I truly and sincerely don't understand why people think that testimony like this is permissible.
Defendant gets charged with molesting his stepdaughter. The stepdaughter says he did it; defendant says he didn't. As in many of these cases, there's not much (if any) physical evidence to support either side. It is a classic credibility context.
As in most of these cases, the prosecution introduces expert testimony to explain why the alleged victim waited years before disclosing the abuse. I've got no problem with that. It makes sense to explain to the jury how the psychological effects of molestation might result in fear, secrecy, accommodation and delay.
But then the expert gets asked: "Is it common for children to make up a story that abuse occurred, when, in fact it did not?” To which the expert responds: "No, that's rare. There’s research that talks about false allegations, and . . . . it’s not a common thing for kids to fabricate such a story.” (In this particular case, the question was asked by a juror; in other cases, it's asked by the prosecutor. The answers by the expert also sometimes take different forms; e.g., that only about one to eight percent of allegations are false, or that studies have shown that only six percent of allegations are typically unfounded.)
The Court of Appeal holds today that it's improper to introduce such testimony, on the ground that it "invades the province of the jury" by essentially telling them that the defendant is guilty. (Though the Court of Appeal here also holds that it was harmless error, so affirms the conviction.)
That makes sense. But I have a more foundational question, one that the Court of Appeal neither asks nor answers:
Why is such testimony even minimally reliable; i.e., why does it satisfy the Daubert (federal) or Sargon Enterprises (California) test?
I get that this testimony derives from studies, and published ones at that. But the methodology of those studies seem absurd. In all of them, to determine whether the molestation allegation was purportedly "true" or "false," all you had was a person (or group of people) who investigated the allegation and who determined for themselves whether or not they believed the victim. Sometimes it was CPS making the call. Sometimes it was the researchers themselves. Sometimes it was a panel. But the point is, it was simply the call of a particular person or group of persons as to whether a particular disputed physical event actually occurred.
How's that "proof" of anything?
Sure, in an extraordinary small number of cases, we might be able to definitively call one side or the other "true" or "false" and rely on such a conclusion. Let's say there's a video that catches defendant in the act. Or maybe the defendant confesses, or the victim recants, or there's physical evidence that the defendant wasn't even in town on the day of the alleged crime. (Even in those extremely rare cases we might be hesitant to come to a definitive conclusion; e.g., the confession might have been coerced, the victim might have falsely recanted, or the victim may have simply been confused about the date.)
Regardless, those are by far the minority of cases. Most times, it's the alleged perpetrator's word versus the victim's, albeit with one side or the other occasionally bolstered by various forms of circumstantial evidence; e.g., whether the victim told others about the molestation, whether other victims made similar complaints about the perpetrator, etc.
Can you come to a conclusion in such cases? Sure. We do so all the time. In both criminal cases as well as in civil cases. We empanel a jury and come to a conclusion. A conclusion that we understand fully might potentially be wrong, but a decision needs to be made, so we make one, pursuant to a given standard of proof that's established according to the competing social values at stake (e.g., beyond a reasonable doubt in criminal cases and preponderance of the evidence in civil cases).
But you can't do that in a study. Or at least you can't do that just by getting together a group of people (or one person) and telling them: "Decide yes or no who's telling the truth." That's not an established or reliable methodology. Or, at a minimum, shouldn't be.
Take Brett Kavanaugh, for example. There's a ton of evidence about what he allegedly did or did not do. It makes absolutely no sense to me for any purported "study" to "find" that the allegations that he sexually assaulted Christine Blasey Ford in high school are either "true" or "false." Are the allegations in fact true or false? Definitely. Could some "study" so establish this fact by putting together even a thousand "experts" to so decide? Absolutely not. It's a judgment call. And what's true for Kavanaugh is even more true for putting together a hundred (or even a thousand) self-selected molestation cases and deciding whether those claims are "true" or "false". Except in outlier cases, yes, there is surely an objective truth, but we don't know what it is, or at least don't know sufficient to establish reliability of the type we generally require for a purportedly scientific study (which then gets passed on to a jury as established truth).
There are tons of objective facts that we simply don't know, and that we don't know regardless of whether we get purported experts to review the evidence and come to a conclusion regarding the truth or falsity of the underlying factual claims. Let's just take, for example, questions about "Who shot first?" Assume I want to test the following hypothesis: "When two armed groups meet, the group with the smallest number of participants generally shoots first." Maybe because I want to testify that based upon scientific studies, it's "rare" for the larger group to shoot first. (Just like it's purportedly "rare" for kids to make up molestation allegations.)
So I go ahead and get a group of experts to examine a variety of prior armed conflicts to see who shot first. Which side shot first at Lexington? Who shot first at Gettysburg? Did members of the Klan or the Communist Party shoot first at the Greensboro Massacre? Things like that. I'm super diligent and get together a huge group of experts -- fifty of the leading historians -- and I get them to say "true" or "false" to whether the smaller group shot first in these various examples. And based on their subjective evaluation of the evidence, I want to testify: "In over 90% of the cases, it's true that the smaller group shot first."
No way. We don't know who shot first in these examples. Regardless of the subjective evaluation of even super-qualified experts. Sure, there's evidence -- tons of evidence, even -- in favor of one side or the other. (Much more evidence than in molestation cases, I might add.) But to try to say that a study like this can reliably come to a conclusion about the truth or falsity of these events is absurd. No one would possibly think that's sufficient, much less that it's sufficiently reliable to be presented as a study with scientific validity at a trial to determine someone's guilt or innocence of a criminal offense.
I get that, at the margins, there are some things that we can reliably declare as "true" regardless of the existence of contemporary disagreement. Humans have, in fact, landed on the moon. (Sorry, Steph Curry.) Lee Harvey Oswald probably did, in fact, act alone. (Though, even there, notice the caveat.)
It's not that nothing can be established as "true" or "false". It's simply that, in the pantheon of historical facts, whether or not a particular act of molestation has occurred is very much closer to the "Who shot first" examples than it is to the "Did we land on the moon" example. And any study that purports to declare otherwise based simply upon the subjective evaluation of human observers isn't the type of study that we should declare reliable enough to admit at trial.
Do I personally believe that most molestation allegations are "true" rather than "false"? Yes, I do. Do I agree that many of the (minority-) "false" allegations occur in custody fights or other common types of cases, rather than in cases like today's? Yes, statistically, I suspect that's probably true.
But no "study" of a type like the ones relied upon by these experts supports my intuition with any type of reliability -- much less the reliability we require at a criminal trial.
So, yeah, was the expert testimony here wrongly admitted? Sure.
But the flaws, in my mind, go much, much deeper than simply invading the province of the jury.
