This paper explores the sources of difficulties that second language (L2) learners encounter when using English articles. Eighty-four Korean college students completed a forced-choice elicitation task before and after receiving instruction on article use and provided written accounts of article choices. The analysis of the task performance and written accounts indicated the participants’ noticeable tendency to prioritize specificity over definiteness, resulting in the overuse of the with specific indefinites. Not infrequently, the participants estimated a “nonspecificity hierarchy” for nonspecific definites, often leading to the infelicitous use of a(n). The overuse of the with modified noun phrases suggests that L2 learners attempt to construe semantic context (i.e., ±definite) on the basis of the syntactic structure. Furthermore, the participants’ correct use of a(n) for singular count indefinites sometimes stemmed from assuming the number of a target noun to be single rather than considering its multiple existence and, thus, its indefinite nature. These findings underline the necessity of teaching the specificity feature to indicate to learners that (1) English articles are prototypical realizations of encoding definiteness, which requires the mutual identifiability of a unique referent, and (2) specificity, which presupposes identifiability assumed by the writer/speaker alone, is not marked by articles in English.

中文翻译：

---

“当具体是不够的”：二语学习者对确定性的感知与其语言定义之间的差异

本文探讨了第二语言 (L2) 学习者在使用英语文章时遇到的困难的来源。84 名韩国大学生在接受文章使用指导前后完成了一项强制选择启发任务，并提供了文章选择的书面说明。对任务表现和书面记录的分析表明，参与者明显倾向于优先考虑特定性而不是确定性，导致过度使用带有特定不确定性的。不常见的是，参与者估计了非特异性定义的“非特异性层次结构”，通常导致 a(n) 的不合理使用。带有修饰名词短语的过度使用表明二语学习者试图在句法结构的基础上解释语义上下文（即±确定）。此外，参与者正确使用 a(n) 表示单数不定数有时源于假设目标名词的数量是单一的，而不是考虑它的多重存在，因此，它的不定性质。这些发现强调了教授特异性特征的必要性，以向学习者表明（1）英语文章是编码确定性的原型实现，这需要唯一所指对象的相互可识别性，以及（2）特异性，以作者假设的可识别性为前提/speaker 单独，没有用英文文章标记。