The development of interrogative forms and functions in early childhood Cantonese

Abstract

This study investigated question acquisition in Cantonese-speaking young children with a focus on the development of interrogative forms and functions. Data from a child Cantonese corpus (492 children aged 36, 48 and 60 months) were analysed. The main results were that: (1) all the interrogative forms and functions were produced by the three age groups and no age difference was found; (2) significant gender differences in favour of females were found; and (3) a significant form-function preference was obtained, with wh- and multiple questions being primarily used to seek information, and yes/no and rising tone/echo questions being largely used to request action.

Keywords

Cantonese early childhood question acquisition

Question acquisition in young children has been an important topic of continued investigation since Davis’s (1932) analyses of the development of interrogative forms and functions in early childhood English. In the past century, a considerable amount of research has been conducted to explore (1) developmental trends in interrogative forms (Berninger & Garvey, 1981; Erreich, 1984; Ervin-Tripp, 1970; Klima & Bellugi, 1966; Olsen-Fulero & Conforti, 1983; Piaget, 1926; Shatz, 1978; Tyack & Ingram, 1977); (2) the pragmatic functions of interrogatives directed at children (Loukusa, Leinonen, & Ryder, 2007; Shatz, 1979) and generated by children themselves (James & Seebach, 1982; Sachs & Devin, 1976); and (3) the development of interrogative forms and functions in first language (L1) acquisition (Clancy, 1989; Davis, 1932; James & Seebach, 1982; Meyer & Shrine, 1973; Vaidyanathan, 1988) and second language acquisition (L2) (Jackson & Bobb, 2009; Myles, Mitchell, & Hooper, 1999; Spada & Lightbown, 1993). These studies tend to conclude that children have acquired all the interrogative forms and functions of English and other European languages prior to starting school.

Although interrogative forms and functions have been examined in the context of a wide range of Western languages, there are very few studies on early acquisition of interrogatives in Cantonese, a major variety of spoken Chinese and a language with varying typologies of interrogative forms (Matthews & Yip, 1994). Wong and Ingram (2003) conducted a very important study reporting a significant age difference across the 1-3 years age group in acquiring three types of questions: particle/intonation, A-not-A and wh-questions. Little is known of developments post age 3 but it is reasonable to assume that older children are able to produce more forms of interrogatives and express different functions; however, empirical evidence about the nature of progress and changes with age is needed. The present study investigated the development of interrogatives in Cantonese-speaking children beyond 3 years of age.

Interrogative forms in Cantonese Chinese

The interrogative form refers to the lingual construction used to ask a question, for example, a sentence or an expression with question mark and/or final particles in some languages. Cantonese has a complex system of question presentation (Matthews & Yip, 1994; Wong & Ingram, 2003). Questions are indicated not by changes in word order, as in English, but by a number of interrogative constructions and sentence-final particles (Gao, 2000; Matthews & Yip, 1994; Wong & Ingram, 2003). Cantonese has many more sentence-final particles than Mandarin (the spoken language of Mainland China) and poses more problems for learners as well as for those who interact with them (Tse & Li, 2011). Wong and Ingram (2003) studied longitudinally the spontaneous questions of eight children between 1 and 3 years of age, focusing on three types of questions: particle/intonation questions (type I), A-not-A questions (type II) and wh-questions (type III). They found a significant age difference in the emergence of these question forms, the order of question acquisition being I > III > II. Their findings suggest that aspects of syntactic complexity, cognitive complexity and personal/social factors interact to determine the pattern of question acquisition.

The present study is a follow-up to Wong and Ingram’s (2003) research and investigated the development of interrogatives in Cantonese-speaking children beyond age 3. As older children may be expected to acquire mastery of a more extensive range of interrogative forms, the present study utilized Matthews and Yip’s (1994) typology. This includes the three types of questions analysed by Wong and Ingram and three other, more complex, types. Essentially, Matthews and Yip’s classification consists of six categories:

Yes/no questions: questions that may adopt different forms due to various applications, such as particle form (by adding a particle like aa1¹ and me1 to a declarative sentence; Wong and Ingram’s [2003] type I questions consist of these questions); A-not-A form (by repeating the verb or adjective while having the negative marker m4 between; similar to Wong and Ingram’s type II questions); copular form (by adding hai6 m4 hai6 or hou2 m4 hou2); mei6 form (by using the negative word mei6 [‘not yet’] to ask whether something has occurred); existential form (by using jau5 mou5 to ask whether something exists or not); tag form (using question tags and questions tends to presuppose a positive answer); and alternative form (questions that offer a choice between two alternatives and are expressed by conjoining two phrases with the conjunction ding6 or ding6 hai6).

Intonation and echo questions: indicated by rising intonation alone and primarily affecting the last word of the sentence, modifying or exaggerating the basic tone. Wong and Ingram’s type I questions also contain these questions.

Wh- questions: questions that involve ‘wh-words’, such as, bin1 (which), bin1 go3 (who), mat1 (what) and so on. These questions are comparable to the type III questions in Wong and Ingram’s classification.

Indirect questions: whose identification often depends on the content of the questions, for example, keoi5 man6 ngo5 soeng2 m4 soeng2 tai2 uk1 (‘She asked me if I wanted to see the house’).

Multiple questions: questions that have two or more question words appearing in a statement, for example, ngo5 dei6 gei2 si4 hai6 bin1dou6 gin3 min6 aa1? (‘Where and when are we going to meet?’).

Exclamatory/ rhetorical questions: used as exclamation and negation.

The first three categories of questions are comparatively simple as questions are formed by adding simple interrogative words or rising intonation. The latter three involve mastery of multiple question words and more complex sentence structures. We expected an interaction of type and age, the first three types expected to have been mastered by age 3 and the latter three types to develop after age 3.

Interrogative functions in early childhood language

The interrogative function refers to the pragmatic function of questioning. Chouinard (2007) hypothesized that ‘request for information’ (RfI) should be a universal function of child questioning. She argued that the ‘information requesting mechanism’ (IRM) is a key process used by children to learn about their world. Conducting a set of studies of the interrogative functions in over 100 American children aged 1;0 and 5;0, she proposed perhaps the most systematic classification of young children’s interrogative functions as follows: (1) information-seeking questions requesting (a) isolated facts without causal component, (b) explanations with causal component; (2) action-beseeching questions requesting (a) attention, (b) action, (c) permission and (d) play; and (3) non-information-seeking questions including (a) those addressing a child or animal who cannot answer and (b) all the questions that cannot be classified into any category. As expected, she found information-seeking to be the most dominant function (81%) of children’s questions.

Interesting and compelling as Chouinard’s findings might be, four major concerns have been raised (Harris, 2007; Maratsos, 2007). One concern is about the cross-cultural generality of the findings, given that all the children studied come from ‘a Western, industrialized culture’ (Chouinard, 2007, p. 107). In Chinese and many other cultures, adults do not converse on equal grounds so routinely with children and the latter are not encouraged to question authoritative adults, such as parents and teachers. Hence, the cultural influence on children’s interrogatives needs to be considered (Maratsos, 2007). The second concern is about the identity of the child’s conversation partner (Harris, 2007). In Chouinard’s (2007) studies, the children were asking questions primarily of their parents. However, young children usually also spend a large portion of their day in other settings interacting with other adults and peers. The third concern is about socio-economic status (SES) differences found to impact on question acquisition (Gullo, 1982a, 1982b). The fourth concern relates to gender differences. Davis (1932), reported a significant gender difference in the mean length of utterance (MLU) of questions, with girls producing longer questions than boys, and Tse, Kwong, Chan, and Li (2002) found the same pattern in early childhood Cantonese. This female superiority in MLU might be associated with some aspect of gender difference in question acquisition that deserves further study.

Other researchers have studied the pragmatic functions of interrogatives as if they are independent of their syntactic form. James and Seebach (1982) assert that this is not the case, at least not in English-speaking children. They classified the pragmatic functions of interrogatives into three major categories: the informational function, the directive function and the conversational function. In their sample of 24 preschoolers, they found a form-function preference. The yes/no questions were used as directives more often than any other question types, with wh-questions found to be particularly useful for initiating and/or maintaining conversations. A similar form-function preference was also found by Shatz (1979), again with English-speaking children. Whether this preference is also evident in childhood Cantonese Chinese has not been empirically addressed.

All the above concerns warrant study of young children whose mother tongue is not English; children who are raised in contrasting cultures; and children whose early years are spent conversing with peers from the same culture and with adults who are not their parents. The early childhood Cantonese corpus (Tse, Li, & Leung, 2007) provides an ideal arena for such a cross-linguistic investigation of early interrogative development. With this in mind, the authors re-analysed all the questions extracted from the corpus, and recoded their functions and forms according to the form and function systems developed by Matthews and Yip (1994) and Chouinard (2007). Two broad research questions guided the investigations:

RQ1. What are the interrogative forms produced by a representative sample of Cantonese-speaking young children? Does the Matthews and Yip (1994) typology apply to early childhood Cantonese?

RQ2. What interrogative functions are present in early childhood Cantonese? Does the English typology proposed by Chouinard (2007) also apply to Cantonese Chinese?

In addition, we tested the following predictions:

H1. There would be an age difference in the development of interrogative form and function in childhood Cantonese, with older children producing more types of interrogative forms and functions.

H2. There would be a gender difference in the development of interrogative form and function in childhood Cantonese, with girls producing more types of interrogative forms and functions.

H3. There would be a form-function preference in childhood Cantonese, with yes/no questions requesting action and wh-questions requesting information.

Method

Participants

The early childhood Cantonese corpus (Tse et al., 2007) represented the utterances produced by 492 Cantonese-speaking preschoolers randomly sampled from 68 preschools (58 kindergartens and 10 nurseries) located in Hong Kong Island, Kowloon and the New Territories. The preschoolers were selected randomly from the 68 preschools participating in the International Educational Achievement (IEA) Pre-primary Project, and consisted of all the Cantonese-speaking preschoolers between age 3;0 and 5;11 in Hong Kong. The number of children randomly drawn from each preschool was limited to 10 so as to minimize possible school effects, the final sample representing children from three age groups (age 3;0, 4;0 and 5;0), with 82 boys and 82 girls in each age group. All the participants were native speakers of Cantonese and their parents and teachers also spoke Cantonese at home and in the preschool. All the children had been learning English in preschools from the age of 3 years and their English language proficiency varied across families, preschools and social classes.

Communication task

The participants were paired randomly (boy/girl, boy/boy, or girl/girl). A toy play context area furnished with a set of toys, including cooking materials, food and fruit, furniture and electrical appliances and hospital materials and vehicles, was set up in the participants’ classroom. One dyad at a time was allowed to play in the room for 30 minutes. Their conversations were audio-recorded unobtrusively and the researchers observed the activities in the free play sessions but made no interventions.

Transcription

All conversations were audio-taped using high-fidelity equipment and assessed by two research assistants using a tape player that allowed automatic rewinding for repeat playing at slow and normal speeds. The research assistants transcribed these conversations to a level of detail that captured all words and word fragments (including overlapping speech) audible to the ear, non-lexical fillers (e.g. uh), as well as other vocalizations (e.g. laughter). Then, other researchers independently proofread the transcripts against the taped audio clips to ensure accuracy of transcription.

Coding

Trained research assistants filed the transcribed utterances onto a computer using Microsoft Chinese Windows and Office. Next, the Chinese script of each child’s oral language was segmented into utterances. A question was defined as a linguistic expression of enquiry that invites or calls for a reply. An interrogative was defined as the linguistic form of the question, and could be a word, a phrase or a sentence. All the interrogatives were analysed and independently coded according to systems developed by Matthews and Yip (1994) and Chouinard (2007) for forms and functions, respectively, by two research assistants. For example, the question ngo5 waan4 ne1 joeng6, nei5 waan4 ne1 joeng6, hou2 m4 hou2 (‘I play this, you play that, ok or not?’) was coded as a yes/no question in copular form (form) and an action-beseeching question which suggested an action (function); while the question nei5 giu3 zou6 me1meng4? (‘What is your name?’) was coded as a wh-question in what form (form) and an information-seeking question that requested a fact (function). The number of characters in each utterance in the transcriptions was used to indicate the length of utterance (Tse et al., 2002). There was a 93.8% level of agreement between the two coders’ coding of interrogative forms and functions, indicating excellent interrater reliability. The coding results were scrutinized by one author of this article; and, a panel of Cantonese linguists carefully went through the coded scripts to ensure consensus over the identification and categorisation of the variables.

Results

Altogether, 3140 types of Cantonese interrogative (13.12%) were elicited from the corpus of 23,936 utterances produced by the 492 children. No English interrogatives were found in this study, although some children did switch codes and produced a few English words. The MLU was 5.90 (SD = 2.79), 5.99 (SD = 2.63) and 6.14 (SD = 2.92) for the 3-, 4- and 5-year-olds, respectively. The MLU was 6.15 (SD = 2.94) and 5.85 (SD = 2.60) for girls and boys, respectively. Two-way ANOVA was applied to test the statistical significance of any differences in MLU, with Age (3) × Gender (2) as between-subject variables. A statistically significant gender difference was found in MLU, F = 8.63, df = 1, p < .005. No significant effects of Age or Age × Gender were found.

On average, each participant uttered 6.38 interrogative types during the half-hour period of communication. Some 1681 (53.5%) of the interrogatives were produced by the girls, the remaining 1459 (46.5%) by the boys. The interrogative types also appeared to be quite evenly distributed across the three age groups, with 1092 (34.8%), 980 (31.2%) and 1068 (34%) being uttered by the 3-, 4- and 5-year-olds, respectively. To explore the main effects of Age (3) and Gender (2) and two-way interactions, we used generalized estimating equations (GEE) in the study. GEE is a method developed for dealing with complex longitudinal, repeated/clustered data, in which the observations within each cluster are correlated with each other (Liang & Zeger, 1986). Regression analyses with the GEE methodology is a common choice when the outcome measure of interest is discrete (e.g. binary or count data, possibly from a binomial or Poisson distribution) rather than continuous. In this set of GEE analyses, separate models with Form and Function as the dependent variables were used, and Age and Gender were entered as covariates. Significant main effects of Gender (B =.24; SE = .11; χ² = 5.83; p = . 016) and non-significant Age (B = .04; SE = 0.55; χ² = .64; p =. 423) were found in the distribution of Form. No significant main effects of Age (B = .09; SE = .06; χ² = 2.63; p = .105) and Gender (B = .10; SE = .09; χ² = 1.00; p = .317) were found in the distribution of Function. Further analyses were executed to explore developmental trends in interrogative form and function and possible form-function preferences.

The development of interrogative form

As shown in Table 1, the three age groups of children produced all the six categories of Cantonese interrogative forms proposed by Matthews and Yip (1994). Yes/no (47%) appeared to be the most popular form used by the participants, followed by wh-questions (42.4%). In particular, wh-questions in ‘what’ form (23.22%) and particle form of yes/no questions (18.73%) were more often used by the children than other subtypes of interrogative forms.

Table 1.

Gender differences in interrogative forms and functions.

Interrogatives		Female		Male
		(N = 1681)		(N = 1459)
		n	%	n	%
Forms
Yes/no questions	Particle	322	19.2%	266	18.2%
	A-not-A	140	8.3%	82	5.6%
	Copular	169	10.1%	115	7.9%
	Mei6	36	2.1%	36	2.5%
	Existential	52	3.1%	37	2.5%
	Tag	117	7.0%	96	6.6%
	Alternative question	6	0.4%	1	0.1%
	Subtotal	842	50.1%	633	43.4%
Wh-questions	Who/whose	35	2.1%	50	3.4%
	What	355	21.1%	374	25.6%
	Which	22	1.3%	29	2.0%
	Where	72	4.3%	59	4.0%
	When	12	0.7%	4	0.3%
	Why	102	6.1%	85	5.8%
	How	76	4.5%	56	3.8%
	Subtotal	674	40.1%	657	45.0%
Indirect questions		1	0.1%	2	0.1%
Intonation/echo		147	8.7%	149	10.2%
Multiple questions		3	0.2%	4	0.3%
Rhetorical		14	0.8%	14	1.0%
Functions
Request for information	Name	14	0.8%	21	1.4%
	Location	103	6.1%	96	6.6%
	Description	37	2.2%	17	1.2%
	Others	469	27.9%	452	31.0%
	Subtotal	623	37.1%	586	40.2%
Request for action	Attention-seeking	93	5.5%	73	5.0%
	Declaratives	73	4.3%	60	4.1%
	Permission for action	23	1.4%	26	1.8%
	Promoting conversation	88	5.2%	72	4.9%
	Confirmation/verification	400	23.8%	345	23.6%
	Suggestion for action	227	13.5%	161	11.0%
	Prohibition/rejection	35	2.1%	26	1.8%
	Subtotal	939	55.9%	763	52.3%
Emotional expression		66	3.9%	57	3.9%
Egocentric speech		53	3.2%	53	3.6%

All the interrogative forms appeared to be quite evenly distributed across the three age groups and two genders, except for those for yes/no and wh-questions. A significant gender (2) difference was found in GEE, χ² = 5.83, df = 1, p = .016. Further analyses of each cell revealed that the girls tended to ask a greater proportion of yes/no questions (adjusted residual = 3.8) than the boys (adjusted residual = −3.8), while the boys (adjusted residual = 2.8) were more likely to ask wh-questions than the girls (adjusted residual = −2.8).

However, a non-significant age difference was found in GEE in the distribution of yes/no and wh-questions, χ² = .64, df = 2, p = .423. Further analyses of each cell revealed that: (1) in the yes/no questions, the adjusted residuals of 3-, 4- and 5-year-olds were 1.7, –2.0 and 0.2, respectively; (2) in the wh-questions, the adjusted residuals of 3-, 4- and 5-year-olds were −1.7, 2.0 and −0.2. These results indicate (although not statistically significantly) that the 4-year-olds tended to use fewer yes/no questions but more wh-questions than other groups. In summary, a significant gender (not age) difference was found in the interrogative forms uttered by the Cantonese-speaking young children.

The development of interrogative function

As can be seen in Table 2, the three age groups of children produced a total of 13 types of interrogative function that could be placed into the three major categories suggested by Chouinard (2007): (1) information-seeking (request for information, RfI), (2) action-beseeching (request for action, RfA) and (3) non-information-seeking. It is important to note that the interrogatives in early childhood Cantonese more often served the function of soliciting action (52.3%) than seeking information (40.2%) and other functions (7.5%). Within the subcategory of RfA questions, requesting confirmation/verification (23.8%) was the most frequently observed interrogative function in the early childhood Cantonese utterances. GEE analysis results indicated no significant Age or Gender differences, χ² s < 2.70, ps > .32.

Table 2.

Distribution of different interrogative functions over the forms (N = 3140).

Forms	Functions				Total
Forms	Request for information	Request for action	Emotional expression	Egocentric speech	Total
Yes/no questions	218 (14.8%)	1192 (80.8%)	41 (2.8%)	24 (1.6%)	1475 (100%)
Wh-questions	960 (72.1%)	274 (20.6%)	64 (4.8%)	33 (2.5%)	1331 (100%)
Indirect questions	1 (33.3%)	2 (66.7%)	0 (0%)	0 (0%)	3 (100%)
Intonation/echo	21 (7.1%)	219 (74.0%)	11 (3.7%)	45 (15.2%)	296 (100%)
Multiple questions	5 (71.4%)	2 (28.6%)	0 (0%)	0 (0%)	7 (100%)
Rhetorical	4 (14.3%)	13 (46.4%)	7 (25.0%)	5 (14.3%)	28 (100%)
Total	1209 (38.5%)	1702 (54.2%)	123 (3.9%)	106 (3.4%)	3140 (100%)

Note: The items in bold type show a significant form-function preference in the chi-square analysis.

It is important to note that, within the ‘non-information-seeking’ subtype of interrogative function, there was a minor difference in the ratios of emotional expression versus egocentric speech. The 3-year-olds produced relatively more egocentric speech than the older children (age 3;0.48, 58%; 4;0.25, 40%; age 5;0.33, 40%). This is in line with the developmental trend proposed by Piaget (1954) for young children in the pre-operational stage of cognitive development.

The form-function preference in Cantonese interrogatives

A form-function preference was evident in the production of interrogatives: the yes/no (80.8%) and intonation/echo (74%) questions were primarily used to request for action, whereas the wh-questions (72.1%) and the multiple questions (71.4%) were employed to request information. These outcomes in Table 2 showed that the young children tended to: (1) seek information more frequently with wh-questions; (2) request action more frequently with yes/no questions and rising tone/echo; (3) less likely to use wh-questions to request action; and (4) there was little use of yes/no questions and a rising tone to elicit information.

Discussion

This study provides robust evidence about the development of interrogative forms and functions in early childhood Cantonese. No significant age difference was found in interrogative form and function, thus H1 is rejected, whereas statistically significant gender differences were found in the forms and MLU providing empirical support for H2. Statistically significant form-function preferences were found, so H3 is also supported by the study. Elaboration on these findings is now offered.

Developmental trends in early Cantonese interrogatives

A most important finding in the study is that gender plays an important role in the incidence of interrogative forms in Cantonese children’s utterances. In line with our previous study (Tse et al., 2002), this study also found a gender difference in the MLU of interrogatives as well as the forms. The girls utilized yes/no questions more often than the boys, whereas the boys asked relatively more wh-questions than the girls. These results resemble the findings of Davis (1932) and Smith (1933) with English-speaking children, where boys were found to ask more causal questions using wh-questions, particularly how and why, than girls, and where girls were found to produce longer questions than boys. This study provides cross-linguistic evidence of females outperforming males during this period of language development, consistent with the findings of Bornstein, Hahn, and Haynes (2004), who followed 329 children from 1;1 to 6;10 and found that girls’ scores were consistently higher than those of boys in their language performance after the first year and before 7 years. It also shows that, during the period of middle childhood (2–6 years), girls on average are more generally advanced in language performance than are boys. A satisfactory explanation of existing gender differences in language performance involves interactions between biological, psychological and social variables.

It is interesting to note that non-significant age and gender differences were found in the interrogative functions produced by the Cantonese-speaking preschoolers. All the functions and forms suggested by Matthews and Yip (1994) and Chouinard (2007) could be identified in the utterances produced by the 3-year-olds. This is particularly in line with the findings of Chouinard (2007), who found no significant difference in interrogative functions across age groups after 30 months. The study also provides cross-linguistic evidence to support Chouinard’s notion that young children have acquired the interrogative functions by age 3 and can use different interrogatives to serve communicative purposes in early childhood.

Another noticeable finding is that the 492 children produced altogether 3140 types of interrogative during the half-hour communication sessions, generating a questioning rate of 12.76 per child per hour. This number is much less than the questioning rate reported in Tizard and Hughes’s (1984) study with English children. These researchers found that the children asked their mothers 26 questions per hour at home. However, when they analysed the same children’s conversations with their teachers in preschool, they found that the questioning rate was 2 per child per hour. They concluded that all children ask far fewer questions in preschool than they do at home (Tizard & Hughes, 1984). The young children in the present study were investigated in a free play context in a preschool, thus the questioning rate of 12.76 is reasonably acceptable.

Interrogative form-function preferences

An interesting form-function preference is found in the production of interrogatives: the young children tended to seek information more frequently with wh- and multiple questions and to request action more frequently with yes/no and rising tone/echo questions. In contrast, they were less likely to use wh-questions to request action and yes/no and rising tone questions to elicit information. Such a pattern is similar to that reported by Celce-Murcia and Larsen-Freeman (1999), in whose study yes/no questions were often used as commands and action requests in English-speaking children, while wh-questions were often used when the speaker missed a specific piece of information. Mehrani (2010) provides empirical evidence to account for this preference: yes/no questions carry a measure of suggestibility load and impose a restriction on children’s responses. They might hence be more suitable for requesting action.

James and Seebach (1982), however, report a different form-function preference in English-speaking children. Yes/no questions were used as directives more often than any other question type, and wh-questions turned out to be particularly useful for initiating and/or maintaining conversations. This form-function preference was verified by Shatz (1979). The difference in the preferred function of yes/no questions between their studies and those of the present authors might be connected with the different coding systems used in the two studies. We followed Chouinard’s (2007) classification that included three major functions: (1) information-seeking questions; (2) action-beseeching questions; and (3) non-information-seeking questions. James and Seebach (1982) classified the pragmatic functions of interrogatives into: the informational function, the directive function and the conversational function. These two different systems just share one common function: information-seeking. Nevertheless, the interaction between interrogative forms and functions deserves further cross-linguistic and cross-contextual study.

Request for action rather than a request for information

We found that 39% of the questions appearing in the early childhood Cantonese corpus were used to elicit information (RfI), whereas more than half (54%) were used to request action (RfA). Although RfA turned out to be the leading function in childhood Cantonese interrogatives collected in the present study, RfI was still frequently used in the free play context during peer conversations. In other words, this study also provides empirical evidence to support the hypothesis that the information-requesting mechanism (IRM) is universal across languages (including Cantonese) and circumstances (including in peer conversations). However, there are minor differences in the percentages of IRM between Cantonese- and English-speaking children that merit further attention.

First, these differences might be a reflection of the social nature of the communicative tasks in the study. The free play areas in the study were furnished with toys for boys as well as girls, and the participants (from the same class) were paired randomly and allowed to play in the room. During the play session, the communicative tasks were basically social and interactive, allowing the participants collaboratively to engage in role-play and to share or exchange interesting toys with their peer. Second, the difference might also be associated with the different interlocutors in the communicative tasks. In Chouinard’s study, the children were hopefully expected to address questions to parents or other adults in a very different learning context, whereas in our study the children were talking with peers. The different interlocutors and the different communicative tasks might have significantly influenced the purposes and functions of the interrogatives produced by the two samples, hence causing the minor differences. Further studies with same design but other Chinese languages such as Mandarin might be needed to see whether this difference is specific to the language of Chinese.

The present study has limitations. First, the sample is cross-sectional rather than longitudinal. Second, the age range examined should have extended both downwards and upwards in order to elicit a more complete picture of the development of interrogatives in the children. Nevertheless, the typology and pattern established in the present study offer a contribution to furthering understanding of the development of interrogative forms and functions in childhood Cantonese Chinese and to broadening our database of question development in different languages.

Footnotes

Acknowledgements

Thanks are expressed to Dr Terry Dolan, Dr Marie Ng, Ms S. K. Leung, Ms W. Y. Chio, Dr Carol Chan, Ms S. M. Kwong, Mr W. J. Lau, Ms L. Y. Fung, Miss Lilian Chau and the children and staff in participating preschools.

Funding

This work was partially supported by the GRF Fund (RGC Ref No. 747109), HKSAR.

Notes

References

Berninger

Garvey

(1981). Complementary balance in the use of the interrogative form by nursery school dyads. Journal of Child Language, 8, 297–311.

Bornstein

M. H.

Hahn

C. S.

Haynes

O. M.

(2004). Specific and general language performance across early childhood: Stability and gender considerations. First Language, 24, 267–304.

Celce-Murcia

Larsen-Freeman

(1999). The grammar book: An ESL/EFL teacher’s course (2nd ed.). Boston, MA: Heinle & Heinle.

Chouinard

(2007). Children’s questions: A mechanism for cognitive development. Monographs of the Society for Research in Child Development, 72, (1, Serial No. 286).

Clancy

P. M.

(1989). Form and function in the acquisition of Korean wh-questions. Journal of Child Language, 16, 323–347.

Davis

(1932). The form and function of children’s questions. Child Development, 3, 57–74.

Erreich

(1984). Learning how to ask: Patterns of inversion in yes-no and wh-questions. Journal of Child Language, 11, 579–592.

Ervin-Tripp

(1970). Discourse agreement: How children answer questions. In Hayes

J. R.

(Ed.), Cognition and the development of language (pp. 79–107). New York, NY: Wiley.

Gao

M. C. F.

(2000). Mandarin Chinese: An introduction. New York, NY: Oxford University Press.

10.

Gullo

D. F.

(1982a). A developmental study of low- and middle-class children’s responses to wh-question. First Language, 3, 211–221.

11.

Gullo

D. F.

(1982b). Facilitating children’s understanding of questions through pictures: Implications for social class language differences. First Language, 3, 121–138.

12.

Harris

P. L.

(2007). Commentary: Time for questions. In M. M. Chouinard, Children’s questions: A mechanism for cognitive development. Monographs of the Society for Research in Child Development, 72, (1, Serial No. 286), 113–120.

13.

Jackson

Bobb

S. C.

(2009). The processing and comprehension of wh-questions among second language speakers of German. Applied Psycholinguistics, 30, 603–636.

14.

James

S. L.

Seebach

M. A.

(1982). The pragmatic function of children’s questions. Journal of Speech and Hearing Research, 25, 2–11.

15.

Klima

Bellugi

(1966). Syntactic regularities in the speech of children. In Lyons

Wales

(Eds.), Psycholinguistic papers (pp. 183–208). Edinburgh, UK: Edinburgh University Press.

16.

Liang

K. Y.

Zeger

(1986). Longitudinal data analysis using general linear models. Biometrika, 73, 13–22.

17.

Loukusa

Leinonen

Ryder

(2007). Development of pragmatic language comprehension in Finnish-speaking children. First Language, 27, 279–296.

18.

Maratsos

M. P.

(2007). Commentary. In M. M. Chouinard, Children’s questions: A mechanism for cognitive development. Monographs of the Society for Research in Child Development, 72, 121–126.

19.

Matthews

Yip

(1994). Cantonese: A comprehensive grammar. London, UK: Routledge.

20.

Mehrani

M. B.

(2010). What is biased? Children’s strategies or the structure of yes/no questions? First Language, 31, 214–221.

21.

Meyer

Shrine

(1973). The form and function of children’s questions. Journal of Genetic Psychology, 123, 285–296.

22.

Myles

Mitchell

Hooper

(1999). Interrogative chunks in French L2. Studies in Second Language Acquisition, 21, 49–80.

23.

Olsen-Fulero

Conforti

(1983). Child responsiveness to mother questions of varying type and presentation. Journal of Child Language, 10, 495–520.

24.

Piaget

(1926). The language and thought of the child. New York, NY: Harcourt, Brace and Co.

25.

Piaget

(1954). The construction of reality in child. New York, NY: Ballantine Books.

26.

Sachs

Devin

(1976). Young children’s use of age-appropriate speech styles in social interaction and role-playing. Journal of Child Language, 3, 81–98.

27.

Shatz

(1978). On the development of communicative understandings: An early strategy for interpreting and responding to messages. Cognitive Psychology, 10, 271–301.

28.

Shatz

(1979). How to do things by asking: Form-function pairings in mothers’ questions and their relation to children’s responses. Child Development, 50, 1093–1099.

29.

Smith

M. E.

(1933). The influence of age, sex, and situation on the frequency, form and function of questions asked by preschool children. Child Development, 4, 201–213.

30.

Spada

Lightbown

P. M.

(1993). Instruction and the development of questions in L2 classrooms. Studies in Second Language Acquisition, 15, 205–224.

31.

Tizard

Hughes

(1984). Young children learning. London, UK: Fontana.

32.

Tse

S. K.

Kwong

S. M.

Chan

(2002). Sex differences in syntactic development: Evidence from Cantonese-speaking preschoolers in Hong Kong. International Journal of Behavioral Development, 26, 509–517.

33.

Tse

S. K.

(2011). Early child Cantonese: Facts and implications. Berlin, Germany: De Gruyter Mouton.

34.

Tse

S. K.

Leung

S. O.

(2007). The acquisition of Cantonese classifiers by preschool children in Hong Kong. Journal of Child Language, 34, 495–517.

35.

Tyack

Ingram

(1977). Children’s production and comprehension of questions. Journal of Child Language, 4, 211–224.

36.

Vaidyanathan

(1988). Development of forms and functions of interrogatives in children: A longitudinal study in Tamil. Journal of Child Language, 15, 533–549.

37.

Wong

Ingram

(2003). Question acquisition by Cantonese speaking children. Journal of Multilingual Communication Disorders, 1, 148–157.