In its ongoing battle with federal law enforcement agencies over its refusal to unlock the iPhone, Apple has mounted a ‘Code is Speech’ defense arguing that “the First Amendment prohibits the government from compelling Apple to make code.” This has provoked some critical commentary, including an article by Neil Richards, which argues that Apple’s argument is “dangerous.”
Richards alludes to some previous legal wrangling over the legal status of computer code, but does not name names. Here is an excerpt from my book Decoding Liberation: The Promise of Free and Open Source Software (co-authored with Scott Dexter) that makes note of a relevant court decision and offers arguments for treating code as speech protected under the First Amendment. (To fully flesh out these arguments in their appropriate contexts, do read Chapters 4 and 5 of Decoding Liberation. I’d be happy to mail PDFs to anyone interested.)
Most commonly, software is deemed to be harmful when it facilitates the breaking of law, as in the DeCSS case, which was the first to test the enforceability of the DMCA’s anticircumvention provisions. In this case, the First Amendment was invoked in defense of the creation and distribution of the “guilty” software:the defendants sought to portray software as a medium for the expression and communication of ideas, not only between humans and machines but also among computer scientists. Implicit in the defendant’s brief is the concern that restrictions on such “speech” have broad and complex implications in technologized society.
In 1999, Jon Lech Johansen, a Norwegian student, wrote and published source code for software intended to circumvent DVD access control mechanisms. As part of the movie industry’s technical protection of intellectual property, DVDs were encrypted using the proprietary Content Scramble System (CSS), which is intended to allow only licensed players to play DVDs. At the time, the only computers with licensed players were Windows- and Macintosh-based; Linux users could not play DVDs — even if legally purchased — on their machines. Called DeCSS because it reverses the operation of CSS, Johansen’s software, circumventing the protective encryption, allowed DVDs to be played on Linux computers. As far as Linux users were concerned, this software merely enabled the fair use of their legally purchased DVDs, though the movie industry had a different perspective. In 2000 American authorities contacted Norwegian police, who raided Johansen’s home; arrested and tried in Norwegian court, he was fully acquitted in 2003.
In the United States, Universal Studios filed suit under the terms of the DMCA against Eric Corley, Shawn C. Reimerdes, and Roman Kazan, Americans who had posted a copy of the DeCSS code on the Web site 2600.com. In the ensuing legal proceedings, the defense argued that source code is subject to First Amendment protections, citing the landmark decision in Bernstein v. US Dept. of Justice, which had found that “the particular language one chooses [does not] change the nature of language for First Amendment purposes. This court can find no meaningful difference between computer language, particularly highlevel languages as defined above, and German or French.” The Court agreed to an extent, saying
It cannot seriously be argued that any form of computer code may be regulated without reference to First Amendment doctrine. The path from idea to human language to source code to object code is a continuum. As one moves from one to the other, the levels of precision and, arguably, abstraction increase, as does the level of training necessary to discern the idea from the expression. Not everyone can understand each of these forms. . . . Only English speakers will understand English formulations. Principally those familiar with the particular programming language will understand the source code expression. And only a relatively small number of skilled programmers and computer scientists will understand the machine readable object code.
But the Court, as a preamble for its finding in favor of placing restrictions on code, continued, “the long history of First Amendment jurisprudence makes equally clear that the fact that words, symbols and even actions convey ideas and evoke emotions does not inevitably place them beyond the power of government.” While this has never been contested, even by First Amendment proponents, the Court seemed to be failing to confront the question of whose rights would be protected, and whose infringed, by such restrictions.
The Court concluded that the functional nature of code overshadows its expressive, speechlike aspects:
The computer code at issue in this case, however, does more than express the programmers’ concepts. It does more, in other words, than convey a message. DeCSS, like any other computer program, is a series of instructions that causes a computer to perform a particular sequence of tasks which, in the aggregate, decrypt CSS-protected files. Thus, it has a distinctly functional, non-speech aspect in addition to reflecting the thoughts of the programmers. It enables anyone who receives it and who has a modicum of computer skills to circumvent plaintiffs’ access control system.
The Court, after weighing the relative importance of consumers’ fair use rights and content providers’ protection, ruled that DeCSS code was entitled only to a weak form of First Amendment protection, and found for the motion picture industry. Stripped of the veneer of a copyright debate, at its essence, this ruling reflects and reinscribes an old chauvinism that stresses the mechanic/organic, natural/synthetic, and biological/technological dichotomies. We suspect the issue is not ultimately one of functionality outweighing speech; it is simply that the Court cannot conceive of human–machine communication as speech.
The finding was appealed, and in an amicus curiae brief filed by a group of computer scientists, this defense was further elaborated, contesting the Court’s argument about the functionality of code:
[M]ost computer code does not function in a manner legally different from how a cake recipe or a music score “function,” but there can be no doubt that the latter are entitled to full First Amendment protection. . . . but in the absence of human intervention . . . [they] do nothing on their own except to inform the reader. So it is with most computer code: a human must give the command to interpret or compile the source code . . . a human must give the command to execute it . . . [and] a human must operate the device which makes the code functional. . . . The law is clear that the protection afforded to speech is not dependent on whether a device is needed to “execute” the speech. (Tyre 2001)
These arguments seek to establish that software (whether source or object code) is not only an avenue of human expression — “that subset . . . which computers can interpret and execute” (Tyre 2001) — but also one that should be subject to regulation only to the extent it is a form of speech. In their brief, the amici point out that code is used not only to communicate with computers but with computer scientists as well. That is, code is an integral part of “a complex system of understood meanings within specific communities” (Tyre 2001). Further, the expressive quality of source code, containing the “ideas, commands, objectives” (Tyre 2001) of the programmer, is carried into the executable code during a translation process. Thus, code has both communicative and expressive aspects. In the context of the cyborg world, the free speech protections for which the amici advocate generalize in the broadest sense to communication among its hybrid denizens.
Like any other form of speech, code can challenge power and ideology. The act of writing DeCSS was a fundamentally political one, contesting a particular unjust restriction on freedom. The Court’s ruling, implicitly recognizing the political implications of this act, upholds that restriction. While the DeCSS ruling is at one level only about intellectual property protection, the implications for the restrictions on expression extend much further. Without doubt, programmers’ expression has been restricted by this ruling’s chilling effect on the creation and application of cryptographic code; much more alarming is the prospect of restriction of speech and expressiveness in the cyborg world.
Tyre, James. 2001. Brief of Amici Curiae, Universal City Studios, Inc., et al. vs. Eric Corley, a/k/a Emmanuel Goldstein, 2600 Enterprises, Inc., Shawn C. Reimerdes, Roman Kazan. http://cryptome.org/mpaa-v-2600-bac.htm.
One thought on “Apple’s ‘Code Is Speech’ Argument, The DeCSS Case, And Free Software”
Would “code as speech” also support Google’s arguments that they can’t be regulated, ala this piece: http://www.volokh.com/wp-content/uploads/2012/05/SearchEngineFirstAmendment.pdf