I take on board many of Swati Srivastava’s suggestions of what the book neglected to do. However she has two main critiques, one methodological and the other theoretical, that I may address.
The first charge is of conceptual looseness regarding the term “sovereignty” in which discrepancies exist between narrower conceptual definitions and broader notions in the empirical cases. This is frequently on account of, as Srivastava has argued in her own book, the idealized and lived versions of sovereignty that are often used interchangeably yet at odds with each other. In my research, leaders and diplomats tended to discuss policies rather than principles (except in broad, non-specific ways), though the implications for sovereignty were usually clear. Their understandings were inferred by combing archival sources or from secondary interviews of others who interacted with them (the costs and barriers of direct interviewing being prohibitive, although the post-Covid Zoom era may alleviate one part of this perennial problem for comparative regionalists). The challenge for me was how to maintain fidelity with such usage or inferred implications against a concise theoretical definition.
The simplest way to map permutations of sovereignty was along an axis between two extremes of a regional organization’s potential influence on the state; i.e., from total non-interference (where the OAU was in the 1990s) to full authority over the state (a hypothetical situation potentially realisable had the “United States of Africa” proposal succeeded). Even so, linearly modelling a locus of agency between state and regional organization has difficulties with some conceptions, such as the case study of the “ASEAN minus X” principle, which was simply a sovereignty bypass rather than power over any state. These reflect the non-linearity of these debates even as they test different conceptions of sovereignty. Ultimately, case variation (methodologically, the need for significant cases with variant outcomes of acceptance, rejection, or qualification) for me was a higher priority than conceptual stringency, which would have greatly limited the empirical set of cases investigable.
A more fundamental question Srivastava raises is the interrelationship between power and utility, as she describes, “divorcing the creation of ‘greater utility’ from normative force.” She is correct that I only tackle it in a single footnote, albeit one that encapsulates a very large debate in sociology (see Dave Elder-Vass, “Developing Social Theory Using Critical Realism,” Journal of Critical Realism 14[1], 2015). The path out of this feedback loop is to use time as a methodological separator. This means that at the outset of each case, I take the initial preferences as fixed, and any changes of preferences thereafter are mapped by the movement of actors across norm circles. Therefore if an actor initially believed a norm held low utility, but became convinced otherwise, it should be observable within the period of the case study by their movement to a different norm circle. This model thus only seeks to explain movement rather than their original positions, and as I show in the book, the practices of agenda control, use of shared norms, and metis are adequate for such explanations.
