This chapter is dedicated to the contribution of the theoretical review of Costa, Pickering, and Sorace (2008) to our understanding of alignment in second-language dialogue, or, more broadly, conversations involving at least one less proficient speaker. We first briefly describe the review’s foundation, the Interactive Alignment Account, and some of its theoretical considerations. We then highlight the main hypotheses put forward by Costa et al. (2008), following each with subsequently obtained relevant evidence. As proposed by these authors, the emerging picture is indeed one in which second-language dialogue is characterized by enhanced non-automatic alignment and, in some cases, reduced automatic alignment, with important roles of audience design, second-language proficiency, and modeling behavior. While recent advances have been made in this area, some of Costa et al.’s (2008) hypotheses have not been tested, and some have produced conflicting evidence; we thus end with a list of research questions aimed to inspire future work on this topic.