By James Grimmelmann
Last updated May 2026
This is a crib sheet on how I approach my academic writing. Being explicit about it helps me be consistent and stick to my goals. It might also be helpful to readers to understand what I am trying to do in my articles, and why. It’s what Tyler Cowen would call “the James Grimmelmann production function.”
But this page is titled “how I write” and not “how to write” for a reason. What works for me will not necessarily work for you, and vice versa. If you do find anything here helpful, that’s great, but it’s not the goal. For actual advice, see my advice for junior scholars.
I believe that “the purpose of research is to ascertain the truth abut the world, as best it can be ascertained, whatever it may be.” I try to approach every research project diligently and with an open mind, and to be scrupulously honest in describing what I believe I have learned. This isn’t an absolute moral commitment; I’m not a Kantian. But as long as the institution of the academy exists, and as long as I serve within it, I will value academic integrity above almost everything else.
I have beliefs about what the law is, and about what it should be. Because I’m a scholar first and a lawyer third, I do my best to keep the two distinct at all times. I often disagree with other people about what the law is and should be. But my loyalty is to the pursuit of truth, not to a side. I will not put my name as a scholar to something I do not believe, even if it would be useful or expedient to other causes I care about. I apply this principle to my own work, and also to amicus briefs or letters I am invited to sign.
I took a lot of math courses in college and deeply internalized the ethos of rigorous proof. I try, very hard, to approach my academic writing with that same ethos. Every claim should be justified in a way that the reader can accept, and the claims should fit together to build an argument with no gaps.
Some proofs read like prose. The point here is to give readers a strong intuition for why the theorem is true. They don’t need to see the details, only to be convinced that they could fill in the details if needed. This is my default writing style. I don’t show all of the machinery, but I want the reader to deeply understand the intuition behind my claim and how it can be grounded in propositions they already accept.
Other proofs are highly formal, as in mathematical logic or high-school geometry. Language is not logic, but sometimes it is important to ensure that every step in an argument is meticulously correct and follows seamlessly from the one before it. This is how I write the most critical passages in my papers, especially when they depend on technical or doctrinal details.
Careful citation practice is also essential. Although I may groan when editors ask for a source to support a proposition, they are right. Good citations strengthen the argument even if no one reads them, because they force me to double-check my assumptions.
I focus, to a fault, on the conceptual architecture of law. I see my job as identifying and clarifying the basic moving parts of our system of laws. As in chemistry, the individual elements may be few, but their combinations and interactions are unbelievably rich. I focus on concepts not because I think they are logically necessary or because the universe requires them, but because I believe that only the simplest, clearest, and most coherent ideas can survive the process of being transmitted down the years from one generation of jurists to the next. Lawyers, legislators, and judges are all groping for a dimly-sensed underlying order; my job is to help them articulate it.
I take doctrine seriously but not literally. I reject the tendency (most prominent among constitutional law scholars) of reading each new decision minutely and issuing oracular pronunciations about how its slightly different wording reshapes existing tests. Most of this will come out in the wash. But I also reject the tendency to downplay doctrine. Courts’ reasoning matters, and so do the details of statutory and regulatory frameworks. It is an ultimate fact for the moment, whatever may be our hopes for a change.
And thus, while I care about policy—because having better and more just laws is profoundly important—I rank it a distant third. Only once we have a clear view of how possible legal rules differ is it possible to make policy arguments about which of them is better. Policy is at the top of my hierarchy of needs as an intellectual: to get there, it is necessary to have a solid foundation of concepts and doctrine. I consider my work well done if I contribute to those foundations, even if I never make it to the top.
I’m a fox, not a hedgehog. I will almost always choose breadth over depth. Partly this reflects my short attention span. But it also reflects my belief that the division of law into discrete subjects obscures more than it reveals. I have made a serious effort to know at least the rudiments of every subject that is commonly taught in U.S. law schools, from family law to antitrust. I could not pass for a scholar in most of them. But if an unexpected issue comes up in one of the areas that I do write about, I want to know where to look.
My casebooks, in particular, reflect my attempts to redefine (and expand) the conventional boundaries of their subjects. Intellectual property includes identifier registries like telephone numbers and radio callsigns. Property includes intangible and intellectual property. Internet law includes platforms’ content-moderation policies, liability for defective software, and much more. I want students—and other scholars—to see the connections and follow them wherever they lead.
I also believe that many incredibly important bodies of law have fallen into a kind of deseutude because they don’t have a clear home in the modern curriclum’s division of courses. Contracts makes a half-hearted stab at restitution; civil procedure makes a half-hearted stab at equity; corporations makes a half-hearted stab at agency. There are scholars who take these areas seriously, but they often labor in undeserved obscurity. I would rather have a properly-cited paragraph drawing on their work than pages upon pages of reinventing the wheel.
Many other scholars work extensively with research assistants. I don’t. My usual practice for solo-authored pieces is to research everything myself, write everything myself, and edit everything myself.
Originally, I didn’t rely on RAs because I never learned how to delegate effectively. I wasn’t employed as an RA in law school. I’ve never worked in a strongly hierarchical legal practice, like a large law firm. I never had a mentor who showed me the ropes of how to work with RAs.
In hindsight, I’m glad I didn’t. Reading all the sources myself gives me a better sense of what they say. Adding all the citations myself gives me more confidence that they’re on point. And so on.
This isn’t an absolute rule. When my students are actual co-authors (which I much prefer to hiring them as RAs), I trust and expect them to do the same kinds of work that I do. I’ve hired RAs for some jobs that are monotonous but still require a careful human eye—like converting my IP materials from one file format to another without breaking hyperlinks or citations. And a few times I’ve worked with outstanding librarians and fellows (like Cas Laskowski) whose help in navigating unfamiliar databases and collections was invaluable. But on the whole, I prefer not to delegate anything.