A defense of writing in books, by someone who has done it for twenty years and has only recently learned how.
I have been writing in my books since I was about fifteen, which makes it twenty-two years of marginalia, give or take. The early years I would not, at this point, defend in writing. I underlined too much. I underlined the wrong things. I wrote, in the margins of pages I have since reread with embarrassment, the kind of comments a fifteen-year-old writes in the margins of pages — emphatic, declarative, often missing the actual point of the passage by a wide and humiliating margin. I have, over the intervening years, kept most of those books. I revisit them occasionally. They are useful as a record of what I used to think.
The practice has changed quite a lot over the two decades. What I want to write about is what the practice is actually for, because in the last two years marginalia has had a strange small renaissance — it has become, of all things, a content category — and most of the popular discourse on the subject is, in my view, missing the point with some consistency. The books I see annotated in the BookTok and Bookstagram videos are beautiful objects. They are also, almost without exception, doing the wrong thing.
What annotation is for
Begin with the question. What is annotation actually for? The popular answer, on the platforms, is some combination of: it makes the book yours, it lets you find favorite passages later, it produces a beautiful artifact you can share. None of these answers is wrong. All of them, however, are downstream of a more important answer that the popular discourse mostly does not articulate.
Annotation is a practice for slowing down. The point is not the marks. The point is the act of making the marks. The book is the occasion. The marks are the residue. What is actually happening, when annotation is being done well, is that the reader is being forced — by the small physical commitment of putting pencil to paper — to articulate, in some form, what she is thinking about a passage before she moves on to the next one. The articulation is the practice. The marks are how you know the articulation happened.
This is why the most beautiful annotated books, the ones that go viral with their color-coded tabs and their painstaking marginalia in three colors of ink, are often, when you actually read them, doing the least. The annotation has become the art. The book has become the canvas. The reader is, in some sense, performing rather than reading. The marks no longer carry the residue of an articulation; they carry the residue of an aesthetic decision. This is a different practice, and it is fine as a practice, but it is not what marginalia in its serious form is doing.
How to do it well
I want to share, mostly because someone might find it useful, the practice I have arrived at after twenty-two years. It is not the only good practice. It is the one that has, for me, finally produced annotations I am willing to revisit.
Use a pencil. I know this is unfashionable. The pencil is, however, the right tool. It commits less. It allows revision. It produces marks that age with the book rather than against it. Highlighter, the tool of choice for the aesthetic version of annotation, has the wrong properties — it is loud, it is permanent, and it imposes the same emphasis on every passage it touches. Pencil is variable. You can press lightly. You can press hard. You can write in different sizes. The variability is the point.
Write in the margins, not on the page. The line of underlining is the lazy mark. It says: this passage seemed important. The margin note is the working mark. It says: this passage seemed important, and here is what I am doing with it. The discipline of writing the margin note, even briefly, even in a fragment, is the discipline that separates annotation as practice from annotation as decoration. If you cannot articulate, in three or four words, why a passage is worth marking, the passage probably does not need to be marked. (This is a high bar. I do not always meet it. Aspiring to it has, however, made my marginalia considerably better than they were ten years ago.)
Keep the underlining sparing. An underlined book is, to a future reader, almost useless. Everything is underlined; therefore nothing is. The reader who will most often encounter your annotations is your future self, two or three years from now, looking for the one passage you remembered. That self will be unable to find it if you have underlined every paragraph. Underline only what you genuinely think you will want to find again. Mark the rest, if you mark it at all, with a pencil dot in the margin. The dot is enough. The dot does not pretend to be more than it is.
Date your major annotations. This is the practice I most wish I had started in my twenties. A margin note dated 4/19/22 is doing different work than the same note undated. It tells your future self when you thought this thing. It lets you watch your reading change. The books I have re-annotated three or four times over a decade are, in some cases, the most valuable books on my shelf — not for what they contain, but for what they record about the person who has been reading them.
Do not annotate every book. This is the most important point and the one the popular practice gets most wrong. Most books do not need to be annotated. Most reading is not a slow-down practice; it is a flow-through practice, and flow-through practice is fine and important and most of how reading actually works. Reserve annotation for the books that are doing something you actively want to think with. The dark-romance shelf, in my reading, is a flow-through shelf — I read those books fast, in single sittings when possible, and I do not annotate them. The literary fiction and the relationship-thinking shelf I do annotate, because I am reading those books to think with them, and the marginalia is the thinking.
What annotation does for the reading itself
The interesting question, beyond the mechanics, is what annotation does to the experience of reading. I have, after twenty years of doing it, three answers.
First, annotation slows the reader down. This is obvious. It is also, for the kind of book that rewards slow reading, a considerable benefit. A page that takes twelve minutes to read with annotation produces, in most cases, more retained material than the same page read in three minutes. The annotation is, in part, a forcing function for attention. You cannot annotate a passage you have not understood. The annotation, by requiring articulation, requires understanding first.
Second, annotation builds a record. Not just of what you read, but of what you thought about it at a particular moment in time. This is harder to value in the moment than in retrospect. The notebook of margin notes from a decade of reading is, when you sit down with it, a record of your own intellectual life that is more honest than almost any other record you might keep. It does not flatter you. It often embarrasses you. It is, however, accurate. This kind of accuracy is rare and worth keeping.
Third, and this is the answer I find most personally important, annotation produces a different relationship with the books on your shelf. An annotated book is a book you have actually inhabited. The reading was not consumption; it was occupation. The book belongs to you in a way the unannotated book does not. This produces, over years, a shelf that means something — that is not just a record of what you have purchased but a record of where you have spent attention. The shelf, in this sense, becomes a kind of autobiography. It is a quiet one. But it is true.
An annotated book is a book you have actually inhabited. The reading was not consumption; it was occupation.
One small final note
I want to address, briefly, a worry I have heard from several readers who have asked me about this practice. The worry is that annotation will damage the resale value of the book — that a marked-up book is a diminished book, and the right thing to do is to keep the books pristine.
I want to say, clearly: this worry is wrong. The premise it rests on, that books are primarily objects whose value is preserved by leaving them alone, is the wrong premise. The book is a tool. The tool is improved by use. A book covered in margin notes, dated, dog-eared, slightly battered — this is a book that has been used. The use is what the book is for. The pristine book on the shelf is not a more valuable object. It is a tool that has not been deployed. There is no virtue in leaving the tool unused. There is, however, considerable practical reward in actually using it.
Read your books. Mark them up, where the marking helps. Date the marks. Keep the books on the shelf. Return to them. The practice is small. The accumulated effect, over a decade or two, is enormous.
I am, as a small admission, partway through annotating an old novel I have read four times without ever marking. I have decided, this winter, that I owe the book a closer reading. I will let you know what I find. The marks, so far, are mostly questions. This seems to me the right way to begin.