I’ve long been interested in copyright and games — an interest that began with copyright and video games, but worked its way backwards to consider games generally. Games exist at the boundary of copyright law: they seem to include much that is protectable, and yet there is a general rule in copyright doctrine that games are not copyrightable. (For more, see my four–part series on PrawfsBlawg in 2008, in particular Part III and Part IV; also this post).
I’ve now uploaded a new paper to SSRN, Games and Other Uncopyrightable Systems, that explains the purpose and argues for the continued vitality of that rule. Some may recognize the paper as what used to be Part I — the “background” section — of my long-awaited video games paper. The questions surrounding the copyrightability of games proved to be so intricate that it required a separate paper just to address them. In short, games are uncopyrightable because they are systems — a conclusion that is only moderately helpful, because systems themselves are not well understood. I therefore tackle that issue as well. Here is the abstract:
This article solves two puzzles in copyright law. First, it has long been blackletter law that games are not copyrightable. But the origins of this rule are lost to history, and the reasons for it are not obvious. Second, it has never been adequately explained what makes something a “system” excluded from copyright protection under Section 102(b) of the Copyright Act. Modern courts interpret “system” as merely a synonym for “idea” or “process,” two other categories of exclusions. Others have interpreted it using the broadest definition in the dictionary, which would sweep in large amounts of copyrightable material as well. Neither definition gives the term any meaningful content.
Like solving a crossword puzzle, this Article uses each of these questions to shed light on the other. Games are uncopyrightable because they are systems. The case law that led to the adoption of Section 102(b) demonstrates that systems are schemes for transforming user inputs into a correlated set of outputs. Games do exactly that. A game is a scheme for transforming player activities into moves within the game. The reason why games and other systems are uncopyrightable then becomes clear: the purpose of a system is to serve as a forum for user activity; it is users, not authors, who provide the primary informational value to the outputs of a system. Games and other systems are excluded in order to fence in copyright protection before it reaches user creation.
Cross-posted at Marquette University Law Faculty Blog.
Cool — really looking forward to reading this!
That makes for an interesting read, particularly in light of the IP issues emerging in the context of online gaming — e.g., can I trademark the name of the business I created in virtual reality? Who owns the rights to my avatar? Can a bard bring a copyright infringement action against someone who made a bootleg tape of a concert that only occurred in the confines of virtual reality (like the concert Ben Fold Five put on in Second Life, attended by paying avatars)?
Gaming seems to be the new font of brain-teasers for intellectual property.
One of the footnotes states that section 102(b) was intended â€œto restate, in the context of the new single Federal system of copyright, that the basic dichotomy between expression and idea remains unchanged.â€ The paper states that certain courts have taken this view. I didn’t understand why there is a pressing need to find a further explanation.
I really enjoyed it. I want to read some of the cases you cite and Pam Samuelson’s paper before I offer anything definitive, but just some quick reactions.
* I, too, have been puzzled by the string of words in 102(b) and how they should be interpreted. The words seem to run in different directions, with little guidance as to how they form coherent groupings — interesting to see you making a stab at that.
* I really enjoyed your ludiological analysis of games (Suits, Huizinga, Zimmmerman/Salen, Juul, etc.)
* I think viewing games as “state machines” is quite helpful in understanding how they are different from other cultural forms
* I agree that the play/performance issue is central to the puzzle here — it’s hard to reconcile the doctrinal divergence between scripted and unscripted (but still scripted) performances. It’s a hard line to draw, I think.
* When you were making your case for game players as different than actors at the end, I started to wonder about other possibilities for the divergence. So, e.g.:
What about fixation? (Games, historically, have not been fixed forms capable of subsequent enjoyment.)
What about utility? (The pleasure of games is not from the textual script but from the action the rules enable — hence, the rules have a sort of functional utility.)
What about collaborative authorship? (Always vexing for copyright.)
What about the fact that we have of a set of discrete elements lacking fixed arrangement yet providing utility when arranged? (Reminds me of typefaces.)
What about scenes a faire? (E.g. ball games tend to share certain rules.)
These just popped into my head as alternative ways to think about copyright and games. I think the gist of what you’re saying is that protecting the value of the play experience requires diminishing the control over the rule set — and I think that’s right. But you could probably find other angles to explore here.
* Finally, couldn’t you take that main observation (about game play trumping game rules in a policy analysis) and bring it to bear on, say, remixes and derivatives? Aren’t both about instances of privileging downstream creative autonomy over upstream labor and investment?
Thanks everyone for your comments. Hugo, I think the problem with taking that sentence literally, as the Toro court did (and other courts have followed), is that it’s being taken out of context. I don’t believe Congress intended to codify only the idea/expression dichotomy, narrowly understood to refer only to “levels of abstraction,” in Section 102(b). It’s also inconsistent with the text, which if read as eight synonyms for “idea,” becomes somewhat ridiculous (reminiscent of the Sesame Street skit I reference, in which the hotel owner takes down an order for “fish, fish, fish, fish, fish, fish”).
The “pressing need” is a bit harder to pin down. I’m not sure it’s correct to say, as some have, that the failure to exclude “systems” means that copyright will run amuck. But I think it is true, as I suggest in the paper (p.21), that it at least puts pressure on courts to jury-rig other doctrines to exclude materials such as parts numbers and blank pitching forms that would be easier to analyze as uncopyrightable systems or products of such a system.
Greg, your last two bullet points — the critical part of your comments — nail down some questions I flew by. (I wanted to sell the paper as being under 20,000 words, and I just made it — but that’s not to say that I couldn’t make it more efficient). There’s a few different ways of characterizing what I’m arguing. One is the way that I focus on in the paper — games provide frameworks for user input, and are therefore copyrightable systems. Other ways are to talk about authorship (e.g., the NBA v. Motorola case on sports, quoted toward the end of the paper) or fixation: the primary part of the content in a “game-instance” is not fixed in the “game-form,” or “game-in-the-box” as I call it. I wanted to get away from describing that as a fixation problem though for a couple of reasons. My sequel to this paper will focus on video games, and there the early courts dealt with a fixation argument — but that has always seemed mistargeted to me. Characterizing the issue as fixation makes it seem like the problem is one of unfixed performances. But I don’t think game play is a performance at all — not of the game, and not of any work by the player either. (I’m talking about normal game play, not machinima or the like.)
I have to get ready for class but will try to respond to the remaining points later.
Two more points–re: utility, I think there’s a general sense that that is the dividing line between various of the categories in 102(b) and copyrightable expression. But I don’t think that works, entirely. The problem is that some things are useful for producing entertainment or information — games, but also sheet music and scripts. Then you start to get somewhat circular — a process is copyrightable expression only if it produces copyrightable expression. Toward the end of the paper, I argue that more than utility is needed to draw the line copyright wants to draw–you need to look at both ends of the communication. If the material in question produces something other than a communication directly from the author to the recipient, then that’s excluded.
Re: remixing/derivative works, I’m actually trying to be very careful with how far the argument extends. I see remixes/DWs as importantly different. Game play is where the game is experienced, similar to reading or watching. And most of that, as I indicate above, is not arising from the fixed expressions of the game designer but from player activities. Remixes and DWs are a different situation. There, you do have a communication from author to recipient. That communication may give rise to various reactions in the head of a user, in a way the user doesn’t have full control over. And one way of continuing or expanding on that experience might be to spell out expression of one’s own incorporating the previous communication. But once that merged expression starts passing out the other side of the user’s head, so to speak, it’s different from gameplay in that it is incorporating large parts of the message from author to user. The issue with game play is that there is no there there — the game-instance is not the game designer’s creation and is not expression at all, although the game designer’s creation allowed it to happen. But it is also true that I just don’t understand what the derivative works right is supposed to be anyway — that’s a subject for another day, perhaps.
Hi Bruce —
Fair points all and thanks for the additional explanation.
Just to be clear, I do think you’re right in not lumping games together with derivative works. But still I am interested, recognizing that we’re talking about two different things, in the doctrinal similarities that might be constructively employed in the derivative works setting.
Re the nature of games, I think the distinctions work really well for, say, Roller Derby or hockey or chess. I’d be interested in seeing what the game theory / game design folks make of this distinction, though. Many of them work in the video game arena, which as you note has separate issues, but what about a game like Brenda Braithwaite’s Train:
That’s a rather unusual game, but I think it points out the way that the game/performance line, while valuable to draw, will inevitably have some fuzziness about it.
Thanks Greg, these comments are great. You’ve zeroed in right on what is unresolved and unexplained in my paper.
I’ll check out that game you link to. I think there are bound to be all sorts of problematic cases out there. In-person role-playing games, for example, at least begin to slide away from the view of game-play as not player expression; that’s not a key element of this paper but is more important a piece of my video games paper. I certainly recall trying to figure out appropriate ways for my character to react, which were not necessarily the same ways I would react. But it’s a fuzzy boundary; I remember having vaguely similar thoughts in hockey (“I don’t feel like punching this guy, but he just hit our goalie, and retaliation is required by the informal norms of hockey”). I haven’t checked it out yet, but I think to the extent the player gets more involved in spinning out a story of their own, or watching a story, it’s less game-like. I view it as somewhat similar to the way complexity exists on the edge of chaos–if the system is too static, you have only “order,” and if it’s too dynamic, you have only “chaos,” but if it’s right in the phase transition between order and chaos, you have complexity, with all sorts of resultant emergent properties. If the players are just watching a story unfold, that’s just “consuming” a work; if they are doing too much creation, that’s extemporaneous expression; but if they are constrained and given a goal to pursue, that’s where game-play occurs.
That’s very consonant with some of the broader notions of “play,” e.g. the “nip is not a bite” idea of animal play; the engineering notion that a machine part that moves a bit too freely “has play”; the deconstructive idea of play as freedom to deviate from an an implicit structure. Fascinating stuff.
But I’m even more fascinated by your hockey comment. So, I take it, in hockey, that there’s this norm that requires players to (even if they would rather not) break an explicit game rule that prohibits fighting — a rule with penalty consequences. So fighting is explicitly not allowed, but it’s actually an essential part of the true game.
I wonder if there’s anything remotely analogous in law? Are there legal rules that we’re reasonably expected to break in that same sort of way?
I sort of accidentally selected an example of role-following that actually involves breaking the rules, rather than following a role assigned by the rules. But yes, in hockey (and probably many other sports) there are a number of norms that are widely known among the players, not all of which are actually consistent with the rules. The goalie one actually makes some sort of sense — since the goalie wears incredibly bulky pads, he/she is a relatively easy target and usually unable to retaliate on his or her own (although there are certainly exceptions). So a norm has developed that if the other team checks your goalie — particularly in the crease, where the rules do not allow it, but even outside the crease, where it’s technically permitted — or worse, punches or otherwise messes with your goalie, some sort of retaliation is required. (I think what constitutes “messing with the goalie” is going to provoke all sorts of disputes over boundary cases, just like “camping” in Counterstrike.) I actually remember at least one case where I enforced the norm, even though I generally tried not to incur penalties.
I think it’s an example of what Salen and Zimmerman describe as an “implicit rule,” although what’s interesting about it is that this “implicit rule” is in conflict with the explicit rules.
Re: your last question, absolutely, there are many. Try driving at or below 55 mph on I-95 sometime during moderately heavy traffic.