even if some authors actually desired oblivion, today’s publishers are irrepressible in the rediscovery of «forgotten minor writers,» even when these never actually wrote a single line.

Posterity, as we know, is voracious and easily pleased. In order to have something to write about, any writing by others will do. Therefore, O writers, you must beware of the use that posterity may make of your work. Naturally, the ideal course would be to leave lying around only the things that, in your lifetime, you had decided to publish, shredding daily any other documentation, including third galleys. But, as we also know, keeping notes as you work is necessary, and death can arrive unexpectedly.

When it does, the first risk is that unpublished material will be published, and will reveal that you were a perfect idiot. And if everyone reads the notes written in the notebook the day before you died, this risk is very strong indeed (particularly because the notes are, inevitably, out of context).

In the absence of notes and notebooks, the second risk is that, immediately post mortem, there will be an epidemic of conferences about your work. Every writer wants to be remembered in essays, doctoral dissertations, critical editions; but these things take time and cash. The immediate conference achieves two results. First, it inspires hordes of friends, admirers, young people in search of fame, to scribble hasty reinterpretations—but, as we know, in such cases they dish up only the familiar gruel, confirming a stereotype. And then, in no time, readers fall out of love with writers so blatantly obvious.

The third risk is that private letters will be published. Writers rarely write private letters that differ from those of ordinary mortals, unless the letters are a pretense, as in the case of Foscolo. They can write «send me some Preparation H» or «I love you like crazy and I thank God that you exist»—which is only normal and natural, and it is pathetic that posterity should seek out these documents simply to conclude that the writer was a human being. What did you think he or she was? A flamingo?

How can such misfortunes be avoided? With notes and notebooks, I would suggest leaving them in an unlikely place, while abandoning in a desk drawer a kind of buried-treasure map indicating the existence of these documents but with undecipherable directions for finding them. This ensures both that the manuscripts will remain hidden and that many dissertations will be written on the sphinx-like impenetrability of those maps.

As for conferences: it might be a good idea to leave precise testamentary instructions, asking, in the name of Humankind, that for every conference held within ten years of your death, the organizers donate twenty million dollars to UNICEF. It will be hard to raise the sum, and few will be so brazen as to go against the express wishes of the deceased.

The love-letter problem is more complex. For those yet to be written it is a good idea to use the computer, thus thwarting graphologists; sign them with affectionate nicknames («Your puppy dog, Fur-rikins»), changing names for each lover, so attribution will prove problematical. It is also advisable to include phrases that, however impassioned, are embarrassing for the addressee («I love everything about you, even your flatulence»), who will thus be dissuaded from publishing.

Letters already written, especially during adolescence, are beyond revision, however. For these, the best course is to track down the recipient and write a note recalling in tranquillity those bygone days; and promising that even after your death you will revisit the scene, still in the thrall of those memories. This doesn’t always work, but a ghost is, after all, a ghost; and the recipient will not sleep well after that.

You could also keep a fake diary, including occasional suggestions that friends and lovers had a tendency to invent and to falsify. «What a delightful liar dear Adelaide is,» or «Today Reginald showed me a fake letter from Pessoa: a really admirable job.»

1990

How to Organize a Public Library

1. The various catalogues must be housed as far apart as possible from one another. All care must be taken to separate the catalogue of books from that of periodicals, and these two from the catalogue by subject; similarly, the recent acquisitions must be kept well away from older collections. If possible, the spelling in the two catalogues (recent acquisitions and older collections) must be different. In the recent acquisitions, for example, pajama should be spelled with an a, in the older, pyjama with a y. Chaĭkovskiĭ in recent acquisitions will follow the Library of Congress system; in the older catalogue the name will be spelled in the old-fashioned way, with Tch.
2. The subjects must be determined by the librarian. On their copyright pages the books must bear no indication of the subjects under which they are to be listed.
3. Call numbers should be impossible to decipher and, if possible, very complex, so that anyone filling out a call slip will never have room to include the last line of numbers and will assume they are irrelevant. Then the desk attendant will hand the slip back to him with the admonition to fill it out properly.
4. The time between request and delivery must be as long as possible.
5. Only one book should be released at a time.
6. The books distributed by the attendant after the request form has been properly submitted cannot be taken into the reference room, so the scholars must divide their working life into two fundamental aspects: reading on the one hand, and reference consultation on the other. The library must discourage, as conducive to strabismus, any crossover tendencies or attempts at the simultaneous reading of several books.
7. Insofar as possible, no photocopier should be available; if such a machine does exist, access to it must be made very time-consuming and toilsome, fees should be higher than those in any neighborhood copy shop, and the maximum number of copied pages permitted should not exceed two or three.
8. The librarian must consider the reader an enemy, a waster of time (otherwise he or she would be at work), and a potential thief.
9. The reference librarian’s office must be impregnable.
10. Loans must be discouraged.
11. Interlibrary loans must be impossible or, at best, must require months. The ideal course, in any event, is to ensure the impossibility of discovering the contents of other libraries.
12. Given this policy, theft must be very easy.
13. Opening hours must coincide precisely with local office hours, determined by foresighted discussions with trade union officials and the Chamber of Commerce; total closing on Saturday, Sunday, evenings and mealtimes goes without saying. The library’s worst enemy is the employed student; its best friend is Thomas Jefferson, someone who has a large personal library and therefore no need to visit the public library (to which he may nevertheless bequeath his books at his death).
14. It must be impossible to find any refreshment inside the library, under any circumstances; and it must also be impossible to leave the library to seek sustenance elsewhere without first returning all books in use, so that, after having a cup of coffee, the student must fill out requests for them again.
15. It must be impossible on a given day to find the book one had been using the day before.
16. It must be impossible to learn who has a book that is currently out on loan.
17. If possible, no rest rooms.
18. Ideally, the reader should be unable to enter the library. If he does actually enter, exploiting with tedious insistence a right, granted on the basis of the principles of 1789, that has nevertheless not been assimilated by the collective sensibility, he must never ever—with the exception of rapid visits to the reference shelves—be allowed access to the sanctum of the stacks.

CONFIDENTIAL NOTE: All staff must be affected by physical defects, as it is the duty of a public institution to offer job opportunities to handicapped citizens (the Fire Department is considering an extension of this rule to their ranks). In particular, the ideal librarian should limp, in order to lengthen the time devoted to receiving the call slip, descending into the basement, and returning. For personnel expected to use ladders to reach the shelves more than eight meters above the ground, it is required that missing arms be replaced by prosthetic hooks, for security reasons. Personnel lacking both upper limbs will deliver the requested volume by gripping it in their teeth (library regulations tend to prevent the delivery of volumes in a format larger than octavo).

1981

How to Speak of Animals

Central Park. The zoo. Some kids are playing near the polar bear tank. One dares the others to dive into the tank and swim alongside the bears; to force them to dive in, the challenger hides the others’ clothes; the boys enter the water, splashing past a big male bear, peaceful and drowsy; they tease him, he becomes annoyed, extends a paw, and eats, or rather chomps on, two kids, leaving some bits lying around. The police come quickly, even the mayor arrives, there is some argument about whether or not the bear has to be killed, all admit it’s not his fault; some sensational articles appear in the press. It so happens that the boys have Hispanic names: Puerto Ricans, perhaps black, perhaps newcomers to the city, in any event accustomed to feats of daring, like all slum kids who hang out in packs.

Various interpretations ensue, all fairly severe. The cynical reaction is fairly widespread, at least in conversation: natural selection, if they were stupid enough to mess with a bear, they got what they deserved; even when I was five, I had enough sense not to jump into a bear tank. Social interpretation: