How many nouns can a verb take, and how does their choice influence the meaning, morphology, and syntax of the verb? What about cases that are only expressed in prepositional phrases? Such questions are usually addressed in terms of predicate and arguments, and they need to be addressed in Hebrew grammar as well. This paper will briefly introduce the linguistic work of Beth Levin and Malka Rappaport Hovav published in Argument Realization (2005) and will illustrate the problem of argument realization for Hebrew verbs such as put and give. It will then bring more traditional approaches to Hebrew valency into play by discussing Michael Malessa's Untersuchungen zur verbalen Valenz im biblishcen Hebräisch (2006) and will relate his findings to the issue of argument realization in current lexical semantics. The paper will then present findings from a particular structuralist-functional linguistic theory which is especially well-suited to help linguists categorize verb classes and their argument alternations as an interplay between syntax and sematics. This theory, Role and Reference Grammar, will be used to show how its semantic metalangue can assist in explaining argument realization. The paper will share some solutions to questions that are usually dealt with as transitivity in traditional grammars. It will also address the question whether a new approach could give us a better understanding of the the verbal stems in Hebrew. This paper builds upon results gained from a research project on a Role-Lexical Module for Biblical Hebrew which was used as a tool to map Hebrew syntax to semantic representation (see http://lex.qwirx.com/lex/clause.jsp).