30
EDY VENEZIANO
4.2. The emergence of phonomorphologically relevant (PMR) variations and the
beginnings of linked verbal morphology
When verb words enter the child’s vocabulary, different occurrences may present
phonological variation (for example, /’be/ and /o'be/ for /tδ'be/, ‘to fall, fallen’; /’kas/
and /'kaʃ/ for /'kaʃ/, ‘hide(s)’). Such variation is not morphologically relevant. From the
morphological point of view, verb-words are produced for some time in only one form.
This can be the infinitive/past participle form for the verb-words of the first group (end-
ing in - er) as, for example, /'ka'ʃe/ for /'ka'ʃe/, ‘to hide/hidden’, /e'ti/ for /'sor'tir/, ‘to get
out/got out’, and /e've/ for /ɛlə've/ ‘to remove/removed’. For other verb-words, the cho-
sen form might be the present indicative, most often homophone of the imperative form
as, for example, /’bwa/ for /’bwa/, ‘drink(s), /'sot/ for /'sot/ , ‘jump(s), and /'plæ/ for /
'plær/ ‘cry(ies)’. Producing these different forms for different types of verb-words can-
not be used as evidence that the child knows about morphological variations. The cho-
sen form may simply be the one the child has retained to produce the corresponding
lexical item. However, when the child produces phonomorphologically relevant varia-
tions for one and the same verb-word, it becomes more justified to consider such pro-
ductions as evidence for the beginnings of verbal grammatical linked morphology.
Single-form verb morphology has been reported in several studies of early lan-
guage acquisition, and for children learning different languages (see, for example,
Gathercole, Sebastian & Soto, 1999; Pizzuto & Caselli, 1994; Tomasello, 1992). The
studies reported here confirm these results. C produces only one PMR form per type
of verb-word until 1;9, and G until 1;11. Afterwards, at least two PMR forms are
found for some types of verb-words. Table 2 presents several examples for the girl and
for the boy.
This phenomenon doesn’t occur for noun-words nor for words belonging to other
grammatical categories in the adult language.
5. Supplementary signs of word categorization
The systemic approach we have taken in this work leads to further information
susceptible of confirming or disconfirming the hypothesis of an initial absence of
categorial differentiation among the words in the child’s early vocabulary.
5.1. Absence of categorial differentiation and the production of PAEs
Results of several analyses performed on C’s production of PAEs strongly point
to the existence of two periods. In the first period PAEs have a premorphological
status, and in the second PAEs are considered protomorphological, as they start to
present some characteristics of French grammatical morphemes (Veneziano, 1998;
Veneziano & Sinclair, 2000).
During the first period, C seems essentially to aim at the realization of an iambic
pattern in the V’CV form11. Similar results are found also in the data of the boy and
11 The iambic pattern is typical of plurisyllabic French words and of monosyllabic words embedded into
a proximal syntactical structure of the type ‘DET+N’ (’la grue’) and ‘PRON+V’ (‘iljoue’).