Years ago, as an undergraduate in sociology, Theda Skocpol’s States and Social Revolutions came up in the best courses, those that raised my interest and left me with a more encompassing understanding of major social processes. When I recently saw a copy of this classic on sale, I decided that it was time to finally catch up a little with my reading list as a Bachelor student.
As a philosopher, I read the book with an eye to methodological and metaphysical considerations and Skocpol clearly has thought about these issues from the perspective of a social scientist. In this post, I want to discuss one important choice Skocpol makes, and I hope to follow up with further comments as my reading progresses.
Skcopol starts the book clarifying her methodology and conceptual framework. In the course of doing so, Skocpol notes that her definition of social revolutions “makes successful sociopolitical transformation–actual change of state and class structure–part of the specification” (1979: 5). Although she sometimes writes of failed social revolutions, there is strictly speaking no such thing according to Skocpol’s analysis. Failed social revolutions are to social revolutions like rubber ducks are to actual ducks.
Skocpol justifies this perhaps surprising decision by pointing to her “belief that successful social revolutions probably emerge from different macro-structural and historical contexts than do either failed social revolutions or political transformations that are not accompanied by transformations of class relations” (1979: 5). In other words, there are regularities that apply specifically to successful events which are colloquially called “social revolutions”. As I understand it, Skocpol’s approach suggests that social revolutions are a natural kind which are characterised by effective political and class transformation among other features.
The claim that (causal) regularities constrain the correct analysis of concepts in the social sciences is of interest to philosophy of the social sciences. It suggests that there are ways to carve the social which are more natural than others and supports a robust realism. At the very least, Skocpol believed in 1979 that regularities of the social put speak in favour of one conceptual analysis over another.
Let me end by noting that Skocpol’s justification for her definition contains one worrying ambiguity which results from the word “probably”. Is it used in an epistemic or a nomic sense? In other words, does Skocpol say that she is uncertain about her proposed regularity, showing some epistemic humility, or does she say that the regularity might not be deterministic, acknowledging that social regularities are usually imperfect? Depending on how imperfect the regularity is, this might affect whether one should consider a social revolution a natural kind.
 It isn’t clear from the passages I cite, but Skocpol’s methodological arguments are deeply steeped in causal thinking. My hunch is that Khalidi’s theory of natural kind.
 One other interesting feature of social revolutions as a social kind is their sparsity. Given the criteria put forward by Skocpol, there aren’t that many. Skocpol (1979: 33-40) herself grapples with the consequences of this feature.