gm»...

3.
\r. «31!:

”mm?
.w.

at»... m... in:
an. .3 . (in?

51.... .1‘ 4 2

t,
v.

"- .
.. . , . . , . . Lam?
V L, Ifmuuwwuauh

.Itl
5.1.

 

 

mmuunliil’illlllllijll‘ll‘lmluml
”BRARY 31293 01774 3935

‘ Michigan State
University

 

 

 

 

This is to certify that the
dissertation entitled
A SOCIOLINGUISTIC ANALYSIS OF FACE—THREAT
AND FACE-MANAGEMENT IN

POTENTIAL COMPLAINT SITUATIONS
presented by

LAURA C. HARTLEY

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Linguistics

 

 

 

 

 

 

 

Major professor

Date December 1, 1998

 

MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

1.32 5390”“

 

@2329:

 

 

APR' 1

 

 

 

DEM mm:

 

 

 

 

 

 

 

 

 

 

 

 

1M WM“

 

A SOCIOLINGUISTIC ANALYSIS OF F ACE-THREAT AND
F ACE-MANAGEMENT IN POTENTIAL COMPLAINT SITUATIONS

By

Laura C. Hartley

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Linguistics

I998

ABSTRACT

A SOCIOLINGUISTIC ANALYSIS OF F ACE-THREAT AND
F ACE-MAN AGEMENT IN POTENTIAL COMPLAINT SITUATIONS

By

Laura C. Hartley

Within the pragmatics literature, a number of studies have investigated the
speech act of complaining. In most cases, the researchers have failed to define a
"complaint" before proceeding with analysis. The purpose of this study is to
provide a theoretical definition of a "complaint" and to examine in-depth the
linguistic details of the speech act of complaining, with a particular view to
illuminating the linguistic devices employed in managing the "face" of the
participants involved in Potential Complaint Situations. In addition to dealing
with micro-level questions regarding how complaints are realized in various
circumstances, this study also sheds light on macro-level issues related to how
individuals create and maintain appropriate kinds of face-management in
conversational interaction.

Data in the study were collected through interviews, discourse completion
tests and ratings tasks and were analyzed using both qualitative and quantitative
means. The situations examined in the study varied in terms of three

sociolinguistic variables: gender, power and social distance. The respondent

group consisted of European-American students at a Midwestern university, aged
18-25.

Results of the study indicate that while complaints can be and sometimes
are made directly, it is more often the case that some kind of face-management is
employed. Options for managing face include the choice of opting out of
performing the face-threatening act, as well as performing the complaint indirectly
or utilizing mitigation. How and when complaints are made is affected by the
gender, power and social distance dynamics, as well as particular characteristics of
the situation.

This study also illuminates the importance of examining speech acts within
the social and cultural contexts in which they occur. It points to the importance of
the bearer as a co-constructor of meaning within the conversation and the effects
of the surrounding discourse on the identity and function of the speech act under

investigation.

ACKNOWLEDGMENTS

I am grateful for the tremendous support that I have received from my
advisor, Dr. Dennis Preston, throughout my graduate career at Michigan State
University. He has taught me much, not only about Linguistics, but also about the
pursuit of knowledge and the importance of mentoring relationships.

I am also very fortunate to have had the opportunity to work with Dr. Bruce
Fraser of Boston University on this project. His input has been invaluable, and
much of the theoretical rigor in this volume is the direct result of my interaction
with him.

Grateful acknowledgment is made to Cambridge University Press for the
right to reprint Figure l (reproduced in this volume as Figure 1, pg. 5) from
Politeness: Some universals in language usage by Penelope Brown and Stephen
C. Levinson.

Finally, I would like to thank my husband, Ben Hartley, for putting up with
all of my complaints throughout this process and my son, Luke, whose birth came
just one week after I finished writing this dissertation. The two of you inspire me

to be all that God has gifted me to be.

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................. vii

LIST OF FIGURES .............................................................................................. x

CHAPTER 1 ,

INTRODUCTION ............................................................................................... l
1.1 Theoretical Foundations: Speech Act and Politeness Theories ................ l
1.2 Previous Research on Complaints ........................................................... 6
1.3 Goals of this Study ............................................................................... 14

CHAPTER 2

METHODOLOGY ............................................................................................. 16
2.1 Defining Complaints ............................................................................. 16
2.2 Data Collection ..................................................................................... 18
2.3 Data Analysis ....................................................................................... 28
2.4 Expected Conclusions ........................................................................... 30

CHAPTER 3

THE F OLKLINGUISTICS OF COMPLAINING .............................................. 32
3.1 Folklinguistic Categories of Complaining ............................................. 32
3.2 Intention and Interpretation in Complaining .......................................... 45
3.3 Factors in the Interpretation of Complaints ........................................... 48

CHAPTER 4

DIRECTNESS AND INDIRECTNESS IN COMPLAINT REALIZATIONS....56
4.1 Face-Management Choices in Potential Complaint Situations ............... 56
4.2 The Choice of Opting Out ..................................................................... 58
4.3 Direct Complaints ................................................................................. 63
4.4 Mitigated Complaints ........................................................................... 70
4.5 Indirect Complaints .............................................................................. 77
4.6 Complaints and Requests for Repair ..................................................... 84

CHAPTER 5

SOCIOLINGUISTIC VARIATION IN COMPLAINING .................................. 94
5.1 Situations and Complaint Realizations .................................................. 94
5.2 Severity, Responsibility and Likelihood of Repair as Independent

Variables .............................................................................................. 99
5.3 The Effects of Sex, Power and Social Distance on Complaint
Realizations ........................................................................................ 107

CHAPTER 6

SUMMARY AND CONCLUSIONS ............................................................... 118
6.1 Summary of Results ............................................................................ 118
6.2 Critiquing Brown and Levinson .......................................................... 125
6.3 Implications for an Understanding of F ace-Threat and Face-

Management ....................................................................................... 130
6.4 Concluding Remarks .......................................................................... 133

APPENDICES
Appendix A: Interview Questions ............................................................... 135
Appendix B: Sample Questionnaire ............................................................ 139

REFERENCES ................................................................................................ 151

vi

LIST OF TABLES

Table 1. Aspects of complaining in folklinguistic definitions .............................. 32
Table 2. Folk labels for complaints ...................................................................... 36
Table 3. Definitions for selected subcategories of complaints .............................. 38
Table 4. Situations and responses judged in phase two ........................................ 39

Table 5. Results from: "Would you consider this statement to be a complaint?" .. 41

Table 6. T-values for intention v. interpetation ratings ....................................... 47
Table 7. Factor groups (with weightings) ............................................................ 49
Table 8. Means scores for each response by factor group .................................... 50
Table 9. Overall directness levels for complaint realizations ............................... 57
Table 10. Frequency of opting out of performing any FTA ................................. 59
Table 11. Use of direct complaints by situation ................................................... 64
Table 12. Components of complaints in each situation ........................................ 67
Table 13. Use of mitigated complaints by situation ............................................. 72
Table 14. Mitigating devices used in complaint realizations ................................ 73
Table 15. Use of indirect complaints by situation ................................................ 77
Table 16. Categories of indirect complaints ......................................................... 79
Table 17. Overall directness levels for requests for repair ................................... 87
Table 18. Mean scores for "likelihood of repair" ratings ...................................... 88

Table 19. Complaints and requests for repair by situation and directness levels .. 90

Table 20. Mean scores for three characteristics in each situation ......................... 95

vii

Table 21. Results of post-hoc Tukey tests for the three characteristics ................ 97
Table 22. Overall effects of severity of offense on complaint realizations ......... 100
Table 23. Overall effects of responsibility of offender on

complaint realizations .................................................................................. 100
Table 24. Overall effects of severity of offense on requests for repair ............... 101
Table 25. Overall effects of responsibility of offender on requests for repair ..... 101
Table 26. Overall effects of likelihood of repair on requests for repair .............. 101
Table 27. Overall effects of responsibility on directness levels in complaints... 102
Table 28. Overall effects of likelihood of repair on directness levels

in requests .................................................................................................... 103
Table 29. Severity of offense for complaints in situation 4 ................................ 104
Table 30. Responsibility of offender for complaints in situation 4 ..................... 104
Table 31. Likelihood of repair for complaints in situation 1 .............................. 105
Table 32. Severity of offense for requests for repair in situation 1 ..................... 106
Table 33. Likelihood of repair for requests for repair in situations 1, 2 and 4 107
Table 34. Complaint realizations by sex, power and social distance .................. 108
Table 35. The effects of power on the choice of opting out of complaining ....... 109
Table 36. The interaction of power and hearer sex in complaint realizations ..... 110
Table 37. The effects of power on directness levels of complaints ..................... 111
Table 38. The effects of speaker sex on directness levels of complaints ............ 112
Table 39. The interaction of speaker and hearer sex in complaint directness ..... 113

Table 40.

Table 41.

The interaction of speaker sex and power in complaint directness ..... 114

The interaction of speaker sex and distance in complaint directness .. 114

viii

Table 42. The interaction of hearer sex and power in complaint directness ........ 115

Table 43. Overall results for complaint realizations in each situation ................ 121

ix

LIST OF FIGURES

Figure 1. Possible strategies for doing FTAs ......................................................... 5
Figure 2. Complaint categorization for each statement ........................................ 43
Figure 3. Is v. Intended to be. v. Interpreted as complaint .................................... 46
Figure 4. Mean scores for each response by factor group .................................... 52

Chapter 1

INTRODUCTION

1.1 Theoretical Foundations: Speech Act and Politeness Theories

In 1962, a series of lectures by the philosopher J.L. Austin was published
under the title How to Do Things with Words. A major premise in this work was
that utterances are not only used to describe or report on situations. Rather it is the
case, in Austin’s words,

that to utter the sentence (in, of course, the appropriate circumstances) is

not to describe my doing of what I should be said in so uttering to be doing

or to state that I am doing it: it is to do it (p. 6, italics his).
Austin calls these kinds of utterances “performatives,” since in saying them, a
person is actually performing some act.

After developing the notion of performatives and the conditions under
which they can be successfully performed (which he calls "felicity conditions"),
Austin moves into a broader discussion of the ways in which to say something is
to actually do something. He distinguishes three kinds of acts that are all
accomplished by the speaking of any single utterance. First, there is the
locutionary act, which is "roughly equivalent to uttering a certain sentence with a
certain sense and reference, which again is roughly equivalent to 'meaning‘ in the
traditional sense" (p. 108). Secondly, an utterance accomplishes an illocutionary

act, which is the force that the utterance has, or the way in which the utterance is

being used. For example, any number of utterances can be used to perform the

illocutionary act of asking a question, giving information, ordering, warning, etc.
Finally, an utterance also performs a perlocutionarjy act, namely "what we bring
about or achieve by saying something" (p. 108), e. g. convincing, inciting to action,
conveying urgency, etc.

John Searle, a student of Austin, elaborated on the notion of using language
to accomplish certain actions in his 1969 work entitled Speech Acts. In the
beginning of the book he explains that to speak a language involves performing
speech acts and goes on to describe the main thesis of his argument as follows:

...all linguistic communication involves linguistic acts. The unit of

linguistic communication is not, as has generally been supposed, the

symbol, word or sentence, or even the token of the symbol, word or
sentence, but rather the production or issuance of the symbol or word or

sentence in the performance of the speech act (p. 16).

In the third chapter of Speech Acts, Searle undertakes an analysis of the
level of "illocutionary acts" suggested by Austin. Using the illocution of
"promising" as a case in point, he lays out the conditions that must hold for an
illocutionary act to be successfully accomplished. Searle points out that these
conditions are similar to Austin's notion of felicity conditions in the use of
performatives.

The work of Austin and Searle has become the basis for a whole body of
research into speech acts. Some scholars have focused attention on further
developing and clarifying a theoretical understanding of speech acts (see, for

example, Sadock, 1974; Bach and Hamish, 1979; Tsohatzidis, 1994). Other

researchers have focused on applications of speech act theory, examining how

speakers in various languages perform a variety of speech acts. Some of the
speech acts that have been studied include requests (e. g. Blum-Kulka & Olshtain,
1984; Gordon & Ervin-Tripp, 1984; Blum-Kulka & House, 1989; Weizman,
1993), apologies (e.g. Borkin & Reinhart, 1978; Olshtain & Cohen, 1983; Blum-
Kulka & Olshtain, 1984; Bergman & Kasper, 1993 ), compliments (W olfson, 1981,
1989), expressions of gratitude (Eisenstein & Bodman, 1986, 1993), and refusals
(Liao, 1994).'

In addition to the foundational work of Austin and Searle, many of these
studies have relied in some part on Brown and Levinson's (hereafter B&L) (1978,
1987) politeness theory. Building on the work of Goffman (1967), B&L develop
the notion that all human beings are constantly at work in the process of
maintaining different aspects of their own and others' "face," i.e. the public self-
irnage that each individual projects to the world. They claim that "face" is
"something that is emotionally invested, and that can be lost, maintained, or
enhanced, and must be constantly attended to in interaction" (p. 61).

B&L divide face into two components: negative face, which is the basic
claim to freedom of action and freedom from imposition, and positive face, the
desire to be appreciated and approved of. The terms "positive" and "negative"
should be understood here as technical labels, without the normal connotations of

"good" and "bad." Every individual has and must attend to both kinds of face.

 

' Some of these examples are not "classical" speech acts, in Austin and Searle's definitions, but
would certamly fall into this general area of study as it has been understood and expanded by
researchers since that time.

Having thus defined face, they argue that there are certain kinds of
communication acts that intrinsically threaten some aspect of face of either the
speaker, hearer or both. They call these acts face-threatening acts, or FTAs. B&L
distinguish FTAs based on both the kind of face threatened (i.e. positive or
negative) and whether the FTA constitutes a threat to the speaker's face or the
hearer's face. They note that these two distinctions give rise to a four-way grid for
classifying F TAs based on the kind of face that is primarily threatened. In an
endnote, they offer examples of speech acts that illustrate threats to the four kinds
of face: promises (threat to speaker negative face), warnings (threat to hearer
negative face), apologies (threat to speaker positive face) and criticisms (threat to
hearer positive face). While this grid provides a neat classification schema, B&L
point out that, for most FTAs, the situation is considerably more complicated than
this grid suggests, since many F TAs can be classified as fitting into more than one
category. Even with the seemingly straightforward examples provided above, it is
evident that this is the case. For example, while a criticism primarily threatens the
hearer's positive face, it is no doubt the case that the speaker's positive face is also
on the line, since the speaker runs the risk of being seen as overly-critical or
unjustified in his or her comments.

Because performing FTAs in the most direct way (i.e. by strictly

adhering to Grice's (1975) conversational maximsz) could easily cause offense,
B&L claim that speakers use a number of strategies to soften or mitigate FTAs.

They lay these strategies out in a kind of decision-making schema, as shown in

Figure l.

1. without redressive action, baldly
on record < 2. positive politeness

Do the FT A < with redressive action
< 4. off record

5. Don't do the FIA

3. negative politeness

Figure 1. Possible strategies for doing FTAs (Brown & Levinson, 1987, p. 69.
Reprinted with the permission of Cambridge University Press.)

According to this schema, the most direct way of performing an FTA is to
do it baldly, without redress3 while the most indirect way to perform the act is to
go off record, i.e. to do it in some way that keeps the attributable intentions of the
speaker ambiguous. In between is the possibility of performing the act on record,
i.e. unambiguously, but with some kind of redressive action directed to either the
positive or negative face of the bearer. By laying out this schema for doing FTAs,

B&L appear to equate levels of politeness with levels of directness, although they

 

2 Grice posited that all participants in conversation adhere to a "Cooperative Principle", which
states that individuals will attempt to cooperate in conversation. He further elaborates this by
claiming that, in particular, interactants will follow four "maxims": Quantity (say neither more
nor less than required); Quality (make your contribution genuine and not spurious); Relation
(make your contribution appropriate to the transaction); Manner (make your contribution clear).

3 "Redress" and "redressive action" as used here are synonymous to what is also referred to in this
volume as "mitigation."

never make this connection explicit.4

The majority of the rest of B&L's volume is devoted to spelling out specific
strategies within the four overall macro-strategies described above. They claim
that these various strategies are universal, i.e. they can be utilized by any speaker,
regardless of the language being spoken. They illustrate each of the strategies
using data primarily from three languages: English, Tamil and Tzeltal. While
these three languages are certainly associated with diverse cultures, a claim of
universality based on such minimal empirical evidence is shaky at best. In the two
decades since the original theory was put forth, B&L's politeness theory has
become the framework through which a good deal of empirical work in a number
of different languages has been done.5 Taken together, these studies provide a rich

resource for the evaluation of B&L‘s claims regarding universality in politeness.

1.2 Previous Research on Complaints
One interesting speech act to investigate in terms of the kinds and extent of

politeness and face-management utilized is that of complaining. Like the speech

 

4 While B&L deal extensively with positive and negative face and the various kinds of mitigation
that can be used in the process of face management, they never actual define what the term
“politeness” in their “politeness theory” refers to. Fraser (1990) identifies Brown and Levinson's
approach as one of four different models of politeness found within the linguistic literature. Their
model is what Fraser calls the "face-saving view." The other models are the "social-nonn View."
the "conversational-maxim view." and the "conversational-contract view." Meier (1995) defines
politeness functionally as “behaving in a socially appropriate manner” (p. 351) and then points
out that both context and participant expectations will play a role in determining what constitutes
appropriate behavior.

5 While not the primary purpose of this paper, it is the author's hope that the results of this study
will contribute to the increasing body of empirical work that investigates speech acts and face-
management strategies.

act of criticism, complaining is a speech act which primarily threatens the positive
face of the hearer, although the speaker's positive face is also on the line. Studies
in the literatures of disciplines such as Linguistics, Communication, Social
Psychology and even Business Marketing have attempted to describe and analyze
various aspects of the complaint event.

Candlin, Coleman and Burton (1983), for example, center their attention on
one particular situation in which complaints often occur, namely dentist-patient
communication. Their focus is on the different discoursal sets that dentists and
patients bring to the communication situation and how that affects the
conversational flow. They claim that patient complaints occur within a three-stage
event: (1) the dentist provides a cue that a report should be made; (2) the patient
presents the complaint, often indirectly; (3) the dentist responds to the complaint
by either ignoring it altogether, disagreeing with it, or agreeing but minimizing it.
Although the researchers provide several examples from naturally occurring data,
they never spell out precisely how they are defining "complaint. "

Beebe and Takahashi (1989) examine Japanese and American styles of
chastisement6 through the use of both ethnographic data collection and discourse
completion tests. On the discourse completion test, subjects were presented with
two situations in which they were asked to place themselves and then respond with

what they would say in the situation. Both situations involved a photo-copying

 

6 The authors of this study do not define what they mean by the term "chastisement.” The
situations that were utilized and responses they received. however. appear quite similar to the
kinds of scenarios present in other research that has claimedlto be studying complaints.

task within a business situation. In one scenario, they were asked to imagine
themselves in the higher status position (i.e. corporation president) and then
respond to a mistake made by their assistant. In the second situation, they were to
play the part of a middle-manager dealing with a miscommunication with their
boss.

The researchers analyzed the responses for each situation as falling into
one of four categories, based on what they call “semantic formulas.” For situation
1, these were statement of error, correction, request for repair, and criticism. They
found that Americans were more likely to state the error and ask for repair, while
the Japanese most often criticized. Situation 2 produced semantic formulas of
self-defense, apology, offer of repair and criticism. In this case, both groups
resorted to statements of self-defense, although the Americans did so at a higher
rate. Americans also apologized more than Japanese did. From the figures that
Beebe and Takahashi present, it is evident that many of their responses were
categorized into more than one semantic formula.7 They make no mention of this
in the text of their article, however, and therefore do not discuss whether certain
formulas often occurred together in specific ways.

Focusing on a different aspect of the complaint situation, Boniskowska
(1988) examines the choice of opting out of performing a complaint. Building on

an analysis of complaints in an unpublished master's dissertation, this article

 

7 This evidence comes from the fact that the percentages do not total 100%. For example, in
Situation 1, semantic formulas used by Japanese respondents are given as: statement of error
(39%), correction (31%), request for repair (46%), and criticism (54%).

investigates the reasons given by British English-speaking students when they
chose not to complain in a situation. The researcher found three main reasons why
students opted out: (1) the act (A) was not seen as offensive because the speaker
(S) believed the bearer (H) was justified in causing it; (2) S did not hold H
responsible for A; and (3) S did not blame H openly for A, feeling that she or he
had no right to do so.

Clyne's (1994) discussion of complaints is situated within a very broad
investigation of intercultural communication within the work environment. His
treatment of complaints is thus much briefer than other researchers who have
focused more specifically on this single speech act type. He divides complaints
into two categories: exercitive complaints, where someone is exercising power,
and whinges, long and repeated expressions of discontent. Of the studies
discussed so far, Clyne comes the closest to defining “complaint” by offering
definitions for these two subcategories of complaint. The results of his analysis
suggest that, while people from all groups in his study whinge, European males
are the most likely to complain, and Southeast Asian women are the least likely to
either complain or whinge.

Newell and Stutrnan (1989) consider the speech act of complaint within the
broader context of conversational negotiation of confrontation, especially with
reference to initiation and response. Thus they focus their analysis on the entire
conversational episode in which a complaint appears. They point out that there is

an inherent difficulty in studying complaints, since a great deal of different kinds

of activity can all be placed within the category of "complaint."

After listing a number of studies which deal with complaints (none of
which provide a definition), they note that “the definition of complaining has
proven problematic for conversational analysts” (p. 146). In their final analysis,

“6

they conclude that to complain’ in fact denotes a variety of different speech
acts” Q). 148) which then may be used to initiate different types of communication
episodes. Thus “complaint” is in reality, for these researchers, merely a cover
term for the various kinds of acts (again, not defined any more specifically then
this) that bring about social confrontation.8 Since social confrontation is the focus
of their interest, the remainder of their article deals with the responses to
complaints and how confrontation is negotiated within a conversation.

While Edmondson (1981) is primarily interested in investigating the speech
act of apology, he discusses complaints in terms of their role in on-going discourse
in providing an opportunity for an apology to be offered. He notes that the

conditions for complaining and apologizing to occur are similar, namely:

APOLOGIZE: S did P, P bad for H
COMPLAIN: H did P, P bad for s9

 

8 Newell and Stutrnan do point out that “[a]lthough confrontations always begin with complaints,
complaints about behaviors do not always function to initiate confrontations” (1989/90, p. 148).
With this acknowledgement, however, they essentially void whatever definition of complaint they
have thus posited (i.e. as an act that initiates social confrontation). Ultimately, these researchers
also proceed with an analysis of complaints without having any definition of complaint to work
from.

9 S here stands for "speaker" and H for "bearer." P refers to the event that occurred.

10

and notes that "in the case of APOLOGY and COMPLAIN, the same event P
might be involved, such that what from one conversationalist's viewpoint
constitutes a ground for a COMPLAIN constitutes a ground for an APOLOGY for
his interlocutor" (p. 278).

As with other studies, Edmondson fails to define the speech act of
"COMPLAIN", although he does make several important observations about the
occurrence of complaints. He points out, for example, that unlike apologies, there
does not appear to be an overt performative for complaining. Instead, he suggests
that indirectness is both appropriate and conventional in the performance of a
complaint, which he ties to the fact that complaints constitute non-H supportive
behavior modes. '0 Furthermore, Edmondson notes the fact that in discourse terms,
a complaint has no conventional "right-pair-part" that will satisfy that complaint.
He claims that this follows from the fact that the perlocutionary intent of a
complaint is negotiable, a point elaborated on by the Newell and Stutrnan article
discussed above. ”

In their discussion of politeness markers in English and German, House and
Kasper (1981) also deal with the speech act of complaint. Their primary goal in

this article is to compare English and German speakers in terms of directness.

They present a schema of eight directness levels of complaints, although the

 

'0 Edmondson is drawing here on terminology from Leech's (1977) "Tact principle" rather than
Brown and Levinson's politeness theory, but the concepts of "non-H supportive behavior" and "H
face-threat" are clearly compatible.

” This point will be more elaborately illustrated in Chapter 3, when the difference between
intention and interpretation of complaints is discussed.

ll

justification for their ordering of these levels is not presented in this article.12
Furthermore, while they lay out the circumstances under which a complaint can
occur, they also do not specifically define complaint. In fact, they claim that their
use of the term “complain” in this study is broad enough to include other
expressive verbs such as “criticize”, “accuse” or “reproach.”

Finally, Olshtain and Weinbach (1993) report on two different studies
involving complaints. To their credit, the first sentences of their article lay out the
definition of complaint from which their analysis will proceed. They say:

In the speech act of complaining, the speaker (S) expresses displeasure or

annoyance—censure—as a reaction to a past or ongoing action, the

consequences of which are perceived by S as affecting her unfavorably.

This complaint is usually addressed to the bearer (H) whom the S holds, at

least partially, responsible for the offensive action. For the purpose of this

study, censure will be assumed to have been expressed whenever S chooses

to verbalize her disapproval of the violation (p. 108).

The researchers then proceed to discuss two studies of complaints made by
speakers in their native languages. In the first study, Hebrew-speaking university
students were presented with 20 situations and asked to respond in a discourse
completion format. From the responses, a five-point scale of complaint realization
was developed in terms of the degree of face-threat that the speaker undertook.

These five categories, from least to greatest amount of face-threat, were: (1)

Below the level of reproach, realizations that enable S to avoid direct mention of

 

'2 For example, they claim that the statement "There‘s a stain on my blouse” is more indirect than
saying “Did you wear my blouse by any chance?” In the first case, the hearer‘s agentive
involvement in the act is avoided. But in the second case, explicit mention of the act itself is lcfi
out. Why the authors consider the second statement to be more direct than the first is not clear.

12

the act (A) or focus on S; (2) Expressions of annoyance or disapproval, do not
explicitly mention A or H, but do express general annoyance at the violation; (3)
Explicit complaint, S makes use of an openly face-threatening act towards H, but
no sanctions against H; (4) Accusation and warning, S uses an open-face threat
and implies potential sanctions; (5) Immediate threat, S openly attacks H. In this
study, Olshtain and Weinbach found that the respondents tended to cluster around
the three central strategies, and that the relative status of the S and H influenced
the choice of complaint strategy. It is difficult to assess the validity of the coding
categories in this study, however, since the authors provide only one or two
examples in each case.

The second study focused more on the actual linguistic realizations of
complaints, but merely presented four categories into which responses were
placed: indirect, unmitigated, mitigated, and opting out. This study examined
British-English, American-English and (Israeli) Hebrew speaking groups. They
found that all three groups chose to express complaint more than they chose to opt
out. They also discovered that overall, the three groups behaved very similarly in
their choice of strategies. Assessment of their work is again difficult given the
fact that they offer no examples of what they consider to be an indirect, a

mitigated, or an unmitigated complaint.

l3

1.3 Goals of this Study

The above discussion about past research on complaints has been
purposefully detailed. In each case, the studies have focused on particular parts of
the complaint event but have often failed to define a "complaint" itself or provide
the necessary support for the framework used within their analysis. It is important
for a strong theoretical foundation to undergird any empirical analysis; indeed
without such a frame, it is difficult to interpret data in any meaningful way.

Accordingly, the goals of this study are:

(1) to construct a working definition of the speech act of complaining. This
definition must minimally include:

(a) a description of the kinds of situations in which complaints can
occur; and
(b) a statement regarding what constitutes a complaint itself;

(2) to investigate the roles that speaker intention and hearer interpretation
play in the evaluation of whether or not a complaint has been made;

(3) to understand the ways in which indirectrress and mitigation are
accomplished within complaint realizations and how they contribute to
the overall face-management accomplished;

(4) to analyze the effect of the sociological variables of gender, power and
social distance on complaint realization;

(5) to situate the above analysis within a broader construct of face-

management, in order to see how complaints interact with other speech

14

acts in potential face-threatening situations.

The organization of chapters in this volume are laid out along the lines of
the research questions presented above. The next chapter first provides a working
definition of the speech act of complaining as well as a description of the pre-
conditions necessary for an utterance to be considered a complaint. It then
describes the methodology utilized for the data collection of the present study,
including hypotheses of the study. Chapter 3 focuses on the results first from the
portion of the data collected through interviews, which reveals a good deal
regarding the folklinguistic notions of complaints. Some of the quantitative data
collected as a follow-up to the interviews are also discussed here, and these data
shed some light on what goes into the evaluation of whether or not a complaint has
been made.

The data collected through discourse completion tests are examined in Chapter
4. This chapter includes a more careful discussion of the strategies employed for
directness, indirectness and mitigation in both complaint realizations and requests
for repair, the two kinds of face-threatening acts that appeared most frequently in
the data. Chapter 5 discusses the kinds and extent of variation found in the data as
a result of the independent sociological variables examined. Finally, Chapter 6
discusses implications of the current study within the theoretical frame of

politeness theory and suggests a possible course for further study.

15

Chapter 2

METHODOLOGY

2.1 Defining Complaints

As noted in the first chapter, any investigation of a speech act must begin
with a working definition of the act under investigation. A problem with many
studies in the speech act literature is that they proceed by asking their readers to
merely rely on some intuitive understanding of the speech act in question. While
native-speaker intuition certainly plays an important role in the analysis of when a
complaint has and hasn't been made, it is still necessary to spell out as explicitly as
possible what constitutes a complaint. A good starting point is the examination of
the conditions under which a complaint can be made.

In the broadest sense, a speaker (S) can make a complaint about any
situation that he or she finds unsatisfactory. It is not the case, as Olshtain and
Weinbach (1993) suggest, that some particular action has to have been committed
by some hearer (H). In fact, one can complain about circumstances brought about
by oneself (e. g. “I can’t believe how horribly I did on that test”) or even about
situations with no apparent agentive cause (e. g. “I’m so tired of this rain”) A
person can also complain about another individual to a third party1 (e. g. “I’m so

mad that Sue did that”) Furthermore, it is not obligatory that S perceive some act

 

1 This kind of complaint is what Boxer (1993) calls an "indirect complaint.” This terminology
should not be confused with my later discussion of directness v. indirectness in complaint
realizations.

l6

as having unfavorable consequences for him or herself, as Olshtain and Weinbach
also claim. One can certainly complain about an act committed against someone
else (e. g. “I can’t believe that she didn't help you”) Thus it appears that the only
prerequisite for complaining is that the speaker finds a particular situation to be
unfavorable in some way. In this study, I will refer to situations where this is the
case as "Potential Complaint Situations."

That being said, however, it is understandable that the past research which
has examined complaints has primarily focused on circumstances in which the
speaker is complaining about some particular act committed against him or herself
for which he or she holds the bearer to be responsible in some way. These are the
situations in which face is threatened most directly and which confrontation and
face-management must be balanced most carefully. Thus they provide the most
interesting scenarios for an analysis of politeness phenomena. As in past studies,
the focus of the current study will be on situations in which the bearer has
committed some particular act that the speaker believes to have adverse
consequences for him or herself.

If the preconditions of a complaint are only that the speaker finds some
situation unsatisfactory, the definition of a complaint is simply the expression of a
negative evaluation of the situation by S. In order for the complaint to be
considered a direct complaint, however, this negative evaluation must be
straightforwardly recoverable from some aspect of the utterance itself. This will

usually be a negatively-loaded word or phrase, but could also be a particular

17

intonation. Paralinguistic devices are an important area of study in the realm of
complaints, but due to scope limitations, they will not be examined in the current
study. Because complaints are so highly face-threatening, a wide variety of
options exist for making a complaint indirectly. This point will be further

elaborated in Chapter 4.

2.2 Data Collection

A major tenet in sociolinguistic research is that if we want to investigate
how people actually use language, we need to examine contexts in which language
occurs naturally. This desire for "naturalness," however, is sometimes sacrificed
in the course of research for a number of reasons. First, there are ethical issues
related to observing and especially recording people's speech surreptitiously,
which may be one of the few ways of obtaining truly naturally occurring data.
Once people are made aware that their speech is under investigation, they often
become more self-conscious of how they are speaking (a phenomenon described
by Labov (1972) as the "observer's paradox"). Second, if language is analyzed in
naturally occurring situations, it may be difficult to account for all of the factors
that are potentially influencing how the language is being used. There is little
experimental control in these cases. Finally, when investigating a specific
linguistic item, using naturally occurring data may result in a tremendous amount
of analysis that yields only a few tokens of the item being studied. In terms of the

study of speech acts, it may be fairly easy to observe instances of, for example,

18

greetings by simply hanging around areas where people are likely to meet and
greet one another. In the case of complaints, however, it is much more
challenging to find ways of observing natural instances of complaint realizations.

Because of these challenges inherent in the study of speech acts, many
researchers have relied on the use of discourse completion tests, in which
respondents are presented with situations and asked to write what they would say
in that situation. For reasons of both experimental control and comparability to
other research in this area, the present study also utilized a discourse completion
test as the primary means of collecting utterances in Potential Complaint
Situations.

The data for this study were collected in two phases. In the first phase,
qualitative interviews were conducted by university-aged students with their peers.
The primary reason for having students, rather than the researcher, interview
respondents was to minimize the effects of the observer’s paradox as much as
possible by reducing the power differential within the interview situation and
increasing the participants’ familiarity. In this way, the researcher was able to
obtain more casual and less guarded responses.2 The interviewers were given a
script to follow in which they first asked their respondents to define a complaint
and to provide other words that could be used to describe a complaint. Follow-up

questions were used to further define and clarify the categories which respondents

 

2 The interviews were often filled with laughter and were sprinkled with both slang terms and in
many cases a good number of Obscenities, all indications of their casual tone.

l9

offered. The interviewers then presented respondents with a number of actual
complaints drawn from a pilot study (Hartley 1996) and asked them to provide
words or adjectives that describe the complaint or the person using the complaint.3
A total of 73 interviews were conducted, with 38 female respondents and 35 male
respondents. The dyads were primarily same-sex, although a number of cross-sex
pairs also participated.

In phase 2 of the data collection, 120 university students (60 males and 60
females) completed a questionnaire that contained two parts.4 The first part was a
discourse completion test, in which respondents were presented with four
situations and asked to write what they would say in those situations. The four
situations chosen varied from each other in terms of two important factors which
Brown and Levinson (1987) claim play a significant role in influencing levels of
politeness--namely, social distance and power.

Social distance (D) in this study is represented by the degree of familiarity
between participants in the situation. A (-D) situation is one in which there is a
high degree of familiarity, as opposed to (+D), where the participants do not know
each other well, if at all. Power has to do with the degree to which one participant
in a situation is able to impose his or her plans or self-evaluation at the expense of
the other. A (-P) situation is one in which the participants are more-or—less social

equals, while a (+P) situation in this study will be one in which the offender has a

 

3 The actual script that interviewers were asked to follow is presented in Appendix A.
4 A copy of the questionnaire as it was presented to respondents can be found in Appendix B.

20

higher rank ascribed to him or her than the person who is offended and thus
complaining.5

Furthermore, this study includes an added sociolinguistic variable, namely

gender. For each situation, half of the male respondents and half of the female
respondents were asked to imagine a female committing the offense, while the
other half of the respondent group was presented with a male offender. The
results were a four-way gender classification for each scenario in terms of the
gender identities of the complainer/speaker and the offender/bearer: female-
female, female-male, male-female and male-male.

The respondents were given the following directions for completing the

discourse portion of the questionnaire:

Directions: You will be presented with four situations. For each situation,

do the following:

1. Read the details of the scenario and then write out a script of what you
would say in that situation on the blank lines given below the scenario.
Please try to be as realistic as possible; that is, write only what you
believe you would actually say, and not what you would merely think or
wish to say.

2. If you wouldn’t say anything in the situation, write “Nothing” on the
first line and then on the remaining lines give a brief explanation as to
why you would say nothing.

3. After you have completed this task, rate each situation by circling a
number on each of the scales which follow it.

The actual scenarios that were presented to the respondents are listed

below, along with an indication of the social distance and power differentials that

 

5 A third possibility for power exists. where the person doing the complaining has more power
than the one being complained to. This situation. however, calls for the least amount of face-
management of the three power possibilities and has thus been excluded from this study in order
to make the scope of the investigation more manageable.

21

they represent (although these markings were not present on the questionnaires).

Only one gender was presented to each respondent, as explained above.

1.

A good fiiend of yours from high school is visiting you for the weekend. In
preparation for his/her visit, you clean the kitchen thoroughly. Shortly after
your friend’s arrival, you tell him/her to make himself/herself comfortable

while you run out to the store. When you arrive home, you notice that your

friend has made a huge mess in the kitchen. You say to your friend:
('1), 'P)

You’ve been waiting in line for two hours to buy tickets to a concert that is
almost sold out. As you're standing there, a man/woman about your age
tries to cut in line in front of you. You say: (+D, -P)

You have worked for your current boss for some time now and you get
along well. When your annual performance review comes around, you find
out that he/she has given you a bad review. You do not think the review is
a fair picture of your work. When you see your boss, you say: (-D, +P)

Last year when you were planning out your courses, you went to talk for
the first and only time to the professor who has been assigned to you as
your academic adviser, in order to ask his/her advice on what courses you
should take for your major. After taking several of the courses he/she said
you had to take, you find out that one of the courses will not count towards
your major after all, and that you will need to take a different course instead
during summer school in order to graduate on time. You go to your
adviser’s office during his/her office hours and say: (+D, +P)

After the description of each situation, respondents were given five blank lines to

write out their responses. It should be noted that the four scenarios were presented

to respondents in different orders, so as to reduce any ordering effects overall.

As mentioned above, the four scenarios in this study were constructed to

represent the four possible combinations of [+/- power] and [+/- social distance],

with the additional factor of gender (of both speaker and hearer). To achieve

slightly better control in this study, a single situation could have been chosen and

22

the participants in the scenario varied to produce the four-way power/distance
distinction. There are several reasons why this research utilized four different
scenarios, rather than this single scenario with varied participants approach.

First, one of the aims in constructing the scenarios was to create situations
in which the respondents could reasonably and realistically find themselves. To
imagine a single scenario in which a fiiend, a stranger, an employer and an advisor
(or some other manifestations of the possible power/distance relationships) would
all commit the same offense proved to be quite a challenge. While a scenario such
as cutting in line might have been a candidate, it does not seem likely that
someone one knows (i.e. a [- distance] relationship) would simply cut in line
without saying anything. In addition, as these respondents were all university
students, it is difficult to imagine that many of them would be waiting in line for a
concert that both their peers and, for example, academic advisors might equally
want to attend. Perhaps a different kind of event, such as a football game, would
be more realistic. In any case, trying to create a single scenario that would find
four different participants equally likely to commit the same offense could easily
have taken away from the plausibility of the scenarios.

Secondly, the practical concern of availability of respondents motivated the
choice to use four different, rather than one single, situation. Were respondents
presented with the same scenario four different times, there may have been some
redundancy effects in their responses. In other words, by the time they reached the

fourth presentation of the same scenario, their responses may have been greatly

23

influenced by the previous responses they had written. Great care would have had
to be taken to randomize the scenarios so that equal numbers of respondents were
presented with the four participants in all 16 possible orderings. Another
possibility would have been to present each respondent with only one of the
possible four participant scenarios. This, however, would have required four times
the number of respondents. As questionnaires for this study were handed out in a
large undergraduate class in which respondents were not required to participate in
the study, either of these possibilities would have posed serious difficulties for the
timely collection of data.

Finally, using only one scenario instead of four would have allowed for
better control, but this control would have come at the expense of some
comparative ability. Even using four different scenarios, this research study is
necessarily quite narrowly focused in terms of the demographics of the respondent
group. To further narrow the scope of investigation to a single scenario would
make any statements of generalizability that much more tentative. By employing
four different situations, comparisons can be made among the responses given for
different kinds of scenarios, and these differences themselves become an
interesting source of variability.

In past studies of complaints, up to 20 different scenarios have been utilized
Within the data for a single study. These studies have usually lumped all responses
together into a single data set for analysis, without taking into account any

variability which may be a result of the differences in how the offenses in question

24

are perceived. The current study thus goes one step further in its analysis by
attempting to understand how the characteristics of the situation itself can
contribute to the realization of complaints.

Since four different situations were used in this study, an important
question is in what ways (beyond the power/distance relationships of the
participants) the situations are significantly different from one another. More
specifically, what are the characteristics of a situation that nright affect the way in
which a complaint is registered? Although a large number could be imagined (and
perhaps even controlled for), three factors were examined in the current study.

First, the perceived severity of an offense is likely to influence complaint
realizations. If an offense is seen as not severe at all, chances are greater that
respondents may choose not to complain at all, or may use a less face-threatening
complaint than if the offense is perceived to be quite severe. Second, how
responsible the speaker holds the bearer for the offense in question may also
influence how a complaint is made. This is particularly true in situations in which
a person is complaining to some representative of the offending party rather than
the person who directly committed the offense, although no scenarios of that sort
were utilized in the current study. Still, it is an interesting question as to how
responsible the offenders in each of the four scenarios are held to be.

Finally, whether or not the speaker thinks that the bearer is likely to repair
the situation may also affect the complaint realization. This characteristic speaks

to the motivation for the complaint. It is an interesting question as to whether

25

complaints differ if the individual making the complaint feels like his or her
complaint is likely to change the situation in question. It is important to note again
that these three characteristics do not represent an exhaustive list of the possible
influences on complaint realizations. They do, however, provide a good starting
point for an investigation of how a situation itself might produce some variation in
complaints, apart even from the participant relations involved in the situation.

In order to examine these three characteristics, respondents were presented
with three questions after they had written their responses for each scenario and

asked to rate them as follows:

How severe is this oflense?

not severe at all 1 2 3 4 5 very severe

How responsible is [the person] for this situation?

not very responsible 1 2 3 4 5 very responsible
How likely is it that [the person] will do something to correct the situation?

not likely at all 1 2 3 4 5 very likely

The four situations above were chosen because they were situations in
which the respondents could realistically find themselves. Thus, while the data are
not naturally occurring, the respondents were not being asked to role play a type of
situation with which they might be completely unfamiliar, a criticism which has
often been made regarding discourse completion data collection techniques.

In the second part of the questionnaire, respondents were presented with

26

three complaints from each of the same four situations (for a total of 12
complaints) which were drawn from the pilot study. For each complaint the
respondent was asked the following questions:

1. Would you consider this statement to be a complaint?

2. If the respondent circled “yes,” they were then asked which category of

complaint the response best fits into.6

3. Do you think the person who made this statement intended for it to be a

complaint?

4. Do you think the other person in the situation will interpret this

statement as a complaint?

The respondents were then asked to rate the complaint on nineteen five-
point semantic-differential scales.7 The word pairs used for these scales were
drawn from the terms offered in the second half of the interview data. For
example, since in the qualitative interviews a number of people had labeled some
of the responses as “rude,” all of the responses presented in the second half of the
questionnaire were rated on a “rude - polite” scale. Once again, the situations and
complaints were presented in different orders to different groups of respondents in
order to eliminate possible ordering effects.

In order to control the number of independent variables within this study,
the respondent group chosen was fairly socially homogenous. The respondents

were all between 18 and 25 years old, from the Great Lakes or Midwest region of

the US, students at Michigan State University, and native speakers of English. In

 

6 The categories offered were five of the most frequently stated terms given by respondents in the
initial portion of the interview data when they were asked about other terms that could be used to
describe a complaint. The respondents were also given the category of “Other.” followed by a
blank line on which they could write an alternative term.

7 The actual scales utilized are presented in Section 3.2.

27

addition, it was primarily respondents of European-American descent within this
demographic group that were studied, although a number of respondents of other
racial groups were included in the interview portion of the data collection.

For the questionnaire portion of the data collection where greater control
over the respondent group was possible, care was also taken to balance the
respondent group in terms of urban-rural origin. Half of the respondents within
each gender group grew up in towns with populations of 10,000 or less, while the
other half came from areas with populations of greater than 10,000 residents.
While some research has shown urban v. rural location to be an important
sociolinguistic variable, the pilot study conducted for this research showed no
significant effects from residence. Thus this variable was not included in the

present study as an independent variable to be investigated.

2.3 Data Analysis

As data in this study were both qualitative and quantitative, a variety of
tools were incorporated in the data analysis. All interviews were transcribed by
the researcher. Analysis of folldinguistic definitions of complaints, as well as
collection of terms to be used in the semantic-differential portion of the
questionnaire, was done using HyperResearch, a qualitative analysis software
program. HyperResearch allows the researcher to develop a coding system and
then code a large number of transcripts using this system. Reports can then be

generated to locate occurrences of specific codes with the corpus. Use of this

28

software allowed a more sophisticated and systematic analysis of themes within
the qualitative data.

Data from the discourse completion’portion of the questionnaire were
coded first according to whether or not the response could be considered a direct
complaint, an indirect complaint or a choice to opt out of complaining.8 Those
responses that contained a direct complaint were then coded for use of mitigators.
For indirect complaints, an analysis of the kinds of speech acts used (c. g. request,
compliment, offer, etc.) was done.

Chi-square tests of independence were run on this portion of the data to
determine the effects on choice of response contributed by the four independent
variables in the study: speaker sex, hearer sex, power and social distance. In
addition, a correlation between ratings of severity of offense, responsibility of the
offender and likelihood of correction and directness levels were explored, also
through chi-square analysis.

For part two of the questionnaire data, responses to the questions asked
about each complaint realization were tabulated and compared. Factor analysis
was run on the semantic-differential ratings to identify the major factors involved
in the interpretation of complaints. Mean scores of each situation for each of the

resulting factor groups were then calculated.

 

8 Opting out included both instances in which the respondent indicated that he or she would say
nothing and cases in which the respondent wrote something that could clearly not be considered a
complaint, e.g. “Hi, how are you doing today?”

29

2.4 Expected Conclusions

A number of hypotheses were examined, based on a pilot study of
complaint realizations (Hartley 1996) as well as other studies of politeness
phenomena, particularly Brown and Levinson (1987), found in the sociolinguistics
and pragmatics literature. These hypotheses, together with their theoretical bases,

are presented below.

HYPOTHESIS 1: Respondents will draw on a variety of politeness
strategies in performing complaints, including the choice of opting out of
performing the FTA, the use of mitigators (both syntactic and lexical), and
the use of indirectness, to accomplish face management. This hypothesis is
based both on the pilot study (Hartley 1996) and Brown and Levinson
(1987)

HYPOTHESIS 2: Situations where the participants are in an on-going
relationship (i.e., -D relationships) will result in more face management
than those where the encounter is brief and passing. This will manifest
itself in the use of more politeness strategies, such as mitigators and use of
indirectness. Similarly, those situations which incorporate a power
differential (+P) will also require greater attention to face management and
the corresponding use of politeness strategies. Considering these factors in
combination, the greatest amount of politeness should be used for the
scenario which includes the boss (-D, +P), while the stranger cutting in line
scenario (+D, -P) should produce the greatest amount of directness and the
least amount of mitigation. This hypothesis is based primarily on the work
of Brown and Levinson (1987).

HYPOTHESIS 3: Gender may influence the results, both in terms of how
complaints are made by men v. women and in terms of how complaints are
made to men v. women. Brown and Levinson (1987) point out that many
of the findings in the sociolinguistic literature which support the idea that
men and women use different “genderlects” may be the result of other
factors, such as power differentials and social networks. Nevertheless,
there is enough empirical evidence within the sociolinguistic literature to
suggest that gender is an important social variable to consider in studies of
language use.

30

In addition to these specific hypotheses, which were examined primarily
through quantitative results in this study, qualitative methods were used to explore
the folldinguistic notions of what constitutes a “complaint” and how various
elements within a situation (participants, setting, etc.) interact to produce

circumstances where different kinds of complaints are seen to be appropriate.

31

Chapter 3

THE FOLKLINGUISTIC S OF COMPLAINING

3.1 Folklinguistic Categories of Complaining

In the last chapter, a working definition of a complaint as any negative
statement made about a situation that one finds unsatisfactory was given.
Folldinguistic studies of a variety of language phenomena, however, reveal that
linguists’ beliefs about language do not always coincide with popular ideas and
attitudes. Because of this, a good place to begin an analysis of what constitutes a
complaint is to compare the technical definition that forms the basis of this study

with how respondents themselves define a complaint.

Table 1. Aspects of complaining in folklinguistic definitions

 

 

 

 

 

 

Feature # of occurrences
Being unhappy, upset, angry 19 Dr 23
Having a problem with something 9 J
Voicing opinion. concern. feelings 15 \
Expressing disagreement 13

Saying you don't like something 13 56

 

Being or saying something negative

 

 

 

Trying to resolve or fix something

 

9
Expressing dissatisfaction 6 J

6

2

 

 

 

Yelling at someone

 

In the interviews, the first question that respondents were presented with

was "Give me your definition of a complaint." Using the HyperResearch softward

32

program, an analysis of responses given to this question revealed ten aspects of
complaining, which can be further reduced to two obligatory and two optional
components.1 These aspects, along with the number of times they occurred in the
interviews, are presented in Table 1.

The most frequently mentioned aspect of complaining was someone being
unhappy, upset or angry with a situation or person. For example, one respondent
said, "Generally it's just when someone's upset about something" and another
offered, "A complaint would be if someone's very unhappy." An additional nine
respondents said that a complaint occurs when someone has a problem with
something. Examples of statement along this line include: "It's basically just when
someone has a problem with something" and "A complaint is something that a
customer or any kind of consumer has a problem with, the product they're buying
or the services they're receiving." These two aspect correspond to the precondition
of a complaint given in the technical definition (i.e. that a speaker finds a situation
to be unsatisfactory in some way).

The next five categories listed in Table 1 are all variants of the technical
definition of a complaint as the negative evaluation of a particular situation.
Fifteen people expressed this in terms of voicing some concern or telling your

feelings to another person, as in the following statement: "A complaint is um

 

l The aspects presented here are drawn from the actual wordings of the responses given. For this
reason, several categories are fairly overlapping in their meanings and could be collapsed
together. Also, it should be noted that a single response could be categorized into more than one
aspect.

33

when you voice your opinion with something that you don't agree with." Another
respondent explained it as: "When someone is not happy about something and
they tell their feelings to another person."

In thirteen cases each, respondents claimed that complaining involved
expressing disagreement or saying that you don't like something. Examples of
disagreement include: "A complaint is someone who doesn't agree with what's
happening in his or her life" and "Someone complains when they're not in
agreement with what's going on." In the "don't like" category, respondents offered
definitions such as "A complaint is when somebody tells another person about
something they don't like" and "Someone who's bitching about something that
they don't like."

The next most frequent category was being or saying something negative.
This wording was used nine times. Examples include: "when people are being
verbally negative" and "expressing detrimental views or opinions about a situation
or person." Expressing dissatisfaction with a situation was the final way in which
the technical definition of a complaint appeared in the folklinguistic data. For
example, "A complaint is when a person is unsatisfied with a certain someone or a
certain thing."

In addition to the preconditions for a complaint and the actual definition of
a complaint, six responses also incorporated the intended outcome of complaining.

This was expressed in terms of trying to resolve or fix something. For example,

34

one respondent said: "A complaint would be something that I have a problem with
that I need to talk to someone about and get fixed."

Finally, two respondents thought that yelling at someone should also be
included in the definition of complaining, although the data in the following
chapter shows that this is clearly not obligatory. Their responses are as follows:
"when someone does something to make you angry and you might yell or scream
at them because they did that to make you angry" and "when somebody's done
something wrong you have the right to yell at them or tell them that they did
something wrong."

This examination of the folk definitions of complaining provides support
for the technical definition of a complaint that forms the basis of this study. It
seems that there is general agreement that for something to be a complaint, the
individual making the utterance must be unhappy or dissatisfied with a particular
situation (which is precisely the precondition for a complaint in the technical
definition). They then choose to voice their feelings or opinions about the matter,
i.e. they produce a negative evaluation of the situation. Sometimes this is done
with an eye towards fixing the situation, although this does not appear to be an
obligatory aspect of complaining. In other words, sometimes complaints are made
for the sole purpose of venting frustration or anger, and this may or may not be
accomplished through yelling at someone.

With a basic folk definition of a complaint established, it is also interesting

to examine the various labels for complaining that exist in the folk wisdom. This

35

Table 2. Folk labels for complaints

 

Term/Phrase

Total #
suggesting
term (n=73)

# of females
suggesting
term (n=38)

# of males
suggesting
term (n=35)

 

Bitching

48

26

22

 

Whining

N
N

13

 

Nagging

 

Griping

 

Arguing

 

Saying something/being negative

 

Moaning

4
1
2
2
4

 

Chewing out

 

Disagreeing

—

 

Voicing opinion/concem

 

Annoying

 

Saying you don’t like something

 

Talking about/discussing the situation

 

Being unhappy

NNNt—NNNONNMUIC

 

Criticizing

h

 

Sharing upsetness/being upset

 

Crying

NNNNNNwwwwh-bhao

 

Throwing a fit

 

Letting off steam

u—nr—ioo

 

Request for some diversity in what
they’re doing

p—n

OOONNt—OOON—

 

Ventilating

 

Trippin’

 

Hassling

 

 

 

Being disruptive

 

Pissin'

 

 

Bothering

 

 

OOOOOO

 

 

36

 

was done by asking the respondents the question: "What else can you call it when
a person complains?" In response to this question, 26 different terms or phrases
were offered. These are presented in Table 2.

By far the most common category of complaining for these university
students is “bitching”, with 48 respondents offering this term. This is followed by
“whining”, given by 22 people. “Nagging” and “griping” come next, with nine
and six respondents respectively suggesting these terms. There were a few gender
effects in this portion of the data, although given the relatively small numbers here
it is unlikely that these differences are significant. “Whining” is a term that more
males than females use, with 13 males (or 37%) offering the term while only nine
females (24%) did. “Moaning” is exclusively a male term, with all four instances
of this item being given by men. In contrast, females tend to use the term
“griping,” with five women but only one man offering this word.

Having established the labels which students use to describe complaints,
the question arises whether these labels represent sub-categories of complaining,
and, if so, how they are defined and differ from each other. In elaborating on the
various labels, aspects such as attitude,2 degree, duration, effect, reason,

specificity, emotion involved and language used all helped to differentiate among

2 . . . . .

I am usrng the term "attitude" here as it IS commonly understood in phrases such as "he has an
attitude problem," rather than the technical linguistic uses associated with either "propositional
attitudes" in Semantics or "language attitudes" in Social Psychology of Language research.

37

Table 3. Definitions for selected subcategories of complaints

 

 

 

 

 

 

 

 

 

‘ “Bitching”

Attitude: “bitching is getting an attitude”

Degree: “bitching is venting your anger in a severe manner”

Eflect: “when you're bitching at someone you're just kind of babbling telling
them you really don't appreciate what they're doing and they're not I don't
really care if they learn anything from it”

Emotion: “bitching is I think um when you're really really really upset about
something that's really really making you mad”

Language: “1 would say that bitching out is far more - like there would be some
swear words in that situation”

Reason: “you're just getting things off your chest”

Specificity: “bitching has a tendency to be more personal, more directed towards a
person instead of - such as their bad habits etcetera”

“Whining”

Duration: "whining 1S uh continually gorng on about something

Eflect: “whining's more like babyish, and annoying and makes me just like think
of obnoxious people”

Reason: “whining's kind of like uh when you're unhappy for silly reasons and you
just whining for little reasons”

Language: “whining you've got these [whining sound] (laugh) just high pitch almost
like a cry”

“Nagging”

Degree: “1 would compare nagging to so- somelike like, if you're complaining
about how your mom won't let you go outside and play and you're like
"come on mom let me play" and bitching is more on the serious note
where like - something like "screw you mom you never let me go out
with my friends" something more serious than nagging”

Duration: “nagging would be just like keep on saying it”

Effect: “1 think nagging is more or less just like wanting your way and just you
know nagging until you get your way”

“Griping”

Degree: “1 also think maybe bitching might be a little harsher than griping”

Reason: “griping to someone is more along the lines of if you have a problem
with what they're doing, like you have something to gripe about it's not
it's not something that you don't have control over”

Language: “griping I guess is kind of like you're just - 1 don't know in your regular

voice”

 

38

 

Table 4. Situations and responses judged in phase two

 

Situation 1: A person returns from running errands to find their visiting friend has
messed up the newly-cleaned kitchen. They say:

 

Response 1.1: “I spent all day cleaning the kitchen. I would appreciate it if you
would pick up your mess.”

Response 1.2: “Boy, someone was in the cooking mood.”
Response 1. 3: “What the hell did you do to the kitchen?”

 

Situation 2: A stranger cuts in line ahead of someone who has been waiting two hours to
buy concert tickets. The person says:

 

Response 2.]: “Excuse me, but the end of the line is back there. We’ve been waiting
all day.”

Response 2.2: “1 think people cutting in line is really rude.”
Response 2. 3: “Who do you think you are?”

 

Situation 3: A person has just received a bad performance review from their boss with
whom they have worked awhile and get along well. They go to see the boss and say:

 

Response 3.1: “Why did you give me a review like that. I really don’t think that was
fair. Why didn’t you talk to me - instead of giving me a bad review. I
though you would be cooler than that.”

Response 3. 2: “I would like to know why you gave me a bad review. I feel that I
have done more than enough to make this company happy.”

Response 3. 3: “I don’t understand why I received a bad review I wish that I had been
told that you thought my performance was ba

 

Situation 4: A student finds out the advice their adviser gave them regarding classes was
wrong and they will now have to go to summer school. They go to see their professor

 

and say:

Response 4.1: “Is there any way I can get out of this requirement since you gave me
some incorrect information?”

Response 4. 2: “What am I supposed to do now that I’ve wasted my time on
unnecessary courses?”

Response 4. 3: “How could you give me the wrong advice? Aren’t you supposed to

 

know this stuff?”

 

 

39

the terms. Table 3 includes selected quotes that exemplify some these aspects for
each of the top four categories.

In general, bitching seems to be thought of as the harshest term and is
accompanied by an attitude. It’s also associated with anger or intense emotion.
Whining and nagging are both complaints that are stretched out over long periods
of time. The difference is that nagging appears to accomplish something whereas
whining might just be done to hear oneself complain. Griping appears to be used
when there are specific issues at hand, but these issues are not of major
importance.

To get a better understanding of the nuances among the various
subcategories of complaints suggested by the respondents, additional data were
collected in the second phase of the research. As described in Chapter 2, this
phase of data collection included the presentation of 12 utterances (three in each of
four situations) to respondents who were then asked to make certain judgments
about them. The statements that were presented to the respondents are given in
Table 4.

The first question that respondents were asked about each utterance was
whether or not they considered it to be a complaint. Table 5 presents the results of

this question.

40

Table 5. Results from: "Would you consider this statement to be a complaint?"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Response Number of respondents Percent answering "Yes"
answer "Yes" (n=120)

1.1 68 56 67%

1.2 23 minim

1.3 105 87 50%

2.1 59 49:7

2.2 98 81.67%

2.3 86 71.67%

3.1 107 89 17%

3.2 74 61. 67%

3.3 50 ; -' * ‘31::

4.1 44 E2§s§s§ 15:51:51: . .. . 513355;

4.2 118 98. 33%

4.3 . 95 79.17%

 

 

 

 

In each of the situations, there was one response that the majority of
respondents did not consider a complaint. These are responses 1.2: “Boy,
somebody was in the cooking mood”; 2.1: “Excuse me, but the end of the line is
back there. We’ve been waiting all day”; 3.3: “I don’t understand why you gave
me a bad review. I wish that I had been told that my performance was bad.”; and
4.1: “Is there any way I can get out of this requirement since you gave me some
incorrect information?” In each of these situations, the responses in question are
clearly the most indirect examples of complaining.3 It may be the case that if
these responses had been presented in isolation, they would have been judged to
be complaints by a larger percentage of respondents. In comparison with much

more direct complaints, however, respondents may have chosen to classify them

 

3 A more detailed discussion of indirectness in complaint realizations is taken up in Chapter 4.

41

as non-complaints, particularly since the choice of categorizing them as "indirect"
complaints was not offered to the respondents.

For those respondents who did consider each statement to be a complaint,
the next task they were asked to do was to categorize the complaint by selecting
one of five labels, which were chosen because they appeared frequently in the
interview data. These categories were: "Whining", "Bitching", "Nagging",
"Griping" and "Chewing out." In addition, respondents were given the option of
selecting "Other" and filling in a term of their choosing. The results from this
question are presented in Figure 2.

For all of the responses, the categorizations were divided between two or
more categories. This is likely because the categories themselves are overlapping
in many ways. In addition, part of the definition of each of these kinds of
complaining has to do with the context. So, for example, what rrright be
considered a “bitch” if said to a friend might sound more like a “whine” if said to
a boss. In any case the best examples of each of the categories seem to be as
follows:

“Whine” — 3.1 “Why did you give me a review like that. I really don’t

think that was fair. Why didn’t you talk to me - instead of giving me a bad

review. I though you would be cooler than that.” (said to boss)

“Bitch” — 1.3 “What the hell did you do to the kitchen?” (said to friend) —
this is almost equally considered to be “chewing out”

“Gripe” — 1.2 “Boy, someone was in the cooking mood.” (said to friend)

“Chew out” — 2.3 “Who do you think you are?” (said to stranger)

42

 

% selecting each category
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%

1 l I I

 

 

 

 

      

      
 

U

 

liti'flwtng:

 

“HHHHHUHU”HHHHHHHHHHHHHHHHUHHHHHHUHHHHHHHHHHHHHHHHHHHHHHHHHHUHHLHJHURHHHHHHHVHHHHHI HHMHHIHWVAKNAIH

 

 

 

 

 

 

     
   
   
 

“flfdffildfkh'H/Hfljfihfliigii

fififiKfiflfifififlfififififiRfi
wHm“NHNHMHWHWHWNNHWHNHMHWUMHMNWHMHWHMMmNW“MNmHmHMHHHMHMHNHWHNHWHNHWHWHWwLIHTWHNHHIlit”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ImmmmmmmmmmmMmmmmmmmmmmmmmmm

 

 

’/////// ///
lmmmmmmmmmmmmmmmmmmmmmmmummmmmmmmmnmmmmmmmmmmmmmmmmmmmmmmmmm.

1

EWhine IBitch INag groiipe nChew out .Othet'

   
 

 

 

 

 

 

 

 

 

 

Figure 2. Complaint categorization for each statement

43

There doesn’t seem to be a very good example of nagging in this data, according
to the respondents, although response 1.2 (“Boy someone was in the cooking
mood”) received the highest percentage for this category.

There are several interesting points to note in regards to the category of
"other." First, the four responses which received the lowest percentages of people
calling them a complaint (see Table 5: 1.2 -- 19.17%; 2.1 -- 49.17%; 3.3 --
41.67%; 4.1 -- 36.67%) also received the highest percentages of "other
categorization" among those who did classify them as complaints (see Figure 2:
1.2 -- 30.43%; 2.1 -- 6.78%; 3.3 -- 20.00%; 4.1 -- 11.36%). In other words, ofthe
ahnost 20% of respondents who considered response 1.2 to be a complaint for
example, almost 1/3 of those individuals were not happy with the categories
presented to them for complaint classification and instead opted to create their
own category for this response.

In fact, the "other" classification of response 1.2 sheds some light on why
such a low percentage of respondents chose to call this a complaint in the first
place. Five of the seven individuals who placed this response in the "other"
category called it "sarcastic." It thus seems that the indirectness utilized in this
complaint comes across as sarcasm rather than a more benign and less face-
threatening hint, even when intonation that might help convey a sarcastic meaning
is lacking in the stimulus data. It may be the case that many of the more than 80%
of respondents who said that this response was not a complaint nonetheless

considered it a sarcastic comment. In fact four individuals who circled "No" to the

44

question "Do you consider this statement to be a complaint" wrote on their
questionnaires that they would classify the statement as sarcasm. It thus seems
then that respondents want to keep separate the illocutionary act of "complaining"
and rhetorical device of "sarcasm."4 Why this is would need to be explored more
thoroughly in future research.

Responses 2.1, 3.3 and 4.1, in contrast to 1.2, seem to derive their low
percentages of classification as complaints from their subtlety or tact. Terms such
as "questioning," "legitimate," "subtle" and "stating the facts" are applied to these
responses, both by those who classified these statements as "other" complaints and

those who said they were not complaints at all.

3.2 Intention and Interpretation in Complaining

In the performance of a face-threatening act, an individual generally has a
number of options in terms of directness, as illustrated in Brown & Levinson's
(1987) decision making tree of possible strategies for performing FTAs (see
Figure 1). Using more indirect strategies, such as going off-record, has the
advantage of allowing the speaker to avoid responsibility for damaging the other
participant's face, while still saying something that will hopefully get his or her
point across. The primary disadvantage, however, in going off-record is that the

point may in fact be missed or ignored altogether by the bearer. Because of this, it

 

4 Although note the discussion in Chapter 4 regarding the use of sarcasm as an indicator of direct
complaints.

45

is possible for any given FTA for there to be a mismatch between the speaker's
intention (i.e. to perform the FTA) and the hearer's interpretation (i.e. whether or
not he or she understood the statement to be the intended FTA).

In order to explore the relationship between speaker intention and hearer
interpretation in complaint realizations, the respondents were also asked to judge
whether they thought that each of the 12 statements discussed in the last section
were intended to be complaints and would be interpreted as complaints. The
results of these questions, combined with the previous question of whether or not

the respondent thought each statement was a complaint, are presented in Figure 3.

 

100% ~
90%
80%
70%
60%
50%
40%
30%
20%
l 0%

0%

% responding "yes"

    

1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3 4.1 4.2 4.3

Response

 

 

 

Is a Complaint I Intended as Complaint E Interpreted as ComplaM

 

 

Figure 3. Is v. Intended to be. v. Interpreted as complaint

46

Paired-sample t-tests reveal that there is a statistically significant (at the .01
level or greater) difference between the ratings of intention and interpretation for

seven of the twelve responses. Table 6 shows the t values obtained.

Table 6. T-values for intention v. interpretation ratings

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Response to... d.f. p

1.1 4.674 119 .000
1.2 -3.345 119 .001
1.3 0.377 119 .707
2.1 3.431 119 .001
2.2 1.420 119 .158
2.3 2.620 119 .010
3.1 3.513 119 .001
3.2 5.596 119 .000
3.3 3.431 119 .001
4.1 0.217 119 .828
4.2 1.000 119 .319
4.3 0.831 119 .408

 

 

 

 

 

Of the seven significantly different responses, it is only response 1.2 (“Boy,
someone was in the cooking mood”) that respondents thought was more likely to
be intended as a complaint than interpreted as one. This is interesting in light of
the previous point regarding the distinction between sarcasm and complaining.

For the remaining six responses, respondents thought that the statement is more
likely to be interpreted as a complaint than intended as one.

This result points to the importance of the bearer in the negotiation of face-
management. It appears that the nature of Potential Complaint Situations is such
that almost any statement made could be interpreted as a complaint, even if it was

not the speaker's intention that it be one. Perhaps this is due to the hearer's own

47

 

recognition that in these situations he or she has violated some social norm or
otherwise threatened the speaker's face (whether intentionally or unintentionally)
and as a result is expecting some kind of negative response from the speaker.

Thus any response can be taken to be a negative evaluation, i.e. a complaint.

3.3 Factors in the Interpretation of Complaints

If hearer interpretation plays such an important role in the overall face-
management of Potential Complaint Situations, then it is necessary to understand
what factors go into the process of interpreting statements to be complaints. The
results of the semantic-differential rating task shed some light on this issue.

As mentioned in chapter 2, fi'om the initial interview data in which
respondents offered adjectives to describe certain complaints, 19 word pairs were
chosen to be included in the second round of data collection. The terms were
selected based on the frequency with which they were mentioned in the interviews
in combination with the breadth of meaning that they provided for a description of

complaints. These word pairs were:

Rude - Polite Effective - Ineffective
Mean - Nice Negative - Positive
Legitimate - Not Legitimate Not Sarcastic - Sarcastic
Diplomatic - Not Diplomatic Defensive - Not Defensive
Indirect - Direct Arrogant - Not Arrogant
Unreasonable - Reasonable Not Assertive - Assertive
Aggressive - Non-aggressive Patient - Impatient
Appropriate - Inappropriate Mature - Immature

Angry - Not Angry Passive - Active

Confrontational - Non-confrontational

48

 

Respondents were asked to rate on a five point scale each of the responses
discussed above in terms of these word pairs. Word pairs were presented in the
order shown above, so that not all positive terms would be on one side or the
other. Polarities were then reversed on a number of the word pairs for analysis, so
that the ratings went from negative (=1) to positive (=5).

Factor analysis on the data set revealed that the 19 word pairs could be
reduced to three major factor groups, accounting for almost 60% of the total
variance. Table 7 presents the factors in each of the three groups, along with their
weightings. .30 was selected as the cut-off point for inclusion in a factor group;
terms that received a weighting slightly less than this value can be considered
marginal terms in the factor group (and are thus indicated with parentheses in

Table 7).

Table 7. Factor Groups (with weightings)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fl “Diplomacy” £22 “Boldness” #_3 “Manner”

0.86 Reasonable -0.81 Assertive [=Not Assertive] 0.83 Nice

0.83 Appropriate -0.79 Active [=Passive] 0.80 Polite

0.81 Legitimate -0.73 Direct [=Not Direct] 0.76 Not Angry

0.56 Diplomatic 0.59 Non—confrontational 0.72 Patient

0.55 Mature 0.58 Non-aggressive 0.7 2 Positive

0.41 Effective 0.30 Not Angry 0.57 Mature

0.41 Polite (0.25 Not Defensive) 0.55 Non-aggressive

0.37 Positive (-0.25 Effective]=lneffective]) 0.54 Not Arrogant

0.35 Nice 0.46 Non-confrontational

(0.29 Not Arrogant) 0.42 Diplomatic
0.39 Not Defensive
0.30 Appropriate
(0.28 Not Sarcastic)

 

 

 

 

 

 

 

49

The first factor group is labeled “Diplomacy”, because these are factors
which relate to the legitimacy of the complaint, the way the complaint is handled,
and the outcome of the complaint. The second factor group has been labeled
“Boldness,” because it appears to deal with how willing the person is to stand up
for him or herself. This factor group is framed in the negative; i.e. the respondents
seem more concerned about a complaint being too “wimpy” than overly bold or
aggressive. The final factor group has to do with the “Manner” in which a
complaint is expressed. According to the respondents, there should be an
appropriate amount of politeness and calm on the part of the complainer, with a
certain level of maturity and lack of sarcasm.

Table 8 shows the mean scores of each of the factor groups for each of the
responses that were rated. Response 3.3 “I don’t understand why I received a bad
review. I wish that I had been told that you thought my performance was bad” is
rated the best for all three factor groups. Apparently, if one is going to complain
to an employer, this response can be used as a model for how best to do it,

according to these respondents.

Table 8. Means scores for each response by factor group

 

 

 

 

 

 

 

 

 

 

 

Response FGI FG2 FG3 Response FGI FG2 FG3 I
1.1 3.73 3.38 3.23 3.1 2.64 2.93 2.49
1.2 3.07 3.39 3.2 3'27 ' 5.1.1.-,2.8.5.l
“3 . 2-90 1-89 3-3 3.193.333.2313:
2.1 4.03 3.42 3.38 4.1 3 94 3 40 3 53
2,2 3.12 l 2.74 l 2.87 4.2 2.59 2.76 2.11
2.3 2.45 2.83 | 1.83 4.3 2.57 2.86 2.16

 

 

 

50

On the other end of the spectrum, three different responses received the
worst rating in each factor group. In terms of diplomacy, “What the hell did you
do to the kitchen?” is not very effective, perhaps due to the presence of an
obscenity (mild though it may be for some of these respondents' vocabularies).

The indirect “I think people cutting in line is really rude” appears to be the
big winner in terms of wirnpiness. And “Who do you think you are?” is rated the
least polite in terms of the manner in which the complaint is expressed. This is
likely due to the fact that the complaint deviates from the actual issue at hand and
personally attacks the offender.

Figure 4 presents these scores in a chart; the responses pattern in several
groups. At the bottom, there are five responses that are seen as relatively bold, but
not very diplomatic and delivered in a fairly poor manner. These are responses 1.3
“What the hell did you do to the kitchen”; 2.3 “Who do you think you are?”; 3.1
“Why did you give me a review like that. I really don’t think that was fair. . .”; 4.2
“What am I supposed to do now that I’ve wasted my time on unnecessary
courses”; and 4.3 “How could you give me the wrong advice? Aren’t you
supposed to know this stuff?”

At the top, there are four responses that are seen as being very diplomatic,
with their boldness and manner ratings similar to each other, although less than the
diplomacy ratings. The responses that pattern this way are 1.1 “I spent all day
cleaning the kitchen. I would appreciate it if you would pick up your mess”; 2.1

“Excuse me but the end of the line is back there. We’ve been waiting all day”;

51

 

3.3
2.1

4.1

 

3.8
1.1

1.2

 

3.3

 

 

2.8 i

Mean Scores

 

 

 

2.3 —

 

 

 

l .8
Diplomacy Boldness Manner

Factor Groups

 

l "Ar-1.1 ------- I ------- 1.2 --a--1.3 ---><---2.1|
l +2.2 --0- 2.3 -----+ ------ 3.1 ”.0...” l
l —e—3.3 ~~o~——4._i_ -~E3---- 4.2 +4.33

 

 

 

 

 

Figure 4. Mean scores for each response by factor group

52

3.3 “I don’t understand why I received a bad review. I wish that I had been told
that you thought my performance was bad”; and 4.1 “Is there any way I can get
out of this requirement since you gave me some incorrect information.”

This leaves the three responses that fall in the middle of the graph. 1.2
“Boy someone was in the cooking mood” and 2.2 “I think people cutting in line is
really rude” are judged to be fairly equal in their respective diplomacy and manner
ratings, but they are substantially lower in their boldness ratings. This makes
sense since these are the two responses that utilize the greatest amount of
indirectness. They avoid direct reference to the offender as the agent of the
offensive act and do not ask for a specific remedy to the situation.

The last response is 3.2 “I would like to know why you gave me a bad
review. I feel that I have done more than enough to make this company happy.”
This response is seen as middle of the road in terms of diplomacy and boldness,
but it slips some in its ratings of manner. In the interview data, when respondents
were asked to describe this complaint and the person who made it, many people
made references to the person being a bit arrogant or too confident in his or her
own abilities. For example, one respondent said “this complaint he’s kind of like
bragging about his accomplishments like he thinks he’s so good for the company
and and I think he’s kind of big-headed,” and another said “I think that person
kind of thinks too highly of themselves um and I don’t think that would be a good

way to approach a boss someone that’s ahead of you and can like fire you.”

53

To summarize, the results of this investigation of folklinguistic ideas of
US. university students about complaints reveal the following:

( 1) American university students use quite a few categories to describe
different kinds of complaints, and these different categories are evoked according
to factors such as who the participants in the complaint situation are, what the
setting is, what is being complained about and such particular characteristics as
degree of intensity and emotion involved in the complaint realization. The most
common kind of complaint for these respondents is “bitching,” with “whining”
being the next most frequent.

(2) There is no definitive idea about what is and what isn’t a complaint.
There appears to be a wide range of directness and indirectness that can be
employed in complaint situations. Presenting respondents with a particular
response and asking them to choose “yes” or “no” as to whether the response is a
complaint thus leads to breakdowns such as 40% of respondents saying it is a
complaint while 60% say it isn’t (or 70-30, 55-45 etc.) A future study might be
able to investigate the question of directness more specifically by asking
respondents to rate a response on a scale as to whether it definitely is or is not a
complaint.

(3) There is often some discrepancy between the intention and
interpretation of complaints. The respondents recognized the possibility of a
statement that was not intended to be a complaint being interpreted as one, and

vice versa. Again, this relates to the range of directness available in the

54

construction of complaints and the subsequent “wiggle room” or benefit of the
doubt that hearers can choose to give or not to give. Interestingly, it appears that
responses that are not intended to be complaints may be interpreted as them more
often than the other way around. This has implications for the degree to which the
hearer is part of the negotiation of meaning within the situation.

(4) For university students, there are three majors factors to consider when
judging the politeness and effectiveness of a complaint. These factors are
diplomacy, boldness (or lack of “wimpiness”) and manner.

This discussion of folldinguistic categories and ratings of complaints
provides a starting point from which to proceed to a larger investigation of the
linguistic details of actual complaint realizations, and that is the task that is taken

up in the following chapter.

55

Chapter 4

DIRECTNESS AND INDIRECTNESS IN COMPLAINT REALIZATIONS

4.1 F ace-Management Choices in Potential Complaint Situations

In Chapter 2, the working definition of a complaint being utilized in this
study was presented. It was noted that the only precondition for a complaint is
that the speaker finds the situation in question unfavorable in some way. When
this occurs, the situation can be considered a "Potential Complaint Situation" or
PCS. Within a PCS, a speaker has a choice as to how direct or indirect he or she
will be in making a complaint.

In the current study, the data from the discourse completion portion of the
questionnaire were analyzed in terms of how direct a complaint 3 speaker chose to
make. Each response was coded as containing: (1) no complaint; (2) an indirect
complaint; (3) a direct, but mitigated complaint; or (4) a direct complaint. In the
few cases where a response contained two distinct complaints that could be coded
at separate directness levels, the more direct code was assigned. Table 9 gives the
total numbers and percentages of respondents who made a complaint at each
directness level for each situation.

While the degree of face-threat involved in performing a given speech act is

not the same thing as the level of directness, it is generally the case that the more

56

direct a speech act, the more face-threat it will produce.1 Thus a direct complaint
such as "You're a real slob" is taken to be more face-threatening than the mitigated

complaint "You're kind of a slob, aren't you?"

Table 9. Overall directness levels for complaint realizations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Situation Directness Level # of respondents % of respondent
choosing this level choosing this
(out of 120) level

1 No congrlaint 37 30.83%
(friend/ Indirect 49 40.83%
messy Mitigated 4 3.33%
kitchen) Direct 30 25.00%
2 No complaint 36 30.00%
(stranger/ Indirect 53 44. 17%
cutting Miti gated 2 l .67 %
in line) Direct 29 24.17%
3 No complaint 33 27.50%
(boss/ Indirect 60 50.00%
bad Mitigated 16 13.33%
review) Direct 1 1 9. 17%
4 No coglaint 19 15.83%
(adviser/ Indirect 6 l 50.83%
bad Mitigated 1 l 9. 17%
advice) Direct 29 24. 17%

 

 

In the sections that follow, each directness level will be analyzed separately
in terms of the kinds of complaints that appear in that category. In addition,
explanation as to how and why the respondents utilized that directness level for

each situation will be offered.

 

1 The converse, however, may not be true. See, for example, Blum-Kulka (1987) for empirical
evidence that the most indirect requests are not always judged to be the most polite.

57

4.2 The Choice of Opting Out

As Brown and Levinson (1987) note, the first choice an individual must
make in the performance of any face-threatening act is whether or not to do the
FTA at all. B&L note that the payoff for choosing to not do the FTA is that the
speaker avoids offending the bearer in any way. They also point out, however,
that the speaker also fails to achieve his or her desired communication goals by
taking this route. They conclude that since "there are naturally no interesting
linguistic reflexes of this last-ditch strategy, we will ignore it our discussion
henceforth" (p. 72).

While it may be the case that choosing to say nothing or to say something
other than a statement that can be interpreted as a face-threatening act may be
uninteresting from a linguistic point of view, it is not the case that these situations
are also uninteresting from a sociolinguistic point of view. That is to say,
examining the frequency with which certain individuals or groups choose to opt
out of performing a face-threatening act such as a complaint, as well as the reasons
that people choose to opt out of doing the FTA, may shed some interesting light on
the social evaluation of a given situation as well as the norms of behavior of
various sociological groups.

As previously described in chapter 2, respondents in the current study were
presented with four situations and asked what they would say in each situation. In
the instructions they were given for this task, the following statement was

included: "If you wouldn't say anything in the situation, write 'Nothing' on the

58

first line and then on the remaining lines give a brief explanation as to why you
would say nothing." Table 10 gives the frequencies and percentages of those who
chose to opt out of complaining in each situation.

As the table indicates, the choice of opting out of performing a complaint is
utilized quite frequently in the first three situations, with about half as many

individuals making this choice in the fourth situation.

Table 10. Frequency of opting out of performing any FTA

 

 

 

 

 

 

 

 

 

 

Situation # of respondents % of respondents
choosing to opt out choosing to opt out
(out of 120)
l (fiiend/kitchen) o
[-P, -D] 37 30.83 /o
2 (stranger/line) o
H), +0] 36 3000/0
3 (boss/review) - o
[+P, -D] 33 27.50 /o
4 (adviser/advice) 0
1+1), +D] 19 1583/0
TOTAL 125 26.04%
(out of 480)

 

Although not every respondent who chose to opt out of performing a
complaint provided a reason for their choice, some interesting trends can be
discovered from those who did. In the first situation, in which a visiting fiiend
messes up the kitchen, the primary reason given for avoiding a face-threatening
act was that it would risk some disruption in the established relationship. Consider

the following quotes that exemplify this point:

59

"I wouldn't want to hurt my friend's feelings."

"I wouldn't say anything because they were a good friend and more
than likely I had not seen them in along time."

"Sometimes you just have to grin and bear it -- especially with a
fiiend you don't see often. "

A reason given almost as often in this scenario was that the speaker either
expected the friend to clean up or at least believed that they would help if asked.
For example:

"I think that when I came home my friend would apologize and say

she would clean up the mess in a minute. I'd then start to help her

clean up the kitchen."

"I'm sure she would offer to help as I was cleaning."

Other respondents expressed the fact that they wouldn't want to make the
situation uncomfortable. For example, one respondent included the comment: "I
don't want to ruin the weekend over a messy kitchen." Others indicated that this
situation was "no big deal." A few even claimed some measure of responsibility
for the offense, with comments like "I told her to make herself at home, so I don't
expect anything from her in the way of cleaning. She is a guest."

In the second situation, in which a stranger is cutting in line, only two
respondents actually provided reasons for their choice to opt out of complaining,
and both agreed that they would not complain because someone else probably

would. One respondent phrased the reason as follows: "By that time I would be

too frustrated to say something and someone else would probably say something

anyway. "

6O

Table 10 indicates that in the third situation, in which the boss gives a bad
employee review, 33 respondents chose not to perform the face-threatening act of
complaining. Interestingly, only four of these individuals indicated this by
claiming they would say nothing, and only one of those individuals gave a reason
for their choice. This person indicated that "[t]his response depends of the type of
job. If it is my permanent career, I would probably say something, but none of my
other jobs have been that important yet. "

The remaining 29 respondents who were coded as opting out of the third
situation wrote something, but their statements were not particularly threatening to
the hearer's face. Most often they were a simple statement or question asking for
permission to speak with the boss about the review, such as "I was wondering if
we could discuss my annual review." It may be the case that in conversations
which followed this introductory remark the individual would have registered a
complaint. However, given the nature of the data collection tool used in this study
there is no way to know this. Still, it is interesting that given the unequal power
relationship in this scenario, many respondents felt the need to begin the encounter
by asking permission to speak -- a move that is clearly a face-management tool in
and of itself.

The final scenario was one in which an academic adviser has given the
student some bad advice that has led to the need for a summer school class. A

reason for opting out given by ahnost half of the respondents who made this

61

choice was that the speaker him/herself was actually responsible for this situation,
rather than the adviser. Consider the following quotes:

"It would be my fault for not having the adviser put in writing, that if

I take these classes, then 1 will graduate. Also, I should have

checked out the requirements myself and made my own schedule."

"It is my responsibility to know what classes I have to take. I

wouldn't need his help and it wasn't his fault the course wouldn't

count. I would just figure out one that would."

Several other respondents seemed to feel that with this situation, the
damage had been done and could not be repaired. For example, one respondent
indicated "It is too late by then and there's nothing you can do about it. . .The
adviser messed up, but she can't fix it, so it's not worth getting worked up over. "

Out of all the reasons provided for all the situations, only one respondent
indicated that the situation in question (which happened to be the messy kitchen
scenario) was unrealistic. This respondent wrote: "A high school fiiend of mine
would not do such a thing. Also I would not go to the store without them." The
fact that out of 480 possible responses only one included such a comment is a
good indication that the data collection tool used in this study, while having
certain limitations, did not suffer from the problem of placing respondents in
unrealistic situations and having them imagine what their response might be. In
fact, one respondent, in opting out of saying anything in the adviser situation,
commented: "Like everything else, I would probably just deal with it and move

on. (It's happened before)"

62

In examining the reasons that individuals chose to opt out of performing a
complaint in this study, it becomes evident that the choice of not performing an
FTA in a Potential Complaint Situation is a valid and effective option for the
management of the participants' face. In some situations, it appears that the
offense itself is not perceived to be severe enough to warrant the risk to both the
speaker's and hearer's face that the performance of a complaint would entail.2 In
other cases, while offense is in fact taken, factors such as maintaining a pre-
established relationship are seen as being more valuable than any relief of
frustration that complaining might provide. Finally, there are situations in which
speakers are frustrated or offended but feel that the situation is not likely to be
changed. In these cases, respondents sometimes also choose to opt out of

performing the FTA.

4.3 Direct Complaints

At the opposite end from opting out in Brown and Levinson's (1987)
decision making tree for performing face-threatening acts is the choice to go on-
record, with no redress. This is the choice to perform an FTA in the most
straightforward, direct manner. The biggest advantage of doing this, according to
B&L, is that the speaker accomplishes his or her communication goals in the most

efficient manner (p. 72). In addition, he or she avoids the potential for being

 

2 This possibility will be explored more in the next chapter.

63

misunderstood. Of course, performing a direct, bald F TA also brings with it the
greatest amount of risk to face of both the speaker and the bearer. It is thus only in
situations where the speaker feels that the face-threat is most needed or most

justified that direct, unmitigated F TAs are likely to be employed.

Table 11. Use of direct complaints by situation

 

 

 

 

 

 

 

Situation # of respondents % of respondents
using a direct using a direct
complaint complaint
(out of 120)
l (friend/kitchen) 0
H), -D] 30 25.00 /0
2 (stranger/line) o
H), +D] 29 2417/0
3 (boss/review) o
[+P,-D] 11 9.17/0
4 (adviser/advice) o
[ +1), +D] 29 2417/0
TOTAL 97 20.21%
(out of 480)

 

 

 

 

As Table 11 shows, respondents in this study chose to use direct complaints
almost equally within situations 1, 2 and 4 -- about a quarter of the time. For the
third situation with the boss, however, respondents complained directly in less
than 10% of the cases. A chi-square test of independence confirms that there is a
significant difference between the number of respondents who utilized direct
complaints in the third situation, as compared to the other three scenarios (x2 =

12.339, d.f. = 3, p < .01). These figures are easily interpretable in light of the

64

previous discussion of the risks inherent in direct FTAs, combined with an
understanding of the social dynamics involved in each situation.

In the scenario with the boss, there is both an unequal power relationship
(in which the speaker finds him or herself in a position of less power than the
hearer) and an on-going relationship that must be maintained. It is in this situation
then that face-management must be most skillfully employed. Uttering a direct
complaint thus appears to be too risky a strategy in the estimation of most
respondents.

If this is a valid explanation for the use of direct complaints, one would
expect the greatest number of direct complaints to show up in the situation that
finds participants in an equal power, but non-on-going relationship. This would be
the second situation, with the stranger cutting in line. Table 11, however, reveals
that this is not in fact the case. In raw figures, situation two has the same number
of direct complaints as situations four and one less than situation two. Chi-square
tests reveal no significant differences between these three scenarios in terms of
their use of direct complaints (x2 = .214, d.f. = 2, p > .80). It appears that where
power between participants is equal or social distance is large, then the likelihood
of a direct complaint increases. The combination of equal power and large social
distance, however, is not enough to free speakers from politeness constraints and
face-management concerns.

Having discussed the use of direct complaints within each of the situations,

it is important to understand what goes into the native speaker's intuitive

65

perception that a complaint has been registered. In chapter 3 we explored the
question of what factors assist in the assessment of a complaint; however, to this
point we have not moved beyond the simple definition of "the expression of a
negative evaluation of a given situation" to explore what precisely constitutes a
direct complaint. An analysis of the responses given in the discourse completion
portion of the data collection in this study contributes to a better understanding of
this issue.

The primary question related to what constitutes a complaint is in what
ways the "negative evaluation" of a situation manifests itself. In the data for this
study, six different reflexes of this negative evaluation emerged from an analysis
of the data. Table 12 presents these reflexes, along with the number of
occurrences within each situation.3

The most frequently utilized strategy for registering a complaint overall
was for the respondent to make a statement regarding his or her disapproval of or
disagreement with the hearer's action. An example of a complaint incorporating
this strategy in each situation follows:

1: "I can't believe you made a mess in my kitchen, after I cleaned it
for you. "

2: "Hey buddy! I've been standing here for two hours. . .no way can
you just waltz into line like that!"

 

3 As in other cases, more than one reflex ofien occurred within a given response. Thus, while a
total of 97 direct complaints were produced (out of 480 total responses), the figures in Table 10
add up to more than the total number of direct complaints that occurred.

66

3: "Listen, about my review, what is wrong with my work. I think
you're way off."

4: "That was very irresponsible telling me to take that course. You
get paid to do this and aren't doing it right. I think I'm going to talk
to your supervisor."

Table 12. Components of complaints in each situation

 

 

 

 

 

 

Situation Disapproval Obscenity/ Excla- Conse- Sarcasm Negative

of action Name- mation quences Feelings
calling

1 9 1 8 l 0 4 2

2 l l 7 12 0 0 l

3 8 0 1 0 2 0

4 l 1 7 2 l l 3 5

TOTAL 39 32 16 1 1 9 8

 

 

 

 

 

 

 

 

 

The next most frequently used strategy to perform a complaint was to
incorporate an obscenity or to directly attack the offender through name-calling.
This strategy did not occur in the third situation, but was used more than any other
strategy in the first situation. Phrases such as "What the hell did you do in here?"
were the most common, but other Obscenities also occurred. Examples include
"Thanks a fuckin' bunch" and "Holy shit!" Name-calling included terms like "pig"
and "jerk," as well as slightly more offensive designations such as "bitch" and
"dumbass."

Each of the situations also contained at least one example of a negatively-

loaded exclamation, such as "I don't think so!," "No way!," "Oh no!" or "Are you

67

kidding?!" These exclamations, either by themselves, or in combination with
other strategies also convey the negative evaluation necessary for a complaint.

One strategy that was used exclusively in the fourth situation in this data
was pointing out the negative consequences of the hearer's action. A good
example of this strategy is found in the following complaint: "I just wasted my
time and money on a useless class. I believe this is your fault because it was your
job to direct me." Another example of this strategy is the utterance: "I can't
believe that I have to stay at school because of your neglagence!4 If all this
information is printed why didn't you give me a copy to double check." Although
this strategy did not occur for the other situations in this data, it is easy to think of
complaints that could be made in this way. For example, in the messy kitchen
scenario, one might say: "All the time I spent cleaning this morning was for
nothing!" Similar kinds of statements could be imagined for the other situations as
well.

A fifth strategy for making a complaint is to utilize sarcasm. This is a more
difficult strategy to analyze, given the fact that the data in this study are primarily
written, since sarcasm often is conveyed through paralinguistic devices. Even

given that fact, however, there are a number of instances of complaining in this

 

4 Since the data in this portion of the study were written. spelling errors (such as the spelling here
of "negligence") occurred at times. Rather than correct the errors, the quotations are presented
here as they were actually written.

68

study that are clearly accomplished through the use of sarcasm. Several examples
from three of the four situations follow:
1: "Good Lord, did you grow up in a barn?"

1: "Don't you clean up after yourself? I'm not the maid around
here."

3: "Do you think you could have been a little harder?"

4: "Why don't you just stick a knife in me. I don't want to stay
another semester."

Finally, sometimes direct complaints are made through the speaker's
expression of some negative feelings he or she holds as a result of the action of the
bearer. Examples of this strategy once again occurred in three of the four
situations. For example:

1: "What's going on. I left this perfectly clean and you go and do
this. This is totally disrespectful."

2: "Excuse me mame but I have been waiting here in line for two
hours now and I don't appreciate you trying to cut in front of me."

4: "I am very upset with the job you have done as my adviser. I
really do not think I should have to take another class since it was
your mistake."

4: "I am sick of you people telling students to take classes that are
wrong."

A direct complaint can thus be said to have been made when any one of the
components described above occurs and conveys a negative evaluation of the

situation on the part of the speaker. Of course, the more that the components are

69

combined, the stronger the complaint may seem, although there is probably no

absolutely objective way to measure the "strength" of a complaint.

4.4 Mitigated Complaints

Mitigation is the process of employing certain linguistic devices or
strategies in an attempt to soften the negative impact of a face-threatening speech
act. Fraser (1980) points out that mitigation and politeness are not the same
phenomenon, although they often occur together. One can be relatively polite
while employing minimal mitigation, such as in a request like "Please sit down."
On the other hand, it is difficult to imagine a mitigated speech act that does not
also imply some attempt at politeness. Thus, Fraser concludes, mitigation entails
politeness, but not vice versa (p. 344).

While Fraser lists what he calls a "partial" and "preliminary" set of
mitigating devices, a more thorough list of the kinds of devices that can be
employed to mitigate the force of an F TA was utilized in the Cross-Cultural
Speech Act Realization Patterns (CC SARP) project. The results of much of the
work in this project, which focused primarily on the speech acts of requesting and
apologizing, appear in Blum-Kulka, House and Kasper (1989), and the appendix
of this volume includes the coding schema employed in the project. Within the
overall coding was the identification of what the researchers call "downgraders,"
which are lexical/phrasal or syntactic devices used to mitigate the face-threat of

the request.

70

While the CCSARP project and coding schema was devised primarily to
study requests and apologies, in an earlier article House and Kasper (1981)
suggest that a very similar list of "downgraders" can be applied equally to
complaints as to requests in order to achieve a desired level of politeness. The
current study thus began the analysis of mitigation in complaint realization using
the categories of downgrading utilized in the CCSARP project. In the current
complaint data, a number of those downgrading categories did not occur, while
several other categories of mitigation emerged. A discussion of these categories
follows shortly.

Although mitigation is an important option for accomplishing face-
management for many speech acts, it was actually not utilized with great
frequency when respondents made complaints in the current study. Table 13
presents the numbers and percentages of mitigated complaints that occur in the
data.

Mitigation occurred most often in the third situation, which suggests an
interesting correlation, since this is precisely the situation where the fewest
number of direct complaints occurred (see Table 11). It was employed slightly less
in situation four and very infrequently in situations one and two.

Chi-square tests of independence reveal that, overall, situation is a
significant factor in the use of mitigation. More specifically, there is significantly
less use of mitigation in situations one and two as compared with three and four

(x2 = 12.666, d.f. = 3, p < .01). There is no significant difference between

71

Table 13. Use of mitigated complaints by situation

 

 

 

 

 

 

 

Situation # of respondents % of respondents
using a mitigated using a mitigated
complaint complaint
(out of 120)
1 (friend/kitchen) o
H), -D] 4 3.33 A)
2 (stranger/line) o
H), +D] 2 1.67%)
3 (boss/review) o
[+P, -D] 16 1333/0
4 (adviser/advice) o
[+P, +D] 11 9.17%)
TOTAL 33 6.88%
(out of 480)

 

 

 

 

situations one and two or situations three and four. From these results, it appears
that mitigation seems to be largely a result of the power component, i.e. it is more
likely to be utilized in complaints when there is a power difference than when
participants are equal in terms of power.

Table 14 shows the kinds of mitigating devices found in the data.5 The
only mitigating device that occurred at least once in every situation was the
interrogative. This is a syntactic device that frames the complaint in the form of a
question, thus presenting the bearer with the illusion that he or she is being given a
choice to modify or at least explain his or her action. Examples of this device

from each situation are as follows:

 

5 Several of complaints utilized more than one mitigating device. Thus the total numbers here do
not match the total number of mitigated responses in Table 12.

72

1: "Would you mind having a little common courtesy and stacking
the dishes in the sink?"

2: "Why don't you have some consideration for those people who
have been here for hours and step to the back of the line?"

3: "Why did you give me a bad review? If you had a problem with
the way I worked why didn't you say something?"

4: "Why did you mis-advise me? Do you know what type of
situation you put me in?"

Table 14. Mitigating devices used in complaint realizations

 

 

 

 

 

 

 

 

 

 

 

 

 

Situation

MitgatiniDevices 1 2 3 4 TOTAL
Subjectivizer 0 0 10 2 12
Interrogative 1 l 2 7 11
Hedge 0 0 l 3 4
Lexical choice 2 0 l 0 3
Conditional 0 0 2 0 2
Tense 0 0 0 2 2
Cajoler 0 0 l 0 1
Hesitator 0 1 0 0 1
Agent avoider 0 0 l 0 1
Relationship marker 1 O 0 0 1
TOTAL 4 2 19 14 39

 

 

 

 

 

 

 

 

The most frequently employed mitigating device was the subjectivizer,
although this was only used in situations three and four. Subjectivizers occur

when the speaker expresses the fact that the statement is only his or her subjective

73

opinion, thus lowering his or her commitment to the propositional content that
follows. Phrases such as "I think" and "I guess" are often utilized in this manner
to downgrade the overall force of the speech act. Examples of subjectivizers in
the data include: "I believe that I deserve a better rating than what I received and I
was wondering why" and "I can 't help but think you're somewhat responsible for
wasting my time and money" (italics mine).

This last example also contains an example of a hedge (i.e. "somewhat"), a
lexical or phrasal addition that allows the speaker to avoid some degree of
preciseness and thus mitigates the overall force of the FTA.6 Another example of
a hedged complaint, which comes out of the third situation in the data, is "You
know, 1 am not real happy about the report you gave me. I was somewhat
confused as to the reasons I was evaluated this way" (italics mine).

Lexical choice is a kind of mitigating device wherein the complaint could
have clearly included a stronger or more face-threatening word, but the speaker
chose instead to use a less face-threatening lexical item. This is the case, for
example, when more mild words are substituted for Obscenities. In the first
situation, for example, we find two cases of "What the heck. . ." instead of the
more direct and face-threatening "What the hell. . . "

Conditionals can also be utilized as mitigating devices, as in the case of the

following complaint which comes out of the third situation: "Can I talk to you

¥

6 According to Fraser (1980), hedging is not the same thing as mitigation; however, some hedges
can be used to mitigate.

74

about the review you gave me? I would really wish if you would've said
something to me about all of this." In this instance, the speaker performs a kind of
odd mitigating hypercorrection by utilizing modals combined and a conditional
"if," the combination of which actually makes the sentence ungrammatical, but
clearly conveys the speaker's tentativeness in putting forth the complaint.

"Tense" permits a point-of-view operation in which a past tense is utilized
where one would expect the present tense. The effect of this switch is to
metaphorically remove the speaker from the complaint that is about to be put
forth. In the complaint "I just wanted to make you aware of the inconvenience
you've caused me involving both time and money," the use of past tense "wanted"
instead of present tense "want" accomplishes this effect.

The remaining categories occur only once each in this data, although a brief
explanation of them is still useful, since they clearly can be and are used to
mitigate complaints. A cajoler is an element that seeks to draw the bearer into a
kind of agreement with the speaker. The phrase "you know" functions this way in
the complaint "You know, I am not real happy about the report you gave me." A
hesitator is a pragmatic marker that indicates some hesitation on the part of the
speaker to put forth the complaint in the first place. In this data, an example is:

" Um, I don't think so. You can move."7

¥

7The relative infrequency of these two categories is likely a result of the fact that the data
collected were written rather than spoken, since pragmatic markers such as these generally occur
rather frequently in spoken data.

75

An agent avoider is a syntactic device by which the speaker removes him or
herself as the agent of the complaint or avoids nanring the bearer as the perpetrator
of the offensive act. In this study, an example of this mitigator occurs in the
complaint: "This review seems to be an unfair representation of my performance."
Finally, a relationship marker is an appeal to a previous or current relationship
between the participants. This acts as a complaint mitigator by conveying the idea
that the speaker doesn't view the offense as severely as he or she might were the
relationship not to exist. The complaint: "Just like old times, you are still messy!"
(italics mine) incorporates this mitigation device.

As shown through the discussion above, there is a wide range of options
available for the mitigation of complaints, although these tend not to be utilized in
this data as much as with other speech acts such as requesting. This may partially
be the result of collecting the data through written means, but it may also be the
case that complaints threaten face to such a greater extent than many other FTAs
that to truly reduce the impact of the utterance, something more than a mitigating
device is required. It is with this suggestion in mind that the discussion now turns

to the utilization of indirectness in making complaints.

76

4.5 Indirect Complaints

The most frequently utilized directness level in the data for all situations
was that of indirectness. As Table 15 shows, indirect complaints occurred in
nearly half of all responses given. A chi-square test of independence indicates no
significant difference between the four situations in the use of indirectness (x2 =

3.308, d.f. = 3, p >30).

Table 15. Use of indirect complaints by situation

 

 

 

 

 

 

 

 

Situation # of respondents % of respondents
using an indirect using an indirect
complaint complaint
(out of 120)
1 (fiiend/kitchen) 0
H), -D] 49 40.83 A)
2 (stranger/line) o
['13, +D] 53 44.17%)
3 (boss/review) 0
1+1), -D] 60 50.00 /o
4 (adviser/advice) o
[+P, +D] 61 50.83 A»
TOTAL 223 46.46%
(out of 480)

 

 

 

The fact that indirect complaints show up so frequently in this data is
undoubtedly tied to the degree of face-threat involved in complaints. To perform a
complaint threatens first the hearer's positive face, since one is pointing to him or
her as the perpetrator of some offensive act. It also, however, is a threat to the

speaker's positive face, since in performing the complaint, he or she opens him or

77

herself up to being seen in a negative light, as one who complains too easily,
without justification, etc.

Brown and Levinson (1987) point out that the major benefit of going "off-
record" (i.e. using an indirect FTA) is that the ambiguity involved provides a kind
of escape-hatch against such criticism, by allowing the speaker to avoid
committing him or herself to a single intent. In terms of complaints, using
indirectness is clearly a face-management strategy, since it (hopefully)
communicates that the speaker is unhappy with the situation without coating out
and directly stating the cause of his or her dissatisfaction.

One of the most interesting things about indirect complaints is that there
seem to be few or no conventional methods for complaining. In fact, the claim
can be made that, given the right circumstances (and perhaps the right inflection),
almost any utterance can be used and/or interpreted as a complaint. In the current
study, an analysis of the semantic content of the indirect complaints revealed
eleven different categories of indirectness. That there are many others is quite
likely. Table 16 presents the categories and frequencies of indirect complaints in
this study.

The most frequently employed strategy overall was for the speaker to
simply make a statement pointing out that the offensive act had been committed.
Very often, the reference to the offensive act itself was somewhat indirect, leading
to a kind of embedded indirectness. Thus, for example, in the messy kitchen

scenario, several respondents made statements such as "It looks like you made

78

yourself right at home." This is clearly a reference to the mess the bearer has
made, albeit an implied, rather than direct one. Similarly, in the situation in which
the stranger cut in line, many people made references to the back of the line (e. g.
"There is a line and it doesn't start there" and "The end of the line is back there"),

implying that the hearer had entered the line at an inappropriate spot.

Table 16. Categories of indirect complaints

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Situation

Caggrry 1 2 3 4 TOTAL
Statement of the Act 11 17 19 28 75
Justification 7 30 13 17 67
Hearer's Explanation l 10 27 20 58
Request for information 21 0 22 ll 54
Lexical/phrasal choice 12 4 l 0 17
Scolding 0 0 0 9 9
Warning 2 1 0 0 3
Protest 0 2 0 0 2
Compliment l 0 0 0 1
Wish I 0 0 0 1
Offer 1 0 0 0 1
TOTAL 57 64 82 85 288

 

 

 

 

 

 

 

 

In the third scenario with the boss, many respondents included statements
that pointed to the fact that they had seen the review and would like to talk about

it. In some cases, the fact that the review was unfavorable was stated, e. g. "I

79

noticed you gave me a bad review, could you please tell me what it is that does not
meet up to your expectations." In other cases, the unsatisfactory nature of the
review was more subtlely mentioned, such as "I got my review back and I didn't
realize that my work was not up to par." Notice that, in this case, part of the
indirectness comes from the avoidance of pinpointing the boss as the evaluator in
the bad review, although obviously both the boss and employee know that this is
the case. The strategy of stating the act is used most frequently in the final
situation, in which the academic adviser gives the student some bad advice on
which courses to take. A representative example in this case is the statement "You
advised me to take several classes that I come to find out are not applicable to my
major. "

The next most frequently employed indirectness strategy is for the speaker
to make a statement that is a justification for why he or she is complaining. The
purpose here is clearly two-fold. As mentioned previously, complaining is a
speech act that threatens not only the hearer's face, but also the speaker's face. By
laying out a justification for the complaint, the speaker not only registers the
complaint, but also points out that he or she is a reasonable person for making the
complaint. Often, these statements of justification are combined with other
statements, particularly requests for the bearer to repair the consequences of the
act. Examples from each of the situations in which this is the case are given

below:

80

1: "I would appreciate it if you would clean up after yourself. I just
cleaned the kitchen a few hours ago."

2: "Excuse me, I was here first so wait in line."

3: "Could you tell me why I received such a bad review...l thought
my performance was just fine."

4: "Is there anyway that I can get around taking this course since you
gave me bad advise [sic] about a different class."

It is sometimes the case that justifications for the complaint actually stand
alone and serve as the complaint itself. This is an interesting situation, since it
appears that a justification statement acts as a kind of mitigating device, which
serves to soften the force of a complaint that isn't ever stated. For example, in the
first situation, one respondent made the statement "Geez I just cleaned this place
up." It is possible to go through an inferencing process from the actual statement
made to an understanding that a complaint about the mess has been registered.

A strategy that is similar to justification is hearer's explanation. The
difference is that while justification works to directly protect the speaker's face,
hearer's explanation is a strategy that on the surface at least looks like it is being
used to manage the hearer's face. In this strategy, the speaker asks the bearer why
he or she would commit the offensive act, thus giving the bearer the chance to
explain his or her actions. Examples from the data are:

1: "What did you do that for?"

2: "Why did you cut in front of me?"

3: "Can you explain these low scores to me please."

81

4: "Why did you suggest that course when I didn't need it?"

In reality, using a question about the hearer's intentions is probably not a
sincere opportunity that the speaker is presenting the bearer to explain what
happened, but rather a way of pointing out that the speaker recognizes that the
hearer has committed an offensive act. Thus it is a very interesting strategy since
it seems to threaten the hearer's face (by being an indirect complaint) at the same
time it gives the appearance of maintaining it.

Since the strategy of hearer's explanation is most often phrased in the form
of a question, it could also be seen as a subset of the next category. In this more
general case, the complaint is disguised as a request for information. As with the
previous examples, however, it is not likely that the speaker is really interested in
an answer to the question but rather is using this as a way of presenting a
complaint. Examples from the various situations include:

1: "You were really hungry, weren't you."

3: "What was I doing wrong to deserve a review like that?"

4: "Do you realize the classes you had me take last semester were not even
required?"

The six strategies listed at the bottom of Table 16 illustrate the fact that
almost any speech act can be used to register a complaint. Even a speech act as
unlikely to be a complaint as a compliment can in fact be used in this way, as in
the case of the respondent who made the following statement about the messy

kitchen: "Did you make something good to cat? You utilized the kitchen well!"

82

The remaining category listed in Table 16 is somewhat unlike the others.
While the other strategies employ speech acts in ways that allow them to act as
something other than the most straightforward illocutionary act that they would
normally be interpreted as, the lexical choice strategy is one in which another
speech act is utilized, but the particular choice of a certain word or phrase is
enough to imply that the speaker is unhappy with the current state of affairs.

For example, the statement "Help me clean up the kitchen" would probably
not be interpreted as a complaint. The addition of a single phrase, however, is
enough to push it over the line and make it an indirect complaint. This is the case
with the respondent who said "Help me clean up the mess you made in the
kitchen" (italics mine). In other cases, the subtle change of a word is enough to
imply the complaint. This happens in the second scenario when one respondent
wrote the utterance "Excuse you" instead of the simple attention-getting phrase
"Excuse me." In the boss scenario, the one instance of this strategy being used is
when one respondent says "Hi, I would like to talk to you about my sub par
review." Here the addition of the adjectival phrase "sub par" is enough to suggest
the employee's dissatisfaction with the boss' rating.

It is important to reiterate the fact that the strategies discussed here are by
no means the only ones available for making an indirect complaint. In fact, one of
the reasons that there was so much variation in the comparison of intention versus
interpretation in complaints in chapter 3 was precisely because there is a kind of

negotiation of meaning that is available when complaints are made in this indirect

83

fashion. The advantages of being so indirect are clearly related to the face-
management of the parties involved in the situation, but the ambiguity can also

lead to complaints being missed altogether or being registered where not intended.

4.6 Complaints and Requests for Repair

Before closing this chapter, it is important to bring up one more aspect
regarding the threat to face that complaints carry. The problem with many past
studies of complaints is that they present the data in such a way as to suggest that
complaints are always made in isolation from other speech acts. What the data in
the current study show, however, is that complaints are often combined with other
speech acts, which work together to produce an overall face-threat to the various
participants involved in the interaction.

Given the fact that the scenarios in this study were purposefully constructed
to be situations which could be potentially repaired, the most common type of
FTA besides a complaint that appeared in the data was a request for repair of the
situation.

Requests are one of the more thoroughly studied speech acts in the
linguistics literature. While a complaint primarily threatens the positive face of the
bearer, a request primarily poses a threat to the hearer's negative face. This is
because a request asks the bearer to surrender a bit of his or her autonomy by

submitting to some imposition. Of course, like a complaint, a request likely also

84

involves some threat to the speaker's positive face, since he or she does not want to
be seen as unreasonable or overly-demanding in making the request.

In the CCSARP project mentioned earlier, an elaborate coding system for
the analysis of requests was developed. One aspect of the coding had to do with
what the researchers labeled "strategy type." This is a nine-category classification
system based on the syntactic and semantic features of the request in question.
These nine categories are further collapsed into three levels of directness, which
the authors term (1) direct strategies, (2) conventionally indirect strategies, and (3)
nonconventionally indirect strategies.

In the current study, requests for repair were identified in the responses and
coded on the same three point scale used for complaints, namely (1) direct
requests, (2) direct, but mitigated requests, and (3) indirect requests. The category
of indirect requests in this case corresponded to Blum-Kulka and House's (1989)
designation of "nonconventionally indirect" strategies, which included both strong
and mild hints. The main difference between the coding in the current study and
that of the CCSARP project is in what kinds of utterances are considered "direct"
requests.

Blum-Kulka and House claim that in addition to "mood derivable"
utterances (which utilize the grammatical mood, or command form, of the verb to
signal a request, e. g. 'Clean up that mess'), four other categories should be
considered direct requests. These categories, with examples, are performatives

('I'm asking you to clean that mess'), hedged performatives ('I would like to ask

85

you to give your presentation a week earlier than scheduled'), obligation
statements ('You'll have to move your car'), and want statements ('I really wish
you'd stop bothering me'). These five categories are combined into the level of
"direct" requests and differentiated from "conventionally indirect" requests, which
include suggestory formulae ('How about cleaning up') and query preparation
('Could you clean up the kitchen, please?)

While I would agree that mood derivable utterances are the most direct
strategy for making a request, I cannot see a principled justification for separating
strategies 2-5 from 6-7, calling the first group "direct" and the second
"conventionally indirect," and the researchers do not offer such a justification.
Rather, it seems to me that strategies 2-7 all represent mitigated forms of the bald
(i.e. "mood derivable") request strategy and should be classified in Brown &
Levinson's terms as on-record, with redress. In the current study, then, only those
requests for repair that utilize the typical command structure were classified as
"direct." Utterances that fall into the other categories mentioned above were
coded as being "direct, but mitigated." Of course the final option, as with
complaints, was to opt out of making a request for repair in a response. Table 17
presents the results for requests for repair by situation and directness level found

in the data for the current study.

86

Table 17. Overall directness levels for requests for repair

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Situation Directness Level # of respondents % of respondent
choosing this level choosing this
(out of 120) level

1 No request 74 61.67%
(friend/ Indirect 3 2.50%
messy Mitigated 25 20.83%
kitchen) Direct 18 15.00%
2 No request 20 16.67%
(stranger/ Indirect 53 44. 17%
cutting Mitigated 20 16.67%
in line) Direct 27 22.50%
3 No request 118 98.33%
(boss/ Indirect 2 1 .67%
bad Mitigated O 0.00%
review) Direct 0 0.00%
4 No request 78 65.00%
(adviser/ Indirect 13 10.83%
bad Mitigated 23 19. 17%
advice) Direct 6 5.00%

 

Several points are readily apparent from the figures given above. First, in
general, respondents chose to opt out of making a request for repair at higher rates
than they opted out of complaining (see Table 9 for a comparison). This is
especially the case in the third scenario, where only two brave souls made indirect
suggestions that the boss reconsider his or her marks on the employee performance
evaluation. The second situation produced the greatest number of requests for
repair, but even here there were twice as many indirect as direct requests.

What explanations can be offered to account for the relative infrequency of
requests for repair in situations constructed so as to be potentially repairable? One
possibility is that the respondents did not, in actuality, see the situations in

question as being easy to repair. This is not a good explanation, however, when

87

one considers the overall ratings given to each situation in terms of how likely it is
that the situation will be corrected. Recall from chapter 2 that after respondents
wrote responses to each scenario, they were asked to rate the situation on a five-
point scale for three factors. One of these factors was "likelihood of repair."
Table 18 gives the mean scores and standard deviations for this category.

As Table 18 shows, on a scale of 1 (= not very likely) to 5 (= very likely),
all four situations received a mean score of higher than 2.5, which means that
overall, respondents considered all of the scenarios to be more repairable than
not.8 Even situation 3, in which only two individuals chose to make a request for

repair, was rated as being relatively correctable.

Table 18. Mean scores for "likelihood of repair" ratings

 

 

 

 

 

Situation Mean Score Standard
Deviation
margi/hglchen) 3.767 1.158
2 (sgggfgqine) 3.483 1.092
3 (talcfsfrggsew) 3,292 0.893
4 (adliiggigllvice) 2.592 1.344

 

 

 

 

 

 

8 It should be noted that while the standard deviations for these ratings are somewhat large given
the scale in question. histograms show fairly normal distributions of ratings.

88

An alternative explanation for the greater use of complaints as opposed to
requests for repair is related to the kind of face threatened by each speech act. As
previously discussed, complaints primarily threaten the hearer's positive face,
while requests threaten hearer's negative face. In her work on cross-cultural
conflict styles and negotiation, Ting-Toomey (1988) puts forth the proposition that
one of the major differences between members of individualistic and collectivistic
cultures is the relative importance of negative versus positive face needs.

She claims that in individualistic cultures, such as the US, the need for
autonomy and privacy outweighs the need for interdependence and inclusion,
while the Opposite is true for members of collectivistic cultures. In terms of
conflict situations, this would mean that members of individualistic cultures
generally perceive a greater degree of face-threat from F TAs that threaten negative
face (such as requests) than those that threaten positive face (such as complaints).
If Ting-Toomey's theory is correct, it may be the case that the respondents in this
study are refraining from making requests for repair more than refraining from
making complaints because requests for repair are considered to be more serious
in terms of face-threat to these US. students.

A third and final explanation that may account for the data here relies not
on the kind but rather the amount of face-threat involved overall. Table 19
provides a comprehensive picture of how the two F TAs in question are used in

combination.

89

Table 19. Complaints and requests for repair by situation and directness levels

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Situation Complaint Repair Level
Level No repair Indirect Mitigated Direct
1 No complaint 30 2 5 0
(friend! Indirect 26 l 10 12
messy Mitigated 2 0 2 0
kitchen) Direct 16 0 8 6
2 No complaint 3 23 2 8
(stranger/ Indirect l3 2 1 l 1 8
cutting Mitigated 0 0 2 0
in line) Direct 4 9 5 H
3 No complaint 31 2 0 0
(boss/ Indirect 60 0 0 0
bad Miti gated 16 0 O 0
review) Direct 1 1 0 0 6
4 No complaint l7 1 1 0
(adviser/ Indirect 32 9 17 3
bad Mitigated 1 1 0 0 0
advice) Direct 1 8 3 5 3

 

 

 

 

 

 

 

The first item to note in this table is the relatively infrequent use of direct
complaints in combination with direct requests for repair (see shaded cells). This
combination clearly involves the greatest amount of face-threat to the hearer. If
we look at the scenarios in terms of the power and social distance components, we
find that the situation requiring the most skillful management of face-concems, i.e.
with the boss (an unequal power but on-going relationship), has no instances of
this combination of F TAs. The adviser scenario ([+power, +distance]) has the
next fewest incidents, followed by the friend scenario ([-power, -distance]). Both
of these situations involve some careful manipulation of face-needs, but not to the
same extent as the boss scenario. The situation which requires the least amount of

attention to face concerns, namely the stranger cutting in line (a [-power,

90

+distance] scenario), also has the highest incidence of the most face-threatening
combination of speech acts.

At the opposite end of the face-threat continuum is the choice to opt out of
performing any FTA at all. Here we find that social distance is the important
factor in determining whether or not some threat to face will be risked. For the
two {-distance] situations which represent on-going relationships that must be
maintained, we find a fairly high occurrence (around 25%) of respondents opting
out. These respondents likely feel that the offenses in question are not so great as
to risk threatening the face of the hearer and thus damaging the relationship. For
the remaining two scenarios, respondents are much more likely to risk some face-
thrcat, although here power seems also to come into play. For the [+distance]
scenario that includes an unequal power component (i.e. with the adviser),
respondents are ahnost six times as likely to opt out of performing any FTA than
with the similar distance, but equal power situation.

In between these two extremes, we find various combinations of indirect,
mitigated, and direct complaints and requests for repair. Since these two F TAs
involve qualitatively different kinds of face-threat, it is not possible to objectively
quantify the precise amount of face-threat that any given combination produces. It
is apparent, however, that for each situation these two kinds of F TAs, with their
varying levels of directness, are used together in managing overall face concerns.

In the first situation, there are, for example, an almost equal number of

respondents who chose to produce a direct complaint with no request for repair as

91

those who chose a direct request for repair with no complaint. A similar statement
can be made about the second and fourth scenarios, although in this case the
combinations in question are direct complaints and indirect requests for repair, and
vice versa (thus producing a slightly higher face-threat than the combination in the
first scenario).

In terms of other combinations, we find a high frequency of indirect
complaints with no requests for repair (a combination that is clearly only one step
removed from choosing to opt out) for the first, third and fourth situations (with
26, 60, and 32 respondents respectively using this combination). In the second
situation, on the other hand, more respondents (23) chose the opposite
combination. In addition, a fairly high number (21) risked slightly more face-
threat by placing an indirect complaint together with an indirect request.

Although more could undoubtedly be said about the particular FTA
combinations for each scenario in this study, the overall point is that it is not
enough to study speech acts in isolation if we want to understand how they are
used in the overall politeness and face-management of actual discourse situations.
Rather it is vital to understand that each speech act used contributes some portion
of the overall threat (or maintenance for that matter) in the negotiation of
participants' face.

This discussion has focused on two main types of face-threatening acts
found in PCSs, namely complaints and requests for repair. To be completely

accurate in claims regarding the overall face-threat would require investigation

92

into additional kinds of speech acts involved in the responses in this study. A scan
of the data was enough to convince the researcher that no other category of speech
acts was substantial enough to warrant separate analysis in this case. Nonetheless,
future studies of Potential Complaint Situations (or any discourse situation for that
matter) should bear this point in mind when examining how speech acts are

manipulated to maintain appropriate levels of politeness and face-management.

93

Chapter 5

SOCIOLINGUISTIC VARIATION IN COMPLAINING

5.1 Situations and Complaint Realizations

In the preceding chapter, we saw how levels of directness for complaints
and requests for repair are used to produce different amounts of face-threat in
various situations. In this chapter, the discussion turns to an analysis of the
primary sociolinguistic variables found in this study -- namely situation, speaker
sex, hearer sex, power and social distance -- and how these variables affect the
realization of complaints in Potential Complaint Situations.

In the discussion in chapter 4, directness levels were analyzed for each of
the four situations, although no justification was given for separating the data in
this way. It is thus helpful to begin this discussion with a more careful look at
how the four scenarios that appear in this study differed from one another. As
mentioned in Chapter 2, respondents were asked to rate each situation in terms of
three characteristics: severity of the offense, responsibility of the offender and
likelihood of correction. The mean scores and standard deviations for each of

these characteristics were calculated, and the results are presented in Table 20.

94

Table 20. Mean scores for three characteristics in each situation

 

 

Situation Severity of Responsibility Likelihood of
Offense of Offender Repair
1 2.308 3.992 3.767

S.D.=l.019 S.D.=l.213 S.D.=l.158

 

2 3.875 4.292 3.483
S.D.=0.984 S.D.=l.337 S.D.=l.092

 

3 3.292 3.667 3.292
S.D.=l.170 S.D.=0.973 S.D.=0.893

 

4 4.075 4.008 2.592
S.D.=l.189 S.D.=l.357 S.D.=l.344

 

 

 

 

 

 

If we compare the situations to one another in terms of the combination of
these three factors, we find some interesting differences among the scenarios. The
first situation, with the friend and the messy kitchen, is seen as being least severe
and most likely to be repaired, while falling in the middle in terms of
responsibility of the offender. The exact opposite can be said for the fourth
situation, in which the advisor gives bad advice. To these respondents, this
offense is both most severe and least likely to be corrected. In situation 2, the
stranger cutting in line is held most responsible for his or her actions, and the
offense is seen as fairly severe but also fairly correctable. Finally, the employer
giving the bad review is seen as least responsible for the offense in question, and
this offense is seen as less severe but also less likely to be repaired than all but one
other.

One-way ANOVA tests reveal that situation is a significant variable for all

three of these characteristics (Severity: F =62.954, d.f.=3, p=.000; Responsibility:

95

P: 5.184, d.f.=3, p=.002; Likelihood: F=23.428, d.f.=3, p=.000). Post-hoe Tukey
tests identify more precisely where the significance lies, as shown in Table 21.

In terms of severity, situations 2 and 4 are not rated significantly differently
from one another, but these two combined are significantly different from both 1
and 3. Situation 1 is seen as least severe, followed by situation 3 and then
situations 2 and 4.

The case is slightly more complex for the responsibility ratings. Situations
1, 2 and 4 are not significantly different, and neither are situations 1, 3 and 4. The
difference between the ratings of situations 2 and 3, however, is large enough to
create overall significance. In terms of the mean scores, the boss is held least
responsible for his or her actions, followed by the fiiend. Next comes the advisor,
and finally the stranger.

In terms of likelihood of repair, situations 1 and 2 are not significantly
differently rated, and neither are situations 2 and 3. Situations 1 and 3, however,
are significantly different from one another, and situation 4 is significantly
different than all three other scenarios. Mean scores reveal that situation 4 is seen
as least likely to be corrected, while situation 1 is most likely to be repaired. In

between these two fall situations 3 and 2 respectively.

96

Table 21. Results of post-hoc Tukey tests for the three characteristics

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mean
De endent (I) (J) . Std. .

Viriable Situation Situation Difffrsnce Error Slg'
2.00 4566791 .141 .000

1.00 3.00 -.9833(*) .141 .000

4.00 -1.7667(*) .141 .000

1.00 1.5667(*) .141 .000

. 2.00 3.00 .5833(*) .141 .000
se‘ffm 4.00 -.2000 .141 .489
Offense 1.00 .9833(*) .141 .000
3.00 2.00 -.5833(*) .141 .000

4.00 -.7833(*) .141 .000

1.00 l.7667(*) .141 .000

4.00 2.00 .2000 . 141 .489

3.00 .7833(*) .141 .000

2.00 -.3000 .159 .232

1.00 3.00 .3250 .159 .171

4.00 -1.6667E-02 .159 1.000

1.00 .3000 .159 .232

Responsibility 2.00 3.00 .6250(*) .159 .000
of 4.00 .2833 .159 .280
Offender 1.00 -3250 .159 .171
3.00 2.00 -.6250(*) .159 .000

4.00 -.3417 .159 .137
1.00 1.667E-02 .159 1.000

4.00 2.00 -.2833 .159 .280

3.00 .3417 .159 .137

2.00 .2833 .146 .213

1.00 3.00 .4750(*) .146 .006

4.00 1.1750(*) .146 .000

1.00 -.2833 .146 .213

. . 2.00 3.00 .1917 .146 .556
““6"?“ 4.00 .8917(*) .146 .000
Raga“ 1.00 -.4750(*) .146 .006
3.00 2.00 -.1917 .146 .556

l 4.00 .7000(*) .146 .000
1.00 -1.1750(*) .146 .000

l 4.00 2.00 -.89l7(*) .146 .000
L 3.00 -.7000(*) .146 .000

 

 

 

 

 

 

 

 

* The mean difference is significant at the .05 level.

97

 

It is, of course, important to note that the ratings of severity, responsibility
and likelihood of repair cannot be completely disentangled from the identities of
the participants involved. What makes the offense in the first scenario least
severe, for example, may be at least in part the fact that it is committed by a friend.
If the scenario was set up so that a roommate's fiiend messed up the kitchen, it is
possible that the offense would have been rated more harshly. It is interesting to
note, as well, that the boss is seen as least responsible for the offense while the
stranger is held most responsible. In reality, it is much more likely that cutting in
line would be an unintentional act, particularly compared to the bad review given.
Clearly some consideration of the power and social distance differentials must
have played a role in these ratings.

Even so, how each situation was rated in terms of the three characteristics
discussed here provides some explanatory power when we examine levels of
directness utilized (see Tables 9, 17, and 19). Returning briefly to some of the
points mentioned in the last chapter, recall that respondents chose to opt out of
complaining the least in the foruth scenario. It is probably not a coincidence that
this is the situation that was considered to contain the most severe offense.
Situations l and 2 also contained the highest numbers of direct and mitigated
requests for repairs, and these are the two scenarios rated most likely to be
repaired.

The fact that the boss is held least responsible for the offense in question

may also partially explain the extremely high number of responses that contain an

98

indirect complaint with no request for repair. If the boss is not considered overly
accountable for his or her actions, then there is no firm ground on which to put
forth a face-threatening speech act that will seem justified. These explanations
should not be divorced from other factors such as power and social distance which
undoubtedly contribute to the overall results; however, used in combination with
these other factors, a better understanding of how and why directness levels are

utilized in the performance of FTAs can be achieved.

5.2 Severity, Responsibility and Likelihood of Repair as Independent
Variables

Up to this point, we have been examining the ratings of severity of offense,
responsibility of offender and likelihood of repair as dependent variables, with the
independent variable being situation. However, it is also possible to examine the
levels of directness utilized by respondents by using these three categories as
independent variables. It is an interesting question, for example, as to whether
those respondents who perceive a situation to be quite severe are more likely to
use a higher level of directness in complaint realization than those who see the
situation as not being very severe. It is possible to examine the directness levels
utilized for both complaints and requests for repair in this way. Although
respondents rated each scenario for these three factors on a scale of l to 5, for the

purposes of this analysis, respondents have been collapsed into three groups: low

99

(ratings of 1 or 2), medium (rating of 3) and high (rating of 4 or 5) for each
characteristic.

It is possible to examine the effects of the factors of severity, responsibility
and likelihood of repair both overall and within particular situations. Overall, both
severity and responsibility proved to be significant factors in the choice of opting
out versus performing a complaint, as Tables 22 and 23 illustrate. Respondents
who judge the situation to be low or medium in terms of severity of offense and
responsibility of offender chose to opt out of performing the complaint at a higher

rate than those who judged the situations to be high in terms of these factors.

Table 22. Overall effects of severity of offense on complaint realizations

 

 

 

Respondents Severity of Offense RatirL
who: Low Medium HIE
Opted Out 42 (32.8%) 36 (31.8%) 47 (19.7%)

 

 

 

 

 

 

Complained 86 (67.2%) 77 (68.1%) 192 (80.3%)
x2=10.077,d.f.=2.p<,01

 

Table 23. Overall effects of responsibility of offender on complaint realizations

 

 

 

Respondents Responsibility of Offender Rating
who: Low Medium High
Opted Out 24 (36.9%) 25 (31.2%) 76 (22.7%)

 

Complained 41(63.1%) 55 (68.8%) 259 (77.3%)
x2=7.081,d.£=2,p<,05

 

 

 

 

 

 

In terms of the speech act of requesting repair, all three factors had a

significant overall effect in the choice of opting out versus performing the FTA.

100

These results are given in Tables 24-26 below. The results for severity of offense

and likelihood of repair are straightforward in their interpretation. Low ratings in

these two categories tend to produce a higher percentage of respondents choosing

to opt out, while high ratings cause respondents to perform the request for repair at

a higher rate.

Table 24. Overall effects of severity of offense on requests for repair
Respondents Severity of Offense Rating
who: Low Medium Hifi
Opted Out 94 ( 73.4%) 68 (60.2%) 128 (53.6%)
Requested Repair 34 (26.6%) 45 (39.8%) 111 (46.4%)

Table 25.

Table 26.

 

 

 

 

 

 

 

 

 

 

x2 = 13.780, d.f. = 2, p < .001

Overall effects of responsibility of offender on requests for repair

 

 

 

 

 

 

Respondents Responsibility of Offender Rating
who: Low Medium High
Opted Out 36 (55.4%) 71 (88.8%) 183 (54.6%)
Requested Repair 29 (44.6%) 9 (11.2%) 152 (45.4%)

 

 

 

 

x2 = 32.238, d.f. = 2, p < .001

Overall effects of likelihood of repair on requests for repair

 

 

 

 

 

Respondents Likelihood of Repair Rating

who: Low Medium High
Opted Out 92 (77.3%) 82 (63.1%) 116 (50.2%)
Requested Repair 27 (22.7%) 48 (36.9%) 115 (49.8%)

 

 

 

 

 

x2 = 24.637, d.f. = 2, p < .001

The figures in Table 25 are harder to interpret. It appears that respondents

who rate the responsibility of the offender as either high or low chose to perform

lOl

the F TA at an almost equal rate. The statistical significance comes from the fact
that the group of medium raters chose to opt out of requesting repair at a much
higher rate. Perhaps this is a sign that this factor was not seen to be as relevant to
requests for repair as the other two factors, leading a large number of respondents
to fall back on the neutral rating of "3."

For respondents who did choose to perform either F TA, we find an overall
significant effect on complaint realizations only for the factor of responsibility of
offender, as shown in Table 27. As in Table 25, the results are somewhat
surprising. Respondents who rated the responsibility of the offender as either high
or low used a fairly similar distribution of indirect versus mitigated versus direct
complaints. It is once again the medium raters who show a slightly different
pattern. They utilized a slightly higher rate of indirectness and mitigation, with a
substantially smaller percentage of direct complaints. It may be their uncertainty
in determining the responsibility of the offender that causes them to err on the side

of caution in this case.

Table 27. Overall effects of responsibility on directness levels in complaints

 

 

 

 

 

Directness Responsibiligy of Offender Ratigg
Level: Low Medium High
Indirect 26 (63.4%) 40 (72.7%) 157 (60.6%)
Mitigated 3 (7.3%) 10 (18.2%) 22 (8.5%)
Direct 12 (29.3% 5 (9.1%) 80 (30.9%)

 

 

 

 

 

 

x3 = 13.636. d.f. = 4. p < .01

102

In terms of requests for repair, only likelihood of repair had an overall
significant effect on directness levels. These results are given in Table 28. In this
case, those who rate the situations as highly likely to be repaired are almost
equally distributed in their choice of directness levels. Perhaps since the situation
is likely to be repaired, a request for repair in any form will do the trick. Which
directness level is actually chosen is then determined by other factors. For those
who are less confident about the reparability of the situation, however, greater
face-management must be employed so as not to push the hearer away from a

willingness to correct the situation due to the directness/rudeness of the speaker.

Table 28. Overall effects of likelihood of repair on directness levels in requests

 

 

 

 

 

 

Directness Likelihood of Repair Rating
Level: Low Medium High
Indirect 12 (44.4%) 21 (43.8%) 38 (33.0%)
Mitigated 6 (22.2%) 22 (45.8%) 40 (34.8%)
Direct 9 (33.3%) 5 (10.4%) 37 (32.2%)

 

 

 

 

 

x2 = 10.701. d.f. = 4, p < .05

In addition to the overall effects discussed above, it is also possible to get a
better idea of how individual situations interact with the three variables of
severity, responsibility and likelihood of repair, if at all. When we examine
individual scenarios, chi-square tests reveal no significant effects for any of the
three variables in terms of level of directness for either complaints or request for

repair. There were a few significant effects, however, in the choice of opting out

103

versus performing the FTAs for each of the characteristics in at least one of the
situations.
In terms of severity of offense, there was a significant effect on complaint

realizations for the fourth situation only. Table 29 presents these results.

Table 29. Severity of offense for complaints in situation 4

 

 

 

Respondents Severi of Offense Rating
who: Low Medium High
Opted Out 10 (52.6%) 1 (12.5%) 8 (8.6%)

 

Complained 9 (47.4%) 7 (87.5%) 85 (91.4%)
x2 = 23.022, d.f. = 2, p < .001

 

 

 

 

 

 

The figures here reveal that respondents who rated the offense in the fourth
situation as being moderately to highly severe were much more likely to produce a
complaint. Of the 19 respondents who saw this offense as being fairly low in
terms of severity, however, just over half chose to opt out of complaining about
the situation.

A similar effect was found for the fourth situation only in terms of

responsibility of the offender, as shown in Table 30.

Table 30. Responsibility of offender for complaints in situation 4

 

 

 

Respondents Responsibility of Offender Rating
who: Low Medium High
Opted Out 8 (38.1%) 5 (38.5%) 6 (7.0%)

 

 

 

 

 

 

Complained 13(61.9%) 8(61.5%) 80 (93.0%)
x2=17.867,d.f.=2,p<,001

 

104

In this case, those who found the advisor highly responsible for the mistake
complained at a rate of 93%. In comparison, only about 62% of those who
thought the advisor to be low or medium in terms of his or her responsibility

complained about the situation.

Table 31. Likelihood of repair for complaints in situation 1

 

 

 

Respondents Likelihood of Repair Rating
who: Low Medium High
Opted Out 10 (58.8%) 8 (36.4%) 19 (23.4%)

 

Complained 7 (41.8%) 14 (63.6%) 62 (76.5%)
x2 = 8.627, d.f. = 2, p < .05

 

 

 

 

 

 

In terms of likelihood of repair, a significant effect for complaining versus
opting out was found only in the first situation. These figures are presented in
Table 31. In this case, the more likely the respondents feel the situation is to be
repaired, the more likely they are to complain about it. Those who see a low
likelihood of repair complain at a rate of only 41.8%. The respondents who rated
the likelihood of repair as medium complain at a rate of 63.3%. Finally, 76.5% of
those who perceive a high likelihood of repair chose to make a complaint.

Turning to the F TA of requesting repair, we find a few more significant
effects based on these three factors. Severity of offense comes into play only in

the first situation, as shown in Table 32.

105

Table 32. Severity of offense for requests for repair in situation 1

 

 

 

Respondents Severity of Offense Rating
who: Low Medium High
Opted Out 52 (74.3%) 18 (47.4%) 4 (33.3%)

 

Requested repair 18 (25.7%) 20 (52.6%) 8 (66.7%)
x: = 12.077, d.f. = 2. p < .005

 

 

 

 

 

 

In this situation, the expected pattern is once again found. Respondents are
most likely to opt out if they find the offense to be low in terms of severity. Only
a third of those who find the situation highly severely offensive chose to opt out.
In between these two extremes lie those who rated the situation as moderately
severe. This group is almost equally split between those who request a repair and
those who opt out of performing the FTA.

There were no significant effects on any of the situations for requests for
repair in terms of the ratings of responsibility of the offender. As might be
expected, however, three of the four situations showed significant effects by
likelihood of repair ratings on the rates at which repairs were requested. These
figures are presented in Table 33.

For the first situation, only the respondents who perceived the situation as
being highly likely to be repaired bothered to request a repair at any substantial
rate (53.1%). In situation 2, the break seems to come between those who perceive
the situation as very low in terms of being repaired (52.2% requested repair) as

opposed to those who rate it medium (88.2%) or high (92.1%).

106

Table 33. Likelihood of repair for requests for repair in situations 1, 2 and 4

 

 

 

 

 

 

 

 

 

 

Situation Respondents Likelihood of Repair Rating
who: Low Medium m
1 Opted Out 17 (100%) 19 (86.4%) 38 (46.9%)
Requested repair 0 (0%) 3 (13.6%) 43 (53.1%)
x2 = 23.702, d.f. = 2, p < .001
2 Opted Out 11 (47.8%) 4 (11.8%) 5 (7.9%)

 

Requested repair 12 (52.2%) 30 (88.2%) 58 (92.1%)
x3 = 20.124. d.f. = 2. p < .001
4 Opted Out 45 (75.0%) 12 (46.2%) 21 (61.8%)

 

 

 

 

 

 

 

Requested repair 15 (25.0%) 14 (53.8%) 13 (38.2%)
x2 = 6.853, d.f. = 2, p < .05

 

 

 

 

In situation 4, we find a slightly different pattern. Here it is those who see
the situation as moderately likely to be repaired that are most likely to request a
repair (53.8%). Those who rate it highly likely to be repaired request a repair at a
rate of only 38.2%. And those who see it as unlikely to be repaired drop to a
request rate of only 25%. The reversal of expected rates of requesting repair
between the medium and high raters is difficult to explain. It is likely that other

factors that are not immediately apparent are influencing the results in this case.

5.3 The Effects of Sex, Power and Social Distance on Complaint
Realizations

The remaining sociolinguistic variables examined in this research were sex
(of both the speaker and hearer), power and social distance. Like the

characteristics of the situation discussed in the previous section, these variables

107

influenced the choice of whether or not to complain, as well as the directness level
when complaints were made, in various ways. The raw numbers for the
breakdown of complaint realizations in terms of the combination of all four of
these variables are given in Table 34. Since it is difficult to say anything
substantial about the figures as presented in the detailed breakdown, the analysis
will proceed to a discussion of each of these more specific variables by collapsing

the numbers given in Table 34 in various ways.

Table 34. Complaint realizations by sex, power and social distance

Distance Speaker Hearer Com Directness Level

Sex Sex Out Indirect ' Direct

F 13
14
8
14
15
16
9
13
22
ll
9
18
l6
l4
13
18

I-‘I—i
Miss“)

M

E :ooxo-n

h-l
OU'

 

Nw-fi-NOOQOOI—NI—h—Ou—N
— ~h—l
qoqquowosoam_ow0x

F
M
F
M
F
M
F
M 8
F 6
M
F
M
F
M
F
M

DJ-kLIIN

The first question we can examine is to what extent the identities and
relationships of the participants influence the choice to make a complaint in the

first place. Chi-square tests of independence reveal that of the four variables

108

discussed here, only power is significant overall in the choice of opting out. Table
35 summarizes the data for this variable.

Without considering the individual situations, participants chose to opt out
of performing a complaint at a rate of 26% overall. Quite unexpectedly, those
situations which include a power differential between participants are actually less
likely to lead participants to opt out of complaining than those where participants

are equal in power.

Table 35. The effects of power on the choice of opting out of complaining

 

 

 

 

 

Power Choice

Level Opt Out Make Complaint
- 73 (30.4%) 167 (69.6%)

+ 52 (21.7%) 188 (78.3%)
Total 125 (26.0%) 355 (74.0%)

 

 

 

 

 

x2 = 4.770, d.f. = 1, p < .05

It is difficult to interpret these results in isolation. This outcome may be the
by-product of the particular situations utilized in this study rather than a general
result about the effects of power on complaint realizations. Table 10 in chapter 4
shows the rates at which respondents opted out of each particular situation. While
there are similar numbers of respondents opting out of situations 1, 2 and 3, a
much smaller number opted out of complaining in situation 4 (which is, in fact, a
[+power] situation). As discussed previously in this chapter, situation 4 is

considered to contain the most severe offense and is least likely to be repaired in

109

the respondents' estimation. These facts may account for the lower percentage of
opting out, and this in turn may be an explanation for this power result.

Actually, explanations for the effects of power on complaint realizations
should also be sought in terms of possible interactions with the other variables.
Looking at the two-way interactions between speaker sex, hearer sex, power and
distance reveals that there is, in fact, a significant interaction between power and

hearer sex.l Table 36 presents the results of this interaction.

Table 36. The interaction of power and hearer sex in complaint realizations

 

 

 

 

 

 

 

Power/Hearer Sex Choice

Combination Opt Out Make Complaint
-power, female 41 (34.2%) 79 (65.8%)
-power, male 32 (26.7%) 88 (73.3%)
+power, female 22 (18.3%) 98 (81.7%)
+power, male 30 (25.0%) 90 (75.0%)
Total 125 (26.0%) 355 (74.0%)

 

 

 

 

 

x2 = 7.907, d.f. = 3, p < .05

The figures in this table reveal that while respondents tend to complain to a
male offender at about the same rates (around 74%), regardless of the power
relationship between the two, there is a striking difference between complaint rates

to females in positions of power versus non-power. To peers, i.e. power equals,

 

1 There are no other significant two-way interactions except for power by distance, which is the
same thing as situation.

110

both male and female respondents2 opt out of complaining at a higher rate if the
hearer is female rather than male. For authority figures, exactly the opposite is
true. This is a very interesting result and may reflect societal attitudes about the
"proper" roles and treatment of women versus men that persist, even among
younger respondents, after 20+ years of mainstream feminist thought promoting
the idea of women as status equals to men in positions of authority.

It is also helpful in this discussion to consider the directness levels utilized
when complaints were made. Once again, a chi-square test shows power to be a

significant factor.

Table 37. The effects of power on directness levels of complaints

 

 

 

 

 

Power Directness Level

Level Indirect Mithggted Direct

- 102 (61.0%) 8 (4.8%) 57 (34.1%)
+ 121 (64.4%) 27 (14.4%) 40 (21.3%)
Total 223 (62.8%) 35 (9.9%) 97 (27.3%)

 

 

 

 

 

 

x2 = 13.718, d.f. =2,p< .001

The figures in Table 37 are quite straightforward in their interpretation.
The two situations in this study in which respondents are in equal power
relationships with the offenders produce a much higher percentage of direct

complaints and a slightly lower rate of indirect complaints than the situations

 

3 It is important to note here that there was no significant 3-way interaction between speaker scx,
heater sex and power (or social distance for that matter).

11]

where power is unequal. Mitigation tends to be used much more in [+power]
situations as well.

In addition to power, the variable of speaker sex was also significant
overall in terms of the directness levels utilized in making complaints (although
not in the choice to opt out). Table 38 shows the comparison. The results here
support other claims in the sociolinguistic literature that women tend to be more
indirect in their use of language than men (e.g. Goodwin 1980). In this case, we
find that two-thirds of the direct complaints in the study come from men, who
utilize the option of directness (rather than mitigation or indirectness) at ahnost

twice the rate of women.

Table 38. The effects of speaker sex on directness levels of complaints

 

 

 

 

 

Speaker Directness Level

Sex Indirect Mitigated Direct
Female 121 (70.3%) 19 (11.0%) 32 (18.6%)
Male 102 (55.7%) 16 (8.7%) 65 (35.5%)
Total 223 (62.8%) 35 (9.9%) 97 (27.3%)

 

 

 

 

 

 

x2 = 12.774, d.f. = 2, p < .005

Once again, the situation is slightly more complex than either Tables 37 or
38 suggest. If we look at interactions between variables, we find a number of
significant results. In fact, chi-square tests reveal two-way interactions between
most of the independent variables. These will be presented and discussed

individually as appropriate.

112

To begin with, we find that speaker sex shows a significant interaction with
hearer sex, as seen in Table 39. One important source of significance in this table
is the behavior of male respondents towards female offenders. It is in this gender
combination that the smallest rate of indirectness and largest percentage of direct
complaints are made. Female respondents are not very likely to be direct to either
females or males (although slightly more so to females); however, they utilize
mitigated complaints at a rate that is substantially higher when they are
complaining to men as opposed to other women. Mitigation in general, therefore,

seems to be largely the result of cross-sex versus same sex-interactions.

Table 39. The interaction of speaker and hearer sex in complaint directness

 

 

 

 

 

 

 

Speaker/Hearer Directness Level

Sex Combination Indirect Mitigited Direct
female to female 66 (29.6%) 5 (14.3%) 18 (18.6%)
female to male 55 (24.7%) 14 (40.0%) 14 (14.4%)
male to female 39 (17.5%) 13 (37.1%) 36 (37.1%)
male to male 63 (28.2%) 3 (8.6%) 29 (29.9%)
Total 223 35 97

 

 

 

 

 

 

x2 = 30.462, d.f. = 6, p < .001

Speaker sex also interacts significantly with both power and social distance.
The figures for these interactions are presented in Tables 40 and 41.

In the first case, three different patterns occur when we examine each of the
directness levels. For indirectness, we find similar percentages of usage for

females in both power configurations and males in [+power] relationships, but

113

males in {-power] situations use much less indirectness and the highest rate of
direct complaints. They also rely on mitigation very little. Mitigation in fact goes
hand-in-hand with power differentials, regardless of speaker sex. In terms of
directness, both females and males utilize it more in {-power] situations than in

[+power] ones, although overall males use it more than females.

Table 40. The interaction of speaker sex and power in complaint directness

 

 

 

 

 

 

 

 

Speaker Sex/Power Directness Level

Combination Indirect Mitigated Direct
female, -power 58 (26.0%) 6 (17.1%) 18 (18.6%)
female, +power 63 (28.2%) 13 (37.1%) 14 (14.4%)
male, -power 44 (19.7%) 2 (5.7%) 39 (40.2%)
male. +power 58 (26.0%) 14 (40.0%) 26 (26.8%)
Total 223 35 97

 

 

 

 

x2 = 28.843, d.f. = 6, p < .001

Table 41. The interaction of speaker sex and distance in complaint directness

 

 

 

 

Speaker Sex/Distance Directness Level
Combination Indirect Mitigated Direct
female. -distance 60 (26.9%) 10 (28.6%) 11 (11.3%)

 

 

 

 

 

female, +distance 61 (27.4%) 9 (25.7%) 21 (21.6%)
male, -distance 49 (22.0%) 10 (28.6%) 30 (30.9%)
male, +distance 53 (23.8%) 6 (17.1%) 35 (36.1%)
Total 223 35 97

 

 

 

 

x2 = 16.102, d.f. = 6, p < 05

114

 

In terms of the interaction between speaker sex and social distance, there
are several interesting observations that can be made. Both females and males
tend to utilize indirectness at similar rates, regardless of how well they know the
other individual (although once again, females use it more than males). When it
comes to directness, however, social distance does figure in, as both females and
males are more direct in [+distance] relationships. This makes sense since there is
no substantial on-going relationship that could be harmed by being too face-
threatening. In on-going relationships, both males and females utilize mitigated
complaints at a higher rate.

The remaining two-way significant interaction in this data comes between

hearer sex and power. This result is presented in Table 42.

Table 42. The interaction of hearer sex and power in complaint directness

 

 

 

 

 

 

 

Hearer Sex/ Power Directness Level

Combination Indirect Mitigated Direct
female, -power 45 (20.2%) 4 (11.4%) 30 (30.9%)
female, +power 60 (26.9%) 14 (40.0%) 24 (24.7%)
male, -power 57 (25.6%) 4 (11.4%) 27 (27.8%)
male, +power 61 (27.3%) 13 (37.1%) 16 (16.5%)
Total 223 35 97

 

 

 

 

 

 

x3 = 16.040. d.f. = 6, p < ,05

The results in this table are again fairly straightforward. Indirectness tends
to be utilized more in [+power] situations regardless of whether the respondents

are complaining to a male or female, although it should be noted that the females

115

in {-power] relationship receive the lowest rate of indirectness as well as the
highest rate of directness in complaint realizations. Conversely, males in positions
of power receive the most deference, reflected in the greatest percentage of
indirectness and smallest rate of direct complaints. The use of mitigation is once
more the consequence of power, even more so than is indirectness.

While it is possible to examine three-way interactions as well, the
interpretation of significance becomes much more difficult, and in fact we do not
learn much more about the data than we already have by examining the effects of
individuals variables and two-way interactions.

In addition, it is possible to examine the other FTA in this study, i.e.
requests for repair, using the same sociolinguistic variables as in the preceding
analysis of complaint realizations; however, doing so does not add significantly to
an understanding of the effects on the performance of face-management which
these variable contribute. The effects, in fact, are similar, with each of the four
variables of speaker sex, hearer sex, power and social distance contributing in
some way to the overall results, either as an independent effect or by interaction
with some other variable. Furthermore, since only two respondents chose to make
a request for repair in the third situation (with the boss), the results of statistical
analysis would be skewed. Because of these reasons, the results and discussion of
the effects of the four sociolinguistic variables on requests for repair have been

excluded.

116

To summarize this section, we can say that, of the variables examined here,
the variable of power seems to exert the greatest influence over complaint
realizations. It has a significant individual effect on both the choice of whether or
not to make a complaint in the first place, as well as on what level of directness is
utilized when complaints are made. In addition, it interacts with the other three
variables to affect complaint realizations in various ways. Speaker sex appears to
be the next most important variable, as it also individually exerts significant
influence over directness levels in complaint realizations. It also interacts with
each of the other three variables in terms of directness levels. Hearer sex and
social distance are less important, although they do have some significant

influence in combination with the other variables.

117

Chapter 6

SUMMARY AND CONCLUSIONS

6.] Summary of Results

The purpose of this chapter is to summarize the main points from this
investigation of responses to Potential Complaint Situations and to situate the
current results within a broader framework of face-threat and face-management.
The current study has examined both what people say within PCSs, as well as how
they define, interpret and perceive complaints.

The study began with a pragmatic definition of a complaint as the
expression of a negative evaluation of some situation that the speaker finds
unsatisfactory. This definition was anchored in folldinguistic data, and the
discussion noted that it is not necessary for the offensive act to have been
committed by the hearer (i.e. the one to whom the complaint is being made). It is
cases such as these, however, where most research (including the current study)
has focused attention, since it is these situations that require the greatest amount of
attention to facework issues.

Having defined complaints, an argument was made that in order for a
complaint to be considered a "direct" speech act, the negative evaluation inherent
in it must be straightforwardly recoverable from some aspect of the utterance
itself. The fourth chapter explored various ways that this direct negative

evaluation can be expressed. In particular, six strategies utilized in the data of the

118

current study were noted. These strategies were: (1) expressing disapproval of the
action, (2) using Obscenities and/or name-calling, (3) incorporating negative
exclamations, (4) pointing out the negative consequences of the hearer's action, (5)
utilizing sarcasm, and (6) stating negative feelings. While this list is most likely
non-comprehensive, it does provide a good starting point for the further analysis
of the content of direct complaints, heretofore not examined in the speech act
literature.

This study also showed that, while direct complaints can and certainly are
utilized in a variety of situations, the most common way of putting forth a
complaint is to do so indirectly, by going "off-record" in Brown and Levinson's
terms. This is no doubt due to the high degree of face-threat inherent in making
complaints, and it sets complaints apart from other speech acts, such as requesting,
apologizing, complimenting, etc., that utilize conventional syntactic formulas
much more regularly.1

The major problem associated with the frequent use of indirectness in
complaint realizations is that it becomes much more difficult to say conclusively
when a complaint has or hasn't been put forth. By using an indirect complaint, the
speaker thus requires the bearer to be more actively involved in the overall

negotiation of meaning within the conversational exchange. The hearer's

 

1 But see Kleiner (1996) for the claim that directness in direction-giving is the least preferred
strategy among middle- (but not working-) class respondents. Status was not a variable in the
current study.

119

interpretation of the utterance, in addition to the speaker's intention, becomes
important in determining when a complaint has been made, and this can
sometimes lead to a mismatch between the speaker's and hearer's understanding of
exactly what has transpired within the exchange.

Because of the importance of bearer interpretation in complaining, we
explored folldinguistic categories of complaints and found that there were a
number of characteristics that helped respondents define a complaint. The aspects
mentioned included the attitude with which the utterance was put forward, the
degree and duration of the statement, the reason it was uttered and the subsequent
effect it had, the amount of specificity it involved, the emotion conveyed, and
language utilized in making the utterance. In terms evaluating utterances as
complaints, we saw three main factors that are employed: diplomacy, boldness and
manner.

Regarding the issue of what people actually say in PC85, we found that
aspects of both the context and participant identities have a significant influence
on whether or not and how complaints are made in response to offensive acts.

This study examined four situations in particular and looked at how a variety of
sociolinguistic factors influenced the choice of complaint realizations. Table 43
presents the overall results of this portion of the data and discusses them in terms

of the hypotheses presented in chapter 2.

120

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

35.4: .5 35.5: 2. 3.5.5: B 35.5: 2 5955
35.8 5 3.5.: N 3.5.2: 2 3.5.: m 2 9 2 39265
3.5.8 8 3.5.: 5 3.5.5: 2 3.5.: e 5 9 2 95
3.5.8 5 3.5.2 5 3.: : 3 3.5.: 5 2 9 a 995 59598
3.5.: 5 3.5.: N 3.52: 2 35.8 5 a 9 a + + e N _ 5
5.55.5: : 35 .5: 2 35.5.: 55 35.5: mm 5955
3.5.: N 3.58 5 3.5.5: 2 35.8 8 2 9 2 5.659
3.5.8 5 3.5.: 5 3.5.: 5 3.5.: m a 9 .2 93
3.5.8 5 3.5.8 5 3.5.8 : 3.5.5: 2 2 9 a 588
3.5.: N 3.58 5 3.5.2: NN 3.5.8 5 a 9 a. I + N 4 m m
3.53 55 35.2 4 35.2.: 2 3.5.52 55 955
3.5.: 5 3.5.8 5 3.5.5: 2 3.5.8 5 2 9 2 9.: a
3.5.: 5 35.8 _ 3.5.: 5 35.8 : a 9 z 5:93
3.5.: 5 3.5.: N 3.52: 2 3.5.8 5 2 9 a 588.8
3.5.: 5 3.5.8 _ 35N: 2 35.: 5 ..N 9 5 + I N 2 - N N
35.5: 55 5.55.: 4 35.5: 5:. 3.5.52 55 5955 .1
35.8 : 3.5.: _ 3.: : 2 3.5.2 5 2 9 2 82293
35.8 8 35.8 5 35.8 5 3.5.5: 2 a 9 .2 as...
3.5.: m 3.58 _ 35.2 : 3 3.5.5: 2 z 9 a :58: 5598
3.5.: e 35.: N 3.5.5: 2 3.5.: 5 a 9 a I I _ m e 2
80.5 touammz—z 32:5:— 25 3O die—:5— mama—=5—
35355 5:33— .? 5:5 "Ev—=3—
:oflafi—aofl «Ea—9.30 new 39cm 3ch 32:35:4— -_m=eamo~— haw—gum Egan—Em

 

 

 

 

 

 

 

 

 

8:535 some 5 28252—52 «Em—Q88 Now $358 =So>O .mv «Bah.

121

The first hypothesis stated that respondents would draw on a variety of
politeness strategies in performing the complaints, and these strategies include the
choice of opting out in addition to the use of mitigation and indirectness. Table
43 shows that this hypothesis is clearly confirmed. Direct complaints were
utilized overall at a rate of only about 20%, and they were not used more than 25%
of the time for any given scenario. The remaining 75-80% of responses were
either indirect, mitigated or a choice to not complain at all. Indirect complaints
were the most frequently utilized, both overall and in each individual situation.
Opting out was the second most common response, while mitigation was utilized
relatively infrequently in these data.

The second hypothesis stated that situations in which the participants are in
on-going ([-distance]) relationships, as well as those that contain a power
differential, will require the greatest amount of face-management. Conversely,
situations with [+distance] and {-power} relationships will demand less facework.
This combination should lead to the greatest amount of attention paid to face in a
[-D, +P] situation, and the least amount of facework in a [+D, -P] scenario.

The situation in this study that represented the [-D, +P] configuration was
the boss/bad review scenario. An examination of Table 43 shows that this is
indeed the situation in which the respondents are most attentive to face concerns.
While they chose to opt out of complaining at the same rate in this situation as in
situations 1 and 2 (and more so than in situation 4), respondents utilized both a

greater amount of indirectness and mitigation as well as a significantly lower rate

122

of directness than in the other scenarios. Furthermore, only two respondents chose
to add the additional face-threat of a request for repair in this situation, as Table 16
illustrates. It is important to note that respondents did not find the offense in this
situation to be particularly severe and they held the boss least responsible for the
offensive act, compared to other situations. These facts certainly contribute to the
overall use of face-management in this situation.

The opposite situation in terms of power and distance in this study was the
stranger cutting in line scenario. It was predicted that respondents would be least
concerned with face matters in this scenario, in which the participants are social
equals in a non-on-going relationship. This should lead to the greatest amount of
directness in complaining. The actual results do not entirely confirm this portion
of the second hypothesis. Rather, it seems that the distribution of complaint
realization choices is ahnost identical to the first situation, which is a [-D, -P]
relationship. In some ways, the fourth situation ([+D, +P]) appears to involve less
face-management than the second. Respondents opted out of complaining to the
adviser at about half the rate that a complaint was made to the stranger in line,
choosing instead to produce ahnost three times the number of mitigated
complaints.

An explanation for this result lies at least partially in the characteristics of
the situations themselves. The offense committed by the adviser was seen to be
the most severe of all the situations, while the friend messing up the kitchen was

not rated particularly severely at all. Furthermore, the friend situation was

123

considered to be most likely to be repaired, while the adviser scenario was rated
least likely on this same scale. These facts, combined with the power and social
distance characteristics of the relationships represented by each situation, worked
together to produce results which make the realization of complaints in the second
situation much more like those of the first and fourth situations.

In addition, these results suggest that social norms require a certain degree
of politeness in interactions as a kind of baseline for socially acceptable behavior.
Factors such as a power differential or lack of social distance may contribute
added degrees of politeness to the exchange, but only in extreme cases will aspects
of the situation override the minimal level of politeness and actually cause
speakers to make use of rudeness conventions. What specifically these aspects
may be is an interesting area for further exploration.

Finally, it is important to note that in at least one way, the second half of
the second hypothesis is indeed confirmed. The results from use of requests for
repair (as seen in Table 17) clearly indicate a willingness to utilize a greater
amount of this kind of face-threat in the second situation than the other three.
Respondents chose to opt out of performing this particular FTA at a rate of only
16.7%, compared to 61.7%, 98.3% and 65% respectively in situations 1, 3 and 4.
Furthermore the use of direct requests for repair was also significantly higher than
in the other situations. This result illustrates the importance of examining the

entire utterance when doing speech act research, rather than simply pulling out for

124

analysis those portions of responses which fit into the category of the speech act
under investigation.

The third hypothesis presented in chapter 3 indicated that gender may
influence the results of the study, both in terms of how complaints are made by
men v. women and to men v. women. As the discussion in chapter 5 showed, this
was in fact the case. More specifically, female speakers tend to opt out of
complaining at higher rates overall than males and utilize a higher percentage of
indirectness as well. Males, on the other hand, use more direct complaints
comparatively. Hearer sex does not seem to be as significant an influence on
complaint realization, although it does contribute some minor effects. It interacts
with power in terms of the choice of opting out, with the greatest number of
respondents opting out of complaining to a female in an equal power relationship,
while the fewest respondents do so when the hearer is a female in a position of
power. Furthermore, hearer sex interacts with both speaker sex and power in the
actual directness levels utilized when complaints are made. Thus the third
hypothesis is confirmed, although gender is not the most influential variable in the

overall results.

6.2 Critiquing Brown and Levinson
While Brown and Levinson's politeness theory has provided a good
framework through which to investigate the speech act of complaining, this study

has illuminated a number of areas of weakness in the theory as well. One of the

125

major critiques of B&L's theory, in fact, has been that it is highly theoretical,
although it purports to be empirically-based by citing examples from three diverse
languages. Many of the empirical studies that have utilized the theory have
attempted to fit their results into the existing theory rather than point out areas in
need of revision or expansion. The purpose of this section is to critique B&L's
theory in light of the results from this investigation of complaints.

The first point worth noting is one mentioned in the first chapter of this
volume. B&L claim that their theory is about "politeness," but they never define
what they mean by this term. It is apparent from their discussion that they take
"politeness" to be the equivalent of "directness," but they give no principled
explanation or empirical evidence as to why they believe this is so. Studies such
as Blum-Kulka (1987) have shown, in fact, that these concepts are not the same;
the most indirect instantiation of a particular speech act is not necessarily viewed
as the most polite way of performing it.

Although the current study did not specifically test this issue, results from
the rating task discussed in Chapter 3 do shed some light on the matter. Recall
that factor analysis of the semantic differential ratings revealed that utterances in
Potential Complaint Situations were evaluated in terms of diplomacy, boldness
and manner. The feature of directness is part of the "boldness" factor group, while
politeness is an important element in the "manner" group (see Table 7). Table 8
and Figure 4 shows how 12 different utterances were judged. Although a few of

the utterances received similar scores for boldness and manner, the majority had a

126

fairly substantial difference between their ratings in these categories. While
"directness" and "politeness" were not being specifically compared, these findings
suggest that for complaints, as in Blum-Kulka's findings for requests, these terms
should not be used interchangeably.

Another problem with B&L's theory is the fact that it is too speaker-
centered. Although in their theory the hearer‘s face is part of what is potentially
threatened and must be maintained in a conversational episode, it seems to be
entirely up to the speaker to assess the amount of face-threat involved with a
particular act and to adjust his or her behavior accordingly. The results of the
current study, however, suggest that with complaints, the bearer is an important
and active participant in the identification of an utterance as a complaint. More
specifically, Figure 3 illustrates the mismatch between respondents' assessment as
to whether an utterance was intended as a complaint and whether it would be
interpreted as one. In most cases, a higher percentage of respondents believed that
the utterance would be interpreted as a complaint than believed that it was
intended as one. Although this result does not specifically spell out the relative
importance of the speaker v. the bearer in the overall negotiation of meaning in
complaint exchanges, it does point out the need to incorporate the bearer into the
analysis to a greater extent that B&L do.

Thirdly, B&L's conception of the weightiness of an FTA is too narrow.
They argue that the assessment of the seriousness of an FTA involves only the

factors of social distance (D), power relations (P), and the absolute ranking (R) of

127

impositions related to the expenditure of goods and services in the particular
culture. They even go so far as to provide a formula for computing the
weightiness (W) of an FTA as: Wx = D(S,H) + P(H,S) + Rx.

While the current study has shown the influence that these factors
(particularly the first two) have on the realization of a complaint, it is quite
apparent that there are other factors that play into this as well. In particular, the
gender of the participants is often a significant factor in how complaints are made,
as are characteristics of the situation such as the severity of the offense, the
responsibility of the offender and the likelihood of correction. As previously
mentioned, it is likely that there are factors beyond these that are influencing how
complaints are realized, but the analysis of these factors alone is enough to show
that an assessment of the weightiness of a given FTA is more complex that B&L's
simple formula suggests.

A further problem with B&L's theory is that it takes an F TA to be the
appropriate unit of analysis for an understanding of how face-management is
accomplished. In other words, it appears that B&L believe either that speaker
utterances never contain more than one FTA or that the amount of face-threat in a
communicative exchange is equal to the sum of the amount of face-threat in each
act that is part of the utterance. Once again, the current study shows the reality to
be far more complex than either of these possibilities.

While the main analysis in this volume centers on complaints, the data in

this study also revealed a large number of another kind of FTA, namely requests.

128

for repair. It would have given an inaccurate picture of the amount of face-threat
and face-management involved in these utterances had the requests for repair been
excluded from the analysis. In particular, Table 19 shows the interaction between
complaints and requests for repair that appears in this data. In some cases, while
there is no complaint present in an utterance, there is nonetheless a request for
repair made at various directness levels (and vice versa). This shows that in
addition to the amount of face-threat, respondents are sensitive to the different
kinds of face-threat that may be appropriate or inappropriate in a given situation.
It also suggests that face-management strategies could come from outside the F TA
itself, by way of other communicative acts in the utterance, although a specific
analysis of these kinds of strategies was not incorporated in the current study.

A related issue has to do with the identity of acts themselves. Again, since
B&L deal with single FTAs as their unit of analysis, there is no discussion in their
work of the possibility of a single utterance functioning as more than one
illocutionary act. In the discussion of indirectness in complaint realizations, this
study demonstrated that an utterance which could be considered a direct
instantiation of a particular speech act (such as statement, request, warning,
compliment, etc.) could nonetheless also be interpreted as an indirect complaint.
The argument here is not that what appears to be one speech act is in reality a
different one; rather, it seems to be the case that an utterance that functions as one
illocutionary act can at the same time function as another. It remains an issue for

further discussion and elaboration as to how precisely the hearer goes through the

129

implication process to recognize multiple functions of a given utterance, but it is
quite clear from the evidence presented in this study that this process is indeed
occurring.

The above critiques of B&L's politeness theory illustrate the importance of
drawing on empirical research to test the validity and assess the strengths and
weaknesses of any theoretical framework. While B&L's work provides a great
deal of explanatory power in an analysis of utterances in Potential Complaint
Situations, the points elaborated above suggest that the theory would be more
useful if its designers would find ways of refining it given the results of various

empirical studies, including the current one.

6.3 Implications for an Understanding of Face-Threat and Face-
Management

One of the stated goals of the current study was to situate an analysis of
complaints within a broader framework of face-management. The results
summarized at the beginning of this chapter offer several new insights into the
construct of facework as it has been previously understood and utilized within the
pragmatics literature.

To begin with, we have seen that not all speech acts are "created equally"
when it comes to the amount of face-threat involved and subsequent face-
management required. While all speech acts can be performed either directly

"on-record") or indirectly ("off-record"), some have more readily-identifiable

130

conventional forms than others. This fact is directly connected to the amount of
face-threat involved with merely performing the speech act at all. In the case of
complaints, we found that the use of indirectness is much more frequent than with
many other speech acts. This makes the analysis of complaints more challenging,
although certainly not less interesting.

While Brown and Levinson suggest an ordering in terms of directness
between redress directed at positive versus negative face needs (with positive
redress being more direct), this study has suggested that positive and negative face
are qualitatively different. Because of this, it is not possible to objectively
measure the amount of face-threat or face-management that corresponds to the
different kinds of face-concerns. Indeed, it may be the case that social and cultural
contexts play a significant role in determining the degree of face-threat involved in
the utterance of a given kind of speech act.

This study, in fact, illustrates the importance of analyzing characteristics of
a situation to determine how they effect the realization of responses within that
situation. In particular, we noted how the evaluation of situations in terms of
severity of offense, responsibility of the offender and likelihood of correction
affected the realization of complaints and requests for repair within Potential
Complaint Situations. It is undoubtedly the case that there are other aspects of
conversational context which affect how complaints are made, not to mention how

other speech acts are realized and face-management accomplished.

131

In addition, the current study has underscored the holistic nature of
discourse that has often been ignored or overlooked in the analysis of speech acts.
While the primary goal of Austin and Searle's pioneering work in this field was to
move researchers beyond a word- or sentence-level analysis towards an
understanding of language as action, quite often studies in this area have isolated
particular speech acts and analyzed their form and content apart from the contexts
in which they occur. This kind of analysis is foundational and thus crucial to the
understanding of a given speech act, but it is not enough. In addition to the form
that it can take, it is important to understand how a particular speech act interacts
with other speech acts in the same utterance or conversation to create the overall
meaning that is being negotiated.

Furthermore, we find that a speech act's very identity is often malleable in
that it can function at a number of directness levels to operate as more than one
speech act at the same time. We saw, for example, that a statement such as "I see
you made yourself at home" is at one level merely a statement, with the
corresponding illocutionary force of stating. And indeed it is possible for a hearer
to choose to interpret this utterance as nothing more than a statement. At another
level, however, we find that this utterance can also be understood as a complaint.
Herein lies the power and flexibility of language to be used not only as a tool for

the conveyance of meaning but also as an instrument of social maneuvering.

132

6.4 Concluding Remarks

While this study is not the first to attempt an analysis of complaints, it has
done so in ways that have addressed a number of shortcomings of previous
research in this area. To begin with, the analysis has proceeded from a specific
theoretical definition of a complaint, rather than merely relying on native-speaker
intuition in determining what should and should not be counted as a complaint. It
has furthermore incorporated folldinguistic understandings of this particular
speech act as a confrrrnation of and supplement to technical concepts. In addition,
while the study retains some of the limitations inherent in using discourse
completion tests as opposed to naturally-occurring data, it has attempted to deal
with past criticisms related to asking participants to respond to unrealistic
situations.

The major contribution that this research makes to the field of language
study is the connections it draws between the more specific concepts of speech act
theory and the broader notions of politeness and facework theory. It has illustrated
the importance of analyzing both the internal content of the speech act and the
external context of the situation to achieve a fuller understanding of how language
is indeed action in the world of conversation. It has also illuminated a number of
both strengths and weaknesses in Brown and Levinson's conceptualization of face-
threat and face-management in communicative exchanges.

This study represents merely the beginning of what could be a broader,

more comprehensive investigation into the speech act of complaining. More

133

specifically, this work has focused on a necessarily homogeneous and well-
defined respondent group as well as a few particular kinds of situations. There is
great potential for expanding this research to include other demographic and
language groups and thus move the research into the realm of cross-cultural

pragmatic studies.

134

APPENDICES

 

APPENDIX A

Interview Questions

 

 

Instructions: You will make a tape recording (on a normal cassette tape) of a
young (18 to 25 year old) Michigan State University student (who is from the
Great Lakes or “Midwest”) and is a native speaker of English.

At the beginning of the tape, state your name and student number, but do
not mention the respondent's name. You should state on the tape, however, the
respondent’s age, gender, and ethnicity. (This is also a good opportunity to
play the tape back and make sure the tape recorder is working.)

After you have put this information at the beginning of the tape, go on to
the interview, which has two parts.

PARTI:

Ask your respondent the following underlined questions (just as they are
written, except as noted):

1) Give me your definition of a complaLin_t.

(Pause for response.)

QWh_at else cam you call it when aperson coglains?

(Pause for response.)

This next part of the interview is a little trickier. You will ask the following
questions about each one of the answers you got to question number 2. For
example, if your respondent said he or she could call a complaint “chewing
somebody out” and “bitching,” then you would ask the following kinds of
questions about both answers:

3) Who can you “chew out”? Who can’t you “chew out”?

(Pause for response.)

4) Where, when. and about what sorts of things can you “chew people out”?
(Pause for response.)

You would then ask these same questions about “bitching.” In your interview,

you will use the actual words or phrases your respondent gave you as answers
to question 2.

 

135

 

APPENDIX A

 

 

Next, you will try to get your respondent to give definitions of each of the
words they gave you as their own responses to question number 2. They can
best do this by comparing the words they gave you to one another. For
example, if you had the responses we mentioned above, you would ask

5) How- exactly, is “chewing out” different from “bitching”? (Of course, you
will use the words your own respondent gave you.)

(Pause for response; then go on to Part 11.)

PART II

In part two we are trying to find out what words (usually adjectives) are used
to describe people and situations. Ifyour respondents ask you to clarify after
you give them the following data to respond to, just say something like “What
words would you use to describe this?” or even “What adjectives would you
use to describe this?”

There are two situations. First, ask the respondent to imagine the following:
Situation #1: You've been waiting line for two hours to buy tickets to a
concert th_at is ahnost sold out. A person cuts in front of the person in front of
You. and the person who got cut in front of sag/g

Excuse me, but the end of the line is back there. We’ve been waitingall day

How would you describe this complaint? How would you describe the person
who made it?

(Pause for response)
.Insteard of that- imagine thngerson 52E
1 think people cattingin line is really rude.

Again, ask your respondent: How would you describe this complaint? How
would you describe the person who mime it?

(Pause for response)

 

136

 

APPENDIX A

 

 

Instead of that, imagine the person said:
Why do you think you can cut in front of me?

Again, ask your respondent: How would you describe this complaint? How
would you describe the person who made it?

(Pause for response)
Instead of thatpimagine the person said:

Who do you think you are?

Again, ask your respondent: How would you describe this complaint? How
would you describe the person who made it?

(Pause for response; go on to the second situation)

Now ask the respondent to imagine the following:

Situation #2: A person has wor1_(§d for their current boss for some time. am!
they get alrmg well. When the employeeis annual performance review comes
around, they find out that their boss has given them a bad review. They do not
think the review is a_fair picture of their work. They go to see the boss and SE

I don’t understand why I received Qad review. I wish that I had been told that
You thoraght my performance was bad.

How would you describe this complaint? How would you describe the person
who anade it?

(Pause for response)
Instead of that. imagine thflerson said:

I was wondering why you gave me a bad review. Can I do anmiag to change
this?
Again, ask your respondent: How would you describe this complaint? How
would you describe the person who made it?

 

137

 

 

APPENDIX A

 

 

Instead of thatgirnagine the person said:

I would lilac to know why you gave me a__bad review. I feel dart I have done
more than enough to make this company happy.

 

Again, ask your respondent: How would you describe this complaint? How
would you describe the person who Image it?

(Pause for response)

Instead of tflt- imagine the person sa_i<_l_:

thdidyvou give me a review like that. I really don’t think that was fair. Why
didn’t you £1th to me. insteard of giving me a_bad review. I thought you would

be cooler than thaa

 

Again, ask your respondent: How would you describe this complaint? How
would you describe the person whmde it?

(Pause for response)

End of interview.

 

138

 

 

APPENDIX B

Sample Questionnaire

 

 

Please provide the following information for classification purposes only. In no way will this
information he used to identifizyou individually in published results of this study.

 

Age: Sex: M / F Ethnicity (circle one):
1 European-American
Highest grade level completed: 2 African-American

 

3 Hispanic-American
4 Asian-American
5 Other

 

Please answer with city and state (and country if not US.) for each of the following questions:

Where were you born?

 

Where did you spend your elementary school years?

 

Where did you spend your high school years?

 

Where did your mother grow up?

 

Where did your father grow up?

 

Has the majority of your life been in a town/city with a population of (check one):
Less than 10,000 people 10,000 people or more
Are you a native speaker of English? Yes No

If "No”. what is your first language? Besides English, what other languages do you know and how
well (check level)?

 

 

Language: Language:
D have studied for _ years El have studied for _ years
13 can read newspaper El can read newspaper
Cl can speak conversationally El can speak conversationally
D can speak fluently El can speak fluently

Please read and sign the consent form on the back of this sheet before answering the questions on
the following pages.

 

139

 

 

APPENDIX B

 

 

Consent Form for Linguistic Study

1, , have consented to participate in a linguistic study involving situations
where complaints are made. I understand that the data obtained from me will be in the form of written
responses to given scenarios and ratings regarding certain beliefs. I understand that all demographic
information about myself in addition to all data I provide is strictly for the purpose of linguistic
research and will be kept confidential. In no way will I be placed at risk in this study. I also
understand that my participation in this study is strictly voluntary and that I can withdraw my
participation at any time during the project, including mid-process. Any questions regarding this
project or inquiries regarding the results of the study may be addressed to: Laura C. Hartley,
Department of Linguistics, A614 Wells Hall, Michigan State University, East Lansing MI 48824;
hartleyl@pilot.msu.edu.

 

Signature

 

140

 

APPENDIX B

 

 

PART I
Directions: You will be presented with four situations. For each situation, do the following:
1. Read the details of the scenario and then write out a script of what you would say in that situation on the blank lines given

below the scenario. Please try to be as realistic as possible; that is, write only what you believe you would actually say,
and not what you would merely think or wish to say.

2. Ifyou wouldn’t say anything in the situation, write “Nothing” on the first line and then on the remaining lines give a brief
explanation as to why you would say nothing.

3. Afier you have completed this task. rate each situation by circling a number on each of the scales which follow it

1. A good friend of yours from high school is visiting you for the weekend. In preparation for his
visit, you clean the kitchen thoroughly. Shortly after your friend’s arrival, you tell him to make
himself comfortable while you run out to the store. When you arrive home, you notice that your
friend has made a huge mess in the kitchen. You say to your friend:

 

 

 

 

 

How severe is this offense?

not severe at all 1 2 3 4 5 very severe
How responsible is your fiiend for this situation?

not very responsible l 2 3 4 5 very responsible
How likely is it that your fi'iend will do something to correct the situation?

not likely at all 1 2 3 4 5 very likely

2. You've been waiting in line for two hours to buy tickets to a concert that is almost sold out. As
you’re standing there, a woman about your age tries to cut in line in front of you. You say:

 

 

 

 

 

How severe is this offense?

not severe at all 1 2 3 4 5 very severe
How responsible is the person who cut in line for this situation?

not very responsible l 2 3 4 5 very responsible
How likely is it that this person will do something to correct the situation?

not likely at all 1 2 3 4 5 very likely

 

141

 

 

APPENDIX B

 

You have worked for your current boss for some time now and you get along well. When your
annual performance review comes around, you find out that he has given you a bad review. You
do not think the review is a fair picture of your work. When you see your boss, you say:

 

 

 

 

 

How severe is this oflense?

not severe at all 1 2 3 4 5 very severe
How responsible is your boss for this situation?

not very responsible l 2 3 4 5 very responsible
How likely is it that your boss will do something to correct the situation?

not likely at all 1 2 3 4 5 very likely

Last year when you were planning out your courses, you went to talk for the first and only time to
the professor who has been assigned to you as your academic adviser, in order to ask her advice on
what courses you should take for your major. After taking several of the courses she said you had
to take, you find out that one of the courses will not count towards your major after all, and that
you will need to take a different course instead during summer school in order to graduate on time.
You go to your adviser’s office during her office hours and say:

 

 

 

 

 

 

How severe is this offense?

not severe at all 1 2 3 4 5 very severe
How responsible is your professor for this situation?

not very responsible l 2 3 4 5 very responsible
How likely is it that your professor will do something to correct the situation?

not likely at all 1 2 3 4 5 very likely

 

142

 

APPENDIX B

 

PART 11
Directions: Imagine that instead of being in the situations in part I, you were instead simply observing
pe0ple in those situations. You will be presented with a number of possible statements that
might be made by people in those situations. For each statement:

 

1. Answer the questions which follow by circling “Yes” or “No".

2 Rate the statement using the word pairs which follow it by placing an “X" in one of the blanks between the two word
opposites. For example, if you thought the statement was quite rude, you would place an “X” in the blank closest to the
word “rude" like this:

Rude X _ __ __ __ Polite
If you thought is was somewhat rude, you would place the “X in the next blank over, and so on.
Rude X Polite

.————._—_—__—_._

Situation A: A stranger cuts in line ahead of someone who has been waiting two hours to buy
concert tickets. The person says:

Response 1: “Excuse me, but the end of the line is back there. We‘ve been waiting all day.”

Would you considered this statement to be a complaint? Yes No

If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

 

 

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate ' Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic ' Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _f __ _ __ __ Non-aggressive Patient _ _ _ __ __ Impatient
Appropriate fl _ f _ Inappropriate Mature pg _ __ __ _ Immature
Angry ________ Not angry Passive ’___ __ _ __ __ Active

C onfrontational Nonconfrontational

 

143

 

 

APPENDIX B

 

Response 2: “I think people cutting in line is really rude.”

Would you considered this statement to be a complaint? Yes No

If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect WWWWW Direct Arrogant flflflflfl Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-a ggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

Response 3: “Who do you think you are?”

Would you considered this statement to be a complaint? Yes No

If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

Rate this statement on the following scales:

I
Rude _____ Polite i Effective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate I Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic I Defensive _____ Not Defensive
Indirect _____ Direct l Arrogant _____ Not Arrogant
Unreasonable A _ _ 7 A Reasonable I Not Assertive _____ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
C onfrontational _____ Nonconfrontational

 

 

144

 

APPENDIX B

 

Situation B: A person has just received a bad performance review from their boss with whom
they have worked awhile and get along well. They go to see the boss and say:

Response 1: “Why did you give me a review like that. I really don’t think that was fair. Why didn’t
you talk to me - instead of giving me a bad review. I though you would be cooler than that.”

Would you considered this statement to be a complaint? Yes No
If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

Rate this statement on the following scales:

Rude _____ Polite Efl’ective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-a ggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

145

 

APPENDIX B

 

Response 2: “I would like to know why you gave me a bad review. I feel that I have done more than
enough to make this company happy.“

Would you considered this statement to be a complaint? Yes No

If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Ineffective
Mean ______ Nice Negative _____ Positive
Legitimate ______ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

Response 3: “I don’t understand why I received a bad review I wish that I had been told that you
thought my performance was bad."

Would you considered this statement to be a complaint? Yes No

If "Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Inefi'ective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _ _ _ _ ___ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non—aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

146

 

 

APPENDIX B

 

Situation C: A person returns from running errands to find their visiting friend has messed up
the newly-cleaned kichen. They say:

Response 1: “I spent all day cleaning the kitchen. I would appreciate it if you would pick up your
mess."

Would you considered this statement to be a complaint? Yes No
If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

Rate this statement on the following scales:

Rude _____ Polite Efl'ective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable ___ fifl _ _ __ Reasonable Not Assertive ______ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

147

 

APPENDIX B

 

Response 2: “Boy, someone was in the cooking mood."
Would you considered this statement to be a complaint? Yes No
If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

Response 3: “What the hell did you do to the kitchen?"
Would you considered this statement to be a complaint? Yes No
If "Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

Rate this statement on the following scales:
Rude ______ Polite Efi'ective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic

‘ Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

 

148

 

APPENDIX B

 

Situation D: A student finds out the advice their adviser gave them regarding classes was wrong
and they will now have to go to summer school. They go to see their professor and say:

Response 1: “Is there any way I can get out of this requirement since you gave me some incorrect
information?“

Would you considered this statement to be a complaint? Yes No
If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

Rate this statement on the following scales:

Rude _____ Polite Efl‘ective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

149

 

APPENDIX B

 

Response 2: “What am I supposed to do now that I’ve wasted my time on unnecessary courses?”
Would you considered this statement to be a complaint? Yes No
If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable Not Assertive _____ Assertive
Aggressive _____ Non-aggressive Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

Response 3: "How could you give me the wrong advice? Aren‘t you supposed to know this stufl?"
Would you considered this statement to be a complaint? Yes No
If “Yes what category of complaint do you think this response best fits in (circle one)?

Whining Bitching Chewing Out Griping Nagging Other

 

 

Do you think the person who made this statement intended for it to be a complaint? Yes No

Do you think the stranger will interpret this statement as a complaint? Yes No

 

 

 

Rate this statement on the following scales:

Rude _____ Polite Effective _____ Ineffective
Mean _____ Nice Negative _____ Positive
Legitimate _____ Not Legitimate Not Sarcastic _____ Sarcastic
Diplomatic _____ Not Diplomatic l Defensive _____ Not Defensive
Indirect _____ Direct Arrogant _____ Not Arrogant
Unreasonable _____ Reasonable 1 Not Assertive _____ Assertive
Aggressive _____ Non-aggressive . Patient _____ Impatient
Appropriate _____ Inappropriate Mature _____ Immature
Angry _____ Not angry Passive _____ Active
Confrontational _____ Nonconfrontational

 

 

 

150

 

REFERENCES

REFERENCES

Austin, J .L. (1962). How to do things with words. Cambridge, MA: Harvard
University Press.

Bach, K. & Hamish, R (1979). Linguistic communication and speech acts.
Cambridge, MA: MIT Press.

Beebe, L.M. & Takahashi, T. (1989). "Sociolinguistic variation in face-threatening
speech acts: Chastisement and disagreement", In M. Eisenstein (Ed), The

dynamic interlanguage: Empirical studies in second language variation (p.
199-218). New York: Plenum Press.

Bergman, M.L. & Kasper, G. (1993). "Perception and performance in native and
nonnative apology", In G. Kasper & S. Blum-Kulka (Eds), Interlanguage
pragmatics (p. 82-107). Oxford University Press.

Bltnn-Kulka, S. (1987). "Indirectness and politeness in requests: Same or
different?" Journal of Pragmatics, 11, 131-146.

Blum-Kulka, S. & House, J. (1989). "Cross-cultural and situation variation in
requestive behaviour in five languages." In S. Blurn-Kulka, J. House & G.
Kasper (Eds), C ross-cultural pragmatics: Requests and apologies (p. 123-
154). Norwood, NJ: Ablex.

Blum-Kulka, 8., House, J ., & Kasper, G. (1989). "Investigating cross-cultural
pragmatics: An introductory overview." In S. Blum-Kulka, J. House & G.

Kasper (Eds), C ross-cultural pragmatics: Requests and apologies (p. l-
34). Norwood, NJ: Ablex.

Blum-Kulka, S. & Olshtain, E. (1984). "Requests and apologies: A cross-cultural
study of speech act realization patterns (CCSARP)". Applied Linguistics, 5,
196-213.

Bonikowska, MP. (1988). "The choice of opting out." Applied Linguistics, 9(2),
169-181.

Borkin, A. & Reinhart, SM. (1978). "'Excuse me' and 'I'm sorry.” TESOL
Quarterly, 12, 57-79.

151

Boxer, D. (1993). Complaining and commiserating: A speech act view of
solidarity in spoken American English. New York: Peter Lang.

Brown, P. & Levinson, S. (1978, I987). Politeness: Some universals in language
usage. Cambridge University Press.

Candlin, C., Coleman, H. & Burton, J. (1983). "Dentist-patient communication:
Communicating complaint." In N. Wolfson & E. Judd (Eds),
Sociolinguistics and language acquisition (p. 56-81). Rowley, MA:
Newbury House.

Clyne, M. (1994). Inter-cultural communication at work: Cultural values in
discourse. Cambridge University Press.

Edmondson, W.J. (1981). "On saying you're sorry." In F. Coulmas (Ed.),
Conversational routine: Explorations in standardized communication
situations and prepatterned speech (p. 273-288). The Hague: Mouton.

Eisenstein, M. & Bodman, J. (1993). Expressing gratitude in American English.
In G. Kasper & S. Blum-Kulka (Eds), Interlanguage pragmatics (p. 64-
81). Oxford University Press.

Eisenstein, M. & Bodman, J. (1986). "'I very appreciate': Expressions of gratitude
by native and non-native speakers of American English." Applied
Linguistics, 7, 167-85.

Fraser, B. (1990). "Perspectives on politeness." Journal of Pragmatics, 14, 219-
236.

Fraser, B. (1980). "Conversational mitigation." Journal of Pragmatics, 4, 341-350.

Goffman, E. (1967). Interaction ritual: Essays on face to face behavior. Garden
City, NY: Anchor Books.

Goodwin, M.H. (1980) "Directive response sequences in girls' and boys' task
activities." In S. McConnell-Ginet, R. Borker, & N. Furman (Eds), Women
and Language in Literature and Society (p. 157-73). New York: Praeger.

Gordon, D. & Ervin-Tripp, S. (1984). "The structure of children's requests." In R.
Schieffelbusch & J. Pickas (Eds), The acquisition of communicative

competence (p. 83-106). Baltimore, MD: University Book Press.

152

Grice, HP. (1975). "Logic and conversation." In P. Cole & J .L. Morgan (Eds),
Syntax and Semantics, vol. 3: Speech Acts (p. 41-58). New York:
Academic Press.

Hartley, LC. (1996). "What’s in a complaint?" Paper presented at NWAV 25
(New Ways of Analyzing Variation), Las Vegas, NV.

House, J. and Kasper, G. (1981). "Politeness markers in English and German." In
F. Coulrnas (Ed.), Conversational routine: Explorations in standardized
communication situations and prepattemed speech (p. 157-185). The
Hague: Mouton.

Kleiner, B. (1996) "Class ethos and politeness." Journal of Language and Social
Psychology, 15 (2), 155-75.

Labov, W. (1972). Sociolinguistic patterns. Oxford: Blackwell.

Leech, G. (1977). "Language and Tact. " Linguistic Agency University ofTrier,
Series A, Paper 46.

Liao, C. (1994). A study of the strategies, maxims, and development of refusal in
Mandarin Chinese. Taipei: Crane.

Meier, A]. (1995). "Defining politeness: Universality in appropriateness."
Language Sciences, 1 7(4), 345-356.

Newell, S.E. & Stutman, R.K. (1989). "Negotiating confrontation: The
problematic nature of initiation and response." Research on Language and
Social Interaction, 23, 139-162.

Olshtain, E. & Cohen, A. (1983). "Apology: A speech act set. " In N. Wolfson &
E. Judd (Eds), Sociolinguistics and language acquisition (p. 18-35).
Rowley, MA: Newbury House.

Olshtain, E. & Weinbach, L. (1993). "Interlanguage features of the speech act of
complaining." In G. Kasper & S. Blum-Kulka (Eds), Interlanguage
pragmatics (p. 108-122). Oxford University Press.

Olshtain, E. & Weinbach, L. (1987). "Complaints: A study of speech act behavior
among native and normative speakers of Hebrew. " In M.B. Papi & J.
Verschueren (Eds), The pragmatic perspective: Selected papers from the
1985 international pragmatics conference (p. 195-208). Amsterdam:
Benjamins.

153

 

Sadock, J. (1974). Toward a linguistic theory of speech acts. New York:
Academic Press.

Searle, J. (1969). Speech acts. Cambridge University Press.
Searle, J. (1979). Expression and meaning. Cambridge University Press.

Takahashi, T. & Beebe, L.M. (1993). "Cross-linguistic influence in the speech act
of correction." In G. Kasper & S. BIum-Kulka (Eds), Interlanguage
pragmatics (p. 138-157). Oxford University Press.

Ting-Toomey, S. (1988). "Intercultural conflict styles: A face negotiation theory."
In Y.Y. Kim & W.B. Gudykunst (Eds), Theories in intercultural
communication (p. 213-235). Newbury Park, CA: Sage.

Tsohatzidis, S.L (Ed.) (1994). Foundations of speech act theory: Philosophical
and linguistic perspectives. London: Routledge.

Weizrnan, E. (1993). "Interlanguage requestive hints." In G. Kasper & S. Blum-
Kulka (Eds), Interlanguage pragmatics (p. 123-137). Oxford University
Press

Wolfson, N. (1981). "Compliments in cross-cultural perspective." TESOL
Quarterly, 15, 117-24.

Wolfson, N. (1989). "The social dynamics of native and normative variation in
complimenting behavior." In M. Eisenstein (Ed.), The dynamic
interlanguage: Empirical studies in second language variation (p. 219-
236). New York: Plenum Press.

154

 

nrcnran smrE UNIV. LIBRRRIES
mII\thmIIIHIIWIMIIHIHIlllllIIIHIHIINIIWHI
31293017743935