Lexical analysis can be a way to remove ambiguity in Arabic text. Resolution of ambiguity is an important task in several Natural Language Processing (NLP) applications. Our proposed resolution method is based essentially on the use of transducers on text automata. Indeed these transducers specify lexical and contextual rules for Arabic. They allow the resolution of lexical ambiguities. Different types of lexical ambiguities are identified and studied to extract an appropriate set of rules. After that, we describe lexical rules in the ELAG system (Elimination of Lexical Ambiguities by Grammars), which can delete paths representing morphosyntactic ambiguities. In addition, we present an experimentation implemented in the Unitex platform with various linguistic resources to obtain disambiguated syntactic structures suitable for parsing. The obtained results are ambitious and can be improved by adding other rules and heuristics.