Monday, August 04, 2014

"Off-topic" Debate and the Socratic Method

by Michael Dorf

When I was an undergraduate thirty-some-odd years ago, I spent a considerable portion of my extracurricular time engaged in organized debate. I participated in parliamentary debate--sometimes called "off-topic" debate--an import from the Commonwealth countries that was seen as something of a refuge from "on-topic" or "NDT" (for National Debate Tournament) debate. In the latter, students would debate the same topic over and over again throughout the year, with an emphasis placed on research (in those pre-computer-age days, symbolized by gigantic files of index cards), and on the ability to speak incredibly fast so as to overwhelm the opposing side. In parliamentary debate, by contrast, topics changed constantly and were announced only minutes before each debate round so that in theory, each debate was extemporaneous. I say "in theory" because many teams cheated by preparing a battery of cases that could be squeezed into the various vague topics. I understand that this trend has continued so that the use of prepared cases is no longer considered cheating.

I recently learned from friends whose son is currently a high school debater that the high school version of extracurricular debate has now diversified to include an intermediate category of "public forum" debate, in which topics change monthly. (High school debate has long included another branch, called Lincoln-Douglas, that is somewhat closer to parliamentary debate.) Presumably the more frequent change of topic means that the files for each debate are smaller, and therefore there is less of a premium on speaking fast--although I could well be wrong about that. The monthly change of topic might just mean that the students participating simply spend more of their between-debate time doing research.

In any event, I have not paid much attention to the world of competitive debate over the last three decades. Although I am occasionally invited to judge one or another college debate, I invariably decline. Given my day job, it would feel too much like a busman's holiday.

Because I have not kept up with the world of competitive high school or college debate, I base my remarks here on my recollection of how things were in the 1980s, but I don't think that will render my observations dated, because I am using that recollection simply as the inspiration for some thoughts on legal education.

Former debaters have gone on to have successful careers, many of them, not surprisingly, in law, including in legal academia. E.g., the two main contributors to this blog were successful parliamentary debaters, while such noted legal academics as NYU President John Sexton and my colleague Steve Shiffrin were once very successful on-topic debate coaches, and my own mentor, Larry Tribe, was a champion on-topic debater. Because on-topic debate has been the dominant form of high school and collegiate debate in the U.S., I think it's fair to say that within the law professoriate, those who were high school or college debaters were more likely to have been on-topic debaters than off-topic debaters. Nonetheless, legal education in the U.S. looks a lot more like parliamentary debate than like on-topic debate: We allot no time for research of the facts; instead we just dive right into the normative argument.

Consider standard Socratic instruction, when it works as it is supposed to. The professor might begin by asking easy questions that aim to ensure that the students understood what the facts, holding, and reasoning of the case were. Then come the harder questions. What if the plaintiff had not consented? Or suppose that both parties had been partly at fault? Whatever. The point of such questions is first to explore the limits of the rule that the case is taken to establish and second to see whether any sensible rule (or standard) that justifies the result in the case is in fact justifiable overall. I have engaged in this sort of exercise countless times as a teacher, and it is very useful in training students to think analytically.

But the Socratic method as just described has a pretty serious limitation: Often the question of what the "best" rule or principle is will depend on the answer to some empirical question. E.g., Should contributory negligence be a complete defense (the traditional common law rule) or should the rule be one of comparative negligence, whereby the plaintiff's own negligence merely reduces recovery (as in many states now)? Should each side bear its own legal fees (the "American rule" absent a fee-shifting statute) or should the loser pay the winner's legal fees (the "English rule")? The answers to such questions depend in part on the incentive effects of each regime, their respective adiministrative costs, etc. A student who says this sort of thing will earn praise from the professor, but the class does not then embark on a review of the empirical evidence. We are in law school, after all, not fact school.

There is some pushback in the other direction. In my own field of constitutional law, the last couple of decades has witnessed some greater effort by legal academics and political scientists to take account of one another's work. And various scholars have engaged in inter-disciplinary empirical work that bears on law. Such work has long been a side interest of mine (including a current grant-funded project on the efficacy of smoking warnings). And even after the premature death of my colleague Ted Eisenberg, the "grandfather of empirical legal studies," Cornell Law School remains a leader in that field. But even as some of us in the legal academy spend some of our scholarly time investigating empirical questions, legal instruction remains a mostly fact-free zone.

In a follow-up post, I'll discuss the limited demand for facts from policy makers, but here I want to conclude with a clarification about what I'm suggesting with respect to legal education. I do not mean to endorse the on-topic debate model for legal education. For one thing, at least as I encountered it, that model was seriously flawed in its own right. The problem was not just the fast talking, which is mostly self-limiting: Former on-topic debaters who grow up to interact with regular humans learn to slow down so that they can be understood. The deeper problem was the indiscriminate use of authority--so that a quotation of Hayek (or of a contemporary Hayekian) could be used to "rebut" a quotation of Keynes (or a contemporary Keynsian), or vice-versa. According to a well-known story, an enterprising debater once wrote a letter to the editor of Time magazine so that he could later invoke "Time magazine" as an authority for a proposition in a debate. The story may be apocryphal but the lesson is true: all sources of authority are not equal, but the debate format gives the debaters an incentive to invoke unreliable sources as though they had greater weight.

Perhaps that caution points to at least a small reform in the way we teach law school classes. As noted, in my experience, the (kinder gentler version of the) Socratic method is useful for teaching analytical thinking. Evaluating the weight of sources should be part of such thinking.


Sam Rickless said...

Maybe it's because I'm a philosopher, but where you see the need for empirical data to determine the right rule or standard (e.g., in contributory vs. partial negligence and in the American vs. English legal fee payment system), I see a need for more philosophical/normative investigation. Take negligence. The main complaint against contributory negligence, if I understand it correctly, is that it does not permit a slightly negligent seriously injured plaintiff from recovering any damages from a very negligent defendant. This is not an empirical argument: it is an argument based on considerations of fairness or justice. To me, the argument seems quite powerful. Sure, if empirical data shows that the system of comparative negligence is unworkable or counterproductive or actually yields unjust results because of the way it is administered, then those data are relevant. But this is true of any legal system. The real action is in the justification for the negligence tort: is it consequentialist (about distributing the burdens of tortious behavior in a way that maximizes utility, or something in the ballpark) or non-consequentialist (about the relative strength of duties of care and such).
Now take the system of legal fee payment: English vs. American. Here it seems to me that even before looking at the data, we need to look at the normative justification for the civil legal system. What we have is a classic case of different normative considerations bumping up against each other, given the facts. On the English system, there is a serious disincentive to file law suits. This cuts down on frivolous litigation, encourages dispute resolution outside the courts, and lowers the cost that companies pay to defend themselves against lawsuits, the savings getting passed along to consumers in the form of lower prices. Consequentialist considerations pretty clearly point in the English direction. On the American system, we lose these good consequences for the sake of principle: that everyone, especially someone poor, who has suffered cognizable injury as a result of the action or omission of a rich entity or person should be able to turn to the legal system for redress without being discouraged from doing so by the possibility of losing. This is a principle of justice, and an argument can be made that violation of this principle is not justified by the good consequences of violating it. A non-consequentialist will be drawn to the American system for this reason. Of course, one can argue on empirical grounds that the advantages of the English system are so huge that it is worth sacrificing the American principle for them. But for many non-consequentialists (like me), the relative gains from the English system would have to be *enormous* to justify the violation of the American principle. And although I am open to the data showing the relative gains to be large enough to overcome the presumption in favor of justice, I think the biggest question here is whether the basic justification for the legal system is consequentialist or non-consequentialist.

Michael C. Dorf said...

Sam: I generally agree that there is a strong normative component to these and other questions. That's why I think that the sort of Socratic inquiry that occurs in law school classrooms is useful. My point is that there is often (I didn't say always or even usually) also an empirical component. The empirical piece can be relevant, as in your examples, to a consequentialist counter-consideration that may outweigh non-consequentialist considerations. But (untested and possibly false) empirical assumptions often underlie arguments that are not mostly consequentialist. For example, the availability of insurance could undercut not only incentive effects of tort liability but also could undercut a corrective justice rationale, if tortfeasors come to think of the harm they do not as a "wrong" (as envisioned in certain conceptions of the tort system) but as the cost of doing business, as amortized by insurance.

Sam Rickless said...

Maybe this is just a matter of emphasis, Mike, but I think I was reacting to the implicatum (perhaps not the actual statement) that such factors as incentive effects and administrative costs could prove significant in determining whether to go with contributory or comparative negligence, or with the English or the American system of legal fees. I guess that ultimately my view, in these cases at least, is that if the ultimate justification of the relevant legal principles is non-consequentialist, then these sorts of considerations end up being unimportant (except perhaps in extreme, non-actual cases). On the other hand, the same considerations are hugely important if the relevant justification is consequentialist.

I'm not sure how the presence of insurance is supposed to undercut the corrective justice rationale. This may just be me not understanding what the corrective justice rationale is supposed to be. I guess that if corrective justice has to do with punishing wrongdoers with an eye to deterrence or to the lessening of future harm through the recognition that one has done wrong, then insurance could indeed undercut the corrective justice rationale. But if the rationale has to do with making sure that the wrongdoer (rather than the wrongsufferer) pays, then I'm not sure that insurance undercuts it. Of course, if one wants the legal system to function so as to minimize or significantly lessen tortious activity, then the presence of insurance will interfere with the accomplishment of this goal. But this claim acquires real significance only within a consequentialist conception of the legal system's function.

Unknown said...

I teach "legal analysis" at UW School of Law (the standard first-year writing/research/analysis course), and I find that teaching students to evaluate the weight of sources is one of the main tasks of the class. Even putting aside empirical sources, most substantive text books present legal materials as if they all have equal weight---a case from the Supreme Court is presented next to a case from the NY Court of Appeals. The point seems to be presenting various legal theories/reasoning, without much (if any) attention to the weight of authority.

I realize you're addressing a different problem here, but I find that students often have a hard time understanding that even different cases have different levels of persuasion---based on jurisdiction, the depth of reasoning, the date, the level of court, etc. Obviously, some of this is just a need to understand stare decisis and related doctrines. But beyond that, if you can't find an obviously binding case, using so-called "persuasive" authority is often a tough concept for students to get their hands around. What cases are persuasive is often a matter of judgment that students are still in the process of obtaining.

I wonder if teaching an ability to weigh empirical/factual sources might somehow go hand-in-hand with greater focus on weight of legal authority in substantive classes, if both skills are more easily addressed in skills-based courses like mine, or if it makes sense to separate them out.

喜洋洋 said...


Emma O'Connell said...

for the sake of principle: that everyone, especially someone poor, who has suffered cognizable injury as a result of the action or omission of a rich entity or person should be able to turn to the legal system for redress without being discouraged from doing so by the possibility of losing. This is a principle of justice, and an argument can be made that violation of this principle is not justified by the good consequences of violating it. A non-consequentialist will be drawn to the American system for this reason. Of course fifa 15 coins 英雄联盟代练 Elo Boost