently, the conclusions
concerning reference in the discourse and exploitative
domination in the rites need to be confirmed by examining
data from other types of rites in other settings.
A second major limitation is that the analyses— both
the discourse analysis and the critiques— are incomplete.
To complete the picture of even one hotel, more work is
needed. Discourse in other rites must be examined; rites
involving only managerial workers need be analyzed to
identify intragroup differences in discourse practices and
to discover if the techniques of reference discussed as
group-defining actually occur as such only in manager-
nonmanager interaction. Similarly, examining rites in
which only nonmanagerial employees participate would help
verify the observations from these three rites concerning
group boundaries.
The analysis of discourse features should be expanded
to identify other linguistic forms reflecting the
inscription of power, domination, and ideology. In this
connection, it seems that enough social changes have
occurred in the English speaking world since the early
1960s to re-examine Brown and Gilman’s work on second
person pronouns, particularly in discourse involving a
multiethnic workforce. A more detailed analysis of
conventions governing the use of names and titles might
bring further insight to the analysis of "we."
3 1 2
Richard K. Brown C1982) argues that individual work
histories are essential to any discussion of career
trajectories of occupational or other groups. Any
statements in this study about probable career impacts of
workers’ language skills and experiences must therefore be
considered tentative ideas until work histories are
collected and analyzed.
Epilogue
This dissertation is not mainly about hotels or about
personal pronouns or even about rites in work
organizations. fly central concern has been to better
understand possible relationships among power, discourse,
and workers’ careers. I am especially concerned to
uncover the mare subtle ways in which discourse
contributes to the structures of exploitative
relationships in work organizations. Clearly, much more
needs to be done. riichael Shapiro C19B4D reminds us that
we must focus not on a speaker’s intentions but on "the
inheritance of practices and conceptions that precede what
comes out of the speaker’s mouth." Often the only way
f
1
analysts have of knowing those practices and conceptions
is to study carefully what comes out of speakers’ mouths.
This study has demonstrated that discourse itself is
inherited practice. Moreover, discourse is both the
primary reflection of power relationships and their
313
essential mode of production and reproduction. Foucault
C1984, p. 110) describes its relevance:
CDHiscourse is not simply that which translates
struggles or systems of domination, but is the thing
for which and by which there is struggle, discourse
is the power which is to be seized.
References
314
Anderson, W. T. C1986). The apostolic Function of the
dentist. In S. Fisher & A. D. Todd CEds.). Discourse
and institutional authoritu: Medicine, education, and
lauj Cpp. 78-90). Norwood, NJ: Ablex.
Auer, J. C. P. C1984). Referential problems in
conversation. Journal of Pragmatics. 8, 627-640.
Austin, J. L. C1962). How to do things with words.
Cambridge: Harvard University Press.
Babbie, E. Cl986). The practice of social research
CFourth Ed.). Belmont, CA: Wadsworth.
Bach, K. & Harnish, R. M. C1979). Linguistic
communication and speech acts. Cambridge: NIT Press.
Bantz, C. R. C1983). Naturalistic research traditions.
In L. L. Putnam & fl. E. Pacanowsky CEds.). _
Communication and organizations: An interpretive
approach Cpp. 31-54). Beverly Hills-. Sage.
Barthes, R. Cl968). Writing degree zero CA. Lavers & C.
Smith, Trans.). New York: Hill and Wang.
Barthes, R. C1974) . S/Z CR. Miller, Trans.). New York:
Hill and Wang.
Benson, J. K. C1983). A dialectical method for the study
of organizations. In S. Morgan CEd.). Beuond method
Cpp. 331-346). Beverly Hills: Sage.
Berger, P. L. & Luckmann, T. C1967). The social
construction of realitu. New York: Anchor Books.
Blam, J.-P. & Gumperz, J. J. C1971). Social meaning in
linguistic structure-. Code-switching in Norway. In J.
J. Gumperz CEd.). Language in social groups Cpp. 274-
310). Stanford: Stanford University Press.
Bourhis, R. Y. C1979). Language in ethnic interaction-. A
social psychological approach. In H. Giles & B. St.
Jacques CEds.). Language and ethnic relations Cpp.
117-142). Oxford: Pergamon Press.
Brown, G. & Yule, G. C1983). Discourse analusis.
Cambridge: Cambridge University Press.
315
Brawn, P. & Levinson, S. C1979). Social structure,
groups, and interaction. In K. R. Scherer & H. Giles
CEds.). Social markers in speech Cpp. 291-34:2).
Cambridge:Cambridge University Press.
Brown, R. H. C197S). Bureaucracy as praxis: Toward a
political phenomenology of Formal organizations.
Administrative Sciences Quarterlu. 23, 365-381.
Brown, R. K. C19S2). Work histories, career strategies
and the class structure. In A. Giddens & G. Mackenzie
CEds.). Social class and the division of labour Cpp. 119-
136). Cambridge: Cambridge University Press.
Brown, R. & Ford M. C1961). Address in American English.
Journal oF Abnormal and Social Fsucholoau. 62. 375-385.
Brawn, R. & Gilman C1972). The pronouns oF power and
solidarity. In P. Giglioli CEd.). Language and social
context Cpp. 252-282). New York: Penguin Books.
Bourdieu, P. C1977). Outline oF a theoru oF practice CR.
Nice, Trans.). Cambridge: Cambridge University Press.
Burke, K. C1969). A rhetoric oF motives. Berkeley:
University oF CaliFornia Press.
Capra, F. C1977). The tao oF phusics. New York: Bantam
Books.
Carbaugh, D. C1985). Cultural communication and
organizing. In U). B. Gudykunst, L. P. Stewart & S. Ting-
Toamey CEds.). Communication, culture, and
organizational processes Cpp. 30-47). Beverly Hills:
Sage.
Carranza, M. A. & Ryan, E. B. C1975). Evaluative
reactions oF bilingual Anglo and Mexican American
adolescents toward speakers oF English and Spanish. _
International Journal oF the Sociologu oF Language, 6,
83-104.
ChaFe, U. C1976). Givenness, contrastiveness,
deFiniteness, subjects, topics, and paint oF view. In
C. N. Li CEd.). Subject and topic. New York: Academic
Press.
Chastain, C. C1975). ReFerence and context. In K.
Gunderson CEd.). Language, mind, and knowledge CVol.
VIII, pp. 194-269). Minneapolis: University oF
Minnesota Press.
3 16
Chisholm, R. II. C19B6). Selecting metaphoric terminology
for the computer industry. Journal oF Technical Writing
and Communication. 16. 3, 195.
Chomsky, N. C1957). Suntactic structures. The Hague:
Mouton.
Chomsky, N. C1965). Aspects of the theoru of syntax.
Cambridge: MIT Press.
Clegg, S. Cl975). Pouier. rule and domination: A critical
and empirical understanding of pouter in sociological
and organizational life. London: Routledge & Kegan
Paul .
Clegg, S. C1977). Power, organization theory, Marx and
critique. In S. Clegg & D. Dunkerley CEds.). Critical
issues in organizations Cpp. 21-40). Lcndon: Routledge
& Kegan Paul.
Cole, P. C1978). On the origins of referential opacity.
In P. Cole CEd.). Suntax and semantics 9: Pragmatics
Cpp. 1-22). New York: Academic Press.
Conrad, C. C1983). Organizational power: Faces and
symbolic Forms. In L. L. Putnam & M. E. Pacanowsky
CEds.). Communication and organizations: An
interpretive approach Cpp. 173-194). Beverly Hills:
Sage.
Conrad, C. & Ryan, !i. C1985). Power, praxis, and person
in social and organizational theory. In P. Tompkins &
R. McPhee CEds.). Organizational communication research
and theoru Cpp. 235-257). Beverly Hills: Sage.
Coulthard, M. C1977). An introduction to discourse
analusis. London: Longman.
Dandridge, T. C. C1985). The life stages of a symbol:
When symbols work and when they can’t. In P. Frost, L.
Moore, M. R. Louis, C. Lundberg & J. Martin CEds.).
Organizational culture Cpp. 141-154). Beverly Hills:
Sage.
D ’Anglejan, A. & Tucker, S. R. C1973). Sociolinguistic
correlates of speech style in Quebec. In R. W. Fasold
CEd.). Variation in the Form and use of language Cpp.
327-353). Washington, D.C.: Georgetown University
Press.
317
de la Zerda, N. & Hopper, R. C1979). Employment
interviewers’ reactions to Mexican American speech.
Communication Monographs. 46, 126-134.
De Vos, 6. (1975). Ethnic pluralism: Conflict and
accommodation. In G. De Vos and L. Romanucci-Ross
CEds.). Ethnic identitu: Cultural continuities and
change. Palo Alto, CA: Mayfield Publishing Co.
Deal, T. E. & Kennedy, A. A. (1982). Corporate cultures:
The rites and rituals of corporate life. Reading, MA:
Addisan-Wesley.
Deetz, S. (1985). Critical-cultural research: New
sensibilities and old realities. In P. Frost (Ed.).
Organizational Symbolism (Special issuel. Journal of
Management. 11. 2, 121-136.
Deetz, S. & Kersten, A. C19B3). Critical models of
interpretive research. In L. L. Putnam & M. E.
Pacanowsky (Eds.). Communication and organizations: An
interpretive approach (pp. 147-171). Beverly Hills:
Sage.
Deschamps, J.-C. C19B1). Social identity and relations of
power between groups. In H. Tajfel CEd.). Social
identitu and interaroup relations Cpp. 85-98).
Cambridge: Cambridge University Press.
DiPietro, R. J. (1980). Linguistics and the professions.
Norwood, NJ: Ablex.
Dolson, D. P. (19B5). The effects of Spanish home
language on the scholastic performance of Hispanic
pupils. Journal of Multilingual and Multicultural
Development. 6, 2, 135-155.
Donnellan, K. S. (1966). Reference and definite
descriptions. Philosophical Review. 60, 281-304.
Donnellan, K. S. (1978). Speaker reference, descriptions
and anaphora. In P. Cole (Ed.). Suntax and semantics
9: Pragmatics (pp. 47-68). New York: Academic Press.
Eca, U. (1976). A theoru of semiotics. Bloomington:
Indiana University Press.
Edmonson, UJ. (1981). Spoken discourse: A model for
analusis. New York: Longman.
Edwards, J. C1985). Language, societu and identitu.
Ox ford.:._Bas i 1_B 1 a c k wel 1_.____________________________________
31B
Edwards, J. & Biles, H. (1984:). Applications of the
social psychology of language-. Sociolinguistics and
education. In P. Trudgill (Ed.). Applied _
sociolinguistics. London:Academic Press.
Eisenberg, E. M. (19B4). Amgiguity as strategy in
organizational communication. Communication Monographs.
51, 227-242.
Eisenberg, E. M. (19B6). Successful socialization:
Managing multiple memberships. Paper presented to the
annual meeting of the International Communication
Association, Chicago, IL.
Eisenberg, E. M. & Riley, P. Cin press). Organizational
symbols and sensemaking. In G. Goldhaber (Ed.).
Handbook of organizational communication. Norwood, NJ:
Ablex.
Evered, R. C19B3). The language of organizations: The
case of the Navy. In L. R. Pondy, P. J. Frost, G.
Morgan & T. C. Dandridge CEds.). Organizational__
symbolism. Greenwich, CT: JAI Press.
Farrell, T. B. & Aune, J. A. C1979). Critical theory and
communication: a selective literature review. The
Quarterly Journal of Speech. 65. 93-107.
Fasold, R. Ui. (Ed.). (1973). Variation in the form and
use of language. Washington D.C.: Georgetown University
Press.
Fisher, S. (19B4). Institutional authority and the
structure of discourse. Discourse Processes. 7, 201-
254 .
Fisher, S. & Todd, A. D. (1983). The social organization
of doctor-patient communication . Washington, D.C . - .
Center for Applied Linguistics.
Fisher, S. & Todd, A. D. (Eds.). (1986). Discourse and
institutional authoritu: Medicine, education, and law.
Norwood, NJ: Ablex.
Fishman, J. A. (1986a). Bilingualism and biculturism as
individual and as societal phenomena. In J. A. Fishman
(Ed.). The rise and fall of the ethnic revival:
Perspectives on language and ethnicitu (pp. 39-56). New
York: Moutan.
319
Fishman, J. A. C1906b). Mother-tangue claiming in the
United States since I960: Trends and correlates. In J.
Fishman CEd.). The rise and fall oF the ethnic revival:
Perspectives on language and ethnicitu Cpp. 107-194).
New York: Mouton
Fishman, J., Gertner, fl. H., Lowy, E. G. & Milan, W. G.
C19B6). Ethnicity in action: The community resources of
ethnic languages in the United States. In J. Fishman
C Ed. ). The rise and Fall of the ethnic revival:
Perspectives on language and ethnicitu Cpp. 195-282).
New York: Mouton.
Forester, J. C1982). Planning in the face of power.
Journal of the American Planning Association. 48, 67-80.
Forester, J. C1983). Critical theory and organizational
analysis. In G. Morgan CEd.). Beuond method Cpp. 234-
246). Beverly Hills: Sage.
Forester, T. C1981). The Jelly bean people of Silicon
Valley. In T. Forester CEd.). The microelectronics
revolution Cpp, 65-71). Cambridge: MIT Press.
Foucault, M. Cl972). The archaeoloou of knowledge & the
discourse on language CA. M. Sheridan Smith, Trans.).
New York: Pantheon.
Foucault, M. C19B0). Power/knowledge: Selected interviews
& other writings 1972-1977 CC. Gordon, L. Marshall, J.
Mepham & J. Soper, Trans.). New York: Pantheon.
Foucault, M. C19B4). The order of discourse. In M.
Shapiro CEd.). Language and politics Cpp. 108-138).
New York: New York University Press.
Fowler, R., Hodge, B., Kress, G. & Trew, D. C1979).
Language and control. London: Routledge & Kegan Paul.
Fraser, B. C1975). Warning and threatening. Centrum. 3,
169-180.
Frege, G. C1952). Philosophical writings. Oxford: Basil
Blackwell.
Frost, P. J., Moore, L. F., Louis, M. R., Lundberg, C. C.
& Martin, J. CEds.). C19B5). Organizational culture.
Beverly Hills: Sage.
320
Garfinkel, H. & Sacks, H. C1970). On the formal
structures of practical actions. In J. C. McKinney & E.
A. Tiryakian CEds.). Theoretical sociologu:
Perspectives and developments Cpp. 337-366). New York:
Appleton-Century-Crofts.
Geach, P. C1962). Reference and generality. Ithaca, NY:
Cornell University Press.
Geertz, C. C1973). The interpretation of cultures. New
York: Basic Books.
Gerbner, G. C1983). The importance of being critical— in
one’s own fashion. Journal of Communication. 33, 3, 355-
362 .
i
Geuss, R. C19B1). The idea of a critical theory: Habermas j
& the Frankfurt school. London: Cambridge University
Press.
Giddens, A. Cl979). Central problems in social theoru: I
Action, structure and contradiction in social analysis. J
Berkeley: University of California Press. j
]
Giddens, A. C1981). A contemporaru critique of historical
materialism. Berkeley: University of California Press.
Giddens, A. Cl981). The constitution of society. Outline
of the theoru of structuration. Berkeley: University of !
California Press. I
Giddens, A. C1986). Action, subjectivity, and the j
constitution of meaning. Social Research. 53, 529-515. j
I
Giles, H. C1977). The social context of speech: A social j
psychological perspective. ITL: A Review of Applied j
Linguistics. 35, 27-12. |
I
Giles, H. C1979). Ethnicity markers in speech. In K. R. j
Scherer & H. Giles CEds.). Social markers in speech
Cpp. 252-290). Cambridge: Cambridge University Press. J
Giles, H., Bourhis, R. Y. & Taylor, D. M. C1977). Towards !
a theory of language in ethnic group relations. In H. |
Giles CEd.). Language, ethnicitu and intergroup ;
relations Cpp.307-318). London: Academic Press.
Giles, H. & Johnson, P. C1981). The role of language in
ethnic group relations. In J. Turner & H. Giles
CEds.). Intergroup behaviour Cpp. 199-213). London:
Basil Blackwell.
321
Giles, H. & Powesland, P. C1975). Speech stule and social j
evaluation. London: Academic Press.
Giles, H., Robinson, W. P. & Smith, P. C19B0). Language:
Social psucholoaical perspectives. Oxford: Pergamon
Press.
Giles, H. & St. Clair, R. N. C1979). Language and social
psuehologu. Oxford: Basil Blackwell.
Giles, H. & Smith, P. M. C1979) . Accommodation theory.
Optimal levels of convergence. In H. Giles & R. St.
Clair CEds.). Language and social psuehologu Cpp. 45-
55). Oxford: Basil Blackwell.
Cambridge:
Frame analysis. New York: Harper and
Glazer, N. C1983). Ethnic dilemmas 1964-1992.
Harvard University Press.
Goffman, E. C1974)
Row.
Goldwyn, E. C1981). Now the chips are down. In T.
Forester CEd.). The microelectronics revolution Cpp.
297-302). Cambridge: NIT Press.
Good, C. & Butterworth, B. C190O). Hesitancy as a
conversational resource: Some methodological
implications. In H. UJ. Dechert & N. Raupach CEds.). _
Temporal variables in speech. The Hague: Nouton.
Gray, B. Bougon, N. G. & Donnellan, A. C1985).
Organizations as constructions and destructions of
meaning. In P. Frost CEd.). Organizational symbolism
CSpecial issue!. Journal of Management. 11. 2, 83-98.
Gregory, K. L. C1983). Native-view paradigms: Multiple
cultures and culture conflicts in organizations.
Administrative Sciences Quarterly. 28, 359-376.
Gregory, M. & Carroll, S. C197B). Language and situation:
Language varieties and their social contexts. London:
Routledge & Kegan Paul.
Grice, H. P. C1975). Logic and conversation. In P. Cole
& J. L. Morgan CEds.). Suntax and semantics 3: Speech
acts Cpp. 107-142). New York: Academic Press.
Grice, P. C19B1). Presupposition and conversational
implicature. In P. Cole CEd.). Radical pragmatics Cpp.
183-198). New York: Academic Press.
3££
Gudykunst, W. B., Stewart, L. P. & Ting-Toomey, S. CEds.).
C1985). Communication. culture, and organizational
identitu■ Beverly Hills: Sage.
Gumperz, J. J. C1S77). Sociocultural knowledge in
conversational inference. In N. Saville-Troike CEd.).
The S8th annual round table monograph series on language
and linguistics Cpp. 191-Ell). Washington, D.C.:
Georgetown University Press.
Gumperz, J. J. CEd.). C198£a). Language and social
identitu. Cambridge: Cambridge University Press.
Gumperz, J. J. C198£b). Discourse strategies. Cambridge:
Cambridge University Press.
Gumperz, J. J. & Hernandez-Chavez, E. C1971).
Bilingualism, bidialectalism and class-room
interaction. In C. Cazden, V. John & D. Hymes CEds.).
The Functions of language in the class room. New York:
Teachers College Press.
Gumperz, J. J. & Wilson, R. C1971). Convergence and
creolization: A case From the Indo-Aryan/Dravidian
border.
In D. Hymes CEd.). Pidginization and creolization oF
languages Cpp. 151-167). Cambridge: Cambridge
University Press.
Haas, J. & Shaffir, W. C198E). Ritual evaluation of
competence: The hidden curriculum of professionalization
in an innovative medical school program. Work and
Occupations. 9, E, 131-154.
Habermas, J. Cl979). Communication and the evolution of
society.. Boston: Beacon Press.
Harragan, G. H. C1977). Games mother never taught uou.
New York: Warner Books.
Hawkins, J. C1980). On surface definite articles in
English. In J. Van der Auwera CEd.). The semantics of
determiners Cpp. 41-66). Baltimore: University Park
Press.
Heath, S. B. C19B3). Waus with words: Language, life, and
work in communities and classrooms. Cambridge:
Cambridge University Press.
Heidegger, N. C1971). On the wau to language CP. D.
Hertz, Trans.). San Francisco: Harper & Row.____________
323
Heller, M. C1982), Negotiations of language choice in
Montreal. In J. J. Gumperz CEd.). Language and social
identitu Cpp. 108-11B). Cambridge: Cambridge University
Press.
Heydebrand, U). V. C19B3). Organization and praxis. In G.
Morgan CEd.). Beuond method Cpp. 306-320. Beverly
Hills: Sage.
Hidalgo, M. C1986). Language contact, language loyalty,
and language prejudice on the Mexican border. Language
in Societu. i5. 2, 193-220.
Hodge, B., Kress, G. & Jones, G. C1979). The ideology of
middle management. In R. Fouler, B. Hodge, G. Kress &
T. Treu CEds.). Language and control Cpp. 81-93).
London: Routledge & Kegan Paul.
Hopper, R. & Williams, F. C1973). Speech characteristics
and employability. Speech Monographs. 4:0. 296-302.
Horkheimer, M. C1972). Critical theoru: Selected essaus
CM. J. O’Connell, Trans.). Neu York: Herder & Herder.
Houze, W. C. C1985). Career veer: Hou to position
uourself for a prosperous future. Neu York: McGrau-
Hill.
I
Hudson, R. A. C19B0). Sociolinguistics. Cambridge: i
Cambridge University Press. j
Husband, C. & Khan, V. S. C19B2). The viability of
ethnolinguistic vitality: Some doubts. Journal of
Multilingual and Multicultural Development. 3, 193-205.
Huspek, M. C1986). Linguistic variation, context, and
meaning: A case of -ing/in’ variation in North American
workers’ speech. Language in Societu. 15, 149-164.
Hymes, D. C1974). Foundations in sociolinguistics: An
ethnographic approach. Philadelphia: University of
Pennsylvania Press.
Jay, M. C1986). Permanent exiles: Essaus on the
intellectual migration from Germany to America. Neu
York: Columbia University Press.
Johnson, R. T. & Ouichi, W. G. C1974). Made in America
Cunder Japanese management). Harvard Business Revieu. _
52, 61-69.
324
Jupp, T. C., Roberts, C. & Cook-Gumperz, J. C1982).
Language and disadvantage: The hidden process. In J. J.
Gumperz CEd.). Language and social identitu Cpp. 232-
25G). Cambridge: Cambridge University Press.
Kalin, R. C1982). The social significance of speech in
medical, legal and occupational settings. In E. B. Ryan
& H. Giles CEds.). Attitudes touiards language variation
pp. 148-163). London: Edward Arnold.
Kalin, R. & Rayko, D. C19B0). The social significance of
speech in the job interview. In R. N. St. Clair & H.
Giles CEds.). The social and psucholooical contexts of
language Cpp. 39-50). Hillsdale, NJ: Lawrence Erlbaum.
Kerr, E. B. & Hiltz, R. S. C19B3). Computer-mediated
communication sustems: Status and evaluation. New York:
Academic Press.
Kersten, A. C1986). Philosophical foundations for the
construction of critical knowledge. In M. L. McLaughlin
CEd.). Communication uearbook 9. Beverly Hills: Sage.
Killman, R. H. C19B2). Getting control of the corporate
culture. Managing. 3, 11-17.
Killman, R. H. C1985). Five steps for closing culture
gaps. In R. H. Killman, M. J. Saxton & R. Serpa CEds.).
Gaining control of the corporate culture Cpp. 351-369).
San Francisco: Jossey-Bass Publishers.
Kramarae, C., Schulz, M. & O ’Barr, UJ. M. C1984). Language
and power. Beverly Hills: Sage.
Krefting, L. A. & Frost, P. J. C19B5). Untangling webs,
surfing waves, and wildcatting: A multiple-metaphar
perspective on managing organizational culture. In P.
Frost, L. Moore, M. R. Louis, C. Lundberg & J. Martin
CEds.). Organizational culture Cpp. 155-168). Beverly
Hills: Sage.
Kress, G. & Hbdge, R. C1979). Language as ideology♦
London: Routledge & Kegan Paul.
I Kripke, S. C1972). Naming and necessity. In D. Davidson
& G. Harman CEds.). Semantics of natural language Cpp.
253-355). Dordrecht: D. Reidel Publishing Company.
Labov, UJ. Cl966). The social stratification of English in
New York Citu. Washington, D.C.: Center for Applied
Linguistics.
325
Labov, UJ. C1972). Sociplinauistic patterns. Phila
delphia: University of Pennsylvania Press.
Lakoff, 6. C1976}. Pronouns and reference. In J. 0.
flcCauiley CEd.). Suntax and semantics 7: Notes from the
linguistic underground Cpp. 276-335). New York:
Academic Press.
Lambert, UJ. E. C1967) . A social psychology of
bilingualism. Journal of Social Issues. 23, 91-109.
Lambert, W. E., Frankel, H. & Tucker, G. R. C1966).
Judging personality through speech: A French-Canadian
example. Journal of Communication. 16. 305-321.
Lambert, UJ. E. & Tucker, 6. R. C1976). Tu. vous. usted.
Rowley, tIA: Newbury House.
Lamy, P. C1979). Language and ethnolinguistic identity:
The bilingualism question. International Journal of
the Sociologu of Language. 20, 23-36.
Lenski, G. E. C1984). Power and privilege: A theoru of
social stratification C2nd Ed.). Chapel Hill: The
University of North Carolina Press.
LePage, R. B. & Tabouret-Keller, A. C1982). Models and
stereotypes of ethnicity and of language. Journal of
Multilingual and Multicultural Development. 3, 161-192.
Levinson, S. C1983). Pragmatics. Cambridge: Cambridge
University Press.
Lind, E. A. a O’Barr, UJ. M. C1979). The social
significance of speech in the courtroom. In H. Giles a
R. N. St. Clair CEds.). Language and social psuehologu
Cpp. 66-87). Oxford: Basil Blackwell.
Linde, C. C1979). Focus of attention and the choice of
pronouns in discourse. In T. Givon CEd.). Suntax and
semantics 12: Discourse and suntax Cpp. 337-354). New
York: Academic Press.
Linsky, L. C1967). Referring. London: Routledge a Kegan
Paul.
Louis, M. R. C1985). An investigator’s guide to workplace
culture. In P. Frost, L. Moore, M. R. Louis, C.
Lundberg a J. Martin CEds.). Organizational culture
Cpp. 73-93). Beverly Hills; Sage.
3 26
Lyons, J. C1977). Semantics. Vols. 1 & 2. Cambridge:
Cambridge University Press.
Mathiassen, L. & Andersen, P. B. C1986). The impact of
computer-based systems upon the professional language of
nurses. Journal of Pragmatics, 10, 1, 1-26.
McLaughlin, M. C1985). Conversation: Houi talk is
organized. Beverly Hills: Sage.
Mehan, H. C1986a). The role of language and the language
of role in institutional decision making. In S. Fisher
& A. Todd CEds.). Discourse and institutional
authoritu: Medicine, education, and lam Cpp. 140-163).
Norwood, NJ:Ablex.
i
Mehan, H. C1986b). Language and power in organizational J
process. Paper presented to the Talk and Social j
Structure Conference, Santa Barbara, CA. i
I
Mey, J. C1985). Whose language? A studu in linguistic !
pragmatics. Philadelphia: John Benjamins. [
Miles, M. B. & Huberman, A. M. C1984). Qualitative Data
Analusis. Beverly Hills: Sage.
Mitroff, I. I. & Kilmann, R. H. C1985). Corporate taboos
as the key to unlocking culture. In R. H. Kilmann
CEd.). Gaining control of the corporate culture Cpp. 184-
199). San Francisco: Jossey-Bass Publishers.
Morgan, G. C1983a). Research strategies: Modes of
engagement. In G. Morgan CEd.). Beuond method:
Strategies for social research Cpp. 19-43). Beverly
Hills: Sage.
Morgan, G. C1983b). Toward a more reflective social
science. In G. Morgan CEd.). Beuond method: Strategies
for social research Cpp. 368-376). Beverly Hills: Sage.
Morgan, G. C1983c). The significance of assumptions. In
G. Morgan CEd.). Beuond method: Strategies for social
research Cpp. 377-382). Beverly Hills: Sage.
Morgan, G. C19B5). Spinning on symbolism: Some
developmental issues in organizational symbolism. In P.
Frost CEd.). Organizational symbolism CSpecial issue!.
Journal of Management. 11. 2, 29-30.
327
Munchus, G. C1983). Employer-employee based quality
circles in Japan: Human resource policy implications for
American Firms. Academu of Management Review. 8, 255-
261 .
□’Barr, UJ. M. C19B4) . Asking the right questions about
language and power. In C. Kramarae, M. Schulz & UJ. M.
□ ’Barr CEds.). Language and power Cpp. 260-280).
Beverly Hills: Sage.
!
□chs, E. C1S79). Planned and unplanned discourse. In T. 1
Givon CEd.). Suntax and semantics 12: Discourse and
suntax Cpp. 51-80). New York: Academic Press.
□chs, E. C1984). A sliding sense of obligatoriness: The
poly-structure of Malagasy oratory. In J. Baugh & J.
Scherzer CEds.). Language in use: Readings in
sociolinguistics Cpp. 167-182). Englewood Cliffs, NJ:
Prentice-Hall.
Drtner, S. C1984). Theory in anthropology since the
Sixties. Journal of the Societu for Comparative Studu of
Societu and Historu. 126-166.
I
□uichi, UJ. G. C1981). Theoru Z. Reading, MA-. Addison-
UJesley .
Pascale, R. T. & Athos, A. G. C1981). The art of Japanese j
management. New York: Simon and Schuster. '
Paulson, C. B. C1984). Pronouns of address in Swedish:
Social class semantics and a changing system. In J.
Baugh & J. Scherzer CEds.). Language in use: Readings
in sociolinguistics Cpp. 268-291). Englewood Cliffs,
NJ: Prentice-Hall.
Peter, L. J. C1975). The Peter plan. New York: Bantam
Books.
Peters, T. J. & UJaterman, R. H. C1982). In search of
excellence. New York: Harper & Row.
Phillips, K. C1986, May 11). New Americans for the next
America. Los Angeles Times. Section V, pp. 1-2.
Pool, J. C1979). Language planning and identity planning.
International Journal of the Sociologu of Language, 20,
5-21.
328
Poole, M. S. & McPhee, R. D. C1983). A structuratianal
analysis of organizational climate. In L. L. Putnam &
M. E. Pacanousky CEds.). Communication and
organizations: An interpretive approach Cpp. 195-220).
Beverly Hills:Sage.
Pryor, R. C1981). On the method of critical theory and its
implications For a critical theory of communication. In
j S. Deetz CEd.). Phenomenologu in rhetoric and
communication Cpp. 25-36). Washington, D.C.: Center for
Advanced Research in Phenomenology & University Press of
America.
Putnam, L. L. C1983). The interpretive perspective: An
alternative to functionalism. In L. L. Putnam & M. E.
Pacanousky CEds.). Communication and organizations: An
interpretive approach Cpp. 31-54). Beverly Hills: Sage.
Putnam, L. L. & Pacanousky, M. E. CEds.). C1983).
Communication and organizations: An interpretive
approach. Beverly Hills: Sage.
Quine, UJ. V. □. C1961). From a logical point of vieu.
Cambridge: Harvard University Press.
Raelin, J. A. C1984). An examination of deviant/adaptive
behavior in the organizational careers of professionals, j
Academu of Management Revieu. 3, 413-427. |
I
Real, M. C1984) . The debate on critical theory and the
study of communications: A commentary an Ferment in the
Field. Journal of Communication. 34. 4, 72-80.
Rey, A. C1977). Accent and employability: Language
attitudes. Language Sciences. 47, 7-12.
Riley, P. C1985). Spinning on symbolism: The spinning
metaphor and dialectical tensions. In P. Frost CEd.).
Organizational symbolism CSpecial issue!. Journal of
Management. 11, 2, 49-50.
Rogers, E. C1982). The empirical and the critical schools
of communication research. In M. Burgoon CEd.).
Communication uearbook 5 Cpp. 125-144). Neu Brunsuick,
NJ: Transaction Books.
Rosen, M. C1985). Breakfast at Spiros: Dramaturgy and
dominance. In P. Frost CEd.). Organizational
symbolism CSpecial issue!. Journal of Management. 11.
31-48.
3£9
Ross, J. A. C1979). Language and the mobilization of
ethnic identity. In H. Giles & B. Saint-Jacques
CEds.). Language and ethnic relations Cpp. 1-13).
Oxford: Pergamon Press.
Rothschild, W. E. Cl984). How to gain Cand maintain) the
competitive advantage in business. New York: McGraw-
Hill.
Ruesch, J. & Bateson, G. C1951). Communication: The
social matrix of psuchiatru. New York: W. UJ. Norton.
Russell, B. C1958). Logic and knowledge. London: Allen
and Unwin.
Sadler, P. C19B1). Welcome back to the "automation"
debate. In T. Forester CEd.). The microelectronics
revolution Cpp. E90-E97). Cambridge: MIT Press.
Sag, I. & Hankamer, J. C1973). Syntactically versus
pragmatically controlled anaphora. In R. UJ. Fasold
CEd.). Variation in the form and use of language Cpp.
189-E04). Washington D. C.: Georgetown University Press.
Sankoff, G. C1974). A quantitative approach to the study
* of communicative competence. In R. Bauman & J. Scherzer
CEds.). Explorations in the ethnooraphu of speaking.
Cambridge: Cambridge University Press.
Sankoff, G. C1980). The social life of language.
Philadelphia: University of Pennsylvania Press.
Schall, M. C1983). A communication-rules approach to
organizational culture. Administrative Sciences
Quarterly. E8, 557-581.
Schein, E. H. C1981). Does Japanese management style have
a message for American managers? Sloan Management
Review, £3, 55—68.
Schein, E. H. C1983) . The role of the founder in creating
organizational culture. Organizational Dunamics
CSummer), 13-EB.
Schein, E. H. C1984). Coming to a new awareness of
organizational culture. Sloan Management Review. £5,
3-16.
Schein, E. H. Cl985). Organizational culture and
leadership. San Francisco: Jossey-Bass.
330
Scherzer, J> C1984). Strategies in text and context: Kuna
kaa kujento. In J. Baugh & J. Scherzer CEds.). Language
in use: Readings in sociolinguistics Cpp. 183-197).
Englewood Cliffs, NJ: Prentice-Hall.
Searle, J. C19B9). Speech Acts. Cambridge: Cambridge
University Press.
Searle, J. C1975). Indirect speech acts. In CoIb , P. &
Morgan, J. L. CEds.). Suntax and semantics 3: Speech
acts Cpp. 59-8S). New York: Academic Press.
Shapiro, M. CEd.). C1984). Language and politics. New
York: New York University Press.
Sherman, H. C1976). Dialectics as a method. Insurgent
Sociologist. 6, 57-64.
Sinclair, J. M. & Coulthard, M. C1975). Towards an
analusis of discourse. London: Oxford University Press.
Sinetar, M. C19B7). Do what uou love, the moneu will
follow. New York: Paulist Press.
Slack, J. D. & Allor, M. C19B3). The political and
epistemological constituents of critical communication
research. Journal of Communication. 33, 3, E08-E18.
Smircich, L. C1983a). Concepts of culture and
organizational analysis. Administrative Sciences
Quarterlu. £8. 339-358.
Smircich, L. C1983b). Studying organizations as cultures.
In 6. Morgan CEd.). Beuond method: Strategies for
social research Cpp. 160-17S). Beverly Hills: Sage.
Smircich, L. C1985). Is the concept of culture a paradigm
for understanding organizations and ourselves? In P.
Frost, L. Moore, M. R. Louis, C. Lundberg & J. Martin
CEds.). Organizational culture Cpp. 55-7E). Beverly
Hills: Sage.
Smircich, L. & Calas, M. B. Cin press). Organizational
culture: A critical assessment. In F. Jablin, L.
Putnam, K. Roberts & L. Porter CEds.). Handbook of
organizational communication. Newbury Park: Sage.
Smith, R. C. & Eisenberg, E. M. Cin press). Understanding
conflict at Disneyland: Analyzing employee
interpretations of root-metaphors. Communication
Monographs■
331
Smythe, D. UJ. & Tran, V. D. C1983). On critical and
administrative research. Journal of Communication. 33,
3, 117-127.
Spender, D. C19B0). tlan made language. London: Routledge
& Kegan Paul.
t
Spender, D. C1984). Defining reality: A powerful tool.
In C. Kramarae, M. Schulz & U). M. 0’Barr CEds.). _
Language and power Cpp. 194-205). Beverly Hills: Sage. |
*
Stablein, R. & Nord, UJ. C1985). Practical and j
emancipatory interests in organizational symbolism: A j
review and evaluation. In P. Frost CEd.). j
Organizational symbolism CSpecial issuel. Journal of i
Management. 11. 13-28. I
Stevenson, R. L. C1983). A critical look at critical |
analysis. Journal of Communication. 33, 3, 262-269. |
I
Stonier, T. C1981). The impact of microprocessors on t
employment. In T. Forester CEd.). The microelectronics !
revolution Cpp. 303-307). Cambridge: MIT Press. ;
Strawson, P. F. C1971). Logico-linguistic papers. j
London: Methuen & Co., Ltd. ,
i
Street, R. & Giles, H. C1982). Speech accommodation
theory: A social cognitive approach to language and
speech behavior. In M. Roloff & C. Berger CEds.). _
Social cognition and communication. Beverly Hills: |
Sage. j
Stubbs, M. C1983). Discourse analusis: The j
sociolinguistic analusis of natural language. Chicago: ■
University of Chicago Press. j
[
i
Sypher, B. D., Applegate, J. L. & Sypher, H. E. C19B5).
Culture and communication in organizational contexts.
In UJ. B. Gudykunst, L. P. Stewart & S. Ting-Toomey j
CEds.). Communication, culture, and organizational i
processes Cpp. 13-29). Beverly Hills: Sage.
t
Tajfel, H. CEd.). C1978). Differentiation between social I
groups: Studies in the social psuehologu of intergroup ;
behaviour. London: Academic Press.
Tannen, D. C19B2). Oral and literate strategies in spoken
and written narratives. Language. 5B. 1, 1-21. j
332
Thompson, J. B. C19B3). Ideology and the analysis of
discourse: A critical introduction to the Michel
Pecheux. Social Research. 212-236.
Ting-Toomey, S. C1985). Toward a theory of conflict and
culture. In Ul. B. Gudykunst, L. P. Stewart & S. Ting-
Toomey CEds.). Communication, culture, and
organizational processes Cpp. 71-86). Beverly Hills:
Sage.
Treichler, P. A., Frankel, R. M., Kramarae, C., Zoppi, K.
& Beckman, H. C1984). Problems and problems: Power
relationships in a medical encounter. In C. Kramarae,
M. Schulz & U). M. O’Barr CEds.). Language and power
Cpp. 62-89). Beverly Hills: Sage.
Trice, H. M. C1984). Rites and ceremonials in
organizational culture. In S. Bacharach & S. Mitchell
CEds.). Perspectives on organizational sociology:
Theoru and research CVol. 4). Greenwich, CT: JAI Press.
Trice, H. M. & Beyer, J. M. C19B5). Studying
organizational cultures through rites and ceremonials.
Academu of Management Review. 9, 653-699.
Trudgill, P. C1974). The social differentiation of
English In Norwich. Cambridge: Cambridge University
Press.
Trudgill, P. C1983). On dialect. New York: New York
University Press.
Trujillo, N. C19B3). "Performing" Mintzberg’s roles: The
nature of managerial communication. In L. L. Putnam &
M. E. Pacanowsky CEds.). Communication and
organizations: An interpretive approach Cpp. 73-98).
Beverly Hills: Sage.
Turner, V. W. C1967). The forest of sumbols. Ithaca, NY:
Cornell University Press.
Turner, V. Ul. C1969). The ritual process: Structure and
anti-structure. Chicago: Aldine.
van Djik Cin press). Discourse and power. In M.
McLaughlin CEd.). Communication Yearbook 10. Newbury
Park: Sage.
Van Maanen, J. C1976). Breaking-in: Socialization to
work. In R. Dubin CEd.). Handbook of work.
organization, and societu Cpp. 67-130). Chicago: Rand
McNally.
333
Van Naanen, J. C1985). Spinning on symbolism:
Disquisition. In P. Frost CEd.). Organizational
symbolism CSpecial issue!] . Journal oF Management. 11.
119-120.
Van Naanen, J. & Schein, E. H. C1979). Toward a theory of
organizational socialization. Research in
Organizational
Behavior. 1., 209-264.
Watson, D. R. C1986). Doing the organization’s work: An
examination of aspects of the operation of a crisis
intervention center. In S. Fisher & A. D. Todd CEds.).
Discourse and institutional authoritu: Medicine,
education, and law Cpp. 91-121). Norwood, NJ: Ablex.
Webster’s Third New International Dictionaru Cunabridged)
C1976). Springfield, NA: 6. & C. Nerriam Co.
Weick, K. E. C1979). The social psuehologu of organizing
C2nd Ed.). Reading, NA: Addison-Wesley.
Weick, K. E. C19B3). Organizational communication: Toward
a research agenda. In L. L. Putnam & N. E. Pacanowsky
CEds.). Communication and organizations: An
interpretive
approach Cpp. 7-12). Beverly Hills: Sage.
West, C. C1984). Routine complications: Trouble with talk
between doctors and patients. Bloomington: Indiana
University Press.
Widdowson, H. G. C1973). Directions in the teaching of
discourse. In S. Corder & E. Roulet CEds.). _
Theoretical linguistic models in applied linguistics
Cpp. 65-76). Paris: Didier.
Widdowson, H. G. C1979). Rules and procedures in
discourse analysis. In T. Neyers CEd.). The
development of conversation and discourse Cpp. 61-71).
Edinburgh: Edinburgh University Press.
Williams, F. C1976). Explorations of the linguistic
attitudes of teachers. Rowley, NA: Newbury House.
William, G. C1979). Language group allegiance and ethnic, i
interaction. In H. Giles & B. Saint-Jacques CEds.). j
Language and ethnic relations. New York: Pergamon. j
!
i
33 4
Wilson, C. C19B4). A communication perspective on
socialization in organizations. Paper presented to the
annual meeting of the International Communication
Association, San Francisco, CA.
Wish, n., O’Andrade, R. G. & Gaodnow, J. E. II C1980).
Dimensions of interpersonal communication:
Correspondences between structures of speech acts and
bipolar scales. Journal of Personalitu and Social
Psuehologu. 39, 848-860.
Wittgenstein, L. C197SD . Philosophical investigations.
Oxford: Basil Blackwell.
Zukav, G. C19793. The dancing wuli masters. New York:
Bantam Books.
U M I Number: DP22411
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a com plete manuscript
and there are missing pages, th ese will be noted. Also, if material had to be removed,
a note will indicate the deletion.
Published by ProQ uest LLC (2014). Copyright in the Dissertation held by the Author.
Dissertation Publishing
UMI DP22411
Microform Edition © ProQ uest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United S tates Code
ProQ uest LLC.
789 E ast Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Building a model of organization acculturation: an interpretive study of organizational culture and stories

PDF

Communication during physical intimacy: A theoretical and methodological study of communication competence

PDF

Cinema-verite in America

PDF

AIDS preventive sexual behavior in college students: An empirical test of the health belief model

PDF

Calculations of electrostatic interactions in proteins

PDF

A formulation and partial test of a rules-based theory of interpersonal persuasion

PDF

Bob Hope and the popular oracle tradition in American humor.

PDF

Cinematic concepts and techniques in the adaptation and staging of chamber theatre

PDF

Clint Eastwood: An ideological study of his films, star image, and popularity

PDF

An empirical investigation of the communicative rules which should guide the use of deceptive messages within non-interpersonal and interpersonal relationships

PDF

A study of the intonation patterns of black and Standard English speaking children in formal and informal situations

PDF

An exploratory investigation of responses of young children to nonverbal auditory cues in messages

PDF

A critical interpretation of China in American educational films, 1936-1963: a historical and statistical analysis

PDF

A ship in a bottle: How communication builds the lifeboats of the world's largest A.A. group

PDF

A critical analysis of the society comedies of Henry Churchill De Mille and their contribution to the American theater

PDF

Clouds of witnesses: A rhetorical analysis of narrated witness in the Gospels.

PDF

A total evidence cladistic analysis of the Haliotidae (Gastropoda: Vetigastropoda)

PDF

A working theory of film genre

PDF

Business process reengineering as communication genres in space electronics

PDF

Connections of the dorsomedial hypothalamic nucleus in the rat

Asset Metadata

Creator Banks, Stephen Preston (author)

Core Title A critical analysis of ideology and discourse in two hotels

School Graduate School

Degree Doctor of Philosophy

Degree Program Communication Arts and Sciences

Degree Conferral Date 1987-07

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,speech communication

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Riley, Patricia (committee chair), [illegible] (committee member), Fisher, Walter R. (committee member), Och, Elinor (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-711218

Unique identifier UC11342928

Identifier DP22411.pdf (filename),usctheses-c17-711218 (legacy record id)

Legacy Identifier DP22411.pdf

Dmrecord 711218

Document Type Dissertation

Rights Banks, Stephen P.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA