As courses that once regularly met face-to-face are increasingly offered in blended or fully online formats, questions have been raised about the comparability of instruction, teacher and student roles, and overall quality of the learning experience. The present study offers a five-inference interpretive argument (Chapelle, Enright & Jamieson, 2008a; Kane, 2006) that may be used to guide the design and implementation of blended language curriculum and instruction in which tasks are viewed as micro-formative assessments. An example compares the language practiced and produced by students during three language tasks that were carried out both online and in a classroom setting. Finally, suggestions to guide the gathering of validity evidence for blended language course design are offered.