Sign In

Communications of the ACM

Communications of the ACM

Viewpoint: The COPA Battle and the Future of Free Speech

Last October, a new battlefront opened in the U.S. war between those who see a risk to childhood's innocence and those who see a risk to free speech. The omnibus budget package includes the Child Online Protection Act (COPA), a successor to the Communications Decency Act (CDA). The very morning after the act was signed, the American Civil Liberties Union (ACLU), last year's victor in the battle over the CDA, filed suit, asking for an injunction against the COPA's enforcement and for its main provision to be declared unconstitutional.

This isn't just one more battle, however. This time the very future of free speech jurisprudence may be at stake. I make this melodramatic statement based on four predictions about the course of the battle. First, the courts provide no speedy protection to speakers this time. Second, there will accordingly be time for Internet speakers to develop their own coping mechanisms: to "route around" the COPA. Third, when the case eventually reaches the Supreme Court (as I am sure it will), it will force the Court to squarely face a watershed choice that it has thus far been avoiding. Fourth, regardless of the choice the Court makes, the option of routing around the COPA will prove to be merely a short-term fix. However, if the Court goes down one of the two paths open to it, the cost for eliminating routing around could be a terrible one for the First Amendment.

Back to Top

No Speedy Resolution

Unlike the CDA, the COPA has no special provision for accelerated judicial review. Therefore, the ACLU's challenge will proceed at the court system's usual stately pace. The ultimate resolution by the Supreme Court will not come soon. As I will explain subsequently, that resolution will hinge on whether the Court chooses to stick with the "least restrictive means" doctrine or switch to an alternative "substantial burden" doctrine. Under existing doctrine, the COPA is likely to be upheld (perhaps with some minor tailoring). I will justify that claim later; for now, the point is though the district court has already granted temporary relief, this does little to reduce the risk that the law may eventually take effect.

Back to Top

Routing Around the COPA

If there is little hope for a quick reduction in legal uncertainty, this raises the question of how Net users will respond. John Gilmore is famous for his remark that "the Net treats censorship as damage and routes around it." What he refers to is at its heart a technical fact: the Internet contains redundant paths between points, so that if one path is blocked, data can flow through an alternate path. However, Gilmore meant more: not just the automated Net routers, but also the humans, will find ways around censorship. The COPA in particular makes this trivially easy. Its authors bent over backward to respond to the Court's criticism of the CDA as overly broad. As a consequence, Congress made the COPA so narrow as to beg for routing around.

For example, the CDA applied not just to the Web but to all forms of Internet use, even those where erecting a barrier to selected users would have been infeasible. In response to the Court's criticism of this defect, Congress limited the COPA to the "hypertext transfer protocol or any successor protocol." (The hypertext transfer protocol, more commonly known as HTTP, is the main mechanism of the Web.) However, it would be easy for commercial sites to move their freely available content that is harmful to minors (such as teaser ads on pay-for-porn sites) to the older file transfer protocol (FTP). This can be done without any user-visible difference, except some some slight added delays. Since FTP predates HTTP, it cannot be considered a successor protocol.

Back to Top

A Watershed Court Case

The main question raised by the COPA concerns its impact on adult-to-adult communication. There is no question that the COPA impairs commercial speakers' ability to cheaply, easily, and broadly communicate material to adults that is constitutionally protected as to the adults (non-obscene), though harmful to minors. The question before the Court will be whether this burden on protected speech is justified by the government's interest in shielding children, or at least in enhancing parental control.

In deciding this question, the Court will be facing more squarely than ever before the doctrinal question posed in recent articles by UCLA law professor Eugene Volokh: Should even substantial burdens on protected speech be allowed if they are necessary to achieve a compelling government interest? The Court orthodoxy is that even substantial burdens are permissible if they are the "least restrictive means" to achieve the government's interest. In deciding the fate of the CDA, the Court reiterated this standard, but claimed that Congress had not properly examined less restrictive means to see if they might be equally effective. No such easy out will be available to the Court in considering the COPA. This time, Congress has quite explicitly considered alternatives by writing them into the law's findings and by taking pains to narrowly tailor the law as the Court itself suggested. Thus the Court will either need to stick with its established doctrine and conclude that the COPA is constitutional, or make a momentous change of doctrinal course. (As Volokh has pointed out, there are some precedents for a doctrine that rejects substantial burdens even when no less restrictive alternative will suffice. So I don't mean to say the doctrinal change of course would be into wholly uncharted waters, only that it would be a radical shift in how the Court analyzes cases such as this one.)

Either option the Court selects will have substantial impact. Even sticking with the "least restrictive means" test under such trying circumstances would cement that test far more firmly in place, and likely lead to a substantial change in its practical application. It will no longer be possible to read this standard as one that generally protects free speech. If instead a new "substantial burdens are unconstitutional" doctrine emerges, that will also be of historic significance. It is in this sense that I see the case as a watershed: either way, the future course of the law is dramatically affected.

Should even substantial burdens on protected speech be allowed if they are necessary to achieve a compelling government interest?

If the Court changes doctrine and strikes down the COPA, routing around the law will become irrelevant. (Moreover, new laws for the Net to route around will be rendered less likely.) However, even if the Court upholds the COPA as a "least restrictive means," the routing around response is doomed in the long run. This is because such a Court opinion would give Congress a green light to pass whatever additional restrictions prove necessary to its ends. Routing around COPA's narrow restrictions will prove that broader restrictions on speech are necessary.

For example, it would now become permissible to regulate all forms of Internet use, since the less restrictive means of regulating only HTTP would have been shown to not be equally effective. Never mind that a blanket restriction on all forms of Internet use would be considerably more burdensome; if the Court stays its course, it will sanction substantial burdens in situations where lesser burdens prove ineffective.

Of course, the Net probably would not give up the fight easily, and so would come up with another creative response. As a result, there could be a few more rounds of ratcheting up the restrictiveness, with a Net that routes around being progressively countered by a Congress authorized by the Court to in effect "do whatever it takes." But ultimately this path leads to a police state, not to techno-anarchy. From this, it should be obvious which option I hope the Court takes when this watershed case reaches it.

Back to Top


Max Hailperin ( an an associate professor at Gustavus Adolphus College, St. Peter, Minn.

©1999 ACM  0002-0782/99/0100  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 1999 ACM, Inc.


No entries found