This paper seeks to use a corpus-based analysis of assessment commentaries on Master's level assignments to shed light on the guidance practices of those who provide feedback. The analysis offers a set of functional categories that emerge from the corpus and uses these to consider the degree of transparency evident in the commentaries. Based on this analysis, the paper discusses implications for feedback providers and offers suggestions for diminishing power imbalances and for re-placing the student writer at the centre of academic discourse. In doing so, I hope to enhance the value placed on individuals' academic contributions and facilitate the process of induction into the academic discourse community, through a notion of critical inclusion, as opposed to a prescriptive notion of convention adoption.