So a North Carolina judge held in Rothrock v. Cooke (N.C. Super. Ct. June 11, 2014) that North Carolina’s torts of “alienation of affections” and “criminal conversation” are unconstitutional. (He also relied on other theories as well.) I think, though, that the court’s analysis is almost entirely unsound. (I do agree that alienation of affections lawsuits brought based on verbal criticism of a person’s spouse, for instance by parents or friends who think the marriage is a bad one, would indeed violate the First Amendment, so long as the criticism consists of true statements or opinions rather than of slanderous falsehoods; but while such lawsuits were at one point not uncommon, they seem extremely rare today.)

1. First, a bit of explanation of the legal issue. Alienation of affections basically consists of a defendant’s (1) wrongfully (2) causing plaintiff (3) to lose the affection and often company of the plaintiff’s spouse. In principle, it could apply to supposedly meddling in-laws and other third parties, and has sometimes been applied that way, though if in-laws are looking out for their married child’s best interest such behavior might not be seen as “wrongful.” In practice, and especially in the recent past, it has generally been applied to people who supposedly seduce away one spouse from the other (if it can be shown that they caused the alienation, rather than that a preexisting alienation of the spouses caused one spouse to be interested in the defendant’s attentions), and engage in adulterous sexual conduct along the way.

Criminal conversation, like the Holy Roman Empire, is neither. It’s a tort (i.e., grounds for a civil lawsuit), not a criminal law, and it requires adultery, not conversation — it basically consists of a defendant’s having adulterous sex with plaintiff’s spouse, though of course such conduct may also often lead to an alienation of affections claim. Generally speaking, the plaintiff’s agreement to the conduct is a defense to these tort claims, so the torts don’t interfere with agreed-to “open marriages.”

Such torts used to be broadly recognized, but not any more. Many state legislatures have abrogated them, and some state supreme courts have done the same. By my last count, Hawaii, Illinois, Mississippi, New Mexico, North Carolina, South Dakota, and Utah still recognize the alienation of affections tort, and at least North Carolina still recognizes criminal conversation.

Moreover, in North Carolina such litigation is quite common. When I last checked into this, back in 2009, it turned out that in fiscal years 2000-2007, there were an average of 230 such filings per year in North Carolina — a bit over 0.5 percent of the number of all divorces, but about twice the number of product liability lawsuits, and about three-quarters of the number of wrongful death lawsuits. And awards can be huge; see this $9 million alienation of affections award.

2. But while civil liability for adultery has been on retreat in most of the nation, and is likely a bad idea, that doesn’t mean that such liability is unconstitutional. To begin with, the judge’s claim that the First Amendment precludes criminal conversation claims strikes me as quite unsound. The judge reasons that,

Regarding expressive conduct, [criminal conversation] punishes all expression of affection or intimacy in the form of consensual sexual conduct between individuals if adultery directly results from the sexual intercourse. As previously stated, adults involved in consensual sexual intercourse intend to convey a message of intimacy [earlier described as “the desire to be intimate with or display affection for one another" -EV] and a great likelihood exists that observers of the conduct would understand the conduct’s intended message of intimacy.

Yet even if this qualifies sex as “expressive conduct” for First Amendment purposes (not completely clear, but let’s assume this for now), the government may still restrict such conduct for reasons unrelated to this message, so long as the reasons are substantial, and the restriction advances those reasons. (This is the so-called O’Brien test.) The court reasons that,

No substantial, important, or even legitimate state interest exists for punishing affection or intimacy expressed through consensual sexual conduct that takes place in private; [criminal conversation]’s restriction cannot further a state interest if no such interest exists. Therefore, [criminal conversation] unconstitutionally infringes upon an individual’s protected First Amendment right of expressive conduct to engage in consensual sexual intercourse.

But that’s just assertion, and not terribly persuasive assertion, it seems to me. A marriage involves certain promises, including fidelity. The other spouse relies on these promises in many ways (by investing emotionally in the relationship, investing financially, engaging in sexual behavior that might be risky if the other spouse was unfaithful, forgoing outside sexual relationships in exchange for a similar promise by the other spouse, and so on). Why wouldn’t there be a substantial interest in punishing third parties for participating in a breach of the promise, in compensating a spouse for the harm caused by the breach, and in deterring such breaches, even if the deterrent effect is modest? (Note that the lawsuit is brought against the third party, not against the cheating spouse; but North Carolina law apparently does take adultery into account in deciding the property settlement in the event of a divorce, so the cheating spouse doesn’t get off scot-free.)

3. Likewise, the judge’s view that alienation of affections is, facially and as applied in this case, content-based (and thus unconstitutional unless it passes the “strict scrutiny” test, which is much more demanding than the O’Brien test) strikes me as unsound as well. To be sure, when the alienation of affections tort is applied to speech rather than to sex, and especially speech that doesn’t constitute solicitation of adulterous sex, the tort might be constitutional. One illustration in the Restatement of Torts, for instance, notes that there might be liability in this situation (though perhaps for the closely related tort of causing spousal separation rather than alienation of affections as such):

A, who is a stranger to B’s wife, sees B drunk in a bar. Annoyed at B’s conduct, A seeks out B’s wife, tells her of the incident, and strongly urges her to leave her husband. This persuasion induces B’s wife to leave him.

I think such speech would be constitutionally protected, and the tort would indeed be content-based as applied to such speech. But this fact pattern is almost never seen in modern alienation of affections cases; nearly all such cases involve alleged adulterous sex (or, in very rare situations, alleged actionable slander rather than the communication of accurate factual statements or of opinions). So it makes much more sense to narrow the alienation of affections tort to avoid First Amendment problems — for instance, by limiting it to adultery — especially since this is a common-law tort that was created by judges, and could quite reasonably be narrowed by judges.

4. The more plausible argument that the judge makes is that the torts interfere with the Lawrence v. Texas right of sexual autonomy. But here too the judge doesn’t explain why adultery should be treated the same as nonadulterous consensual sex. The Court’s discussion in Lawrence dealt with bans on behavior that was fully consensual, and that broke no promises to third parties. Yet the law — especially civil law as opposed to criminal law — has long distinguished behavior that breaks promises (and behavior that induces another to break a promise, or that aids another in breaking a promise) from behavior that doesn’t break a promise. Indeed, the law has drawn this distinction even as to behavior that, absent a promise, would be constitutionally protected; this is why there’s generally no First Amendment problem with imposing liability for breach of a promise of confidentiality (see Cohen v. Cowles Media (1991)).

Yet the judge, as best I can tell, doesn’t squarely confront this at all, and simply asserts that Lawrence applies as much to promise-breaking sexual behavior (and inducement or assistance in promise-breaking) as to other sexual behavior. That strikes me as a very hard position to defend.

(The judge also reasserts that, even “[i]f subjected to a rational basis test, … [the torts] still fail because, as shown earlier in this opinion, no legitimate state interest exists in punishing the sort of conduct required for sustaining either action.” But again this doesn’t deal with the counterarguments noted in item 2 above.)

So the judge’s decision seems to me wrong, and I hope the North Carolina appellate courts reverse it (though I think it would also make sense for the North Carolina Supreme Court to abrogate the tort altogether, though as a matter of state tort law, not of federal constitutional command).