IP: Cato Institute paper on Net-speech regulation, by

[Dave Farber <farber () cis upenn edu>]

X-Received: (from majordomo@localhost) by vorlon.mit.edu (8.8.2/8.8.2) id
WAA26063 for fight-censorship-announce-outgoing; Sun, 17 Nov 1996 22:39:14
-0500
Date: Sun, 17 Nov 1996 19:38:42 -0800 (PST)
From: Declan McCullagh <declan () well com>




[Among other things, Solveig's paper talks about the "harmful to minors" 
standard in a future CDA that we've discussed before and I wrote about in
June: http://www.hotwired.com/netizen/96/24/declan4a.html --Declan]


---------------


http://www.cato.org/pubs/pas/pa-262es.html


Cato Policy Analysis No. 262
November 4, 1996


                   BEYOND THE COMMUNICATIONS DECENCY ACT:
                   CONSTITUTIONAL LESSONS OF THE INTERNET 
                                      
                   by Solveig Bernstein (sberns () cato org)
                                      
     Solveig Bernstein is assistant director of telecommunications and
                 technology studies at the Cato Institute.
                                      
     _________________________________________________________________
                                      
                             Executive Summary
                                      
   On February 8, 1996, the Communications Decency Act was enacted into
   law. The law criminalizes the use of any computer network to display
   "indecent" material, unless the content provider uses an "effective"
   method to restrict access to that material to anyone under the age of
   18. But there is no affordable, effective way for nonprofit or
   low-profit speakers to restrict children's access to such a broad,
   ill-defined category of material. Thus, the statute effectively bans
   much speech from the Internet and other networks. The Internet
   promised the ordinary citizen a low-cost method of reaching an
   audience beyond immediate family, friends, and neighbors. Legislation
   like the CDA betrays that hope and is clearly unconstitutional.
   
   No regulation of computer network indecency, however carefully
   tailored, should pass constitutional scrutiny. First, no legislator
   has been able to define indecency coherently. Such regulation is
   inherently unfair, especially as applied to spontaneous, casual speech
   of the sort that the Internet facilitates between unsophisticated and
   noncommercial speakers. Second, government cannot legitimately claim
   that it has any interest in content control, when civil society has
   solved the perceived problem on its own. Here, private sector
   solutions include both software filters that parents can use to screen
   out offensive material and Internet service providers who provide
   access only to child-safe materials.
   
   
[...]
   


         Why Indecency on Computer Networks Should Not Be Censored 
                                      
   One can be certain, however, that the censors will not give up. If the
   CDA ultimately is declared unconstitutional, the censors will try to
   craft new legislation along similar lines.
   
   Possible Alternatives to the CDA 
   
   Legislation somewhat less broad than the CDA would cover only material
   that is "harmful to minors." This option would essentially still ban
   much amateur speech because of the technical and economic difficulties
   of restricting access. More sophisticated plans have also been
   suggested.
   
   Use of site rating labels could be added to the available defenses
   (which would in effect make labeling of sexually explicit sites
   mandatory) to ease this problem somewhat. Labels can be used to rate
   newsgroups, Web sites, and content posted on online networks. Eugene
   Volokh of the University of California Los Angeles Law School has
   suggested that governments could require all content providers to rate
   their own sites. [77] Parents could then buy software filters that
   would reject adult-rated content.
   
   In defending the CDA, the Department of Justice crafted a similar
   argument. Under one proposal supported by the Department of Justice,
   all "indecent" materials would be tagged "L18," for "not less than
   18." At the first CDA hearing in Philadelphia, the Department of
   Justice explained that computer network users would be registered as
   "adults" or "minors," and that information would be encoded in their
   online personas. Network servers (the computers on which content is
   stored) would be customized to deny minors access to Web sites tagged
   "L18." [78] The proposal would require all Internet service providers
   to reprogram a substantial number of their servers. The CDA does not
   require Internet service providers to undertake any such project;
   generally, only those that control content are liable under the law.
   Thus, the argument that the L18/server scheme could alleviate the
   burdens of the CDA on speakers was essentially absurd, as Judge
   Sloviter noted. [79] Additionally, courts have recognized that
   advanced speaker registration requirements stifle the spontaneity of
   free expression. [80] And advanced registration would threaten the
   existence of electronic forums operated for the benefit of those most
   anxious to protect their identity, such as victims of sexual abuse.
   [81] Finally, the suggestion that servers be restructured is eerily
   reminiscent of the Singapore government's insistence that Internet
   communications be routed through "proxy servers" to facilitate
   intensive political censorship.
   
   By the second CDA hearing, the government had apparently abandoned the
   server/registration approach to tagging, and explained that the tags
   could work with filtering software controlled by the end user. But
   that would not satisfy the CDA's effectiveness requirement, as Judge
   Cabranes noted, because many parents do not use filtering software;
   the Department of Justice's assertions at the hearing that it would
   not prosecute labeled sites were not binding on any prosecutor. [82]
   As a defense of the CDA, both incarnations of the L18 scheme failed.
   But they might suggest a direction for future legislative efforts.
   
   Any form of mandatory labeling, however, is objectionable for several
   reasons. First, it is compelled speech, which should not be
   constitutionally permissible. [83] It would place an extraordinary
   burden on entities with large collections of works, such as libraries.
   [84] It would be oppressive to expect such labels to be applied to
   casual or intimate speech, such as statements in chat rooms, private
   e-mail, or individual newsgroup or bulletin board postings. For
   spontaneous computer speech, mandatory tagging would be the equivalent
   of requiring the labeling of conversations around a backyard barbecue.
   
   Second, mandatory labeling as unsophisticated as the L18 scheme
   proposed by the Department of Justice would prevent older children
   from accessing information about reproduction, art, and other topics,
   or from contributing to discussions of those topics. Minors, too, have
   free speech rights. Sixteen-year-olds should not be restricted to
   viewing what is fit for six-year-olds.
   
   Third, because there is so much content on computer networks, the only
   practically feasible kind of universal labeling scheme would require
   content providers to rate their own material. A substantial number of
   amateur or casual speakers would, out of an excess of caution or as an
   act of civil disobedience, deliberately give their sites a more or
   less restrictive label than the law requires. Libraries might be
   forced to slap an "adult" label on their entire collection, because
   they could not afford to rate all their content. There are so many
   thousands of communications traveling over computer networks every day
   that only a very small proportion of the labels would be checked by
   third parties. Thus, ironically, a mandatory labeling regime is more
   likely than voluntary labeling to be substantially inaccurate and
   unhelpful to parents. Under the market-driven voluntary systems that
   will work with the new rating standards known as PICS (Platform for
   Internet Content Selection), unrated sites can be blocked
   automatically by filter software; a greater proportion of those fewer
   sites that are rated can be checked by private ratings groups. Only
   voluntary rating would be consistently undertaken with care.
   
   The Fallacy Motivating the Search for CDA Alternatives 
   
   Proposing any legislative alternative to the CDA makes a fundamental
   error: such proposals assume that government has constitutional
   authority to regulate nonobscene sexually explicit computer network
   speech. Judge Dalzell identified this as the central issue at the
   hearings concerning the constitutionality of the CDA, stating that:
   
     from the Supreme Court's many decisions regulating different media
     differently, I conclude that we cannot simply assume that the
     Government has the power to regulate protected speech over the
     Internet....Rather, we must decide the validity of the underlying
     assumption as well, to wit, whether the Government has the power to
     regulate protected speech at all. [85]
     
   The analysis below shows that this assumption is not valid. Even if we
   assume that the precedents that allow the government to regulate
   nonobscene sexual speech on other media are correct, these precedents
   do not supply any convincing rationale for regulation of computer
   networks. Communication over computer networks does not raise entirely
   new constitutional issues. But it raises two particularly important
   issues in such a way that they cannot be avoided.
   
   First, computer networks empower millions of ordinary citizens to
   become speakers. As censorship laws are enforced, the court's failure
   to coherently define categories of forbidden talk about sex will look
   more and more obviously unjust and arbitrary.
   
   Second, the power of the private sector to offer alternatives to
   censorship erodes arguments that government has any legitimate
   interest in this problem. Without a constitutionally cognizable
   interest in imposing the regulation, government cannot act.
   
   These are both sound reasons to believe that indecency (or its cousin,
   material that is "harmful to minors") on computer networks cannot
   constitutionally be regulated at all. First Amendment jurisprudence
   must evolve to address these issues or become divorced from the
   reality of the marketplace of ideas.
   
   Defining Forbidden Speech 
   
   Unwilling to rule that government simply may not censor any speech,
   the Supreme Court has struggled to distinguish between speech about
   sex that may be censored, and speech that may not be. Early on, the
   Court decided that obscene speech was not entitled to First Amendment
   protection. But what was obscene? The Court's attempts to define this
   category coherently have important implications for regulation of
   indecency or material that is "harmful to minors" on computer
   networks.
   
   This is not because obscenity and indecency are the same thing.
   Whatever is obscene is almost certainly indecent; a wide range of
   material that is indecent is not obscene. But our judgments about what
   is obscene and what is indecent are closely tied to subjective moral
   judgments. If the Court cannot define one category coherently, it is
   unlikely to make much headway with the other. Nor is it likely to make
   headway with the in-between category of "harmful to minors."
   
   For years, the Supreme Court struggled to create a national definition
   of obscenity. It failed. At bottom, the question of what is "obscene"
   is a matter of taste. No power in the world can convert a subjective
   question into an objective one, even by abstracting from the myriad
   subjective tastes of members of a national community. Under the
   "national" approach, ultimately, a work was obscene if it offended
   enough Justices of the Supreme Court. This was evidenced by hilarious
   yet deeply troubling statements such as that of Justice Potter
   Stewart, who, in attempting to define hard-core pornography declared,
   "I know it when I see it." [86] In 1963, Chief Justice Earl Warren
   stated, "I believe there is no provable 'national standard.'" [87]
   Still later, in abandoning the national standard, the Court explained:
   
     it is neither realistic nor constitutionally sound to read the
     First Amendment as requiring that the people of Maine or
     Mississippi accept public depiction of conduct found tolerable in
     Las Vegas, or New York City. People in different states vary in
     their tastes and attitudes, and this diversity is not to be
     strangled by the absolutism of imposed uniformity. [88]
     
   Similarly, the FCC has failed to craft a coherent national standard of
   broadcast indecency. According to the FCC, broadcast indecency is to
   be judged according to the tastes of the "average broadcast viewer."
   But who is this "average" viewer? In a country with local standards as
   diverse as those of San Francisco or Iowa, there can be no such
   animal. The national standard boils down to what offends the FCC.
   
   Insofar as interpreters of the CDA are directed by the legislative
   history to craft a national indecency standard, they will be no more
   successful than the FCC. The early print media precedents are no more
   helpful. It is possible for any court to string together words in an
   important sounding way, crafting phrases such as "prurient interest,"
   or a mythical national consensus, and claim to have created a uniform
   definition of indecency. What it will have done, in effect, is to
   impose its tastes on the rest of the nation.
   
   Nor can the Court resolve the problem by referring to a hypothetical
   "average" computer network user. A First Amendment that protected only
   "average" speech would provide little or no protection at all to
   unpopular minorities. Part of the reason that computer networks are
   special is that they empower an extraordinary range of speakers from
   diverse communities. The tastes of the "average" user are thus not
   only hard to identify, but should be of no relevance.
   
   If there cannot be a national standard for forbidden speech about sex
   on computer networks, can there be local standards? The Supreme Court
   allowed states to adopt community standards to alleviate the
   embarrassment of its failure to craft a national obscenity standard
   for the print media. [89] The question of what was obscene was largely
   left to local juries. [90]
   
   But some members of the Court long resisted adopting a local community
   standard, for good reason. Justice William Brennan argued that the
   local community standard could not serve as a constitutional standard:
   
     We do not see how any "local" definition of the "community" could
     properly be employed in delineating the area of expression that is
     protected by the Federal Constitution....It would be a hardy person
     who would sell a book or exhibit a film anywhere in the land after
     this Court had sustained the judgment of one "community" holding it
     to be outside the constitutional protection. [91]
     
   His fear was that an adverse judgment in a few restrictive local
   communities would chill the national distribution of speech.
   
   The Supreme Court has since flatly refused to recognize the
   constitutional dimensions of this problem. In one case, the Court
   considered a dial-a-porn operator's argument that Congress could not
   force it to tailor its messages to the least restrictive community,
   because such a requirement in effect created a national standard of
   obscenity. The Court explained, "While Sable [the operator] may be
   forced to incur some costs in developing and implementing a system for
   screening the locale of incoming calls, there is no constitutional
   impediment to enacting a law which may impose such costs on a medium
   electing to provide these messages." [92]
   
   Computer networks will raise this issue again, this time with a
   vengeance. The impact of the law will be felt, not by the narrow,
   unpopular community of professional pornographers, but by ordinary
   citizens able to reach a wide audience for the first time. The local
   standard will not suffice in any country that takes free speech
   seriously.
   
   If the national standard is inherently incoherent, and the local
   standard inherently unfair, what is the Court to do? The answer is
   that the Court should admit that government, especially the federal
   government, has no place regulating the display of sexual imagery in
   cyberspace, especially if it is neither obscene nor categorized as
   child pornography. If it cannot be done consistent with the
   Constitution, it should not be done.
   
   But will this mean that the United States' children are to be exposed
   to a never-ending stream of sexually explicit images? It will not mean
   that at all. And the dispute surrounding the constitutionality of the
   CDA is the perfect opportunity for the Court to make this clear.
   
   Market Alternatives Erode the Government Interest 
   
   The Supreme Court's indecency jurisprudence requires that a statute
   choose the least restrictive means to serve a compelling state
   interest. The Court's accumulated indecency cases, however, do not
   make clear what that interest is. It is either government's interest
   in helping parents protect their children, or an independent interest
   of government in protecting the children themselves. [93] The analysis
   below shows that the latter interest cannot be viewed as
   constitutionally compelling. And, where computer networks are
   concerned, parents are capable of taking care of their own children.
   With computer networks, government's interest falls away.
   
   An Interest in Helping Parents. The Supreme Court has described the
   government's interest in regulating indecency as an interest in
   helping parents supervise their children--not in protecting children
   from indecency when their parents believe the materials in question
   would do their children no harm:
   
     Constitutional interpretation has consistently recognized that the
     parents' claim to authority in their own household to direct the
     rearing of their children is basic in the structure of our
     society.... The legislature could properly conclude that parents
     and others, teachers for example, who have this primary
     responsibility for children's well-being are entitled to the
     support of laws designed to aid discharge of that
     responsibility...the prohibition against sales to minors does not
     bar parents who so desire from purchasing the magazines for their
     children [emphasis added]. [94]
     
   It is not rational to argue, however, that government can have a
   compelling interest in helping concerned parents when concerned
   parents do not need help. Government should not be able to argue that
   it has a compelling solution to a problem that has effective private
   solutions.
   
   Computer networks offer an excellent private solution to parents who
   want to protect their children from indecency, but who do not want to
   deny access to online services altogether. As with any media, parents
   can control their child's access to computerized indecency by
   exercising a little sense. Some parents, for example, do not allow
   their children access to online services in the privacy of their own
   rooms; access is available only by means of a computer in the family
   room, where anyone walking by can see what is on the screen.


[...]
   
   An Independent Interest in Protecting Children? Perhaps government
   could claim a compelling interest in protecting unsupervised children,
   children whose parents do not purchase or use filtering software?
   Justice Stephen Breyer, writing for the plurality in Denver Area
   Educational Telecommunications Consortium v. FCC, a case involving the
   constitutionality of restrictions on the transmission of indecent
   material over cable television, restates that protection of children
   is a compelling or at least important interest, and suggests, without
   further analysis, that such interest allows the federal government to
   intervene to protect children of "inattentive" parents. [125]
   
   There are substantial reasons to believe that protecting children from
   a danger that the childrens' parents do not recognize as particularly
   grave should not amount to a compelling interest. As pointed out
   above, filtering software is affordable to anyone who can afford a
   computer system. Nonsupervising parents have implicitly decided that
   exposure to material of a sexual nature probably will not harm their
   children enough to bother with. If the parents do not find the
   interest sufficiently compelling to take action, there is no reason to
   think that government should.
   
   Indeed, there may be parents who believe that their children should be
   exposed to materials that might be considered indecent, including
   information about disease prevention, birth control, reproduction,
   works of literature and art, and so on. Government's claim of an
   independent interest in restricting indecency contradicts government's
   claim of an interest in helping parents control their children's
   education. [126]
   
   If government did have an independent compelling interest in keeping
   children from viewing all sexually explicit or vulgar material, it
   could pass a law that parents must lock all the indecent materials in
   their home (Playboy, romance novels, Lady Chatterley's Lover) in
   special safes to ensure that their children never access it. Or that
   parents must use software filters to prevent teenagers from using the
   Internet to read about sex.
   
   Imagine police searching through private residences to enforce this
   law. The reaction would be public outrage. In short, when it comes
   down to it, there is nothing compelling about government's alleged
   interest in protecting children from indecency. In this context, we
   recognize that parents have the right and responsibility to make
   decisions about such matters for themselves.
   
   So why do we pretend that the interest becomes compelling when the
   burden of complying with the law is placed on someone other than the
   parents? We pretend it because we place the burden of complying with
   the law on unpopular speakers--pornographers, purveyors of smut.
   
   The application of indecency laws to computer networks will throw the
   issue into stark relief. First, under the CDA, it is possible that
   parents and teachers could be prosecuted for allowing children in
   their charge to use computers to access material that the parents
   believe the child is mature enough to handle. Second, the easy
   availability of private solutions for parents who are concerned about
   indecency makes it obvious that the CDA is nothing but a convenience
   for parents who will not take the trouble to supervise their
   children--not a compelling problem that the government must step in to
   solve.
   
   Private solutions might not always be available to solve "indecency
   problems." On public property, for example, which everyone must access
   from time to time, one faces more difficult questions. But computer
   networks are not public parks. They are sophisticated user-controlled
   private spaces. And private solutions clearly should be part of the
   constitutional analysis.
   
[...]






--------------------------------------------------------------------------
To subscribe to the moderated fight-censorship-announce list, send
"subscribe fight-censorship-announce" to majordomo () vorlon mit edu.
More information is at http://www.eff.org/~declan/fc/