Sensitivity to Parasitic Gaps Inside Subject Islands in Native Speakers of English and Najdi Arabic Learners of English

This study examined second language (L2) learners’ sensitivity to parasitic gaps (PGs) in English. In PG constructions, an unacceptable gap inside an island becomes acceptable when it is combined with an acceptable gap (e.g., whati did [ISLAND the attempt to repair __pg] ultimately damage__i?). Linguists have often viewed PGs as marginally acceptable by native speakers, but recent studies have shown that they are fully acceptable. This phenomenon, however, has received little attention in L2 research. Thus, this study’s purpose was to test sensitivity to PGs in L2 learners. In an acceptability judgment task, native speakers of English (n = 32) and Najdi Arabic learners of English (n = 38) used a 10-point scale to rate their acceptability of wh-questions with PGs, wh-questions with gaps inside subject islands, and wh-questions with gaps inside non-island structures. Like native speakers of English, Najdi Arabic learners of English rated wh-questions with PGs not only more acceptable than they rated ungrammatical wh-questions with gaps inside subject islands but also as highly as they rated grammatical wh-questions with gaps inside non-island structures. These results suggest that PGs are fully acceptable by Najdi Arabic learners. The Najdi Arabic learners’ sensitivity to PGs in English supports the Full Transfer/Full Access Hypothesis (Schwartz & Sprouse, 1996), which claims that advanced adult L2 learners can acquire L2 properties regardless of L1.