|
Article Excerpt
ABSTRACT. With the intent to examine the universality and applicability of a well-documented article acquisition sequence, this study investigated the acquisition of English articles by a 9-year-old beginning Chinese ESL learner in the American context. The longitudinal data collected during the 13 months of the study were first thoroughly indexed using NUD*IST 3.0.4d (QSR 1998), a computer software program for qualitative data analysis. The data were then statistically analyzed by using the Statistical Package for the Social Sciences (SPSS 6.1; see Nousis 1996). Brown's (1973) coding scheme, Pica's (1983) coding scheme, and a revised scheme were used in scoring the spoken data. Results showed that a was acquired before the. The findings from this study, on the one hand, challenge several claims documented in the current L2 research literature and on the other hand, provide new insight into the understanding of acquisition processes of English articles by Chinese ESL learners. **
INTRODUCTION. The English article is 'one of the notorious problem areas that foreign learners of the language are faced with' (Robberecht 1983:61). It has also been reported that the English article is the ESL teacher's toughest teaching problem (Celce-Murcia & Larsen-Freeman 1983). Quite ironically, articles 'are never taught to the native speakers of English because of their obviousness' (Kaluza 1981:7) in spite of the facts that a and the constitute two of the top five most frequently used words in the English language (Carroll, Davies & Richman 1971, Taylor & Taylor 1990) and that the English article system is a complex syntactic phenomenon from the linguistic point of view (Christophersen 1939, Chomsky 1962, Hewson 1972, Quirk, Greenbaum, Leech & Svartvik 1985). Because of their frequent use, their complex syntactic nature, and their difficulty in second-language learning and teaching, English articles and their acquisition sequence have been the focus of a number of important second language acquisition studies (Bailey, Madden & Krashen 1974, Dulay & Burt 1974, Hakuta 1976, Huebner 1983, Parrish 1987, Master 1988, Thomas 1989, Chaudron & Parker 1990, Lee, Cameron, Linton & Hunt 1994). Research to date has consistently reported that the definite article the is acquired before a by ESL learners.
Although there have appeared so far no findings that challenge this proposed sequence, several issues relating to this sequence still remain unsettled and call for further research. First, this sequence needs to be tested among learners with different language backgrounds for its validity and universality. The baseline data of Chinese ESL learners, for example, are useful in this regard since there is no article system in Chinese. Moreover, in the current literature there is little longitudinal performance data available concerning Chinese-speaking children's acquisition of English articles. As Master (1988:9) rightly points out, 'a true picture of article acquisition should be based on longitudinal studies' and not on cross-sectional studies. Second, no explanation has been offered regarding why the is acquired before a, even though this sequence runs counter to current pedagogical practice wherein the is often considered more complex and difficult to teach, and it conflicts with learners' retrospection when they usually feel that the is more difficult to learn. Third, the term acquisition appears to be defined by different researchers in different ways. Understanding it in a more comprehensive way requires further studies and explorations. As an attempt to address those issues, the present study investigates longitudinally one Chinese child's acquisition of the English article system.
1. LITERATURE REVIEW. Research on the acquisition or the use of articles by L2 learners has appeared only relatively recently. There is a history of contrastive studies in articles: English versus Slavic in Kaluza (1963), English versus Persian in Jafarpur (1979), and English versus Arabic in Al-Johani (1982). Similarly, pedagogical analyses of the article system have appeared: Hok (1970), Grannis (1972), McEldowney (1977), and Master (1990). There have been descriptions of the use of English articles by learners sharing a given L1: Kharma (1981) for Arabic, Yamada and Matsuura (1982) for Japanese, Agnihotri, Khanna, and Mukherjee (1984) for Hindi and Punjabi, and Liao (1984) for Chinese. In the context of variation research, evidence has been found showing that variation on article selection and use can be affected by task (Tarone, 1985, Tarone & Parrish 1988), by semantic function of noun phrase (Huebner 1983), by interlocutor (Young 1991), by planning condition (Crookes 1989), and by different levels of the target-language proficiency (Huebner 1983).
Compared with the L1 acquisition order of morphemes, evidence has been found showing that in L2 acquisition, articles a/the (in the same group as the auxiliaries) are acquired after the group consisting of -ing, plural, and copular and before the group consisting of irregular past as well as the group consisting of regular past, 3rd person singular, and possessive -s (Krashen 1977). As for the acquisition sequence of articles, research findings have quite consistently suggested that the is acquired before a(n) (Hakuta 1976, Andersen 1977, Huebner 1983, Parrish 1987, Master 1988, Crookes 1989, Thomas 1989, Chandron & Parker 1990, Gorokhova 1990, Lee et al. 1994, Shannon 1995).
Hakuta (1976) conducted a longitudinal study of a Japanese girl learning English as a second language. This Japanese girl was observed over a period of 60 weeks, from age 5;4, which was five months after her exposure to English began, until age 6;5. Spontaneous speech was recorded once every two weeks. Extending Brown's method of the attainment of the 90% criterion in obligatory contexts, Hakuta scored for what he called errors of commission (supplying articles in nonobligatory contexts) as opposed to errors of omission (not supplying articles in obligatory contexts) and found a large number of such errors. This suggests that a learner may recognize the form of an article before recognizing its function. He also found that the was acquired before a by his research participant.
Andersen (1977) discussed the performance of 89 first-year university students in the use of English articles. All his participants were native speakers of Spanish. In an innovative way, Andersen identified the following four articles: a, the indefinite article; the, the definite article; [0.sub.1], the absence of an article or other determiner in English in those contexts where Spanish requires a definite article; [0.sub.2], the absence of an article or other determiner in English in those contexts where there is also no article or other determiner in Spanish. From his data, Andersen was able to find the following: (a) Most of the participants performed quite well with the in obligatory contexts, but their performance with a ranged between almost 0% and 100%; (b) the indefinite article a was frequently omitted in contexts where it is omitted in Spanish and one was often used where a is required; and (c) the participants performed well when the articles are the same in each language (i.e., [0.sub.2]) hut not nearly so well when the articles are different in the two languages (i.e., [0.sub.1]). Based on those findings, Andersen concluded that 'wherever English and Spanish both require the definite article or do not permit an article (the and [0.sub.2]) the participants perform well. When Spanish and English require different articles (a and [0.sub.1]) considerable variability is evident and many of the errors seem to be interference errors' (Anderson 1977:72).
Yamada and Matsuura (1982) investigated article usage in Japanese students. Although they looked for target-like usage rather than a non-target interlanguage system, the study was an improvement on earlier work in that the articles were broken into separate tallies for a(n), the, and 0. In general, Yamada and Matsuura found that the overall difficulty order (easiest to hardest) for intermediate-level Japanese ESL students was the [right arrow] a(n) [right arrow] whereas the difficulty order for advanced-level Japanese ESL students was the [right arrow] [right arrow] a(n).
It is not difficult to see that the studies done by Hakuta, Andersen, or Yamada and Matsuura were more or less based on Brown's criterion paradigm, which was criticized by Bickerton (1981) as the tendency 'to look to the goal, rather than the path, to ask "What has the child acquired?" rather than "How has he acquired it?'" (Bickerton 1981:143).
In an effort to abandon this criterion paradigm, Huebner conducted a longitudinal study in 1983. His lone participant was a Hmong refugee learning ESL in an untutored setting. The participant, in his early twenties, had never had any formal instruction in English, and data collection began within a few weeks of his arrival in Hawaii. The data were collected on an average of every three weeks for one year. Huebner developed a system of analysis that accounted for article use in all pre-nominal positions. Using Bickerton's (1981) proposed universal features of referentiality [SPECIFIC REFERENT] and [ASSUMED KNOWN TO HEARER], Huebner divided environments for articles according to whether the noun phrase on which the article was dependent was used referentially or nonreferentially [[+ or -] Specific Referent] ([[+ or -] SR]), and whether or not that the noun phrase was identifiable by the listener [[+ or -] Assumed Hearer's Knowledge] ([[+ or -] HK]). These two binary features varied independently, yielding a 'semantic wheel' or more accurately speaking, four cross-classified environments...
|
