In order to investigate children’s production of the Chinese modal verb hui, we collected the utterances from a corpus, called Chinese Corpora – CHILDES, and then analyzed each token of hui according to its meanings. In the present study, the second question focuses on the acquisition of the Chinese modal verb hui found in children’s utterances. Section 4.2.1 presents the overall findings of the Chinese modal hui, and Section 4.2.2 discusses the results.
4.2.1 Overall Findings
Table 4-4 illustrates the overall production of the five meanings of hui by the child groups and the adult group:
Table 4-4 Subjects’ overall Productions of the Five Meanings of Hui
Meaning Group
Dynamic Deontic Epistemic Generic Futurity
f % f % f % f % f % p-value
Child groups 55 34.38 5 3.125 61 38.13 28 17.5 11 6.87 2.2e-16***
Adult group 53 31.36 6 3.55 56 33.14 30 17.75 24 14.20 1.655e-10***
60
As can be seen in Table 4-4, the child groups and the adult group produced the Chinese modal hui differently, and the overall p-values showed that their production data were significantly differently (p = 2.2e-16 and p = 1.655e-10). Concerning the production of hui, the children uttered as many times the dynamic hui, the epistemic hui and the generic hui as the adults did. However, both the child groups and the adult
groups produced few sentences with the deontic hui and the futurity hui.
Table 4-5 shows the subjects’ overall productions of the five meanings of hui analyzed by the Chi-square test in R-studio:
Table 4-5 Overall Chi-square for the within-group between-type Differences Dynamic Deontic Epistemic Generic Futurity Child
As Table 4-5 shows the child groups produced the four meanings of hui similarly (dynamic, p = 0.7064; deontic, p = 0.2585; generic, p = 0.2676; futurity, p = 0.815;).
However, their production of the epistemic hui was significantly different (p = 0.0003).
Let us turn to the results of the adult group, their production of the Chinese modal hui was similar (dynamic, p = 0.6315; epistemic, p = 0.0572; generic, p = 0.406; futurity, p
= 0.145). Nevertheless, their utterance of the deontic hui was significantly different (p
61
= 0.0334). Therefore, the child groups and the adult group uttered the sentences with the five meanings of hui in a different degree.
Figure 4-2 shows the overall percentage of the five meanings of hui by the 3-, 4-, 5-, 6-year-old groups and the adult group:
Dynamic Deontic Epistemic Generic Futurity
f % f % f % f % f %
3 years 14 29.17 0 -- 25 52.08 7 14.58 2 4.17
4years 17 31.48 3 5.56 19 35.19 11 20.37 4 7.41
5 years 13 43.33 1 3.33 8 26.67 6 2 2 6.67
6 years 11 39.29 1 3.57 9 32.14 4 14.29 3 10.71
Adults 53 31.36 6 3.55 56 33.14 30 17.75 24 14.20 Figure 4-2 Overall Percentages of Each Group for the Five Meanings of Hui
As can be seen in Figure 4-2, the 3-, 4-year-old groups and the adult group produced more sentences with the epistemic hui (f = 25, 19, 56) than those with the dynamic hui (f =14, 17, 53). On the contrary, the 5- and 6-year-old children uttered more sentences
D Y N D E O E P I G E N F U T
3 years 4 years 5 years 6 years Adult
62
with the dynamic hui (f =13, 11) than those with the epistemic hui (f =8, 9). Regarding the generic hui and the futurity hui, all child groups (3-, 4-, 5-, and 6-year-olds) produced the utterances with these two hui’s similarly (i.e., the generic hui, f = 7, 11, 6, 4; the futurity hui, f =2, 4, 2, 3). As for the production of the deontic hui, all age groups uttered only few sentences with this hui (f = 0, 3, 1, 1, 6).
Figure 4-2 shows the overall production of the five meanings of hui by our five age groups. In general, the child groups and the adult group produced more utterances with the dynamic hui and the epistemic hui than those with the deontic hui, the generic hui and the futurity hui. It is also clear that the younger child groups’ and the adult
group’s utterances (3- and 4-year-olds) of the five meanings of hui outnumbered those of the older child groups (5- and 6-year-olds).
Table 4-6 shows the overall p-values for the within-type between-group differences of hui by the 3-, 4-, 5-, 6-year-olds and the adults:
63
Table 4-6 Overall P-values for the within-type between-group Differences of Hui Dynamic Deontic Epistemic Generic Futurity younger child groups and the older child groups were similar (3 vs. 4, p = 0.7769; 3 vs.
5, p = 0.8461; 3 vs. 6, p = 0.7432; 4 vs. 5, p = 0.8175; 4 vs. 6, p = 0.7332; 5 vs. 6, p = 0.9513). However, their production tokens of the dynamic hui were different from those of the adult group (3 vs. Adult, p = 0.0190; 4 vs. Adult, p = 0.0275; 5 vs. Adult, p = 0.0235; 6 vs. Adult, p = 0.0152). Second, concerning the deontic hui, both the child groups and the adult group produced sentences with this hui similarly, and the p-values were higher than 0.05 (3 vs. 4, p = 0.6782; 3 vs. 5, p = 0.9720; 3 vs. 6, p = 0.8130; 3 vs.
Adult, p = 0.0724; 4 vs. 5, p = 0.9028; 4 vs. 6, p = 0.9028; 4 vs. Adult, p = 0.6782; 5 vs. 6, p = 1.000; 5 vs. Adult, p = 0.1895; 6 vs. Adult, p = 0.2175). Third, regarding the production of the epistemic hui, it was found that the four child groups’ production data were similar (3 vs. 4, p = 0.9975; 3 vs. 5, p = 0.6430; 3 vs. 6, p = 0.6934; 4 vs. 5, p =
64
0.4382; 4 vs. 6, p = 0.2185; 5 vs. 6, p = 0.9999). Table 4-6 also shows that the younger child groups and the adult group produced the epistemic hui similarly in quantity (3 vs.
Adult, p = 0.0983; 4 vs. Adult, p = 0.1948), but the older child groups did not (5 vs.
Adult, p = 0.0019; 6 vs. Adult, p = 0.0025).
Finally, concerning of the generic hui and the futurity hui, the younger child groups
and the older child groups produced the utterances with the former hui (3 vs. 4, p = 0.9859; 3 vs. 5, p = 0.9971; 3 vs. 6, p = 0.9952; 4 vs. 5, p = 0.9679; 4 vs. 6, p = 0.8975;
5 vs. 6, p = 0.9310) and the latter hui similarly (3 vs. 4, p = 0.9988; 3 vs. 5, p = 1.000;
3 vs. 6, p = 0.8853; 4 vs. 5, p = 0.42678; 4 vs. 6, p = 0.71123; 5 vs. 6, p = 0.57663).
Nevertheless, their production of the generic hui (3 vs. Adult, p = 0.0240; 4 vs. Adult, p = 0.0485; 5 vs. Adult, p = 0.0167; 6 vs. Adult, p = 0.0078) and the futurity hui (3 vs.
Adult, p = 0.0360; 4 vs. Adult, p = 0.0206; 5 vs. Adult, p = 0.0360; 6 vs. Adult, p = 0.0408) were different from the adult group’s.
4.2.2 General Discussion
According to our findings, it is obvious that both the child groups and the adult group produced sentences with the dynamic hui and the epistemic hui more frequently.
However, they faced a challenge to utter the deontic hui and the futurity hui in their conversations. Moreover, as regards Greenburg’s (1966) Markedness Theory, the unmarked meaning is acquired earlier. The dynamic hui should be unmarked since it is
65
concrete and more frequently said in children’s daily life. On the contrary, the deontic and the epistemic meanings are more marked than the dynamic one. Since, acquiring these meanings require personal knowledge toward situations; therefore, children might feel them difficult to interpret and produce.
With respect to the productions of the deontic hui and the epistemic hui, Fan (2012)
argues that the former hui endows a less complex notion (an event feature) than the latter hui (a proposition feature); hence, the deontic hui may be easier to acquire than the epistemic hui. However, the present study did not support this claim because most of our child participants’ production of the epistemic hui outnumbered the deontic hui.
It is to say that the epistemic hui was easier than the deontic hui for the children in the present study. Hirst & Weil (1982) claim that children are able to produce the dynamic and the epistemic meanings, but that sometimes they may fail to produce the deontic meaning. This is because the deontic meaning requires children’s acquaintance or knowledge of the certain events (i.e., regulations, duties, customs, etc.). By contrast, the epistemic meaning can be used by children easily, since it only relates to the speaker’s or the subject’s individual opinion toward things. With respect to the generic hui and the futurity hui, it is noticeable that our participants found it easier to produce
sentences with the generic hui than those with the futurity hui because the futurity hui involves planning for the future which makes the children confused with the epistemic
66
hui. Therefore, it was difficult for our children to acquire.
As for the age factor, Guo (1995), Modyanova et al. (2010), Wang (2014), and Moscati et al. (2017) claim that children acquire the dynamic meaning when they are approximately at the ages of 3 to 4 and the deontic at 5 to 6, and finally the epistemic hui by 8. In the present study, the younger child groups’ (3- and 4-year-olds) production
of the dynamic hui was found in the corpus, indicating that the 3- and 4-year-old children have acquired this hui successfully.
Example (2) shows a 3-year-old child’s dynamic hui utterance in Chang’s (1998)
database:
(1) EXP: Ayi deu mei ting dao ni shuohua ai.
‘I haven’t heard your story yet.’
CHI: Wo bu hui jiang gushi.
‘I don’t know how to tell a story.’ (Chang 1998)
In the above conversation, the experimenter asked the kid to tell a story to her, and, the child then uttered a sentence with the dynamic hui to show that he did not know how to tell a story.
As for the production of the deontic hui, Noveck (2001), Modyanova et al. (2010), and Moscati et al. (2017) claim that children can acquire the deontic hui when they are 5-year-olds; however, we found that our 4-year-old children have already had an ability to produce sentences with this hui. Although, in the present study, the utterances of the deontic hui produced by the child groups were marginal, it is still clear that our
4-year-67
olds could produce the deontic hui utterances in their conversations.
Thus, our findings did not support these previous studies. Regarding to the epistemic hui, it is noticeable that our child participants produced a great number of sentences
with this hui. Moreover, they used this hui as early as the 3- and 4-year-olds. In other words, the epistemic hui did not cause them any difficulty in acquiring or producing it.
Example (3) is a conversation between an experimenter and a child in Chang’s
(1998) database:
(2) EXP: Ta mama zenyang?
What will her mother do next?
CHI: Ta mama hui ba nage [ dragon] da si!
Her mother will kill the dragon. (Chang 1998)
The experimenter asked the child to tell her a story about animals and what her mother did if a dragon was in the forest in the story. The kid replied that she would kill it. If children can produce utterances with the epistemic hui when they are 3- to 4-year-old, it means that the child may have already acknowledged and acquired this hui.
Consequently, our findings of the corpus study did not support the studies of Guo (1995), Noveck (2001), Modyanova et al. (2010), Wang (2014), and Moscati et al.
(2017). On the contrary, the present study supported Hirst & Weil’s (1982) and Gaidagri’s (2014) studies. Since they claimed that children before 5-year-old can actually acquire and produce the epistemic meaning rather than the deontic meaning.
Even though, there is no literatures on the generic hui and the futurity hui, the
68
utterances of these two hui’s in the corpus showed the 3-year-olds uttered seven sentences with the generic hui and two with the futurity hui. This is to say that they may have already acquired these two meanings at this age. In conclusion, our findings only supported the findings of Hirst & Weil (1982) and Gaidagri (2014).