In sum, Senator Exon and others supporting government restriction of sexually explicit images and discussion on computer networks have pointed to the "societal decay" that such communication must foster. Most research and public concern about potential harms has focussed on visual imagery, despite evidence that suggests conversational speech about a wide variety of sexual topics is much more common online. Nevertheless, because images of violence fused with sexuality may promote rape myths or more tolerance for sexual violence (see generally Linz & Malamuth, 1993), the extent to which online pictures and stories are violent warrants investigation.

Perhaps more important, however, are investigations of effects of online pornography that take into account audience differences among those within relevant newsgroup subcultures versus those observing from outside. Lea and Spears' (1991) work on social identity in computer-mediated communication provides a useful beginning for examining the conditions under which sexual deviance or violence may be encouraged among those who strongly identify with newsgroups devoted to these interests. However, differences between posters and lurkers as well as the capacity in newsgroups for conflicting group trends need to be considered, as these factors may greatly mediate the influence of extreme positions within the group.

There is also evidence that the virtual communities (Rheingold, 1993) created by newsgroups and bulletin boards have many prosocial benefits, such as social support, information exchange, and honest communication among people of similar interests, particularly those otherwise marginalized by society. Such communities may well also provide an outlet for sexual release, which may help alleviate or prevent the repression that is commonly associated with sexual predators (Meyer, 1995), although this has yet to be examined empirically. In addition, explicit sexual expression is likely to contribute to important debate on many sexual issues and to dissemination of accurate sexual information, although probably within the limits of predominantly male participation. Research efforts should be directed at systematically addressing the potential for sex education, even (perhaps especially) for adolescent participants.

Conclusions

Computer-mediated communication is not one technology but rather a number of technologies, including USENET newsgroups (available through the Internet), commercial bulletin board systems, real-time chat, and a worldwide Internet library of textual and graphic resources (available through the world wide web, ftp, gopher, and telnet searching systems). Furthermore, sexual expression online is not one form of communication, but many, including viewing or reading widely distributed pictures or stories, conversing in real-time, and participating in a social community through reading and posting lines of discussion.

Policy makers and researchers that attempt to treat the sexual content of all of these media as mere "cybersex" or "cyberporn" may ultimately encourage regulatory efforts based on inaccurate assumptions about the pervasiveness, intrusiveness, or social effects of very different content. For example, most political and public concern about sex-related materials has centered on pornographic visual imagery, despite evidence that it is only a small part of online sexual content. Unfortunately, where social science might have served to illuminate the public and Congress about online imagery, conflating findings and misrepresented data in Marty Rimm's study of download patterns of images from private "adult" BBSs (i.e., BBSs that require payment and proof of age before subscribing) and readership data for a small subset of Usenet newsgroups unfortunately has only served to cloud the debate.

Thus, I have argued that rigorous empirical research can address several assumptions about online sexual materials that have permeated public discourse and calls for limited First Amendment protection for online sexual communication. Specifically, I argued that three broad research areas emerge from concerns about "cyberporn" that reflect prominent legal and social arguments for restricting speech in other media. Using Senator Exon's language, the fear that computer networks are becoming a "red light district" reflects an empirical question about the pervasiveness and unseemliness of computer pornography; the idea that pornography is "only a few clicks away from any child with a computer" presents opportunity to examine the intrusiveness of computer pornography and its potential accessibility to children and adolescents; and finally, Exon's comment that "society is going astray" reflects a need to address the potential influence of computer content in a socially harmful versus beneficial direction.

Limited systematic research has examined the pervasiveness of different forms of online sexuality. There is evidence, albeit mostly anecdotal, that sex-related material of any kind (pictures, stories, or discussion) comprises only a small portion of the total Internet activity, and is thus less pervasive than much popular press and Congressional panic would suggest. However, it is likely that extreme forms, such as bestiality, may be more available online than in the local adult book store (certainly in discussion form such topics are readily available online). More important research on the complex nature of sexual discussion online, such as investigations of how sexual encounters are negotiated in chat rooms, how "burn in Hell" posts are treated in newsgroups, and how personal and commercial sexual stories or images are traded and critiqued, could bring attention to the richness of online communication and extend regulatory debate beyond mere outcries about offensive pictures.

Drawing on judicial precedent and legal definitions for electronic media, I have also argued that computer-mediated communication is not "intrusive" in the broadcast sense of confronting people, particularly children, unaware in their homes. In fact, online communication may be the least intrusive of all electronic media (P. Miller, 1993), especially where children are concerned (Corn-Revere, 1995b), because kids must have both skills and motivation necessary to access sexual content, and they are not likely to come across such material accidentally:

Put plainly, cyberspace is a voluntary destination--in reality, many destinations. You don't just get "onto the net"; you have to go someplace in particular. That means that people can choose where to go and what to see (Dyson, 1995, p. 26).

Research in this area would probably be best suited to addressing concerns that might help parents anticipate and deal with motivated, computer-wise offspring, such as what are the developmental skills and motivations that are likely to mediate very young versus older children and adolescent Internet participation. Investigation of strategies for coping with offensive online behavior would also help parents protect their children from potential abuse. Finally, for the truly panicked parent, teacher, or politician, a number of technological parental controls, while not perfect, provide alternatives to government censorship that can help adults continue the American tradition of altogether circumventing having to confront children on issues of sexuality.

On the final issue of the harms or benefits of online sexual materials, there is some theoretical support (Durkin & C. Bryant, 1995) for the argument that sexually deviant preferences, such as pedophilia, may be reinforced and encouraged in newsgroups in which they strongly identify, because of the potential for influence toward group norms among de-individuated persons (Lea & Spears, 1991). However, because newsgroups, unlike the real-time decision making groups of de-individuation experiments, have the capacity for simultaneous, conflicting group trends and social differences between lurkers and posters, research needs to further examine this potential for fostering potentially criminal deviant behavior.

On the other hand, whether or not newsgroups do encourage "deviant" behavior only becomes a concern inasmuch as a given sexual preference or practice is considered harmful or immoral. I have argued that, with the possible exceptions of pedophilia and sexual violence, many forms of online "deviance" (e.g., sadism and masochism) are subject to tolerance or rejection based on competing moral value judgments that are not easily reconciled.

Finally, I have argued that there are prosocial values of explicit sexual expression online. Even for deviant practices that are almost universally reviled, unrestricted communication among their advocates may provide invaluable knowledge about their rationale. Indeed, "the power to protect a child may come from understanding what makes a pedophile tick" (Tamosaitis, 1995, p. 140). In addition, sex-related newsgroups provide the potential for more mainstream sexual information exchange, development of personal relationships, and debate on many important sexual issues. Research efforts should be directed at systematically addressing the potential for accurate sex education online, for adults, adolescents, and possibly even children.

In sum, the panic in public and political spheres about online sex appears largely unwarranted. Regarding Senator Exon's specific concerns, online pornographic images and discussion do not appear to be particularly pervasive, are certainly not intrusive (as they must be actively sought by both adult and child audiences), and may be harmful in some ways but beneficial in others. Given these arguments, government censorship of sexual materials on the Internet or other computer networks cannot be adequately justified. Nevertheless, some important dimensions of each of the three above concerns warrant further research attention that might help alleviate fears and help parents and politicians better address concerns about online sexuality.

I should note that government restrictions on sexual communication on the Internet may not inherently limit all of the forms of sex online that I have discussed, such as educational or controversial talk about sex. Explicit talk without a visual image, for example, is not likely to meet current legal definitions of obscenity (according to at least one district attorney [see Elmer-Dewitt, 1994], words alone have virtually no chance of being successfully prosecuted). On the other hand, if, as the Exon amendment proposes, "indecent" material is prohibited, any sexually explicit language in a discussion group, educational or not, may have criminal ramifications, given the legal standards for indecent language that have been used in the broadcast medium (see e.g., Pacifica, Action for Children's Television III).

Therefore, with such ramifications on the line, it is crucial that we take seriously the concerns outlined in this paper, but also that we approach them with sound empirical research and legal reasoning over sensationalism. As Hoffman & Novak (1995) put it, "the critically important national debate over First Amendment rights and restrictions on the Internet and other emerging media require facts and informed opinion, not hysteria" (online).

Notes

1. According to Miller, states could deem "obscene" works which "the average person, applying contemporary community standards" would find "taken as a whole, appeal to the prurient interest in sex, which portray sexual conduct in a patently offensive way, and which taken as a whole, do not have serious literary, artistic, political, or scientific value" (Miller, 413 U. S. at 24 - 25, 1973).
2. Following the language of the Supreme Court in FCC v. Pacifica Foundation (1978), the FCC currently defines "indecency," at least for broadcasting, as all "language or material that, in context, depicts or describes, in terms patently offensive as measured by contemporary community standards for the broadcast medium, sexual or excretory activities or organs" (1993 Enforcement Rules).
3. In the Thomas case, federal statutes prohibiting the transmission of obscenity across state lines were applied to the couple's commercial transmissions over the Internet. Such prosecution of even obscene materials on-line has been legally challenged, as the "community standards" requirement of Miller becomes problematic when applied to computer networks (the Thomases uploaded material in California, but were charged based on downloads in Tennessee). See Sergent (1994) for a discussion of this problem.
4. I should note that, although his position on online pornography has not changed, Senator Grassley has since distanced himself somewhat from the actual Rimm study (Marty Rimm was removed from the list of witnesses scheduled to testify before Congress on this issue; see Lewis, 1995d).
5. Corn-Revere (1995b) points out that since Sable, lower courts have upheld FCC rules that require customers to request dial-a-porn services in writing before a common carrier may provide access, in order to limit children's access (Information Providers' Coalition for Defense of the First Amendment v. FCC, 1991).
6. In an attempt to help teachers deal with online profanity in the classroom, one Internet server recently posted on the web an example of how certain "cyber-terrorists" might be effectively handled (Rogers, 1995). In one instance, offensive e-mail had been sent "anonymously" to children in a classroom Internet project who had sought online responses to their questions. Through some interesting network backtracking, the server was able to track down the university student whose "prank" had upset the children, report the terrorist's behavior to his own server, and was able, in addition to teaching the prankster a lesson in "netiquette," to extract a detailed apology that was sent to all of the children in the class.
7. The prohibitions in the Exon amendment itself are not limited to photographic representations. Content in personal stories, newsgroup posts, and interactive conversations would be actionable if considered "obscene, lewd, lascivious, or indecent."
8. Not only issues about sexuality have raised such controversy. Gun-oriented and militia newsgroups, for example, have also come under fire for posting, among other things, information about how to make bombs (see Gleick, 1995).
9. Some authors and journalists (e.g., Rheingold, 1993; Webb, 1995) argue that public concern about cyberporn focusses attention away from the vast majority of Internet traffic that does not concern deviant sexual practices (or, for that matter, satanism, white-separatism, and other anti-social fascinations). For example, although one newsgroup does, as many people fear, contain explicit images and discussion of bestiality, there are dozens more animal-related newsgroups where one can find information on less prurient topics such as dog behavior ("rec.pets.dogs.behavior), aquarium health ("rec.aquaria"), and the eating habits of cats ("rec.pets.cats"). The Internet is also rife with resources geared towards self-improvement (e.g., dieting, speed reading), recovery (e.g., from alcohol or drug addiction), and just making friends (e.g., a romantic guide for "cyber-nerds" or companionship and independence for senior citizens). Researchers are just beginning to address the use and effectiveness of newsgroup participation in these prosocial areas (e.g., Baym, 1995; Czaja, Guerrier, Nair, & Landauer, 1993; James, Wotring, & Forrest, 1995; Ogan, 1993).

continues