The first chapter of my dissertation is about political dogwhistling [the title is redacted to comply with anonymous peer review regulations]. Here's the abstract:

In this paper I argue that a necessary (but insufficient) condition for something to be a dogwhistle is that an utterance exhibits favoritism toward a social ingroup, and a sufficient condition is that it does so covertly. In addition to this, the main argument of this paper is that dogwhistles that covertly exhibit favoritism sometimes also encourage favoritism within that social ingroup. I argue that two paradigm instances of dogwhistling work this way: the “Willie Horton” ad campaign and political criticisms of the “inner city”. The literature has recognized that dogwhistling often involves ingroup dynamics, but the focus has been on how dogwhistling hides something from an outgroup. This overlooks how dogwhistles can be so powerful: they can also hide something from the ingroup, namely their own ingroup bias. In short, the literature has not recognized how dogwhistles induce and encourage ingroup thinking in a targeted audience without their knowledge. 

I'm always grateful for feedback; please email me if you'd like a copy.

The rest of my dissertation argues that crucial parts of human communication rely on awareness of contextual social norms.