The Challenge of Writing a Constitutional Law Exam in a History-Centric Regime
by Michael C. Dorf
In a Verdict column last month, I decried the recent turn to history in SCOTUS cases involving unenumerated rights, firearms, and the Establishment Clause. I contended that, in these particular subject matter areas and as a general matter: history standing alone is too under-determinate to adjudicate contemporary constitutional controversies; therefore it will be used (as it was in recent cases) chiefly as a mask for decisions reached on ideological grounds; and unless the Justices are prepared to overrule just about all of their precedents, they will need some means of deciding when a case presents a novel question to be decided under the history-only rubric versus when a case presents a sufficiently interstitial question to call for the application of older precedents that are preserved per stare decisis.
The turn to history presents a further problem that I raise here only partly tongue-in-cheek. It poses a serious challenge to those of us who teach constitutional law: how to write a good exam question?
To my mind, a good exam question tests students' knowledge of extant case law and their ability to identify and explain where some novel set of facts falls within the relevant case law. I typically try to write exam questions that rest along the fault lines of existing case law. For example, in my 2015 exam, I wrote a question in which, depending on how one characterized the facts, the actions of a doctor constituted either abortion (then a constitutional right) or assisted suicide (not a constitutional right). In that exam, as in all of my exams, I gave full credit to student answers that identified the salient precedents, explained how and why those precedents could be construed in various ways, and then made a plausible argument for one conclusion or another. I don't care (for purposes of grading) which of a variety of plausible conclusions a student reaches.
Note that the kind of exam question just described aims at testing the core skill that legal education has sought to instill since the late 19th century (when law schools as we now know them first emerged): analogical reasoning. Is the set of facts F provided in the hypothetical question on the exam more like those in precedent A or B? To be sure, I am not, by any stretch, a wholly Langdellian instructor. That is, I don't pretend or claim for my students that there is a uniquely correct answer one can invariably reach when reasoning according to the common law method. On the contrary, I (and I'd venture most law professors these days) accept what Oliver Wendell Holmes, Jr. wrote in the opening paragraph of The Common Law:
The life of the law has not been logic: it has been experience. The felt necessities of the time, the prevalent moral and political theories, intuitions of public policy, avowed or unconscious, even the prejudices which judges share with their fellow-men, have had a good deal more to do than the syllogism in determining the rules by which men should be governed.
Nonetheless, I tell my students that Holmes was taking what H.L.A. Hart would later describe as an external perspective. He certainly was not saying that illogic counts as a virtue in legal reasoning; he was merely stating that logic alone rarely decides contested cases. In writing a good exam answer, a student can and should make policy arguments, but not as stand-alone points. Rather, the goal of a good exam answer, like the goal of a good brief or a persuasive judicial opinion, is to make effective policy arguments within the framework--the logic, if you will--of the extant legal doctrine.
Or at least that had been the goal until recently. If instead the right answer to a novel question of constitutional law is supposed to be given by a deep dive into the relevant historical materials, then analogical reasoning is largely beside the point.
Wait a minute! What about what Justice Thomas wrote for the Court in this year's Second Amendment case, NYS Rifle & Pistol v. Bruen? There he said that to determine whether a contemporary regulation of firearms is valid, one must ask whether in 1791 (and/or 1868 for state and local laws) there were accepted analogous regulations. Isn't that analogical reasoning?
Well, yes, after a fashion it is, but it only occurs after one has first exhaustively canvassed the relevant history. Of course, even then, as Justice Breyer's dissent explains at length, the determination whether a particular historical regulation counts as analogous to a modern-day one is chock full of normative content. That is, after all, inherent in the common-law method. Whether case A is relevantly similar to case B depends on what counts as relevantly similar--a normative question; likewise with analogical reasoning under Bruen. So the turn to history in Bruen and other contexts hardly eliminates analogical (and thus normative) reasoning, but it does require the additional and time-consuming initial step of investigating history.
That initial step is essentially impossible for a constitutional law exam, when students have a few hours to analyze the problems and write their answers. I give 8-hour take-home exams, but even that's well short of the time needed to do the history right. By way of comparison, last year I wrote a law review article on the question whether the First, Second, or both amendments in combination protect a right of private armed groups to assemble. (The article was cited and I was quoted about the subject in this NY Times story on Saturday.) It will surprise no one to learn that it took me much more than eight hours to do the historical research I needed to do to write the paper. I couldn't possibly give such a question on an exam.
Well, so what? The point of constitutional law isn't to make life easy for constitutional law professors creating exams for students.
Fair enough, but the problem ramifies for practicing attorneys. It's also nearly impossible for lawyers who are not trained as historians to do the sort of research that would be required even to identify the relevant body of material under a history-first-and-often-history-only approach. For that reason, perhaps, Justice Gorsuch has hired as one of his law clerks (beginning in October Term 2024), a law professor with a Ph.D. in history. But that's hardly a practical proposal for all lawyers whose practice from time to time implicates questions of constitutional law.
And, of course, for the reasons noted in the first paragraph of today's post, even if all constitutional lawyers were also trained historians, that would hardly justify a history-first-and-often-history-only methodology in constitutional law. Well trained historians frequently disagree with each other about what happened in the past, and even when they agree, what they agree upon will often be that it was messy. Ultimately, hiring a well-trained historian to resolve contemporary questions of constitutional meaning is like hiring a well-trained violinist who has only ever performed the works of others to compose a symphony. The expert's skill set is somewhat relevant but hardly sufficient for the task.
Were the turn to history truly justified or even remotely sensible, complicating the task of writing a law school exam would be a small price to pay. Because it is neither--because the Supreme Court will engage in law office history while pretending otherwise--the cost to us constitutional law instructors is gratuitous. Thanks a lot, SCOTUS!