Lessig Blog Archives

Enough for the moment on fair use; I’ll get back to that.

I’m interested in the report of the 9/11 Commission on the intelligence failures that led up to the 9/11 attack. I was asked to do a book review of by it the New York Times, and I agreed (the review will appear in next Sunday’s New York Times book review section, but as the Sunday Times Book Review is published the preceding Monday, my review was actually published today) because of my interest in how the nation should be responding to catastrophic risks (this turns out to be, to a considerable extent, a law and science issue). My book Catastrophe: Risk and Response will be published in November by Oxford University Press, and besides taking the opportunity created by my guest blogging to plug the book shamelessly, I am going to be discussing some of the issues raised in it.

One is how to defend against terrorism. Although the 9/11 Commission’s report is a good read, and has other virtues as well, one of its greatest weaknesses is its failure to address, other than in passing, terrorist risks that are even greater than that of another 9/11: in particular the risks of bioterrorism, nuclear terrorism, and cyberterrorism. The Commission’s recommendations are concerned essentially with preventing a more or less exact repetition of the 9/11 attacks, which are any event the least likely form of a future terrorist attack, since surprise has been lost. We give our adversaries little credit if we suppose that the only attack they can launch is the one we’ve anticipated. They didn’t make that mistake on 9/11; why should they now?

The Commission’s report, because of its timing in relation to the election and because of the unusual promotional (and self-promotional) efforts of the Commission’s members, has attracted enormous attention, particularly with respect to its marquee recommendation (as one commentator put it) for the creation of a new position of National Intelligence Director.

With all due respect for Senator Roberts, his proposal, at least as described by the New York Times, seems to me to highlight the problems with the Commission’s report rather than to solve them. He proposes to break the CIA into three parts, and also to create a new agency from within the DIA (Defense Intelligence Agency), and to place all the intelligence agencies under the direction of the new NID (National Intelligence Director). It is unclear what he wants to do with the uniformed intelligence services (Army, Navy, etc.), so it’s unclear just how many intelligence agencies he envisages the NID directing, but it is at least 15 and could go as high as 18.

The agencies are highly diverse. One for example designs and launches spy satellites; another analyzes intelligence for the State Department; another does satellite mapping; another is the FBI’s counterrorist division; and so on and on. No one seems to have considered whether it makes good management sense to place all these agencies under the direction of a single official. There is some optimum span of control, which the proposal seems to exceed. It would be highly unusual in any organization for 18 divisions to report directly to the president, especially if they were as diverse as our intelligence agencies. Probably the agencies should be grouped. Well, they are grouped; most of them are part of the Defense Department. The proposal would spin them off, them put them all under the NID. Sounds pretty dubious.

There is also a point that readers of Larry’s blog should particularly appreciate: the dangers of centralization. Of course to have multiple overlapping intelligence agencies creates a degree of chaos (and there are limits, which breaking up the CIA might well exceed); but it also creates competition among organizations having different organizational cultures, personnel policies, traditions, methods, etc. Were the agencies to be welded into a single organization, the inevitable tendency would be the substitution of uniformity for heterogeneity and centralized control for a competitive “market” in which rival agencies strive to get their views accepted by the President. Would there be a net improvement? Neither the 9/11 Commission, with its less ambitious project of centralization, nor Senator Roberts, with his more ambitious one, has explored this vital question in depth, at least so far as an outsider is able to judge.

Against all this it can be argued that we have to do something, because we were surprised in 9/11, so we must have been doing something wrong. Not necessarily. Yesterday my wife and I happened to attend a surprise 60th birthday party thrown by the woman’s husband. He had thrown a surprise party for her on her 50th birthday. Nevertheless, she was completely surprised–indeed, stunned–by the party yesterday.

It is easy to surprise people, including entire nations.

I’ve received some interesting comments on my last posting. One of the commenters asked me to explain the fair use doctrine; here goes.

The doctrine, which has close counterparts in patent and trademark law, permits a degree of unauthorized copying of copyrighted works. Shocking! If a teenager takes a joyride in my car and is arrested, can he defend by arguing that it was a “fair use”? No, but the example points up an important difference between physical and intellectual property, a difference obscured by the use of words like “theft” and “piracy” to describe unauthorized copying. If someone takes my car, he deprives me of its use. If he copies my copyrighted book, I still can read, use, and sell the book, although my publisher’s and my income may be less because one fewer copy will be sold. But maybe not–maybe the copycat wouldn’t have bought the book if he’d had to pay the retail price. And if instead of copying an entire book, a book reviewer quotes a paragraph from it, I may well be better off (in contrast, say, to someone who doesn’t want to drive my car but just store stuff in the trunk); and if he had to get my permission to quote, I might be worse off, especially since reviews would lack credibility if reviewers needed the author’s permission to quote. Unauthorized quotation by book reviewers is an example of fair use.

The fair use doctrine originated as a judicial doctrine, and like many judicial doctrines was general in terms, setting forth a standard rather than a set of precise rules. Congress put a clumsy thumb in the pudding in 1976, when it “codified” the doctrine. It wasn’t really codification, because Congress did not set down a legislative rule to supplant the judicial one; it just listed four factors for courts to consider, and it made clear, as is sometimes overlooked but as I tried to make clear in my opinion in one of my court’s countless (and fascinating) “Beanie Baby” copyright cases, Ty, Inc. v. Publications Int’l Ltd., that the four factors are neither exclusive nor mandatory.

Where does that leave us? The general view, which you find in advice to authors, as in “Copyright and Fair Use,” is that fair use is very vague: “The only guidance is provided by a set of fair use factors outlined in the copyright law. These factors are weighed in each case to determine whether a use qualifies as a fair use. For example, one important factor is whether your use will deprive the copyright owner of income. Unfortunately, weighing the fair use factors is often quite subjective. For this reason, the fair use road map is often tricky to navigate.”

That’s certainly true if fair use is applied on a case by case basis, using either the four factors or some more general–and more accurate–standard such as that unauthorized copying is permissible when it does not seriously infringe the legitimate interests of the copyright owner. But that isn’t the right approach; it’s much too vague. The right approach is categorical. Book reviewers can quote from the books they’re reviewing. Period (unless, perhaps, the book is so short, and the review so long, and the review quotes so much of the book, that the review becomes an actual substitute for the book–but I’ve never heard of such a case). Parodists can copy a big swatch of the parodied work–otherwise it won’t be recognized as a parody. And so on–and, Patry and I believe, copying of old copyrighted works should be permitted if the copyright owner has not taken minimal steps to place copiers on notice of his interest.

“Fair use” is a doctrine of copyright law (with counterparts in patent and trademark law as well) that allows a degree of unauthorized copying of copyrighted works. Shocking! Squatters’ rights! Can a teenager take my car for a joy ride and when he’s caught plead “fair use”–that I wasn’t using my car and didn’t miss it for a second! But there’s a big difference between physical and intellectual property. Only one person at a time can use my car, so, in general if not in every instance, someone who takes it interferes with my use of it. But if someone copies my copyrighted book, that doesn’t interfere with anyone else’s use of the book, or prevent my publisher from continuing to sell copies, though it may reduce his and my income. That’s why “theft” of intellectual property, and such synonyms as “piracy,” are merely analogies, and often misleading ones. It’s hard to see how I could benefit from someone taking a joyride in my car without my permission, but easy to see how I could benefit if a reviewer of one of my books quoted a paragraph of the book in his review without having to get my permission (which would undermine the credibility of the review).

Bill Patry, a distinguished copyright lawyer and treatise writer, and I have written an article soon to be published in the California Law Review in which we advocate an interpretation of “fair use” that would solve the major problem that extending the copyright term creates. We argue that it should be considered fair use to copy an old work if the copyright owner hasn’t taken reasonable steps to provide notice of his continued rights, as by entering his name and address in a copyright registry. Given such a rule, such registries (which have counterparts in the case of works of visual art) would spring up overnight. Then if an Eldred wanted to publish some old work, he would consult the registry or registries and if no owner was listed (which would usually be the case, because most old works have no commercial value and so their owners won’t bother to try to keep them from falling into the public domain), he could publish it without a license.

The conventional economic objection to copyright and other IP propertization is that it limits access and by doing so causes a misallocation of resources: making and distributing another copy of some piece of software might cost nothing (well, virtually nothing), and yet if the copyright owner charges a price of say $100 for the software, people who value it at more than zero but less than $100 will be deflected to substitute products that may be more costly or of inferior quality; and that’s inefficient. Of course, limiting access may (or may not) be the least objectionable way of incentivizing producers of IP, but, as critics of Eldred and the Sonny Bono Act correctly point out, tacking on years at the end of an already long copyright term has only negligible effects on the incentive to create a copyrighted work in the first place. Who would refuse to create a work unless he could count on his heirs’ receiving income from it more than 50 years after his death?

But, contrary to appearances, limiting access because copyright enables copyrighted works to be sold at prices in excess of marginal cost (the cost of one more copy) is not the main problem created by the Sonny Bono Act. Very few works of art, literature, or entertainment retain significant value 50 or 70 years after the death of their author. As Lessig’s brief in the Supreme Court pointed out, the problem for people like Eldred who want to publish old works (works that would have entered the public domain by now were it not for the Act) is transaction costs, not license fees: the costs of locating and negotiating with the current copyright owner. Those costs may well exceed the modest commercial returns from publishing a public domain work (which anyone can copy). The beauty of the old (pre-1976) copyright system, with its requirement of renewal beyond a shortish initial term (like 28 years), was that most copyrights, lacking commercial value by the end of their initial term, were not renewed, and so fell into the public domain, and so licensing costs fell to zero.

The old system is gone, perhaps forever. What to do? Fair use to the rescue (next).

Larry Lessig from time to time flagellates himself about losing the Eldred case in the Supreme Court. He shouldn’t; it was unwinnable for a host of reason (the lopsided vote–7-2–is a clue). Yes, Congress can confer copyrights only “for limited Times,” but what’s “limited” is a matter of perspective. If the Sonny Bono Copyright Term Extension Act had been in force in Shakespeare’s time (there was no copyright then, in fact), then, since Shakespeare died in 1616, his works would have entered the public domain in 1686–more than 300 years ago. So Larry focused on the retroactive feature of the law (that it extended existing copyrights), but that carries the issue away from “limited Times”–the extended works were extended for only a liimited time–and ignores the fact that there social benefits from “propertization”–from the fact that a property owner has an incentive to conserve his property; that is why toll roads are less congested than “open access” highways; and why poor old Mickey Mouse would be even less free if anyone could employ him without compensation. Then too, if the Supreme Court had invalidated the Act, Congress could have retaliated by allowing states to grant copyright–perpetual copyright, if they wanted, which was the regime for most unpublished works until 1976.

All this said, the net effects of the Act and therefore of the Eldred decision are probably bad. But the worst of them should be remediable fairly easily. Stay tuned.

Sad to say, there are two things that, as a judge, I can’t talk about in public (and this is in public). One is pending cases, which means (in any court, not just mine) cases in which all possibilities of further proceedings, such as an appeal to the Supreme Court, have not yet been exhausted. Oh I can mention a case, e.g., Grokster, which Tim discussed and linked to, or even talk around it a bit, but I can’t comment directly or indirectly on the merits of the decision. I can, though, point you to a case, related to Grokster, that I wrote and that the court in Grokster mentioned in footnote 9 of its opinion: the Aimster case.

And, second, I can’t comment about any current political campaigns, the presidential or any other, or candidate, though I can make a nonpolitical observation about the upcoming election–and as it happens a technological observation. It is to repeat a point I made in my book on Bush v. Gore and the 2000 election fiasco (Breaking the Deadlock is the name of the book): it’s that technological fixes for the sorts of problem that emerged in Florida in the last election, the simplest fix being to replace the punchcard ballot (especially when the votes are counted at the county rather than precinct level) by the optical-scanning (Marksense) system, are likely to be opposed by incumbents. Incumbents have adapted to whatever system they were elected under and are reluctant to take their chances with a new system. This is one example of the uneasy relation between law and technology, and another is the indifference (at least until 2000) of most students of election law to the serious technological issues involved in our election methods.

Tim having vacated the premises early, I thought I should accelerate my entry, especially to correct his statement that “Eight Heads in a Duffel Bag” is my favorite movie. It was my favorite movie, but that was before I saw “The Matrix,” which is my current favorite (though just the first of the three Matrix films), and which seems to me a portent of one of the directions in which technology is moving us. The social and legal impact of technology is going to be the principal theme of my week as Larry’s guest blogger.