Such a development is in line with previous findings demonstrating that
gender marking for natural gender has been reported to be the earliest stage
of gender acquisition in (near-)monolingual Welsh children (Gathercole et al.,
2001, 2005; Gathercole & Thomas, 2009). Demonstrative pronouns, on the
other hand, were produced frequently by all children, bilingual and (near-)
monolingual alike (see Table 3.8). However, as with the personal pronouns,
the use of the masculine form dominated, even in contexts involving femi-
nine nouns:
(CP1 2;11,6):
5(a) na, hwnna ddim mam Erin [= toy female]
no, that(m)-not-mother-Erin
‘no, that [is]not Erin’s mother’
(CP2 2;7,14):
5(b) hwnna doli
that(m)-dolly
‘that [is the] dolly’The children’s exclusive use of the default masculine form in the case of
the demonstrative pronoun could be influenced by the corresponding
German form, which is always das (neuter) (das ist der Hund ‘that is the dog’
(masc.), das ist die Frau ‘that is the woman’ (fem.), das ist das Haus ‘that is the
house’ (neut.)), irrespective of the gender of the noun to which it refers. At
the same time, (near-)monolingual Welsh children also use an overwhelm-
ingly large number of masculine demonstratives, indicating a possible influ-
ence of the English ‘it’, with first uses of the feminine at a more advanced
stage in language development beyond the ages of the bilinguals in this
study. The use of a masculine pronominal form to modify a feminine noun
has also been documented as usual practice among adults, both for Welsh
(Jones, 1993; Jones, 1998) and for Scottish Gaelic, which has a similar gender
system (Dorian, 1976). Our bilinguals’ utterances may therefore simply
reflect the nature of their input rather than cross-linguistic influence per se.
In terms of adjectives, the bilinguals produced more forms in Welsh than
in German, although frequency of use was still low. The adjective forms
produced were all used to modify masculine nouns and all appeared in their
target, non-mutated form.
Summary: Gender
Together, the patterns emerging from our initial analysis of these three
data points in both languages are not suggestive of cross-linguistic influ-
ence with respect to gender acquisition. Although our bilinguals are dem-
onstrating greater knowledge of gender in German than in Welsh, their
58 Part 2: Bilingual Language Development