from the actually,-that's-the-point dept
For a few years now we've
discussed a few
times some of the confusion as to why Sherlock Holmes isn't considered in the public domain in the US, even though he probably should be. As we've explained, all of Sir Arthur Conan Doyle's Sherlock Holmes books
except for one are in the public domain. The Conan Doyle estate claims that having that single book under copyright means that the entire character is covered by copyright. Earlier this year, we pointed out that a noted Sherlock Holmes scholar (such things exist!) named Leslie Klinger had decided to
file for declaratory judgment that Sherlock Holmes is in the public domain, following a legal nastygram from the Estate, arguing that it needed a license fee for Klinger's latest book.
The Conan Doyle Estate has now filed its response to the motion for summary judgment, and it's an
astounding study of ignorance concerning copyright law and the public domain. While it admits that there are only ten short stories (from that one remaining book) that are under copyright, it still argues that those ten stories lock up pretty much everything else. First, it argues that the characters of Sherlock Holmes and Dr. Watson continued to grow as personalities in those last ten stories, and that the stories were non-linear (i.e., some took place earlier in their fictional lives), it more or less encompasses everything, even those public domain works.
The facts are that Sir Arthur continued creating the
characters in the copyrighted Ten Stories, adding significant aspects of each character’s background,
creating new history about the dynamics of their own relationship, changing Holmes’s outlook on
the world, and giving him new skills. And Sir Arthur did this in a non-linear way. Each of the Ten
Stories is set at various points earlier in the two men’s lives—and even late stories create new aspects
of the men’s youthful character. In other words, at any given point in their fictional lives, the
characters depend on copyrighted character development.
Of course, if that's true, it basically presents a way to make copyright on characters perpetual. You just need to have someone continue to release new works that have some minor change to the character, and they get to pretend you have a new starting point for the public domain ticker. That can't be what the law intended.
Astoundingly, the filing suggests that the public domain clock only begins
at the point where the "creation of the characters was complete." That is, so long as you never "complete" the character creation, they can never go into the public domain.
If the creation of the characters was complete in works published in the United
States before 1923, the characters are in the public domain. If, however, the characters as works of
authorship were only completed in copyrighted stories published in 1923 or after, the characters are
works of authorship protected by United States copyright law.
But, if that's true, then the whole concept of the idea/expression dichotomy is shot to hell. Copyright is only supposed to cover the specific expression, not the wider ideas, which would include the facets of the character that were added in those last ten stories. But, the Estate is trying to flip this all on its head by arguing that those minor character changes in the small number of covered works retroactively flows back to the character in the earlier stories.
And then it gets really wacky. The Estate seems to freak out that if the elements of Holmes and Watson were to be declared public domain, then you'd get (*gasp*!) a splintering of their characters -- a fate that seems so horrific to the Estate, which they assume must also horrify everyone else. To the Estate, there can be only one version of those characters:
Plaintiff’s position would create multiple personalities out of Sherlock Holmes: a “public
domain” version of his character attempting to only use only public domain traits, next to the true
character Sir Arthur created. But there are not sixty versions of Sherlock Holmes in the sixty stories;
there is one complex Sherlock Holmes. To attempt to dismantle Holmes’s character is not only
impossible as a practical matter, but would ignore the reality that Sir Arthur Conan Doyle created a
single complex character complete in sixty stories.
But, of course, it's not at all true that there is only one version of Holmes and Watson. There are infinite versions. There's the version that Conan Doyle intended. There are different versions that appear in each book. There are different versions that appear in various movies and television programs. There are different versions that live on in the minds of millions of readers and listeners and viewers who have experienced Sherlock Holmes's stories. In fact, the very nature of the public domain is
to encourage that sort of thing by enabling everyone to build off of those cultural artifacts to create something new and different, using those original works as resources. To pretend that it's one singular cultural concept, only encapsulated in the whole of the published works, such that the public domain clock does not start ticking until the final work is published is entirely illogical.
The Estate also has a pretty serious problem, in that previous case law is against them on this. In particular, the
Silverman v. CBS Inc. case, concerning the public domain status of "Amos 'n' Andy." In that case, the court ruled that the original radio scripts
and the characters were in the public domain, even though some later television shows were not. The Estate tries to distinguish this case by arguing that Amos 'n' Andy didn't change over time and were "flat entertainment characters" who could be described as "without the depth and complexity of a living person." But that, too, makes little sense. Again, it's clear from that ruling that the characteristics of the parts that went into the public domain are, also, considered in the public domain. That those characters didn't change merely means that there were no remaining characteristics that remain under copyright protection even as the TV shows might. That still works in this particular case, where all of the characteristics outside of the final ten stories are also in the public domain. That there are character developments in those final ten stories, any copyright should only apply to the
direct expression of those character developments.
The Estate also rejects the (fairly compelling) argument that a character is created in the initial work, and future works fleshing out that character are simply derivative works off of the initial copyright. That's a reading of the law that makes sense, but the Estate will have nothing to do with it, complaining that "it's not fair" to think of later works in a series as derivative. Honestly, that's about the extent of the Estate's argument here: that it's somehow insulting to suggest that later books in a series can be called "derivative" for the sake of copyright. But they seem to offer no legal theory to support this.
Of course, it's kind of ridiculous we're even having this fight. At the time those last ten stories were published (1927), the copyright law stated, clearly, that the
maximum amount of time that the works could be under copyright was 56 years. Since the structure of copyright law is that it's supposed to act, in part, as an incentive to get someone to publish (for the purpose of promoting the progress of science and learning), clearly that incentive was
enough. Any second beyond 56 years is taking the work away from the public domain, effectively going back on the deal with the public. The deal was "you get 56 years of exclusivity, and we get your book." Thus, it's insane and nonsensical that all Sherlock Holmes works haven't bee in the public domain since 1983. However, hopefully this case will show that nearly all of Holmes and Watson are, in fact, in the public domain.
Filed Under: arthur conan doyle, copyright, dr. watson, leslie klinger, public domain, sherlock holmes
Companies: conan doyle estate
