Google – Steamboats Are Ruining Everything

Pedestrianism in novels

I am perennially curious about the distances that characters in nineteenth-century novels are happy to walk. Turgenev's Torrents of Spring happens to offer some geographic clues. The hero, Dimitri Pavlovitch Sanin, stays at the White Swan Hotel in Frankfurt, Germany, in 1840, when there was as yet no railroads to carry him home to Russia. The novella fails to locate Sanin's hotel precisely, and Google doesn't yet index fictional travel accommodations as far back as 1840, but on his first night in town, Sanin takes a stroll.

He went in to look at Danneker's Ariadne, which he did not much care for, visited the house of Goethe, of whose works he had, however, only read Werter, and that in the French translation. He walked along the bank of the Maine, and was bored as a well-conducted tourist should be.

Johann Heinrich von Dannecker's statue Ariadne on the Panther is lodged in a museum known as the Liebieghaus, and the house where Goethe was born is also easy to locate, so it's safe to say that Sanin was staying in what is today downtown Frankfurt.

Later in the novel, after Sanin has fallen in love, his beloved orders him to stay away from her for a day. He passes the time with her brother:

After drinking coffee, the two friends set off together—on foot, of course—to Hausen, a little village lying a short distance from Frankfort, and surrounded by woods. The whole chain of the Taunus mountains could be seen clearly from there. The weather was lovely; the sunshine was bright and warm, but not blazing hot . . . The two young people soon got out of the town, and stepped out boldly and gaily along the well-kept road.

The family dog accompanies them; they play leap-frog, run races, sing songs; and they space out the walk by drinking and eating at three inns. They're not, in other words, in any hurry. How far did they go? If you ask Google Maps for walking directions from the Liebieghaus to Goethe's house, and thence to the district of Hausen (which is now part of Frankfurt, and no longer a separate village), the trip is about 4 miles one way, and should take about an hour and twenty minutes on foot. An eight-mile, three-hour round trip is not a terribly taxing walk, though few today would take it uncomplainingly. An equivalent walk would take me from my neighborhood, Park Slope, Brooklyn, to the Soho shopping district in downtown Manhattan.

Copywrongs

“Terms of Infringement,” my review of Adrian Johns’s new history Piracy: The Intellectual Property Wars from Gutenberg to Gates, appears in this weekend’s issue of The National (Abu Dhabi).

In the opening paragraph of my review, I refer to an essay by Immanuel Kant. Its title is “Of the Injustice of Counterfeiting Books,” and in Google Books, you can see a Google technician’s hand obscuring the text of it at the bottom of page 234. Above: My scan of a print-out, corrected by me in green ink, of Kant’s essay as scanned by Google, along the scanning technician’s hand.

Objection, your honor

Because the community of authors seem to be greeting the advent of the Google Books Settlement with uncharacteristic silence, here's the letter of objection I recently sent to the clerk of the court settling the case. You can only object if you opt in to the settlement, so this means I'm in. (The deadline for opting out or for objecting to the settlement is May 5.) The letter incorporates ideas that I laid out in earlier posts; apologies for the repetition. I'm sure I've mangled a few legal details along the way; the only thing I'm confident of is that the settlement will have important effects that have not been foreseen by anyone, let alone me.

Brooklyn, NY 11215

28 March 2009

Office of the Clerk, J. Michael McMahon
U.S. District Court for the Southern District of New York
500 Pearl Street
New York, New York 10007

Dear Mr. McMahon:

As an author whose works appear in the Google Book Search database, I am writing to object to some of the provisions in the proposed Google Books Settlement. The concerns that motivate my objections came to me as I was navigating the settlement website, and broadly speaking, my objections fall into two categories. The first consists of defects in the settlement that seem to be caused by a mismatch between legal thinking, which is abstract and precise, and the diffuse and sometimes sloppy way that information is actually managed online. I've grouped these objections below under the heading "Practical objections." The second category is caused by the disparity between author's rights as legally defined and the influence and persuasive power practically available to authors heretofore. I’ve grouped these under the heading "Moral rights." There's some overlap between the two categories.

A. Practical objections.

1. I object to the restriction of the settlement to books registered with the U.S. Copyright Office.

According to the Google Books Settlement list of Frequently Asked Questions, “The requirement that United States works must be registered with the U.S. Copyright Office in order to be covered by the Settlement was included in the Settlement in order to comply with a decision of a U.S. court.” Perhaps, then, it’s a provision that the two parties felt they needed to comply with, but that your court will be able to set aside. I hope so. Until this settlement was proposed, it was unnecessary to register one's work with the U.S. Copyright Office in order to own copyright in it. If a copyright was infringed, an author could sue provided he was able to prove that the work in question was his and that he had not sold the rights to it. As a preliminary to a suit, an author did have to file a copyright claim with the Copyright Office, but it was not necessary to file this claim upon publication or for unpublished works in order to deter infringement. If approved, therefore, this settlement would have the practical effect of retroactively adding a bureaucratic requirement in order to secure ownership of copyright.

A personal example might explain the dilemma. Searching the Google Books database, I found a citation of a thesis I wrote as an undergraduate and deposited, as a requirement for my degree, in my college library. I never registered my thesis with the U.S. Copyright Office, because I never expected it to be published. Nonetheless I believed I had copyright in it and would have the power to decide whether it was to be published. As far as I can tell, Google has not yet digitized it, but by the terms of this settlement, the company is only restrained from digitizing it by its wish to maintain good will in the marketplace. By a common-sense definition, my undergraduate thesis is a "book"—it is bound, there are multiple copies (though only two, I’m pretty sure, in all the world), and it is publicly available (to anyone willing to trek to my college library).

Legal precedent would require me to register my undergraduate thesis with the Copyright Office in order to protect it. But I would argue that the sheer volume of Google's proposed encroachments ought to shift such a bureaucratic burden away from authors and onto Google. In other words, I believe that as a practical matter the courts should waive the formal requirement of copyright registration and offer the protections in the settlement to any book that Google finds on a library shelf.

2. I object to the lack of guidance in the instructions on the settlement website as to how authors should claim a book or an insert that appears several times in Google’s database.

By digitizing the holdings of many libraries, Google has inadvertently scanned a number of works several times. Sometimes the Google database treats the multiple digitizations as belonging to the same underlying work; sometimes it doesn’t. (It almost never knows when an insert has been reprinted.) Should an author claim every version of his work, so as to be sure that he protects his rights in every instance? Or should he only claim one version of each work, so as to avoid seeming to ask for settlement payments he doesn't deserve? As it stands, there is no way for an author to inform Google that it has digitized multiple copies of one work and that the author wants to make a single claim covering all such copies.

This is not merely a bureaucratic flaw. Under some of the ways that Google proposes to make money off of digitized books, different digital versions of a work would correspond to different revenue streams. If I were to get a share of the revenue only from one of several digitized versions of my work, I would be short-changed. I think this is a problem that could be easily remedied if Google were to modify its online claim form.

3. I object to the failure to share future revenue streams with the authors of inserts.

Google has offered to make one-time cash payments to authors of books and inserts for the infringements of copyright it has made by digitizing works and for the right to sell digitizations of their works to institutional subscribers. However, it also proposes to make money off of these digitizations in other ways—perhaps by placing advertisements next to search results or by selling downloads of digitized books. Google proposes to share these alternative revenues with the authors of whole books, but according to schedule C of the settlement agreement, it is not planning to share these profits with the authors of inserts. I object to this as grossly unfair. Why should the author of a novel be entitled to a class of compensation that an author of short stories is denied? In the case of an anthology, a book consists entirely of inserts, and Google would apparently be entitled to keep for itself all such revenues.

Though this seems to me a matter of simple fairness, I have classed it as a practical matter because I think it's the sort of problem whose resolution looks cumbersome to a legal mind but that would in practice be quite easy to remedy with a modification of the database procedures.

4. I object to the failure of the agreement to cover periodicals.

Again this seems to me a matter of simple fairness, but I'm classing it as a practical objection because it seems to me that the distinction between a book and a magazine is much more perspicuous to a lawyer than it is to the scanners in operation at Google. As a practical matter, the Google Books Database does not often distinguish between periodicals and books, especially with older periodicals that are no longer publishing. As researchers who use the Google Books database know to their dismay, periodicals are the database’s Achilles heel; in many cases, older volumes of a bound periodical are treated as if they were a series of books with the same title, and volume numbers and year of publication are rarely included in the metadata.

But my concern here is not as a researcher but as an author to maintain his rights. I have written much more for magazines than for books, and over a much longer period of time, and Google seems to have digitized a number of these works, sometimes for magazines that are defunct and no longer around for Google to negotiate with. I feel that I ought to be able to assert control over and be compensated for Google’s use of all these writings, and if they are excluded from the settlement, it will be an opportunity missed.

B. Moral rights

The idea of moral rights is an attempt to address the fact that works of art are not like, say, eggs or lumps of coal. The creator of a work of art cares about what happens to it, even after he's sold the right to publish it or display it or even own it, in a way that a keeper of chickens does not care about the fate of eggs, or a miner about the fate of coal. Unlike monetary rights, the moral rights in a work of art cannot be transferred. I understand that American law does not protect authors' moral rights. However, in the community of American publishing, these rights have nonetheless been to some extent respected, because a publisher who fails to respect them is considered scurrilous and loses the good will of the community of authors.

Google has introduced itself into the publishing world in a new way. Unlike other publishers, it is not in the business of signing up new authors but is exclusively in the business of repackaging works that have been published by others. It is therefore less subject to direct moral suasion by authors, and this settlement may be one of the few occasions when authors have the power to exercise their moral force with Google. I believe it would be appropriate for authors to insist that Google recognize explicitly some of authors’ moral rights.

1. I object to the failure of the settlement to allow an author to restrict display and redistribution of creative work even if he signed a "work-for-hire" contract with his original publisher.

In the case of a poem, a work of fiction, or a memoir, I believe that an author retains effective control over republication of a work, even if he originally signed a work-for-hire contract.

Again, a personal example may clarify the issues. A decade ago, a short story of mine was published in an anthology now out of print. I'm happy it's out of print, and I'd like to keep it that way, but when I searched for my name in the Google Books database, a "snippet view" of the story popped up and panicked me. I could not remember whether I had signed a work-for-hire contract for it. After subsequent research, I discovered that I had not signed a work-for-hire contract and that I do retain all legal control over the copyright. But what if I had signed one? Would I really have no control over my short story? Before the advent of Google Books, a traditional publisher might have been able to buy the legal right to reprint my story but I doubt that very many publishers would dare republish it over my objections. If any tried to, I would call foul, alerting my friends in the community of authors and asking my agent to apply pressure. Please note that I'm not claiming that I could win money for a reprinting even if I'd signed a work-for-hire contract. I'm claiming that over creative work, I retain an effective veto over republication, especially if the work has subsequently fallen out of print, no matter what kind of contract I originally signed. I believe the settlement should make some kind of provision for a veto based on moral rights, to respect an author's wish to decide which of his works he would like to be judged by in the future.

A side matter: On the settlement website as it is operational today, 28 March 2009, when an author views the details of an insert that he has claimed, the "Display Uses Authorized" button stays clicked when he saves his changes, but the "Display Uses Not Authorized" button does not. The anomaly needs to be fixed.

2. I object to the failure of the settlement to treat the creator of a "work-for-hire" as the copyright holder when the purchaser of the work-for-hire is defunct and no subsequent owner of the rights comes forward.

Publishers go out of business, and sometimes they take their rights with them. The Google Books Settlement exists to remedy exactly this problem. I believe that if a publisher doesn't sell his copyrights before expiring, the rights effectively revert to the creator, even if that creator signed a work-for-hire contract. This may be more a matter of jeopardy than of strict accounting. In other words, if a traditional publisher were to try to republish such a work without compensating the original creator, he'd be risking a lawsuit. The Google Books Settlement, however, excludes books written as works-for-hire and therefore does not make Google accountable to such creators. As a practical matter, such books will probably be treated by Google as if they were non-work-for-hire books whose rights holders had not come forward.

My personal concern here is with translation. Most of my translation work was done as work-for-hire, and in at least one case, the publisher is now extinct. Who now owns the rights to those translations? I’d argue that I have a stronger claim on them than Google does. (Of course, if it transpires that the publisher before closing shop donated the rights elsewhere, then the donee has a stronger claim than either of us.) I suggest that Google modify its claim form to allow translators and other work-for-hire creators to state the nature of their contribution in a work and that Google compensate them (though perhaps not at the same rate as full-on authors) if no other claimants come forward.

I hope these objections are useful to you in deciding a final settlement, and I appreciate your attention to them.

Respectfully submitted,

Caleb Crain

Moral rights vs. work-for-hire

Responding to my post about the Google Book Settlement, a commenter mentioned that he hadn’t claimed his translations because he had done them as work-for-hire, and it occurs to me that the way work-for-hire contracts will play out in the settlement is worth another few words.

In Britain and in Europe, authors own not only a copyright in their works but also moral rights to them. The idea of moral rights is an attempt to address the fact that works of art are not like, say, eggs or lumps of coal. The creator of a work of art cares about what happens to it, even after he’s sold the right to publish it or display it or even own it, in a way that a keeper of chickens does not care about the fate of eggs, or a miner about the fate of coal. Unlike monetary rights, the moral rights to a work of art cannot be transferred.

American law does not similarly protect the moral rights of its authors. In fact, it has a legal convention called “work-for-hire” that is to moral rights what peonage is to citizenship. If you sign a contract with a “work-for-hire” clause, you agree that what you’ve written is a thing without any more integrity than a lump of coal, and that the purchaser can do whatever he wants to it, editorially, without any need to consult you, and that no matter how much or under what circumstances the work is republished, you have no rights to demand further payment. In my opinion, work-for-hire contracts are disreputable acts of force majeure on the part of publishers. Nonetheless, it is almost impossible for a novice writer to avoid signing them, and in the last few years, it has been difficult even for established writers to avoid them. To its shame, the New York Times insists that its freelance writers, including book reviewers, sign work-for-hire contracts, and even The New Yorker insists that shorter pieces like Talk of the Town essays and capsule reviews be works for hire.

But do you really lose rights forever in work because you signed a work-for-hire clause? Let me offer two counterexamples. More than a decade ago, I translated three pieces of Czech fiction for Catbird Press, which were published in an anthology titled Daylight in Nightclub Inferno. I don’t remember whether my translation contract stipulated that it was work for hire; such terms are common with translations, alas, but the publisher of the press, Robert Wechsler, cared deeply about translation, so the contract might have been more generous. In any case, the book went out of print some years ago, and Catbird Press went out of business some time after that. As far as I know, the rights to the anthology were never sold to anyone before or after the press was shut down; certainly no new edition has appeared. So who owns the rights to my translations? Obviously the original authors own the rights to the underlying works of fiction. But the rights to the translations, I would argue, have reverted to me. Let me put it this way: no reputable publisher would try to reissue the book without negotiating some kind of arrangement with the authors and the translators, if they could be located. If they did, I would sic my agent on them in a New York minute. So I have placed a claim on the “inserts” in these books that correspond to my translations.

A second, perhaps more important example. I’ve published a number of short pieces of fiction over the years, but I’ve decided, in my own mind, that my work as a fiction writer officially begins with a novella that I published in the winter 2008 issue of the journal n+1. I’m not going to go from library to library ripping pages out of old journals and anthologies in order to erase my past, but I’ve nonetheless decided that the novella “Sweet Grafton” is opus 1, number 1, and that what came before should be quietly left behind to rot. Lately, though, I have begun to catch glimpses of earlier pieces of my fiction in the Google archive, digitized but not yet released to the public. I don’t care about the money Google might make and withhold from me; I don’t think there’s any serious money to be made. But I don’t want these pieces suddenly to become readily available. Did I sign away my control to these works, with work-for-hire contracts? Again, I don’t remember, and again, I’d argue that it doesn’t matter whether I did. Because even though American law doesn’t respect an author’s moral rights, the community of American publishing heretofore mostly has respected the most important among them. If Saul Bellow had mistakenly signed a work-for-hire contract on an early short story, would anyone have dared reprint it without his permission, or over his protests? No, they wouldn’t have, because by doing so, such a publisher would call down upon himself the opprobrium of not only Bellow but all other writers who are careful about their presentation of their work—which is to say, all other writers that any reputable publisher might want to sign up. Google isn’t part of that community, so it may not be subject to that kind of moral discouragement. And therefore I think this new settlement ought to allow authors to enforce some of their moral rights, whether or not the contract contained a work-for-hire clause. To get started as a writer, I was willing to write almost anything, sometimes under the most absurd terms, and Google is welcome to much of it without any interference from me. But what has my name on it, and took some of my artistry, even where the artistry didn’t succed, I want to retain at least a veto over.

Why not ask for more?

A couple of days ago, in a successful attempt to sabotage my own efforts to meet a deadline, I decided to look into the Google Book Settlement. The settlement is an agreement, hammered out last fall between Google and the Authors Guild, about how Google will share with authors some of the money it hopes to make from its digitization of books in copyright. The agreement itself is very long (you can download it here) and rather complicated. It isn't set in stone quite yet, but the cement is hardening. In order to opt out, you have to notify the settlement administrator by 5 May 2009. You can also stay in the settlement but object to some of its terms, if you make your objections by 5 May 2009. That's only a few months away, so it's not too early to start forming an opinion.

I haven't yet read the agreement all the way through. I didn't think I was going to need to, because I have warm, fuzzy feelings both about Google and the Authors Guild. Also, the site that the settlement administrator has set up for authors to claim their work looks streamlined and friendly and is in fact very easy to use. But now that I've used it, I have some questions, and I'm not sure how to answer them.

For one thing, I'm pretty sure that I filled out the online claims form "wrong," but I felt that I had little choice if I wanted to protect my rights. Then again, I may not have filled them out "wrong"; I'm not sure. Here are some of the dilemmas I found myself facing.

First, under the terms of the settlement, I allegedly don't have rights to my published work unless it was registered with the U.S. Copyright Office. The settlement's fine print claims that this is in conformity with a court decision. I don't think this fine print matters much in my case, because I suspect that most of my published work was copyrighted on my behalf by my publishers, but if it did matter, it would be more than a little enraging. When I started life as a writer, the law of the land rendered it unnecessary to register one's work with the U.S. Copyright Office in order to own copyright in it. In fact, the consensus was that only fussbudgets bothered to. Copyright of one's expression was a common-law claim that didn't need bureaucratic imprimatur; if challenged, you only needed to be able to prove that you and no one else had written the words in question. Listed in Google's database, though not yet digitized, is my undergraduate thesis on Nelson Algren. I know I never registered the copyright. I'm also fairly sure that there are only two surviving copies of it, one on my bookshelf here at home and another in the bowels of Widener Library at Harvard. But it's nonetheless distressing to imagine that if Google were to digitize it, I might not be able to control what happened to it, or make money off it if suddenly a great number of people wanted to know what I thought about Chicago realism when I was twenty. I've also never registered the copyright to any of my magazine articles, ever, but I've felt confident until this week that I owned copyright in them nonetheless, and continued to own copyright when they were reprinted in books, and would not lose that copyright if someone scanned and uploaded it.

Another problem is the settlement's division of the literary world into books and "inserts." An "insert," in the terms of the settlement, is a part of a book that an author owns a right to. For example, the introduction and notes to the Modern Library edition of Royall Tyler's Algerine Captive are copyrighted in my name, so they're my "inserts" in that edition. Since the book is still in print, I told Google that Modern Library still has the rights, and I presume this means the Modern Library will get the lump-sum cash payment for its digitization, not me. But an article that I wrote on Milan Kundera for the magazine Lingua Franca was reprinted in the anthology Quick Studies, which is now out of print, so presumably I will get some money off of that. Not as much as I think I deserve, though. Google is offering to reimburse authors in several ways: first through lump-sum payments for digitization, and later through revenue sharing, based on the money Google makes by selling subscriptions to its database to libraries and colleges, by placing ads on webpages that display the digitized material, and perhaps by selling downloads of books otherwise out of print. As an insert, my old Lingua Franca article will bring me a $15 lump-sum payment and later, perhaps, a $50 payment for inclusion in databases that Google sells to libraries and colleges. But according to Attachment C of the settlement agreement, my insert will bring me nothing from any of Google's other revenue-sharing programs. If Google sells ads next to my Kundera article, or sells someone a download of it, I get zilch. Since Quick Studies is an anthology, it consists entirely of inserts. So who's this revenue going to be shared with? The magazine Lingua Franca, by the way, is defunct. As a writer, I've made far more money off of magazine articles than books at this stage of my career, and I still make money off the reprinting of some of them. It seems to me that excluding "inserts" from substantial revenue sharing is an element of the settlement agreement worth objecting to.

A confusing element of the system: multiple digitized versions. Google's database seems to know that it has scanned both the hardcover and the paperback versions of a short story collection that I helped to translate, Josef Skvorecky's The Tenor Saxophonist's Story. I claimed inserts in both versions, even though the instructions told me not to, because I figured Google would be able to figure out that they were the same book. I claimed both of them for a reason: how else am I to be be sure that Google knows that I have a rights claim (in this case, as a translator, a pretty limited rights claim, but still, something) to both versions? For some reason, Google has scanned two versions of my book American Sympathy, and its database doesn't seem to know they're the same book. Moreover, it also has a reference to what seems to be a free-standing copy of one of my book's chapters, not yet digitized, which I never published separately. I claimed that, too. And I claimed an "insert" in a scholarly anthology that reprints a journal article that overlaps a great deal with one of the book's chapters. I know for a fact that no one else has any right to that insert. Google's instructions say that if an insert reprints material also published in a book, the author should only claim either the book or the insert, but not both. Well, that makes sense as far as the lump payments go. But if Google is later going to sell ads on webpages or sell downloads, it doesn't make sense. The income that Google will be making off my content will be split between the various versions of my work that are in its databases, and I should be able to claim revenue from all versions they hold of everything I've written. (By the way, this book, too, remains in print, so as I understand it, I won't be getting any lump-sum payments for it no matter how I fill out the forms.)

I'll end by saying that this agreement is so complex that it seems destined to have unintended consequences, and that I welcome corrections to any misunderstandings I may have made here. I look forward to learning other writers' reactions to the agreement and the claims process, because my sense is that most of us in the rank and file have yet to weigh in on them.