The purpose of this paper is to apply theEntrenchment-and-Conventionalization Model(EC-Model hereafter; seeSchmid 2014,2015,2016,2017,2018;Schmid & Mantlik 2015) of language knowledge to genre, with the aim of showing how a unified theory of the relation between usage and linguistic knowledge and convention can shed light on the way genre knowledge becomes entrenched in the individual and shared conventional behavior in communities. The EC-Model is a usage-based and emergentist model of language knowledge and convention rooted in cognitive linguistics and usage-based approaches. It sees (...) knowledge as emerging from language usage, and explains the processes underlying the intertwining of social practice and cognition. However, so far, no suggestion has been advanced on how to extend the model to account for entrenchment and conventionalization at the supra-sentential level. In the area of genre studies various attempts have been made by scholars to develop or apply theories belonging to different scientific domains to understand the nature of genre. However, so far, there has been no research that applies a unified model in the attempt to link entrenchment of genres in individuals to their conventionalization at the societal level. I largely focus on the long tradition of rhetorical studies of genre, one among the different approaches that, over time, have regarded genre as their main topic of investigation. I concentrate on this tradition as it opens up the entire field of enquiry that defines contemporary genre research. To these I add by showing how the explanations provided so far can be cognitively clarified and unified under the EC-Model. The paper, then, argues that the EC-Model is theoretically apt to address questions about the nature of genre, capturing in an elegant way the interplay between cognition and social interaction in genre emergence, evolution, stabilization and variation. (shrink)