To date, no convincing explanation of the historical development of the internal passive verbs of Semitic has been proposed. This dissertation avoids obstacles that have impeded previous scholarship by viewing them in their broad relational context of detransitive stems more generally, and by adapting a novel framework for understanding language diachrony from the discipline of sociolinguistics. Internal Passives are found to have resulted from the extension of a voice-marking height contrast that first appeared in participle forms, and to have arisen as variants to fill out gaps in the verbal paradigm, thereby creating a more functionally symmetrical system. Their history is largely one of extended (and usually abortive) competitive variation with other constructions, rather than language change, properly speaking. Apart from the immediate subject matter of internal passives, the dissertation highlights several competing motivations at work elsewhere in the development of Semitic verbs and illustrates how the variation-and-change model may be applied to historical Semitic languages in contexts where other theoretical frameworks have not proved fruitful.