Being affected: the meanings and functions of Japanese passive constructions

Citation preview

Published by LINCOM GmbH 2007.

LINCOM GmbH Gmunderstr. 35 D-81379 Muenchen

[email protected] http://home.t-online.de/home/LINCOM.EUROPA www.lincom-europa.com webshop: lincom.at All rights reserved, including the rights of translation into any foreign language. No part of this book may be reproduced in any way without the permission of the publisher.

Printed in E.C. Printed on chlorine-free paper

Die Deutsche Bibliothek - CIP Cataloguing-in-Publication-Data

A catalogue record for this publication is available from Die Deutsche Bibliothek (http://www.ddb.de)

full text research abstracts of all titles monthly updates

LINCOM webshop www.lmcom-europa.com'

Being Affected: The meanings and functions of Japanese passive constructions Mami Iwashita

2007 LINCOM EUROPA

11

CONTENTS Table o f Contents

ii

Abstract

vi

Acknowledgements

vii

Abbreviations

ix

List o f Figures and Tables

x

C hapter t

1

Introduction

1.1 Overview 1.2

About the data used in this study

1 6

1.2.1 Written data

6

1.2.2 Spoken data

8

1.2.3 Examples in the text of this study 1.3 Motivation for the use of authentic data

11 13

1.4

Definition of passive

14

1.5

~(r)are constructions in Japanese

22

1.6

Basic terms and concepts

28

1.6.1 Verb types in Japanese

28

1.6.1.1 Transitive and intransitive verbs

29

1.6.1.2 Unergative and unaccusative verbs

30

1.6.2 Actor and undergoer 1.7

Sentience and passive constructionsin Japanese 1.7.1 Sentience of the participants in intransitive-based passives 1.7.2 Sentience o f the participants in direct passives

33 37 39 39

1.7.2.1 Sentience o f the ‘actor’ in the direct passive

40

1.7.2.2 Sentience of the subject in the direct passive

41

1.7.2.3 Direct passive with a non-sentient subject

42

iii

Chapter 2

2.1 2.2

2.3

Classification of Japanese passives - Previous analyses

Introduction

43

Earlier approaches

44

2.2.1

Yamada’s approach

44

2.2.2

Matsushita’s analysis

46

Mainstream approach - direct passive and indirect passive 2.3.1

51

Generative transformational grammarian’s approach to the dichotomy of direct and indirect passives

2.4

43

More recent approaches

56 61

2.4.1

Masuoka's approach

61

2.4.2

Nitta’s approach

74

Chapter 3

Classification used in this study - Syntactic distinction and semantic

distinction

78

3.1

80

3.2

Syntactic distinction 3.1.1 Accessibility Hierarchy

80

3.1.2 Direct passive

81

3.1.3 Semi-direct passive

83

3.1.4 Indirect passive

86

Semantic distinction

88

3.2.1 Plain passive 3.2.1.1 Demotional passive

88 89

3.2.1.1.11s the demotional passiveinherent to Japanese?

92

3.2.1.1.2Verb types and the demotional passive

96

3.2.1.2 Attributive passive 3.2.2 Passive of interest

102 106

3.2.2.1 Passive with latent affectee

107

3.2.2.2 Sentient passive

115

IV

C hapter 4

The lapanese passiv ex w ith special emotive affectedness

4.1 The emotive undertone of indirect passives 4.1.1 The indirect passive of an intransitive verb 4.1.2 The indirect passive of a transitive verb 4.2

119

120 121 135

Semi-direct passives with special emotive nuance

140

4.3

Direct passives with emotive undertone

144

4.4

The emotive nuance

146

4.4.1 Involvement

148

4.4.2 Semantic integration

150

4.4.3 Relevance

153

4.4.4 Why adversative?

162

Chapter 5

5.1 5.2

Japanese passive in authentic texts

Data and methodology Basic findings 5.2.1 Proportions of each category in the data

167

167 168 168

5.2.1.1 Syntactic categories

168

5.2.1.2 Semantic categories

170

5.2.2 Actor

174

5.2.3 Subject

178

5.2.3.1 Subject types

178

5.2.3.2 Degree of topicality of the subject

183

5.2.4 Noun-modifying clauses 5.2.5 Propositional meaning

186 190

5.2.5.1 Propositional meaning in each syntactic group

195

5.2.5.2 Propositional meaning in each semantic group

200

5.3 Detailed discussion on the findings 5.3.1 Degrees of centrality of the passive subject to the event 5.3.1.1

206 206

Subject’s degree of centrality to the event in the syntactic classification 206

5.3.2 Objective affectedness vs direct affectedness in the semantic category 5.3.3 Three types of affectedness in Japanese passive constructions

( ’haptcr 6

Conclusion

210 220

226

6.1 Aims and findings of this study

226

6.2

234

Limitations o f this study and recommendations for further research

References

236

Sources o f examples

240

VI

ABSTRACT

Amongst the multiple and diverse meanings and functions passive constructions hold, this study considers that the primary function of passives in Japanese is to portray an event from the point of view o f an affected entity. The study identifies three types of affectedness in Japanese passive constructions: emotive affectedness, direct / physical affectedness, and objective affectedness. Emotive affectedness, often referred to as ‘adversative’ meaning, has drawn attention from many researchers. It has been strongly associated in the past with the syntactic category called the ‘indirect passive’, but is actually also observed in many instances of the ‘direct passive’. Direct / physical affectedness is detected mainly in the construction here referred to as the ‘direct sentient passive’. This meaning is common in passives in many other languages, including English. The last type - objective affectedness - is primarily associated with ‘non-sentient passives’, more specifically with what is here called the ‘plain passive’. Many previous researchers have claimed a complete and apparently transparent correlation between syntactic and semantic distinctions of the Japanese passive. The present study rejects these direct correlations. In analysing authentic data, it becomes evident that the correlation is much more subtle than has generally been recognised, and that is a matter of degree or continuum, rather than a discrete, black and white issue. To reflect this view, this study proposes separate sets of categories for syntactic and semantic distinctions. The ultimate aim o f this study is to reveal how Japanese passives are actually used in real contexts. In order to achieve this aim, detailed examination o f authentic written and spoken data is conducted. Some findings of the data analysis in the present study contradict previous claims, such as the finding of a large proportion of passives with a non-sentient subject and very low frequency of occurrence of indirect passives. This research also finds that, although more than half o f the propositional meanings in the passive data examined are ‘negative’, a considerable number of passives still appear in a proposition with a neutral or positive meaning. Another prominent finding regarding propositional meaning is that it seems to be related to the degree of centrality of the passive subject to the event. With regard to the syntactic classification of passive, in particular, it is observed that the lower the degree of the centrality of the subject of the passive to the event, the greater the likelihood that the passive clause involves a negative proposition.

vii

ACKNOWLEDGEMENTS

Without the guidance and support of numerous people, this book could not have been completed. 1 would like to express my sincere gratitude to all these people.

Firstly, I would like to thank my principal supervisor, Professor Hugh Clarke for his wise guidance and support throughout the time of my candidature. I greatly appreciate his broad and scholarly input to my work, and value highly his comments and contributions, stemming from his vast knowledge in various fields.

My heartfelt gratitude goes to Dr Nerida Jarkey, my associate supervisor, who has also been my acting principal supervisor for six months in 2002 and for the last three months of my candidature. I could not have come this far without her unfailing support, constant warm encouragement, and friendship. I am deeply indebted to Nerida for the great deal of time and effort she has generously put into our frequent and stimulating discussions, and for her detailed and abundant comments on every aspect of my book.

I am also most grateful to Ms Kazumi Ishii, my mentor and friend, for her consistent support, invaluable advice and warm, hearty encouragement. Sincere thanks also to Ms XiangDong Liu, my best friend and comrade, for her kind caring and sharing the hardships and joys, and Ms Yukako Kamashita for her unwavering friendship and encouragement.

I was supported throughout my candidature by my colleagues at the University of Sydney: Professor Tony Stephens, Head of the School of Languages and Cultures, Associate Professor Elise Tipton, Chair of the Department of Japanese and Korean Studies, Dr Yasuko Claremont, Dr Hiroko Kobayashi, Associate Professor Sakuko Matsui, Dr Keizo Nanri, Mr Colin Noble, Dr Chun Fen Shao and Ms Seiko Yasumoto.

viii

My deep gratitude also extends to my former colleagues at the University of Tasmania: Dr Maria Flutsch, Mrs Takame Ueki-Sabine and Ms Hisako Umeoka for their warm support and encouragement. Also to my former teachers at University of Tsukuba Associate Professor Sumiko Horiguchi, Professor Toshiko Ishida, Professor Yutaka Kusanagi, Associate Professor Yuriko Sunakawa, Professor Makoto Takada, Professor Tadayuki Yuzawa, and the late Professor Hideo Teramura - I would like to convey my sincere thanks. I have also greatly appreciated valuable comments and advice from Professor Takashi Masuoka, Professor Hisashi Noda, and Professor Masayoshi Shibatani.

To all my friends I cannot begin to express my gratitude: Hiromi Ayano, Chikako Shigemori-Bucar, Mary Cemy, Rachel Clarke, Liz Cruse, Mike Dowman, Ralph Forehead, Anki Gustaffson, Flitomi Hatano, Shinji Ido, Kyoko Kanda, Yuko Katabami, Hyunhee Kim, Noriko Kitamura, Ayumi Miyamoto, Shoko Ono, Kimie Takahashi, Yuri Takahashi, Yoshiko Wada and June Williams!

Immeasurable thanks are also due to Michael Kirk for his caring and support, especially over the last six months of my candidature; he helped me get through the difficult times, and gave timely help in proof reading the final draft.

Last, but by no means least, I would like to express my sincere gratitude to my parents, Toshio and Michiko Iwashita for their everlasting, generous support, warm-hearted encouragement and patient understanding. I cannot even begin to list the numerous ways in which they have helped me.

ABBREVIATIONS

ABL

ablative

NEG

negative

ACC

accusative

NML

nominalizer

AGT

agent

NOM

nominative

CAUS

causative

NONPAST

non-past

CLF

classifier

OBL

oblique

CLTZ

collectivizer

OCOMP

object of comparison

CMPL

complementizer

PART

particle

COMP

comparative

PASS

passive

CONJ

conjunctive

PAST

past

COP

copula

PERF

perfective

DAT

dative

PLR

plural

DESID

desiderative

POL

polite

DO

direct object

POTEN

potential

EMPH

emphatic

PRES

present

EXHT

exhortative

Q

question

GEN

genitive

QUOT

quotative

HON

honorific

RPTV

repetitive

IMP

imperative

SPON

spontaneous

INSTR

instrumental

SFP

sentence final particle

LOC

locative

TOP

topic

X

List of Figures and Tables

Figures: 1. Syntactic categories 2. Semantic categories 3. ‘actor’ 4. The nature of an overt‘actor’ 5. Subject types 6. Subject types 2 7. Subject’s degree of topicality 1: in each syntactic group 8. Subject’s degree of topicality 2: in each semantic group 9. Noun-modifying clauses 10. Noun-modifying clauses in written data 11. Propositional meaning 1: in each syntactic group 12. Propositional meaning 2: in each semantic group

page 169 172 175 177 180 182 184 185 187 188 195 201

Tables: 1. Matsushita’s classification 2. Relation between Matsushita’s and Teramura’s classifications 3. Relation between Matsushita’s, Masuoka’s and Teramura’s classifications 4. Classification used in this study compared with previous analyses 5. Combination of ni, ni-yotte, and kara NPs with verbs in passivesentences 6. Classification used in this study 7. Syntactic classification used in this study 8. Semantic and syntactic classifications used in this study 9. Emotive affectedness 10. Passives with direct affectedness 11. Objective affectedness

46 52 63 79 97 1] 7 207 211 221 221 224

1

Chapter 1

1.1

Introduction

Overview

Passive expressions occurring in a wide range o f languages have multiple diverse meanings and functions: personal or impersonal passive, transitive-based or intransitive-based passive, passives with the meaning of adversity or without any such meaning, etc. In previous research, such as Mikami (1953/1972: 98-112), Kuno (1973: 24), Shibatani (1978: 133-142) and Teramura (1982: 214-217), the assumption has been that the Japanese passive has dual semantic functions: adversative and neutral. Objecting to this view, Klaiman (1987: 429) maintains that all passive -rare expressions in Japanese convey a nuance of ‘affect’, and basically fall into a single semantic type. However, the present research takes a stance that differs from both o f the views above. It considers the Japanese passive neither as having the semantic dichotomy o f adversative and neutral nor as carrying a single semantic role, but, as with passives in many other languages, to be semantically multi-functioned.

Amongst the diverse meanings and functions passive constructions hold, such as defocusing the ‘actor’, describing the subject’s attribute, and so on, the primary one is related to the perspective from which the event is described. When you try to describe a scene, you have several options depending on the perspective you would take. This study considers that passives in Japanese portray an event or situation from the point of view of

2

an affected entity.

Previous research, such as Kuno (1973: 24) and Teramura (1982: 214-217), has claimed a simplistic and apparently transparent correlation between syntactic and semantic distinctions of the Japanese passive. The present study rejects these direct correlations, but nevertheless maintains that a correlation between syntactic features and semantic types does exist. In fact, this study demonstrates complex and sophisticated correlations between syntax and semantics in the case of Japanese passive constructions. In examining authentic data, it becomes evident that the correlation is much more subtle than has generally been recognised, and that is a matter of degree or continuum, rather than a discrete, black and white issue.

From the standpoint declared above, the current study recognises three types of affectedness in Japanese passive constructions: emotive affectedness, direct / physical affectedness, and objective affectedness. The special meaning of emotive affectedness associated with only some Japanese passives, often referred to as ‘adversative’ meaning, and has drawn attention from many researchers. This meaning is primarily syntactically motivated; however, several parameters of semantic transitivity also play an important role. Direct / physical affectedness is detected mainly in the construction here referred to as the direct sentient passive. This meaning is quite common in passives in many other languages, including English. The last type - objective affectedness - is primarily associated with ‘non-sentient passives’, more specifically with what is here called the plain passive. In this

3

type a non-sentient entity appears in the subject position. This meaning is also related to Aktionsart1. Boundaries between these three types of affectedness are not discrete: instead we can see these notions of affectedness as a matter of degree.

The present study has three main aims. The first aim is to clarify and classify all the functions of passives in Japanese. The second aim is to tackle and solve the well-known issue of the ‘adversative meaning’, referred to in this work as the ‘special meaning of emotive affectedness’, which is associated with only some types of passive in Japanese. The final aim is the ultimate goal o f this study. That is to reveal how the multi-functioned Japanese passives actually appear in real contexts.

This research uses authentic written and spoken data for analysis, in order to portray how Japanese passives are actually used. Only a limited number o f examples have been generated on the evidence of native-speaker intuition, for the sake of simplifying the explanation. Ways of simplifying include omission of adverbial clauses, noun-modifying clauses and sentence final particles that do not affect the inteipretation of passive. Previous research on these constructions has had tendency to rely heavily on inauthentic data, made up by the researchers themselves. Once again, it is only by looking at authentic examples, in the context in which they originally occurred, that we can fully understand the real meanings and functions of these constructions.

1 The term Aktionsart is used here, following Klaiman (1987: 401 & 432), to refer to ‘the inherent aspectual character of the verb’, or ‘the range of temporal characters which are ascribable to some verbally denoted action purely in virtue of the verb’s lexical sense’.

4

For the written data for this study, we used the CD-Rorn collection of Japanese novels, Shincho Bunko no Hyakusatsu (One hundred Shincho Paperbacks). For the spoken data, the collection of Josei no Kotoba - Shokubahen (Women’s Language - Workplace section) and Dansei no Kotoba - Shokubahen (Men’s Language - Workplace section), commercially available on CD-Rom, were used.

The book consists of six chapters. The current chapter gives an overview of the background to the study. It has three main parts. After outlining the research statement and taking a general view of the study in this section, Section 1.2 deals with the data used in the study in detail, and Section 1.3, motivation for the use of authentic data. The rest of Chapter 1, Sections 1 .4 - 1.7, is devoted to the theoretical background, including the notion of passive itself, and other basic and essential terms and concepts related to the study of Japanese passive constructions.

As mentioned above, the passive ~(r)are construction in Japanese has multiple semantic functions, from describing the subject’s attribute to depicting a situation in which the subject is somehow affected by an event. In order to clarify these functions, in Chapter 2, the controversial issue of the classification o f Japanese passives in previous research is reexamined. Chapter 3 then presents the new classification used in this study. The current research raises an objection to the previous claim o f a direct correlation between the syntactic and semantic distinctions - more specifically the correspondence between the indirect passive and the adversative passive on the one hand, and that between the direct

5

passive and the neutral passive on the other. Instead we propose separate sets of categories for each syntactic and semantic distinction. This study then demonstrates that correlation between syntax and semantics in Japanese passives is much more subtle and complex than has generally been recognised. In dealing with real life data, it becomes apparent that this issue of correlation between syntax and semantics can never be seen as a simple, black and white matter.

After presenting the classification of Japanese passives in Chapter 3, Chapter 4 investigates the type with the special emotive affectedness, widely known as the ‘adversative passive’ in previous research. The term ‘adversative passive’ has presumably been used because, in most cases, the emotive nuance is adversative. However, we note cases in which the emotive undertone cannot be considered ‘adversative’. It is therefore referred to in this study as having a ‘special meaning of emotive affectedness’. We will focus on examining under what circumstances the special nuance appears, and suggest why such a nuance occurs at all.

Finally, in Chapter 5, the findings o f the data analysis are discussed in detail. In Section 5.2, numerical findings on the syntactic and semantic categories of passive, the nature of the ‘actor’ and the subject, the occurrence in noun-modifying clauses, and the kind of propositional meanings that occur are fully discussed. Some findings confirm the claims made in previous research, such as the high frequency of passives without an overt ‘actor’, and of passives with a negative propositional meaning.

6

Other findings completely contradict previous claims, such as the large proportion of passives with a non-sentient subject. This research also finds that the frequency o f the occurrence of the indirect passive, the major focus of previous studies of Japanese passives, seems, in fact, to be very low.

The last section of Chapter 5 deals mainly with the issue o f how central the role played by the referent of the subject is to the event: how important this is and how it affects the propositional meaning of the passive sentence.

1.2

About the data used in this study

In this research we have primarily used authentic written and spoken data for analysis, in order to reveal how Japanese passives are actually used.

1.2.1 Written data

For the written data of this study, we collected 679 passive examples from ten different novels. Using the CD-Rom collection of Japanese novels, Shincho Bunko no Hyakusatsu (One hundred Shincho Paperbacks), we first selected the fifteen most recent novels in the collection. The author’s gender and age at the time s/he wrote the novel are then noted, and

7

ten were chosen from among the fifteen, aiming for the best balance possible with regard to these variables. Authors of the selected novels were all in their 20s to 50s at the time s/he wrote the novel. Although there are some other more recent novels written by male authors, Takano’s novel was chosen in order to keep a good balance in the numbers of male and female authors. Details of the novels selected are given below: Author’s Name Akagawa, Jiro

Year Age

Gender

1984

36

Male

Fujiwara, Masahiko

1978

35

Male

Miyamoto, Teru Murakami, Haruki

1982

35

Male

1985

36

Male

Sawaki, Kotaro Shiina, Makoto

1981

34

Male

1991

47

Male

Shiono, Nanami

1991

54

Female

Sono, Ayako Takano, Etsuko

1979

48

Female

1971

22

Female

Tsutsui, Yasutaka

1981

47

Male

Title

Passive Nos Narrative set in 31 1980s Tokyo Type of novel

Onna Shacho ni Kanpai! ‘Cheers to Madam President!’ First person Wakaki Sugakusha no novel set in America ‘A Young Mathematician in America’ 1970s US Narrative set in Kinshu ‘Gold Brocade’ 1970s Japan Narrative set in Sekai no Owari to Hadoboirudo Wandarando future Japan ‘The End o f the World and the Hard-boiled Wonderland’ Narrative set in Isshun no Natsu ‘Summer 1970s Tokyo in a Split Second’ First person Shinbashi novel set in Karasumoriguchi Seishun-hen ‘Shinbashi 1980s Tokyo Station, Karasumori Exit In My Young Days Historical Konsutantinopuru no novel set in Kanraku ‘Fall of 15th century Constantinople’ Constantinople Narrative set in Taro Monogatari ‘Taro’s 1970s Japan Story’ First person Hatachi no Genten ‘The Origin of Twenty Years of novel set in 1960s Japan Age’ Narrative set in Edipusu no Koibito future Japan ‘Oedipus’s Lover’

85

69 69

74 31

157

51 39

73

We searched the middle 100 pages of each novel. However, note that this CD-Rom version of the novels contains 320 characters per page, half that of a normal Japanese paperback. The written corpus was thus equivalent in length to approximately 500 pages o f a hard copy Japanese paperback.

1.2.2 Spoken data

For the spoken data, the collections of Josei no Kotoba - Shokubahen (Women’s Language - Workplace section) and Dansei no Kotoba - Shokubahen (Men’s Language - Workplace section), commercially available on CD-Rom2, were used. 169 passive examples out of a total of 16921 utterances were found in these spoken data collections.

The ‘Women’s Language’ corpus was originally compiled for the study of the actual conditions o f the spoken language of women by ‘Gendai Nihongo Kenkyukai’ (Society o f Contemporary Japanese Language Studies). Recordings were made during the period September 1993 to November 1993 in the Tokyo Metropolitan area. The subjects are 19 working women in their 20s to 50s. A breakdown of the subjects by age group and occupation is shown below:

2 When the data analysis for this research was conducted, CD-Rom for ‘Men’s Language’ corpus was not yet commercially available. I am indebted to one of the compilers, Dr Haruko Hayakawa, for the use of the pre-publication version of the CD-Rom.

9

Occupation

Subject’s code 01

Age group 20s

Company employee (office job)

02

20s

Company employee (office job)

03

30s

Company employee (editing)

04

50s

University Lecturer

05

40s

Company executive

06

40s

Company employee (editing)

07

40s

University assistant

08

50s

Primary school teacher

09

30s

Senior high school teacher

10

40s

Public servant (office job)

11

20s

Company employee (sales)

12

50s

Public servant (office job)

13

20s

Company employee (office job)

14

20s

Public servant (University office)

15

30s

Company employee (planning)

16

30s

Company employee (editing)

17

30s

Company employee (editing)

18

40s

Public servant (research assistant)

19

40s

Public servant (research assistant)

In this collection 11421 utterances were searched, and 76 passive examples were found.

The ‘Men’s Language’ corpus was recorded during the period October 1999 to December 2000 in the Tokyo Metropolitan area, also by ‘Gendai Nihongo Kenkyukai’

10

(Society of Contemporary Japanese Language Studies). The subjects are 21 working men in their 20s to 50s. Detailed information of the subjects’ age group and occupation is presented below:

Subject’s code 01

Age group

Occupation

40s

Pharmacy owner

02

50s

University employee

03

30s

Company employee (sales)

04

20s

Company employee (sales)

05

50s

Company employee (engineer)

06

40s

University Lecturer

07

30s

Company employee (sales)

08

40s

Company employee (office job)

09

40s

Car manufacturer/ mechanic

10

50s

Company employee (technical)

11

40s

Senior high school teacher

12

50s

Company owner

13

20s

Research institute employee - part-time

14

30s

Hairdresser

15

20s

Insurance company employee (sales)

16

40s

Insurance company employee (sales)

17

30s

Company employee (call centre)

18

20s

University employee (library)

19

20s

University employee (library)

20

50s

Senior high school teacher

21

30s

Musician

II

In this database, 5500 utterances were searched3, and 103 instances of the passive were found.

If written and spoken data are combined, altogether 848 passive examples have been analysed in this study. Both the written and the spoken data are consistent in terms of writers’/ speakers’ age group, as they are all in their 20s to 50s. Moreover, all the data were fairly contemporary; written or recorded during the period of 1971 to 2000.

1.2.3 Examples in the text of this study

This research mainly uses authentic written and spoken data for analysis, described in detail in the sections above. Only a limited number of examples have been generated on the evidence of native-speaker intuition, for the sake of simplifying the explanation. With each example from the written data, the author and the page number are given in square brackets, as in the example (1) below:

(1) Meimon

to

iw-are-ru

kono si

Prestigeous quot say-PASS-PRES this

no

Koowa-tyuugaku, ...

[Tsutsui 376]

city gen Kowa.Junior.High.School

‘The Kowa junior high school in this city, which is said to be a prestigious school, ... ’

3 The pre-publication version used contains 5500 utterances. However, the commercially available version has a total of 11000 utterances.

12

For the spoken data, the following information is given in square brackets: the abbreviation of the source (‘Josei’ for Josei no Kotoba - Shoknbahen (Women’s Language - Workplace section) and ‘Dansei’ for Dansei no Kotoba - Shokubahen (Men’s Language - Workplace section)), the utterance number, and the speaker’s gender, age and occupation, as in example (2) below:

(2) Ee,

Nihonzin

no

kanari ooku no

hito

ga

desu nee, Bukkyoo

Well Japanese.people gen fairly many gen people nom cop-pres Buddhism sinzya tosite kazoer-are-te blieveras

i-mas-u. [Dansei 2704: male, 45, University Lecturer]

count-PASS-coNj be-POL-PRES

‘Well, quite a large number of Japanese people are, uh, counted as Buddhists.’

In the few cases in which no source is noted, the example has been created by the current author on the evidence o f native-speaker intuition, as in example (3) below:3

(3) Nobuko Nobuko

wa keisatu ni utagaw-are-te top

police

ir-u.

by suspect-PASS-coNJ be-PRES

‘Nobuko is suspected by the police.’

13

1.3

Motivation for the use of authentic data

Chomsky (1957) claimed that a fundamental goal of linguistic enquiry should be to develop a theory which mirrors a cognitively plausible model of language. What has to be observed is language competence - internalized knowledge of a language - rather than performance - external evidence of language competence (McEnery & Wilson 1996: 5). Chomsky, therefore, suggested that the observation of naturally occurring data could never be meaningful to linguistics enquiry.

Gathering data by using introspection has a great advantage. You can gather data anytime you want, and gather only data that is relevant to your study. However, the process of a speaker’s introspective judgement is unverifiable. In contrast, as McEnery & Wilson (1996: 12) suggest, authentic data is both more public and more objective. Anyone can observe the naturally occurring data. Observations of actual data are more objectively verifiable than observations based on introspective judgement. This is the reason why it is important to look at naturally occurring data.

Furthermore, the ultimate goal of this study is to reveal how Japanese passives actually appear in real contexts. Authentic data, in the context in which they originally occurred, is the most reliable source that portrays the actual usage of these constructions. By observing natural examples in context, we can fully understand the real meanings and functions of Japanese passive constructions.

i

14

1.4

Definition of passive

The definition of passive has been a centre o f controversy for quite some time. Many attempts have been made to characterise the passive from a number of different points of view: morphological, syntactic, semantic, functional/pragmatic, and so forth. In this research, Shibatani’s (1985) prototype approach is adopted, which begins by defining the passive prototype, and then considers how close various constructions are to that prototype. However, in order to view this approach in a broader context, first let us take a general look at some of the other main positions.

Transformational Grammarians describe passives in relation to changes brought about in the structural characteristics of the clause, such as linear ordering and relative dominance (Chomsky 1965, Lakoff 1971, Hasegawa 1968, and Langacker & Munro 1975). Linear order is related to the position of a NP relative to the verb. Keenan (1975: 343) points out that this characterisation must be highly language specific. Under the transformational characterisation, English passive applies to any structure that requires ‘an NP immediately followed by a verb, followed by another NP’. It would clearly not be applicable in many other languages, such as SOV and VSO languages. In terms of relative dominance, Keenan (1975: 343) also suggests that, although subject is usually ‘the highest’ NP, the distinction between subject and direct object in terms of dominance is not clear in ‘free word order’ languages, such as Tagalog (Schachter and Otanes 1972) and Walbiri

15

(Hale 1967).

Relational Grammarians, such as Perlmutter and Postal (1974, 1977: 399) have also argued against the Transformational approach. They stress the change in grammatical relations in characterising the passive. Their emphasis is on the direct object nominal at a transitive level becoming a subject nominal at the next level, passive. As a result, ‘the active SUBJECT case ceases to bear any grammatical relation to its verb’. This approach, however, does not accommodate cases where a non-direct object can be passivised.

In criticising the Transformational Grammarians’ view, discussed above, and modifying the theory of Perlmutter and Postal (1974), Keenan (1975: 340) regards passivisation primarily as a process of the demotion of the agent from the subject position, with the promotion of a non-agent to subject status viewed as a consequence. In this approach the emphasis is on the demotion of agent.

However, Keenan’s view still can not be adapted to instances that do not involve promotion of any participant to the subject position, such as impersonal passives, as indicated by Comrie (1977: 47-48). Comrie suggests the idea of spontaneous demotion, or removal, of a subject in the impersonal passive. In this construction, the ‘underlying subject’ has been demoted spontaneously, not related to the promotion of some other participant to subject, and appears as an oblique object. Consider the following examples from Welsh (Comrie 1977: 55):

16

(4) a.

Fe’i lladdod

ddraig.

him killed

dragon

‘A dragon killed him.’ b.

Fe’i lladdwyd him was-killed

(gan ddraig). by

dragon

‘He was killed by a dragon.’

A direct object in the corresponding active sentence (4a), f e ’i ‘him’, stays as a direct object in the impersonal passive sentence (4b). No other participant is promoted to the subject position, therefore (4b) does not have a subject. Nevertheless the ‘underlying subject’, draig ‘a dragon’, is removed from the subject position in (4a), and appears as an oblique object in (4b). Comrie (1977: 55) claims that this is a lucid illustration of spontaneous demotion o f the ‘underlying subject’.

Givon (1979:186), who takes an explicitly functional stance in his discussion of passive, also criticises Keenan’s (1975: 340) theory since it ‘disregards the function of passives’. Givon (1979:186) defines the passive as follows, focusing on its functional characteristics:

Passivization is the process by which a nonagent is promoted into the role of main topic of the sentence. And to the extent that the language possesses coding

17

properties which identify main topics as subjects and distinguishes them from topics, then this promotion may also involve subjectivalization. (Emphasis in the original.)

In the definition above, Givon (1979:186) manages to portray the functional properties of passive to a certain extent. However, his approach meets opposition from Shibatani (1985: 830), since it still prioritises the promotion of a non-agent.

As opposed to Givon’s argument (1979:186), and further developing the line of Comrie (1977: 48) and Keenan’s (1975: 340) approaches, Shibatani (1985: 830) claims that the primary pragmatic function of the passive is ‘agent-defocusing’. His argument is that, first of all, there are some passives that do not involve the promotion of any non-agent participant to subject, as Comrie (1977: 47-48) suggests regarding impersonal passives. Shibatani (1985: 831) also claims that the fact that passive sentences do not usually involve an overt ‘agent’ in the clause shows that ‘their fundamental function has to do with the defocusing of agents’.

Following Shibatani (1985: 830), Givon (2001: 125) later revises his earlier view, and claims that the primary function of the prototypical passive voice is to demote or defocus the agent. Givon (2001: 126) also mentions the high frequency of the passive without an overt agent in four languages. This is illustrated in the table below (Givon 2001: 126):

18

Percent of non-anaphoric zero agents in active and passive clauses in narrative text Voice construction Language

Active

Passive

Chamorro (Cooreman 1987)

0%

93.5%

Modem Greek (Roland 1994)

0%

93.0%

Karao (Brainard 1994)

0%

90.5%

English (Givon 1979a)

/

80.0%

As will be seen in Section 5.2.2, in our Japanese data search also, nearly 80% of the passive clauses do not involve an overt ‘agent’ in the clause. Along with Shibatani (1985: 831), Givon (2001: 126) also suggests that the fact that a large proportion of passives do not involve an overt ‘agent’ demonstrates that ‘agent’ suppression is the foremost function of the passive voice.

In analysing the correlations o f passives to other related constructions, Shibatani (1985: 821-822) also proposes a prototype approach to characterise the passive. He claims that it is meaningless to discuss whether or not a given construction should come under the passive domain. Rather, since various constructions are lined up along a continuum, the real question is how much the construction is related to or differs from the prototype. Shibatani (1985: 837) characterises the passive prototype as follows:

19

Characterization of the passive prototype a. Primary pragmatic function: Defocusing o f agent. b. Semantic properties: (i)

Semantic valence: Predicate (agent, patient).

(ii)

Subject is affected.

c. Syntactic properties: (i)

Syntactic encoding: agent -> 0 (not encoded). patient -> subject.

(ii)

Valence of Predicate]: Active = P/n; Passive = P/n-1.

d. Morphological property: Active = P; Passive = P [+passive].

This prototype is advocated with a view to the universal characterisation of the passive prototype. This study, however, deals principally with Japanese passives. It is therefore necessary, in this study, to modify Shibatani’s characterisation of passive slightly in order to accommodate the features of Japanese passive constructions. Japanese passives do not include impei'sonal passives (like those Comrie exemplifies in Welsh), and all the passives in Japanese involve a subject from whose perspective the event is described, although this subject is often elided in the clause. In this research, therefore, the fundamental function of Japanese passives is considered as bringing the focus to the subject, regardless of whether it is promoted from non-agent or not. The defocusing of the agent is regarded as the second important function of passives in this language. In the case of Japanese passives, then,

L

>

20

criterion (a) above is revised as follows:

Characterisation of the passive prototype (in Japanese) a. Primary pragmatic function: Bringing focus to subject. Secondary pragmatic function: Defocusing of agent.

As an example o f the prototypical passive, consider the following example from the data used from this study:

(5) a. Teki

wa ooku no

Enemy

top

hito

o

totunyuu tyokugo.ni korosi-ta.

many gen people acc break-in

right.after

kill-PAST

‘The enemy killed many people right after the break-in.’ b. Ooku no

hito

wa totunyuu tyokugo.ni (teki

Many gen people top break-in right.after

ni) koros-are-ta.

(enemy by)

k ill-P A S S -P A S T

‘Many people were killed right after the enemy’s break-in.’ [Shiono 409, modified]

Example (5b) satisfies all the criteria for the prototypical passive of Japanese passive: its subject (many people) is focused, even topicalised; its ‘agent’ (the enemy) is defocused and is not encoded; its active counterpart (5a) involves an ‘agent’ (the enemy) and a patient (many people); and the patient becomes the passive subject which is affected by the event; its valence is decreased by 1 compared to its active counterpart (5a); its verb involves the passive morpheme, -(r)are.

21

Next, let us consider a non-prototypical case. The indirect passive in Japanese differs from the passive prototype above in several respects. Consider the example below:

(6) a.

Kodomo ga bonnetto no

ue ni

not-te

tatioozyoos-ita koto nado

Children nom bonnet gen top on get.on-coNJ be.stuck-PAST mo

case etc.

at-ta.

even there.is-PAST b. (Watasi wa) (kodomo ni) bonnetto no (I

top) (children

koto nado mo case etc.

by) bonnet

at-ta.

gen

ue ni nor-are-te

tatioozyoos-ita

top on get.on-PASS-coNJ be.stuck-PAST

[Fujiwara 340]

even there.is-PAST

‘Furthermore there even was a time when I was stuck because (the kids) got on the bonnet (of my car).’

Since an indirect passive like example (6b) is an intransitive-based passive, it does not involve a patient. It also differs from the prototype in that it increases rather than decreases the valence of the verb root by 1, compared to the closest active equivalent (6a). Despite these facts, in this study, this kind of passive is regarded as passive, as it fulfils the most important factors of the passive prototype: the focus is brought to the subject, I, (although elided in the sentence (6b)); the event is described from the point of view of the subject. Moreover, the ‘agent’ (‘the kids’ in example (6b)) is defocused and is not encoded. This kind of passive also satisfies the last criterion of passive prototype: its verb includes the

.

22

passive morpheme, -(r)are.

Grounded on the discussion above, this research considers the subject focusing and the ‘agent’ defocusing to be the key functional criteria of the passive. The indirect passive in Japanese (as in example (6b)), along with the direct passive (as in example (5b)), therefore, is regarded as passive to the extent that it involves these primary and secondary functions of passive, bringing focus to the subject and defocusing an ‘agent’, and it shares the passive morpheme, -(r)are.

1.5

~(r)are constructions in Japanese

In Japanese, the verbal morpheme -(r)are is used in spontaneous, potential and honorific constructions, as well as in the passive construction. However, the main focus of discussion in previous research has been the passive use of -(r)are, and the non-passive uses of -(r)are have not attracted much attention. In this section, we will briefly examine the relationship between the passive use and the non-passive uses of -(r)are. Examples o f these four uses of the -(r)are construction — passive, spontaneous, potential and honorific follows:

(7) passive use Otoko wa keibiin Man

top

ni mise kara hooridas-are-ta.

security.guard by shop from throw.out-RARE-PAST

23

‘The man was thrown out from the shop by a security guard.’ (g) spontaneous use Syoogatu ni

naru

to

(watasi ni

New.year dat become when (I

wa) hurusato

dat top)

ga

omoidas-are-ru.

home.town nom recall-RARE-PRES

‘When the New Year comes, I (always) remember my home town.’ (9) potential use Watasi (ni) I

wa niku ryoori ga

(dat) top meat dish

nom

tabe-rare-nai. eat-RARE-NEO

‘I cannot eat meat dishes.’ (10) honorific use Takada sensei

wa sensyuu

Doitu

ni tat-are-ta.

Takada teacher top last.week Germany to leave-RARE-PAST ‘Professor Takada left for Germany last week (honorific).’

In terms of the historical development of -(r)are morphology, in the Nara period (7th century) the antecedent of -(r)are, -yu / rayu, was mainly used, and later -ru / raru became dominant. Shibatani (2000: 163) mentions that because -yu / rayu did not have an honorific use at all, and that the honorific use of -ru / raru was not acquired until after the 81,1 century. For these reasons the honorific use is thought to have developed later than the spontaneous, potential and passive uses.

Shibatani also observes that before the Heian period (9lh century), the potential

24

use o f -ru/raru only occurred in a negative context. This is the reason why he infers that the potential use developed later than the spontaneous and passive uses. Before the 7lh century, therefore, Shibatani suggests that -yu/rayu was primarily used for the spontaneous and the passive.

For the period for which there are no surviving records, Hosoe (1928) maintains that, in Japanese, the spontaneous and the passive formed the middle voice, and were represented by the same morpheme, -yu/rayu. Shibatani (2000: 166) further develops Hosoe’s view and takes a stance that even before that period, the opposition between active and spontaneous was held, and the passive use was derived from the spontaneous use.

To sum up the historical development of the ~(r)are constructions, the primary use o f -(r)are, or its antecedents -y u /ra y u and ru /ra ru , in classical Japanese was for the spontaneous, and all other uses were developed later. First, the passive was derived from the spontaneous, then the potential and finally the honorific. In this sense, the passive, potential and honorific are all connected to the spontaneous in some way. However, as Shibatani (2000: 161) points out, this is a historical matter, and it has not yet been determined whether or not the common basis or core meaning of -(r)are constructions can be accepted as a synchronic grammatical knowledge.

Shibatani (2000: 162) states that, in general, morphological or syntactic similarities between constructions are due to the fact that one construction was historically

25

developed from the other. In this case, however, the similarities may be merely historical inheritance, and the connection between the constructions might not have any synchronic meaning. Ambiguity between independent constructions can be due to morphological polysemy. Therefore, Shibatani (2000: 162) claims that the fact that one sentence can be interpreted as passive or honorific, does not prove that these two constructions share a common basis today. He suggests that we need to examine how these constructions coexist in the modem Japanese language (the synchronic aspect) as well as the historical development of these constructions (the diachronic aspect).

Next, let us examine the differences and similarities of the four uses of -(r)are constructions in contemporary Japanese, in order to clarify the connection between the constructions, more specifically to elucidate the relationship between the passive use and other uses o f-(r)are in the modem Japanese language (the synchronic aspect). Consider the examples (7) - (10), cited again below:

(7) passive use Otoko wa keibiin Man

top

ni mise kara hooridas-are-ta.

security.guard by shop from throw.out-RARE-PAST

‘The man was thrown out from the shop by a security guard.’ (8) spontaneous use Syoogatu ni

nam

to

(watasi ni

New.year dat become when (I

wa) hurusato

ga

omoidas-are-ru.

DAT top) home.town nom recall-RARE-PRES

26

‘When the New Year comes, I (always) remember my home town.’ (9) potential use Watasi (ni) I

wa niku ryoori ga

(dat) top meat dish

nom

tabe-rare-nai. eat-RARE-NEG

‘I cannot eat meat dishes.’ (10) honorific use Takada sensei

ga

sensyuu

Doitu

ni tat-are-ta.

Takada teacher nom last.week Germany to leave-RARE-PAST ‘Professor Takada left for Germany last week (honorific).’

The honorific use seems totally different from other three in that it has the same perspective as that of a non-honorific active clause. In the honorific example (10), the agent is marked by ga, in the same way as in an unmarked active clause without ~(r)are. In fact, it does not involve any change in case marking compared to the corresponding active clause without -(r)are.

The spontaneous example (8) and the potential example (9) are almost identical in terms of case marking. They both have the agent (watasi T in both sentences) marked by dative particle ni when it appears, this ni often being complemented or replaced by the topic marker wa. The passive example (7), in fact, also has the agent (keibiin ‘security guard’) marked by dative ni. However, unlike the agent in the spontaneous or the potential clause, it is very rare for the dative ni marked agent in the passive clause to be topicalised.

27

This is clearly related to the fact that the passive serves to defocus this participant. Another noticeable point is that the passive clause, the spontaneous clause and the potential clause all have an Undergoer marked by nominative particle ga. In the case of the passive use, this ga is often replaced by topic marker wa. The Undergoer in the potential use (niku tyoori ‘meat dishes’ in (9)) can also be topicalised, whereas the Undergoer in the spontaneous use (hurusato ‘home town’ in (8)) rarely can. We, therefore, conclude that these three uses the passive, the spontaneous and the potential - are fundamentally different in terms of the perspective from which the event is described. The passive clause usually describes the situation from the point of view of the Undergoer, whereas in the spontaneous clause the event is portrayed from the agent’s perspective. The potential clause can portray the dual perspective of the agent and the Undergoer.

Moreover, in modem Japanese (synchronically), Shibatani (2000: 173) claims that the passive and the spontaneous are closely related in terms of forming a voice opposition with the active. However, he considers the potential and the honorific as existing on another level as they were developed through pragmatic motivation. Shibatani also states that the potential construction describes a state, and it belongs to the domain of modality rather than that of voice, although it is close to the spontaneous semantically. In contrast, it is hard to find synchronic affinity in the development from spontaneous to honorific.

In considering the discussion above, in this research, we recognise the way each

A

28

of the four functions o f -(r)are developed historically, and the similarities and differences among the four. We also acknowledge that there probably is some core meaning common to all four uses of -(.r)are, but we regard each of the four uses of-(r)are as an independent construction. Therefore, we will only deal with the passive use of -(r)are in this study.

1.6

Basic terms and concepts

In this section, we will define and illustrate the basic terms and concepts used in this research. First, Section 1.6.1 deals with clarifying the difference in verb types in Japanese: the transitive verb and the intransitive verb, and the unaccusative verb and the unergative verb. In Section 1.6.2, we will discuss the concepts o f ‘Actor’ and ‘Undergoer’, introduced in Foley and Van Valin (1984: 28-29), in comparison with thematic roles defined in previous research. Section 1.6.3 describes how sentience constrains certain constructions in Japanese, especially passive, and considers the concepts o f ‘sentient’ and ‘non-sentient’.

1.6.1 Verb types in Japanese

One way of classifying Japanese verbs is to divide them into three groups: the transitive verb, the unaccusative intransitive verb and the unergative intransitive verb. First, we will divide verbs into two groups using the traditional distinction of transitive and intransitive

29

verbs. I will then introduce the notion of unaccusative and unergative verbs, following Kageyama (1995: 43) which roughly corresponds to Mikami’s distinction of inactive and active verbs (in Mikami 1953, 1972).

It is the intransitive verbs that are subdivided into

these two groups: unaccusative verbs and unergative verbs. Finally, we lay down a definition of these two subcategories of intransitive verb used in this study, applying Dixon’s (1994: 6-8) concepts of S, A, and O.

1.6.1.1 Transitive and intransitive verbs

There are several theories with regard to the distinction between syntactically transitive verbs and intransitive verbs in Japanese. I will define syntactically transitive

verbs here

simply as those which have an object marked by particle o (excluding deictic motion verbs for which the o-marked NP signals the ground covered, such as (miti o) iku ‘go along the street’ and (heya o) deru ‘leave the room’), and those which have a single dative object marked by particle ni, such as hoeru ‘to bark’, tobituku ‘jump at’, horeru ‘fall in love’, etc. Nomura (1995) calls this second group w'-transitive verbs. There are also quite a large number of verbs which have both an accusative object marked by o and a dative object marked by ni, such as susumeru ‘to recommend’, osieru ‘to teach’, tanomu ‘to ask (a favour)’. They are often called ditransitive verbs and are grouped here in the broad category of transitive verbs. If the verb is not identified as a syntactically transitive verb by these criteria, then it would be considered as an intransitive verb.

30

1.6.1.2 Unaccusative and unergative verbs

The subgroups of syntactically intransitive verbs - unaccusative verbs and unergative verbs - are widely recognised by GB theorists. Relational Grammarians and Lexical Functional Grammarians. Among Japanese linguists, these categories have also been widely discussed. Motoori Haruniwa noted the distinction in the early 19th century, of ‘onozukara sikaru' (something happens by itself, something comes about naturally) and ‘mizukara sika suru’ (do something by oneself) (Motoori 1828). Mikami (1972) also classified Japanese verbs into ‘inactive’, which correspond to the unaccusative verbs, ‘active’, which

correspond to

the unergative verbs together with the transitive verbs. More recently, Miyagawa (1989), Sagawa (1991), Kageyama (1995: 43) and some other linguists have begun to look again at these two basic types o f verbs in Japanese.

Although Dixon does not utilize the terms unaccusative verbs and unergative verbs, he, too, mentions the distinction between these two basic types of intransitive verb (Dixon 1994: 53). Since his definition is very clear and useful in comprehending the concept as it relates to Japanese passives, we will consider it below.

First, Dixon (1994: 6-8) states the premise that all languages work in terns of three primitive relations:

31

S - intransitive subject A - transitive subject O - transitive object

Dixon (1994: 7) maintains that in an intransitive clause, the single core argument is mapped onto S. In terms of the two core arguments in a transitive clause, one is mapped onto A and the other onto O.

In investigating verb classes, Dixon (1994: 7) recognises a number o f ‘semantic roles’ associated with predicates —different ‘semantic roles’ with different ‘semantic types of predicates. ‘Some of the semantic types of verbs which appear' in all languages are summarised below (with Dixon’s examples from English):

SEMANTIC TYPES

Semantic Roles

AFFECT, c.g. hit, cut, bum

Agent, Manip (thing manipulated), Target

GIVING, c.g. give, lend, pay

Donor, Gift, Recipient

SPEAKING, c.g. talk, tell, order

Speaker, Addressee, Message

ATTENTION, e.g. see, hear, watch

Perceiver, Impression

Dixon (1994: 8) claims that it is ‘the Agent for an AFFECT verb, Donor for a GIVING verb, Speaker for a SPEAKING verb, Perceiver for an ATTENTION verb and so on, that is placed in the A function. If there are just two core roles, then the one not identified as A

32

will be mapped onto O.

Regarding intransitives, Dixon (1994: 53) states that ‘for some intransitive verbs the referent o f the S NP would be likely to be the controller o f the activity, e.g. “jump”, “speak”; this subtype of S can be called Sa. For other verbs the referent of the S NP is not likely to control the event but may be affected by it, e.g. “break”, “die”, “yawn” - this can be called So’.

In the light o f Dixon’s definition, we will define the two types o f intransitive verbs as follows:

Unergative verb: is one that has Sa, which

performs the action described by

the verb intentionally, and could “initiate or control” the event in the same way as A, the subject of an archetypal transitive clause, e.g. okiru ‘to get up’, nigeru ‘to run away’, yasumu ‘to rest’, etc.

Unaccusative verb: is one that has So, which is semantically more like O, the object of a transitive clause, in that is not thought o f as controlling the event but is involved in the event described by the verb unintentionally, e.g. kimaru ‘to be decided’, sakeru ‘to tear’, tokeru ‘to melt’, etc.

Note that there are some intransitive verbs, such as agaru ‘to go up/rise’ and utum ‘to

33

move / permeate’, which can hold both unergative and unaccusative interpretations, generally depending on the animacy of S.

Mikami’s (1972) notion of inactive and active verbs is based on the applicability of passivisation. He classifies verbs that cannot appear in passive clauses as ‘inactive’ verbs, and ones that can occur in passives as ‘active’ verbs. Kageyama (1995: 59) also mentions that unaccusative verbs cannot occur in passive constructions, especially in indirect passives. In Chapter 4, we will argue against this claim, showing evidence of such use in Japanese novels. We also examine the context in which such verbs are used and the implications of their use in such context.

1.6.2 Actor and Undergoer

In adopting Dowty’s (1979) scheme of lexical decomposition based on Vendler’s (1967) verb classification system, Foley and Van Valin (1984: 27-32) introduce the notion of the semantic macroroles, ‘Actor’ and ‘Undergoer’. They mention that ‘actor and undergoer are the two arguments in a transitive predication, either one of which may be the single argument of an intransitive verb’ (Foley and Van Valin 1984: 27). If one assumes, in classifying verbs, that there are several semantic types, as Dowty does, each type of verb assigns a different thematic role to the subject and to the object. (This is similar to Dixon’s view as seen in Section 1.6.1.2.) For instance, the basic thematic role of the subject of a

34

type I verb might be ‘Agent’, and that of its object, ‘Patient’. A type II verb might have different thematic role for the subject and for the object, and so forth. Foley and Van Valin (1984: 28-32) propose the macroroles ‘Actor’ and ‘Undergoer’ to represent all the roles basic to the two core arguments in all types of transitive clause proposed by Dowty (1979). Either ‘Actor’ or ‘Undergoer’ may denote ‘the single argument o f an intransitive verb. Foley and Van Valin (1984: 30) list the examples in (11) to illustrate types of the Actor, and those in (12) to illustrate types of the Undergoer:

(11) a. Colin killed the taipan.

(Agent)

b. The rock shattered the mirror.

(Instrument)

c. The lawyer received a telegram.

(Recipient/ Goal)

d. The dog sensed the earthquake.

(Experiencer)

e. The sun emits radiation.

(Source)

(12) a. Phil threw the ball to the umpire.

(Theme)

b. The avalanche crushed the cottage.

(Patient)

c. The arrow hit the target.

(Locative)

d. The mugger robbed Fred of $50.00.

(Source)

e. The announcer presented Mary with the award.

(Recipient/ Goal)

In terms of the case role/ thematic relation, Colin in (11a) is an agent, the rock in (lib ) is an instrument, the lawyer in (11c) is a recipient/ goal, the dog in (lid ) is an experiencer, the sun in (lie ) is a source, and yet they are all Actors. In the same way, the ball in (12a) is a

35

theme, the cottage in (12b) is a patient, the target in (12c) is a locative, Fred in (12d) is a source, Mary in (12e) is a recipient/ goal, nevertheless they are all Undergoers. Foley and Van Valin (1984: 29) state that, on one hand, the Actor is ‘the argument of a predicate which expresses the participant which performs, effects, instigates, or controls the situation denoted by the predicate’. On the other hand, the Undergoer is ‘the argument which expresses the participant which does not perform, initiate, or control any situation but rather is affected by it in some way’.

In the case of the intransitive clause, the unergative4 verb is now characterised as one that has Actor as subject (or Sa), and the unaccusative verb is one that has Undergoer as subject (or S0). Some Japanese examples follow:

(13) Kodomo.tati wa kuruma no Children

top

car

gen

bonnetto no

ue ni n o t-ta.

bonnet

top on get-PAST

gen

‘The kids got on the bonnet of my car.’ (14) Obaatyan no Granny

gen

yo-nin no

musuko wa sensoo de sin-da.

four-CLFgen son

top

war

in die-PAST

‘Four of granny’s sons died in the war.’

The verb in example (13), noru ‘to get on’, is an unergative verb. The Sa, kodomo lad ‘children’, is an Actor, and it intentionally initiates the activity of getting on the bonnet of 4 See Section 1.6.1.2 for the definition of the unergative verb and the unaccusative verb, and that of Sa and S0.

36

the speaker’s car. In example (14), the verb, sinu ‘to die’, is regarded as an unaccusative verb. It has an Undergoer S0, yonin no musuko ‘four sons’, which does not initiate or control the event o f dying, but is affected by it.

Foley and Van Valin (1984: 31) also suggest that one can simplify the characterisation of passivisation, using the notions of Actor and Undergoer. They portray the English passive as ‘the undergoer occurring as syntactic subject and the actor as the object of by, if it occurs at all'. However, as seen in Section 1.4, there are some passives in Japanese that do not involve promotion of any participant to the subject position, such as example (16b) below. Therefore, Foley and Van Valin’s (1984: 31) characterisation of English passive works with examples o f the direct passive in Japanese, like (15b), but not with examples o f the indirect passive, like (16b).

(15) a.

Keiko ga

Hitosi

o

nagut-ta.

Keiko nom Hitoshi acc hit-PAST 'Keiko hit Hitoshi.' b. Hitosi ga

Keiko

Hitoshi nom Keiko

ni

nagura-re-ta.

by hit-PASS-PAST

'Hitoshi was hit by Keiko.' (16) a. Kinoo

ame ga

hut-ta.

Yesterday rain nom fall-PAST ‘It rained yesterday.’

37

b. Takasi

ga

kinoo

ame ni hur-are-ta.

Takashi nom yesterday rain by fall-PASS-PAST ‘Takashi was adversely affected by the rain falling yesterday.’

In example (15b), the Undergoer, Hitoshi, occurs as passive subject, and the Actor, Keiko, as the object of the particle ni, which corresponds to English ‘by’. In the case of example (16b), however, it is not so straightforward. The corresponding active sentence, example (16a), involves an unaccusative verb, hunt ‘to fall’, and the S„ is the Undergoer. The object of ni in (16b) (ame ‘rain’) is, therefore, the Undergoer. The passive subject, Takashi, does not correspond to any participant in (16a); it is neither the Undergoer nor the Actor. We will discuss various issues of this type of passive, the indirect passive, in detail in Sections 3.1.4 and 4.1.

1.7

Sentience and passive constructions in Japanese

The issue of the animacy or sentience of the subject and animacy or sentience of the ‘actor’5 in passive constructions has been widely discussed amongst researchers of Japanese passives. The sentience of the subject of the passive, especially, has been an issue not only in recent times but also for traditional Japanese linguists. Yamada (1908: 374)

5 The term ‘actor’ (with the lower-case) is used in this study in a general way to refer to the participant marked by ni or niyotte in a passive sentence. They are mostly Actors (in Foley and Van-Valin’s terms), except those with an unaccusative verb. (See Section 1.6.2.)

38

claims that ‘the Japanese passive is very closely related to sentience, and in most cases, if a non-sentient NP (in his words seisin naki mono ‘a NP without mind’) takes the subject position, the passive cannot be constructed’. More recently, Song (1993: 85) also states clearly that ‘the Japanese passive is strongly constrained by animacy. The passive with an inanimate subject is very restricted.' Among traditional Japanese linguists, the passive with an inanimate, or to be more precise, non-sentient, subject is called the hijo no ukemi ‘non-sentient passive’. The term ‘sentient’ is defined in this research as having character, intuition or perception, and being able to see or feel things through the senses. The sentient being differs from the animate being as it includes mainly human beings, and only some animals and personified inanimates that are perceived as having senses, such as those described above. It is sentience rather than animacy that has close connection with Japanese passive constructions. Therefore, the relationship between sentience and Japanese passives is considered in this section.

The traditional analysis deals mainly with sentience of the subject of the direct passive. In this section, we will cover the sentience of the subject and the ‘actor’ (the participant marked by ni) in the intransitive-based passive, as well as in the direct passive. This is because, in these cases, the sentience is particularly relevant. In this section, however, intransitive-based passives are mentioned only briefly (Section 1.7.1), since they are dealt with in detail in Section 4.1. We will then focus (Section 1.7.2) on the direct passive and thus on transitive verb clauses.

39

1.7.1 Sentience of the participants in intransitive-based passives

As will be seen in Section 4.1, the sentience of the ‘actor’ of an intransitive verb has been widely acknowledged to have a great influence on the acceptability of a passive clause in Japanese. The sentience of the subject of the intransitive-based passive, however, does not seem to have received much attention in the literature, compared to that o f the ‘actor’ of the passive. The primary characteristic of the intransitive-based passive in Japanese is that ‘the speaker describes an event in terms of the concerns of a participant denoted by the subject NP’ (Song, 1993: 98). The subject therefore is most often sentient. This may be the reason why the subject of the indirect passive seems less likely to be mentioned in the sentence at all. We will discuss these issues further in Section 4.1 in relation to the issue of adversative meaning.

1.7.2 Sentience of the participants in direct passives

It seems that in direct passives the sentience of the subject o f the passive has more influence upon the acceptability of passivisation than the sentience of the ‘actor’. We will begin, however, by briefly discussing the issue o f the sentience o f the ‘actor’, and then go on to address that of the sentience of the subject in more detail.

40

1.7.2.1 Sentience of the ‘actor’ in the direct passive

The ‘actor’ in a direct passive clause would be the subject of a transitive verb in the corresponding active clause. As seen in Section 1.6.1.2, Dixon (1994: 6-8) uses the term A to refer to the subject of a transitive verb: the participant that is ‘most relevant to the success of the activity’ and thus ‘could initiate or control the activity’. He notes that ‘this can be something inanimate (as in The wind wrecked the house, The midday sun melted the butter); most often the role mapped onto A will be human’.

It is interesting that Dixon’s examples o f inanimate As are both so-called ‘weather words’ (the wind, the midday sun). As will be seen in Section 4.1, in Japanese, weather words are sometimes treated like animate NPs. They are quite easily personified. For example, they can be the addressee in imperative sentences.

Dixon’s (1994: 8) claim that ‘most often the role mapped onto A will be human’ is certainly true for Japanese. For this language at least it would also be reasonable to add that those that are not human are ‘animate’ in a broad sense (including weather words and personified inanimate NPs). If most of the subjects of transitive clauses are at least animate, if not human, there is no need to argue the animacy of the ‘actor’ of their passive counterparts.

41

1.7.2.2 Sentience of the subject in the direct passive

As mentioned above, the sentience of the subject of the passive has been a major point of discussion for many researchers o f Japanese passives. Traditional Japanese grammarians argued that the passive with a non-sentient subject is not inherent to the Japanese language, but began to occur under the influence of seiyo-go ‘Western languages’. (See Yarnada 1908: 374 and Hashimoto 1931: 276). Recently Okutsu (1992: 7) argued against this theory, showing evidence of some use of passives with non-sentient subject in Japanese classical literature.

It is true, however, that the subject of the passive in Japanese is most likely to be sentient. This is not only because of the fact that the Japanese passive with an non-sentient subject is very restricted, but also because o f the Empathy Hierarchy (Kuno 1977: 646 & 652, Silverstein 1986, and Okutsu 1992). (See also Section 5.2.3). Generally speaking, the speaker is more likely to adopt the viewpoint of a sentient NP over that of a non-sentient NP. The NP whose viewpoint is adopted, in turn, is most often identified as the subject NP.

It is also true, however, that quite a number of passive sentences in Japanese do have a non-sentient subject. We will make introductory remarks on the context in which such sentences are used in the next section. However, detailed examination will be made in Section 3.2.1. We will also discuss the actual distribution of this type of passive in our data

42

in Section 5.2.3.

1.7.2.3 Direct passive with a non-scntient subject

Song (1993: 103) classifies passives with a non-sentient subject into two groups, ‘anticausative passive’ and ‘attributive passive’. Kinsui (1992: 14) divides Japanese passives first into two categories, koyu no ukemi ‘inherent passive’ and hi-koyu no ukemi ‘non-inherent/ imported passive’. He then divides both of these categories into two further subcategories, yujo no ukemi ‘sentient passive’ and hijo no ukemi ‘non-sentient passive’ resulting in four subcategories. Song’s anticausative passive seems to correspond to Kinsui’s hi-koyu-hijd no ukemi ‘non-inherent, non-sentient passive’, or the so-called ‘ni-yotte’ passive.1 We use the term ‘demotional passive’ for this type of passive. Song’s other category, attributive passive seems to correspond to Kinsui’s koyu-hijo no ukemi ‘inherent, non-sentient passive’. We will adopt Song’s term ‘attributive passive’ for this type of passive, since the name reflects the function of this type of passive very well. In Section 3.2.1, we define each of these two types of passives with non-sentient subject and discuss their function.

43

Chapter 2

2.1

Classification of Japanese passives - Previous analyses

Introduction

Over the course of the twentieth century, many researchers, such as Yamada (1908: 371-380), Matsushita (1930/1977: 157-161), Mikami (1953/1972: 98-112), Kuno (1973: 22-24), Shibatani (1978: 133-142), Teramura (1982: 214-217), Nitta (1991: 31-35) and Masuoka (2000: 55-68), have attempted to classify Japanese passives into several basic types. We will go over their views, dividing them into three main groups: earlier approaches, the mainstream approach and more recent approaches, following the chronological development of ideas. Among earlier approaches, first, we will examine Yamada’s view. Although he did not actually try to categorise Japanese passives, his work is a good starting point for comparison with later studies. We will then introduce Matsushita’s analysis as the major work in this early period. Matsushita divides Japanese passives into two groups, the passive o f interest and the plain passive, according to their semantic features. Following these early approaches comes what we refer to as the mainstream approach. Scholars such as Mikami, Kuno, Shibatani, and Teramura, who adopt this approach, apply a syntactic dichotomy in analysing Japanese passives, resulting in their classification of direct and indirect. This syntactic approach is still the one most commonly used. In a more recent approach, however, Masuoka seems to go back in time to something similar to Matsushita’s early semantic analysis. We will see, however, how

44

Masuoka’s approach has improved and polished Matsushita’s original analysis. Lastly, Nitta’s recent analysis is considered, especially with respect to the treatment of possessor passives.

2.2

Earlier approaches

2.2.1

Yamada's approach

Yamada Yoshio, one of the first researchers to adopt a scholarly approach to the study of language in Japan, contributed to making modem Japanese grammar an academic field. However, in terms of the study of Japanese passive constructions, his focus was on describing the differences between Japanese passives and those in English and German. For instance, Yamada observes that in Japanese, unlike in English, there are passive sentences involving not only transitive verbs but also intransitive verbs. However, he does not go on to explain the difference between the passive o f a transitive verb and that of an intransitive verb in Japanese.

Yamada also states that the Japanese passive is restricted to cases in which the subject is a sentient NP, that is one that can be thought o f as having consciousness. He gives the following, unacceptable example. (Yamada 1908: 373)

45

(1) *Kano hasi

wa w agayuuzinni

That bridge top my

friend by

tukur-are-tari. build-PASS-PAST

‘That bridge was made by my friend.’

Although the sentence above is unacceptable, if the agent marker ni were replaced by ni-yotte, as in example (2) below, the sentence would be grammatical. This is the type of passive that Matsushita (1930) recognises as the ‘plain passive’ or so-called ‘ni-yotte passive’.

(2) Kano hasi

wa waga yuuzin ni-votte

That bridge top my

friend by

tukur-are-tari. build-PASS-PAST

‘That bridge was made by my friend.’

This kind of passive is said not to be inherent to the Japanese language, but to have developed fairly recently under influence from the translation of Western languages. Kinsui (1997: 762) claims that ni-yotte was not used to mark the agent in a passive sentence until the word-for-word translation of Dutch grammar appeared in the late 19th century. We can, thus, speculate that in Yamada’s days, around 1908, ni-yotte was not yet widely accepted as an agent marker in a passive sentence, as it was by Matsushita’s time, around 1930. We will discuss this matter further in relation to Matsushita’s analysis in the next section.

46

2.2.2

Matsushita’s analysis

Matsushita Daizaburo’s work is said to be one of the first major comprehensive studies of passive constructions in modern Japanese. Matsushita (1930 / 1977: 157-161) classifies Japanese passives, firstly, into two groups - ‘A. Plain / simple passive’ and ‘B. Passive of interest’ - according to their semantic features. He then subdivides B, passive of interest, into four groups. Matsushita’s classification is summarised in the following table:

Table I: Matsushita’s classification

A. Plain passive B. Passive of interest i.

one in which the subject itself is affected by the action

(self passive)

ii.

one in which possessions/relations of the subject are affected by the action (possessor passive)

iii. one in which the subject is affected by the action o f his/her possessions/relations (possessor self passive) iv.

one in which the subject is affected by the action of a third party passive)

Matsushita describes each group as follows:

(third party

47

A. Plain / simple passive: is one that does not have a special meaning of affectedness at all. This type is not inherent to the Japanese language.

(3) Ie

goto.ni kadomatu

House each

ga

tate-rare-ta.

decorative.pine.trees nom set.up-PASS-PAST

‘New Year’s gate decoration pines were put up in front of every house.’ (4) Ziti

seido

ga

sik-are

kokkai

ga

syoosyuu-s-are-ta.

Autonomy system nom place-PASS the.Diet nom call-PASS-PAST ‘A system of self-government was promulgated, and the Diet was assembled.’

Matsushita (1930 / 1977) claims that the plain passive ‘is not inherent to the Japanese language’, and ‘it is used in a spoken sentence only when it is derived from a written sentence which has the style of a literal translation of a Western language’. However, Masuoka (1987) and other more recent researchers claim that, among the passives with an inanimate subject (in Matsushita’s words ‘plain passive’), there actually is a type that is inherent to the Japanese language. I will discuss this in relation to Masuoka’s analysis (1987, 2000) in Section 2.4.1.

B. Passive of interest: is one in which the subject is treated as an individual and is affected adversely or beneficially by the action o f somebody or something else. Matsushita identifies the following four subgroups in this type of passive (Matsushita 1930: 159; his examples):

48

i)

one in which the subject itself is affected by the action (self passive)

(5) Kodomo ga Child

nom

inu

ni

kam-are-m. (transitive)

dog

by

bite-PASS-PRES

‘A child is bitten by a dog.’ (6) Kodomo ga Child

nom

inu ni

tobituk-are-ru. (intransitive1)

dog by jump.at-PAST-PRES

‘A child is affected by a dog’s jumping (at him).’

ii)

one in which possessions/relations of the subject are affected by the action (possessor

passive) (7) Busi

ga

teki

ni katana o

Samurai nom enemy by sword

otos-are-ru. (transitive)

ACC drop-PASS-PRES

‘A samurai is affected by having his sword slashed out o f his hand by his enemy.’ (8) Busi

ga

teki

ni temoto e

Samurai nom enemy by hand

tobikom-are-ru2. (intransitive)

at jump.in-PASS-PRES

‘A samurai is affected by his enemy’s jumping right in front of him.’

1 Although according to the definition of transitive and intransitive verbs in this study the verb lobiluku ‘jump at’ would be categorised as transitive, Matsushita identifies it as intransitive. 2 This example does not seem to illustrate Matsushita’s point well. I assume that, by giving this example, Matsushita means that the samurai’s “personal safety” is affected by the action. If one’s opponent is too close, it is difficult to use one’s sword, so presumably in that way the samurai is adversely affected.

49

iii)

one in which the subject is affected by the action of his/her possessions/relations

(possessor self passive) (9) Teisyu

ga

nyoobo ni syaku o

Husband nom wife

okos-are-ru. (transitive)

by temper acc have-PASS-PRES

‘A husband is affected by his wife’s losing her temper.’ (10) Teisyu

ga

nyoobo ni sina-re-ru. (intransitive)

Husband nom wife

by die-PASS-PRES

‘A husband is affected by his wife’s dying (on him).’

iv)

one in which the subject is affected by the action of a third party (third party passive)

(11) Tanin

ni na

o

nas-are-m. (transitive)

Stranger by name acc make-PASS-PRES ‘One is affected by another’s making a name (for himself).’ (12) Tanin

ni seikoos-are-ru. (intransitive)

Stranger by

succeed-PASS-PRES

‘One is affected by another’s succeeding.’

Matsushita states that all passives with an animate or personified inanimate subject, in his words passives of interest, have the meaning of adversative or beneficial effect. This view is still basically shared by some current researchers, such as Masuoka (1987) and Song (1993). However, Shibatani (2000: 175) counters that it is possible to describe an event from a neutral point of view in a passive with an animate subject, and he

50

gives the following example (Shibatani 2000: 176):

(13) Kodomo ga Child

inu ni kam-are-ta

nom

no

o mi-te

wareware wa issei.ni

dog by bite-PASS-PAST nml acc see-coNJ we

top

all.at.once

tobidasi-ta. rush.out-PAST ‘We saw that a child was bitten by a dog, and rushed out all at once.’

Example (13) is a variation of Matsushita’s example (5), an example of type (i) of the passive of interest (those in which the subject itself is affected by the action - ‘self passive’). It is cited again below.

(5) Kodomo ga Child

nom

inu ni kam-are-ru. (transitive) dog by bite-PASS-PRES

‘A child is bitten by a dog.’

Example (13) includes example (5) in a subordinate clause in the past tense. Shibatani claims that example (13) does not have the same kind o f adversative reading as Matsushita’s examples of type (iv) o f the passive of interest do (those in which the subject is affected by the action of a third party), as in example (11), cited again below.

51

(11) Tanin

ni na

o

nas-are-ru. (transitive)

Stranger by name acc make-PASS-PRBS ‘One is affected by another’s making a name (for himself).’

Shibatani (2000) states that if one can detect some adversative meaning in example (13), it would merely come from the lexical meaning of the verb kamu ‘to bite’, and it is essentially different from the one that is recognised in type (iv) of the passive o f interest. This is a very important issue. We will examine this matter in more detail in Section 5.1.5.

2.3

The mainstream approach - dichotomy of direct passive and indirect passive

Studying Matsushita’s analysis, later researchers, like Mikami (1953/1972: 98-112), Kuno (1973: 24), Shibatani (1978: 133-142) and Teramura (1982: 214-217), notice that there are some major syntactic and semantic differences between type one of Matsushita’s passive of interest - ‘i. one in which the subject itself is affected by the action / self passive’ - and all of the other three types: ‘ii. one in which possessions/relations of the subject are affected by the action’, ‘iii. one in which the subject is affected by the action of his/her possessions/relations’ and ‘iv. one in which the subject is affected by the action of the third party’. This leads them to attempt to reclassify Japanese passives, again into two basic groups, but with a significant difference from Matsushita’s classification. These scholars take

52

Matsushita’s type (i) together with his ‘plain passive’ in one group and his ‘passive of interest’ types - (ii), (iii) and (iv) - in the other. This approach is currently the one most commonly used. Although the researchers mentioned above all apply different names to the two groups of passive, the basic stance o f their analyses is very similar. We will examine Teramura’s analysis here.

Teramura (1982: 214-217) calls the two groups of Japanese passive ‘direct passive’ and ‘indirect passive'. The direct passive corresponds to Matsushita’s plain passive and type (i) of the passive o f interest (i. one in which the subject itself is affected by the action). The relationship between Matsushita’s and Teramura’s classifications is shown in Table 2 below: Table 2: Relation between Matsushita’s and Teramura’s classifications

Matsushita’s classification A. Plain passive

Teramura’s classification

Direct passive

B. Passive of interest i. self passive

ii. possessor passive iii. possessor self passive iv. third party passive

Indirect passive

53

Teramura defines the direct passive as one that has a corresponding active sentence, and in which the subject is affected by the act directly, as in example (14) below (Teramura 1982: 214; his example):

(14) Naotaka wa sobo

ni sodate-rare-ta.

Naotaka top grandmother by bring-up- pass-past 'Naotaka was brought up by his grandmother.’

Example (14) satisfies both of Teramura’s criteria for a direct passive. Firstly, it has a corresponding active sentence, given as (15) below:

(15) Sobo

ga

Naotaka o

sodate-ta

grandmother nom Naotaka acc bring-up-PAST ‘(His) grandmother brought Naotaka up.’

Example (14) also satisfies Teramura’s criterion that a direct passive has a subject that is affected directly by the act described. In Example (14), Naotaka is the Undergoer3 of the action; he is directly affected by sobo’s (‘his grandmother’s’) action, sodateru ‘to bring up’.

3 As explained in 1.6.2,1 use the term ‘Undergoer’ here as introduced in Foley and Van Valin (1984). In contrast to the notion ‘Actor’, ‘Undergoer’ is characterised ‘as the argument which expresses the [core] participant which does not perform, initiate, or control any situation but rather is affected by it in some way’. (Foley and Van Valin 1984: 29)

54

On the other hand, Teramura claims that an indirect passive is one that does not have an active counterpart, and the subject is affected by the event described by the verb only indirectly, as in example (16) below (Teramura 1982: 214; his example):

(16) Naotaka wa Naotaka top

go-sai

no

five years old gen

toki hubo

ni

sin-are-ta.

time parents by die-PASS-PAST

'Naotaka was adversely affected by his parents dying when he was five.'

Teramura claims that example (16) satisfies his first criterion for an indirect passive, in that it does not have an active counterpart. However, it must be acknowledged that example (17) below could be thought of as an active counterpart o f example (16).

(17) Naotaka ga

go-sai

no

toki, hubo

ga

sin-da.

Naotaka nom five years old gen when parents nom die-PAST ‘Naotaka’s parents died when he was five.’

However, note that Naotaka does not appear in the main clause, hubo ga sin-da ‘his parents died’ in example (17). It thus can be said that example (16) satisfies Teramura’s first criterion in the sense that it does not have an active counterpart that has Naotaka in the main clause. However, one could also suggest (18) as an active counterpart for example (16).

55

(18) Naotaka ga

go-sai

no

toki,

Naotaka no

hubo

ga

sin-da.

Naotaka nom five years old gen when Naotaka gen parents nom die-PAST ‘Naotaka’s parents died when he was five.’

In (18), Naotaka appears in the main clause, in this case, as the possessor NP. In order to eliminate (18) as an active counterpart for (16), we need to clarify Teramura’s first criterion for indirect passive as follows: an indirect passive is one that does not have an active counterpart in which the NP denoting the subject of the passive appears as a core argument. Now we can say that example (16) satisfies the first criterion for an indirect passive.

Example (16) also satisfies Teramura’s second criterion for indirect passive, in that the subject, Naotaka, is not an Undergoer o f the event, the death of his hubo ‘parents’. Naotaka is still affected by the event, but not directly. I agree with Teramura’s analysis to this extent.

Teramura (1982: 215), also, claims that the indirect passive generally implies that the subject is adversely affected by the event, and therefore, he also calls it the ‘adversative passive’. Many previous researchers, such as Mikami (1953 / 1972: 98-112), Kuno (1973: 24) and Shibatani (1978), have held the same view. Kuno (1973: 24), for example, distinguishes the syntactic subgroups, ‘direct passive’ and ‘indirect passive’, and the semantic subgroups, ‘neutral passive’ and ‘adversative passive’. He then states that the correlation between the syntactic distinction and the semantic distinction - more

56

specifically the correspondence between the indirect passive and the adversative passive on the one hand, and that between the direct passive and the neutral passive on the other occurs due to their different derivation processes. Although Kuno later retracts this claim (Kuno 1982), it has been a very influential view in the field of the study of Japanese passives. For this reason I will investigate the derivation processes that have been proposed for Japanese passives in the next section.

2.3.1

Generative transformational grammarian's approach to the dichotomy of direct and indirect passives

In the framework of generative transformational grammar, there are two analyses proposed to account for the derivation process of the direct passive and the indirect passive. Some researchers, like McCawley (1972), Kuno (1973) and Shibatani (1978), claim that the direct passive and the indirect passive have different types of underlying structure and that the two types of passive need to be distinguished on this basis. Other researchers, like Kuroda (1965) and Howard and Niyekawa-Howard (1976), argue that all Japanese passives can be derived in the same way, and should be syntactically treated uniformly. The former is called the ‘non-uniform theory’ and the latter is called the ‘uniform theory’.

Both the uniform theory and the non-uniform theory agree on the derivational process of the indirect passive. According to both analyses, an indirect passive, such as

57

example (19), is derived from an underlying structure that involves two sentences, as in ( 20).

(19) Watasi I

wa Butyoo top

ni kinoo

no

hanasi o

division chief by yesterday gen story

acc

kik-are-ta. hear-PASS-PAST

‘I was adversely affected by the division chiefs hearing yesterday’s story.’

(20) [Watasi wa [Butyoo ga kinoo no hanasi o kik] are ta]

The suffix -(r)are functions as a main clause predicate, and the extra noun phrase (always the subject / topic of the passive), in this case watasi ‘I’, is considered to be the subject / topic of the main clause. This main clause takes an embedded sentential complement, Butyoo ga kinoo no hanasi o kik(u) ‘the division chief hears the story’. To convert (20) to (19), it is necessary to apply the Agentive-Ni Attachment rule (Kuno 1982: 193) to the subject of the embedded clause, butyoo ‘the division chief, and the Verb Raising rule (Kuno 1982: 193) to the verb of the embedded clause.

However, the uniform theory and the non-uniform theory take different approaches in accounting for the derivation process of the direct passive. According to the uniform theory, a direct passive, such as example (21) below, can be derived from the underlying source (22) below, in the same way as an indirect passive.

58

(21) Hitosi

ga

Keiko ni

nagur-are-ta.

Hitoshi nom Keiko by

hit-PASS-PAST

‘Hitoshi was hit by Keiko.’

(22) [Hitosi ga [Keiko ga Hitosi o nagur] rare-ta.]

There is only one difference between the structures in (20) and (22). That is, in the structure in (22) the main subject / topic, Hitoshi, is identical to the object of the embedded clause, Hitoshi. To generate sentence (21), this object must be deleted at some point in the derivation process.

On the other hand, proponents of the non-uniform theory apply a permutation transformation, which Kuno (1973: 345) refers to as ‘pure passive formation’, to a simplex underlying structure as in (23).

(23) Keiko ga

Hitosi o

nagut-ta.

Keiko nom Hitoshi acc hit-PAST ‘Keiko hit Hitoshi.’

Thus Kuno (1973), Shibatani (1978) and many other proponents of the non-uniform theory claim that the direct passive and the indirect passive need to be treated separately. One of the main reasons for their claim is that they believe that it is because the indirect passive

59

and the direct passive are derived differently that they therefore function differently. The indirect passives, they say, are semantically adversative passives, whereas the direct passives are neutral passives. Kuno (1973) mentions that a direct passive is derived from a simplex underlying structure by applying the simple passivisation rule. The suffix - ( r)are of a direct passive is a transparent grammatical form, and, therefore, it does not have an adversative meaning. The adversative meaning only accompanies the suffix -(r)are in the complex underlying structure from which indirect passives are derived.

More recently, Hoshi (1999) has re-examined the uniform versus non-uniform hypotheses. Although he supports the uniform hypothesis with regard to the relationship between the ni direct passive and the ni yotte passive, Hoshi agrees with the non-uniform hypothesis, in relation to the relationship between the ni direct passive and the ni indirect passive in particular. Hoshi (1999: 195) claims that we have to recognise that the ni direct passive and the ni indirect passive are different from one another in an important respect. He presents the evidence of the antecedent of zibun ‘se lf, which was originally suggested by N. A. McCawley (1972) and Kuno (1973) separately Consider the following examples from N. A. McCawley (1972) and Kuno (1972: 299, 304), which Hoshi gives:

(24) a.

Johni-ga

Maryj-ni zibun^j-no uti-de

John-NOM Mary-by self-GEN

koros-are-ta.

house-in kill-PASS-PAST

‘Johni was affected by Mary 's killing him in selfj/*j’s house.’

60

b.

Johnj-ga

Maryj-ni zibunj/j-no koto-o

John-NOM Mary-by self-OEN

zimans-are-ta.

matter-ACC boast-PASS-PAST

‘John, was affected by Mary/s bragging about self/j.’

In example (24a), an example of the direct passive, John can be the antecedent of an anaphor, zibun ‘self, but Mary cannot. In the case of example (24b), an example of the indirect passive, however, zibun can take either John or Mary as its antecedent. It is clear that the direct passive and the indirect passive behave differently in this respect.

There is a problem, however, with the non-uniform theory in its assumption of a correspondence between the direct passive and the neutral passive: the assumption that all the direct passives have neutral meaning. This problem arises because there are a considerable proportion of sentences that would clearly qualify as direct passives, according to the definition given above, and yet have an adversative meaning. Look at example (25) below.

(25) a. John ga

Mary

o

mi-ta.

acc

see-PAST

John

ni

mi-rare-ta.

Mary nom John

by

see-PASS-PAST

John nom Mary 'John saw Mary.' b. Mary ga

'Mary was seen by John.'

61

Example (25b) is a direct passive according to the criterion that there is an active counterpart ((25a). However, all native speakers would agree that this sentence is interpreted as having an adversative meaning. This is actually one of the main reasons why Kuno later withdraws from the non-uniform theory and adopts the uniform theory (Kuno (1982: 196)). We will consider this matter in Section 4.3 below in connection with the special emotive reading associated with some Japanese passives.

2.4

More recent approaches

2.4.1

Masuoka’s approach

Masuoka (1987, 20004: 55-68) classifies Japanese passives, firstly, into three semantic groups: A. Demotional passive; B. Attribute-describing passive; and C. Affected passive. He, then, subdivides C, Affected passive, into two groups: direct passive and indirect passive. The main change from previous analyses is that his analysis is based on the concept that there are several types of description. According to Masuoka (2000: 39), there are two basic description types: ‘attribute description’ and ‘event description’. Attribute description, on the one hand, is for expressing that an entity exhibits certain features or

4 Although Masuoka first presented this analysis in Masuoka (1987) and then in Masuoka (1991), I will quote from the most recently revised version, Masuoka (2000).

characteristics. Event description, on the other hand, is for denoting events that occur at a specific time and place. The attribute-describing passive is a type of ‘attribute description’, and the demotional passive and the affected passive are kinds of ‘event description’. As a result, Masuoka’s analysis of Japanese passives is similar to Matsushita’s (1930 / 1977: 157-161), described earlier in Section 2.2.2. Masuoka’s affected passive basically corresponds to Matsushita’s ‘passive of interest’, and his demotional passive corresponds to Matsushita’s ‘plain passive’. The significant point here is that, in his analysis, Masuoka recognises another type, the attribute-describing passive, which can be categorised as a part of Matsushita’s plain passive. Consequently, the plain passive includes not only the type of passive not inherent to the Japanese language, the demotional passive, but also the one inherent to Japanese, the attribute-describing passive.

This study acknowledges that the classification of Japanese passive constructions is a very complex issue, and one of its aims is to clarify this classification. In order to help making things as clear as possible at this stage of the discussion, the following table shows the relation between Matsushita’s distinction and Masuoka’s classification. For the purpose of comparison, Teramura’s classification is also included, to represent the mainstream approach.

63

T able 3: R elatio n betw een M a ts u s h ita ’s, M a s u o k a ’s an d T e r a m u r a ’s classifications

Masuoka’s classification

Matsushita’s classification A. Plain passive

Teramura’s classification

A. Demotional passive (.N i-y o tte passive) B. Attribute describing P Direct passive C.

B. Passive of interest

Affected passive ii. direct passive

(includes some non-sentient passives

i. self passive ii. possessor passive

with a latent affcctec)

Indirect passive

iii. possessor self passive iv. third party passive

i.

indirect passive

(includes possessor passives)

(does not include possessor passives)

Masuoka claims that ‘A. Demotional passive’, the so-called ‘ni-yotte’ passive5, is a type of event-describing passive. Its main purpose is to background the ‘agent’. This type of passive is generally a non-sentient passive67- one that has a non-sentient subject. Many of the demotional passives are expressed as topicless sentences, in which the subject is marked simply by the nominative particle g a . This is because the demotional passive

5 This is the one that does not occur traditionally in Japanese. See Section 3.2.1.1. 6 Passives with an inanimate subject are called hizyoo no ukemi ‘non-sentient passive’ by Japanese linguists. 7 Very often this kind of sentence with particle ga is interpreted as an objective description, as opposed to a sentence where the subject is selected as a topic and therefore expresses the situation from a certain perspective.

64

usually describes an objectively observed fact, as it is, without representing somebody’s subjective view. Examples follow (Masuoka 2000: 64; his examples):

(26) Onazi hizuke de hakubutukan

syomu

Same date

on Museum

gizyutuin

seido ga

technologist

system nom set.up-PASS-PAST

kitei ga

kaiteis-are atarasiku

general.affairs regulation nom revise-PASS newly

mooke-rare-ta.

‘On the same date, the regulations concerning the general affairs of the museum were revised, and a group o f technologists was set up for the first time.’ (27) Masako no

me no

mae.ni itirin

no

akai bara ga

Masako gen eye gen in.front single gen red

sasidas-are-ta.

rose nom thrust-PASS-PAST

1 A red rose was thrust under Masako’s very nose.’

However, Masuoka also states that there still are some demotional passives where the subject is topicalised by the topic particle wa. Following are examples (Masuoka 2000: 65; his examples):

(28) Umi wa nannen.ka mae.ni sukkari Sea

top

umetate-rare, ...

some.years before completely reclaim-PASS

‘The sea was completely filled in several years ago, and ...’

65

(29) Sumoo wa Kennin-zi Sumo

top

no

keidai

no

akiti

de okonaw-are-ta.

Kennin.Temple gen precinct gen open.space at hold-PASS-PAST

‘Sumo matches were held in an open space in the precincts of the Kennin Temple.’

As mentioned above, Masuoka states that, in a demotional passive sentence, the ‘agent’ is backgrounded: in many cases, it does not appear in the sentence at all. If it is necessary to state the agent, it is indicated as a ni-yoffe-NP, not as a m'-NP as in an affected passive. This is illustrated in examples (30) and (31) below (Masuoka 2000: 67; his examples):

(30) Sitai wa sarani keisatui ni-votte syoosai.ni sirabe-rare-ta. Body top again police

by

in.detail

investigate-PASS-PAST

‘The corpse was examined thoroughly by a police doctor.’ (31) Sore ga

Mizuta-hakase ni-votte mitome-rare-ta.

That nom Dr.Mizuta

by

approve-PASS-PAST

‘That was approved by Dr Mizuta.’

It is for this reason that this type of passive is widely called the ‘ni-yotte passive’.

Masuoka (2000: 56) recognises a second type of non-sentient passive: ‘B. Attribute-describing passive’. He defines the attribute-describing passive as one in which an entity, the referent of the subject, is described as possessing certain attributes, as in

66

examples (32) to (35) below (Masuoka 2000: 56; his examples):

(32) Kono bun This

wa zyudoobun

sentence

top

ni hukum-arc-ru.

passive.sentencein include-PASS-PRES

‘This sentence is categorised as a passive sentence.’ (33) (Sono suyaki

wa) ima made haniwa no

issyu

da

to

(that unglazed.pottery top) now until Haniwa gen one.kind cop quot mi-rare-te

i-te, ...

see-PASS-coNJ be-coNj ‘(This unglazed pottery) has been seen as a kind of a Haniwa (clay image), and ...’ (34) Kono maturi wa maitosi This

siti-gatu ni okonaw-are-ru.

festival top every.year July

in hold-PASS-PRES

‘This festival is held in July every year.’ (35) “Karei-naru itizoku” wa ooku no

tyuugoku zinmin ni yom-are-te

“The.Great.Family” top many gen Chinese people by read-PASS-coNJ i-mas-u. be-POL-PRES ‘ “The Great Family” has been read by many Chinese people.’

This type is generally a non-sentient passive, with a non-sentient subject. It does not depict the occurrence or existence of an event at a specific time and place. Sentences o f this type do occur historically in Japanese, as in example (36) below (cited in Okutsu 1992: 10):

67

(36) Mukasi

koso Naniwa inaka

to

Long.ago emph Naniwa little.town quot

if-are-keme

ima miyako

call-PASS-PAST now capital

hiki moved

miyakobini-keri. [Manyoshu: No. 312] looks.like.capital-PAST ‘Long ago, Naniwa was called a little town, but now the capital has been moved and it looks like a real capital.’

In example (36), the subject, Naniwa, is non-sentient, and the rest of the sentence describes an attribute of the subject: it is called a little town. It was partly because of ignorance of this fact that they were not recognised earlier by Matsushita (1930 /1977).

Masuoka defines the third type, ‘C. Affected passive’, as one that describes a situation in which the subject is somehow affected by the event. According to Masuoka’s classification, if a passive is not classified as either a demotional passive or an attribute-describing passive, it should be classified as an affected passive. Almost all the passives with a sentient subject, therefore, belong to this group.

Masuoka calls the first subtype of the affected passive ‘direct passive’, which has a corresponding active sentence. Masuoka’s direct passive is not as broad as Teramura’s direct passive. Masuoka’s direct passive does not include the attribute-describing passive and demotional passive because they generally have a non-sentient subject, as opposed to

68

his other categories, which usually have sentient subjects. Teramura groups all of these together precisely because they all have an active counterpart. (See Table 3 above for a comparison of Masuoka’s and Teramura’s classification.)

All o f these passives involve a transitive verb. (Refer to Section 1.6.1.1 for the definition of transitive verbs.) The following three examples o f Masuoka’s direct passive all have corresponding active sentences with an object marked by the accusative particle o.

(37) a.

Ano hito

ga

watasi o

That person nom I

acc

mi-te

i-ta

no

ka.

see-coNJ be-PAST nml q

‘(Oh no), that person was watching me.' b.

Ano hito

ni mi-rare-te

i-ta

no

ka.

That person by see-PASS-C0NJ be-PAST nml q ‘(Oh no), I was being watched by that person.’ (38) a. Kuukai ga Kukai

Atoo Taisoku o

nom

Ato

tasikani higos-ita.

Taisoku acc definitely protect-past

‘Kukai definitely protected Ato Taisoku.’ b. Atoo Taisoku Ato

Taisoku

wa

tasikani Kuukai

ni

higo-s-are-ta.

TOP

definitely Kukai

by

protect-PASS-PAST

‘Ato Taisoku was definitely protected by Kukai.’ (39)

a. Ie

no

kagi o

mot-te

hait-te

ki-ta

sinseki no

obasan

House gen key acc have-coNJ enter-coNJ come-PAST relative gen aunt

69

ga

watasi o

NOM I

hakkens-ita.

ACC fmd-PAST

‘My aunt, who got in with the house key she had, found me.' b.

Watasi wa ie I

top

no

kagi o

mot-te

hait-te

ki-ta

sinseki no

House gen key acc have-coNJ enter-coNJ come-past relative gen

obasan ni hakken-s-are-ta. aunt

by find-PASS-PAST

‘I was found by my aunt, who got in with the house key she had.’

The corresponding active sentence is not necessarily a cardinal transitive sentence with the object marked by the accusative particle o. Sometimes the object is marked by the dative particle ni. Examples that have an object marked by particle ni in the corresponding active sentence are as follows:

(40) a.

Tanaka Syooni to.yuu mono ga Tanaka Shoni

Kuukai ni irais-i...

called person nom Kukai

DAT ask-coNJ

‘Aperson called Tanaka Shoni asked Kukai. and...’ b. Kuukai wa Tanaka Syooni to.yuu mono ni irais-are... Kukai

top

Tanaka Shoni

called person by

ask-PASS

‘Kukai was asked by a person called Tanaka Shoni, and... (41) a.

Oisya-sama ga watasi ni

Hokurikuno onsen

Doctor

Hokuriku gen hot.spring nom good quot

nom

I

dat

ga yoi

to

70

susume-mas-ita node, ... advise-POL-PAST because ‘The ddctor advised me that a hot spring in Hokuriku was good (for me), so... ’ b. Oisya-sama ni Hokuriku no Doctor

onsen

ga

yoi

to

susume-rare-mas-ita

by Hokuriku gen hot.spring nom good quot advise-PASS-POL-PAST

node, ... because ‘I was advised by the doctor that a hot spring in Hokuriku was good (for me), so ...’

According to Masuoka’s classification, the direct passive also includes the so-called possessor passives, which are given as examples (42) and (43) below (Masuoka 2000: 59; his examples):

(42) Hukoona.koto.ni Unfortunately ie

o

syooidan

ga

kinzyo

ni rakkas-i tatimati

incendiary.bomb nom neighbor in fall-coNj instantly

yak-are-ta.

house ACC bum-PASS-PAST ‘Unfortunately, a firebomb fell in the neighborhood, and, in an instant, I had my house burnt down.' (43) Kono hito

ni te

o

hik-are-te

watasi wa Kyooto no

This person by hand ACC lead-PASS-cowl

top

Kyoto

gen

teradera o temples acc

71

megut-ta. tour-PAST Being led by the hand by this person, I had a tour of the temples of Kyoto.’

Masuoka (2000: 59) defines the possessor passive as one that expresses the situation in which possessions of the subject (including related people) undergo the action denoted by the verb, and as a result, the possessor (the subject) is affected. In example (42), the house underwent the action of burning, and the possessor of the house, I, was affected by the situation. In example (43), te ‘hand’ underwent the action of leading, and as a result, the possessor of the hand, I, was affected. As mentioned above, Masuoka characterizes this possessor passive as one type of direct passive. This is different from the view of many other researchers. I will examine this issue in relation to Nitta’s analysis in the next section.

The fundamental difference between Masuoka’s categorisation o f direct passives and that of Teramura is that Masuoka’s direct passives mostly have sentient subjects. This is because Teramura has no problem with direct passives having non-sentient subjects. In spite of that fact, then, Masuoka does concede that there are some that he would classify as direct passives that have a non-sentient subject. The following are the examples he gives:

(44) Ano e

ga

kodomo ni hikisak-are-ta.

That picture nom child

by

tear-PASS-PAST

‘That picture was torn by a child.’

72

(45) Taisetuna okane ga

doroboo ni nusum-are-ta.

Important money nom thief

by Steal-PASS-PAST

‘Important money was stolen by a thief.’

Masuoka states that this kind of passive is acceptable only when one can assume a latent affectee. A latent affectee is a sentient party that does not appear in the passive sentence, but is affected by the event in some way. For example, in sentence (44), we can assume a latent party, such as the owner o f the picture, who was affected by the child’s tearing the picture. In the case of example (45), it would be the possessor of the money who was affected. This type of passive differs from the possessor passive in that the owner or the possessor does not occur in the sentence and remains latent. In the possessor passive, on the other hand, as in examples (42) and (43), the possessor appears in the sentence as the subject. In example (42), the owner of the house (I), although elided in the sentence, is the subject of the passive clause. In example (43), the possessor of the hand (I), also elided in the sentence, appears as the subject. We will discuss this matter in more detail in Section 3.1.3.

Masuoka’s second subtype of the affected passive is called the ‘indirect passive’. Masuoka’s indirect passive basically corresponds to Teramura’s. Masuoka defines it as one that describes a situation in which the subject is affected by an event in which s/he does not take part, as in examples (46) and (47) below:

73

(46) (Watasi wa) (I

top)

kare ni nige-rare-ru he

to komar-u

node,...

by run.away-PASS-PRES if be.in.trouble-PRES because

‘I’d be in trouble if he (goes and) runs away (on me), so ...’ (47) Sono tame

ukauka.to sin-are-te

simai-mas-ita.

That because carelessly die-PASS-coNJ complete-POL-PAST ‘Because of that I was off my guard and (he) (went and) died on me.’

In example (46), the subject, I, would be affected by the event, his running away, even though ‘I’ is not involved in the event. In the case of example (47), also, the subject, I, was affected by the event of his dying, again an event in which ‘I’ was not directly involved.

As mentioned above, Masuoka defines the affected passive as one that describes a situation in which the subject is affected by an event in some way. He also states that the relationship between the subject and the event can be direct or indirect. In the direct passive, as in example (38) above, the subject (Ato Taisoku) bears a direct relation to the event (‘Kukai’s protecting Ato Taisoku’). Compared to this, in the possessor passive, as in example (42), the relationship between the subject (‘I’ (elided in the sentence)) and the event (‘having my house burnt’) is more indirect. Furthermore, in the indirect passive, as in example (46), the involvement of the subject (watasi ‘I’) in the event (someone else’s running away) is extremely indirect. In the end, it is apparent that Masuoka views the degree of the subject’s affectedness as a continuum, with the possessor passive in the middle of this continuum. This is the main reason why he puts direct passive, possessor

74

passive and indirect passive all together under the same heading of affected passive.

2.4.2

Nitta's approach

Nitta (1991: 31) classifies Japanese passives into three groups: normal passive, possessor passive and third party passive. Basically, Nitta’s normal passive corresponds to Teramura’s direct passive, Nitta's Third Party passive to Teramura’s indirect passive, and Nitta’s possessor passive to Masuoka’s possessor passive. Nitta states that the normal passive should be placed on one end, the third party passive on the other, and the possessor passive should be positioned in the middle.

A significant point in Nitta’s analysis is that he sets up the possessor passive group as an independent group. Teramura (1982: 244-245), on the one hand, categorizes the possessor passive, along with the third party passive, as a type of indirect passive. He suggests that this is because, given that the following construction is considered to be an indirect passive, it is usually the case that ‘Z is X ’s something’ (Teramura 1982: 245):

(48) X ga

Y ni

Z

X NOM Y GEN Z

o

-rare-ru.

ACC

-PASS-PRES

‘X is affected by Y’s doing something to Z.’

75

‘Z’ could be X’s body part, relative, possessions, or occupied space, in other words, something that has some relation to X. Teramura also states that the possessor passive has an adversative reading, and the degree of adversity depends on the nature of Z, decreasing in the order given above: body part —» occupied space.

Masuoka (2000), on the other hand, regards the possessor passive as a type of direct passive, as briefly mentioned in the previous section. He claims that this is because, unlike indirect passives, whether or not the effect on the subject is perceived as desirable basically depends on the verb’s lexical meaning. For instance, compare the following examples (Masuoka 2000: 60; his examples):

(49) Taroowa sensei ni

atama o

Taro top teacher by head

tatak-are-ta.

acc

hit-PASS-PAST

‘Taro was hit on the head by his teacher.’

(50) Taroo wa sensei Taro

top

ni repooto o

teacher by report

acc

takaku hyooka-s-are-ta. highly

value-PASS-PAST

‘Taro had his report highly valued by his teacher.’

Masuoka claims that in example (49), the subject, Taro, is understood to be affected undesirably because the verb, (atama o) tataku ‘to hit (on the head)’, expresses an undesirable action. On the contrary, example (50) is interpreted as describing an event that

76

had a desirable effect on the subject, Taro, as the verb, (repooto o takaku) hyooka-suru ‘to value (his report highly)’, denotes a desirable situation.

According to Masuoka, the third party passive is totally different, in that the adversative interpretation is independent o f the lexical meaning of the verb. This is illustrated in example (51) and (52) (Masuoka 2000: 60; his examples).

(51) Suzuki-san wa sensei Mr.Suzuki

top

teacher

ni Yoshida-san no musuko o by Mr.Yoshida gen son

acc

home-rare-ta. praise-PASS-PAST

‘Mr Suzuki was adversely affected by the teacher’s praising Mr Yoshida’s son.’ (52) Taroo wa Taro

top

sensei

niYoshida-sanno

repooto o

takaku hyooka-s-are-ta.

teacher by Mr.Yoshida gen report acc highly value-PASs-PAST

‘Taro was adversely affected by the teacher’s highly valuing Ms Yoshida’s report.’

Although the verbs in both (51) and (52) describe desirable situations, these sentences can only be interpreted as involving subjects that are adversely affected.

The reasons Teramura and Masuoka give to determine whether the possessor passive should be considered a direct passive or an indirect passive contradict each other. However, both argue on semantic grounds: whether or not a possessor passive supports an adversative interpretation. Considering the problem o f some direct passives having an adversative reading, as mentioned in Section 2.3.1,1 conclude that neither Teramura’s nor

77

Masuoka’s explanation is convincing. I, therefore, adopt Nitta’s view on this matter, recognising the possessor passive8 as an individual group, and placing it in between the direct passive and the indirect passive.

This chapter reexamined the controversial issue of the classification of Japanese passives in previous research. Several problems have been identified. In particular, this study disagrees with the claim o f a direct correlation between the syntactic and semantic distinctions made by the mainstream approaches - more specifically the correspondence between the indirect passive and the adversative passive on the one hand, and that between the direct passive and the neutral passive on the other. Instead, in the next chapter, we propose separate sets of categories for the syntactic and semantic distinctions. The syntactic distinction is developed from the mainstream approach, and the semantic classification is based on Masuoka’s analysis. The study then demonstrates complex and subtle correlations between syntax and semantics in the case of Japanese passive constructions.

8 In this study, the possessor passive forms a part of the group called ‘semi-direct passive’. See the definition in Section 3.1.3.

78

Chapter 3

Classification used in this study - Syntactic and semantic distinctions

After having examined various approaches to classifying Japanese passive constructions in the previous section, in this section, we will present the classification used in this study.

It seems that all the analyses we examined above have some of their own problems. These arise mainly because both semantic and syntactic features form part of the definitions used. The evidence presented shows that it is vital to begin by treating the two types of distinction - the syntactic distinction and the semantic distinction - separately, and then to consider the correlations between each classification. Because o f the problem of some direct passives having adversative reading, in particular, Kuno (1982) is inclined to use a semantic distinction only. Song (1993), also, suggests that the ‘traditional’ distinction between direct and indirect passives does not provide any explanation for the adversative meaning that accompanies some direct passive sentences. However, I propose to redefine the terms ‘direct passive’ and ‘indirect passive’ in a purely syntactic way, and to add one new syntactic group, ‘semi-direct passive'. These notions are, then, distinguished from the semantic concepts, ‘plain passive’ and ‘passive of interest’. Passives with an ‘adversative meaning’, or what is here referred to as the special meaning o f ‘emotive affectedness’, will be discussed in the next chapter. The summary o f the classification used in this study compared with the previous analyses is illustrated in Table 4 below:

79

Table 4: Classification used in this study compared with previous analyses

Masuoka’s

Matsushita’s

semantic

classification

classification

Plain Passive

Demotional P

Plain Passive

i. demotional P

( N i-y o tie passive)

ii. attributive P

Attribute describing P

P of interest i. P with latent affectee ii. Sentient P

Affected Passive

Teramura’s

Classification used in this study

classification

syntactic

Direct Passive

Direct Passive

i. direct Passive

P of interest i. self P

(includes some non-sentient passives with a latent affectee)

Semi-direct P Indirect Passive ii. possessor P iii. possessor self

(includes possessor passives)

Indirect Passive

ii. indirect P (docs not include possessor passives)

iv. third party P

80

3.1

Syntactic distinctions

3.1.1

Accessibility Hierarchy

Before we start looking into each syntactic category of Japanese passive, let us first examine the notion of the ‘Accessibility Hierarchy’, as it is vital to the syntactic definition of passive constructions used in this study.

Languages differ according to which NP positions can occur as a passive subject, just as they do in the case of other syntactic processes such as relativisation. Keenan and Comrie (1977: 66) propose the notion of the Accessibility Hierarchy below, which they first applied to restrictions on relative clause formation. The hierarchy expresses the relative accessibility to relativisation of NP positions in simplex main clauses.

Accessibility Hierarchy

(Keenan and Comrie 1977: 66)

SU > DO > IO > OBL > GEN > OCOMP

The symbol *>’ means ‘is more accessible’; SU stands for ‘subject’, DO for ‘direct object’, IO for ‘indirect object’, OBL for ‘major oblique case NP’ (NPs that express arguments of the main predicate), GEN stands for ‘genitive’ (or ‘possessor’) NP, and OCOMP for ‘object of comparison’ (Keenan and Comrie 1977: 66). The hierarchy above indicates, for example,

L

81

that the subject NP is the most easily relativized of all the NPs in the main clause, and the direct object is more accessible to relativization than an oblique NP.

Keenan and Comrie (1977: 96) indicate the possible extension o f the applicability of this notion of an Accessibility Hierarchy to passive constructions. The Accessibility Hierarchy may indeed be relevant in determining which NP positions can be passivised in a given language. For instance, Keenan and Comrie (1977: 96) claim that ‘if a language has a rule advancing OBL to passive subject, then it must also have a rule advancing DO to passive subject’. That is because OBL is farther down the Accessibility Hierarchy than DO is. Now let us consider the syntactic definition of Japanese passive constructions, applying this notion of the Accessibility Hierarchy.

3.1.2 Direct passive

A direct passive is syntactically defined in this study as one that has a corresponding active clause, and whose subject would correspond, in the active clause, to a direct object (that is an NP placed in the second position on the Accessibility Hierarchy) or an indirect object (that is an NP in the third position on the Hierarchy). (Note that the first position on the Accessibility Hierarchy, the subject of the corresponding active clause, cannot appear as the subject of the passive clause.) Let us consider the following examples:

82

(1) a. Keiko Keiko

ga

Hitosi

o

nagut-ta.

nom

Hitoshi acc hit-PAST

'Keiko hit Hitoshi.' b. Hitosi ga

Keiko

ni

nagura-re-ta.

Hitoshi nom Keiko

by

hit-PASS-PAST

'Hitoshi was hit by Keiko.'

(2) a.

Keiko ga

Hitosi

ni

tegami o

Keiko nom Hitoshi dat letter

acc

dasi-ta. send-PAST

‘Keiko wrote a letter to Hitoshi.’ b. Hitosi ga

Keiko

ni

tegami o

das-are-ta.

Hitoshi nom Keiko

by

letter

send-PASS-PAST

acc

‘Hitoshi was sent a letter by Keiko.’

In example (1), the passive subject corresponds to the noun in the accusative case, the direct object, Hitosi, and in example (2), it corresponds to the noun in the dative case, indirect object, Hitosi.

As mentioned earlier, the direct passive is defined in this study as one whose subject would be either a direct object or an indirect object of the verb in the corresponding active clause, which are the first and the second positions on the Accessibility Hierarchy. This is exactly the same as the syntactic part o f Teramura’s definition of the direct passive,

83

as seen in Section 2.3. However, among the remainder of the NP positions on the hierarchy, in the corresponding active clause only the genitive case, that is a ‘possessor’ NP, has been widely recognised as appearing as a subject o f a passive clause in Japanese. In this study, a passive whose subject corresponds, in the active clause, to any of the NP positions on the hierarchy other than direct object or indirect object, including a genitive NP, is regarded as a ‘semi-direct passive’.

3.1.3 Semi-direct passive

As pointed out above, which NP positions can occur as the subject / topic of passive varies from language to language. In Japanese, a wide range o f NP positions can appear as the topic / subject of passive. In order to investigate this further, let us return to Keenan and Comrie’s Accessibility Hierarchy:

Accessibility Hierarchy

(Keenan and Comrie 1977: 66)

SU > DO > IO > OBL > GEN > OCOMP

In the previous section, we saw that the core argument(s) o f the verb in the corresponding active clause, a direct object or an indirect object, can occur as the topic / subject of the direct passive. The rest of the NP positions on the hierarchy, major oblique case NP (OBL), genitive case (GEN), and object of comparison (OCOMP), can be called peripheral

84

participants. We will define a ‘semi-direct passive’, in this study, as one that has a corresponding active clause, and whose subject would be one of the peripheral participants in the corresponding active clause. The following example is one in which the subject corresponds to an oblique participant of the active clause.

(3) a.

Keiko ga

Hitosi kara pen o

Keiko nom Hitoshi abl

tot-ta.

pen acc take-away-PAST

‘Keiko took his pen away from Hitoshi.’ b.

Hitosi ga

Keiko ni

Hitoshi nom Keiko

by

pen o

tor-are-ta.

pen acc take-away-PASS-PAST

‘Hitoshi was adversely affected by Keiko’s taking his pen away from him.’

In example (3), the passive subject corresponds to the referent marked by an ablative postposition1, kara ‘from’. Example (4) and (5), below, are examples of the so-called ‘possessor passive’, and also classified here as semi-direct passives, since the passive subject corresponds to the genitive or ‘possessor’ NP of the corresponding active sentence.

(4) a.

Keiko ga

Hitosi no

Keiko nom Hitoshi gen

kata

o

tatai-ta.

shoulder acc tap-PAST

‘Keiko tapped Hitoshi on the shoulder.’

1 As a reader of the early draft has suggested, the passive whose subject corresponds to an ablative NP is unusual. Nevertheless, it must be acknowledged that it is not impossible in Japanese.

85

b. Hitosi ga

Keiko

ni

kata

o

Hitoshi nom Keiko

by

shoulder acc tap-PASS-PAST

tatak-are-ta.

‘Hitoshi was tapped on the shoulder by Keiko.’

(5) a. Keiko ga

Hitosi

no otooto

o

okut-ta.

Keiko nom Hitoshi gen younger-brother acc give.a.lift-PAST ‘Keiko gave a lift to Hitoshi’s younger brother.’ b. Hitosi ga

Keiko

Hitoshi nom Keiko

ni

otooto

o

okur-are-ta.

by younger-brother acc ride-pass-past

‘Hitoshi was adversely affected by Keiko’s giving a lift to his younger brother.’

In Japanese even the NP in the lowest position on the Accessibility Hierarchy, OCOMP, can appear as the passive subject, as in example (6) below:

(6) a. Keiko

ga

Hitosi

vori sakini

Keiko nom Hitoshi than early

sono hon o

yonda.

that book acc read-PAST

‘Keiko read that book earlier than Hitoshi did.’ b. Hitosi

ga

Keiko

ni

sakini

sono hon o

Hitoshi nom Keiko

by

early

that book acc read-PASS-PAST

yom-are-ta.

‘Hitoshi was adversely affected by Keiko’s reading that book earlier than he did.’

86

Note that examples o f the (6b) type include the passive o f an intransitive verb, as in example (7b) below.

(7) a.

Keiko ga

Hitosi

yori sakini kaet-ta.

Keiko nom Hitoshi than early

go-home-PAST

‘Keiko went home earlier than Hitoshi.’ b. Hitosi ga

Keiko

ni

sakini kaer-are-ta.

Hitoshi nom Keiko

by

early

go-home-PASS-PAST

‘Hitoshi was adversely affected by Keiko’s going home earlier than him.’

As you can see in examples (1) to (7) above, the Japanese language is very flexible concerning the accessibility to passivisation of NP positions. I will reconsider this matter in Section 4.2 in relation to the ‘emotive affectedness’ accompanying some Japanese passives.

3.1.4 Indirect passive

Like the direct passive and the semi-direct passive, the indirect passive is here defined in purely syntactic tenns. An indirect passive is one whose subject does not correspond to any of the arguments o f the active verb. In other words it is one which has one extra norm phrase as the newly introduced subject, such as Watasi ga in example (8b) and Takasi ga in example (9b), compared to the closest active equivalent, (8a) and (9a) respectively.

87

Passives of this kind can be made both from transitive verbs (example (8)) and intransitive verbs (example (9)).

(8)

a. Butyoo

ga

kinoo

no

hanasi o

kii-ta.

Division chief nom yesterday gen story acc hear-PAST ‘The division chief heard the story about yesterday.’ b. Watasi I

wa

butyoo

ni kinoo no

hanasi o

top

division chief by yesterday gen story

acc

kik-are-ta. hear-PASS-PAST

‘I was adversely affected by division chiefs hearing the story about yesterday.’

(9)

a. Kinoo

ame ga

hut-ta.

Yesterday rain nom fall-PAST ‘It rained yesterday.’ b. Takasi

ga

kinoo

ame ni hura-re-ta.

Takashi nom yesterday rain by

fall-PASS-PAST

‘Takashi was adversely affected by the rain falling yesterday.’

This type of passive, put together with what is here called the semi-direct passive, has been classified as ‘indirect passive’ in previous research. It has been a major focus of research on Japanese passive constructions. This is because the indirect passive can occur with an intransitive verb as well as with a transitive verb, and it is usually accompanied by a special emotive nuance. This issue of the emotive nuance will be discussed further in detail in

88

Chapter 4, and the proportion of the indirect passive in our data in Section 5.2.1.1.

3.2 Semantic distinction

In this study, Japanese passive constructions are semantically classified into two basic groups, ‘plain passive’ and ‘passive of interest’. The plain passive describes an event objectively, whereas the passive of interest illustrates an event in terms of the subject’s interest. The plain passive is then divided into two subgroups, ‘demotional passive’ and ‘attributive passive’. The attributive passive is said to be inherent to the Japanese language, while the demotional passive is not. The passive of interest, the second of the two broad semantic groups, is also divided into two subgroups: ‘sentient passive’ and ‘passive with latent affectee’. The following sections examine the characteristics of each semantic group of Japanese passive and how each one differs from those proposed in previous analyses.

3.2.1 Plain passive

The plain passive is defined in this study as one generally having a non-sentient subject, and describing a situation or event objectively with only the meaning of objective affectedness of the Undergoer, or the passive subject. In this type of passive, the situation is portrayed objectively. However, still the sense of something being done to the Undergoer is detected. This type o f affectedness is referred to as ‘objective affectedness’ in this study.

89

The plain passive is divided into two groups: demotional passive and attributive passive.

3.2.1.1 Demotional passive

The first subgroup of plain passive is called ‘demotional passive’. It corresponds to Masuoka’s demotional passive. Table 4 is given again below to show the relationship between Masuoka’s analysis and the classification used in this study: Table 4: Classification used in study compared with previous analyses

Teramura’s

Classification used in this study

Masuoka’s

Matsushita’s

classification

syntactic

semantic

classification

classification

Direct Passive

Direct Passive

Plain Passive

Demotional P

Plain Passive

i. demotional P

(Ni-yotte passive)

ii. attributive P

Attribute describing P

P of interest i. P with latent affectee ii. Sentient P

Affected Passive i. direct Passive

P of interest i. self P

(includes some non-sentient passives with a latent affectee)

Semi-direct P Indirect Passive (includes possessor passives)

ii. possessor P iii. possessor self Indirect Passive

iv. third party P ii. indirect P (does not include possessor passives)

90

The primary function of the demotional passive is to marginalise, or ‘demote’ the ‘actor’. Because of that, the ‘actor’ is often eliminated from the passive sentence, as in examples (10) and (11) below:

(10) Zassi

ga

kitinto tukur-are-te

Magazine

nom

iw-anai...

[Shiina 234]

i-r-eba kaisya

wa nani.mo

properly make-PASS CONJ be-PREs-coND company top nothing

say-NEG ‘As long as the magazine is properly produced, (our) company would not say anything...’ ( 11) Iina Good

to

omot-tara kono botan

quot

think-coND this

insatu-s-are-mas-u.

o

os-eba kono 40-mai ga

button acc push-coND this

40-clf nom

[Josei 3488: female, 43, University assistant]

print-PASS-POL-PRES ‘If you think it’s ready, and press this button, then these 40 sheets will be printed.’

However, when it is necessary to indicate the agent in this kind of sentence, it is marked by ni-yotte, not by the dative ni as it is in other types of Japanese passive. Examples follow:12*

(1 2 ) Sarawaku hakubutukan wa rnoto Sarawak

Museun

top

Sarawaku no

former Sarawak

gen

Igirisu-jin

no

syuutyoo

English-CLF gen chief

91

dat-ta

SaaZyeemusu Burukku ni-votte tate-rare-ta

cop-past

Sir James

Brook

by

mono dear-u.

build-PASS-PAST thing cop-nonpast

‘The Sarawak Museum was built by Sir James Brook who was an English chief of Sarawak.’

[Sono 1026]

(13) Daga sono ai But

that

mo motomoto kanozyo ni-votte atae-rare-ta

love also originally she

by

monodeat-ta.

give-PASS-PAST thing cop-past

‘However, that love was also something that was originally given by her.’

[Tsutsui

396]

It is for this reason that this type of passive is widely called the ‘ni-yotte passive’. As mentioned above, however, the ni-yotte NP (the ‘actor’) is often eliminated. For this reason, we adopt the name ‘demotional passive’ in this study.

Song (1993: 105) claims that the primary function of the demotional passive (in his words ‘anticausative passive’) is ‘to eliminate the causative Agent’ (the ‘actor), ‘or background it’. This claim might perhaps be too strong, since quite a number of examples can be found in which the ni-yotte phrase is clearly essential to the meaning of the clause, and thus not fully backgrounded. Such examples include sentence (12) above. Here are two more examples:14

(14) Kono syoosetu wa Natume Sooseki ni-yotte kak-are-ta. This novel

TOPNatsume Soseki by

write-PASS-PAST

92

‘This novel was written by Natsume Soseki.’

(15) Kono kikai This

wa syoogakusei

ni-yotte tukur-are-ta.

machineTOP primary.school.student by

make-PASS-PAST

‘This machine was made by a primary school student.’

In Examples (12), (14) and (15), the ni-yotte phrase cannot be dropped. The ni-yotte phrases in (12) and (14) both refer to special and famous persons, and they are essential to the context. As for sentence (15), a primary school student is not special, but the fact that ‘a primary school student made a machine’ is surprising. Therefore the phrase syoogakusei ni-yotte ‘by a primary school student’ is indispensable in this context.

3.2.1.1.1 Is the demotional passive inherent to Japanese?

It is said that the demotional passive is not inherent to the Japanese language. According to Kinsui (1997: 762), this type of passive first appeared under the influence of word-for-word translations o f Chinese classics and Buddhist scripture. This translation technique gradually spread after the ninth century, and it became part of an established literary style, called the 'kanbun kundoku style’. The ‘actor’ is generally not indicated in this kind of passive sentence in the kanbun style writing. The ‘actor’ marker ni-yotte only started to be used as an ‘actor’ marker when the literal translation of Dutch came into practice in the 19lh century. The ‘actor’ marker ni-yotte is derived from a dative particle ni plus the participa! form of a

93

verb yoru ‘be due to / owing to’. In modern Japanese, however, ni-yotte is completely grammaticised, and means ‘by’, ‘by means o f, ‘according to’, ‘depending on’, or ‘be based on’.

As mentioned in Chapter 2, Yamada (1908) does not recognise the non-sentient passive as a Japanese passive construction. He claims that the Japanese passive is restricted to cases in which the subject is a sentient NP that can be thought of as having consciousness. Since the demotional passive only started to be used in 19“ century (Kinsui 1997: 762), it is likely that in Yamada’s day, around the turn of the century, the demotional passive had not yet been fully recognised as an acceptable passive construction. However, just a few decades later, Matsushita (1930) identifies the demotional passive as the ‘plain passive’, and classifies it as a type of Japanese passive construction.

Kinsui (1992: 18) suggests that the attributive passive, another type o f non-sentient passive, is also a ‘new’ type of passive whose usage has increased in modem times. However, on analysing 81,1 century to 19th century data, Kinsui (1997: 776) retracts this claim, saying that ‘there were non-sentient passive sentences as well as sentient passive sentences in Japanese before modem times’.

Okutsu (1992: 7) also claims that it must be recognised that non-sentient passive is indigenous to Japanese. He examines three of the best known ancient Japanese texts, and finds that, amongst the direct passives that occur, 38.8% in Tsurezuregusa (1030) are

94

non-sentient, 27% in Makuranosooshi (1000), and 17.2% in the Manyooshuu (8th century). Let us examine his examples below (Okutsu 1992: 10)

(16) Saion-zi______no

kane. wausiki.deu ni i-raru-besi-tote,

amata.tabi

Saion-temple OEN bell

‘la’.in.scale to cast-PASS-must-CMPL many.times

i-kafe-rare-keredomo,

kanafa-zari-keru wo, won.goku

yori

cast-change-PASS-though agree-NEO-PAST acc far.country from tadune-idas-are-keri. frnd.OUt-PASS-PAST

‘Because the bell of the Saionii temple should be cast to produce ‘la’ on the musical scale, casting was changed many times in vain, and then a bell of the ‘la’ scale was found in a distant province." (17) Sebaki en

ni tokoro seki

on-sauzoku

no

sitaeasane nado

Narrow veranda in space crowded POL-formal.costume gen underrobes etc. hikitiras-are-tari. [Makura no soshi: 108] dangle-PASs-PAST ‘The veranda was too narrow for the men’s formal Court costumes, and their under-robes were dangled all over the floor.’ (18) Mukasi

koso Naniwa inaka

to

Long, ago emph Naniwa little.town quot miyakobini-keri. [Manyoshu: No. 312] looks.like.capital-PAST

if-are-keme

ima miyako

call-PASS-PAST now capital

hiki moved

95

‘Long ago, Naniwa was called a little town, but now the capital has been moved and it looks like a real capital. ’

As Okutsu (1992: 10) points out, there are three passive clauses in example (16) whose subjects are all Saionzi no kane ‘the bell o f the Saionji temple’. They all have a latent affectee2, the monks and/or parishioners of the Saionji temple, who can be thought of as parties who were affected by the event of the bell’s being cast, the bell’s casting being repeated, and the right bell being finally found.

However, Kinsui (1997: 769) suggests that the function of -(r)are in example (16) should not be regarded as passive but as ‘honorific’. He claims that there are a large number of examples of honorific (r)are that may be mistaken for non-sentient passive (r)are in classical Japanese. As seen in Section 1.5, in the modem Japanese language the honorific -(r)are seems totally different from passive -(r)are in that the honorific clause has the same perspective as that o f an active clause. In fact, it does not involve any change in case marking compared to the corresponding active clause without -(r)are. However, in classical Japanese the formal properties were closer than they are now. This is presumably the reason why it is harder to distinguish the passive -(r)are from honorific -(r)are in classical Japanese.

To clarify this matter, however, further historical research on the data of Meiji and 2 As will be discussed in Section 3.2.2.1, a latent affectee is a sentient entity that one can assume is affected by the event in some way, but is not a participant in the passive sentence.

96

Taisho period (1868 - 1925) is essential. Such historical research is beyond the scope of this study.

3.2.1.1.2 Verb types and the demotional passive

Another way in which the demotional passive differs from other passives is with regard to the type of verb that tends to occur with this type of passive. In previous analyses, there have been two interesting contributions to the understanding of this issue. Song (1993 : 104) states that the demotional passive, in his terms the ‘anticausative passive’, occurs mainly with factitive verbs, such as kensetu-suru ‘to build’, suteru ‘to dump’, erabu ‘to choose’, and orosu ‘to lower’. The term ‘factitive verb’ is used here as characterised in Song (1993: 7). Song explains the term ‘factitive verb’, in contrast to the term ‘operative verb3’, as denoting an event in which the direct object NP, or the subject of the passive, undergoes a specific change (of position, possession or condition4). The verbs used in our examples (10) to (13), tukuru ‘to make’, insatu-suru ‘to print’, tatem ‘to build’ and ataeru ‘to give’, are thus all factitive verbs. Examples (10), (11) and (12) describe events in which the subject of the passive (zassi ‘magazine’, kono 40-mai ‘these 40 sheets’ and Sawaraku hakubutukan ‘The Sawarak Museum’) undergoes a change o f condition. In example (13), the subject (ai

3 In Song (1993), ‘operative verb’ is characterised as one that describes an event in which the subject perfonns an operation, and this operation affects the direct object NP. However, the direct object NP does not undergo any specific change as a result. Examples include hit, touch, strike, slap, shoot and so forth. 4 Following Song (1993: 7), I use the term ‘condition’ here as introduced in Ikegami (1975). It encompasses Jackendoffs three distinct semantic parameters: the identificational, existential, and circumstantial.

97

‘love’) undergoes a change of possession.

Kinsui (1997: 763), based on Teramura’s (1982) discussion, takes a slightly more detailed approach. He divides Japanese verbs into five groups and examines the compatibility of each group of verb with ni, ni-yotte and kara5 as ‘actor’ marker in all types of Japanese passive sentence. His observations throw some light on the issue of verb types in demotional passives, and are summarised as in Table 5 below (Kinsui 1997: 763): Table 5: Combination of n i, n i-y o tte , and k a r a NPs with verbs in passive sentences

Verb type

ni

ni-yotte

kara

(ok)

*

ok

*

ok

Type3: denotes direct (but superficial) contact with a ok

*

*

*

ok

*

*

ok

*

Typel: denotes physical/psychological effects (on a ok patient) e.g. korosu ‘kill’, tukamaeru ‘catch’, kowasu ‘break’ Type2: denotes transition o f emotion and perception e.g. aisuru ‘love’, nikumu ‘hate’, miru ‘see’

patient

e.g. sawaru ‘touch’, tataku ‘hit’, naderu ‘rub’

Type4: verbs of creation e.g. tateru ‘build’, tukuru ‘make’, kaku ‘write’ TypeS: denotes abstract and neutral relationships many of them of Sino-Japanese origin

5 The ablative particle kara is also used to mark the ‘actor’ in a Japanese passive with some types of verb.

98

e.g.

syoyuu-suru

‘possess’,

un’ei-suru

‘manage’,

hazimeru ‘begin’

According to Kinsui, ni-yotte can be used with type l 6, type 4 and type 5 verbs, but not with type 2 and type 3 verbs. Let us compare Song’s and Kinsui’s observations.

If we apply Song’s view to Kinsui’s categorisation, type 1, 4 and probably 5 verbs above would be factitive verbs and type 2 and 3 verbs would not be. Although it is not very clear from Kinsui’s definition, at least all the examples he has given for types 1,4 and 5 can be categorised as factitive. Type 3 verbs seem to fit the prototypical definition o f the operative verb. Type 2 verbs do not seem to be classified as operative verbs, however, they are certainly not factitive verbs either. In summary, Kinsui’s type 1, 4 and 5 verbs can be categorised as factitive verbs, and type 2 and 3 verbs cannot be classified as factitive verbs. To this extent, therefore, Song’s view fits with Kinsui’s observations; both Song and Kinsui indicate that ni yotte passives can occur with these types of verbs.

After examining the data collected for this study, we can add another characteristic to Kinsui’s type 5 verbs. As Kinsui describes them, type 5 verbs include many of the NP-suru verbs that have Sino-Japanese origin. Examples include insatu-suru ‘to print’ in example (11) above and, kaisi-suru ‘to begin’, in example (19) below:

6 Although ni-yotte with typel verbs sounds rather stiff and formal, it is acceptable.

99

(19) Dai-go raundo ga Fifth

kaisis-are-ta. [Sawaki 659]

round nom Start-PASS-PAST

‘The fifth round was started. ’

In our data, we also found a number of NP-suru verbs that originate from Western languages. The following are examples o f verbs of Western origin: nooto-suru ‘to note down’ in example (20) and fairu-suru ‘to file’ in example (21) below:

(20) Zibun

no

yuu

koto ga nootos-are-te i-ru

Oneself gen say-PRES thing nom wa kairaku top

no

hitotu to

pleasure gen one

note-PASS-coNj

i-er-u

Quot

no

o

mi-ru

koto

be-PRES nml

dear-oo.

[Fujiwara 274]

say-POTEN-PRES cop-exht

‘To see what one has said being noted down could be said to be one of (life’s) pleasures.’ (21) Anoo kookoku Well

tor-u

toki

no

tame.ni zenbu

advertisement get-PRES when gen for

all

fairus-are-te file-PASS-coNj

mas-u

kara. [Josei 2765: female, 43, Company employee (editing)]

pol-pres

because

‘Well, it’s because they were all filed, ready for when we’d be getting advertisements. ’

acc

see-PRESnm

100

Note that Kinsui (1997: 763) indicates with regard to type5 verbs that they describe ‘abstract and neutral relationships’. Although many verbs of this type are NP-suru verbs of Sino-Japanese or Western origin, as in examples (11) and (19), and examples (20) and (21) respectively above, there are cases of verbs of this semantic type that are not NP-suru verbs, as in examples (22) and (23) below:

(22) Tasya ni-yotte utusidas-are-ru onore, Others by no

de

zibun

wa

reflect-PASS-PRES oneself oneself top

wa nai ka.

nani.mo nai nothing not.exist

[Takano: 173]

NML COP TOP NEO Q ‘(Seeing) yourself reflected in other person; you wonder if perhaps you are nothing at all.’ (23) Utyuu

no

tituzyo wa aru

Universe gen order tamot-are-te

ki-ta

top

tan’itu no

utyuu

isi

ni-yotte

certain single gen universe will by no

da. [Tsutsui: 351]

retain-PASS-coNj come-past nml cop ‘The thing is that the order of the universe has been retained by a (certain) single universal will.’

To sum up the discussion above, the demotional passive is used primarily to marginalise the ‘actor’. The ‘actor’ is, therefore, often eliminated. However, if it is needed, it is marked by ni-yotte. As for the types of verb that appear in the demotional passive, they

101

are mainly factitive verbs. To be more specific, they are ones that denote changes in the Undergoer, whether that change involves creation, change of possession or position, physical or psychological change. Many of them are NP-suru verbs of Sino-Japanese or Western origin.

Concerning stylistic constraints, Kinsui (1997: 764) states that the demotional passive, in his words the ‘ni-yotte passive’, is ‘rather a stiff literal expression’, and, therefore, it cannot be used to describe a trivial event such as that in example (24) below (Kinsui 1997: 764):

(24) ?? Uraniwa

no

ana wa John no

kaiinu

ni-yotte hor-are-ta.

Back.yard gen hole top John gen pet.dog by

dig-PASS-PAST

‘The hole in the backyard was dug by John’s pet dog.’

Kinsui also mentions that this type of passive is not used in the spoken language. However, we found quite a number of demotional passive examples in spoken data. This issue will be discussed in more detail in Section 5.2.1.2.

3.2.1.2 Attributive passive

The second type of plain passive is called ‘attributive passive’ in this study. It corresponds to Masuoka’s attribute describing passive. This is another type of non-sentient passive, like

102

the demotional passive. However, the attributive passive is different from the demotional passive in that its ‘actor’ is marked by ni (not ni yotte), and is not restricted to factitive verbs. This type is used only ‘when the rest of the sentence attributes some property to the participant denoted by the passive subject’ (Song, 1993: 110). Moreover, as Song (1993: 109) states, an attributive passive is semantically static, while a demotional passive is dynamic. It does not depict the occurrence or existence of an event at a specific time and place. Masuoka (1987: 188) points out that the attributive passive is used to foreground the ‘direct object’ NP by moving it into the subject position. In an attribute-describing sentence, it is normal to place the entity whose attribute is accounted for in the position of the syntactic subject. This type of passive is a kind of attribute-describing sentence, used to describe or imply some attribute of the referent of the subject by the rest o f the elements of the sentence. Observe the following example:

(25) Kono manga ga This

kodomo-tati ni yoku

comic nom children

by often

yomareru. read-PASS-PRES

‘This comic is often read by children.’

The fundamental feature of this type of passive is that a property o f the subject is conveyed by the circumstances denoted by the verb. Example (25), for instance, implies that the subject, ‘this comic’, has the property of being ‘popular among children'. The attributive passive thus differs from attribute-describing constructions like adjectives, as in example (26) below, that generally describe an inherent quality of the subject.

103

(26) Kono manga wa omosiroi. This

comic top interesting

‘This comic is interesting.’

In example (25), the property of the subject (this comic) is defined by the circumstance described by the rest of the passive sentence: it is often read by children. This is the reason why this type of passive still has a nuance of affectedness of the subject - objective affectedness - compared to an adjective clause like (26). In (26), no meaning of affectedness is detected.

Furthermore, if we compare example (27) below to example (25), the concept of ‘attribution’ in the definition of attributive passive becomes clearer:

(27) ? * Kono manga ga

Tarooni yoku yom-are-ru.

This comic nom Taro by often read-PASS-PRES ‘This comic is often read by Taro.’

This sentence is unacceptable, or at least unnatural, because it can hardly admit the interpretation of attributing any characteristics to the comic.

The “Nihongo Kihon Doshi Yoho Jiten” gives a list of verbs that are said not to

104

occur in direct passive sentences. Some of these can be categorised as transitive. Examples include kagu ‘smell’, gaman suru ‘endure, be patient’, tanosimu ‘enjoy’, tugoo suru ‘manage’, nugu ‘take o ff, benkyoo suru ‘study’ and yameru ‘quit’. Most of these verbs cannot take sentient objects. This is generally considered to be the main reason why they cannot appear in the direct passive sentences. However, if we examine these verbs again, verbs like benkyoo suru ‘study’ and tanosimu ‘enjoy’ can be used in attributive passive sentences as in Examples (28) and (29):

(28) Nihongo

wa takusan no

Japanese.language top many

gen

Oosutoraria no

gakusei ni benkyoos-are-te

Australian

student by study-PASS-coNj

gen

i-ru. be-PRES ‘Japanese language is studied by many Australian students.’

(29) Kono geemu wa takusan no This game top many

gen

wakamono

ni tanosim-are-te i-ru.

young.people by enjoy-PASSCONJ be-PRES

‘This game is enjoyed by many young people.’

It can thus be claimed that when one considers the applicability of the direct passive to transitive verbs, one should examine the possibility of their occurrence in the context of attributive passives, which readily allow specific non-sentient subjects.

L

105

As stated earlier, this type of passive usually has a non-sentient subject. However, Masuoka (2000) claims that there are some attributive passives with a sentient subject. In our data, we also found a few of this kind, such as examples (30) and (31) below:

(30) Venetsia kyoowakoku Venice zokusuru

Republic

de

in top

wa,

kokusei

o

tantoos-uru

state.politics acc take.charge-PRES class

hitobito wa kizoku to

yob-are-ta

ga, ...

kaikyuuni to

[Shiono454]

belong-PRES people top nobility quot call-PASS-PAST but ‘In the Republic of Venice, people who belong to the class in charge of governing of the country were called nobility, but ...’

(31) Nanisiro toruko-zin wa zitu.no oya Anyway Turkish-CLF top real ut-te

kane

o

o

koros-ita mono sae,

dorei ni

parent acc kill-PAST person even slav

mooke-ru hoo o

erab-u

to

as

iw-are-te

sell-coNj money acc eam-PRES one acc choose-PRES quot say-PASS-coNJ i-ru.

[Shiono410]

be-PRES ‘Anyway, the Turkish people are said to choose to earn money by selling even those who have killed their real parents as slaves.’

Although these subjects are, indeed, sentient, as Masuoka claims, they are certainly not the kind of referent that attracts a high degree of speaker empathy. A generic NP, like these, is

106

far more likely to occur in an attributive passive than specific human NP.

Masuoka (2000) considers the attributive passive peripheral, claiming that the frequency of the actual usage of this kind of passive is low. However, we found 111 examples of the attributive passive out of 848 passive examples. The number is actually much larger than that o f the widely recognised indirect passives (44 / 848) (including semi-direct passives as has been done in previous research), and therefore, the attributive passive should not be treated as peripheral. We will discuss this issue together with all other findings resulting from the analysis o f our data in Chapter 5.

3.2.2 Passive of interest

In Section 3.2.1, we examined the first semantic type of Japanese passive - the plain passive - and its two subtypes: the demotional passive and the attributive passive. The second semantic type of Japanese passive is called, in this study, the passive of interest. It corresponds to Masuoka’s affected passive. The passive of interest is defined here, in contrast to the plain passive, as one which portrays an event in terms of the concern of a referent of the subject with the meaning of subject’s being affected by the event directly and / or emotively. In the majority o f cases, therefore, the referent of the subject is sentient, and most likely human. However, a passive of interest can occur with a non-sentient subject

107

if one can assume a latent affectee7. We will therefore divide the passive of interest into two subcategories: the passive with latent affectee and the sentient passive.

3.2.2.1

Passive with latent affectee

The first subgroup of the passive of interest is referred to, in this study, as the passive with a latent affectee. As mentioned in Section 2.4.1, Masuoka (2000) points out that there are some passives of interest, in his term ‘affected passive’, that have a non-sentient subject. This is the category in which one can assume a latent affectee, as in examples (44) and (45) in Section 2.4.1, cited again as examples (32) and (33) below.

(32) Ano e

ga

kodomo ni hikisak-are-ta.

That picture nom child

by

tear-PASS-PAST

‘That picture was torn by a child.’ (33) Taisetuna okane ga

doroboo ni nusum-are-ta.

Important money nom thief

by Steal-PASS-PAST

‘Important money was stolen by a thief.’

Although these examples have non-sentient subjects, because each one immediately brings to mind an affected entity - a latent affectee - they are all acceptable. In example (32), the 7 A latent affectee is a sentient entity that one can assume is affected by the event denoted by the passive clause in some way, but is not actually a participant in the passive sentence.

108

owner o f the picture can be thought of as the latent affectee, who was affected by the event of a child’s tearing his/her picture. In example (33), the latent affectee would be the possessor o f the money.

However, Masuoka does not clearly specify in what circumstances a latent affectee can be assumed. Amano (2001) examines this issue comprehensively. According to Amano, whether or not the latent affectee can be supposed depends on the meaning of the event and the meaning of the ga-marked NP, the participant marked by the nominative particle ga (the subject), in the passive sentence.

The latent affectee can be assumed if the event described by the passive sentence is easily perceived as having the meaning of affecting a certain party psychologically. For instance, Amano (2001: 4) points out that if the event describes either ‘evaluation’, ‘loss’, or ‘change of the state of an entity’, the assumption that someone is affected by the event, although that ‘someone’ does not appear in the sentence, can easily be made, as in examples (34), (35) and (36) below (Amano 2001: 4; her examples):345

(34) Atarasii kokoromi ga New

trial

kyaku

nom

ni

kibisiku hihans-are-ta.

[Evaluation]

customer by severely criticise-PASS-PAST

‘The new trial was severely criticised by customers.’ (35) Genkin yusoo.sya Cash

ga

keikan o

yosoot-ta

gootoo ni

transport.car nom police acc impersonate-PAST burglar by

109

ubaw-are-mas-ita.

[Loss]

rob-PASS-POL-PAST ‘A cash transport car was robbed by burglars impersonating policemen.’ (36) Tatiiri.kinsi no Keep.out

gen

teepu ga tape

nozaru

ni hikitigir-are-ta.

nom wild.monkey

by tear-PASS-PAST

[Change of state]

‘A ‘keep out’ tape was torn away by a wild monkey.’

Example (34) describes a type of evaluation of the new trial by the customers. We can easily imagine a participant who was psychologically affected by the evaluation, although the participant does not appear in the sentence. In this case, it would be the person in charge of the trial. In example (35), the loss of the cash transport car is described. The latent party who is affected by the event would be the owner of the car. In the case o f example (36), a change of state occurs - that of the ‘keep out’ tape. Here again a latent participant who is affected by the event, the person in charge o f putting the tape around, can easily be assumed.

Amano (2001) claims that the three types of event discussed above - ‘evaluation’, ‘loss’, and ‘change of the state of an entity’ - are not the only conditions in which one can assume a latent affectee in the passive sentence. It might be possible to add some other event types. However, she states that the event type is one important aspect in deciding whether or not a latent affectee can be assumed in a passive sentence.

110

The second criterion Amano suggests for assuming a latent affectee is the meaning of the participant denoted by the go-marked NP (the subject) in the passive sentence. The following examples are all the same in that they do not fit any of the event types mentioned above. However, example (37) is slightly more acceptable than examples (38) and (39) (Amano 2001: 5; her examples):

(37) ? Osiri Bottom

ga

siranai

otoko ni sawar-are-ta.

nom

unknown man

by touch-PASS-PAST

‘(My) bottom was touched by a strange man.’ (38) ?# Mizutamari no Puddle

gen

doromizu

ga

kodomo ni haner-are-ta.

muddy.water nom child

by

splash-PASS-PAST

‘Muddy water in a puddle was splashed by a child.’ (39) ?# Kusai

tabako

ga

syoodan

Smelly cigarette nom business

aite partner

ni by

suw-are-ta. smoke-PASS-PAST

‘A smelly cigarette was smoked by a business partner.’

This is because in example (37), the subject (osiri ‘bottom’), in Amano’s terms the ‘go-marked NP’, must be somebody’s body part. The mention of a body part as subject immediately evokes the thought of a latent affectee. Amano suggests three more types of ga NP with which a latent affectee can easily be assumed: ‘a possession’, ‘an NP which describes an action' and ‘an entity that can bring to mind a certain related party’. The ga NPs in examples (38) and (39) do not match any of the four types o f ga NP mentioned

above, and do not easily evoke a latent affectee. This is the reason why their acceptability is very low.

In the following examples, the ga NPs indicate somebody’s body part, possession, action and an entity that evokes a certain related participant, respectively (Amano 2001: 5; her examples):

(40) Ziman no

kamigata

Proud oen hairstyle

ga

yuuzin

ni

homer-are-ta.

NOM

friend

by

praise-PASS-PAST

‘(My) favourite hair style was praised by my friend.’ (41) Sono tegami wa Tarooni yabur-are-ta.

(Takami 1995: 39a)

That letter top Taro by tear-PASS-PAST ‘That letter was torn by Taro.’ (42) Sigoto ga

tonari

no

zyuunin ni samatager-are-ta.’

Work nom next.doorGEN resident by disturb-PASS-PAST ‘(My) work was disturbed by the next door neighbour.’ (43) Kono mati wa K-taisa This

ni kowas-are-ta.

town top Colonel.K by destroy-PASS-PAST

‘This town was destroyed by Colonel K.’

In example (40), the ga NP (the subject) is a body part (kamigata ‘hair style’) for which a latent affectee, the owner of the hair style, can easily be assumed. Note that example (40) is

112

much more acceptable than example (37). This is because, although both examples (37) and (40) have a g a N P that indicates a body part, only example (40) also fits Amano’s first criterion, describing evaluation.

Example (41) has a ga NP indicating somebody’s possession (that letter), and the possessor o f the letter can be thought of as a latent affectee. In example (42), the ga NP represents an action (work), and the participant who does the action is assumed to be a latent affectee. In the case of example (43), there is an entity (this town) that evokes a certain related party as ga NP, and that related party (the residents of the town) can be recognised as the latent affectee.

In addition to her discussion of the elements in a clause involving a passive with a latent affectee, Amano (2001: 7) also compares this passive type with the attributive passive, and claims that the two constructions have something in common. She maintains that whether or not a latent affectee can be assumed influences the acceptability of not only a passive of interest with a non-sentient subject (a passive with a latent affectee), but also of an attributive passive sentence. She suggests that this is because they both involve non-sentient subjects and agents marked by the particle ni. Let us examine Amano’s examples below (Amano 2001: 8). (Example (44) is originally cited in Masuoka (1982).)4

(44) Kono zassi This

wa

10-dai

no

wakamono

ni yoku

yom-are-te

magazine top teenage gen young.ppeople by widely read-PASs-coNJ

113

i-ru.

(Masuoka 1982: 14a)

be-PRES ‘This magazine is often / widely read by teenagers.’ (45) Kono.aida kai-ta Recently

sakuhin wa ooku no

write-PAST piece

top

hito

ni tooyoo-s-are-te

many gen people by plagiarise-PASS-coNJ

i-ru. be-PRES ‘A piece I wrote recently has been plagiarised by many people.’

Amano states that event types of examples (44) and (45) function to represent ‘evaluation’ and Toss’, respectively. Moreover, the subject (in Amano’s words, ga NP), kono zassi ‘this magazine’, in example (44) is ‘an entity that evokes a certain related participant’, the editor of the magazine. In example (45), the subject, sakuhin ‘a piece’, is considered to be ‘a possession’. In both cases, a latent affectee can easily be assumed. Amano claims that this is the reason why the acceptability of examples (44) and (45) is high.

Because of the fact that attributive passives, such as examples (44) and (45), can be thought of as having a latent affectee, Amano suggests that all the attributive passives and the passives of interest with a non-sentient subject (passives with a latent affectee) should be categorised into a same group as non-sentient passives with an agent marked by particle ni.

114

However, there are many examples of attributive passive that cannot be thought of as having a latent affectee. Let us examine the following examples:

(46) Kono bun This

wa zyudoobun

ni hukum-are-ru.

(Masuoka 2000: #1)

sentence top passive.sentence in include-PASS-PRES

‘This sentence is categorised as a passive sentence.’ (47) Kono ie

wa itabei

ni

kakom-are-te

i-ru.

(Inoue 1976: #84)

This house top wooden.fence with enclose-pass-conj be-PRES ‘This house is enclosed with a wooden fence.’

Examples (46) and (47) differ from Amano’s examples (44) and (45) in that, while they both involve ni NPs, these ni NPs do not indicate agents. In fact, neither example (46) nor (47) can involve an agent. This is because this type of passive describes the circumstances of the subject rather than a specific event in which it is involved.

As discussed above, some attributive passives with an agent marked by the particle ni can be thought of as having a latent affectee, as in examples (44) and (45). However, these examples are completely different from the passive with latent affectee in that they depict a static situation, not the occurrence of an event at a specific time and place; neither are they associated with the meaning of direct or emotive affectedness. Moreover, there are also some attributive passives that do not have an agent marked by ni and in which a latent affectee cannot be assumed, such as examples (46) and (47). In this

115

study, therefore, the attributive passive is not treated as the same as the passive o f interest with a non-sentient subject as a group of non-sentient passive with an ‘actor’ marked by the particle ni. The attributive passive remains as a separate group.

3.2.2.2 Sentient passive

The second subgroup of passive of interest, the sentient passive, has a sentient (most likely human) subject. This type of passive is used to describe a situation in which the subject is directly or emotively affected by the event. Examples follow:489

(48) Sono tame,

tiisai toki

kara kinzyo

no

kodomo nakama ni yoku

That because little when from neighbour gen children group izimer-are-ta

soo-da. [Miyamoto: 275]

bully-PASS-PAST

s e e m -c o p

by often

‘Because of that, evidently she was often bullied by neighbourhood children since the time she was little.’ (49) Taroo ni-totte ryoori Taro

for

wa geizyutu no

cooking t o p art

hitotu.na no da

gen

zairyoo

o

toriager-are-ru

ingredient

acc

take.away-PASS-PRES

one

nm l

to urami mo hukai no if grudge also

kara cop

sono because that

dear-u. [Sono: 1006] deep

nml

‘Cooking is a kind of art for Taro, so if he has his ingredients taken away, he deeply

c o p - pres

116

resents it.’ (50) ... sara.ni

wa bonnetto no

furthermore top bonnet nado mo etc.

gen

ue ni nor-are-te

tatioozyoos-ita koto

top on get.on-PASS-coNJ be.stuck-PAST

case

at-ta. [Fujiwara: 240]

even there. is-PAST

‘... furthermore there even was a time when 1 was stuck because (the kids) got on the bonnet (of my car).’

In example (48), the subject (she) was bullied, and therefore, was directly affected by the event. Example (49) describes how, if the subject (Taro) had his ingredients (for cooking) taken away from him, he would be adversely affected by the event. In the case of example (50), it portrays the event in which the subject (I) was in trouble because the children climbed onto the hood of ‘my’ car, and ‘I’ could not move the car.

Syntactically, examples (48)-(50) are classified into different groups in this study: direct passive, semi-direct passive and indirect passive, respectively. These examples, therefore, vary according to the way the subject is affected by the event in each one. Examples (48)-(50), however, all portray the event in terms of the interest of the referent of the subject. This is the reason why they are all categorised in the same semantic group in this study - the sentient passive.

The main focus o f previous research has been the last two types of the sentient

117

passive - the semi-direct and indirect passive. In terms of numbers found in the data, it is actually the first type - the direct sentient passive - that is by far the most prominent numerically. Furthermore, in terms of the numbers of tokens of each semantic type, sentient passives represent the highest proportion in the data. They are discussed in considerable detail in Sections 5.2.1.

This chapter has presented the new classification of Japanese passives used in this study. Raising an objection to the previous claim of a direct correlation between the syntactic and semantic distinctions, it has proposed separate sets o f categories for the syntactic and semantic domains. The summary of the classification of Japanese passives used in this study is illustrated in the following table:

Table 6: Classification used in this study

attributive j demotional passive passive Plain passive

w ith la te n t

I n d ir e c t p a s s iv e

Semantic groups

Scmi-dir ect passive

Direct passive

Svntactic groups

sentient passive

afT c ctc e

Passive of interest

In this chapter, it has been made clear that correlation between syntax and semantics in Japanese passives is much more subtle and complex than has generally been recognised.

118

Chapter 4

The Japanese passives with special emotive affectedness

The special meaning of emotive affectedness associated with some Japanese passives has drawn attention from many researchers. These passives with special emotive undertone are often referred to as ‘adversative’ passives. This is presumably because in most cases, the emotive nuance is adversative. However, there certainly are other cases in which the emotive undertone cannot be considered ‘adversative’. We therefore use the expression the ‘special meaning of emotive affectedness’. The primary characteristic of this type of passive in Japanese is that the speaker describes an event in terms of the concerns o f a referent denoted by the subject NP. The party that is responsible for the occurrence of the event appears, if at all, as a passive ‘actor’. Therefore, the passive subject must be an entity that can be concerned about something and the passive ‘actor’, one that can be held responsible for it. This may be the reason why it is traditionally said that both the subject of the passive and the passive ‘actor’ are most often sentient in this passive type.

It has been traditionally claimed that the indirect passive in Japanese has an adversative meaning and that the direct passive does not. (See Mikami 1953 /1972: 98-112, Kuno 1973: 24 and Teramura 1982: 214-217). As discussed in Section 2.1.2.1, however, the actual situation is not quite so straightforward. Some direct passives have an adversative

119

connotation, while some indirect passives' do not have an adversative reading. In this chapter, therefore, we first examine passives with an adversative meaning within the framework of each syntactic category of Japanese passive: the indirect passive, the semi-direct passive and the direct passive. We then discuss in detail where the adversative reading actually comes from.

4.1

An emotive undertone of indirect passives

The indirect passive, according to the definition used in this study, generally has an adversative reading, and invariably has some sort of emotive nuance. As seen in section 3.1.4, the indirect passive is here defined in purely syntactic terms. An indirect passive is one whose subject does not correspond to any of the arguments of the active verb. Passives of this kind can involve both from intransitive and transitive verb stems, as in examples (1) and (2) respectively:

(1)

a. Kinoo

ame ga

Yesterday rain nom

hut-ta. fall-PAST

‘It rained yesterday.’ b. Takasi

ga

kinoo

ame ni hura-re-ta.

Takashi nom yesterday rain by fall-PASS-PAST 1 According to the definition used in this study, these are categorised as semi-direct passive.

120

‘Takashi was adversely affected by the rain falling yesterday.’

(2)

a. Butyoo

ga

kinoo

no

hanasi o

Division chief nom yesterday gen story

acc

kii-ta. hear-PAST

‘The division chief heard the story about yesterday.’ b. Watasi wa I

top

Butyoo

ni kinoo no

hanasi o

division chief by yesterday gen story

acc

kika-re-ta. hear-pass-past

‘I was adversely affected by division chiefs hearing the story about yesterday.’

The indirect passive, therefore, can be divided into two syntactic subgroups: indirect passive of an intransitive verb and indirect passive of a transitive verb. It is, however, not the case that all intransitive verbs and transitive verbs can occur in indirect passives. In the following sections, we will examine when intransitive verbs or transitive verbs can appear in indirect passives.

4.1.1 The indirect passive of an intransitive verb

As seen in Section 1.6.1.2, intransitive verbs are divided into two semantic subgroups: the unergative verb and the unaccusative verb. These subgroups are highly relevant to the applicability of passivisation in Japanese. The definition of two subtypes of intransitive verb is repeated here:

121

Unergative verbs: are those whose subject (Sa) is an Actor which performs the action described by the verb intentionally, and could initiate or control the event in the same way as the subject of an archetypal transitive clause (A). e.g. aruku ‘to walk’, nigeru ‘to run away’, yasumu ‘to rest’, etc.

Unaccusative verbs: are those whose subject (So) is an Undergoer, which is semantically like the object of a transitive clause (O) in that it is involved in the event described by the verb only unintentionally, and is not thought of as controlling the event. e.g. kimaru ‘to be decided’, sakeru ‘to tear’, tokeru ‘to melt’, etc.

Note that there are some intransitive verbs, such as agaru (go up / rise) and uturu (move / permeate) that can have both unergative and unaccusative interpretations, depending on the nature of the subject. Note also that the subject of the intransitive verb would be the ‘actor’ in a passive construction. Mikami (1972 ), Kageyama (1995: 59) and Tsuboi (1997: 290) all claim that unaccusative verbs do not occur in indirect passives. Tsuboi (1997: 288) claims that sentences like (3), (4) and (5) are not possible, even with an adversative connotation such as that suggested in the translations below:

(3) *Taroo Taro

wa kyonen top

hidoi

zisin

ni okir-are-ta.

last year devastating earthquake by occur-PASS-PAST

122

‘*Adevastating earthquake occurred last year, adversely affecting Taro.’ (4) T o kyo no

hitobito wa

kinoo

Tokyo

people top

yesterday lightningby fall2-PASS-PAST

gen

kaminari

ni oti-rare-ta.

‘♦Lightning struck in Tokyo yesterday, adversely affecting the people.’ (5) *Taroo wa Taro

top

pan

ni urikire-rare-ta.

bread by be sold ouL pass-past

‘♦Bread was sold out, adversely affecting Taro.’

According to Tsuboi (1997: 290), in order for an adversative passive to occur, the event described by the verb has to involve an Actor, since the Japanese adversative passive expresses not only the sense of being subjected to an annoying/upsetting event, but also having an accusing feeling toward the person who is responsible for the occurrence o f that event. Tsuboi claims that (3), (4) and (5) are not acceptable as their verbs are unaccusatives, and they describe events that “just happen”. The m'-marked NPs in those sentences, therefore, do not have control over the event, and cannot be held responsible for the event.

We will argue against this claim, showing evidence of such use in Japanese novels. We also examine the context in which such verbs are used and the implications of their use in such a context.

Having examined all the verbs listed in the “Dictionary of the Usage o f Basic 2 N ote that, even though the English translation in this exam ple involves an unergative verb ‘struck’, the Japanese expression k a m i n a r i g a o ti r u (lit: ‘lightening fe ll’) is unaccusative.

123

Japanese Verbs”, it became apparent that most of the verbs that the dictionary lists as not being able to occur in any passive constructions are, indeed, unaccusative verbs. Examples include: amaru ‘to remain’, aru ‘to exist’, ukabu ‘to float’, tasukaru ‘to survive’, tariru ‘to suffice’ and waku ‘to come to the boil’. To this extent, then, I basically agree with the analyses of Mikami, Kageyama and Tsuboi. However, there are some unaccusative verbs that do seem to be able to occur in indirect passives. The following are some examples3:

(6)

kawaru ‘to change’ Yakuin

no

menbaa ni

tairvoo.ni kawar-ar-te

Executive gen member by a.lot.of

yowat-te

i-ru.

change-PASS-coNJ be.in.a.fix-coNJ be-PRES

T am in a fix because a lot of members of the committee have been changed.’ (7)

seityoosuru ‘to grow up’: Titioya wa suemusume_______ni seityoos-are-te

sukosi sabisi soo

Father top youngest.daughter by grow.up-PASS-coNj a.little lonely seem dat-ta. COP-PAST 'The father seemed a bit lonely, having had his youngest daughter grow up.' (8)

taoreru ‘to fall down’: Butvoo_______ ni karoo

de

taore-rare-te

simat-ta.

Division.Chief by overwork abl fall.down-PASS-coNJ complete-PAST 3 As a reader of this study has noted, these unaccusative verbs can be interpreted as having an unergative sense, if they are forced into a certain context, as in Doozo sonnani hayaku seityoo sinaide kudasai ‘Please don’t grow up so soon’ and Sonnani sigoto site byooki de taoreru nayo ‘Don’t don’t collapse from working so hard!’.

124

'We are in trouble because our manager has ended up falling ill (on us) through overwork.’ (9)

sagaru ‘drop / go down’:

Kabu

no ne

ni sagara-re-te

ooyowari da.

Share o f price by drop-PASS-coNJ very.weak cop 'I am in a very weak position with the value of mv shares having (gone and) fallen.' (10) tum oru ‘be piled up / lie on’ Yuki

ni konna ni tumora-re-te

wa soto ni

dera-re-nai.

Snow

by like.this pile.up-PASS-coNj top outside dat go.out-POT-NEG

'I can’t go out with the snow piled up like this.' (11) huru ‘fall’: Kono aida

wa totuzen

ame ni hura-re-te

The.other.day top suddenly rain

taihen dat-ta.

by fall-PASS-CONJ terriblecop-PAST

'The other day was terrible with the rain (going and) falling so suddenly.'

If it is basically true that unaccusative verbs tend not to occur in the passive in Japanese, then we need to find some explanation as to why these verbs seem to occur quite freely, as in the examples above.

Most of these verbs have a sentient ‘actor’, as in Examples (6), (7) and (8). We

125

might hypothesize that even though the activity described by an unaccusative verb in an example such as those above is unintentional, if the ‘actor’ is sentient, it might be considered to have some kind of responsibility for the activity, and that is why these verbs are able to occur in passive sentences. In fact, in examples like (6), (7) and (8), there is a strong sense that the passive subjects actually blame the ‘actors’ in some way for the event described. In example (7), for instance, although the daughter could hardly help the fact that she grew up, there is still a sense that her father feels that by doing so, she has somehow abandoned him.

In the case o f (6) kawaru ‘to change’ and (8) taoreru ‘to fall down’, for example, if the ‘actors’ of the passive were non-sentient NPs, such as hon no nedan ‘price of the book’ and niwa no ume no ki ‘the plum tree in the garden’, as in (12) and (13) below, the sentences would not be acceptable:

(12) * Hon no

nedan ni kawar-are-te

book gen price

okane

ga

tari-nakat-ta.

by change-PASS-coNj money nom be.enough-NEG-PAST

‘Since the price of the book had been changed, I did not have enough money for it.' (13) * Niwa no

ume no

ki

ni taore-rare-te

gakkari

da.

Garden gen plum gen tree by fall.down-PASS-coNJ be.disappointed cop ‘I am disappointed as the plum tree in mv garden has fallen down.’

At this stage, then, we can say that even if the verb is considered to be unaccusative, if the

126

‘actor’ is sentient, it is more likely to be able to appear in the passive.

Amongst propositions with unaccusative verbs that can be used in the passive, as well as those with sentient ‘actors’, there are also quite a few examples that have ‘weather phenomena’ as the ‘actor’ (as in examples (1), (10) and (11) above). In fact, this is not the only circumstance in which speakers o f Japanese treat weather phenomena like sentient noun phrases. For example, words describing weather phenomena are quite easily personified, as in the example below:

(14) H ru

no

so y o k a z e

ga

sp rin g gen g e n tle .b re e z e nom

w a ta si n o

I

hoo

o

n a d e-ta .

gen c h e e k acc stroke-PAST

‘T h e sp rin g b re e z e s tro k e d m y c h e e k .’

Weather words can also be the addressee in imperative sentences:

(15) Ame ame hur-e rain

hur-e!

rain fall-lMP fall-lMP

‘Rain, rain, fall on me!’ (16) K a z e

yo

huke!

wind part blow-imp ‘Blow, wind, blow!’

127

It is, therefore, evident that expressions describing weather phenomena can behave like sentient NPs, and also be thought of in an anthropomorphic way, as having some measure of control or blame for the event described. Once again, there is a strong sense of blaming the agent in these sentences. It is presumably this feature of personification that allows the events to be described in the passive in these cases.

As seen above, however, Tsuboi (1997: 288) claims that sentences like (3) and (4), cited again below, are unacceptable even though they involve weather phenomena as the ‘actor’:

(3) *Taroo Taro

wa kyonen

hidoi

zisin_______ ni okir-are-ta.

TOP last year devastating earthquake

by

occur-PASS-PAST

‘A devastating earthquake occurred last year, adversely affecting Taro.’ (4) *Tokyo Takyo

no

hitobito wa kinoo

kaminari

GEN people TOP yesterday lightning

ni

oti-rare-ta.

by

fall-PASS-PAST

‘Lightning struck in Tokyo yesterday, adversely affecting the people.’

Tsuboi (1997: 295) suggests, however, that the difference between examples like (10) and (11) above and examples like (3) and (4) is the nature of the events. An event like raining and snowing, as in (10) and (11), usually continues for some considerable duration of time, whereas one like lightning or an earthquake, as in (3) and (4), happens suddenly and passes within a very short period o f time. According to Tsuboi, raining and snowing can be

128

thought of as self-sustained activities, and therefore they are more easily personified, compared to dynamic and spontaneous events like lightning and earthquakes. This is the reason, Tsuiboi claims, why events like raining and snowing can occur in an adversative passive.

However, consider t : following example4:

(17) Konna

tokoro ni

this.kind.of place

in

made

kaminari

ni

oti-rare-ta

as.far.as lightning

by

fall-PASS-PAST

n

zya

doosiyoo.mo.nai yo

ne.

NML

INSTR.TOP

helpless

SFP

sfp

‘If we can be struck h lightning even in a place

i this, I don’t suppose there’s

anything we can do abc t it, is there?’

Most native speakers would agree that this sentence is acceptable. We need to seek for an alternative explanation. The difference between examples (3) and (4) and example (17) is that, in examples (3) and (4), the passive is used in a main clause, whereas in example (17), it appears in a subordinate clause. It seems to be for this reason that example (17) is acceptable. This issue will be discussed further, later in this section.

As seen above, the sentience or ‘blame-worthiness’ of the agent with an

4 I am indebted to one of the readers o f the early draft for considering this example.

129

unaccusative verb does not seem to be the only factor relevant to whether the sentence can be put into the passive. In Example (9), kabu no ne ‘the price of shares’ cannot be thought of as having any potential control over the event described, and yet the indirect passive sentence is acceptable. Our next focus is, therefore, on the indirect passive of unaccusative verbs with a non-sentient (and non-personified) ‘actor’ like this one.

Ding (1997) examines indirect passives, especially the passive of intransitive verbs, with inanimate passive 'actors’ (in his words "moto no bun no ga kaku" [the participant marked by ga in the underlying sentence]) (Ding, 1997). He considers each indirect passive sentence in a broader context. Ding claims that an intransitive verb with an inanimate, and even immobile, ‘actor’ can occur in an indirect passive sentence. This is possible in a context in which there is a negative value judgement by the subject of the passive sentence against the realisation of the event described by the verb. That is, before the event occurs, the subject of the passive forms a negative judgement about the event, or desires that the event should not happen. If the event then occurs contrary to the subject's judgement or desire, that event can be described in an indirect passive sentence. To this extent, I agree with Ding's analysis. It is certainly very important to consider factors such as the judgement or attitude of the subject participant when indirect passive sentences are analysed. Let us see some of Ding’s examples.18

(18) “Zairu wa kireru mono de wa nai” .... “Sorya, oogoto da yo. Sakura Seikoo to site wa, doo atte mo zairu ni kirerarete wa komaru! Hyakuman’en mo kane o kakete, zisya

130

no seihin ga warui to hi zikken o suru baka wa inai kara na.”

[Source: Hyoheki]

'"Climbing ropes just don’t break".... "That’s a serious matter. As Sakura Steel Mill [the manufacturer of the rope], we'll be in trouble if the rope (goes and) breaks (on us)! Nobody would be so foolish as to spend as much as a million yen to conduct an experiment which proves that his own company's products are bad."' (19) “Hidoku hutte kita ne. Kono yoosu zya keiben no miti ga koware ya sinai ka ne” kare wa sikata nasi ni Tuda no mimi e mo hairu yoo na ookina koe o dasite koo itta. “Nani daizyoobu yo. Nanbo namae ga keiben datte, soo keiben ni kowarerareta hi nya noru mono ga sainan daa ne”

[Source: Meian]

‘"It's started raining heavily. I wonder if the light rail track might be damaged,” he said in a loud voice so that Tsuda could hear, too. “It should be O.K. Even though it’s called a light rail, the day the track (goes and) gets damaged so ‘lightly’ will be a disaster for the passengers, won't it?”’

In Example (18), it is clear from the context that “the climbing rope must not break”. The subject of the passive thinks that it cannot happen, it should not happen, and yet “if the climbing rope were to break, we would be in terrible trouble.” A similar situation applies in Example (19). Ding’s claim about this negative judgement of the subject regarding the event described is clearly correct. I would add, however, that there is, in these examples too, a sense of blame. However unreasonably or illogically, the subject is ready to blame the passive agent if the situation were to arise.

131

Ding also states that this kind o f indirect passive sentence is more acceptable and more natural if the event described by the verb is expressed as one that has not yet been realised, as you can see from comparing the following examples.

(20) Sakura Seikoo

to sitewa zairu ni

Sakura Steel.Mill as

top rope

kire-rare-te

wa

koma-ru

by break-PASS-coNJ top be.in.trouble-NON-PAST

‘As the Sakura Steel Mill (the manufacturer of the rope) we'll be in trouble if the rope (goes and) breaks (on us).’ (21) ??Sakura seikoo wa Sakura Steel.Mill top

zairu ni kire-rare-te

komat-te

i-ru

rope by break-PASS CONj be.in.trouble-coNJ be-NON-PAST

‘The Sakura Steel Mill is in trouble because the rope (they made) has (gone and) broken (on them).' (22) Keiben

no miti ni koware-rare-ta hi

nya

noru mono ga

sainan

Light.rail of track by break-PASS-PAST day loc-top ride people nom disaster daa ne COP SFP ‘The day the light rail truck (goes and) gets damaged will be a disaster for the passengers, won't it?’ (23) ??Keiben no Light.rail of nat-ta become-PAST

miti ni koware-rare-te noru mono ga

taihen-na koto ni

track by break-PASS-coNJ ride people nom terrible

thing dat

132

‘It was a disaster for the passengers because the light rail truck (went and) got damaged.’

In addition I have also observed that, as in examples (20) and (22), and also in example (17) above, most of the indirect passives o f intransitive verbs with an inanimate ‘actor’ occur in subordinate clauses, especially in counterfactual clauses. As I mentioned above, in an indirect passive sentence, the passive ‘actor’ is portrayed as the party that is responsible for the event. It seems that in the unreal world of the future or counterfactual clause, it is easier to blame something that is not animate.

To further test if the use o f the indirect passive is really acceptable with an inanimate agent and an unaccusative verb, I made up a series of passive sentences that satisfied all of the conditions mentioned, and conducted a small survey of Japanese native speakers to determine their acceptability. All of the examples clearly involve events that had not (yet) occurred, and negative value judgements by the subjects. I also tested only examples in which the passive verb appears in subordinate, counterfactual clauses.

Most of the sentences tested were considered acceptable by the majority of informants (7 out of 11). However, there was one example that was felt to be unnatural by most of the informants (Example (24)), and another (Example (25)) that all of the informants regard as very unnatural, if not unacceptable.

133

(24) ?Daizi-na okyaku-san ga Special

guest

koge-rare-te

aru node,

nom

kondo

koso

kono keeki ni

have because this time top-emphat this

cake

by

wa komaru.

bum-PASS-C0Nj top be.in.trouble-NON-PAST ‘Because we're having a special guest, I'll be in (real) trouble this time if this cake gets burnt.’

(25) *Hon no nedan ni kawa-rare-te Book of price

i-te

wa komaru

node,

by change-PASS-coNJ be-coNJ top be.in.trouble-NON-PAST because

arakazime denwa de

kakunin sita.

beforehand phone instr check

do-PAST

‘If the price of the book had been changed, I'd have been in trouble, so I checked the price on the phone beforehand.’

Our hypothesis concerning the reason for the unacceptability or awkwardness of these sentences is that, in each case, the real responsibility for the unfortunate event - if it were in fact to occur - would be attributable not to the passive subject, nor to the ‘actor’, but to a third party - one that is clearly involved in the event, even though not mentioned in the sentence. In the case of Example (24), it would presumably be some factor such as an unreliable oven, rather than the cake itself, which might be held responsible for the cake getting burnt, resulting in the low acceptability of this sentence for many native speakers. In the case of example (25), the party ultimately responsible for a price rise is quite clearly not

134

the book but the owner of the bookstore. In this case, where a third party is in total control of the outcome, the acceptability of sentences with a non-sentient ‘actor’ is even lower.

In this section, we have seen that the indirect passive of an unaccusative verb with a non-sentient ‘actor’ can, in fact, occur, despite claims to the contrary, provided that it fulfdls certain conditions. These are that it appears in a context in which there is a negative value judgement by the subject, in which the subject somehow blames the ‘actor’ for the situation (however unreasonably), that it occurs in an unrealised or counterfactual clause, and that there is no other party involved in the event that might be perceived as having control over its outcome.

4.1.2 The indirect passive of a transitive verb

In this section, we will examine the indirect passive of a transitive verb which has an o-NP that cannot in any way be thought of as a possession of the passive subject (as in the case of a semi-direct passive). This type of indirect passive is illustrated in example (26) below.26

(26) Ozima no

tczvoo

o

mi-rare-te

wa taihen

da.

Ojima gen handcuffs acc see-PASS-coj top trouble cop ‘(We) will be in trouble if Ojima’s handcuffs are seen.’

[Akagawa 587]

135

The o-NP is the direct object of the active verb in the corresponding active sentence. In the passive sentence (26), the subject, watasi-tati ‘we’, which is elided in the sentence, does not correspond to any argument o f the active verb. Example (26), therefore, is an indirect passive and it has an adversative meaning. The o-NP remains as an o-NP, and it cannot be thought of as a possession o f the subject in any way. ‘Ojima’s handcuffs’ are not the possession of watasi-tati ‘we’. I will consider only this type of passive here, and exclude the one that has an o-NP that is the subject’s possession (e.g. belongings, relatives, feelings, etc.), such as example (27) below.

(27) Naitoo wa 71-nen Naito

top

Yanagi ni yabure-te

taitoru o

1971-CLF Yanagi by be.defeated-coNJ title

acc

ubaisar-are-ta. take.away-PASS-PAST

‘Naito was defeated by Yanagi in 1971 and was adversely affected by Yanagi’s taking his title away.’

[Sawaki 600]

The o-NP, here is taitoru ‘title’ and it belongs to Naito. This type is considered to be a semi-direct passive. We will examine it in the next section.

In this section we will discuss in what context passives like example (26) adversative passives with an o-NP that is not a possession of the subject - occur. In the data for this study, most of them occur under one of the following circumstances:

(i)

the sentence describes a situation in which the subject is adversely affected by an

136

unexpected or sudden event or change (ii)

the sentence describes a situation in which the subject is adversely affected by something that is decided by some kind of authority

In the case of example (26), when it is observed in context, as shown in (28) below, it is clear that it belongs to the circumstance (i).

(28)

Tobira ga

ak-u

to

wakai onna

ga

de-te

ki-ta.

Door nom open-PREs when young woman nom come.out-coNJ come-PAST Ak-u

nari

norikom-oo to

s-ita

Ozima to

open-PRES as.soon.as get.into-EXHTQUOT do-PAST “Ara, gomennasai”, to Oh sorry mi-rare-te

onna

quot

wa taihen

ga

it-ta.

ayauku hatiawase.

Ojima with

Ozima no

tezvoo

nearlybump.into o

woman nom say-PASTOjima gen handcuff acc da. Nobuko ga

awatete

Ozima o

waki e

see-PASS-coNJ top in.troublecop Nobuko nom in.a.flurry Ojima acc side to hippat-te, zibun ga

sono mae ni tat-ta.

pull-coNJ herself nom that

[Akagawa 587]

in.front stand-PAST

‘When the door opened, a young woman came out. She nearly bumped into Ozima who tried to get into (the lift) as soon as the door opened. The woman said, “Oh, I’m sorry.” (We) will be in trouble if Oiima’s handcuffs are seen. Nobuko, in a flurry, pulled Ozima to the side and stood in front of him.

137

‘Nobuko’ was “in a flurry” because she didn’t expect the young woman to come out. Both Nobuko and Ojima would be in trouble if the woman saw Ojima’s handcuffs, the sentence appears in a context in which the subject is affected adversely by an unexpected, sudden incident. The following, example (29), is another example of this kind.

(29) At-ta.

Himo o hik-u

to, katiri.to

oto wa s-ita

There.is-PAsr string a cc pull-PRES if click tuk-anai.

Suruto,

yahari

anzenki

come.on-NEG then as.I.thought safety.device Doo sur-eba

ii

How d o-coN D good

da-roo?

sound o acc

to p

ga, akari wa

do-PAST but light

kir-are-te

i-ru

top

rasii.

cut-PASS-coNj be-PRES seem

[Akagawa 607]

c o p -e x h t

‘I’ve found it! When (Nobuko) pulled the string, there was a click, but the light didn’t come on. It seems that the safety device is cut off. What shall I do?’

When Nobuko pulled the string on the floor lamp, she expected the light to come on. It didn’t, and she realized that the safety device might be cut off. The passive is used again here, in the context of this unexpected event.

Examples (30) and (31) below are the examples of circumstance (ii), a situation in which the subject is adversely affected by something that is decided by some kind of authority:

138

(30) Singuru no Single

g en

heya ga ai-te

i-nai

toki.ni wa, sikata.ga.nai node

room

be-NEG

when

n o m b e .v a c an t-co N J

tuin

no

heya ni ire-rare-ru. ... seiki

tw in

gen

ro o m in

put-PASS-PRES

no

no.choice

top

tuin no rvookin o

re g u la r g e n tw in g e n c h a rg e

tor-are-te_____ sima-u_______ kara

turai.

because

acc

[Shiina 251 ]

take-PASS-coNj com plete-PRES b e c a u s e b e .to u g h

‘When there is no single room (available), they have no choice but put me in a twin room....... it is tough because you would be charged regular twin rate.’

In Japan, it is usually the case that if you stay at a hotel, you have to pay the rate for the room. Even if you stay in a twin room alone, you have to pay for the twin room. It is this rule decided by some nebulous hotel authority, that affects the subject adversely.31

(31) Kizoku

ni umare-ta

Aristcrat

top

kokkai

ni wa syussekis-uru

as be.bom-PAST because Parliament to

Sore wa kakutaru that

kara

to p

attend-PRES

riyuu mo naku-tekessekis-uru to,

particular reason even

bun

no

kyuuryoo gurai no

worth

gen

salary

n e g -c o n j

bakkin o

about g e n fine

acc

tor-are-ru

but

kare no

miss-PRES

ga,

2-nen

if

kara

he

gen

two-CLF

de,...

take-PASS-PRES because

cop

‘(Nicolo) attends Parliament because he was bom as an aristocrat. However, he does not have any particular reason for that, but it is only because if he misses (Parliament), he would be fined about two years worth of his salary.’

[Shiono 454]

139

This sentence, again, concerns some sort of rule imposed by an authority, in this case, one decided by Parliament, That is if a member o f Parliament fails to attend, he has to pay about two years worth of his salary. The passive is used to express the fact that the subject is adversely affected by such a rule.

To sum up the discussions above, an intransitive verb can appear in the indirect passive if it is an unergative verb. In the case of the unaccusative verb, it can occur in the indirect passive if it fulfills certain conditions: it appears in a context in which there is a negative value judgement by the subject; in which the subject somehow blames the ‘actor’ for the situation (however unreasonably); that it occurs in an unrealised or counterfactual clause, and that there is no other party involved in the event that might be perceived as having control over its outcome. In terms of transitive verbs, most of them occur in the indirect passive under one of the following circumstances: (i) the sentence describes a situation in which the subject is adversely affected by an unexpected or sudden event or change; (ii) the sentence describes a situation in which the subject is adversely affected by something that is decided by some kind o f authority.

4.2

Semi-direct passives with special emotive nuance

As proposed in Section 3.1.3, the semi-direct passive is defined in this study as one that has a corresponding active clause, and whose subject would be one of the peripheral

140

participants of the verb in the corresponding active clause. Examples o f the so-called ‘possessor passive’, whose passive subject corresponds to the genitive case or ‘possessor’ NP of the corresponding active sentence, belong to this category of the semi-direct passive. Since the possessor passive, including the body-part passive, forms a large part of the semidirect passive, we will discuss the emotive nuance in the possessor passive in this section.

Teramura (1982: 244-245) suggests that, in this type of passive, it is usually the case that ‘Z ’ is ‘X ’s something’, as illustrated in (32) below (Teramura 1982: 245):

(32) X ga

Y ni

Z o

-(r)are-ru.

X NOM Y GEN Z ACC -PASS-PRES ‘X is affected by Y’s doing something to Z.’

‘Z ’ could be X ’s body part, relative, possessions, or occupied space, in other words, something that has some relation to X. Teramura also states that the possessor passive has an adversative reading, and the degree of adversity depends on the nature of Z, decreasing in the order given above: body part —> occupied space.

As seen in Section 2.4.2, Masuoka (2000), however, claims that in the possessor passive, unlike indirect passives, whether the effect on the subject is perceived as favorable or adversative basically depends on the verb’s lexical meaning. For instance, compare the following examples, cited w again below (Masuoka 2000: 60; his examples):

141

(33) Taroo Taro

wa

sensei

ni atama o

top

teacher by head

acc

tatak-are-ta. hit-PASS-PAST

‘Taro was hit on the head by his teacher.’ (34) Taroo wa sensei

ni repooto o

Taro top teacher by report

acc

takaku hyookas-are-ta. highly value-PASS-PAST

‘Taro had his report highly valued by his teacher.’

Masuoka claims that in example (33), the subject, Taro, is understood to be adversely affected because the verb, (atama o) tataku ‘to hit (on the head)’, expresses an unfavorable action. On the contrary, example (34) is read as describing an event that had a favorable effect on the subject, Taro, as the verb, (repooto o takaku) hyooka-suru ‘to value (his report highly)’, denotes a desirable situation.

Klaiman (1987:402-403) also points out that some possessor passives do not have any meaning of emotive affectedness. Consider the following examples (Klaiman 1987: 403):35

(35) Sono sitai that

wa atama o

corpse top head

acc

siroi nuno de/ni

oow-are-te

i-ta.

white cloth with/DAT cover-PASS-coNJ be-PAST

‘The head of that corpse was covered with a white cloth.’

142

(36) Sono hako wa hanbun o that

box

top

half

acc

siroi nuno de/ni

oow-are-te

i-ta.

white cloth with/DAT cover-pa ss -c o n j be-PAST

‘That box was half covered with a white cloth.’

Klaiman (1987: 402-403) indicates that examples (35) and (36) do not carry the nuance of the subjects’ emotive affectedness. She claims that it is because (35) and (36) describe some state or attribute of the subjects, not some specific event. Example (35) indicates an attribute of ‘that corpse’ being covered with a white cloth, and example (36) is, similarly, about ‘that box’ being half covered with a white cloth.

However, although Klaiman’s claims do seem applicable to the examples she gives above, consider the following example:

(37) Sono sita i th a t

wa

k a ib o o

ga

owar-u

to

sugu

z e n s in

o

siroi

c o rp s e t o p a u to p s y n o m fmish-NONPAST w h e n a t.o n c e w h o le .b o d y a c c w h ite

nuno d e

oow-are-ta.

c lo th DAT cover-PASS-PAST

‘The whole body of that corpse was immediately covered with a white cloth when the autopsy was finished.’

Example (37) denotes an event that occurs at a specific time and place, and yet it still does not carry the special nuance of the subject’s affectedness. Contrary to Klaiman's view,

143

example (37) does not imply the subject’s emotive affectedness even though it indicates a specific event.

This issue of the special emotive nuance in the possessor passive will be discussed further in Section 4.4.3 in relation to one of Shibatani’s (1996: 7) criteria for his concept of ‘relevance’.

4.3

Direct passives with an emotive undertone

In previous analyses, such as Mikami (1953/1972: 98-112), Kuno (1973: 24), Shibatani (1978: 133-142) and Teramura (1982: 214-217), the Japanese passive has been considered to have an adversative meaning when it is an indirect passive (see Section 2.3). According to the non-uniform theory in the framework of generative transformational grammar (McCawley (1972), Kuno (1973) and Shibatani (1978)), the indirect passive has a meaning not shared by the direct passive because it is derived from an underlying structure that involves two sentences rather than one (see Section 2.3.1). However, we have seen evidence that some direct passives also have an adversative reading (Example (24)-b in Section 2.2.1). The following are further examples:

Keiko ga

sono ie

Keiko

that

nom

no

tikaku de Hitosi

house GEN near

at

Hitoshi

o

mi-ta.

acc

see-PAST

144

‘Keiko saw Hitoshi near that house.’ b. H ito s i g a

so n o ie th a t

H ito sh i n o m

no

tik a k u de K e ik o ni

h o u se g e n n e a r

at

K e ik o

mi-rare-ta.

by see-PASS-PAST

‘Hitoshi was adversely affected by Keiko's seeing him near that house.’ (Lit. Hitoshi was seen by Keiko near the house.)

(39) a. Keiko ga

Hitosi

o

eki

no

mae de ni-jikan

K e ik o n o m H ito s h i a c c s tatio n g e n fro n t in

mat-ta.

tw o h o u rs wait-PAST

‘K e ik o w a ite d fo r H ito s h i in fro n t o f th e sta tio n fo r tw o h o u rs .’

b. H ito s i

ga

K e ik o n i

H ito s h i n o m K e ik o ‘H ito sh i

eki

by sta tio n

d e n i-jik a n

mat-are-ta.

no

m ae

gen

fro n t in tw o h o u rs wait-PASS-PAST

was adversely affected by K e ik o 's waiting fo r h im in fro n t o f th e station

fo r tw o h o u rs .’

(Lit. Hitoshi was waited for by Keiko in front of the station for two hours.)

Examples (38b) and (39b) are direct passive sentences; their subjects would be one of the core arguments of the active verb in the corresponding active clause. In each case, however, they are interpreted as having adversative reading5.

It may be thought that the adversative reading in example (38b) arises from context. However, as seen in Section 2.3.1, example (25b) (simplified version of example (38b)), Mary ga John ni mi-rare-ta ‘Mary was seen by John’, is also interpreted as having an adversative meaning. This shows that the adversative reading is a part of linguistic meaning o f the sentence, not pragmatic implicature.

145

Song (1993: 93) suggests that the ‘traditional’ distinction between direct and indirect passives does not provide any explanation for the adversative sense in cases like these. However, if you define the ‘direct passive’ and the ‘indirect passive’ as syntactic subgroups, as has been done in the previous sections of this book, and distinguish them from the semantic subgroups, then there is no confusion. Examples (38b) and (39b) are syntactically classified as direct passive sentences and the fact that they have an adversative reading does not cause any problem. Song’s argument here, like the one advanced in this study, is against the traditional idea that indirect passives are adversative passives and direct passives are neutral passives. If we treat the two types of distinction, the syntactic and the semantic, separately, and do not insist on the complete correspondence of these two types of distinction, then cases like examples (38b) and (39b) - direct passives with an adversative meaning - should not be a problem.

Now, however, we have lost the advantage o f the traditional analysis, that of being able to give a simple account regarding which cases of the passive have the adversative meaning and which do not. We need to seek for an alternative explanation. This explanation will be advanced in Section 4.4 below.

4.4

The emotive nuance

First, let us compare example (38b) cited again below, with an example of a passive

146

without any special emotive connotation, example (40b).

[Passive with an emotive connotation] (38)

a. Keiko ga

sono ie

Keiko n o m that

house

no

tikaku de Hitosi

gen

near

at Hitoshi

o acc

mi-ta. see-PAST

‘Keiko saw Hitoshi near that house.’ b.

Hitosi ga

sono ie

Hitoshi

that house

nom

no tikaku d e Keiko gen

near

at Keiko

ni m i- r a r e - ta . by see-PASS-PAST

‘Hitoshi was adversely affected by Keiko's seeing him near that house.’ (Lit. Hitoshi was seen by Keiko near the house.)

[Passive without an emotive nuance]

(40) a. Keiko ga Keiko

nom

Hitosi Hitoshi

o acc

obut-ta. carry-PAST

‘Keiko carried Hitoshi on her back.’ b. Hitosi Hitoshi

ga

Keiko ni obuw-are-ta.

nom

Keiko by carry-PASS-PAST

‘Hitoshi was carried by Keiko on her back.’

Example (38b) is a direct passive that has an adversative meaning. Example (40b) is a direct passive and unlike (38b), it has no adversative connotations. It seems that the main

147

semantic difference between the adversative passive, (38b), and the ordinary passive, (40b) is the way in which the subject of the passive is involved in the event described by the verb. In example (40b), the subject, Hitoshi, is directly involved in the event and is directly affected by Keiko’s carrying him. In example (38b), on the other hand, Hitoshi is clearly not involved in the event, nor affected by the event, as directly as in example (40b). Let us consider each of the parameter of involvement- in more detail below.

4.4.1 Involvement

The notion of involvement appears in Wierzbicka’s (1979/1988: 270) and Kuno’s (1982: 205) analyses in accounting for an adversative meaning in a passive sentence. Wierzbicka states that:

Thus, it appears that Japanese treats actions which affect us without involving us in one way and those which both affect and involve us, in another..... ‘If an action (or event) does not involve us directly and yet affects us, it will probably affect us negatively’, is the view embedded in Japanese syntax. (Wierzbicka, 1988: 270)

While Wierzbicka focuses on the contexts in which an adversative meaning occurs with a passive, Kuno tries to explain why the adversative meaning has to be read in a passive sentence. Kuno proposes the adversative passive meaning rule as follows,

148

considering the subject’s involvement in the event described by the verb:

[Adversative passive meaning rule] In the deep structure of the w'-passive, the more directly the subject of the main clause is involved in the event described in the embedded clause, the more easily the passive can be interpreted as a neutral passive. The less the involvement is, the stronger the adversity interpretation arises in the passive. In other words, if one has an NP which is not expected to be able to occur in a passive sentence as the subject o f the passive, one needs to somehow supply the interpretation that the NP is directly involved in the action or the psychological state in the embedded clause. This interpretation has to come from somewhere other than the element of the meaning of the verb. This is why the adversative reading obtains. (Kuno 1982: 205, translation by the author.)

Kuno actually withdraws from the non-uniform theory and adopts the uniform theory (See Section 2.3.1.) over this issue o f the meaning of the adversative passive, especially the adversative meaning of some direct passive sentences. He opts to explain the distinction of the passive without an emotive nuance and the passive with an emotive nuance semantically rather than representing it structurally.

According to Kuno’s hypothesis, example (40b), cited again below, does not have a special emotive connotation, because the subject, Hitosi, is directly affected by the event

149

described in the embedded clause, Keiko ga Hitosi o obuu ‘Keiko carries Hitoshi’.

(40) b. Hitosi Hitoshi

ga

Keiko ni obuwa-re-ta.

NOM

Keiko b y carry-PASS-PAST

‘Hitoshi was carried b y Keiko on her b a c k .’

In example (38b), on the other hand, Kuno claims that the subject, Hitosi, is not involved in the event directly. Therefore, it is interpreted as an adversative passive.

(38) b. Hitosi ga Hitoshi

nom

sono ie that

h ouse

no

tikaku de Keiko ni mi-rare-ta.

g en

near

at Keiko by see-PASS-PAST

‘Hitoshi was adversely affected by Keiko’s seeing him near that house.’ (Lit. Hitoshi was seen by Keiko near the house.)

We have examined Kuno’s analysis of the circumstances under which the adversative meaning occurs in a passive sentence. However, it is still not clear exactly why Kuno suggests that we need to ‘supply the interpretation that the NP is directly involved in the action or the psychological state in the embedded clause’ if we have ‘an NP which is not expected to be able to occur in a passive sentence as the subject of the passive’ (Kuno, 1982: 205). We will consider this issue in the next section.

150

4.4.2 Semantic integration

According to Shibatani (1996: 3), the adversative meaning in question is not exclusive to the Japanese passive. Although it is mainly restricted to body part passives, both Chinese and Korean have passive constructions which are associated with an adversative meaning. The so-called Ethical Dative construction in Indo-European languages also has an adversative meaning. Shibatani suggests that we need to seek for some principle to account for this issue of adversative meaning within a larger framework, which encompasses all of the constructions mentioned above.

The constructions mentioned above, adversative passives in Japanese, Chinese and Korean and the Ethical Dative in Indo-European languages, have a common syntactic feature. That is that they all have an ‘extra-thematic’ argument. An ‘extra-thematic’ argument is the one ‘that is not part of the case frame o f the verb with which it occurs, or that does not bear a theta role specified by the verbal head’ (Shibatani 1994: 465). Kuno’s notion of ‘an NP which is not expected to be able to occur in a passive sentence as the subject of the passive’ can be thought of as corresponding to this ‘extra-thematic’ argument.

Shibatani (1996) proposes the principle of semantic integration as follows:

151

[Principle of semantic integration] In a given construction, all the related NPs which bear referring function have to be semantically integrated properly by making some kind o f semantic contribution. (Shibatani 1996: 8, translation by the author.)

In other words, Shibatani claims that even if an NP does not bear a theta role assigned by the verb, if the NP is interpreted as making some sort of semantic contribution, it is semantically integrated into the proposition denoted by the rest of the sentence, and is thus grammatically acceptable.

In many examples of the Japanese passive with the special emotive nuance, the subject does not bear any theta role assigned by the verb. That is, in the terms used in this study, it is not an argument of the verb. Therefore, some ‘extraneous semantic support’ (Shibatani 1994: 467), in this case the emotive overtone, is needed in order to properly integrate the subject into the clausal semantics. Now another question arises. That is, ‘why is it that the “extraneous semantic support” is, in most o f the cases, adversative?’. We will consider this question in Section 4.4.3 below.

152

4.4.3 Relevance

As seen in section 4.4.1, Kuno and Wierzbicka tried to explain the semantic difference between the passive with an emotive nuance, (33b), and the passive without an emotive connotation, (35b), using the notion of involvement. However, there are some examples that cannot be explained with Kuno’s and Wierzbicka’s notion o f involvement. These include some possessor passives, especially the body part passives, such as examples (41) and (42) below.

(41) Hitosi

ga

Keiko ni te

o

nigi-rare-ta.

Hitoshi nom Keiko by hand acc

hold-PASS-PAST

‘Hitoshi was adversely affected Keiko’s holding his hand.’

(42) Hitosi

ga

Keiko ni tume

Hitoshi nom Keiko by nail

o acc

ki-rare-ta. cut-PASS-PAST

‘Hitoshi was adversely affected by Keiko’s cutting his nails.’

In these examples the subject is interpreted as the possessor of the body part. For example, in (41), the hand that Keiko held is assumed to be Hitoshi’s hand, and in example (42), the nails are taken to be Hitoshi’s. If we take this natural interpretation, the subject - “Hitosi”, in both examples above - is very much involved and affected by the event described by the verb. However, these passive sentences are also interpreted as having an adversative

153

meaning, contrary to Kuno and Wierzbicka’s claim that if the subject is directly involved in the event, the passive clause does not have a special emotive connotation. We need, therefore, to reconsider the notion of involvement.

First, let us compare examples (41) and (42) with other examples o f body part passives that are not likely to be interpreted as having an adversative meaning, such as example (4b) seen in Section 3.1.3, cited again below.

(4)

b. Hitosi

ga

Keiko ni kata

o

tatak-are-ta.

Hitoshi nom Keiko by shoulder acc tap-PASS-PAST ‘Hitoshi was tapped on the shoulder by Keiko.’

The difference between examples (41) and (42) and example (4b) is that, in examples (41) and (42), the body parts that appear in the sentence are peripheral compared to the one in example (4b). That is, the hand and the fingernail in (41) and (42) respectively, are located more peripherally on the body than the shoulder in (4b). Shibatani (1996: 6) claims that whether the body part is central or peripheral affects whether or not the passive sentence has an adversative meaning. In other words, how the possessor of the body part, the subject of the passive, relates to the event changes depending on the body part that appears in the passive sentence.

This issue of the body part passive leads Shibatani (1996: 7) to propose the notion

154

of ‘relevance’, refining Kuno’s and Wierzbicka’s notion of ‘involvement’. Shibatani states that the referent indicated by each NP has to be relevant to the event described by the verb in some way. He defines the notion of ‘relevance’ by delineating the properties of optimally relevant participants as follows:

The participants constituting the event itself, say an agent and a patient of a scene construed as involving a prototypical transitive activity, are optimally relevant to the scene of the event. (Shibatani 1994: 470)

He suggests two parameters that decide each NP’s degree of relevance to the event. These are the participants’ physical presence (proximity) and the affecting and affected roles they play in constituting a transitive event. Shibatani claims that ‘the higher the relevance of the referent of an extra-thematic argument is to the described scene, the easier it is to integrate it. Conversely, the more difficult to integrate an extra-thematic argument, the more required are semantic augmentation and morphological trappings supporting and overtly indicating its relevance’.

If we apply Shibatani’s analysis to the body part passive in Japanese, it can be said that if the body part is high in proximity to the possessor, the possessor of the body part, the subject of the passive, is central to the event and very much affected by the event. That is, if the subject is highly relevant to the event, it does not require any supplementary semantic input to increase its relevance to the event. The passive sentence, therefore, is not

155

likely to have an adversative meaning.

According to Shibatani’s analysis, example (42), cited below again, has an adversative meaning. That is because the fingernails’ proximity is fairly low compared to other body parts, which means the relevance to the possessor of the body part, the subject of the passive, is low.

(42) Hitosi

ga

Keiko ni tume

o

ki-rare-ta.

Hitoshi nom Keiko by nail acc cut-PASS-PAST 'Hitoshi was adversely affected by Keiko’s cutting his nails.’

In example (4b), cited below again, on the other hand, kata ‘the shoulder’ is much more central than the fingernail, and the proximity to the possessor is higher. Thus, this sentence does not have an adversative meaning.

(4)

b. Hitosi Hitoshi

ga

Keiko ni kata

nom

Keiko

o

tatak-are-ta.

by

shoulder

acc

tap-PASS-PAST

‘Hitoshi was tapped on the shoulder by Keiko.’

We can expand this Shibatani’s first criteria of proximity to be applied not only to the ‘body part’ passive, but also to other ‘possessor’ passives, combining with Teramura’s (1982: 244-245) analysis.

156

As seen in Section 4.2, Teramura (1982: 244-245) suggests that in the ‘possessor’ passive, it is usually the case that ‘Z ’ is ‘X ’s something’, as illustrated in (32) and cited below again (Teramura 1982: 245):

(32) X ga

Y ni

Z o

~(r)are-ru.

X NOM Y OEN Z ACC -PASS-PRES ‘X is affected by Y’s doing something to Z.’

‘Z’ could be X’s body part, relative, possessions, or occupied space, in other words, something that has some relation with X. Teramura also states that the possessor passive has an adversative reading, or an emotive nuance, and the degree of adversity depends on the nature of Z, decreasing in the order given above: body part —> occupied space. The crucial point here is also how close the relation between X and Z is; namely the degree of proximity of the possessor (X) to the possession (Z). Compare the following examples:

(43) Ikuraka

no

kane

o

okur-u to ...

Some.amount gen money acc give yomiager-are-ru.

if

ringu anaunsaa ni namae o ring

announcer by name ACC

[Sawaki 614]

Call-PASS-PRES. ‘If you gave some money, your name would be called out by the ring announcer (44) ... sono zairvoo that

o

toriager-are-ru

to urami mo hukai no

ingredient acc take.away-PASS-PRES if grudge too deep

NML

157

dear-u. COP-NONPAST ’... if the ingredients are taken away, his grudge too will be deep.’

In examples above, the possessions (Z) are the underlined NPs, and the possessors (X) are the subjects of the sentences, elided in both cases. It is clear that the relation of ‘name’ in (43) with its ‘possessor’, ‘you’, is closer than the one of ‘ingredients’ in (44) with its ‘possessor’, ‘he’. It is also observed that if the proximity to the possessor of the ‘possession’ is high, the ‘possessor’ or the subject of the passive is more central to the event. This is presumably one o f the reasons why the special emotive nuance in (44) is stronger than that in (43). It is now evident that the lower the proximity of the ‘possession’, and thus the lower the degree of the centrality o f the subject to the event, the stronger the special emotive undertone arises.

Shibatani’s second parameter, the affectedness, also seems to be related to the centrality of the subject to the event, and highly important in explaining the case of the direct passive that has the special emotive nuance, as in examples (38b) and (39b), cited again below:38

(38) b. Hitosi

ga

sono ie

no

tikaku de Seiko ni mi-rare-ta.

Hitoshi nom that house gen near

at Seiko by see-PASS-PAST

'Hitoshi was adversely affected by Seiko's seeing him near that house.'

158

(39) b. Hitosi Hitoshi

ga

Keiko ni eki

no

mae de ni-jikan

raat-are-ta.

nom

Keiko by station gen front in two.hours

wait-PAss-PAST

'Hitoshi was waited for by Keiko in front of the station for two hours.'

In (38b) the degree of the centrality of the subject (Hitoshi) to the event is low, and therefore the affectedness of the subject in the event described by the verb (to see) is also very low. The passive sentence is thus interpreted as having quite a strong emotive nuance. In the case of (39b), also, the subject (Hitoshi) does not play a central role in the event (Keiko’s waiting for him for two hours). He might not be even aware o f the fact that Keiko was waiting for him, and the event does not directly affect Hitoshi. It therefore also has a special emotive undertone.

Let’s have a look deeper into this issue, examining some real examples.

(45) (Zyunkowa) hait-te (Junko

top)

kur-u

Ozima-tati

enter-coNJ come-PRES Ojiza-PLR

o

kagame-ru.

acc

crouch-PRES

ni

mi-rare-nai

yoo.ni mi

DAT see-PASS-NEG so.that body

[Akagawa 496]

‘Junko bent herself forward so that she was not seen by Ojima and others who came in.’ (46) Ozima no

tezyoo

Ojima gen handcuff

o

mi-rare-te

wa

taihen

acc

see-PASS CONj top trouble

da. [Akagawa 587] cop

159

‘(We) will be in trouble if Ojima’s handcuffs are seen.

Example (45) is a direct passive with an emotive nuance just like example (38b). The subject of the passive, Zyunko ‘Junko’, does not play a main part in the passive clause nor is she physically affected by the event, even though she would be a core argument of the verb miru ‘to see’ in the corresponding active sentence. She is, however, clearly emotionally affected by the event. If the event, Ozima-tati ni mi-rare-ru ‘being seen by Ojima and others’, ever happens, Junko will be in trouble. Example (46), on the other hand, is an indirect passive; its subject, watasi-tati ‘we’ (elided in the sentence), does not correspond to any of the arguments of the active verb, mi-ru ‘to see’. The subject of the passive, therefore, is not involved nor affected physically by the event in any way, but it is emotionally affected by the event. Although (45) is a direct passive and (46) is an indirect passive, the degree of emotive nuance in (45) and (46) is very similar. The following is a similar case with a different verb,yom-u ‘read’.

(47) ... Tuneo wa

sono tame.ni zibun

no

minikui kokoroga

Tsuneo top that because oneself oen ugly yom-are-ru read-PASS-PRES

koto

0

NML ACC

hidoku

osor-e.....

heart

nom

Nanase ni Nanase dat

[Tsutsui]

terribly be.afraid-coNJ...

‘Tsuneo is terribly afraid o f his own ugly mind being read by Nanase because of that, and

160

Example (47) is a direct passive with a clear emotive undertone. This is because neither the subject of the passive, zibun no kokoro ‘his own mind’, nor the possessor of the mind, ‘Tsuneo’, which is the subject o f the main clause, is central to the event nor directly affected by the event. However, ‘Tsuneo’ would clearly be emotionally affected, if the event, zibun no minikui kokoro ga Nanase ni yom-are-ru ‘his own ugly mind being read by Nanase’, ever occurs. Let us compare (47) with an indirect passive involving the same verb, yom-u ‘to read’, in example (48).

(48) Tsuneo w a ..... terepasu Tsuneo top

ni kokoro o

telepathist by mind

kokuhukus-ita.....

acc

yom-are-ru

kyoohu o

read-PASS-PRES fear

acc

[Tsutsui 379]

overcome-PAST ‘Tsuneo overcame the fear of his mind being read by a telepathist..... ’

The subject of the passive, Tsuneo, here as well, will be emotionally affected by the event, Tsuneo wa terepasu ni kokoro oyom-are-ru ‘Tsuneo’s mind being read by a telepathist’, if it ever occurs. The degree of emotive nuance in (47) and (48) is very similar.

Both (45) and (48) are examples of the direct passive that have a special emotive nuance. This is because the verbs, mi-ru ‘to see’ and yom-u ‘read’, in (45) and (48) respectively, are both low in transitivity. The subjects of the passive are, therefore, not central to the event nor directly affected by the event described. It is, however, the nature of

161

the passive subject that is understood to be somehow affected by the event. In order to interpret their subjects as being affected in some way, the passive clauses (45) and (47) have an additional emotive undertone (see Section 4.4.2).

Although they are syntactically direct passives, the function of examples (45) and (47) are extremely similar to that of the indirect examples (46) and (48). All of these sentences have a special emotive nuance. It thus can be said that the direct passive whose passive subject is not central to the event, nor directly affected by the event, has a special emotive undertone, and its function is very similar to that o f the indirect passives, which invariably have the emotive nuance.

4.4.4 Why adversative?

In the previous sections, we have seen that, if an NP’s relevance is low in the event described and it is not directly involved in the event physically or emotionally, some extra interpretation has to be supplied to show that the NP is indirectly or emotionally affected. In doing so, the NP’s relevance is increased and it is thus integrated appropriately into the event. However, we have not considered why, in most cases of Japanese passive constructions, this extra interpretation is adversative . First we will reconsider Wierzbicka’s analysis of the ‘view embedded in Japanese syntax’ below:

‘If an action (or event) does not involve us directly and yet affects us, it will

162

probably affect us negatively’ (Wierzbicka, 1988: 270)

Shibatani (1996: 15) comments that, as Wierzbicka states above, whether or not we are involved in the event directly affects whether we consider the event positively or negatively. He suggests that we tend to regard an intentional act as favourable, whereas we regard an unintentional act as unfavourable. Shibatani states that this is based on our universal world view, which is that the way the world works does not in itself lead to good outcomes, and that in order to get a good outcome, our intentional intervention is needed. He considers the following examples (Shibatani 1996: 16):

(49) Hanako wa Taroo ni Hanako

to p

Taro

agt

kaminoke o

ki-rase-ta.

hair

cut-CAUS-PAST

acc

‘Hanako made Taro c u t her hair.’ (50) Hanako wa Taroo ni Hanako

top

Taro

by

kaminoke o

k i- r a r e - ta .

hair

cut-PASS-PAST

acc

‘Hanako had her hair cut b y Taro.’

A causative sentence, (49), describes an intentional action, and it indicates that the result of the haircut would have been a favourable one, or at least one that the subject expected. A passive sentence, (50), on the other hand, implies an action that was not intended by the subject of the sentence and therefore an unfavourable result. Shibatani claims that in the prototypical adversative passive construction, the event happens unrelated to the subject’s

163

intention. In order to secure the relevance of the referent of the subject, the construction also describes the situation that the subject is somehow affected psychologically. Shibatani states that the adversative meaning of a passive sentence is produced by the following three factors: the subject’s lack of intention, the subject’s emotional affectedness, and our universal world view that our intentional intervention is needed to lead to a favourable outcome.

Now let us consider some other constructions in Japanese that involve the subject’s lack of intention. It is said that Japanese has ‘the general tendency towards richness of intransitive expression’ (Jacobsen 1992: 83). Although it does not have an extra-thematic NP as in the indirect passive, lack of intention is one of the key features of the spontaneous intransitive construction6, such as examples (51) and (52) below.

(51) Koogi w a m o o le c tu re

top

h a z im a tte -iru .

a lre a d y begin-PERF

‘The le c tu re h a s a lre a d y b e g u n .’

(52) K in o o y e ste rd a y

k o k o d e z ik o h e re

at

ga

a c c id e n t n o m

o k o t-ta. happen-PAST

‘An accident happened here yesterday.’

6

In this study, these are referred to as ‘unaccusative constructions’. (See Section 1.6.1.2.)

164

Japanese also has a so-called ‘spontaneous passive construction7’ which is made by adding the suffix -(r)are to the verb (see Section 1.5). It is used to express emotions or perceptions that involuntarily arise in the perceiving subject. Examples are shown below.

(53) Gakusei.zidai ga Schooldays

nom

natukasiku omoidas-are-ru. longingly

rem em ber-spoN -PR Es

‘I remember my school days fondly.’ (54) Samusa c o ld n e ss

ga

y a w a ra i-d a

yoo

ni

k a n z i-ra re -ru .

nom

abade-PAST

com p

dat

feel-spoN-PRES

‘It felt (to me) as if the cold had abated.’

As we have seen in the prototypical adversative passive construction in Japanese, in the spontaneous intransitive construction and the spontaneous passive construction the event is also described as occurring without the subject’s intention. However, in spite of this, in these latter two constructions, it is not necessarily understood to be unfavourable.

The difference between spontaneous constructions, such as those illustrated directly above, and the adversative passive construction is that the former do not imply anybody’s intention, neither the subject’s nor that of some other party. Rather than being caused or controlled by any intentional participant, these constructions present the event as 7 As a reader of the early draft has noted, this construction is more commonly referred to simply as the ‘spontaneous construction’. However, the term ‘spontaneous passive construction’ is used here to distinguish it from the ‘spontaneous intransitive construction’, discussed in the previous page.

165

happening spontaneously. In the adversative passive, on the other hand, the event is certainly described as happening unrelated to the subject’s intention, but, in addition, it implies the control of the ‘actor’ over the event. It can thus be said that the adversative meaning occurs when the event is described as something that happens unrelated to the subject’s intention but under the control of some other party. The subject’s sense of powerlessness about somebody (or something) else having control over the event is expressed as an additional adversative interpretation.

This chapter has investigated the type of passive with special emotive affectedness, widely referred to as the ‘adversative passive’ in previous research. We examined passives with an emotive undertone within the framework of each syntactic category of Japanese passive: the indirect passive, the semi-direct passive and the direct passive. We then investigated under what circumstances the special nuance appeared.

On the basis of this examination we found that, if a passive is syntactically classified as an indirect passive, it is always associated with this emotive nuance. Examples that have previously been presented as instances of ‘indirect passives’ that do not have the special emotive affectedness are actually better categorised as semi-direct passives - more specifically the so-called ‘possessor passive’, whose subject corresponds to the genitive case or ‘possessor’ NP in the active counterpart.

In the case of the direct passive, as well as the semi-direct passive, the degree of

166

the centrality of the subject and the impact of the event on the subject were found crucial to the degree of the emotive affectedness. The lower the degree of the centrality of the subject or the impact o f the event on the subject is, the stronger the special emotive undertone is perceived to be.

167

Chapter 5 Japanese passive in authentic texts

5.1

Data and methodology

As described in detail in Section 1.2, this research has primarily used authentic written and spoken data for analysis, in order to portray how Japanese passives are actually used. For the written data of this study, we collected 679 passive examples from ten different contemporary novels. Using the CD-Rom collection of Japanese novels, Shincho Bunko no Hyakusatsu (One hundred Shincho Paperbacks), we selected the ten most recent novels in the collection and searched the middle 100 pages of each novel. For the spoken data, the collection of Josei no Kotoba - Shokubahen (Women’s Language - Workplace section) and Dansei no Kotoba - Shokubahen (Men’s Language - Workplace section), commercially available on CD-Rom, were used. 169 passive examples out of a total of 16,921 utterances were found in these spoken data collections. Altogether, 848 passive examples have been analysed in this study.

The size of the spoken data (169 examples) is relatively small compared to the written data (679 examples). When it is necessary to take an overview of spoken and written data together on a certain issue, the percentages of spoken and written data are averaged1rather than giving the raw figures.

1 I am indebted to one of the readers of the early draft for suggesting the use of this method.

168

5.2

Basic findings

5.2.1 Proportions of each category in the data

5.2.1.1

Syntactic categories

As discussed in Chapter 3, in this book, Japanese passive constructions are syntactically classified into three basic groups; direct passive, semi-direct passive and indirect passive. Overwhelmingly the focus of previous study of Japanese passives has been on the indirect passive. However, the data shows that, in fact, the vast majority of instances of passive in Japanese, at least in the genres examined in this study, are the direct passive2. This is a highly significant finding of this study.

The following figure shows the proportion o f each syntactic type of passive accounted for in the data. It includes causative passives and spontaneous passives which

2 Yoshida (1996: 38) presented a similar finding in her research on passives in Japanese conversation. In her data, there are 172 out of a total of 243 instances of the direct passive (71%) compared to only 29 cases of the indirect passive (12%). Note that Yoshida’s category of indirect passive incorporates both the categories of indirect and semi-direct passives in this study.

L

169

have been treated separately from other types of passive in previous research. In this study, also, they are not included as Japanese passives in the strict sense. However, in order to compare the proportions of these two constructions with other types of passive, they are included in the following figure. The figure also includes a set of passive examples that cannot be clearly placed in any one group because some critical parts3 of the sentence are elided. In the figure, the top bar indicates the proportion of each syntactic type of passive accounted for in the spoken data, and the middle bar in the written data. The bottom bar shows the average of the percentages o f spoken and written data.

Figure 1. Syntactic categories Spoken data (n=186) Written data (n=743)

AVERAGE (%)

0%

□ N o t classified ■Spontaneous □Causative P □Indirect ■Semi-direct □Direct Passive

20%

40%

60%

80%

AVERAGE (%)

Written data (n=743)

Spoken data (n=186)

4.2 2.4 2.4

6 31 27

14

1.2 2.2 87.8

10 32 637

100%

1 2 2 0 167

3 In Japanese, mainly in spoken language, particles are often elided. Sometimes you cannot even tell if a NP is in the accusative case or nominative case, and therefore, cannot distinguish the indirect passive from the direct passive. This is the reason why spoken data has a sizable proportion in the ‘not-classified’ group.

170

The most striking feature of this figure is the predominant proportion of the direct passive. Both in the spoken and the written data, almost 90% of all the passive clauses are, in fact, examples of the direct passive. In contrast, the examples of the indirect passive found in this data search account for only 1.2%. Even when these are put together with examples of the semi-direct passive, as has been done in previous research, the proportion is still only 3.4%.

The indirect passive can occur with an intransitive verb, and it is usually accompanied by a special emotive nuance. Since these characteristics are unusual from the Indo-European point o f view, and are very intriguing, the indirect passive has, understandably, been widely discussed amongst researchers of Japanese passives. However, the frequency of its occurrence seems, in fact, to be very low. Even though the indirect passive has been a major focus of research on Japanese passive constructions, the results show that it can hardly be taken to represent Japanese passive constructions in general.

5.2.1.2

Semantic categories

Next, let us consider semantic categories. The basic semantic groups o f Japanese passive constructions used in this study are the plain passive and the passive of interest, as seen in Section 3.2. The plain passive is then divided into two types: the demotional passive and the attributive passive. The passive of interest is also divided into two groups: passives with a latent affectee and sentient passives. The following factors are considered to classify the

171

tokens:

Demotional passive is one that is used primarily to marginalise the agent. The agent is, therefore, often eliminated. However, if it is needed, it is marked by ni-yotte. As for the types of verb that appear in the demotional passive, they are mainly factitive verbs. (See Section 3.2.1.1.) Attributive passive is one that is used to describe or imply some attribute of the referent of the subject. It does not depict the occurrence or existence of an event at a specific time and place. (See Section 3.2.1.2.) Passive with a latent affectee is one in which one can assume a latent affectee. A latent affectee is a sentient entity that one can assume is affected by the event in some way, but is not a participant in the passive sentence. (See Section 3.2.2.1.) Sentient passive is one that has a sentient (most likely human) subject, and is used to describe a situation in which the referent of the subject is directly or emotionally affected by the event. (See Section 3.2.2.2.)

In previous research, the demotional passive4 is regarded as being used mainly in the written language. First, we will consider the difference between the proportion of the demotional passive in the written data with that in the spoken data. The following figure shows the proportion of each semantic type of passive accounted for in the data.

4 It is not inherent to the Japanese language. Example: Sono hakubutukan wa yuumei-na kentikuka ni-yotte tater-are-ta. ‘That museum was built by a famous architect.’

172

Figure 2. Semantic categories

Spoken data (n=169)

■Demotional (Plain)

104

F

Written data (n=679)

■Attributive (Plain)

AVERAGE (%)

0%

20%

40%

60%

80%

100%

In the figure above, it is clear that the proportion of demotional passive in the written data (28.4%) is more than twice as large as in the spoken data (12.4%). This finding supports the view of previous researchers, such as Kinsui (1997: 764), that the demotional passive (in Kinsui’s terms the ‘ni-yotte passive’) is ‘rather a stiff literal [sic]5 expression’ and therefore is mainly used in the written language.

However, Kinsui also claims that this type o f passive ‘is not used in the spoken language'. It is true to say that this type of passive mainly occurs in the written language. However, the data shows that there are 21 out of a total o f 169 examples of demotional passives (12.4%) in our spoken data.

5 Kinsui probably intended to use the word ‘literary’ rather than ‘literal’.

173

Next, let us have a look at the attributive passive6. As seen in Section 3.2.1.2, Masuoka (2000) claims that the attributive passive is peripheral, as the frequency o f the actual usage of this kind of passive is low. In this data search, however, we found 78 examples of the attributive passive out of 679 passive examples in the written data, and 33 out o f 169 examples in the spoken data. It thus constitutes about 15.5% of the data on average. The number is actually much larger than that of the widely recognised syntactic category of indirect passives (less than 5%). We suggest, therefore, that the attributive passive should not be treated as peripheral.

Looking more closely into the data, the difference between the proportion of the attributive passives in the spoken data and that in the written data is apparent. The proportion of attributive passive in the written data (11.5%) is only about half that of the spoken data (19.5%). This is completely opposite from the case of the demotional passive. This is an important point, supporting the analysis that the attributive passive should be considered separate from the demotional passive.

Another striking fact regarding the distribution o f the various semantic categories of passive is the high proportion of non-sentient passives. The non-sentient passive consists of the demotional passive, the attributive passive, and the passive with a latent affectee. It has been marginalised in the research as, like the demotional passive, it is said not to be inherent to the Japanese language (Yamada 1908, Mitsuya 1908, Matsushita 1930 and 6 Example: Kono manga wa kodomo-tali niyoku yom-are-ru. ‘This comic is often read by children.’

17 4

Hashimoto 1931). However, it constitutes more than 40% of this data set. Even in the spoken data, about 38% (65 out of 169 examples) are non-sentient passives. Although sentient passives are in the majority, it is apparent that non-sentient passives cannot be treated as peripheral, as they have been in previous research. This issue will be discussed in more detail in Section 5.2.3 in relation to the subject types.

5.2.2 ‘actor’

An ‘actor7’, or what many scholars refer to as the passive ‘agent’, is one of the main elements of the passive sentence. However, it is a common observation that an ‘actor’ argument is often elided in a passive sentence. In order to verify this generally accepted view, the proportions o f the passive data with and without an overt ‘actor’ in the sentence are compared in the following figure:

7 See Section 1.6.2 for the definition of Actor.

175

Figure 3. 'actor'

T

___ (

1

1_ _ _

1

153

Spoken (n=169)

t

16 1 °

■■ Written (n=679)

M—

500

AVERAGE (%)

■ 1 0%

.

1 20%

82,1

..............

1

40%

□ n o 'actor' ■w ith 'actor'

| 1

60%

17.9

1 .......... 80%

| ^ 100%

The first thing we notice in the figure above is the large proportion of the passive without an overt ‘actor’ overall. More than 80% of all the passive sentences in the data do not involve an overt ‘actor’ in the sentence. This finding confirms the claims made in previous research, such as Siewierska (1984: 35), Givon (2001: 126), Yoshida (1996: 41), Masuoka (1982) and a number of others. Siewierska claims that ‘Statistical data reveal that agentless passives are far more common than those with an agent’. We have already seen Givon’s (2001: 126) investigation on ‘Percent of non-anaphoric zero agents in active and passive clauses in narrative text’ in Section 1.4. He also recognises the high frequency of the passive without an overt agent in some languages. Siewierska (1984: 35) mentions a number of studies on various languages, such as Huddleston (1971) and Krauthamer (1981) on English, Duskova (1972) on Czech and Brinker (1971) and Schoenthal (1976) on German. Siewierska (1984: 35) states that an overt ‘actor’ cannot be involved at all in a passive clause in many languages, such as Latvian, Urdu, Kupia, Classical Arabic, Amharic, Igbo, Tera, Sonrai, Fijian, Atjnjamathanha, Cupeno, Cora, Huichol, Cahuilla, Shoshoni and

176

Pepecano. She also maintains that in some other languages, a passive clause can, but need not involve an overt ‘actor’. According to this data search, it is clear that Japanese belongs to this latter group.

Another aspect that should be considered with regard to the overt ‘actor’ in a passive is the nature of the ‘actor’. Shibatani (1998: 137) suggests that, in some passive constructions, NPs low in the Reverse Empathy Hierarchy (cited below) do not easily occur as an ‘actor’. He claims that those NPs ‘are most likely to be placed in the syntactically most prominent position, namely subject position, because the speaker tends to empathise with them and to treat them as central entities in the described event’ (Shibatani 1998:137).

Reverse Empathy Hierarchy natural force > instrument > institution > generic human > specific human > 3Id person > speech act participants (Shibatani 1998: 134)

To test Shibatani’s view in our data, the proportion of the nature o f each ‘actor’ that overtly appears in a passive sentence is summarised in the following figure. The category of non-sentient NPs is added before that of natural force, since it is regarded here as the highest in the Reverse Empathy Hierarchy for Japanese. There was only one instrument ‘actor’ found in the data. It is included in the non-sentient group.

177

Figure 4. The nature of an overt 'actor' El non-sentient □natural force

Spoken (n=16)

■institution □g eneric human

Written (n=179)

■specific human □ 3 rd person

A V E R A G E (%)

□S APs 0%

20%

40%

60%

80%

100%

In the figure above, it is clear that the speech act participants (first person and second person), the lowest category in the Reverse Empathy Hierarchy, do not often occur as overt ‘actor’. Their frequency is, in fact, very low (4%). This supports Shibatani’s observation. However, the second lowest in the Reverse Empathy Hierarchy, the 3rd person ‘actor’, constitutes almost 30% of the data. It is, in fact, the largest group of all. This finding contradicts Shibatani’s view.

The 3rd person ‘actor’ has a special role in a Japanese passive. As seen in Section 5.2.1.2, more than 60% of all the data belong to the semantic category o f the passive of interest, which describes a situation in which the subject is in some way affected by the event. It is also clear that the effect is more likely to be an adverse one. (See Section 5.2.5, Propositional Meaning.) The ‘actor’ of a passive is the one who is responsible for the adverse effect on the subject, and the one who is to blame. The easiest target for the subject

178

(or the speaker) to blame is a 3rd person; it is easier than to blame the 1st person, the 2nd person or a non-human entity. It is presumably this aspect of Japanese passives that allows the 3rd person ‘actor’ to occur so often.

5.2.3 Subject

5.2.3.1

Subject types

The subject, very often the topic in Japanese, is the most prominent factor in a passive sentence. In this section, therefore, we will examine the nature of a subject used in a passive sentence.

First, let us consider Kuno's Principles of Empathy Relations (Kuno, 1977: 646 & 652, 1987: 179 & 207, 1990: 45), as they are highly relevant to the current discussion. Kuno’s Principles are that:

(i)

the speaker is more likely to adopt the viewpoint of a human NP than that o f a non-human NP (Humanness Empathy Hierarchy); Human > Nonhuman animate > Inanimate

(ii)

the speaker is most likely to adopt the viewpoint of the subject (Surface

179

Structure Empathy Hierarchy) Subject > Other NPs

Based on Kuno’s Principles of Empathy Relations, we can hypothesise that NPs low in the Empathy Hierarchy (developed from Shibatani’s (1998: 134) Reverse Empathy Hierarchy and cited below), such as non-sentient NPs, do not often appear as a subject of a passive clause.

Empathy Hierarchy Is1 person > 2nd person > 3rd person > specific human > generic human > institution > (animal) > instrument > natural force > non-sentient

To put the above hypothesis to the test, the proportion of each group of subject in the Empathy Hierarchy that appears in a passive construction is summarised in the following figure:

180

Figure 5. Subject types

□non-sentient ■animal

AVERAGE (%)

Written (n=679)

Spoken (n=169)

36.8

279 10 10

55 2

□third person

1.3 2.2 5.5 2.1 11.2

■second person Sfirst person

4.3 36.6

in s titu tio n ■generic human □specific human

50 24 120 10 176

5 6 1 8 12 80

From this data search, it is clear that almost 40% of all the passive examples collected have a non-sentient NP as subject. In previous research, however, this numerically prominent category has once again been marginalized. It has been called ‘hizyoo no ukerni (non-sentient passive)’, and it is said not to be inherent to the Japanese language by many traditional Japanese grammarians (Yamada 1908, Mitsuya 1908, Matsushita 1930 and Hashimoto 1931). (See Section 3.2.1.1 Demotional passive.) However, our findings show that, inherent or not, the non-sentient passive constitutes almost 40% of all the data gathered from the authentic contemporary sources examined. Two of the examples are cited

below:

(1) Nihoneo zvaanaru wa sekaizyuu Japanese journal

top

de ur-are-ru

n

da

yo.

all.over.the.world in sell-PASS-PRES nml cop sfp

‘You know, the Nihoneo Journal is sold all over the world.’ [Josei 9320: female, 31, Company employee (editing)] (2) Utvuu

no

tituzvo wa aru

Universe gen order tamot-are-te

ki-ta

top

tan’itu no

utyuu

isi

ni-yotte

certain single gen universe will by no

da. [Tsutsui: 351]

retain-PASS-coNj come-PAST nml cop ‘The thing is that the order of the universe has been retained by a (certain) single universal will.’

This finding is contrary to the hypothesis proposed above, based on Kuno’s Principles of Empathy Relations that NPs low in the Empathy Hierarchy, such as non-sentient NPs, do not easily occur as a passive subject.

Since the nature of a subject is an important factor even in their definition, next, let us look at the nature of the subject used in each of the two semantic types o f Japanese passive: the plain passive8 and the passive of interest9, separately. Based on the nature of

8 The plain passive is defined in this study as one generally having a non-sentient subject, and describing an event objectively without any special meaning of affectedness. 9 The passive of interest is defined here, in contrast to the plain passive, as one that portrays

182

the plain passive, we might hypothesize that most of the non-sentient subjects used in the data actually appear in the plain passive. The following figure shows the proportion of each group of subject used in each semantic type o f Japanese passive. In the figure, the percentages of spoken and written data are averaged. The bottom bar shows the average of the percentages o f the passive of interest and the plain passive.

Figure 6. Subject types 2

Passive of interest (%)

Plain passive (%)

AVERAGE (%)

0%

□non-sentient ■animal □institution

20%

40%

60%

AVERAGE (%)

Plain passive (%)

47.3 1.1 3.2

85 0.4

80%

100%

Passive of interest (%) 9.6 1.9 0

■generic human □specific human □third person

5.5 1.9 9.5

6.5 4.8 1.1 1.8

6.3 2.6 17.1

■second person □first person

3.3 28.2

0.2 0.2

6.3 56.2

From the figure above, it is apparent that the types o f subject used in the plain passive and an event in terms of the concerns of the referent of the subject.

183

in the passive of interest are totally different - almost opposite in fact. In the plain passive, non-sentient subjects constitute almost 90% of the data. This verifies the hypothesis above. In the passive of interest, on the contrary, non-sentient subjects are used in only about 10% of occurrences, and first, second and third person subjects, NPs high in the Empathy Hierarchy, together constitute nearly 80% of the data. The result shows that the hypothesis set up before based on Kuno’s Principles of Empathy Relations - that NPs low in the Empathy Hierarchy, such as non-sentient NPs, do not often appear as a subject of a passive sentence - can only be applied to the passive of interest, not to the plain passive.

5.2.3.2

Degree of topicality of the subject

Next, let us consider how the subject appears (or does not appear) in a passive clause. Figure 7 below summarises how the passive subject appears in each syntactic group of Japanese passive; if the subject is marked by the topic particle wa, appears as a head noun of a noun-modifying clause, is marked by the nominative particle ga, or does not appear at all. In the figure, the percentages of spoken and written data are averaged.

184

Figure 7. Subject's degree of topicality 1: in each syntactic group indirect (%)

0 n o subject ■w a (TOP) □head noun □ g a (NOM) ■ unidentified

semi-direct (%)

direct (%)

Average (%)

0%

20%

40%

60%

80%

100%

The most striking feature of this figure is the predominant proportion of subjects in the indirect passive that do not appear in the clause. Unlike the case of the ‘actor’, the fact that the passive subject does not appear in the clause does not mean that it is defocused. On the contrary, it is highly topicalised. The subject in the indirect passive is an emotively affected entity. It is therefore most likely to be a topic o f the discourse. It is an important discourse tracking mechanism to elide the topic of the discourse in Japanese. When it is not elided, the subject in an indirect passive is marked by topic particle wa. It is clear from the result above that the subject of the indirect passive is the most topicalised of the three syntactic types.

In the case o f semantic groups, Figure 8 below illustrates how the subject appears in the passive clause. The numbers in the figure indicate the average of the percentages of

185

spoken and written data.

Figure 8. Subject's degree of topicality 2: in each semantic group

There are some interesting points in Figure 8. The first thing we notice in the figure above is the large proportion of the sentient passive without an overt subject. This supports the Humanness Empathy Hierarchy proposed by Kuno (1977: 646 & 652, 1987: 179 & 207, 1990: 45). The subject in a sentient passive is a sentient NP. As discussed in Section 5.2.3.1 above, according to the Humanness Empathy Hierarchy, the speaker is more likely to adopt the viewpoint of a sentient NP over that of a non-sentient NP. The NP whose viewpoint is adopted, in turn, is most often identified as the subject NP. This is presumably the reason why the subject in a sentient passive is highly topicalised; almost 80% are either elided or marked by topic particle wa.

186

Another striking feature of Figure 8 is the large proportion of nominative go-marked subjects in the passive with a latent affectee and the demotional passive. As seen in Section 2.4.1, Masuoka (2000: 55) claims that many of the demotional passives are expressed as topicless sentences, in which the subject is marked simply by the nominative particle go111. The results of this data analysis confirm Masuoka’s claim. This is because the demotional passive usually describes an objectively observed fact, as it is, without representing somebody’s subjective view.

In terms of the passive of a latent affectee, the subject is not often topicalized, and simply marked by the nominative particle ga, since the topic of the passive clause is presumably the latent affectee", not the subject. Thus this construction is unusual amongst passives in Japanese in that it has a separate topic and subject.

5.2.4 Noun-modifying clauses

Since they constitute a considerable proportion of the data, next we will examine the proportion of passives used in a noun-modifying clause. It is apparent that these examples appear much more commonly in written than in spoken data, as seen in Figure 9 below. In10

10 Very often this kind o f sentence with particle ga is interpreted as an objective description, as opposed to a sentence where the subject is selected as a topic and therefore expresses the situation from a certain perspective. 11 A latent affectee is a sentient entity that one can assume is affected by the event in some way, but is not a participant in the passive sentence.

187

the figure, the bottom bar shows the average of the percentages of spoken and written data.

Figure 9. Noun-modifying clauses S p o ken (n =1 69 ) ElNoun-m odifying ■ N o t noun-m odifying

W ritten (n=679)

A verage (%)

0%

20%

40%

60%

80%

100%

As apparent from the figure above, passives appear in a noun-modifying clause in the written data, nearly three times as often as in the spoken data (35% and 13%, respectively). We therefore consider the examples in the written data in this section. Figure 10 below summarises the proportion of the passives used in a noun-modifying clause for each semantic type of passive in the written data:

188

Figure 10. Noun-modifying clauses in written data

■Noun­ m odifying □ N o t noun­ modifying

a

0%

20%

40%

60%

80%

100%

If we look at the figure above, one thing is very clear: the attributive passive occurs in a noun-modifying clause more often than not (just over 60%) whereas other types of passive occur in a noun-modifying clause less than 35% of the time.

The primary function of the attributive passive is to describe or imply some attribute of the referent of the subject. (See example (4) below.) In an attribute-describing sentence, the entity whose attribute is accounted for is the syntactic subject. The subject of an attributive passive in a simple sentence corresponds to the relativised NP of a noun-modifying clause.

The main function of the noun-modifying clause is to modify or qualify the meaning o f the relativised NP (head noun). The noun-modifying clause is usually divided

189

into two basic types: restrictive type and non-restrictive type. The restrictive type is used to identify one NP, which meets a certain condition, out of several NPs, whereas the non-restrictive type simply provides additional information about the NP. Our data shows that passives appear equally in both of these two types of noun-modifying clause.

It is notable that the function o f the noun-modifying clause, either restrictive or non-restrictive, is very similar to that of the attributive passive. This is presumably the reason why the attributive passive is so often used in a noun-modifying clause. Let us consider the following examples:

(3) Meimon P re s tig e o u s

to

iw-are-ru

quot

kono si

say-P A S S -P R E S

th is

no

Koowa-tyuugaku, ... [Tsutsui 376]

c ity

G EN

‘ T h e K o w a j u n i o r h ig h s c h o o l in th is c ity , w h ic h

(4)

Kono

si

no

K o o w a -ty u u g a k u

T h is

c ity

gen

K o w a . J u n io r . H ig h . S c h o o l

wa

top

is said

K o w a . J u n io r . H ig h . S c h o o l

t o b e a p r e s tig io u s s c h o o l, . . . ’

m e im o n

p r e s tig io u s

to

quot

iw - a r e - te

s a y -P A S S -C 0 N J

i- r u ...

be-PRES ‘The Kowa junior high school in this city is said to be a prestige school ... ’

Example (3) is an example from the data of an attributive passive used in a noun-modifying clause. It qualifies the relativised NP, Kowa junior high school, as being said to be a prestigious school. Example (4) is made up, using the attributive passive equivalent to

190

Example (3) in a simple sentence. It describes the attribute of the subject, Kowa junior high school, that it is said to be a prestigious school.

5.2.5 Propositional meaning

In previous research, such as Shibatani (1990) and Kuno (1973), a crucial factor for categorising Japanese passive constructions has been to know whether or not the passive has an adversative meaning. Shibatani (1990: 318) suggests that one type of adversative reading arises when the whole passive clause designates that the referent of the subject is adversely affected psychologically, even though the verb itself has a neutral lexical meaning without any adverse nuance, as in Shibatani’s example, cited below:

(5) a.

Tomodatiga friend

kaet-ta.

nom retum-PAST

‘The friend returned.’ b. Taroo wa tomodati ni kaer-are-ta. Taro

top

friend

by retum-PASSPAST

‘Taro was adversely affected by his friend’s returning.'

In example (5a), the verb, kaeru ‘to return’, itself is neutral, and the clause does not have any adversative connotation. Example (5b), the corresponding passive clause, however,

191

indicates that the referent of the subject, Taro, was bothered by the event of his friend’s returning.

Shibatani distinguishes this type of adversative reading from one originating in the verb’s lexical meaning, as in example (6) below:

(6) a.

Kodomo ga child

Takasi o

nom Takashi acc

nagut-ta. hit-PAST

‘His(/A) child hit Takashi.’ b. Takasi wa kodomo ni nagur-are-ta. Takashi top child

by hit-PASs-PAST

‘Takashi was hit by his(/a) child.’

Example (6b) has an adversative reading. Shibatani (1990: 318) claims that it is due to the fact that the verb, naguru ‘hit’, has an adverse meaning itself, and this kind of lexical adverse meaning should be distinguished from the adversative reading of the entire passive clause seen in example (5b).

In some cases, however, it can be a fairly subjective matter to decide whether or not a passive clause has an adversative reading in Shibatani’s sense. This is especially true now that we know some direct passives have an adversative reading too, (see Section 4.4.3), making it sometimes almost impossible to distinguish the adversative reading in Shibatani’s

192

sense from a lexical adversative meaning of the verb. As seen in Section 4.4.3, attempting to explain why some direct passive clauses have an adversative reading while others do not, Shibatani (1996: 7) suggests that if the subject o f the passive is high on the scales of physical presence (proximity) and physical affectedness, namely highly relevant to the event, it does not require any supplementary semantic input to justify its relevance to the event. The passive sentence, therefore, is not likely to have an adversative meaning. Consider the following example:

(7) a. Keisatu wa

Nobuko o

utagat-te

ir-u.

Police top Nobuko acc suspect-coNJ be-PRES ‘The police suspect Nobuko.’ c. Nobuko wa Nobuko top

keisatu ni

utagaw-are-te

ir-u.

police by suspect-pass-conj be-PRES

‘Nobuko is suspected by the police.’

Example (7) involves a verb that has a lexically adversative meaning, utagau ‘suspect’. Example (7a), thus, has an adversative meaning, as well as its passive counterpart, example (7b). However, example (7b) also satisfies Shibatani’s (1996: 7) criteria for a passive clause to have an adversative meaning; namely the subject is low in physical presence and physical affectedness. It is, therefore, not clear whether the adversative reading of example (7b) arises from the verb’s lexical meaning or from something else.

193

Attempting to find a more objective measure, we will investigate the kind of proposition in which a passive appears. It is ‘taken to represent a “once-off ’ interpretation of a text-sentence as it is used in a context’. In order to clarify the relationship between the propositional meaning and each type of passive, in this section, the meaning of the proposition of the sentence in which each passive clause occurs will be considered. Proposition are divided into three types: negative proposition, neutral proposition and positive proposition. Examples (8), (9) and (10), cited below, represent a negative proposition, neutral proposition and positive proposition, respectively:

(8) Zibun

ga

kakusi.mot-te

i-ta

zettaiteki tabuu ni koosyuu no

Myself nom secretly.hold-coNJ be-PAST absolute taboo on public sawar-are-te

i-ru

ka

no

yoona, aru

touch-PASS-cONJ be-PRES as.ifNMLlike aruiwa syuutisin or

syu no

gen

in.ffont

toowakukan,

certain kind gen embarrassment

no gotoki mono ni osow-are-te

sense.of.shame genlike

menzen.de

i-ta. [Fujiwara 326]

thing by seize-PASS-coNJ be-PAST

‘I was seized with some kind of embarrassment or something like a sense o f shame as if I was being publicly touched on some absolute taboo that I secretly held.’ (9) Ee,

Nihonzin

no kanari ooku no hito

ga

desu nee, Bukkyoo

Well Japanese.people genfairly many gen people nom cop-pres Buddhism sinzya tosite kazoer-are-te bliever as

i-mas-u. [Dansei 2704: male, 45, University Lecturer]

count-PASS-coNJ be-POL-PRES

194

‘Well, quite a large number of Japanese people are, uh, counted as Buddhists.’ (lO)Firentse

syoonin Tedarudi wa, seikoku

Itaria de yori.mo Huransu de

Florentine merchant Tedaldi top birth.country Italy in than na

o

sir-are-ru

France

in

yoo.ni nar-u. [Shiono 442]

name acc Ioiow-pass-pres like

become-PREs

‘The Florentine merchant, Tedaldi, comes to be better known in France than in Italy, the country of his birth.’

Example (8) is syntactically an indirect passive and semantically a sentient passive, and its propositional meaning is negative. The subject (I), in talking to the mother of a neighborhood child, discovers that the child is adopted. The subject was embarrassed because this kind of topic was an absolute taboo for him. This, therefore, is regarded as a negative proposition. Example (9) is syntactically categorised as a direct passive and semantically an attributive passive, and it has a neutral propositional meaning. The speaker, a university lecturer, is giving a lecture on religious sociology. It is an objective statement that many Japanese people are regarded as Buddhists from the point of view of religious sociology. It is, thus, considered as a neutral proposition. Example (10) is an example of the semi-direct passive, syntactically, and of the sentient passive, semantically. We understand Tedaldi had witnessed the fall of Constantinople, and had told the whole story to a certain French man. His story got around quickly in France, and he became famous there. The propositional meaning of example (10) is, thus, considered a positive one.

195

5.2.5.1

Propositional meaning in each syntactic group

In this section, we will look at what kind of propositional meaning passive clauses of each type tend to hold: negative proposition, neutral proposition or positive proposition. It is one of the syntactic groupings of passive, the indirect passive, that previous research has generally regarded as having an adversative reading. For this reason, let us first consider the propositional meaning for each syntactic group of Japanese passive. In the figure, the percentages of spoken and written data are averaged.

Figure 11. Propositional meaning 1: in each syntactic group Indirect (%)

Semi-direct (%)

Direct (%)

Average (%)

On average, about 77% of passive clauses are used in a sentence that has a negative proposition, 20% have a neutral proposition, and only about 3% a positive proposition.

196

One of the most noticeable features in the figure is that all o f the indirect passive clauses and more than 70% of the semi-direct passive clauses hold a negative propositional meaning. One such example of the indirect passive is example (8), cited above. This finding supports the claims made on the basis of previous research. Since the indirect passive (which includes the semi-direct passive in definitions used in previous studies) is said to generally have an adversative meaning, it is natural that it mainly has a negative propositional meaning.

However, it is important to note that we do find two examples of the indirect passive that are used in a neutral and a positive context12 in our data, even though they still have a negative propositional meaning. First, let us consider such examples of the indirect passive, cited below:

(ll)Obaatyan Granny noni,

wa yo-nin top

musuko ni sensoo de sin-are-ta

four-CLF gen son

doosite kono yo

even.though why ir-u

no

this

musuko.tati no

be-PRES son-CLiz

gen

by war

in die-PASS-PAST should

de mata a-eru

to

world in again meet-POTEN quot kazu

ga

yonin

hazu.na

de

wa nak-u

sinzi-te believe-coNJ sannin.na no

number nom four-CLF cop top neg-conj three-CLF nml

12 We follow Brown and Yule (1983: 25-26 & 35-46) in defining the term ‘context’ in a broad sense to refer to the ‘environment’ or ‘circumstances’ in which a sentence or a clause is used.

197

da-roo

to.iu koto dat-ta. [Miyamoto 280]

COP-EXHT CMPL fa c t

COP-PAST

‘It was the fact that even though she had been deprived by the death of four of her sons in the war, why was it that the number of the sons granny believed she could meet again in this life13 was three, not four.’

(12)... sara.ni

wa bonnetto no

furthermore nado mo etc.

top

bonnet

gen

ue ni nor-arc-te

tatioozyoos-ita koto

top on get.on-PASS-cONJ be.stuck-PAST

case

at-ta. [Fujiwara: 240]

even there.is-PAST

‘... furthermore there even was a time when I was stuck because (the kids) got on the bonnet (of my car).’

There is only one example each of an indirect passive used in a neutral context (example (11)) and in a positive context (example (12)). Example (11) is a part o f a fairy-tale-like story that ‘my’ grandmother always told ‘me’ when ‘I’ was a child. When the story came to the part about her sons, T always wondered why ‘my’ grandmother believed she could see only three of her deceased sons again, reincarnated during her lifetime, even though she had lost four sons. It is just a child’s innocent question, and is therefore, regarded overall as a neutral context, even though the proposition itself is negative. The context in which example (12) is used is that the subject (I) was very popular among the neighborhood 13 ‘My’ grandmother believes in reincarnation. She is convinced that her already deceased sons will be reincarnated in her life time, and she will see them again.

198

children, and when ‘I’ got home in ‘my’ car from work, they usually gave ‘me’ a warm welcome, and sometimes they even got on the bonnet of ‘my’ car to show their affection. It often cheered ‘me’ up, especially when T had problems at work and was tired. This context, thus, is considered as a positive one, even though we might assume that children climbing on one’s car would normally be thought of as something negative.

A point in common between example (11) and example (12) is that the passive clause itself still has an adversative reading, as common in the indirect passive, and moreover, its propositional meaning is negative. In example (11), the passive clause, obaatyan wa yonin no rnusuko ni sensoo de sin-are-ta ‘four of her sons died on my grandmother in the war’, has an adversative meaning and negative propositional meaning, since it is a part of recollection of a query T had in ‘my’ childhood. Even so the broader context is regarded as neutral. In case of example (12), the propositional meaning of the whole sentence is negative. Nevertheless it appears in a positive context. The use of the indirect passive in this context implies that the subject ‘I’ is trying to be modest. It is almost as if ‘I’ is not allowing himself to be openly happy about the fact that he is very popular among the children, and is pretending to be a little annoyed. Even though the context is positive, one can still detect something negative about it.

Although a considerably large proportion of semi-direct passives, more than 70%, also have a negative propositional meaning, the proportion is not as striking as with the indirect passive. The number of semi-direct passives whose propositional meaning is

199

neutral or positive is not insignificant. Consider the following:

(13)

Taitoru Title siai

o

take-PASSCONJ since mainly

acc

o

kara syu.to.site Kankoku toka Toonan.Azia

ubaw-are-te

s-ite

i-ta

to

ki-ite

Korea

or

i-mas-u

South.East.Asia

de in

kedo ... [Sawaki 602]

match acc do-coNJ be-PAST quot hear-coNJ be-POL-PRES but ‘Since your title was taken, I heard that you mainly had fights in Korea and South East Asia ...?’ (14)

Ikuraka

no

kane

o

okur-u to siai

no

tyokuzen.ni nanigasi ni

Some.amount gen money acc give-PRES if match gen right.before so-and-so dat dare.sore.san

kara

gekireisyoo ga

todoi-te

ori-mas-u

to

Mr.so-and-so from encouragement.prize nom be.received-coNJ be-POL-PRES quot ringu anaunsaa

ni namae o

yomiager-are-ru.[Sawaki 614]

ring.side.announcer by name acc call-PASS-PRES ‘If you gave some money, your name would be called out by the ring side announcer, saying “The encouragement prize was given to so-and-so from Mr so-and-so”, just before the match.’

Example (13) involves a semi-direct passive, and has a neutral propositional meaning. Although the verb itself has a negative meaning of taking something away, the propositional meaning of the sentence is considered to be neutral, since it is a part of an interview held four years later, just before the return match of the protagonist (you). The

200

interviewer is asking about what the protagonist has been doing during his four-year break. Example (14) also involves a semi-direct passive, but its propositional meaning is positive. The person who gave the money presumably felt proud when his name was called before the match. Although far less frequently than the direct passive, it is clear that the semi-direct passive can be used in a sentence that holds a neutral or positive propositional meaning as well as negative one. In this aspect, the semi-direct passive is distinguished from the indirect passive, and positioned in between the direct passive and the indirect passive.

S.2.5.2

Propositional meaning in each semantic group

Next, let us examine the propositional meaning for each semantic group of Japanese passive. This is summarised in the following figure. In the figure, the percentages of spoken and written data are averaged.

L

201

Figure 12. Propositional meaning 2: in each semantic group

^Negative UNeutral ^Positive

Overall, more than 55% of passive clauses in the entire data set are used in a sentence that has a negative proposition, nearly 40% have a neutral proposition, and only about 5% a positive proposition.

From the figure above, it is evident that more than 80% of the sentient passives hold a negative propositional meaning. The sentient passive is one that describes a situation in which the subject is somehow affected, and has a sentient subject. All the indirect passive and the semi-direct passive, and part of the direct passive are semantically classified into the sentient passive. Our findings show that, whether they have a special adversative meaning or not, the majority of the sentient passives are used in a sentence that has a negative propositional meaning.

202

If we look at the figure from another angle, it is obvious that the majority o f the attributive passives have a neutral propositional meaning and only about 15% hold a negative propositional meaning. In the written data, in particular, the number is very small. Only nine such examples (just over 10%) are found in the written data. Three of them are given below:

(15) Nanisiro toruko-zin

wa zitu.no oya

Anyway Turkish-CLF top real ut-te

kane

o

o

koros-ita mono sae,

dorei ni

parent acc kill-PAST person even slav as

mooke-ru hoo o

erab-u

to

iw-are-te

sell-coNj money acc eam-PRES one acc choose-PRES quot say-PASS-coNJ i-ru. [Shiono410] be-PRES ‘Anyway, the Turkish people are said to choose to earn money by selling even those who have killed their real parents as slaves.’ (16)Kane

to tii

to

‘ikka itimon’

no

gakureki

no

meiyo ni

Money andposition and the.whole.family gen academic.background gen honour by i-ta

sasaer-are-te

Huziwara-ke

ni

wa ookina gata

maintain-PASS-cow be-PAST Fujiwara.familyDAT top big k-ita

ato

de

ga

damage nom

wa at-ta. [Sono 1063]

come-PAST after cop top cop-past ‘It was, in fact, after the Fuiiwara family, that was maintained by its money, position

203

and the honor of ‘the whole family’s’ academic background, got all rickety.’ ri71Asoko no That ar-u

g en

uti

wa donna

family

top

kara

ziken

ga

okot-ta-tte

kane

ga

what-sort.of incident n o m happen-PAST-even.if money n o m

sukuw-are-te

i-rut-te

iw-are-ta

have-PRES because Save-PASS-CONJ be-PRES-QUOT

n

say-PASS-PAST

da-tte. nm l

‘I heard that they said that that family had been saved because they had money, no matter what sort of incident occurred.’ [Sono 1069]

Examples (15), (16) and (17) all have a negative propositional meaning. However, they do not describe the fact that the subject is affected by the event in some way. They still denote the event objectively. This is the reason why these examples o f the attributive passive can hold a negative propositional meaning.

The attributive passive, in fact, belongs to the plain passive which is defined in this study as one, which (i) generally has a non-sentient subject, and (ii) describes a situation objectively without any special meaning of affectedness. The subjects in examples (15), (16), and (17) - ‘Turkish people’, ‘the Fujiwara family’ and ‘that family’, respectively - are not non-sentient NPs, but third parties. On this point, these examples do not fit the definition of the prototypical attributive passive. However, they satisfy the second criterion for the plain passive, in that they describe an objectively observed fact without representing any meaning of affectedness.

c o p -Qu o t

204

Next, let us examine examples of the attributive passive used in a sentence from our spoken data that has a negative propositional meaning. We found just seven such examples (just over 20%) of the attributive passive in the spoken data. Two o f them are cited below:

(18)Zentai no

yappari

anoo. denrvokurvoo

tte

no

wa kagir-are-te

Whole gen as.you.know well quantity.of.electricity cmpl nml top limit-PASS-C0NJ ru

karaa,

anmari sonoo, ensyuu nadoni wa tuke-te

be-PRES because not.very well nai tte.yuu no neg cmpl

seminar etc. for top turn.on-coNj get-DESiD-PRES

ga, soo keiyaku-gakari

nml nom

so

morai-tak-u

no

hoo

in.charge.of.contract gen person

no

kiboo-na n

gen

request

des-u

kedo ne. [Dansei 2381: male, 45, University Lecturer]

cop-pres

but

nml

SFP

‘The whole, well, as you know, the quantity of electricity is limited, so the request from the person in charge of the contract is, well, that s/he does not want us to turn on the air-conditioning for things like seminars.’ (19)Asoko

wa datte Supein, Porutogaru ni senryoos-are-te

That.place top but zya

nai desu

Spain

Portugal

ta

by occupy-PASS-coNj be-PAST

ka.

COP-TOP NEG COP-PRES Q

‘But that place was occupied by Spain and Portugal, wasn’t it.’ [Dansei 4001: male, 27, Car manufacturer / mechanic]

205

In the same way as examples from the written data (examples (15), (16) and (17)) both examples (18) and (19) have a negative propositional meaning. They also denote the event objectively, and do not function to indicate that the subject is affected by the event in any way. This is the reason why these examples of the attributive passive in the spoken data, as well, can be in a sentence whose propositional meaning is negative.

In examples (18) and (19), the subjects are both non-sentient NPs. In fact, all five examples of the attributive passive that have a negative propositional meaning in the spoken data have non-sentient subjects. They, therefore, satisfy both criteria for a prototypical plain passive, having a non-sentient subject, and describing an event objectively without any special meaning of affectedness.

Another prominent feature of Figure 11 and Figure 12 above is that the proportion of the passives that hold a negative propositional meaning declines as you go down the vertical axis of the figure. In Figure 11, the syntactic categories of the Japanese passive are lined up in order of indirect passive, semi-direct passive and direct passive, along the line of the vertical axis. In Figure 12, the order of the semantic groups on the vertical axis is the sentient passive, the passive with a latent affectee, the demotional passive, and the attributive passive. We will next introduce the notion of the centrality of the subject of the passive to the event to try to explain these differences in proportions of negative propositional meaning. Our hypothesis concerning this issue is that the lower the degree of

206

the centrality of the subject of the passive to the event is, the more often the propositional meaning of the passive clause is negative. This matter will be discussed in detail in the next section.

5.3

Detailed discussion on the findings

5.3.1 Degrees of centrality of the passive subject to the event

On the basis of the results shown in Figure 11 and Figure 12, we proposed a hypothesis: the lower the degree of the centrality o f the subject of the passive to the event, the more often the propositional meaning is negative. In this section, first, we will test this hypothesis with regard to our syntactic classification of passive.

5.3.1.1

Subject’s degree of centrality to the event in the syntactic classification

In this study, Japanese passive constructions are syntactically classified into three groups: direct passive, semi-direct passive and indirect passive. Recall that a direct passive is syntactically defined here as one that has a corresponding active clause, and whose subject

207

would correspond to a core argument in the active clause (that is a direct object or an indirect object). We define ‘semi-direct passive’, in this study, as one that has a corresponding active clause, and whose subject would be one of the peripheral participants of the corresponding active clause (that is a major oblique case NP, a genitive NP or an object of comparison). An indirect passive is defined here as one whose subject does not correspond to any of the arguments or participants of the active clause. The syntactic classification used in this study is summarised in the following table which also reflects the distribution of each syntactic type accounted for in the data search, as seen in Section 5.2.1.1:

Table 7: Syntactic classification used in this study _

Svntactic groups

D i r e c t p a s s iv e

S e m i- d i

a. S'

re e l p a s s iv e

Taking a general view of all three syntactic categories, there appears to be a scale throughout regarding the passive subject’s degrees of centrality to the event. In a direct passive, since the subject corresponds to the core argument in the active clause, the referent of the subject plays a very important role in the event. In a semi-direct passive, compared with the case of a direct passive, the subject’s degree of centrality to the event is low, as the subject would correspond only to one of the peripheral arguments of the active clause. In the case of the indirect passive, the degree of centrality o f the subject is lowest, as the

208

subject does not correspond to any of the arguments of the active clause.

(20)K o d o m o n o C h ild

k o ro s in s y u u n o

gen

nooka

ni

azuke-rare-te i-ta

k o to

tim e S h in s h u g en fa rm in g .fa m ily d a t leave-PASS-coNjbe-PAST case

no

a r-u

sei

k a . . . [F u jiw a ra

nom

there.is-PRES

b e c a u s e .o f o r

300]

‘It probably because I was once left in a farming family’s care in Shinshu when I was a child (21)... ringu (b o x in g ) rin g

anaunsaa ni

namae o

a n n o u n ce r dat n am e

yomiager-are-ru. [Sawaki614]

acc

read-PASS-PRES

‘...his name was called out by the ring-side announcer.’ (22)Ato.de bakuhatu-s-are-te mo L a te r

komar-u

explode-PA SS-coN j e v e n be.in.trouble-PR E S

no nml

yo. sfp

‘I will be in trouble if (you) explode (emotionally) later.’ [Josei 7971: female, 22, Company employee (office job)]

Examples (20), (21) and (22) are examples of the direct passive, semi-direct passive and indirect passive, respectively. If we compare the three subjects ( T , a boxer and ‘I’) (although they are all elided in the sentences), the degree o f centrality to the event for each subject is clearly different. The subject of example (20) plays the most central part among three, and the subject o f indirect passive, the least central. Now it is evident that as you go more to the right in the table above, the passive subject’s degree of centrality to the event

209

declines.

In Figure 11 in Section 5.2.5.1, given again below, the result shows that the distribution of the passives that appear in a proposition with a negative meaning declines as you go down the vertical axis of the figure: the indirect passive is more likely to be used in a proposition with a negative meaning than the direct passive, while the semi-direct passive is somehow in between.

Figure 11. Propositional meaning 1: in each syntactic group

Combining the results in Figure 11 and the discussion above, we can now confirm the hypothesis fonnulated in the end of Section 5.2.5.2: the lower the degree of the centrality of the subject of the passive to the event, the more likely the passive clause holds a negative

21 0

propositional meaning. However, this does not suggest that whenever the degree of subject’s centrality to the event is high, the passive clause involves a neutral or positive proposition. It simply means that the proposition in a direct passive clause, whose subject’s centrality is the highest of all, is less often negative than that in an indirect or semi-direct passive. In fact, nearly 60% of the direct passive still hold a negative propositional meaning.

The fact that, in Figure 11, the proportion of passives that appear in a proposition with a negative meaning declines as you go down the vertical axis o f the figure is. in fact, compatible with the degree of emotive affectedness in each syntactic category of Japanese passives. The indirect passive is always associated with the meaning of emotive affectedness, the semi-direct passive less often than the indirect passive, and the direct passive even less than the semi-direct passive. As seen in Section 4.4.3, the degree of emotive affectedness varies even within the framework of the semi-direct passive and the direct passive. This issue will be discussed further in Section 5.3.3.

5.3.2 Objective affectedness vs direct affectedness in the semantic categories

In this section, let us first consider whether or not the subject’s degree of centrality to the event is also related to the semantic categories of Japanese passive. The semantic classification used in this study is summarised in the following table in relation to the

211

syntactic classification:

Table 8: Semantic and syntactic classifications used in this study

Semantic groups

Direct passive

attributive passive

demotional passive

Plain passive

with

S cm i-d ir ect p a ssiv e

Indirect passive

Syntactic groups

sentient passive

affectee

Passive of interest

The following examples illustrate each semantic categoiy of Japanese passive: the sentient passive14 (example (23)), the passive with latent affectee15 (example (24)), the demotional passive16 (example (25)), and the attributive passive17 (example (26)):

(23)Sentient passive

14 The sentient passive has a sentient (most likely human) subject, and is used to describe a situation in which the referent of the subject is directly or emotionally affected by the event. A latent affectee is a sentient entity that one can assume is affected by the event in some way, but is not a participant in the passive sentence. 16 As seen in Section 3.2.1.1, the demotional passive is used primarily to marginalise the agent. The agent is, therefore, often eliminated. However, if it is needed, it is marked by ni-yotte. As for the types o f verb that appear in the demotional passive, they are mainly factitive verbs. 17 The attributive passive is used to describe or imply some attribute of the referent of the subject. It does not depict the occurrence or existence of an event at a specific time and place. (See Section 3.2.1.2.) Example: Kono manga wa kodomo-tati niyokuyom-are-ru. ‘This comic is often read by children.’

212

Kore o

keiki

ni

This acc tuming.point as

ta

no

zyuusin-tati

mo kotogotoku

other gen senior, statesmen too entirely

koros-are-ta. [Shiono 426] kill-PASS-PAST ‘With this as a turning point, other senior statesmen, too, were all killed.’ (24)

Passive with latent affectee

Hisasiburi.de

Koosinkan e hait-ta

ga kokimi.no.ii hodo

After.a.long.timeKoshinkan to enter-PAST but satisfactory extent hakaisitukus-are-te

i-ta. [Takano218]

completely.destroy-PASS-coNJ

be-PAST

‘I went to Koshinkan after a long time, and it was completely destroyed in a very satisfactory way.’ (25)

Demotional passive

Referii

ni-yotte kata

Referee by siai

doori no

model as

kaisi no gongu ga

match start

gengong

nom

gen

tyuui

ga

ataer-are-ru

to

sugu.ni

warning nom give-PASS-PRES when immediately

nat-ta. [Sawaki 647] ring-PAST

‘When the usual warning was given by the referee, the gong rang for the beginning of the match.’ (26)

Attributive passive

Nihongo zyaanaru wa sekaizyuu Japanese journal

top

de ur-are-ru

n

da

yo.

all.over.the.world in sell-PASS-PREs nml cop sfp

213

‘You know, the Nihongo Journal is sold all over the world.’ [Josei 9320: female, 31, Company employee (editing)]

These are all syntactically regarded as direct passive. The degree of the centrality of the passive subject to the event, therefore, is higher than that o f a semi-direct passive or indirect passive, since their subjects correspond to core arguments in the corresponding active clauses. However, among the four examples above, the degree of the passive subject’s centrality to the event is considered to be more or less the same. Therefore, our hypothesis (the lower the degree of the centrality of the subject of the passive to the event, the more likely the passive clause is to be used in a proposition with a negative meaning) may not fit the case of the semantic categories.

As seen in Figure 12, however, cited again below, the results show that the proportion of passive clauses that hold a negative propositional meaning declines as you move down the vertical axis of the figure. An alternative way to account for this must be sought.

214

Figure 12. Propositional meaning 2: in each semantic group Sentient (%)

Latent affectee (%) H Negative ^N eutral H Positive

Demotional (%)

Attributive (%)

Average (%) 0%

20%

40%

60%

80%

100%

In an attempt to account for the results shown in Figure 12, let us have a look at factors related to semantic transitivity18 proposed by Hopper & Thompson (1980: 251-253). They state that a semantically transitive event is generally seen as one which ‘an activity is “carried-over” or “transferred” from’ one participant, an agent, to another, a patient, and it consists of a number of elements. They identify the following ten semantic parameters of semantic transitivity:

A. Participants

HIGH 2 or more participants, A and O19.

LOW 1 participant

18 We are not dealing here with syntactic transitivity, or valency. 19 Hopper and Thompson (1980) state that they ‘follow Dixon (1979) in using “A” (for Agent) and “O” (for Object) to refer to the two participants in a two-participant clause’.

215

B. Kinesis C. Aspect D. Punctuality E. Volitionality F. Affirmation G. Mode H. Agency I. Affectedness of O J. Individuation of O

action telic punctual volitional affirmative realis A high in potency O totally affected O highly individuated

non-action atelic non-punctual non-volitional negative irrealis A low in potency O not affected O not-individuated (Hopper and Thompson 1980: 252)

If a clause has more features in the ‘high’ column in the table above, it is considered to be more transitive semantically, and it is more akin to an archetypal transitive clause.

In terms of semantic transitivity, examples (23) and (24) - an example o f the sentient passive and the one of the passive with a latent affectee, respectively - are much more highly transitive than examples (25) and (26) above, which involve a demotional passive and an attributive passive respectively. Examples (23) and (24) both describe a volitional and affirmative action, and the action is portrayed as a completed and real event. The passive subjects of examples (23), other senior statesman, and (24), Koshinkan, are totally and directly affected as ‘they were all killed’ in (23) and ‘it was completely destroyed’ in (24). In contrast, examples (25) and (26) are less transitive semantically than examples (23) and (24). Although they both express affirmative, volitional action, the events are not viewed as completed, and the passive subjects are not really affected in Hopper and Thompson’s sense. They are only objectively affected.

In comparison with the two subgroups of the plain passive - demotional passive

216

and attributive passive —the subcategories of the passive of interest —the sentient passive and the passive with a latent affectee - in fact occur in semantically more transitive clauses more often. This is evident even by simply contrasting the verbs that occur in the data in these types of passive. Compare the following lists:

(A) Verbs that appear in the sentient passive: hooridasu ‘throw out’, odokasu ‘threaten’, otosu ‘drop’, ayasimu ‘doubt’, uttaeru ' sue’, yament ‘quit’, talakiokosu ‘knock up’, koogisuru ‘protest’, donaru ‘shout’, korosu ‘kill’, taihosuru ‘arrest’, osou ‘attack’, kyuudansuru ‘denounce’, sikaru ‘scold’, osieru ‘teach’, taberu ‘eat’, izimeru ‘bully’, umeru ‘bury’, toraeru ‘capture’, attoosuru ‘overwhelm’, sasu ‘stab’, utu ‘hit’, mini ‘see',yobu ‘call’, nikumu ‘hate’, sukuu ‘save’, homeru ‘praise’, osiyam ‘push away’, hum ‘ditch’, sutem ‘throw away’, sokubakusuru ‘restrain’, yookyuusum ‘demand’, nagum ‘strike’, batoosum ‘abuse’, sasou ‘invite’,yuukaisum ‘kidnap’, aisum ‘love’, nagetobasu ‘fling away’, hoohukusum ‘retaliate’, turesam ‘take away’, mamom ‘protect’, kirau ‘dislike’, ositubusu ‘crush’, etc.

(B) Verbs used in the passive with a latent affectee: hakaisitukusu ‘completely destroy’, kyuusyuusum ‘suck in’, suiagem ‘suck up’, ubaisam ‘take away’, hikiyosem ‘pull something near’, tukikuzusu ‘break’, arasu ‘ruin’, bakumsum ‘reveal’, utu ‘hit’, ositukem ‘press something against’, hurimawasu ‘swing something about’, kim ‘cut’, simetukem ‘squeeze’, sarasu

217

‘expose’, sawagitateru ‘make a great fuss’, etc.

(C) Verbs used in the demotional passive: kaku ‘write’, hiraku ‘open’, nootosuru ‘note down’, kabaasuru ‘cover’, sekkeisuru ‘plan’, teisyutusuru ‘submit’, ataeru ‘give’, mookeru ‘provide’, kataru ‘talk’, kokoromiru ‘try’, tukuriageru ‘make up’, gooseisuru ‘compose’, kaisisuru ‘begin’, insatusuru ‘print’, hosyoosuru ‘guarantee’, sakuseisuru ‘make out’, noonyuusum ‘supply’, sitekisuru ‘point out’, kaisyuusuru ‘collect’, etc.

(D) Verbs used in the attributive passive: kakomu ‘enclose’, iu ‘say’, yobu ‘call’, tutumu ‘veil’, hasamu ‘put something between’, komeni ‘contain’, kau ‘keep’, kagiru ‘limit’, kuwaeru ‘add’, situkeru ‘discipline’, hakensuru ‘dispatch’, miru ‘see’, hootisuru ‘neglect’, katuyoosuru ‘utilise’, hukumu ‘include’, okonau ‘carry out’, omou ‘think’, tukau ‘use’, oou ‘overlay’, kazoeru ‘count’, etc.

Most of the verbs listed in (A) and (B), ones used in the sentient passive and the passive with a latent affectee, are action verbs, and the majority o f them denote an event that makes a profound impact on the Undergoer (or the passive subject), such as hooridasu ‘throw out’, tatakiokosu ‘knock up’, korosu ‘kill’, nagetobasu ‘fling away’, ositubusu ‘crush’ in list (A), and hakaisitukusu ‘completely destroy’, suiageru ‘suck up’, tukikuzusu ‘break’, kiru ‘cut’, simetukeru ‘squeeze’ in list (B). It must be acknowledged that some verbs that appear, such

218

as ayasimu ‘doubt’ and mint ‘see’ in list A, and bakurosum ‘reveal’ and sawagitatem ‘make a great fuss’ in list B, are in fact very low in transitivity. However, verbs of this kind are quite clearly in the minority in the data examined.

Verbs that appear in the demotional passive, as discussed in Section 3.2.1.1, are mainly factitive verbs: a type of verb that denotes an event in which the direct object NP, or the subject of the passive, undergoes a specific change (of position, possession or condition). To be more specific, they are ones that designate physical / psychological effects on a patient or abstract and neutral relationships, or they are verbs o f creation. Many o f them are NP-suru verbs of Sino-Japanese or Western origin. The verbs in list (C) are all, in fact, regarded as factitive. They depict a specific change in the Undergoer, or the passive subject. However, although some kind of change is involved, the affect on the passive subject is generally not as drastic as in the sentient passive or the passive with a latent affectee.

The attributive passive is not restricted to factitive verbs. Even though the majority of the verbs, listed in (D), denote an action, they are ones that would only make a small impact on the Undergoer, the passive subject. The primary function o f the attributive passive is to describe or imply some attribute of the referent of the subject by the rest o f the elements of the sentence. This type o f passive is also often used to describe a scene. Moreover, Song (1993: 109) suggests that an attributive passive is semantically static, and it does not depict the occurrence or existence o f an event at a specific time and place. It is,

219

therefore, expected that the attributive passive generally appears in an even less transitive clause than the demotional passive does.

Now let us have a look at Figure 12 again. Semantic groups are lined up on the vertical axis in order of the sentient passive, the passive with a latent affectee, the demotional passive, and the attributive passive. The sentient passive, thus, occurs the most often in a negative proposition, and the attributive passive the least often. We have also observed above that the demotional passive and the attributive passive appear more often in a less transitive clause, compared to the sentient passive and the passive with a latent affectee. It is now apparent that the more semantically transitive the passive clause is, the more frequently it occurs in a proposition with a negative meaning.

Another point noticed in Figure 12 is that there is a big gap between the proportion of negative propositional meanings for the demotional passive and for the passive with a latent affectee. It is quite natural for the plain passive - the demotional passive and the attributive passive - to mainly occur in a neutral proposition. It is the nature of the plain passive to describe a situation or event objectively with only the meaning of objective affectedness of the Undergoer, the passive subject. It is now evident, therefore, that the two semantic groups, the plain passive and the passive of interest, have to be treated separately in terms of the distribution of propositional meaning.

This study acknowledges three types of affectedness in Japanese passive

220

constructions: emotive affectedness, direct / physical affectedness, and objective affectedness. It seems that the degree of each type of affectedness is also influenced by semantic transitivity. We will, therefore, discuss this issue in detail in the next section.

5.3.3 Three types of Affectedness in Japanese passive constructions

This study considers that the primary function o f passives in Japanese is to portray an event or situation from the point of view of an affected entity. All Japanese passives, therefore, convey the meaning of affectedness in some sense, although the degree and nature of the affectedness varies considerably. In this study, we recognise three types of affectedness in passives in Japanese: emotive affectedness, direct / physical affectedness, and objective affectedness.

As seen in Chapter 4, the special meaning of emotive affectedness is often referred to as ‘adversative’ meaning, and has drawn attention from many researchers. This type of affectedness accompanies only the passive of interest in Japanese. Within the semantic group of the passive of interest, if a passive is syntactically classified as an indirect passive, it is always associated with this emotive nuance. In contrast, some semi-direct passive and considerable proportion of the direct passive do not convey the emotive nuance. In the case of these two types o f passive, the semantic transitivity of the clause is highly related to the degree of the emotive affectedness. More specifically, if the

221

passive subject is not central to the event and / or directly affected by the event, the passive sentence has a special emotive undertone. This is illustrated in Table 9 below:

Table 9: Emotive affectedness

Semidirect passive

Syntactic groups

Semantic groups

attributive demotional passive passive Plain passive

w ith la te n t a ff e c te e

n d ir e c t p a s s iv e

Emotive affectedness: low high ---------------------------------------------------------------► Direct passive

sentient passive Passive of interest

The second type of affectedness, direct / physical affectedness, is detected mainly in passives that are syntactically classified as direct and semantically categorised as the passive of interest, as indicated in Table 10 below:

Table 10: Passives with direct affectedness

Emotive affectedness: low

high

Semidirect passive

Semantic groups

attributive demotional passive passive Plain passive

iaU’Ht

sentient passive Passive of inkiest

Direct affectedness: high • --------------------------------

high ----- ►

Semantic groups

attributive demotional passive passive Plain passive

;

n d ir c c t p a s s iv e

Semidirect passive

Syntactic groups sentient passive Passive of interest

---------------------------------------------L _____________________ i____________________________

Direct affectedness: high ^ ______________________________________ low Semantic transitivity: high M ---------------------------------------------------------- low

In the plain passive, a non-sentient entity normally appears in the subject position, and the situation is portrayed objectively. However, it still is a fact that something has been done to the Undergoer. In this sense, the Undergoer is affected in a businesslike manner. This type of affectedness is referred to as ‘objective affectedness’ in this study.

As discussed in the last section, the plain passive, which is accompanied by the objective affectedness, is semantically lower in transitivity than the passive o f interest. The verbs usually used in the plain passive do not impact on the Undergoer as drastically as the verbs that most commonly appear in the passive o f interest. It thus can be said that objective affectedness is considerably weaker than direct or emotive affectedness, both of which are associated with the passive of interest.

In this chapter, the findings o f the data analysis are discussed in detail, in order to reveal how Japanese passives are actually used in real contexts. We first discussed the

L

225

proportion of each syntactic and semantic category of passive in the data, the nature o f the ‘actor’ and the subject, the occurrence in noun-modifying clauses, and the kind of propositional meanings that occur.

Some findings confirmed the claims made in previous research, such as the high frequency of passives without an overt ‘actor’, and of passives with a negative propositional meaning. Other findings contradicted previous claims, such as the large proportion of passives with a non-sentient subject. Another striking finding was the predominant proportion of the direct passive (almost 90% o f the data). In contrast, the frequency of the occurrence of the indirect passive, the major focus of previous studies of Japanese passives, was, in fact, very low (1.2% of the data).

The second section o f this chapter dealt mainly with the issue of the degree of centrality of the passive subject to the event. We observed that, with regard to our syntactic classification of passive, the lower the degree of the centrality of the subject of the passive to the event, the more likely it is that the passive clause holds a negative propositional meaning. In terms of the semantic categorisation of the Japanese passive, we found that it was the type of affectedness - emotive affectedness, direct / physical affectedness, or objective affectedness - that is highly relevant to the propositional meaning of the passive clause.

226

Chapter 6

Conclusion

The passive is used primarily to depict the event from the point of view of the Undergoer. Givon (1990: 566) states that ‘the notion of voice is fundamentally pragmatic’; in a transitive clause, a semantically identical event can be portrayed from several different perspectives using ‘the very same verb, agent and patient’. The passive voice is one of these perspectives.

In the case of Japanese passive, in particular, this study maintains that it is the perspective o f the affected entity from which the event is portrayed. Although the degree and nature of the affectedness varies, all Japanese passives convey the meaning of affectedness in some sense. Furthermore, this study recognises three types of affectedness in Japanese passive constructions: emotive affectedness, direct / physical affectedness, and objective affectedness.

6.1

Aims and findings of this study

This study set out three main aims and eight more specific research questions related to these three aims. The aims and related questions of this research were as follows: (i)

Firstly, the study aimed to clarify and reclassify the wide range of

227

functions of passives in Japanese. (a) What specific syntactic features and semantic functions do Japanese passives have? (b) What broad types can these features and functions be classified into? (c) What sorts of correlations exist between the syntactic and the semantic distinctions? (ii)

Secondly, the study aimed to tackle and solve the well-known issue of the 'adversative meaning’, referred to in this research as the ‘special meaning of emotive affectedness’, which accompanies only some types of passive in Japanese. (a) With what types of passive does the special emotive nuance appear? (b) Under what circumstances does the special emotive nuance occur?

(iii)

Finally, the study aimed to reveal how the multi-functioned Japanese passives are actually used in real contexts. (a) What types of passive are used most in Japanese and why are these more common than other types? (b) What are the characteristics of the participants in Japanese passive constructions? (c) What types of proposition do Japanese people express using the passive voice?

The final aim (aim (iii)) was the ultimate goal for this research. The first and the

228

second aims should actually be seen as parts of the final one. The investigation started by achieving the first aim: to clarify and categorise the functions of Japanese passives. This was carried out by reclassifying Japanese passives according to their syntactic features and semantic functions, and by reexamining the relationships between them (related to the research questions (i-a) to (i-c)).

In relation to syntactic features (questions (i-a) and (i-b)), Japanese passives were divided into three groups: the direct passive, the semi-direct passive and the indirect passive, depending on to which argument in the active counterpart the passive subject corresponds. If the subject corresponds to a core argument in the active clause, the passive is classified as the direct passive. In a semi-direct passive, on the other hand, the subject would correspond to a peripheral participant in the active clause. The subject in an indirect passive does not correspond to any of the arguments in the active clause, core or peripheral.

The semi-direct passive is a new category. This type has been included in the indirect passive in previous research. This distinction between the indirect and semi-direct passive, in this study, has been found very helpful when it comes to teasing out the issue of the special emotive nuance.

A wide range of semantic functions of Japanese passives were recognised (question (i-a)): depicting a situation in which the subject is somehow affected by the event, defocusing the ‘actor’, describing the subject’s attribute, etc. These functions were first

229

classified into two basic groups, the plain passive and the passive of interest (question (i-b)). The plain passive is one that generally has a non-sentient subject and describes an event objectively. Nevertheless, it portrays the situation as something that is done to the referent of the subject, and therefore implies the nuance of affectedness. This type of affectedness is called ‘objective affectedness’ in this study. The plain passive is further divided into two types: the attributive passive and the demotional passive. The attributive passive is used to describe some attribute of the subject by the rest of the sentence. The primary function of the demotional passive is to marginalise the ‘actor’. It is a comparatively new type of passive in the Japanese language, whose usage has increased during the last century.

In contrast to the plain passive, the second broad group, the passive of interest, is one that portrays an event in terms of the concerns of the referent o f the subject. The referent of the subject is directly and / or emotionally affected by the event. In the majority of cases, therefore, the referent of the subject is sentient, and most likely human. The passive of interest is categorised into two subtypes: the sentient passive and the passive with a latent affectee. The sentient passive is used to describe a situation in which the subject is somehow affected by the event. The second type, the passive with a latent affectee, has a non-sentient subject, and yet it delineates the direct affectedness of the subject. This is the category in which one can assume a latent affectee1.

In terms of correlations between the syntactic and the semantic classifications 1 A latent affectee is a sentient entity that one can assume is affected by the event denoted by the passive clause in some way, but is not actually a participant of the passive sentence.

230

(question (i-c)), this study argued against the previous view of the straightforward correlation between the syntactic and semantic distinctions - more specifically the correspondence between the indirect passive and the adversative passive on the one hand, and that between the direct passive and the neutral passive on the other. This is because some direct passives have the special emotive nuance while, at the same time, some tokens of what has previously been called the ‘indirect passive’2 do not have the emotive reading. While it is true that the indirect passive - according to the definition used in this study - is accompanied by the special emotive nuance, the correlation in the case of the other syntactic types is much more complicated. This issue relates to the second aim of this study. We, therefore, set out to investigate this issue to tackle and solve the question of the ‘special emotive affectedness’, or the ‘adversative meaning’, next.

In order to achieve the second aim, the present research first examined passives with an emotive undertone within the framework of each syntactic category of Japanese passive: the indirect passive, the semi-direct passive and the direct passive (research question (ii-a)). It then investigated in detail where the special meaning of emotive affectedness actually came from (question (ii-b)).

The study observed that the controversial cases of ‘indirect passives’ that do not have the special emotive nuance are actually better categorised as semi-direct passives. More specifically, the cases concerned all belong to the so-called ‘possessor passive’, 2 According to the definition used in this study, these are actually categorised as semi-direct passives rather than indirect passives.

231

whose subject corresponds to the genitive case or ‘possessor’ NP in the active counterpart. Not all instances of the ‘possessor passive’ have the special emotive undertone. Modifying Shibatani’s (1996: 7) and Teramura’s (1982: 244-245) analyses, this study found that if two criteria are satisfied, then the sentence is not likely to have the special meaning of emotive affectedness. These two criteria are: (i) the o-NP in the passive - the ‘possessed’ NP - is high in proximity to the possessor, and therefore the ‘possessor’ NP, or the subject of the passive, is central to the event; and (ii) the event has a significant impact on the subject. The reason why the sentence is not accompanied by the special meaning of emotive affectedness when these criteria are satisfied is that the subject is highly relevant to the event, and therefore it does not require any supplementary semantic input to increase its relevance to the event. On the contrary, if the ‘possessed’ NP is low in proximity to the possessor, and thus the ‘possessor’ NP is not central to the event, and if the event denoted by the verb does not have a strong impact on the subject, then the sentence is more likely to be associated with the special emotive undertone.

In the case of direct passives that have the special emotive nuance, the present study noted that it is the degree of the centrality of the subject and the impact o f the event on the subject, again, that is the crucial factor. The lower the degree of the centrality of the subject or the impact of the event on the subject is, the stronger the special emotive undertone is perceived to be.

Finally, the investigation for the ultimate goal o f the present study was performed.

232

Aim (iii) was to reveal how the multi-functioned Japanese passives are actually used in real contexts. This aim had already been partly achieved in a sense, since aims (i) and (ii) are considered to be parts of aim (iii). Further examinations were carried out in relation to research questions (iii-a), (iii-b) and (iii-c).

In terms of the proportion of each category o f passive in the data (question (iii-a)), the most striking finding was the predominant proportion of the direct passive (almost 90% of the data). In contrast, the frequency of the occurrence of the indirect passive, the major focus o f previous study of Japanese passives, was, in fact, very low (1.2% of the data). Even if examples o f the semi-direct passive are taken into account as well, as has been done in previous research, the proportion would still be only 3.4%. The results show that the case of indirect passive can hardly be taken to represent Japanese passive constructions in general.

The main findings on the characteristics of the participants of the passive (question (iii-b)) are related to the ‘actor’ and the passive subject. The first finding on the ‘actor’ confirmed the view in previous research: an ‘actor’ argument is often elided in a passive sentence. This study observed that more than 80% of the passive sentences in the data do not involve an overt ‘actor’ in the sentence.

The most prominent finding on the passive subject was the substantial number of non-sentient subjects (almost 40% of the data). Types of passive that have a non-sentient

233

subject are the attributive passive, the demotional passive, and the passive with a latent affectee. The demotional passive, in particular, takes up more than 20% of the data gathered for this study, even though it is said not to be inherent to Japanese and to have come into use with the ‘actor’ marker ni-yolte only in the 19th century. It is apparent from this study that its usage has dramatically increased over the course of the 20th century.

Regarding the propositional meaning in a passive clause (research question (iii-c)), this research obtained the result that, although more than half of the propositional meanings in the passive are ‘negative’ (about 55%), there still are considerable number of passives that have a neutral or positive propositional meaning.

Another significant finding on the propositional meaning was that it seemed to be related to the degree of centrality of the passive subject to the event, that is how central the role played by the referent of the subject is to the event. We observed that, with regard to our syntactic classification of passive, the lower the degree of the centrality of the subject of the passive to the event, the more likely it is that the passive clause holds a negative propositional meaning. Consequently, the propositional meaning of a direct passive clause, whose subject’s centrality is the highest o f all, is less often negative than that of an indirect or semi-direct passive. The subject o f a semi-direct passive, o f course, is far less centrally involved in the event than that of a direct passive clause, and the subject of an indirect passive is not involved at all.

234

In terms of the semantic categorisation o f the Japanese passive, it is the type of affectedness that is highly relevant to the propositional meaning of the passive clause. The passive of interest, which is associated with the special meaning of emotive affectedness or direct / physical affectedness, appears mainly in propositions with a negative meaning. In contrast, the plain passive, which involves objective affectedness, mostly holds a neutral propositional meaning.

6.2

Limitations of this study and recommendations for further research

All the eight research questions specified at the beginning of this study have been addressed in the course of this study. However, there are a number of important limitations of this research project, and they must be recognised.

First of all, all three aims set for this research were achieved only with respect to the genres examined in this study, that is novels and workplace conversations. It is therefore essential to investigate and compare with other genres, such as newspapers, scientific writing, casual conversation, and so on.

The amount of data, in particular the spoken data, was also somewhat limited. A larger scale of investigation would add weight to the claims made in this study.

235

The present study has mostly been done principally at the level of the sentence rather than the level of the text. However, in investigating constructions that express the attitude of the speaker, portray an event in terms of the concern of a participant, or imply a special emotive nuance, it is essential to look beyond the sentence to the wider context.

In acknowledging the limitations of the current study, the most important recommendation for the further studies on the Japanese passive is that comparative studies be undertaken with wide range of languages, especially with some other Asian languages, such as Chinese and Vietnamese that are also said to have an ‘adversative passive’ construction. It is also essential to identify the similarities and differences between the passive and other, functionally related constructions in Japanese, such as the ‘-te shimau’ construction, the ‘-te aru’ construction, and the unaccusative intransitive construction.

236

REFERENCES

Amano, Midori. 2001. Museibutsu shugo no ni-judobun - imiteki kankei no sootei ga hitsuyona bun (Ni-passive with an inanimate subject; the sentences that needs assumption o f semantic connection). Kokugogaku (Japanese Philology) 52:1-15. Brinker, K. 1971. Das Passiv in Heutigen Deutsch. Form undFunktion. Munchen: Heuber Schwann. Brown, Gillian , and Yule, George. 1983. Discourse Analysis. London: Cambridge University Press. Chomsky, Norm. 1957. Syntactic Structures. The Hague: Mouton. Chomsky, Norm. 1965. Aspects o f the Theory o f Syntax-. MIT Press. Comrie, Bernard. 1977. In defense of spontaneous demotion: the impersonal passive. In Syntax and Semantics Volume 8 Grammatical Relations, ed. Peter Cole & Jerrold M. Sadock, 47-58. New York: Academic Press. Comrie, Bernard. 1981. Language Universals and Linguistics Typology. Oxford: Basil Blackwell. Ding, Yi-Xiang. 1997. Kansetsu ukemi ni kansuru ichi-kosatsu (A study on the indirect passive in Japanese). Nihongogaku (Japanese Linguistics) 93. Dixon, R. M. W. 1979. Ergativity. Language 55:59-138. Dixon, R. M. W. 1994. Ergativity. London: Cambridge University Press. Dowty, David. 1979. Word Meaning and Montague Grammar. Dordrecht: Reidel. Dowty, David. 1991. Thematic proto-roles and argument selection. Language 67:547-619. Duranti, Alessandro. 1997. Linguistic Anthropology. Cambridge: Cambridge University Press. Duscova, L. 1971. The passive voice in English and Czeck. Philologica Pragensia 15:93-118. Foley, William A., and Jr., Robert D. Van Valin. 1984. Functional Syntax and Universal Grammar. Cambridge, London, New York, New Rochelle, Melbourne, Sydney: Cambridge University Press. Givon, Talmy. 1979. On Understanding Grammar. New York, San Francisco, London: Academic Press. Givon, Talmy. 1981. Typology and functional domains. Studies in Languages 5:163-193. Givon, Talmy. 1984. Direct object and dative shifting: semantic and pragmatic case. In Objects, ed. Frans Plank (ed.), 151-182. New York: Academic Press. Givon, Talmy. 1984. Syntax: A Functional-Typological Introduction Vol. /.vol. I. Amsterdam: John Benjamins. Givon, Talmy. 1990. Syntax: A Functional-Typological Introduction Vol. II. Amsterdam, Philadelphia: John Benjamins Piblishing Company. Givon, Talmy. 1995. Functionalism and Grammar. Amsterdam, Philadelphia: John Benjamins Piblishing Company. Givon, Talmy. 2001. Syntax An Introduction Vol. I & //.vol. I & II. Amsterdam/Philadelphia: John Benjamins. Hashimoto, Shinkichi. 1931. Judooshi no kenkyuu (A Study of auxiliary verbs). Reprinted

237

in 1969. In Joshi Jodooshi no Kenkyuu (A Study o f Particles and Auxiliary Verbs), 266-292. Hopper, Paul J. and Sandra A. Thompson. 1980. Transitivity in grammar and discourse. Language 56:251 -299. Hoshi, Hiroto. 1999. 7. Passives. In The Handbook o f Japanese Linguistics, ed. Natsuko Tsujimura, 191-235. Malden & Oxford: Blackwell Publishers. Hosoe, Itsuki. 1928. Waga kokugo no dooshi no soo o ronji, dooshi no katsuyookeishiki no bunki suru ni itarishi genri no ittan ni oyobu (A study on Japanese voice - in view of revealing the principle of divergence of Japanese verb conjugation). In Okakura Sensei Kinen Ronbunshuu (Essays in celebration o f Professor Okakura's 60th Birthday)'. Okakura Sensei Kanreki Shukugakai. Howard, Irwin and Agnes M. Niyekawa-Howard. 1976. Passivization. In Syntax and Semantics Vol. 5: Japanese Generative Grammar, ed. Masayoshi Shibatani. New York, San Francisco, London: Academic Press. Huddleston, R. D. 1971. The Sentence in Written English. London: Cambridge University Press. Ikegami, Yoshihiko. 1975. Imiron (Semantics). Tokyo: Taishukan. Inoue, Kazuko. 1976. Henkei Bunpo to Nihongo (Transformational Grammar and Japanese). Tokyo: Taishukan. Iwashita, Mami. 1996. Verb type and Japanese passive constructions. Paper presented at 20th Anniversary Conference, ASAA., Melbourne, Australia. Iwashita, Mami. 1997. Animacy and passivisation in Japanese. Paper presented at 1997 JSAA Conference., Melbourne, Australia. Jacobsen, W. M. 1992. The Transitive Structure o f Events in Japanese.vo\. 1: Studies in Japanese Linguistics. Tokyo: Kuroshio Publishers. Jarkey, Nerida. 1999. Case marking o f objects of stative predicates in Japanese. Australian Journal o f Linguistics 19:197-224. Kageyama, Taro. 1995. Bunpo to Go-keisei. (Grammar and Word Formation) '. Hitsuji Publishing. Keenan, Edward L. 1975. Some universal of passive in relational grammar [18-20 April]. Papers from the eleventh regional meeting Chicago Linguistic Society:340-352. Keenan, Edward L. and Bernard Comrie. 1977. Noun Phrase Accessibility and Universal Grammar. Lingustic Inquiry 8:63-99. Kinsui, Satoshi. 1992. Bamen to shiten - ukemibun o chushin ni - (Scene and perspective focusing on passive sentences). Nihongogaku 11:12-19. Kinsui, Satoshi. 1997. The influence of translation on the historical development of the Japanese passive construction. Journal o f Pragmatics 28:759-779. Klaiman, M. K. 1981. Toward a universal semantics of indirect subject. Proceedings o f the seventh annual meeting o f Berkeley Linguistics Society February 1981:123-135. Klaiman, Miriam H. 1987. Aktionsart, semantics, and function in the Japanese "passive". Studies in Language 11:401434. Klaiman, M. K. 1988. Affectedness and control: a typology of voice systems. In Passive and Voice, ed. Masayoshi Shibatani. Amsterdam, Philadelphia: John Benjamins Piblishing Company.

238

Koizumi, Tamotsu ed. 1993. Nihongo Kihon Doshi Jiten (A Dictionary o f the Usage o f Basic Japanese Verbs)-. Taishukan. Krauthamer, H. 1981. The prediction of passive occurrence. Linguistics 19:307-324. Kuno, Susumu. 1973. The Structure o f the Japanese Language. Cambridge, Massachusetts and London, England: The MIT Press. Kuno, Susumu. 1977. Nihongo no shugo no tokushusei (Peculiarity of the Japanese subject). Gekkan "Gengo" (Monthly "Language") 6:11-18. Kuno, Susumu, and Kaburaki, Etsuko. 1977. Empathy and Syntax. Linguistic Inquiry 8:627-672. Kuno, Susumu. 1982. Shin Nihon Bunpo Kenkyu (New Research in Japanese Grammar). Tokyo: Taishukan. Kuno, Susumu. 1987. Functional Syntax. Anaphora, Discourse and Empathy. Chicago and London: The University of Chicago Press. Kuno, Susumu. 1990. Passivization and thematization. In On Japanese and How to Teach It. In Honor o f Seiichi Makino, ed. Osamu Kamada & Wesley M. Jacobsen, 43-66: The Japan Times. Kuroda, S-Y. 1965. Generative Grammatical Studies in The Japanese Language, MIT: Ph. D. Lakoff, Robin. 1971. Passive resistence. Chicago Linguistic Society VTI: 149-162. Langacker, Ronald and Pamela Munro. 1975. Passives and their meaning. Language 51:789-830. Li, Charles N. ed. 1976. Subject and Topic. New York, San Francisco, London: Academic Press. Lyons, John. 1977. Semantics Volumes I & 2.vol. 1 & 2. Cambridge: Cambridge University Press. Masuoka, Takashi. 1982. Nihongo Judoobun no imi bunseki (An analysis of the meaning of Japanese passives). Gengo Kenkyuu (Studies in Language) 82:48-64. Masuoka, Takashi. 1987. Meidai no Bunpo (Grammar o f Proposition). Tokyo: Kuroshio Publishing. Masuoka, Takashi. 1991. Modality no Bunpo (Grammar o f Modality). Tokyo: Kuroshio Publishing. Masuoka, Takashi. 2000. Nihongo Bunpo no Shoso (Aspects o f Japanese Grammar). Tokyo: Kuroshio. Matsushita, Daizaburo. 1930. Hyojun Nihon Koogohoo (Standard Japanese Spoken Language). Reprinted in 1977. Tokyo: Chubunkan Shoten. [Reprinted in 1961 Hakuteisha], McCawley, Noriko. 1972. On the treatment of Japanese passives. In Papers from the Eighth Regional Meeting o f the Chicago Linguistic Society, ed. P. M. Peramteau, J. N. Levi and G. C. Phares, 256-270. Chicago: University of Chicago Department of Lnguistics. McEnery, Tony & Andrew Wilson. 1996. Corpus Linguistics. Edinburgh: Edinburgh University Press. Mikami, Akira. 1953. Gendai Gohoo Josetsu (Preface to Modern Grammar). Reprinted in 1972. Tokyo: Kuroshio Publishers.

239

Mitsuya, Shigematsu. 1908. Koto Nihon Bunpo (High-class Japanese Grammar): Meiji Shoin. Miyagawa, Shigeru. 1989. Syntax and Semantics 22: Structure and Case Marking in Japanesse. New York: Academic Press. Motoori, Haruniwa. 1828. Kotoha no Kayoimichi (The Ways o f Language). Nitta, Yoshio. 1991. Voice teki hyogen to jikoseigyosei. In Nihongo no Voice to Tadosei, ed. Yoshio Nitta. Tokyo: Kuroshio. Nomura, Tsuyoshi. 1995. Jido tado ukemidoshi ni tsuite. In Doshi no Jita, eds. K. Suga and E. Hayatsu: Hitsuji Publishing. Okutsu, Keiichiro. 1992. Nihongo no ukemibun to shiten. Nihongogaku 11:4-11. Perlmutter, David M., and Postal, Paul M. 1974. Lectures on Relational Grammar. Paper presented at Summer Linguistic Institute o f the Linguistic Society o f America, Amherst. Perlmutter, David M., and Postal, Paul M. 1977. Toward a universal characterization of passivization [19-21 Feb], Proceedings o f the 3rd annual meeting o f Berkeley Linguistics Society394-4\7. Sagawa. 1991. GB-riron ni okeru nookakusei ni tsuite (A study on ergativity in GB theory). Hosei University Bulletin 36. Schoenthal, G. 1976. Das Passiv in Der Deutchen Standardsprache. Munchen: Max Hueber Verlag. Shibatani, Masayoshi ed. 1976. Syntax and Semantics: The Grammar o f Causative Constructions, vol. 6. Syntax and Semantics. New York, San Francisco, Philadelphia: Academic Press. Shibatani, Masayoshi. 1978. Nihongo no Bunseki. Tokyo: Taishukan. Shibatani, Masayoshi. 1985. Passives and related constructions: a prototype analysis [Dec]. Language 61:821-848. Shibatani, Masayoshi ed. 1988. Passive and Voice. Typological Studies in Language (TSL). Amsterdam, Philadelphia: John Benjamins Publishing Company. Shibatani, Masayoshi. 1990. The Languages o f Japan. London: Cambridge University Press. Shibatani, Masayoshi. 1994. An integrational approach to possessor raising, ethical datives and adversative passives. Paper presented at Proceedings o f the 20th Annual Meeting o f the Berkeley Linguistics Society. Shibatani, M. 1995. The Languages o f Japan.: Cambridge University Press. Shibatani, Masayoshi. 1996. Meiwaku ukemi no imiron (Semantics of the adversative passive). In Watanabe Minoru Koki Kinen Ronshu. Shibatani, Masayoshi. 1998. Voice parameters. In Typology o f Verbal Categories, ed. Leonid Kulikov and Heinz Vater, 117-138. Tubingen: Max Niemeyer Verlag. Shibatani, Masayoshi. 2000. Voice. In Bun no Kokkaku, ed. Shinjiro Muraki Yoshio Nitta, Masayoshi Shibatani and Masato Yazawa. Tokyo: Iwanami Publishing. Siewierska, Anna. 1984. The Passive. A comparative Linguistic Analysis. Sydney: Croom Helm. Silverstein, M. 1986. Features and Projections. In Hierarchy o f features and Ergativity, eds. Muysken and van Riemsdijk: Foris.

240

Song, Nam Sun. 1993. Thematic Relations and Transitivity in English, Japanese and Korean. : Center for Korean Studies, University of Hawai'i. Takami, Kenichi. 1995. Kinooteki Koobunron niyoru Nichieigo Hikaku (Functional Syntax: A Comparative Study o f Japanese and English). Tokyo: Kuroshio Publishers. Teramura, Hideo. 1982. Nihongo no Shintakusu to Imi (Syntax and Semantics in Japanese) Vol. I & //.vol. 1. Tokyo: Kuroshio Publishers. Tokunaga, Misato. 1992. Dichotomy in the structures of honorifics of Japanese. Pragmatics 2:127-140. Tsunoda, Tasaku. 1985. Remarks on transitivity. Journal o f Linguistics 21:385-396. Tusboi, Eijiro. 1997. Cognitive models in transitive construal in the Japanese adversative passive. In Constructions in Cognitive Linguistics, eds. Ad Foolen and Frederic Van Der Leek: Joh Benjamins. Vendler, Zeno. 1967. Linguistics in Philosophy. Ithaca: Cornell University Press. Wierzbicka, Anna. 1979. Are grammatical categories vague or polysemous (Japanese "adversative" passive in a typological context)? Papers in Linguistics 12:111-162. Wierzbicka, Anna. 1988. The Semantics o f Grammar. Amsterdam: John Benjamins. Yamada, Yoshio. 1908. Nihon Bunpooron (Theory o f Japanese Grammar). Tokyo: Hobunkan. Yamada, Yoshio. 1922. Nihon Bunpoo Koogi (Lectures on Japanese Grammar). Tokyo: Hobunkan. Yamada, Yoshio. 1936. Nihon Bunpogaku Gairon (Introduction to Japanese Grammar). Tokyo: Hobunkan. Yoshida, Tosaku. 1976. Bunpo kenkyu no rekishi (2) (History of Study of Grammar). In Bunpo I, 299-356. Tokyo: Iwanami Publishing. Yoshida, Eri. 1996. A Discourse Study of Passives in Japanese Conversation, East Asian Languages and Cultures, University of California. Yule, Gillian Brown and George. 1983. Discourse Analysis. Cambridge: Cambridge University Press.

Sources of examples Gendai Nihongo Kenkyukai (Society of Contemporary Japanese Language Studies). (1999). Joseino Kotoba - Shokubahen (Women's language - Workplace section). (CD-Rom collection of spoken Japanese). Tokyo, Hitsuji Publishing. Gendai Nihongo Kenkyukai (Society of Contemporary Japanese Language Studies). (2002). Dansei no Kotoba - Shokubahen (Men’s language - Workplace section!. (CD-Rom collection of spoken Japanese). Tokyo, Hitsuji Publishing. NEC. (1995). Shincho Bunko no Hvakusatsu (One hundred Shincho Paperbacks). (CDRom collection of Japanese novels). Tokyo, Shincho Publishing. Inoue, Yasushi. Hvoohek/. Tokyo, Shincho Bunko. Natsume, Soseki. Meian. Tokyo, Shincho Bunko.

LINCOM Studies In Asian Linguistics In this series 01 Karen Ebert

A Grammar of Athpare

02 Stefan Georg

Marphatan Thakali

03 Tsutomu Akamatsu

Japanese Phonetics. Theory and Practice

04 D.N.S. Bhat& M.S. Ningomba

Manipuri Grammar

05 Siew-Yue Killingley

Learning to Read Pinyin Romanization and its Equivalent in Wade-Giles: A Practical Course for Students of Chinese

07 Barbara Niederer

Les langues Hmong-Mjen (Miao-Yao). Phonologie historique

12 Duck-Young Lee

Korean Phonology. A Principle-based Approach

22 Margaret Mian Yan

Introduction to Chinese Dialectology

25 Heinrich Werner

Probleme der Wortbildung in den Jenissej-Sprachen

26 Yogendra Yadava

Issues in Maithili Syntax A Government-Binding Approach

27 John Newman & Anand V. Raman

Chinese Historical Phonology

28 Ming Chao Gui

Yunnanese and Kunming Chinese: A Study of the Language Communities, the Phonological Systems, and the Phonological Developments

30 Yavar Dehghani

A Grammar of Iranian Azari Including Comparisons with Persian

32 Xiaonong Sean Zhu

Shanghai Tonetics

33 Nikolai Vakhtin

The Old Sirinek Language

38 Tsutomu Akamatsu

Japanese Phonology. A Functional Approach

39 Ago Kunnap

Contact-induced Perspectives in Uralic Languages

40 Lau Chunfat

The Decline of the General Hakka Accent in Hong Kong

44 Shengli Feng

The Prosodic Syntax of Chinese

45 Ago Kunnap

Main Language Shifts In the Uralic Language Group

46 Zhongmin Chen

Studies on Dialects in the Shanghai Area Their Phonological Systems and Historical Developments

47 Janet Zhiqun Xing

Grammaticalization of Verbs in Mandarin Chinese

48 Hyung-Soon Yim

The Intonational Phonology of Direct and Indirect Imperative Sentence Types in Seoul Korean

49 Choi-Yeung-Chang Flynn

Intonation in Cantonese

50 Andras J. E. Bodrogligeti

An Academic Reference Grammar of Modern Literary Uzbek. Vol. I

51 Andras J. E. Bodrogligeti

An Academic Reference Grammar of Modern Literary Uzbek. Vol. II

52 Noriko Katsuki-Pestemer

Japanese Postpositions: Theory and Practice

53 Hsin-Yun Liu

A Profile of the Mandarin NP

55 Kenichi Takashima & Jiang Shaoyu (eds.)

Meaning and Form: Essays in Pre-Modern Chinese Grammar

56 Ming Chao Gui

The Phonology of Guangzhou Cantonese

57 Wendan Li

Topic Chains in Chinese

58 Paul Sidwell

The Katuic Languages

59 Mine Guven

Adverbials in Turkish

60 Andras J. E. Bodrogligeti

The Ten Letters

62 Picus Sizhi Ding

Studies on Ba Resultative Construction

63 Tatsuya Fukushima

Ga: Japanese Conjunction Its Functions and Sociolinguistic Implications

64 Anvita Abbi

Endangered Languages of the Andaman Islands

65 Mikhail S. Andronov

Brahui, a Dravidian Language

66 Xiaonong Zhu

A Grammar of Shanghai Wu

68 Guido Oebel

Japanische Beitrage zu Kultur und Sprache Studia laponica Wolfgango Viereck emerito oblata

70 Tsung Chin

Sound Systems of Mandarin Chinese and English: a comparison

71 Mami Iwashita

Being Affected: The meanings and functions of Japanese passive Constructions