A Case Study in the Limits of Generalization in Comparative Politics

If you’re here because you’re interested in comparative politics, you ought to go chug yesterday’s rich post on Nate Jensen’s blog about what legislatures in authoritarian regimes do and then chase it with Tom Pepinsky’s addendum on his own Indolaysia blog. Nate solicited thoughts on this topic from some of the best minds in the field, including Tom, and you can learn a lot in just a few minutes of reading.

What struck me in reading those posts was the sharp tension in comparative politics between the general and the specific. Nate’s question has been a focal point for comparativists for nearly a decade now, and the arc of that inquiry it has inspired should be familiar to anyone who works in this field.

  1. A particular case or new line of deductive theorizing suggests an intriguing puzzle.
  2. Some solutions to that puzzle are tabled in the form of broad conjectures.
  3. Patterns are found in cross-national data that seem to confirm those conjectures. (Note, though, that the preceding two steps do not necessarily occur in that order).
  4. The field buzzes with the promise of a new covering law, the Holy Grail of political science. Conference panels abound.
  5. People with expertise in specific cases start to nibble at that law, identifying permutations or exceptions in the parts of the world they know. By the time they’re done, there’s practically nothing left.
  6. Distracted by the next big puzzle, the field fails to notice what just happened.
  7. The policy community discovers the covering law and incorporates a few of its prescriptive implications (“Consociationalism good! Presidential systems bad!”) into current practice.

Okay, so those last two may be a little mean spirited, but you get the idea: there’s some meta-irony here. The quest for covering laws keeps serving up failures, but the quest itself ends up following a recurring pattern.

On the specifics of how institutions shape politics, I think Nate’s virtual round-table also provides a nice reminder of the limits of functionalist theories of institutionalism. Functionalist theories see institutions as intelligently designed solutions to specific political and economic problems. For any formal political institution, we should be able to work backwards from the fact of the institution’s existence to the problem it was intended to solve and then show in practice how the institution’s origins support that line of reasoning. Voilà.

The problem is the disconnect between the assumed version and the reality of how institutional arrangements come into being. In some cases, historical institutionalists have mustered compelling evidence to show how the original institutional innovation was deliberately and successfully designed to solve a specific problem in a specific case. Most of the time, though, the backstory on how a particular institution actually came to exist in any instance chosen at random looks rather different from that first instantiation. And when a functionalist theory gets the “how” part wrong, it’s going to get the “why” part wrong, too.

Scott Desposato’s contribution to Nate’s post on legislatures in authoritarian regimes provides a nice case in point: Brazil. That country’s 20-year, military-led authoritarian interlude didn’t proactively decide to create a legislature as a way to credibly commit to the protection of its subject’s property rights or to craftily divide its opponents. Instead, it inherited an elected legislature from the democratic regime it toppled, which inherited its basic contours from the colonial metropole that built it. If the Brazilian junta can be said to have made a choice here, it was the decision not to demolish an extant legislature, but that choice surely had as much to do with inertia and the expected costs of such a destructive act as it did with the anticipated benefits of keeping an elected legislature around.

These path dependencies are less problematic for softer versions of institutionalist theory that emphasize the marginal and contingent effects of specific institutions under a variety of conditions, as I think Jennifer Gandhi does in her award-winning book on political institutions under dictatorship.

Even there, though, the layers of contingency pile up pretty quickly. Institutions that seem to function a certain way in a certain set of cases for a certain period of time start to do different things or simply become less relevant as the economic, social, and technological context changes around them, and as wily humans looking for an edge keep innovating. For example, in an era of ubiquitous and instant communication, the information-gathering and opposition-moderating functions that authoritarian legislatures are sometimes said to serve may now be satisfied in other ways, as Gary King, Jennifer Pan, and Molly Roberts suggest in a widely-cited paper on the selective censorship of China’s Sina Weibo social-media platform.

So where does that leave us? I think the opening line of Barbara Geddes’ contribution to Nate’s blog post sums it up nicely:

Autocratic legislatures play different roles and serve different functions in different dictatorships.

In other words, “It depends” is about as close to a covering law as we’re going to get.

Leave a comment

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 5,709 other followers

%d bloggers like this: