Thoughts on the Scrabulous Lawsuit, Part III: The Question of the Copyrightability of Games

In Part I of this series, "The Complaint," I considered Hasbro's complaint in its suit against the creators of Scrabulous, with a particular focus on the copyright claim. In Part II, "The Mystery of Alfred Mosher Butts," I discussed a potential historical problem with that claim, namely that it is not clear that the board was published by Butts, the original author, with the required copyright notice, which would invalidate the copyright in at least the game board (Hasbro's strongest claim).

Today I want to focus on the rule that's forcing Hasbro to locate its copyright in the game board, rules, tile set, Scrabble dictionary, etc. -- everywhere except the Scrabble game itself. To wit: "Games are not copyrightable." See Nimmer on Copyright, § 2.18[H][3][a] (citing sources). Why is that the rule, and does it make any sense?

One thing that immediately stands out about the rule that "games are not copyrightable" is that it's a very narrow rule. As the Nimmer treatise puts it, "It is said that games are not copyrightable, but this general proposition is subject to qualification." Id. (emphasis added). The qualifications are extensive. As I mentioned on Friday, while the "game" is not copyrightable, all of its constituent elements are -- the board, the playing pieces, the statement of the rules, the box art, playing card designs, etc. This makes the rule a rather odd one. It's a bit as if there was a rule that "novels are not copyrightable," but a novel's plot, characters, setting, dialog, and cover art all were. What would be the point of such a rule? Nimmer's introduction of the rule -- "it is said" that game are not copyrightable -- seems to indicate his skepticism of its value as well. However, I think the rule has a purpose; it demarcates a difficult boundary for the copyrightability for games, which otherwise would be susceptible to simplistic analogies to other creative works.

The cases in which the traditional rule is stated offer little help in determining a precise theoretical basis for it. Many of those cases date from the 1920s and 30s, and they often involved simple games without their own boards or playing pieces -- Acey-Deucy, roller derbies, promotional contests, basketball tournaments, and variants of tiddly-winks. Almost none of the cases provide a reason for the rule, other than that previous cases have said so. The justification that is most often given in the literature is that the game itself is essentially equivalent to its rules; and the rules of a game are instructions, like recipes, which are also uncopyrightable.

That is, the story goes, it's Section 102(b) that bars copyright in games. Section 102(b) bars copyright in "any idea, procedure, process, system, method of operation, concept, principle, or discovery." The rules of a game are like the instructions in a recipe -- both are commands to human readers about how to behave. Recipes are excluded under § 102(b) as "processes," a series of steps to accomplish a certain result, namely the finished dish. Games might be thought of as "processes," namely a series of steps necessary to produce entertainment for the players; or perhaps a "method of operation" of the game pieces (if any); or perhaps a "system," like Selden's accounting system in Baker v. Selden, for governing leisure activity. See Nimmer, § 2.18[H][3][a] ("[N]o copyright may be obtained in the system or manner of playing a game....").

There are two problems with this theory. One is that it is not at all clear which, if any, of the § 102(b) exclusions games fit under. The rules of a game are not instructions for play in the same way that recipes are instructions for making a cake; they do not fully specify what occurs during play. Game rules are thus not a "process" or "procedure" for carrying on a game. Think about Scrabble. The rules provide the initial state of the game -- X players, 7 tiles per player randomly drawn from the fixed set of 100 tiles, with a procedure to determine who goes first. The rules then state it is the first player's turn, and place some boundary constraints on what the player may do (e.g., he or she must place a word on the center square). But within those very broad constraints, the rules do not specify at all what word the player should put down. He or she is free to put down "it" or "tea" or "hate" or "tithe," all from the same draw of 7 tiles. The second player has even more freedom. If a computer were executing the Scrabble "process," it would grind to a halt without further input.

Perhaps, then, as Nimmer suggests, the game is a "system" of playing. The meaning of "system" in § 102(b) is a little obscure; the dictionary definitions include "1: a regularly interacting or interdependent group of items forming a unified whole <a number system> ... 3a: an organized or established procedure <the touch system of typing> b: a manner of classifying, symbolizing, or schematizing <a taxonomic system> <the decimal system>...." A game doesn't seem to be an "organized or established procedure" for the reasons considered above; and while it may be an "interdependent group of items forming a unified whole," excluding all "unified wholes" from copyright protection (e.g., movies?) seems overly broad.

Pamela Samuelson, in her fascinating recent study of the history and meaning of § 102(b), was able to conclude only that "the game case law ... is consistent with [the § 102(b)] exclusions" (emphasis added), suggesting that games might be excluded under the spirit of § 102(b) even if they are not within the letter. But even that is not obvious. The § 102(b) exclusions, as Samuelson makes clear, have their origins in Baker v. Selden, a case that at bottom is about keeping the realms of patent and copyright distinct. Selden's copyright in his book about a method of accounting did include the practice of that method, including any forms he devised that were necessary for that method. The reason was that such methods were subject to the more stringent requirements of getting a patent in order to be protected. Under modern caselaw, however, the boundary of patentable subject matter may not include a game per se. In order to be patentable, a process or algorithm ("procedure") must produce a "useful, concrete and tangible result." State Street Bank & Trust Co. v. Signature Fin. Group, Inc., 149 F.3d 1368 (Fed. Cir. 1998). The rules of a game, however, just produce the entertainment of the game. State Street Bank stretched its test a little to find that updated mutual fund prices are a "useful, concrete, and tangible result." But it would seem to stretch that principle past the breaking point to say that having fun is a "useful, concrete, and tangible result."

If "process" and "system" are broad enough to encompass games as methods of producing entertainment, then that leads to the second difficulty with § 102(b) as the basis for the traditional rule -- namely, that interpretation would swallow forms of works that are clearly copyrightable. Two venerable forms of copyrighted works come in the form of instructions: sheet music and scripts for plays. Why should those be protected, but games are not? Even the terminology used to describe all three activities indicates similarity. Games, music, and dramatic works are all "played" by "players" who follow the instructions they are given. The Copyright Act clearly recognizes the play of music or a script as within the scope of the copyright owner's exclusive rights -- the right of public performance. Yet there's no rule that "music composition is not copyrightable" or "plays are not copyrightable." Perhaps, as Thomas Hemnes suggests, "'Copyright does not protect games' is an example of a principle of law that may have grown larger and more rigid than the facts of the seminal cases warranted."

This post was harder to churn out than I thought it would be, given that I've been mulling over this question for 5 years. So I'm going to end here, and pick up again tomorrow in Part IV of my 3-part series: "The Answer to the Copyrightability of Games."