Acquisition of copulas ser and estar in Spanish: learning lexico-semantics, syntax and discourse

Although there are a very large number of proposals in the literature to account for the complex distribution of the copulas ser and estar in Spanish and Portuguese (Bello 1951; Gili Gaya 1955; Querido 1976; Luján 1981; Lema 1995; Schmitt 1992, 2004; Maienborn 2000, 2003), very little is known about children’s acquisition of these two highly frequent copula verbs. The only acquisition study we are aware of is on natural and elicited production by Sera (1992). To our knowledge nothing has been published on children’s comprehension of these copulas and therefore we don’t even have preliminary comprehension data that could help us in answering basic questions such as the following: (i) do young children have problems interpreting the distinctions between ser and estar when both options are grammatical? (ii) if their performance is not adult-like, do they default to one of the copulas and if so, which one? Our paper aims at providing preliminary answers to these two questions. We present the results from two studies on children’s comprehension/acceptance of the copulas ser and estar followed by adjectival predicates in particular contexts. Study 1 examined children’s ability to use the pragmatic implicatures associated with each of the copulas in the context of permanent vs. temporary properties via Picture Matching Tasks (PMT). Study 2 is an Acceptability Judgement Task (AJT) that attempts to verify the extent to which children are aware that lexical and syntactic properties of certain adjectives and/or preceding discourse are crucial pieces of information for determining which copula should be used. As we will see below, because the choice between the two copulas is governed by an intricate set of factors that go from semantic and syntactic properties of certain adjectives to semantic and/or pragmatic factors that involve larger discourse contexts, the acquisition of the copulas can be used as a testing ground for proposals that have been made concerning children's different abilities to incorporate syntactic, semantic, pragmatic and discourse information in comprehension at different stages in their development. More specifically, in study 1 we will investigate Noveck’s (2001) hypothesis that children first master the logical meaning of expressions and only later compute the implicatures associated with different items. In study 2 we will investigate the hypothesis that children have more trouble incorporating discourse information than syntactic and semantic information by asking children to judge the acceptability of ser and estar when different types of information (discourse or syntactic or semantic) must be taken into consideration. On a more general level, our results show that 4-5 year-olds have not yet mastered the ser and estar distinction in comprehension. The results also provide empirical support for both hypotheses: children do not always use the implicatures associated with particular items to make a choice between two elements, although they seem to distinguish between the two copulas. Second, in the AJT, children had a lot of trouble incorporating discourse information to judge the acceptability of sentences with ser and estar, but again were more willing to accept estar in the inappropriate context than vice-versa. This paper is organized as follows: section 2 will provide a brief description and analysis of the properties of ser and estar; section 3 provides the acquisition background against which our results should be discussed; section 4 presents study 1; section 5 presents study 2 and section 6 summarizes the results.