The root and word distinction: an experimental study of Hebrew denominal verbs

Abstract

The morpho-syntactic structure of Semitic languages, traditionally seen as based on abstract root morphemes, has been analysed by some as being fully word based. Others have proposed a root-based system which allows for word-based derivation as well. A distinction between word-based and root-based morpho-syntactic derivations has previously been posited in both morpho-syntactic and lexical semantic literature. Under this distinction the semantic and phonological access to a root morpheme during morpho-syntactic construction is fully available in a root-based derivation, but is restricted by the category bearing head in a word-based derivation. However, there has to date been no behavioural evidence for the distinction, and it is by no means universally accepted that words are morphemically decomposable into root morphemes. The current study utilized a masked priming experimental paradigm of word recognition in an attempt to differentiate root-based derivation from word-based derivation in Hebrew, proposing an analogy between availability under Marantz (2000) and Arad (2003), and linkage within the mental lexicon model under Frost et al. (2005). The results strongly support the proposed analogy, and the cognitive reality of the root morpheme as the basis of Hebrew morphological derivation. In addition they provide a first experimental verification of the theoretically motivated distinction between root derivation and word derivation.