| Writing ability comprises a significant component of second language competence and has been a topic of extensive research in the field of applied linguistics in the recent 30 years. Its development, especially in a language other than LI, involves a wide range of factors such as L2 proficiency, language aptitude, working memory capacity, motivation, and learning strategies. Therefore, it becomes notoriously difficult to promote L2 writing development by instructional means. The difficulties notwithstanding, research efforts have never ceased. A considerable batch of studies have emerged to explore the underlying rules and trajectory of L2 writing development, with a focus primarily on feedback and interlanguage development.However, those initial research endeavors have generated mixed findings only, leaving many issues unresolved but increasingly complicated. A case in point is the role of written corrective feedback (WCF) in L2 writing development, which has been heatedly debated, but still remains highly controversial even though error correction has long been a common practice in second language classrooms. Another instance is the incompletely revealed picture of L2 writing development where there are even conflicting findings concerning the development of various dimensions of L2 writing. Therefore, more in-depth research investigations are needed to elucidate second language writing development as well as its contributing factors. Theoretically, such studies can put an end to controversies over the effectiveness of written corrective feedback and thus establish the role of negative evidence in second language acquisition; in the meanwhile, they can also delineate a clearer picture of the routes and rules of L2 writing development. Pedagogically, the research findings can justify some traditional teaching practice like error correction and help breed innovations in L2 writing instruction.The inconsistency and even conflicts in previous research attempts can be attributed to deficiencies in research design and differences in research foci. For WCF-related research, it is mostly undertaken quantitatively in laboratory settings, which potentially divorces research from pedagogy. In contrast, there is an obvious shortage of classroom-based investigations which are in urgent need to lend empirical support to some pedagogical practice in real L2 classrooms. A related deficiency resides in the scant attention to unfocused (comprehensive) error correction even though it has long been practiced in L2 writing classrooms worldwide. This lack of attention, in sharp contrast to the plethora of research on focused error correction, may potentially compromise the value of the existing research findings, and thus should be redressed so that research and pedagogy can be further wed. In addition, feedback is treated predominantly as a remedial measure, rather than a preventative one, to errors in second language writing, as is manifested in most, if not all, empirical studies. It will be pedagogically more sensible and valuable if written feedback can be delivered in a way as to both remedy and prevent errors in L2 writing. Besides that, it is noted that insufficient attention has been granted to the inherent shortcomings of written corrective feedback. For example, compared with oral feedback, the delivery of written corrective feedback always lags in time, which might compromise its effectiveness in promoting L2 writing development. Such "bugs" should be circumvented in future research so as to boost the effectiveness of WCF.For research on second language writing development, it leaves much to be desired in two aspects, according to our observation. Firstly, out of practical concerns, most studies available in the literature choose a cross-sectional design rather than a longitudinal one. Within the small number of longitudinal studies, most of them turn out to be case studies and therefore their generalizability is compromised. Instead of the advantages of being time and resource efficient, cross-sectional designs blur and even ignore inter-subjects variations and thus likely jeopardize the validity of the research findings. So there is an urgent need for more longitudinal studies in the real sense to generate more robust findings about L2 writing development. Secondly, the majority of development studies, if not all, are disassociated from the real pedagogical treatment, which will surely curtail the practical value of the research itself. In other words, little regard is afforded to the influence of specific types of instruction over L2 writing development. The disintegration of research and instruction may ultimately take its toll on the values of the investigations so far. Now it becomes clear that L2 writing development still calls for more insightful and more fine-grained empirical studies in order to guarantee the reliability and validity of the findings. And undoubtedly, these problems or deficiencies should be circumvented or overcome before a new trail can be blazed in the research on written corrective feedback and L2 writing development.On the basis of a clear understanding of the deficiencies of previous investigations, the current study proposes a brand new type of instructional treatment named ’joint production’which aims to improve learners’linguistic performance in L2 writing. Embedded organically in second language classrooms, joint production fulfills form-focused instruction by integrating oral feedback and written feedback and fully mobilizing teacher-learner and peer interactions in class. Besides, as a form of feedbacking, it distinguishes itself from other regular types in that by nature, it is both preventative and remedial on one hand and it conforms to the principle of "here and now" on the other. It works on learners’learning strategies and on their awareness to language quality in L2 writing so much so that their ability to control language use in writing is improved. Endowed with these features conducive to language development, joint production is hypothesized to exert positive influence over the linguistic performance of L2 writers. In addition to this pedagogical concern, the current research aims to explore L2 writing development from a multi-dimensional perspective as well. Specifically, this query of the current research can be boiled down to a couple of research questions below:Which dimensions of L2 writing are likely to develop over time? Are there interactions among the dimensions? The first question addresses the longitudinal development of various dimensions of L2 writing; the second seeks to untangle the controversies over the relationship between human cognition and language performance and that between syntactic complexity and lexical complexity among second language learners.Methodologically, the current study is empirical by nature and adopts a quasi-experimental design. All participants come from two intact classes of approximately 30 students each in a Chinese university. The experimental group receives joint production as an instructional treatment while the control group does not. Except for the difference in treatment, the two groups are no different from each other in terms of the courses taken, reading and writing assignments, and more importantly, English language proficiency. The study lasts 19 weeks and altogether 14 sessions of joint production are delivered. In all, nine compositions are collected from each participant from both groups. These writings are first coded and quantified in terms of accuracy, syntactic complexity, fluency, and lexical complexity before the data are fed into SPSS for further analysis. Besides the compositions, other types of quantitative data are also collected via questionnaires and a paper-and-pencil language proficiency test on learners’ motivation, learning strategies, working memory capacity, language analytic ability, and language proficiency levels. In addition to the quantitative data, some qualitative data are collected by means of videotaping and interview so that the evidence can be triangulated to support the findings. Statistical analysis of the data undergoes three stages (1) to establish the effectiveness of joint production, (2) to explore the individual difference factors mediating the effectiveness of joint production, and (3) to outline the rules and trajectories of L2 writing development.The research has generated some important and interesting findings. Firstly, there is evidence to show that as an instructional treatment, joint production is effective in promoting second language writing development. To be specific, learners receiving joint production find significantly more progress in accuracy and lexical diversity of L2 writings than those who do not. However, no similar effect is detected in the other dimensions of L2 writing such as syntactic complexity, fluency, and lexical sophistication. Secondly, the effectiveness of joint production is mediated by such individual factors as learning strategies and language proficiency. However, the effect size is much smaller than those reported in the WCF-related literature. Thirdly, L2 writing development is not linear all the time. Instead, it is characterized by progress, regressions and fluctuations. Developmentally, learners progress significantly along all the dimensions of writing except for syntactic complexity and lexical sophistication. In addition, accuracy is found to be negatively correlated with structural complexity, but positively correlated with fluency, which lends partial support to the hypothesized trade-off effect in the Limited Capacity Hypothesis proposed by Skehan (1998). Another interesting finding is that there is a negative correlation between structural complexity and lexical sophistication, though not to a significant level. This shows that different from native speakers, nonnative speakers find it hard to be syntactically and lexically elaborate and complex simultaneously in L2 writing.Some conclusions can be drawn on the basis of the above findings. Firstly, joint production is indeed a highly effective feedbacking strategy to promote second language writing development. Its effectiveness can be expounded by the inherent features of joint production:the integration of oral and written corrective feedback, the "here and now" principle of feedback delivery, the accommodation of learner needs, and the preventative and remedial nature of error treatment. Secondly, individual differences play a limited role in mediating the effectiveness of joint production. It follows that joint production can cater to the needs of different learners. Thirdly, L2 writing development is never smooth or linear. It is characterized by an upward trend with constant fluctuations and a competition between language accuracy and structural complexity.Experimental by nature, the current study is limited in several ways. The first point is concerned with the duration of the study. As it takes much time for writing to develop, the time of observation had better be lengthened to a full year or longer so as to generate more robust findings about L2 writing development. Besides that, the research scope can be broadened to include speech which might also benefit from joint production and learners of different proficiency levels other than the intermediate learners in the current study. The last point is concerned with the choice of measures in the study. Some measures of language performance turn out to be very tricky and even unreliable. For example, W/T (the number of words per T unit) is more likely a measure of structural complexity than of fluency, as is revealed in the analysis section. Therefore, such measures had better be replaced. These are not only the limitations of the current research but also future directions for related investigations. It is expected that more research can be undertaken to confirm the effectiveness of joint production before it is promoted to the real classrooms and to enhance the understanding of L2 writing development of SLA researchers and teachers. |
PDF Full Text Request