24
EDY VENEZIANO
1982; Bates, Bretherton & Snyder, 1988; Bassano, 1998), for semantic (nouns have
more identifiable and more concrete referents), formal (are less diverse morphologi-
cally, present greater regularities in their context of appearance, e.g. Maratsos & Deak,
1995) or for structural reasons (predicates presuppose arguments, e.g. Macnamara,
1986). These variables might account for the fact that children more easily learn
certain words than others, words that in languages like English, French, Italian and
German, where the bias is found, function as nouns1.
But does it make any sense to call the words these young children produce „nouns”
and „verbs”? If these words are nouns and verbs in the language children are acquir-
ing, are they also nouns and verbs in their emerging language systems? Or do they
start out as semantic units, undifferentiated from the point of view of their grammati-
cal category?
This paper contributes analytical tools to clarify this issue, looking for indices
pointing to the presence of a noun/verb categorization within the child’s emerging
language system. This is an important issue that needs to be clarified before we can
even start making hypotheses about how children learn nouns and verbs from the
language they hear, or about how they use the information contained in such identifi-
cation for retrieving the meaning of words, and vice versa.
2. Some considerations about nouns and verbs in French
Lazard (1984) states clearly that nouns and verbs in French are „purely lin-
guistic objects whose reality exists only within the language” and are defined „by
their place and their properties in the structure of the language, namely, in mor-
phology and syntax”2 (op.cit., p.29). The semantic distinctions that correspond in
general to nouns and verbs may provide the cognitive and functional support to
the formal distinctions between them. Diachronically, semantic and formal dis-
tinctions may have risen at the same time. However, if semantic distinctions didn’t
correspond to formal differences, we would be simply confronted by words differ-
ing in meaning, without the possibility, nor the need, to identify categories of
words (e.g., Clairis, 1984 ). Lexical acquisition would then be reduced to acquir-
ing the meaning of individual words without any bootstrapping coming from the
category to which the word belongs.
Thus, when considered from the point of view of the knowledge children have of
grammatical categories like nouns and verbs, the question asked at the beginning is
more complex. In order to attribute to the child knowledge of part of speech categories
what is needed is evidence of formal distinctions between words such that they are
pulled together under some similarities and distinguished across some differences. I
will argue that such evidence can be provided by using a systemic and multidimen-
sional approach that takes into account the development of different aspects of chil-
dren’s language production at the same time (Veneziano, 1999).
1 The bias is not found in all languages, in particular, not in Korean (e.g., Gopnik & Choi, 1995).
2 translation by the author