Florida State University Libraries
Electronic Theses, Treatises and Dissertations The Graduate School
2017 Yes-No Question Intonation in Puerto Rican Spanish and Beijing Mandarin Linxi Zhang
Follow this and additional works at the DigiNole: FSU's Digital Repository. For more information, please contact [email protected] FLORIDA STATE UNIVERSITY
COLLEGE OF ARTS & SCIENCES
YES-NO QUESTION INTONATION
IN PUERTO RICAN SPANISH AND BEIJING MANDARIN
By
LINXI ZHANG
A Thesis submitted to the Department of Modern Languages and Linguistics in partial fulfillment of the requirements for the degree of Master of Arts
2017 Linxi Zhang defended this thesis on April 07, 2017. The members of the supervisory committee were:
Carolina González Professor Directing Thesis
Lara Reglero Committee Member
Antje Muntendam Committee Member
The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.
ii
This thesis is dedicated to Dr. Carolina González for above all, convincing me that I can do this, and for all the support and guidance she offered me throughout the process.
iii ACKNOWLEDGMENTS
I would first like to thank my advisor Dr. Carolina Gonzalez for being such a patient and supportive advisor over the process of my research and thesis. She has also been an encouraging professor over the two years of my M.A. study who ignited my interest in Phonetics and
Phonology.
I would also like to express my gratitude for the other professors I have had at Florida
State: Dr. Leeser, Dr. Muntendam, Dr. Reglero, and Dr. Sunderman. I have learnt so much from them, not just knowledge, but also the passion and professionalism in what they are doing. I would like to thank Dr. Leeser for also being such a kind and helpful graduate advisor who listens to my problems with the greatest patience and always offers the best advices.
I also appreciate the constructive suggestions that I was given from Professor Martínez-
Gil, Professor Campos-Astorkiza, and Professor Erin O´Rourke and many others while presenting this project at the OSUCHill conference at the Ohio State University, as well as the very helpful comments I got from my fellow graduate students at the Terra Incognita Graduate
Student Conference at Florida State University.
This project would not have been possible without my participants who dedicated their precious time. I would like to give special thanks to my fellow graduate student Linda Núñez, for reading the scripts for my experiment and helping with recruiting participants. Also many thanks to my dearest classmates and friends with whom I have had so much good time even in the hardest struggles as a graduate student. I sincerely thank them for their love and support.
Last but not the least, I would like to thank my parents and other family members for their unconditional love and support in whatever I do.
iv TABLE OF CONTENTS
List of Tables ...... vi List of Figures ...... vii Abstract ...... ix
1. INTRODUCTION ...... 1
2. INTONATION ...... 10
3. TONE AND INTONATION IN CHINESE ...... 21
4. METHODOLOGY ...... 35
5. RESULTS-SPANISH ...... 48
6. RESULTS-CHINESE ...... 64
7. CONCLUSION ...... 77
APPENDICES ...... 85
A. INFORMED CONSENT FORM ...... 85 B. MATERIALS ...... 86 C. HUMAN SUBJECT APPROVAL LETTERS ...... 91
References ...... 92
Biographical Sketch ...... 98
v LIST OF TABLES
Table 2.1 Summary of intonational patterns of basic utterances in Puerto Rican Spanish from Armstrong (2010: 185-187) ...... 16
Table 3.1 Pitch contours and notations of lexical tones in Mandarin adapted from Lee (2005:11, Table 2.1) ...... 24
Table 3.2 Question categories considered in this thesis (*indicate the expected categories) ...... 32
Table 4.1 Participants information: Spanish ...... 44
Table 4.2 Participants information: Chinese ...... 46
Table 5.1 Distribution of final contours for information-seeking questions ...... 49
Table 5.2 Distribution of final contours for echo-surprise questions ...... 52
Table 5.3 Distribution of final contours for confirmatory questions ...... 55
Table 5.4 Distribution of final contours for echo-repetition questions ...... 59
Table 6.1 Distribution of different syntactic structures in each pragmatic context ...... 64
vi LIST OF FIGURES
Figure 2.1 Intonation of statements in Spanish (Northern Peninsular female) ...... 12
Figure 2.2 Intonation of wh-questions in Spanish (Northern Peninsular female)...... 12
Figure 2.3 Intonation of yes-no questions in Spanish (North Peninsular female)...... 14
Figure 2.4 Intonation of yes-no questions in Caribbean varieties of Spanish, by Puerto Rican speaker ...... 14
Figure 3.1 Pitch contour of syllable ma with 4 tones ...... 22
Figure 5.1 ¿Tus padres viven in Guayama? “Your parents live in Guayama?” Information- seeking question with L+H*L% ...... 51
Figure 5.2 ¿Hay bizcocho de vainilla? “Is there vanilla cake?” Information-seeking question with L+H*H% ...... 51
Figure 5.3 ¿Tu apartamento tiene lavadora? “Does your apartment have washer?” Information- seeking question with H+L*L%% ...... 52
Figure 5.4 ¿De verdad no tienen Mondongo? “You really don´t have Mondongo?” Echo-surprise question with L+¡H*L% ...... 53
Figure 5.5 ¿Guillermo de verdad se ganó la lotería? “Did Guillermo really win the lottery?” Echo-surprise question with L*L% ...... 54
Figure 5.6 ¿En verdad tienes que trabajar el domingo? “Do you really have to work on Sunday?” Echo-surprise question with H+L*L% ...... 54
Figure 5.7 ¿Fuiste a la bodega? “You went to the wine shop?” Confirmatory question with L+H*L% ...... 56
Figure 5.8 ¿Fuiste a la bodega hoy? “You went to the wine shop today?” .. Confirmatory question with L+H*H% ...... 57
Figure 5.9 ¿Fuiste a la Muralla Romana? “You went to the Roman Wall?” Confirmatory question with H*+LL% final contour ...... 57
Figure 5.10 ¿Tú vas para Aguagdilla? “(Did) you say you go to Aguadilla?” Echo-repetition question with L+H*L% ...... 60
Figure 5.11 Intonational variation across contexts P2 ...... 61
vii Figure 5.12 Intonational variation across contexts P1 ...... 61
Figure 5.13Intonational variation across contexts P4 ...... 62
Figure 6.1 Ni de yin liao xu yao jia bing ma? “Do you want ice in your drink? Information- seeking ma-question with H*H% ...... 69
Figure 6.2 Ni shi que de Huangshan ma? “Did you go to Mountain Huang?” Confirmatory ma- question produced with! H*H%...... 69
Figure 6.3 Xiao Wang, ni zhou liu hai yao jia ban ma? “Xiao Wang, you have to work (even) on Saturday?” Echo-surprise ma-question with H*H% ...... 70
Figure 6.4 Ni shi yao jin nian ba yue hui lao jia ma? “(Did you say) you are returning home in August?” Echo-repetition ma-question with H*H%...... 70
Figure 6.5 Qing wen you mei you Disanxian? “Do you have Disanxian?” Information-seeking A- not-A question with L*H% ...... 72
Figure 6.6 Xia wu you mei you ke? “Do you have class in the afternoon?” Information-seeking A-not-A question with L*L% ...... 72
Figure 6.7 Xiao Wang, ni shi bus hi gan mao le ya? “Xiao Wang, did you catch a cold?” Confirmatory A-not-A question with L+H*L% ...... 74
Figure 6.8 Shi bu qu na er gou wu la? “You went shopping somewhere?” Confirmatory A-not-A question with H*L% ...... 74
Figure 6.9 Shi bus hi you dian er gan mao? “Did you catch a cold?” Confirmatory A-not-A question with L*L% ...... 74
Figure 6.10 Jin tian shi bus hi you you men da xia chi? “(Did you say) we are esting shrimp today?” Echo-repetition A-not-A question with L*L% ...... 75
viii ABSTRACT
The intonation of yes-no questions in Puerto Rican and other Caribbean varieties of
Spanish has provoked great interest of the investigators, partly because of its unique circumflex contour that is different from the final rising contour common in other dialects (Sosa 1999;
Armstrong 2010, 2012). Previous research has also shown that in Puerto Rican Spanish, the nuclear accent in echo-yes-no questions that express surprise or ask for confirmation are represented with different tones than the information-seeking questions (Armstrong 2010).
On the other hand, the yes-no questions in Mandarin Chinese have been studied more for its syntactic variations. Two syntactic structures are believed to be alternative in the formation of
Chinese information-seeking yes-no questions: 1) the use of the sentence-final particle ma, known as the question marker and 2) the A-not-A structure (Huang et.al, 2009). Nonetheless, little is known about syntactic variation across pragmatic contexts and the intonation of the questions (Lee
2000, 2005).
The present study aims to investigate, above all, the intonational differences in yes-no questions of four different pragmatic purposes: information-seeking, echo-surprise, confirmatory, and echo-repetition, in PR Spanish and Beijing Mandarin Chinese. It also considers any syntactic variation across the question types, especially in Chinese. Lastly, it considers the effect of different degrees of bilingualism of the participants on their intonation.
For the study, an elicitation task with visual and audio guidance by means of a PowerPoint is used. The task has a Spanish section and a Chinese section. Each section consists of 20 contexts triggering Yes-No questions. Target items are divided into 4 blocks corresponding to the four contexts.
ix Spanish results show that as expected, most of the utterances were realized with falling intonation. At the same time, there are intonational differences among questions of different pragmatic contexts. Contradicting previous literature on PRS intonation (Armstrong 2010, Sosa
1999), the ´circumflex ´structure is preferred in information-seeking, confirmatory, and echo- repetition contexts, while echo-surprise context favors H*LL% final contour. In terms of bilingualism, the Spanish dominant speaker shows greater intonational variation across questions types. Some instances of rising intonation are attested probably due to influence of English or other varieties of Spanish.
Chinese results show syntactic variations in the questions of different pragmatic contexts.
The ma particle structure is favored in information-seeking and echo-repetition contexts, while A- not-A structure is preferred in confirmatory context. Yes-no question is scarcely found in echo- surprise context. In terms of intonation, there are effects of syntactic structure and narrow focus.
.
x CHAPTER 1
INTRODUCTION
This thesis is a comparative study of the intonation of Yes-no questions in Puerto Rican
Spanish and Beijing Mandarin Chinese It examines yes-no question intonation in different pragmatic contexts in the two languages, and the phonological and syntactic strategies that are involved. Chapter 1 addresses the guiding research questions of the study and reviews previous related studies. It also introduces the intonational framework used for the analysis of the data in order to provide a better understanding of discussion in subsequent chapters. Finally, an overview of the organization of this thesis is provided.
1.1 Guiding topics of research
The main objective of the thesis is to investigate how Yes-No questions are formed and their intonational properties in different languages. Yes-no questions are questions that requires yes/no as their answer, also known as polar interrogatives (Ginzberg and Sag, 2000), closed interrogatives (Huddleston & Pullum, 2002), polar questions (Ladd 1981; Buring & Gunlogson
2000), or absolute interrogatives (Sosa,1999). In English, yes-no questions are generally formed with the use of auxiliaries like do and subject-auxiliary inversion, known as the ‘do-support’, e.g.
Do you know him? (Kaplan 1989).
We focus on Spanish and Chinese, using Puerto Rican Spanish and Beijing Mandarin
Chinese as representatives of each language. More specifically, we focus on the intonation patterns of yes-no questions in these two languages, in both neutral information-seeking and echo contexts.
Neutral information-seeking questions are interrogatives that truly expect an answer in an unbiased, out of blue context, e.g. Do you know him? Echo questions, on the other hand, do not
1 expect an answer to a true question, but fulfill pragmatically different functions, such as asking for confirmation or repetition, or express the speaker’ emotion and/or attitude towards the meaning of an utterance, such as surprise or disbelief. (Noh 2005), e.g. a friend says that he knows Donald
Trump in person --- Do you know him? (with interpretation of surprise or disbelief).
Recent studies have shown that intonation differs between echo questions and information- seeking questions, as well as among different types of echo questions. Gonzalez & Reglero (2016) examined the intonation of information-seeking, echo-surprise, and echo-repetition wh-questions in North Peninsular Spanish. They found that rising contours are present most often in information- seeking questions and least often in echo-repetition contexts. They also found that global and local pitch range and the duration of the wh-word are the greatest in the echo-surprise questions, followed by echo-repetition and information-seeking questions.
In this thesis, we examine 3 types of echo questions: surprise, confirmatory, and repetition.
See examples (1) to (3):
(1) John tells Mary he knows Donald Trump in person, Mary was surprised by this, and
says:
“Do you know him?”
(2) John sees a photo of his friend Peter on Mary’s desk, he guesses they must know each
other, and says to Mary:
“(Do) you know him?”
(3) John tells Mary that he knows her neighbor Tim, Mary did not hear clearly due to the
noise, and says:
“(You said) you know him?”
2 This thesis aims to describe the main intonational properties of information-seeking and these 3 types of echo-questions in both Spanish and Mandarin Chinese. As we know, Chinese is a tonal language (with lexical tones that change the meaning of words) and is rich in syntactic variation in its interrogative questions (Huang et al. 2009). On the contrary, Spanish is an intonational language and its yes-no questions share mostly the same syntactic structure with its statements (Sosa 1999). Therefore, it is interesting to compare how the different pragmatic meanings of yes-no question are expressed in these two languages.
In addition, we examine the specific roles of phonological and syntactic cues in the question forming in the two languages, and aim to access the interaction between phonology and syntax in the variation of intonation, especially in Chinese.
Besides, since Puerto Rican Spanish is one of the Spanish varieties that have the most contact with English, we will also look at the effect of speakers’ status of bilingualism on the yes- no question intonation. The participants of the study include native speaker of PRS, early bilingual speaker, and heritage speaker. We will examine if there is any difference in their intonational patterns in the distinct types of Yes-No questions.
1.2 Related studies
Intonation, as defined by Ladd (1996), is the use of suprasegmental phonetic features to convey ‘postlexical’ or sentence-level pragmatic meanings in a linguistically structured way.
Intonation has many functions, one of them is to distinguish meanings such as interrogative and declarative. (Bolinger 1989, Ladd 1996). In the following conversation in English: “What would you like to drink? Water?” “Yes, water.”, the same word ‘water’ is said with a rising intonation and falling intonation respectively, to indicate if it’s a question or a statement.
3 Previous cross-linguistics studies have found that there seem to be some universal tendencies of intonational form-meaning mapping, the so called “intonational universals”. One good example of an “intonational universal” is the common observation that in most of the languages, there seems to be a consistency in the meanings associated with rising tones versus falling tones: the falling tones are associated with statements, while the rising tones are associated with questions. This hypothesis is borne out by the fact that in most languages, declaratives show a declining pitch contour from the beginning to the end of the sentence, while yes-no questions usually show a final rise. (Bolinger 1978; 1986; 1989; Cruttenden , 1986; Miller and Tench 1981, among others).
However, some languages or dialects, such as PRS and other Caribbean varieties of
Spanish have been found to be an exception of the ‘intonational universal’ mentioned above. While most other Spanish dialects including Castilian Spanish and Mexican Spanish show final rising intonation in yes-no questions (Face 2008, de la Mota, Butragueño & Prieto, 2010), PRS typically show terminal falling tone (Cruttendem‐ 1986;‐ Quilis 1987, 1993; Sosa 1999, 2003;
Armstrong 2010, 2012).
This falling intonation has been reported in early literature like Navarro Tomas (1944) and
Quilis (1987, 1993). Later, it was described with more detail in Sosa (1999) under the Auto- segmental Metric (AM) model, in comparison with other dialects of Spanish. As the Sp_ToBI system was developed and revised to better account for the intonation labeling in Spanish,
Armstrong (2010) presented a more comprehensive description of intonation in PRS as a chapter of the Transcription of Intonation of the Spanish Language project (Prieto & Roseano (eds.),
2010). The chapter applied the revised version of the Sp_ToBI (Estebas-Vilaplana and Prieto ,
2008). It listed the 3 monotonal and 6 bitonal tones that PRS consists of, and analyzed the
4 intonation contour in basic utterances in PRS such as broad and narrow statements, broad-focused and echo-yes-no questions, wh-questions, among others.
However, the chapter on PRS did not provide data for the echo-repetition type of questions.
At the same time, previous works on PRS are based on corpus from native Puerto Rican speakers, there is hardly any research on the intonation of bilingual population. This thesis aims to fill these gaps.
On the other hand, less is known about the intonation of yes-no questions in Beijing
Mandarin Chinese (BMC). BMC is spoken natively in the capital of the People’s Republics of
China and vicinity. It is the most prestigious dialect of modern Chinese and the basis for Modern
Standard Chinese (MSC), and is logically one of the most studied varieties of Mandarin Chinese
(Wang. X, 1992; Wang. J, 2003;). However, works on its intonational patterns are rather scarce.
Early works like Karlgren (1918) claimed that because tonal languages such as Chinese make use of sentence final particles for functions that are often provided by intonation in English, MC lacks intonation. Such assumption is still held by many scholars today. Besides, as Lee (2000) summarized in her research on the intonation of ma-particle questions in MC, early prosodic studies in Chinese tend to focus on the interaction between tone and intonation instead of intonation itself.
With a few exceptions, hardly any research has been done on the pragmatic functions of
Chinese questions and their relationship to intonation. Some early works attempted to compare the intonation of Chinese neutral information seeking questions and echo questions (Chao, 1933a,
1968; Ho, 1977; Lin 1985; etc.) However, their conclusions were very much divided in terms of whether the rising movement of pitch is global (Chao, 1933a; Ho, 1977; etc.) or local (Liu et al.
5 1983; Yuan, 1993; etc.), and whether the echo questions have declarative-like intonation (Wu,
1982; Hu, 1991; etc.) or show an overall rise (Shen 1990).
Moreover, the majority of these studies were based only on impressionist observation, without any experimental evidence. There also lacked a standard and systematic tonal labeling framework that can efficiently describe the tone and intonation in Mandarin Chinese. In recent years, as the M_ToBI system for Mandarin was proposed and revised (Peng et al. 2005), it becomes feasible to give a clearer description of the tone types in Chinese intonation. Nonetheless, still little work is being done on this aspect. Therefore, two goals of the study are to provide production data of BMC yes-no questions, and to describe the tones using the M_ToBI labeling system.
1.3 Labeling frameworks
1.3.1. The AM theory. The autosegmental-metrical theory (AM theory) began developing in the late 1970s, based on PhD theses by Liberman (1975), Bruce (1977), and Pierrehumbert
(1980). It represents pitch contours phonologically as sequences of discrete intonational events.
For intonation languages like English and Spanish, it distinguishes two main types of such events: pitch accents and edge tones (boundary tones). Pitch accents are associated with prominent syllables in the segmental string; and edge tones are associated with the edges of prosodic domains of various sizes. Pitch accents and edge tones are analyzed using level tones or pitch targets: High
(H) and Low (L). (Ladd 1996:42).
Within the AM model as currently applied to Spanish, there are two general types of pitch movements (i.e. changes in F0) associated with stressed syllables called pitch accents: the nuclear pitch accent (pitch movement on the final stressed syllable) and prenuclear accents (pitch movements associated with non-final stressed syllables) (Henrisken 2010). It is observed that in
Spanish, the prenuclear accents, especially the first one, tend to have a delayed peak that is located
6 in the postonic syllable, while the nuclear accents fall within the last stressed syllable (Hualde
2014, Prieto & Roseano 2010).
1.3.2. The ToBI transcription framework. ToBI (Tones and Break Indices, Beckman et al. 2005) was first initiated for Mainstream American English as a system for transcribing the intonation patterns and other aspects of the prosody of English utterances. It was devised by a group of speech scientists from various different disciplines (electrical engineering, psychology, linguistics, etc.) who wanted a common standard for transcribing an agreed-upon set of prosodic elements, in order to be able to share prosodically transcribed databases across research sites in the pursuit of diverse research purposes and varied technological goals. (Beckman & Elam, 1993).
Since the prosodic system in each language has its own characteristics, later on, ToBI systems for other languages have been created. Up to this day, complete systems for 14 languages and some of their varieties have been established, including Korean (Jun 2006), Catalan (Prieto, P. et al.
2009), Spanish (Beckman et al. 2002), and Mandarin Chinese (Peng et al. 2006).
Following the AM theory, the system uses two tones: L (low) and H (high) to represent the prenuclear and nuclear pitch accents. Each content word of the sentence should carry a pitch accent and an asterisk is used to indicate where the pitch accent (or tones) starts in relation to the stressed syllable. The nuclear configuration is a combination of the last (or nuclear) pitch accent and the final boundary tone. Intermediate phrases and word boundaries are marked with four numerical break indices (Beckman, et al. 2006: 24).
The tones and break indices model for Spanish is known as Sp_ToBI (Beckman et al.
2002). The system was created with the intention to apply to all dialects and varieties of Spanish.
In the original SP_ToBI, 6 tiers were included in the annotation of Spanish utterances: words tier,
7 syllable tier, break indices tier, tones tier, misc tier and code tier. The annotation of data in this thesis basically follows this format.
The ToBI system for Mandarin Chinese is known as the M_ToBI. Mandarin Chinese is spoken as first or second language by well over a billion people, and encompasses many varieties in mainland China, Taiwan, and Singapore. Along the years, independent efforts have been made to develop systems for tagging Mandarin corpora within the ToBI framework. These efforts have bear some fruits including one system developed at Academia Sinica (AS) in Taipei, which was also adopted by the Institute of Linguistics at the Chinese Academy of Social Sciences in Beijing
(Li et al. 1999). The other was developed in the Ohio State University (OSU) in Columbus. In the need of accounting for the many varieties of Mandarin Chinese with a unified convention, an initiation was started towards a Pan-Mandarin System for prosodic transcription by merging the two systems. (Peng et al. 2006). The preliminary merged system presented by Peng et al. (2006) consists of 8 tiers: words, Romanization (pinyin), syllables (with resilabification), stress, Sandhi
(change of a tone as result of adjacent tones), tones (H/L), break indices, and code (marking code- switching, etc.) Given that the current study and analysis do not require such detailed information about elements like lexical tones, tone Sandhi, and stress, here we only apply a simplified system similar to the Spanish ToBI system.
1.4 Contribution
As one of the very few studies that look at the intonation of Echo-yes-no questions, this thesis will contribute to the field by presenting and describing data of the intonation patterns in
Echo-Yes-no questions. It is also probably the first work that compares Spanish and Chinese intonation. Bringing these two languages together could provide us with a better understanding of the degree to which different languages would apply similar or distinct strategies in questions
8 formation and intonation manipulation. The analysis of syntactic variation in Chinese yes-no questions will reflect instances of Syntax/Phonology interface. And the inclusion of bilingual speaker in the Spanish part of the study will shed light on the influence of bilingualism on intonation.
Apart from the theoretical contribution, this research may also provide us with some pedagogical implications. The potentially different strategies used in the two languages for question formation might suggest certain difficulty in the acquisition of Spanish question intonation by the native Chinese speakers, or vice versa.
1.5 Organization
This thesis contains seven chapters organized as follows. Following the first chapter,
Chapter 2 reviews and discusses the previous literature on the definition and function of intonation and intonation in Spanish. It also summarizes previous findings on intonation in Puerto Rican
Spanish. Chapter 3 introduces tone and intonation in Chinese, explaining the interaction between the two elements, as well as the syntactic categorization of Chinese questions and their pragmatic uses and intonation. Chapter 4 introduces the research methodology used in the study, including information about the materials, task, participants, procedure, and data analysis. Chapter 5 presents the results from the Spanish data, focusing on the pitch contour and the effect of level of bilingualism on intonation. Chapter 6 presents results from the Chinese data, focusing on the syntactic variation in yes-no questions of different pragmatic uses and their intonation pattern.
Chapter 7 wraps up the main findings of the study and talk about limitations and future studies.
9 CHAPTER 2
INTONATION
This chapter reviews previous literatures on intonation. Section 2.1 talks about its definition and functions, Section 2.2 gives an overview of the intonation in Spanish, and Section
2.3 reviews studies on intonation in Puerto Rican Spanish.
2.1 Intonation: definition and functions
Intonation, as defined by Ladd (1996), is the use of suprasegmental phonetic features to convey ‘postlexical’ or sentence-level pragmatic meanings in a linguistically structured way.
These suprasegmental phonetic features are mainly the variation of the Fundamental Frequency
(F0) (Bolinger 1955, Quilis 1981, among others), as well as other variables like duration and intensity (Koch and Kennedy 1963). Variation of F0 is perceived as the rise and fall of pitch (Jones
1909), and is produced with the increase and decrease of the airstream and tension of the vocal cords. Pitch rises when the tension increases, and falls when it decreases (Quilis 1981).
Intonation has many functions. Linguistically, it conveys linguistic information that can be distinctive (to distinguish statement vs. interrogative; finite vs. infinite), integrative (to integrate words to form a sentence), and delimitative (to delimitate the sentences and separate the speech to certain number of units for it to be easier pronounced and understood) (Quilis 1981).
Extralinguistically, intonation as a means of communication can convey attitude, emotion, and meaning (Trager and Smith, 1935; H. van Schooneveld, 1961; among others). It also carries personal information (characteristics like age, sex, personality, etc.) as well as sociolinguistic information of (origin geographic, socio-class, education level, etc.) the speaker (Quilis 1981). In
Spanish, Prieto et al. (2011) found that Peninsular Spanish also makes sub-divisions in question types through the use of intonation. In Central Catalan and Buenos Aires Spanish, increased pitch
10 expansion is used either globally or locally for incredulity questions than in unbiased yes-no question. (Catalan: Crespo-Sendra 2011; BAS: Ar Lee et al. 2008).
2.2 Intonation in Spanish
Contrasting tonal language like Chinese, (see Chapter 3), Spanish is an intonational language, meaning that the variation of pitch does not change the meaning of words, but can modify the pragmatic meaning of a sentence. (Hualde 2014, Navarro 1944, among others). The same sentences can be pronounced with different intonation to convey different meanings. E.g
Llegaron los amigos. vs. ¿Llegaron los amigos?
Statements in Spanish, like in many languages, are characterized by an overall descending tendency, referred to as ‘declination’ (Hualde 2014, Sosa 1999). Despite of the general declination, the first prenuclear pitch accent has been reported to be ascending with delayed peak in the postonic syllable (L+>H*). Depending on the dialect, the nuclear accent can be ascending to a down-stepped peak (L!H*) or low (L*). The final boundary tone is typically a low tone (L%).
(Hualde 2014:265). See Figure 2.1 for example.
Questions in Spanish are generally divided into wh-questions and yes-no questions. Wh- questions are questions that include a wh word e.g. ¿Qué dice Mariano? (What does Mariano say?)
Yes-no questions are questions that requires yes/no as answer, often referred to as absolute interrogatives ‘interrogativas absolutas’ (Sosa,1999) in the Hispanic linguistic convention. E.g.
¿Ves a Mariano? (Do you see Mariano?).
Similar to the declaratives, the wh-questions typically have a descending intonation. The pitch reaches to its highest at the first accented word (usually the wh word) and gradually declines to a L*L% final contour. (Sosa 1999:144).
11
Figure 2.1 Intonation of statements in Spanish (Northern Peninsular female)
Figure 2.2 Intonation of wh-questions in Spanish (Northern Peninsular female)
In Spanish wh-questions, the interrogative meaning is carried partly by the wh-word and the subject-verb inversion, so intonation is not the only cue that tell question part from statements.
The same is true in yes-no questions in many languages. As stated by Sosa (1999), in many other languages, cues other than intonation are used to tell questions apart from statements, such as particles, affix, verbal morphemes, and changes in sentence order. In Mandarin Chinese, for 12 example, sentence-final particle ‘ma’ (glossed as Q). is largely used to convert a statement into yes-no question. See example (1) (See also Chapter 3 for more knowledge about Chinese yes-no questions).
(1) a. Ni renshi ta.
You know him
‘You know him’
b. Ni renshi ta ma?
You know him Q
‘Do you know him?’
In Spanish, however, syntactic strategies are not as prevalent. Possible syntactic variations include subject-verb inversion, insertion of tags like ´que´ (that), ´de verdad´ (really) and/or negation word ´no´ at the beginning of the questions. (Hualde and Olarrea 2014; Reglero 2007).
Except for these few strategies, intonation is basically the only manner of distinguishing yes-no questions from statements.
Studies on various dialects of Spanish including the Castilian and Mexican varieties have shown that Spanish yes-no questions typically have a ‘suspension of declination’. The pitch stops from going down and show a final rising intonation, labeled H*H%, L*H%, L*+HH%. (Face 2008,
Hualde 2014, Prieto & Roseano 2010, Sosa 1999, etc.). Navarro Tomas (1944) stated that the suspension of declination in interrogative intonation is the crucial factor that makes it possible for the interlocutor to distinguish a question such as ‘¿Estuvieron esperando?’ from the statement
‘Estuvieron esperando ‖ hasta las cinco de la tarde’, even if the declarative sentence is not finished.
See Figure 2.3 for example.
13 Figure 2.3 Intonation of yes-no questions in Spanish (Northern Peninsular female)
However, there exist some dialectal differences. In particular, it is reported that yes-no questions in the Canarian and Caribbean dialects, including the Puerto Rican Spanish, have final falling intonation, known as the ‘circumflex’pattern (Hualde 2014, Quilis 1987, Sosa 1999), labeled as L H*L%. (Figure 2.4).
Such pattern is also prevalent in yes-no questions in Bilbao Spanish, spoken in the Basque county in Northern Spain (Gonzalez & Reglero, under review; Robles Puente 2011, 2012).
Figure 2.4 Intonation of yes-no question in Caribbean varieties of Spanish, by Puerto Rican speaker
14 In addition to dialectal differences, question type has also been found to have an effect on the tone choices in Spanish yes-no questions. Sosa (1999) stated that even in peninsular varieties, the ending is not necessarily rising. The choice of final rising or falling depend on factors such as motivation, pragmatic contexts, and other extra-linguistic factors. Navarro Tomas (1944) distinguished several different subclasses of questions. Among them are Relative Question (biased information-seeking questions that expect certain positive or negative answer), Reiterative
Question (repeating what is said to assure proper understanding), and Hypothetic Question (to confirm a hypothesis). He pointed out that intonation of each of these type of questions is somehow different. Quilis (1981) also intonational variation in yes-no questions of different pragmatic uses.
He gave neutral information-seeking yes-no question a default final rising intonation. However, he noted that in Reiterative Questions, Relative Questions, and Confirmative Questions, final falling pattern is more common.
2.3 Intonation in Puerto Rican Spanish
The intonation of PRS was first studied as a variety of Caribbean Spanish in works of cross- dialectal comparison (Quilis, 1987, 1993; Sosa, 1999). Sosa (1999) compared intonation of speakers from several cities including Buenos Aires, Madrid, Ciudad de Mexico, and San Juan,
Puerto Rico for the basic utterances. He noticed the prevalence of L*+H prenuclear tone and
L*+L% boundary tone in the statements in PRS, showing similar pattern with some other
American varieties such as the Habana variety, and Peninsular varieties of Sevilla, Pamplona, and
Madrid. In wh-questions, PRS shows a H*H prenuclear tone and a H+L*L% nuclear configuration.
It resembles the other dialects under study in terms of general curve shape. The only difference is that in PRS it maintains the H tone until the last stressed syllable and the falling largely concentrates within the nuclear accent, while in the Caraquenian dialect, the falling starts earlier.
15 However, it is in the yes-no question intonation that PRS deviates the most from other dialects, especially from the non-Caribbean varieties. As is discussed in the previous section, a final rising intonation has been found in yes-no questions in the majority of the Spanish dialects. In PRS, on the other hand, a final falling tone is used to mark the question, similar to its declarative nuclear configuration (Quilis, 1987, 1993; Sosa, 1999; Armstrong, 2010, 2012 ;)
Recent work on the intonation of Puerto Rico highlights the chapter of Puerto Rican
Intonation (Armstrong 2010) as part of the Transcription of Spanish Intonation project initiated by Prieto and Roseano in 2010. In this chapter, the author analyzed data from an elicitation task with questionnaire and included much wider array of utterance types: broad and narrow focus declaratives, yes-no questions, wh-questions, commands, etc. She described the tone patterns in
PRS for different utterances, summarized as in Table 1:
Table 2.1: Summary of intonational patterns of basic utterances in Puerto Rican Spanish (Armstrong 2010: 185-187). 1 Sentence Prenuclear Nuclear Boundary Nuclear Notes Type Tone Tone Tone Configuration Broad-focus L*+H ¡H+L* L% ¡H+L*L% M boundary tone found statements at intermediate phrase boundary Narrow focus L+>H* L+H* L% H*L% / statements L+H*L%
1 Table 2.1 is reproduced with the permission of the author Meghan. Armstrong.
16 Table 2.1 continued Sentence Prenuclear Nuclear Boundary Nuclear Notes Type Tone Tone Tone Configuration Exclamative L*+H L* / M% / L% L*M% / statements L+H* L*+HL%
Yes-no L*+H H* ¡H* L% ¡H* L% question Echo Yes-no L*+H L+¡H* L% L+¡H*L% Like narrow focused (surprise) statements &exclamative but higher rise Echo Yes-no L*+H L* HL% L*¡HL% (incredulity) Imperative L*+H ¡H* HH% ¡H* HH% (polite) yes- no Confirmation L*+H H* H+L* L% H+L* L% With negation e.g. No yes-no hay un lugar que vende (positive) piononos? Tag questions L*H%
Info-seeking L*+H H* H+L* L% H+L* L% wh-question Echo-wh L+>H* L+¡H* L% L+¡H*L% (clarify) Command L* HL% L*HL%
17 As for yes-no questions, PRS is reported to have a falling intonation in the yes-no questions
(Armstrong 2010, 2012; Quilis 1987, Sosa 1999, among others). Based on the utterances from a speaker from San Juan, Sosa (1999:204) described yes-no questions in PRS as showing
‘suspension of declination’ in the prenuclear accents, labeled as L*+H or H*. The pitch then rises to a high peak at the syllable nucleus and drops to a low target (H+H*L%). Alternatively, the pitch does not rise again at the nucleus but maintains high before it falls at the end (H*L%).
Armstrong (2010) confirmed the findings in Sosa (1999) but found the first type of contour
(H+H*L%) being more common. However, she points out that the pitch rises at the syllable nucleus to an up-stepped high peak located around the mid-point of the nucleus, labeled as ¡H*, forming a nuclear configuration of ¡H*L%.
In terms of echo-yes-no questions, Armstrong (2010) found the typical ‘circumflex’ pattern in echo-questions that express surprise, with an up-stepped peak at the syllable nucleus
(L+¡H*L%). This is not surprising since Sosa (1999:232) proposed the rise-fall ‘circumflex’ patter, labeled as L*+HL% for exclamative, e.g. ¡Ellos entran en Cuba y se pierden! (They entered Cuba and get lost!, or ‘expressive/emotional wh-questions’ (translation mine) e.g. ¿Con quién llegó a la fiesta? (With whom (he/she) got to the party!). The question is probably produced in a context where a person arrives at a party with someone famous, and is uses to express admiration and surprise. As Armstrong (2010) points out, given the semantic similarity between exclamatives and echo-questions, the similarity in their intonation is expected. Some types of biased yes-no questions with positive or negative presupposition were also studied. For Confirmation questions, a falling tone through the syllable nucleus was found, followed by a low boundary tone (
H+L*L%).
18 These proposed for neutral and non-neutral yes-no question intonation in PRS were verified in Armstrong (2012). The author looked at the semantic-pragmatic aspect of PRS interrogative intonation from speaker and hearer’s perspective. She confirmed that speakers rely on the default contour ¡H*L%, for questions that does not encode any sort of bias. However, speakers have the option of encoding bias directed through choice of intonation contour. The
L*HL% configuration is used to express incredulity and disbelief. The H+L*L% configuration, on the other hand, has a positive belief value, used to transmit a personal belief.
Based on previous studies, this thesis aims to get a more comprehensive picture of the intonation of yes-no questions of different pragmatic uses in PRS by re-examining the contexts of information-seeking and echo-surprise, and at the same time exploring two other contexts: confirmatory (somehow different from Confirmation questions), and echo-repetition questions
(See definitions of the contexts in Chapter 4 Methodology).
2.4 Bilingualism and intonation
In previous research, language contact has been shown to affect lexicon, syntax, and prosody. For Spanish, contact with languages like German or Quechua has been found to affect prosodic cues such as peak alignment (Lleó, Rakow and Kehoe 2004, O’Rourke 2003, among others). Intonation may also be affected by the contact language. Alvord (2010) observes that second-generation Cuban immigrants in Miami might be transferring the final rising intonation from English to Cuban Spanish yes-no questions. Robles Puente (2012) found young bilingual speakers from Guernica, Basque using very similar intonational strategies in Spanish and Basque to convey two distinctive pragmatic meanings (statements vs questions). They apply rise-fall
‘circumflex’ final contour to Spanish yes-no questions probably due to influence of Basque, a language that shows final falling intonation in yes-no questions.
19 The large Puerto Rican population in the U.S provide a good source for us to understand the role of bilingualism in intonation variation. In this study, two of the participants (P1 and P4) reported American English as their dominant language. Both were both born in Puerto Rico but moved to the U.S at a very young age: P1 at 14 months, and P4 at 3 years old. According to Alvord
(2010), both of them should be categorized as second generation immigrants. Therefore, we would expect them to show some probable effect of language contact.
20 CHAPTER 3
TONE AND INTONATION IN CHINESE
3.1 Lexical tones in MC
Lexical tones involve the use of pitch to distinguish the meaning of words Yip (2002). The difference between lexical tone and intonation is that the former conveys meaning at a lexical level, while the latter does so at the phrasal or sentential level. Although tone and intonation have different domains, the primary acoustic correlate of both is the Fundamental Frequency (F0), the lowest resonant frequency of the vocal cords (Duanmu 2000, 2007; Shih 1988; Shen 1990; Xu
1999; Yip 2002; among others).
Early works looking into its articulatory mechanism hypothesized that tone is controlled by a single articulatory mechanism: the vocalis muscles which controls the tension of the vocal cords (Halle and Stevens, 1971). Later, another mechanism, the cricothyroid, was identified to be responsible as well in tone production. Zemlin (1981) demonstrates that the two mechanisms function together. The cricothyroid controls the elongation and thickness of the vocal cords, thus controlling the F0 of its vibration, causing changes in pitch. The vocalis muscles control the
‘isometric tension’ of the vocal cords, which corresponds with register (stiff/slackness of vocal quality).
According to Yip (2002:1), an estimate of 60 to 70 percent of the world’s languages is tonal. Mandarin Chinese is a typical tonal language in that the pitch contour over a syllable can distinguish word meanings. In Mandarin Chinese, there are 4 tones on full syllable, which are often referred to as Tone 1 through Tone 4. Tone 1 is a high-level tone, tone 2 is perceived as a rising
21 tone, tone 3 is referred to as the ‘fall-rise’ tone, with a perceptively rather salient dip of the pitch, and tone 4 is a falling tone (Duanmu 2000, 2007).
The following figure of the author pronouncing the same segment with the 4 different tones shows their pitch contours:
mom hemp horse scold
Figure 3.1 Pitch contour of syllable ma with 4 tones
With the differentiation of the 4 tones, the same segments in Chinese can have different meanings: mā – mother, má – hemp, mă – horse, mà – to scold. In addition to the 4 tones, it is observed that in Chinese, the unstressed syllables are toneless. One of the examples is the sentence- final question-particle ma. This particle is unstressed and realized without any of the 4 tones. Some researchers refer it as a neutral tone, or Tone 0 (Chao 1968, Duanmu 2000, 2007). Chao (1968:36) found that the tone of the weak syllable depends on the tone of the preceding syllable. He further pointed out that T0 is high after T3, and low after other tones.
22 Tones in Chinese has been transcribed in many ways. Early attempts of labeling the four tones highlight Chinese linguist Chao Yuen Ren’s tone letters (Chao 1930, 1968). The letters consist of a vertical bar on the right indicating the pitch range, and a contour line on the left indicating pitch movement relative to the pitch range. In this system, Tone 1 to Tone 4 are transcribed using the following symbols: , , ˩ , ˩ (Chao 1930, 1968; Duanmu 2000, 2007). The
Chao tone letters were later adopted by the International Phonetics Association (IPA). A modified version of the tone letters digitalizes the pitch range with a five-level numerical scale (Chao 1933,
1968). In this scale, number 1 to 5 are used to relatively represent the value of a speaker’s pitch range, with "1" being the lowest pitch value and "5" being the highest pitch value. Thus, Tone 1 through Tone 4 are represented as "55", "35", "214", and "51", respectively. This system was later adopted by the M_ToBI labeling system (Peng et al. 2006).
During the nineteenth century, several transcription systems were invented and utilized by occidental scholars and missionaries to transcribe Chinese words using Latin letters, among them the Wade–Giles system produced by Thomas Wade in 1859 (Ao, 1997). Inspired by those systems, the Pinyin system was developed in the 1950s and has been widely used as the official
Romanization system in the People's Republic of China (Lee 2005). Different from Chao’s digits, this system marks the tones using iconic diacritics above the letters, as in ā, á, ă, and à, for Tone 1 to Tone 4 respectively. 2
Lee (2005) labeled the 4 tones using H and L and their combination to describe the ‘’Tone
Target’’, with H referring to a high tone, and L referring to a low tone. She summarized the pitch contours and notations of the 4 tones in the following table:
2 The Chinese examples in this thesis are written in Pinyin, with the exclusion of diacritical tone symbols.
23 Table 3.1 Pitch contours and notations of lexical tones in Mandarin adapted from Lee (2005:11, Table 2.1) Tone 1 Tone 2 Tone 3 Tone 4 Tone Shape High-level High-rising Low-dipping High-falling Tone Target H LH L(H) HL Pitch Value in Chao’s 55 35 214 51 Scale Tone Diacritics in Pinyin ā á ă à
As introduced above, the M_ToBI system uses Chao’s digits to label tones in its annotation.
However, for the purpose of convenience, in this thesis, the 4 full tones and the ‘neutral tone’ are represented with numbers from 0 to 4 attached to the end of each syllable, for example ma1, ma2, ma3, ma4, ma0. Tones of each syllable are marked in the syllable tier in the annotation in Praat.
3.2 Interaction between tone and intonation
From the previous introduction, we can see that in a tonal language like Chinese, the pitch contour is greatly influenced by lexical tones. Keeping that in mind, a question then arises: in languages like Chinese, since word tones are lexically determined and there is little flexibility in varying Hs and Ls independently, how does it express intonational meaning, like statement, doubt, surprise, query, command, etc. (Duanmu 2000, 2007)
Chao (1933) proposed that tone and intonation can be combined through 2 ways of tonal
‘addition’: successive addition and simultaneous addition. Successive addition is when a rise or a fall is added to the end of an utterance, and the resulting syllable will be lengthened in order to carry an extra tone. (See (1) and (2) from Duanmu (2000, 2007)). Simultaneous addition, on the other hand, predicts that intonation is superimposed on lexical tones and the pitch range of an
24 utterance is raised, lowered, expanded, or compressed as a result of the co-function of tone and intonation. In other words, if a rising intonation is applied to a falling tone (e.g. Tone 4), the pitch contour is still falling to maintain the lexical meaning of the word, but does not fall quite so low; on the other hand, if a rising tone (Tone 2) appears at the end of a statement, where the intonation is falling, the pitch contour is still rising but no quite so high. As Chao (1933) describes, the final pitch contour is the algebraic sum or resultants of two factors, the original word tone and the sentence intonation proper. This hypothesis has been supported by phonetic studies of S. Shen
(1989), He and Jing (1992)), among others, which found that question intonation raises the pitch height of the entire sentence, not changing the distinctiveness of word tones.
(1) Tone Intonation LH + L → LHL nan nan ‘difficult’ ‘affirmation’ ‘Surely difficult!’
(2). Tone Intonation HL + H → HLH mai mai ‘sell’ ‘question’ ‘Sell?’
(examples of successive addition, Duanmu 2000:250-251, (37), (38))
At the same time, it is believed that in Chinese, syntactic cues are also used to express sentential meaning. (Chao 1933; Lee 2000, 2005). Chao (1933) suggests that many functions of intonation in other languages are fulfilled in Chinese by the use of particles. In fact, Chinese is a language that is rather rich in particles. Thompson and Li (1981) summarized the usage of 5 particles commonly used in Chinese to express pragmatic meanings such as interrogative (ma
25 [ma]), suggestion and solicitation of agreement (ba [ba]), friendly warning (ou [o]), interest/exclamation (ne [nə]), and so forth. What is not explained here is whether these particles in expressing pragmatic meanings are used together with prosodic cues like intonation, or to the exclusion of it.
When creating the materials for this thesis, effort was made to control the influence of lexical tones on intonation. Most of the contexts were created so that the intended answers have their last accented syllable in Tone 1 (the level-high tone). However, we are aware that in the elicitation task, participants might produce sentences ending with other tones. Therefore, although tone and intonation interface is not the focus of this thesis, hypotheses for the interaction between them will be applied to explain any abnormality of the pitch contour shape.
3.3 Categorization of Chinese questions
Statement in Chinese follow a SVO order, as in (1):
(1) Ni renshi ta
You know him
‘You know him’.
Unlike in some languages in which interrogative sentences have different word order,
Chinese questions follow the same order, and are classified into three categories: (i) yes-no questions (‘particle questions’); (ii) disjunctive or alternative questions, and (iii) wh-questions
(‘constituent questions’) (Huang et al. 2009, Li and Thompson 1981, Zeng 1996, among others).
An example of each of these categories is provided as below (from Huang et al. 2009:236 examples
(1)-(3)):
(2) Ni renshi ta ma?
26 You know him Q3
‘Do you know him or not?’
(3) Ni xiang chu-qu kan dianying haishi zai jia da majiang?
You want go out see movie or at home play mahjong4
‘Do you want to go out to see a movie or stay at home and play mahjong?
(4) Ni xiang gen shui shangliang zhe-jian shi?
You want with who discuss this-CL thing?
‘Who would you like to discuss this matter with?’
As is illustrated in the (3), in Chinese, the disjunctive question is formed by coordinating two sentences with the conjunction ‘haishi’ (or). Wh-questions are formed using wh words that stay in-situ, like the ‘shui’ (who) in example (4). Yes-no question, on the other hand, is formed by simply adding a sentence-final Q particle ma to the end of the corresponding statement. (See example (2). As mentioned in section 3.1, this particle is unstressed and toneless. Its tone depends on the tone of the preceding syllable. Out of various sentence-final particles in Chinese, ma is the least controversial in its meaning, due to its strong interrogative flavor that marks the utterance a question (Lee 2000).
In addition to the 3 types, researchers have recognized a special question form, the A-not-A questions:
(5) Ni renshi ta bu renshi ta? (Huang et al. 2009: 236 (4))
You know him not know him
‘Do you know him?’
3 Q: Q-particle or question-particle.
4 Mahjong: a Chinese tile-based game, which is similar to the Domino in some Latin-American countries.
27 To form A-not-A questions, the general rule is to put the affirmative and the negative form of the sentence together (Li and Thompson 1981:535). Although this type of question is translated as yes-no question, most researchers classify it as a type of disjunctive question. (Huang et al.
2009, Li and Thompson 1981, Zeng 1996, among others). The choice presented to the respondent is the affirmative sentence and its negative counterpart. In example (4), the listener is supposed to make a selection between ‘knowing him’ and ‘not knowing him’.
The ‘A’ in the term A-not-A questions generally refers to the predicate of the utterance, usually the verb or the adjective, or part of it. Depending on what ‘A’ accounts for, there can be three sub-classes of the A-not-A question: VP-not-VP, V-not-VP, VP-not-V (Huang et. al 2009).
The example (5) is a VP-not-VP question, where both the structures before and after the negation are full VPs ‘renshi ta’ (know him). Examples of V-not-VP and VP-not-V questions are given in
(6) and (7) below.
(6) Ni renshi bu renshi ta? (Huang et al. 2009: 246 (46a))
You know not know him?
‘Do you know him?’
(7) Ni renshi ta bu renshi? (Huang et al. 2009: 246 (46b))
You know him not know?
‘Do you know him?’
In these sentences, the V is ‘renshi’ (to know), and the VP is ‘renshi ta” (to know him). In
(5), what comes before the negation is only the V, without the NP ta ‘him’, which is the complement of the verb, but what follows the negation is the full VP. In (6), on the other hand, there is a full VP before the negation, but just a V after the negation.
28 Although not widely recognized, there exists another type of yes-no question that is syntactically unmarked, also known as the intonation-only questions (Lee 2000, 2005). This type of question, as its name suggests, uses solely intonation to differentiate statement and yes-no questions, same as in Spanish and many other languages. See example in (8):
(8) Ni renshi ta?
You know him
‘Do you know him?’
Literature on this type of question is scarce. Based on my intuition, lacking the question marker ‘ma’, intonation-only questions are used more in non-information seeking contexts like surprise, confirmatory, or repetition, at least in the Beijing dialect.
In this thesis, three types og questions that can function as yes-no questions in Chinese, namely ma-question, A-not-A question, and intonation-only question, will be examined in terms of their pragmatic uses and intonation.
3.4 Pragmatic functions of the Chinese questions
Although both ma-questions and A-not-A questions can function as yes-no questions in
Chinese, there has been a discussion about their neutrality. Some researchers observed that the ma- particle questions tend to have negative presupposition and are not completely neutral. In many situations, the answer ‘no’ is expected from the addressee (Chao 1968, Lyu 1980, Liu et al. 1983,
Hu 1991, Fang 1992, Chu 1998). In this sense, the A-not-A questions are more neutral. However, other researchers point out that A-not-A questions actually have a narrow focus on the verb, since such questions cannot not be answered with ‘yes’ or ‘no’, but with the verb (Huang et al. 2009,
Lee 2000), as illustrated in (8):
29 (9) Ni xihuan bu xihuan chi pinguo?
You like not like eat apples
‘Do you like eating apples?
a. * Shi./Bu shi.
Yes/No.
b. Xihuan / bu xihhuan.
Like / not like
However, from my intuition as a native speaker of MC, both ma-particle questions and A- not-A questions can be used in both neutral information - seeking and biased contexts or as echo questions that express certain emotions. In (8), I provide a context from my experimental material that elicits an ‘out of blue’, neutral information-seeking question. I show that ma-particle question
(9a) and A-not-A question (9b) are actually exchangeable in this case, and can both be completely neutral:
(10) Context:
You are having some friends at home for a party. As you are preparing drinks for them,
you want to ask them if they want ice in their drinks. What do you ask to get this
information?
a. Nimende yinliao jia bing ma?
Your drinks add ice Q
‘You (want) ice in your drinks?’
b. Nimende yinliao jia bu jia bing?
Your drinks add not add ice
‘You (want) ice in your drinks?’
30 As for non-information-seeking contexts, Lee (2005) pointed out that ma-questions can be used both as information-seeking questions, and as echo-questions (surprise), same as intonation- only questions. However, the possibility of A-not-A questions being used as echo questions was not discussed. In example (10), I provide example to show that ma-questions can be used to express surprise (10a) while A-not-A question cannot (10b).
(11) Context: A friend tell you that she wants ice in her drink, but you know that she
normally does not like iced drinks, that takes you by surprise. What do you say to ask
her if she really wants ice in her drink?
a. Nide yinliao jia bing ma?
Your drink add ice Q
‘You (want) ice in your drinks?!’
b. *Nide yinliao jia bu jia bing?
Your drink add not add ice?
‘You (want) ice in your drinks?’
However, based on my observation, A-not-A question, with certain intonation and/or the use of adverbs ‘daodi’ or ‘hai’5, can be used to express impatience or anger (10a), while ma- questions are less acceptable in this context:
(12) Context: Your roommate promised you to go out with you for dinner. However, it
seems that she hasn’t started preparing to go out when it was already very late. What
do you ask her to see if she still wants to go or not?
5 ‘Daodi’ according to Huang et al. (2009) literally means ‘reach bottom’, but conveys urgent desire to know the answer, or even impatience. ‘Hai’ literally means still, but has the same pragmatic meaning as ‘daodi’ in this case.
31 a. Ni (daodi) / (hai) qu bu qu chifan (a)?
You Adv. go not go eat (Particle)
‘Do you go out to eat or not!’
b. ??Ni (daodi) / (hai) qu chifan ma?
You Adv. go eat Q
‘Do you go out to eat or not!’
Nonetheless, although ma-questions and intonation-only questions have been confirmed to be used in both neutral information seeking and echo-surprise contexts, to my knowledge, there has not been study that investigate if these question types, along with A-not-A questions, can be used in confirmatory and echo-repetition contexts. In fact, it is one of the goals of this thesis to get a more comprehensive picture of the pragmatic functions of these different types of yes-no questions in Chinese.
In the elicitation task used in my study, participants are not restricted in terms of which syntactic structure they should use. Therefore, I am expecting to see correspondence between the three sypes of questions under examination, and the four pragmatic contexts. The categories that will be considered in this thesis are summarized in Table 3.3.
Table 3.3 Question categories considered in this thesis (* indicate the expected categories)
Pragmatic Functions
Echo Info-seeking Surprise Confirmatory Repetition
ma-particle questions * * * * Syntactic A-not-A questions * Types Intonation questions * * *
32 3.5 Chinese interrogative intonation
Research on Chinese interrogative intonation agrees on that Chinese yes-no questions have rising intonation. However, there is a discrepancy on whether the rising is global or local. Many studies view a global intonation trend (e.g., raised F0 register) to be the most crucial F0 cue to distinguish questions from statements (Shen, X.N, 1989, 1990; Shi 1980). In this view, question intonation is manifested over either the entire utterance or the constituent that is larger than a word unit. One of the models proposed under this view is Gårding's grid model (Gårding 1984, 1987).
It treats the primary prosodic difference between statements and questions as distinct slopes of grid. The statements have level grids while questions have an upward slope.
Other researchers supported a localized rise at the terminal partition of the questions, which can be the last syllable (Chao 1968, Ho 1977, Wu 1982, etc.); the last stressed/full-toned syllable
(Shi 1980, 1981, Hu 1985, Shen, J. 1985, etc.); or the last word (Rumjancev 1972).
Lee (2000) studied intonation-only questions and ma-questions in neutral information- seeking and echo-surprise contexts in comparison with corresponding statements. She found both global rising trend and local rise at the last syllable (ma in the case of ma-questions). She also found that intonation-only questions have greater pitch range in than ma-questions, and echo- surprise questions have greater pitch range than neutral information seeking questions.
In terms of the intonation of A-not-A question, Lee (2000) reported information-seeking
A-not-A questions to have a final falling at the last syllable. The first verb of the A-not-A phrase is realized as a narrow focus with expanded pitch range, and the following part of the sentence has compressed pitch.
33 In this study, we aim to investigate the intonation of A-not-A questions in non-neutral contexts (if they will be used in those contexts). We also aim to verify the intonation pattern of ma-questions and intonation-only questions and describe them using the AM framework.
34 CHAPTER 4
METHODOLOGY
4.1 Goal
The objective of this thesis is to examine the intonational properties of the information- seeking and echo-yes-no questions in Puerto Rican Spanish and Beijing Mandarin Chinese. We use these two dialects as representatives of Spanish and Chinese to compare the use of phonological and syntactic cues in the two languages in the formation of questions and the expression of certain pragmatic meanings.
4.2 Research questions and hypotheses
The research questions concerning the PR Spanish are double folded:
Firstly, are there intonational differences in Yes-No questions of different pragmatic uses, namely neutral information-seeking, echo-surprise, confirmatory, and echo-repetition in Puerto
Rican Spanish? What are the intonational contours like in each of the pragmatic context?
Secondly, given that the Spanish participants recruited in this study include native and heritage speakers of PR Spanish, this study also aims to investigate the following question: Does participants’ level of bilingualism affect their Spanish yes-no question intonation?
It is hypothesized firstly, that a final falling intonation will be found in the information- seeking yes-No questions in Puerto Rican Spanish. PR Spanish, like other Caribbean dialects, is characterized by lacking a final rise and using a nuclear rise-fall to encode yes-no questions (Sosa
1999; Armstrong 2010, 2012). For the un-biased information-seeking yes-no questions, a ¡H* nuclear pitch accent and L% phrase boundary is expected.
35 On the other hand, echo questions are likely to have different nuclear pitch accents followed by a low boundary tone. Armstrong (2010) proposed a L+¡H*L% nuclear configuration for echo-surprise questions and a H+L* L% for confirmatory questions. It is likely that similar patterns will be found in this study. Alternatively, it is also likely that the results will be different since the two studies differ in materials and language status of the participants. As for the echo- repetition questions not explored by Armstrong (2010), it is hypothesized that similar contour would be used as in confirmatory questions, as the repetition questions also seek confirmation about what has not been heard well. However, it is expected that some strategy like the expansion of pitch range at the repeated item will be used to distinguish the two types of questions.
As for the effect of participants’ level of bilingualism, it is likely that early bilinguals and heritage speakers would apply different tone types than native speakers. Alvord (2010) found that the second generation of Cuban immigrants in Miami use a final rising tone in Yes-no questions, different from the first generation who apply a final falling tone, probably because of the influence of English intonation. Alternatively, since the Puerto Rican Spanish speakers generally return to their home country more frequently than the immigrants from Cuba and other Latin-American countries, it is also possible that the early bilingual and heritage speakers of PR Spanish are able to use native-like tone types, but are less capable of making the subtler distinctions in their intonation of Yes-no questions that express different emotions. It is also likely that differences will be found between the early bilingual and the heritage speaker as a result of their different age of initial contact with English, etc.
As for Chinese, we are mostly concerned with the following questions: Do similar intonational differences exist in Mandarin Chinese? Is there syntactic variation across these question types in this language?
36 Within the second research question, this study also intends to investigate what might be the factors that constraint the selection of the three possible syntactic structures in Mandarin
Chinese Yes-No questions, namely the 1). sentence-final particle ma (e.g. Ni chifan-le ma? You eat-aux-past. PRT ‘Have you eaten?’), 2). the A-not-A structure (e.g. Ni chi mei chi fan? You eat- not-eat food? ‘Have you eaten?’) and 3). the intonation only (e.g. Ni chifan-le? You eat-aux-past?
‘You have eaten?’). Is the selection determined by question type or by personal preferences? Do all participants prefer using one structure over the others in a certain question type, or each participant use the same structure for all different question types?
Based on previous literature, it is very likely that the syntactic structures used would vary across the four questions types, namely information-seeking, echo-surprise, confirmatory, and echo-repetition. The two syntactic variants that are believed to be strong indicator of interrogative meaning will likely to be observed in the information-seeking questions: 1) sentence-final particle ma (e.g. Ni chifan-le ma? You eat-aux-past. PRT ‘Have you eaten?’) and 2) A-not-A Structure
(e.g. Ni chi mei chi fan? You eat-not-eat food? ‘Have you eaten?’) and its variant A-not (e.g. Ni chi fan mei? You eat-not? ‘Have you eaten’), with the sentence-final particle ma being the preferred form. (Huang et al. 2009). On the other hand, echo questions, which do not emphasize inquiry, are hypothesized to be produced without such question particle, and be realized with changes of intonation only. Alternatively, based on my own intuitive as a native speaker, it is also possible that personal preference might be a stronger predictor of the selection of the syntactic structure in the forming of all types of questions. Participants would keep using the same type of structure across questions of different pragmatic uses.
In terms of intonation, a rising tone at the sentence final question particle ma is expected in the information-seeking questions, since studies showed that a rising of the base line on the last
37 stressed syllable in ma-particle questions expresses a speaker's expectation of hearing a response from the addressee (e.g., emphasizing a speaker's inquiry) (Hu 1985; Shen, J. 1985; Gao 2001).
As for the intonation of the echo Yes-No questions, previous researches revealed that both syntactic (with or without the presence of question particle ma) and pragmatic factors may affect the boundary tone of Chinese Yes-no questions. (Lee 2000, 2005; Qiu and Song, 1994). However, little is known about the specific intonation of echo-questions in Chinese, which is left to be explored in this thesis.
4.3 Materials
Stimuli for this study include a Chinese section and a Spanish section. Each section is composed of four blocks representing four types of yes-no questions of different pragmatic purposes: information-seeking, echo-surprise, confirmatory, and echo-repetition. We followed
Armstrong (2010) in the classification of neutral information-seeking questions and echo-surprise question. However, out of the two types of echo-questions discussed in Armstrong (2010): contra- expectation and incredulity, the echo-surprise questions in this study are closer to the first category.
The confirmatory questions are different from the confirmation questions in Armstrong (2010) in that what is being confirmed is something being observed instead of told. This question type is proposed based on observation that both Spanish speaking and Chinese people sometimes start a conversation with a question that confirms the situation they are observing. The inspiration for the echo-repetition question came from study on the echo-repetition wh-questions in Gonzalez and
Reglero (2016). It is observed that yes-no questions can also have such function.
The exact definition of the contexts is listed as below:
(I). Neutral information seeking: un-biased questions that require a neutral answer.
(II). Echo-surprise: echo-questions that express surprise and do not expect an answer.
38 (III). Confirmatory: biased questions with positive presupposition to confirm a hypothesis
that the speaker formed based on what is seen.
(IV). Echo-repetition: questions that repeat part of what the speaker heard to confirm some
information.
Each of the blocks contains five contexts that aim to create real-life situations that trigger the production of the corresponding type of yes-no question. In each block, there are also 5 distractors (statements or wh-questions), each inserted between every two target contexts.
The Spanish materials were created with words that are commonly used in the Puerto Rican
Spanish, including the names of some Puerto Rican cities like Guayama and Aguadilla, and names of typical food such as Mondongo6. The inspiration of some of the target items and distractors came from González & Reglero (2016) and the Encuesta de Español Puertorriqueño (Armstrong,
M.E. (2010)). The study was piloted with some native Puerto Rican speakers to test the materials.
Contexts were carefully designed so that the expected utterances mostly have two prenuclear accents and a nuclear accent, and so that the last content word of each sentence contains only voiced segments in order to prevent any voiceless segment from breaking the pitch contour.
Number of syllables in the expected questions range from 9 to 13.
For the reason of cultural differences, the Chinese and Spanish parts are not translation equivalents. The Chinese materials were created by the investigator, who is a native of the language. Contexts that are common in the daily life in Beijing, China were used, and verified by other native speakers of Chinese. Apart from choosing words with voiced segments as the final
6 Mondongo, or Sopa de mondongo, is a soup made from diced tripe (the stomach of a cow or pig) slow-cooked with vegetables such as bell peppers, onions, carrots, cabbage, celery, tomatoes, cilantro (coriander), garlic or root vegetables. The dish is generally prepared in former Spanish colonies in Latin America and the Caribbean, and in the Philippines.
39 words, effort was made to assure that the lexical tone of the final syllable of each sentence is the high-level Tone 1 (Chao 1930, 1933, 1968), to control the potential interference of lexical tones with the sentential intonational pitch contour (Chao 1933).
All target contexts and distractors were presented through PowerPoint slides in Spanish and Chinese to the participants, each accompanied by one or two pictures that describe the situation and an audio of a native speaker reading the context aloud.
An example of the material for each section and block is provided below. For reasons of space, the Chinese scripture is not included in the examples. Complete examples of the materials are available in Appendix B.
Examples:
Section I Spanish:
A: Neutral Information-seeking
Context: En el restaurante, quieres saber si tienen bizcocho de vainilla. ¿Qué le preguntas al camarero?(In the restaurant, you want to know if they have vanilla cake. What do you ask the waiter?)
Expected answer: ¿Tienen bizcocho de vainilla? (Do you have vanilla cake?)
B: Echo-Surprise
Context: Tu amiga Andrea te ha regalado dos entradas para el concierto de Lady Gaga. Tu hermana la odia pero dice que quiere ir contigo al concierto. Te sorprende mucho. Pregúntale si de verdad a ella le gusta Lady Gaga. (Your friend Andrea gave you two tickets to the Lady Gaga concert.
Your sister wants to go with you but you know she dislikes the singer. This takes you by surprise.
Ask if she really likes Lady Gaga.)
40 Expected answer: ¿De verdad te gusta Lady Gaga? (Do you really like Lady Gaga?)
C: Confirmatory
Context: Ves a tu amigo mirando fotos de la Muralla Romana. Sabes que acaba de regresar de un viaje a Europa, y parece que visitó la Muralla Romana. ¿Qué le preguntas para confirmarlo?
(You see a friend looking at his pictures of the Roman Walls. You know that he has just got back from a trip to Europe. What do you ask to confirm it?)
Expected answer: ¿Fuiste a la Muralla Romana? (You went to the Roman Walls?)
D: Echo-Repetition
Context: Tu amigo te dice que va a viajar a Aguadilla. No sabes si escuchaste bien porque en ese momento tu perro comenzó a ladrar. Pregúntale si va a viajar a Aguadilla. (Your friend told you that he is going to travel to Aguadilla. You don’t know if you heard it clearly because at that moment your dog began to bark. Ask him if he is traveling to Aguadilla).
Expected answer: ¿Vas a viajar a Aguadilla? (You are traveling to Aguadilla?)
Section II Chinese
A: Information-seeking
Context: At noon, you met Xiao Liu in the hallway; you want to know if she has eaten, you ask ______?) Expected answer: a) Sentence-final particle: Ni chifan-le ma?
You eat-aux-past. PRT
‘Have you eaten?’
b) A-neg.-A Structure: Ni chi mei chi fan?
You eat-not-eat food?
‘Have you eaten?’
41 c) Intonation-only: Ni chifan-le?
You eat-aux-past;
‘You have eaten?’
B: Echo-Surprise Context: Saturday evening you call Xiao Wang and he says he’s working in the office, you are surprised that he had to work on Saturday, you say _____?
Expected answer: Ni zhouliu hai jiaban?
You Saturday still work？
‘You still work on Saturday?’
C: Confirmatory
Context: You see Xiao Wang sneezing; you want to ask if she´s caught a cold, you say
______?)
Expected answer: Ni ganmao-le (a)?
You catch a cold aux-past (PRT)?
‘You have caught a cold? ’
D: Echo-Repetition
Context: Your friend tells you that Xiao Li is going to study in Singapore, you did not hear her well because of the noise in the cafe, so you repeat it to confirm, you say _____?)
Expected answer: XiaoLi yao qu Xinjiapo?
XiaoLi will go Singapore?
‘XiaoLi will go to Singapore’
4.4 Task
The Spanish elicitation task consisted of 47 PowerPoint slides with instructions that guided the participant in how to respond to the audio input and the visual aid provided with the picture. 42 The voice that the participants heard on the PowerPoint came from a native speaker of Puerto
Rican Spanish, whose voice was recorded in a quiet office room in the Department of Modern
Languages and Linguistics of the Florida State University prior to the study.
Depending on the prompt read aloud by the voice, participants were asked to respond with a Yes-No question (target item), a wh-question, or a declarative sentence (distractors). There was a practice slide for each of the sentence types.
In one of the slides (information-seeking no.1), the participants were shown the picture of a waiter in a restaurant and given the following context: “En el restaurante, quieres saber si tienen bizcocho de vainilla. ¿Qué le preguntas al camarero?” (In the restaurant, you want to know if they have vanilla cake, what do you ask the waiter?) The majority of the participants responded with the expected Yes-No question “¿Tienen bizcocho de vainilla?” (Do you have vanilla cake?)
The Chinese elicitation task had generally the same design. The voice in the PowerPoint came from the researcher who is a native speaker of Beijing Mandarin.
4.5 Participants
A total of 8 participants (4 male, 4 female) were recruited and recorded in Florida State
University, Tallahassee, Florida. Among them, 4 are native speakers of Mandarin Chinese from the region of Beijing, China, including 1 male and 3 females. They are all graduate students at
Florida State University. The other 4 (3 males, 1 female) are speakers of Puerto Rican Spanish who are studying or working in Tallahassee: two of them (P2 and P3) can be categorized as first generation immigrants and native speakers of PR Spanish, according to the sociolinguistic criteria for generation division in Silva-Colvarán (1996), as they moved to the U.S at age 10 and 23 respectively. The other two (P1 and P4) were also born in Puerto Rico but lived in the U.S since they were 3 years old and younger. Based on information in their language questionnaire, these
43 two participants are both more proficient in English and learned Spanish at home as a ‘heritage language’ (minority language learned at home, of which the acquisition is often interrupted or incomplete) (Montrul 2008; Polinsky & Kagan 2007; Valdés, G. 2000, 2005; among others).
Information about the participants is summarized in Table 4.1.
Table 4.1 Participants information: Spanish
Participant Sex Age Age when moved to the U.S
Number (year)
P1 male 24 1
P2 female 27 23
P3 male 28 10
P4 male 19 3
P1 is a heritage speaker of PR Spanish. He was born in Puerto Rico but moved to Florida,
U.S.A. with his family when he was 14 months old, and lived in the states since then. However, as he reported, his parents spoked to him in Spanish before he started school at age 4. He reported visiting PR for a few weeks every few years, and he received his BA degree in Spanish 2 years before the study. He self-rated ‘very good’ (6 out of 7) for listening, speaking, reading and writing in Spanish.
P4 is a 19-year-old heritage speaker of Spanish born in Puerto Rico and acquired Spanish as his first language until having contact with English when moving to the U.S at age 3. He reported
44 having used Spanish as language of instruction in Elementary school, Middle school and High
School. He self-rated himself native-like in listening and speaking in Spanish, and ‘very good’ (6 out of 7) in reading and writing, while native-like in all aspects in English.
P2 is 27-year-old native speaker of PRS and lived in the island until she was 24 years old when she moved to Florida. She reported that she started learning to speak English as a foreign language at age 15 and to write at age 18. She attended college but did not finish her BA degree.
Table 4.2: Participants information: Chinese
Participant Number Sex Age Age of moving to the U. S
(yaer)
C1 female 26 24
C2 female 24 23
C3 female 22 21
C4 male 27 21
The participants either responded to the advertisement posted on social network, or were informed in a “friend of a friend” manner. Before the study, participants read and signed a consent form (Appendix D). They were told that they would be recorded for a study on some speech habits of native Chinese/Spanish speakers. After the recording, they were asked to complete the
Language History Questionnaire (LHQ 2.0) (Zhang et al., 2013), accessible online at http://blclab.org/language-history-questionnaire/. The questionnaire was available in English,
45 Spanish, and Chinese. Participants could choose the language by their preference. Participants were given a small gift at the end of the study.
4.6 Data collection and analysis
Participants were recorded one at a time at the FSU Phonetics Lab. A TASCAM DR-05 digital voice recorder was used and the participants´ utterances were recorded in WAV format, on the mono setting. The sample rate was set for 44,100 Hz. The PowerPoints were displayed on the personal laptop of the investigator. A low cut of 40 Hz was set to minimize the effect of noise generated by the laptop ventilator and other factors.
Participants were told to read and sign the consent form. Then they were told to put on headphones and follow the instruction in the audio prompt. Before starting with the target items, they were given three sentences for practice. During the recording, feedback was given to the participants if they misinterpreted the context, or if the volume of their voce was too low to obtain a good recording. Participants were encouraged to speak as naturally as possible and ask any question they had about the task. After participants finished all the utterances, they were told to repeat the process all over again. Due to time constraints, only the utterances in their second repetition were analyzed and included in this study.
The whole process last around 35 minutes.
Recordings were transcript and annotated using the phonetic software Praat where analysis of tones are done based on the AM framework and SP_ToBI system (see Chapter 1 Section 1.3.1,
1.3.2). A total number of 120 tokens were analyzed (5 items x 4 contexts x 3 participants x 2 languages). No token was excluded in the study. In one context, P2 used a disjunctive question that contains two yes-no questions separated by a pause: “Juan, ¿en serio tienes trabajo, o es que
46 no quieres pasar tiempo conmigo?” (Juan, you really have work or you don´t want to spend time with me?). In this case, both questions were analyzed.
47 CHAPTER 5
RESULTS-SPANISH
This chapter presents the findings from the Spanish data. Section 5.1 presents the final contour of the utterances in the 4 pragmatic contexts: neutral information-seeking, confirmatory, echo-surprise, and echo-repetition. Contour patterns are described in terms of boundary tones, nuclear accents, and nuclear configurations. Section 5.2 presents the individual differences among the 3 participants: a native speaker and two heritage speakers, and discusses the potential effect of level of bilingualism on intonation.
5.1 Final contour
5.1.1 Boundary tones. Out of the total 60 tokens, 53 (88%) were realized with a low boundary tone (L%). The pitch falls at the end of the utterances to a low target. This pattern was most consistent in the echo-repetition questions. The prevalence of L% is expected since it is reported to be the most frequent in yes-no questions of various pragmatic uses in PRS (Sosa 1999,
Armstrong 2010).
High boundary tone (H%) was also found in 5 utterances in which the pitch rises at the end of the utterances. P1 produced 2 high boundary tones in information-seeking questions and P2 produced 3 confirmatory questions with final rising intonation. It is worth mentioning that one of these confirmatory questions was realized as a tag question, ending with the tag ‘¿de verdad?
(right?). The tag follows another yes-no question that ends with a low boundary tone.
In addition, 2 cases of middle-boundary tones (M%) were also attested, both in the echo- surprise questions produced by P2. In one of them the M% appeared between a pause in a disjunctive question because of suspension.
48 5.1.2 Nuclear accent. Four types of nuclear accent were attested:
1). L+H*: pitch rising through the syllable nucleus.
2). H*L/HL*7: pitch falling though the syllable nucleus.
3). H*: high plateau at the syllable nucleus.
4). L*: flat low tone at the syllable nucleus.
The best attested nuclear accents are L+H* and H*+L/H+L*, with the L+H* nuclear accent being the most frequent pattern overall. P1 consistently produced it in all 4 types of questions. P2 mostly used this tone type in all contexts but the echo-surprise questions. In a few utterances, the pitch rises to an extra high peak, marked with up-stepped ¡H*. On the other hand, the H*L/H+L* tone is the dominant tone type in the utterances produced by P4 in all 4 contexts. It was also applied by P2 in 4 out of her 5 echo-surprise questions.
5.1.3 Nuclear configuration. Combining nuclear accents with boundary tones, we see 5 types of nuclear configuration: L+H*L%, H*LL%, HL*L%, H*L%, L+H*H%. The distribution of nuclear configuration in each context is summarized in the following sections.
5.1.3.1 Information-seeking questions. Table 5.1 illustrates the distribution of the 5 types of nuclear configurations in neutral information-seeking questions.
Table 5.1 Distribution of final contours for information-seeking questions
L+H*L% H*+LL% H+L*L% H*L% L+H*H%
P1 3/5 2/5
P2 3/5 1/5 1/5
7 The * marks whether the nuclear accent is more related to low/high.
49 Table 5.1 continued
L+H*L% H*+LL% H+L*L% H*L% L+H*H%
P4 2/5 3/5
Total 6/15 2/15 4/15 1/15 2/15
(40%) (10%) (27%) (7%) (13%)
As we can see in Table 5.1, the most common tone type for information-seeking questions is a rising nuclear pitch accent followed by a low boundary tone (L+H*L%). The pitch contour is generally characterized by a rise-fall ‘circumflex’ pattern at the last accented syllable.
The prevalence of the L+H* L% final contour deviates from the findings of Armstrong
(2010:167). She reported a ¡H*L% configuration with a fall from an up-stepped peak occurring very early in the syllable nucleus or around its midpoint. In this study, however, we saw that P1 and P2 mostly produced a rise-fall ‘circumflex’ pattern within the last accented syllable. See Figure
5.1 for an example.
However, although an overall of 40% of the information-seeking questions show the L+H*
L% final contour, there is a rather great degree of variability in their tonal preference. For instance,
P1 (the heritage speaker) applied final rising tone in 2 tokens, which can be best transcript as
L+H*H%. This contradicts the general final-falling pattern in information seeking yes-no questions in PRS. See Figure 5.2 as an example of the final-rising tone.
P4 showed a distinct pattern. In all the cases, he applied a H*+LL%/H+L*L% final contour, with a falling throughout the syllable nucleus from the highest peak. The final postotic syllable maintains at a flat low level instead of reaching to a very low point. It is worth noticing that this
50 pattern prevails in his utterances across all types of questions. Whether this is a characteristic of personal speech will be discussed with more detail in the following section. See Figure 5.3 for an example of the ¡H*LL% tone.
Figure 5.1 ¿Tus padres viven en Guayama? “Your parents live in Guayama?” Information- seeking question with L+H*L%
Figure 5.2 ¿Hay bizcocho de vainilla? “Is there vanilla cake?” Information-seeking question with L+H*H%
51
Figure 5.3 ¿Tu apartamento tiene lavadora? “Does your apartment have washer?” Information- seeking question with H+L*L%
5.1.3.2 Echo-surprise questions. As shown in Table 5.2, most of the echo-surprise questions are produced with a final falling tone, except for one token by P2 which ended with a mid-tone8. Two types of final falling contour H+L*L% and L+H*L% are the most frequent.
Table 5.2 Distribution of final contours for echo-surprise questions
L+H*L% H*+LL% H+L*L% L*L% H+L*M%
P1 5/5
P2 3/5 1/5 1/5
P4 1/5 4/5
Total 5/15 1/15 7/15
(33%) (7%) (47%)
8 There is another question with L+H*M% nuclear configuration that is the first half of a disjunctive question, and is excluded in this table.
52 Looking at each participant separately, P1 mostly produced the ‘circumflex’ structure best transcribed as L+H*L%, and in two of the cases reached to an up-stepped peak (L+¡H*L%). An example from P1 is provided as in Figure 5.4.
P2 applied various tone types across utterances, but mainly a HL*L% contour falling through the syllable nucleus. What draws special attention is that in one utterance, she uses a L*L% final contour, characterized by a flat low final tone. This is not found in the utterances of the other participants, nor is it reported in previous literatures. The use of flat low tone is also against our hypothesis, since there has been evidence that echo-surprise wh-questions in northern peninsular
Spanish showed expanded pitch range to express the feeling of surprise. (González & Reglero
2016). One possible account for this abnormality is the intervention form the tag phrase ‘de verdad’ inserted at the beginning of the sentence. Contour of this utterance is provided as in Figure 5.5.
P4 produced mostly the HL*L% final contour, which as mentioned, is found to be his preferred tone type in all question types. The pitch contours are characterized by a gradual rising from the beginning until the falling starts around the mid-point the syllable nucleus. An example is provided as below in Figure 5.6.
Figure 5.4 ¿De verdad no tienen Mondongo? “You really don’t have Mondongo?”Echo-surprise question with L+¡H*L%
53
Figure 5.5 ¿Guillermo de verdad se ganó la lotería? “Did Guillermo really win the lottery?” Echo-surprise question with L*L%
Figure 5.6 ¿En verdad tienes que trabajar el domingo? “Do you really have to work on Sunday?” Echo-surprise question with H+L*L%
Armstrong (2010) discussed the intonation of two types of echo-surprise questions distinguished by nuances in their pragmatic meanings. One type of questions express that what the speaker heard goes against his/her expectations. The tone produced for this type of question of
‘surprise’ or ‘disbelief’ is transcribed as L+¡H*L%, consistent with the findings about this type of
54 question in Sosa (1999). Another type is the ‘questions of incredulity’, which implicates that the speaker is unwilling to admit or accept the propositional content as true. They are different from the first types of questions in that they also implicate ‘doubts’. Such questions are reported to have a L* nuclear accent, and a HL% bitonal boundary tone (L*HL%).
In the results of this study, only P1 showed similar pattern with the finding of Armstrong and Sosa (Armstrong 2010, Sosa 1999) for the echo-surprise questions that express contra- expectation, but without the sense of incredulity.
5.1.3.3 Confirmatory questions. For confirmatory questions, all utterances are produced with final-falling tones, except for 3 final-rising tones from P2. However, as shown in Table 5.3, there is much variation in the nuclear configurations.
Table 5.3 Distribution of final contours for confirmatory questions
L+H*L% H*+LL% H+L*L% H*L% L+H*H%
P1 5/5
P2 1/5 1/5 3/5
P4 4/5 1/5
Total 6/15 4/15 1/15 1/15 3/15
(40%) (27%) (7%) (7%) (20%)
55 P1 mostly produced the ‘circumflex’ pattern, best transcribed as L+H* L%. The pitch contours rise at the beginning then remains flat until rise around the mid-point of the syllable nucleus and falls. See Figure 5.7 for example.
Somehow surprisingly, P2 produced final-rising tone in 3 out of the 5 tokens. The pitch contours showed a gradual rising from the very beginning all the way to the end. (See Figure 5.8 for example). It is necessary to mention, however, that one of them was produced as a tag question with ‘¿de verdad? (right?)’ attached to the end of the question. In the other 2 utterances realized with final-falling, she produced (L)H*L% final tone with observingly reduced pitch range.
Very similar to his behavior in other types of questions, P4 consistently produced the
H*LL% final tone. The pitch contour rise at the beginning, then remains flat until falling starts at the end of the nuclear accent. The pitch stays flat at the last syllable with a slight rising at the end.
See Figure 5.9 for example.
Figure 5.7 ¿Fuiste a la bodega? “You went to the wine shop?” Confirmatory question with L+H*L%
56
Figure 5.8 ¿Fuiste a la bodega hoy? “You went to the wine shop today?” Confirmatory question with L+H*H%
Figure 5.9 ¿Fuiste a la Muralla Romana? “You went to the Roman Wall?” Confirmatory question with H*+L L%
For Confirmation yes/no questions, Armstrong (2010: 173) reported a fall through the nuclear accented syllable to a low target, labelled H+L*L%. The fall is reported to begin in the pretonic syllable of the utterance-final prosodic word, and the low target is reached in the tonic
57 syllable. She noted that this contour was not found when the speaker did not have any specific belief about the propositional content.
As can be seen from Table 5.3, both P1 and P2 deviate from this pattern. This is not surprising since the contexts designed for ‘Confirmation/Confirmatory yes-no questions’ are not the same in nature. In Armstrong (2010:173), a Confirmation yes-no question refers to a question that confirms a proposition the speaker believes is true based on some information he/she has been given. In the example provided by the author, the participants were told in the preceding contexts that they believed that there was indeed a place nearby that sold piononos9. The target utterance produced by participants was ¿No hay por aquí un lugar que vende piononos? (Is´t there a piononos place around here?). The utterance was produced with an abrupt fall through the syllable neucleus and reached a low target (see Armstrong (2010: 175, Figure 13). The negation used in the utterance suggest that the belief that there is indeed a pionono place is very strong, with probably more than
90% of possibility. The speaker might not be even expecting an answer.
However, in this study, the ´Confirmatory yes-no questions´ is defined as ´a question that is used to confirm a hypothesis that the speaker generates based on what he-she has seen and-or the situation encountered´ (see Chapter 4). For instance, in one of the context, the participants were given a context in which he/she she knew that a friend was about to get married and finds that person looking through an engagement ring catalog and it seems like he´s selecting a ring for his girlfriend. The participants were then asked what he/she would say to confirm this. The target utterance is ´ ¿Vas a comprar un anillo para tu novia?´ (You are buying a ring for your girlfriend?).
In situations like this, the question is also strongly biased, with expectation of a positive answer.
However, the belief is not as strong as in the context of Armstrong (2010). Therefore, participants
9 Pionono: A fried food common in Puerto Rico.
58 seem to have used similar final contours as they did for information-seeking questions, with
L*H*L% as a dominant tone type. However, whether the Confirmatory questions in this study are produced differently from neutral Information-seeking questions in terms of pitch range is left to be examined in future analysis.
Interestingly, the final contour used by P4 are very similar to the Confirmation yes-no questions described in Armstrong (2010:173). Nonetheless, since the same patterns repeats in his utterances for all types of questions discussed, this similarity is more probably a coincidence.
5.1.3.4 Echo-repetition questions. For echo-repetition questions, there is a rather high consistency of trend. All the utterances were realized with final-falling tones. At the same time, the pitch contours appear to be rather flat, with the final falling being slight and gradual. The dominant tone is again L+ H*L% (53 %), followed by H+L*L% (40%).
Table 5.4 Distribution of final contours for echo-repetition questions
L+H*L% H+L*L% H*L%
P1 5/5
P2 3/5 1/5 1/5
P4 5/5
Total 8/15 6/15 1/15
53% 40% 7%
P1 and P2 performed similarly as they both consistently produced the (L) H* L% nuclear configuration. The pitch contours often have an elevated beginning, and rise slightly and gradually
59 until around the mid-point of the nuclear accent, then fall slightly. See Figure 5.10 examples. P4 preferred H+L*L% tone, same as in the other three question types. See Figure
Figure 5.10 ¿Tú vas para Aguadilla? “(Did) you say you go to Aguadilla?” Echo-repetition question with L+H*L%
5.2 Effect of bilingualism
5.2.1 Degree of variation. Comparing the results from the three participants, we found that the Spanish dominant speaker P2 showed greater intonational variation across the different contexts, while the two English dominant speakers P1 and P4 kept using the same or mostly the same final contour in all utterances. This could possibly be a result of differences in proficiency.
Both P1 and P4 speak Spanish as a heritage language only with family. It is probable that although they produce the target final falling contour for most of the utterances, they are not capable of distinguishing the subtler pragmatic differences among questions through intonation. Nonetheless, they might be using other strategies to make those distinctions, like tonal range, but further
60 measurements should be done to verify it. Figure 5.11-5.13 illustrate the degree of intonational variation of each speaker across contexts.
P2
100% 80% 60% 40% 20% 0% Info-seeking Echo-surprise Confirmation Echo-repetition L+H*L% H+L*L%/H*+LL% L+H*H% H*L% L*L%
Figure 5.11 Intonational variation across contexts P2
P1
100%
50%
0% Info-seeking Echo-surprise Confirmatory Echo-repetition
L+H*L% L+H*H%
Figure 5.12 Intonational variation across contexts P1
61 P4
100%
80%
60%
40%
20%
0% Info-seeking Echo-surprise Confirmation Echo-repetition
H+L*L%/H*+LL%
Figure 5.13 Intonational variation across contexts P4
5.2.2 Final rising contour. Although 88% of the utterances are realized with final falling intonation, instances of final rising contour were attested both form heritage speaker P1 (in information-seeking questions, see Figure 5.2) and native speaker P2 (in confirmatory questions, see Figure 5.8). Final rising is rare in Puerto Rican Spanish yes-no questions; however, it is not surprising in this study since we hypothesized the effect of contact with American English. As we know, American English have a final rising intonation in yes-no questions (Pierrehumbert 1980).
Based on previous findings of intonational transfer (Alvord 2010; Robles Puente 2012, among others), it is possible that participants are transferring the rising intonation in English to their
Spanish. The usage of rising intonation in yes-no questions was also found among the second- generation Cuban immigrants in Miami in Alvord (2010). He proposed the influence of English as
62 a possible cause to it, as well as potential influence of the contact with non-Caribbean varieties of
Spanish which have final rising intonation. This alternative explanation is also likely to be applicable in this study, since contact with other varieties such as Colombian Spanish and Mexican
Spanish is very likely in Florida. This hypothesis is especially plausible with P1 since he worked for 9 months in Chile very shortly before the study took place, and he also received formal instruction on Spanish in university as he pursued a BA degree in Spanish, when he took class with teachers of various dialects of Spanish. However, to verify the impact of English and/or other varieties of Spanish, it is necessary to study their intonation when speaking English, and more details about their social network.
63 CHAPTER 6
RESULTS: CHINESE YES-NO QUESTIONS
This chapter presents the results found for the Chinese yes-no questions of different pragmatic uses (information-seeking, echo-surprise, confirmatory, and echo-repetition). Out of the three predicted syntactic structures that could occur in these pragmatic contexts (ma-structures; A- not-A questions, and intonation-only questions), only the first two were found in the results.
Exclamations and wh-questions were found in the echo-surprise questions, but their intonation will not be discussed here. Section 6.1 presents the distribution of syntactic structures used for each function. Section 6.2 discusses the pitch contour of ma-questions and A-not-A questions in each pragmatic context, with a focus on the boundary tone.
6.1 Syntactic structure As hypothesized in chapter 1, results show syntactic variation across the pragmatic contexts. Besides, there appears to be an individual preference in the selection of syntactic structure.
Table 6.1 shows the distribution of ma-question and A-not-A question. Ma-structure are favored in information-seeking and echo-repetition contexts, while A-not-A questions are preferred in confirmatory context. On the other hand, in surprise contexts, participants mostly uttered exclamations rather than yes-no questions.
Table 6.1 Distribution of different syntactic structures in each pragmatic context Info-seeking Surprise Confirmatory Repetition Ma A-not- Excl Wh Ma Ma A-not- Ma A-not- Tag- A A (A) Q C2 4/5 1/5 4/5 1/5 5/5 4/5 1/5
64 Table 6.1 continued Info-sseking Surprise Confirmatory Repetition Ma A-not- Excl Wh Ma Ma A-not- Ma A-not- Tag- A A (A) Q C3 5/5 3/5 2/5 5/5 5/5 C4 1/5 4/5 1/5 4/5 5/5 5/5 Tot. 10/15 5/15 8/15 6/15 1/15 5/15 10/15 9/15 5/15 1/15 (67%) (33%) (53%) (40%) (7%) (33%) (67%) (60%) (33%) (7%)
*Excl: exclamation; Tag-Q: tag question
6.1.1. Info-seeking. For neutral Information-seeking questions, C2 and C3 use the same ma-structure. C4, on the other hand, used A-not-A in 4 out of the 5 utterances, and ma-question in only one of the utterances.
This shows that both ma-questions and A-not-A questions can be used as neutral information-seeking yes-no questions in Mandarin Chinese. The same question can be realized with ma particle (as shown in (1)) or with A-not-A questions, as shown in (2).
(1) Ni xia wu you ke ma? (From C2)
You afternoon have class Q
‘Do you have class in the afternoon?’
(2) Ni xia wu you – mei - you ke? (From C4)
You afternoon hav- not - have class?
‘Do you have class in the afternoon?’
From the results of this study, it seems that both C2 and C3 prefer ma-questions to A-not-
A questions, while C4 prefers A-not-A questions. More data from more speakers are necessary to see if there is a general preference for ma-questions in Chinese in this pragmatic context, and 65 whether the preference for A-not-A questions shown by C4 is idiosyncratic, or a result of sociolinguistic factors such as gender (C4 is the only male participant in the study).
6.1.2 Echo-surprise. In the Surprise contexts, the majority of the utterances were produced as exclamations (as in (3)), or wh-questions (shown in (4)). Ma-questions were scarce, with only one token out of a total of 15, and the A-not-A structure was not attested in this context. This suggests that in MC, at least in the Beijing dialect, Echo yes-no question might not be a very common strategy of expressing surprise. Indeed, exclamation and wh-questions with ‘zenme’ (how come) are more commonly used for such pragmatic purpose.
(3) zheme zao-jiu sheng-le a! (From C2)
This early-CL give birth-CL Q
‘You gave birth this early?!’
(4) zenme zheme zao-jiu sheng-le (From C4)
How come this early-CL give birth-CL
‘How come you gave birth so early!’
6.1.3 Confirmatory questions. For Confirmatory questions, there seems to be an individual preference in the choice of syntactic structure. C2 used ma-question in all cases, while both C3 and C4 used A-not-A structure exclusively. Interestingly, C2 and C4 showed the same preference in the Information-seeking and the Confirmatory questions, however, C3 preferred ma-questions in the neutral Information-seeking block and A-not-A questions in the Confirmatory block.
It is worth mentioning that for both C3 and C4, the most frequent A-not-A phrase they used was with the verb ‘shi’ (to be), which also means ‘yes’ in Chinese. The phrase ‘shi-bu-shi’ (be- not-be, or yes-not-yes) followed by a subordinate clause or a VP is a common way of expressing
66 confirmation in MC. Some sentence-final particles like ‘a’, and ‘ya’ ([ia])10 are often added to the end of the A-not-A questions. (See (5) for example).
(5) XiaoWang ni shi-bu-shi ganmao-le ya? (From C3)
XiaoWang you be-not-be catch a cold-CL Q?
‘XiaoWang you caught a cold?’
6.1.4 Repetition questions. For Echo-Repetition questions, C2 used ma-questions in 4 out of the 5 tokens, the remaining target sentence including a tag-question with tag ‘dui ba?’ (right?) as in (8).
C3 exclusively used ma-questions, and consistently with the verb ‘shi’ (to be/yes). In fact, the structure of ‘shi…ma?’ is commonly used to confirm something the speaker is not sure of or did not hear clearly. See (9).
C4, on the contrary, exclusively used A-not-A structures in this context, also with the ‘shi- bu-shi’ structure. See (10).
(8) Ni shi-yao yi juan shuangmianjiao, dui ba? (From C2)
You be-want one CL tape, right?
‘What you want is one tape, right?’
(9) Ni shi-yao ba-yue hui lao jia ma? (From C3)
You will eight-month returen hometown Q?
‘Is it in August you are returning your hometown?’
(10) Ni gang cai shi-bu-shi shuo XiaoLi yao-qu Xinjiapo? (From C4)
You just now be-not-be say XiaoLi will-go Singapore?
10 Sentence-final particles a/ya were presented by Chao (1968) to function as marker of Confirmation, Commands, Vovative, and Warning. However, Li & Thompson (1981) claimed that the particles themselves do not have those functions, but are used to reduce the forcefulness in those contexts, glossed as RF (Reduced Forcefulness).
67 ‘Did you just said that XiaoLi would go to Singapore?’
6.2 Intonation contour
This section describes the pitch contour of ma-questions and A-not-A questions in each context, focusing on the boundary tones and intervention of syntactic elements like sentence-final particle and narrow focus.
6.2.1 ma-questions. As discussed in the previous section, questions with the final particle ma ‘Q’ are found in neutral information-seeking, echo-surprise, confirmatory and echo-repetition contexts.
In general, for neutral Information-seeking questions, the pitch maintains high at the final particle ma regardless of the rising or falling lexical tone of the preceeding segment. The contour can be best transcribed with a high boundary tone (H%), as is shown in Figure 6.1. This boundary tone is consistent with what was reported in Lee (2000).
As for the nuclear accent, the pitch is observed to rise to a down-stepped high peak at the syllable nucleus, best labeled as (!H*). The most frequent nuclear configuration for neutral ma- questions is !H*H%.
For the Confirmatory questions, only C2 produced 5 utterances of ma-questions. In all of these utterances, the final particle ma maintains a fairly high pitch, and a final trend of rising is often noticed. See Figure 6.2 for example
For the ‘Surprise’ contexts, only one exemplar of Echo-surprise ma-question was produced by C2. Again, the pitch maintains high at the final particle ma. See Figure 6.3.
68
Figure 6.1: Ni de yin liao xu yao jia bing ma? “Do you want ice in your drink?” Information ma-question with H*H%
Figure 6.2: Ni shi qu de Huangshan ma? “Did you go to Huangshan?” Confirmatory ma- question with !H*H%
For the Echo-repetition questions, only C2 produced four yes-no questions. In all them, a final rising boundary tone was found. See Figure 6.4.
69
Figure 6.3 Xiao Wang, ni zhou liu hai yao jia ban ma? “Xiao Wang, you have to work (even) on Saturday? Echo-surprise ma-question with H*H%.
Figure 6.4: Ni shi yao jin nian ba yue hui lao jia ma? “(Did you say) you are returning home in August? Echo-repetition ma-question with H*H%.
Early impressionist works on ma-question intonation in Chinese proposed both a general rising trend and a localized rising at the final-particle ma. (Chao 1933, 1968). However, later experimental researches confirmed the general rising trend but did not find local rise on ma. (Ho
70 1977, Garding 1987). Lee (2000) was the first study to observe local rise at the particle. With few exceptions, the results of this thesis are consistent with the findings of Lee (2000) in that a final local rise was observed at the end of the ma particle. However, a general rising trend described as an ‘up-stepped slope’ in previous literature (Garding 1987) was not found.
Lee (2000) observed that neutral and non-neutral Information-seeking yes-no questions may have different boundary tones, as one of her three speakers produced final falling intonation in Confirmation-seeking questions. However, the other two speakers used similar boundary tones in neutral Information-seeking and Confirmation-seeking questions. The results of the
Confirmatory questions in this thesis, although also drawn from one speaker, confirmed that final rising at ma is also used in Confirmation-seeking questions, similarly as in neutral Information seeking questions.
For Echo-surprise questions, Lee (2000) proposed similar pitch pattern with the neutral
Information seeking questions, but with a more exaggerated local rising trend. However, in the only Echo-surprise question attested in this study, a flat high tone is observed without exaggerated local rising. Since it is reflected in this study that Echo-surprise yes-no question may not be very common in the Beijing dialect, this example is probably not representative. It might be more desireable to study the intonation of echo-surprise yes-no questions in other dialects of MC.
6.2.2 A-not-A question intonation. Both high and low boundary tones were found in A- not-A questions. For the Information-seeking A-not-A questions, the only utterance produced by
C2 was realized with final rising tone (H%). See Figure 6.5. Similarly, C4 also produced two A-
71 not-A questions with high boundary tone. However, the other two uttarances produced by him were realized with low boundary tone (L%), marking a falling ending. See Figure 6.6:
Figure 6.5 Qing wen you mei you Disanxian? “Do you have Disanxian (a Chinese dish)?” Information-seeking A-not-A question with L*H%
Figure 6.6 Xia wu you mei you ke? “Do you have class in the afternoon?” Information-seeking A-not-A question with L*L%
72 For the Confirmatory A-not-A questions, C3 consistently added a sentence-final particle to the end of the utterances, such as the ‘ya’ in the following example. A low boundary tone (L%) is applied at the final particle. See Figure 6.7.
Figure 6.7 Xiao Wang, ni shi bu shi gan mao le ya? “Xiao Wang, did you catch a cold?” Confirmatory A-not-A question with L+H*L%
Same as C3, C4 used final falling tone, both when there is a sentence-final particle at the end (See Figure 6.8) and when there is not a final particle (See Figure 6.9).
For Echo-repetition A-not-A questions, utterances only come from C4. All the sentences were realized with a low boundary tone (L%) although no sentence-final particle was present. The pitch reaches its highest peak at the first verb of the ‘A-not-A’ phrase, falls, and rises again at the nuclear accent, and then falls at the end. See Figure 6.10.
73
Figure 6.8 Shi bu qu na er gou wu la? “You went shopping somewhere? Confirmatory A-not-A question with H*L%
Figure 6.9 Shi bu shi you dian er gan mao? “Did you catch a cold?” Confirmatory A-not-A question with L*L%
74
Figure 6.10 Jin tian shi bu shi you you men da xia chi? “(Did you say) we are eating shrimp today?” Echo-repetition A-not-A question with L*L%
It is found in previous studies that the first verb in A-not-A phrase in realized as a narrow focus and show an expansion of pitch range (Lee 2000, 2005). This confirms that A-not-A question is a verb-focused type of question (Huang et al. 2009, Lee 2000 etc.). This is also proved in this study by the high pitch at the beginning of the A-not-A phrase in most of the target utterances examined. At the same time, we also noticed that pitch range expands also at the stressed syllable of the item that is being repeated, which implies that the repeated word may also be a narrow focus.
6.3 Summary
As shown in this section, there is a variation of syntactic structures across the pragmatic contests investigated. On the one hand, judging from the total number of usage, it seems that ma- questions are favored in the neutral information-seeking and echo-repetition contexts while A-not-
A structures are favored in the Comfirmatory contexts. In addition, there appears to be some individual differences in the preference for certain syntactic structures. Besides, we also found that in Beijing Mandarin, echo yes-no question is rarely used for expressing surprise.
75 In terms of the pitch contours, we found final rising boundary tones in the ma-questions, and a local rise at the last syllable ma. The A-not-A questions, on the other hand, showed both rising and falling intonation in information-seeking context, and generally falling intonation in echo-surprise, confirmatory and echo-repetition contexts. There also seems to be an interaction with the narrow focus. The first verb of the A-not-A phrase and the repeated NP in the Echo- repetition questions are realized as a narrow focus.
76 CHAPTER 7
CONCLUSION
This chapter offers a final conclusion of the study. Section 7.1 recaps the main goals and of the study, and wraps up the results in light of the research questions. Section 7.2 points out the limitations of this study. Finally, Section 7.3 indicates some directions for future research.
7.1 Summary of study and results
7.1.1 Objectives. The objective of the study is to examine the intonation of yes-no questions in Puerto Rican Spanish and Beijing Chinese in four different pragmatic contexts: information- seeking, echo-surprise, confirmatory, and echo-repetition questions. While intonation of yes-no question in PRS has been investigated in some detail (Armstrong 2010, 2012), Chinese has not.
For Puerto Rican Spanish, a secondary goal was to investigate the impact of bilingualism in the intonation pattern of these four question types (see also Alvord 2006, 2010). In addition, for
Chinese, we examine the interaction of syntax with intonation in the pragmatic contexts under study, contributing to recent studies of the syntax-intonation interface (González and Reglero
2016).
7.1.2 Revisiting the Spanish results. The first research question of the study addresses what the intonation of yes-no questions are like in different pragmatic contexts in PRS. Based on the total percentages of the final contours, information-seeking, confirmatory, and echo-repetition questions have the L+H*L% nuclear configuration, while echo-surprise questions show a
H+L*L% final contour. These results are not consistent with previous findings. The L*+H L%
‘circumflex’ feature was proposed by Sosa (1999:232) for exclamations and emphasized wh- questions. In Armstrong (2010), circumflex pattern with an up-stepped H (L+¡H*L%) was found in echo-yes no questions (surprise). The H+L*L% configuration, on the other hand, was found in
77 broad-focused statements and confirmation yes-no questions in Armstrong (2010). However, in this study, we found a preference for the ‘circumflex’ patter without up-stepped H peak as the dominant final contour for information-seeking, confirmation and echo-repetition questions, while the echo-surprise questions are found to have a similar pattern as the broad-focused statements and confirmation questions in the previous studies (Armstrong 2010, Sosa 1999).
Our second research question is concerned about the effect of level of bilingualism on intonation. Comparing the behavior of the 3 participants, we see rather striking differences. Firstly, we find that the two speakers with English as their dominant language (P1 and P4) show a preference for the same type of final contour across different contexts. This ‘monotone’ is especially obvious for P4 as he used the H*LL%/HL*L% final contour for all his utterances. P1, on the other hand, applied the circumflex pattern L+H*L% for most of the tokens, except for having two L+H*H% final rising contour in two information-seeking questions. On the other hand,
Spanish dominant P2, showed the greatest level of variation across contexts. She preferred a final contour of L+H*L% for information-seeking and echo-repetition questions, H+L*L% for echo- surprise questions, and L+H*H% for confirmatory questions. Although it’s hard to draw any conclusion from only three speakers, these results suggest that the lack of native proficiency in
Spanish might have prevented the heritage speaker and the early bilingual speaker from effectively distinguish questions of different pragmatic uses by means of intonation.
Secondly, it is known that yes-no questions in PRS typically have falling intonation
(Armstrong 2010, Sosa 1999). However, in this study, we observe some instances of rising intonation in the questions. Considering that the speakers live in Florida where English and a large variety of Spanish dialects are spoken, the final rising tone could possibly come from influence of
English questions or other varieties of Spanish that the speakers are exposed to. The influence of
78 other Spanish dialects is especially probable for P1 since he had spent 9 months working in Chile shortly before the study. It is worth noticing that the rising intonation was attested from both P1 (a heritage Spanish speaker) and P2 (a native speaker who had lived in the U.S for about 3 years). To investigate exactly which factor played a more important role, it is necessary to verify if they use the same intonation when speaking English. It is also desirable to gather more information about their social network to see if they socialize much with speakers of English and/or of other Spanish dialects.
7.1.3 Revisiting the Chinese results. For the Chinese yes-no questions, we aimed to look at its syntactic and phonological variation across contexts. We found syntactic variation across different contexts and participants. Generally speaking, information-seeking and echo-repetition contexts favor ma-questions, while confirmatory context favors A-not-A structures. On the other hand, we also found a seemingly individual preference for one syntactic structure over the other.
Besides, the scarcity of yes-no questions in echo-surprise context suggests that echo yes-no questions may not be commonly used to express surprise in Mandarin Chinese, at least in the
Beijing dialect.
As for the phonological variation, we found that ma-questions have final rising boundary tone that maintains the pitch high at the last syllable ma regardless of the preceding tone in all contexts. The A-not-A questions intonation, on the other hand, showed both rising and falling intonation in information-seeking context and falling intonation in confirmatory and echo- repetition contexts. Consistent with Lee (2000), the first verb of the A-not-A phrase is often realized as a narrow focus. The pitch rises at the narrow focus, falls and rise again at the nuclear accent. In the echo-repetition questions in particular, the item being repeated is realized as another narrow focus.
79 7.1.4 Comparison between Chinese and Spanish. Putting together results of the two languages, we confirm our hypothesis that syntactic and phonological cues play different rules in
Spanish and Chinese. In Spanish, intonational cue is stronger than syntactic cue. In at least one of the participants (P2), different final pitch contours were used for different pragmatic contexts. We expect further measurements to reveal more differences in tonal range, segment duration, among other aspects. On the other hand, syntactic variation in Spanish yes-no questions in this study are limited to a few cases of insertion of tags such as ‘de verdad’, mainly in the confirmatory questions, as well as the use of the conjunction ‘o’ to form a disjunctive question. Subject-verb inversion could not be easily attested since most utterances address to second person (you) and do not contain overt subject as it is common for pronouns to be omitted in Spanish. In utterances where there is overt subject, a few instances of subject-verb inversion were found solely from P1. This is consistent with the claim in Hualde and Olarrea (2014) that Puerto Rican Spanish is one of the dialects in which subject-verb inversion does not occur in absolute interrogatives. The reason why
P1 applied inversion in several cases could be due to contact with other dialects spoken by his coworkers in Chile. Although how these minor syntactic changes may interfere with intonation needs to be looked at with more detail, it’s safe to conclude that syntactic variation is not as crucial as phonological variation in Spanish yes-no questions, and it’s basically through intonation changes that pragmatic meanings are conveyed.
On the contrary, in Chinese, we found syntactic cues to be more essential. There is a clear alternation between ma-particle and A-not-A structure and at least one participant (C3) used different structure in different contexts. Intonation, however, mainly change in accordance with the syntactic structures. Ma-questions showed final rising intonation and A-not-A questions have final falling intonation. The pragmatic meanings are basically expressed through certain syntactic
80 structures. Both ma particle and A-not-A structure can express interrogative meanings. A-not-A structure with the verb ‘shi’ (to be) ‘shi-bu-shi’, is commonly used to seek confirmation; while the verb ‘shi’ (to be) followed by the ma particle ‘shi…ma’ is used to seek repetition. Some particles added at the end of A-not-A questions also help to distinguish the meanings of the sentences.
7.2 Limitations
One of the limitations of the Spanish part of this study is the reduced number of participants and tokens analyzed. Four participants were initially recorded for Spanish, and four for Chinese, and target sentences were recorded twice per participant. However, due to time constraints, data from one repetition was analyzed for three participants of each language. Taking into consideration of the various factors that potentially affect intonation in this study, namely level of bilingualism
(divided into 3 levels) and pragmatic uses (divided into 4 different contexts), the number of tokens for each condition are far from desirable. Obviously, data from additional participants would provide deeper insight into the patterns described here. Especially, we only had one participant representing each of the 3 level of bilingualism. Although we can already attest some differences in their intonation, it’s not safe to draw the conclusion that those behaviors are caused by differences in language background, or contact with English and/other varieties of Spanish.
Another drawback of the methodology design is that we did not include declarative utterances to compare with the interrogative utterances. One of the key questions involving interrogative intonation, especially the yes-no questions in Spanish, is how the interrogative meaning is expressed through intonation to differentiate it from statements. Moreover, since the yes-no questions typically have final falling intonation that is similar to declarative sentences, it would be important to compare their contours to see how intonation is manipulated to tell questions from statements.
81 The study also needs to be improved in terms of the measurement of bilingualism of the participants. For that, the LHQ (Li et al. 2014) language history questionnaire was used. This questionnaire includes comprehensive questions that reflect the language history background of the speakers. However, its questions are mostly descriptive and give more qualitative information of the participants, such as how old they were when they started their L2 acquisition, and in what contexts they use one or the other language, etc. For the division of bilingualism in this study, however, we need a more quantitative and comparable numeric index that reveal the level of bilingualism of the participants and that permits a reliable comparison across individuals. For this purpose, the BLP (Birdsong et al. 2012.) may be a better questionnaire as it generates a score for each participant that indicates their language dominance.
For the Chinese part of the study, the limitations are also the limited number of participants and tokens, and the lack of comparison with statements. Besides, we admit that the effect of lexical tone on intonation has not been explored or taken care of to a desirable degree. Since it is known that the tone level of the stress-less sentence-final particle depends on the tone of the preceding syllable, the contexts were created to elicit utterances with only the level high Tone 1 before the final particle to control the intervention from the rising or falling of lexical tones. However, some utterances were produced ending with syllables with other tones.
Apart from that, the wording for eliciting in the Chinese materials might need to be reconsidered. Unlike in the Spanish materials where the verb ‘preguntar’ (to ask) was used to trigger the question, e.g. ¿Qué le pregunta a ella? (What do you ask her?), in the Chinese materials, the eliciting sentences contain the verb ´shuo´ (to say) and an underline, e.g. 你说____? (you say______?). Although it seems more natural for the author to elicit an utterance in this way in
82 Chinese, the verb ‘to say’ instead of ‘to ask’ might be a factor that led to the scarcity of yes-no questions in the echo-surprise context.
It is also worth mentioning that in the Chinese data, numerous examples of vowel and/or consonant reduction were noticed, many paired with the typical insertion of the ‘er’ (IPA [ɚ]) noun-suffix in Beijing dialect (Chinese Academy of Social Sciences (2012)). e.g.
‘jintian’(IPA[t͡ ɕin.tian] (today)) pronounced as ‘jier’ (IPA [t͡ ɕi ɚ]. Another example frequently found in the utterances of C3 and C4 is the A-not-A phrase ‘shi-bu-shi’ (IPA [ʂi-bu- ʂi], (be-not- be, translated as ‘is it...’)) pronounced as ‘shi-buer’ (IPA [ʂi-bu ɚ]). Although this is expected in
Beijing dialect, its potential impact on intonation is left for future study.
7.3 Future Study
Future studies should improve the limitations of the study by including declarative utterances and more participants and tokens. To start with, the current study can be expanded by including data from the 4th participant recorded and the first repetition of all four participants. This would double the number of tokens and gives us more suggestion on the consistency of the participants’ behavior. For further investigation, more participants should be recruited and at the same time more contexts should be created to get more tokens from each participant.
For the Spanish part of the project, more participant of different level of bilingualism should be recruited both from Puerto Rico and the U.S. We should also have monolingual control.
BLP should be used to better reflect their language profile. With enough participants, the bilingual index we get from them would probably allow us to see a correlation between the level of bilingualism and how much their intonation is affected by English. We expect to see that higher degree of dominance in English predicts greater influence of English.
83 For the Chinese part of the study, more speakers of the Beijing dialect should be included to get a more reliable pattern of the syntactic and phonological variation. Although Chinese-
English bilingualism is not a factor understudy in this thesis, it would be interesting to collect data from the monolingual Beijing Chinese speakers living in China and compare their performance with the bilingual Chinese living in the U.S. It is also desirable to include speakers from other dialects of Mandarin to see if they would use some other syntactic structures, like the syntactically unmarked or intonation-only yes-no questions, as well as if in any dialect yes-no questions are used in the echo-surprise contests. Also, how stress, lexical tone, and intonation interact also calls for more investigation.
In both languages, future study can also be done to examine the perception of yes-no question intonation. Perception test can be done with both monolingual native speakers and bilingual speakers to see if from intonation they would be able to tell the dialect of the speaker and the pragmatic meaning that the questions intend to convey.
Last but not the least, the differences between Spanish and Chinese in terms of yes-no question formation strategies found in this study may set up a theoretical and experimental platform for future research on the acquisition of Spanish question intonation by learners of L1
Chinese. It is interesting to see whether speakers of Chinese can successfully perceive and produce the intonational differences in Spanish yes-no questions of different pragmatic functions.
Especially in the case of echo-surprise questions, if indeed yes-no question is not a common way of expressing surprise in Chinese, would they have difficulty perceiving and producing echo- surprise questions in Spanish? These questions should be answered with future research with the growing population of L1 Chinese learners of Spanish.
84 APPENDIX A
INFORMED CONSENT FORM
85 APPENDIX B
MATERIALS
Spanish Material:
Information-seeking:
1. En el restaurante, quieres saber si tienen bizcocho de vainilla. ¿Qué le preguntas al
camarero?
2. Te encuentras con Natalia en la biblioteca, quieres saber si tiene clase por la mañana.
¿Qué le preguntas a ella?
3. Estás buscando un apartamento que tenga lavadora. Te comunicas con Karla porque
quieres saber si su apartamento tiene lavadora. ¿Qué le preguntas para saberlo?
4. Charlas con tu nueva compañera, quieres saber si sus padres viven en Guayama. ¿Qué le
preguntas para saber esta información?
5. Tu amigo tiene muchos libros en casa. Pregúntale si tiene Pasión de Gavilanes.
Echo-surprise:
1. Te encuentras con tu cuñada en la calle y te cuenta que tu hermano Guillermo ganó la
lotería, te sorprende mucho. Pregúntale si de verdad ganó Guillermo.
2. Tu amiga Andrea te ha regalado dos entradas para el concierto de Lady Gaga. Tu hermana
la odia, pero dice que quiere ir contigo al concierto. Te sorprende mucho. Pregúntale si de
verdad a ella le gusta Lady Gaga.
3. Vas a tu restaurante favorito con un amigo y pides mondongo, pero el camarero dice que
no tienen. Te sorprende mucho porque es uno de los platos típicos del restaurante.
Pregúntale al camarero si de verdad no tienen mondongo.
86 4. Llamas a Juan para ver si quiere pasar un rato contigo el domingo, pero te dice que tiene
que trabajar ese día. Eso te toma por sorpresa. Pregúntale si de verdad tiene que trabajar
el domingo.
5. Tu mejor amiga Mía te dice que su hermana Amanda está preñada, te sorprende mucho
porque solo tiene 17 años. Pregúntale si de verdad está preñada.
Confirmatory:
1. Ves a tu amigo mirando fotos de la Muralla Romana. Sabes que acaba de regresar de un
viaje a Europa, y parece que visitó la Muralla Romana. ¿Qué le preguntas para confirmarlo?
2. Ves a Natalia salir corriendo de la oficina con su teléfono, parece que tiene una llamada
¿Qué le preguntas para confirmarlo?
3. Ves a Carlos ojeando el catálogo de anillos. Tú sabes que pronto va a comprometerse con
su novia, parece que le va a comprar un anillo. ¿Qué le preguntas para confirmarlo?
4. Tu vecina viene a casa con botellas de coñac y vino. Tú sabes que va a dar una fiesta,
parece que fue a la bodega. ¿Qué le preguntas para confirmarlo?
5. Visitas la casa de tu amiga y ves que en el cuarto de su hijita hay muchas fotos de la Barbie
por las paredes. Parece que a su hijita le gusta mucho la Barbie. ¿Qué le preguntas para
confirmarlo?
Echo-repetition:
1. Tu amigo te dice que en mayo regresa a su país, no le has oído bien si dijo mayo porque
alguien acaba de tocar la bocina. Pregúntale si regresa en Mayo.
2. Tu amigo te dice que va a viajar a Aguadilla. No sabes si escuchaste bien porque en ese
momento tu perro comenzó a ladrar. Pregúntale si va a viajar a Aguadilla.
87 3. Tu madre te dice que va a hacer limonada para la fiesta. No sabes si escuchaste bien porque
en ese momento tu teléfono sonó y te distrajo. Pregúntale si va a hacer limonada.
4. Tu colega argentino Manuel te pide prestado una regla. No sabes si escuchaste bien porque
en ese momento otro compañero empezó a hablar por el teléfono. Pregúntale a Manuel si
quiere una regla.
5. Tu compañero de clase te dice que este semestre está tomando una clase de biología. No
sabes si escuchaste bien porque en ese momento un carro tocó la bocina y te distrajo.
Pregúntale si está tomando biología.
Chinese Material
Information-seeking:
1. 中午，你在 过道遇到同事小刘，你想问她是否 经吃 了，你说______?
2. 你在 书馆看见小孙，你想问他今天 午是否 课，你问______?
3. 你看见王老师领着她的儿子来学校了，你想问王老师儿子是否 经 学了，你问
______?
4. 你在餐厅点餐，你想问服 员是否 地 鲜，你问 ______?
5. 朋 到你家做客，你为他准备 料，想问他是否想在 料里 冰，你说______?
Echo-surprise:
1. 早 十点半，你见到小王在炒菜，你惊讶他 么早就吃午 了，你说______?
2. 周 晚 你打电话给小王，他说他在办 室工作，你很惊讶他周 要 班，你说
______?
88 3. 表 打电话说自 生了宝宝，你记得预产期在一个 以后，你惊讶她 么快就生
了，你说______?
4. 同事夫 在谈论儿子的高考，可你印象中他们的孩子才 初中，惊讶他怎么都高考
了，你说_____?
5. 新来的学妹告诉你她买车了，你知道她刚来美 不到一个 ，很惊讶她 么快就买
车了，你说______?
Confirmatory:
1. 你看见小王打喷嚏，你想问她是不是感冒了，你说______?
2. 你看见小杨在翻看自 在黄山拍的照 ，你知道她假期去安徽玩了，想问她是不是
去了黄山，你说______?
3. 家庭聚会间， 父拿着烟盒和打火机向门外走去，你想问他是不是出去抽烟，你说
______?
4. 午，你看见邻居王大 提着商场的袋子回家，你想问她是不是去买东西了，你说
______?
5. 你看见邻居一家拿着游泳圈出门，你想问他们是不是带孩子去海边，你说______?
Echo-repetition:
1. 你和室 在厨房做 。室 告诉你她明年 回老家。你没听清她是否说的 因
为此时微波炉响了。于是你问她是否是 回老家。 你说______?
2. 朋 告诉你小李要去新 坡留学，此时咖啡厅比较嘈杂，你没听清她是否说的新
坡。于是你向她确认小李是否要去新 坡。你说_____?
89 3. 你 跟你说今天吃油焖大虾。 时你的手机响了，你没听清 说的是不是油焖
大虾。于是你问她今天是否吃油焖大虾。你说______?
4. 办 室外地的同事请你帮她拿一卷 面胶。由于他的口音你没 听清。于是你向他
确认一 是否是要一卷 面胶，你说______?
5. 朋 在电话里跟你相约在肯德基，由于信 不好你没听清他说的是不是肯德基，于
是你向他确认是不是约在肯德基，你说______?
90 APPENDIX C
HUMAN SUBJECTS APPROVAL LETTERS
91 REFERENCES
Alvord. Scott M. (2006). Spanish Intonation in Contact: The case of Miami Cuban Bilinguals. PhD dissertation. University of Minnesota. Available at:http://works.bepress.com/scott_alvord/1/
Alvord, S. M. (2010). Variation in Miami Cuban Spanish Interrogative Intonation. Hispania 93(2), 235-255. The Johns Hopkins University Press. Retrieved March 8, 2017, from Project MUSE database.
Ao, Benjamin (1997). "History and Prospect of Chinese Romanization". Chinese Librarianship: an International Electronic Journal. 4.
Armstrong, Meghan E. 2010. Puerto Rican Spanish intonation. Transciption of intonation of the Spanish language, ed. by Pilar Prieto & Paulo Roseano, 155-189. München, Germany: Lincom EUROPA.
Armstrong, Meghan E. 2012. The development of yes-no question intonation in Puerto Rican Spanish. Columbus, Ohio: Ohio State University dissertation.
Beckman, M. & Hirschberg, J. (n.d.). The ToBI annotation conventions. http://www.ling.ohio- state.edu/~tobi/ame_tobi/annotation_conventions.html
Beckman, M., Díaz-Campos, M., McGory, J. T., & Morgan, T. A. 2002. Intonation across Spanish, in the Tones and Break Indices framework. Probus, 14(1), 9-36.
Beckman, M., Hirschberg, J., & Shattuck-Hufnagel, S. 2006. “The Original ToBI system and the evolution of the Framework.” In S. Jun (ed.) Prosodic Typology: The Phonology of Intonation and Phrasing, 84-117. Oxford: Oxford University Press.
Bolinger, Dwight. 1978. Intonation Across Languages. In: Universals of Human Language. vol. II. Phonology. Ed. by Greenberg, Joseph H. Stanford: Stanford University Press. pp.471-524.
Bolinger, Dwight. 1989. Intonation and Its Uses: Melody in Grammar and Discourse. Stanford: Stanford University Press.
Chao, Yuen Ren. 1930. A System of Tone Letters. Le Maître Phonétique, troisième série 30: 24- 27. (The Chinese translation in Fangyan 1980.2: 81-83.)
Chao, Yuen Ren. 1932. Guoyu yudiao [Intonation of Mandarin]. Lecture. Printed in Chao Yuen Ren yuyuanxue lunwenji. Eds. by Wu, Zongji and Zhao, Xinna. 2002. Beijing: Shangwu yinshuguan.
Chao, Yuen Ren. 1933. Tone and Intonation in Chinese. Bulletin of the Institute of History and Philology 4: 121-134.
92 Chao, Yuen Ren. 1948. Mandarin Primer. MA: Harvard University Press.
Chao, Yuen Ren. 1968. A Grammar of Spoken Chinese. Berkeley and Los Angeles: University of California Press.
Chinese Academy of Social Sciences (2012). Zhōngguó yǔyán dìtú jí (dì 2 bǎn): Hànyǔ fāngyán juǎn 中 语言地 集(第 2 ):汉语方言卷 [Language Atlas of China (2nd edition): Chinese dialect volume]. Beijing: The Commercial Press. p. 11.
Cruttenden, Alan. 1986. Intonation. Cambridge: Cambridge University Press.
Duanmu, San. 2000, 2007. The Phonology of Standard Chinese. Second Edition. Oxford University Press.
Elordieta, Gorka. (2003). The Spanish intonation of speakers of a Basque pitch-accent dialect. Catalan Journal of Linguistics 2.67-95.
Elordieta, Gorka, and Nagore Calleja. (2005). Microvariation in accentual alignment in Basque Spanish. Language and Speech 48.359-396.
EscandellVidal, Victoria. 1998. Intonation and procedural encoding: The case of Spanish interrogatives. In Rouchota and Jucker (eds.), pp. 169‐203.
EstebasVilaplana, Eva, and Pilar Prieto. 2008. La notación prosódica del español: una revisión del Sp_ToBI. Estudios de Fonética Experimental 17: 265‐283.
Estebas-Vilaplana, E. & Prieto, P. (2010). Castilian Spanish intonation. In Pilar Prieto & Paolo Roseano (eds.), Transcription of Intonation of the Spanish Language, 17-48. München:Lincom Europa.
Face, Timothy, (2008). The Intonation of Castilian Spanish Declaratives and Absolute Interrogatives. Lincom Europa.
Face, Timothy, (2011). Perception of Castilian Spanish Intonation: Implications for Intonational Phonology. Lincom Europa.
García Riverón, Raquel. 1996. Aspectos de la entonación hispánica: II análisis acústico de muestras del español de Cuba. Anejos del anuario de estudios filológicos, N.o 21. Cáceres: Universidad de Extremadura, Servicio de Publicaciones.
García Riverón, Raquel. 1998. Aspectos de la entonación hispánica: III las funciones de la entonación en el español de Cuba. Anejos del anuario de estudios filológicos, N.o 22. Cáceres: Universidad de Extremadura, Servicio de Publicaciones.
Gårding, Eva. 1983. A Generative Model of Intonation. In: Prosody: Models and Measurements. Eds. by Anne Cutler and Robert Ladd. Berlin: Srpinger Verlag. pp.11-26.
93 Gårding, Eva. 1984. Chinese and Swedish in a Generative Model of Intonation. In Nordic Prosody III. Eds. by Elert, Claes-Christian, Johansson, and Eva Strangert. Stockholm: University of Umeå. pp. 79 91.
Gårding, Eva. 1987. Speech Act and Tonal Pattern in Standard Chinese: Constancy and Variation. Phonetica 44: 13-29.
Gonzalez, C., & Reglero, L. (in press). Dime una cosa: Are wh-in-situ questions different in Spanish? Evidence from intonation. In Francisco Ordoñez, & Lori Repetti (Eds.), Linguistics Symposium of Romance Languages 46, Georgetown, Washington D.C. Amsterdam: John Benjamins.
Halle, M. and Stevens, K. (1971). ‘A note on Laryngeal Features’, RLE Quarterly Progress Report, 101> 198-213 (MIT).
Ho, Aichen T. 1977. Intonation Variation in a Mandarin Sentence for Three Expressions: Interrogative, Exclamatory and Declarative. Phonetica 34: 446-457.
Hu, Mingyang. 1985. Yuyan yu yuyanxue [Languages and Linguistics]. Wuhan: Hubei Jiaoyu Chubanshe.
Hualde, J. (2013). Los sonidos del español: Spanish Language edition. Cambridge: Cambridge University Press. doi:10.1017/CBO9780511719943.
Hualde, José Ignacio, Antxon Olarrea and Erin O'Rourke (eds). The Handbook of Hispanic Linguistics. Blackwell Publishing, 2012. Blackwell Reference Online. 17 April 2017 http://www.blackwellreference.com/public/book.html?id=g9781405198820_9781405198820
Huang et al. 2009. The Syntax of Chinese. Cambridge: Cambridge University Press.
Kaplan, Jeffrey P. (1989), English Grammar: Principles and Facts, Englewood Cliffs, New Jersey: Prentice-Hall
Kratochvil, Paul. 1998. Intonation in Beijing Chinese. In: Intonation Systems: A Survey of Twenty Languages. Eds. by Daniel Hirst and Albert Di Cristo. Cambridge: Cambridge University Press. pp. 417-431.
Ladd, D. Robert. 1996. Intonational Phonology. Cambridge: Cambridge University Press.
Lee, Ok Joo. 2000. The Pragmatics and Intonation of Ma-Particle Questions in Mandarin. M.A.thesis. The Ohio State University
Lee, Ok Joo. 2005. The Prosody Of Questions In Beijing Mandarin. Ph.D. dissertation. The Ohio State University.
94 Li, Charles N. and Thompson, Sandra A. (1981). Mandarin Chinese: a Functional Reference Grammar. Berkeley: University of California Press.
Li, P., Zhang, F., Tsai, E., Puls, B. (2014). Language history questionnaire (LHQ 2.0): A new dynamic web-based research tool.Bilingualism: Language and Cognition, 17(3), 673-680. DOI: 10.1017/S1366728913000606.
Lin, Tao. 1985. Tantao Beijinghua Qingsheng Xingshi de Chubu Shiyan [A Preliminary Experiment on Characteristics of Neutral Tone in Beijing Mandarin]. Beijing University Working Papers in Phonetics. Beijing: Beijing University Press. pp. 1-26.
Lleó, Conxita, Martin Rakow and Margaret Kehoe. 2004. Acquisition of language specific pitch accent by Spanish and German monolingual and bilingual children. Laboratory Approaches to Spanish Phonology, ed. by Timothy L. Face, 3-27. The Hague: Mouton.
Lyu, Shuxiang, 1985. “Yiwen, Fouding, Kending.” [Questioning, Negation, Affirmation]. Zhongguo Yuwen 4:241-250.
Miller, J. and Tench, P. (1981). ‘Aspects of Hausa Intonation, I: utterances in isolation. ‘Journal of the International Phonetic Association, 10, 45-63.
Montrul, Silvina (2008). Incomplete Acquisition: Re-examining the Age Factor. Philadelphia: John Benjamin North America. pp. 249–275.
Navarro Tomás, Tomás. (1944). Manual de entonación española. New York: Hispanic Institute of the United States.
O’Rourke, Erin. (2003). Peak alignment in Peru: Spanish intonation in contact with Quechua. Paper presented in the 33rd Linguistic Symposium on Romance Languages. Indiana University, Bloomington. April 24-27.
Peng, S., Chan, M., Tseng, C., Huang, T., Lee, O.J., & Beckman, M. (2006). “Towards a Pan- Mandarin system for Prosodic Transcription.” In S. Jun (ed.) Prosodic Typology: ThePhonology of Intonation and Phrasing, 84-117. Oxford: Oxford University Press.
Prieto, P. & Roseano, P. (eds.) (2010). Transcription of the Intonation of Spanish. Munich: Lincom Europa.
Prieto, P.; Aguilar, L.; Mascaró, I.; Torres-Tamarit, F.; Vanrell, M.M. (2009)
Polinsky, Maria; Kagan, Olga (2007). "Heritage languages: In the 'wild' and in the classroom". Language and Linguistics Compass. 1 (5): 368–395. doi:10.1111/j.1749-818x.2007.00022
Quilis, Antonio. 1987. Entonación dialectal hisp nica. In Lopez Morales and Vaquero (eds.), pp. 117 164.
95 Quilis, Antonio. 1993. Tratado de fonética española. Madrid: Gredos.
Robles-Puente, S. 2011. Absolute Questions Do Not Always Have a Rising Pattern: Evidence from Bilbao Spanish. In Selected Proceedings of the 5th Conference on Laboratory Approaches to Romance Phonology, ed. Scott M. Alvord, 98-107. Somerville, MA: Cascadilla Proceedings Project. www.lingref.com, document #2638.
Robles-Puente, S. 2012. Two Languages, Two Intonations? Statements and Yes/No Questions in Spanish and Basque. In International Journal of Basque Linguistics and Philology (ASJU), eds. Rebeka Campos-Astorkiza & Jon A. Franco. 46 (1), 252-262.
Shen, Xiao-Nan. (1990). The Prosody of Mandarin Chinese. California: University of California Press.
Sosa, Juan Manuel. 1999. La entonación del español. Madrid: Cátedra. Ultan, R. (1978). ‘Some general characteristics of interrogative systems.’ In J. H. Greenberg, C.A. Ferguson and E. A. Moravscsik (eds.), Universals of human language. Volum 4: Syantax. Stanford: Stanford University Press.
Valdés, G. (2005). "Bilingualism, heritage language learners, and SLA research: Opportunities lost or seized?". The Modern Language Journal. 89 (3): 410–426. doi:10.1111/j.1540- 4781.2005.00314.
Wang, Xu-Dong. 1992. Beijinghua de qingsheng quhua ji qi yingxiang [Change of neutral tone to falling tone in Beijing Dialect and its consequence]. Zhongguo yuwen 1992.2: 124-128.
Wang, Jing. 2003. Rhythmic Grouping, Tone Sandhi and Stress in Beijing Mandarin. Ph.D. dissertation. Macquarie University, Australia.
Willis, Erik W. 2003. Prenuclear low tone alignment in Dominican Spanish. Proceedings of the 15th ICPhS, pp. 2941‐2944.
Wu, Zong-Ji. 1982. Putonghua yuju zhong de shengdiao bianhua [Tonal Changes in Mandarin Sentences]. Zhongguo yuwen 1982.6: 439-450.
Yip, Moira. 1980. The Tonal Phonology of Chinese. Ph.D. dissertation. MIT.
Yip, Moira. (2002). Tone. Cambridge Textbooks in Linguistics. Cambridge: Cambridge University Press.
Yuan, Jiahong. 2004. Intonation in Mandarin Chinese: Acoustics, Perception, and Computational Modeling. Ph.D. dissertation. Cornell University.
Zemlin, W. R. (1981). Speech and Hearing Science: Anatomy and Physiology, 2nd edn. (Englewood Cliffs, NJ: Prentice-Hall).
96 Zeng, Suzanne Marie. 1996. A Pragmatic Study of Chinese Interrogatives. Ph.D.dissertation. University of Hawai’i.
97 BIOGRAPHICAL SKETCH
Linxi Zhang is originally from Beijing, China. She completed her B.A. in Spanish Language and Literature in Beijing Language and Culture University. She furthered her study towards a M.A. degree in Hispanic Linguistics in the Florida State University. She is interested in Phonetics and
Phonology, Sociolinguistics (including Sociophonetics) and Second Language Acquisition. She is also interested in the cross-language variations and has been working with her own language
Mandarin Chinese as well. She is going to start her PhD in Hispanic Linguistics in Georgetown
University and will focus on second and third language acquisition.
98