Go to the Library of Congress catalogue. It is online at http://catalog.loc.gov/. This is an astounding repository of material—not just books and periodicals, but pictures, films, and music. The vast majority of this material, perhaps as much as 95 percent in the case of books, is commercially unavailable.6 The process happens comparatively quickly. Estimates suggest that a mere twenty-eight years after publication 85 percent of the works are no longer being commercially produced. (We know that when U.S. copyright required renewal after twenty-eight years, about 85 percent of all copyright holders did not bother to renew. This is a reasonable, if rough, guide to commercial viability.)7 26

Yet because the copyright term is now so long, in many cases extending well over a century, most of twentieth-century culture is still under copyright—copyrighted but unavailable. Much of this, in other words, is lost culture. No one is reprinting the books, screening the films, or playing the songs. No one is allowed to. In fact, we may not even know who holds the copyright. Companies have gone out of business. Records are incomplete or absent. In some cases, it is even more complicated. A film, for example, might have one copyright over the sound track, another over the movie footage, and another over the script. You get the idea. These works—which are commercially unavailable and also have no identifiable copyright holder—are called "orphan works." They make up a huge percentage of our great libraries' holdings. For example, scholars estimate that the majority of our film holdings are orphan works.8 For books, the estimates are similar. Not only are these works unavailable commercially, there is simply no way to find and contact the person who could agree to give permission to digitize the work or make it available in a new form. 27

Take a conservative set of numbers. Subtract from our totals the works that are clearly in the public domain. In the United States, that is generally work produced before 1923. That material, at least, we can use freely. Subtract, too, the works that are still available from the copyright holder. There we can gain access if we are willing to pay. Yet this still leaves a huge proportion of twentieth- and twenty-first-century culture commercially unavailable but under copyright. In the case of books, the number is over 95 percent, as I said before; with films and music, it is harder to tell, but the percentages are still tragically high. A substantial proportion of that total is made up of orphan works. They cannot be reprinted or digitized even if we were willing to pay the owner to do so. And then comes the Internet. Right now, you can search for those books or films or songs and have the location of the work instantly displayed, as well as a few details about it. And if you live in Washington, D.C., or near some other great library, you can go to a reading room, and if the work can be found and has not been checked out, and has not deteriorated, you can read the books (though you probably will not be able to arrange to see the movies unless you are an accredited film scholar). 28

I was searching the Library of Congress catalogue online one night, tracking down a seventy-year-old book about politics and markets, when my son came in to watch me. He was about eight years old at the time but already a child of the Internet age. He asked what I was doing and I explained that I was printing out the details of the book so that I could try to find it in my own university library. "Why don't you read it online?" he said, reaching over my shoulder and double-clicking on the title, frowning when that merely led to another information page: "How do you get to read the actual book?" I smiled at the assumption that all the works of literature were not merely in the Library of Congress, but actually on the Net: available to anyone with an Internet connection anywhere in the world—so that you could not merely search for, but also read or print, some large slice of the Library's holdings. Imagine what that would be like. Imagine the little underlined blue hyperlink from each title—to my son it made perfect sense. The book's title was in the catalogue. When you clicked the link, surely you would get to read it. That is what happened in his experience when one clicked a link. Why not here? It was an old book, after all, no longer in print. Imagine being able to read the books, hear the music, or watch the films—or at least the ones that the Library of Congress thought it worthwhile to digitize. Of course, that is ridiculous. 29

I tried to explain this to my son. I showed him that there were some works that could be seen online. I took him to the online photograph library, meaning to show him the wealth of amazing historical photographs. Instead, I found myself brooding over the lengthy listing of legal restrictions on the images and the explanation that reproduction of protected items may require the written permission of the copyright owners and that, in many cases, only indistinct and tiny thumbnail images are displayed to those searching from outside the Library of Congress "because of potential rights considerations." The same was true of the scratchy folk songs from the twenties or the early film holdings. The material was in the Library, of course—remarkable collections in some cases, carefully preserved, and sometimes even digitized at public expense. Yet only a tiny fraction of it is available online. (There is a fascinating set of Edison's early films, for example.) 30

Most of the material available online comes from so long ago that the copyright could not possibly still be in force. But since copyright lasts for seventy years after the death of the author (or ninety-five years if it was a corporate "work for hire"), that could be a very, very long time indeed. Long enough, in fact, to keep off limits almost the whole history of moving pictures and the entire history of recorded music. Long enough to lock up almost all of twentieth-century culture. 31

But is that not what copyright is supposed to do? To grant the right to restrict access, so as to allow authors to charge for the privilege of obtaining it? Yes, indeed. And this is a very good idea. But as I argue in this book, the goal of the system ought to be to give the monopoly only for as long as necessary to provide an incentive. After that, we should let the work fall into the public domain where all of us can use it, transform it, adapt it, build on it, republish it as we wish. For most works, the owners expect to make all the money they are going to recoup from the work with five or ten years of exclusive rights. The rest of the copyright term is of little use to them except as a kind of lottery ticket in case the work proves to be a one-in-a- million perennial favorite. The one-in-a-million lottery winner will benefit, of course, if his ticket comes up. And if the ticket is "free," who would not take it? But the ticket is not free to the public. They pay higher prices for the works still being commercially exploited and, frequently, the price of complete unavailability for the works that are not. 32

Think of a one-in-a-million perennial favorite—Harry Potter, say. Long after J. K. Rowling is dust, we will all be forbidden from making derivative works, or publishing cheap editions or large-type versions, or simply reproducing it for pleasure. I am a great admirer of Ms. Rowling's work, but my guess is that little extra incentive was provided by the thought that her copyright will endure seventy rather than merely fifty years after her death. Some large costs are being imposed here, for a small benefit. And the costs fall even more heavily on all the other works, which are available nowhere but in some moldering library stacks. To put it another way, if copyright owners had to purchase each additional five years of term separately, the same way we buy warranties on our appliances, the economically rational ones would mainly settle for a fairly short period. 33

Of course, there are some works that are still being exploited commercially long after their publication date. Obviously the owners of these works would not want them freely available online. This seems reasonable enough, though even with those works the copyright should expire eventually. But remember, in the Library of Congress's vast, wonderful pudding of songs and pictures and films and books and magazines and newspapers, there is perhaps a handful of raisins' worth of works that anyone is making any money from, and the vast majority of those come from the last ten years. If one goes back twenty years, perhaps a raisin. Fifty years? A slight raisiny aroma. We restrict access to the whole pudding in order to give the owners of the raisin slivers their due. But this pudding is almost all of twentieth- century culture, and we are restricting access to it when almost of all of it could be available. 34

If you do not know much about copyright, you might think that I am exaggerating. After all, if no one has any financial interest in the works or we do not even know who owns the copyright, surely a library would be free to put those works online? Doesn't "no harm, no foul" apply in the world of copyright? In a word, no. Copyright is what lawyers call a "strict liability" system. This means that it is generally not a legal excuse to say that you did not believe you were violating copyright, or that you did so by accident, or in the belief that no one would care, and that your actions benefited the public. Innocence and mistake do not absolve you, though they might reduce the penalties imposed. Since it is so difficult to know exactly who owns the copyright (or copyrights) on a work, many libraries simply will not reproduce the material or make it available online until they can be sure the copyright has expired—which may mean waiting for over a century. They cannot afford to take the risk. 35