Developmental Psychology 1983, Vol. 19, No. 3, 440-444 Copyright 1983 by the American Psychological Association, Inc. Initial Verbs of Yes-No Questions: A Different Kind of General Grammatical Category Stan A. Kuczaj II Southern Methodist University Michael P. Maratsos University of Minnesota The nature of grammatical categories has long been a subject of dispute for theorists of language development, just as the nature of concepts has long been a concern for theorists of cognitive development. In our study, we examined 16 children to determine their knowledge of placement privileges of auxiliary verb forms (e.g., am, can, have) in yes-no questions. Our purpose was to specify the grammatical category (if any) the auxiliary verb forms constituted early in de- velopment. The results suggested that the category of concern is a peculiar mixture of general and specific knowledge. The implications of this finding for theories of language and cognitive development are briefly considered. As young children acquire their first lan- guage, they gradually create a productive grammatical system that provides the basis for the production and comprehension of novel utterances (i.e., ones that the children have not previously encountered). Although much of the work in language development during the 1960s and early 1970s focused on the development of the productive gram- matical system (Bloom, 1970; Brown, 1973; McNeill, 1970), a consensus about the nature of the system and the nature of the processes underlying its development has still not been reached (e.g. Kuczaj, 1982; Maratsos & Chalkley, 1980). Thus, the literature on syn- tactic development is replete with disagree- ments about the nature and development of grammatical rules and grammatical classes (see the following for varied explanations of errors such as did I missed it?; Fay, 1978; Hurford, 1975; Kuczaj, 1976; Maratsos & Kuczaj, 1978; Prideaux, 1976). In this short article, we will discuss and illustrate a kind of knowledge of grammar in young children that has not been widely studied: knowledge of a grammatical cate- gory denned by obligatory use in a sentential A version of this article was presented as part of a symposium, The Child's Formulation of Grammatical Categories and Rules, at the meeting of the Society for Research in Child Development, San Francisco, March 15-18, 1979. Requests for reprints should be sent to Stan A. Kuczaj II, Department of Psychology, Southern Methodist Uni- versity, Dallas, Texas 75275. position, but not yet analyzed by the child for the general related properties that predict a term being a candidate for membership in this grammatical category. In particular, we will argue that in acquiring the correct use of English auxiliary verbs in yes-no ques- tions, children realize early in their devel- opment that some relational term must ap- pear in the sentence's initial position (i.e., before the subject-noun phrase, as in "will the boy eat the ice cream?", will being the sentence-initial-relational term, and the boy being the subject-noun phrase). This realiza- tion is an abstract structural knowledge that goes beyond the specific knowledge that par- ticular individual terms can occur in this po- sition. At the same time, however, children have not yet correlated yes-no question uses of auxiliary verbs with appropriate declara- tive uses. The result is a consistent failure to predict from declarative uses which terms might constitute members of this yes-no aux- iliary class. Thus, for a particular time during the course of development, children appear to have a peculiar mixture of specific and general knowledge about the use of auxiliary verbs in yes-no questions, similar to that ev- idenced during the acquisition of the modal auxiliary placement rule in wh questions (Kuczaj & Brannick, 1979). In the developmental period of interest (roughly corresponding to the mean length of utterance [MLU] range of 3.0 to 4.5), chil- dren reliably produce yes-no questions in which auxiliary verbs are used in correct sen- 440