LIBRARY

Michigan State
University

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
To AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 campus-m4

 

USE OF SILENCE IN JAPANESE DISCOURSE
By

bhnsuko()nn

hAAHYTEmiiTIESJS

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

MASTER OF ARTS
Department of Linguistics and Germanic, Slavic, Asian, and African Languages

1998

ABSTRACT

USE OF SILENCE 1N JAPANESE DISCOURSE
BY
Natsuko Orai

The present study investigates the use of silence in Japanese private discourse.
In the present study, silence is treated as a component which structures discourse in much
of the same way as speech does. I lmve provided the functional explanations of silence
in private discourse in Japanese, considering variables such as gender difference and the
degree of intimacy: (1) To make an appropriate verbalization, (2) To avoid imposing on
the other speaker’s turn. The findings of this study are summarized as follows. First,
the more intimate the relation is between the speakers, the more silence is used both in
length and in frequency in Japanese private discourse. Secondly, female pairs use long
pauses than male pairs. It is evident that female pairs tolerate pauses compared to male
pairs. In addition, the number of pauses in the conversations of mixed pairs is greater
than those in male pairs, and smaller than those in female pairs. Thus, a female uses
silence less when talking to a male, while a male uses silence more when conversing with
a female. This is considered to be a result because a female does not tolerate silence
when the interlocutor is male, and a male uses silence to avoid imposing upon a female

interlocutor.

Copyright by
Natsuko Orai
1998

ACKNOWLEDGEMENTS

I am greatly indebted to a number of very generous people who supported me
during the various stages of this research. I would like to express my sincere gratitude
to the members of my guidance committee: To Dr. Denise Troutrnan for being there
when I desparately needed her help and her warm encouragement; to Dr. Dennis Preston
for his helpful criticism; and to Dr. Mutsuko Endo Hudson for her help during the course
of this study. Thanks are also due to my friends in Michigan and in Calgary, all of
whom encouraged me in one way or another aiding the completion of this thesis. I
could not have finished this research without their emotional support. Finally, I would
like to dedicate this thesis to my family, my father Noriyoshi, my mother Yasuko, my

brother Taro and my sister Aiko.

iv

Chapter 1

Chapter 2

Chapter 3

Chapter 4

CONTENTS

Introduction

1.1 Outline of the study

1.1.1 Definitions of silence

1.1.2 Type of discourse to be examined

1.2. Theoretical background

1.2.1. Discourse analysis

1.2.2. The framework of the study

1.2.3. Conversation Analysis

1.3. Research questions

1.4. Organization of the present study

Previous studies on silence

2.1. Preliminaries

2.2. The nature of silence

2.3. Psychological views of pausing in narratives

2.4. The interpretation of silence in conversation

2.5. The significance of silence in religious practices

2.6. Gender and silence in public discourse

2.7. Summary

Features of silence in Japanese

3.1. Preliminaries

3.2. Silence as a component of discourse

3.3. Impact of silence in Japan

3.4. Silence as a communicative strategy in business
setting

3.5. Summary

Use of silence in spoken discourse

4.1. Data

4.1.]. Thetypeofdataanalyzed

4.1.2. Procedures for data collection

4.1.3. Interpretation of data

4.2. Methods of analysis

4.3. Results and analysis

4.3.1. The total length of pauses in each conversation

4.3.2. The total number of pauses in each conversation

O‘O‘M-thJUJNNN-d

10
12
12
13
15

16
16
16
18

20
22

24
24
24
24
26
26
27
27
28

4.3.3. The number of tum-intemal and between-turn

pauses 33
4.3.4. The correlation between the number and length
of pauses 33
4.3.5. The correlation between the degree of intimacy
and gender 35
4.3.6. Functions of silence 35
4.3.6.1. Preliminaries 35
4.3.6.2. The environments in which pauses occur 35
4.3.6.3. Functions of silence in Japanese private discourse 47
4.4. Summary 48
Chapter 5 Conclusion 50
Appendices 52

Bibliography 56

vi

Chapter 1

Introduction

1.1. Outline of the study

The present study investigates the use of silence in private discourse in the
Japanese language. Although silence is phonetically null, it is not an empty or
meaningless linguistic phenomenon. In the present study, silence is not considered a
mere absence of speech; rather it is treated as an active ingredient which structures
discourse in much of the same way as speech does.

Silence is prevalent in Japanese conversation, and is considered to be one of its
common communicative strategies. There have been several studies about the use of
silence in Japanese; however, none of them are based on the naturalistic data taken from
private discourse, which is the type examined in the present study. The purpose of this
study is to provide some functional explanations of silence in private discourse in
Japanese and show that variables such as gender difference and the degree of intimacy do

affect the use of silence.

1.1.1. Definitions of silence
“Silence” is defined here as a period within a conversational act in which no
sound is uttered, regardless of its length. In this study, two types of silences are
identified: (1) turn-internal pause and (2) between-tum pause. The latter is fiu'ther
divided into “gap”, “lapse”, and “attributable silence”. This classification of between-
tum pause is based on one by Levinson (1983: 298-299), but is different in its details.
1) A “gap” is a pause that occurs when the current speaker (hereafter C) does
not select the next speaker (hereafter N), and any party, including C, may self-
select and speak.

2) A “lapse” is a pause that occurs when C selects N, and stops speaking, but N
does not take the next turn, and C continues speaking.

3) An “attributable silence” is a pause that occurs between the time when C

selects N and stops speaking and the time when N takes his/her turn.
1.1.2. Type of discourse to be examined

In the past, the cultural significance of silence in Japanese has been discussed in
the field of Japanese literature (e.g. Miyoshi, 1974), anthropology (e.g. Lebra, 1987) and
business discourse (e.g. Yamada, 1992). Researchers have devoted a great deal of
attention to silence and illustrated the features of silence in Japanese from their
respective standpoints. These researchers suggest that the prevalence of silence is
salient in spoken Japanese discourse.

The present study examines data taken from naturally-occurring conversations in
order to provide functional explanations of silence in private discourse in Japanese. For
the purpose of this study, eight conversations involving 16 participants were audiotaped

The participants are native speakers of Japanese in their 20’s and 30’s. They were

3

asked to talk about their childhood dreams in dyadic conversations. At the time of
recording, the focus of the study was not revealed to the participants.

According to Levinson (1983:319) , Conversation Analysis (hear after CA) tries
to model the procedures and expectations actually employed by participants in producing
and understanding conversation. It is therefore considered important to use naturalistic

data to investigate the functions of silence.

1.2. Theoretical background
1.2.1. Discourse analysis

Schifi‘rin (1987:1-2) states that discourse analysis is one of the most vast, but
also one of the least defined areas in linguistics, providing the following definitions of
discourse analysis given by Brown and Yule (1983: 1) and Stubbs (1983a:1).

The analysis of discourse, is necessarily, the analysis of language use. As such,
it cannot be restricted to the description of linguistic forms independent of the
purposes or functions which these forms are designed to serve in human affairs.
(Brown and Yule, cited in Schiffrin l987:1)

Discourse analysis refers to attempts to study the organization of language above
the sentence or above the clause, and therefore to study larger linguistic units,
such as conversational exchanges or written texts. It follows that discourse
analysis is also concerned with language in use in social contexts, and in
particular with interaction or dialogue between speakers.

(Stubbs, cited in Schiffrin in 1987 :1)

Following the definitions given above, Schiffrin (1987: 3) outlines the following
key assumptions in discourse analysis, and contends that context and commmication are
central to discourse analysis.

1. Language always occurs in a context.

2. Language is context sensitive.
3. Language is always communicative.

4

4. Language is designed for communication.
I argue that these key assumptions about language are also applicable to the use of
“silence”. Silence is a highly context dependent phenomenon and therefore it is
important to place context and communication central to the study of silence. In
conversation, much inference is required on the part of the hearer to interpret what the
speaker means by silence. As such, an analysis of the functions of silence needs to
consider the notion of implicature as well. Given that, my key assumptions about
silence is as follows.

1. Silence always occurs in a context.

2. Silence is context sensitive.

3. Silence is always communicative.
The fourth assumption about language proposed by Schiffrin, which is ‘Language is
designed for communication,’ is not applicable to silence because silence is recognized

as a part of language during the conversational interaction. Therefore it cannot be said

that ‘Silence is designed for communication.’

1.2.2. The framework of the study

Schiffiin (1994) views the vastness of discourse analysis not as a weakness, but
as a strength, and as an indication of interest and development. She claims, however,
that the scope of discourse analysis needs to be clarified in such a way that it can
continue dealing with a wide range of problems and phenomena, and in a systematic and
theoretically coherent way. In order to clarify the scope of discourse analysis, she

describes and compares several different approaches to the linguistic analysis of

5
discourse: speech act theory, interactional sociolinguistics, ethnography of
communication, pragrnatics, conversation analysis and variation analysis.

Amongst several approaches to discourse, Conversation Analysis has been
chosen for the present study, because CA focuses on the local management organimtion
of tum-taking and adjacency pairs, in which silence crucially figures.

I investigate the use of silence in Japanese private discourse in the Conversation
Analysis framework, as I believe that a language phenomenon such as silence must be

studied within the contextualized, situated settings.

1.2.3. Conversation Analysis

Schiffrin (1994:232) states that CA offers an approach to discourse that has been
extensively articulated by sociologists, beginning with Garfinkel (1967), who developed
the approach known as ethnomethodology, and Sacks, Schegloff and Jefferson (1974),
who specifically examined conversation, and developed CA. Just as the central concern
for ethnomethodology is to uncover what we know about language, the focus of CA is
placed on the details of actual events. Shiffrin goes on to state that CA considers the
way participants in talk construct systematic solutions to recurrent organizational
problems of conversation. The two basic goals in CA investigation proposed by
Levinson (19832319) are as follows.

(a) We should attempt to locate some particular conversational organization, and
isolate its systematic features by demonstrating participants’ orientation to it.

(b) We should ask, (i) what problems does this organization solve, and (ii) what
problems does this organization raise and therefore what implications does it
have for the existence of further solutions to further problems?

6

This paralles Mendoza-Denton(l995) who claims that to study a particular language
behavior within the context, conducting microlevel analysis along side large scale
statistical studies, enables us to understand the different fimctions and meanings of a
pariticular linguistic behavior as well as the ways in which that linguistic behavior might

fit into the dynamics of a contextualized interaction.

1.3. Research questions
The following are the research questions to be investigated in the present study.

1) What kinds of functions can be observed in the use of silence in
private discourse in Japanese?

2) What role does gender difference play in the use of silence, if any?
3) What role does the degree of intimacy play in the use of silence, if
any?

In the present study, the following three hypotheses are posited

l) The degree of intimacy correlates with the frequency of the use of
silence.

2) Gender difference correlates with the frequency of the use of silence.

1.4. Organization of the present study

The rest of this study is organized in the following way. Chapter 2 reviews
previous studies on silence in general, illustrating a wide range of silence phenomena
from different perspectives. Chapter 3 focuses on the use of silence in Japanese and
reviews relevant literature. First, within chapter 3 a brief history of how silence came to

be recognized as a component of discourse is presented . Then, some aspects of silence

.7

peculiar to such an interdependent society as Japan are reviewed. Finally, the functions
of silence in business discourse are presented. In chapter 4, silence as it occurs in the
natural data taken from eight audiotaped conversations in Japanese are examined in the
framework of Conversation Analysis and chapter 5 summarizes the findings in the

present study.

Chapter 2

Previous studies on silence

2.1. Preliminaries

With regard to silence in general, there have been several studies. As Saville-
Troike (1985:3-4) observes, within linguistics, silence has traditionally been ignored
except for its boundary-marking function, delimiting the beginning and ending of
utterances. The following researchers shed light on the study of silence from their
standpoints by showing that silence is not a simple unit of communication, but is

composed of complex dimensions and structures.

2.2. The nature of silence

Saville-Troike (1985: 16-17) investigates the complex nature of silence in a wide
range of data, from private discourse to public discourse. She contends that
communicative behavior consists of both sounds and silences, and argues that adequate
description and interpretation of the process of communication require that we
understand the structure, meaning, and functions of silence as well as sound. In her

study, the emphasis is placed on the fact that silence can be used to fulfill the functions

 

9

of most speech acts, as well as larger discourse functions. Saville-Troike (198524)
makes a major distinction between the absence of sound, when no communication is
going on, and silence which is part of communication. She goes on to state that silence
which structures communication but serves to organize and regulate its social
relationships is to be distinguished from communicative silence, and that silences, which
do not have propositional contents, are to be distinguished from those which carry
illocutionary force.

Regarding acquisition of silence, that is, how children learn when not to talk,
and what silence means in their speech commrmity, Saville-Troike observes the
following: Children talk more when they are being enculturated into society which places
a high value on individual achievement (e. g., British and American), and less when
family and group’s achievement is more valued (e.g., Chinese and Japanese). Further
she points out that interpretations of patterns of sound and silence are dependent on their
cultural norms. Given that, Saville-Troike (1985: 16-17) offers a broad classification of

silence as follows.

A. Institutionally-determined silence
1. Locational: temples, libraries
2. Ritual (except for duly authorized speakers and occasions):

religious services, legal proceedings, funerals, classes in
school, public performances (operas, movies)

3. Membership: religious groups (nuns, monks) with a vow of
silence

4. Hierarchical/structural: lower-status (less-powerful) members
of society or organization versus higher-status persons

5. Taboo: persons in certain statuses (e.g., chief) or defined
relationships (e.g., wife’s mother) with whom communication
is proscribed

B: Group-determined silence

1. Situational: access to speaking privilege is allocated by group

10

decision, which may be delegated (legislative bodies,

committees)

2. Normative: differential speaking privileges allocated to
individuals or classes of individuals (shunning as punishment,
children, audience members)

3. Symbolic: communicative actions

C: Individually-determined / negotiated silence

I . a. Socio-contextual

1) role-indicative (e.g., auditor in conversation)
2) status-indicative (e.g., deference, superiority)
3) situation-indicative (e.g., context-structuring,
tension—management, social control)
4) tactical-symbolic/attitudinal (nonparticipation,
anger, sorrow, respect, disapproval, dislike,
indifference, alienation, avoidance, mitigations,
concealment, mystification, dissimulation, image
manipulation)
5) phatic (emotional sharing)

b. Linguistic
1) discursive (prayer, fantasizing, rehearsing)
2) propositional (negation, affirmation, consent,

agreement, refusal, acknowledgment)
3) didactic (‘fill in the blank’)
c. Psychological (timidity, embarrassment, fear,
neurosis)
2. Noninteractive
a. Contemplative/meditative
b. Inactive

Although Saville-Troike (1985217) herself does not consider the classification above to
be a definitive framework, but rather a suggestive one, it illustrates the scope of the study

of silence, and suggests potential areas of further study.

2.3. Psychological views of pausing in narratives

The meaning of silence at the micro level of hesitations, or pauses during verbal

ll

interactions, has been examined by a linguist from psychological perspective. Chafe
(1985278) claims that the speaker’s main goal is in pausing to get across what he/she has
in his mind, but not to strive toward grammaticality of his/her utterances. He cites the
following statement by Chomsky (1965z4).

‘A record of natural speech will show numerous false starts, deviations from

rules, changes of plan in mid-course, and so on. The problem for the linguist

[...] is to determine from the data of performance the underlying system of rules

that has been mastered by the speaker-bearer and that he puts to use in actual

performance’
Chafe finds that this view is misleading in that it suggests that the speaker has some
grammatical ideal toward which he is striving, and which he is often prevented from
attaining because of various imperfections in the psychological processes involved in
‘performance’. He proposes on the contrary that the speaker is interested in the
adequate verbalization of his/her thoughts. In that sense, pauses, false starts,
afierthoughts, and repetitions are considered to he steps on the way to achieving it. He
states that hesitation phenomena provide good evidence that speaking is not a matter of
regurgitating material already stored in the mind in linguistic form, but that it is a
creative act, relating thought and language, which require adjustments and readjustments
to each other. In his study, he distinguishes in-tum pauses, which occur within the
conversational turn of a speaker, from switching pauses, which occur at the boundaries of
speaking turns. He examines hesitations in ‘the pair stories’ narratives told about what
happened in a short film. His findings show that hesitations may reflect mental search

processes in the speaker’s attempt to construct a description of an event from memory.

He concludes that a considerable proportion of hesitating stems from the need to

12

verbalize something that is low in codability.

2.4. The interpretation of silence in conversation

Tannen (1985) offers an analysis of the conversation style of three New Yorkers
of Eastern European Jewish background In discussing the features of the conversation,
she considers how their attitudes toward silence are accommodated and reflected in
conversational interaction. She finds that in the conversation with New York Jewish
people, it is important to show their high involvement to avoid silence which is seen as a
lack of rapport. Accordingly, high involvement (e.g., overlapping) is valued, but silence
has a negative value in their conversation. Tannen states that silence has both positive
and negative values depending on the context in which silence occurs. In terms of
positive values, she illustrates the two major benefits: rapport and defensiveness.
According to Tannen, the former comes from being understood without verbalization,
and the defensive value of silence comes from omitting something to say. She views
silence has, as positive value, as a way of serving negative politeness which is not
imposing others. On the other hand, silence is negatively valued when it is too long or

appears to be at the wrong time and in a wrong place.

2.5. The significance of silence in religious practices
Maltz(1985:1 13-137) consider silence and noise to be purposeful means in
religious practices. He compares attitudes toward silence and noise in worship among

Pentecostals and Quarkers, and investigates the variations in the interpretation of noise

13

and silence. According to Maltz Quakers streess the ‘silence’ of inner religious
experience in reponse to the superficiality and externality of mere talk. Pentecostals, in
contrast reponeded to the spiritual silence of a religiously indifferent world by advocating
the making of a ‘joyful noise.’ Noiseness is practically a defining characteristic of
Pentecostal worship, and silence is that of Quakerism. He finds that the silence of the
Quakers and the noisiness of the Pentecostals serve similar religious functions such as an
expression of committment and joy , participation in congregational worship and giving
free rein to the Holy Sprit He concludes that the potential meanings of noise and
silence are not arbitrary; rather they are constrained in their meanings by their inherent
logics. What varies are the contexts in which the symbols ‘silence’ and ‘noise’ are
interpreted, the situations in which they are applied, and the seriousness with which

multiple meanings are explored.

2.6. Gender and silence in public discourse

From the perspective of gender study, Lakoff (1995) and Mendoza-Demon (1995)
have made a contribution to the study of silence, particularly that observed in public
discourse. Lakofi‘ ( 1995) emphasizes that there is an advantage of analyzing the
meanings and the functions of silence from both semantic and pragmatic perspectives.
A semantic perspective helps to analyze ‘attributable silence”, while a pragmatic one is

useful in analyzing ‘silencing’. Further, silencing is divided into private (or

 

1 According to Lakofl‘s classification (1995:26), gaps, lapses and pauses are all ‘attributable
silence’.

l4
conversational) and public (or persuasive). Lakoff offers an analysis of various types of
discourse in public settings (e. g. courtrooms, workplaces, political arenas and media
settings). She emphasizes that the superficial form of silencing that occur in the private
discourse may also occur in public settings, however, although the form may be identical,
the function and consequences are often more serious for the silenced person, and more
permanent in public sinece it it expected to have effects beyond that of mere socialization.
Specifically, she analyzes the following cases.

1) the Clarence Thomas confirmation hearings and their sequels

2) the role of Hillary Rodharn Clinton as presidential candidate’s wife and the

first lady

3) the Babbit affair

4) the Nancy Kerri gan-Tonya Harding interaction

5) the OJ. Simpson murder case
She observes different types of silencing, such as interruption, topic control and
nonresponse. She points out that these pragmatic silencing acts show male dominance
in the conversation although some of those acts are very subtle.

Mendoza-Demon (1995) examines the use of silence during the Anita Hill-
Clarence Thomas hearings. The hearings reveal complexity of issues including race,
gender, sexuality, and the power that has affected the way in which politics is conducted.
She views the hearing as a site for the reproduction of power relations. Considering the
complexity of issues revealed in the hearings, she emphasizes that language behavior
should be studied within contextualized, situated settings, conducting microlevel analysis

alongside large-scale statistical studies. She contends that it enables us to understand

the different functions and meanings of a particular interruption, for example, as well as

15

the ways in which that interruption might fit into the dynamics of a contextualized
interaction. Focusing on how microlevel linguistic features worked subtly to construct a
discourse that enraged many of its listeners, she finds that Thomas’s statements were
periodically followed by “pregnant pauses and poignant silences”. According to
Mendoza-Denton(1995:54) Thomas’s discourse setting was characterized by 1) numerous
and lengthy expressions of support and sympathy from the senetors, 2)questions that were
answerable with a simple ‘yes, Senetor’ or ‘no, Senetor’ and 3) weighty, respectful
silences following his hyperbolic turns of phrase. Further, she argues that during the
hearings speech is always coconstructed so that the silences are not “owned” by any
single person. She concludes that the silences in this political discourse are complex

and connected.

2.7. Summary

In this chapter several previous studies on silence were reviewed The above
discussions of research on the topic is not considered to be a comprehensive one, but they
illustrate a wide range of silence phenomena from several levels of linguistic analysis:
psycholinguistics, conversational analysis, pragrnatics, and sociology of language. It is
also aimed to show how complex the nature of silence is and that it is a worthwhile

object of linguistic investigation as silence always means and communicates something.

Chapter 3

Features of silence in Japanese

3.1. Preliminaries

As described in chapter 2, past studies show that silence is not a simple unit of
communication, but is composed of complex dimensions and structures. In this chapter,
the focus will be placed on some of the aspects of silence that are characteristic of
Japanese. I discuss the features of silence in Japanese in terms of the following: 1)
recognition of silence as a component of discourse, 2) impact of silence in Japan, and 3)

silence as a communicative strategy in business settings.

3.2. Silence as a component of discourse

The recognition of silence as a component of discourse emerged relatively
recently in the literature. Miyoshi (1974) has noted that there are four phenomena about
the Japanese language. First, there is a noticeably large gap between the spoken and the
written language. The former is called Koogo, or vernacular language which was
considered an inappropriate medium for literature. The latter is Bungo, or literary

language, which was well-suited for writing. Second, the use of the honorific language

16

l7

promotes people’s perception of the relative social standings of the speaker, listener and
referent. Third, what he calls the “vague” tense system and “loose” syntactic form
allow for subtlety and ambiguity. Fourth, according to Miyoshi the “typical Japanese
dislike” of verbal communication causes a certain type of style such as ‘Translation style’,
which tends to be stiff and far from natural conversation. He observes that since the
Japanese culture is “primarily visual”, and not verbal in orientation, the social decorum
rewards reticence, but not eloquence. In his opinion, Haiku, which is one type of
Japanese poetry consisting of seventeen moras, is the perfect embodiment of the subtlety
of silence. He claims that silence is a powerful compulsion throughout the whole
society, and, in contrast, writing in Japanese is always something of an act of defiance.
Writing in K oogo, the way people talk, was indeed a great attempt against the written
standard during the early Meiji era

When Japanese writers made an attempt to reproduce the use of silence in
conversation, they wrote ‘ ...... ’ , ‘silence’ or ‘pause’ in Japanese. Saville-Troike (1985)
cites Hokari (1980), who shows the frequent use of silence markers in Japanese literature.
For example, ‘ ...... ’ is used 173 times in 103 pages in Kaze Tachinu by Tatsuo Hori, and
in K igadoomei by Kobo Abe, it is used as frequently as 15 times per page. This usage
shows that the writers were aware of the prevalence of silence in natural conversation.
The silence marker ‘ ...... ’ enabled them to reproduce natural conversation more

accurately.

18

3.3. Impact of silence in Japan

Lebra (1987) investigates intended and perceived meanings of silence in
Japanese communication, as well as its cultural values. She proposes four dimensions
of silence that are culturally salient, though they may be mutually contradictory; i.e.,
truthfulness, social discretion, embarrassment and defiance.

First, according to Lebra, the Japanese view themselves as split into inner and
outer parts: Truth lies in the inner part and in her view “truthfulness” resides in silence.
In contrast, the outer self consists of face, mouth and spoken words. To illustrate that a
person of few words is trusted in Japanese society, she provides proverbs of warning
about the inner—outer duality of a speaker.

K uchr' m' mitsu ari, hara ni ken ari.

Honey in the mouth, a dagger in the belly.

Aho no hanashr‘ gui.
A fool eats (believes) whatever is said

Hanashi hanbun.
Believe only half of what you hear.

Bigen shin narazu.

Beautiful speech lacks sincerity.
She observes that implicit in these cautions against spoken words is the image of a
trustworthy person characterized as kuchi gatai (hard-mouthed).

Second, “social discretion” refers to the use of silence when it is considered
necessary or desirable in order to gain social acceptance or to avoid social penalty.
Discretional silence requires knowledge of what can and cannot be said to whom in what

situations. Lebra also cites proverbs concerning “social discretion” .

d:

hr

l9

Iwanu ga hana.
Better to leave things unsaid.

K uchr' wa wazawor‘ no molo.

The mouth is the gate of trouble (talking causes trouble)

K r'jr'r' mo nakazuba utaremajr'.
Ifthe pheasant did not sung, it would not be shot.

Mono ieba kuchr’br'ru samushi aki no kaze.

If you talk, your lips will feel cold ( It is safer not to talk).

While the first dimension of silence attaches “truthfulness” to silence, and falsity to
spoken words, the second dimension explains why silence is sometimes associated with
social masking of conflict or embarrassment, for instance. Thus the first and the second
dimensions function differently.

Lebra’s third aspect of silence refers to “embarrassment’. It can be found
particularly in conjugal relationships where intimacy exists. Silence here is considered
to be one of the strategies that allows people to avoid embarrassment when talking about
love. She illustrates conjugal silence as follows.

I am referring to the husband and wife who are in love but too embanassed

to express their feelings in speech. Embarrassment extends to address

terms, including personal names (until recently) for a spouse, so that husband

and wife may end up terrnless in addressing one another.

Lebra contends that intimacy inhibits verbalization of emotion, just as social distance
demands discretional silence.

Lebra’s fourth dimension refers to the use of silence expressing estrangement,

hostility, or “defiance”. According to Lebra, in order to say “I disagree with you’, ‘I

object’, ‘I am angry with you’, or ‘I hate you’, one may invoke silence, usually

20

accompanied by some facial cues. She claims that this type of silence is considered to
be the logical inverse of the second type which is a means of creating or maintaining
sociability, and also of the third type which is a sign of embarrassment emanating from
intimacy.

As shown above, silence is not only polyfunctional but can express opposite
meanings or emotions. Lebra (1987) cautions that this may cause confusion and
misunderstanding to a cultural outsider, but to a native speaker as well. In addition, she
points out that in actual conversation, social hierarchy plays a key role in the distribution
of speech and silence. Lebra’s study is insightful as well as informative. It illustrates
the features of silence which are considered to be culturally-specific. The functional
explanations for each dimension of silence would be strengthened if the data taken from
naturally-occurring conversations are provided, which is what I attempt to do in section
4.3.6.

3.4. Silence as a communicative strategy in business setting

Yamada (1992:81-83) offers an analysis of silence as a communicative strategy
in Japanese business meetings among three bank officers. Her study shows that long
silences (pauses greater than 1.5 seconds) occur more frequently in the Japanese meeting
than in the American meeting. Below is Yamada’s summary of the use of silence in the
two business discourses.

Table 3.1 Use of silence in the American and the Japanese discourses

 

 

 

 

 

 

 

 

[participants Japanese American
[Eta] number of pauses 103 20

kotal length or pauses 107.45 sec / 20 min 41.2 sec / 27 min.
Igvergge length of pause 5.15 sec 0.74 sec
[longest pause used 8.2 sec 4.6 sec

 

1n additi
silences

from (h

Yam:

1h.

21

In addition to the results above, she illustrates how the series of topics are punctuated by

silences despite the use of talk throughout the meeting. The following is an example2

from the Japanese business meeting (Yamada: 82).

(1) Yamada’s example 3.7

Ikeda: nihon da to saikoo issh- isshuukan deshita kara>
Japan is if at most one-week- one-week it was so

Shimizu: n> isshuukan desu yo>
mhm a week it is (E)

..-> {8.20}

Tanaka: zenzen hanashi ga chigaun desu kedo> kondo
not-at-all talk (S) different is but next-time
mata beishuu kaigi arun desu yo ne<
again American-States meeting have is (E) (PF)
hachigatsu goro ni>
August around (TFN)

Ikeda: Because in Japan it was a week at the most.

Shimizu: Mhm, it’s A WEEK.
1820}

Tanaka: This is about something completely different, but there will

be another regional meeting around August.

Yamada states that a topic shift is identifiable by the explicit mention of that shift in the

 

2 The transcription convention in Example 3.7 is one by Yamada (1992: xxi).

> falling intonation
< rising contour

- clipped spwch

{ } discernible pause
CAPS emphatic stress

hold point of study

B

S
PF
S
TFN

final particle

time-flame (ni)

22

subsequent topic opening. According to Yamada, that is where “Tanaka” says, “this is
about something completely different but,” and then goes on to open the next topic on the
regional meeting in August.

Based on the results above, Yamada argues that the Japanese participants shift
topics through long silences and take a longer time to complete the shift compared to the
American participants. Further, she contends that these long silences enable the
participants to obscure the topic endings. She claims that the implicit topic-shifting
strategy has the important twofold function of not only drawing the attention of meeting
members away from “what was being talked about” but also from “who was talking
about it”. Yamada (1992:83) concludes that “obscuring talk” is a necessary strategy in
Japanese conversation because talk is viewed unfavorably. I, however, feel that her
reason for this obscurity may be too impressionistic to support her claim. In addition, a
question arises with regard to her view that the participants intended to make use of the
twofold function during the meeting. Her claim would be strengthened if an
explanation is provided on what motivates such a communicative strategy as “obscuring

talk”.

3.5. Summary

This chapter illustrated several aspects of silence in Japanese. The issues
addressed were the importance of recognizing silence as a component of discourse, the
impact of silence in an interdependent society such as Japan, and silence as a

communicative strategy. The recognition of silence as something that structures

C0111
K00;
1mm

net

5113
lilfii

116C

in“

and

23

conversation could not have emerged if Japanese writers did not attempt to write in
Koogo, vernacular language. The use of silence in Japanese is best understood with the
knowledge of Japanese society where harmony is emphasized. From an anthropological
viewpoint, Lebra (1987) proposes four dimensions of silence in Japanese; truthfulness,
social discretion, embarrassment and defiance. The use of silence as a commtmicative
strategy in the Japanese business meeting differs greatly, from that in the American
meeting. Yamada (1992) contends that “obscuring talk” associated with long pause is
necessary in the Japanese discourse to draw the attention of meeting members away from
“what was being talked about” as well as “ who was talking about it”.

In chapter 4, use of silence in naturally-occurring conversations will be
investigated. My spoken data will be examined in terms of the length, the frequency

and functions of silence.

 

Chapter 4

Use of silence in spoken discourse
4.1. Data
4.1.1. The type of data analyzed
The data collected for this study contain conversational interaction between
eight pairs of native Japanese Speakers, which was audio-recorded with no outsider
present The subjects were provided with a topic of “childhood dreams” and asked to

talk as naturally as possible, with no other directions given.

4.1.2. Procedures for data collection

The procedures for data collection in this study were based on those by Maynard
(1989:15-18). All the data were collected as follows:

Participants:

The subjects were native speakers of Japanese and in their 20’s and 30’s. Each
conversation was performed by two speakers, totaling 8 pairs, involving 16 speakers.
Eight pairs were formed and each participant was labelled as follows: A-B, C-D (Female-
Female), E-F, G-H (Male-Male), I-J, K-L, M-N, O-P (Female-Male). All were graduate

students at an American university except for M and 0, who were housewives. M and

24

25

N, and O and P were married to each other, while other pairs were close fiiends. Paired
subjects had conversed with each other numerous times prior to the time of my data
collection. Subjects were informed of the general purpose of the research, that was,
Conversation Analysis. They were recnrited through a network of college friends, and

they voluntarily participated in the study.

Setting:

The focus of the research was not revealed to the participants prior to the taping.
Although some crucial sociolinguistic information was known prior to the recording,
subjects were asked to fill out forms, which asked for demographic and other relevant
information, such as age, sex, educational background . (A list of subjects with
demographic information is provided in Appendix A, a consent form in Appendix B and
a respondent identification in Appendix C.)

Two tape recorders were used at the time of recording: one with a voice
activated microphone, and the other without. This was done as it was helpful to
measure the total length of pauses which were observed in each conversation, and to
count the number of pauses. The tape recorders were placed in front of the subjects.
The subjects sat across a table from each other. They were given instructions initially,
and asked to talk about their childhood dreams as naturally as possible. The genre of
the data was casual conversation, but it was somewhat serious in the sense that
participants were aware of being taped. Recordings lasted 30 minutes for each pair.

The first three-minute segment of each conversation was not included in the present

26

study as the subjects might have been conscious of the tape recorder.

4.1.3. Interpretation of data

In this study, analysis of the data was made solely by the researcher without
soliciting the participants’ opinions about what they experienced at the time. As
Maynard (1989215) cautions, “while subjects’ reflections and interpretations may be
useful, incorporating their input into the data with equal weight without being subjective
would require the kind of care which is difficult to achieve”. The data were analyzed by
measuring the number and the length of pauses, and transcribing the conversations
recorded The key transcription methods and conventions are based on those by
Maynard (1989: 17-18), but different in details. (A of the key transcription methods and

conventions is provided in Appendix D.)

4.2. Methods of analysis
In the present study, pauses greater than 0.5 second are analyzed. First, the

data are examined in terms of the following.

1) The total length of pauses in each conversation

2) The total number of pauses in each conversation

3) The number of tum-intemal pauses and between-turn pauses
4) The correlation between the number and the length of pauses
5) The correlation between the degree of intimacy and gender

Second, based on the results given above, speculations are made about the
functions of silence in Japanese private discourse. The relevant examples are provided

with transcriptions in explaining the functions that are found in the data.

27

4.3. Results and analysis
4.3.1.The total length of pauses in each conversation

In this study, pauses longer than 0.5 second are considered significant, and
thoese pauses less than 0.5 second are not included in the data3 . The pause lengths are
totaled cumulatively for each conversation. Table 4.1 shows the total length of pauses.
The figure in the parentheses indicates the ratio of silence in the whole conversation of
27 minutes long.

Table 4.1 The total length of pauses

 

 

 

 

 

 

 

pair participants sex relationship total length ofpauses
l A-B F—F friends 02:17:10 (8.4%)
2 OD F-F fiiends 01 :05:50 (4.0%)
3 BF M-M friends 00:12:10 (0.7%)
4 G-H M-M friends 00:12:05 (0.7%)
5 I-J F-M friends 00:37:50 (2.3%)
6 K-L F-M fiiends 00:24:80 (1.5%)
7 M-N F-M married 03:01 :30 (l 1.2%)
8 OP F—M married 06:31 :10 (24. 1%)

 

Female pairs use more silence than male pairs in conversation between friends.
Specifically, while the total length of silence in each of the male pairs is 12 seconds,
which is less than 1 % of the whole time of the conversation, that in the female pairs 1
and 2 are 2:17:10 (8.4%) and 1:05:50 (4 % ), respectively.

In the mixed-sex conversations, the total pause length are 00:37:50 and 00:24:50,
approximately 2 % in both pairs 5 and 6. In other words, the length is greater than that

of male pairs 3 and 4, and less than female pairs 1 and 2. This suggests that a kind of

 

3 The voice activated microphone does not catch pauses less than 0.5 second.

 

28

adjustment occurs more with respect to the length of silence when participants converse
with those of the opposite sex.

The comparison of pair between 5 and 6, and between 7 and 8 shows that long
silence occurs between individuals married to each other than friends. Especially, in the
conversation in pair 8, the total length of pauses is 6.5 minutes, which is approximately
24% of the whole conversation. It is the longest pause in my entire data.

Those results show that not only gender difference, but also the degree of

intimacy are crucial factors in terms of the total length of pauses.

4.3.2. The total number of pauses in each conversation

As the pauses in each conversation range in length, I classified them into a
“short” pause (greater than 0.5 second, but 1.0 second or less), a “substantial” pause
(greater than 1.0 second, but 1.5 seconds or less), and a “long” pause (greater than 1.5
seconds) based on the classification byYamada (1992: 96). In the Table 4.2, the figure
in the parentheses indicates the percentage of each type pause in each conversation.

Table 4.2 The total number of pauses

 

 

 

 

 

 

pair participants sex relationship short pause substantial long total
0.55 x<l.0 pause pause number
LDS x<l.5 1.55 x of mum
1 A-B F-F friends 22 (25%) 28 (32%) 37 (43%) 87
2 OD F-F friends 25 (58%) 10 (23%) 8 (19%) 43
3 E—F M-M friends 2 (25%) 2 (25%) 4 (50%) 8
4 G-H M-M friends 3 (27%) l (9%) 6 (54%) ll
5 H F-M friends 1 (6%) 4 (25%) 11 (69%) 16
6 K-L F-M friends 7 (35%) 10 (50%) 3 (15%) 20
7 M-N F-M married 7 (8%) 21 (23%) 61 (69%) 89
8 OP F-M married 0 (0%) 9 (14%) 62 (86%) 72

 

 

29

The total numbers of pauses in conversation of female pairs 1 and 2 are 87 and
43, respectively, whereas, those of male pairs 3 and 4 are 8 and 11, respectively.

Female pairs 1 and 2 use silence more frequently compared with male pairs 3 and 4 (X2
=82.691 P<0.01).

Mixed pairs friends 5 and 6, used 16 and 20 pauses in their conversations,
respectively. These numbers are greater than the ones in male pairs and less than the
ones in female pairs (X2 =16.221 P<0.01). It is possible that a kind of adjustment
occurs in the use of silence when participants converse with partners of the opposite sex.

A comparison between mixed pairs 5 and 6, and married couples 7 and 8 shows
that married couples use silence more frequently (X2 =79.315 P<0.01). The total
numbers of silence in conversation of married couples 7 and 8 are 89 and 72, respectively,
while those of mixed pairs 5 and 6 are 16 and 20, respectively. In terms of the duration
of pause, the use of “long” pauses (greater than 1.5 seconds) is remarkable in both
married couples. Pair 7 used 61 “long” pauses, which is close to 70 % of the pauses
found in this conversation. In the case of pair 8, 62 “long” pauses are observed This
includes a 31- second pause, which is the longest found in my entire data.

The results above confirm that gender and the degree of intimacy are both

significant factors that influence the frequency of pause in Japanese conversation.

30

4.3.3. The number of turn-internal and between-turn pauses

In this section, the pauses in the data are further classified into tum-intemal and
between-tum types. First, turn-internal pauses are examined in terms of the total
number and the number of turn-internal pauses owned by each participant. An analysis
of between-tum pauses is then offered concerning the distribution of three types: a “gap”,
a “lapse,” and an “attributable silence”. In the present study, three types of between-
tum pauses are defined as follows.

1) A “gap” is a pause that occurs when the current speaker (hereafter C) does
not select the next speaker (hereafter N), any party, including C, may self-select

and speak.

2) A “lapse” is a pause that occurs when C selects N, and stops speaking, but N
does not take the next turn, and C continues speaking.

3) An “attributable silence” is a pause that occurs between the time when C

selects N and stops Speaking and the time when N takes his/her turn.

The number of turn-intemal pauses and the number of pauses owned by each
participant are summarized in Table 4.3. The figure in the parentheses indicates the
ratio of the turn-internal pauses to the total number of pauses in each conversation.

Table 4.3 The number of turn-internal pauses

 

 

 

 

 

 

pair participants sex relationship number of total number of
pauses owned turn-internal
pauses
l A-B F-F friends 29-32 61 (70%)
2 C-D F-F friends 8-1 1 20 (47%)
3 E-F M-M friends 1-1 2 (25%)
4 G-H M-M friends 1-2 3 (27%)
5 I-J F -M fiiends 4-3 7 (44%)
6 K-L F -M friends 3-5 8 (40%)
7 M-N F-M married 6—47 53 (60%)
8 O-P F-M married 11-6 17 (24%)

 

 

31

Female pairs use turn-internal pauses more frequently compared with male pairs
(X2 =68. l4 P<0.01). The total number of between-tum pauses in the female
conversations l and 2 are 61 and 21, respectively, while those of male pairs 3 and 4 are 2
and 3, respectively. With regard to the ratio to the total number of pauses, the greater
percentage was observed in female pairs.

The total number of turn-internal pauses in the female pairs 1 and 2 are 61 and
21, respectively, while those of male pairs 3 and 4 are 2 and 3, respectively. The total
number of turn-internal pauses in the mixed pairs 5 and 6 are 6 and 8, respectively. The
results shows that the number of turn-internal pauses in mixed pairs is greater than that in
male pairs and less than that in female pairs (X’=17.364 P<0.01).

In comparison with married couples 7 and 8, mixed pairs of friends use a
smaller number of turn-internal pauses (X’=28.594 P<0.01). In terms of the number
of pauses owned by each participant, participant N stands out. Although it is not clear
what motivates him to use turn-internal pauses frequently, it is possible that the pauses
help them find words or recall something. In order to support this claim, the
environments in which turn-internal pauses occur is examined in relation to the functions

of silence in section 4.3.6.

32

Table 4.4 shows the number of each type of between-tum pauses found in each

conversation. The figure in the parentheses indicates the ratio of the total number of

between-turn pauses in each conversation.

Table 4.4 The number of between-turn pauses

 

 

 

 

 

 

pair participants sex relationship gap lapse attributable total number

silence of between-

turn pauses
1 A-B F-F friends 16 (62%) 2 (8%) 8 (30%) 26
2 C-D F-F friends 13 (59%) 2 (9%) 7 (32%) 22
3 E-F M-M friends 5 (83%) 0 (0%) l (17%) 6
4 G-H M-M friends 8 (100%) 0 (0%) 0 (0%) 8
5 I-J F-M friends 8 (70%) 1 (10%) 1 (20%) 10
6 K-L F-M fiiends 6 (50%) 1 (8%) 5 (42%) 12
7 M-N F-M married 34 (94%) 1 (3%) 1 (3%) 36
8 O-P F-M married 46 (85%) l (2%) 7 (13%) 54

 

Female pairs use between-tum pauses more frequently than do male pairs (X2

=18.645 P<0.01). The total number of between-tum pauses in the female

conversations 1 and 2 are 26 and 22, respectively, while those of male pairs 3 and 4 are 6

and 8, respectively.

This parallels the finding that female pairs used turn-internal

pauses more frequently that did male pairs on page 31.

Mixed pairs 5 and 6 used 10 and 12 between-tum pauses, respectively.

These

numbers are greater than those in male pairs (pair 3=6, pair 4=8) and less than those in

female pairs (pair 1=26, pair 2=22); however, a chi-square test indicates that these

numbers are not significant in showing that gender of interlocutor is a variable in the use

of between-turn pauses (X2=2.0717 P>0.05) .

In terms of mixed-pair conversations, married couples 7 and 8 use between-turn

pauses more frequently than friend pairs 5 and 6 (X2 =41.28 P<0.01).

With regard to

 

33

the types of between-tum pauses in each conversation, the following is observed.
Regardless of the difference in the nature of relationship or gender, a gap occurs more
frequently than a lapse or an attributable silence in each of the conversations. It is
plausible that the frequent occurrence of gap is a result of negotiation in choosing the
next speaker. In order to support this claim, the environment in which between-turn

pauses occurs is examined in relation to the functions of silence in section 4.3.6.

4.3.4. The correlation between the number and length of pauses
The findings regarding the correlation between the number and the length of

pauses are as follows.

1) When the total number of pauses is greater, the total
length of the pauses is longer. (See Table 4.1 and Table 4.2)

2) In conversations of pairs who use the greater number of pauses, the
frequent use of long pauses (greater than 1.5 seconds) is observed.
(See Table 4.2)

Although it is not clear why there was not evidence of a case in which the number of
pauses is small, and the length of pause is long, it might suggest that the number and the

length of pauses trigger each other during the conversation.

4.3.5. The correlation between the degree of intimacy and gender
The correlation between the degree of intimacy and gender are analyzed below.
1) A comparison between mixed pairs who are friends and those of
married couples showed that the degree of intimacy is a significant
factor in the use of silence both in length and frequency.
(See p.28 and p. 29)

2) Female pairs use more silence both in length and in frequency.

34
(See p. 27 and p. 29)
3) A comparison between pairs of the same sex and pairs of the opposite
sex shows that the gender of the interlocutor is a significant factor in
the use of silence in length. (See p.27 and p.28)

First, it is interpreted that the closer the relationship is between the speakers
results in more pauses that occur in relation to the use of silence in conjugal relationships
as proposed by Lebra (1987), who argues that silence is used to avoid the embarrassment
of talking about love. The first analysis allows one to say that married couples use silence
more regardless of the topic of conversation in which they are engaged.

The second analysis offers two interpretations: One is that female pairs are more
intimate and thus are able to tolerate more silence during conversation. The other is
that female pairs use silence to be polite as a way to avoid imposing on each other’s turn.
According to Tannen (1985), silence is positively valued when it is used such as in the
latter case as a negative politeness. This, however seems not to be applicable to
conversations of married couples where politeness is not as strongly required.

Finally, the third analysis indicates that a female uses silence less when talking
to a male, while a male uses silence more when conversing with a female. This is

considered to be a result because a female does not torelate silence when the interlocutor

is male, and a male uses silence to not impose upon a female interlocutor.

35

4.3.6. Functions of silence
4.3.6.1. Preliminaries

In the previous section 4.3.1. , it was speculated that turn-internal pauses helped
participants find words or recall something. Also, it was posited that the frequent
occurrence of gap can be a result of negotiation in choosing the next speaker. In order
to support these claims, the environments in which pauses greater than 1.5 seconds
occurred were examined The environments for turn-intemal pauses are discussed first,

then those for between-turn pauses.

4.3.6.2. The environments in which pauses occur
Table 4.5 shows the environments for turn-internal pauses. The figure in the
parentheses is the percentage of the number of instances in each environment to the total

number of turn-internal pauses, that is 87.

Table 4.5 The environments for turn-internal pauses

 

 

 

I environments number of instances

fore making a point 34 (39%)
when recalling something 25 (29%)
when groping for words 17 (20%)
before a quote 6 (7%)

fore asking questions 3 (3%)
after laugh 2 (2%)
L 37 (100%)

 

 

Turn-intemal pauses occurred 34 times when the speaker tried to make a point.
In such cases, conjunctions such as dakara ‘so’, done ‘because’, demo, ‘but’ immediately

precede or follow the pause. See Example (2) below.

36

(2) Pair 3. E and F (male friends)
(E and F are talking about their plans for Chrismas break.)

E: Kono huyuyasumi doo sun no?
this winter break how do [P

F: Teyuuka, dakaraa (1.8) sono sanshuukan no aida dakara aru yoobi ni
well so the three weeks LK during so certain days on

kooyattete sukoshi dete kaunseringu shinakucha ikenai karnoshinnai.
like this little do counseling do must maybe

E: What’s your plan for this winter break?
F: Well, (1.8) the thing is that I might have to do counseling on certain days
during the three weeks.

The total number of 25 pauses occurred in the environment when the speaker tried to
recall something. Those pauses were often preceded by the phrase nan datta ke naa
‘What was it?’ See example (3).

(3) Pair 1, A and B (female friends)

(A has just told B that most stewardesses suffer from a kind of occupational
disease.)

B: Donna shokugyoobyoo desu ka?
what kind of occupational disease BE Q

A: Nanka ne, nantettakkena? Nanka oshikko ga denakunatchau.
like IP what is it called like urine S come out NEG

B: Eee, kawaisoo.
ID what a pity

A: Nantendakke? (1.7) A, demo wasurechatta.
what is it called ID but forgot

B: What kind of occupational disease is it?

A: Like, what is it called? Like, you can’t urinate, the disease that causes urine
not to come out.

B: Oh, no. What a pity.

A: What is it called? (1.7) Oh, well, I forgot.

37

A 17 second pause occurred when the speaker was groping for words, in this case, the

phrase nan re yuu Ira ‘How can I say?’ appeared often. See example (4).
(4) Pair 1, A and B (female friends)

(B is telling A how her mother dealt with her childhood dreams.)

B: Yappari ne, nan te yuu ka, (2.8) 500 na n desu yone.
indeed IP how QT say so BE NOM BE IP

J iko jitsugen to yuu ka. J ibun o osaeta bubun ga aru n desu tte.
self achievement QT say SelfO control part S have NOM BE QT

B: Indeed, what can I say? (2.8) That’s right. It can be a kind of self

achievement. (According to her,) she controlled herself.

88% of all turn-internal pauses appeared in one of the three environments:
before making a point, when recalling something, and when groping for words. Other
environments in which tum-intemal pauses occurred were: before quoting someone (See
example 5), asking a question (See example 6) and after a laugh (See example 7).

(5) Pair 1, A and B (female friends)

(B has been telling A that Y, an actress, loves tenpura ‘deep fried vegetables’.)

B: Nanka, (1.6) ironna shinseki no hito ga atsumaru toki
like many relatives LK peple S get together time

yamamori ni shite taberu n desu tte.
bowlful make eat NOM BE QT

B: Like, (1.6) (Actress Y said that), plenty of tenpura was served when there
was a family reunion.
(6) Pair 1, A and B (female friends)

(A and B are talking that B’s mother has been supportive about what her
daughter is doing.)

38

A: Okaasan wa ooenshite kuretemasu ne, soo yuu imi dewa.
mother T support give IP so say meaning in
B: 800 desu ne.
So BE IF

A: Uun, soo ka. (1.7) Otoosama wa doo desu ka?
Hm, so father T how BE Q

A: Your mother is supportive in that sense.

B: Right.

A: I see. (1.7) What about your father?

(7) Pair 5, I and J (mixed friends)

(I is talking about her friend who lives in Poland).

1: Moshi mada Poorando ni sundetara awashitageru.
If still Poland at live if let see

Hahaha.(3.0) Maemo konna hanashi shitetan.
laugh before like this talk did

I: If (she) still lives in Poland, I will introduce you to her.
Laugh. (3.0) I talked about this before, too.

In this section, the environments where turn-internal pauses occurred were
examined. It is evident that turn-internal pauses enable the speaker to continue to speak.
In other words, those turn-intemal pauses contributed to the speakers attempt to make an
appropriate verbalization regarless of whether or not the attempt was successful.

Table 4.6 shows the environments for between-tum pauses. The figure in the
parentheses is the percentage of the number of instances in each environment to the total

number of each type of silence: gap, lapese, and attributable silence.

39

Table 4.6 The environment in which between-turn pauses occurs

 

 

 

 

 

 

 

 

 

types of silence environments number of
instances
gap alter a backchannel 27 (31%)
after a verb endings with IP (e. g. ne, yo) 25 (29%)
after incomplete sentences 21 (24%)
after verb endings without 1P 7 (8%)
after postposed sentences 7 (8%)
87 (100%)
lapse alter questions 5 (71%)
after tag questions 2 (29%)
7 (100%)
attributable silence after questions 9 (82%)
after a tag question 1 (9%)
after a request 1 (9%)
I 11 (100%)

 

The results are as follows.

Gaps occurred 87 times, lapses 7 times, and

attributable silence 11 times. The environments for each of the types of silence in Table

4.6. are shown below with examples.

Environments for gaps:

31 % of all gaps occured alter a backchannel such as Uun or hee ‘Hummn’.

example 8 below.

(8) Pair 7, M and N (married couple)

(N and M are talking about the main current in psychology.)

N: Soo yuuto saa, sugu

so say

IP immediately evidence

gakusha wa. ‘Doo yatte jisshoosuru n da ne?’ toka
scholar T how do prove

M: Aa.
I see

N: Sugu

sooyuu koto yuu hito

‘jisshoosee ni kaku’ toka yuu n da yo
lack such as say NOM BE [P
ne.

NOM BE [P such as IP

gairukara (3.1)

immediately such thing say people S exist

See

 

40

kanari muzukashiku naru yone.
quite diflicult become IP

 

M: Huun.
hmmn
(1 1.2)
M: Ato wa desu ne, watakushi no, ma, kodomo no toki no
other thing T BE IP I LK well child LK time LK
yurne toka ne.
dream such as [P
N: If I say such a thing, scholars would say like ‘(lt) lacks evidence’, or ‘How do -
you prove it.’ V!
M: I see.
N: There are such kinds of people, (3.1) so things are getting complicated
M: Hmmn.
(11.2)

M: Let me see. Well, (talk about) my childhood dreams.

After M gave a backchannel to N, an 11 second gap occurred. M’s signal was not clear
enough to indicate N as the next speaker, and after the pause M took a turn again with the
topic of childhood dreams.
29% of all gaps occured after a verb ending with IP (interactional particle). See
example 9 below.
(9) Pair 4, G and H (male friends)
(G is talking about his job searching experience.)

G: Un, de, onnanoko ga ryuugakusuru toka daigakuin iku toka itte.
right then girl S study abroad or graduate school go or said

H: Aa, tashikani sore wa ano jiki deshita ne.
yes indeed that T that time BE 1P

(1.8)

41

G: Sookoo shiteiru uchini desu ne, ano, sotsugyooshite,
this and tha do while BE 1P well graduate

ikko kimatteta n desu yo, oroshino tokoro ga.

one got job NOM BE IP multiple purpose dealer place S

G: Right, then, female students tended to go to study abroad or to graduate school.
H: Right, indeed, it happened at that time.
((1:81: the meantime, I graduated from college, and got a job at a multiple purpose
dealer.
IPs helped show the interactants’ involvement, in this case, ‘agreement,’ but they did not
necessarily mark the end of a turn.
21% of all gaps occurred after an incomplete sentence. See example 10 below.
(10) Pair 3, E and F (male friends)
(E and F are talking about their mutual friend’s car.)

F: Sooda yonaa, maa toorasu de.
that’ sright IP well Taurus with

(1.2)
E:Aredattara,maanandemo ii n dakara
that BE if well anything good NOM because

F: That’s right. (He can go anywhere) with his Taurus.

g:.'zl')hat car is unstoppable.
F ended his turn with the postpositional phrase, providing no predicates. A 1.2 second
pause, which occurred after F’s turn as a result of the ambiguous cue did not clearly mark

the end of his turn.

follows.

42

7% of all gaps occurred after a verb ending without IP. An example is as

(11) Pair 5, I and J (mixed friends)
(I is talking about the conflict between I and her friend.)

I: Atashi betsuni nanimo kanojo no sei ni suru ki nai.
I specially nothing her LK fault BE make intention NEG

(2.0)

J: Hum.
hmmn

I: I’m not saying that (that) was her fault.
(2.0)
J: Hmmn.

I ended her turn with the verb nai ‘do not have’ without an IP. This, again did not

provide any cue to J, and a 2.0 second pause occurred.

In addition to the case above, 7% of all gaps occurred after a postposed sentence.

See example 12 below.

(12) Pair 6, K and L (mixed friends)
(K is telling L that she finds that female six graders have big feet.)

K: Onnanoko saikin ookii yone, ashi toka.
Girls these days big IP foot such.

(1.0)

L: Hahaha. Miteru ne.
Laugh. See IP.

K: Girls lately have big feet, don’t they?
(1.6)
L: Laugh. You are paying attention to it, aren’t you?

43

K’s turn was a postposed sentence. Again, this did not clearly mark the end of her turn
and a 1.6 second pause occurred before L took the next turn.

These linguistic cues in the five environments for gaps are ambiguous in that
they do not clearly mark the end of the current speaker’s turn, and therefore I speculate

that a gap is a result of negotiating the next turn.

Environments for lapse:
The second type of a between-tum pause, a lapse occurs when the current speaker
chooses the next speaker, providing linguistic cues such as questions or tag questions, but
the next speaker does not take the turn There were two environments in which lapses
occurred. One was after a question (See example 13), and the other was after a tag
question (See example 14).

(13) Pair 1, A and B (female friends)

(A is asking B about Japanese high school students.)

A: Nihon no kookoosee tte doo desu ka?
Japan LK high school student T how BE Q

(2.2)

A: Kotchi no kookoosee toka to.
here LK high school students such and

B: A, zenzen chigaimasu.
ID completely different

A: How are the Japanese high school students?

(2.2)

A: Compared to the American high school students?
B: Oh, they are completely different.

44

After A asked B how the Japanese high school students were different from the American
high school students, a 2.2 second lapse occurred. Despite the fact that A chose B as

the next speaker, B did not take the turn.

(14) Pair 8, M and N (married couple)

N: Betsuni yume nante, yaritai koto datte iishi,
well dream such thing to want things BE okay

zenzen chigattemo ii it ja nai kane?
completely different good NOM TAG IP

(23)

N: Tatoeba yume no naka de miru yume ne.
for example dream LK in at see dream IP

N: Huun.
hmmn

N: Well, such a thing as a dream can be something you want to do, or something
totally different, can’t it?

(2-3)

N: For example, the dream which you see while sleeping.

M: Hmmn.
A 2.3 second lapse occurred after N asked M a tag question since M did not take the next
turn. After the lapse, N took the turn again and provided an example of dreams with N.

In my data lapses were not as frequently used as gaps were. Gaps occurred 87

times, while lapses occurred 7 times. This may be due to the fact that a lapse is often

times avoided so as not to embarrass the current speaker who expects the next speaker to

take a turn.

45

Environments for attributable silence:

Attributable silence occurs when the current speaker chooses the next speaker
providing linguistic cues such as questions or tag questions and the next speaker does
take a turn, but after a pause. In my data, attributable silence occurred in the following
environments: after a question (82%) , after a tag question(9%) and after a request (9%).
In what follows, examples for each environments are given.

(15) Pair 3, E and F (male friends)

(E is explaining what his comprehensive exam is like.)

E: Marutipuru choisu dattara are janai no?
Mulitiplechoice BE if that TAG 1P

Sonnani nagai no, jikan toka?
that long IP time for example

(3.5)

F: U::n, doregurai daro? Unto ne, ichioo sanjikan waku wa tottearu.
well how long wonder let’s see IP generally three hour frame T reserved

E: Isn’t it multiple choice, though? Time wise, is it that long?

Iglswonder how long it is. Let’s see, generally three hours are reserved for that
A 3.5 second attrlbutable silence occurred after B asked F a question. After the pause, F
gave B an answer for the question. This might suggest that the pause was needed for F

to come up with an appropriate answer.

(16) Pair 7, 0 and P (married couple)
(P has just talked to 0 about how he raced with someone on the highway.)

0: Yokatta, notte nakute. Datte inochi nakatta n janai no?
good ride NEG Because life NEG NOM TAG IP

46

$00 ja nai?
So TAG

(3.1)

P: Okkanaimon. Aa naru to omowanakatta ne.Yappa kodomo datta n da ne.
scary that become Q think NEG IP Indeed child BE NOM BE [P

O: Glad I did not ride with you. You could have been killed. Isn’t that right?
(PiSLary. 1 did not expect that would happen. After all, I was just a kid
After P asked 0 a tag question, a 3.1 second pause occurred. After the pause, P took the
next turn and told P his impression at that time. The pause seemed to provide P time to

recall how he felt when the incident happened.

(17) Pair 8, O and P (married couple)
(0 is trying to let P guess O’s childhood dream.)

0: Jaa, atashi wa nani ni naritakatta ka? Ichioo atta wa,
then I T what to become Q actuallly had IP

ima kangaete miru to. Jaa, atete.
now think try then guess

(5.6)

P: Pianisuto?
pianist
0: Then, what I wanted to be? In fact I had one. Guess?
(5.6)
P: A pianist?
The example above was the only case in which a between-pause appeared alter a
linguistic cue, which was a request. After 0 requested P to guess what O’s childhood

dream was, a 5.6 second pause occurred and then P guessed what it was. The pause

47

seemed to be used for the time to come up with an answer.

The examples above illustrated the environments in which attributable silences
occurred. Each case showed that an attributable silence enabled the next speaker to
come up with the next utterance that would be spoken. Therefore, attibutable silences

occurred less than gaps.

4.3.6.3. Functions of silence in Japanese private discourse

Based on the results given above, I propose that silence in Japanese private
discourse have the following two major functions.

1) to make an appropriate verbalization

2) to avoid imposing on the other speaker’s turn

The first type of function is frequently observed in the turn-internal pauses (See
example 2 to 5 in section 4.3.6.1) and attributable silences (See example 16 and 17 in
section 4.3.6.2). Regardless of the length of the pause, and / or, whether or not the
current speaker’s attempt to make an appropriate verbalization was sucessful, it is
plausible that the pauses that occurred in the examples above could provide the time to
come up with the next utterance that would be spoken. In this sense, the use of silence
follows the co-operative principle proposed by Grice (1975): Make your contribution
such as is required, at the stage at which it occurs, by the accepted purpose of direction of
the talk exchange in which you are engaged. This suggests the area of future study
could be extended to the application of Grice’s co-operative principle and the four

Maxims to the use of silence.

48

The second type of the function was found in gaps (See example 6 to 12 in
section 4.3.6.2.) and in lapses (See example 13 and 14 in section 4.3.6.2.). In section
4.3.6.2., I argued that the linguistic cues in the environments (e. g. after a backchannel)
for gaps were ambiguous in that they did not clearly mark the end of the current
speaker’s turn, and therefore I speculated that a gap was a result of negotiating the next
turn. In my data lapses were not as frequently used as gaps were. Gaps occurred 87
times, while lapses occurred 7 times. This may be due to the fact that a lapse is often
times avoided so as not to embarrass the current speaker who expects the next speaker to
take a turn. The function enables the interactants to be polite by avoiding to impose on

the other’s turn during the conversation in which they are engaged.

4.4. Summary

In this chapter, the use of silence in Japanese conversation was examined. The
findings are as follows. First, the degree of intimacy correlates with the frequency of
the use of silence. In comparison between pairs of the opposite sex, married couples
used more silence both in length and frequency (See p. 28 and p. 29). Secondly, gender
difference correlates with the frequency of the use of silence. Comparing between pairs
of the same sex showed that female pairs used more silence than male pairs both in
length and frequency (See p.27 and p.29). This offered two interpretations: One is that
female pairs are more intimate and thus are able to tolerate more silence during their
conversations. The other is that female pairs use silence to be polite as a way to avoid

imposing on each other’s turn. Third, the number of pauses which occurred in

49

conversation between mixed pairs who are friends is greater than that of male pairs, and
smaller than that of female pairs (p.27 and p.28). This indicates that a female uses
silence less when talking to a male, while a male uses silence more when conversing with
a female. This is considered to be a result because a female does not tolerate silence
when the interlocutor is male, and a male uses silence to avoid imposing upon a female
interlocutor.

Based on the results, two major functions of silence in Japanese private discourse
are proposed; One is to make an appropriate verbalization, and the other is to avoid
imposing on the other speaker’s turn. It is also proposed that the use of silence in
Japanese private discourse follows the co-operative principle in that the use of silence

contributes to the talk exchange in which people are engaged

Chapter 5

Conclusion

The present study investigates the use of silence in Japanese private discourse.
Although silence is phonetically null, it is not an empty or meaningless linguistic
phenomenon. In the present study, silence is treated as a component which structures
discourse in much of the same way as speech does. Silence is prevalent in Japanese
conversation, and is considered to be one of the common communicative strategies in
Japanese language. I have provided the functional explanations of silence in private
discourse in Japanese, considering variables such as gender difference and the degree of
intimacy: (1) To make an appropriate verbalization, (2) To avoid imposing on the other
speaker’s turn

The findings of this study are summarized as follows. The more intimate the
relation is between the speakers, the more silence is used both in length and in frequency
in Japanese private discourse. In my data, married couples used a greater number of
pauses in their conversations and longer pauses. Secondly, female pairs use long pauses
than male pairs. It is evident that female pairs tolerate pauses compared to male pairs.

In addition, the number of pauses in the conversations of mixed pairs is greater than

50

51

those in male pairs, and smaller than those in female pairs. Thus, a female uses silence
less when talking to a male, while a male uses silence more when conversing with a
female. This is considered to be a result because a female does not tolerate silence
when the interlocutor is male, and a male uses silence to avoid imposing upon a female
interlocutor.

Finally, it is proposed that the use of silence in Japanese private discourse
follows the co—operative principle proposed by Grice (1975) in that silence contributes to
the talk exchange in which speakers are engaged. The area of future study could be
extended to the application of Grice’s co-operative principle and the four Maxims to the
use of silence. Admittedly, samples are very small, but I hope this study will be the first

step in uncovering the use of silence in Japanese private discourse.

APPENDICES

APPENDIX A

Appendix A

List of Japanese subjects

 

 

 

 

 

 

 

 

 

pair subject sex gage occupation martial status
1 A F 30 graduate student single
B F 30 graduate student single
2 C F 26 graduate student single
D F 28 graduate student single
3 E M 27 graduate student single
F M 23 graduate student siggle
4 G M 26 graduate student single
H M 30 JEduate student single
5 I F 23 graduate student single
J M 30 gLaduate student single
6 K F 30 graduate student single
L M 27 graduate student single
7 M F 35 housewife married
N M 36 graduate student married
8 O F 27 housewife married
P M 28 graduate student married

 

52

APPENDIX B

Appendix B

CONSENT FORM

I agree to let Natsuko Orai, for the purpose of linguistic analysis, record converation of
me, including casual conversations on a ntunber of topics of social interest. I
understand that any publication or presentation she might make as a result of these
recordings would not personally identify me. I also understand that my participation is
purely voluntary and that I may withdraw from the entire program of research at any time
or at any time ask to have the tape recorder turned off with no repercussions whatsoever.

Signed

Date

53

APPENDIX C

Appendix C

RESPONDENT IDENTIFICATION

RESPONDENT INFORMATION

Name

Address

Phone (or other contact means)

Age Sex

Profession

Education

54

 

APPENDIX D

1)

2)

3)

4)

5)

6)

7)

Appendix D

The key transcription methods and conventions

Japanese transliteration is given in phonetic orthography referred to as the
Hepburn style, except that fu is spelled hu. In presenting double consonants,
before cha, chi, cho, and chu, instead of adding an extra c, r is added, thus
instead of icchi ‘agreement’, itchr‘ is used Syllabic n is written n unless it
immediately precedes a vowel, in which care it is written it '.

Speakers in each pair are identified by capitalized letters.

Examples taken from the data sets are presented with assigned numbers given in
parentheses.

When proper nouns appeared in the data, either random initials are used.

Paralinguistic and other interactional symbols used are:
( ) pause, actually indicated with figure

. utterance-final contour

? rising interrogative contour

LAUGH various types of laughs and laugh-like utterables
bold point of analysis

The Following abbreviations are used:
BE copulative verb, be

IP interactional particle
ID ideophones (including onomatopoetic [giseigo] and mimetic [gitaigo]
words)

LK linker (linking nominals and nominal adjectives)

NEG negative

NOM nominalizer

0 direct object marker

Q question marker (assigned to final particle ka only)

QT quotative marker

S subject marker

T topic marker

TAG tag-question-like auxiliary verb forms (including jan, ja-nai, desho, etc.)

A word-by-word gloss is provided directly below the transcribed Japanese.
Then, the translation of the whole excerpt is given. For the most, the
translations are idiomatic.

55

BIBLIOGRAPHY

Bibliography
Chafe, Wallace. (1983). Some reasons for hesitating. In Tannen, D., & Saville-Troike, M.

(Eds). 77-89.

Crown.C.L., & Feldestein.S. (1985) Psychological correlates of silence and sound in
conversational interaction. In Tannen, D., & Saville-Troike, M. (Eds). 31-54.

Grice, HP. (1975) Logic and conversation. In P. Cole and J .Mogan (Eds), Speech Acts
(Syntax and Semantics, Volume 9). New York: Academic Press, 41-58.

Hall, K., & Bucholts, M. (1995) Gender articulated: Language and the socially
constructed self New York and London: Rontladge.

J anicki, K. (1990) Toward non-essentialist sociolinguistics. Berlin: Mouton de Gruyter.

J aworski, A. (1993) The power of silence: Social and pragmatic perspectives. Newbury
Park, London, New Delhi: Sage publications.

Kurzon, D. (1997) Discourse of silence. Philadelphia: John Benjamin’s North America.

Lakoff, R (1995) Cries and whispers: The shattering of the silence. In Hall, K, &
Bucholts.(Eds.). 25-50.

Lebra, T. S. (1987) The cultural significance of silence in Japanese communication.
Multilingua, (6-4), 343-357.

Leech, ON. (1983) Principle of pragmatics. London and New York: Longrnan.
Levinson, S. C. (1983) Pragmatics. Cambridge, UK: Cambridge University Press.

Maltz, D (1985) Joyful noise and reverent silence: the significance of noise in Pentecosta
worship. In Tannen, D., & Saville-Troike, M. (Eds). 113-137.

Maynard, S. (1989) Japanese conversation: Self-contextualization through structure and
interaction management. Norwood, NJ: Ablex.

Maynard, S. (1997) Japanese communication: Language and thought in context.
Honolulu: University of Hawaii Press.

McGuire, PC. (1985) Speechless dialect: Shakespeare 19 open silences. Berkeley:
University of California Press.

56

57

Mendoza-Demon, N.(1995) Silence and Authority in the Anita Hill-Clarence Thomas
hearings. In Hall, K, & Bucholts, M. (Eds). 51-66.

Miyoshi, M. (1974) A ccomplices of silence: The modern Japanese novel. Berkeley:
University of California Press.

Nakamura, H. (l995)1wanaideoku: Nihongo o migaku. Tokyo: Kindaibungeisha.

Ono, T.& Yoshida, E. (1996) A study of co-construction in Japanese: We don’t finish
each their sentences. In Akatsuka, N., Iwasaki, S.,& Strauss, S. (Eds), Japanese
Korean Linguistics, 5, 115-127. California: CSLI Publications.

Sacks, H., Schegloff, E., & Jefferson, G. (1974) A simplest systematics for organization
of turn- taking in conversation. Language, 50, 696-735.

Saville-Troike, M. (1985) The place of silence in an integrated theory of communication.
In Tannen, D., & Saville-Troike, M (Eds). 3-18.

Schiffrin, D. (1987) Discourse markers: Cambridge: Cambridge University Press.
Schit’frin, D. (1994) Approaches to discourse. Oxford: Basil Blackwell.

Sperber, D., & Wilson, D. (1986) Relevance: Communication and cognition. Oxford:
Basil Blackwell.

Stubbs, M. (1983) Discourse analysis. Chicago: University of Chicago Press.

Tannen, D. (1985) Silence: Anything but. In Tannen, D., & Saville—Troike, M (Eds). 93-
111.

Tannen, D. (1994) Gender & Discourse. Oxford: Oxford University Press.
Wardhaugh, R. (1986) An introduction to sociolinguistics. Oxford: Basil Blackwell.

Yamada, H. (1992) American and Japanese business discourse: a comparison of
interactional styles. Norwood, NJ: Ablex.

nrcutcaN sran UNIV. LIBRARIES
I“WINill"Willi”W“HINIIIHIWWHI
31293017120134