Most research on gender and foreign language textbooks has focused solely on texts, independent of the use to which they are put. And though the implication is usually that gender bias in texts will have an effect on learners' gender identities and/or language-learning opportunities, we suggest that it is in fact impossible to predict the “effect” of a given text. One thing that seems to have been missed in these studies is an exploration of teacher behavior in relation to textbook texts, what might be called “teacher talk around the text.” This mediation of the text may be more relevant to any “effect” of the text than the text itself. To investigate this, a study was carried out which drew on concepts underpinning literacy studies and critical discourse analysis. Data was collected in three separate investigations, in three foreign language teaching settings. The focus in each investigation was a set of lessons in which the textbook covered some aspect of gender. Using lesson transcripts, as well as the teachers' own perceptions of their behavior, we show how teacher talk cannot be predicted from the textbook text itself. In particular, our data shows cases of the “traditional” focus of texts being endorsed, “progressive” texts being undermined, those parts of texts that can be seen as containing “traditional biases” being passed over, and talk about gender issues within texts being characterized by “gender blindness.” Drawing on these possibilities, we propose a working analytical framework, which can be used in and developed by future studies. We conclude that since textbooks themselves may be less sexist than they used to be, one particularly useful focus of research might now be teacher treatment of “progressive” texts.