Over the last seven years, I have spent time across three continents talking to scientists and mathematicians about their beliefs and attitudes and experiences related to writing in their respective disciplines.I have been impressed by the passion and insight with which most have talked about writing and its relationship to critical thinking, and I have often been surprised by how they engage in these practices.For example, rather than working from an a priori hypothesis, many researchers in the STEM disciplines compose backwards, from the results to the introduction.And when reading, many seem to move from the middle of a paper outwards, beginning with the results and method, using an extremely critical eye, and then perhaps scanning out to the introduction and the discussion, or dispensing with these sections altogether.Over and over again, I heard this same story from different scientists, as if it were a secret each alone had stumbled upon.In addition, collaboration, conversation and peer review are very much part of the language of composition that takes place in the sciences (co-authorship, the hierarchies of disciplinary or interdisciplinary teams, the drafting process and the use of technology), but we who work in WID (writing in the disciplines) and WAC (writing across the curriculum) programs are constantly challenged: "How do we teach process in ways that are disciplinarily appropriate?"Historically, we haven't done this well.As Burton and Morgan observed on the training of mathematicians as writers,