Perception of Articles in L2 English

It is well established in the relevant literature that the second language (L2) acquisition of articles is difficult, particularly if the L1 lacks articles (Huebner 1985; Robertson, 2000; Thomas, 1989; Ionin, Ko & Wexler, 2004; Trenkic, 2007). Typical errors include omissions in obligatory contexts; inappropriate substitution of one article for the other (typically, definite for indefinite, see Thomas 1989). There are several hypotheses that account for the errors encountered in the inter-language of L2 English learners. The Failed Functional Features Hypothesis (FFFH) (Hawkins & Chan, 1997) assumes that if the L1 lacks a particular functional category, it cannot be learned in the L2. On the Missing Surface Inflection Hypothesis (Prévost & White, 2000), functional categories are present in the inter-language, but learners have difficulty retrieving the corresponding surface morphology. More recently, Goad, White & Steele (2003) and Goad & White (2004) have attributed omission of articles and other functional elements to constraints imposed by the prosodic system of the L1. Their Prosodic Transfer Hypothesis posits that the acquisition of free clitics, such as articles, in the L2 is constrained by the rhythmic qualities of the L1. Finally, in the realm of semantics, the Fluctuation Hypothesis (Ionin et al. 2004) asserts that L2 learners have access to the semantic universals of definiteness and specificity, but have difficulty learning from the input which one of these is encoded by English articles, and therefore fluctuate between them, overusing the with specific indefinites and overusing a with non-specific definites. All of the foregoing hypotheses are based on production data, written or spoken, but relatively little experimentation has been carried out to understand what learners actually perceive in the input. Perception of individual phonemes has been relatively well studied and has provided strong evidence that the L1 phonemic inventory has a significant effect on the perception of phonemes in the L2 (Logan, Lively & Pisoni, 1991; Lively, Logan & Pisoni, 1993; Bradlow, Pisoni, Yamada, & Tohkura, 1996; McCandliss, Fiez, Protopapas, Conway & McClelland, 2002). Given that the features of individual sounds are perceived through a filter of the phonemic inventory of the L1, it is reasonable to assume that other aspects of the L1 phonology, e.g. prosodic or rhythm qualities would also influence perception of the L2. The research reported on here was designed to address the following questions: 1. Do learners whose native languages are prosodically different from English (in this case Mandarin and Korean) perceive English articles accurately? 2. Does overall L2 English proficiency and/or grammatical knowledge of English articles correlate with accurate perception of English articles? And finally, 3. Neither Mandarin nor Korean has articles, but given that the two languages have different prosodic systems, do L1 Mandarin and L1 Korean learners of L2 English differ in their perception of English articles? Given that neither Mandarin nor Korean has definite or indefinite articles (see Li & Thompson, 1981 for Mandarin; Kim-Renaud, 1994 for Korean), the acquisition patterns of English articles should be similar and indeed, prior work on L2 English learners’ article choice in written tasks found similar error patterns among Korean speakers (Ionin et al. 2004) and Mandarin speakers (Trenkic, 2008).