Monday, October 08, 2007

The Second Circuit Goes to the Dark Side

I try not to be discriminatory among the circuits, either in my praise or criticism. As things fall out, for me and for others, the Ninth Circuit tends to fall more than others into the criticism category for reasons that many have pondered and which may remain insoluble. The Second Circuit’s forays to the dark side are less often, for reasons that may also remain insoluble, but a decision on October 5th, in Davis v Blige, 2007 WL 2893003 (2d Cir. Oct. 5, 2007), is about dark as indeed. The decision opens up to challenge untold numbers of settlement agreements, and even more broadly eliminates one of the central tenets of the 1976 Act: the alienability of one co-owner’s rights without the permission of the other co-author(s). In this latter respect, the Second Circuit has unwittingly joined the Ninth Circuit’s widely lambasted opinion in Gardner v. Nike, Inc., 279 F.3d 774 (9th Cir. 2002). The Davis v. Blige opinion is by an excellent judge, Jose Cabranes, reversing that of another excellent judge, District Judge Charles (Terry) Haight. What is more inexplicable is that the circuit once vacated part of an earlier opinion that had incorrectly came to a similar conclusion, Morris v. Business Concepts, Inc., 283 F.3d 502 (2d Cir. 2002).

The case involved an infringement claim brought by Sharice Davis against a number of individuals and companies over two songs on defendant Mary Blige’s album “No More Drama.” After suit was filed, two of the co-authors, Bruce Chambliss (whose testimony on key issues was admittedly less than a model of clarity), executed a transfer to another of the co-authors, Bruce Miller. The transfer was retroactively effective to the date of the works’ creation (an unspecified date). Miller in turn had licensed the other defendants. Defendants’ claim, which was correct, is that Miller’s license as a co-author eliminated an infringement claim by plaintiff. Plaintiff was not left out in the cold, however, because she still had a right to an accounting of the profits from the licenses Miller entered into. Judge Haight agreed that such a retroactive license barred plaintiff’s infringement claim, citing prior case law in the SDNY. The court of appeals vacated and remanded. A full discussion of the opinions shortcomings is well beyond even a long blog like this, so I will focus on the essentials.

First, the court held, “we conclude that … retroactive transfers violate basic principles of tort and contract law, and undermine the policies embodied by the Copyright Act.” Call me old fashioned, but if there are explicit provisions in the Copyright Act that address the question at hand, that is where one finds the relevant answers and policies, and I don’t give a hoot whether basic principles of tort and contract law are in conflict. The relevant provisions are Section 201(d) (cited by the court only in passing) and the definitions provision of Section 101. Congress explicitly permitted an exclusive licensee's right to transfer rights without the author's permission in these sections.

Section 101 defines a “copyright owner as the owner of any particular exclusive right.” No distinction is made among the author, a licensor of the author, or a subsequent licensee. The statute further defines a transfer of copyright ownership as an assignment or exclusive license, excluding only nonexclusive licenses. Section 201(d)(1) ties these two definitions together by expressly indicating that a transfer of copyright ownership may be made “by any means of conveyance.” Under these provisions, every holder of an exclusive right is regarded as a “copyright owner” to the extent of the right possessed. There may be, and typically are, multiple copyright owners of a single work, e.g., a novelist sells to a book publisher the right to publish a sequel, but retains the right to authorize the making a movie of based on the book, but there can be as in Davis v. Blige, co-authors of one work.

Important economic consequences flow from a joint authorship relationship: Each joint author is a co-owner of the work and, as a tenant in common, possesses an equal, undivided interest in the whole, absent a contractual agreement to the contrary, and regardless of the relative extent or quality of their contributions. Thus, two joint authors each own a 50% interest in the whole, even if one author contributed only 10% of the work. As a co-owner in the whole, each joint author may utilize the work him- or herself without the other's permission and indeed over the other author's objection. A co-owner may also sue for infringement by third parties without joining the other co-owner in litigation.

A license from one co-owner is a defense to a claim of infringement brought by the other, nonlicensing joint owner. This is a significant point: a co-owner may unilaterally grant nonexclusive licenses; a joint author (or co-owner by assignment) is immune from an infringement claim by the other author or co-owner; and, a joint author has an independent right to use or license others to use the work.

The only obligation of a co-owner is to account for any profits earned from the exploitation. One joint author may also transfer all of his or her own exclusive interest in the work without the other joint author's consent. A joint author (or co-owner) may not, however, transfer all interest in the work without the other co-owner's express (and written) authorization, since that would result in an involuntary transfer of the other joint owner's undivided interest in the whole. But the statute explicitly covers the ability of one co-author to transfer his or her proportional share in the work, and regardless of whether such a transfer is called a transfer or an exclusive license: the two terms are synonyms.

The Second Circuit, contrary to Section 201(d), held that one co-author cannot transfer his or her exclusive rights in the work without all other co-authors’ permission: “a co-owner cannot unilaterally grant an exclusive license”: to do so, the court held would conflict with “the venerable law of property that, while an owner may convey any of his rights to others permanently or temporarily, he may nor convey more than he owns.” This misses entirely the point that a co-author is only conveying his or proportional share as a tenant in common and that Section 201(d) was written precisely to permit such transfers. Such a co-owner is not, as the court of appeals incorrectly stated, trying to “convey the interests of his fellow co-owners without their express written consent… .” Under the court of appeals’ approach, there can be no copyright divorce without consent: a co-owner is the equivalent of an agunah in Jewish religious law, even though the Copyright Act lets spouses go at will.

Here is the simple analysis the court of appeals missed. Section 101 states: ”’Copyright owner’”, with respect to any one of the exclusive rights comprised in a copyright, refers to the owner of that particular right. A copyright owner is therefore merely someone who owns an exclusive right. Co-authors own proportional share of the exclusive rights. The question is then, can they transfer their exclusive right? Is there anything in the statute that bans such transfers? No there isn’t. Is there anything that permits such transfers? Yes there is. Section 201(d)(1) reads: “The ownership of a copyright may be transferred in whole or in part by any means of conveyance or by operation of law, and may be bequeathed by will or pass as personal property by the applicable laws of intestate succession.” Is there anything that permits partial transfer of interests? Yes there is. Section 201(d)(2) reads: “Any of the exclusive rights comprised in a copyright, including any subdivision of any of the rights specified by section 106, may be transferred as provided by clause (1) and owned separately. The owner of any particular exclusive right is entitled, to the extent of that right, to all of the protection and remedies accorded to the copyright owner by this title.” The section expressly contemplates that there will be multiple owners of exclusive rights, and is not limited to copyright in works authored by a single author. A co-author who transfers his or her proportional interest is merely transferring his or right “to the extent of that right.” It is telling indeed that the court of appeals failed entirely to quote or apply Section 201(d)(2), referencing it only as a cite and referencing Nimmer for the absurd and irrelevant proposition that one may not license more than one owns.

The court’s failure is to understand the statute is profound and will have profoundly negative consequences: no business can now purchase rights to a work unless it gets an assignment from all authors: imagine trying to buy rights to collective works that contains the works of thousands of authors, many of whom are co-authors. This is a disaster of major proportions, all the more so in the face of a clear statute to the contrary.

The second part of the court’s opinion painfully reveals its result-driven, legal realist nature. The issue was the validity of retroactive licenses and transfers. This section was unnecessary analytically because of the first part, but was necessitated by the court’s unwillingness to overrule its contrary opinion in Eden Toys, Inc. v. Florelee Undergarment Co., Inc., 697 F.2d 27 (2d Cir. 1982). Eden Toys held, in part, that a later writing could confirm an earlier oral agreement. Chambliss claimed there had been such an agreement, but the evidence on this point was confusing at best. But in the end, who cares? Unless a co-author can never assign rights, he or she can assign them at any time and make them retroactive to cover things like pre-existing causes of action. It is here that other courts have permitted, rightly, transfers from one-co-owner as a part of a settlement agreement that then ends the suit. The co-author(s) who didn’t transfer or settle are not out of luck, because they still have a right to an accounting from exploitation of the work, past or future. Yet, mistakenly believing non-transferring or settling co-authors would be harmed, the court of appeals held that such settlement transfers “can only waive or extinguish claims held by a settling owners; it can have no effect on co-owners who are not parties to the settlement.” Why not? Because of “a venerable principle of New York co-tenancy law … .” This is copyright law, not state law. The court also pulled out of thin air the assertion that “Licenses and assignment … are prospective… .” That will be news to millions of licenses and assignees. The court apparently thought it terrible that “A retroactive license or assignment purports to authorize a past use that was originally unauthorized.” And so what? That happens all the time.

Davis’ holding on retroactive active licenses is as disastrous as its holding on the nonassignability of proportional exclusive licenses, and moreover reflects a stunning deafness to the commercial realities of life: one would expected much much more from a court that sits in the commercial hub of the country.

One can only hope that a petition for rehearing is granted, the original opinion is vacated, and a proper one substituted.


Anonymous said...

I can't argue with your legal reasoning about this case, and am at a loss as to understanding how the court could have so badly misread the statute.

There is one thing about joint authors I have never understood, however. You write, "Plaintiff was not left out in the cold, however, because she still had a right to an accounting of the profits from the licenses Miller entered into." But what if Miller had granted non-exclusive licenses for free? No profits, nothing to account for - and the other co-owners are left out in the cold.

Does the responsibility to account for profits also entail a responsibility to maximize profits for the other co-owners?

William Patry said...

Hi Peter H. My view is no, there is no duty to maximize profits or conversely to avoid waste. This opinion doesn't reach these issues but I think it comes very close to providing arguments there are such obligations.

CEP said...

One thing that bothers me about the 2d Circuit opinion is its poor organization, reflective of the low standard of writing in the law. Bluntly, the opinion would have been much better grounded had it inverted its argument:

(1) "Retroactive" (and the quotes are there for a reason: they reflect the reality that these are not arms-length bargains between parties with relatively equal power and information) transfers under the Copyright Act — even when purportedly confirming a prior oral transfer — are disfavored, particularly in the face of a claim, as an improper means of evading the "signed writing" requirement.

(2) Even if retroactive transfers did not implicate critical Copyright Act concerns, general principles of tort and contract law would disfavor this particular transfer's validity.

I'm not saying that I entirely agree with point 1 above, but if it had been point 1 of the opinion instead of the alternate ground, I think the actual issues involved could have been both more readily understood and accepted... and any exploration of them could have been better subjected to second-guessing from those of us who have not seen all of the discovery in the matter.

All of that said, I think the real problem is that the court reached a possibly defensible result in this matter (again, depending upon what was actually in the discovery) that was not properly considered when the 1976 Act was drafted... an analog of the Morris problem. (I still disagree that a certificate of registration is a jurisdictional element: It is better treated as an element of the plaintiff's case, based on recent SCt jurisprudence on the reach of "jurisdictional element." That's for another time, though.) The reality is that the statute was written without much, if any, persuasive input from the actual creators of copyrightable material; instead, the middlemen got to speak, and that has resulted in lacunae like this one.

Bad writing trumps bad law most of the time, and this is just another example.

Ampersand said...

I nominate the Blige decision for the Copyright Hall of Shame where it can lounge about with:

Rano v. Sipa Press, 987 F.2d 580 (9th Cir. 1993);

Gardner v. Nike, 279 F.3d 774 (9th Cir. 2002).

Other nominations are welcome.

Seriously, if the Blige decision isn't withdrawn or modified, its mischief will be far-reaching. Co-authors (and successors of co-authors) often have different attitudes toward exploitation of the work. If unanimity is now required, and prospectivity is mandatory, the universe of works that can be licensed has just contracted. This is the antithesis of the goal of copyright law.

William Patry said...

Vincent, amen!

RT said...

Thanks for the analysis. I had the opinion queued up for later examination; it looked quite odd on a first read. The signed writing/exclusivity part is quite interesting; I hadn't considered this before, but it seems under the proper reading of the statute the first coauthor to jump can, in an industry that requires exclusive licensing, disable other coauthors from licensing to someone else ... except that if the industry really did require exclusive licensing, then the consent of one coauthor wouldn't be sufficient in the first place; this is only useful if there has been some mistake (or wrongdoing) and the agreement is by way of settlement. The first-acting coauthor can constrain the other coauthors with exclusive licensing of her share of the rights; but it was always true that the first-acting coauthor could constrain the other coauthors.

The question of whether a coauthor has a duty not to waste the copyright often comes up in my classes. (See Wilchcombe v. Teevee Toons, Inc., 2007 U.S. Dist. LEXIS 6205 (N.D. Ga. Jan. 26, 2007) for a potential example.) I can see reasons to imply such a duty, but it need not restrain the coauthor from making a good exclusive deal.

TruePath said...

The profit accounting bit raises some interesting possibilities.

Suppose I am a joint author in a work and the other author grants a license for 'free' in return for the beautiful women who is the licensee sleeping with him. Further suppose there is ample evidence that she only slept with him to get the license and that he was unwilling to transfer the license without the lure of sexual services.

Would the accounting of profits issue force the courts to put a monetary value on sleeping with the woman in question? Would experts on the woman's beauty and the going market rate for prostitution end up being called?

Anyway I suppose this consideration applies to a lot more than copyright law but I'm just wondering if these non-monetary compensations would have to be monetized by the court.

William Patry said...

I'll have to sleep on that one.

lawnun said...

Ignoring the whole mess regarding retroactive transfers for a moment, I'd like to touch on what you said regarding the grant of exclusive licenses by co-owners.

The italicized section expressly contemplates that there will be multiple owners of exclusive rights, and is not limited to copyright in works authored by a single author. A co-author who transfers his or her proportional interest is merely transferring his or right “to the extent of that right.” ...

The court’s failure is to understand the statute is profound and will have profoundly negative consequences: no business can now purchase rights to a work unless it gets an assignment from all authors ...

How is the court's opinion any different than what's already recognized? In this case, the co-owner (Chambliss) assigned an exclusive license to the defendant , and that's precisely what the court said was a no-no. Unless I'm fundamentally misreading the case, how is this any different than what you said here (

"A joint author may sell its proportional interest without the permission of the others, but cannot engage, solo, in exclusive licensing (which is the same thing as an assignment of the right licensed)."

How is that any different than what the court concluded?

William Patry said...

Lawrun, the difference is this.

There is no difference between an exclusive license and an assignment of an exclusive right, or a transfer of copyright. The court held, I think, that one cannot exclusively license, a/k/a assign, a/k/a assign your proportional share of the copyright in the work although it is hard to tell how far the court meant to go because it is so poorly written and organized.

What I was saying in the earlier post you quote is that while you can exclusively license, a/k/a assign, a/k/a assign your proportional share, you cannot exclusively license, a/k/a assign, a/k/a assign the whole of the work because that would involve an impermissible effort to sell others' shares too.

Ethan said...
This comment has been removed by the author.
bruce said...


Infringer is criminally convicted under 18 U.S.C. 2319. Improbably, he's sentenced to a year in prison.

The copyright owner then grants the infringer a "retroactive license." Does the infringer still have to serve his prison sentence, or can he get his conviction tossed out? After all, he didn't do anything wrong!

Now think about whether the statutory range of damages has a punitive component...

William Patry said...

Bruce, I don't understand the punitive part of your comment. As to the retroactive license and its effect on the criminal sentence, I would have to defer to a specialist in criminal law, but it sounds like a great hypo.

bruce said...

My apologies for being vague.

We learn early and often that the beneficiary of the patent and copyright systems is the public. This line of thinking seems to stem from the express wording of the IP clause.

The beneficiary of criminal law, too, is the public. In criminal law, the state prosecutes violations on behalf of all. While a decision to prosecute may often come down to whether a private citizen desires to "press charges," this is not necessary.

In copyright, we allow copyright owners to prosecute civil actions to protect the integrity of the system for the benefit of the public. The copyright owners are in a better position to notice infringement, and have a larger incentive to vigorously enforce and protect their copyrights.

The answer to my hypothetical question is "no." The infringer could not get his conviction thrown out because he committed the crime. A grant of forgiveness by the copyright holder will not extinguish the infringer's criminal culpability. In this context, the astounding fictional nature of the "retroactive license" is on full display. Saying a license is retroactive is just a tidy way to say that claims for past infringement are released, and a prospective license is being granted as well. It does not change the fact that the infringer committed a willful and bad act.

Letting a willful infringer off the hook simply because one co-owner is willing to grant a retroactive license flies in the face of these concepts. Statutory damages are provided for in copyright law for two reasons. One is that actual damages may be hard to calculate. The other is that limiting actions to actual damages would make the lawsuits not worth filing in many instances, and so would insufficiently deter infringement. The legislative history of the amendments increasing the statutory damages talk repeatedly about deterrence and punishment. "As the court considers just" is a clear instruction to weigh culpability, and 504(c)(2) reads like a sentencing guideline. Statutory damages clearly have a punitive element intended to punish the infringer's bad acts against society, not simply the copyright owner.

As long as one co-owner is willing to maintain a suit against a willful infringer, and seek punishment against that infringer on behalf of all of us, it is unconscionable to let the unilateral actions of another co-owner defeat the claim. The legal fiction of a "retroactive license" does not erase from history the willful and bad acts which our society seeks to punish and deter.

Crosbie Fitch said...

How many 'criminal' works of art, I wonder, have involved copyright infringement in their production, that are subsequently legitimised by retroactive license obtained via 'clearance'?

Should their authors be prosecuted irrespective of the 'forgiveness' by other authors whose permission was only sought after the act of infringement?

After all, it is not for authors to be lenient on the public's behalf. The public's selfless grant of copyright must be protected, and perpetrators of its infringement must be punished.

No member of the public should enjoy a liberty properly suspended for the privilege of publishers - unless the publisher gives them permission BEFOREHAND!

The public is a beneficiary of copyright and yet hoist by the very same petard.

Anonymous said...

I'm confused by your use of the term "proportional share." Do not joint holders of a copyright have an undivided interest in the whole? Whatever "proportions" exist would be not be 1/2 and 1/2 for two authors, 1/3-1/3-1/3 for three authors, etc. The proportional split would be spelled out in some kind of agreement between the joint authors or between them and their publisher as to how to split up the earnings of the joint copyright, not how to split the copyright itself. Two joint authors might decide to split the earnings 90-10.

William Patry said...

Dear Anon, you are, I think, confusing sharing of royalties under a contract with ownership of a copyright interest in the work. Two joint authors automatically share an undivided 50% interest in the whole, three authors a third, etc. They can change this, but to do so will represent a transfer of ownership, and thus has to comply with 17 USC 204. By contrast, a decision how to share royalties is distinct from ownership: two authors could for example leave intact 50-50 ownership but permit one to get 60% of the royalties. In short, the two concepts are different, with a split on royalties being a contract, not a copyright matter.

Jane Daniel said...

I left a question about "proportional share" of ownership of a joint copyright a while ago. I've read and reread your answer but I still don't understand. I am a publisher so I know what an instrument for the purpose of licensing a copyright looks like. What does same look like for changing the proportional ownership of a three proprietor joint work from 1/3-1/3-1/3 as you describe, to, say, a 40-40-20 split. Is it the same?

In William Strong's excellent tome, "Copyright Book", he writes, "Where joint authorship exits...each of them owns an undivided share in the entire work...This share is not necessarily equal, for the authors can slice the pie unevenly if they so choose." So how does the unequal arrangement appear on the copyright notice on a book or on the copyright registration with the Library of Congress?

William Patry said...

Jane, an agreement to alter the default proportional share is an agreement to transfer rights, so it has to comply with the ordinary rules for transfers of rights. Such an agreement may be recorded in the Copyright Office. I don't give advice, but speaking personally, I don't see why one would want to reflect any share on a copyright notice, and I have never seen one reflected in a registration.

Anonymous said...

What I'm having a problem with here is that you seem to be saying that every time a co-author issues an exclusive license of a portion of his rights to another, that exclusive license should be treated as a transfer of that owner's share to the licensee, and the exclusive licensee is therefore a copyright owner.

I do not understand why this alternative reading is not just as likely -- the co-author's issuance of an exclusive license encompasses the whole of that right within the work, meaning that the co-author has also transferred his other co-author's rights. Obviously, the exclusive licensee, who may have no knowledge of the other co-author, believes he is obtaining an exclusive license as to the whole interest -- not that he is sharing that interest with another co-author/co-owner.

Why must we leap to the conclusion that the co-author is only transferring his rights (and is therefore no longer a copyright owner with respect thereto), rather than concluding that he simply exceeded the permissible bounds, and issued a license which is not limited to his own rights, and impinges upon the rights of his co-author/co-owner?

This has crucial implications for accountings between co-authors. If one co-author issues an exclusive license to someone and gets royalties for that, must he divide those royalties with his co-author? Under your theory, the answer would be no, right -- because the first author has transferred his rights, the exclusive licensee would be a copyright owner and would have a duty to account to the other co-author, but the co-author who issued the exclusive license (and is being paid therefor) would not have to account to his co-author because the income he is receiving is income attributable to a transfer of his own rights, rather than a royalty stream subject to division. The co-author who is left out would then be limited in his recovery to a share of profits (and an accounting) from the exclusive licensee, who doesn't even know he or she exists.

This outcome to me does not seem right. The better view would be that the co-author had no right to issue an exclusive license, and therefore the license is void; the licensor would have to account to his co-author for any profits received, and the other co-author could then sue the "exclusive licensee" for infringement, since the license was void ab initio.