Hsieh & Hsu (2010)

Hsieh & Hsu (2010) was the first empirical research to especially focus on the contextual effect and linguistic convention in Mandarin-speaking children’s idiom comprehension.

Although there is abundant previous literature discussing first language acquisition of idiom comprehension and production, these studies investigated the native languages of English (Abkarian et al. 1992) or Italian (Levorato & Cacciari 1992, 1995, 1999, 2002, 2004) speakers.

In addition, Hsieh & Hsu (2010) highlighted the fact that the characteristics of Chinese idioms are different from idioms in other languages. There are certain syntactic constraints within the structures of Chinese idioms. In (16), the syntactic structure in the Chinese idiom ‘qingmei zhuma’ is adjective with complement.

(16) qing mei zhu ma green plum bamboo horse

‘childhood friends’ (Hsieh & Hsu 2010:506)

Hsieh & Hsu (2010) targeted three major effects including linguistic convention familiarity, and context since these three factors were reported to be significant in previous studies. First of all, the non-literal meaning of figurative expression comes into being when the language community agreed upon the convention (Hsieh & Hsu 2010). The idiom in (9), for

instance, is used metaphorically to represent the relationship between childhood friends. This metaphorical meaning ‘childhood friends’ was derived from linguistic convention since

qing-mei-zhu-ma originally referred to the name of a childhood game. However, there was a

contradictory finding about the children’s age at which linguistic convention could affect idiom comprehension. Cacciari & Levorato (1989) suggested that the age should be 7 years old but Ackerman (1982) reported the age was 10 years old.

In previous research, contextual inference had been reported to be an influential factor in children’s idiom comprehension (Abkarian et al. 1992, Gibb 1991, Levorato & Cacciari 1995,

1999). The crucial point was at which age children can search for cues in context since there was a discrepancy in the findings which were reported in the related studies. The last issue was familiarity of idioms since previous studies (Nippold & Rudzinski 1993, Levorato & Cacciari 1992, Nippold & Taylor 1995, Nippold et al. 1996) had different opinions regarding the

familiarity effect on children idiom comprehension. The aim of Hsieh & Hsu (2010) was to investigate how and when these three linguistic factors affected children’s idiom

comprehension.

The first experiment was a word-card task with Mandarin-speaking children in two groups.

This experiment examined how 6-year olds and 9-year olds interpreted the tested idioms. The first group was composed of 16 children (Mean age=6;1) and the second group also had 16 children (Mean age=9;6). The designed content of the task was based on plant-related Chinese

idioms since Hsieh & Hsu (2010:507) stressed that “plant names are popular vehicles in languages.” That is, plant-related idiomatic expressions serve as a necessary role in children’

perception and comprehension. The control group in Experiment 1 was composed of 16 adults whose age was between 23 to 35.

Regarding the materials in Experiment 1, 27 elementary school teachers and 19 kindergarten teachers were chosen as raters to ensure the familiarity of each idiom. Based on a four-point scale (1 point being very unfamiliar; 4 points being very familiar), they were asked to rate a list of thirty idioms chosen from the Academia Sinica Balanced Corpus of Mandarin Chinese. Seven familiar idioms and seven unfamiliar idioms were selected as the tested idioms in Experiment 1. The subjects were presented with one word card with a printed idiom each time and asked to explain the meaning of the target idiom on the word card. The experimenter would read aloud each tested idiom first and then the child subjects had to respond with their

interpretation of the idiom. All responses were recorded for the purpose of data coding.

According to the results in the first experiment, both age and familiarity were reported to significantly affect the subjects’ responses. Regarding the age effect, the adult group expectedly

had the best performance among all the subjects and the youngest group had more incorrect answers than the 9-year-olds. As for familiarity, all three groups responded with more correct answers in the familiar idiomatic type than in the unfamiliar idiomatic type.

The design of the second experiment followed the same design in Laval (2003), also being a comprehension task combined with a meta-pragmatic task. The comprehension task required the subjects to choose the pictures which were the literal or idiomatic meanings of the figurative expressions. In addition, the subjects also needed to explain the tested idioms and their verbal explanations were recorded for the analysis of pragmatic part. The outcome of the meta-pragmatic part was then used to compare with the results of the French children in Laval (2003).

Based on the findings of the picture selection task in Experiment 2, all groups displayed a tendency to choose the literal picture in the literal context. However, if the non-literal context was provided, the 9-year-old group tended to give more idiomatic answers. The youngest group also depended on the aid of context to select the idiomatic interpretation, indicating that idiomatic context was indeed a significant factor for the child groups even as young as six years old. In addition, two-way significant interaction was also found in familiarity and age. The six-year-old group had the lowest percentage of idiomatic responses in both the familiar and

unfamiliar conditions. This result was significantly different in the adult and the 9-year-old groups.

As for the meta-pragmatic analysis, the recorded answers were categorized into four categories: explanation related to linguistic convention, literal meaning, context, and other factors. Based on a Chi-square test, the age factor was found to be significantly different in these four categories of explanations. For the linguistic convention, the adult group had the most explanation related to this category and the six-year-old group had the fewest. The adult group possessed the ability to relate linguistic convention to the idioms but the youngest group mostly relied on the context. Overall, Hsieh & Hsu (2010:519) proposed that “the effect of meta-pragmatic knowledge is apparent after age 6 and was clearly established by age 9.”