The Lexicon Issue: A Retrospective

by Dan H

Dan is up to date as ever
So a little while ago a concerned citizen popped up to say that they (I'd say she from the LJ handle but one doesn't like to presume) felt they should tell me that they had “laughed so hard at my cluelessness” in this article. She/he also kindly provided me with some links to discussions of the case, which I duly read and from which I was forced to conclude that, in the language of this commenter “cluelessness” means “being broadly correct about everything.”

What I objected to at the time was the fact that Rowling's objections were stupid and irrelevant, and people who have a better understanding of the law seemed to agree. There's a rather good list here of the various points people brought up at the trial, conveniently broken down into “stuff that is legally relavent” and “stuff that isn't”. You might notice that high up on the list of things that aren't relevant is “how lazy, sloppy or inaccurate JKR considers the Lexicon to be”.

Now this is mostly water under the bridge, the verdict is in – although it's possible there's appeals going on. A lot of people had good reasons to think that the court's rulings were a bit shonky in places (several people with actual law degrees were confused by the fact that the court ruled that the Lexicon wasn't a derivative work) but to be honest I think it all came out right in the wash – some decent analysis here sums up the key points pretty well and what it boils down to is “the Lexicon copied too much.”

Ironically, the Lexicon was – as the judge observed – a victim of its own enthusiasm. The basic concept was legitimate but it used too much original language (a good example from one blog is describing the sounds made by the “clankers” - the things that scare away the blind dragon in Deathly Hallows – as “like tiny hammers on anvils”). If Vander Ark had been more willing to mess with the Holy Writ, he'd probably have been absolutely fine.

But mostly, what I want to talk about here is copyright law, because I think it's kinda cool, and I'm kind of on a roll here with articles about stuff I know jack shit about.

Derivative Works

One of the most peculiar things about the Lexicon ruling was that it held that the lexicon was not a derivative work. The basis for this is rather obscure, but it seemed to be that a compilation of information about a work was considered sufficiently transformative that it was not considered a derivative work. This seems reasonable to me, but also seems to conflict with US legal precedent (specifically with Castle Rock vs Carol Publishing in which a Seinfeld trivia book was ruled to constitute a derivative work, and therefore to be in violation of copyright).

I suspect that this basically comes down to that old aphorism about laws and sausages. The definition of a derivative work is unclear (and the definition of fair use even more so). The strongest interpretation seems to be that even making reference to copyrighted material makes a work derivative – this seems peculiar to me, unless you're going to argue that “derivative” means “contingent upon the existence of”. This seems to be the logic that held sway in Castle Rock - the court ruled that the “fictional facts” of Seinfeld were protected by copyright. This is a peculiar idea in and of itself and one to which I will return shortly.

Of course the weakest definition of a derivative work – a direct adaptation to another medium – is also unsatisfying. Intuitively, it seems reasonable that fanfiction, for example, be considered derivative (in the legal as well as the literary sense), and certainly few people would dispute the fact that only JK Rowling has the right to produce an eighth Harry Potter book.

Where this becomes problematic is that “derivative work” is actually an extremely powerful term in copyright law. Authors are assumed to have an absolute monopoly on derivative works. This is a big deal – monopolies are generally a bad thing and it's relatively rare for them to be protected by law. Given the stakes, it's entirely predictable that while one court feels that Castle Rock Entertainment has the right to control the production of Seinfeld trivia books, another feels that JK Rowling does not have the general right to control books of information about her fictional world. It's a murky area of law and one with no clear right answers. Most people would – I think – accept that JK Rowling has the right to decide who can make movies out of her books (although Derek Bambauer argues here that she shouldn't – at least from an economic perspective) but I suspect most people would also accept that she has no right to decide what people write about her books.

Fictional Facts

One of the strangest aspects of US copyright law I dug up in my recent trawling through the intarwebs was the notion of “fictional facts” - this was a key element in the Castle Rock case, in which it was ruled that authors (or in this case entertainment companies) do retain copyright over matters of fact in their fictional worlds.

I have a policy when it comes to matters of law, which is to assume that if it looks like the law is made of stupid that it's probably covering up something else which is even more made of stupid.

Because on the face of it, the idea of “fictional facts” seems – well it seems pretty made of stupid.

We'll leave aside for now the fact that it's an oxymoron of the highest order, and focus on the weird implications. If I'm understanding the precedent correctly, the statement “Harry Potter is a Wizard” (or for that matter “Dumbledore is Gay”) is protected by copyright, due to its being a “fictional fact” created by JK Rowling. Now most uses of that statement will wind up being protected under fair use but it still seems to be based on the principle that authors (at least in theory) have the right to control information about the contents of their books, which seems perverse.

I don't want to go too far into slippery slope arguments here, but it does strike me that treating “facts” as copyrightable puts spoilers in a difficult legal position. I don't actually think that anybody will ever get sued for spoilering, or that any court in the land would uphold an anti-spoilering case on copyright grounds, but by a strict application of logic, spoilering looks a lot like it breaches copyright. A spoiler consists of the repetition of a fictional fact (which is copyrightable material), the act of spoilering cannot be said to have transformative value (indeed many argue that spoilers detract from the value of the original work), and a case can be made that spoilers directly compete with the author's original product (insofar as a person could, quite reasonably, decide not to read a book or see a film as a result of having been spoilered for it). A sign bearing the legend “Snape Kills Dumbledore” is, in essence, a derivative work which – since it consists only of copyrighted material repeated without commentary – may not be protected by fair use.

Again I should clarify that I don't necessarily think this is a problem. In America at least, a Snape-Kills-Dumbledore sign would be protected by freedom of expression, and the constitution trumps copyright last time I checked, but it does highlight some of the weirder implications of this idea of “fictional facts”. I also suspect that the distinction between a “fictional fact” and a – for want of a better term - “factual fact” is a narrow one. Part of the reason that the Castle Rock ruling went against the defendants seemed to be that their Trivia book had focused exclusively on episode content and not on questions about (for example) the cast or sets – such questions would clearly have been matters of factual fact and not protected by copyright. So perhaps what it boils down to is that while “Dumbledore is Gay” is a fictional fact protected by copyright, “JK Rowling declared in interview that Dumbledore was gay” is a factual fact and therefore fair game. This seems like a silly distinction, but it probably matters rather a lot.

For a start, people will in fact pay for fictional facts. One of the biggest points against the Lexicon back in 2008 was the fact that it had reproduced a lot of information from Quidditch Through the Ages and – that one about magical beasts the name of which I can't be bothered to look up – both of which were sold primarily as books of fictional information. The fact is that people do like to know More Stuff about fictional realities, and they will pony up real cash to find out More Stuff. The Harry Potter Lexicon does tell the reader a lot of Stuff about the Wizarding World, and much as I hate to admit it, some people really do read novels purely or primarily to acquire facts about a secondary reality (I think these people are culturally moribund, but they seem to exist) so from a certain point of view it does make sense to see the “fictional facts” of the Potterverse as having value and requiring protection.

On top of this, if “fictional facts” are not protected, then it becomes very difficult to see how the law protects authors from things like unauthorised sequels. If “Harry Potter is a Wizard” is not on some level protected by copyright, then it becomes difficult to see why I cannot write a book about a Wizard called Harry Potter with as much impunity as I could write a book about, say Napoleon Bonaparte or Abraham Lincoln. If we accept the (seemingly common sense) idea that the basic facts of fictional settings should be fair game for use and commentary we tacitly allow people to recreate other people's work from whole cloth. If I have unlimited license to refer to the facts of somebody else's creation, then in practice I have unlimited license to reproduce their work (since after all, any text is just information about the content of that text).

All of this leads to a rather difficult situation. Copyrighting facts seems dangerously close to copyrighting ideas (which would be a terrible, terrible precedent), but not copyrighting facts seems dangerously close to not copyrighting anything.

Who is Copyright For Anyway

Copyright is one of those areas of law that everybody thinks they understand but in fact nobody does. Intuitively we all get it. You create something, it's yours and you get to control it. A lot of people take this as a kind of moral axiom: these are the people who literally believe that JKR has the right to call the shots in all things Potter related, be it the production of reference guides, Dumbledore's sexuality, or whether or not Snape was redeemed. This is the “it's her world, we're just playing in it” doctrine.

The thing is that this is a naïve approach to the law. We can't just say “playing with other people's toys is naughty, so ban it”. Copyright exists for quite important economic reasons and, contrary to popular belief, those reasons have comparatively little to do with stopping people from bootlegging stuff.

Copyright is generally considered necessary because in a perfectly competitive marketplace, the price of any good tends towards its marginal cost of production, the price of creating one more unit. As long as you can sell a unit of a good for more than you spent to make it, you should and somebody will. This works great for bananas, coffee tables, bricks and door handles, but it doesn't work so great for books, music and video games. The marginal cost of producing a copy of a book is very small indeed (and the marginal cost of producing a copy of an ebook is effectively zero). If authors were not allowed a monopoly over their work, they wouldn't be able to sell it, because any price they could sell it for, somebody else could sell it cheaper.

Just to be clear here, this very much isn't about piracy. Piracy is a crime (well actually it's probably a tort, but let's not split hairs here) and crimes, sort of by definition aren't prevented by the law. Copyright doesn't stop people illegally reproducing copyrighted material because, well duh. FACT and its associated bodies would have you believe that Copyright Is Good because it Protects Authors from Bad People. This is stuff and nonsense. Copyright is good because it protects publishers from other, better funded publishers.

Consider: you are Bloomsbury, on the verge of bankruptcy you discover a promising children's author by the name of Rowling. These books get inexplicably popular. You celebrate.

Consider: You are every other publishing company in the world. You notice that Ms Rowling's books are getting extremely popular, you also notice that Bloomsbury, having paid the author an advance, paid the salaries of editors and proofreaders, hired cover artists and so on, has incurred a great many costs which you can avoid, simply by taking their product and reproducing it (using the resources which, as a large and established publishing company, you most certainly possess). The resulting competition drives down prices, which is fine for you but not so great for the company that has paid the substantial setup costs. Every other publisher in the world makes a tidy profit, Bloomsbury goes bust.

Worse, nobody wants to pick up the option on the next book in the series, because everybody knows that their competitors can sell flawless copies of the book more cheaply than they themselves can afford to sell them. Even if JK Rowling wanted to sell her next book, nobody would buy it, because everybody would know that whoever published first would incur large costs for little reward (this is true even given the substantial first-day sales for popular books, many ordinary customers would rather wait a couple of weeks and pay half the price).

Conventional publishing and distribution models are founded on the notion of copyright. The problem is not, as many assume, that people wouldn't write books if they weren't sure they'd make a lot of money from it. Thousands, probably millions of people are working on novels right now with no guarantee of financial reward. Thousands of people put their work online for free as a matter of course, and an awful lot of people actually pay vanity publishers for the pleasure of seeing their work in print. Copyright isn't there to reassure authors that they'll be paid, it's there to reassure publishers that they'll recoup their losses.

Now of course you can argue that the conventional distribution model for novels and the like is inherently broken because, well take your pick, information should be free, corporations shouldn't tell us what to like, whatever. Speaking personally, though, I actually have a lot of faith in the conventional models of publishing, at least for the mass market. Indie and self-publishing is great for niche materials (the indie-RPG industry, for example, works well because it services a small community and everything it puts out is effectively peer reviewed by the community it serves) but not so great for novels and the like. Anecdotal evidence here at Ferretbrain supports the observation that self-published books really are less good than those that are published conventionally.

The public interest is generally served by allowing authors, and by extension publishers, to control distribution of their work. This means that the commercial interests of publishing companies are served by seeking out high quality authors (allowing them exclusive control of a valuable resource) rather than by seeking out more efficient means of distribution (allowing them to better exploit the resources developed by others). It is not so easy to see how the public interest is served by allowing authors to control derivative works, particularly if the term is defined so broadly as to include things like reference guides. Put simply, I do not believe that one single person has ever been dissuaded from writing a novel by the fear that somebody might write a reference guide to it at some point in the future. Nor do I believe that any publisher has ever refused to publish a book on such grounds.

From this perspective, derivative works rights are a lot harder to justify. While it feels intuitively right that you shouldn't be able to make Harry Potter tie-in material without JKR's say-so, it's not immediately clear why: sure it might make her upset, but “it will upset people” is generally not a good reason for legislation. A hard economic argument would say that if there is a market for something, and the production of that thing will not be generally detrimental to the public good, then people should be allowed to make that thing. Ultimately, shouldn't it be up to the Invisible Hand to decide whether – say – a fan-made guide to a fictional setting is worth producing? This might lead to a market inundated with trashy cash-ins, but if there's a market for trashy cash-ins then those cash-ins have real economic value. Of course they might harm the value of the individual property but to my (admittedly limited) understanding, that becomes a trademark rather than a copyright issue (I can't go around putting the coca-cola logo on things, but that isn't because it's copyright, it's because it's a trademark which is a subtly different thing).

This article doesn't really have a conclusion beyond “good lord copyright law is complicated and unintuitive”. I shall end, therefore, by sharing the irrelevant tidbit that “uncopyrightables” is the longest word in the English language which uses all of its letters exactly once.

bookmark this with - facebook - delicious - digg - stumbleupon - reddit


Show / Hide Comments -- More in March 2011