The aim of this paper is to provide a broad overview into studies that have focused on the acquisition of articles (the and a) and to tell the story of article acquisition, from early first language (L1) studies to recent second language (L2) studies. The overview specifically examines the possible cause of difficulties for L2 learners by focusing on different areas of core grammar (syntax, morphology, phonology and semantics) and how core grammar interacts with discourse / pragmatics.

This paper presents second language (L2) studies of bare plural DPs as well as a set of related linguistic phenomena that present potentially fruitful avenues for exploration in second language acquisition (SLA). So far, DPs have been at the centre of L2 studies in the nominal domain. The focus in L2 studies has revolved around some well-known properties such as article use, agreement features and the mass/count distinction. Shifting the focus away from these properties for a moment, in this paper, we will present some relatively less well-established properties associated with bare plural DPs in L2 acquisition. We first start by looking at how BPs, in their generic use, can offer an insight into L2 learners' sensitivity to lexical semantic constraints. Then we take on the notion of plurality — do plural nouns denote necessarily a plurality of things (i.e., more than one)? In English, it seems not, and this has been a topic of intensive debate in semantics in recent years (Sauerland et al., 2005; Farkas & de Swart, 2010; Kane et al., 2015). Following a recent study on this question by Renans et al. (2018), we show how a similar issue can be exploited in SLA. This can offer a new way of testing L2 learners' knowledge of plural DPs amid the mass/count distinction. In turn, this will also have implications for SLA and L1 transfer among obligatory vs. optional plural marking languages.

This article explores the effect of the count-mass distinction in English article acquisition and argues for its importance in the research of second language (L2) article acquisition. While L2 article use has extensively been discussed in terms of semantic contexts (i.e., definiteness, specificity) and L1 effects, such factors as input and lexical information of nouns have not been given much attention in the literature. The article first overviews the previous research based on semantic contexts and points out the L2 phenomena which cannot be attributed solely to definiteness or specificity. Turning to the data of input frequency and cues, such phenomena seem to be plausibly explained by the frequency distribution of noun types and forms in the input. Next, the article argues that the count-mass distinction poses a more persistent problem than definiteness, by examining to what extent each feature involved in article choice (e.g., [±definite], [±count], [±plural]) contributes to L2 learners' use of articles. After claiming the prolonged difficulty with the count-mass distinction, the article addresses acquisition issues of abstract nouns, for which L2 learners often have trouble choosing correct articles, especially indefinite ones (a/an or ø). An experiment was set up to investigate whether or not the lexical- semantic property of boundedness and/or the derivation type of verb-derived abstract nouns were potential factors to influence the countability of nouns. As a result, the two factors did not categorically affect the countability judgment of abstract nouns, rather indicating a possibility of L1 influence to be further investigated. At last, the article attempts to verify the claim that L2 learners, especially with low proficiency, determine the countability of nouns by intuition without considering contextual information. The comparison of L2 learners' intuitive countability judgment of decontextulised nouns with the use of articles in context revealed that a majority of learners demonstrated no such correlation between countability intuition and article use. Although there were some who showed reliance on intuition, their L2 proficiency was not the discriminative factor to identify what kind of learners would strongly draw on intuition for countability judgment of nouns. In conclusion, this article emphasises that the effect of nominal properties, which are susceptible to L1 influence, can interact in the use of articles and require careful empirical scrutiny in order to grasp a whole picture of L2 article acquisition.

This paper tests the predictions of the Article Choice Parameter (Ionin, Ko, & Wexler, 2004) against individual data from Japanese child second language (L2) learners of English. Ionin et al. (2004) and Ionin, Zubizarreta, and Philippov (2009) proposed the Fluctuation Hypothesis for explaining L2 learners' choice of English articles. Ionin et al. (2009) argued that Russian child L2 learners of English can access the Article Choice Parameter but they fluctuate between the two settings, definiteness and specificity. In a replication study, Yamada and Miyamoto (2010) showed that Japanese child L2 learners of English incorrectly chose definite article the for a [-definite, +specific] context in their L2 English, as found in previous child L2 acquisition studies (Zdorenko & Paradis, 2008). However, Hawkins et al. (2006) and Trenkic (2008) point out that the status of the Article Choice Parameter as a theoretical construct and the validity of it are unclear. As a result, the current study re-examined child data in Yamada and Miyamoto (2010) by focusing on the individual results rather than focusing on group results. We reveal that the children's article interpretation actually varied, which the Article Choice Parameter cannot explain sufficiently. We show how a feature-based, the Distributed Morphology account (Halle & Marantz, 1993; Harley & Noyer, 1999), could clarify the process of acquiring the correct composition of features relating to English articles.

In a sentence like Tom ate an apple/apples, (a)telicity of the event expressed by the verb phrase is determined by number and definiteness features of the object. This article discusses the second language (L2) acquisition of such formal features in terms of L1 transfer and semantic/pragmatic computation, based on the interpretation of telicity by Japanese learners of English (JLEs). It is known that JLEs find it difficult to interpret telicity of a sentence with definite plurals like the apples (Kaku, 2009; Kimura, 2014; Wakabayashi & Kimura, 2018). It has also been shown that elementary JLEs are insensitive to number and definiteness features in interpreting telicity (Kimura, 2014; Wakabayashi & Kimura, 2018). Previous studies have explained these facts in terms of computational complexity with definite plurals or the absence of relevant functional categories in the developing interlanguage grammar. Highlighting problems with those studies, this article offers an alternative proposal: Following Filip (2008), who proposes that telicity is derived by scalar implicature, it is argued that the insensitivity to formal features by elementary learners arises from a failure in the application of scalar implicature. Further, it is also argued that the difficulty intermediate learners have with the interpretation of definite plurals lies in lexical learning of formal features (definiteness) because of the lack of the lexical item the in their first language (L1).