31 Speech Learning in
Bilinguals: Consonant Cluster
Acquisition
Robert Mayr, Danna Jones and
Ineke Mennen
Introduction
A key issue in the study of bilingual speech learning is to understand
along which phonetic and phonological dimensions languages can systemati-
cally differ, and how such differences are generated in speech production,
evaluated in speech perception and ultimately acquired by those learning to
speak two (or more) languages. Identifying these dimensions is important as
it will lead to a better understanding of how bilinguals manage two –
partially competing – sound systems. This, in turn, will help identify what
is learned phonetic behaviour and inform models of phonetic/phonological
representation. Many studies have shown that one of the consequences of
bilinguals having to manage or ‘juggle’ two competing sound systems is that
their speech production differs at least to some extent from that of monolin-
gual speakers in each language. We see such interaction between the two
languages in bilinguals as a perfectly normal phenomenon resulting from
language contact, and one that does not only occur in bilinguals who started
learning a second language (L2) later in life (i.e. sequential or adult bilingualism)
(e.g. Flege et al., 1995), but also in bilinguals whose exposure to both lan-
guages started at birth or soon thereafter (i.e. simultaneous or childhood
bilingualism) (e.g. Kehoe, 2002).
The Speech Research group of the ESRC centre for research on bilingual-
ism in theory and practice aims to investigate the nature and extent to which
languages differ systematically in their phonetic and phonological properties,
and how these differences are realised and acquired by bilinguals. Languages
are known to differ in their phoneme inventories and their fine phonetic