The Volokh Conspiracy - Use of Social Science in Judging:

On the Comments Board to my "Economics of Kelo" post (and let me echo David and others on how much I have learned from reading those comments), one participant raised an excellent question, that I think is worthy of a longer response (sorry to write so long, which I have been doing a lot lately for some reason).

Bill writes on the Comment Board:

Todd's economic analysis here is great. [TZ--I think this part of the post is especially brilliant.] But I am reminded of his previous reluctance to embrace social science in considering whether Ten Commandment displays or litigation about them is more divisive. [I wrote a long post under his original one there.]

Todd: You call Breyer's analysis "second rate". Is there any other relevant difference (besides quality of analysis) between your Kelo analysis and the social scientific inquiry into divisiveness that you forswore? I can think of two or three that I mentioned in the other post. But none of them really seem decisive to me.

Perhaps I'm just deluding myself, but I think these situations are distinguishable. So let me take a crack at explaining why I think so, and when and what type/quality of social science evidence may be relevant. Note that it is a seperate question whether this sort of evidence should ever be relevant to constitutional decision-making, a different question that arises in many cases (such as the reliance on social science evidence in Brown v. Board of Education and whether the reliance on the social science evidence there was appropriate and improved the Court's decision and decision-making).

In my "Economics of Kelo" post, I am trying to argue that economics can help us to understand why the "public use" clause is written into the Fifth Amendment and should be enforced. In the "divisiveness" context (here and here), recall, my argument is that the Court should either skip the divisiveness inquiry or actually use real social science to determine whether or not religious displays are truly divisive (just like they rely on empirical evidence in reviewing dormant commerce clause cases, as in the wine cases).

The problem in the religion context, I think, is that Breyer is doing neither--he is postulating that religious displays are "divisive" and that some case rulings will reduce this divisiveness (but not all)--but the Court has no empirical evidence to support these claims. Certainly, such empirical evidence could be produced through surveys and studies, it just hasn't (or at least Breyer doesn't rely on it). Instead, he is relying on his own gut assumptions about what is or is not divisive, rather than real social science research.

Given the lack of any good social science evidence on this point, I believe that the Court should rely on principle, rather than Breyer's personal speculations, formed by a relatively cloistered and narrow social and intellectual life unrepresentative of the overwhelming majority of Americans. If valid social science evidence were available, and it tended to a determinate answer on the question of divisiveness (one way or the other), there would still be the more fundamental question as to whether this sort of social science evidence should be relevant to the decision (a question, of course, that is relevant to all "Brandeis brief"-type evidence). The only polling data I have seen suggests that these displays are not that divisive. Certainly I wouldn't rely on this one weak poll to conclude that the Ten Commandments are not divisive, but there is also no strong evidence to back up Breyer's argument that they are divisive (or to clarify, there may be such evidence out there, but he doesn't rely on any in his decision). So until such evidence is produced one way or the other, it is largely just my hunch against his. And, my own personal hunch would be that if such research were produced, it wouldn't tend to support the distinctions that the Court is drawing. But, of course, my personal hunches don't seem any more valid or relevant to Supreme Court decision-making than Justice Breyer's personal hunches.

So, in this light, it would certainly be valid to reject my Kelo analysis on the ground that the economic analysis that I put forth is incorrect or incomplete, in the same spirit as my criticism of Breyer's gut-feeling estimates of divisiveness in the religion context.

In my defense, however, I would offer that using economic analysis to support the general proposition that the Court should consider the "public use" clause in a case like Kelo is different from saying that the Court should use this evidence to drive results in any given case, as Breyer seems to be suggesting. I am not saying that the economic analysis can be used to balance subjective value with holdout problems in any given case; rather, I am supporting the more general proposition that there are very good reasons why we should take seriously the "public use" requirement in the Constitution.

To illustrate the distinction, consider freedom of speech. There is a difference between saying that the concept of a "free marketplace in ideas" provides a general presumption in favor of protecting freedom of speech versus saying that in any given case a particular idea is a valid contribution to the marketplace of ideas. I would argue that my economic analysis is more like the former (explaining the principle, but not the outcome). And that Breyer's analysis in the Ten Commandments cases is more like the latter (outcome-dterminative), only badly done. And that the Supreme Court's analysis in the context of facially-discriminatory bans on interstate commerce (such as in Granholm and older cases such as Maine v. Taylor) is in the second category (outcome-determinative), but I think generally well done.

	[Todd Zywicki, July 1, 2005 at 10:24am] Trackbacks Use of Social Science in Judging: On the Comments Board to my "Economics of Kelo" post (and let me echo David and others on how much I have learned from reading those comments), one participant raised an excellent question, that I think is worthy of a longer response (sorry to write so long, which I have been doing a lot lately for some reason). Bill writes on the Comment Board: Todd's economic analysis here is great. [TZ--I think this part of the post is especially brilliant.] But I am reminded of his previous reluctance to embrace social science in considering whether Ten Commandment displays or litigation about them is more divisive. [I wrote a long post under his original one there.] Todd: You call Breyer's analysis "second rate". Is there any other relevant difference (besides quality of analysis) between your Kelo analysis and the social scientific inquiry into divisiveness that you forswore? I can think of two or three that I mentioned in the other post. But none of them really seem decisive to me. Perhaps I'm just deluding myself, but I think these situations are distinguishable. So let me take a crack at explaining why I think so, and when and what type/quality of social science evidence may be relevant. Note that it is a seperate question whether this sort of evidence should ever be relevant to constitutional decision-making, a different question that arises in many cases (such as the reliance on social science evidence in Brown v. Board of Education and whether the reliance on the social science evidence there was appropriate and improved the Court's decision and decision-making). In my "Economics of Kelo" post, I am trying to argue that economics can help us to understand why the "public use" clause is written into the Fifth Amendment and should be enforced. In the "divisiveness" context (here and here), recall, my argument is that the Court should either skip the divisiveness inquiry or actually use real social science to determine whether or not religious displays are truly divisive (just like they rely on empirical evidence in reviewing dormant commerce clause cases, as in the wine cases). The problem in the religion context, I think, is that Breyer is doing neither--he is postulating that religious displays are "divisive" and that some case rulings will reduce this divisiveness (but not all)--but the Court has no empirical evidence to support these claims. Certainly, such empirical evidence could be produced through surveys and studies, it just hasn't (or at least Breyer doesn't rely on it). Instead, he is relying on his own gut assumptions about what is or is not divisive, rather than real social science research. Given the lack of any good social science evidence on this point, I believe that the Court should rely on principle, rather than Breyer's personal speculations, formed by a relatively cloistered and narrow social and intellectual life unrepresentative of the overwhelming majority of Americans. If valid social science evidence were available, and it tended to a determinate answer on the question of divisiveness (one way or the other), there would still be the more fundamental question as to whether this sort of social science evidence should be relevant to the decision (a question, of course, that is relevant to all "Brandeis brief"-type evidence). The only polling data I have seen suggests that these displays are not that divisive. Certainly I wouldn't rely on this one weak poll to conclude that the Ten Commandments are not divisive, but there is also no strong evidence to back up Breyer's argument that they are divisive (or to clarify, there may be such evidence out there, but he doesn't rely on any in his decision). So until such evidence is produced one way or the other, it is largely just my hunch against his. And, my own personal hunch would be that if such research were produced, it wouldn't tend to support the distinctions that the Court is drawing. But, of course, my personal hunches don't seem any more valid or relevant to Supreme Court decision-making than Justice Breyer's personal hunches. So, in this light, it would certainly be valid to reject my Kelo analysis on the ground that the economic analysis that I put forth is incorrect or incomplete, in the same spirit as my criticism of Breyer's gut-feeling estimates of divisiveness in the religion context. In my defense, however, I would offer that using economic analysis to support the general proposition that the Court should consider the "public use" clause in a case like Kelo is different from saying that the Court should use this evidence to drive results in any given case, as Breyer seems to be suggesting. I am not saying that the economic analysis can be used to balance subjective value with holdout problems in any given case; rather, I am supporting the more general proposition that there are very good reasons why we should take seriously the "public use" requirement in the Constitution. To illustrate the distinction, consider freedom of speech. There is a difference between saying that the concept of a "free marketplace in ideas" provides a general presumption in favor of protecting freedom of speech versus saying that in any given case a particular idea is a valid contribution to the marketplace of ideas. I would argue that my economic analysis is more like the former (explaining the principle, but not the outcome). And that Breyer's analysis in the Ten Commandments cases is more like the latter (outcome-dterminative), only badly done. And that the Supreme Court's analysis in the context of facially-discriminatory bans on interstate commerce (such as in Granholm and older cases such as Maine v. Taylor) is in the second category (outcome-determinative), but I think generally well done.