Tuesday, May 31, 2005

Is There a Hole in the First Sale Doctrine?

Section 109 of the Copyright Act contains what is commonly called in the United States the "first sale doctrine": if you own a lawfully made copy of a work you can resell that copy (and publicly display it without liability; there is no first sale limitation on the reproduction right, the right to prepare derivative works, or on the performance right). Elsewhere in the world, the doctrine is known as the "exhaustion doctrine" on the theory that the copyright owner, having parted with title to the physical object, has exhausted rights with respect to that copy. Because the doctrine in the U.S. applies to copies that are given away, "exhaustion" is the better name, but "first sale" is what is used, so I will use it too.

The first sale doctrine is essential to libraries and to a functioning resale market, like secondhand books and CDs; its application to the on-line environment has been the subject of much debate. Although now statutory, the first sale doctrine's origins lie in the common law's dislike for restraints on alienation, Bobbs-Merrill Co. v. Straus, 210 U.S. 339 (1908).

This posting deals with the relationship between the right to import copies in Section 602 and Section 109. Last week, Judge Cote in U2 Home Entertainment, Inc. v. Lai Ying Music & Video Trading, Inc., 2005 U.S. Dist. LEXIS 9853 (S.D.N.Y. May 25, 2005) found that the importation and sale of unauthorized reproductions of certain Chinese language films violated rights held by the exclusive U.S. distributor. The case was correctly decided and is fairly straightforward. It is the type of opinion one would read through quickly and nod your head, including at this sentence: "[T]he importation of copies into the United States of a work manufactured in a foreign country can form the basis for a copyright infringement claim by an exclusive licensed U.S. distributor without regard to the first sale doctrine."

The highlighted language here might not give people pause because the copies were made without permission, thus the first sale defense, which is limited to ownership of lawfully made copies, doesn't apply. But the entire quote is not so limited, that is, it is not limited to authorized or unauthorized copies, and that is where a fascinating unresolved question rears its head.

In Quality King Distributors v. L'anza Research Int'l, Inc., 523 U.S. 135 (1998), the Supreme Court construed the interrelationship among Section 106(3) - the exclusive right to distribute copies - Section 109, and Section 602, in addition to construing the meaning of the language in Section 109, "lawfully made under this title." The short answer to how these sections work is this: unauthorized copies made overseas can be seized by Customs and prohibited entry. Lawfully made copies overseas imported into the U.S. in violation of a U.S. distributor's exclusive distribution right (so-called "parallel imports") cannot be seized by Customs but can be barred by a federal district court.

Quality King dealt with a "round-trip" scenario: copies made in the U.S. but not sold here, sent overseas, and reimported and then sold in the U.S. The Supreme Court held these copies could not be barred by anyone since the first sale doctrine applied: the copies had been lawfully made in the United States. The round trip overseas was an irrelevant factual and legal point. Once lawfully made here, the copyright owner's rights were exhausted as to distribution in the U.S. of those copies.

I thought then and still think that is correct. The Court reached its decision by holding that "lawfully made under this title" in Section 109 means lawfully made according to title 17, which is not extraterritorial: thus, Section 109 is limited to copies made in the U.S. But here is the potential loophole created by this straightforward construction of that language. Copies made overseas are not subject to the first sale doctrine. People usually limit this principle to parallel imports, where its application is non-controversial.

But why is it so limited? Let's say someone in the United States buys, through Amazon.com in the United Kingdom, the next installment of a hot series, like the Harry Potter books, released in the UK, but not in the U.S. The copies come into the country without the U.S. distributor knowing about it. They are then advertised for sale once in the U.S. The importation right isn't helpful because the copies are already in the country. Does the first sale doctrine excuse the sale, or does it not apply because, as Quality King held, "lawfully made under this title" applies only to copies made in the U.S.? Since these copies were made in the UK, they weren't made under title 17, and thus fall outside Section 109. The first sale doctrine thus doesn't apply, and as a straight statutory matter it would appear the copyright owner could bar the sales.

Thursday, May 26, 2005

Why there is no Copyright Royalty Tribunal

After the Grokster debate in Palm Springs (see earlier posting), Judge Kozinski asked me what happened to the Copyright Royalty Tribunal (b1976, d1993). Here's the story.

Compulsory licenses aren't really licenses in the sense that copyright owners are compelled to "give" them to those wishing to use a copyrighted work. "Statutory" license is a better (although still not ideal) term: the statute places a limitation on copyright owners' rights in the form of terms and payments that a qualifying entity must meet and make. If the appropriate documents and money are timely deposited with the Copyright Office, as far as the user is concerned, that's it. The user doesn't have to negotiate with the copyright owner. (Voluntary licensing is though the norm in the music industry).

What happens to the money deposited with the Copyright Office? It is sent to the U.S. Treasury for later distribution. How the money flows back to copyright owners and how royalty rates were adjusted was the job of the CRT. But the CRT wasn't necessary until the 1976. Until the 1976 Act, there was only one statutory license, to make mechanical reproductions of non-dramatic musical works. The rate was set in the statute at 2 cents (and it stayed there for 69 years). Since the license was work-specific, there was no problem tracing payment to the correct copyright owner.

This changed in the 1976 Act, when three new non-work specific statutory licenses were added: Section 111 for cable retransmissions, Section 116 for jukeboxes, and Section 118 for public broadcasting. Because these licenses covered all works within the defined class, one couldn't trace a particular "license" to a particular work as under the 1909 Act. At the same time, Congress decided that adjusting rates in the statute was not the way to go: too much time would go by, there had to be bills introduced, hearings, haggling, the whole political thing. Better to fob the job off to someone else (and then Congress could come in as the white knight and save the day if the decision was too favorable to one side: bad agency, bad agency!). From 1976 until 1993, that someone else was the Copyright Royalty Tribunal. The CRT was given the responsibility both to periodically adjust rates according to loose statutory factors, and to decide who should get how much (if the parties couldn't agree among themselves).

The CRT was originally a five member tribunal, in the legislative branch, with members appointed by the President. Other than the very first chair, Tom Brennan, the former chief counsel of the Senate IP Subcommittee, who helped write the legislation that created his chairmanship, no subsequent permanent CRT member had any real experience with copyright law; some had none; some weren't even lawyers (there were former boxers, for example, and I don't mean Barbara or the dog). As one Senator described the situation, the CRT was a dumping ground for unqualified people to whom the President owed a small favor.

The first years of the CRT were marked by incessant appeals of its decisions. It was only the deep compassion of the DC Circuit in affirming decision after decision that kept the CRT afloat. Eventually, with the initial rate setting and challenges out of the way, the CRT was reduced to three members, but even three was too many; even one full-time member was too many. In data I compiled for its abolition, I discovered that on average, the CRT had only two weeks of hearings a year. This cried out for ad hoc adjudication. That opportunity came in 1993.

In 1993, as a result, in my opinion, of the appointment as chair of the then-wife of a former Congressman from Nebraska, the CRT simply imploded. It was a soap opera worthy of prime-time television, but for the substantial amounts of money that were involved. We didn't set out to abolish the CRT, the CRT invited it. I was minding my own business in the subcommittee's offices in the Cannon House Office Building. We began to get visits from different CRT members and the General Counsel, complaining about the fights that were destroying the place. For example, two members (a majority) would vote to change a regulation and send it to the Federal Register. The Federal Register would send it back because the change had not been transmitted by the chair; the chair, who was on the losing side of the vote, refused to transmit it. Moreover, we were told by the private sector that members were lobbying the private sector ex parte. No agency can function like that; the CRT was dangerously out of control.

We told the CRT to clean up itself up, or we would. They didn't, and we did. We did it by abolishing the tribunal and replacing it with ad hoc arbitration panels called "copyright arbitration panels" or "carps" (originally it was "copyright royalty arbitration panels," but the acronym didn't seem right). In addition to fitting the staffing to the workload, we also had the idea that if we required the parties to pay the arbitrators' fees (which would be high), we would encourage settlement and maybe even approximate market rates, which we thought was desirable; after all a "compulsory" license need not be a subsidized license, it could be a license that merely eliminates transaction costs. There is no reason, for example, why copyright owners should subsidize cable operators.

We were wrong the carps would encourage settlement, and for a few reasons. First, we should have tightened up standing issues and the process to ensure that small holdouts could not cause the whole carp mechanism to be invoked. In our defense, I observe that we had zero assistance or input from the private sector, despite requesting it. (There was an ongoing CRT proceeding). A second flaw is that we bought into the belief that companies will act in an economically rational way: if litigating is much more expensive than settling, companies will settle. This didn't happen, in part because of the process flaw, but also for other reasons.

At the end of last session, the carps were replaced with full-time government "Copyright Royalty Judges" appointed eventually to six-year terms (the initial appointments are staggered). The process was also improved. The new regime certainly saves the private sector a lot of money by putting the decisionmakers on the government payroll at government salaries. Whether it results in fewer or shorter disputes, and whether it results in better decisionmaking cannot be known by anyone. But as the number of compulsory licenses increase (and calls for more are a constant drumbeat), a properly functioning royalty setting and distribution mechanism is in everyone's interest and worthy of our attention.

Wednesday, May 25, 2005

The Constitutionality of GATT Restoration

Yesterday, the DC Circuit affirmed the district court's decision in Luck's Music Library, Inc. v. Gonzalez. The district court had upheld the constitutionality of Section 514 of the 1994 Uruguay Round Agreements Act; that is, 17 USC 104A. The DC Circuit's action follows on the heels of a similar decision by Chief Judge Babcock of the District of Colorado on April 20, 2005 in Golan v. Gonzalez.

This post will not attempt to convince anyone of the merits of the dispute over the constitutionality of the GATT restoration: the lines are too well drawn and hashed over in litigation, articles, and blogs. I shall merely supply some background as a participant in the drafting of the provision. I previously gave some of that background in my posting on bootlegs.

The origins of GATT restoration are found in the decision by Congress in 1988, when passing the Berne implementing legislation, not to provide, as Article 18(1) of that Convention requires, protection for "all works which, at the moment of its coming into force, have not yet fallen into the public domain in the country of origin through the expiry of the term of protection." This refers, in our case, to retroactive protection solely to the works of other Berne members. Berne, after all, is a law for foreigners; Berne doesn't care how a country treats works of domestic origin.

The purpose of Article 18(1) is straightforward: if a country has been outside the club (like the U.S. had been), when it joins the club, it has to provide, retroactively, protection to the works of existing club members (that is, all Berne members other than the U.S. itself). This requirement ensures that those who have remained outside the club don't benefit from having stayed outside. Article 18 benefits U.S. copyright owners when other countries join Berne after our adherence (March 1, 1989) and those countries haven't previously protected U.S. works.

The decision of Congress in 1988 not to amend our law to comply with Article 18(1) was made in the face of a requirement that when a country joins Berne, its laws must comport with the treaty's requirements. This makes sense: a country shouldn't be able to join a treaty, get the benefits that the treaty confers for your own authors, but not provide those benefits to the works of other countries' authors. This is, though, exactly what the U.S. did in 1988. Our only excuse for not protecting foreign works retroactively as Article 18(1) requires is that doing so posed "difficult issues." Well, so what?

The chickens came home to roost in the 1994 GATT Uruguay Round Agreements, when the U.S. wanted Berne standards incorporated into the TRIPS portion of those agreements. And the real kicker was the ability to haul miscreants before a WTO panel which could, in turn, order cross-sector penalties (as later happened with the 1998 amendment to 17 USC sec. 110(5)). In 1994, the United States Trade Representative's negotiators told me (as copyright counsel to the House IP subcommittee) and the subcommittee's chief counsel (Hayden Gregory) that if the GATT implementing legislation did not include a provision complying, at long last with Article 18(1), the European Union had indicated the first WTO panel convened would be over that issue. We believed them, and properly so. But even aside from this, I wanted to include a retroactive provision in GATT because I thought that we were wrong to have joined Berne on the sly and, therefore, that we had an obligation to rectify the mistake made in 1988. I still believe that: if Article 18(1) was too big a pill to swallow, we shouldn't have joined Berne.

These then were the reasons for the inclusion of current 17 USC 104A. (There had been a 104A in the NAFTA Agreement, but limited to Mexico and Canada; that provision was repealed by the GATT legislation). Congress was quite aware of the constitutional issues raised by retroactivity and, as noted in the bootleg posting, the IP subcommittees held a joint hearing at which the issues were fully explored. We attempted to take criticisms raised at the hearing into account (including those of Eugene Volokh) in the "reliance party" provisions.

I did not believe that the legislation provided a direct incentive for the creation of new works, nor that, divorced from the Berne obligation, restoration represented good policy. But honesty in treaty adherence is good policy, and retroactive protection for U.S. works overseas is of benefit to both U.S. copyright owners and to the system as a whole. I thought those two objectives sufficient, from a policy and constitutional standpoint. And I say this as someone who believes, more fervently than even Larry Lessig, that the "Promotion of the Progress of Science" language in Article I, section 8, clause 8 is a real substantive limitation on Congress's power. But I also believe that one must look to the system as a whole to see whether there has been a benefit provided by legislation, and in the case of GATT restoration, I answer that question affirmatively.

I respect that others disagree, but I would add that any disagreement over GATT should be divorced from term extension: term extension was four years off, and to Hayden's and my credit, when we were approached about term extension in 1994, we rejected it out-of-hand. I thought then, and still think, term extension was very bad policy. The Supreme Court, I believe, also regarded term extension as dubious policy, although a majority thought it passed constitutional muster.

This is not to say that one cannot disagree with both GATT and term extension: I only ask that two issues which are very different, temporally, substantively, and policy-wise, should be considered separately.

Destruction of Works of Visual Art

On May 13, 2005, Judge Deborah Batts issued her fourth opinion in Board of Managers of Soho Int'l Arts Condominium v. City of New York et al, 2005 U.S. Dist. LEXIS 9139 (SDNY)(prior decisions are at 2004 U.S. Dist. LEXIS 17807 (SDNY Sept. 8, 2004), 2003 U.S. Dist. LEXIS 13201 (SDNY July 13, 2003), and 2003 U.S. Dist. LEXIS 10221 (SDNY June 17, 2001)). The case raises important issues in perennial battles over public art, in particular whether works of art must remain publicly installed, as well as the perennial battle between art and commerce given the desire of the building owners to use the space for advertising.

This long running saga began in 1972 when defendant Forrest ("Frosty") Myers was asked to create and install a three-dimensional work of art on the northern wall of a building at 599 Broadway, Manhattan. (The building is a 12 story loft-structure located on the west side of Broadway between Spring and West Houston streets). The work is a non-representational, minimalist creation consisting of four-foot aluminum bars bolted perpendicularly to 42 steel braces, with the wall on which it is mounted being painted in various colors. The work was commissioned by City Walls, Inc., a private not-for-profit organization. There was no written agreement between the then-owner of the building (Charles Tannenbaum) and anyone regarding ownership of rights in the work or regarding title to the physical object. The work was installed in late 1973.

In 1987, the building's owners petitioned the Landmarks Commission to remove the work, but the request was withdrawn after the Commission said the work was an integral part of a landmarks protected building. In 1997 permission was again sought to remove the work. The Commission approved an interim removal due to its unstable condition, and the work was removed. The building owners then asked for permanent removal; permission was denied, but the work has not been reinstalled. Litigation was then brought under a wide variety of theories. Counterclaims were filed. The case became a cause celebre, and remains one. This post only concerns claims under the 1990 Visual Artists Rights Act (VARA).

Beginning in 1979, members of the Massachusetts Congressional delegation began introducing bills to give visual artists rights to prevent the distortion or mutilation of their works. VARA is, however, an offshoot of efforts by film directors to obtain droit moral ("moral rights") in the 1988 Berne Convention implementing legislation. Film directors were unsuccessful, but Congress did subsequently pass VARA, granting very limited, hobbled rights to a defined class of "works of visual art" (defined in Section 101). The rights are found in Section 106A. The definition makes clear that VARA is, in large part, really a preservation act, more concerned with preserving the original and limited edition copies than with giving the artist true moral rights. For example, there is a signing and numbering requirement, something totally inconsistent with Article 5(2) the Berne Convention. And aside from the limited category of subject matter covered (not all pictorial, graphic, or sculptural works), the rights don't extend to mutilations of reproductions. VARA is the Mini-Me of moral rights laws.

One way VARA is broader than Berne, though, is in providing a right against destruction of copies of a work of recognized stature (Section 106A(a)(3)(B)). This right was at issue in Judge Batt's case, in particular, whether the work there had been destroyed or removed. (Two other cases which confronted this issue in the SDNY are English v. BFC&R East 11th Street LLC, 1997 U.S. Dist. LEXIS 19137 (SDNY Dec. 13, 1997 (Baer, J.), aff'd summary order, 1999 U.S. App. LEXIS 23697 (2d Cir. Sept. 27, 1999) and Flack v. Friends of Queen Catherine, Inc.,139 F. Supp.2d 526 (SDNY 2001)(Stein, J.).

In her most recent opinion, Judge Batts reviewed Woody Allen, Annie Hall-like testimony about what destroyed versus removed means, including variations such as "disbanded," "not existing now," "recreated." ("Dismantled" is one that comes to my mind). Judge Batts resorted to the safety of various dictionary definitions in her earlier opinions, with "remove" being defined as "to move from a position occupied ... to convey from one place to another," and "destroyed" being "to tear down or break up."

Why did this matter? Recall I said VARA is really a preservation statute. It is. Congress wanted to preserve works of art as enduring parts of our culture; that explains the limitation of the destruction right to works of recognized stature: lesser works can be destroyed without any loss to society. If VARA were truly a moral rights act, the interests of artists in lesser works would have been protected too in this regard. But they weren't.

Art installed in buildings presents special challenges. If a work of art is installed in a building, does this mean that it has to remain there forever? Does it mean that the building can't be torn down? An affirmative answer might well provide an incentive to building owners to stay a country mile away from art. Yet, depriving installed art of any rights would be a major exception from the very concept of VARA as a preservation statute.

I worked on VARA while a Policy Planning Advisor to the Register of Copyrights. (The principal drafters were House IP counsel and John Podesta, later President Clinton's chief of staff. I first met John when he was a staffer for Senator Leahy and he remained a strong advocate for artists after leaving the Senator). Unlike other issues in VARA (e.g., whether there could be a waiver of rights, a sticking point between the House and the Senate), how to deal with installed art was not the subject of real disagreement.

The basic approach is this, found in 17 USC 113(d) : if an installed, post-enactment work of protected art can not be removed without being destroyed, unless the artist and the owner of the building agree in writing that the work can be destroyed anyway, it can't be removed. This places quite a burden on building owners, but as between them and artists, it was felt building owners are better represented and have more leverage. If the work can be removed without being destroyed, then the artist is given an opportunity to have it removed. If the artist doesn't avail him or herself of that opportunity, then the work can be removed without liability. There is thus no right to keep a removable work of art installed over the building owner's objections and efforts to contact the artist. I spent countless hours with the talented House Office of Legislative Counsel attorney Sandy Strokoff trying to articulate clearly in statutory language what was a non-controversial policy, but in hindsight I think the drafting, while not in the least ambiguous, could have more elegant in the sense of being more simply expressed.

The distinction drawn between "removed but destroyed" and "removable and not destroyed" was at the heart of the VARA claim in the Myers case. Judge Batts found the work removed and destroyed. There are some other wrinkles in the case, though. Congress faced takings issues if it protected, retroactively, pre-enactment works. Myers' work is such a work. One way around a takings problem is to protect the subject matter but to excuse any pre-enactment conduct. VARA does this, but the removal in Myers' case occurred post-enactment. If Congress extended rights to pre-enactment subject matter but reached post-enactment conduct, takings problems are still there in the form of expectations that having owned the work (or the physical copy) at a time when there was no liability one could continue in futuro to act without liability. (This is also an issue for GATT restored works).

For pre-enactment installed works like Myers', our solution was as follows: if, as in Myers' case the work was installed with consent before the effective date, 113(d)(1) does not apply in the event the work was destroyed by removal. This is what Judge Batts held. (Since there is no right to prevent removal of a work that can removed without destruction, 113(d)(2) is of no use either in Myers' case).

There is another point on retroactivity. The retroactivity section of VARA is contained in the effective date provision and is not codified in title 17. (This is almost always the case for effective dates). Instead it is found only in the Public Law version, here P.L. 101-650, section 610(b):

"The rights created by section 106A of title 17, United States Code, shall apply to--
(1) works created before the effective date set forth in subsection (a) but title to which has not, as of such effective date, been transferred from the author, and
(2) works created on or after such effective date, but shall not apply to any destruction, distortion, mutilation, or other modification (as described in section 106A(a)(3) of such title) of any work which occurred before such effective date."

What does "title" here mean? Representative Kastenmeier, the floor manager, explained that the term "title" referred to the physical copy of the work, not to title to the copyright. 136 Cong. Rec. H13314 (daily ed. Oct. 27, 1990). As reported out by the House Judiciary Committee, the legislation had used the term "copyright" in place of "title." H.R. Rep. 514, 101st Cong., 2d Sess. 5, 23 (1990)(There is no Senate report). The change from "copyright" to "title" was designed to eliminate any remaining takings issues, but it also helps artists: an artist could have transferred copyright in a pre-enactment work and transferred title to some copies but not all. As to those where title to the physical object has not been transferred, VARA applies.

In whom title vested in Myers was subject to confusing, inconsistent testimony, at least to me. In her recent opinion, Judge Batts stated that the building owner takes the position that not it, but City Walls has always possessed title. That seems to help the artist on the effective date problem since City Walls supports him. But then there is another question: who is the author? This question is relevant for two reasons. First, the effective date refers to title being transferred from the author. If City Walls is the author then it couldn't have transferred title to itself and the work isn't knocked out on effective date grounds (although as we shall see it is on another ground). But if Myers is the author and it is true that title is in City Walls, then title was transferred by the author, and there is no VARA protection. Thus, Myers can overcome this particular aspect of the effective date problem only if he is the author and he didn't transfer title to the physical object.

The second problem is this: there is an exclusion from VARA coverage for works made for hire. City Walls commissioned the work under the 1909 Act which, in the Second Circuit had a very liberal view of work made for hire. If City Walls is the author as a work for hire, there was no transfer of title but no rights.

Monday, May 23, 2005

Grokster Debates

On Friday (May 20th), I moderated a debate on Grokster at a joint annual meeting of the San Francisco and Los Angeles IP associations in Palm Springs. It was 105 degrees: appropriately for the temperature, on the panel were Judge Alex ("Easy Rider") Kozinski of the Ninth Circuit (and one of the Article III Groupie's Super Hotties), Stanley Pierre-Louis, Director of Litigation for the RIAA (and top on anyone's list of people you would want to hang out with), and Michael Page, who represented Grokster at the lower court level.

In November, at the Copyright Society of the U.S.A. in New York, I had moderated a similar debate between Judge Kozinski and Judge Posner, billed as the "Transylvanian Terminator versus the Chicago Crusher." That bill lived up to its name. It was a wild, wildly amusing, over-the-top, mano-a-mano theoretical combat between two of our most insightful judges. Judge Kozinski's libertarian streak was on full display, and that had the effect of making Judge Posner appear very conventional. The Kozinski-Posner debate took place before the oral argument in Grokster and in Palm Springs I was interested to see how the oral argument might have changed crystal ball looks at the outcome before the Supreme Court.

I tried to set an appropriate tone in Palm Springs by wearing my Buttshark t-shirt, but the rudeness of the deadly-boring speaker before us in not ending anywhere near on time (droning on in yet another useless Power Point presentation consisting entirely of text that was read verbatim), and the absence of Judge Posner had the effect of dampening Judge Kozinski's naturally infectious nature. Presumably too, beating up on lawyers poses an unfair advantage for him, as well as being a bit of a busman's holiday.

The bottom line Palm Springs crystal ball for the Supreme Court's outcome was this: unlike past Supreme Court opinions, where one would race to the end to read "reversed" or "affirmed" (the last rare in the case of the 9th Circuit, but cf. Stewart v. Abend in which the Supreme Court favored the 9th Circuit over the 2d Circuit), in Grokster the view expressed was that such a label would be of considerably less consequence. One reason for this is that a reversal is unlikely (in my opinion), so the spin will be over the nature of the affirmance (if that is what happens).

I say this under the assumption that the inducement part of the case is in large part not (and maybe not at all) before the Court as a discrete ground. That is, if it is true that had the Court not granted certiorari there would still be a proceeding before the trial court on past alleged inducement conduct ("use our software and please commit copyright infringement") then the Court can't "reverse" on inducement grounds, and that was the most likely ground for reversal, in my opinion.

Indeed, while there was confusion on this point at oral argument, the real focus was on the "Sony" part of the case, i.e, the staple article of commerce problem. On this central point, there are a number of problems, and I note that both sides moved for summary judgment so the idea that the Sony part would be sent back for trial is unlikely.

The Court could reverse under any approach adopted. The Court could affirm the result (no liability), apply a "straight" Sony test (whatever that is) but trash the Ninth Circuit's interpretation of Sony)(which I expect, and think is a trashing well-deserved). The Court could modify Sony to fit the Internet. Or, the Court could announce new standards (hence Scalia's admonition to forget about stare decisis). Under either of these last two, the court could vacate in light of the new standards and ask the Ninth Circuit to reconsider, but given the summary judgment posture I don't see why they would do that rather than decide the case themselves, unless they thought summary judgment shouldn't have been granted at all. Justice Souter might have that view, given his statement that it was bizarre to separate the inducement part of the case. And it is entirely possible that enough Justices have discomfort with the posture of the case and the state of the record that such an approach provides an easier route than dumping it because "we made a mistake in taking this case" or the 4 to 4 split in Lotus v. Borland. If I were voting, I would attempt to craft a modified Sony test, vacate the summary judgment, and tell the parties to fill in the gaps in the record.

Aside from questions about the posture of the case, much of the oral argument centered around what "test" to use. The copyright owners' didn't help themselves by not helping the Court. Here are some excerpts from the transcript of questions asked the copyright owners' counsel: Justice Kennedy: "It's not clear to me from your brief what your test is" (p. 9); after an effort to explain one, Justice Kennedy then stated, "Well, I don't quite understand." (p. 15); Justice Souter, after trying to pin counsel down: "You're not answering the question." (p. 15); Justice Scalia on counsel's confusing the inducement and the Sony arguments: "I think that's your second argument. I thought you were going to stick with the first one. I mean, that's an inducement argument," to which Justice Souter added, "Yeah, that's inducement" (p. 16). Justice O'Connor, adopting Capitol Hill lingo, straight out announced, "I don't understand your pitch." (p. 17).

The P2Ps' counsel was much more direct, advocating application of the "clear" Sony test, which got him in trouble with Justice Ginsburg, who found the test far from clear. But at least the Justices knew where he stood. Unfortunately for him, it is probably not a place that will command 5 votes, and some, such as Justice Kennedy, were downright hostile.

Where does this leave the Court? Acting like common-law judges, which is to say making law, and by themselves, not in public, which is why I thought before cert. was granted that the Court was a dangerous venue for the copyright owners. My belief, expressed for many months before the argument, was that there are not five votes for reversal, but not five votes for applying Sony for a straight affirmance, and instead a murky area in between, murky enough that neither side will be able to declare a full victory and therefore not enough to require Congressional intervention either. The lawyer panelists in Palm Springs agreed this is a possible scenario, after of course expressing optimism for a decisive victory.

Judge Kozinski favored Congressional guidance. Leaving aside the unlikelihood that Congress would get involved after a murky decision, I question how much guidance Congress can give. Certainly Congress can resolve discrete issues, like the split between the 7th and 9th Circuits over whether one looks to present uses or whether one can factor in possible future ones. Congress can also flesh out the knowledge requirement (but how is an intensely political issue even more than a drafting one). Beyond that, unless there is to be a detailed DMCA safe-harbor-like exemption (and that is as likely as the Democrats developing a backbone), there is little Congress can do that will be of help in future cases.

In short, the area may well remain as murky as it is now.

Thursday, May 19, 2005

Bootlegs Continued

Thanks for the many fine comments on yesterday's blog. I will use this posting to reply to them (as some requested). I have, however, an unrelated question I would love some feedback on. I have a 4,000 page hard copy (double spaced text, single space footnotes) new treatise on copyright. BNA was going to publish it, but we parted ways over format after 13 chapters were in page proofs. BNA wanted it in two bound hard copy volumes, no looseleaf, bound annual updates, no commitment to ever putting it online, and CD ROM Appendix only (but no CD ROM for the text). I thought that was a bad idea. I have since pitched it to another publisher, and got a good reaction in February. Since then, however, radio silence. I am fed up. As time goes by, the need to update a work that isn't even out yet increases. I am seriously thinking of publishing it myself in a purely on-line format, leaving legal publishers to their dinosaur world. The work would be formatted like a hard copy book and have page numbers. It would be searchable, and you could cut and paste. I would sell it by a flat annual subscription so that a law firm would pay one price and everyone within the firm could use it; ditto a law school and its faculty and students. There would be free updates, probably once a week. Is this a good idea, a good idea but too difficult time-wise and technically? Or a terrible idea. Any comments would be most appreciated. I am particularly concerned with any unwillingness to cite the work in briefs or court opinions. (My idea on having it formatted like a hardcopy book would take care of how to cite it).

On to bootlegs. Aaron Perzanowski takes me to task for supposedly telling only half the story, the fixation part, and not telling the distribution part of the statute. (Martignon was accused of only the latter). I disagree: I carefully set out all three possible violations: fixation, transmission, and distribution; indeed since the transmission prong did not involve a fixation I asked whether that provision might be regarded differently by critics.

Aaron's substantive criticisms involve his statement that live musical performances are "original and expressive." On this ground, he sees no difference between performances fixed with permission and those fixed without permission. He adds a quote from a Supreme Court opinion that writings include any "physical renderings" of creativity. He concludes that because the statute bans selling bootlegs in a physical rendering it is within the Copyright Clause. And because it is perpetual it is unconstitutional.

In my opinion, Aaron confuses a number of important points. First, originality is irrelevant. The bootleg statute doesn't care if a performance is original or not, no more than the trademark statute (another perpetual federal intellectual property statute) cares if a logo or other mark is original or not. More importantly, he confuses the subject matter of the bootleg statute with its violation. The subject matter of the bootleg statute is not the illicit copy; it is the live musical performance. The Act is violated, in one prong, when there is an illicit copy made, but the fact that there is a violation doesn't mean that the violation is what is protected; its not. The subject matter of the Act is an unfixed live musical performance: that's what you have to measure up for constitutionality. And when you do, there is no conceivable way that it falls within the ambit of the Copyright Clause because that Clause is limited to "writings." "Writings" means "fixed" subject matter: unfixed live performances are not writings. The only way to make the subject matter of the bootleg statute a writing is to impermissibly shift the subject matter from what it truly is to what is a violation of the Act.

In my opinion, Aaron further confused this point by positing equivalence between "fixation" and "physical renderings" (as the Supreme Court used that final term.) The two concepts aren't the same. What the Supreme Court meant is that if there is a creative work we are not going to judge the merits of its creativity. Aaron wants the passage to mean that any time someone fixes something, even without permission, there is a "physical rendering," therefore a fixation therefore it falls within the Copyright Clause. That's not what the Supreme Court meant and it would be bad policy. It would mean that if a baseball player's image is photographed without permission and put on baseball cards, the Copyright Act applies and state right of publicity laws are preempted. That's why I said yesterday the professors' brief fails to distinguish between the subject matter and the copy. Aaron's argument is the same; he just adds the additional point that a live musical performance is original, to which I say, so what? The subject matter is not an original work of authorship, and no defendant can make it one by fixing it or distributing an unauthorized fixation.

Brian Carver and Aaron both pointed out an inconsistency in the drafting: since the statute is in title 17, and therefore uses the definition of "fixed" in Section 101, you have an "unauthorized authorized" embodiment. I think they're right and I wish now that we had left the statute in title 15 for that reason (and many others). Brian also says though that in making my fixation swinging door comment I didn't address contrary arguments. I think I did, but apparently not clearly or persuasively enough. Here's my argument: the dormant effect of the Copyright Clause only extends so far as encompassing "writings" since that is the reach of the Clause. It doesn't impact on things that aren't writings. Unfixed live musical performances aren't writings and therefore are unaffected by the Copyright Clause.

To me the dormant argument says, we're not going to let Congress wear the Commerce Clause hat and screw up the Copyright Clause, say by protecting unoriginal works. We do that because if we didn't we would be interfering with the purposes of the Copyright Clause. But since the Copyright Clause has no purpose with respect to unfixed works, nothing Congress does with respect to them can interfere with the Copyright Clause. That's my argument. I accept that others may disagree.

Brian also notes that I didn't talk about the First Amendment issue in the professors' brief. I didn't for two reasons. First, the issue is not particular to bootlegs and applies equally well to fixed copyrighted works. Second, I have never given it any credence with fixed works. I don't think the First Amendment gives others a right to use my speech, where speech is defined as the substantial taking of expression and without an available fair use defense. The courts agree. See furthermore Harper & Row, which expressly adopted my view that if the First Amendment is in play in copyright litigation, then there are important First Amendment values vesting in the author's right to remain silent.

Finally, Brian asks for some more citations to legislative history. I wish they existed. As a fast track bill controlled in the House by the Ways & Means Committee (and its equivalent in the Senate), there was no Judiciary Committee report. Indeed, as I noted in yesterday's post, the Clinton Administration's USTR wanted no hearing, no published bill, a complete blackout. I can't speak for Senators, but I do know that Senate Judiciary staff understood it was a Commerce Clause provision (I spoke with the Senate IP subcommitte's chief counsel almost every day then), and so too did my bosses in the House.

Aside from the constitutionality issue, Anonymous states that it was a mistake not to put a time limit on protection. In hindsight, I agree.

Finally Fred von Lohman (I am a huge fan of his; he has done sterling oral advocacy in the 9th Circuit), asks whether we considered amending the simultaneous fixation provision in the Copyright Act instead. We didn't because it would have required all live performances to be simultaneously transmitted. We knew some were (hence the second prong) but it seemed too much to actually require all of them to be. Plus, we also would have had to amend the definition of "fixed" to include unauthorized fixations . But then we might have avoided the inconsistencies noted above.

Wednesday, May 18, 2005

The Constitutionality of the Bootleg Statute

A general note about posting comments: the default was, unknown to me, set to require a registered account. I have changed it so anyone can post a comment. I encourage all comments and I hope this posting will generate more than a few.

While I am a full-time litigator in private practice, I have a keen interest in statutory interpretation, having spent eight years as a copyright attorney in the federal legislative branch, including a stint as copyright counsel to the U.S. House of Representatives IP subcommittee. I have drafted a fair number of bills, including quite a few that became law. Three that became law were included in the 1994 GATT implementing legislation. Two of those three concern the unauthorized fixation, transmission, or subsequent sale of unauthorized live musical performances. (The third is the 104A restoration section in the same legislation). This posting concerns the bootleg provisions.

There is a civil provision in 17 USC 1100 and a criminal provision in 28 USC 2319A. The heart of both provisions is the same: a violation occurs when, without the permission of the performer(s) someone engages in: (1) a fixation of a live musical performance, or (2) a transmission of such a performance to the public, or (3) a distribution in copies or phonorecords of (1).

These provisions were inserted into the GATT legislation to comply with TRIPS obligations (Article 14(a)), but the means of compliance as well as the details were a matter of U.S. law. Hayden Gregory (the subcommittee's outstanding chief counsel) and I spent alot of time hashing out the best way to draft appropriate legislation. We carefully and conscientiously reviewed the constitutional issues of legislating under the Copyright Clause or the Commerce Clause. We consulted with the Justice Department's Office of Legislative Counsel, the Copyright Office, RIAA, and academics. Our decision was motivated by the need to comply with a treaty obligation and what we believed the best public policy to be. We were, though, open to other ideas and to criticism.

To get such ideas and criticism, the House IP subcommittee then held a joint public hearing with the Senate IP subcommittee in August 1994. The hearing was particularly noteworthy because both the Office of United States Trade Representative (which had general responsibility for drafting the GATT legislation on behalf of the Administration) and the private sector did not want a hearing on the legislation, nor did they want drafts of the Administration's version of the IP parts of the legislation circulated to the public at large. Bear in mind that this was fast track legislation, meaning it could not be amended after introduction. What was introduced would be what passed: if you couldn't see it before introduction, you had no way demanding changes. My experience with USTR's heavy-handed conduct convinced me that no President should have fast track authority.

By holding a hearing, and also by persuading Senator DeConcini to "out" the Administration's draft by introducing it as a bill and placing it in the Congressional Record, we struck a blow for open government. I mention this because we believed it important to provide those wishing to challenge particular provisions to be heard before introduction so that necessary changes could be made.

Witnesses at the hearing included government officials, private sector representatives, and academics. At my request, two law professors, one quite prominent testified. While they testified at length on perceived constitutional problems with Section 104A, neither they nor anyone else expressed any problems with the bootleg provisions. (One of the two recently signed on to a professors brief discussed below.) The record was held open after the hearing to receive written comments. There were none on the bootleg provisions.

Given the opportunity we provided anyone to comment on the proposals, it surprising to learn that some who were there, but silent, are now critics. Academics seem particularly prone to public choice theory, perhaps in direct relationship to their lack of real knowledge about or participation in the political process. It is one thing to subscribe to that theory when you are excluded from participating in the process, but it is quite another when you are invited to participate and don't. Those who constantly whine about the effect of the private sector on copyright legislation should look at themselves and ask why, when they had a genuine chance in 1994 to influence legislation so allegedly flawed, they didn't.

I have dealt with the details of the bootleg statutes in a 1994 book published by BNA ("Copyright and the GATT"); this posting only concerns the three court opinions issued since then, United States v. Moghadam, 175 F.3d 1269 (11th Cir. 1999); United States v. Martignon, 346 F. Supp.2d 413 (SDNY 2004); and, Kiss Catalog v. Passport Int'l Prods., 350 F. Supp. 823 (CDCAL 2004). As the captions reveal, the first two are criminal, the last civil.

The legislation was enacted deliberately under Congress's Commerce Clause powers. We did so because of doubt that an unfixed performance was a writing and therefore encompassed by Congress's Copyright Clause power. Evidence of legislating under the Commerce Clause is seen in Section 1101(a), which states that a violator of the right "shall be subject to the remedies provided in sections 502 through 505 to the same extent as an infringer of copyright." Obviously, if we had regarded the right as a "copyright" right (in the broadest sense), the italicized language would have been unnecessary, even wrong. Second, we made the right perpetual, both because we saw no need to limit it and as another obvious sign that we weren't legislating under the Copyright Clause.

In 1994, I had been practicing copyright law for 13 years. I was well aware of the limited Times restriction. Everyone involved was aware of it. Do critics think that in making the bootleg right perpetual we meant to legislate under the Copyright Clause but just had a memory lapse, or that we said, "Hell, let's draft an unconstitutional provision; why not, its bound to be fun?" The answer is, no, we didn't draft a copyright or copyright-like provision at all. We drafted a sui generis right under the Commerce Clause. (For those who are wondering, Congress is not in the habit of saying in a statute, "hey this is the power we are legislating under." See also Woods v. Taylor, 333 U.S. 138 (1948)).

Why then did we place the bootleg provision in title 17? For ease of use of definitions and other administrative parts of that title. We did not put it in the Copyright Act. That Act constitutes only chapters 1-8 of title 17. Instead, we put the bootleg provisions in a new chapter 11, in the sui generis outposts of title 17, which begin with chapter 9, another sign of their non-copyright nature. The placement in title 17 has no substantive meaning at all. Congress could put a bill on the interstate commerce regulation of GPS chips in dogs in title 17 if it wanted to, but that wouldn't make the provision a copyright provision. We could have put the bootleg provisions in title 15 (in fact we did put them in title 15 in an early draft, but we took them out after trademark owners complained), or we could have put them in no part of the USC at all, just made them a freestanding Act; neither decision would have had any impact on the substantive nature of the right.

Moghadam upheld the law as a valid exercise of Commerce Clause power, but it was dead wrong in saying "Congress thought it was acting under the Copyright Clause." 175 F.3d at 1275. The court didn't say why we supposedly thought that, and we most definitely didn't. Before a court makes a statement like that, it should back it up, if it can, and Moghadam didn't. Inconsistently, Moghadam thought that live musical performances weren't writings. If so, why did the court think we were legislating under the Copyright Clause? I find this particularly annoying because the writing issue was the very reason we decided to legislate under the Commerce Clause. The court's inability to grasp the structure of title 17 or even the import of something so simple as the reference in the remedies part of "to the same extent as an infringer of copyright" is disappointing, but then courts know very little about legislation. (What they lack in knowledge they make up in imagination).

Judge Baer's Martignon opinion is quite another story. If one were to grade judicial opinions, Moghadam would get a C- and Martignon would get an "F" (and that applies also to the two law students who apparently drafted the opinion, see 346 F. Supp.2d at 416 n.1). First, Judge Baer gave substantive significance to the placement of the provision in title 17, a fundamental error. And more: the placement in title 17 was deemed by Judge Baer to support the view that the statute is one "directed at protecting the interests of artists, rather than commerce, and therefore further sustains the view that the statute is copyright-like in nature." Id. at 422. Geez, I thought that artists were engaged in interstate commerce, especially live musical performers who go on tour throughout the country and have their performances taped while engaged in such commerce; you know, the very ones the bootleg provisions try to protect. Perhaps performers are like baseball players, but if so they should enjoy the same antitrust exemption.

Judge Baer's next noteworthy error (there are too many to detail in any blog, regardless of size) was his belief that the bootleg provision was both not a copyright provision -- because it did not protect writings -- and a copyright provision because the rights were copyright-like. This confuses subject matter with rights. Even Section 301, the preemption provision in title 17, doesn't make that error: states can grant identical exclusive rights to those granted in Section 106 and 106A so long as the subject matter is not the same. (Conversely, states can grant non-equivalent rights in copyright subject matter.) The subject matter of the bootleg provisions is not the same as those protected by the Copyright Act; it can't be, there is no writing. Judge Baer's collapsing of non-copyright material into copyright material is a fundamental error.

His dormant Copyright Clause argument misses the whole point: "Congress may not .. enact copyright or copyright-like legislation, which conflicts with the fixation or durational limitations of the Copyright Clause, even if another clause provides the basis for such power because Congress' power to enact copyright or copyright-like legislation... is only as broad as the Copyright Clause allows." Id. at 426-427. The reach of the Copyright Clause stops at the fixation door: it doesn't expressly or impliedly swing open to things that aren't fixed.

This fixation on fixation is also seen in a brief submitted on May 12th by a number of law professors in support of the hapless Judge Baer on appeal, (Law Professors' Brief). The brief makes a number of points. I focus on one: The professors assert that the bootleg provisions are copyright rights because they criminalize (in the case of 2319A) or find infringing (in the case of 1101) fixation of live musical performances in copies or phonorecords. The professors need to brush up on a few fundamentals. One is the distinction between a material object and any intellectual property embodied therein (17 USC 202). The material object is not the intellectual property. The professors err by saying that the subject matter of the bootleg provision is the copy. Its not: the live musical performance is the live musical performance; the copy of the phonorecord in which an individual captures an unauthorized fixation of the performance does not become the live musical performance; the copy remains just a piece of tape or another storage medium.

I note too that the definition of "fixed" in Section 101 is limited to authorized fixations; thus Section 1101 falls outside of the Copyright Act on that ground too. I also note that Section 301 saves from preemption other federal statutes. Moreover, the "transmission right" in the bootleg provision does not require any fixation: does this mean that provision is constitutionally OK?

The Seventh Circuit originally made the same error as the professors in its first Toney v. L'Oreal opinion, discussed in an earlier posting. The panel then reversed, correctly holding that a photograph which served as the fixation for a model's image was not the subject matter for purposes of preemption of state right of publicity laws. The model's image was still her image, regardless of how her right of publicity was violated. Under the law professor's approach, right of publicity laws must be struck down too (under the complete preemption doctrine), even though the Supreme Court upheld them in the Zacchini case. So too all state laws on purely oral works, despite the lack of statutory preemption for them.

The dormant Copyright Clause argument only extends to material that is either subject to copyright, was under copyright, or which conflicts with a clear objective of the Copyright Clause having to do with the operation of that Clause. Live musical performances, having nothing to do with that clause, are not affected.

Tuesday, May 17, 2005

Wine, The Constitution, and Database Legislation

For those who own a "Free the Grapes" t-shirt (and as a Northern Californian from Tiburon living in snowy Connecticut I do), yesterday's 5-4 Supreme Court decision in Granholm v. Heald was cause to uncork a good bottle of wine (red, of course, see accompanying photo). But does the opinion have any relevance to ongoing efforts to enact database legislation under the Commerce Clause? Maybe.

Granholm was a dormant Commerce Clause case (at least that was the question accepted for cert. and quoted approvingly in Justice Kennedy's majority opinion). Unlike the database proposals, which would be federal if enacted, Granholm dealt with the power of states to act in a discriminatory way toward out-of-state shipments of wine, the Twenty-First Amendment (repealing Prohibition), and federal legislation passed before that Amendment.

Nevertheless, the database proposals do raise fundamental questions about the limitation of one constitutional power on another, for example, the Copyright Clause on the Commerce Clause (a "dormant Copyright Clause" perhaps). I explored this question in a 1999 law review article, "The Enumerated Powers Doctrine and Intellectual Property: An Imminent Constitutional Collision," 67 George Washington L. Rev. 359. In that article I first noted the Supreme Court's statement in Feist that there is a constitutional right to copy unprotected material. I wondered whether this meant only when Congress is legislating under the Copyright Clause, or whether it was a broader, structural statement indicating that Congress could not pass legislation extending copyright-like protection under the Commerce Clause. For that possibility, I relied on Railway Labor Executives Assn. v. Gibbons, 455 U.S. 457 (1982), which struck down an attempt to circumvent the constitutional uniformity requirement in the Bankruptcy Clause by legislating under the Commerce Clause.

Some of my colleagues, such as the great Eugene Volokh, disagree with me. But aside from Granholm, subsequent to my article, there have been three opinions on the 1994 bootleg statute (unauthorized taping of live musical performances and the subject of tomorrow's posting) and one on the DMCA that have taken the argument seriously. One, United States v. Moghadam, is a circuit opinion, 175 F.3d 1269 (11th Cir. 1999), involving the the criminal bootleg provision, 28 USC 2319A. The legislation was enacted under the Commerce Clause. Citing Railway Labor Executives, the court of appeals agreed the opinion "suggests that in some circumstances the Commerce Clause cannot be used to eradicate a limitation placed upon Congressional power in another grant of power," id. at 1279-1280, and noting a tension with earlier Commerce Clause cases such as Heart of Atlanta Motel. The bootleg statute passed muster according to the circuit because live musical performances weren't deemed writings and because there was no constitutional purpose of permitting their reproduction, as there is with facts.

The bootleg statute was, however, held unconstituional in two district court opinions, both of which relied on Railway Labor Executives, United States v. Martignon, 346 F. Supp.2d 413 (SDNY 2004)(holding that Congress may not legislate under the Commerce Clause that which it cannot under the Copyright Clause); Kiss Catalog v. Passport Int'l Prods., 350 F. Supp.2d 823 (CDCAL 2004)(same).

By contrast, in United States v. Elcom Ltd., 203 F. Supp.2d 1111, 1138 (C.D. Cal. 2002), a case under the DMCA, even the Government, according to the court "agree[d], that as broad as Congress' Commerce Clause is, Congress may not use that power in such a way as to override or circumvent another constitutional restraint," citing Railway Executives. (The court rejected the argument that the DMCA did exceed Congress' power).

The above cases illustrate, at a minimum, that the relationship of the Copyright and Commerce Clauses is more complex than previous naysayers wish it to be. For example, few would deny that if Congress granted protection to public domain works in a form exactly like that int he Copyright Act, but under the Commerce Clause, the effort would be struck down. The real cases will be much harder, but line drawing at least accepts the principle that there is a problem.

Monday, May 16, 2005

Work for Hire, the Constitution, and Termination Rights

Section 201(b) of the Copyright Act provides that in the case of a work made for hire, the employer or commissioning party is the author; not just the owner of all rights, but the author. There are two important consequences that flow from this distinction: term of protection, and termination of transfer rights, found in Sections 203 and 304. (Lack of moral rights under Section 106A is another but that right is very limited).

The term of protection for a work made for hire is not measured by the life of the author, but from the date of first publication or creation (whichever is shorter), and there is no termination right.

In its expansive treatment of work for hire (all works of all employees created in the scope of their employment, and nine enumerated categories of specially ordered or commissioned works), U.S. copyright law is quite exceptional: most countries for most works follow a natural law tradition, pursuant to which only living, breathing individuals can be authors. This is not to say other countries don't recognize the economic realities of the employment relationship. They do, by vesting most rights in the employer, but the employer is not the author. This distinction between authorship and ownership in the employment field emanates from a natural rights conception of authorship. That conception posits a strong bond between the individual and his or her work, hence the strong continental droit moral tradition.

The U.S. Supreme Court has variously described the rationale for our copyright law. Sometimes the Court emphasizes a utilitarian purpose (bringing works to the market). At other times, the
Court emphasizes the need to protect the fruits of authors' mental and creative labors. These two rationales need not be in conflict, and indeed James Madison believed them to be harmonious. But no one seriously believes that U.S. copyright law has a natural rights foundation. The Supreme Court's 1834 Wheaton v. Peters opinion, finding there is no federal common law of copyright and only statutory rights, eliminated that possibility.

The exclusively statutory basis for copyright has not eliminated all doubts, though, about Congress's power to sweep into work for hire any old arrangement. One basis for doubt is the Constitution, Article I, section 8, clause 8 of which limits Congress's power to grant exclusive rights to "authors."

But who are "authors?" The Constitution doesn't say. Should we take an originalist approach and say authors are those whom the Founding Fathers would have regarded as authors (assuming we could figure that out)? Do we take the Learned Hand "living Constitution" approach and say authors are those who would be regarded as authors under today's standards (assuming again we could figure that out)? Is there any difference between 1790 and 2005 on that score? Or, might we take some other approach, perhaps a law and economics approach (an author is the entity that can most efficiently exploit the work), a contract approach (an author is whomever the parties decide the author is), or should we defer entirely to Congress under the very generous Eldred standard of review?

One view that Congress can't do what it wants came from an unexpected source, an opinion by the legendary Second Circuit Judge Henry Friendly in Scherr v. Universal Match Co., 417 F.2d 497, 502 (2d Cir. 1969). Judge Friendly wrote that "the course of decision has gone past the point of argument" that employers cannot be authors, but he added that it would be "quite doubtful that Congress could grant employers the exclusive right to writings of employees regardless of circumstances." (He did not go further into what those other circumstances might be).

Twenty two years later, another legendary Second Circuit Judge, Jon O. Newman, citing Scherr, stated in dictum that the work for hire doctrine "is not constitutionally suspect." Childress v. Taylor, 945 F.2d 500, 506-507 n.5 (2d Cir. 1991), and presumably he meant under all circumstances. Are we to assume then that authorship is merely a social construct that Congress is free to adapt as it deems the circumstances require? Judge Posner, citing Michael Foucault ("What is an Author?") appears to adopt this view. See Law, Pragmatism, and Democracy 607 (2003); Law & Literature 381 (Rev. ed. 1998). It is a view that might also mesh with Eldred's standard of review.

But it may not appeal to originalists, and to others, Judge Friendly being one example. The enormous economic consequences of termination of transfer rights, particularly those for legacy sound recordings in the 1972-1977 period of Section 304 terminations, for which the more liberal work for hire approach of the 1909 Act governs, as well as for 1978 and beyond works, for which Section 203 governs (actually, an author can have both a Section 203 and 304 termination right if rights in a 1909 Act work were conveyed on or after January 1, 1978) may be sufficient to see a resurgence of interest in the constitutional meaning of "author."

Thursday, May 12, 2005

Statute of Limitations: Part Two, Authorship Claims

Wednesday's posting concerned statute of limitations issues for copyright infringement claims. Today's posting concerns limitations issues that arise when plaintiff asserts joint or sole authorship, or a renewal interest. The governing provision is the same in both cases, 17 USC 507(b). That provision states that no civil cause of action may be brought later than three years from the date the claim accrued. The statute doesn't distinguish between infringement and authorship actions, nor does the legislative history. So why should we?

The courts have. But based on what? Recall from yesterday's posting that the legislative history is clear that Section 507(b) treats the statute of limitations as remedial, not as a bar on the right. This means that if you fail to bring an action within the limitations period against a particular defendant, you haven't lost your copyright, it just means you're not entitled to relief. But isn't a right without a remedy no right (as first year law students are told)?

Limitations defenses don't distinguish between meritorious and unmeritorious claims: if you miss the period for filing, no claim. You had the right, but you blew your ability to sue by waiting too long. But if you miss it once, does this mean you can never bring a claim ever, against anyone? What if you don't want money or injunctive relief?

Let's say my co-author claims sole authorship. I know we wrote the work together and intended it to be a work of joint authorship. But I don't discover the outrageous claim of sole authorship until three years after I should have known about it (under discovery accrual) or from the date the claim is first made (violation accrual). I bring a declaratory judgment after the limitations period asking that from this day forward I should be declared a co-author. Under a remedial statute of limitations I should get the declaration. If a limitations period is a bar on the right, I can't. Courts have, unfortunately, treated claims for authorship as if 507(b) was a bar on the right, while they regard the same section as a bar only on the remedy if a claim for infringement is brought. That's a distinction not supported in the statute, its legislative history, or pragmatically (in the Posnerian meaning of that term).

Things started out on the right foot in Stone v. Williams, 970 F.2d 1043 (2d Cir. 1992), which involved a very late claim for renewal rights (and thorny laches issues). Stone properly held that Hank Williams Jr.'s illegitimate daughter could obtain renewal rights going forward, but not recover money for exploitation preceding the limitations period, because the statute was remedial. Stone is correct.

Things then went south four years later with Zuill v. Shanahan, 80 F.3d 1366 (9th Cir. 1996), amended 1996 U.S. App. LEXIS 14516 (9th Cir. June 14, 1996) and Merchant v. Levy, 92 F.3d 51 (1996). Zuill attempted to justify treating infringement limitations as remedial and ownership limitations as a bar to the right by declaring that an infringement occurs every time the work is exploited without permission, but that creation, not infringement, is the gravamen of an ownership claim, 80 F.3d at 1371. That's wrong: its not creation that causes the dispute, but a subsequent alleged false claim of sole authorship.

Merchant faced a different problem since it had prior precedent, Stone. Merchant attempted to distinguish Stone as involving "highly idiosyncratic facts," 92 F.3d at 56, but that is beside the point. The relevant holding in Stone was a legal construction of the nature of the statute as remedial or not, and that determination had nothing to do with facts.

The policy implications of treating ownership claims as being forever barred if untimely can be terrible: someone who has not contributed expression and cannot constitutionally be any kind of author will be one effectively, if the real author misses the limitations period.

So when does accrual run for ownership claims, where discovery accrual is used? There is no clear answer. Some courts have referred to filing a copyright registration in one author's name, but in Estate of Burne Hogarth v. Edgar Rice Burroughs, Inc., 342 F.3d 149, 164 (2d Cir. 2003), Judge Jon Newman, the author also of Merchant, tried to put an end to that route. Publishing a book with one author's name only is a pretty good way, but in the end the inquiry is likely to be fact-specific.

Wednesday, May 11, 2005

Statute of Limitations: Part One, Infringement

A risk to lawyers specializing exclusively in one area of law over a long period of time is tunnel-vision: one may miss trends in general law that impact significantly on your area. A new wind is blowing from the Supreme Court in statute of limitations questions that conflicts with very well-established case law in copyright.

Section 507(b) requires that a civil action under title 17 be commenced within three years from the date the claim accrued. Until 1957, there was no federal limitations period for civil actions, with state law supplying the relevant period. The legislative history states that Congress intended application of the limitations defense to be a bar on the remedy, not the right. This distinction is critical not only in infringement actions, where it affects recovery of damages, but also for declaratory judgment actions on ownership. Ownership claims and the statute of limitations will be the subject of Thursday's posting.

Like almost all limitations periods, Section 507(b) does not define accrual or (commencement). It is common in copyright opinions for courts to refer to the limitations period as running from the last act of infringement, but that's a sloppy way of talking about the problem. It is not always true that if defendant begins infringing in 2000 and stops in 2005, plaintiff may sue in 2008 and collect damages going back to 2000. Plaintiff may sue in 2008 sure enough, but damages may well be limited to 2005, depending upon a number of factors, including plaintiff's knowledge of the infringement and which theory of accrual is used. It is also not true that there is a split in the circuits on this issue. It is folklore in copyright that Judge Posner's 1983 opinion in Taylor v. Meirick, 712 F.2d 1112 (7th Cir.) authorized going back to 2000 in our hypo, while all the other circuits would cut off damages at 2005.

Here's how limitations works in copyright cases. A complaint is filed in 2005. Under what circumstances may plaintiff collect damages for acts of infringement that began in 2000 and continued into 2005? (If the infringement began in 2000 but stopped in 2001, a suit brought in 2005 is untimely and no recovery is possible). All circuits permit recovery reaching back to 2002. The only differences in approach would be whether to permit recovery going back to 2000, a date outside the three year period. Taylor is mistakenly construed as permitting recovery back to 2000, while the other circuits are considered to cut off recovery before 2002.

Taylor did permit recovery going back to 2000, but for an entirely different reason: defendant was found to have fraudulently concealed the infringement so the clock hadn't begun to run. Had there been no fraudulent concealment and had plaintiff known of the infringement in 2000, the Seventh Circuit, like all courts, would have cut off the recovery at 2002. And, if there is fraudulent concealment, all the other circuits would permit recovery back to 2000. Hence, there is no circuit split.

Aside from tolling doctrines, when does a claim for infringement "accrue"? There are two basic approaches, violation accrual and discovery accrual. Violation accrual runs the limitations period from when infringement first occurred, regardless of whether plaintiff knew of the infringement or could have known of it. Violation accrual's value is certainty. Discovery accrual runs from when a reasonable person in plaintiff's position knew or should have known of facts which would support a claim of infringement. Discovery accrual's value is fairness to plaintiffs.

Courts have historically used discovery accrual in copyright infringement actions. That may be a thing of the past. The Supreme Court has been extremely critical of discovery accrual as of late. In TRW, Inc. v. Andrews, 534 U.S. 19 (2001), Justice Ginsburg noted the Court had only approved of it in two situations, latent disease and medical malpractice. Justices Scalia and Thomas went further, calling discovery accrual "bad wine of a recent vintage." If so, the drek continues. In 2004, the Ninth Circuit reaffirmed its commitment to discovery accrual in copyright cases, Polar Bear Productions, Inc. v. Timex Corp., 2004 U.S. App. LEXIS 22131 (9th Cir. Oct. 25, 2004). Polar Bear didn't cite TRW, though, relying on an earlier circuit copyright case. Polar Bear makes my point about relying on copyright decisions to the exclusion of trends in general law.

A notable exception is Auscape International v. National Geographic Society, 71 USPQ2d 1874 (S.D.N.Y. Aug. 12, 2004). Judge Lewis Kaplan, in an exhaustive opinion citing TRW and the 1957 legislative history, held that it is "strikingly clear that Congress intended to adopt a three-year limitations period running from the date of infringement, as a discovery rule would have defeated its overriding goal of certainty." Auscape is a very good indication that reliance on past precedent in the area may be a big mistake.

Monday, May 09, 2005

The Seventh Circuit Reverses: Itself; Toney v. L'Oreal

On Friday, May 6th the Seventh Circuit reversed its earlier (and very controversial) opinion which had held that the Copyright Act preempted Illinois' right of publicity statute. Here's the link to the original opinion (Original Opinion). Here's the link to Friday's opinion (Friday's opinion). The new opinion correctly holds that the statute is not preempted on the facts presented (and probably most others). Both opinions are by Judge Kanne, but they are "Night and Day" as the classic song goes.

The error in the original opinion was principally its failure to distinguish between a fixation of a likeness in a copyrighted work (there a photograph) and the right granted under publicity statutes: the right to prevent an unauthorized, commercial exploitation of the likeness. While the unauthorized likeness may be fixed in a photograph (just as an unauthorized use of a voice may be fixed on a phonorecord), the fixation is not the equivalent subject matter for Section 301 purposes. Instead, it is merely the means by which the likeness is captured. Moreover, as the Seventh Circuit pointed out in its opinion Friday, no one is the author of their own likeness, a fitting observation to make on Mother's Day weekend.

Firday's opinion is also helpful in clarifying the meaning of the circuit's 1986 Baltimore Orioles opinion (805 F.2d 663), itself much criticized for its seeming holding that baseball games as games are protectible under the Copyright Act. On Friday, Judge Kanne took the opportunity to construe Baltimore Orioles as holding "that state laws that intrude on the domain of copyright are preempted even if the particular expression is neither copyrighted nor copyrightable," giving public domain and unoriginal works as examples. This should be helpful in overcoming so-called "partial preemption" claims, claims that states can protect facts, ideas, etc. since they fall outside the domain of protected material in 17 U.S.C. 102(a). (They do fall within 102(b) and are preempted for that reason too). In this connection I draw attention to footnote four, which states that the opinion was circulated to all active (that is non-senior) judges on the circuit and no one requested en banc on the preemption question.

It is rare for a panel to so completely reverse itself and so masterfully too. Congratulations to Judge Kanne and to plaintiff's attorneys: Thomas Westgard (who told me of the opinion), Simon Frankel, and Blake Lawit.

Friday, May 06, 2005

Parody v. Satire: Should There be a Difference?

In a previous posting, I discussed "Fun with Dick and Jane," and the distinction drawn by the Supreme Court in Campbell between parody and satire. Campbell regarded parodies as directing their ridicule toward the original: "Parody needs to mime an original to make its point" about the original. Satire was regarded as directing its ridicule toward society at large, or at least not solely toward the original. Because "satire can stand on its own two feet [it] requires justification for the very act of borrowing."

Justice Kennedy's concurring opinion made the same distinction as Justice Souter's majority opinion, but took a more restrictive approach to what might qualify as a privileged parody. Justice Kennedy expressed concern that defendants might get off the hook merely by being humorous. My earlier posting expressed a similar concern about political correctness. I felt (and still do) that it should not be enough, for fair use purposes, to justify an unauthorized derivative work merely on the ground that the original was racist (Gone With the Wind), white-bread (Pretty Woman) or Waspy suburbanite (Dick and Jane).

But I do not ascribe to Justice Kennedy's restrictive view of fair use. What should matter is whether the use is transformative or in its synonyms, productive (Sony) or complementary (Posner). If the degree to which the use made is transformative is the true (or at least initial) focal point of the fair use inquiry, then the parody-satire distinction, if regarded as erecting a higher threshold for satire, is wrong.

I also question the utility of the distinction as a matter of classification: if a satirical musical group takes a Bob Dylan song from the mid-1960s and substitutes new lyrics for the purpose of showing how at least some former hippies are now Republicans living in the Virginia suburbs, is the work a parody or a satire? Isn't the work at least partly directed toward Dylan? Most importantly, for fair use purposes why should we care what label is affixed so long as the public is benefited with a transformative use that provides new commentary or insights?

Justice Souter was aware of the problems of his own classification scheme, observing that "parody often shades into satire - when society is lampooned through its artifacts." And he also cautioned against bright-line rules. Indeed, Campbell's principal contributions to fair use jurisprudence were its jettisoning of the Sony "presumptions," in particular Sony's false dichotomy between non-commercial and commercial uses. That dichotomy haunted fair use decision making for ten years after Sony.

Let's hope that the parody-satire distinction does not have the same effect. It is possible Justice Souter meant no more than an observation about the problems of proof that a satirist would face, rather than setting out a legal hurdle to overcome.

A reader of the blog, Annemarie Bridy recently published an excellent look at these issues in the Winter 2004 Journal of the Copyright Society.