[Cross-posted at Concurring Opinions.]
Brian Leiter notes this news story about a South Korean law which has just taken effect, requiring large web sites to obtain real names and the equivalent of Social Security numbers from everyone who posts content. He compares this approach to that taken in the US where, he says, “there exist only private remedies against Internet sociopaths and misogynistic freaks who hide behind anonymity. I suppose time will tell which is the better approach.”
Personally, I don’t need to wait for the passage of time to criticize the South Korean initiative (which has been under discussion there for some four years). Obviously, this law arises in a cultural context very different from our own, which I believe explains a good deal of the difference in approach. And it may not even be as different as it first appears. But there are principled reasons, distinct from cultural ones, to oppose a “show me your papers” internet.
First and foremost, it should be no surprise that China reportedly is looking at a similar model — as a technique to curb dissent, not just cyberbullying. The ability to remain anonymous protects unpopular speakers who might otherwise be unable to spread their ideas. In some countries, anonymous bloggers risk life and limb. Despite massive internet filtering by governments, blogging still provides dissidents a powerful tool. Even in more democratic countries, whistleblowers, political outsiders, and unhappy employees use anonymous blogging to avoid retribution. An outright ban on anonymity will curtail such often-useful speech.
Second, I am pretty confident a law like South Korea’s would be unconstitutional here. There is a clear right to anonymous speech under McIntyre v. Ohio Elections Commission. An influential district court case from 1997 (the web’s early days) relied on McIntyre to strike down as overbroad a Georgia law banning pseudonymous internet communications. (This is one of many cases the ACLU has brought in defense of internet anonymity.)
Finally, there are many technical methods to increase the chances of remaining anonymous online, such as using proxy servers or onion routing. (Ethan Zuckerman summarizes them in this guide aimed at dissidents and published by Reporters Without Borders). A demand for a real name and ID number makes these tactics somewhat harder to use, but certainly not impossible for a determined anonymity-seeker. The result: while most of us who follow the rules would lose the realistic capacity to stay anonymous online, the truly dedicated cyberbully, slanderer, harasser, or spammer can still remain in the shadows. Indeed, it’s not clear to me that a legal requirement to offer a (phony) name and ID number adds any real teeth against those with the knowledge and incentive to remain anonymous.
Without question, many in the US abuse online anonymity. They can cause real harm. So, are we just stuck with these abuses as the price we pay for free speech? Not entirely. There are more modest remedies available, and they are pretty much in place already. For one thing, individual sites are free to allow more or less anonymity as they see fit. Many ban anonymous comments. Many others, if they allow screen names or pseudonyms, still require registration. Lots of sites also guard against the prevalence of “Internet sociopaths and misogynistic freaks” by moderating their comments.
More fundamentally, in situations where a site does not choose to do any of the above, there is the John Doe subpoena. A complaint that states a reasonable claim against John Doe defendants for unlawful online speech usually will result in a judicial order that a web site or ISP provide available information about the speaker. Generally that information will include at least an IP address, and often the full name and postal address. Such data has been plenty for the recording industry and the government to identify defendants and proceed against them. I predict a similar outcome in the AutoAdmit case. (Indeed, Concurring Opinions’ own Frank Pasquale recently found his name in use as a pseudonym for a blogger running a gripe site critical of a health care company; in that case the court has allowed the plaintiff to “pierce the pseudonymous veil” and discover the blogger’s identity.) We may need some better procedures for discerning when such “piercing” on the basis of a complaint is appropriate, but I believe this doctrine will develop, just as it did in “corporate veil” cases. In that sense, we are evolving toward a model that allows psudonymity, but rarely true anonymity.
That may not be so different from South Korea’s approach in most cases. As I understand the new law there, large web sites must collect the information, but need not turn it over unless the victims sue. Except for a somewhat stronger mandate for collecting information, that is pretty similar to the John Doe subpoena. And since such subpoenas appear to have proven adequate in the US to identify speakers in most cases, this may not be as important in the real world as it first appears. That said, to the extent there are differences, leaving some breathing room for anonymity is a better fit with our free speech traditions.