Literature reviews play an important role in the development of knowledge. Yet, we observe a lack of theoretical underpinning of and epistemological insights into how literature reviews can contribute to knowledge creation and have actually contributed in the IS discipline. To address these theoretical and empirical research gaps, we suggest a novel epistemological model of literature reviews. This model allows us to align different contributions of literature reviews with their underlying knowledge conversions - thereby building a bridge between the previously largely unconnected fields of literature reviews and epistemology. We evaluate the appropriateness of the model by conducting an empirical analysis of 173 IS literature reviews which were published in 39 pertinent IS journals between 2000 and 2014. Based on this analysis, we derive an epistemological taxonomy of IS literature reviews, which complements previously suggested typologies.