(Fabiano-Smith & Barlow, 2010). Evidence for acceleration is, for instance,
found in Lléo et al. (2003). This study revealed greater accuracy in the pro-
duction of final consonants in Spanish by Spanish-German bilingual chil-
dren than age-matched Spanish-speaking monolinguals, perhaps because the
less restricted patterns of syllable-final consonants in German had a facilita-
tive effect. Similarly, Grech and Dodd (2008) found that children exposed to
both Maltese and English at home were significantly more accurate in their
overall percent consonant correct scores than those only exposed to one lan-
guage at home.
However, bilinguals are not ‘the sum of two complete or incomplete
monolinguals, but have a specific and unique linguistic configuration’
(Grosjean, 1995: 259). Some studies have therefore focused, not on compari-
sons between monolinguals and bilinguals, but between different types of
bilinguals (Goldstein et al., 2005; Law & So, 2006; Munro et al., 2005). Munro
et al. (2005), for instance, examined the acquisition of singleton consonants
in Welsh-English bilingual children differing in language dominance. The
results revealed significantly greater accuracy by the Welsh-dominant chil-
dren on consonants occurring in Welsh words, such as /ɡ ŋ v θ ʃ ɬ w j r/.
Interestingly, the advantage of the English-dominant bilinguals on conso-
nants occurring in English words was somewhat less equivocal, with greater
accuracy than the Welsh-dominant bilinguals on some consonants, that is,
/s ʃ tʃ dʒ/, but worse performance on others, that is, /t h/. Interestingly, con-
sonants shared by Welsh and English were sometimes acquired earlier in one
language than the other. Language-specific differences of this kind are also
reported in many other studies (Dodd et al., 1996; Holm & Dodd, 1999).
Together, they indicate that bilingual children have differentiated language
systems.
This does not mean, however, that their phonological systems are
entirely independent of each other. Rather, the evidence suggests that the
two languages of bilinguals interact with each other, exhibiting cross-
linguistic effects (Fabiano-Smith & Barlow, 2010; Grech & Dodd, 2008;
Holm & Dodd, 1999; Paradis, 2001; So & Leung, 2006; Stow & Pert, 2006;
Yang & Hua, 2010). Keshavarz and Ingram (2002), for instance, report that
the Farsi-English bilingual child in their study used patterns specific to one
language in the other language. Likewise, the two sequential Cantonese-
English bilingual children in Holm and Dodd (1999) generalised language-
specific simplification patterns, such as final consonant backing in Cantonese
and final consonant deletion in English, to the other language. This resulted
in errors that would be considered atypical in monolingual children of either
language. Although most studies report cross-linguistic effects of this kind,
their incidence in individual bilingual children appears to be comparatively
low (Fabiano-Smith & Barlow, 2010).
One aspect of phonological acquisition in bilingual children that has
received little attention thus far is consonant clusters. Consonant clusters are
Speech Learning in Bilinguals 5