I have a column up on the homepage this morning about what may be, in E. Jean Carroll’s civil rape allegation against Donald Trump, the pivotal role of so-called propensity evidence. This calls into question the rule of evidence (which is rife with exceptions — see Rule 404(b)) that instructs courts to exclude proof of similar but uncharged bad acts that are offered to prove that the defendant likely committed the bad act charged. The point of that presumption against propensity evidence is that we only want a defendant found guilty (in the criminal context) or liable (in the civil context) because the charged bad act has been proven in isolation, rather than because a jury figured the defendant must be guilty because he’d done the same kind of thing before.

My purpose here is not to rehash the column but to correct something I said when Rich and I recorded the podcast last week.

As we discussed (and as the column elaborates), the presumption against propensity evidence has been eradicated in sexual assault cases. What rule-makers were finding, in these cases particularly, was that victims typically did not come forward for years (even decades, as in Carroll’s accusation against Trump) after a rape or other form of sexual abuse — long past the time when a useful forensic examination could be done. There are also theories offered that sex offenders are unique in their inability to control their compulsions, and thus that the propensity to offend is especially relevant in sexual-assault cases — theories I do not find persuasive because I believe sociopaths tend to recidivate across the board, not just in the sex-offense context, and are responsible for most crime. (I also don’t like the notion of denying people’s agency. If a person can’t control himself, the justification for punishing him for his actions is weaker.)

The mistake I made in the podcast was to suggest that this easing of the rule against admitting propensity evidence occurred mainly in the context of civil cases. In general, we tolerate lower evidentiary burdens in civil cases rather than criminal cases because all that is at stake (usually) is money damages — we are not talking about removing someone’s fundamental constitutional right to be at liberty by incarcerating them, as in criminal cases. Thus, to take the most prominent example, a plaintiff wins a civil trial by proving the claim by mere preponderance of the evidence (showing the allegation is more likely than not true), whereas the government must prove guilt beyond a reasonable doubt to convict someone of a crime.

At the time we did the podcast, I thought the easing of the rule against propensity proof had been done for the benefit of civil plaintiffs. That is not the case — which goes to show you, dear reader, how long it’s been since your over-the-hill correspondent (who, as a prosecutor, never tried a sexual-assault case) has been in court. As explained by Judge Lewis Kaplan, who is presiding over the Carroll/Trump trial, the federal rule so evidence have been amended to permit propensity evidence not only in civil sexual-assault cases (Rule 415), but also in cases of child-molestation crimes (Rule 414), and in other sexual-assault crimes (Rule 413).

We’ll try to remember to talk about this when we do the next podcast later this week. (I say “try” because you never know anymore what crazy pressing legal or national-security story will push everything else to the bottom of the pile.) But to my mind, the lowering of barriers to propensity evidence is a good development in the law. It is consistent with common sense and how crime actually works. To the extent it could be abused (i.e., defendants could be found guilty or liable because of the propensity proof rather than the proof of the charged crime), that is best handled by letting the trial judge (a) guard against admitting too much propensity evidence, and (b) dismiss a case in which the government’s evidence of the charged bad act is too weak, rather than letting it go to the jury.