The acquisition of control crosslinguistically: structural and lexical factors in learning to licence PRO 1
Rules for interpreting empty category (EC) subjects of complement clauses vary crosslinguistically across structural and lexical dimensions. In adult Greek, a distinction is made between the verbs meaning WANT and TRY, the former but not the latter permitting the EC subject of its subjunctive complement to refer outside the sentence. The EC is pro for WANT and PRO for TRY. In adult Spanish, both the verbs meaning WANT and TRY require the EC subject (pro) to refer outside when the complement is in the subjunctive, and require the EC subject (PRO) to refer to the main clause subject when the complement is in the infinitive. Twenty-three Greek-speaking four- to five-year-olds and 10 adults, 29 Spanish-speaking four- to five-year-olds, 18 six- to seven-year-olds and eight adults took part in act-out experiments. The results indicate an awareness of language-particular distinctions governing the interpretation of EC complement subjects. However, child speakers of both languages experience difficulty in giving sentence external reference, leading to error in the case of subjunctive sentences for Spanish-speaking children. We argue that the data overall is most compatible with children having access to the empty category PRO by age four, and that failure to give external reference of an EC when required can plausibly be treated as performance error. A picture verification task produced less clear results, but points to the need for data from younger children to establish whether there is an early stage in which lexical semantics dominates children's interpretation of ECs.(Received November 6 1998)
(Revised February 17 2000)
c1 Address for correspondence: Helen Goodluck, Department of Linguistics, University of Ottawa, 70 Laurier E. Ottawa, ON, K1N 6N5, Canada. e-mail: firstname.lastname@example.org.
c2 Address for correspondence: Arhonto Terzi, Department of Speech Therapy, TEI Patras, Meg. Alexandrou 1, Koukouli, 26334, Patras, Greece. e-mail: email@example.com.
c3 Address for correspondence: Gema Chocano Díaz, English Department, Universidad Autónoma de Madrid, Cuidad Universitaria de Cantoblanco, 28049, Madrid, Spain. e-mail: firstname.lastname@example.org.
1 This research was supported by Social Sciences and Humanities Research Council of Canada grant number 410-94-0707. Two anonymous reviewers provided very helpful commentary. Various parts of the work have been presented at the 20th Annual Boston University Conference on Language Development, the 28th Annual Child Language Forum at Stanford University, the 18th meeting of the Linguistics Department of the University of Thessaloniki, the 3rd International Conference on Greek Language in Athens, and the Workshop on the Acquisition of Syntax and Semantics in Trieste, in Summer 1998. We would like to thank the participants to these events for their comments, as well as students in course L72 at the Ortega y Gasset Institute, Madrid, in Spring 1995. A particular thanks to Maria Teresa Guasti, Andreas Papapavlou, Maria-Luisa Rivero, Melanie Sellar and Ken Wexler. Our heartfelt thanks go to the staff and children at the kindergartens and daycare centres where we did our testing: N. Smirni and EPOIZO in Athens, Vivi Bakalarou and the children at Kagadi, Patras and Colegio Nuestra Señora de la Consolación, Madrid.