gregh  2007-07-07 18:17  fourth_amendment  Law  ninthcircuit  penregisters  privacy   

Until yesterday, my paper on Fourth Amendment and statutory protections against interception of Internet communications had little directly related case law. Now I have the Ninth Circuit's decision in U.S. v. Forrester, in the matter of the second defendant, Alba. When I first read the headline, I was a bit concerned. After reading the rest of Prof. Solove's commentary, as well as the opinion itself, I am no longer.

That is to say, I am no longer concerned about the state of my paper. I am concerned about where the courts are going to take constitutional protections on the Internet. Many of my complaints continue to come down to many of the overly simplistic analyses of email transport on the Internet. Prof. Solove trumpets Prof. Kerr's work in the above post:

Orin Kerr has usefully analogized the distinction between the non-content / content information to that between an envelope and the contents of a letter. The envelope contains addressing information that is exposed to others; the contents of the letter are concealed. Envelope information falls outside Fourth Amendment protection, but content information is fully protected by the Fourth Amendment.

Kerr's work does often make this distinction, but it often does so with an apparent misunderstanding of how email transportation works. Judge Fisher, in the Forrester decision, follows a path of similar technical missteps.

First, e-mail and Internet users, like the telephone users in Smith, rely on third-party equipment in order to engage in communication. Smith based its holding that telephone users have no expectation of privacy in the numbers they dial on the users’ imputed knowledge that their calls are completed through telephone company switching equipment. 442 U.S. at 742. Analogously, e-mail and Internet users have no expectation of privacy in the to/from addresses of their messages or the IP addresses of the websites they visit because they should know that these messages are sent and these IP addresses are accessed through the equipment of their Internet service provider and other third parties.

In fact, all telephone users rely on third-party equipment. However, there are decided differences between telephone numbers opening a circuit and the routers that route traffic to destination IP addresses. But the problem runs deeper, and this is a significant complaint of mine in the literature. This peering of email and Internet use, as if sending an email is somehow different, is what allows for this butchery. For instance, the next paragraph begins:

Second, e-mail to/from addresses and IP addresses constitute addressing information and reveal no more about the underlying contents of communication than do phone numbers.

An Internet email address without an IP network attached goes nowhere.

There is a useful (to me) diversion into an email/snail mail comparison:

The government’s surveillance of e-mail addresses also may be technologically sophisticated, but it is conceptually indistinguishable from government surveillance of physical mail. In a line of cases dating back to the nineteenth century, the Supreme Court has held that the government cannot engage in a warrantless search of the contents of sealed mail, but can observe whatever information people put on the outside of mail, because that information is voluntarily transmitted to third parties.
. . .
E-mail, like physical mail, has an outside address “visible” to the third-party carriers that transmit it to its intended location, and also a package of content that the sender presumes will be read only by the intended recipient. The privacy interests in these two forms of communication are identical. The contents may deserve Fourth Amendment protection, but the address and size of the package do not.

Oh, it does, indeed. Email has an outside address called an IP address. But beyond that, it has outside addresses called envelope addresses, as specified by the RFCs. These, the technical envelope specifications, are ignored in the literature, but they are key to using Prof. Kerr's envelope terminology. For something to be an envelope, it must surely have an impact on the delivery of a message. The RFC content of a message has no bearing on delivery, yet Prof. Kerr (and now the Ninth Circuit) appear to be perfectly willing to allow it to be intercepted.

And so, this is just another lousy decision, crafted not with an understanding of Internet communications, but with the more typical understanding, that seems to reflect the belief that a "send button" magically whisks an email message to some remote spot on the globe. In the case of the Ninth Circuit, that spot appears to be someplace that need not have even heard of IP addresses.