According to the Automation Services Department, there was at least one earlier purge report, run in 1995, covering discards from some point in the past through the end of 1994. That purge report was itself purged, however; it doesn’t exist on any backup tape or disk. “That’s a report that hasn’t existed on the system for a long time,” I was told. “We did do a purge of withdrawals in May of 1995, but generally we don’t keep those files around for very long, because once they’re withdrawn and gone, there’s really no need to keep a history of that.”
I can’t agree.
(1996)
If Libraries Don’t Do It, Who Will?
Remarks Delivered at the Ribbon-Cutting Ceremony for the
Library Service Center, Duke University Libraries
Thank you, and good afternoon, everyone. I’ve never commemorated the opening of a building before, and I must say it’s an enormous pleasure and an honor to be here, standing in front of this large beige building, to talk about the storage of paper. Paper storage has been on my mind a great deal lately, because last year I started a little library that has in it twenty or thirty tons of bound newspapers, all sold off by the British Library. When the sale actually went through—I didn’t want it to go through, I wanted the library to keep the papers—but when it did, I began to have yearning thoughts about storage. I would drive by some undistinguished steel-sided building, painted some shocking color, and I’d spot those beautiful words FOR LEASE on it, and it would call out to me: storage. I saw a FOR LEASE sign on a converted mill one day and I called the number; the developer said, “I can show you the mill, but I’ve got something better for you. I want to show you something that is a dream space, top of the line, and I know it may be more than you can do costwise, but I just want you to see this and I want you to think about it.” I said okay, and my wife and I took a drive with the developer to a navy base, and we parked in front of an enormous stone building with towers and parapets. It looked like a gigantic medieval fortress. What was it? It was the navy prison. There was a vast rusting cell block with prison cages that went up many stories, and a crumbling men’s room that in its bleak ruination stretched back into the shadows, with maybe thirty sinks along one wall, none of which worked. I was very tempted—but in the end, it just didn’t seem like the right place to keep the last surviving twentieth-century runs of the Chicago Tribune, the New York Herald Tribune, and the New York World.
So I now know, more than I ever did before, about the deep and abiding joy that comes from having enough space—and even now I sometimes feel a slight envious resentment rising within when I cruise down a big highway near New York City and I see buildings that have fifty truck bays. What are they holding? They’re holding cheese products, or truck parts, or Happy Meal toys, or Pentium computers that will be scrap in five years. They’re not holding books. One tank depot or tire warehouse would hold everything that our national library has been sent, free, by publishers and has rejected every year. Our national library says that they don’t have enough space, and they are unwilling to lease space, even though they’re willing to budget 94 million dollars for digital projects.
So here we have a building that has one purpose—to store books—books that we can carry around, flip through, and read just as they were meant to be read by their creators. There’s a cherry picker machine inside, a state-of-the-art cherry picker, that lifts a book retriever up thirty-two feet, where he or she gets the book out of a cardboard tray and comes down with it. And there will be two and a half million books in here. The cost was seven and a half million dollars—so this brand-new place cost about three dollars a book to build. Very few of the books that are going in here have been digitally scanned—and here’s the dramatic comparison. To store a nineteenth-century book, it costs three dollars a book, plus an estimated seventeen cents a book in maintenance and staffing; to scan a nineteenth-century book, it costs a hundred dollars a book. And the book doesn’t even need batteries! Not that it’s a bad thing to take digital pictures of books, as long as the picture-taking doesn’t require that the book be cut out of its binding—the electronic versions can be extraordinarily useful. The point is that offsite book storage, even traditional storage in call number order, is cheap, and any scanning or microfilming we do should be done with the expectation that the original book go back into the collection when the copying is done. And it’s compact, too—2.5 million books go in here, and across the street, an even bigger building is devoted to doing the laundry. Besides being things of intrinsic beauty and interest, books are marvelously compact.
Now there are some futurists, some central planners, who don’t agree with any of what I’ve just said. There’s a man named Michael Lesk, of the National Science Foundation in Washington, who is in charge of giving away millions of dollars in federal grants for digital library projects, who told me that he routinely says to libraries, hey, maybe you shouldn’t repair your library building, you could scan everything in that building, and let the building fall down, and you would save money. Lesk refers to an analysis by a library director from Minnesota who claims that libraries would save about 44 billion dollars over the next one hundred years if they digitally scanned about twenty million books and got rid of more than four hundred million duplicate books. Our libraries would be better off, in other words, if they dismantled about 95 percent of their accumulated collections, according to this analysis. Many—not all, but many—in the digital library world believe that the destruction of local research collections will help hurry us toward the far digital shore. They inflate the cost of keeping things, and they denigrate the durability of paper, because it’s distressing to them that it is so inexpensive to store what was long ago bought, cataloged, and shelved.
Research library collections grow. That’s what this fine building recognizes. Your children’s feet grow, and you buy them new shoes—the bigger feet do not represent a “growth problem” but a developmental fact, something to be proud of. For the past half century or more, though, growth has been an embarrassment to some Washington visionaries. They were swept by a kind of Cold War fervor of informational reform, and they wanted all growth to stop. Libraries would reach a certain fixed size, a few million volumes, and then the weeding parties would gather and the microcopying would crunch down the excess, and when the microfilm spools themselves took up too much room, then they could microcopy the microcopies at ultra-high resolution, and crunch things down more, and the stacks would function like a vast trash compactor, squeezing the words. Because words were squeezable, weren’t they? They were disembodied astral presences that had nothing to do with the ink that formed them or the paper that they were printed on or the bindings that held the paper together; they could be “reformatted”—preserved by being destroyed—because they were immaterial; the books would still exist, they would just not exist; they would be there, but they wouldn’t be there; you could hold your head high and say you had the finest U.S. newspaper collection in the world, when in fact you had gotten rid of 90 percent of it and replaced it with microfilm, much of it unchecked for quality.
What was the source of this thinking? There was one especially influential person some of you may have heard about. His name was Fremont Rider, head librarian at Wesleyan. Rider’s first book, published in 1909, was about the amazing discoveries of spirit rappings and table turnings and levitation—he felt these things deserved serious study and that the tables did in fact turn. He wrote pulp fiction and he was the managing editor of Library Journal, and when he went bankrupt in 1929 after a manic episode in which he spent a small fortune founding a high-society supper club on Long Island, he wrote an indignant pamphlet in which he said that people were fed up with being indebted to banks, and they wanted a new deal. He sent the pamphlet to Franklin D. Roosevelt, and Roosevelt shot him back a letter with a handwritten note saying you’re right—keep it up!—and then a few months later Roosevelt, in his nomination speech, pledged himself to a New Deal for the American People. So Fremont Ri
der was an influential person—and his new deal for librarians was this: make or buy microcopies of your book collection, sell off the book collection to dealers at scrap prices, and you will make, in his words, “an actual cash profit on the substitution.” You’ll enrich your library by getting rid of its books. Rider got the Librarian of Congress and the Deputy Librarian of Congress and the head librarian at Michigan and the head librarian at Harvard and other big-time leaders all over the library world to blurb his book and serve on his Microcard committee. It’s a mathematical fact that book collections double every sixteen years, Rider said (he was wrong about that), and if we didn’t start buying Fremont Rider’s Microcard reading machines and selling off the collections, the stacks were going to overrun the entire square footage of New Jersey. Building a storage warehouse was, according to Rider, “a confession of past failure”—it was unmanly, somehow.
This way of thinking continues in some circles, and it was very powerful in the 1980s, when the Library of Congress had high hopes for its optical disk pilot project, which could, according to the Deputy Librarian, squeeze down the library’s three buildings to one. But the optical disk pilot program didn’t work out—nobody uses those big platters anymore—and over the past decade or so, some enlightened librarians have begun to accept the fact that the easiest way to keep a research collection is to keep the research collection. There is no shame in growth—it is not a confession of failure. Putting up shelves sufficient to hold what’s there is the crucially important primary task that research libraries must fulfill—they must do this because no other institutions, public or private, can be depended on to keep these things—the obscure things, the cumbersome things that even though they’re used only once in ten years or thirty years or fifty years are valuable because they are what people published and read. To a researcher, the fact that something is little used is a positive attribute—if a photo editor for a documentary on, say, Ellis Island pages through a forgotten autobiography and finds a picture that has never been reproduced before, she is overjoyed, because the picture is interesting, and because it is unused. We till around in great collections looking for things that have lain unnoticed—the urge to search through obscurity is basic to scholarship. And if the research libraries don’t keep it—don’t keep copies of the stuff that we as a people publish—nobody else is going to do it. It just won’t happen. We can’t depend on businesses to save our past. The New York Times has kept no run of its own paper, for instance.
We understand why fragile old flags and old presidential letters are valuable as things—we don’t believe that taking a snapshot of Plymouth Rock amounts to a “reformatting” of Plymouth Rock, and after some long and painful decades of urban renewal we’re doing better with old mills and train stations. There are very nice postcards of Whistler’s painting of a woman in a white dress for sale in museum giftshops, but Whistler’s woman in white is still on the wall. Storage! That’s what this building is about. Keep it cool, keep it dry, but above all—keep it. Nice going, Duke.
(2001)
Reading the Paper
An Address Delivered at the Annual Meeting
of the Bibliographical Society of America
Early one morning not long ago, I put on my coat over my pajamas and went out to the end of the driveway to get the paper. It was in a blue bag that said “The New York Times Home Delivery Service,” and “Warning: Keep This Plastic Bag Away From Children.” The bag’s knot was untiable—tied by the deliveryman in the knowledge that each recipient would tear it open and pour out the newspaper, which I did when I got inside. The paper was curled around itself, and when I opened it and began paging through it I could feel in every section the timed-release coolness that is always associated with newsprint. You keep getting outside air on your hands as you read. Newsprint is its own insulation. A single page makes a rattling sound when you turn it, but the whole issue is quiet, muffled by its own layered pulp.
Because newspapers are such patchworks of visual miscellaneousness, we read them differently than we read books, which are, except for an occasional excursion to check a footnote, linear experiences. The newspaper’s front page is both binding and title page at once, and it offers its above-the-fold headlines first, so big, often, that you take them in without even knowing you’re reading them—and then the underworld below the fold comes up out of the shadows into view with a quick turn of the wrist. Next the unfolding begins, and once you open up a section and hear the rattly sounds of the singled-out pages, the rest of the world falls away—the newspaper is so big now that it becomes the landscape. Your eye loops and leaps, lighting on a photo and then dropping to read the caption and then circling to find the article that is associated with it; and you jump from page one to an inner page to finish the article, and then hop across to the adjacent page while you’re there, where you notice an ad with a funny image and an article that looks interesting that is continued from page one, and so you return to page one. And when you turn the page, you don’t turn it as you would turn a book page—you close the whole paper and hand off the right-hand page to the left hand—and then you open the paper again. And at the top of every page is the date: all this happened now.
That constant assertion of nowness is precisely what is so appealing and instructive about old newspapers, yellowing and fragile though they may be. Great libraries turned the newspapers into books—big, heavy books with, in some cases, vellum corners and marbled boards—by binding fifteen or thirty consecutive todays in one. And then the policy changed, beginning in the ’50s, and the libraries got rid of most of their twentieth-century newspaper collections—meaning that the remaining runs are unspeakably rare, rarer than early Chaucers or Dantes. We are very close to losing our own twentieth century. So what I’ve been doing is opening volumes up and taking pictures of pages that interest me. Like a microfilmer of the 1930s, I’ve set up a digital imaging workstation, which consists of a wooden pallet on the floor over which I’ve put a sheet of plywood and some foam core and some white banner paper from Staples. For lighting I use the clamp-on utility lights that you can buy for five dollars apiece at the hardware store; these are supported by an old coat tree and a cast-off intravenous drip-bag stand on rollers. I could use the tripod to steady the camera, but I don’t, because you have to be able to make tiny angle adjustments that are clumsy with a tripod. I just bend over the open volume and frame the picture and hold my breath and try not to quiver at the wrong moment and I end up with a reasonably good four-megapixel digital picture. Now, a four-megapixel picture is better than a three-megapixel picture, and it is in color, but the resolution isn’t as good as 1940s black-and-white microfilm. But even if I had a four-hundred-megapixel camera, and could record from five feet up the ink-slippage marks on each Linotyped line, and the faint fuzz of paper-hairs that fringe a tiny tear in the margin, would I feel that I had successfully reformatted the pages and could now throw them away? Of course not. I like old things because they are old—their oldness and their fragility is part of what they have to say. They hold the record of the time in which they were printed, and the record of the years that have passed between that time and now. The copying of an old thing is, or should be, like the publishing of a scholarly edition, an act of homage to the physical source from which one is working—a way of saying thank you for holding the riches you hold.