Published online by Cambridge University Press: 26 September 2008
The traditional assumptions that novel word compounds fill lexical gaps and allow speakers to convey an intended meaning more precisely were explored. Examples from spontaneous language demonstrated that some novel compounds occur in the absence of a lexical gap and that not all compounds serve a communicative function. The relation between communicative demands and novel noun–noun compound use was explored experimentally also. Twenty-eight five-year-old children and 16 adults participated in referential and non-referential communication tasks in which they were exposed to referents whose elements were inherently and non-inherently related. Both children and adults produced more compounds for inherent than for non-inherent referents. However, although the children demonstrated that they were sensitive to the need for greater communicative precision in the referential compared to the non-referential task, there was no difference in frequency of compound use across tasks. These results suggest that the functions ascribed to novel compounds warrant closer scrutiny.
This study is based on a dissertation completed under the direction of Laurence Leonard at Purdue University. His comments and suggestions directly influenced aspects of the study and are gratefully acknowledged. Also, grateful thanks are extended to Gwendolyn Pennington for drawing the stimulus materials and to Patricia Arndorfer, Barbara Drobes, and Lucia Mendez-Buckley for their help in running subjects. Helpful comments by Garth Holloway, George Allen, Mark McDaniel, Richard Schwartz, Gerald Siegel, and two anonymous reviewers are appreciated.