Both of the Wrights are gone now, but their files of local history are not. Judy Haven, the Historical Association’s librarian until 2000, is not sure how much longer they can hold on to their mass of papers — they have no money, there have been water problems in the basement, and people don’t understand that a historical society needs public support to function. “Everybody wants us to hang on to the newspapers,” she said, “but nobody wants to help us with the cost of storing them in any way.” The pages of the Daily Standard’s mummy issue rattle when you turn them.
CHAPTER 7. Already Worthless
And that’s why, when I was reading Brother Lemberg’s quietly titled work, “A Life-Cycle Cost Analysis1 for the Creation, Storage, and Dissemination of a Digitized Document Collection,” which proposes to pay for the digitization of American libraries by annihilating thousands of established book collections, I thought of Dr. Deck. Four hundred million library discards? Of course not that many will end up going, just as Dr. Deck’s entrepreneurial idea didn’t in the end consume five hundred million mummies; instead American papermakers learned how to handle wood pulp, while in under-forested England, esparto grass became (Munsell writes) “the most valuable fibre2 yet discovered as a substitute for that of linen.” (Esparto, not wood pulp, was the dominant English papermaking furnish even after the Second World War, until Swedish wood-pulp suppliers dumped their product on the British market and forced esparto out.)
But people were willing to give Egyptian linen a try — mummies were incontestably imported and unwrapped, their coverings tossed into papermaking macerators and, possibly, inked into issues of the New York Tribune, and then volumes of the New York Tribune were microfilmed by Kodak’s Recordak Corporation in the thirties, and all over the country old volumes of the Tribune and the Sun and the Times were thrown away or auctioned off, and now much of the transfigured shroudage is reburied once again, though in less pharaonic company, in the company of a million old, undecaying phone books. (The Library of Congress is now microfilming and tossing its early phone-book collection, too, by the way.)
“Duplication in libraries is a real problem,” Brother Lemberg explained to me when I called him. “We all have the same thing! What if just one library owned a paper copy, and we had an electronic library for the rest of them? They could all get rid of theirs. It’s an exciting kind of idea for retrospective work.” Of the books that would be purged in accordance with Lemberg’s cost analysis, about thirty million of them, according to one of his statistical tables, would date from 1896 or before.
I learned about Lemberg’s dissertation from a recent textbook by Michael Lesk,3Practical Digital Libraries: Books, Bytes and Bucks. Lesk works at the National Science Foundation, where, as division director of Information and Intelligent Systems, he helps administer the government’s omni-tentacled Digital Library Initiative, which has dispensed many millions of federal dollars for library projects. The National Science Foundation began its sponsorship of the digital initiative in 1992, with the help of NASA and DARPA, the Defense Advanced Research Projects Agency; now the Library of Congress and the National Endowment for the Humanities are also participating. In an early round of funding, six universities — including Harvard, Berkeley, Stanford, and the University of Michigan — received about four million dollars apiece to help them find their way to the all-electric word kitchen. Lesk agrees with Brother Lemberg: in order for libraries to provide the best service to the most people for the least money, they ought to begin large-scale scanning projects right now, and simultaneously divest themselves of their originals.
When I interviewed Lesk (in December 1998), he offered me an example of what a library might do to help finance a digitization project. Stanford was spending, he said, about fifty-five million dollars to repair its on-campus library building, damaged in the 1989 earthquake. “You could scan the entire contents of that building for less. Now the problem is that letting that building fall down is not acceptable to the Stanford community.” Later he said: “I routinely suggest to libraries, ‘You know, gee, maybe you should think about repairing your building. Maybe you don’t want to do it, because maybe you want to do something else.’ ” Lesk would not get rid of all scanned duplicate books, however. “There are various caveats,” he said.
You might feel that you wanted more than one copy, just in case there was a fire in Washington. Suppose the book had been bound by a famous bookbinder, you might not want to tear that copy apart. But basically, yes, for the vast majority of nineteenth-century material, where you don’t particularly care about the binding or the paper, the book is falling apart, there are no illustrations, I would say, yes, you would be better off with one digital copy and one carefully watched paper copy, than you are with relying on eight different or ten different decaying paper copies and no digital access.
Why not both? Why can’t we have the benefits of the new and extravagantly expensive digital copy and keep the convenience and beauty and historical testimony of the original books resting on the shelves, where they’ve always been, thanks to the sweat equity of our prescient predecessors? We can’t have both, in Michael Lesk’s view, because the destruction of the old library will help pay for the creation of the new library. The fewer books that remain on the shelves, the lower the storage cost — that’s the first-order “benefit” to Lesk’s government-financed plan. And the fewer physical books that are on the shelves — the more they must fly by wire — the more the public will be obliged to consent to the spending of ongoingly immense sums necessary for global conversion, storage, networked delivery, and “platform migration.” Lesk used to work at Bellcore, the research group owned until recently by NYNEX and other Baby Bell phone companies; it isn’t entirely surprising that millions of dollars of the National Science Foundation’s grant money are going to telecommunications networks, notably to the creation of a joint MCI WorldCom/NSF “very high performance Backbone Network Service,”4 or vBNS, and hookups thereto, that will connect phone companies, university science labs, and university libraries, for the benefit of all concerned.
None of this would disturb me — who can quarrel with high-performance backbones? — if an attack on low-tech book spines weren’t also part of the plan. The attack is part of the plan, though, just as it was when the Library of Congress began destroying the newspapers. The single “carefully watched paper copy” that Lesk thinks we should keep will generally not be the one that is actually scanned, because the scanned book is thrown away afterward: big projects like JSTOR (Journal STORage, the Andrew W. Mellon Foundation’s digitally copied database of scholarly periodicals, many going back to the nineteenth century) and the Making of America (Cornell’s and the University of Michigan’s growing collection of digital books) routinely prepare for digitization5 by cutting up the book or journal volume they have in hand, so that the pages can lie flat on the scanner’s glass. Michigan’s librarians choose digital conversion, according Carla Montori, the head of preservation, “knowing that the original will be disbound,6 and that there will be little chance it can be rebound.” The disbound Making of America7 books are, some of them, uncommon mid-nineteenth-century titles; e.g., Henry Cheever’s The Island World of the Pacific (1856), The American Mission in the Sandwich Islands (1866), Josiah Parsons Cooke’s Religion and Chemistry; or, Proofs of God’s Plan in the Atmosphere and its Elements (1865), John C. Duval’s The Adventures of Big-Foot Wallace, the Texas Ranger and Hunter (1870), and Mary Grey Lundie Duncan’s America As I Found It (1852). Michigan’s preservation department maintains that the thousands of books they have scanned were all terminally brittle — but the term “brittle” has shown itself to be remarkably pliant in recent decades, and nobody now can evaluate Michigan’s diagnoses, since most of the scanned remnants went in the trash.