When a bilingual speaker has a larger linguistic sub-system in their L1 than their L2, how are L1 categories mapped to the smaller set of L2 categories? This article investigates this “subset scenario” (Escudero, 2005) through an analysis of laterals in highly proficient bilinguals (Scottish Gaelic L1, English L2). Gaelic has three lateral phonemes and English has one. We examine acoustics and articulation (using ultrasound tongue imaging) of lateral production in speakers’ two languages. Our results suggest that speakers do not copy a relevant Gaelic lateral into their English, instead maintaining language-specific strategies, with speakers also producing English laterals with positional allophony. These results show that speakers develop a separate production strategy for their L2. Our results advance models such as the L2LP which has mainly considered perception data, and also contribute articulatory data to this area of study.

当双语使用者的 L1 语言子系统比 L2 语言更大时，L1 类别如何映射到较小的 L2 类别集？本文通过对高度熟练的双语者（苏格兰盖尔语 L1、英语 L2）的侧向分析来研究这种“子集情景”（Escudero， 2005）。盖尔语有三个横向音素，英语有一个。我们检查了说话者两种语言中横向产生的声学和发音（使用超声舌成像）。我们的结果表明，说话人不会将相关的盖尔语侧音复制到他们的英语中，而是保持特定于语言的策略，说话人也会产生具有位置同位异音的英语侧音。这些结果表明，说话人为他们的 L2 开发了单独的生产策略。我们的结果推进了 L2LP 等模型，该模型主要考虑了感知数据，也为这一研究领域提供了发音数据。