Interrogative Strategies
An Areal Typology of the Languages of China
Dissertation zur Erlangung des akademischen Grades eines Doktors der Philosophie
vorgelegt von
Tianhua Luo
an der
Geisteswissenschaftliche Sektion Fachbereich Sprachwissenschaft
Tag der mündlichen Prüfung: 23. September 2013
1. Referent: Prof. Dr. Frans Plank 2. Referentin: Prof. Dr. Nicole Dehé
Contents
Acknowledgements... v
Zusammenfassung... vii
Abstract... xi
Notational conventions... xiii
Chapter 1. Introduction ... 1
1.1. The grammar of interrogatives...1
1.1.1. Interrogative forms... 1
1.1.2. Assymetries in form and meaning...11
1.2. Motivation ...16
1.3. Material ...19
1.4. Methodology ...24
1.5. Outline of the work ...28
Chapter 2. A survey of polar interrogative strategies ... 31
2.1. Sino-Tibetan languages ...32
2.1.1. Sinitic languages ... 32
2.1.1.1. Standard Chinese... 32
2.1.1.2. Yongxin Gan... 48
2.1.1.3. Comparative Sinitic... 56
2.1.2. Tibeto-Burman languages ... 71
2.1.2.1. Tibetan languages... 71
2.1.2.2. Yi languages ... 74
2.1.2.3. Jingpo languages ... 88
2.1.2.4. Burmese languages... 93
2.1.2.5. Qiangic languages ... 95
2.1.3. Kam languages ... 105
2.1.4. Hmong-Mien (Miao-Yao) languages ...117
2.2. Altaic languages ...122
2.2.1. Turkic languages ... 122
2.2.2. Mongolian languages ... 128
2.2.3. Manchu-Tungusic languages... 133
2.3. Austronesian languages...136
2.4. Austro-Asiatic languages ...146
2.5. Indo-European ... 154
2.6. Creole languages... 155
2.7. Summary... 157
Chapter 3. Question particles and final particles ... 159
3.1. The position of question particles... 162
3.2. ma ne polar questions in Sinitic languages... 165
3.2.1. ma ne polar questions ... 165
3.2.2. The nature of ma ne polar questions... 172
3.3. Final particles in wh-questions ... 174
3.3.1. Final particles in wh-questions ... 174
3.3.2. Final particles in reduced wh-questions... 177
3.4. Summary... 182
Chapter 4. Disjunctions and alternative questions... 183
4.1. Alternative vs X-neg-X questions... 185
4.2. Alternative questions: the or vs or/or? typology... 188
4.2.1. Introducing the or vs or/or? typology ... 188
4.2.2. The or vs or/or? typology... 189
4.2.3. The position of or and or/or?... 193
4.2.4. The or vs or/or? typology and clause order ... 196
4.3. Particles as disjunctions... 198
4.3.1. Alternative questions through particles ... 198
4.3.2. Patterns of particle disjunctions in alternative questions... 200
4.4. Alternative islands in Sinitic... 202
4.5. Summary... 204
Chapter 5. Wh-phrases and wh-questions ... 205
5.1. The position of wh-phrases... 206
5.2. Wh-fronting in Standard Chinese... 208
5.3. The syntax of wh-questions ... 212
5.3.1. Wh-questions and word order change... 212
5.3.2. Wh-questions with coordination in Sinitic... 216
5.4. The reduplication of wh-phrases... 217
5.4.1. Languages with reduplication in wh-phrases... 218
5.4.2. Which wh-phrases can be reduplicated?... 219
5.4.3. Semantics of reduplicated wh-phrases... 221
5.4.4. Syllable pattern of reduplicated wh-phrases... 223
5.5. Summary... 226
Chapter 6. Three types of verb-related questions ... 227
6.1. Q-VP questions ...228
6.1.1. Q-VP in Sinitic... 228
6.1.2. Q-VP in Tibeto-Burman... 234
6.1.3. Summary ... 239
6.2. Verb-reduplicating questions...241
6.2.1. Verb-reduplication in Sinitic ... 241
6.2.2. Verb-reduplication in Yi ... 250
6.2.3. Verb-reduplication in Hmong-Mien... 252
6.2.4. Summary ... 253
6.3. Interrogative verbs ...255
6.3.1. Interrogative verbs in Sinitic... 255
6.3.2. Interrogative verbs in Formosan ... 257
6.3.3. Hagège (2008)... 258
6.4. Conclusion...260
Chapter 7. Typological and area-historical assessment ... 263
7.1. Correlations of interrogativity...265
7.1.1. Correlations with interrogative strategies ... 266
7.1.2. Interrogatives and word order ... 270
7.1.3. Interrogatives and locus of marking... 277
7.1.4. Interrogatives and alignment... 280
7.1.5. Conclusion... 289
7.2. Changes in questions: areal and historical perspectives ...290
7.2.1. Yes-no questions... 290
7.2.2. X-neg-X questions ... 291
7.2.3. Alternative questions... 293
7.2.4. Three types of verb-related questions in Sinitic... 295
7.3. Further topics: interrogation and negation ...310
7.3.1. Negation and interrogation in ask-and-answers... 310
7.3.2. Diachronic negation and interrogation... 313
Chapter 8. Conclusion ... 315
References... 319
Appendix I. Features of 138 languages of China... 341
Appendix II. Atlas of interrogative strategies... 351
Acknowledgements
My work in Konstanz could not have been realized without the help of various people. I am deeply indebted to everyone who helped directly with this project, as well as to all who provided inspiration, support, and encouragement along the way.
First of all, I am particularly grateful to my supervisors Frans Plank and Nicole Dehé. I would like to thank Frans Plank for his invaluable advice on the project and kind support over the years, and thank Nicole Dehé for her constructive comments on earlier versions of the manuscript. Sincere thanks also go to Ka Yin Benjamin Tsou for agreeing to serve as the external examiner on my evaluation committee.
I would like to thank the Department of Linguistics at the University of Konstanz, particularly Aditi Lahiri (ständige Gastprofessorin), Maribel Romero, and Heike Zinsmeister for their useful and informative seminars, and for Heike’s help with the R project. Friends in offices G111-112 and G115 deserve special thanks for their constructive assistance, conversation, and laughter. I am especially grateful to Thomas Mayer for his support in general and for plotting the atlas. Thanks to Muna Pohl and Florian Schönhuber, too, for their kind help.
I also wish to sincerely thank Bingfu Lu, Jue Wang, and Elizabeth Zeitoun.
I thank Lu Laoshi and Wang Laoshi for leading me into an academic career in linguistics and for their kind support during my years of study. Thanks to Elizabeth Zeitoun for providing me with a detailed list of publications on Formosan interrogatives. I am also indebted to native language informants for having provided me with various linguistic data for my research.
Part of this work was presented at ICSTLL 43 (Lund 2010), ALT 9 (Hong Kong 2011), and ALT 10 (Leipzig 2013). I am grateful to the participants for their useful comments.
For financial support, I owe special thanks to the China Scholarship Council (CSC) affiliated with the Ministry of Education.
Finally, I would like thank my family for their love and support. Sadly, my father passed away during my studies in Konstanz and could not witness the completion of this work. I would like to dedicate the present dissertation to his loving memory.
Zusammenfassung
Diese Dissertation untersucht Interrogativstrategien von 138 Sprachen in China in Bezug auf ihre Vielfalt und auf ihre Gemeinsamkeiten. Dabei wird ein räumlich-typologischer Ansatz verfolgt, der eine quantitative Analyse von 20 strukturellen Kriterien beinhaltet (zumeist morphosyntaktische Parameter).
Diese Arbeit zeigt wesentliche strukturelle Eigenschaften der Interrogative in den einzelnen Sprachen auf und versucht Korrelationen zwischen den strukturellen Merkmalen herauszustellen. Des Weiteren wird untersucht, ob eine spezifische räumliche Verteilung von Strukturmerkmalen das Ergebnis von Sprachkontakt zu benachbarten Sprachen ist. Dabei machen Erkenntnisse im Bereich der Strukturmerkmale, Korrelationen zwischen den Parametern und räumliche Faktoren interrogativer Strategien den maßgeblichen Beitrag dieser Arbeit aus.
Inhalt und Ergebnisse werden wie folgt zusammengefasst:
Kapitel 2 beinhaltet eine Studie der interrogativen Strategien der Sprachen in China. Zudem enthält dieses Kapitel eine Analyse zweier einzelner Sprachen, Standard Chinesisch und Yongxin Gan, sowie ein Profil ihrer Interrogativ- strategien.
Mehrere spezifische Themen im Bereich Interrogativstrategien werden in den Kapiteln 3 bis 6 behandelt.
Kapitel 3 macht deutlich, dass sich die Position der Fragepartikeln der Sprachen in China deutlich von einer weltweiten Sprachstichprobe unterscheidet (Dryer 2005b), denn die Fragepartikeln der meisten chinesischen Sprachen werden bevorzugt an das Satzende gesetzt, wohingegen in Dryers Stichprobe ein sehr viel kleinerer Teil der Sprachen Fragepartikeln an das Satzende stellt.
Ebenso zeigt eine Untersuchung der Entscheidungsfragen, die in den sinitischen Sprachen mit zwei benachbarten/adjazenten Finalpartikeln gebildet werden, z.B.
ma ne Fragen, dass solche Fragen mittels einer Fragepartikel und einer Finalpartikel gebildet werden. Damit wird der Mythos der sogenannten ‚Fragen mit zwei Fragepartikeln’ widerlegt.
In der Diskussion über Disjunktionen und Alternativfragen stellt Kapitel 4 Kriterien für die Unterscheidung zwischen X-neg-X Fragen und alternativen Fragesätzen in den sinitischen Sprachen vor, da X-neg-X Fragen in der
Fachliteratur meist als Subtyp alternativer Fragesätze behandelt werden. Dieses Kapitel geht zudem von einer or vs. or/or? Typologie für alternative Fragen aus und untersucht zugleich, welche Faktoren für eine solche Typologie von Bedeutung sind. Denn in den Sprachen Chinas sind Disjunktionen in Deklarativen und Interrogativen verschieden, jedoch zeigen nicht alle Sprachen eine derartige Unterscheidung. Diese typologische Unterscheidung ist wichtig, weil sie (zumindest) mit einigen Wortstellungsparametern korreliert; sie wird in den Generalisierungen zusammengefasst (Abschnitt 4.2). Neben den “normalen Disjunktionen“ gibt es einige Sprachen, welche Finalpartikeln als disjunktive Strategie verwenden. Eine weitere Studie zeigt, dass Partikeln nach dem zweiten gewählten Token häufiger als nach dem ersten wegfallen.
Im Gegensatz zu den Ergebnissen der weltweiten Sprachstichprobe (Dryer 2005c), ist wh- in situ ein allgemeines Charakteristikum der Sprachen in China.
Dies wird in Kapitel 5 näher beschrieben. Das seltene Phänomen der wh-Reduplikation in interrogativen Phrasen taucht in 30 Sprachen (meistens im Tibetobirmanischen) auf. Nichtsdestotrotz variieren einzelne wh-Phrasen in ihren Möglichkeiten der Reduplikation, insbesondere Wörter wie who, what, where, und which können leicht redupliziert werden, bei how, when, und why gibt es einige Schwierigkeiten und how many/much und how long (time) können kaum redupliziert werden. Dies lässt sich damit erklären, dass wh-Phrasen nur redupliziert werden können, wenn eine bestimmte Phrase semantisch eine pluralische Bedeutung in sich tragen kann.
Inhalt des sechsten Kapitels sind drei Arten von Fragesätzen: diejenigen, die sich auf das Verb beziehen, diejenigen, die mit interrogativen Adverbien gebildet werden und diejenigen, die mittels Verb-Reduplikation und interrogativen Verben gebildet werden. Bei den ersten beiden Arten handelt es sich um Entscheidungsfragen, beim dritten Typus um Inhaltsfragen. Sinitische Sprachen werden generell als isolierende Sprachen betrachtet, denen morphologische Strategien fehlen. Dieses Kapitel widerlegt diese Ansicht, indem hier die drei Typen der Fragesätze, die sich auf das Verb beziehen (alle in sinitischen Sprachen belegt) gemeinsam betrachtet werden und weist darauf hin, dass interrogative Strategien klare Fälle morphologischer Prozesse sind.
Interrogativstrategien können nicht alleine funktionieren. In einer typologischen und räumlich-historischen Untersuchung von Interrogativen werden im siebten Kapitel 20 morphosyntaktische Parameter zusammengebracht, wobei in einem auf Häufigkeiten basierenden Ansatz Korrelationen aufgezeigt
werden.
Als System korrelieren die einzelnen Strategien untereinander, als Teil eines größeren Sprachsystems korrelieren sie mit vielen anderen morphosyntaktischen Parametern. All diese Korrelationen werden in Form von 30 Generalisierungen zusammengefasst (Abschnitt 7.1). Interrogative Strategien variieren und verändern sich über Raum und Zeit. In diesem Zusammenhang werden geographische Faktoren vorgestellt, insbesondere der Kontakt mit dem Standard Chinesischen (und seinen Entlehnungen) innerhalb vieler Sprachgruppen. Zudem werden historische Entwicklungen, hauptsächlich die Geschwindigkeit der Veränderung, die Stabilität/ der Erhalt und der Rückgang individueller interrogativer Strukturen und Sprachgruppen dargestellt (Abschnitt 7.2). Kapitel 7 zeigt somit, dass einerseits Interrogative untereinander sowie mit anderen Parametern korrelieren und andererseits, dass die Vielfalt der Interrogative das (instabile) Ergebnis und die Repräsentation von Sprachwandel über Raum und Zeit sind.
Abstract
This dissertation explores the diversity and unity of the interrogative strategies in 138 languages of China. It adopts an areal-typological approach and presents a quantitative analysis on 20 structural features (mostly morphosyntactic parameters).
This work provides substantive structural features of interrogatives in individual languages and seeks to establish correlations in different structural features. It also seeks to establish whether particular areal distributions of structural features are the result of language contact among neighboring languages. Hence, structural features, correlations in parameters, and areal factors in interrogative strategies are the major contributions of this work.
The major contents and findings are summarized as follows:
After an introduction in Chapter 1, Chapter 2 gives a survey of the interrogative strategies in the languages of China, with also studies in two individual languages, Standard Chinese and Yongxin Gan, presenting a profile of interrogative strategies of these languages.
Several specific topics on interrogative strategies are discussed in Chapters 3-6. Chapter 3 suggests that the positions of question particles in the languages of China are very different from the worldwide language sample (Dryer 2005b) in that most languages of China prefer sentence-final positions, while Dryer’s sample reports that a much smaller proportion of other languages have sentence-final question particles. Also, the discussion on polar questions formed by two adjacent final particles in Sinitic languages, i.e. ma ne questions, reveals that such questions are formed by a question particle plus a final particle, which dispels the myth of the so-called ‘questions formed by two question particles.’
In the discussion on disjunctions and alternative questions, Chapter 4 provides criteria for distinguishing X-neg-X questions and alternative questions in Sinitic, as X-neg-X questions are frequently treated as a subtype of alternative questions in the literature. This chapter also proposes an or vs or/or? typology in alternative questions and discusses what is important to such a typology because disjunctions in declaratives and interrogatives are different in many languages of China, while some other languages do not demonstrate such a difference. The typology matters in that it correlates with (at least) some word order parameters,
which are summarized in several generalizations (Section 4.2). Except for those with normal disjunctions, some languages are found to use particles as disjunctive strategies. A further study shows that particles to be chosen after the second disjunct are more likely to be dropped than those after the first disjunct.
In Chapter 5, quite different from the findings in a worldwide language sample (Dryer 2005c), wh- in situ is found to be a general characteristic of the languages of China. A rare phenomenon, wh-reduplication in interrogative phrases, is found in 30 languages (mostly Tibeto-Burman). Nevertheless, individual wh-phrases vary in the capability of reduplication, particularly words like who, what, where, and which can be reduplicated very easily, while how, when, and why bear some difficulty, and how many/much and how long (time) can hardly be reduplicated. The explanation is that reduplicating wh-phrases is only possible when a certain phrase can semantically carry a plural meaning.
Three types of verb-related questions, in particular questions formed by a pre-verb interrogative marker, verb-reduplication, and interrogative verbs, are brought together in Chapter 6. The first two types are polar questions and the third type is content questions. Sinitic languages are generally considered to be
‘isolating’ languages that lack inflectional morphology. This chapter fine-tunes such claims by bringing three types of verb-related interrogatives (all reported in Sinitic languages) together and proposes that such interrogatives are clear cases of morphological operations.
Interrogative strategies correlate with each other and with other categories and parameters. In a typological and areal-historical assessment of interrogatives, Chapter 7 brings 20 parameters together and finds some correlations in a frequency-based approach. As a system, individual strategies correlate with each other; as a part of larger systems of language, they correlate with many other morphosyntactic parameters and these correlations are summarized in 30 generalizations (Section 7.1). Interrogative strategies vary and change in space and time. Geographical factors, notably contact with (and borrowing from) Standard Chinese in many groups of languages, as well as historical factors, notably the pace of change, i.e. pertinacity or transience in individual interrogative structures and in individual groups of languages, are presented (Section 7.2). Chapter 7 hence holds that diversity in interrogative strategies is the (unstable) result and representation of language change in space and time.
Notational conventions
Abbreviations
Basically, the conventions followed are those given in The Leipzig Glossing Rules (Comrie et al. 2008). The following abbreviations are used in the interlinear glosses of language examples. In examples taken from descriptive literatures, the glosses are generally the same as those used in their original forms.
1 first person
2 second person
3 third person
A agent-like argument of canonical transitive verb ABS absolutive
ACC accusative Adj adjective Adv adverb(ial) AGT agent
AF agent focus (actor focus)
AP adjective phrase
CAUS causative CL classifier COP copula
D demonstrative DAT dative
DEF definite DIR direction
DISP disposal (construction)
EMP emphatic marker
ERG ergative F feminine
FP sentence-/disjunct-final particle FUT future
FV verb-final vowel
GEN genitive
HON honorific IMP imperative INCL inclusive IRR irrealis LAC Language Atlas of China LOC locative M masculine
NEG negation, negative
NOM nominative NOMIN nominalization
NP noun phrase
NUM/Num numeral OBJ object OBL oblique
P patient-like argument of canonical transitive verb PASS passive
PFV perfective
PIPCQ Position of interrogative phrases in content questions PL plural
Po postposition
PPQP position of polar question particles
PQ polar questions
PREF prefix
Pr preposition PRF perfect PROG progressive PRO(N) pronoun
PN proper name
PRS present PRT particle PST past
Q question particle/marker
RDP reduplication
S sole argument of the intransitive verb SUFF suffix
VP verb phrase
WALS The World Atlas of Language Structures
IPA and Standard Chinese Pinyin
Pinyin IPA Pinyin IPA
b p ai ai, aɪ, Ai
p ph an an, An
m m ang aŋ
f f ao au, aʊ
d t ei ei, eɪ
t th en ən
n n eng əŋ, ʌŋ
l l ia iA, ia
g k ian iɛn, iæn
k kh iang iaŋ
h x iao iau, iaʊ
j tɕ ie iɛ, ie
q tɕh in in
x ɕ ing iŋ
z ts iong iʊŋ, yəŋ
c tsh i(o)u iou, iəʊ
s s ong ʊŋ, uŋ
zh tʂ ou ou, əʊ
ch tʂh u(e)i uəi, uəɪ
sh ʂ u(e)n uən
r ɹ ua uA, uɑ
uai uai, uaɪ, uAi
a A,ɑ, ɛ uan uan, uAn
e ɣ, e, ɛ, ə uang uaŋ
o o, u ueng uəŋ, uʌŋ, uʊŋ
i i, ɿ, ʅ üan yan, yɐ, yɛn
u u, y üe yɛ, ye
ü y ün yn, ün
er ɚ
Notes:
1. i is written as y after zero initial and is written as yi in isolation;
u is written as w after zero initial and is written as wu in isolation;
ü is written as yu after zero initial and is written as u after initials j, q, x.
2. In this thesis most examples of Standard Chinese and other Sinitic languages are given in Pinyin, while those of Yongxin Gan (Sinitic) and minority languages are given in IPA.
Tone system of Standard Chinese
Tone Symbol Graph Pitch Example
high level
¯
55 mā [ma55] 妈 ‘mother’high rising
ˊ
35 má [ma35] 麻 ‘hemp’falling-rising
ˇ
214 mǎ [ma214] 马 ‘horse’high falling
ˋ
51 mà [ma51] 骂 ‘(to) scold’Notes:
1. This table is adapted from Li and Thompson (1981: 6-9) and Sun (2006: 39-40).
2. ‘Falling-rising’ tone is also known as ‘dipping’ tone.
3. In each tone graphs the vertical line on the right serves as a reference for pitch height, which is divided into five levels, 5 is the highest and 1 is the lowest.
4. The unstressed neutral tone (qīngshēng) is not included in the table. Cf. dōngxī ‘west and east’ (xi in the high level tone) and dōngxi ‘thing’ (xi in the unstressed neutral tone).
1 Introduction
Chapter 1 Introduction
1.1. The grammar of interrogatives 1.1.1. Interrogative forms
The sentence types (or types of grammaticalized speech acts) consist of three members, declarative, imperative, and interrogative (though exclamatives are also very often included in the literature). The strategies of their form vary, but the declarative is the default sentence type and is typically left unmarked, the imperative is generally shown by verbal affix(es), and the interrogative has many forms. The core issue of the present thesis is the forms, or “strategies”, for asking questions, in particular polar questions (also known as yes/no questions).
Content questions (also known as wh- questions, information questions, constituent questions) are also covered, but less central.
Three kinds of interrogative forms should be distinguished at first place:
prosodic, morphological, and syntactic. Cross-linguistically, the prosodic question marking is suggested by an intonation contour which is different from the one in declarative, normally a terminal rising one, at times also falling, or some other contour patterns. The morphological and syntactic forms, i.e. the non-prosodic question marking may take a variety of forms, ranging from full or reduced clauses or phrases over independent words or vocal noises to clitics and affixes (Plank 2009).
The inventory of question marking varies among individual languages and individual linguists. For example, in their introductory typological studies on sentence types and/or interrogative strategies, Sadock and Zwicky (1985), König and Siemund (2007), and Dryer (2005a) suggest three different lists of strategies
1 Introduction
for polar interrogative sentences (S & Z, K & S also include content questions, which is not the topic here; cf. Siemund 2001, Koptjevskaja-Tamm and Liljegren 2013).
(1) Polar interrogative strategies according to Sadock and Zwicky 1985 a. Intonation contour
b. Question particles
c. Interrogative verb morphology d. Alternative structures e. Word order change
(2) Polar interrogative strategies according to König and Siemund 2007 a. Intonational marking
b. Interrogative particles c. Verbal inflection
d. Disjunctive-negative structures e. Change in word order
f. Special tags
(3) Polar interrogative strategies according to Dryer 2005a a. Interrogative intonation only
b. Question particles
c. Interrogative verb morphology d. Interrogative word order
e. Absence of declarative morphemes f. No interrogative-declarative distinction g. Question particle and verb morphology
(1) and (2) are basically the same, except that König and Siemund (2007) noticed that certain languages spoken in Papua New Guinea (Amele, Kobon) and some Asian languages, e.g. Standard Chinese, use a disjunctive-negative structure to phrase questions (see below; see also Sections 2.1.1.1.4 and 4.1 for discussions of X-neg-X questions in Standard Chinese).
(3e)-(3f) are different from (1)-(2) in that Dryer (2005a) is aware of some languages which express questions by omitting certain morphemes that are used in corresponding declarative sentences, e.g. Zayse (Omotic, Afro-Asiatic;
Ethiopia), Kabardian (Northwest Caucasian; Russia; Colarusso 1992: 125-126),
1 Introduction
Puquina (isolate; Bolivia; Adelaar 2004: 354), Dinka (Nilotic; Sudan; Nebel 1948: 58-61), and Huichol (Corachol, Uto-Aztecan; Mexico).
(4) Zayse (Hayward 1990b: 307; cited in Dryer 2005a) a. hamá-tte-ten ‘I will go’
b. háma-ten ‘Will I go?’
c. hamá-tt-isen ‘She will go.’
d. háma-ysen ‘Will she go?’
(5) Huichol (Grimes 1964: 27; Palmer 2001: 54)
a. pée-t ʌ́a
ASSERTIVE-direction go
‘He left.’
b. mázá tikuucúu
deer asleep
‘Is the deer asleep?’
In (4), the morpheme -tt(e)- is missing in the interrogatives (4b, 4d), but is kept in corresponding declarative sentences (4a, 4c). In (5a), an ‘assertive’ marker pée- is employed to indicate the sentence is a statement, while the form used as a question is often the unmarked form (5b).
Moreover, there are languages simply demonstrate no formal marking in polar interrogatives, such as Chalcatongo Mixtec (Mixtecan, Oto-Manguean;
Mexico) and Gooniyandi (Australian, Australia; McGregor 1990: 485, 382-3, 369-71; see Miestamo 2011).
(6) Chalcatongo Mixtec (Macaulay 1996: 126) ñábaʔa-ró librú-ro(?)
have-2 book-2
‘You have your book. / Do you have your book?’
(6) can be interpreted as either a declarative sentence or an interrogative sentence, with no difference in intonation associated with the two meanings.
Yet there are a number of languages, e.g. Blackfoot (Algonquian, Algic) and Greenlandic (Eskimo, Eskimo-Aleut), both interrogatives and declaratives are marked by special verb morphology, although not the identical ones (see Sadock and Zwicky 1985).
1 Introduction
The markedness pattern of declaratives and polar interrogatives is summarized in Table 1.1.
Table 1.1. Markedness pattern in declaratives (D) and polar interrogatives (Q)
Markedness pattern Frequency Languages attested (i) D unmarked, Q unmarked rare Chalcatongo Mixtec (ii) D unmarked, Q marked common Germanic, Sinitic, etc.
(iii) D marked, Q marked rare Blackfoot, Greenlandic (iv) D marked, Q unmarked rare Dinka, Huichol, Kabardian,
Puquina, Zayse
It can be seen that polar interrogatives are cross-linguistically more marked than declaratives.
The major forms for asking polar questions in the languages of China can be found in the three lists (1-3) mentioned above.
Interrogative intonation only
Languages form questions by interrogative intonation only (IIO) suggests that they involve same words, morphemes and word order as the corresponding declarative sentence, but with a distinct intonation pattern as the sole indication signaling that it is a question.
Most languages employ a distinctive intonation in questions, in collaboration with some other morphosyntactic strategies. Nevertheless, some of these languages cannot form questions by a distinctive intonation only. Moreover, the IIO languages vary in the frequency in employing the strategy. In other words, although most languages use a distinct intonation pattern in interrogatives, only some of them use the device as the sole indication – this is a matter of “yes” or
“no”; in the IIO languages, some use it often, others not – this is a matter of
“more” or “less”.
There are two extreme cases. One is that the distinctive intonation is in complementary distribution with some other formal markers of interrogation (e.g.
Chrau, a Mon-Khmer language of the Austro-Asiatic family), and another is that some languages simply do not use distinctive intonation at all (e.g. Greenlandic;
cf. Sadock and Zwicky 1985).
1 Introduction
There seems to be a hierarchy in the employment of interrogative intonation cross-linguistically:
(7) Hierarchy of interrogative intonation
IIO in complementary distribution with other strategies > IIO (common >
less common) > Distinctive intonation and others strategies > No distinctive intonation
In many languages of China, distinctive intonation is used to form questions, without the participation of other strategies, although so far no language ranked the highest in the hierarchy. For example, many Sinitic languages can form polar questions by IIO, but not very common, e.g. Standard Chinese, where a special context is required (see Section 2.1.1.1.2). In most Austro-Asiatic languages, polar questions can be formed solely by a terminal rising intonation, e.g.
Bulang.1
(8) Bulang (Li et al. 1986: 73) miʔ2 kɔʔ4 l̥aʔ1 hɣl1? ↗
2SG also want go
‘Do you also want to go (there)?’
Question particles
Question particles are invariable items with the function of forming questions. As a marker of sentence types, question particles signal that a certain sentence is a question. Question particles are different from question tags because no predicate meaning ‘is’ or ‘true’ is involved (which is typical in question tags; see below), and different from interrogative verb inflection in that they are not verb-related but sentence-related, i.e. such particles question the complete statement, not the verb. They are also different from the “particles in questions” in that the latter may not necessarily form questions (see the introduction of Chapter 3 for more discussion on the notion “question particle”).
1 In the nine Austro-Asiatic languages in China, six languages of the Mon-Khmer group invariably use IIO. Nevertheless, only one language from the Viet-Muong group, Lai, use such a question-signaling device, and the other two languages either do not use IIO (Jing) or is still not clear (Mang).
1 Introduction
Many if not most languages with question particles favor to place them sentence-finally. In Dryer’s (2005b) 777-languages data, 272 languages use sentence-final particles, which is much more frequent than other positions, e.g.
initial (118 lgs), second (45 lgs), other (8 lgs), etc. (Cf. Greenberg’s 1966 Universals 9-10 and Ultan’s 1978 Universal 7 on order.)
Sentence-final question particles are especially common in the languages of China, where particles of other types of illocutionary forces are widely reported.
This brings difficulties in distinguishing the question particles and other particles, now that they are in similar forms and occur at the same position. Take Standard Chinese for example, controversial remains in whether final particles ne, ba, and a are question particles or not, although ma is generally accepted to be a question particle.
(9) Standard Chinese
a. ni shi ta gege a/ba/ma?
2SG be 3SG elder.brother FP
‘Are you the elder brother of her/him?’
b. ni shi-bu-shi ta gege a/ne?
2SG be-not-be 3SG elder.brother FP
‘Are you the elder brother of her/him or not?’
In (9a), the interrogative meaning is expressed by using a terminal rising intonation, and (9b) relies on the X-neg-X structure. In both cases, final particles a and ne are not indicators of polar interrogatives (but ma and ba are). In other words, a and ne are common final particles (here, in questions), and ma and ba are final particles signaling questions (see Section 2.1.1.1.3 for more discussion).
Interrogative verb morphology
Interrogative verb morphology covers a variety of question-signaling devices, ranging from affixes or clitics to verb inflection as well as some minor strategies like verb-reduplication and tonal change on the verb.
The use of affixes is a very common interrogative strategy cross- linguistically. For example, in Dryer’s (2005a) 842-languages data, 155 languages employ such a strategy. In the languages of China, this is especially common in the Tibeto-Burman languages. Some languages use interrogative
1 Introduction
infixes, which is a typological rarity, e.g. Muya (Tibeto-Burman, Sino-Tibetan), where an infix æ55 is added between the verb stem and its suffix(es) (see Section 2.1.2.4 for more examples and discussion).
In a number of languages, polar questions can also be formed by reduplicating the verbs or adjectives (adjective-reduplication is rare, see Section 6.2.2). This is reported in a number of Sinitic and Yi languages (Tibeto-Burman, Sino-Tibetan).
(10) Yi (Chen et al. 1985: 94; Chen et al. 2007: 265) a. la33 ‘come’, la44la33 ‘Come?’
b. lɔ55pɔ21 ‘help’, lɔ55pɔ21pɔ33 ‘Help?’
In (10), the verb is reduplicated and expresses an interrogative meaning (note that 10a is total reduplication and 10b is partial, although both are changed in tones). A similar strategy is found in many other Sinitic and Yi languages as well, some with tonal changes, others do not. Such verb-reduplicating polar questions will be addressed in Section 6.2.
Interrogative verbs2 are also a kind of interrogative verb morphology. Such words are normally composed of a verb stem and a morpheme indicating the pronouns (cf. Idiatov and van der Auwera 2004 “interrogative pro-verb”), although they invariably involve content questions. For example, in some Sinitic languages and most Formosan languages, interrogative verbs are employed in forming content questions.
(11) Colloquial Standard Chinese (Hagège 2008) nǐ zài gànmá?
2SG PROG do.what
‘What are you doing?’
Interrogative verbs are synchronically unanalyzable. In (11), gànmá as a whole cannot be analyzed as a verb plus an interrogative pronoun, that is, it is a verb(-complex), not a verbal phrase.
The issue of interrogative verbs will be addressed in Section 6.3.
At times, the term “pre-verb interrogative marker” (for short, Q-V) is used
2 An interrogative verb is “a kind of word which both functions as predicate and questions the semantic content of this predicate.” (Hagège 2008)
1 Introduction
in this thesis. It covers the interrogative prefixes and clitics which occur before the verb, e.g. a- in many varieties of Wu (Sinitic, Sino-Tibetan) and ke- in many varieties of Mandarin. This is unfortunate because in some languages such interrogative markers are clearly prefixes or clitics, but hard to make the distinction in other languages. In this sense, a more general label, though less accurate, is employed.
X-neg-X structures
X-neg-X structures are also known as disjunctive-negative structures (Thompson 1998, König and Siemund 2007), V-not-V structures (Chao 1968), and A-not-A structures (Li and Thompson 1981), etc. “Disjunctive-negative” is not precise because no disjunction, which is a feature of alternative/disjunctive questions, is involved in the structure, although it has close affinity to disjunctive questions both in form and meaning. (For V-not-V, A-not-A, and some others, see Section 2.1.1.1.1.)
The label “X-neg-X” is adopted in this thesis, and “disjunctive-negative” is used only in marginal cases to refer the structure, but not as a (sub)type of questions. Admittedly, X-neg-X questions are semantically similar to alternative questions in that both of them involve a choice of one item from more possibilities, but the differences remains in that (i) the former involves only two items, one positive and one negative, but the latter may involves more than two, not necessarily with one positive and one negative, (ii) less restrictions are placed on the former than the latter in syntactic alternations, (iii) the former can be answered by a less clear statement than the latter (see Section 4.1 for more discussion).
The X-neg-X structure seems to be a typological rarity which is exclusively reported in the languages of China (König and Siemund 2007 also mention some other Asian languages and languages spoken in Papua New Guinea). In the literature of Chinese linguistics, such questions are also known as zheng-fan wenju ‘positive-negative questions’ (sometimes also termed fanfu wenju
‘repetitive questions’). As the label suggests, such questions are characterized by the positive-negative structure.
X-neg-X questions are used in most languages of China, although they are not found in the Altaic languages and some Austronesian and Tibeto-Burman languages. The following example is taken from Sulong (Tibeto-Burman, Sino-Tibetan).
1 Introduction
(12) Sulong (Li 2004: 167)
a. na55 ɬa33sa55 wu55ga31 ba31 wu55ga31?
2SG Lhasa go NEG go
‘Are you going to Lhasa or not?’
b. na55 ɟe33 a31ŋwa33ȵiaŋ55da31 ba31 ŋwa33ȵiaŋ55da31? 3SG.M TOP good.looking NEG good.looking
‘Is he handsome or not?’
In Sulong, the X can be a verb or an adjective. In some other languages, for example in some Sinitic languages, like Standard Chinese and Yongxin Gan, it can also be a noun (phrase) (see Section 2.1.1.1.4).
Alternative structures
Alternative questions (also known as disjunctive questions) are formed with two or more constituents conjoined by disjunction(s), i.e. alternative structures.
The structure provides a list of propositions from which, the speaker suggests, the right answer might be drawn (Sadock and Zwicky 1985).
Semantically, alternative questions are similar to content questions in the answer set because both types seek for information, not logical polarity.
Nevertheless, the present thesis treats such structures as polar questions in regard of three reasons. First, semantically, the addressee is requested to choose which one in the two or more alternatives holds, i.e. alternative questions seek a yes/no value of one disjunct among the two or more. Second, formally, alternative questions do not employ any wh- phrases. Third, also formally, a similar type, X-neg-X questions, is treated as polar questions.
In might be equally sensible to claim that yes/no questions are a subtype of alternative questions because the addressee is asked to choose a “yes” or “no”
value, i.e. yes/no questions are “definite” alternative questions because there are only two alternatives. More radically, one may claim that wh- questions are also a subtype of alternative questions in that the addressee is asked to choose one item from a variety of choices, i.e. wh- questions are “indefinite” alternative questions, although the wh- phrases help to determine what kind of information that the questioner is requested to supply, but the range of choice is not as fixed as alternative questions. The controversial in the taxonomies of interrogatives in Standard Chinese is discussed in Section 2.1.1.1.1.
1 Introduction
The differences between alternative questions and X-neg-X questions have been addressed before, yet there is another matter deserves to be mentioned here, namely the alternative structures conjoined by particles. The following example is taken from Achang (Tibeto-Burman, Sino-Tibetan).
(13) Achang (Dai and Cui 1985: 78) a. nuaŋ55 lɔ35 ma21 lɔ35?
2SG go NEG go
‘Are you going (there) or not?’
b. nuaŋ55 lɔ35 neʔ 55 la21, ma21 lɔ35 la21?
2SG go PRT QP NEG go QP
‘Are you going (there) or not?’
(13a) is a X-neg-X question, and (13b) is an alternative question because the particle la21 functions as a disjunction. Particle disjunctions have been found in a number of languages in China, the issue will be addressed in Section 4.3.
Special tags
Question tags are constituents consisting of an auxiliary verb plus pronoun, attached at the end of a statement in order to convey a negative or positive orientation, e.g. English isn’t it or innit, German nicht wahr. Question tags normally contain a predicate with meaning like ‘is’ or ‘true’ (Sadock and Zwicky 1985; Plank 2009; see also Dehé and Braun 2013 for a detailed definition and references given therein).
According to König and Siemund (2007), question tags are different from question particles in that (i) “[question] tags, apart from characterizing sentences as questions, also contribute a certain bias by raising expectations toward either a positive or negative answer”, and (ii) “[question] tags almost exclusively occur at the end of a sentence, quite independently of basic word order pattern”. That is, cross-linguistically, question particles are an invariable parts-of-speech which occur not necessarily sentence-final (see Section 3.1 for Dryer’s 2005b data), while question tags are constituents (normally containing a predicate) attached at the end of a sentence, although both of them characterize sentences as questions.
Nevertheless, sentence-final seems to be only a preferred position for question tags because in natural language, e.g. English, they may take utterance-final, sentence-final, XP-final, and XP-medial positions, though the XP-final/-medial
1 Introduction
positions are much less common (see Dehé and Braun 2013). Two further possible criteria for the distinction between question particles and question tags are (iii) prosodic separation is more frequent before (non-phrase-medial) question tags compared to question particles, and (iv) question tags are more complex syntactically than question particles, because question tags normally contain verbs or modals while particles are invariable items.
In many languages of China, notably the Sinitic languages, X-neg-X structures are frequently used as question tags.
(14) Standard Chinese
ni yao qu Beijing, dui bu dui?
2SG FUT go Beijing right not right ‘You are going to Beijing, right?’
Li and Thompson claim that the tag question in Mandarin Chinese is composed of a statement followed by an A-not-A (= X-neg-X) form, e.g. dui bu dui ‘right not right’, hao bu hao ‘good not good’, xing bu xing ‘OK not OK’, shi bu shi ‘be not be’ (Li and Thompson 1981: 521). Nevertheless, this is not the whole story.
There are at least two equally common strategies, namely (i) to add a final particle at the end of a copula or an evaluation adjective, sometimes also with a negation word before the structure, e.g. (bu) dui/hao/xing/shi ma/ba? ‘(not) right/good/OK/be QP’, and (ii) to use a negation word after a copula or an evaluation adjective, e.g. dui/hao/xing/shi bu? ‘right/good/OK/be NEG’.
1.1.2. Asymmetries in form and meaning
Different questions are signaled by different forms. Intonational questions, particle questions, alternative questions, X-neg-X questions, tag questions, and wh- questions are typically signaled by distinctive intonation, question particles, alternative structures, X-neg-X structures, tags, and wh- phrases, respectively.
Nevertheless, a number of asymmetrical phenomenon is found in the form and meaning of interrogatives.
Distinctive intonation
Terminal rising intonation is a common signaling device of polar questions
1 Introduction
cross-linguistically. In a number of languages, final rising intonation is used both in polar and content questions, e.g. Diola (Atlantic, Niger-Congo; Sapir 1965, Sadock and Zwicky 1985). In Puyuma (Formosan, Austronesian), terminal falling interrogative intonation is reported in the polar questions without interrogative particles (Huang 2000: 151-3).
(15) Puyuma (Huang 2000: 151-2) a. sagar=yu kanku amáw?↗ like.AF=2.NOM 1 QP
‘Do you like me?’
b. a-ekan=yu Da biTénum?↘
RDP-eat.AF=2.NOM OBL egg
‘Would you like to eat eggs?’
In the particle questions (15a), final rising intonation is used, which is the same in declarative sentences. However, in the polar questions without interrogative particles (15b), terminal falling intonation is adopted.
Question particles
In Standard Chinese and most other Sinitic languages, (final) question particles resemble other final particles both in form and distribution. In particular, their phonological structures are CV, with nasal or bilabial consonants /n/ /m/ /p/
and open (mid) front vowels /a/ /ɛ/; and, as the label ‘final particle’ suggests, they occur sentence-finally. However, only some of them form polar questions, e.g. ma, ba, while some others cannot, e.g. a, ne.
Final particles also behave differently in wh- questions in Standard Chinese and most other Sinitic languages as well. As they occur in polar questions, final particles also express various pragmatic meanings in wh- questions, depending on the context, although they are not necessary in forming wh- questions.
(16) Standard Chinese
a. shui zhidao? (wh- question)
who know
‘Who knows this?’
1 Introduction
b. shui zhidao a/ne? (wh- question, rhetorical/‘on earth’/politeness)
who know FP
‘Who knows? (I don’t know.)’ / ‘Who on earth knows this?’ /
‘(Please tell me) Who knows this?’
c. shui zhidao ma? (polar question)
who know QP
‘Does someone know this?’
Note that a wh- question turns into a polar one if question particle ma is used (16c). (16c) is a case of indefinite-interrogative affinity (see, e.g. Li 1992, Haspelmath 1997, Bhat 2000, Gärtner 2009). In fact, in Standard Chinese the wh-phrases turn to be indefinite pronouns if question particle ma (and the like) is added at the end of a content question, cf. shui ‘who ~ someone’ (16c), nali
‘where ~ somewhere’, shenme ‘what ~ something’, etc.
(16c) has both a wh- word shui ‘who’ and a question particle ma, but it is a polar question, not a wh- one. The rule is that wh- questions invariably turn into polar questions by taking question particle ma (and sometimes ba), in other words, polar question particles override wh- words and form polar questions.
ne is different from other final particles in Standard Chinese in that it forms the so-called reduced wh- questions but the others cannot. In most other Sinitic languages and a number of minority languages in China, only some final particles can form such reduced wh- questions, and the number is much smaller than those used in normal wh- questions, although same particles are used in both normal wh- questions and reduced wh- questions in some languages, e.g.
Yongxin Gan (Sinitic) ne, Ningbo Wu (Sinitic) ȵi, Dulong (Tibeto-Burman) da55, Biao (Kam) ni1, etc. Such particles are collected in Section 3.3.2.
(17) Yongxin Gan
a. Zhangsan ne? ‘Where is Zhangsan?’
b. Zhangsan a? ‘Are you talking about Zhangsan?’
(17a) is a reduced wh- question (with particle ne), and (17b) is a polar question (with particle a, which is similar to Standard Chinese ma in this case).
Alternative structures
The form-meaning asymmetry in alternative questions can be seen at least
1 Introduction
in two aspects. First, in many languages, there is in fact no choice need to be made in alternative structures. For example, in English a question in the form of alternatives can be a yes/no question in reality, depending on intonation and prosody, compare: Do you like [apples]↗ or [oranges]↘? (alternative question), Do you like [apples or oranges]↗? (yes/no question) (see e.g. Han and Romero 2004 and references given therein; see also Jennings 1994: 27). Nevertheless, such yes/no reading does not exist in Standard Chinese and most other Sinitic languages (see the introduction in Chapter 4 for more discussion).
Second, alternative questions are hard to be neatly placed in the polar/content taxonomy. The issue has been mentioned before in Section 1.1.1, as such structures resemble both content questions and polar questions. The situation becomes more complicated if wh- phrases are involved, in which the alternative structures normally occur sentence-finally like a tag, e.g. Which course do you like better, Syntax or Semantics?
X-neg-X structures
In the literature of Chinese linguistics, X-neg-X questions are frequently classified as a subtype of alternative questions (see Section 2.1.1.1.1). This is majorly because the semantic relationship between X and neg-X is also one of disjunction.
Formally, the employment of disjunctions, a signal of alternative questions, is not found in such structures (see Section 4.1 for more discussion on the distinctness between the two structures). A radical formal view would claim that X-neg-X structures are declaratives because no interrogative readings can be directly inferred from the appearance. Nevertheless, this point is not adopted here. The present thesis treats alternative and X-neg-X as different structures not in regard of their forms only.
The form-meaning asymmetry in X-neg-X questions brings difficulties in the taxonomy, namely whether it is a subtype of polar questions or content questions. Formally, wh- phrases, a basic feature of content questions, are not involved in X-neg-X questions. Semantically, however, it requires not a yes/no answer but a X/not-X answer (a content information). For example, Standard Chinese ni qu-bu-qu Bolin? (2SG go-NEG-go Berlin) ‘Will you go Berlin?’ cannot be answered by a polar reply shi/bu ‘yes/no’, but can be answered by qu/bu qu
‘go/not go’, which provides content information.
It is obviously not an easy field to deal with the formal and
1 Introduction
semantic(-pragmatic) distinctions of interrogatives, as is seen in the approaches to the taxonomies (cf. Section 2.1.1.1.1). In this thesis, ‘polar questions’ and
‘yes/no questions’ are used as general labels that cover the interrogatives formed by the strategies listed in (18).
(18) Polar interrogative strategies in the languages of China a. Interrogative intonation only
b. Question particles
c. Interrogative verb morphology d. X-neg-X structures
e. Alternative structures f. Reduplicating structures g. Special tags
It can be seen that the subtypes of polar questions are basically based on the forms, that is, the descriptive concepts being used in this thesis are formal, although semantic-pragmatic factors are also considered.
1 Introduction
1.2. Motivation
Language typology studies what the languages of the world are like (Shopen 2007: xiv). In other words, typology is expected to present the diversity and its patterns (and it is expected to find some unity as well). For the present purpose, it is too early to conclude that a certain list of interrogative strategies covers most or all human languages: The list becomes longer when languages with different strategies are reported. In this sense, providing substantive structural features of interrogatives in individual languages is a first step for further generalization.
To compare generalizations of individual structural features, or rather, to seek to establish correlations in different structural features, is another major interest of typology. Some earlier typological studies on interrogatives have provided extensive results regarding the ways in which languages vary structurally and regarding correlations among different features (e.g. Greenberg 1966 universals No. 8-12, and 23 universals proposed by Ultan 1978; see below).
Two main typological views of the unity of human languages, that it demonstrates common features in structures, and that it correlates in structural features, are directly applicable to interrogatives.
This work contributes to both diversity and unity. It provides structural features of interrogativity in 138 languages in China, many of which have not been presented or presented well in the theoretical literature (see Chapters 2-6), and it correlates features of interrogativity with various structural features, in particular clause order, alignment, and locus of marking (see Section 7.1). It also contributes to areal typology by seeking to establish whether particular geographical distributions of different values for interrogative features are the result of contact among neighboring languages (see Section 7.2; for ‘areal typology’, see e.g. Dahl 2001, Comrie et al. 2005, Koptjevskaja-Tamm 2011).
Correlations or unities are not self-explanatory. In other words, language universals cannot be explained simply by claiming that a certain feature or category has something to do with another. For example, Greenberg’s (1966) Universal 10 notes that “Question particles or affixes, when specified in position by reference to a particular word in the sentence, almost always follow that word.
Such particles do not occur in languages with dominant order VSO.” Curious readers would go beyond the numbers of (un)attested languages and ask why question particles or affixes occur later and why don’t they exist in VSO
1 Introduction
languages. Similar problems are found in his Universals 8-9 and 11-12 concerning interrogatives. Such correlations or universals do not make a significant contribution to our knowledge about interrogatives without further explanation, albeit Greenberg (1966) provides two general principles, namely harmony and dominance, to account for such universals, but the principles need further explanation, too.
One obvious shortcoming of macro typology is that such approach does not tell us the history of individual languages or language groups, because it looks at samples of unrelated languages. Micro typology or areal typology remedies the disadvantage by examining languages that are very closely related, like languages or dialects within one language family, or languages that are related to each other geographically or historically.
All the history of language is areal history. For a complete understanding of the diversities and unities in interrogatives (and other grammatical categories as well) in a certain linguistic area (“area-versals”), several issues, namely (i) how do languages contact and change, (ii) how do certain interrogative strategies influence the others, and (iii) what is the origin of certain interrogative markers, are needed to examine.
The motivation or general research questions of the present work accordingly cover three aspects: (i) What the structural features of interrogative strategies are in the languages of China? (ii) How do the structural features of interrogatives correlate between themselves and with other grammatical categories? (iii) Why there are certain interrogative features in certain groups of languages or linguistic areas?
Interrogativity deserves more attention than it has been given by typologists.
As has been pointed out, “there have been few general, cross-linguistic discussions of questions” (Dixon 2012: 429). The major English typological works on interrogatives in the past 50 years are listed following. In his pioneering work on word order universals, Greenberg (1966/1963) proposes some patterns in interrogatives, which are summarized in his universals No. 8-12.
These universals manifest the placement of interrogative intonation (No. 8), the relation between question particles/affixes and adpositions (No. 9) or basic word order (No. 10), as well as the relation between question word/affixes and constituent sequencing (No. 11-12). Moravcsik (1971) presents some generalizations regarding yes/no questions and their answers. In a Greenbergian
1 Introduction
approach, Ultan (1978) examines the interrogative systems of 79 languages, and presents altogether 23 universals with reference to interrogatives, among which 3 deal with intonation, 3 with word accent, 10 with word order, and 7 with segmental elements. Chisholm (1984) collects information regarding interrogativity in 7 different languages. Bencini (2003) presents a diachronic typology of yes/no question constructions with particles. Dryer (2005a, b, c) conducts three general surveys of polar questions, the position of polar question particles, and the position of interrogative phrases in content questions using worldwide language samples. Rialland (2007) examines yes/no question prosody in 78 African languages. Miestamo (2011) focuses on polar interrogatives in Uralic languages by examining 20 Uralic languages from a typological perspective. Dixon (2012: 376-433) includes one chapter on the typology of interrogatives.
The list is too short compared to the studies of other major fields of linguistic typology, although it is by no means a complete list or a near complete one.
1 Introduction
1.3. Material
There are several theoretical and practical reasons for devoting the present thesis to the languages in China.
As noted, “China possesses rich linguistic resources which remain relatively untapped” (Chappell et al. 2007). Except for Standard Chinese, most other languages spoken in China have not been well represented in the theoretical literature. This thesis attempts to present an outline of the interrogative strategies of the languages of China, including in-depth discussions on topics related to interrogatives, in an effort to present a comprehensive view of the interrogative mechanisms at work in these languages.
The 138 languages with which this work is concerned include 10 Sinitic languages, 46 Tibeto-Burman, 22 Kam, 7 Hmong-Mien (Miao-Yao), 22 Altaic, 16 Austronesian, 9 Austro-Asiatic, 1 Indo-European, and 5 Creole, covering diverse language families.
The practical reason for choosing such a focus is that the languages of China, and the descriptive literature about them, are the languages with which I am most familiar. Although there is a potential risk that linguists may to some extent turn a blind eye to familiar material, the advantage of doing so are obvious, especially in analyzing the areal skewings and historical factors of certain structural features.
This work covers 138 languages currently spoken in China.3 Basically, they are classified into two groups: the Sinitic languages and the ‘minority’ languages.
The classifications and names of individual languages used in this work mainly follow those employed by Sun et al. (2007).
Sinitic languages
There are ten Sinitic languages of China: Cantonese, Gan, Hakka, Hui, Jin, Mandarin, Min, Ping, Wu, and Xiang, covering about 93% of the population of
3 Note that the 138 languages covered here do not comprise the total number of languages currently spoken in China. More languages have been and are being discovered. Some known cases include Ainu, Bumang, Sadu, Younuo, and Zhaba, which are collected in Sun (ed.) (1997– ).
1 Introduction
China (according to the 2010 census, China has a population of 1.37 billion).
Their geographical distribution and number of speakers are roughly demonstrated in Figure 1.1.
Figure 1.1. The Sinitic languages (source: Wikipedia; URL:
http://en.wikipedia.org/wiki/Chinese_language)
1 Introduction
Each language has various dialects or varieties. For example, Mandarin has basically six sub-dialects, including Beijing Mandarin, Jiang-Huai (Central-East) Mandarin, Central Mandarin, Northeastern Mandarin, Southwestern Mandarin, and Lan-Yin (North-Central) Mandarin. Each sub-dialect of Mandarin has millions of native speakers, with notable differences in phonology, lexicon, and to some extent also in grammar (see e.g. Cao 2008). It is not precise to use the label ‘Mandarin (Chinese)’ or ‘Beijing Mandarin’ to refer to the Sinitic languages, nor is it proper to refer to the standard Chinese language, Putonghua (Taiwan Guoyu).4 By definition, Putonghua’s phonological system is based on Beijing Mandarin, its vocabulary is drawn from the large and diverse group of Mandarin varieties spoken across northern, central, and southwestern China, and the grammar is based on modern good literary works written in vernacular Chinese. With regard to this, throughout this thesis, the label ‘Standard Chinese’
is used to refer to Putonghua, while individual Mandarin languages maintain their own labels.
Some comprehensive works on Sintic languages mainly include: Li et al.
(1987), a detailed atlas of the languages of China (a revised edition will be published soon); Li et al. (1991-2003), including 42 monographs on the lexicon of individual Sinitic languages; Hou et al. (1995-1999), a sound archive of 40 Sinitic languages; Huang et al. (1996), a dialectal grammar of Sinitic languages, grouping by grammatical topics; and Cao (2008), three volumes (phonology, lexicon, and grammar) of the linguistic atlas of Sinitic languages. (See also Chappell et al. 2007 for some studies and projects in the Western world.)
Minority languages
The minority languages of China include at least 128 languages from various families/groups. ‘Minority’ does not necessarily imply a small number of
4 Putong-hua ‘common-language’ and Guo-yu ‘national-language’ are basically the same, though there are some differences in the pronunciation of a small number of words (see e.g. Duanmu 2000: 263-7 for an overview of the phonology of Taiwanese accented Standard Chinese).
In fact, most Chinese people acquire their local ‘dialect’ (dialects of Sinitic languages) as their first language and Standard Chinese (Putonghua) as their second language.
‘Standard Chinese’ is a standardized artificial language that no one actually speaks, since everyone speaks it in more or less dialectal ways. In this sense, what people speak is a specific Sinitic language (or, dialect of Chinese), but not Standard Chinese. (According to Duanmu’s 2009: 86 estimation, only about 1% of Chinese people can speak Standard Chinese without any obvious accent.)
1 Introduction
speakers. Minority languages are spoken by 7% of the population of China, which is a minority in comparison with the other 93% (the Han Chinese), but China has a population of about 1.37 billion and many so-called minority nationalities actually have a large population. For example, Tibetan has about 3.3 million native speakers, Mongolian has 5.8 million, Uighur has 8 million.
Nevertheless, many minority languages do indeed have a small number of native speakers (e.g. Dulong, Bola, Langsu, Leqi), or are highly endangered (e.g.
Manchu, Hezhen, Gelao, Tujia, Xiandao), or died out in a recent history (e.g.
some Formosan languages in Taiwan). Sun et al. (2007) provide further information regarding the number of speakers and areas in which individual languages are spoken. The following table provides some basic information about the minority languages. (Note that the classification of languages follows Sun et al. 2007, not necessarily in accordance with the popular classifications, especially in Kam languages, Austro-Asiatic languages, and Hmong-Mien (Miao-Yao) languages.)
Some comprehensive studies on the minority languages of China include:
Sun et al. (1980-1987), which includes 57 monographs and documents 59 languages (revised edition in 6 volumes published 2009, with one language, Manchu, added); Sun (ed.) (1997- ) intends to publish the newly-discovered languages and has already produced more than 40 books; Thurgood and LaPolla (2003), which collects dozens of Sino-Tibetan languages of China and beyond;
and Sun et al. (2007), which introduces 129 languages5 of China in one big book.
A list of the languages of China is given in Table 1.2.
5 In Sun et al. (2007), “Chinese” is a general label for all the Sinitic languages, which is different from the present work.
