By James Grimmelmann
Last updated December 2025
I like to help junior scholars find their footing and their voice. Here are some of the suggestions I find myself giving again and again.
There are many ways to be a scholar, and none of them is the best. There is only what is best for you. Everyone has a different writing style, a different sense of what arguments are worth making, a different way of responding to objections. You will do better work—and be far happier—if you try to be the best version of yourself, rather than trying to be someone else.
Well-meaning people, including me, will tell you what to do. All of it is good advice for someone, but it may not be good advice for you. Listen carefully, and weigh it seriously—and then, if you know it won’t work for you, quietly set it aside and move on. Nothing anyone tells you is set in stone. That includes everything on this page. If you know I’m talking nonsense, ignore me.
Think about the scholarship you admire. What about it inspires you? Is it clarity? Empathy? Meticulousness? Imagination? Elegance? Precision? Humility? Wit? Rigor? Whatever it is, you can do it too. Through effort and practice, you can imbue your own work with the same qualities.
Now think about the scholars you admire. Whose way of being in the world seems like something you would like for yourself? Some people are ambitious; others are humble. Some are intense; some are relaxed. Some are serious; others are playful. Some live on the road; some are homebodies. They all made life choices to become the way that they are. You can make those choices too.
The same goes for negative examples. When you’re disappointed by a paper you read, ask yourself what you would have done differently—and then do that instead. If you see a senior figure in your field behaving in a way that strikes you as stressful and exhausting, remember that you are free to imitate only what you admire about them, and leave the rest behind.
It’s important to act like the kind of scholar and person you want to be, because acting like one is how you become one. (Vonnegut: “We are what we pretend to be, so we must be careful about what we pretend to be.”) Do it now, because “next time” has a way of turning into “never.”
When you’re reading for a specific project, you’ll need to read everything in the literature that bears directly on it. The rest of the time, do your best to read widely rather than deeply. It’s much easier to bring in something useful from another part of your field than it is to invent something new in your subspecialty. You’ll be more creative, you’ll be part of more interesting conversations, and you’ll have more fun.
A good heuristic is the principle of maximum surprise. Find the weak spots in your knowledge, and steer into them. If you know exactly what a paper is going to say based on the abstract, reading it is a waste of your time. Look for articles where you have no idea what to expect, where every sentence comes as a revelation.
The end of a good research agenda (in law, at least) should include a list of article ideas, described in a sentence or two each. It’s fine if they’re unrelated to each other, and it’s fine if many of them don’t directly bear on your main research interests. The point is not that you intend to write all of them up immediately, in order. The point is to demonstrate that you are a bottomless fountain of ideas and you will never be at a loss for something worthwhile to write about.
Now forget about the research agenda that you share with others. Make yourself a list of ideas to write about, a list so long you couldn’t possibly cover all of them in two lifetimes. Some of them you’ll get around to. Some will be preempted because someone else did it first and better. Some you’ll look into and realize that the idea itself was bunk. And some will sit there because you had more urgent projects to take on. All of this is fine. The list isn’t there to tell you what to write next. It’s there to inspire you, to remind you that you have options, to keep you from falling into a rut.
Many articles say that they “fill a gap” in the literature. The phrase is a cliché, but the impulse is good. There’s a common tendency in legal scholarship to write a paper that is 90% recap and 10% novel. Fight it! Write papers that are 10% recap and 90% novel.
It’s easy to break this rule unintentionally. You’ll learn something exciting, or have a revealing insight, and want to share it with the world. But just because something is new to you, it’s not automatically new to the community of scholars. I’ve found that my view of a topic is often shaped by the way I first learned it from a course or a book—along with whatever biases and blinders that first source had. Distrust your own knowledge. No idea is worth working up until you have done a literature review.
In particular, you always need to ask whether a new idea is just an old one in a mustache and glasses. Is there a doctrine that already does the necessary work? If there were, where would it show up in the caselaw and commentary? What would it be called? Most of the time—not all but most—I find that there’s already a good answer to my question. Instead of writing something up myself, I cite the existing work and move on.
There is no point in writing what everyone expects. Surprise yourself; surprise your readers. You don’t need to radically reimagine everything—if that’s your goal, you should probably be an artist rather than a scholar—but an article that never startles or unsettles probably isn’t worth writing. Fields of study tend to run in the same grooves again and again; steer off the road and it may be harder going but the scenery is far nicer.
For obvious reasons, this is a hard habit to cultivate. Making yourself into a bottomless fountain of ideas is a good start, because having more choices of what to work on helps you pick out the more interesting ones. Another helpful technique is to write in haste and edit at leisure, because you’re more likely to shake loose interesting ideas if you’re not overthinking it. There will be plenty of time later to see whether a quirky example or startling juxtaposition really works—but if you don’t write it down while you’re thinking of it, you won’t even get the chance to test it properly.
If you want people to understand your paper and build on it, everything about it needs to be as clear, as memorable, and as simple as it can possibly be. Literal elevator pitches aren’t the best way to convey ideas, but as a genre of writing, the elevator pitch rightly emphasizes these virtues. Of the three, I find simplicity to be the best starting place, because without it the other two are impossible. Here are a few techniques I use.
First, don’t be afraid to go visual and tactile as you work out the moving parts of your argument. I’m a big fan of putting everything on a whiteboard, or writing each piece on an index card and moving them around the floor. If you want to make a conspiracy wall, go for it. Arrange and rearrange until you have found the minimal set of concepts that explain your idea. Circle them, and then put everything else in its proper ordered relationship to them.
Second, the highest-stakes parts of your paper—the title, thesis, and abstract—need to be the most straightforward. Wordsmith them intensely. Try out variations. Read them aloud. Try rewriting them in Up Goer Five style (interactive dingus). Don’t rest until you feel confident you could explain the paper to a half-drunk friend of a friend at a noisy party. (Side benefit: it will make parties more fun.)
Third, give each paper as much time in the rock tumbler as it needs to shine. Read through the paper, top to bottom, fixing everything you can find that is awkard, confused, or wrong: spelling, wording, metaphors, paragraph flow, section order, and everything else large or small. Think about the paper on and off for two weeks but don’t look at it—then come back and go through it top to bottom again. Repeat as many times as you can stand. This is how you clean off the grit that obscures the elegance of your argument.
The quality of a life is not measured in years; the quality of an article is not measured in pages.
Never write filler. Make every section and every sentence worth reading. If the genre demands certain conventions, deal with them as succinctly as possible and move on. Respect your readers’ time and intelligence, and they will respect your efforts.
Some ideas are big; some are small. Don’t try to cram a big idea into a short paper, or to stretch a small idea into a long one. Some questions need to be worked out calmly and methodically; others are best tackled in one sitting in a white heat. Write in the genre that the idea itself tells you that it belongs in.
Sometimes, a paper surprises you as you’re writing it—your examples turn out to be redundant with each other, say, or a doctrinal point you thought would be easy to pin down wriggles out of your grasp. It’s not always within your control, but to the extent that you can, let the paper guide you to its new form. Don’t be afraid to insert an entire section even after the paper has been accepted; don’t be afraid to cut out pages of material and save them for another occasion. Some of my favorite articles started out as a cutting from another piece a decade before.
We’re in this business because we care about sharing knowledge. It’s almost willfully perverse to go through the immense effort of writing a work of scholarship, only to let it languish in obscurity where only a few people can read it. Although it’s not always possible to publish your work under fully open-access terms, you should do your best to make it as open as you can. I have found that almost all law reviews have author agreements that are compatible with Creative Commons licenses and are happy to put an explicit license grant in the author footnote.
Relatedly, do what you can to make your scholarship easy to download. Although law-review websites typically make articles available as PDFs, they are notoriously unstable, and tend to make breaking changes to the URLs in their archives every five to ten years. Repositories are good, and you should make sure that your articles are available in appropriate ones, like SSRN and arXiv. If you post a draft, come back after publication and add the full citation and the most authoritative version you’re able to. Even better, self-host your papers too, with direct PDF download links.
A personal website is your best chance to make a good first impression on potential colleagues and students; present yourself the way you want to be seen. Get a custom domain (your full name is a good default) and put up a professional homepage. The absolute minimum is a brief biography, a short summary of what you work on, your CV, and working links to all of your publications. A picture of you is optional, but I recommend it. (If you think a photograph is too boring, a painting, sketch, pixel-art rendition, or other human-made artistic depiction can also work.)
Treat your website as a long-term commitment. Update it as you add publications or professional accomplishments. You can build out other content over time, but in general you shouldn’t add something unless you’re willing to take on the associated maintenance for years to come. Above all else, do not let the domain name expire.
A social-media presence is optional. If you do have one, think about what tone you want to project, because it’s going to stick with you for a long time. Sarcastic or earnest? Pugnacious or even-tempered? You do you.
Every few years, a famous professor at a fancy school publishes an article that gets the law badly wrong or plagiarizes from the souces it discusses. On investigation, it transpires that the professor simply copy-pasted text from a research assistant’s memo into an article draft. It’s not a good look, but it’s an obvious and forseeable risk of putting an RA’s work product anywhere near your own.
The same is true of generative AI. If you want to be a serious scholar with academic integrity, you should not let generative AI anywhere near your writing, not ever. If you use generative AI for writing-adjacent tasks, you will face a constant temptation to accept its outputs without reviewing them carefully line by line. You may even trick yourself into a kind of normaized deviance based on the fact that you’ve been using generative AI and nothing has gone catastrophically wrong yet. Emphasis on yet. Sooner or later you will end up with listeria in the liverwurst.
The essential promise of any honest scholar is that you are personally responsible for every word you publish. Overreliance on generative AI can will lure you into complacency and inattentiveness—only to betray you with hallucinations at the worst possible moment. Don’t put yourself in a position where you run the risk of screwing up this badly.
If that doesn’t convince you, consider the following, from T.W. Korner’s The Pleasures of Counting:
The plans for the San Francisco metro BART called for the trains to be fully automated but to carry a conductor in case of emergencies. It was pointed out that the kind of person who would be happy to ride back and forth without doing anything for nine years would be precisely the person least capable of coping with an emergency in the tenth.
Don’t deskill yourself. If you use AI to produce good-enough prose with decent arguments, you will never become capable of writing great prose with ironclad arguments. Even if there’s never a disaster, do you want to be the kind of scholar who is happy to ride back and forth for nine years without doing anything?
Typesetting your writing is like showering before going out in public. Even people who don’t consciously notice will appreciate it, and you will feel more confident and professional. Spend some time with Matthew Butterick’s Typography for Lawyers. The website is good, but it’s worth owning the book in hard copy. (If you want to go deeper, also read Robert Bringhurst’s The Elements of Typographic Style.) Essential rules to follow include using curly quotes, avoiding system fonts, proper line and paragraph spacing, and a reasonable line length.
It goes without saying but is worth emphasizing that you should always use people’s preferred names and pronouns. That includes spelling their names correctly, something that I’m particularly attuned to as a two-ns-at-the-end Grimmelmann. For non-English names in Latin scripts, use proper accents and diacritics. For names in other writing systems, use a typeface in that writing system where appropriate, and use the person’s preferred Romanization or an authoritative system (e.g., pinyin) when transliterating.
People with power will sometimes behave terribly toward you. They may make unreasonable demands on your time, pass off your work as their own, casually belittle your efforts, leave you out of conversations you should have been in, neglect their promises to you, and much more. But these are all them problems, not you problems.
Your first responsibility is take care of yourself. A close second is to take care of those around you. A distant third is to jump through the hoops that authority figures hold up for you. Do what you need to make it through. But don’t let it convince you that there is something wrong with you, and don’t let it convince you that this is how academia is supposed to be. You can be better, in every sense of the word.
Prestige and money sing a siren song. It’s easy to make career choices based on what the crowd considers important. But you aren’t the crowd. You’re you. Prestige and money can be means to an end. They can give you influence, if that matters to you, or flexibility, if that matters to you, or comfort, if that matters to you. But the things that matter to you matter because they matter to you. Far better to know what those things are, and to live toward them. Sometimes that means grabbing the brass ring; sometimes it means letting the ring go by.