emerging knowledge of the system in German does little to facilitate their
development of the system in Welsh. Likewise, their lack of development
with respect to gender marking in Welsh has no noticeable effect on their
development in German, at these early ages, although it is striking that
their production of determiner forms in German is lower than for mono-
linguals, suggesting that their emerging knowledge of the determiner
system is not ‘bolstered’ by the transparency of the determiner system in
Welsh. Any errors produced in both languages seem to reflect natural
default strategies that relate to the nature of the respective systems and are
unlikely to be the outcome of cross-linguistic transfer. When children are
learning two overlapping systems that are both complex in different ways,
the likelihood of cross-linguistic transfer may be reduced.
Word Order
One area that has demonstrated sensitivity to cross-linguistic influ-
ence in language acquisition is word order (cf. Müller et al., 2002; Müller,
1993, 1998). Many of the utterances produced by our German-Welsh bilin-
guals involved word order violations that are not necessarily typical of
monolingual speakers of those languages. However, as we will see,
although many of these utterances may reflect possible patterns in the
‘other’ language, it is not always easy to identify whether such violations
are the result of cross-linguistic transfer of some kind. The following sec-
tion presents some of these word order violations in German (with a pos-
sible influence from Welsh) and in Welsh (with a possible influence from
Germanic word order patterns in German and possibly in English). This
discussion is merely suggestive, outlining the potential for further studies
of patterns and sources of word order transfer between these two lan-
guages. More data are needed in order to address more detailed questions
regarding this structure.
Cross-linguistic overlap: Word order violations
A number of German utterances produced by our German-Welsh bilin-
guals involved word order patterns that are rarely produced by German
monolinguals, including an exceptionally high proportion of V1 sentences
among the sentences of CP2. For example, CP2 produced the following at age
2;9,8:
6(a) *Haben wir nicht, Salz Ta r ge t st r uc t u r e: Wir haben kein Salz
have we not salt we have no salt
V S Neg O S V Neg O
‘we have no salt’ ‘we have no salt’Cross-linguistic Influence and Patterns of Acquisition 59