of words are correct (Langacker 1987, 1991; Lakoff 1987). If such patterns exist and
children do learn them, it may be possible to learn new meanings for words that have
been defined by one of the learned patterns. Precious little research (Gropen, Ep-
stein, and Schumacher 1997) has been done on the acquisition of polysemes, so any
proposal about how polysemes are learned remains speculative. Nonetheless, this
topic may shed a good deal of light on the way we learn the meanings of words and
therefore is worthy of investigation.
Conclusion
The research described in this chapter supports the idea that communicative goals of
language are reflected in learning biases. I have demonstrated that children disprefer
learning a different, unrelated meaning for a known word when that word is used in a
linguistic context that fails to bias strongly for a new meaning. Children appear to
have much less difficulty in learning homonyms when the syntactic context clearly
indicates that a new meaning is required (cf. the verb homonym condition verses the
two noun homonym conditions). I do not suggest that children are incapable of learn-
ing homonyms; clearly that is not the case. Nonetheless, the experiment presented in
this chapter does demonstrate a bias against homophony in the acquisition process.
Acknowledgments
I am very grateful for the advice and guidance of Adele Goldberg in the planning
stages of this study as well as in the preparation of this chapter. I am also indebted to
Dan Silverman, whose thought-provoking phonology courses planted the initial
seeds for this research. Finally, I thank the teachers, students, and parents at Country-
side and Next Generation schools for assisting with and participating in this study.
This chapter is an abbreviated version of “Children’s resistance to homonymy: An
experimental study of pseudohomonyms,” which appears in the
Journal of Child
Language
, 32, no. 1 (2005), © Cambridge University Press.
NOTES
1. Gurevich (2004) did not verify that the changes did result in homophony of any kind.
2. Compare this result with the 25 percent of all participants who responded with the correct meaning
of the homonym (the remaining 44 percent of participants invented answers, could not respond, or
gave answers the experiments could not classify).
REFERENCES
Bréal, Michel. 1897.
Essai de sémantique
. Paris: Hachette.
Campbell, Robin N., and Theresa Bowe. 1977. Functional asymmetry in early language understanding. In
Salzberger Beitrage für Linguistik,
ed. Gaberell Drachman. Tubingen, Germany: Gunter Narr.
Campbell, Robin N., and Theresa Bowe-Macdonald. 1983. Text and context in early language comprehen-
sion. In
Early child development and education
, ed. Margaret Donaldson, Robert Grieve, and Chris
Pratt, 115–26. Oxford: Blackwell.
Clark, Eve. 1983. Meanings and concepts. In
Handbook of child psychology,
4th ed., vol. 3:
Cognitive de-
velopment
, ed. John H. Flavell and Ellen M. Markman, 787–840. New York: John Wiley and Sons.
Donaldson, Margaret C. 1978.
Children’s minds
. London: Fontana.
Gathercole, Virginia C. 1989. Contrast: A semantic constraint?
Journal of Child Language
16:685–702.
34
Do'stlaringiz bilan baham: