The purpose of this study is to investigate the written errors made by native English speakers studying intermediate Arabic. The entire corpus consisted of 196 errors taken from journal entries. As a first step, the errors were classified according to general linguistic category, of which four emerged: lexicon, morphology, syntax, and use. Then, the errors were divided into traditional grammatical category and function: adjective, adverb, idafa construction, noun, participle, particle, prepositional phrase, superlative, verb, verbal noun, and verbal phrase. The majority of errors, 82.7%, were syntactic.
Agreement errors amounted to 37.24% of the entire corpus. Three types of agreement emerged: definiteness, case, and gender.
With regard to idafa construction errors (8.16% of the errors analyzed), three types surfaced: the first term being made definite, the second term being made definite, and the second term made indefinite. In addition, three underlying causes became apparent: hypercorrection, language transfer from English to Arabic, and use of the wrong expression. Even though language transfer involving the first term being made definite amounted to 68.75% of the relevant errors, statistics show that it does not play a significant role.
Errors involving the use of the wrong preposition amounted to 6.12% of the corpus. The misused prepositions involved [?an], [bi], [fi], and [ma?a]; the latter occurred six times (50% of the preposition errors). Language transfer was primarily responsible for these errors. Case errors comprised 16.84% of the entire corpus of errors. Four factors were found to influence them: language transfer between English and Arabic, students' lack of knowledge about target language structures, students' lack of knowledge about grammatical structures in general, and miscellaneous noun-adjective agreement. Language transfer does not occur often enough to be considered statistically significant.