This is a big one. As I’ve written about in the past (recursive link), there is a huge upheaval in the stock photography industry over the use of photos in textbooks. In Palmer/Kane LLC v. Rosen Book Works LLC, plaintiff Palmer/Kane licensed the rights in its photos through several agencies, the relevant one here being Corbis.
The textbook industry claims that it is common practice for publishers to use photos in books before obtaining the license for their use. In fact, the Palmer/Kane court said that it was a "persuasive showing."
For those of you who don’t remember contracts class, there is a hierarchy for how one goes about interpreting contracts:
In the interpretation of a promise or agreement or a term thereof, the following standards of preference are generally applicable: … (b) express terms are given greater weight than course of performance, course of dealing, and usage of trade, course of performance is given greater weight than course of dealing or usage of trade, and course of dealing is given greater weight than usage of trade.
Restatement 2d of Contracts, § 203. So standard industry practice can be relevant to understanding the terms of an agreement.
But the defendant’s theory that "we’ve always done it that way!" didn’t matter here, as explained by the court:
Nonetheless, evidence of custom and course of dealing cannot displace rights conferred by the copyright laws. See Dun & Bradstreet Software Servs., Inc, v. Grace Consulting, Inc., 307 F.3d 197, 211 (3d Cir. 2002) ("A defense of industry custom and practice in the face of the protective provisions of the Copyright Act could undermine the purposes and objectives of the statute and reduce it to rubble."); Weinstein Co. v. Smokewood Entm’t Grp., LLC, 664 F. Supp. 2d 332, 348 (S.D.N.Y. 2009) ("[N]otwithstanding plaintiff’s claims about ‘custom and practice’ in the entertainment industry, federal copyright law dictates the terms by which an exclusive license can be granted."). Rosen does not appear to seriously contend that its agreements with Corbis authorized pre-license use (something of a contradiction in terms), but rather that Corbis essentially looked the other way in the face of infringement. Cf. Spelman Decl. dated June 25, 2016, ¶ 6 ("In my experience Corbis did not treat a customer’s use of a licensed photo in excess of or beyond the terms of the license in question as copyright infringement, but rather as a contractual matter to be resolved as part of the ongoing, mutually beneficial relationship between the publisher and Corbis."). But that Corbis may have chosen not to view defendant’s pre-license use as "infringing" is neither here nor there. Infringement is infringement, regardless of what Corbis and Rosen may have chosen to call it.
But wait, there’s more. The Palmer/Kane court, in the Southern District of New York, an influential copyright venue, also disagreed with the conclusion of four of its sister courts that the license eventually taken by the publisher was effective retroactively. Instrumental to the conclusion here is the proper interpretation of Davis v. Blige, 505 F.3d. 90 (2d Cir. 2007), a controversial decision involving the ability of a joint owner of a copyright to retroactively license an infringer. The four earlier courts distinguished Davis on the basis that it was a co-ownership situation, not sole ownership. But the Palmer/Kane court wasn’t buying it (forgive the lengthy block quote but it’s worth it):
As a threshold matter, this Court is not of the view that it has the discretion to narrow the scope of Davis‘s holding on the basis that Davis was animated by a set of considerations that are arguably not relevant here. That, of course, begs the question whether what Davis described as its holding – "that a license or assignment in copyright can only act prospectively" – was truly its holding in the precedential sense. See Davis, 505 F.3d at 104. While the metaphysical line between dictum and holding "is not always easy to draw," "’where a panel confronts an issue germane to the eventual resolution of the case, and resolves it after reasoned consideration in a published opinion, that ruling becomes the law of the circuit, regardless of whether doing so is necessary in some strict logical sense.’" Cetacean Cmty. v. Bush, 386 F.3d 1169, 1173 (9th Cir. 2004) (quoting United States v. Johnson, 256 F.3d 895, 914 (9th Cir. 2001) (Kozinski, J., concurring)). Such is the case here, where the principle established by the Second Circuit in Davis – and identified by that Court as its holding – is unambiguous and the result of full judicial consideration.
Moreover, the policy considerations allegedly motivating Davis do in fact apply to the sole-ownership context. As for "the need for predictability and certainty," many copyrights owners contract with a licensing agent to license their works, as plaintiff did with Corbis here. Davis, 505 F.3d at 105. If a licensing agent may license works retroactively, the sole copyright holder is not all that differently situated – in terms of its ability to "reliably and definitively determine if and when an infringement occurred" – than a copyright co-owner whose infringement claim may be extinguished by another co-owner’s action. Id. The licensing agent is of course acting as the copyright holder’s agent, but the copyright holder is still placed in the unenviable position of being generally unable to know, with certainty, that its infringement claim will not be extinguished by the grant of a retroactive license. See id. ("If retroactive transfers and licenses were permissible, one could never reliably and definitively determine if and when an infringement occurred, because an infringement could be ‘undone’ by the very sort of maneuver attempted by defendants in this case.").
As for the desirability of discouraging infringement, it is difficult to see how the availability of a retroactive license does not "lower the cost of infringement to infringers" even in the sole-ownership context, particularly given that the licensing fees that have been contractually predetermined between the infringer and the licensing agency will often be dwarfed by the statutory damages that would have been available in an infringement action. Id. at 106; see 17 U.S.C. § 504(c) (providing for statutory damages of up to $30,000 per work infringed in general, and up to $150,000 per work infringed if the Court finds willful infringement). Thus, while these two policy considerations might be more stark in the context of copyright co-ownership, they are at play in the sole-ownership context as well. See N. Jersey Media Grp. Inc., 2015 U.S. Dist. LEXIS 29191, 2015 WL 1086566, at 4 n.6 (finding that "the policy reasons underpinning the court’s holding in Davis are similarly present" where there is a sole copyright holder).
The fact that "[a]n owner who wishes to release unilaterally his own accrued claims may do so using whatever language he chooses — including by calling the negotiated settlement the proverbial ‘banana’ — ", Davis, 505 F.3d at 104, is of no help to Rosen. In obtaining the February 2008 License, Rosen plainly was not entering into a "negotiated settlement" with Corbis. There was no recognition by either Rosen or Corbis that infringement had occurred and, in fact, Rosen denies that it infringed. Notwithstanding its reference to the "proverbial ‘banana,’" the Davis Court explicitly cautioned against conflating settlements and retroactive licenses:
Licenses and assignments function differently from settlements and releases, and the use of the term "retroactive license" for "settlement" or "release" by the parties causes unnecessary confusion and potentially creates legal mischief.
. . .
In its simplest form, a license means only leave to do a thing which the licensor would otherwise have a right to prevent. A retroactive license or assignment purports to authorize a past use that was originally unauthorized. Unlike a settlement, which recognizes an unauthorized use but waives a settling owner’s accrued claims of liability, a retroactive license or assignment would — if given legal effect – erase the unauthorized use from history . . . .
Id. at 102-03 (emphasis added) (citation and internal quotation marks omitted). That parties may call a negotiated settlement whatever they wish does not mean that a retroactive license is, by definition, a negotiated settlement. This Court thus cannot bless an allegedly retroactive license by pretending that the license was in fact a negotiated settlement. The Second Circuit might one day limit the scope of Davis in the manner Rosen seeks. But until such time, this Court cannot disregard Davis‘s categorical conclusion "that a license or assignment in copyright can only act prospectively." Id. at 104.
The court, “constrained to disagree” with the earlier decisions, is either prescient or half-baked, I’m not sure which.
And this case is one that keeps on giving. I first started to read it because it also has a section on copyright registrations. I’ll post about that later too. But the licensing discussion was too interesting to not write about first.
Palmer/Kane LLC v. Rosen Book Works LLC, No. 15-cv-7406 (JSR) (Aug. 30, 2016 S.D.N.Y.)