Judge Alex Kozinski Website

Finding Justice in the Internet Dimension

Seattle University Law Review, Spring 1997, 20 Seattle Univ. L. R. 619

An Interview Conducted by Seattle University School of Law Dean James E. Bond

Honorable Alex Kozinski, Circuit Judge, U.S. Court of Appeals, Ninth Judicial Circuit

Dean James Bond (Bond): A three-judge Philadelphia district court recently issued its decision in ACLU v. Reno, n1 invalidating the Internet "indecency" provisions of the Communications Decency Act. Judge Dalzell writes, "The Internet is a far more speech-enhancing medium than print, the village green, or the mails. The Internet may fairly be regarded as a never-ending worldwide conversation. As the most participatory form of mass speech yet developed, the Internet deserves the highest protection from governmental intrusion. Any content-based regulation of the Internet, no matter how benign the purpose, could burn the global village to roast the pig." Judge Kozinski, do you agree or disagree with Judge Dalzell's view of the Internet and his expansive reading of the First Amendment? [*620]

Judge Alex Kozinski (Kozinski): I was amused by the "roast pig" reference in Judge Dalzell's opinion. If you take the phrase "information superhighway" and create an anagram by rearranging its letters, you can spell out "enormous hairy pig with fan." You can also spell out "a rough whimper of insanity" and "I swamp horrify huge nation." Try it out.

I share the reaction of Richard Dooling, writing for the New York Times, n2 who described the decision as a two hundred-plus page opinion from the third branch of government, telling us that an obviously unconstitutional statute is obviously unconstitutional. Generally, it seems to me the result was correct, and I share many of Judge Dalzell's sentiments.

Unfortunately, when people talk about the Internet, they tend to talk in terms of analogies or metaphors - the "information superhighway," the "global village," the "continuing conversation." I don't think that these do much to advance the analysis. Indeed, they tend to hide what the real issues are.

Consider the idea of the Internet as the most awesome means of communication the world has ever known. Of course, we know that television is that, and not the Internet. It's hard to know how many people are on-line these days, but a recent report I've seen estimates around 30 million people worldwide, out of 5 billion. In the United States, estimates zoom all the way up to 20 million people. But it is to some extent an overstatement to claim that the wave of the future is already here.

People with means and an interest in engaging in this kind of communication have been willing to spend a lot of disposable time and dollars using their computers essentially for entertainment. There are other uses, legitimate and illegitimate, of the Internet, but most of you here use it for discretionary, leisure activities. If we take into account this reality, any serious expansion, if there is to be one, is still in the future.

What we're seeing now is similar to the early days of radio. If you were into radio communication, were technically inclined, and could afford it, you got a transmitter and did a little broadcasting. To regulate the airways, Congress came up with the Radio Act. n3 At that time, too, there were cries of censorship. As much as I'm of a libertarian bent, I recognize that there are few human interactions [*621] outside the privacy of the home that will not involve some type of legal gatekeeping, some rules of the road, some governmental involvement either to encourage or to prohibit certain transactions.

I am on the "Net" frequently. I often cruise, looking for islands of humor, to which I then alert some friends. Lately, I've been receiving long and contentious messages, phrased in extreme terms, hysterical that the Communications Decency Act will subjugate freedom on the Internet. Once I found time to read these interminable messages, I started asking myself and others some questions.

In fact, one of the messages was sent to me by a law student. I decided this was a good time to teach him a lesson about government and constitutional law. So I wrote back and said, I've read the letter you sent me and 120 other people, and that you asked us to forward to Congress, objecting to the CDA. There are some pretty extreme claims in there. Do you really mean to suggest that communications on the Internet can't be controlled at all? For example, how about blackmail? How about espionage? How about child pornography? Are you in favor of downloading "snuff flicks"? Is the Internet so different from all other forms of communication that it stands beyond all format control? I thought that I did a respectable job, trying to arouse this particular young lawyer's interest, and hoping to challenge him. Well, he wrote back to me a day or so later and said, "I have your very interesting letter, Judge Kozinski. I will certainly study it after my exams, and I will get back to you." As you can guess, he never did.

To bring this answer to a close, I generally think that the Internet community - just like all other speech communities - ought to be afforded First Amendment protections. I don't see any reason why Internet speech should be treated any less favorably than other kinds of speech. But the vastly overblown claim that the communications medium somehow deserves to be put outside normal legal constraints - because it's so global, or because it's so different - is self-defeating. It substitutes generalities and sentiments for real thinking. The kind of analysis we've seen at this conference - the kind of debate we've had here - is very useful, because we're talking about the specifics of what legal constraints should be allowed. Not whether there should be regulation, for clearly there must be. The real question is, what should they be and how far should it go?

Bond: So, in essence, you reject Justice Black's view of the First Amendment that "no law" means "no law?" n4 You're willing to [*622] acknowledge that in some, as yet undefined, set of circumstances, it's wholly appropriate for the government to regulate the use of the Internet to protect interests that might otherwise be injured?

Kozinski: It's a compelling question. I want to make sure that by signing on to it, I'm not signing on to too much. But, I think the answer is yes. I believe that even Justice Black himself did not mean to be absolute when he said "no law" means "no law." I think that all speech is subject to some constraints. I take for granted the assumption that, when we are talking about speech, we ought to think very carefully about what the constraints ought to be. But, I do believe there are other public interests, and that it is possible to hurt people with speech. For example, child pornography videos inflict injury on the children that act in them.

Bond: You mentioned earlier that you were distrustful to some degree of analogies and metaphors. Yet, we all know that the law tends to grow into new areas by analogy and by metaphor. In the area of the Internet, of course, we already have a huge body of law called intellectual property. And there are a host of difficult questions about the relationship, if any, between doctrines in that area and regulation of the Internet. John Perry Barlow, in an article entitled "The Economy of Ideas," n5 claims that existing intellectual property law cannot be reworked to apply to digitized expression, and that there should be no legally enforceable rights of ownership in digital products. How do you react to the general claim that much of current intellectual property law, particularly trademark and copyright law, simply cannot be applied by analogy to the problems raised by the Internet?

Kozinski: When I first saw Barlow's statement, I was a bit repelled. But, the more I think about it, the more I believe there is some wisdom to what he says, without signing on entirely. Years of experience, judicial and otherwise, have taught me that intellectual property advocates will grab whatever rights they can. If you stroll down the street humming someone's tune, he will sue you and demand that you pay him a royalty. Very often, what owners of intellectual property don't realize, or aren't willing to grasp, is that they are traveling a two-way street. A fairly broad margin of fair use often generates interest in the copyrighted work.

I harken back ten, twelve, or fifteen years ago. When VCRs first came out, the movie industry went berserk. The movie moguls said, "We'll never make another movie again. Who will bother to make [*623] another movie if, as soon as you put it on tape, everyone can copy the tape?" Recall the BetaMax case? n6 The movie industry was up in arms that rental stores would buy a movie for eighty dollars and rent it out for two or three dollars. Now they were saying, "Who will ever buy another movie again for eighty dollars, when they can rent it for three dollars?" Anyway, the movie industry lost that battle, and what has happened as a result? The market in tape-recorded movies has become vastly greater than anybody ever imagined, precisely because people are able to rent movies at the local video store. Because they do not have to say, "Gee, if I buy a VCR and I want to see any movies, I'm going to have to pay eighty dollars a pop every time." They don't have to make that large economic decision. They can buy a VCR for two or three hundred dollars, and every time they want to see a movie, they can rent it. Once the movie industry lost the BetaMax battle, its market for tapes largely became the movie rental market. It's an inherent part of the industry. I think that anyone who looks at the situation objectively would conclude that the user and rental stores have been good for the movie industry.

Intellectual property is not just the thing itself. It's not just the film. It's not just the television program. It is not just the written text of a book. It is the entire package. With software, for example, it is software support. Let's face it, most sophisticated software doesn't work in a manner that's obvious to the average consumer. You've got to make that "support service" phone call, and ask, "Why won't "Duke Nukem' load on my machine?" There's a big difference between having a shareware copy, or a pirated copy, or a borrowed copy, and actually having the retail item, complete with manuals and technical support.

Much the same is true of books. You can download a lot of books on the Internet. But who wants to read them that way? Do you want to sit in front of a CRT and burn your eyes out? Do you want to print it out on your laser printer, and then manage a load of several hundred loose pages? No, you want the book. It's worth the twenty or thirty dollars to own a real book.

Dave Barry gets circulated on Internet gag lists all the time. I bought a couple of his books because someone sent me some of his articles. I said, "Hey, this guy's pretty funny. I think I'll go out and look for his books." Happens all the time.

To some extent, then, I join Barlow in his sentiment. The fears that the global communications network will destroy copyright are [*624] vastly overblown. In many ways, I think, there are flip sides. There are advantages to the copyright holder when Internet users get to try out software, get to see bits and pieces of books, or the like.

Bond: Allow me another follow-up question. As I listened to what you were just saying, I wondered whether the economics of the Internet will ultimately determine the scope of legal protection of various aspects of the Internet. Until someone demonstrably is making money at the expense of someone else's initial investment, will there likely be any need to regulate the use of the Internet with respect to trademark and copyright?

Kozinski: I don't mean to be predictive, for I don't know what will actually happen. I'm only saying that I'm wary about over-regulation of intellectual property on the Internet. I have become wary in court, where I've witnessed lawyers say, "If you don't issue a TRO, if you don't give us this or that, the value of this intellectual property will be destroyed." Lo and behold, the court doesn't give the thing they ask for, and the property is not destroyed. Predictions of what the future will bring are always apocalyptic, particularly by people who think their proprietary interests are being infringed. My suggestion would be that the law tread very carefully in this area. Before we start adopting intellectual property legislation for the Internet, before we start trying to solve problems, we need to see some documented instances of these problems. We may be doing exactly what the owners of intellectual property want us to do, but we very well may end up with a policy that's neither good for them nor good for the rest of us.

Bond: The floor's yours, ladies and gentlemen. This is a rare opportunity. Seldom do laypersons get an opportunity to direct questions to a sitting judge. Usually, the questions go the other way. So take advantage of this opportunity.

Question: Does a U.S. court have jurisdiction over a web server located outside the United States just because it can be accessed from the United States? If the operator of the web site has no U.S. assets or connection to the United States, why should the courts have jurisdiction?

Kozinski: Oy veh! It's hard for me to answer that question for a couple of reasons. First, it's a specific legal question that might conceivably come up one day in my court. Second, it also doesn't have quite enough facts for me to give a meaningfully specific answer.

Generally, however, jurisdiction is a question of real-world power. The courts may enjoy all the theoretical jurisdiction you can imagine, but if they can't get their hands on some person or some set of assets, [*625] it won't do you any good at all. To use a non-Internet analogy, for years and years the United States transmitted capitalist propaganda behind the Iron Curtain in Eastern Europe; it was highly illegal under Soviet law, but what could the U.S.S.R. do about it? You can't exercise jurisdiction over someone whose body you don't have or whose assets you don't have.

We may care very much that someone offshore is providing a web site that downloads pornographic or other undesirable materials. But there's nothing much that we can do about it under domestic law. I think it's well worthwhile for us to pursue international treaty relationships with other countries to try to stamp such things out. If we are interested in keeping these things away from our borders, that's one way of doing it. Then, you have to realize that we must submit, to some extent, to the sensibilities of people in other parts of the world.

But I haven't given up on technology. Often, the Internet is presented as a fait accompli. It is here and there's not much we can do about it. America can keep ballistic missiles from entering and hitting our cities, but we can't keep pornographic messages off our shores? I don't think so.

It's not clear to me at all that having a wide-open, single Net where people can get on and say whatever they want, whenever they want, free of any governmental or private control, is the way things are going to go. This may simply be the way things are evolving now, so long as only those of us in the intelligentsia or affluent society are able to get on. But, when you start getting Internet access by people who have to spend only a few hundred dollars for a platform, perhaps using a television or a telephone, we may well see a breaking up of the whole-body Internet into specific smaller components, intranets of sorts, where access will be restricted by private contract. It is nowhere written that every road has to connect to every other road. And in solving a lot of these problems, both internationally and domestically, the answer may well be private contract, not government intervention.

Bond: The judge has used the phrase "the information highway" several times. I must share with you the story of an e-mail message that I received a couple of weeks ago from a former colleague. He related that a North Carolina legislator had asked the governor's office precisely where the information highway was going to be located within the state, and whether the state police force would have to be expanded in order to patrol it. A question from someone perhaps more sophisticated than that? [*626]

Question: Ought there to be a constitutionally protected right to be anonymous on the Internet? Do I have a constitutionally protected right to refuse to disclose my true identity on the Net?

Kozinski: We have a culture that cherishes anonymity to a large extent. This very problem comes up in a variety of communications contexts. One of them is the "caller-ID" service on the telephone. Many people are incensed that, if they call, someone on the other end may figure out who they are. Somehow, this is perceived as an invasion of privacy.

What do I think of this? Do you have the right to knock on someone's door with a bag over your head? I'm not at all sure that you are entitled to access people's homes by telephone or by computer, and not make it known who you are. In the currently wide-open electronic regime, I have a serious problem with anonymous e-mailers. I think much mischief can be done in an environment where most of us are knowledgeable enough to run a web browser, or knowledgeable enough to send and receive e-mail, but not knowledgeable enough to change our domain names. I might feel differently if we submitted to anonymity contractually. If you belong to an intranet or subnetwork where you sign onto sending or receiving anonymous messages, that would be just fine.

Again, we might worry about the kind of things that can be done once you don't know who the sender is. You might worry about blackmail, espionage, pornography, or simple harassment. It's a bit intimidating to have your mailbox flooded with a thousand anonymous messages. "Well, how about changing mailboxes," you might think, but that's a huge inconvenience. And you should not have to suffer that.

Again, in an environment where everything is wide open, it may well be that we will come to the conclusion that only legislation will be able to deal with it. But, I'm not quite sure, for there may well be private solutions.

Bond: I have three questions here, all of which raise issues of First Amendment protection for particular kinds of messages or images. I'm going to give the judge all three, and he might evaluate any or all of them within the time frame that remains. The first question asserts as a premise that the activities of the Nazi party and the Ku Klux Klan have exploded on the Net, and asks how much protection such messages ought to get. More particularly, is this a type of hate speech that might be prohibited? The second question hypothesizes fully computer-generated images, which would otherwise be illegal if they were actual photographs, such as child pornography. [*627] What if a computer user generated the image of a child so that there was no photographic subject who has been personally victimized? The final question focuses on the difference between print texts and electronic texts. It asks that, unless the electronic medium is entitled to the same constitutional protection as the print medium, how might we deal with the fact that an increasingly large numbers of texts will be in both print and electronic forms, or perhaps only in electronic form? With these three specific issues in mind, the general question is how do we wrestle with such First Amendment problems in the context of the Internet?

Kozinski: Regarding hate speech, I would not apply any special law to the Internet. I haven't seen much hate speech on the Internet myself, but I know there are forums where people engage in hate speech. I can't see how that is constitutionally capable of being prohibited. If hate speech occurs on a bulletin board, or in a chat room, where people come and go, the sanction may be that people will leave and won't come back. But I can't imagine how it could be legally prohibited. In many ways, there is a far less strong case for prohibiting hate speech on the Internet than for prohibiting the burning of a cross in front of a black family's home. Assume that the cross is not on their property, so you don't have the hook of a property crime. Let's say someone is walking on the sidewalk with a burning cross. That invades the privacy of the home. It makes it difficult for the targets to leave their home. The homeowner would have to confront the hate speaker in entering and leaving the house, and that presents a much stronger case for prohibition. Nevertheless, we have a lot of difficulty prohibiting even that, consistent with the First Amendment. It's much harder to justify prohibiting hate speech which is confined to a chat room or a Usenet group, and is seen by no one who doesn't go there.

Things change, it seems to me, once hate speech is directed to particular people. Getting hate e-mail should be no different from getting hate any other kind of mail. Whether this could be prohibited depends on whether the same letter, if put in the U.S. mail, would be subject to punishment. I can't see any reason why we should apply a different standard. I'd have to hear the argument in the context of a particular case. Off the top of my head, however, it doesn't seem to me there's any particularly strong reason to punish or prohibit one more than the other.

Again, I want to emphasize that currently we do have a relatively "open Net." If we keep thinking about it like the Post Office, if we keep thinking that the streets leading from everyone's doorstep must [*628] connect to everyone else's doorstep, we are going to keep running into these kind of problems. This is not necessarily the case. As a contrast, let me give you the example of the intranet we have in our court. This net reaches judges all the way from Fairbanks, Alaska, to Billings, Montana, and from Hawaii all the way down to San Diego. We generate tens of thousands of e-mail messages a month. We all write to each other, and we write to other court personnel, but nothing ever comes through there that is not in some way court-related. And in the eleven years of being a Judge, I have never seen the system misused, because the intranet is so tightly regulated and because every message goes first through our server in San Francisco, so everyone knows exactly where the message comes from. Any attempt to run a prank or to clog up the system would be detectable immediately. I am not at all sure that a lot of problems with hate speech, as with copyright infringement or pornography, can't be solved simply by having people connect through technology only to those people they really want to talk to.

It sometimes gives you a great sense of freedom to be able to get on your computer and connect to anywhere else in the world. What you've got to realize is that when you connect with everybody else in the world, everybody else in the world is connected to you as well. And there's a cost involved in that.

The question of computer-generated images has been sticking in the back of my mind, and it's difficult to come to grips with. I certainly think that computer-generated child-pornographic images raise more limited concerns than actually photographing real children. Let me approach this more gingerly by considering, first, computer-generated pornography depicting adult subjects. Arguably, in such cases, one's privacy interests stop with one's own body. Accordingly, I would be very wary of First Amendment protection for Esquire's antics of twenty or thirty years ago, when the magazine ran a picture of Henry Kissinger in the nude, by grafting a photo of Kissinger's head onto a photo of the unclothed body of another man.

When dealing with minors, however, it seems to me that there are several forms of potential injury. Being involved in the real sexual act is certainly the worst of them. But children may have to live for the rest of their lives with wrongs that happened in childhood. If you use the faces of real children, and then manipulate them in such a way as to make it appear that the children were engaged in sexual activities, I would be concerned about continuing harms inflicted on them as they grow up by having images on the Internet that make it look like they engaged in child pornography. I would have to think long and hard [*629] as to whether the First Amendment would entitle a pornographer to morph the faces of real children on computer-generated bodies. Certainly, the problem is different if the child pornography consisted entirely of made-up faces and bodies.

Finally, regarding the question of print versus electronic texts, there are clear distinctions between the media forms. Presumptively, I would start with the idea that they are to be afforded the same kind of safeguards, both in terms of copyright and First Amendment protections. Yet, I think that if we don't account for the differences between the various media forms, we're likely to go astray. Arguably, the differences are very easily bridged. For example, you can take an excerpt of a printed book, and using a scanner and OCR software, you can turn it into a digitized electronic counterpart within minutes. But we ought to remember what we were talking about earlier - John Perry Barlow's suggestion that the packaging may be the most important part of a copyrighted work. When you upload a book, what you get is not exactly a book. You have a different thing. You can't carry it in a pocket or read it in bed. It may not be something people are willing to use in lieu of a book. And the author, the copyright holder, might not have lost very much at all.

Bond: Judge, we are at the witching hour. Thank you very much.


n1. 929 F. Supp. 824 (E.D. Pa. 1996).

n2. Richard Dooling, Most of These Guys Are Lawyers, Right? N.Y. Times, June 15, 1996, 1, at 19.

n3. 47 U.S.C. 151-613 (1994).

n4. Smith v. California, 361 U.S. 147, 157 (1959) (Black, J., concurring).

n5. John Perry Barlow, Wired 2.03: The Economy of Ideas (visited Mar. 26, 1997) .

n6. Sony Corp. of America v. Universal City Studios, Inc., 464 U.S. 417 (1984).

Judge Kozinski was appointed United States Circuit Judge for the Ninth Circuit on November 7, 1985. Prior to this appointment, Judge Kozinski served as Chief Judge of the U.S. Claims Court, 1982-85; Special Counsel, Merit Systems Protection Board, 1981-82; Assistant Counsel, Office of Counsel to the President, 1981; Deputy Legal Counsel, Office of President-Elect Reagan, 1980-81; Attorney - Covington & Burling, 1979-81; Attorney - Forry Golbert Singer & Gelles, 1977-79; Law Clerk to Chief Justice Warren E. Burger, 1976-77; and Law Clerk to Circuit Judge Anthony M. Kennedy, 1975-76. Judge Kozinski received both his AB degree (1972) and JD degree (1975) from UCLA where Judge Kozinski was graduated first in his law class of 300. He served as managing editor of the UCLA Law Review.