January 2011I, Ingrid Lossius Falkum, confirm that the work presented in this thesis is my own. Where information has been derived from other sources, I confirm that this has been indicated in the thesis.

result would be an overall rate of near zero in a sample of two to three years. Our samples are so small (usually an hour a week) that they would fail to catch these occurrences. Only less frequent verbs, discounted by Marcus and colleagues, might show evidence of strong overregularization. In fact, that they do was shown for R. Brown’s two low-overregularizing subjects Adam and Sarah (see, e.g., 1973). For Adam, for example, the average overregularization rate was a strong 55 percent for his 21 lowerfrequency verbs. The same rate was found even in samples after the child first produced the correct irregular form of a verb. Arguments from sampling considerations indicated that such overregularizations were still persisting after tens or even hundreds of uses. Recent work from a more intensively recorded subject, Peter (Maslen et al. 2004), has strongly supported these analyses and extended them to noun plurals. These data indicate

Accessing multiple meanings: the case of zeugma