Prof. Douglas Lichtman, UCLA School of Law
Sept. 24, 2009

Section 106(2) of the Copyright Act recognizes in authors the exclusive right to prepare derivative work.  This is in addition to the other exclusive rights listed in Section 106, such as the exclusive right to make reproductions and the exclusive right to engage in public performance.  In the definitions section of the Act, we’re given a bit of information about what it means to be a derivative work.  We’re told that a derivative work is a “work based upon one or more preexisting works”; and we’re instructed that a work consisting of “editorial revisions, annotations, elaborations” and the like qualifies so long as those modifications “as a whole” constitute “an original work of authorship.”

But what does all that mean?

On one reading, the derivative-work right would sweep in basically anything that might be based on a prior copyrighted work.  The idea would be that the fair use doctrine could then trim the right back, excluding from protection works and situations where author control seems unwise on policy grounds.  On another reading, however, those definitions I just summarized could be interpreted to establish some clear lines — excluding (say) work that isn’t analogous to that statutory list of examples, or excluding work that is not creative enough to qualify as “an original work of authorship.”

Over at the Intellectual Property Colloquium this month, we are struggling with the challenge of Section 106(2).  We’re trying to understand the relationship between the derivative-work right under Section 106(2) and the reproduction right under Section 106(1).  We’re also trying to figure out how the copyright system allocates work as between Section 106(2) and the fair use doctrine codified at Section 107; that is a relevant question because many of the same concerns could be read to animate both provisions.  Lastly, we are trying to apply all that to what industry insiders call “unobjectionable uses” — that is, work that is clearly derivative, probably not fair use, but work that most copyright holders are nevertheless perfectly happy to allow.  Fan fiction is among the best examples here.

My reaction to the conversation thus far has been to think that the derivative-work right is for the most part a lot of noise about little substance.  The reason?  Most cases where Section 106(2) is at issue are already fully resolved by Sections 106(1) and 107 anyway.

Take, for example, last year’s high-profile fight over RDR Books’ unauthorized encyclopedia of all things Harry Potter.  One of the charges in the case was that RDR Books violated Section 106(2) because the encyclopedia was based on the copyrighted Potter novels.  The court surprised many of us by ruling that, no, the encyclopedia was not a derivative work.  And, indeed, at the time I was somewhat dumbfounded by that result.  What did it mean to say that the encyclopedia was not derivative of the Potter books, given the obvious relationship between the two?

Today, though, my puzzlement is fully replaced by indifference.  After all, even though the court held that the encyclopedia was not derivative of the Potter stories, the court nevertheless issued an injunction.  The court’s logic: The encyclopedia violated Section 106(1) because the encyclopedia pervasively reproduced “fictional facts” like the attributes of imaginary creatures and the traits and undertaking of Potter and his pals.

Understood in that light, Section 106(2) was basically redundant to Section 106(1).  Even without a finding under Section 106(2), the encyclopedia was still vulnerable to an injunction by virtue of its copying, and the encyclopedia was going to avoid that injunction only if fair use excused the copying.  In the Potter case, then, the derivative-works issue was more about a label than about substance.  Expressive material was copied.  The struggle was only over whether to call that copying Section 106(1) reproduction, Section 106(2) derivative work, or both.  Nothing more was at stake; and the case in the end turned fully on the court’s fair use analysis.

Interestingly, there are (rare) Section 106(2) cases where Section 106(2) is not redundant to Section 106(1); those are the cases that strike me as the important ones for understanding the derivative-work right.  So, for instance, there’s a famous line of cases about cut-and-paste projects, where (say) I take your original artwork, cut it up, frame the pieces, and sell those as stand-alone art.  These cases are interesting in that Section 106(1) is absolutely off the table.  The original art is not reproduced, it’s just cut up.  So, if we think original authors should have some say over cut-and-paste-type activity — a question I’ll leave for another day — the only way to do it would be through the derivative-work right.

Another line of cases where Section 106(2) is clearly not redundant to Section 106(1) is a line of cases that arguably involve no expressive overlap between the original and accused works.  An answer key to a textbook might be one example, in that it is easy to imagine crafting an answer key that doesn’t repeat any expression from the associated college textbook.  And so, again, if we think that textbook authors ought to have a say over what happens in the answer key market, the only way to get there might well be Section 106(2).

There are more examples along these lines.  My point, though, is that while in most cases Section 106(2) is redundant to Section 106(1), there is at least a pocket of cases where Section 106(1) is not plausibly implicated and hence Section 106(2) would need to do its own work.  It is in those instances that it’s particularly important that we draw the right boundaries around the derivative-work right.  The rest, while interesting fights to be sure, are distractions from the heavy lifting that Section 106(2) might need to do.