On the Authorship of Law II: The Social Life of Building Codes

By Gordon Hull

As I noted last time, the Supreme Court has decided to take up a case about copyright in state codes.  Specifically, Georgia contracts with Lexis to produce an annotated version of its code, which is the state then blesses with the title “Official Georgia Code Annotated” and claims copyright in.  The question is whether the annotations are part of the code; if they are, they are public domain because the law is public domain.  The 11^th Circuit said that they are, because the legislature officially adopts them, courts refer to them, etc.  If it walks like a duck…

One of the decisions cited in the 11^th Circuit opinion cites along the way establishes that model building codes, once incorporated into statute, lose whatever copyright protection they had.  In Veeck v. Southern Building Code Congress International (293 F.3d 791 (5th Cir. 2002)), SBCCI was “a non-profit organization consisting of approximately 14,500 members from government bodies, the construction industry, business and trade associations, students, and colleges and universities.”  SBCCI’s purpose was to develop model building codes for municipal governments to adopt, which the small north Texas towns of Anna and Savoy did.  Veeck ran a web site about northern Texas, and wanted to put the building codes online.  When he had some difficulty getting them from Anna and Savoy, he paid SBCCI $72 for the codes, and then posted that online, correctly labeling them as the building codes of Anna and Savoy.  The question, then, was whether in being enacted as part of the municipal law of Anna and Savoy, the codes lost the copyright protection they enjoyed as products of SBCCI.  The 5^th Circuit, relying on the premises that “law” is not copyrightable, the copyright idea/expression dichotomy, and extant caselaw, ruled that the codes were no longer copyrightable.

If the Georgia case attempted to think through what the conceptual underpinnings of the thesis that law is not copyrightable, SBCCI offers a chance to think about what that means in practice, and how it interacts with the more commercial IP system.  Indeed, one of SBCCI’s arguments in favor of protection was quite precisely the commercial incentives justification for copyright.  I want to approach all this somewhat elliptically.  Quite some time ago, I used Deleuze’s critique of Platonism (in Difference and Repetition and Logic of Sense) to suggest that the original/copy distinction in copyright functions like the eidos/copy distinction in Platonism.  For Deleuze this distinction isn’t about metaphysics so much as police work: it’s about knowing how to distinguish legitimate copies from illegitimate simulacra.  Deleuze writes:

“The true Platonic distinction ... [is] not between the original and the image but between two kinds of images, of which copies are only the first kind, the other being simulacra.  The model-copy distinction is there only in order to found and apply the copy-simulacra distinction .... The function of the notion of the model is not to oppose the world of images in its entirety but to select the good images, the icons which resemble from within, and eliminate the bad images or simulacra” (Difference and Repetition, 127).

Similarly in copyright, I argued, the point is to distinguish between authorized copies and illegitimate ones.  Digitization then made this distinction very difficult, putting enormous pressure on the law itself, which was tending to devolve into the view that all copies were bad.

Here it seems to me that we see the reverse.  When Savoy, TX adopts the model building code into its law, it is no longer possible to determine whether a given copy is the law, or the model code.  Since the law is, by definition not copyrightable, then the police function points in the other direction, and declares that all copies are acceptable.  Indeed, this is precisely why Veeck was able to buy the building code from SBCCI, strip the copyright notice, and put it online: it is impossible to determine if he derived his copy from the statute at the Courthouse, SBCCI, or some third source.  Both are expressions of the building code in some medium or another.  Indeed, since they are identical, the entire original/copy distinction starts to break down.  Thus for an initial Deleuzian analysis.

At the same time, the case presents an opportunity to refine the analysis a bit.  If the original/copy distinction doesn’t come up the idea/expression one does.  Basically, copyright covers the “expression” of an “idea,” but not the idea itself.  So for example “play about star-crossed lovers” would be an idea and not copyrightable, but the text of Romeo and Juliet would be (it’s too old for copyright of course).  Similarly, the idea of ‘reality TV’ is not copyrightable, but Survivor is.  The distinction can be messy, but the reality-TV idea has been used to show that the dichotomy itself can be quite generative: after all, there has been a lot of post-Survivor reality TV.  Here, the Court invokes the limiting “Merger Doctrine:” if there is only one possible expression of an idea, then it becomes uncopyrightable.  As the Court puts it, here, “because there is only one way to express the meaning of the building codes, the ‘idea’ embodied in the law merges with SBCCI's expression, and at that point, renders copyright protection unavailable.”  As this suggests, the Merger doctrine graphically illustrates the collapse of the idea/expression distinction.

In this sense, idea/expression and original/copy are in the same line of police work: they are there to do the same work of policing illegitimate copies.  In the case of original/copy, it provides a standard on the basis of which to distinguish good from bad copies: good copies are those that correctly imitate the original.  In the case of idea/expression, it provides a way to elevate something to the level of eidos, so that it can then be used to police good/bad copies.  The two doctrines thus work in tandem.  The Merger Doctrine gives the game away: the idea/expression dichotomy embodies some metaphysical niceties that do some serious juridical work. 

This is not to condemn the work that these concepts do!  After all, the criticism of “transcendental nonsense” was meant to direct attention to the work that legal concepts do.  The more interesting point here, I think, is that the difficulty in grounding an idea/expression distinction in the case of code books cum law has as an unacknowledged consequence the inability to make the original/copy distinction.  In some sense, Veeck quite clearly copied the code.  The problem is that the copying didn’t produce a “copy,” but instead, as it were, another “original,” because the law is defined as the sort of thing for which the good/bad copy policework is defined ex ante as incoherent.  Put differently, from the point of view of copyright, it’s actually not possible to copy the code: all one can do is present it or present something else.  If the code book text and the law are the same, then it’s not possible to tell which copies are “good” or “bad,” because reference to the eidos returns the same thing every time: all the copies are perfect.  But that is true in this case partly because there is no expression of the eidos that is not identical to the eidos itself.  In that sense, there is neither original nor copy.

This suggests one might borrow a page from Arjun Appadurai, one might think about the codes as things, as artifacts with social lives.  Appadurai is thinking about “traditional” exchange relations as found in places like bazaars, but his point works here too.  He writes, “the commodity situation in the social life of any ‘thing’ [can] be defined as the situation in which its exchangeability (past, present, or future) for some other thing is its socially relevant feature” (Social Life of Things, 13).  He then notes that “things can move in and out of the commodity state, that such movements can be slow or fast, reversible or terminal, normative or deviant” (ibid.). At any given moment, a thing might possess “candidacy” to be a commodity, which refers to “the standards and criteria (symbolic, classificatory, and moral) that define the exchangeability of things in any particular social and historical context” (14).  From this point of view, the Copyright Act establishes candidacy criteria by which expressions of ideas can be considered commodities.  Among them is that they not also be law, since law is excluded by rule.

In one moment, it functions as a commodity, as when it is produced by SBCCI and promulgated in their various publications.  When Savoy incorporates it into their code, it enters a different phase of its social life, no longer commodity but law.  Savoy’s incorporation of law accomplished by judicial act what would otherwise happen with age and the expiry of copyright: the end of the life of the commodity as a commodity.  In the meantime, SBCCI can still sell all the code books that it wants.  But the code itself has entered into an afterlife as law.