In United States v. Comstock, a unanimous panel of the Fourth Circuit (consisting of a Clinton appointee, a George W. Bush appointee, and a senior district court judge appointed by Reagan) held unconstitutional 18 U.S.C. § 4248, which "authorizes the federal government to civilly commit, in a federal facility, any 'sexually dangerous' person 'in the custody' of the Bureau of Prisons -- even after that person has completed his entire prison sentence." The panel held that Congress's enumerated powers do not reach this far, because Congress lacks a general police power aimed at protecting the public at large from crime.

Here's an excerpt, though it focuses on only part of the government's argument:

Federal commitment of "sexually dangerous persons" may well be -- like the suppression of guns in schools or the redress of gender-motivated violence -- a sound proposal as a matter of social policy. But policy justifications do not create congressional authority....

The Government ... contends that § 4248 constitutes a necessary and proper exercise of its power to prevent "sex-related crimes." But the federal government simply has no power to broadly regulate all sex-related crimes, as § 4248 purports to do.

Consistent with Congress’s limited powers, federal statutes regulating sex crimes are limited in number and breadth, specifically requiring a connection to interstate commerce or limiting their scope to the territorial jurisdiction of the United States. In contrast, § 4248 targets "sexual dangerousness" generally, without any requirement that this undefined danger relate to conduct that the federal government may constitutionally regulate. Because most crimes of sexual violence violate state and not federal law, many commitments under § 4248 would prevent conduct prohibited only by state law. Section 4248 thus sweeps far too broadly to be a valid effort to prevent federal criminal activity....

At its core, the Government’s argument attempts to "pile inference upon inference" so as to "convert congressional authority under the Commerce Clause to a general police power of the sort retained by the States." Were we to accept the Government’s logic, Congress could authorize the civil commitment of a person on a showing that he posed a general risk of any sexually violent conduct, even though not all, or even most, of this potential conduct violated federal law. This argument would convert the federal government’s limited power to criminalize narrow forms of sexual violence into the general power to regulate all sexual violence, including acts which violate no criminal statute. Congressional power does not reach so far.

Here are some tentative thoughts, which might well change as I think about the matter more.

1. On the merits: I sympathize with the panel's concerns. The Constitution clearly provides that Congress shall have only the limited powers that it grants -- not including a general power to prevent crime -- and it seems to me that courts should enforce those constitutional constraints as much as they enforce others. To be sure, there are enough precedents authorizing very broad assertions of Congressional power that I doubt that courts ever will return to the original understanding. But some policing of the outer boundaries of Congressional power is proper, as Lopez, Morrison, and City of Boerne v. Flores make clear.

At the same time, presumably civil commitment of sexual offenders is aimed at preventing repeat sexual offenses. (Let's set aside whether such civil commitment after the end of a sentence may sometimes deny people liberty in violation of the Due Process Clause; that's an issue unrelated to the federal power question, since it would apply equally to states.) And presumably someone who committed a federal sex crime (e.g., possession or trafficking of child pornography) is pretty likely to commit another crime of much the same variety -- which will likely be a federal crime -- and not just some other random state sex crime. If the Commerce Clause power to regulate commerce authorizes Congress to ban commerce in child pornography, and the Necessary and Proper Clause therefore authorizes Congress to ban even private possession of child pornography, then it's hard to see why the Necessary and Proper Clause wouldn't authorize continued detention of people who have shown a willingness to commit such federal crimes.

One way of thinking about it might be to think about the historically established practice of civil commitment of people found not guilty by reason of insanity. If someone is tried for a federal crime and found insane, he won't be imprisoned for the crime -- since he's not criminally guilty -- but he will be locked up in a mental hospital so long as he is thought to be dangerous. I think that's right, but how does it fit the panel's decision?

After all, the person is not guilty, so Congress can't appeal to its power to punish federal criminals (just as the people in this case can't be further criminally punished, since there terms are up). True, we worry that this insane person will commit another crime, but under the panel's reasoning, that might well be a state crime. So must Congress release such people unless it gets a state to agree to take custody of them? Perhaps that's the right answer, since Congress lacks the enumerated power to detain them -- but I'm skeptical that this is so.

Alternatively, perhaps Congress can detain these not-guilty-for-reason-of-insanity people, as part of whatever federal power justified their criminal trial in the first place. If someone insanely commits a federal crime, Congress should be able to lock him up to prevent him from committing more such federal crimes in the future. But why wouldn't this equally be so for the people adjudged sexually violent predators? (I should stress again that this is a separate question from the broader Due Process Clause question of whether any government, state or federal, may civilly lock people up because of their future dangerousness, after their criminal term has expired.) In any case, that's why I'm tentatively skeptical about the panel's reasoning.

2. On the future of this case: The panel reports that this is the first federal court of appeals decision passing on the constitutionality of the statute. There's thus no circuit split of the sort that would normally signal a likely decision by the Supreme Court to hear the case. But there is a federal statute being struck down, and that sort of interbranch split -- the legislature thought the statute was constitutional, the executive is defending the statute as constitutional, but the judiciary is saying it's unconstitutional -- will often lead to a Supreme Court hearing even without an inter-circuit split.

So if the Fourth Circuit doesn't rehear the case en banc -- and en banc seems unlikely, given the general presumption against en banc and the mixed political makeup of the unanimous panel -- and if the Obama Administration asks the Supreme Court to hear the case, I predict the Supreme Court will indeed agree to hear the case.

Thanks to How Appealing for the pointer.