Rejecting Gracen’s fundamentals, the court of appeals reversed. As long as Schrock was authorized to make the photos, as he was, and as long as the photos were sufficiently original, as they were, he owned the copyright. (In other words, post-1976 Act, it’s a mistake to use “copyright” as a verb, as Gracen did, because copyright inheres in the work once fixed.) The court “clarif[ied]” an aspect of Gracen “that is prone to misapplication.” Gracen didn’t require a heightened standard for originality for copyright in a derivative work. All that is required is “enough expressive variation from public-domain or other existing works to enable the new work to be readily distinguished from its predecessors.” Bucklew v. Hawkins, Ash, Baptie & Co., LLP, 329 F.3d 923, 929 (7th Cir. 2003). Here, Schrock’s photos of Learning Curve’s “Thomas & Friends” toys possessed sufficient incremental original expression to qualify for copyright.
The court remanded for findings on whether the parties had agreed to vary the default rule of ownership, or whether defendants had an implicit license to continue to use the photos.
People disagree vigorously over whether a photo of a copyrighted work is a derivative work. The court refused to resolve the issue, assuming for purposes of argument that a photo is a derivative work (and along the way calling Posner’s quite explicit statement that a photo of a Beanie Baby is a derivative work of a Beanie Baby in Ty v. PIL a mere reflection of the parties’ concessions).
Gracen worried that a derivative work could be so similar to the underlying work that it could create a harassment problem for subsequent creators of derivative works based on the same underlying work—two versions of the Mona Lisa, for example; if creator 2 had access to creator 1’s work, it would be very hard to decide (much less avoid) an infringement suit. The Schrock court thought this was a valid concern, but the Copyright Act doesn’t hold derivative works to a more exacting originality requirement than other works of authorship. The standard is not high. “[T]he key inquiry is whether there is sufficient nontrivial expressive variation in the derivative work to make it distinguishable from the underlying work in some meaningful way.”
I don’t think this is a problematic conclusion writing on a blank slate. (Though one would want to adjust the infringement standard upwards in such a case, even though the Act is silent on things like merger, scenes a faire, and other doctrinal refinements that allow/counsel us to make it hard to find that derivative work 2 infringes derivative work 1.)
The problem is this: How could the painting in Gracen have possibly failed to meet the “nontrivial ‘distinguishable variation’” standard? Everyone conceded that the scene Gracen painted did not exist in the movie; she made something new:
By contrast, the panel concludes that Schrock’s pictures of unaltered copyrighted toys have the necessary originality because he made decisions to make the toys look good, maybe a little better than they do in real life—compare to Gracen’s success at representing “our Judy” in Gracen. What is the “distinguishable variation” from the underlying toys, especially since the court emphasizes that it also follows the principle that a mere shift in medium doesn’t produce a copyrightable derivative work?
And this difficulty suggests that the court was wrong when it thought it didn’t need to resolve the question of the photograph’s status—independent work or derivative work. From the decision: “If the photographer’s rendition of a copyrighted work varies enough from the underlying work to enable the photograph to be distinguished from the underlying work (aside from the obvious shift from three dimensions to two …), then the photograph contains sufficient incremental originality to qualify for copyright. Schrock’s photos of the “Thomas & Friends” toys are highly accurate product photos but contain minimally sufficient variation in angle, perspective, lighting, and dimension to be distinguishable from the underlying works; they are not ‘slavish copies.’” As independent work, Schrock’s choices of lighting and composition may well have creative components sufficient to sustain a copyright—what the court terms originality in rendition. But those choices are orthogonal to whether there is a distinguishable variation from the underlying copyrighted toy in the photo. No matter how it’s lit and placed and tilted for perspective, it’s still the same toy. As far as I can tell, the way to distinguish them is that one is 2-D and one is 3-D, precisely the thing the court said wasn’t sufficient. Given that the distinguishable variation test seems either impossible or trivial (if you accept 3-D to 2-D), I’m inclined to say that we shouldn’t be treating photos as derivative works of their copyrighted subject matter at all.
This case is a clear rejection of, not gloss on, Gracen, and the right thing to do would have been to en banc it if Gracen is wrongly decided—as many have argued. Nonetheless, the panel concluded that the “nontrivial ‘distinguishable variation’” test adequately deals with Gracen’s concerns about holdup, especially since the result is a thin copyright in the derivative work extending only to the incremental original expression contributed.
And then there was Gracen’s other key principle, that the author of a derivative work needs permission to copyright it. See 698 F.2d at 303-04 (“[T]he question is not whether Miss Gracen was licensed to make a derivative work but whether she was also licensed to exhibit [her] painting and to copyright it. . . . Even if [Gracen] was authorized to exhibit her derivative works, she may not have been authorized to copyright them.”). The district court relied on this language, but this was error because that part of Gracen was (1) dicta, and (2) wrong. (For what it’s worth, I’m in complete agreement with (2).)
From what the panel could tell, the photos and their use relied on a series of oral agreements. The toymaker, which licensed Thomas the Tank engine from HIT, was arguably contractually required to protect HIT’s IP rights explicitly in everything it did, but “it apparently failed to do so.” Remand was required to sort out the relevant facts on ownership and licensing.