Discourse Effect on Children’s Interpretation of Pre-Subject Exclusive Zhiyou ‘Only’

ABSTRACT

It has been reported for decades that preschool children (age 4-7) tend to assign non-adult-like interpretations for sentences with pre-subject exclusive only. This study reports findings from two experiments investigating (1) the effects of (in)congruent implicit questions in discourse contexts and (2) word order transformation on children’s interpretations of Mandarin pre-subject zhiyou ‘only’. Experiment 1 showed that children performed non-adult-like in the incongruent discourse condition, similar to the patterns found in previous studies, but adult-like in the congruent discourse condition. Experiment 2 showed that using sentences similar to English ‘Those who ate apples was only the boy’ where the VP precedes the subject NP with zhiyou ‘only’ was not sufficient to overcome the processing difficulty caused by the incongruent discourse for preschool children. The results demonstrate that a congruent context taking the VP as the background (i.e., given and salient information) in Experiment 1, but not the syntactic manipulation in Experiment 2, is sufficient for children to interpret sentences with pre-subject zhiyou ‘only’ at an adult-like level of accuracy. The findings demonstrate that children have the syntactic knowledge for sentences with exclusive zhiyou ‘only’, but the discourse incongruence hinders them from correctly interpreting sentences with exclusive zhiyou ‘only’ when it is in pre-subject position.It has been reported for decades that preschool children (age 4-7) tend to assign non-adult-like interpretations for sentences with pre-subject exclusive only. This study reports findings from two experiments investigating (1) the effects of (in)congruent implicit questions in discourse contexts and (2) word order transformation on children’s interpretations of Mandarin pre-subject zhiyou ‘only’. Experiment 1 showed that children performed non-adult-like in the incongruent discourse condition, similar to the patterns found in previous studies, but adult-like in the congruent discourse condition. Experiment 2 showed that using sentences similar to English ‘Those who ate apples was only the boy’ where the VP precedes the subject NP with zhiyou ‘only’ was not sufficient to overcome the processing difficulty caused by the incongruent discourse for preschool children. The results demonstrate that a congruent context taking the VP as the background (i.e., given and salient information) in Experiment 1, but not the syntactic manipulation in Experiment 2, is sufficient for children to interpret sentences with pre-subject zhiyou ‘only’ at an adult-like level of accuracy. The findings demonstrate that children have the syntactic knowledge for sentences with exclusive zhiyou ‘only’, but the discourse incongruence hinders them from correctly interpreting sentences with exclusive zhiyou ‘only’ when it is in pre-subject position.

Acknowledgement

I would like to thank the three anonymous reviewers and the editor for helpful comments and suggestions on earlier versions of this manuscript. Thanks also to the teachers and child/adult participants in the experiments and Chia-hsing Chen for the help with statistical analyses. Any remaining errors are of course my own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ Note that whether the prejacent, i.e., the part of the sentence without only as in (1a), is the presupposition of the whole proposition with only might be under debate among semanticists. There have been different analyses proposed, ranging from taking the prejacent as either the entailment, the strong presupposition, the weak presupposition, or the implicature. A detailed discussion regarding problems associated with these different analyses is beyond the scope of this paper but can be found in Ippolito (2007).

² Since the scenarios used in the experiment of pre-subject zhiyou “only” in Notley et al. (2009) were the same as those used in Zhou and Crain (2010), for the rest of this paper, we will refer to this type of scenario as the Zhou and Crain (2010) contexts. In the story corresponding to (8) used in Zhou and Crain (2010), the props were two gold coins and one silver coin.

³ In (9a) and (9b), the underlined parts indicate the background in prior contexts whereas the italics correspond to the members in the set for evaluation.

⁴ This study was approved by the Research Ethics Committee at the author’s institution (approval number 10304HS011).

⁵ Children’s justifications were either “Because X (the character mentioned in the test sentence) also performed action B” (the VP-association response), or “Because Y (the character not mentioned in the sentence) did that too” (the subject NP-association response), and hence not difficult to categorize. Similar analyses were also adopted in Experiment 3 of Notley et al. (2009) and Sugawara (2016). Occasionally children repeated the story when they were asked why they thought the puppet’s statement was not what happened in the story, and those were not categorized as either subject NP-association responses or VP-association responses.

⁶ In traditional Chinese syntax, sentences like (i) below are analyzed as pseudo-clefts, with de and shi functioning like what and the copula in English pseudo-clefts (e.g., Huang, 1982; Tang, 1979). However, alternative analyses also exist, e.g., predicate inversion as proposed by Cheng (2008). For a pseudo-cleft sentence without pre-subject zhiyou “only” as in (i), the copula shi is required, but a pseudo-cleft with pre-subject zhiyou “only” as in (12), shi is usually omitted. Note that the sentence in (12) cannot be a relative clause because zhiyou “only” must appear before a relative clause rather than between de and the head noun in Chinese.

(i) Dadao nanhai de shi na-ke qiu

hit boy DE BE that-CL ball

“What hit the boy was that ball.”

Additional information

Funding

This work was supported by the Ministry of Science and Technology, Taiwan [MOST 104-2410-H-007-041].