Here is my on what Judge Kozinski wrote:
a. You have anonymous snail-mail and telephone calls--why not e-mail? The truth is, anonymous snail-mail and telephone calls are also an invasion of privacy, but there is not much we can do about them.
1. But do anonymous snail-mail or phone calls invade privacy? Is an anonymous phone call an invasion of privacy in a way that an ordinary phone call is not? Is anonymous snail-mail more invasive of privacy than ordinary junk snail-mail? Is it the anonymity that is objectionable? Many people object to spam even when they know who it's from. It's certainly reasonable to define privacy as including an interest in disattention or not being drawn into communication. But isn't that interest implicated every time someone calls you? And not just when the caller has ID blocking, or doesn't tell you who he or she is? Normally, when I make a phone call, I don't identify myself. If I'm asked, who is this, then we have a decision point. Until I do, isn't the call anonymous? Once I do, it's not - any more - but it was until then. Put slightly differently, in daily life we interact with those we know little about, and we learn more (usually) as the interaction unfolds. Interaction is sequential, involves turn-taking; trust is built, not warranted at the outset. So what is distinctly privacy-invasive in being anonymous?
Someone asked whether I objected to getting anonymous snail mail if it was not threatening. The answer is YES, just like getting an anonymous phone call is objectionable. When the person who communicates with you insists on retaining anonymity, there is always an implicit threat--they know who you are, but you don't--you feel vulnerable, you doubt their motives, you have difficulty knowing whether to trust their representations. Anonymous complaints against co-workers and supervisors was a standard way to get people into big trouble in Communist Romania when I was growing up.
The thrust of his comment seems to be (a) that when the caller is anonymous, there is an imbalance: you know less than the caller. You interact on an unequal footing. This may well be, as he puts it, "objectionable"; but the appeal is to some norm of informational reciprocity, of fair disclosure. What I don't get is how that is a violation of *privacy.* Does "being named" dispel the objectionable quality of the interaction? Don't we often interact with people whom we hardly know? It seems neither necessary nor sufficient. I can feel vulnerable, doubt motives, etc., even when I know someone's name. Isn't he really talking about reputation? To what extent does nymity (I just coined this as the opposite of anonymity) matter? Name may permit you to assess reputation later, but may be of little use at the time.Is that a privacy issue? Does Kozinski assume (b) that the anonymous are more dangerous, that anonymity maps to unaccountability, and unaccountability maps to higher likelihood of threat. Even if that's true, is that a matter of privacy? 2. As for anonymous snail-mail, consider that an anonymous envelope isn't necessarily an anonymous letter. The anonymity is relative to what's exposed. Map that onto e-mail. Isn't that equivalent to anonymizing the header and encrypting the message itself? That the header doesn't reveal sender identity doesn't entail that the message is itself anonymous. The anonymity may merely be against third parties. Kozinski himself says:
2. I also agree that it should be possible to have mutually agreed-upon anonymity--i.e. I write to you and you write to me and we both know who we are, but nobody else does. No problema--it's nobody else's business.
So from that POV, wouldn't a ban on anonymous remailers, to the extent that they don't preclude nymity w/i the message body, reach too far? Blatant self-promotion: BTW, I recently published an article on anonymity and the McIntyre case in the Oregon Law Review. 75 Or.L.Rev. 117 (1996). Lee