In Hebrew VS constructions with unaccusative verbs, the verb and its internal argument sometimes fail to exhibit φ-agreement. This results in 3M.SG (default) morphology on the verb, while the internal argument may be feminine, plural, or both. Furthermore, in several cases, the internal argument may be introduced by the direct object case marker. It has been claimed that what allows this lack of agreement is the structural intervention of a possessive dative between the verb and its postverbal subject (Preminger, 2009). However, lack of φ-agreement with unaccusatives in VS order is attested in corpora even without intervention. To which extent is the intervention required for lack of φ-agreement? Can the acceptability difference between lack of φ-agreement with and without intervention arising from informal judgments, be corroborated by experiments? Should this difference sensed by speakers be interpreted as difference in grammaticality, as argued by Preminger? What is the nature of this intervention? How frequent is the direct object marker in these constructions and how should they be analyzed? That is, what licenses a direct object marker introducing the post-verbal argument?
Hebrew has two types of N+N constructions: compounds and construct state nominals. These constructions differ from each other in their level of cohesion, productivity and syntactic structure, inter alia. Are the syntactic differences among these constructions reflected in different prosodic structures? Will these constructions differ from, say, N+A constructions? And does frequency play a role in the prosodic structures both within and between these construction groups?
Traditionally, one of the main characteristics of Semitic languages that differs them from other language families has been their non-concatenative morphology. As a Semitic language, Hebrew is known for its root-and-pattern morphology, meaning that words are composed of three or four consonants, referred to as root, combined in a vocalic pattern, often with additional affixes. Even though roots are formally taught from second-grade on, the question remains as to why students demonstrate low achievements in this subject, even in high school, and whether this has to do with the instructional methods adopted by the education system. In order to detect the weak spots in the process of root extraction, we carried out an intervention study in which we incorporated a five-lesson online root instruction program. We examined the program’s effects on children’s morphological awareness by evaluating their knowledge prior to and after the program. Then, we analyzed the children’s responses and concluded the factors that made root extraction more difficult. Pattern, either verbal or nominal, was found to be the best predictor of failure, while the second and the third predictors were the presence of the infix vowel i and internal-prefixation, accordingly.
Child’s play: teaching young children Hebrew morphology using e-learning technologies