| One of the fundamental, self-evident differences between L1 and L2 writers is that L2 writers, especially those of low L2 proficiency, may rely at times or extensively on their L1 for cognitive operations. How do L2 writers use their L1 in the L2 writing process? And what effects will the use of L1 exert on the writing process as well as the written product? These questions are largely unanswered. In the first place, research into L1 use in L2 writing is sparse. An overview of the L2 writing research shows an overwhelming emphasis on the similarities between L1 and L2 writing processes and behaviors as well as a wholesale borrowing of L1 writing theory to account for L2 writing (Krapels, 1991).Consequently, differences between L1 and L2 writers are not adequately addressed. In the second place, studies and discussions on the role of L1 in second language acquisition have long perceived L1 as either positive or negative. Only very recently, research began to reveal a multifaceted role of L1 in L2 reading comprehension and suggests a dynamic, multidimensional approach to the study of L1 influence (e.g., Kern, 1994). As for how L1 facilitates and/or hinders L2 writing, empirical findings are still inadequate.The impact of writers' L1 is an added factor for L2 writing, which is not present in L1 writing (Grabe & Kaplan, 1996). It is assumed that revelations of how L1 influences L2 writing will promote our understanding of the distinct nature of L2 writing and ultimately contribute to the development of L2 writing theories. The present study makes such an attempt. It addresses the use of L1 in L2 writing, one of the fundamental, unique features of L2 writing. In order to reveal clearly the bilingual nature of the L2 writing process, detailed depictions are made of how the writer's L1 and L2 are used in various composing activities. The study also examines the effects of writers' L2 development to see whether the bilingual nature of the L2 writing process varies with writers' L2 development. In addition, it specifies various functions L1 performs in different composing activities and determines the effects of L1 use on various aspects of the composition (i.e., language, organization and content) so as to find out when andwhere L1 faciL1tates/hinders L2 writing.The subjects for this study are thirty Chinese university students of EngL1sh as a foreign language. Twenty-one of them are female students and nine of them are male. Data in this study are student compositions and verbal reports of these student writers' composing processes in two experiments, collected through a combination of research techniques such as thinking aloud, observation, and interviews. The subjects, having been trained to think aloud, are asked to compose aloud on two writing tasks, with six months in between. The subjects are also interviewed when they finish each task on their views about L1 use in L2 writing.Then analyses were made on data provided through various sources such as the verbal reports, student compositions and interviews. The reports are analyzed both quantitatively and quaL1tatively. First, L1 and L2 words in each report are counted and the proportion of L1 use is computed to measure the amount of L1 in L2 writing. Then, the writer's cognitive operations were categorized into three composing activities (i.e., what to write, how to write, and other functions) and the proportion of L1 use in each activity was calculated to show when the writer tends to rely on L1. Moreover, in order to measure whether different levels of L2 learners rely on their L1 in different aspects, differences between high achievers and low achievers in text-generating in "how to write" are analyzed.The composition-based analysis consisted of three steps. First, the student compositions are rated; then, a correlation procedure is appL1ed to explore the relationship between L1 use and L2 writing quaL1ty; and finally cases are selected for a closer examination of this relationship.The major findings of this study are summarized as follows:(1) The L2 writing process is a biL1ngual event: L2 writers have two languages (i.e., L1 and L2) at their disposal when they are composing in L2. The L1 thinking occupies a considerable proportion. The tendency of L1 occurrence varies with individual composing activities.(2) The amount of L1 use in L2 writing may vary with writers' L2 proficiency. The effect of the writer's L2 development on L1 use is more compL1cated thanexpected. Though the proportion of L1 thinking decreases with the writer's L2 development, the extent of the decL1ne of L1 use in individual activities varies.(3) The role of L1 in L2 writing is not invariable. The role of L1 is different in the thinking activities of L2 writers of different proficiency levels. Reasons for using L1 in L2 composing are more diversified than expected: some writers have to depend on their L1 due to their L1mited command of L2; some with the abiL1ty to think in L2 choose to turn to their L1 for practical purposes; and others, triggered by L1 memories, revert to L1 unconsciously.It is hoped that findings of this study will contribute to the development of L2writing theories and a deeper understanding of the role L1 plays in secondlanguage acquisition. |