© Koninklijke Brill NV, Leiden, 2012 DOI: 10.1163/18776930-00400011 Brill’s Annual of Afoasiatic Languages and Linguistics 4 (2012) 232–251 brill.com/baall Te Assignment of Gender in L2 Hebrew: Te Role of the L1 Gender System Sharon Armon-Lotem and Orit Amiram Department of English, Bar-Ilan University sharon.armon-lotem@biu.ac.il and oritile@gmail.com Abstract Grammatical gender poses a serious problem to second language (L2) learners as well as to proficient speakers of L2. Tis paper tests what contributes to this difficulty in L2 Hebrew. Gender identification in the absence of lexical information was tested for 30 Hebrew L2 learners and 20 Hebrew near-native speakers whose L1 has gender morphology, e.g. Russian, or not, e.g. English, as well as 10 adult native speakers of Hebrew. Te participants were tested on assigning grammatical gender to novel animate nouns in Hebrew. Te findings show that difficulties are due to L1 interference, by indirect reliance on L1 strategies in determining gender for novel animate nouns in L2, rather than access to Universal Grammar or across the board reliance on native adult strategies. Keywords agreement; grammatical gender; morphology; second language acquisition; Hebrew 1. Introduction Hebrew grammatical gender poses a problem to second language (L2) learn- ers as well as to L2 proficient speakers, but not to first language (L1) acquirers. Native speakers of languages that have morphological gender and speakers of lan- guages that do not have morphological gender, all find it difficult to use gender when acquiring Hebrew as a second language. Tis paper aims at clarifying what contributes to this difficulty. How does natural gender interact with morpho- phonological gender in determining the grammatical gender of nouns in L2? Do L1 features which are transferred into L2 influence L2 learners in acquir- ing this system, or is it all about learning individual lexical items? Does Univer- sal Grammar (UG) play any role in the process, and if so, what is this role? Can parameterized features be reset? As far as L1 acquisition is concerned, it has been found that in gender assignment to novel animate nouns young children rely on morpho-phonological cues. By contrast, adults, when tested in their L1, rely on semantic information (Amiram 2002, Karmiloff-Smith 1978, Levy 1980). Pre- vious studies have found that adult L2 learners give priority to morphological