The purpose of a trial is not to discover the truth. Sorry. Whether civil or criminal, bench or jury, the purpose of a trial is to decide on an “official version” of the facts. The purpose of the justice system is to make an enforceable, hopefully final, decision about a dispute. The system does this by applying the law to the facts, and determining what the appropriate outcome is. The system already knows what the law is, presumably. But it can’t apply that law — it can’t do anything — until it has a set of facts to work with. We’d like the official facts to be as close to the truth as possible, of course, but one way or another we need to decide what they are.
That’s what the jury is for. That’s all the jury is for, most of the time: to be the “finder of fact.” Obviously, there are competing versions of the facts to choose from, or else there wouldn’t be a trial. The jury has to decide which facts the justice system will get to use. (And as Scott Greenfield pointed out this morning, once that official version of the facts has been determined, the system is extremely loath to revisit them. These are the facts we’ll rely on for damages, for sentencing, for appeals, forever.)
The jury’s job is important. It is sacred. The idea that twelve honest members of the community can assess the evidence and figure out what was proven and what was not proven is integral to our concept of justice. And in a jury trial it’s important that only the jury gets to perform its sacred task. We don’t let anyone else decide the facts for them. That would mean replacing the jury of twelve with a jury of one. We tightly control who gets to testify, what questions they can be asked, and what they’re allowed to say. We limit the evidence only to relevant testimony, and try to exclude categories of evidence that are too unreliable to use — especially evidence that cannot be challenged.
And we certainly don’t let witnesses or lawyers vouch for the truthfulness of their testimony. “You have to believe this because, in my opinion, it’s true” is not something you can say to a juror.
Unless you’re an expert witness, that is. Then you get to not only opine on what evidence means, on facts the jury needs to decide, but also on the reliability of your opinion. Ideally, an expert is an objective witness with no stake in the proceedings, who has knowledge of a subject that is just too complex or arcane to expect of jurors. Whatever they’re testifying about, they’re needed because it’s not common knowledge. So the expert gets to summarize the arcane subject, draw factual conclusions for the jury, and also give his opinion about how reliable he is — how confident he is in his conclusions.
When he gives his expert opinion, the expert witness does the jury’s job for them. The jury needn’t assess his reliability — the judge called him an expert, and the expert himself said the basis for his opinion was reliable. The jury doesn’t need to assess his summary or his conclusions about the evidence — if they could do that, they wouldn’t have needed an expert in the first place. And besides, the expert gave them his expert opinion of how reliable that conclusion is. The expert witness can easily become a jury of one.
So we are really really careful about who we allow to testify as an expert witness, and strictly limit what they can testify about.
Of course there are exceptions. And of course they’re mainly to be found in criminal trials, where the stakes are highest, and the jury’s role is most important.
One exception you’ve probably heard about before. [Heck, I’ve probably griped about it here a couple of times, only just as I can’t be bothered to edit these posts before I post them, I can’t be arsed right now to look up whether I’ve written on it already.] This is the exception for expert police testimony. Instead of establishing valid academic and professional credentials to ensure that this expert knows what the heck he’s talking about, and instead of having him specify the resources and data on which he relies (and thus give the other side a chance to challenge the validity of those sources), we pretty much let the police witness provide his own opinion about whether he’s an expert, and then we call him an expert, and then we let him tell the jury what the evidence means.
Cop: “I was trained by other police officers about how drug deals work. I’ve participated in lots of arrests that involved drugs. In my opinion, I am super-familiar with how drug deals work.”
Judge: “Okay, jury, this guy’s an expert.”
Cop: “In my expert opinion, those apparently innocent bits of evidence really mean the defendant sold that other guy some drugs.”
Jury: “Well, that’s that. When’s lunch?”
It’s self-serving testimony by the government, deciding for the jury the ultimate issue of the case. It’s not from a disinterested witness, but usually from the same officer who made the arrest. He’s telling the jury that, in his opinion, he was right. And the judge is telling the jury he’s giving this opinion as an expert. The witness isn’t giving any details of what he’s basing this expert opinion on, and so its reliability cannot be challenged. All that’s happening is he’s getting to vouch for his own expertise, and the government is getting to vouch for the reliability of its evidence.
But this isn’t what I wanted to complain about today. At least the officer was first screened and offered as an expert. Eyewitnesses, on the other hand…
The eyewitness gets to give expert opinion testimony without even being admitted as an expert.
Eyewitness testimony is notoriously inaccurate. On average, it’s no better than a coin toss. More than 3/4 of the death-row exonerations to date have been from convictions based on eyewitness testimony. Humans just don’t see everything as accurately as we think we do, our memories are malleable as hell, and we’re really bad at pointing out the culprit in a lineup. But boy howdy are we confident! We think our memories really happened, even when they’re demonstrably false. We think we’re right. How can we not? But we’re wrong an awful lot.
Confidence has zero to do with the reliability of an eyewitness. Almost every eyewitness is confident that they’re remembering things accurately. Even when they’re wrong.
And yet, when an eyewitness says they’re confident that they’re right, it has a huge effect on juries. When a witness vouches for her own testimony, studies show that jurors tend to believe her. The effect is powerful — it is almost impossible to overstate it. In fact, even when a witness has contradicted herself on the stand, and has been shown to be clearly unreliable, if she tells the jury that she is very confident that she’s right, the jurors will say they found her credible.
Courts also put a lot of stock in an eyewitness’s opinion of her own accuracy. The more confident the witness feels, the more likely she’ll be allowed to testify. The Supreme Court itself has gone out of its way to say that eyewitness confidence is a factor that should be considered when assessing the reliability of her testimony.
In other words, if an eyewitness says “I’m sure I’m right,” the jury’s not only going to be more likely to believe her, but they’re supposed to believe it more.
How stupid is that?
This is opinion testimony. The eyewitness is giving an opinion on an issue of fact — the reliability of her testimony. This is a decision the jury needs to make. She doesn’t have an unusual background that lets her assess this any better than the ordinary juror. On the contrary, because it’s her own memory and perceptions, she’s the one person least qualified to give a reliable assessment of its accuracy. But she’s the one who gets to tell the jury how accurate she is.
Not only that, but in many states the defense is not allowed to even present their own expert testimony to rebut her opinion. No expert opinion that this eyewitness got it wrong. (Two main reasons: one, eyewitness unreliability is considered such common knowledge –ha! — that expert testimony is unnecessary; and two, general testimony about how people in general get it wrong isn’t evidence that this person got it wrong.) All that can be done is cross-examine to find inconsistencies and reasons to argue later that the witness was wrong. But that’s not very helpful, because again, when an eyewitness testifies that she has high confidence in her accuracy, those inconsistencies and other indicia of unreliability get ignored, and the jury tends to believe her regardless.
It’s not just a confrontation issue, but an evidentiary one as well. Witness confidence just isn’t a reliable indicator of witness accuracy. We’re wrong far too much of the time. In general the rules of evidence exclude categories of evidence that have a substantial risk of being unreliable. Everywhere else, we exclude evidence that cannot be challenged. If we were consistent, then, our rules of evidence would preclude witness assessments of their own accuracy rather than encouraging them.
It is the jury’s job to assess whether this witness got things right, and nobody else’s. Yet we go out of our way to take that decision away from them, and let the witness herself decide whether her testimony is correct. She’s giving an expert opinion on the reliability of her evidence. And not only is she not an expert, she is in fact the one person least qualified to assess the reliability of her memory.
But we let her say it. She gets to give her inexpert testimony, and do the jury’s job for them. And that snapping sound we hear right afterwards? That’s the jury’s brains turning off.