Sorry, Not Sorry

Judges and juries believe they can tell when a defendant is remorseful. They can’t.

The role of remorse in sentencing.
Remorseful, or not?

Photo illustration by Slate. Photo by Thinkstock.

In our everyday lives, it’s a natural response: If a person who has done something wrong tells us he is genuinely sorry, we are more inclined to feel merciful toward him than if he is unrepentant. But our ability to accurately detect remorse, and the logic behind the conclusions we draw when someone appears to feel it (or not), deserves special scrutiny in the context of the criminal justice system, argues a new paper in the journal Emotion Review. Too often, writes DePaul University College of Law professor Susan Bandes, judges and jurors make high-stakes decisions about punishment—how long convicts should go to prison for, and even whether or not they should be executed for their crimes—based on faulty perceptions about remorse.


These perceptions, Bandes writes, routinely “influence sentencing hearings; parole, probation, and clemency determinations; forensic evaluations; decisions on whether to try a juvenile as an adult; and even (counter-intuitively) determinations of guilt or innocence.” And yet there is no legal consensus about how to define remorse, or how to properly detect its presence or absence based on a person’s demeanor. Citing a growing body of literature on the central role that evaluations of remorse play in our justice system, Bandes argues it’s time to stop pretending we know what remorse looks like and purge it from the moral calculus that goes into legal decision-making.


I spoke to Bandes, who served as a civil rights lawyer for the Illinois American Civil Liberties Union before becoming a specialist on the role of emotion in the law, by phone this week. Our interview has been edited and condensed for clarity.


Your area of specialization, the intersection of emotion and the law, is not something a lot of people think about very much. What drew you to it?

I think I was like most lawyers, in that I knew emotion played a role in jury trials and the way lawyers made arguments, but I believed that if you were really being rational and logical and rigorous you should put emotion aside. But then I read a couple of Supreme Court cases—one about victim impact statements [which are typically presented as part of the sentencing process], and the other about a young boy who had been very badly beaten by his father and was suing the state of Wisconsin [for allowing it to happen]. In the case of the abused boy, the court said that although it felt compassion for him, compassion had no legitimate role to play in legal decision-making. But then that same year, the court upheld the use of victim impact statements, and I thought, Wait, but victim impact statements are all about evoking compassion!


I was very perplexed, and I sort of naively thought that I could do a few hours of research and come up with an answer as to whether compassion has a role in law or not. That was about 20 years ago. It took me down this path that involves psychology and philosophy and sociology, and more recently, neuroscience—just trying to understand the role that emotions play in the law. I concluded that emotion pervades the law, and that often we’re so used to it that we often don’t even see it.

Your new paper is focused on remorse, and the way judges and jurors tend to think about it. Why is this an important issue?

People think they know what remorse looks like, and that they can tell just by observing someone’s facial expressions or body language whether they feel remorse or not. They also think remorse is really important when it comes to deciding what kind of punishment someone deserves—in part because they think that a person who is remorseful is less likely to commit another crime, and in part because they think that a remorseful person, or a person who looks remorseful, is someone who has a better character.


But once I started studying remorse, I found that there’s really no evidence at all to support the notion that we actually can evaluate it by looking at somebody. And so it’s a really powerful example of what goes wrong when we don’t face the fact that we are letting emotion—and our assumptions about emotion—affect our legal decisions.

What consequences can people’s perceptions of remorse have in legal settings?

Well, in capital cases, for instance, there is substantial evidence that whether or not a defendant looks remorseful is one of the main factors in determining whether a jury will sentence him to death or not.

Do people generally have similar intuitions about what remorse looks like?


No. There have been some very interesting survey-based studies showing that, for example, of all the judges who believed they could evaluate remorse based on facial expression and body language, about half thought that a failure to make eye contact showed remorse, and the other half thought that a failure to make eye contact showed a lack of remorse.

So it’s easy to be wrong about whether someone feels remorse or not.

There’s not yet any good evidence that we can read remorse from facial expressions or body language. There might be some eventually, but no one’s come up with any yet. What we do know is that there are a number of situations where we read it wrong. And so for example, race becomes a huge issue. There’s one study where the same white jurors looking at a white defendant versus a black defendant read their expressions differently. What is read as remorse or regret in a white defendant is read as coldness and arrogance and lack of remorse in a black defendant.


In your paper, you mention that being on psychotropic drugs can make it hard for people to express remorse.

Yes, the Supreme Court actually weighed in on this one—they held that if somebody is taking psychotropic drugs, it may interfere with his ability to exhibit remorse, and that forcing someone to take psychotropic drugs could make his trial unconstitutional because it’s so important to be able to show remorse to the jury.

Having certain kinds of mental disabilities or mental illnesses can make it difficult to show what people think of as appropriate remorse, too. Same with juveniles—juveniles are often misunderstood because their tough exterior is misread as a lack of remorse. Actually, there’s another thing that goes on with kids, too, which is that there’s evidence that sometimes it takes them a long time to really understand the gravity of what they’ve done. Once they get it, they get it. But that might be after they’ve already been on the witness stand. So that’s another big problem with remorse: If it unfolds slowly, you’re putting a lot of stock in the idea that it’ll be present and available for people to see at exactly the right time.


What’s the argument for considering a person’s level of remorse in the first place, though? Even if we could reliably identify its presence or absence, what would we be seeing evidence of?

I think we believe a remorseful person is either someone who committed an act that was very unlike them—that was inconsistent with their real character—or that they have seen the error of their ways and really want to change their character. There’s an assumption that someone who is remorseful is less likely to commit another crime in the future—that what he did was an aberration, a one shot deal and that he’s normally a really good person. But even if we could actually read character by looking at facial expression and body language, it’s not clear that we should be punishing people for having bad character, as opposed to their bad acts. And if you’re punishing someone for the crime they committed, then arguably, how they feel afterwards really shouldn’t matter at all.

There’s also just no evidence at all that someone who is remorseful during a trial is less likely to commit another crime. Because even if it’s sincere remorse, many people backslide. Someone might have the best of intentions and just not act on them. It’s also very hard to know whether someone’s apparent remorse is sincere.

And whether it’s being correctly interpreted as remorse.

Exactly. One important point is that there’s no evidence people can distinguish shame from remorse when they’re looking at someone, and yet we know that certain types of shame have been linked to an increased likelihood of committing another crime, because certain types of shame make people feel stigmatized and that actually pushes them away from society. So if you’re looking at someone’s face and you have no good way of telling whether they’re feeling shame or remorse, then you have a real problem if the whole reason you care about it is that you’re trying to determine whether they’re likely to offend again.


It sounds like you think that assessments of remorse don’t really have any place in the courtroom. What are some ways it could be neutralized as a factor?

I think it would be difficult to take remorse off the table entirely, because people are just so prone to wanting to determine whether someone is remorseful or not, that it would really be an uphill battle. But there are several things I think we could do. One is there could be jury instructions that say, “You can’t judge this person’s remorse from his facial expressions and body language. ”  I think you’d have to call people’s attention to the fact that they are doing it, and then call people’s attention to the difficulty of doing it.

Secondly, expert testimony would be really helpful here. You could have testimony about the difficulty of reading facial expressions in general and you could have testimony about the difficulty of reading remorse in particular, and the specific problems that come along because of race or psychotropic drugs and so on.

Of course, judges aren’t exempt from these misconceptions either, and for judges you really have to incorporate this into judicial training, and start training them on things like demeanor and body language and facial expression and what you can and can’t know from those.

It sounds like we can know very little.

Yeah. It’s such a great example of something that people are so sure they know. And sure enough that they’re actually willing to make decisions about life and liberty based on it. And there’s just so little to support it, and so much to contradict it.