Empirical Scholarship In and Around Constitutional Law
by Michael Dorf
Today and tomorrow I will be participating in a conference at the University of Chicago Law School called "Testing the Constitution." The core idea, as expressed in the draft paper of the organizers of the conference--Lee Epstein (political scientist at Wash U St Louis Law), Barry Friedman (NYU Law), and Geof Stone (U Chicago Law)--is that constitutional law is chock full of doctrines that rest on largely untested empirical claims. The conference organizers (collectively "EFS") give a number of examples, including the following:
The resulting papers will be published in the NYU Law Review. Today we will have the following sessions:
1) Overview paper by EFS.
2) Maggie Lemos (Duke Law) and Kevin Quinn (political scientist at Berkeley Law): To what extent, and in what ways, is state attorney general partisanship reflected in the positions “the states” take, and the interests they assert, before the Supreme Court?
3) Lee Epstein, William Landes (economist at U Chicago Law) and Adam Liptak (NY Times legal correspondent): Whether the Supreme Court only departs from precedents when there is some “special justification” for doing so, as it has claimed.
4) Richard Posner (federal appeals court judge and polymath at U Chicago Law, The New Republic, etc.): Keynote Address discussing all of the papers and more.
5) Brandice Canes-Wrone (political scientist at Princeton) and Yours Truly (Cornell Law): whether the Supreme Court overbreadth doctrine is justified in nominally applying only to freedom of speech on the supposition that free speech rights are especially susceptible to a chilling effect by looking at whether the abortion right is also subject to a chilling effect.
6) Dan Ho (political scientist and lawyer at Stanford Law) and Fred Schauer (U Virginia Law): Whether speech "buffer zones" (at polling places and abortion clinics) impede speech and, more broadly, whether the Holmesian idea that a "marketplace of ideas" will lead to truth.
Tomorrow we will have the following additional sessions:
7) Rebecca Brown (USC Law) and Andrew Martin (political scientist at U Michigan): Whether "the appearance of influence or access [will] cause the electorate to lose faith in our democracy." (The Supreme Court in Citizens United v. FEC said it would not.)
8) Stephen Ansolabehere (political scientist at Harvard) and Nate Persily (political scientist and lawyer at Stanford Law): Whether the creation of majority-minority districts causes expressive harms to voters (as claimed by the Supreme Court in the line of cases beginning with Shaw v. Reno).
9) Concluding discussion among all of the conference participants
Because the circulated drafts are not yet final, I'll hold off on reporting the answers the papers give to the questions they investigated, and will surely follow up about my own project in a later post when Professor Canes-Wrone and I have a revised version that is ready to share. In the meantime, I'll just say that from my vantage point the papers have more than lived up to expectations. Here I will make a set of brief critical observations about the nature of the conference and about constitutional scholarship more broadly.
As EFS acknowledge and document in their paper, the application of empirical methods to questions relevant to law is a substantial and ongoing trend. It is not entirely obvious, however, that their further claim that constitutional scholars have been late to the party is accurate. Indeed, the claim seems belied by the fact that a substantial number of the empirical scholars participating in the conference hold positions on law faculties.
To be sure, there's a selection bias in favor of empirical scholars on law faculties when one is organizing a conference such as this one, but nonetheless, I don't think that the conference participants are major outliers. Consider Cornell Law School, where the first-year sections of constitutional law are regularly taught by: Josh Chafetz (a lawyer who has a D. Phil in Politics); myself (a mere lawyer but I co-author with an economist and various political scientists, such as this one and this one); Sheri Johnson (another mere lawyer but one who has done substantial empirical work on the death penalty and other topics in collaboration with specialists in statistics, experimental psychology, social psychology, and other subjects); Aziz Rana (a lawyer and political scientist); and Jed Stiglitz (yet another lawyer and political scientist). Yes, Cornell has had a strong empirical legal studies group for some time, but none of the con law faculty is part of the core of that group. Perhaps we are a bit more empirically oriented than con law faculty elsewhere, but probably not by all that much: The legal academy in general has gone in for JD-PhDs in a big way over the last couple of decades, and in con law I would be surprised if the modal PhD were not in political science.
Having said that, I have no reason to doubt the evidence that EFS present in their paper showing that constitutional law scholarship relies on or undertakes empirical work only about half as frequently as other legal scholarship. Indeed, the empirical quotient for constitutional law may be even lower than EFS suggest if, as I suspect, most empirical work about the Supreme Court (in constitutional law and other subjects) takes as its aim explaining the Court as an institution. Why does the Court decide cases the way it does? What broader social, economic, and political phenomena explain the pattern of its decisions? These are empirical questions about the Court but not the sort of empirical questions that the Court itself needs answered to test the soundness of its doctrines. (Interestingly, even two of the conference papers--those by Lemos/Quinn and by Epstein/Landes/Liptak--investigate questions that mostly provide information about why the Court decides how it does rather than investigating questions directly relevant to the Court's own decisions.) So why have constitutional scholars paid relatively little attention to the latter sort of empirical question?
Ironically, the answer may be that empirical scholarship about the Court reveals that the Court doesn't really care very much about whether the factual assumptions underlying its decisions are correct. According to the "attitudinal model," political attitudes and ideology of the Justices supply a complete explanation for results in the Court, with legal doctrines (and thus their attendant factual predicates) serving as mere post-hoc rationalizations. I think that's an overstatement but no sensible observer of the Court's work can deny that values and ideology play a large role in Supreme Court decisions.
EFS consider the possibility that Justices don't really care about facts in their paper but dismiss it by noting that normative judgments about better or worse states of the law, and thus the world, are themselves based in substantial measure on factual propositions. I think that's right but I also think it may be irrelevant to the underlying phenomenon, which is largely psychological: Justices' views about facts do indeed underwrite their normative views, but it doesn't follow that their views about facts would change if confronted with new evidence. Suppose it were demonstrated empirically either that race-based affirmative action does not stigmatize its beneficiaries or that the "mismatch" hypothesis is true. Would conservative Justices change their views about affirmative action in the first case? Would liberals change their views in the second case? Count me as dubious.
I do not mean to say that judges and Justices never care about whether the empirical assumptions underlying their doctrines are true. There may well be a substantial subset in which they do care, and (as Professor Buchanan's post earlier in the week underscored) some particular judges are more open to evidence than others. But if the sort of work being done for this conference is to have a practical impact, we need to identify those empirical questions that the courts are likely to actually care about. Doing so is itself an empirical project.
Finally, I would add that plenty of normative legal scholarship is undertaken with no realistic expectation that it will affect the work of the Court. That is not necessarily the point. Scholars may legitimately criticize the Supreme Court for its performance in constitutional (and other) cases with an eye towards mobilizing political constituencies to take action, perhaps through the confirmation process. Or they may criticize the Court simply for the sake of criticizing the Court or, to put it in nobler terms, because telling the truth has inherent value. If critique for its own sake is a worthwhile enterprise with respect to normative arguments (as I believe it is), then it is also worthwhile with respect to empirical propositions.
Today and tomorrow I will be participating in a conference at the University of Chicago Law School called "Testing the Constitution." The core idea, as expressed in the draft paper of the organizers of the conference--Lee Epstein (political scientist at Wash U St Louis Law), Barry Friedman (NYU Law), and Geof Stone (U Chicago Law)--is that constitutional law is chock full of doctrines that rest on largely untested empirical claims. The conference organizers (collectively "EFS") give a number of examples, including the following:
Think of the Miranda rule, familiar to anyone who watches crime drama on television. Here are [some] obvious questions that arise almost instantly, and that matter to the Miranda doctrine, or ought to. Do people generally know those rights without being told of them? After individuals are warned of their rights, do people generally assert the right to silence or do they talk to the police? Do Miranda warnings reduce the number of confessions?Similar questions can be posed about many other constitutional doctrines. Thus, as EFS also explain in their draft paper:
We [matched] notable constitutional thinkers with equally notable empiricists. We asked the con law scholars to identify a core question, assumption, or doctrine from constitutional law, and we asked their partner to take a cut at answering it.
1) Overview paper by EFS.
2) Maggie Lemos (Duke Law) and Kevin Quinn (political scientist at Berkeley Law): To what extent, and in what ways, is state attorney general partisanship reflected in the positions “the states” take, and the interests they assert, before the Supreme Court?
3) Lee Epstein, William Landes (economist at U Chicago Law) and Adam Liptak (NY Times legal correspondent): Whether the Supreme Court only departs from precedents when there is some “special justification” for doing so, as it has claimed.
4) Richard Posner (federal appeals court judge and polymath at U Chicago Law, The New Republic, etc.): Keynote Address discussing all of the papers and more.
5) Brandice Canes-Wrone (political scientist at Princeton) and Yours Truly (Cornell Law): whether the Supreme Court overbreadth doctrine is justified in nominally applying only to freedom of speech on the supposition that free speech rights are especially susceptible to a chilling effect by looking at whether the abortion right is also subject to a chilling effect.
6) Dan Ho (political scientist and lawyer at Stanford Law) and Fred Schauer (U Virginia Law): Whether speech "buffer zones" (at polling places and abortion clinics) impede speech and, more broadly, whether the Holmesian idea that a "marketplace of ideas" will lead to truth.
Tomorrow we will have the following additional sessions:
7) Rebecca Brown (USC Law) and Andrew Martin (political scientist at U Michigan): Whether "the appearance of influence or access [will] cause the electorate to lose faith in our democracy." (The Supreme Court in Citizens United v. FEC said it would not.)
8) Stephen Ansolabehere (political scientist at Harvard) and Nate Persily (political scientist and lawyer at Stanford Law): Whether the creation of majority-minority districts causes expressive harms to voters (as claimed by the Supreme Court in the line of cases beginning with Shaw v. Reno).
9) Concluding discussion among all of the conference participants
As EFS acknowledge and document in their paper, the application of empirical methods to questions relevant to law is a substantial and ongoing trend. It is not entirely obvious, however, that their further claim that constitutional scholars have been late to the party is accurate. Indeed, the claim seems belied by the fact that a substantial number of the empirical scholars participating in the conference hold positions on law faculties.
To be sure, there's a selection bias in favor of empirical scholars on law faculties when one is organizing a conference such as this one, but nonetheless, I don't think that the conference participants are major outliers. Consider Cornell Law School, where the first-year sections of constitutional law are regularly taught by: Josh Chafetz (a lawyer who has a D. Phil in Politics); myself (a mere lawyer but I co-author with an economist and various political scientists, such as this one and this one); Sheri Johnson (another mere lawyer but one who has done substantial empirical work on the death penalty and other topics in collaboration with specialists in statistics, experimental psychology, social psychology, and other subjects); Aziz Rana (a lawyer and political scientist); and Jed Stiglitz (yet another lawyer and political scientist). Yes, Cornell has had a strong empirical legal studies group for some time, but none of the con law faculty is part of the core of that group. Perhaps we are a bit more empirically oriented than con law faculty elsewhere, but probably not by all that much: The legal academy in general has gone in for JD-PhDs in a big way over the last couple of decades, and in con law I would be surprised if the modal PhD were not in political science.
Having said that, I have no reason to doubt the evidence that EFS present in their paper showing that constitutional law scholarship relies on or undertakes empirical work only about half as frequently as other legal scholarship. Indeed, the empirical quotient for constitutional law may be even lower than EFS suggest if, as I suspect, most empirical work about the Supreme Court (in constitutional law and other subjects) takes as its aim explaining the Court as an institution. Why does the Court decide cases the way it does? What broader social, economic, and political phenomena explain the pattern of its decisions? These are empirical questions about the Court but not the sort of empirical questions that the Court itself needs answered to test the soundness of its doctrines. (Interestingly, even two of the conference papers--those by Lemos/Quinn and by Epstein/Landes/Liptak--investigate questions that mostly provide information about why the Court decides how it does rather than investigating questions directly relevant to the Court's own decisions.) So why have constitutional scholars paid relatively little attention to the latter sort of empirical question?
Ironically, the answer may be that empirical scholarship about the Court reveals that the Court doesn't really care very much about whether the factual assumptions underlying its decisions are correct. According to the "attitudinal model," political attitudes and ideology of the Justices supply a complete explanation for results in the Court, with legal doctrines (and thus their attendant factual predicates) serving as mere post-hoc rationalizations. I think that's an overstatement but no sensible observer of the Court's work can deny that values and ideology play a large role in Supreme Court decisions.
EFS consider the possibility that Justices don't really care about facts in their paper but dismiss it by noting that normative judgments about better or worse states of the law, and thus the world, are themselves based in substantial measure on factual propositions. I think that's right but I also think it may be irrelevant to the underlying phenomenon, which is largely psychological: Justices' views about facts do indeed underwrite their normative views, but it doesn't follow that their views about facts would change if confronted with new evidence. Suppose it were demonstrated empirically either that race-based affirmative action does not stigmatize its beneficiaries or that the "mismatch" hypothesis is true. Would conservative Justices change their views about affirmative action in the first case? Would liberals change their views in the second case? Count me as dubious.
I do not mean to say that judges and Justices never care about whether the empirical assumptions underlying their doctrines are true. There may well be a substantial subset in which they do care, and (as Professor Buchanan's post earlier in the week underscored) some particular judges are more open to evidence than others. But if the sort of work being done for this conference is to have a practical impact, we need to identify those empirical questions that the courts are likely to actually care about. Doing so is itself an empirical project.
Finally, I would add that plenty of normative legal scholarship is undertaken with no realistic expectation that it will affect the work of the Court. That is not necessarily the point. Scholars may legitimately criticize the Supreme Court for its performance in constitutional (and other) cases with an eye towards mobilizing political constituencies to take action, perhaps through the confirmation process. Or they may criticize the Court simply for the sake of criticizing the Court or, to put it in nobler terms, because telling the truth has inherent value. If critique for its own sake is a worthwhile enterprise with respect to normative arguments (as I believe it is), then it is also worthwhile with respect to empirical propositions.