A friend of mine wrote to ask:

The Supreme Court overturned the Jaynes conviction on First Amendment grounds, yes? I’m wondering what that could mean from the spam filtering perspective.

Spam filters, and in particular DNS blacklists are intended to prevent e-mail from being delivered. Doesn’t the First Amendment make it illegal to block speech? The short answer is no, but of course it’s slightly more complicated than that in practice.

If I may put on my constitutional lawyer hat for a moment (it doesn’t fit very well since I’m not any kind of lawyer, but nothing I’m about to say is particularly new or controversial,) let’s review what the speech part of the First Amendment does. Like the rest of the Bill of Rights, the First Amendment limits the actions of the US government. Originally it just affected the Federal government, but the 14th amendment was eventually interpreted to apply it to the states as well. The key Supreme Court case was Gitlow vs. New York in 1925, an interesting case about an anarchist in New York who had been convicted of advocating overthrowing the government, and the court held in 1925 that the First Amendment precluded state laws against speech, even speech advocating overthrow of the government, short of a clear short-term threat to public safety. (Wikipedia has more details.)

The First Amendment problem in the Jaynes case was quite specific to the wording of the Virginia anti-spam law, and had nothing to do with the merits of the specific case, as I blogged at the time (blog, on CircleID).

The one other case I know of where the First Amemdment forbade state laws against Internet speech was a 2002 Pennsylvania state law that required ISPs to block addresses on a state-run blacklist of IP addresses that allegedly contained child porn. A federal juge threw it out for obvious reasons, even though child porn is unambigiously illegal, the state run blacklist was way too broad. The key fact here is that the state itself managed the blacklist.

Courts have not found the First Amendment protection absolute, even when a state is doing the blocking: In White Buffalo Ventures v. University of Texas, dating service White Buffalo sued the U of T for blocking its spam, and the court held that even though U of T is part of the Texas government, it was OK for them to block commercial spam sent to their users.

To return to the original question, as I would hope was obvious, ISPs, even very large ISPs, are not part of any government, so the First Amendment is just irrelevant. Beyond that, the Congress has provided specific protection to ISPs that block unwanted content in 47 USC 230. It was part of the Communications Decency Act of 1996, and although most of the CDA was voided on 1st amendment grounds, Sec. 230, which has been litigated and challenged many times, has survived all challenges. So spam blocking is legal, even if it blocks speech that a government might not be able to block.

One recent development that does make me worry is Net Neutrality. The US used to have a lot of competitive ISPs, but due to a series of ill-considered actions by the FCC, in most of the US there are only two viable broadband ISPs, one owned by the local phone company and one owned by the local cable company. Rather than undo the mistakes so there can be competitive ISPs (as the UK and other countries have done), Net Neutrality accepts the duopoly as a fact of nature, and attempts to regulate them to prevent various sorts of abuses. People with experience dealing with the FCC insist that Net Neut wouldn’t affect spam filtering, but it’s not hard to imagine a political or religious spammer arguing that the pair of broadband ISPs offer an essential way to communicate with people, that his spam is no worse than mail from other mailers whose mail isn’t blocked (because they only send it to people who want it, or because unlike the spammer they do not send it in vast server-crushing volume), so Net Neut should require that all mail of that category be treated the same, i.e., delivered. To people familiar with the way that ISPs work, this is a silly argument, but few courts understand network technology, and it has a superficial plausibility that could be persuasive to a judge. I hope it doesn’t come to that, but I’m not sanguine.