The end of the Cold War brought widespread optimism about the future of civil-military relations. But as Michael Desch argues in this thought-provoking challenge to Harold Lasswell's famous "garrison state" thesis, the truth is that civilian authorities have not been able to exert greater control over military policies and decision making. In wartime, civil authorities cannot help but pay close attention to military matters. In times of peace, however, civilian leaders are less interested in military affairs―and therefore often surrender them to the military. Focusing on a wide range of times and places, Desch begins with a look at changes in U.S. civil-military relations since the end of the Cold War. He then turns to the former Soviet Union, explaining why it was easier for civilians to control the Soviet military than its present-day Russian successor. He examines the Hindenburg-Ludendorff dictatorship in World War I Germany, Japan during the interwar era, and France's role in the Algerian crisis. Finally, he explores the changing domestic security environment and civil-military relations in South America.
Very good. Desch presents a structural theory to explain differences in civilian control over the military in various times and places. Essentially, he argues that civilian control will be strongest in states with high external and low internal threat environments. It will be weakest where there is low external threat and high internal threat. In between, low/low and high/high threat situations are structurally indeterminate; situational-specific characteristics like military doctrine are likely to have significant effects on the strength of civilian control.
I do have a number of quibbles and questions throughout, and my margins ended up pretty furiously scrawled-in. Like any good social scientist, Desch closes with a section of "this is stuff I didn't study that would be useful for others to research." There are some useful notes in there, and there's also plenty of discussion to be had about some of his assumptions and the framing of his argument. This is good work and a useful theory, but there's enough gaps and wiggle room that I remain not totally convinced. Absolutely worth reading if the topic remotely interests you, and it's a quick and easy read.