to do or to know

Back in the days of my Master’s Program in Social Justice Education, we spent a great deal of time studying how to facilitate our own, each other’s, and student’s growth along the continuum of social identity.
The core model in the program was Jackson and Hardiman’s (1992, 1997) model of racial social identity based upon “white” and “black” identification in the US context. A social identity model (SID) provides a paired rubric for processes that individuals undergo as members of the “target” or “oppressed” group who seek to become empowered, as well as the processes that individuals take on if they want to understand themselves as members of the “agent” or “oppressor” group. That model has been adapted to apply to many other “isms,” including for instance, sexual orientation. My emphasis tends to the “agent” side of the pairing – how do people who are members of dominant cultures come to grips with the reality of privileges (access to resources and ways-of-being which are not equally available to members of non-dominant groups) and the fact of unconscious collusion with systems of discrimination and prejudicial beliefs that work together to keep oppression real?
My own interests have moved beyond the US domestic context to inter- and transnational issues involving migration and especially the role of language in empowerment processes – those that enable individuals to develop agency and assert voice. I find social identity is still a useful construct, although I now understand identity as the result of communication patterns at multiple levels (the interpersonal, the level of groups such as culture, organization, and/or religion, and mass-mediated (e.g., via television, radio, the internet).
Presently, I’m working on a chapter for a book on media, education and dialogue. Some of the other authors are relying on social identity theory from Tajfel and Turner, which seeks “to understand the psychological basis of intergroup discrimination. Tajfel et al (1971) attempted to identify the minimal conditions that would lead members of one group to discriminate in favor of the ingroup to which they belonged and against another outgroup” (italics removed).
The difference between the two approaches (Jackson & Hardiman, and Tajfel & Turner), at least on this surface reading, is that the latter is geared to understanding while the former is geared to action. Debates during the mid-1990s involved whether or not social identity models are descriptive (i.e., distanced, theoretical, “what is”) or prescriptive (as in, this is the normal way people grow, through all of the stages and roughly in this sequence). I’d love to know of work combining these two different theoretical bases: what are the practical, applied uses of this kind of knowledge?
[Perhaps this 2005 overview by Briodo and Reason situates these two – and perhaps additional – approaches in relation to each other?]