Our understanding of the history and development of Arabic is greatly challenged by the dearth of sources. Most scholars who have undertaken this task in the past have relied almost exclusively on data from the Arabic grammatical tradition. These sources claimed that Arabic originated as the language of the Bedouin of the central and southern Arabian Peninsula, and was only introduced to the settled areas of the Levant and Mesopotamia following the Islamic conquests. In this dissertation, I argue that there is no scientific basis for this view. The first chapter of part one of the dissertation begins with an examination of the evidence for Arabic in the Fertile Crescent from the pre-Islamic period. This evidence suggest that Arabic was part of the linguistic milieu of the Levant and Mesopotamia as early as the Iron Age. Next, I discuss the Arabic grammatical tradition and how it has influenced the production of modern theories regarding the history of Arabic.