'Wh'-Words: Existential or Universal Quantifiers in Child Mandarin?
Saved in:
| Title: | 'Wh'-Words: Existential or Universal Quantifiers in Child Mandarin? |
|---|---|
| Language: | English |
| Authors: | Haiquan Huang (ORCID |
| Source: | Journal of Psycholinguistic Research. 2024 53. |
| Availability: | Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/ |
| Peer Reviewed: | Y |
| Publication Date: | 2024 |
| Document Type: | Journal Articles Reports - Research |
| Descriptors: | Foreign Countries, Mandarin Chinese, Language Acquisition, Preschool Children, Language Usage, Semantics, Language Universals |
| Geographic Terms: | China |
| DOI: | 10.1007/s10936-024-10079-4 |
| ISSN: | 0090-6905 |
| Abstract: | "Wh"-words have been analysed as existential quantifiers (Chierchia in Logic in grammar: polarity, free choice, and intervention. Oxford University Press, Oxford, 2013; Fox, in Sauerland U, Stateva P (eds) Presupposition and implicature in compositional semantics (Palgrave studies in pragmatics, language and cognition). Palgrave MacMillan, Houndmills, pp 71-120, 2007; Liao in Alternative and exhaustification: non-interrogative uses of Chinese "wh"-words . Harvard University, 2010) or universal quantifiers (Nishigauchi, in: Theoretical and applied linguistics at Kobe Shoin 2, Kobe Shoin Institute for Linguistic Sciences, 1999). These two accounts have distinct predictions on how children initially interpret "wh"-words. The universal account predicts that children should initially interpret "wh"-words as universal quantifiers, whereas the existential account anticipates that children should start out with the existential interpretation. To adjudicate between the two accounts, the present study was designed to explore pre-schoolers' semantic knowledge of "wh"-quantification. Specifically, it investigated the interpretation of the "wh"-word "shenme" 'what' with 4-and 5-year-old Mandarin-speaking children and a control group of adults. Using a Truth Value Judgment Task (Crain and Thornton in Investigations in universal grammar: a guide to experiments on the acquisition of syntax and semantics. MIT Press, Cambridge, 1998), Experiment 1 evaluated whether children interpret the "wh"-word "shenme" 'what' as closer in meaning to the polarity sensitive item "renhe" 'any' or the universal quantifier "suoyou" 'all' in the antecedent of "ruguo" 'if' conditionals. Using a Question-Answer Task, Experiments 2 & 3 respectively investigated whether children interpret "shenme" 'what' as closer in meaning to "renhe" 'any' or "suoyou" 'all' in two types of questions: yes-no questions with the particle "ma" and A-not-A questions. It was found that both children and adults interpret "shenme" 'what' as closer in meaning to "renhe" 'any' than "suoyou" 'all'. The findings suggest that Mandarin-speaking pre-schoolers already have adult-like semantic knowledge of "wh"-quantification: "wh"-words are existential quantifiers rather than universal quantifiers. Due to the paucity of primary linguistic input, children's early mastery of the non-interrogative "wh"-words appear to support the biolinguistic approach to language acquisition (Chomsky in Aspects of the theory of syntax. MIT Press, Cambridge, 1965; Pinker in Language learnability and language development. Harvard University Press, Cambridge, 1984; Crain et al. in Language acquisition from a biolinguistic perspective. Neurosci Biobehav Rev, 2016. https://doi.org/10.1016/j.neubiorev.2016.09.004). |
| Abstractor: | As Provided |
| Entry Date: | 2024 |
| Accession Number: | EJ1427651 |
| Database: | ERIC |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwEyAhqS3hYrkG7pX6FsRz4_AAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDMax0fla0RCebnvQ9wIBEICBmko1r_QNecv5kOhgArUVD8wHaZTbIOw7Jd8RXCW4TCMu3UXTaVBv522jo9BfDIESg_EEt08hOAiIH7kxSfp-jYi4doN1w7UEIdIFheLI4ruMtsscgm_Yq2kN7H9y6r0Lh4HtLs0nXGRPcxp9AjDWQYCvZeoMGzZPKWhG3rkWOGVBbGRM5f_uJTuULF44O2cmyz-sZYgzJ52DpNM= Text: Availability: 1 Value: <anid>AN0177233171;41p01jun.24;2024May31.13:00;v2.2.500</anid> <title id="AN0177233171-1">Wh-Words: Existential or Universal Quantifiers in Child Mandarin? </title> <p>Wh-words have been analysed as existential quantifiers (Chierchia in Logic in grammar: polarity, free choice, and intervention. Oxford University Press, Oxford, 2013; Fox, in Sauerland U, Stateva P (eds) Presupposition and implicature in compositional semantics (Palgrave studies in pragmatics, language and cognition). Palgrave MacMillan, Houndmills, pp 71–120, 2007; Liao in Alternative and exhaustification: non-interrogative uses of Chinese wh-words. Harvard University, 2010) or universal quantifiers (Nishigauchi, in: Theoretical and applied linguistics at Kobe Shoin 2, Kobe Shoin Institute for Linguistic Sciences, 1999). These two accounts have distinct predictions on how children initially interpret wh-words. The universal account predicts that children should initially interpret wh-words as universal quantifiers, whereas the existential account anticipates that children should start out with the existential interpretation. To adjudicate between the two accounts, the present study was designed to explore pre-schoolers' semantic knowledge of wh-quantification. Specifically, it investigated the interpretation of the wh-word shenme 'what' with 4-and 5-year-old Mandarin-speaking children and a control group of adults. Using a Truth Value Judgment Task (Crain and Thornton in Investigations in universal grammar: a guide to experiments on the acquisition of syntax and semantics. MIT Press, Cambridge, 1998), Experiment 1 evaluated whether children interpret the wh-word shenme 'what' as closer in meaning to the polarity sensitive item renhe 'any' or the universal quantifier suoyou 'all' in the antecedent of ruguo 'if' conditionals. Using a Question–Answer Task, Experiments 2 &amp; 3 respectively investigated whether children interpret shenme 'what' as closer in meaning to renhe 'any' or suoyou 'all' in two types of questions: yes–no questions with the particle ma and A-not-A questions. It was found that both children and adults interpret shenme 'what' as closer in meaning to renhe 'any' than suoyou 'all'. The findings suggest that Mandarin-speaking pre-schoolers already have adult-like semantic knowledge of wh-quantification: wh-words are existential quantifiers rather than universal quantifiers. Due to the paucity of primary linguistic input, children's early mastery of the non-interrogative wh-words appear to support the biolinguistic approach to language acquisition (Chomsky in Aspects of the theory of syntax. MIT Press, Cambridge, 1965; Pinker in Language learnability and language development. Harvard University Press, Cambridge, 1984; Crain et al. in Language acquisition from a biolinguistic perspective. Neurosci Biobehav Rev, 2016. ).</p> <p>Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</p> <hd id="AN0177233171-2">Introduction</hd> <p>This paper reports experimental results on Mandarin-speaking children's semantic knowledge of <emph>wh</emph>-quantification. In Mandarin Chinese, <emph>wh</emph>-words exhibit interpretive variability in that they may have an interrogative or non-interrogative interpretation, depending on the linguistic environments in which they appear (Cheng, [<reflink idref="bib2" id="ref1">2</reflink>], [<reflink idref="bib3" id="ref2">3</reflink>]; Huang, [<reflink idref="bib21" id="ref3">21</reflink>]; Li, [<reflink idref="bib31" id="ref4">31</reflink>]; Lin, [<reflink idref="bib34" id="ref5">34</reflink>], [<reflink idref="bib35" id="ref6">35</reflink>]). To illustrate, consider sentences (<reflink idref="bib1" id="ref7">1</reflink>)—(<reflink idref="bib3" id="ref8">3</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(1)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chile&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left"&gt;&lt;p&gt;(interrogative interpretation)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="9"&gt;&lt;p&gt;'What did zhangsan eat?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(2)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xiang&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qing&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gaosu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;(existential interpretation)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;want&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;please&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;tell&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="9"&gt;&lt;p&gt;'If you want to eat something, please let me know.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(3)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left"&gt;&lt;p&gt;(universal interpretation)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="9"&gt;&lt;p&gt;'Zhangsan eats everything.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>In sentence (<reflink idref="bib1" id="ref9">1</reflink>), the <emph>wh</emph>-word <emph>shenme</emph> 'what'[<reflink idref="bib1" id="ref10">1</reflink>] receives an interrogative interpretation. In sentence (<reflink idref="bib2" id="ref11">2</reflink>), the <emph>wh</emph>-word <emph>shenme</emph> 'what' resides in the antecedent of a conditional statement, and it yields an existential interpretation, with a meaning corresponding to <emph>something</emph> in English. In sentence (<reflink idref="bib3" id="ref12">3</reflink>), the <emph>wh</emph>-word <emph>shenme</emph> 'what' occurs to the left of the quantificational adverb <emph>dou</emph> 'all'. The quantificational adverb <emph>dou</emph> 'all' turns the sentence into a universal declarative statement, and the <emph>wh</emph>-word <emph>shenme</emph> 'what' receives a universal interpretation, with a meaning corresponding to <emph>everything</emph> in English.</p> <p>As illustrated in the above examples, <emph>wh</emph>-words in Mandarin Chinese typically receive an interrogative interpretation, but they yield a non-interrogative interpretation in certain linguistic contexts. When interpreted non-interrogatively, a Mandarin <emph>wh</emph>-word may receive an existential or universal interpretation. Which interpretation it receives is determined by the linguistic contexts where it occurs.</p> <p>In addition to Mandarin Chinese, Japanese <emph>wh</emph>-words also exhibits interpretive variability (Nishigauchi, [<reflink idref="bib37" id="ref13">37</reflink>]). Specifically, Japanese <emph>wh</emph>-words typically receive an interrogative interpretation, but they can also function as a quantificational expression. To illustrate, consider Japanese sentences (<reflink idref="bib4" id="ref14">4</reflink>) and (<reflink idref="bib5" id="ref15">5</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(4)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;D&amp;#257;re-mo-ga&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ki-ta&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" /&gt;&lt;td align="left"&gt;&lt;p&gt;-Nom came&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="3"&gt;&lt;p&gt;'Everyone came.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(5)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Dare-mo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ko&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;na-ka-ta&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" /&gt;&lt;td align="left"&gt;&lt;p&gt;Come not Past&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="2"&gt;&lt;p&gt;'Nobody came.'&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;(Adopted from Nishigauchi, &lt;xref ref-type="bibr" rid="bibr37"&gt;1999&lt;/xref&gt;: 45)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As illustrated, the <emph>wh</emph>-word <emph>dare</emph> 'who' combined with the particle <emph>mo</emph> generates two interpretations<emph>: dāre-mo,</emph> with the first mora accentuated, yields a universal interpretation<emph>,</emph> whereas <emph>dare-mo</emph> without accentuation is interpreted as a negative polarity indefinite.</p> <p>Due to the interpretive variability of <emph>wh</emph>-words across languages, there has been a long-standing debate on the nature of <emph>wh</emph>-quantification. On one account, <emph>wh</emph>-words are analysed as universal quantifiers (Nishigauchi, [<reflink idref="bib37" id="ref16">37</reflink>]). We refer to this as the universal account. In line with this account, it has been proposed that children start out with the universal interpretation of <emph>wh</emph>-words based the experimental findings of Roeper and de Villiers ([<reflink idref="bib40" id="ref17">40</reflink>]). Specifically, Roeper and de Villiers ([<reflink idref="bib40" id="ref18">40</reflink>]) conducted a series of experiments on English-speaking children's interpretation of <emph>wh</emph>-words. In one study, they contrasted children's interpretation of double <emph>wh</emph>-questions such as (<reflink idref="bib6" id="ref19">6</reflink>) with single <emph>wh</emph>-questions such as (<reflink idref="bib7" id="ref20">7</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;(6)&lt;/p&gt;&lt;/td&gt;&lt;td align="left" colspan="2"&gt;&lt;p&gt;Who ate which fruit?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="2"&gt;&lt;p&gt;(7)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;a.&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Who ate fruit?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;b.&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;The family ate what?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As shown above, (<reflink idref="bib6" id="ref21">6</reflink>) differs from sentences in (<reflink idref="bib7" id="ref22">7</reflink>) as it calls for exhaustive pair-list answers. (7a) can be replied by pair-list answers, but it is not obligatory. (7b) should be answered by providing the information queried by the <emph>wh</emph>-word.</p> <p>It was found that 78.1% of the 4–6-year-olds provided an exhaustive pair-list answer to double <emph>wh</emph>-questions like (<reflink idref="bib6" id="ref23">6</reflink>), indicating that by age 4, the paired exhaustive interpretation has already been in place. However, there were also some curious observations. That is, 57.1% of the 2–3-year-olds provided a pair-list answer to sentences like (7a), and a subset of 4–6-year-olds also provided a pair-list answer to sentences like (7a). This pattern of responses was not provided by adults, so the puzzle is why children performed differently from adults.</p> <p>Due to the puzzle, Roeper and de Villiers ([<reflink idref="bib40" id="ref24">40</reflink>]) proceeded to examine whether children distinguish the sentences that permit the pair-list interpretation from those that do not. The test sentences were illustrated as in (8a) and (8b).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="2"&gt;&lt;p&gt;(8)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;a.&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Who did everyone pull?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;b.&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Who pulled everyone?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>The experimental hypothesis was that if the child had adult-like grammar, s/he would be expected to provide a pair-list answer to sentences like (8a), but not to sentences like (8b). It was found that 72.9% of the 3–5-year- olds provided a pair-list answer to sentences like (8a), and 69% of the children provided the same pattern of responses to sentences like (8b). Based on these findings, it has been proposed that children start out with the universal interpretation of <emph>wh</emph>-words (Nishigauchi, [<reflink idref="bib37" id="ref25">37</reflink>]).</p> <p>In contrast to the universal account, <emph>wh</emph>-words have been analysed as existential quantifiers by nature since the seminal work of Karttunen ([<reflink idref="bib28" id="ref26">28</reflink>]). Specifically, it has been proposed that <emph>wh</emph>-phrases are semantically equivalent to existentially quantified noun phrases such that <emph>wh</emph>-phrases like <emph>who</emph> and <emph>what</emph> have the same translation as <emph>someone</emph> and <emph>something</emph> (Karttunen, [<reflink idref="bib28" id="ref27">28</reflink>]: 19). We refer to this as the existential account.</p> <p>According to Hamblin ([<reflink idref="bib17" id="ref28">17</reflink>]) semantics, a <emph>wh</emph>-word introduces a set of alternatives of some logical type, and such alternatives propagate upward until they are bound by a certain operator. To illustrate, consider sentence (<reflink idref="bib9" id="ref29">9</reflink>), which is uttered in a finite domain with two individuals: Tom and Bill. The <emph>wh</emph>-word <emph>who</emph> introduces a set of alternatives, as illustrated in (<reflink idref="bib10" id="ref30">10</reflink>). These alternatives propagate until they are bound by Existential Closure, which is an operation of existential quantification that unselectively binds a set of free variables (Heim, [<reflink idref="bib18" id="ref31">18</reflink>]). In the framework of alternative semantics, Existential Closure is akin to an operation that introduces existential quantification over a set of propositional alternatives (Kratzer &amp; Shimoyama, [<reflink idref="bib29" id="ref32">29</reflink>]). Based on this analysis, sentence (<reflink idref="bib9" id="ref33">9</reflink>) can be formulated as in (<reflink idref="bib11" id="ref34">11</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;(9)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Who left?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;(10)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;{Tom left; Bill left; Both Tom and Bill left}&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;(11)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;&amp;#955;p &amp;#8707;x [person (w) (x) &amp;#8743; p = &amp;#955;w'. left (w') (x)]&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As shown in (<reflink idref="bib11" id="ref35">11</reflink>), the denotation of a <emph>wh</emph>-question contains a set of propositions (the Hamblin set), with each being a true answer to the question. A <emph>wh</emph>-word is an existentially quantified noun phrase, which is truth-conditionally equivalent to an existential item ∃. The answerhood operator applies to the Hamblin set and the evaluation world <emph>w</emph>, returning a/the complete true answers in <emph>w</emph>, where <emph>p</emph> has the same truth value in both <emph>w</emph> and <emph>w'</emph> (Xiang, [<reflink idref="bib47" id="ref36">47</reflink>], class notes). In fact, such an existential account of <emph>wh</emph>-words has been adopted by a group of semanticists, including Kratzer and Shimoyama ([<reflink idref="bib29" id="ref37">29</reflink>]), Chierchia ([<reflink idref="bib5" id="ref38">5</reflink>]), Fox ([<reflink idref="bib13" id="ref39">13</reflink>]), Huang et al. ([<reflink idref="bib24" id="ref40">24</reflink>]), among others. These semanticists tend to contend that <emph>wh</emph>-words are intrinsically existential quantifiers, which are always existentially quantified.</p> <p>In summary, there are two competing accounts of <emph>wh</emph>-words in theoretical linguistics, namely the existential account[<reflink idref="bib2" id="ref41">2</reflink>] and the universal account. The universal account predicts that children should initially interpret <emph>wh</emph>-words as universal quantifiers, while the existential account predicts that children should start out with the existential interpretation of <emph>wh</emph>-words. The thesis of the present study is to adjudicate between these two accounts by probing young children's semantic knowledge of <emph>wh</emph>-quantification.</p> <p>In Mandarin Chinese, the non-interrogative <emph>wh</emph>-words can be licensed in non-veridical contexts[<reflink idref="bib3" id="ref42">3</reflink>] such as conditionals, as in (<reflink idref="bib12" id="ref43">12</reflink>), yes–no questions, as in (<reflink idref="bib13" id="ref44">13</reflink>), and A-not-A questions, as in (<reflink idref="bib14" id="ref45">14</reflink>) (see e.g., Huang, [<reflink idref="bib21" id="ref46">21</reflink>]; Cheng, [<reflink idref="bib2" id="ref47">2</reflink>]; Li, [<reflink idref="bib31" id="ref48">31</reflink>]; Lin, [<reflink idref="bib34" id="ref49">34</reflink>], [<reflink idref="bib35" id="ref50">35</reflink>]).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(12)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;likai,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qing&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gaosu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;leave,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;please&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;tell&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="7"&gt;&lt;p&gt;'If anyone leaves, please let me know.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(13)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;mai-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;buy-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;book&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="7"&gt;&lt;p&gt;'Did John buy any book?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(14)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;mai-mei-mai&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shu?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;buy-not-buy&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;book&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="7"&gt;&lt;p&gt;'Did John eat any book or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>In the acquisition literature, it has been documented that Mandarin-speaking children are sensitive to the licensing environments of the non-interrogative <emph>wh</emph>-words (Huang et al., [<reflink idref="bib24" id="ref51">24</reflink>]; Zhou, [<reflink idref="bib50" id="ref52">50</reflink>]). Specifically, Zhou ([<reflink idref="bib50" id="ref53">50</reflink>]) investigated Mandarin-speaking pre-schoolers' interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' using a Question-Statement Task. In the task, each participant witnessed six trials, three of which contained questions while the other three included statements. The number of questions and statements was counterbalanced. The participants were evenly divided into two groups such that one group heard questions while the other heard statements on a given trial. All of the trials were ordered to make participants hear questions and statements in an alternating sequence. Participants' task was to judge the truth or falsity of the test sentence if they perceived it as a statement, but to provide an answer to the test sentence if they perceived it as a question.</p> <p>On a typical trial, three pandas went to have breakfast, which included four food options: lemons, pears, strawberries, and eggs. Each of the pandas chose an egg, but none of them had a lemon. Panda One ate a big pear, but Panda Two and Panda Three didn't like it. As the strawberries looked nice, each of the three pandas ended up having a strawberry. When the story concluded, the puppet either produced a statement as (<reflink idref="bib15" id="ref54">15</reflink>), coupled with a filler statement, or a question as (<reflink idref="bib16" id="ref55">16</reflink>), paired with a filler question. As illustrated below, (<reflink idref="bib15" id="ref56">15</reflink>) and (<reflink idref="bib16" id="ref57">16</reflink>) are a minimal pair of test sentences. To avoid participants from getting any prosodic cues, these two types of test sentences were produced in a global level intonation.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(15)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Meiyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xiongmao&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Not-have&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;panda&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="6"&gt;&lt;p&gt;'No panda ate any fruit'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(16)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Mei-zhi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xiongmao&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Every-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;panda&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="6"&gt;&lt;p&gt;'What kind of fruit did every panda eat?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>It was found that children behaved like adults in the study. That is, children rejected test sentences like (<reflink idref="bib15" id="ref58">15</reflink>) 100% of the time. For example, they consistently rejected (<reflink idref="bib15" id="ref59">15</reflink>) and justified their rejections by making reference to the fact that every panda ate a strawberry. By contrast, children responded to test sentences like (<reflink idref="bib16" id="ref60">16</reflink>) by providing an answer 100% of the time. For example, they responded to (<reflink idref="bib16" id="ref61">16</reflink>) by specifying what fruit each of the three pandas ate, i.e., Panda One ate a strawberry and a pear, Panda Two ate a strawberry, and Panda Three ate a strawberry. Taken together, the findings suggest that apart from the typical interrogative interpretation, both children and adults interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential quantifier when it appears in the scope of the negative quantificational expression <emph>meiyouren</emph> 'nobody'.</p> <p>Using the same paradigm, Huang et al. ([<reflink idref="bib24" id="ref62">24</reflink>]) evaluated preschool Mandarin-speaking children's interpretation of the <emph>wh</emph>-word <emph>shei</emph> 'who' when it occurs to the left of the quantificational adverb <emph>dou</emph> 'all'. On a typical trial, the child participants were presented with a minimal pair of test sentences (<reflink idref="bib17" id="ref63">17</reflink>) and (<reflink idref="bib18" id="ref64">18</reflink>) using a global level intonation, in scenarios where Donald Duck, Minnie Mouse, and Snoopy all ate ice-cream.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(17)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Shei&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;bingqilin?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Who&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ice-cream&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Who ate ice-cream?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(18)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Shei&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;bingqilin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Who&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ice-cream&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Everyone ate ice-cream.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As expected, the child participants had adult-like performance. More specifically, they provided an answer in response to test sentences like (<reflink idref="bib17" id="ref65">17</reflink>) 100% of the time. For example, they responded to (<reflink idref="bib17" id="ref66">17</reflink>) by listing the names of the characters that had eaten the ice-cream, i.e., Donald Duck, Minnie and Snoopy. By contrast, the child participants accepted test sentences like (<reflink idref="bib18" id="ref67">18</reflink>) 76% of the time. Considered together, the findings reveal that in addition to the typical interrogative interpretation, both children and adults interpret the <emph>wh</emph>-word <emph>shei</emph> 'who' as a universal when it occurs to the left of the quantificational adverb <emph>dou</emph> 'all.</p> <p>Following the experimental study, Zhou ([<reflink idref="bib50" id="ref68">50</reflink>]) did a survey of the Child Language Data Exchange System (CHILDES) database (Mac, [<reflink idref="bib36" id="ref69">36</reflink>]), including the Beijing 2 corpus, the Zhou 1 and Zhou 2 corpora, in order to check whether there was sufficient adult input of both the interrogative and non-interrogative uses of <emph>wh</emph>-words. The three corpora contain conversational exchanges between adults and individual children aged from 1 to 6. It was found that there were a total of 8194 utterances that contained interrogative uses of <emph>wh</emph>-words in adult input, and only 258 adult utterances in which <emph>wh</emph>-words were used non-interrogatively. The corpus study revealed an asymmetry of adult input between the interrogative and non-interrogative uses of <emph>wh</emph>-words, with the non-interrogative use being significantly less frequent than the non-interrogative use.</p> <p>Taking stock, the two studies we just reviewed reveal that Mandarin-speaking children are sensitive to the licensing environments of the non-interrogative <emph>wh</emph>-words. More specifically, Mandarin-speaking children understand that <emph>wh</emph>-words yield an existential interpretation when they appear in the scope of the negative quantificational expression <emph>meiyouren</emph> 'nobody', but generate a universal interpretation when they occur to the left of the quantificational adverb <emph>dou</emph> 'all'. Presumably, young children almost hear <emph>wh-</emph>words in the input as interrogatives, yet they appear to assign an adult-like interpretation in contexts in which it is interpreted non-interrogatively that infrequently appear in the input. This is very surprising from the perspective of learnability. In particular, the observed behaviour of children on the interpretation of <emph>wh</emph>-words invites us to wonder how the inherent semantics of <emph>wh</emph>-words are acquired, which is another motivation of the present study.</p> <p>So far, no study has been conducted to investigate Mandarin-speaking children's understanding of the semantic nature of <emph>wh</emph>-words. To fill the gap in the existing literature, the present study was designed to investigate how Mandarin-speaking children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' in non-veridical contexts where the non-interrogative interpretation is licensed. More specifically, we were interested to see whether children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' as existential or universal quantifiers. To this end, we conducted three experiments. Using a Truth Value Judgment Task (Crain &amp; Thornton, [<reflink idref="bib11" id="ref70">11</reflink>]), Experiment 1 evaluated children's interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' in the antecedent of conditional statements. Using a Question–Answer Task, Experiments 2 &amp; 3 respectively probed children's interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' in yes–no questions (Experiment 2) and A-not-A questions (Experiment 3).</p> <p>The remainder of the paper is structured as follows. Section 2 presents the investigation of children's interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' in the antecedent of conditionals. Section 3 reports the evaluation of children's interpretation of the <emph>wh</emph>-word <emph>shenme</emph> in yes–no questions. Section 4 presents the examination of children's interpretation of the <emph>wh</emph>-word <emph>shenme</emph> in A-not-A questions. Section 5 discusses the implications of the present findings, and Sect. 6 concludes.</p> <hd id="AN0177233171-3">Children's interpretation of the wh-word shenme 'what' in the antecedent of conditionals</hd> <p>As mentioned in the introduction, the <emph>wh</emph>-word <emph>shenme</emph> 'what' is licensed in the antecedent of <emph>ruguo</emph> 'if' conditionals. Interestingly, the polarity sensitive item <emph>renhe</emph> 'any' is also licensed in such a linguistic environment (see Ladusaw, [<reflink idref="bib30" id="ref71">30</reflink>]; Zwarts, [<reflink idref="bib52" id="ref72">52</reflink>]; von Fintel, [<reflink idref="bib46" id="ref73">46</reflink>]; Giannakidou, [<reflink idref="bib14" id="ref74">14</reflink>]).[<reflink idref="bib4" id="ref75">4</reflink>] For adult speakers of Mandarin, the <emph>wh</emph>-word <emph>shenme</emph> 'what' is interpreted as interchangeable with <emph>renhe</emph> 'any' in the antecedent of <emph>ruguo</emph> 'if' conditionals. To illustrate, consider sentences (<reflink idref="bib19" id="ref76">19</reflink>) and (<reflink idref="bib20" id="ref77">20</reflink>), where both expressions reside in the antecedent of <emph>ruguo</emph> 'if' conditionals.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(19)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;pinguo,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gei&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;apple,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;then&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If anyone ate an apple, then give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(20)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;pinguo,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gei&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;apple,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;then&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If anyone ate an apple, then give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(21)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;pinguo,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gei&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;apple,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;then&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If everyone ate an apple, then give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As illustrated, the <emph>wh</emph>-word <emph>shenme</emph> 'what' can be interpreted as having the meaning similar to <emph>renhe</emph> 'any'. However, these two expressions cannot be interpreted as interchangeable with the universal quantifier <emph>suoyou</emph> 'all' in (<reflink idref="bib21" id="ref78">21</reflink>), albeit the three expressions can all be licensed in the antecedent of <emph>ruguo</emph>-conditionals.</p> <p>In formal semantics, there has been a controversy on the quantificational nature of the polarity sensitive item <emph>any</emph>. On one account, <emph>any</emph> is analysed as an existential quantifier that should be interpreted within the scope of its licensor (cf. Carlson, [<reflink idref="bib1" id="ref79">1</reflink>]; Chierchia, [<reflink idref="bib5" id="ref80">5</reflink>]; Kadmon &amp; Landman, [<reflink idref="bib25" id="ref81">25</reflink>], among others). Adopting a compositional analysis, Chierchia ([<reflink idref="bib5" id="ref82">5</reflink>]) interpreted the polarity sensitive item <emph>any</emph> as having the equivalent truth-conditions with the existential expression <emph>some</emph>, which is schematized as in (<reflink idref="bib22" id="ref83">22</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="2"&gt;&lt;p&gt;(22)&lt;/p&gt;&lt;/td&gt;&lt;td align="left" colspan="2"&gt;&lt;p&gt;[|any|] = [|some|] = &amp;#955;P &amp;#955;Q &amp;#8707;&lt;sub&gt;x&lt;/sub&gt; &amp;#8712; D[P&lt;sub&gt;w&lt;/sub&gt; (x) &amp;#8743; Q&lt;sub&gt;w&lt;/sub&gt; (x)]&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" /&gt;&lt;td align="left"&gt;&lt;p&gt;(Adopted from Tieu &amp; Lidz, &lt;xref ref-type="bibr" rid="bibr44"&gt;2016&lt;/xref&gt;: 313)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As illustrated in (<reflink idref="bib22" id="ref84">22</reflink>), Chierchia ([<reflink idref="bib5" id="ref85">5</reflink>]) analysed <emph>any</emph> as truth-conditionally equivalent to an indefinite expression such as <emph>some</emph>. That is, <emph>any</emph> is an ∃-item that denotes a set of indefinite individual entities in the evaluation world <emph>w</emph>. However, <emph>any</emph> differs from a plain indefinite as it activates domain alternatives, i.e., alternative domains of quantification, which widens the quantificational domain (Tieu &amp; Lidz, [<reflink idref="bib44" id="ref86">44</reflink>]). In other words, <emph>any</emph> has a domain-widening effect, as illustrated in Note 5. On an alternative account, <emph>any</emph> is treated as a universal quantifier that obligatorily takes scope over its licensor (Dayal, [<reflink idref="bib12" id="ref87">12</reflink>]; Quine, [<reflink idref="bib39" id="ref88">39</reflink>]; Shimoyama, [<reflink idref="bib42" id="ref89">42</reflink>]).[<reflink idref="bib5" id="ref90">5</reflink>] Nevertheless, there is some empirical evidence that supports the existential account, rather than the universal account. To illustrate, consider sentence (<reflink idref="bib23" id="ref91">23</reflink>), where the polarity sensitive item <emph>renhe</emph> 'any' is licensed by the negation marker <emph>bu</emph> 'not'.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(23)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Guowang&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jueding&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;bujian&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chiyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wuqi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;de&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;nanren&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;King&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;decide&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;NEG meet&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;have&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;weapon&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;De&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;man&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="8"&gt;&lt;p&gt;'The king decided not to meet a man who has any weapons.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>Sentence (<reflink idref="bib23" id="ref92">23</reflink>) means <emph>The king decided not to meet a man who has any weapons</emph>, which supports the existential account. A further piece of evidence that supports the existential account comes from an instruction that is frequently heard on aircrafts. The instruction is illustrated in (<reflink idref="bib24" id="ref93">24</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="4"&gt;&lt;p&gt;(24)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;bu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;mingbai&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gei&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jieshi guo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;de&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe/yixie&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shiqing,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qing&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;gaosu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wo&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;NEG&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;understand&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;I&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;to&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;explain-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;De&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any/some&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;thing,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;please&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;tell&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="14"&gt;&lt;p&gt;'If you do not understand anything/something I have explained to you, please tell me.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" colspan="6"&gt;&lt;p&gt;(Adopted from Huang &amp; Crain, &lt;xref ref-type="bibr" rid="bibr19"&gt;2014&lt;/xref&gt;: 866)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As shown in (<reflink idref="bib24" id="ref94">24</reflink>), the polarity sensitive item <emph>renhe</emph> 'any' can be interpreted as interchangeable with the existential expression <emph>yixie</emph> 'some', which is not truth-conditionally equivalent to a universal quantifier, i.e., <emph>suoyou</emph> 'all'. Considered together, it is plausible to assume that the polarity sensitive item <emph>renhe</emph> 'any' is an existential quantifier. This assumption is consistent with the unified semantic analysis of the ∃-items proposed by Chierchia ([<reflink idref="bib5" id="ref95">5</reflink>]) and Fox ([<reflink idref="bib13" id="ref96">13</reflink>]). On the unified analysis, the <emph>wh</emph>-word <emph>shenme</emph> 'what' is truth-conditionally equivalent to the polarity sensitive item <emph>renhe</emph> 'any' such that the two types of expressions are existential quantifiers, the so called ∃-items of Chierchia ([<reflink idref="bib5" id="ref97">5</reflink>]).</p> <p>As discussed earlier, the <emph>wh</emph>-word <emph>shenme</emph> 'what', the polarity sensitive item <emph>renhe</emph> 'any' and the universal quantifier <emph>suoyou</emph> 'all' can all be licensed in the antecedent of <emph>ruguo</emph> 'if' conditionals. This provides a good testing ground to probe Mandarin-speaking children's semantic knowledge of the <emph>wh</emph>-word <emph>shenme</emph> 'what'. As proposed by the unified semantic analysis of ∃-items, the <emph>wh</emph>-word <emph>shenme</emph> 'what' and the polarity sensitive item <emph>renhe</emph> 'any' are existential quantifiers (Chierchia, [<reflink idref="bib5" id="ref98">5</reflink>]; Fox, [<reflink idref="bib13" id="ref99">13</reflink>]). Moreover, the expression <emph>suoyou</emph> 'all' is a universal quantifier. If Mandarin-speaking children analyse the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential quantifier, they will be expected to interpret it as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any' rather than the universal quantifier <emph>suoyou</emph> 'all' in the antecedent of <emph>ruguo</emph> 'if' conditionals. Experiment 1 puts the unified analysis of ∃-items into an empirical test.</p> <hd id="AN0177233171-4">Experiment 1</hd> <p>Experiment 1 was designed to probe whether Mandarin-speaking children's interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' was closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any' or the universal quantifier <emph>suoyou</emph> 'all'[<reflink idref="bib6" id="ref100">6</reflink>] in the antecedent of <emph>ruguo</emph> 'if' conditionals. The experiment included three types of test sentences, with each containing one of the three expressions in the antecedent of a conditional statement, as illustrated in (<reflink idref="bib25" id="ref101">25</reflink>),[<reflink idref="bib7" id="ref102">7</reflink>] (<reflink idref="bib26" id="ref103">26</reflink>) and (<reflink idref="bib27" id="ref104">27</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(25)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shoubiao,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;geiwo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;watch,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="10"&gt;&lt;p&gt;'If anyone chooses a watch, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(26)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shoubiao,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;geiwo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;watch,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="10"&gt;&lt;p&gt;'If anyone chooses a watch, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(27)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ren&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shoubiao,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;geiwo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;person&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;watch,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="10"&gt;&lt;p&gt;'If everyone chooses a watch, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>The experimental hypothesis was that if participants accessed the existential interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected to assign the same truth-conditions to sentences with <emph>shenme</emph> 'what' as sentences with <emph>renhe</emph> 'any', but assign different truth-conditions to sentences with <emph>suoyou</emph> 'all'. Alternatively, if participants accessed the universal interpretation of the <emph>wh</emph>-wor<emph>d shenme</emph> 'what', they would be expected to assign the same truth-conditions to sentences with <emph>shenme</emph> 'what' as sentences with <emph>suoyou</emph> 'all', but assign different truth-conditions to sentences with <emph>renhe</emph> 'any'.</p> <hd id="AN0177233171-5">Participants</hd> <p>We interviewed 44 Mandarin-speaking children, who were recruited from the kindergarten affiliated with Hubei University of Technology (HBUT), Wuhan, China. In addition, we adopted a between-subjects design, so the child participants were split into two groups, with 24 in Group A (5;2–5;9, M = 5;6) and 20 in Group B (5;1–5;11, M = 5;5). We also tested 20 Mandarin-speaking adults as controls, who were undergraduates at HBUT.</p> <hd id="AN0177233171-6">Procedure</hd> <p>The child participants were presented with a Truth Value Judgment Task (Crain &amp; Thornton, [<reflink idref="bib11" id="ref105">11</reflink>]). The task was conducted in a between-subjects design. Specifically, the two groups of children were presented with the same test stories, but different target sentences. Group A was presented with sentences containing <emph>renhe</emph> 'any' such as (<reflink idref="bib26" id="ref106">26</reflink>), and sentences containing <emph>suoyou</emph> 'all' such as (<reflink idref="bib27" id="ref107">27</reflink>). Group B was presented with sentences containing <emph>shenme</emph> 'what' such as (<reflink idref="bib25" id="ref108">25</reflink>), and sentences containing <emph>suoyou</emph> 'all' such as (<reflink idref="bib27" id="ref109">27</reflink>).</p> <p>Prior to the formal experiment, we conducted a pre-test to see whether the child participants in Group A understand the semantic meaning of <emph>renhe</emph> 'any' as the target sentences contained the expression. In the pre-test, the child was presented with the Truth Value Judgment Task in the description mode. Two experimenters were involved in the task. One experimenter acted out stories using toy characters and props in front of the child. The other experimenter manipulated a puppet, Kermit the frog, who watched the story together with the child. When the story concluded, Kermit described what had happened. The child's task was to judge whether Kermit's description was right or wrong. If the child indicated Kermit was wrong, then s/he was asked to explain what had really happened in the story.</p> <p>All of the child participants received the pre-test individually in a quiet room. On each trial, the child was presented with a minimal pair of test sentences: One with <emph>renhe</emph> versus one without <emph>renhe</emph>. The presence or absence of <emph>renhe</emph> results in different truth-values of the test sentences for adult speakers of Mandarin. Therefore, the child's responses to the two types of sentences were used as evidence to see whether they understood the semantic contribution of <emph>renhe</emph> (Huang &amp; Crain, [<reflink idref="bib19" id="ref110">19</reflink>]; Huang et al., [<reflink idref="bib24" id="ref111">24</reflink>]). To illustrate, here is a typical trial.</p> <p> <emph>This is a story about house-moving. One day, Small Pig</emph> <emph>, </emph> <emph>Big Pig and Mum Pig were moving to a new house. There were three pieces of furniture to be carried: A suitcase, a pot, and a table. Before carrying the furniture, Mum Pig said: "Small Pig, you are not strong enough to carry heavy items. So, you are allowed to carry the pot, but you are not allowed to carry the suitcase or the table. These two items are too heavy for you. Big pig, you look very strong. So, you can carry the suitcase and the table, but you cannot carry the pot as it's too light for you." When the two pigs started to carry the furniture, they forgot Mum Pig's proclamations. So, they asked Kermit to recall the proclamations.</emph> </p> <p>At that point, Kermit produced the filler sentence in (<reflink idref="bib28" id="ref112">28</reflink>) and the test sentence in (<reflink idref="bib29" id="ref113">29</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(28)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Xiaozhu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;keyi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ban&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-jian&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiaju&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Small pig&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;may&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;carry&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;furniture&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="6"&gt;&lt;p&gt;'The small pig is allowed to carry one piece of the furniture.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(29)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Dazhu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;keyi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ban&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-jian&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiaju&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Big pig&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;may&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;carry&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;furniture&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="6"&gt;&lt;p&gt;'The big pig is allowed to carry any piece of the furniture.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>In (<reflink idref="bib28" id="ref114">28</reflink>), the indefinite NP <emph>yi-jian jiaju</emph> 'one-CL-furniture' appears in the scope of the deontic modal verb <emph>keyi</emph> 'is allowed to'. In the absence of <emph>renhe</emph> 'any', this sentence simply means that Small Pig is allowed to carry one piece of furniture. This interpretation is consistent with Mum Pig's proclamations, so the child participants were expected to accept (<reflink idref="bib28" id="ref115">28</reflink>). By contrast, sentence (<reflink idref="bib29" id="ref116">29</reflink>) contains <emph>renhe</emph>, which appears in the scope of the deontic modal verb <emph>keyi</emph> 'is allowed'. This configuration generates a universal Free Choice Inference that Big Pig is allowed to carry every piece of furniture on offer. This interpretation is incompatible with Mum Pig's proclamations. Therefore, if the child participants understood the semantic contribution of <emph>renhe</emph> 'any', they were expected to reject (<reflink idref="bib29" id="ref117">29</reflink>).</p> <p>The main findings were as follows. Twenty out of twenty-four children accepted the filler sentences like (<reflink idref="bib28" id="ref118">28</reflink>) 100% of the time, but rejected the test sentences like (<reflink idref="bib29" id="ref119">29</reflink>) 100% of the time. On the example trial, they rejected sentence (<reflink idref="bib29" id="ref120">29</reflink>) on the grounds that the big pig was not permitted to carry the pot. The findings reveal that these children understood the semantic contribution of <emph>renhe</emph> 'any'. Therefore, they were invited to proceed to the formal experiment. Four children displayed accuracy below 50% on the test items, so they were excluded from further participation.</p> <p>In the formal experiment, the 20 child participants were presented with a Truth Value Judgment Task in the prediction mode. In this task, two experimenters were involved. One experimenter manipulated a puppet, Kermit the frog. The other experimenter introduced some characters and toys before he hid them behind the curtain of a small theatre. Following that, the experimenter invited Kermit to predict what would happen on the stage. In making its prediction, Kermit used a conditional statement, and a reward for making an accurate prediction was integrated into the consequent clause of the conditional statement. Immediately after Kermit's prediction, the contents of the stage were revealed. The child's task was to judge whether to reward Kermit or not based on the truth or falsity of the prediction. If the child refused to reward Kermit, then s/he was asked to explain what had really happened. The participants in Group B were tested using the same procedure as Group A except that they did not receive the pre-test.</p> <p>Adopting a within-subjects design, the adult participants were directly tested (without any pre-test) using a video-taped version of the test stories that the two groups of children had respectively witnessed. More specifically, they first received the test session of Group A, then followed by that of Group B. Unlike children, the adult participants were tested in groups of ten at a time. On each trial, they were asked to indicate individually whether to reward Kermit or not on a scorecard. If they refused to reward Kermit, then they were asked to provide justifications on the scorecard. Finally, both children's and adults' responses were respectively coded and analysed.</p> <hd id="AN0177233171-7">Materials</hd> <p>There were four trials in the formal experiment. On each trial, Group A was presented with two types of test sentences that respectively contained <emph>renhe</emph> 'any' and <emph>suoyou</emph> 'all' in the antecedent of <emph>ruguo</emph> 'if' conditionals. Group B was presented with another two types of test sentences that respectively included <emph>shenme</emph> 'what' and <emph>suoyou</emph> 'all'. In total, each participant in both groups heard 8 test items, including 4 target sentences and 4 filler sentences. In each group, the two types of test sentences were presented in a pseudorandomised order across trials. As expected, the positive and negative responses were counterbalanced throughout each trial. To illustrate, here is a typical trial for Group A.</p> <p> <emph>This is a game about choosing balloons. There are three animals: A penguin, a dog and a panda, and they are going to choose some balloons for decoration. There are red, yellow and green balloons for them to choose from</emph> (see Fig. 1-a)<emph>. I will put the animals and the balloons behind the curtain</emph> (see Fig. 1-b)<emph>. Kermit will predict what will happen on the stage. If his prediction is right, then you should reward him. Otherwise, you should not reward him."</emph></p> <p>Graph: Fig. 1A typical test trial of Experiment 1</p> <p>At that point, Kermit made a prediction by producing the test sentence (<reflink idref="bib30" id="ref121">30</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(30)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dongwu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;hongsede&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qiqiu,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;geiwo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;animal&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;red&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;balloon,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If any animal chooses a red balloon, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>Immediately after Kermit's prediction, the experimenter unveiled the curtain so that the child could see the outcome: Only the penguin chose a red balloon (see Fig. 1c). Following that, Kermit asked: "Do I get a coin?</p> <p>After the child responded, the experimenter put the animals and the balloons behind the curtain, and then said: "Let's play the game again." Then, Kermit made another prediction by producing the control sentence (<reflink idref="bib31" id="ref122">31</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(31)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dongwu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;hongsede&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qiqiu,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu geiwo yi-ge yingbi&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;animal&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;red&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;balloon,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If every animal chooses a red balloon, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>The experimenter unveiled the curtain immediately after Kermit's prediction such that the child could see the outcome: Only the penguin chose a red balloon. Then, the child decided whether or not to reward Kermit.</p> <p>On the typical trial, both Group A and Group B witnessed the same scenarios of the story, but Group B was presented with two different types of test sentences, as illustrated in (<reflink idref="bib32" id="ref123">32</reflink>) and (<reflink idref="bib33" id="ref124">33</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(32)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dongwu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;hongsede&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qiqiu,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;geiwo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;animal&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;red&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;balloon,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If any animal chooses a red balloon, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(33)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Ruguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dongwu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;xuanze&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;hongsede&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;qiqiu,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ni&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;jiu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;geiwo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yi-ge&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;yingbi&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;If&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;animal&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;choose&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;red&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;balloon,&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;you&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;give&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;me&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;one-CL&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;coin&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="11"&gt;&lt;p&gt;'If every animal chooses a red balloon, you give me a coin.'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <hd id="AN0177233171-8">Predictions</hd> <p>If participants accessed the existential interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected to reward Kermit upon hearing sentences with either <emph>shenme</emph> 'what' or <emph>renhe</emph> 'any', but refuse to reward Kermit upon hearing sentences with <emph>suoyou</emph> 'all'. Alternatively, if participants accessed the universal interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected not to reward Kermit upon hearing sentences with either <emph>shenme</emph> 'what' or <emph>suoyou</emph> 'all', but reward Kermit upon hearing sentences with <emph>renhe</emph> 'any'. For example, the child participants in group A were expected to reward Kermit in response to sentence (<reflink idref="bib30" id="ref125">30</reflink>), but refuse to reward Kermit in response to (<reflink idref="bib31" id="ref126">31</reflink>). In addition, the child participants in Group B were expected to reward Kermit upon hearing (<reflink idref="bib32" id="ref127">32</reflink>), but refuse to reward Kermit upon hearing (<reflink idref="bib33" id="ref128">33</reflink>). Alternatively, the child participants in group A were expected to reward Kermit in response to (<reflink idref="bib30" id="ref129">30</reflink>), but refuse to reward Kermit in response to (<reflink idref="bib31" id="ref130">31</reflink>). The child participants in Group B were expected not to reward Kermit upon hearing either (<reflink idref="bib32" id="ref131">32</reflink>) or (<reflink idref="bib33" id="ref132">33</reflink>).</p> <hd id="AN0177233171-9">Results</hd> <p>The responses from children (Group A) and adults in the two test conditions are presented in Fig. 2.</p> <p>Graph: Fig. 2The percentage of rewarding by children (Group A) and adults in response to the test items</p> <p>As shown in Fig. 2, in response to the test sentences with <emph>suoyou</emph> 'all', children refused to reward Kermit 95% of the time, and adults did so 96% of the time. In particular, justifications for their rejections were made reference to the fact that the other two characters did not perform the same action. For example, children consistently refused to reward Kermit upon hearing sentences (<reflink idref="bib31" id="ref133">31</reflink>) by making reference to the fact that the other two characters did not choose a red balloon. By contrast, children rewarded Kermit 91% of the time in response to the test sentences with <emph>renhe</emph> 'any', and the adults did so 99% of the time.</p> <p>In addition, the responses from children (Group B) and adults in the two test conditions are summarized in Fig. 3.</p> <p>Graph: Fig. 3The percentage of rewarding by children (Group B) and adults in response to the test items</p> <p>As indicated in Fig. 3, both children and adults rewarded Kermit 99% of the time in response to the test sentences with <emph>shenme</emph> 'what'. By contrast, children refused to reward Kermit 95% of the time in response to the test sentences containing <emph>suoyou</emph> 'all', and adults did so 96% of the time.[<reflink idref="bib8" id="ref134">8</reflink>]</p> <p>We fitted a general linear model on children's responses to examine the fixed effects of group and sentence type, using IBM SPSS Statistics 21. Specifically, the group, sentence type (and initially, their interaction) were taken as fixed factors, and the subject and item were taken as random factors. A comparison between children's responses to test items with <emph>renhe</emph> 'any' and ones with <emph>suoyou</emph> 'all' suggested a substantive variance (<emph>t</emph> (<reflink idref="bib1" id="ref135">1</reflink>) = 8.118, <emph>p</emph> &lt; 0.001). A comparison between children's responses to test items with <emph>shenme</emph> 'what' and ones with <emph>suoyou</emph> 'all' revealed a substantive variance (<emph>t</emph> (<reflink idref="bib1" id="ref136">1</reflink>) = 10.384, <emph>p</emph> &lt; 0.001). A comparison between children's responses to test items with <emph>renhe</emph> 'any' and ones with <emph>shenme</emph> 'what' revealed no significant difference (<emph>p</emph> = 0.253). In addition, a comparison on both children's and adults' responses to the three test items revealed that there was no significant difference (<emph>p</emph> = 1).</p> <p>In summary, both children and adults consistently rewarded Kermit upon hearing the test sentences with either <emph>shenme</emph> 'what' or <emph>renhe</emph> 'any', but refused to reward Kermit upon hearing the test sentences with <emph>suoyou</emph> 'all'. The findings suggest that both children and adults interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any', but not the universal quantifier <emph>suoyou</emph> 'all'. We interpret the findings as evidence that preschool Mandarin-speaking children have adult-like semantic knowledge of the <emph>wh</emph>-word <emph>shenme</emph> 'what'. That is, they analysed the <emph>wh-</emph>word <emph>shenme</emph> 'what' as an existential quantifier. This provides empirical support for the unified semantic analysis of the ∃-items proposed by Chierchia ([<reflink idref="bib5" id="ref137">5</reflink>]), according to which both <emph>wh</emph>-words such as <emph>shenme</emph> 'what' and polarity sensitive items such as <emph>renhe</emph> 'any' are existential quantifiers. However, the findings challenge the universal account that <emph>wh</emph>-words are universal quantifiers (Nishigauchi, [<reflink idref="bib37" id="ref138">37</reflink>]).</p> <p>Besides the antecedent of conditionals, <emph>wh</emph>-words such as <emph>shenme</emph> 'what' can also be licensed in non-veridical contexts such as yes–no questions. The next section discusses the interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' in yes–no questions with the sentence-final particle <emph>ma</emph>.</p> <hd id="AN0177233171-10">Children's interpretation of the wh-word shenme 'what' in yes–no questions</hd> <p>The <emph>wh</emph>-word <emph>shenme</emph> 'what' is also licensed in yes–no questions (see e.g., Huang, [<reflink idref="bib21" id="ref139">21</reflink>]; Cheng, [<reflink idref="bib2" id="ref140">2</reflink>]; Li, [<reflink idref="bib31" id="ref141">31</reflink>]; Lin, [<reflink idref="bib34" id="ref142">34</reflink>], [<reflink idref="bib35" id="ref143">35</reflink>]). To illustrate, consider sentence (<reflink idref="bib34" id="ref144">34</reflink>). In (<reflink idref="bib34" id="ref145">34</reflink>), the particle <emph>ma</emph> is attached at the sentence-final position, and it marks the sentence a yes–no question, querying about the truth of the proposition that John ate some fruit (Hagstrom, [<reflink idref="bib16" id="ref146">16</reflink>]; Huang et al., [<reflink idref="bib23" id="ref147">23</reflink>]). The <emph>wh</emph>-word <emph>shenme</emph> 'what' generates an existential interpretation in such a linguistic environment, with a meaning corresponding to <emph>any</emph> in English. Interestingly, the presence of the <emph>wh</emph>-word <emph>shenme</emph> 'what' does not change the nature of the yes–no question. Likewise, the polarity sensitive item <emph>renhe</emph> 'any' is also licensed in yes–no questions, as illustrated in (<reflink idref="bib35" id="ref148">35</reflink>), and it generates an existential interpretation, corresponding to <emph>any</emph> in English.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(34)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did John eat any fruit?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(35)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did John eat any fruit?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>To recap, for adult speakers of Mandarin, the <emph>wh</emph>-word <emph>shenme</emph> 'what' can be interpreted as interchangeable with the polarity sensitive item <emph>renhe</emph> 'what' in yes–no questions. However, these two expressions cannot be interpreted as interchangeable with the universal quantifier <emph>suoyou</emph> 'all', as illustrated in (<reflink idref="bib36" id="ref149">36</reflink>), even though the three expressions are all licensed in yes–no questions.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(36)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shiwu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;food&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did John eat all of the food?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>It is worth noting that sentence (<reflink idref="bib34" id="ref150">34</reflink>) is interpreted as a yes–no question in adult Mandarin regardless of the presence of the <emph>wh</emph>-word <emph>shenme</emph> 'what'. However, young children might have interpretive ambiguity between a yes–no question and a <emph>wh</emph>-question for such a sentence. This is because (<reflink idref="bib34" id="ref151">34</reflink>) contains two interrogative words, namely the yes–no question particle <emph>ma</emph> and the <emph>wh</emph>-word <emph>shenme</emph> 'what'.</p> <p>A yes–no question queries about the truth of a proposition, so its answer is supposed to inform the truth or falsity of the proposition under question (Hagstrom, [<reflink idref="bib16" id="ref152">16</reflink>]; Huang et al., [<reflink idref="bib23" id="ref153">23</reflink>]). For example, if (<reflink idref="bib34" id="ref154">34</reflink>) is interpreted as a yes–no question, then its answer will be either a positive one (e.g., <emph>Yes, he did</emph>) or a negative one (e.g., <emph>No, he didn't</emph>). By contrast, a <emph>wh</emph>-question is information-seeking, and it can be typically answered by naming an entity of the set introduced by the <emph>wh</emph>-complement (Spector, [<reflink idref="bib43" id="ref155">43</reflink>]; Xiang, [<reflink idref="bib48" id="ref156">48</reflink>]). To illustrate, reconsider (<reflink idref="bib34" id="ref157">34</reflink>). If (<reflink idref="bib34" id="ref158">34</reflink>) is interpreted as a <emph>wh</emph>-question, then it can be spelt out as (<reflink idref="bib37" id="ref159">37</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(37)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'What fruit did John eat?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>In (<reflink idref="bib37" id="ref160">37</reflink>), the <emph>wh</emph>-word <emph>shenme</emph> 'what' partitions the denotation of <emph>shuiguo</emph> 'fruit' into kinds such that it introduces a set of food kinds, i.e., apple, banana, orange and so on (Huang &amp; Crain, [<reflink idref="bib19" id="ref161">19</reflink>]). In this sense, (<reflink idref="bib37" id="ref162">37</reflink>) inquires what kind of fruit John ate. To answer this question, one needs to fill up the information gap by specifying the fruit kind that John had, i.e., <emph>An apple or John ate an apple</emph>. Therefore, it depends on one's answer to determine whether s/he interprets sentences like (<reflink idref="bib34" id="ref163">34</reflink>) as a yes–no question or a <emph>wh</emph>-question. In the present study, we incorporate this observation into the research of child language acquisition and dub a research technique called Question–Answer Task, which is a modified version of the Truth Value Judgment Task. This task will be introduced in details in the Experiment section. Adopting the task, we are interested to address the following two research questions:</p> <p></p> <ulist> <item> Do preschool Mandarin-speaking children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' interrogatively or non-interrogatively in yes–no questions?</item> <p></p> <item> If children interpret <emph>shenme</emph> 'what' non-interrogatively in yes–no questions, do they interpret it as intrinsically existential or universal?</item> </ulist> <p>For a child acquiring Mandarin, s/he needs to integrate multiple levels of linguistic knowledge to correctly interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' non-interrogatively in yes–no questions so that s/he could provide a correct answer. Precisely, the child needs to integrate syntactic, semantic, pragmatic and prosodic knowledge to correctly identify (<reflink idref="bib34" id="ref164">34</reflink>) as a yes–no question. First, the child needs to know that the non-interrogative existential <emph>wh</emph>-word <emph>shenme</emph> 'what' is licensed by the yes–no question particle <emph>ma</emph>, which typically appears at the sentence-final position. Second, the child also needs to know that a <emph>wh</emph>-question invokes an answer that specifies the entity of the set introduced by the <emph>wh</emph>-phrase (Xiang, [<reflink idref="bib48" id="ref165">48</reflink>]). Only if the child has mastered such linguistic knowledge will s/he be able to identify (<reflink idref="bib34" id="ref166">34</reflink>) as a yes–no question. Third, the child needs to understand the semantics and pragmatics of yes–no questions such that s/he is able to provide a correct answer. That is, s/he needs to know that a yes–no question inquires the truth of a proposition so that s/he can provide an answer to inform the truth or falsity of the proposition, i.e., 'Yes, he did' or 'No, he didn't'. Fourth, the child needs to know that a yes–no question with the <emph>wh</emph>-word <emph>shenme</emph> 'what' in the object position, as in (<reflink idref="bib34" id="ref167">34</reflink>), typically receives a global rising intonation, with its accent on the particle <emph>ma</emph>. In addition, the child also needs to know that a <emph>wh</emph>-question with <emph>shenme</emph> 'what' in the object position, as in (<reflink idref="bib37" id="ref168">37</reflink>), typically has a global rising intonation, with its pitch on <emph>shenme</emph> 'what' (Yuan, [<reflink idref="bib49" id="ref169">49</reflink>]). Only if the child integrates such syntactic and prosodic knowledge will s/he be able to interpret <emph>shenme</emph> 'what' correctly in yes–no questions. We expect this ability of interacting with different levels of linguistic knowledge to emerge early in the course of language development because it is very important for young children's general communication abilities, i.e., whether a child understands that a <emph>wh</emph>-word is used to raise an information-seeking question or a yes–no question. Since the <emph>wh</emph>-word <emph>shenme</emph> 'what', the polarity sensitivity item <emph>renhe</emph> 'any', and the universal quantifier <emph>suoyou</emph> 'all' are all licensed in yes–no questions, which provides another good testing ground to probe children's semantic knowledge of the <emph>wh</emph>-word <emph>shenme</emph> 'what'. Specifically, if children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' non-interrogatively in yes–no questions, we would be interested to see whether they interpret it as closer in meaning with the polarity sensitive item <emph>renhe</emph> 'any' or the universal quantifier <emph>suoyou</emph> 'all'. That's why Experiment 2 adopted yes–no questions as the test items.</p> <hd id="AN0177233171-11">Experiment 2</hd> <p>Experiment 2 examined: (<reflink idref="bib1" id="ref170">1</reflink>) whether Mandarin-speaking children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' interrogatively or non-interrogatively in yes–no questions; (<reflink idref="bib2" id="ref171">2</reflink>) whether children interpret <emph>shenme</emph> 'what' as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any' or the universal quantifier <emph>suoyou</emph> 'all' if they interpret it non-interrogatively in yes–no questions. The experiment included three types of test sentences, as illustrated in (<reflink idref="bib38" id="ref172">38</reflink>), (<reflink idref="bib39" id="ref173">39</reflink>) and (<reflink idref="bib40" id="ref174">40</reflink>), which respectively contained <emph>shenme</emph> 'what', <emph>renhe</emph> 'any' and <emph>suoyou</emph> 'all'.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(38)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;maile&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wanju&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;buy-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;toy&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did Zhangsan buy any toy?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(39)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;maile&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wanju&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;buy-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;toy&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did Zhangsan buy any toy?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(40)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;maile&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wanju&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Zhangsan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;buy&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;toy&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did Zhangsan buy all toy?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <hd id="AN0177233171-12">Participants</hd> <p>We tested another group of 22 Mandarin-speaking children, and they were aged from 4;2 to 5;6, with a mean age of 4;11. These children were also recruited from the kindergarten affiliated with Hubei University of Technology (HBUT), Wuhan, China. In addition, we tested 20 Mandarin-speaking adults as controls, who were undergraduates at HBUT.</p> <hd id="AN0177233171-13">Procedure</hd> <p>Experiment 2 adopted a Question–Answer Task. In this task, the child was introduced to a puppet called Kermit the frog, who appeared on a laptop. The experimenter acted out stories in front of the child and Kermit, using toy characters and props. The child was told that Kermit was very absent-minded, so he didn't pay close attention to the stories. At the end of each story, Kermit asked the child some questions. To avoid the prosodic cues, all of the test items were pre-recorded in a global level intonation without accent on any word[<reflink idref="bib9" id="ref175">9</reflink>] and the recordings were presented to the child via a video clip aligned with Kermit who appeared on the laptop. The child's task was to answer Kermit's questions. The answer provided by the child revealed how s/he interpreted the test sentence, i.e., whether s/he interpreted the test sentence as a <emph>wh</emph>-question or as a yes–no question.</p> <p>To avoid interference, the child participants were tested individually in a quiet room. Prior to the formal experiment, we ran a pre-test to see whether the child participants were able to distinguish between a yes–no question and a <emph>wh</emph>-question. The pre-test included two trials. On each trial, there was a minimal pair of test sentences, namely one with <emph>shenme</emph> 'what' alone, versus one with both <emph>shenme</emph> 'what' and the particle <emph>ma.</emph> To illustrate, here is a typical trial.</p> <p> <emph>This is a story about a spotted cow and a yellow cow. The spotted cow and the yellow cow were good friends. One day, they went to have dinner together. When arrived at the restaurant, the spotted cow said: "This restaurant serves very good pizza, so I'd like to order some pizza." The yellow cow said: "Hmm, I want to have a light dinner, so I will have some noodle." In the end, the spotted cow ordered pizza, and the yellow cow ordered noodle.</emph> </p> <p> <emph>Kermit said: "I wasn't paying attention just now. I only know this is a story about ordering food, in the end...</emph> </p> <p>At that point, Kermit produced the test sentences (<reflink idref="bib41" id="ref176">41</reflink>) and (<reflink idref="bib42" id="ref177">42</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(41)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Huaniu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dian-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;cai?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Spotted cow&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;order-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dish&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'What did the spotted cow order?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(42)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Huangniu&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dian-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;cai&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Yellow cow&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;order-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;dish&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did the yellow cow order any dishes?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>The presence or absence of the particle <emph>ma</emph> resulted in different interpretations of the test sentences. The child's response was then taken as indicator of whether s/he interpreted the test sentence as a yes–no question or a <emph>wh</emph>-question. For example, (<reflink idref="bib41" id="ref178">41</reflink>) only contained the <emph>wh</emph>-word <emph>shenme</emph> 'what', so it was a <emph>wh</emph>-question, asking the food name that the spotted cow ordered. Therefore, the child was expected to answer the question by naming the food, i.e., <emph>pisa</emph> 'pizza'. By contrast, (<reflink idref="bib42" id="ref179">42</reflink>) contained both the <emph>wh</emph>-word <emph>shenme</emph> 'what' and the particle <emph>ma</emph>. In this case, the non-interrogative <emph>wh</emph>-word <emph>shenme</emph> 'what' is licensed by the particle <emph>ma</emph>. Therefore, (<reflink idref="bib42" id="ref180">42</reflink>) is a yes–no question, which queries about the truth of the proposition that the yellow cow ordered some dishes. The child was expected to provide a positive answer by saying <emph>Dianle</emph> /<emph>Shide</emph> 'Yes, he did' on the grounds that the yellow cow ordered noodle.</p> <p>The main findings were twofold. First, the child participants interpreted sentences like (<reflink idref="bib41" id="ref181">41</reflink>) as <emph>wh</emph>-questions 100% of the time. For example, they either provided the food name by saying <emph>pisa</emph> 'pizza' or pointed to the pizza in response to (<reflink idref="bib41" id="ref182">41</reflink>). Second, the child participants interpreted sentences like (<reflink idref="bib42" id="ref183">42</reflink>) as yes–no questions 82% of the time. For example, they responded to (<reflink idref="bib42" id="ref184">42</reflink>) by saying <emph>dianle</emph> 'Yes, he did'. Individual analysis revealed that eight out of the 22 child participants interpreted sentences like (<reflink idref="bib42" id="ref185">42</reflink>) as <emph>wh</emph>-questions 50% of the time. For example, these children responded to (<reflink idref="bib42" id="ref186">42</reflink>) by saying <emph>miantiao/ta dianle miantiao</emph> 'Noodle/He ordered noodle'. Taken together, the findings revealed that the majority of the child participants were able to tease apart yes–no questions and <emph>wh</emph>-questions. More specifically, they interpreted sentences with both the <emph>wh</emph>-word <emph>shenme</emph> 'what' and the particle <emph>ma</emph> as yes–no questions, suggesting that they assigned a non-interrogative interpretation to the <emph>wh</emph>-word <emph>shenme</emph> 'what'. In contrast, they interpreted sentences with the <emph>wh</emph>-word <emph>shenme</emph> 'what' alone as <emph>wh</emph>-questions.</p> <p>Following the pre-test, the child participants received the main test session, which were recorded for subsequent data analysis. The adults directly received the main test session, and they were tested as a group, using a video-taped version of the stories that the children had witnessed. On each trial, the adults were asked to indicate their responses individually on a scorecard.</p> <hd id="AN0177233171-14">Materials</hd> <p>In total, there were four trials in the main test session. On each trial, there was one test story followed by three test sentences and a filler sentence. The three test sentences all included the particle <emph>ma</emph> at the sentence-final position, and they respectively contained <emph>shenme</emph> 'what', <emph>renhe</emph> 'any', and <emph>suoyou</emph> 'all' in the object position. The story was created to have the truth-condition that validated the existential interpretation, but not the universal interpretation of the test sentences. The filler sentence was designed to elicit a negative response. Therefore, the expected positive and negative responses were counterbalanced across trials. Altogether, the participants heard 16 test items, including 12 target sentences, and 4 filler sentences. To illustrate, here is a typical trial.</p> <p> <emph>This is a story about Small rabbit, Big rabbit and Mum rabbit. One day, Small rabbit and Big rabbit were a little bit hungry, so Mum rabbit took out some fruits for them to eat. There was a date, a cherry tomato, and some grapes</emph> (See Fig. 4-a). <emph>Small rabbit took a look at the fruits and said: 'Wow, the date looks good. I would like to have it. Neither the grapes nor the cherry tomato looks nice, so I don't want to have them.' After that, she ate the date. Big rabbit said: 'Well, the grapes are not ripe yet, but the cherry tomato looks OK. I would like to have the cherry tomato." Following that, he ate the cherry tomato</emph> (See Fig. 4-b).</p> <p> <emph>Kermit: "I wasn't paying attention just now. I only know this is a story about eating fruits, in the end...</emph> </p> <p>Graph: Fig. 4A typical test trial of Experiment 2</p> <p>At that point, Kermit produced three test sentences as in (<reflink idref="bib43" id="ref187">43</reflink>)–(<reflink idref="bib45" id="ref188">45</reflink>), and a filler sentence as in (<reflink idref="bib46" id="ref189">46</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(43)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Xiaotuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Small&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did the small rabbit eat any fruit?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(44)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Datuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Big rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did the big rabbit eat any fruit?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(45)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Xiaotuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Small rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did the small rabbit eat all fruit?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(46)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Datuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-le&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;putao&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;ma?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Big rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-ASP&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;grapes&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Q&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="5"&gt;&lt;p&gt;'Did the big rabbit eat the grapes?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <hd id="AN0177233171-15">Predictions</hd> <p>In the pre-test, it was established that the child participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' non-interrogatively in sentences with the particle <emph>ma</emph>. Therefore, if participants accessed the existential interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected to provide an affirmative response to the test sentences with either <emph>shenme</emph> 'what' or <emph>renhe</emph> 'any', but provide a negative response to the test sentences with <emph>suoyou</emph> 'all'. Alternatively, if participants accessed the universal interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected to provide a negative response to the test sentences with either <emph>shenme</emph> 'what' or <emph>suoyou</emph> 'all', but provide an affirmative response to the test sentences with <emph>renhe</emph> 'any'. For example, if participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential, they were expected to provide an affirmative answer to (<reflink idref="bib43" id="ref190">43</reflink>) and (<reflink idref="bib44" id="ref191">44</reflink>) by saying <emph>chile</emph> 'Yes, (he did)', but provide a negative answer to (<reflink idref="bib45" id="ref192">45</reflink>) by saying <emph>meiyou</emph> 'No, (he didn't)'. Alternatively, if participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as a universal, they were expected to provide a negative answer to (<reflink idref="bib43" id="ref193">43</reflink>) and (<reflink idref="bib44" id="ref194">44</reflink>) by saying <emph>meiyou</emph> 'No, (he didn't)', but provide an affirmative answer to (<reflink idref="bib44" id="ref195">44</reflink>) by saying <emph>chile</emph> 'Yes, (he did)'. The filler sentence (<reflink idref="bib46" id="ref196">46</reflink>) was clearly false, so the child participants were expected to provide a negative answer to it by saying <emph>meiyou</emph> 'No, (he didn't)'.</p> <hd id="AN0177233171-16">Results</hd> <p>All of the participants displayed 100% accuracy on the filler items. For example, both children and adults provided a negative response to (<reflink idref="bib46" id="ref197">46</reflink>) by saying <emph>meiyou '</emph>No, (he didn't)' on the typical trial. However, six out of the 22 child participants failed to compute the semantic meaning of <emph>renhe</emph> 'any' 100% of the time. These children responded to the test items with <emph>renhe</emph> 'any' by saying <emph>wo buzhidao renhe de yisi</emph> 'I don't know the meaning of <emph>renhe</emph>'. In addition, two of the 22 child participants responded to the test items with <emph>renhe</emph> 'any' by saying <emph>wo buzhidao</emph> 'I don't know' on half of the trials. Therefore, the data of these children were excluded from the final analysis, leaving only 14 valid child data. These data, together with those of the 20 adults, were finally analysed. A summary of these participants' responses to the test items is depicted in Fig. 5.</p> <p>Graph: Fig. 5The percentage of 'Yes' responses by both children and adults to the test items</p> <p>As shown in Fig. 5, children provided an affirmative response to the test sentences with <emph>shenme</emph> 'what' 86% of the time; adults did so 87% of the time. For example, both children and adults consistently provided an affirmative answer (<reflink idref="bib43" id="ref198">43</reflink>) by saying <emph>chile</emph> 'Yes, he did' on the typical trial. Individual analysis revealed that two out of the 14 children analysed the test sentences with <emph>shenme</emph> 'what' as <emph>wh</emph>-questions on half of the trials. For example, they responded to (<reflink idref="bib43" id="ref199">43</reflink>) by producing a short answer <emph>yingtao fanqie</emph> 'cherry tomato' or a full sentence <emph>Xiaotuzi chi-le yingtao fanqie</emph> 'The small rabbit ate the cherry tomato'. Additionally, another child interpreted the test sentences with <emph>shenme</emph> 'what' as <emph>wh</emph>-questions 100% of the time. For example, the child consistently responded to (<reflink idref="bib43" id="ref200">43</reflink>) by saying <emph>yingtao fanqie</emph> 'cherry tomato'. The findings revealed that a subset of children interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' interrogatively in yes–no questions. For adults, three out of 20 analysed the test items as <emph>wh</emph>-questions on a chance level and another four did so 25% of the time. On those trials, these adults behaved like the subset of children, i.e., they responded to (<reflink idref="bib43" id="ref201">43</reflink>) by producing a short answer <emph>yingtao fanqie</emph> 'cherry tomato' or a full sentence <emph>Xiaotuzi chi-le yingtao fanqie</emph> 'The small rabbit ate the cherry tomato'. The findings suggest that a subset of adults occasionally displayed interpretive ambiguity between the interrogative and the existential interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' in yes–no question. In response to the test sentences with <emph>renhe</emph> 'any', children provided a positive answer 100% of the time; adults did so 95% of the time. For example, both children and adults consistently provided an affirmative answer to (<reflink idref="bib44" id="ref202">44</reflink>) by saying <emph>chile</emph> 'Yes, he did' on the typical trial. By contrast, children provided a negative response to the test sentences with <emph>suoyou</emph> 'all' 95% of the time; adults did so 100% of the time. For example, both children and adults consistently provided a negative answer to (<reflink idref="bib45" id="ref203">45</reflink>) by saying <emph>meiyou</emph> 'No, (he didn't)' on the typical trial.</p> <p>As Experiment 1, we fitted the same general linear model on children's responses. A multiple comparison revealed a significant difference between children's responses to test items with <emph>shenme</emph> 'what' and ones with <emph>suoyou</emph> 'all' (<emph>t</emph> (<reflink idref="bib1" id="ref204">1</reflink>) = 8.850, <emph>p</emph> &lt; 0.001), as well as test items with <emph>renhe</emph> 'any' and <emph>suoyou</emph> 'all' (<emph>t</emph> (<reflink idref="bib1" id="ref205">1</reflink>) = 10.423, <emph>p</emph> &lt; 0.001), but no significant difference between children's responses to test items with <emph>shenme</emph> 'what' and ones with <emph>renhe</emph> 'any' (<emph>p</emph> = 0.369). In addition, a comparison between children's and adults' responses to the three types of test items revealed no significant difference between groups (<emph>p</emph> = 0.443).</p> <p>In summary, the majority of the child participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' interrogatively in sentences without the particle <emph>ma</emph>, but interpreted it non-interrogatively in sentences with the particle <emph>ma</emph> in the pre-test. The findings reveal that the child participants are able to distinguish between yes–no questions and <emph>wh</emph>-questions. Nevertheless, a subset of children interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' interrogatively in yes–no questions, suggesting that they have difficulty teasing apart yes–no questions and <emph>wh</emph>-question. In the formal experiment, the majority of the child and adult participants consistently interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any', but not the universal quantifier <emph>suoyou</emph> 'all'. As mentioned earlier, if the <emph>wh</emph>-word <emph>shenme</emph> 'what' is an indefinite variable without any inherent force, then both children and adults will not be expected to interpret it as semantically equivalent to the existential expression <emph>renhe</emph> 'any'. Therefore, the findings from Experiment 2 suggest that both Mandarin-speaking children and adults analysed the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential quantifier, but not a universal quantifier or an indefinite variable. This lends further support for the unified semantic analysis of the ∃-items proposed by Chierchia ([<reflink idref="bib5" id="ref206">5</reflink>]) and Fox ([<reflink idref="bib13" id="ref207">13</reflink>]).</p> <p>In addition to yes–no questions, the <emph>wh</emph>-word <emph>shenme</emph> 'what' can also be licensed in A-not-A questions. The next section discusses the interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what' in A-not-A questions.</p> <hd id="AN0177233171-17">Children's interpretation of the wh-word shenme 'what' in A-not-A questions</hd> <p>Besides the typical yes–no questions with the particle <emph>ma</emph>, there is another kind of question that has a similar function as yes–no question in Mandarin Chinese. This kind of question is called A-not-A question as it is formed by a duplication of a predicate with one of the duplications negated (Huang, [<reflink idref="bib22" id="ref208">22</reflink>], [<reflink idref="bib20" id="ref209">20</reflink>], [<reflink idref="bib21" id="ref210">21</reflink>]). Like typical yes–no questions, an A-not-A question is used to inquire about the truth of a proposition (Hagstrom, [<reflink idref="bib16" id="ref211">16</reflink>]). For example, consider sentence (<reflink idref="bib47" id="ref212">47</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(47)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;wufan?&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;(A-not-A)&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;lunch&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did John have lunch or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>As shown above, (<reflink idref="bib47" id="ref213">47</reflink>) contains an A-not-A construction in which the predicate verb <emph>chi</emph> 'eat' precedes and follows the negation maker <emph>mei</emph> 'not'. This configuration constitutes a question that is used to inquire about the truth of the proposition that John had lunch. Probably the most salient difference between yes–no questions and A-not-A questions is that only yes–no questions can be naturally replied with either <emph>shide</emph> 'yes', <emph>duide</emph> 'correct', or <emph>bushide</emph> 'no' (Hagstrom, [<reflink idref="bib16" id="ref214">16</reflink>]). However, A-not-A questions are typically answered by repeating the questioned predicate, i.e., <emph>xiang</emph> 'miss', or its negated form, i.e., <emph>buxiang</emph> 'not miss' (Hagstrom, [<reflink idref="bib16" id="ref215">16</reflink>]).</p> <p>It is noteworthy that both the <emph>wh</emph>-word <emph>shenme</emph> 'what' and the polarity sensitive item <emph>renhe</emph> 'any' are licensed in A-not-A questions. For example, consider sentences (<reflink idref="bib48" id="ref216">48</reflink>) and (<reflink idref="bib49" id="ref217">49</reflink>).</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(48)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did John eat any fruit or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(49)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did John eat any fruit or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(50)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Yuehan&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;John&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did John eat all fruit or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <p>For adult speakers of Mandarin, the <emph>wh</emph>-word <emph>shenme</emph> 'what' in (<reflink idref="bib48" id="ref218">48</reflink>) can be interpreted as interchangeable with the polarity sensitive item <emph>renhe</emph> 'any' in (<reflink idref="bib49" id="ref219">49</reflink>). However, both items cannot be interpreted as interchangeable with the universal quantifier <emph>suoyou</emph> 'all' in A-not-A questions like (<reflink idref="bib50" id="ref220">50</reflink>).</p> <p>This linguistic phenomenon invites us to wonder how Mandarin-speaking children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' in A-not-A questions. More specifically, we are interested to see whether children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning with the polarity sensitive item <emph>renhe</emph> 'any' or the universal quantifier <emph>suoyou</emph> 'all'. As mentioned earlier, the above-mentioned three expressions are all licensed in A-not-A questions, which provides a third good testing ground to verify the unified semantic analysis of the ∃-items proposed by Chierchia ([<reflink idref="bib5" id="ref221">5</reflink>]) and Fox ([<reflink idref="bib13" id="ref222">13</reflink>]). That's why Experiment 3 adopted A-not-A questions as the test items.</p> <hd id="AN0177233171-18">Experiment 3</hd> <p>Experiment 3 was designed to further examine whether Mandarin-speaking children interpret the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any' or the universal quantifier <emph>suoyou</emph> 'all' in A-not-A questions.</p> <hd id="AN0177233171-19">Participants</hd> <p>We tested the same 20 Mandarin-speaking children in Group A of Experiment 1, who had been confirmed that they were able to compute the semantics of <emph>renhe</emph> 'any'. In addition, we tested 20 adult speakers of Mandarin, who were undergraduates at HBUT.</p> <hd id="AN0177233171-20">Procedure</hd> <p>Both the child and adult participants were tested using the same methodology and procedure as Experiment 2.</p> <hd id="AN0177233171-21">Materials</hd> <p>There were four test trials in Experiment 3. On each trial, the participants were presented with the same test story as Experiment 2, but followed by three different test items and a filler item. The three test items were illustrated as in (<reflink idref="bib51" id="ref223">51</reflink>), (<reflink idref="bib52" id="ref224">52</reflink>) and (<reflink idref="bib53" id="ref225">53</reflink>), which were all A-not-A questions that respectively contained the <emph>wh</emph>-word <emph>shenme</emph> 'what', the polarity sensitive item <emph>renhe</emph> 'any' and the universal quantifier <emph>suoyou</emph> 'all'. In addition, a clearly false filler item, as in (<reflink idref="bib52" id="ref226">52</reflink>), was created to counterbalance the number of expected positive and negative responses throughout the trial.</p> <p></p> <p> <ephtml> &lt;table frame="hsides" rules="groups"&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(51)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Xiaotuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shenme&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Big rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;what&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did the big rabbit eat any fruit or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(52)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Datuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;renhe&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Small rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;any&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did the small rabbit eat any fruit or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(53)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Xiaotuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;suoyou&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;shuiguo?&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Small rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;all&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;fruit&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did the small rabbit eat all the fruit or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" rowspan="3"&gt;&lt;p&gt;(54)&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;Datuzi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;chi-mei-chi&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;putao?&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left"&gt;&lt;p&gt;Big rabbit&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;eat-not-eat&lt;/p&gt;&lt;/td&gt;&lt;td align="left"&gt;&lt;p&gt;grapes&lt;/p&gt;&lt;/td&gt;&lt;td align="left" /&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td align="left" colspan="4"&gt;&lt;p&gt;'Did the big rabbit eat the grapes or not?'&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt; </ephtml> </p> <hd id="AN0177233171-22">Predictions</hd> <p>If participants accessed the existential interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected to provide a positive answer to the test sentences with either <emph>shenme</emph> 'what' or <emph>renhe</emph> 'any', but a negative answer to the test sentences with <emph>suoyou</emph> 'all'. Alternatively, if participants accessed the universal interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what', they would be expected to provide a negative answer to the test sentences with either <emph>shenme</emph> 'what' or <emph>suoyou</emph> 'all', but a positive answer to the test sentences with <emph>renhe</emph> 'any'. For example, if participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential, they would be expected to provide a positive answer to (<reflink idref="bib51" id="ref227">51</reflink>) and (<reflink idref="bib52" id="ref228">52</reflink>) by saying <emph>chile</emph> 'eat-ASP', but a negative answer to (<reflink idref="bib53" id="ref229">53</reflink>) by saying <emph>meichi</emph> 'not eat'. Alternatively, if participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as a universal, they would be expected to provide a negative answer to (<reflink idref="bib51" id="ref230">51</reflink>) and (<reflink idref="bib53" id="ref231">53</reflink>) by saying <emph>meichi</emph> 'not eat', but a positive answer to (<reflink idref="bib52" id="ref232">52</reflink>) by saying <emph>chile</emph> 'eat-ASP'. The filler sentence (<reflink idref="bib54" id="ref233">54</reflink>) was clearly false, so participants were expected to reject it by saying <emph>meichi</emph> 'not eat'.</p> <hd id="AN0177233171-23">Results</hd> <p>The child and adult participants responded as expected to the filler items, with both groups scoring 100% accuracy. For example, both children and adults provided a negative answer <emph>meichi '</emph>not eat' to (<reflink idref="bib54" id="ref234">54</reflink>) on the typical trial. Therefore, the data of both groups were included into the final analysis. The results from the three test trials are presented in Fig. 6.</p> <p>Graph: Fig. 6The percentage of positive responses from children and adults to the test items</p> <p>As shown in Fig. 6, children responded to the test sentences with <emph>shenme</emph> 'what' by providing a positive answer 100% of the time;[<reflink idref="bib10" id="ref235">10</reflink>] adults did so 90% of the time. For example, both children and adults provided a positive answer to (<reflink idref="bib51" id="ref236">51</reflink>) by saying <emph>chile</emph> 'eat-ASP'. Individual analysis revealed that three of the 20 adults didn't correctly respond to the test sentences with <emph>shenme</emph> 'what' on half of the trials, and another 2 adults did so 25% of the time. Also as expected, children correctly responded to the test sentences with <emph>renhe</emph> 'any' 95% of the time; adults did so 100% of the time. For example, both children and adults consistently provided a positive answer to (<reflink idref="bib52" id="ref237">52</reflink>) by saying <emph>chile</emph> 'eat-ASP'. Individual analysis revealed that two of the 20 children provided a negative answer to the test sentences with <emph>renhe</emph> 'any' 25% of the time, and another child provided a negative answer to the test sentences on half of the trials. By contrast, both children and adults provided a negative answer to the test sentences with <emph>suoyou</emph> 'all' 100% of the time. For example, both groups consistently provided a negative answer to (<reflink idref="bib53" id="ref238">53</reflink>) by saying <emph>meichi</emph> 'not eat'. We fitted a logistic regression model on the children's and adults' responses to examine the effect of target sentences, groups and their interactions.</p> <p>We did the same statistical analysis as Experiment 2. A multiple comparison revealed a significant difference between children's responses to test sentences with <emph>shenme</emph> 'what' and ones with <emph>suoyou</emph> 'all' (<emph>t</emph> (<reflink idref="bib1" id="ref239">1</reflink>) = 25.131, <emph>p</emph> &lt; 0.001), as well as test items with <emph>renhe</emph> 'any' and ones with <emph>suoyou</emph> 'all' (<emph>t</emph> (<reflink idref="bib1" id="ref240">1</reflink>) = 23.875, <emph>p</emph> &lt; 0.001), but no significant difference between children's responses to test items with <emph>shenme</emph> 'what' and ones with <emph>renhe</emph> 'any' (<emph>p</emph> = 0.641). A comparison between children's and adults' responses to the three test items revealed no significant difference between groups (<emph>p</emph> = 0.549).</p> <p>Taken stock, both Mandarin-speaking children and adults consistently interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' non-interrogatively in A-not-A questions. In addition, both groups interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any', but not the universal quantifier <emph>suoyou</emph> 'all'. The findings suggest that Mandarin-speaking children analyse the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential quantifier rather than a universal quantifier. Once again, the findings lend further support for the unified semantic analysis of the ∃-items (Chierchia, [<reflink idref="bib5" id="ref241">5</reflink>]; Fox, [<reflink idref="bib13" id="ref242">13</reflink>]).</p> <hd id="AN0177233171-24">General Discussion</hd> <p>The main findings of the present study were threefold. First, both children and adults interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning to the polarity sensitive item <emph>renhe</emph> 'any', but not the universal quantifier <emph>suoyou</emph> 'all' in the antecedent of <emph>ruguo</emph> 'if' conditionals. Second, both children and adults interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' non-interrogatively, with a meaning closer to <emph>renhe</emph> 'any', but not <emph>suoyou</emph> 'all' in yes–no questions with the particle <emph>ma</emph>. By contrast, the child participants interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' interrogatively in the absence of the particle <emph>ma</emph>. Third, both children and adults interpreted the <emph>wh</emph>-word <emph>shenme</emph> 'what' as closer in meaning to <emph>renhe</emph> 'any', but not <emph>suoyou</emph> 'all' in A-not-A questions. Taken together, the findings suggest that by age 5, Mandarin-speaking children already have adult-like linguistic knowledge that the <emph>wh</emph>-word <emph>shenme</emph> 'what' is an existential quantifier, which is truth-conditionally equivalent to the polarity sensitive item <emph>renhe</emph> 'any'. Moreover, the findings reveal that Mandarin-speaking pre-schoolers are able to distinguish between <emph>wh</emph>-questions and yes–no questions. We interpret the findings as empirical evidence to support the unified semantic account that both <emph>wh</emph>-words such as <emph>shenme</emph> 'what' and polarity sensitive items such as <emph>renhe</emph> 'any' are uniformly analysed as ∃-items (Chierchia, [<reflink idref="bib5" id="ref243">5</reflink>]; Fox, [<reflink idref="bib13" id="ref244">13</reflink>]). However, the findings appear to challenge the universal account that <emph>wh</emph>-words such as <emph>shenme</emph> 'what' are inherent universal quantifiers (Nishigauchi, [<reflink idref="bib37" id="ref245">37</reflink>]). In a nutshell, the present findings help to constrain the theoretical debate on the semantic nature of <emph>wh</emph>-quantification.</p> <p>Our experimental findings suggest that by age of 5, Mandarin-speaking children have acquired the adult-like non-interrogative meaning of <emph>wh</emph>-words. Presumably, young children almost hear Mandarin <emph>wh</emph>-words as interrogatives, and infrequently receive the input of the non-interrogative uses. We believe that this conjecture can be confirmed by the Zhou ([<reflink idref="bib50" id="ref246">50</reflink>]) corpus study, which revealed that preschool Mandarin-speaking children hardly receive the input of the non-interrogative uses of <emph>wh</emph>-words compared with that of the interrogative ones. However, an anonymous reviewer pointed out that the corpus study by Zhou ([<reflink idref="bib50" id="ref247">50</reflink>]) only searched the Beijing 2 corpus, the Zhou 1 and Zhou 2 corpora, whereas the latter two corpora lack of longitudinal data that contain rich adult input such that it is unfair to assert that that young children receive insufficient input of the non-interrogative interpretation of <emph>wh</emph>-words. In response to this concern, we did a follow-up corpus study, which included the Erbaugh corpus and the Zhou 3 Corpus. These two corpora contain longitudinal data with conversational exchanges between adults and children (age range: 2;00–4;11). It was found that there were a total of 907 utterances that contained the <emph>wh</emph>-word <emph>shenme</emph> 'what' in adult input, and only 40 adult utterances in which the <emph>wh</emph>-word was used non-interrogatively. Combined with the findings of the Zhou ([<reflink idref="bib50" id="ref248">50</reflink>]) corpus study, it is plausible to assume that young children rarely receive any input of the non-interrogative interpretation of the <emph>wh</emph>-word <emph>shenme</emph> 'what'. In view of the paucity of primary linguistic input, it seems very surprising that young children have acquired the non-interrogative meaning of the <emph>wh</emph>-word, i.e., the existential meaning in the experimental contexts of the present study. This actually leads to a learnability question regarding the nature of language acquisition.</p> <p>In the acquisition literature, there are two main approaches that have been proposed to account for the nature of child language acquisition. On one approach, it has been claimed that children are biologically equipped with Universal Grammar, which is a system of principles and parameters for all human languages. Principles contain the deep-seated regularities that all human languages share, whereas parameters encode information on language variations. Children acquire linguistic knowledge using domain specific learning mechanisms rather than domain general learning mechanisms such as the analysis of analogy and distribution (Chomsky, [<reflink idref="bib6" id="ref249">6</reflink>]; Crain &amp; Khlentzos, [<reflink idref="bib7" id="ref250">7</reflink>]; Crain &amp; Pietroski, [<reflink idref="bib9" id="ref251">9</reflink>], [<reflink idref="bib10" id="ref252">10</reflink>]; Crain et al., [<reflink idref="bib8" id="ref253">8</reflink>]; Pinker, [<reflink idref="bib38" id="ref254">38</reflink>]). This is called the biolinguistics approach to language acquisition. On an alternative approach, it has been proposed that young children depend on adult input to acquire linguistic knowledge, i.e., they acquire linguistic knowledge using domain-general learning mechanisms such as the analysis of analogy and distribution (Lieven &amp; Tomasello, [<reflink idref="bib33" id="ref255">33</reflink>]; Saxton, [<reflink idref="bib41" id="ref256">41</reflink>]; Tomasello, [<reflink idref="bib45" id="ref257">45</reflink>]). This approach is referred to as the usage-based approach to language acquisition.</p> <p>Following the biolinguistic approach, we propose a learning scenario of the non-interrogative meaning of <emph>wh</emph>-words. In line with the seminal work of Chierchia ([<reflink idref="bib5" id="ref258">5</reflink>]) and Fox ([<reflink idref="bib13" id="ref259">13</reflink>]), we assume that <emph>wh</emph>-words are always existentially quantified, which needs a certain operator for licensing. For a child acquiring Mandarin, s/he is initially endowed with the linguistic knowledge that a class of words belong to ∃-items, i.e., one candidate of the ∃-items is the <emph>wh</emph>-word <emph>shenme</emph> 'what'. As the child frequently receives input of the interrogative meaning, this triggering experience subsequently helps her/him to map the logical concept of ∃-items onto the <emph>wh</emph>-word <emph>shenme</emph> 'what'. Suppose that the child also understands that the existential non-interrogative meaning must be licensed by a non-veridical operator, just as the existential expression <emph>any</emph> needs to be licensed (Huang &amp; Crain, [<reflink idref="bib19" id="ref260">19</reflink>]). In fact, previous studies have shown that children are sensitive to the licensing environments of the non-interrogative <emph>wh</emph>-words (Zhou &amp; Crain, [<reflink idref="bib51" id="ref261">51</reflink>]; Huang et al., [<reflink idref="bib24" id="ref262">24</reflink>]). Therefore, once the child learns that there is a set of nonveridical operators that license an existential non-interrogative interpretation of <emph>wh</emph>-words, s/he is able to draw such an interpretation in the corresponding licensing environments. For example, once the child learns that the nonveridical operator <emph>ruguo</emph> 'if' licenses an existential interpretation of <emph>wh</emph>-words, s/he is able to draw such an interpretation from a <emph>wh</emph>-word that appears in <emph>ruguo</emph> 'if' conditionals. Similarly, the child is able to draw an existential interpretation from a <emph>wh</emph>-word that occurs in either yes–no questions with the particle <emph>ma</emph> or A-not-A questions once knowing the particle <emph>ma</emph> and A-not-A operators license the existential non-interrogative meaning. However, the corpus study by Zhou ([<reflink idref="bib50" id="ref263">50</reflink>]) has shown that children have hardly ever received adult input of the non-interrogative use of <emph>wh</emph>-words in Mandarin Chinese. Due to the paucity of primary linguistic input, it is hard-pressed for the advocates of the usage-based account to claim that young children acquire the existential non-interrogative meaning of <emph>wh</emph>-words merely via input.</p> <p>Moreover, it remains unknown how young children establish adult-like semantic mapping between the <emph>wh</emph>-word <emph>shenme</emph> 'what' and the polarity sensitive item <emph>renhe</emph> 'any'. The proponents of the usage-based account might claim that young children depend on relevant input to learn that both <emph>shenme</emph> 'what' and <emph>renhe</emph> 'any' are truth-conditionally equivalent to each other. For example, if the target linguistic structures respectively containing <emph>shenme</emph> 'what' and <emph>renhe</emph> 'any' are highly frequent in the interactions between young children and adults, then children might take the advantage of the input to learn that both expressions are logically equivalent ∃-items. To evaluate this possibility, we surveyed the seven Mandarin corpora of the CHILDES database (the Beijing corpus, the Chang corpus, the Context corpus, the Tong corpus, the XuMinChen Corpus, the Zhou 1 and Zhou 2 corpora), which include 480,727 tokens of conversational utterances between adults and individual children aged from 1 to 6. No adult utterances of <emph>renhe</emph> were found in these corpora,[<reflink idref="bib11" id="ref264">11</reflink>] not to mention the exact combinations such as <emph>ruguo</emph> + <emph>renhe, renhe</emph> + <emph>ma, and A-not-A</emph> + <emph>renhe</emph>. As mentioned earlier, the two corpus studies found that there were only 298 adult utterances containing the non-interrogative use of <emph>shenme</emph> 'what' in the five Mandarin corpora, including the Beijing 2 corpus, the Erbaugh corpus, the Zhou1, Zhou 2 and Zhou 3 corpora). Due to the asymmetry of the primary input of <emph>shenme</emph> 'what' and <emph>renhe</emph> 'any', it is unlikely that young children depend on the relevant input to learn that both expressions are logically equivalent ∃-items. In particular, our experimental findings reveal that the emergence of the meanings of both expressions synchronizes in the course of language development. In addition, it has been found that preschool Mandarin-speaking children understand the existential meaning of the polarity sensitive item <emph>renhe</emph> 'any' regardless of the scarcity of adult input (Huang &amp; Crain, [<reflink idref="bib19" id="ref265">19</reflink>]). Taken together, the findings pose a challenge to the usage-based approach to language acquisition (Lieven &amp; Tomasello, [<reflink idref="bib33" id="ref266">33</reflink>]; Tomasello, [<reflink idref="bib45" id="ref267">45</reflink>]). In conjunction with the previous findings by Huang and Crain ([<reflink idref="bib19" id="ref268">19</reflink>]), the present experimental results appear to support the biolinguistic approach to language acquisition that claims young children depend on an innate logic faculty to acquire logic expressions (Chomsky, [<reflink idref="bib6" id="ref269">6</reflink>]; Crain &amp; Khlentzos, [<reflink idref="bib7" id="ref270">7</reflink>]; Crain &amp; Pietroski, [<reflink idref="bib9" id="ref271">9</reflink>], [<reflink idref="bib10" id="ref272">10</reflink>]; Crain et al., [<reflink idref="bib8" id="ref273">8</reflink>]; Pinker, [<reflink idref="bib38" id="ref274">38</reflink>]).</p> <hd id="AN0177233171-25">Conclusion</hd> <p>This paper reports experimental findings showing that 4-and 5-year-old Mandarin-speaking children analysed the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential quantifier rather than a universal quantifier. The findings provide empirical evidence to support the unified semantic analysis of the ∃-items of Chierchia ([<reflink idref="bib5" id="ref275">5</reflink>]) and Fox ([<reflink idref="bib13" id="ref276">13</reflink>]). In addition, the corpus study reveals an asymmetry of the primary linguistic input of <emph>shenme</emph> 'what' and <emph>renhe</emph> 'any' in the interactions between children and caregivers. Therefore, it is unlikely that young children merely rely on adult input to learn that both items are logically equivalent ∃-items. Considered together, the findings pose a challenge to the usage-based approach to language acquisition (Lieven &amp; Tomasello, [<reflink idref="bib33" id="ref277">33</reflink>]; Saxton, [<reflink idref="bib41" id="ref278">41</reflink>]; Tomasello, [<reflink idref="bib45" id="ref279">45</reflink>]), but appear to support the biolinguistic approach to acquisition of logical expressions (Chomsky, [<reflink idref="bib6" id="ref280">6</reflink>]; Crain &amp; Khlentzos, [<reflink idref="bib7" id="ref281">7</reflink>]; Crain &amp; Pietroski, [<reflink idref="bib9" id="ref282">9</reflink>], [<reflink idref="bib10" id="ref283">10</reflink>]; Crain et al., [<reflink idref="bib8" id="ref284">8</reflink>]; Pinker, [<reflink idref="bib38" id="ref285">38</reflink>]). We interpret the contribution of the present study as twofold. First, the present study provides experimental evidence showing that young children initially analyse the <emph>wh</emph>-word <emph>shenme</emph> 'what' as an existential quantifier, not a universal quantifier. Second, the present study exemplifies how experimental studies on child language acquisition can resolve theoretical debates.</p> <hd id="AN0177233171-26">Acknowledgements</hd> <p>This research was funded by National Social Science Foundation of China (Grant Number: 21BYY108), and we are grateful to the funder. For valuable feedback and discussion, we would like to thank Stephen Crain, Rosalind Thornton, Lyn Tieu, Corry Bill, and the audiences at the 44th Annual Boston University Conference on Language Development. In addition, we are also thankful to the kindergarten affiliated with Hubei University of Technology, Wuhan, China.</p> <hd id="AN0177233171-27">Declarations</hd> <p></p> <hd id="AN0177233171-28">Conflict of interest</hd> <p>Haiquan Huang has received research grant from National Social Science Foundation of China (Grant Number: 21BYY108), and has completed the present experimental study with the help of Peng Zhou. Hui Cheng, Lina Qian and Yixiong Chen participated in the data collection and data analysis. All authors declare that they have no conflict of interest with one another.</p> <hd id="AN0177233171-29">Ethical Standards</hd> <p>This study was funded by National Social Science Foundation of China (Grant Number: 21BYY108). All procedures performed in the present study involving human participants were in accordance with the ethical standards of the institutional and /or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.</p> <hd id="AN0177233171-30">Publisher's Note</hd> <p>Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.</p> <ref id="AN0177233171-31"> <title> References </title> <blist> <bibl id="bib1" idref="ref7" type="bt">1</bibl> <bibtext> Carlson GN. Polarity any is existential. Linguistic Inquiry. 1980; 4: 799-804</bibtext> </blist> <blist> <bibl id="bib2" idref="ref1" type="bt">2</bibl> <bibtext> Cheng, L. L.-S. (1991). On the typology of wh-questions. Unpublished doctoral dissertation, MIT.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref2" type="bt">3</bibl> <bibtext> Cheng LL-S. Wh-words as polarity items. Chinese Languages and Linguistics. 1994; 2: 615-640</bibtext> </blist> <blist> <bibl id="bib4" idref="ref14" type="bt">4</bibl> <bibtext> Chierchia G. Broaden your views: Implicatures of domain widening and the "Logicality" of language. Linguistic Inquiry. 2006; 37: 535-590. 10.1162/ling.2006.37.4.535</bibtext> </blist> <blist> <bibl id="bib5" idref="ref15" type="bt">5</bibl> <bibtext> Chierchia G. Logic in grammar: Polarity, free choice, and intervention. 2013; Oxford University Press. 10.1093/acprof:oso/9780199697977.001.0001</bibtext> </blist> <blist> <bibl id="bib6" idref="ref19" type="bt">6</bibl> <bibtext> Chomsky N. Aspects of the theory of syntax. 1965; MIT Press</bibtext> </blist> <blist> <bibl id="bib7" idref="ref20" type="bt">7</bibl> <bibtext> Crain S, Khlentzos D. Is logic innate?. Biolinguistics. 2008; 2: 24-56. 10.5964/bioling.8619</bibtext> </blist> <blist> <bibl id="bib8" idref="ref134" type="bt">8</bibl> <bibtext> Crain S, Koring L, Thornton R. Language acquisition from a biolinguistic perspective. Neuroscience &amp; Biobehavioral Review. 2016. 10.1016/j.neubiorev.2016.09.004</bibtext> </blist> <blist> <bibl id="bib9" idref="ref29" type="bt">9</bibl> <bibtext> Crain S, Pietroski P. Nature, nurture and universal grammar. Linguistics and Philosophy. 2001; 24; 2: 139-185. 10.1023/A:1005694100138</bibtext> </blist> <blist> <bibtext> Crain S, Pietroski P. Why language acquisition is a snap. The Linguistic Review. 2002; 19; 1-2: 163-183</bibtext> </blist> <blist> <bibtext> Crain S, Thornton R. Investigations in universal grammar: A guide to experiments on the acquisition of syntax and semantics. 1998; MIT Press</bibtext> </blist> <blist> <bibtext> Dayal V. Any as inherently modal. Linguistics and Philosophy. 1998; 21: 433-476. 10.1023/A:1005494000753</bibtext> </blist> <blist> <bibtext> Fox D Sauerland U, Stateva P. Free choice and the theory of scalar implicatures. Presupposition and implicature in compositional semantics (Palgrave studies in pragmatics, language and cognition). 2007; Palgrave MacMillan: 71-120. 10.1057/9780230210752_4</bibtext> </blist> <blist> <bibtext> Giannakidou A. Polarity sensitivity as (non) veridical dependency. 1998; John Benjamin Publishing Company. 10.1075/la.23</bibtext> </blist> <blist> <bibtext> Giannakidou, A. (2002). Licensing and sensitivity in polarity items: From downward entailment to nonveridicality. In The Proceedings of 39th Chicago Linguistics Society.</bibtext> </blist> <blist> <bibtext> Hagstrom P Everaert M, van Riemsdijk H. A-not-A questions. The Blackwell companion to syntax. 2006; Blackwell: 173-213. 10.1002/9780470996591.ch7</bibtext> </blist> <blist> <bibtext> Hamblin CL. Questions in Montague English. Foundations of Language. 1973; 10: 41-53</bibtext> </blist> <blist> <bibtext> Heim, I. (1982). The semantics of definite and indefinite noun phrases. Doctoral Dissertation: University of Massachusetts, Amherst.</bibtext> </blist> <blist> <bibtext> Huang A, Crain S. Acquisition of the polarity sensitive item renhe 'any' in Mandarin Chinese. Journal of Child Language. 2014; 2014; 4: 861-889. 10.1017/S0305000913000275</bibtext> </blist> <blist> <bibtext> Huang C-TJ. Move wh in a language without wh-movement. Linguistic Review. 1982; 1: 369-416. 10.1515/tlir.1982.1.4.369</bibtext> </blist> <blist> <bibtext> Huang, C.-T. J. (1982b). Logical relations in Chinese and theory of grammar. Doctoral Dissertation: MIT.</bibtext> </blist> <blist> <bibtext> Huang C-TJ Carol G, Roberta I. Modularity and Chinese A-not-A questions. Interdisciplinary approaches to language. 1991; Kluwer: 305-322. 10.1007/978-94-011-3818-5_16</bibtext> </blist> <blist> <bibtext> Huang C-TJ, Li Y-HA, Li Y. The syntax of Chinese. 2009; Cambridge University Press. 10.1017/CBO9781139166935</bibtext> </blist> <blist> <bibtext> Huang H, Zhou P, Crain S. Wh-questions, universal statements and free choice inferences in child Mandarin. Journal of Psycholinguistic Research. 2017; 47: 1391-1409. 10.1007/s10936-017-9535-6</bibtext> </blist> <blist> <bibtext> Kadmon N, Landman F. Any. Linguistic and Philosophy. 1993; 16: 353-422. 10.1007/BF00985272</bibtext> </blist> <blist> <bibtext> Kamp H. Free choice permission. Proceedings of Aristotelian Society. 1973; 74: 57-74. 10.1093/aristotelian/74.1.57</bibtext> </blist> <blist> <bibtext> Kamp H. Semantics versus pragmatics. 1978; Springer</bibtext> </blist> <blist> <bibtext> Karttunen L. Syntax and semantics of questions. Linguistics and Philosophy. 1977; 1: 3-44. 10.1007/BF00351935</bibtext> </blist> <blist> <bibtext> Kratzer, A., &amp; Shimoyama, J. (2002). Indeterminate pronouns: The view from Japanese. In O. Yukio (Eds.), Proceedings of the third Tokyo conference on psycholinguistics (pp. 1–25). Hituzi Syobo.</bibtext> </blist> <blist> <bibtext> Ladusaw WA. Polarity sensitivity as inherent scope relations. 1980; Garland Publishing</bibtext> </blist> <blist> <bibtext> Li AY-H. Indefinite wh in Mandarin Chinese. Journal of East Asian Linguistics. 1992; 1: 125-155. 10.1007/BF00130234</bibtext> </blist> <blist> <bibtext> Liao, C.-H. (2010). Alternative and exhaustification: Non-interrogative uses of Chinese wh-words. Doctoral dissertation: Harvard University.</bibtext> </blist> <blist> <bibtext> Lieven E, Tomasello M Robinson P, Ellis N. Children's first language acquisition from a usage-based perspective. Handbook of cognitive linguistics and second language acquisition. 2008; Routledge: 168-196</bibtext> </blist> <blist> <bibtext> Lin, J.-W. (1996). Polarity licensing and wh-phrases in Chinese. Unpublished doctoral dissertation: University of Massachusetts, Amherst.</bibtext> </blist> <blist> <bibtext> Lin J-W. On existential polarity wh-phrases in Chinese. Journal of East Asian Linguistics. 1998; 7: 219-255. 10.1023/A:1008284513325</bibtext> </blist> <blist> <bibtext> Mac WB. The CHILDES project: The database. 2000; Psychology Press</bibtext> </blist> <blist> <bibtext> Nishigauchi, T. (1999). Some preliminary thoughts on the acquisition of the syntax and semantics of wh-constructions. In Theoretical and applied linguistics at Kobe Shoin 2. Kobe Shoin Institute for Linguistic Sciences.</bibtext> </blist> <blist> <bibtext> Pinker S. Language learnability and language development. 1984; Harvard University Press</bibtext> </blist> <blist> <bibtext> Quine WV-O. Word and object. 1960; MIT Press</bibtext> </blist> <blist> <bibtext> Roeper, T., &amp; de Villiers, J. (1991). The emergence of bound variable structures. In T. Maxfield, &amp; B. Plunkett (Eds.), Proceedings of the University of Massachusetts Roundtable (UMOP) (pp. 225–266). Amherst MA.</bibtext> </blist> <blist> <bibtext> Saxton M. Child Language: Acquisition and Development. 2010; Stage Publications</bibtext> </blist> <blist> <bibtext> Shimoyama J. Japanese indeterminate negative polarity items and their scope. Journal of Semantics. 2011; 28: 413-450. 10.1093/jos/ffr004</bibtext> </blist> <blist> <bibtext> Spector, B. (2007). Modalized questions and exhaustivity. In M. Gibson, &amp; T. Friedman (Eds.), Proceedings of Semantics and Linguistic Theory17 (pp. 282–299). CLC Publications. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/view/2962</bibtext> </blist> <blist> <bibtext> Tieu L, Lidz J. NPI licensing and beyond: Children's knowledge of the semantics of any. Language Acquisition. 2016; 4: 311-332. 10.1080/10489223.2016.1176172</bibtext> </blist> <blist> <bibtext> Tomasello M. Constructing a language: A usage-based theory of language acquisition. 2003; Harvard University Press</bibtext> </blist> <blist> <bibtext> von Fintel K. NPI licensing, Strawson entailment, and context dependency. Journal of Semantics. 1999; 16: 97-148. 10.1093/jos/16.2.97</bibtext> </blist> <blist> <bibtext> Xiang, Y. (2019). Notes for semantics of questions. Rutgers University class notes for Semantics. https://yimeixiang.files.wordpress.com/2019/01/ho2-hamblin-karttunen.pdf</bibtext> </blist> <blist> <bibtext> Xiang Y. Higher-order readings of wh-questions. Natural Language Semantics. 2021; 29: 1-45. 10.1007/s11050-020-09166-8</bibtext> </blist> <blist> <bibtext> Yuan, J. (2006). Mechanisms of question intonation in Mandarin. In Q. Huo, B. Ma, E. S. Chng, &amp; H. Z. Li (Eds.), Proceedings of the fifth symposium on Chinese spoken language processing (pp. 19–30). Springer.</bibtext> </blist> <blist> <bibtext> Zhou P. Children's knowledge of wh-quantification in Mandarin Chinese. Applied Psycholinguistics. 2013; 36: 411-435. 10.1017/S0142716413000283</bibtext> </blist> <blist> <bibtext> Zhou P, Crain S. Children's knowledge of the quantifier dou in Mandarin Chinese. Journal of Psycholinguistic Research. 2011; 40: 155-176. 10.1007/s10936-010-9161-z. 21061067</bibtext> </blist> <blist> <bibtext> Zwarts F. Nonveritical contexts. Linguistic Inquiry. 1995; 25: 286-312</bibtext> </blist> </ref> <ref id="AN0177233171-32"> <title> Footnotes </title> <blist> <bibtext> Note that <emph>wh</emph>-words in Mandarin Chinese stay in situ, which is probably the most salient typological feature compared with other languages. For example, English <emph>wh</emph>-questions are typically formed via overt <emph>wh</emph>-movement: moving a <emph>wh</emph>-constituent to the sentence initial position. By contrast, <emph>wh</emph>-questions in Mandarin Chinese are formed by leaving a <emph>wh-</emph>constituent in situ. Nevertheless, it has been claimed that <emph>wh</emph>-questions in Mandarin Chinese are subject to covert <emph>wh</emph>-movement (Huang, [20]). Therefore, Mandarin Chinese has been generally assumed as a <emph>wh-in-situ</emph> language (see Huang, [20], [21]; Huang, et al., [23]).</bibtext> </blist> <blist> <bibtext> We are grateful to an anonymous reviewer's request for the acquisition literature that supports the existential account as the universal account is based on acquisition evidence in this paper. However, no direct acquisition evidence has been documented for the existential account even though it has been theoretically claimed by a group of semanticists (Kratzer and Shimoyama, [29]; Chierchia, [5]; Fox, [13]; Huang et al., [24], among others). Actually, the thesis of the present study is to probe the acquisition evidence for the existential account of <emph>wh</emph>-words.</bibtext> </blist> <blist> <bibtext> Note that the distinction between veridicality and nonveridicality is established in the light of truth, as schematized in (i) by Giannakidou ([15]: 5). "A propositional operator F is veridical iff Fp entails p : Fp → p; otherwise, F is nonveridical". To illustrate, consider sentences (ii) and (ii). If the truth of the proposition (ii) is not entailed in the context of (iii), then the context of (iii) is typically considered as nonveridical. John speaks English. I hope that John speaks English. Note also that there has been a continuing controversy on the distribution of the non-interrogative <emph>wh</emph>-words (see Huang, [21], Cheng, [2], Li, [31], Lin, [34], [35], and among others). The precise distribution is beyond the scope of this paper, and interested readers can refer to Li ([31]) and Lin ([35]) for a detailed discussion.</bibtext> </blist> <blist> <bibtext> Polarity sensitive items such as <emph>renhe</emph> 'any' in Mandarin Chinese can also be licensed in other downward entailing contexts, i.e., negative sentences (Tieu &amp; Lidz, [44]). In particular, <emph>renhe</emph> 'any' is not licensed in affirmative sentences (Huang &amp; Crain, [19]), so it is also referred to as a Negative Polarity Item (NPI). To illustrate, consider (i) and (ii). (i) Zhangsan mei chi renhe shiwu Zhangsan Neg eat any food 'Zhangsan didn't eat any food." (ii) *Zhangsan chile renhe shiwu Zhangsan eat-ASP any food 'Zhangsan ate any food.' As illustrated in (i), <emph>renhe</emph> is licensed in the scope of the negation marker <emph>mei</emph> 'not', and it generates a domain widening effect. That is, the sentence entails that Zhangsan didn't any food, including food that is atypical in common sense (Chierchia, [4]; Tieu &amp; Lidz, [44]).</bibtext> </blist> <blist> <bibtext> Note that polarity sensitive items such as <emph>renhe</emph> 'any' can generate a universal interpretation in certain linguistic contexts. For example, the polarity sensitive item <emph>renhe</emph> 'any' licenses a universal Free Choice Inference when it appears in the scope of a deontic modal verb (Chierchia, [5]; Kamp, [26], [27]). For example, consider sentence (i). (i) Zhangsan keyi kan woshufangli de renhe yiben shu Zhangsan may read my study inside DE any one-CL book 'Zhangsan is allowed to read any book in my study.' As illustrated in (i), <emph>renhe</emph> 'any' is licensed in the scope of the deontic modal verb <emph>keyi</emph> 'is allowed to', and the sentence gives rise to a universal Free Choice Inference that Zhangsan is allowed to eat any book in my study.</bibtext> </blist> <blist> <bibtext> Note that both <emph>suoyou</emph> 'all' and <emph>mei</emph> 'every' are universal quantifiers in Mandarin Chinese. However, they differ from each other in terms of distribution. That is, <emph>suoyou</emph> 'all' can appear in either subject or object position. By contrast, <emph>mei</emph> 'every' can only appear in subject position. It sounds odd if <emph>mei</emph> 'every' occurs in object position. In the present study, we adopted <emph>suoyou</emph> 'all' as the universal quantifier because the test sentences in Experiments 2 &amp; 3 require a universal quantifier in the object position. In so doing, we can keep consistency in terms of the use of the universal quantifier across the three experiments in the present study.</bibtext> </blist> <blist> <bibtext> An anonymous reviewer raises concern of the naturalness of the test sentences like (25). We are thankful to this valuable comment. To ensure the naturalness of the test sentences like (25), we did a follow-up survey with 20 Mandarin-speaking adults (10 males and 10 females; age range 31–48; mean 36). In the survey, the participants were invited to judge the naturalness of the test sentences, using a 5-point Likert scale, with 5 representing the perfectly grammatical and 1 representing completely ungrammatical. The mean naturalness score for the test sentences including DE was 4.12 (SD = 0.95), suggesting that the test sentences sound grammatical and acceptable.</bibtext> </blist> <blist> <bibtext> Besides the test sentences like (25), an anonymous reviewer finds the test sentences like (30) and (33) unnatural without the perfective aspectual marker <emph>le</emph>, and he believes that this might explain some adults' non-target-like responses. We agree with this view due to the fact that a subset of adults had a few non-target-like responses to the afore-mentioned test sentences. We also appreciate this friendly reminder as it will help us improve our future experimental design.</bibtext> </blist> <blist> <bibtext> Note that all of the test items were pre-recorded in a sound-attenuated lab at School of Foreign Languages, Hubei University of Technology, Wuhan, China. In addition, we did a post-recording survey on 10 Mandarin-speaking adults and they all confirmed that the test items were produced in a global level intonation without stress on any word.</bibtext> </blist> <blist> <bibtext> Compared with Experiment 2, children had better performance in Experiment 3. We attribute the reason to the fact that A-not-A questions have stronger yes–no question force than yes–no questions with the sentence-final particle <emph>ma</emph>.</bibtext> </blist> <blist> <bibtext> Note that the result from the corpus study is consistent with what Huang &amp; Crain ([19]) has found in a similar survey.</bibtext> </blist> </ref> <aug> <p>Reported by Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib21" firstref="ref3"></nolink> <nolink nlid="nl2" bibid="bib31" firstref="ref4"></nolink> <nolink nlid="nl3" bibid="bib34" firstref="ref5"></nolink> <nolink nlid="nl4" bibid="bib35" firstref="ref6"></nolink> <nolink nlid="nl5" bibid="bib37" firstref="ref13"></nolink> <nolink nlid="nl6" bibid="bib40" firstref="ref17"></nolink> <nolink nlid="nl7" bibid="bib28" firstref="ref26"></nolink> <nolink nlid="nl8" bibid="bib17" firstref="ref28"></nolink> <nolink nlid="nl9" bibid="bib10" firstref="ref30"></nolink> <nolink nlid="nl10" bibid="bib18" firstref="ref31"></nolink> <nolink nlid="nl11" bibid="bib29" firstref="ref32"></nolink> <nolink nlid="nl12" bibid="bib11" firstref="ref34"></nolink> <nolink nlid="nl13" bibid="bib47" firstref="ref36"></nolink> <nolink nlid="nl14" bibid="bib13" firstref="ref39"></nolink> <nolink nlid="nl15" bibid="bib24" firstref="ref40"></nolink> <nolink nlid="nl16" bibid="bib12" firstref="ref43"></nolink> <nolink nlid="nl17" bibid="bib14" firstref="ref45"></nolink> <nolink nlid="nl18" bibid="bib50" firstref="ref52"></nolink> <nolink nlid="nl19" bibid="bib15" firstref="ref54"></nolink> <nolink nlid="nl20" bibid="bib16" firstref="ref55"></nolink> <nolink nlid="nl21" bibid="bib36" firstref="ref69"></nolink> <nolink nlid="nl22" bibid="bib30" firstref="ref71"></nolink> <nolink nlid="nl23" bibid="bib52" firstref="ref72"></nolink> <nolink nlid="nl24" bibid="bib46" firstref="ref73"></nolink> <nolink nlid="nl25" bibid="bib19" firstref="ref76"></nolink> <nolink nlid="nl26" bibid="bib20" firstref="ref77"></nolink> <nolink nlid="nl27" bibid="bib25" firstref="ref81"></nolink> <nolink nlid="nl28" bibid="bib22" firstref="ref83"></nolink> <nolink nlid="nl29" bibid="bib44" firstref="ref86"></nolink> <nolink nlid="nl30" bibid="bib39" firstref="ref88"></nolink> <nolink nlid="nl31" bibid="bib42" firstref="ref89"></nolink> <nolink nlid="nl32" bibid="bib23" firstref="ref91"></nolink> <nolink nlid="nl33" bibid="bib26" firstref="ref103"></nolink> <nolink nlid="nl34" bibid="bib27" firstref="ref104"></nolink> <nolink nlid="nl35" bibid="bib32" firstref="ref123"></nolink> <nolink nlid="nl36" bibid="bib33" firstref="ref124"></nolink> <nolink nlid="nl37" bibid="bib43" firstref="ref155"></nolink> <nolink nlid="nl38" bibid="bib48" firstref="ref156"></nolink> <nolink nlid="nl39" bibid="bib49" firstref="ref169"></nolink> <nolink nlid="nl40" bibid="bib38" firstref="ref172"></nolink> <nolink nlid="nl41" bibid="bib41" firstref="ref176"></nolink> <nolink nlid="nl42" bibid="bib45" firstref="ref188"></nolink> <nolink nlid="nl43" bibid="bib51" firstref="ref223"></nolink> <nolink nlid="nl44" bibid="bib53" firstref="ref225"></nolink> <nolink nlid="nl45" bibid="bib54" firstref="ref233"></nolink> |
|---|---|
| Header | DbId: eric DbLabel: ERIC An: EJ1427651 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: 'Wh'-Words: Existential or Universal Quantifiers in Child Mandarin? – Name: Language Label: Language Group: Lang Data: English – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Haiquan+Huang%22">Haiquan Huang</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0001-9674-3367">0000-0001-9674-3367</externalLink>)<br /><searchLink fieldCode="AR" term="%22Hui+Cheng%22">Hui Cheng</searchLink><br /><searchLink fieldCode="AR" term="%22Lina+Qian%22">Lina Qian</searchLink><br /><searchLink fieldCode="AR" term="%22Yixiong+Chen%22">Yixiong Chen</searchLink><br /><searchLink fieldCode="AR" term="%22Peng+Zhou%22">Peng Zhou</searchLink> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="SO" term="%22Journal+of+Psycholinguistic+Research%22"><i>Journal of Psycholinguistic Research</i></searchLink>. 2024 53. – Name: Avail Label: Availability Group: Avail Data: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/ – Name: PeerReviewed Label: Peer Reviewed Group: SrcInfo Data: Y – Name: DatePubCY Label: Publication Date Group: Date Data: 2024 – Name: TypeDocument Label: Document Type Group: TypDoc Data: Journal Articles<br />Reports - Research – Name: Subject Label: Descriptors Group: Su Data: <searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Mandarin+Chinese%22">Mandarin Chinese</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Acquisition%22">Language Acquisition</searchLink><br /><searchLink fieldCode="DE" term="%22Preschool+Children%22">Preschool Children</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Usage%22">Language Usage</searchLink><br /><searchLink fieldCode="DE" term="%22Semantics%22">Semantics</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Universals%22">Language Universals</searchLink> – Name: Subject Label: Geographic Terms Group: Su Data: <searchLink fieldCode="DE" term="%22China%22">China</searchLink> – Name: DOI Label: DOI Group: ID Data: 10.1007/s10936-024-10079-4 – Name: ISSN Label: ISSN Group: ISSN Data: 0090-6905 – Name: Abstract Label: Abstract Group: Ab Data: "Wh"-words have been analysed as existential quantifiers (Chierchia in Logic in grammar: polarity, free choice, and intervention. Oxford University Press, Oxford, 2013; Fox, in Sauerland U, Stateva P (eds) Presupposition and implicature in compositional semantics (Palgrave studies in pragmatics, language and cognition). Palgrave MacMillan, Houndmills, pp 71-120, 2007; Liao in Alternative and exhaustification: non-interrogative uses of Chinese "wh"-words . Harvard University, 2010) or universal quantifiers (Nishigauchi, in: Theoretical and applied linguistics at Kobe Shoin 2, Kobe Shoin Institute for Linguistic Sciences, 1999). These two accounts have distinct predictions on how children initially interpret "wh"-words. The universal account predicts that children should initially interpret "wh"-words as universal quantifiers, whereas the existential account anticipates that children should start out with the existential interpretation. To adjudicate between the two accounts, the present study was designed to explore pre-schoolers' semantic knowledge of "wh"-quantification. Specifically, it investigated the interpretation of the "wh"-word "shenme" 'what' with 4-and 5-year-old Mandarin-speaking children and a control group of adults. Using a Truth Value Judgment Task (Crain and Thornton in Investigations in universal grammar: a guide to experiments on the acquisition of syntax and semantics. MIT Press, Cambridge, 1998), Experiment 1 evaluated whether children interpret the "wh"-word "shenme" 'what' as closer in meaning to the polarity sensitive item "renhe" 'any' or the universal quantifier "suoyou" 'all' in the antecedent of "ruguo" 'if' conditionals. Using a Question-Answer Task, Experiments 2 & 3 respectively investigated whether children interpret "shenme" 'what' as closer in meaning to "renhe" 'any' or "suoyou" 'all' in two types of questions: yes-no questions with the particle "ma" and A-not-A questions. It was found that both children and adults interpret "shenme" 'what' as closer in meaning to "renhe" 'any' than "suoyou" 'all'. The findings suggest that Mandarin-speaking pre-schoolers already have adult-like semantic knowledge of "wh"-quantification: "wh"-words are existential quantifiers rather than universal quantifiers. Due to the paucity of primary linguistic input, children's early mastery of the non-interrogative "wh"-words appear to support the biolinguistic approach to language acquisition (Chomsky in Aspects of the theory of syntax. MIT Press, Cambridge, 1965; Pinker in Language learnability and language development. Harvard University Press, Cambridge, 1984; Crain et al. in Language acquisition from a biolinguistic perspective. Neurosci Biobehav Rev, 2016. https://doi.org/10.1016/j.neubiorev.2016.09.004). – Name: AbstractInfo Label: Abstractor Group: Ab Data: As Provided – Name: DateEntry Label: Entry Date Group: Date Data: 2024 – Name: AN Label: Accession Number Group: ID Data: EJ1427651 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1427651 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s10936-024-10079-4 Languages: – Text: English Subjects: – SubjectFull: Foreign Countries Type: general – SubjectFull: Mandarin Chinese Type: general – SubjectFull: Language Acquisition Type: general – SubjectFull: Preschool Children Type: general – SubjectFull: Language Usage Type: general – SubjectFull: Semantics Type: general – SubjectFull: Language Universals Type: general – SubjectFull: China Type: general Titles: – TitleFull: 'Wh'-Words: Existential or Universal Quantifiers in Child Mandarin? Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Haiquan Huang – PersonEntity: Name: NameFull: Hui Cheng – PersonEntity: Name: NameFull: Lina Qian – PersonEntity: Name: NameFull: Yixiong Chen – PersonEntity: Name: NameFull: Peng Zhou IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 06 Type: published Y: 2024 Identifiers: – Type: issn-print Value: 0090-6905 Numbering: – Type: volume Value: 53 Titles: – TitleFull: Journal of Psycholinguistic Research Type: main |
| ResultId | 1 |
