Sometimes while browsing at a bookstore I will come across a volume that has eluded me for ages – only to find, once back home with it, that a copy is already on my shelves. Is that a look of pity on the clerk’s face when I answer why the book is being returned? The mistake happens much too often for someone on this side of old age, and it foreshadows a time, years hence, when I will stockpile canned goods and breakfast cereal by accident, through sheer absentmindedness.
In a few cases, I have accumulated two complete sets of an author’s work. This has been deliberate, which may not make it seem any less eccentric. One set of volumes is heavily marked up. Key passages are underscored, with lines so straight it looks as if a ruler had been used; the margins filled with cross-references; the inside covers and other blank pages turned into a customized index. Often, there is a list of unfamiliar words (with the pages on which they occur) to be looked up later. Such was the intention, at least.
The other set of the same volumes is pristine, and I intend to keep them that way. The authors in question are, so to speak, household gods: it is necessary to revisit them every so often for renewal, and to ask for favors. I have no more than a half dozen of these duplicated pantheons. (One such is devoted to C.L.R. James,  the historian and political thinker; another is for the literary theorist Kenneth Burke. ) My old notes are still useful, and the marked-up copies recall the initial, decisive encounter with each volume.
The marks left in the books themselves were nothing, compared to the marks that each book left on the reader.
But annotation is also a kind of interpretation – and someone else’s interpretation can get in the way, at times, even if that “someone” is an earlier version of oneself. So I’ve ended up acquiring those doppelgänger editions. They are the quickest shortcut to feigning a naive encounter with the author’s work.
Some of you may have indulged the same quirk. At least I hope so.
At one time, my wife would express surprise when I brought home a book by some figure she knew preoccupied me. "I thought you already had all of his books," she would say -- a reasonable point. Then I would explain that this one would be my "reading copy." It is, as the expression goes, a gentle madness.
To many people, of course, the idea of marking up a book seems distasteful – a violation of the text, a sign of disrespect for the author’s authority. The structuralist literary theorist Roland Barthes divided readers into two categories: those who produced marginalia, and those who left the book as they found it, instead writing their glosses elsewhere. Barthes himself belonged to the second category. He copied out extracts from his reading on what looks (in the reproductions I’ve seen) like small pieces of graph paper.
No doubt Barthes developed a whole theory around the contrast -– though the distinction need not be so airtight as it might sound. Consistency in such matters is not necessary. Over time, my own attitudes and routines have certainly changed, growing ever more nuanced but also more specific to the kind of text open in front of me.
The thought of using a yellow highlighter in a book, for example -– let alone the sight of such markings on a page of something in a used bookstore or (still worse) a library – fills me with disgust. Yet since getting JSTOR access, my weekly routine includes going over a stack of printouts with ... a yellow highlighter.
Buying books online, I make sure they are in new condition. If an item is rare, or if the price is right, then light pencil markings are tolerable. But there is one exception. While doing research on the history of American radicalism, I have built up quite a collection of obscure pamphlets – and really don’t mind annotations in them. At times, the marks are more interesting than the text itself. What might otherwise look like an unremarkable exercise in rhetorical overkill turns out to have had a particular, intense meaning to one reader.
Once in a while, my expectations about the condition of a book will be stood on their head. A few years ago, a copy of Studies in the Scope and Method of “The Authoritarian Personality” was listed for sale online. It was, in its day, about 50 years ago, a major volume in the social sciences – the response of American scholars to the research in social psychology done by German emigres of the Frankfurt School. The Authoritarian Personality itself remains in print, but I’d been looking for the volume of responses for some while. Finding a clean but very affordable copy – with the original dust jacket, no less! – seemed too good to be true.
When the book arrived, I opened it to find, on its flyleaf, something that the bookseller had not considered important enough to mention: the neatly penned signature of the original owner, Stanley Milgram. 
For once, acquiring a scarce book in excellent condition proved anticlimactic. It would have been fascinating to read the debates over method and interpretation over the shoulder of the man who did the legendary series of experiments on authority and obedience. Apart from his name, however, Milgrim left the book unmarked.
According to Anthony Grafton, a professor of history at Princeton University, there was a time when the proper method for annotating was a basic skill taught in the classroom. (I forget where he discusses this, but can recommend The Footnote  or Bring Out Your Dead  to anyone unfamiliar with Grafton’s work.) During the Renaissance, the senior scholar would provide guidance on what marginal references should be added to one’s edition of Aristotle, or whatever the text for the class might be.
This was not a matter of whim or individual expressiveness. There were systems for doing it properly. Upon ripening in erudition, you could presumably be trusted to go solo. It helped that, for a long time, a reader could order a new volume from the bookseller with blank pages sewn into it at various points. (This was a standard option in the 18th century, and I've seen some 19th century volumes similarly customized.)
Today, of course, annotation is entirely a matter of personal preference. Any method for doing it is bound to be the product of improvisation. My own system, for what it’s worth, has become ever simpler and more efficient over time.
There are advantages to being able to mark significant portions of a text with extremely neat underlinings. (When you are done, the book looks not so much annotated as fileted.) Should the author in question be, say, Hegel, then the tendency to re-read the passage as you underscore it is all to the good. But when the volume in question is one meant for a popular audience -- or if it is, for that matter, some contemporary professor offering a Zizekian riposte to a Derridean critique of a Lacanian interpretation of “Green Acres” -- then the benefits of slowing down are not always clear.
Not that such books are devoid of insight. But life is short. So to get a move on, I’ve taken up a speedier alternative to neat horizontal underlining. With passages that stand out as important, interesting, or otherwise worthy of note – but not necessarily meriting long study and thorough contemplation – I make a line in the margin. To highlight a particular significant sentence within that passage, I might underscore it. (Likewise with striking phrases or unfamiliar words.) But now, making the marginal hashmark is my default habit.
That is the first level of emphasis. A few abbreviations come in handy as intensifiers – indicating what to reread, and why. They also expedite things once I actually sit down to write.
The most generic is “NB” (nota bene) – a suggestion to give the passage more attention, for any of various reasons. Slightly more focused in intention is “CK,” meaning that it makes sense to check a fact or citation. If the book is on a topic of long-term interest, the annotation may point to a reference it is worth following up on. But often CK implies doubt. Something in the text feels “off” – if not enough so to merit the more emphatic and clearly defined “BS.”
It is worth cultivating a knack for apt quotation. The marginal note “Q” is an important tool. I’ll often identify several times as many quotable bits as I ever use, but that is not such a disadvantage, and learning to keep an eye out for them is a good practice. Likewise with “def,” which goes next to definitions of crucial terms.
Even more important are the passages marked with the Greek capital letter Sigma. In mathematics, it is the symbol for the sum of an operation or series. I use Sigma next to statements defining the book’s intention, method, structure, or conclusion (or any combination thereof). Such passages sum everything up. I also use Sigma to indicate moments that clarify for me what is at stake in an argument -– whether or not it seems likely that the author would concur.
The emphasis of any given mark can be qualified by either a question mark (“say what?”) or an exclamation point (“oh man!”). The symbols also occur in various combinations, including the dreaded “Sigma BS.”
Such are the tools in one annotator’s kit. My reflexes as a reader tend to be more academic than not: The instinct to turn first to the footnotes and bibliography of a book is uncommon among the civilian population. But some of the annotating habits just mentioned took shape over years of writing about books for magazines and newspapers. I list them with the thought that some might be useful for people facing other sorts of deadlines.
Now that I’ve fine-tuned the system, though, it seems anachronistic. Some formats of digital publication allow the reader to interact with the text, so that you can insert, not just compact marginal jottings, but glosses as extensive as anyone deems necessary.
The flexibility this offers -– the virtually infinite margins it can create -– could yet also have another effect. It might bring back the practice of treating annotation as a skill to cultivate in a formal way, guided by a teacher, rather than as a private matter. If Erasmus were alive today, he would probably use a laptop.
My own preference is still for tangible text: the flesh and blood of ink and paper. But space limitations may yet oblige a rethinking of that policy. Especially if I keep piling up duplicate copies of the same books.