Second language speech production

The study of second langiage (L2) speech production has a long history in second language (SLA) research, both for what it can tell us about the development of the specific skill and how it might illuminate the general processes of SLA. Over time, studies in L2 phonological attainment have become a battleground for perennial issues in SLA research encompassing cognitive, psychological, and socio-cultural factors as varied as age-related constraints on ultimate attainment, and the role of identity in perceptions of accentedness. SLA research has also benefited from the ongoing development of phonological theory which has been consistently applied to L2 production throughout the decades. The investigation of L2 speech production has been part of second language acquisition research from its beginnings. Following the prevailing linguistic paradigm of the time, researchers employed Contrastive Analysis (CA) (Lado, 1957) and the “difference = difficulty” hypothesis to explain the shape of L2 accents (Weinreich, 1953). Despite this belief that transfer was at the root of non-native accents, it became evident that the prognoses made by CA with regard to L2 phonology, as with other language systems, were not predictive of learner error. Recognizing this difficulty, Eckman supplemented CAwith theMarkedness Differential Hypothesis (1977) and the Structural Conformity Hypothesis (1991). Throughout the 1980s, speech researchers followed the shift in the field of SLA away from CA toward a closer investigation of learner language through Error Analysis (Corder, 1967) and the recognition of stable, transitional grammars or interlanguages (ILs) (Selinker, 1972). Major’s Ontogeny Model (OM) (1987) for example, proposed a three-part structure underlying IL comprising influences from L1, L2 and universal processes. The OM further stated that these influences would be more salient during different phases of learner development in phonology with L1 transfer initially frequent in the IL and then decreasing as developmental processes increased. During this time, few had challenged the presumption of a critical period (Lenneberg, 1967) for linguistic development, and Scovel (1988) predicted that post-pubescent L2 learners would retain permanently accented speech. In the early 1980s, however, Flege and his colleagues began to question the role of the CPH as the primary explanatory factor for the differences between the production of L2 phonetic segments by children and adults (see later in this chapter for a detailed discussion of the debate regarding the CPH). Based on an ongoing series of studies (Flege, 1981,