Kalyvas, S. N. (2006). The Logic of Violence in Civil War, Cambridge, Cambridge University Press.
This week’s reading, Kalyvas’ The Logic of Violence in Civil War, appears to be a comprehensive attempt at theorizing the use of violence in intrastate conflict, including most if not all variables that determine the contours of that violence. Instead, it comes across as a single embedded case study on the contours of violence in the Greek Civil War, with enough anecdotal evidence about other conflicts to sprinkle in a veneer of generalizability. While Kalyvas is an outstanding writer, and has certainly marshaled a vast diversity of sources, the lack of systematic study of anything in the book renders his conclusions suspect.
This week’s reading, Kalyvas’ The Logic of Violence in Civil War, appears to be a comprehensive attempt at theorizing the use of violence in intrastate conflict, including most if not all variables that determine the contours of that violence. Instead, it comes across as a single embedded case study on the contours of violence in the Greek Civil War, with enough anecdotal evidence about other conflicts to sprinkle in a veneer of generalizability. While Kalyvas is an outstanding writer, and has certainly marshaled a vast diversity of sources, the lack of systematic study of anything in the book renders his conclusions suspect.
At every point, I
was left wondering whether Kalyvas was cherry-picking his sources to support
his analysis. Sometimes he draws on sources discussing the America Civil War,
or the French Revolution, or ancient Greek rebellions, or Russian rebellions,
or the English Civil War, but the cases he draws on change with every
assertion, and somehow, always manage to support the assertion he’s making.
This strongly suggests that his research method is deeply flawed, and that he’s
engaged in selection bias.
To some extent,
this can be regarded as a specious criticism; after all, Kalyvas in this book
is not as interested in explaining particular cases of civil war as he is in
theorizing about civil war more generally. Evaluating disconfirming cases
doesn’t serve the purpose of the book, which is to frame the next generation of
civil war research. But if Kalyvas’ theories only apply to certain cases, and
then, only apply to those cases within specified circumstances, then what is
the point in elucidating the theories to begin with?
Ultimately, this
is a broader problem with the discipline that I have noticed in my readings
this semester across several classes. When authors are writing articles, they
are careful about their claims, meticulous in their evidence gathering and
research methods, and quick to offer and explain away alternative hypotheses.
When they write books (thus far in the reading, except for Scott Straus), they
get more sloppy. This suggests two things: first, that peer review works,
mostly. If it didn’t, then we would not see the difference in research quality
between articles, where peer review is required, and books, where it is not.
Second, it
suggests that book editors need more help. Unfortunately, our reliance on
market-driven mechanisms for book publication means that most political science
books have to be at least stylistically written for a general audience.
Unapproachable books (also known as “most rigorous scholarship”) simply don’t sell.
Publishers who put out a lot of books that don’t sell don’t stay in business
for very long. And since most political science books don’t sell like
gangbusters, the staff that put them together are undermanned and overworked,
leading to sloppy scholarship from authors not being fact-checked.