Atmospheric politics

by Henry Farrell on January 10, 2011 · 9 comments

in Environmental Politics

To provide a slightly different twist on John’s post below, one useful way to think about the relationship between violent rhetoric and violent action is to borrow from arguments about climate change. Very often, people engaged in debate over climate change either argue that a specific event (e.g. a cold winter) disproves or demonstrates the reality of climate change. But this is to misunderstand the debate. As I (as a non climate scientist) understand it, the scientific consensus about climate change suggests changes to average temperatures (and changes to the associated likelihood of certain weather events), but it is usually going to be next to impossible to tell whether any given event is ‘caused’ by climate change (it may simply be the result of random fluctuation). Testing arguments about climate change involves multiple data points and the usual problems of statistical inference etc.

Similarly, it is probably a bad idea to attribute any particular violent action to an overall climate of violent rhetoric without some strong evidence of a direct causal relationship. E.g., if the assassin had quoted some of the violent rhetoric that has been widely criticized as an inspiration, had listened to Michael Savage’s radio shows several hours a day or whatever, one would not be able to prove a causal relationship, but it would not be an unreasonable inference. There does not seem to be evidence of that sort in this case. John points to some evidence that is suggestive of a broader statistical relationship between violent rhetoric and attitudes towards violence. This is obviously much weaker than the kind of evidence that climate scientists have gathered pointing to global warming. But, to the extent that it does point to a possible relationship between violent rhetoric and violent action, it is to a probabilistic relationship. One can say that there is (moderate) evidence supporting the argument that violent rhetoric makes violent action more likely. But this does not and cannot show, in the absence of other evidence, that any particular violent action is the product of a general atmosphere of violent rhetoric.


LFC January 10, 2011 at 1:43 pm

That “there is (moderate) evidence supporting the argument that violent rhetoric makes violent action more likely” is an important point and is one reason that historians spend time reconstructing climates of opinion and rhetoric in certain eras, I would argue.

Nazgul35 January 10, 2011 at 8:32 pm

Perhaps this is evidence of changing climate?

James January 11, 2011 at 9:27 am

Well, there’s a big difference between not having any evidence of association because the research hasn’t been done, and not having any evidence because the research has been done and the evidence doesn’t support it.

In this case, I’d say the former applies. Climate scientists can point to actual trend data to make their case for climate change, but political scientists haven’t yet done the work to support or refute the relationship between violent rhetoric and violent action. The research that Sides presented is in no way comparable or adequate to draw any conclusions. You are right, one can’t draw conclusions on a single incident. But the assassination in Arizona isn’t the only instance of political violence that we have seen the past two years.

Here’s some data that begins the process of data gathering.
This link documents a number of the gun-related political violence that we’ve seen the past two years. It’s a start. I’d like to see trends of reported political violence and violent political rhetoric examined in a rigorous way, going back 50 years or more. I hypothesize that an increased “atmosphere” of violent rhetoric is strongly associated with increased incidents of politically-motivated violence.

LFC January 11, 2011 at 9:37 am

James –
You appear to have misread Henry’s post. He says explicitly that there is evidence supporting the argument that violent rhetoric makes violent action more likely. This seems, frankly, obvious to anyone who has read any history, and I therefore would respectfully suggest that investigation of your ‘hypothesis’ might be, at best, not an optimal use of scholarly resources, if not simply a colossal waste of time.

Jeff January 11, 2011 at 10:45 am

I agree that nobody could confidently assert that Jared Loughner’s hideous action was caused by any particular public figure or even by the general climate of hostility.

But for this not to be a wake-up call regarding what is possible is irresponsible. At this point, to defend the intemperate, dishonest, and yes, violent rhetoric that has been employed especially by the right-wing opposition is moving into the territory of reckless and immoral.

James January 11, 2011 at 11:49 am

On the contrary, @LFC, I contend that the available research with respect to violent political rhetoric and politically-motivated violence is woefully inadequate. In fact, Sides’ post stated as much, and came forth with an almost irrelevant study as the best he could find. He linked to a post by Professor Blattman that begins with “I’m sorely disappointed with the political science and economics literature on violence. Why do some people riot, destroy property, beat on others, or get into scraps? Is there any relationship between the propensities for interpersonal and communal or political violence?”

So I wouldn’t agree that it would be a “waste of scholarly resources” at all. In fact, a rigorous, definitive investigation on this specific question, it seems to me, is exactly what the field of political science should be doing. The very fact that following this tragedy that this fundamental question of association is still being raised lends support to my argument.

James January 11, 2011 at 12:38 pm

In fact, I’d like to expand on the idea of “waste of scholarly resources.” I’m an epidemiologist, so my field is public health and population research, not political science. But I would note that it took more than 20 years of research to establish the definitive causal link between smoking and lung cancer, to use a widely recognized example. It started out much like this rhetoric-action association: “Gee! People with yellow fingers and teeth seem to get cancer more often than people without yellow fingers and teeth!” “Preposterous! There is no link at all!” said the tobacco companies, and proponents of smoking. Likewise, it took, not 20 years but a considerable amount of “scholarly resources” to establish the causal link and risk factors associated with (at the time) AIDS – now HIV disease. It took another 10 years to develop effective treatments for HIV disease, including prevention where we aren’t quite there yet.

So I’d just say that political science is still in the “Gosh! yellow fingers!” stage in linking violent political rhetoric with political acts of violence. And I think that for the price we pay in political violence — grief, tragedy, economic cost, and the damage to our political system — is well worth some investment of “scholarly resources.” I’m sorry that you don’t share that opinion.

LFC January 11, 2011 at 1:34 pm

@James -
Perhaps I overstated in using the phrase “waste of scholarly resources.” If political scientists want to try to establish a definitive link between an atmosphere of violent rhetoric and increased likelihood of violent acts, far be it from me to say they shouldn’t.

My argument, however, is that anyone who has a nodding acquaintance with history should already be convinced that there is an association between an atmosphere of violent rhetoric and an increased likelihood of violent acts. You evidently disagree with this proposition.

I will say, however, that the fact that Professor Blattman is “sorely disappointed with the political science and economics literature on violence” is not relevant to my contention. I referred not to the political science and economics literature but rather to history and historical sociology. Perhaps you believe that political scientists should restrict the sources of their knowledge to scholarship by other political scientists (and economists), and that they should ignore everything that historians write. If you believe that, fine. You are certainly entitled to your opinion.

James January 11, 2011 at 5:45 pm

On the contrary, I agree wholeheartedly with you that if one looks at history, the associations between violent political rhetoric and politically motivated violence are unmistakable. I do, however, think that the academic field of political science has a role to play here in further exploring those associations, qualitatively and quantitatively. The fact that there are questions, disclaimers, and uncertainty on that question after the events in Tucson points to the need for more definitive research.

It’s kind of a straw man to assert that political scientists would seek to “restrict” the sources of their knowledge and “ignore historians.” I don’t think anyone is trying to restrict the sources of their knowledge. But after all, political science is a “science” and not a “field of study” and so it is altogether appropriate that they conduct the research to attempt to answer the questions relevant to political violence, beyond the historical record and what one can deduce and infer from that.

Comments on this entry are closed.

Previous post:

Next post: